american journal of critical care -...

Accuracy and Clinical Relevance of Physiological Monitor Alarms in the ICU Sinusitis Associated With Endotrachealand Nasogastric Tubes

Sedation During Weaning of Adults From Mechanical Ventilation

Rectal and Bladder vs Noninvasive Forehead Core Temperatures

Nurses’ Reflections on Family-Centered Care in PICUs

Predictors of Moral Distress in Critical Care Nurses

Outcomes of Acute Kidney Injury in ARDS Patients

January 2018 • Volume 27, Number 1

American Journal ofCritical Care

www.ajcconline.org

Evidence-based interdisciplinary knowledge for high acuity and critical care


AMERICAN ASSOCIATION OF CRITICAL-CARE NURSESPresident, CHRISTINE SCHULMAN, RN, MS, CNS, CCRN-K; President-elect, LISA RIGGS, RN, MSN, APRN-BC, CCRN-K; Secre-tary, MICHELLE KIDD, RN, MS, ACNS-BC, CCRN-K; Trea surer, LOUISE SALADINO, RN, DNP, MHA, CCRN-K; Directors, ELIZABETH BRIDGES, RN, PhD, CCNS; KIMBERLY CURTIN, RN, DNP, APRN, ACNS-BC, CCRN, CEN, CNL; JUSTIN DiLIBERO, RN, DNP, CCRN, CCNS, ACCNS-AG; NIKKI DOTSON-LORELLO, RN, BSN, CCRN, CPTC; WENDI FROEDGE, RN-BC, MSN, CCRN-K; DEBORAH JONES, RN, MS, PhD; MARY BETH FLYNN MAKIC, RN, PhD, CNS, CCNS, CCRN-K; ROSEMARY TIMMERMAN, RN, DNP, CCNS, CCRN-CSC-CMC; BETH WATHEN, RN, MSN, APRN, CCRN; Chief Executive Officer, DANA WOODS, MBA

EDITORIAL OFFICEAmerican Association of Critical-Care Nurses, 101 Columbia, Aliso Viejo, CA 92656. (800) 899-1712, (949) 362-2000. E-mail address: [email protected]. Web address: www.ajcconline.org Publishing Manager, MICHAEL MUSCAT; Managing Editor, KATIE L. SPILLER, MS; Art and Production Director, LeROY HINTON; Copy Editors, BARBARA HALLIBURTON, PhD; JULIE HENDERSON, RN, MS, ELS; LAURIE ANNE WALDEN, DVM, ELS; Graph ics Special-ist, MATT EDENS; Peer-Review Coordinator, DENISE GOTTWALD; Publishing Assistant, SAM MARSELLA

ADVERTISING SALES OFFICESLACK Incorporated, 6900 Grove Rd, Thorofare, NJ 08086. (800) 257-8290, (856) 848-1000. National Account Manager, NICOLE RUTTER; Recruit ment Sales Representative BERNADETTE HAMILTON; Admin-is trator, JOANN CAMPISI

AMERICAN JOURNAL OF CRITICAL CARE® Telephone: (949) 362-2000. Fax: (949) 362-2049. Copyright 2018 by AACN. All rights reserved. The AMERICAN JOURNAL OF CRITICAL CARE is an official peer-reviewed publication of AACN. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mech an ical, including photo copying, recording or by any information storage retrieval system, without per mission of AACN. For all permission requests, please contact the Copyright Clearance Center, Customer Ser-vice, 222 Rosewood Dr, Danvers, MA 01923. (978) 750-8400. Prices on bulk reprints of articles available on request from AACN at (800) 899-1712. Printed on acid-free paper. The AMERICAN JOURNAL OF CRITI-CAL CARE is indexed in MEDLINE/PubMed, Psy-cINFO, Cum u lative Index to Nursing & Allied Health Literature (CINAHL), Science Citation Index Expanded, ISI Alerting Services, Current Contents/Clinical Medi-cine, EBSCO collections, Ovid collections, and Scopus.

Annual individual subscriptions: US and possessions, $59; Canada and Mexico, $120 (US); all other countries, $120 (US). Institutional rates: US, print and online, $640; print only, $465; online only, $440. Institutional rates for all countries outside the US: print and online, $780; print only, $605; online only, $440. Single copies: US, $50; all other countries, $60 (US). For subscription questions: AACN members, please call toll-free (800) 899-2226 or (949) 362-2000; nonmembers, please call toll-free (800) 336-6348 or (818) 487-2075. Change of ad dress: Send notices to AMERICAN JOURNAL OF CRITICAL CARE, Subscrip-tion Service Depart ment, 101 Columbia, Aliso Viejo, CA 92656. Allow 4 to 6 weeks for change to take effect.

The statements and opinions contained in the articles in the AMERICAN JOURNAL OF CRITICAL CARE are solely those of the individual contributors and not of the editors or the American Association of Critical-Care Nurses. The edi-tors and the American Association of Critical-Care Nurses assume that articles emanating from a particular institu-tion are submitted with the approval of the requisite authority, including all matters pertaining to human stud-ies and patient privacy requirements. Advertisements in this journal are not a warranty, endorsement, or approval of the products by the editors of this journal or the Amer-ican Association of Critical-Care Nurses, who disclaim all responsibility for any injury to persons or property result-ing from any ideas or products referred to in the articles or advertisements.

Editors in Chief CINDY L. MUNRO, RN, PhD, ANPDean and Professor, School of Nursing and Health Studies, University of Miami, Coral Gables, Florida

RICHARD H. SAVEL, MDAdjunct Professor of Clinical Medicine and Neurology, SUNY Downstate College of Medicine, New York, New York

Clinical Advisers LINDA BELL, RN, MSNAmerican Association of Critical-Care Nurses Aliso Viejo, California

Founding Coeditors CHRISTOPHER W. BRYAN-BROWN, MD, and KATHLEEN DRACUP, RN, DNSc

Editorial Board

SARAH A. DELGADO, RN, MSN, ACNP-BCAmerican Association of Critical-Care NursesAliso Viejo, California

MICHAEL H. ACKERMAN, RN, DNS

Rochester, New York

THOMAS AHRENS, RN, DNS, CCRN

St Louis, Missouri

JOANN GRIF ALSPACH, RN, MSN, EdD

Annapolis, Maryland

JUDY L. BEZANSON, RN, DSN

Dallas, Texas

STIJN I. BLOT, RN, PhD

Ghent, Belgium

ELIZABETH J. BRIDGES, RN, PhD, CCNS, CCRN

Seattle, Washington

TIMOTHY G. BUCHMAN, PhD, MD, MCCM

Atlanta, Georgia

LINDA L. CHLAN, RN, PhD

Rochester, Minnesota

MARIANNE CHULAY, RN, DNSc

Southern Pines, North Carolina

MARTHA A. Q. CURLEY, RN, PhD

Boston, Massachusetts

RHONDA D’AGOSTINO, ACNP-BC

New York, New York

LYNN DOERING, RN, DNSc Los Angeles, California

BARBARA DREW, RN, PhD San Francisco, California

LEWIS A. EISEN, MD

Bronx, New York

DOUG ELLIOTT, RN, PhD

Sydney, New South Wales, Australia

SUSAN K. FRAZIER, RN, PhD Lexington, Kentucky

DORRIE K. FONTAINE, RN, DNSc

Charlottesville, Virginia

MARJORIE FUNK, RN, PhD

New Haven, Connecticut

MICHAEL A. GROPPER, MD, PhD

San Francisco, California

SANDRA HANNEMAN, RN, PhD

Houston, Texas

KATHRYN HAUGH, RN, PhD

Charlottesville, Virginia

STEVEN HOLLENBERG, MD

Camden, New Jersey

CONNIE JASTREMSKI, RN, MS, CNAA

Syracuse, New York

RUTH KLEINPELL, RN, PhD

Chicago, Illinois

CONSTANTINE MANTHOUS, MD

Bridgeport, Connecticut

PETER E. MORRIS, MD

Winston Salem, North Carolina

DEBRA K. MOSER, RN, DNSc Lexington, Kentucky

JANET D. PIERCE, DSN, ARNP Kansas City, Kansas

KATHLEEN PUNTILLO, RN, DNSc

San Francisco, California

MARY LOU SOLE, RN, PhD

Orlando, Florida

THEODORE A. STERN, MD

Boston, Massachusetts

M. CHRISTINE STOCK, MD

Chicago, Illinois

KATHLEEN M. VOLLMAN, RN, MSN, CCNS, CCRN

Detroit, Michigan

DOUGLAS WHITE, MD, MAS

Pittsburgh, Pennsylvania

SUSAN WOODS, RN, PhD

Seattle, Washington

Printed in the USA.

www.ajcconline.org AJCC AMERICAN JOURNAL OF CRITICAL CARE, January 2018, Volume 27, No. 1 1

Evidence-based interdisciplinary knowledge for high acuity and critical care

January 2018, Volume 27, No. 1

AMERICAN JOURNAL OF CRITICAL CARE® (Print ISSN 1062-3264, Online ISSN 1937-710X) is published bi monthly (January, March, May, July, September, Nov ember) by the American Association of Critical-Care Nurses (AACN), 101 Columbia, Aliso Viejo, CA 92656. Periodicals postage paid at Laguna Beach, CA, and additional mailing office(s). Postmaster: Send address changes to the AMER ICAN JOURNAL OF CRITICAL CARE, Subscription Service Depart ment, 101 Columbia, Aliso Viejo, CA 92656.

Coming in March …Vest and colleagues examine the association between enteral feeding, weight status, and mortality in a medical intensive care unit.

On the CoverDetail from “Burst”

Sharon Kleiman36'' x 48''

Acrylic2017

To view other works bySharon Kleiman,

visit her website at www.sharonkleiman.com

or contact (561) 305-9329,[email protected]

Delray Beach, FL

Patient Safety Issues

Critical Care Evaluation

Pulmonary Critical Care

Critical Care Techniques

11 Measurement of Physiological Monitor Alarm Accuracy and Clinical Relevance in Intensive Care UnitsHalley Ruppel, Marjorie Funk, and Robin Whittemore

24 Incidence of Sinusitis Associated With Endotracheal and Nasogastric Tubes: NIS DatabaseNorma A. Metheny, Leslie J. Hinyard, and Kahee A. Mohammed

32 Nurses’ Sedation Practices During Weaning of Adults From Mechanical Ventilation in an Intensive Care UnitMarta Borkowska, Sonia Labeau, Tom Schepens, Dominique Vandijck, Katrien Van de Vyver, Daphné Christiaens, Christelle Lizy, Bronagh Blackwood,

and Stijn I. Blot

43 Rectal and Bladder Temperatures vs Forehead Core Temperatures Measured With SpotOn Monitoring System Hildy M. Schell-Chaple, Kathleen D. Liu, Michael A. Matthay, and Kathleen A. Puntillo

Articles are available exclusively online at www.ajcconline.org

2 AJCC AMERICAN JOURNAL OF CRITICAL CARE, January 2018, Volume 27, No. 1 www.ajcconline.org

1 Idris et al. Circulation 2012; AHA Ress abstract #LBRS-352.2 Lurie KG et al. Chest 1998;113(4):1084-1090.3 Langhelle A et al. Resuscitation 2002;52:39-48.4 Yannopoulos D, et al. Critical Care Med. 2006;34(5):1444-1449.

Studies available upon request. The generally cleared indication for the ResQPOD ITD available for sale in the United States (US) is for a temporary increase in blood circulation during emergency care, hospital, clinic, and home use. The studies referenced here are not intended to imply specific outcomes-based claims not yet cleared by the US FDA.

Copyright © 2017 ZOLL Medical Corporation. ResQPOD and ZOLL are trademarks or registered trademarks of ZOLL Medical Corporation in the United States and/or other countries. All other trademarks are the property of their respective owners.

MCN HP 1702 0242

For more information and to watch a free educational animation, please visit www.zoll.com/animation

Use of the ResQPOD® Impedance Threshold Device (ITD) with high-quality CPR has been shown in clinical studies to improve survival by 25% or more1. In addition, pre-clinical studies have shown:

• Increased blood flow to the brain by 50%2

• Doubled blood flow to the heart3

• Doubled EtCO24

Attached to an airway during CPR, the ResQPOD ITD enhances negative pressure in the chest to pull more blood back to the heart and lower intracranial pressure non-invasively. This results in improved vital organ blood flow.

52 Nurses’ Reflections on Benefits and Challenges of Implementing Family-Centered Care in Pediatric Intensive Care UnitsHeather Coats, Erica Bourget, Helene Starks, Taryn Lindhorst, Shigeko Saiki-Craighill, J. Randall Curtis, Ross Hays, and Ardith Doorenbos

59 Predictors of Moral Distress in a US Sample of Critical Care NursesCatherine A. Hiler, Ronald L. Hickman, Jr, Andrew P. Reimer, and Kimberly Wilson

67 Outcomes of Acute Kidney Injury in Patients With Severe ARDS Dueto Influenza A(H1N1) pdm09 Virus Christopher J. Tignanelli, Arek J. Wiktor, Cory J. Vatsaas, Gaurav Sachdev, Michael Heung, Pauline K. Park, Krishnan Raghavendran, and Lena M. Napolitano

Families in Critical Care

Healthy Work Environments

Renal Issues

6 Editorial History Is a Tangled Mess: Medical Progress, But at What Cost? Richard H. Savel and

Cindy L. Munro

10 Clinical Pearls Rhonda Board

22 Evidence-Based Review and Discussion Points

Ronald L. Hickman

51 Patient Care Page Choosing Wisely for

Monitoring Patients Cindy Cain

74 Commentary Thoughts on the Immigration

Debate and Health Care: A Personal Story

Harold A. Fernandez

77 ECG Puzzler Prognostic ECG Changes in

a Preoperative Assessment Teri M. Kozik, Salah S. Al-Zaiti,

Mary G. Carey, and Michele M. Pelter

80 Education Directory

Visit AJCC’s website, www.ajcconline.org, to submit a manuscript or for author guidelines, full text of selected articles, OnlineNOW articles, digital edition access, links to AACN’s online continuing education tests, and more.

An Official Publication of the American Association of Critical-Care Nurses


Elevate Your Standard

of Care for Patient Handling

and Pressure Injury

Prevention with the

New Highly Breathable

HoverMatt SPU.

For your lateral patient transfer and positioning needs,

visit www.HoverMatt.com


HISTORY IS A TANGLED MESS: MEDICAL PROGRESS, BUT AT WHAT COST? By Richard H. Savel, MD, and Cindy L. Munro, RN, PhD, ANP

Editorial

This year’s events1 have caused us to revisit and reevaluate our nation’s checkered his-tory. Profoundly important and deeply fundamental issues such as the right to free speech while not inciting violence

are front and center in contemporary headlines. The seminal events highlighting these issues took place in Charlottesville, Virginia, on August 11-12, 2017, and began as a protest of the removal of a statue of Robert E. Lee from its location in a park. Similar con-troversies regarding other statues honoring Confeder-ate soldiers followed. This controversy reminded us all that the evil of slavery is one of the main reasons the United States fought the Civil War. It logically follows that the heroes of the Confederacy were, at least tacitly, proponents of slavery and should not be glorified in our current society. The controversial issue of removing statues celebrating Confederate Civil War heroes has spread throughout the nation and as far north as New York City. It now has touched medicine.

A large statue stands at 5th Avenue and 103rd Street in Manhattan, New York, dedicated to Dr. J. Marion Sims.2 Similar statues exist in Columbia, South Carolina, and Mobile, Alabama. Sims (1813-1883) is considered by many to be father of the

field of modern gynecology, helping to give this field gravitas and create it as a separate specialty. But con-troversy exists regarding his history and legacy because it is clear that his many accomplishments came about as the result of experimental surgery performed on enslaved black women—without their consent and without anesthesia.3-5 This information has given rise to protests in Manhattan to have his statue removed from its current neighborhood (but not destroyed), where some consider it offensive.6 In this editorial, we explore the challenging and provocative question of how we, in historical retrospect, judge the medical contributions of Sims.

Sims was originally from South Carolina, gradu-ated from Jefferson Medical College in 1835, and moved with his family to Mobile, Alabama, where he started his practice of surgery and gynecology. His controversial period—and coincidentally, some of his most significant work—began in 1845 and lasted until 1849. During this time, he performed opera-tions on slaves that he kept in a “hospital” (recorded by many as a shack) behind his house. According to multiple accounts, these operations were done with-out the women’s consent and without anesthesia.7-11 These women were treated primarily for vesicovagi-nal fistulae, a physically and socially debilitating complication of childbirth, for which, at that time, there was no cure. Sims came up with innovative

©2018 American Association of Critical-Care Nursesdoi:https://doi.org/10.4037/ajcc2018994


techniques to visualize the relevant anatomy and developed novel surgical procedures where none had been previously successful. He invented or cre-ated new equipment, techniques, and positions for patients for surgery; his name remains affixed to the Sims position and Sims speculum.

Sims then moved to New York City in 1853 (evidently in an attempt to improve some of his own chronic health issues). His career skyrocketed, and he became one of the founding physicians of the New York City “Woman’s Hospital,” described as the first hospital for women in America. He was in Europe from 1861 to 1871 (encompassing the Civil War period), where again he received great praise and accolades from the medical com-munity. Returning to New York in 1871, he remained an important, though often highly contentious, figure in medicine. As an example, at his hospital’s holiday party, he threatened to resign because of oper-ative restrictions that were being placed on him. The medical center took him up on his presumed bluff, separating him from the medical center he had helped found 20 years earlier. Undaunted, he went on to become president of the American Medical Association and was one of the originators of the New York Cancer Hospital (which eventually became Memorial Sloan Kettering Cancer Center). He remained busy and productive, despite having serious health issues with angina pectoris and typhoid, operating until a day before his death (presumably from a heart attack) at age 70.

The most vexing and thought-provoking parts of an exploration of Sims, his career, and his acco-lades are that the more one reads, the less clear the situation becomes. The facts surrounding his life and accomplishments are not controversial. It is the analysis of these events or, rather, determining the best manner in which to interpret his legacy, that remains of great debate. There also exists a

confusing and somewhat disconcerting disconnect between the perspective of the medical literature and the nonmedical literature regarding the merits of Sims’ work, with the medical literature (in general) shining a more favorable light on the physician.14 Given the complex history and our current vantage point, should Sims continue to be extolled as a groundbreaking surgeon, or should his actions be placed in the same category as Nazi physician atrocities of World War II12 or the Tuskegee syphi-lis experiment?13

As we stated in the beginning of this editorial, the disagreement surrounding Sims does not ques-tion any of the eventual outcomes: he did in fact develop new lifesaving techniques with a zeal and passion verging on the messianic. But it becomes evident upon evaluating Sims’ actions over the course of his lifetime that he did not feel that it was neces-sary to obtain informed consent or use appropriate analgesia and anesthesia during experimental sur-gery on patients who were slaves. Although the 2 potential perspectives that fall out of an analysis of Sims’ life are fairly clear, they are somewhat difficult to reconcile:

Perspective 1: He performed experimental surgery on enslaved women without informed consent and without anesthesia. This behavior is reprehensible, and he should not be the lauded hero of medical history that he has become. There is simply nothing to discuss: his immoral deeds speak for themselves.

Perspective 2: Sims’ actions should be interpreted in the context of the history in which he lived. This is not to excuse him simply because the actions and behaviors occurred in an era with different values—a concept known as “moral relativism.” These women had vesicovaginal fistulae, and Sims was doing what he could to help them and relieve their suffering. There were just no other effective alternative thera-pies known at the time. In terms of general anesthe-sia, most medical historians state that Sims did not consider these procedures major surgery, and one must recall that general anesthesia was only first safely documented by Morton in 18469—neverthe-less, his lack of use of anesthesia in these patients remains extraordinarily controversial. The issue of informed consent is an even more complex one to tackle: the ugly truth is that slaves at the time were considered property, and the idea behind consent

About the AuthorsRichard H. Savel is coeditor in chief of the American Journal of Critical Care. He is director, Adult Critical Care Services, Maimonides Medical Center and adjunct professor of clinical medicine and neurology, SUNY Down state College of Medicine, both in New York City. Cindy L. Munro is coeditor in chief of the American Journal of Critical Care. She is dean and professor, School of Nursing and Health Studies, University of Miami, Coral Gables, Florida.

[Sims] did in fact develop new lifesaving techniques with a zeal and passion verging on the messianic.


simply did not exist or apply. Let us not forget that the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research did not release their important Belmont report (summarizing ethical principles and guide-lines for research involving human subjects) until 1979, more than 130 years after Sims’ experimen-tal surgery.

Evidence also shows that Sims did not hide his racial or social biases in the development of his innovations. As one author states, “Throughout his medical career, Sims maintained a class bound pre-scription for the use of anesthesia with an unspoken premise that those women in the wealthy tier were by far the most vulnerable to pain.”15 It is important to note that there were contemporary rivals of Sims who strongly disagreed with his approach at the time, such as Dr James Simpson, who performed his surgical experiments on animals with appropri-ate anesthesia, rather than on enslaved human beings without anesthesia.7

We have delved into history to carefully examine the controversy and gain more facts regarding the rationale behind the laudatory statues of Dr J. Marion Sims, considered one of the fathers of gynecology. A large statue in his honor stands in a predominantly African American and Latino community in New York City.6 His experimental operations on enslaved black women named Anarcha, Betsey, Lucy, and others allowed him to receive numerous accolades and honors, as well as have a long and prestigious medical career.

Although the exploration of the relevant liter-ature did provide more facts regarding Sims and his career, we, as the editors of this journal, wish to make our stand on the issue quite clear: viewed through the lens and the laws of the 21st century, what Sims did was immoral and illegal, no matter how important the work turned out to be. He plainly performed experimental surgery on an extremely vulnerable patient population. We must use this case as an example in the medical community of where the ends cannot justify the means. Although

Sims’ work was of great medical import, it came at a dreadful—many would say unforgivable—moral cost, a cost that we in the medical and nurs-ing community should not make excuses for. The ends may have justified the means in Sims’ era, but they do not and cannot in ours. The statements and opinions contained in this editorial are solely those of the coeditors in chief.

FINANCIAL DISCLOSURESNone reported.

REFERENCES1. Unite the Right Rally | Wikipedia. 2017. https://en.wikipedia

.org/wiki/Unite_the_Right_rally. Accessed October 16, 2017.2. J. Marion Sims Statue NYC | Central Park Conservancy. 2017.

http://www.centralparknyc.org/things-to-see-and-do /attractions /dr-j-marion-sims.html. Accessed October 16, 2017.

3. Kaiser IH. Reappraisals of J. Marion Sims. Am J Obstet Gyne-col. 1978;132 (8):878-884.

4. Axelsen DE. Women as victims of medical experimentation: J. Marion Sims’ surgery on slave women, 1845-1850. Sage. 1985;2(2):10-13.

5. Ojanuga D. The medical ethics of the ‘father of gynaecology’, Dr J Marion Sims. J Med Ethics. 1993;19(1):28-31.

6. Gonzalez D. An antebellum hero, but to whom? N Y Times Web. 2017. August 18, 2017. https://www.nytimes.com/2017 /08/18/nyregion/j-marion-sims-statue-removal.html. Accessed October 16, 2017.

7. Sartin JS. J. Marion Sims, the father of gynecology: hero or villain? South Med J. 2004;97(5):500-505.

8. Wall LL. The medical ethics of Dr J Marion Sims: a fresh look at the historical record. J Med Ethics. 2006;32(6): 346-350.

9. Spettel S, White MD. The portrayal of J. Marion Sims’ con-troversial surgical legacy. J Urol. 2011;185(6):2424-2427.

10. Rosenbloom JM, Schonberger RB. The outlook of physician histories: J. Marion Sims and ‘The Discovery of Anaesthe-sia’. Med Humanit. 2015;41(2):102-106.

11. West MJ, Irvine LM. The eponymous Dr James Marion Sims MD, LLD (1813-1883). J Med Biogr. 2015;23(1):35-45.

12. Savel RH, Goldstein EB, Savel I, Savel H. Time does not heal all wounds: medical luminaries, national socialism, and the American College of Chest Physicians. Chest. 2007; 132(6):2064-2065.

13. Lerner BH, Caplan AL. Judging the past: how history should inform bioethics. Ann Intern Med. 2016;164(8): 553-557.

14. Washington HA. Medical Apartheid: The Dark History of Medical Experimentation on Black Americans from Colo-nial Times to the Present. New York, NY: Anchor; 2008.

15. McGregor DK. Sexual Surgery and the Origins of Gyne-cology: J. Marion Sims, His Hospital and His Patients. New York, NY: Garland Publishing; 1989.

To purchase electronic or print reprints, contact American Association of Critical-Care Nurses, 101 Columbia, Aliso Viejo, CA 92656. Phone, (800) 899-1712 or (949) 362-2050 (ext 532); fax, (949) 362-2049; e-mail, [email protected].

“ Although Sims’ work was of great medical import, it came

at a dreadful—many would say unforgivable—moral cost. ”

Universal ICU DecolonizationWithout Antibiotics

1. Huang SS et al. Targeted versus universal. N Engl J Med, 2013: 368(24): 2255-65. Mupirocin and CHG used in study. 2. Huang SS et al. Cost Savings of Universal Decolonization to Prevent Intensive Care Unit Infection. ICHE, 2014: 35 (S3): S23-S31. 3. Steed L, et al. Reduction of nasal Staphylococcus aureus carriage. AJIC, 2014: 42(8): 841-846. ©2017 Global Life Technologies Corp. All rights reserved. Made in USA. Nozin®, Nasal Sanitizer®, Leader in Nasal Decolonization™ are trademarks of Global Life Technologies Corp. Nozin® Nasal Sanitizer® antiseptic is an OTC topical drug. No claim is made that it has an effect on any specific disease. Savings estimates are for example only. Any actual savings may vary.

Compared to targeted screen and isolate, universal nasal and skin decolonization of ICU patients has shown superior infection outcomes.1 Also, reducing MRSA contact precautions can help lower costs while improving patient care.2 Now you can have fast, effective nasal decolonization without antibiotics.

• Safely Reduce MRSA Contact Precautions• 30 Bed ICU Could Save $600k per Year• Research Shows 44% HAI Reduction1

• Improve Patient Care and Satisfaction

Alcohol-based Nozin® Nasal Sanitizer® antiseptic is proven to decolonize nasal S. aureus day one.3 Team up with your Nozin advisor to assess the advantages of this program. Learn how you can implement universal ICU decolonization to help protect patients, improve care and lower costs.

LEADER IN NASAL DECOLONIZATION™

Evidence shows fewer infections and lower costs with universal nasal and skin decolonization.1, 2

Nozin.com 877-669-4648

Ask a Nozin advisor for a custom savings analysis and for professional samples.

Better Care. Lower Costs. ICU • MRSA CP

Clinical Pearls Rhonda Board, RN, PhD, CCRN, Section Editor

Clinical Pearls is designed to help implement evidence-based care at the bedside by summarizing some of the most clinically useful material from select articles in each issue. Readers are encouraged to photocopy this ready-to-post page and share it with colleagues. Please be advised, however, that any substantive change in patient care protocols should be carefully reviewed and approved by the policy-setting authorities at your institution.

Putting Evidence-Based Care in Your Hands

Nurses’ Reflections on Family- Centered Care in Pediatric Intensive Care Units

Family-centered care (FCC) enables and empowers families

through improved communi-cation and involvement in their child’s care. However, recent changes in FCC have incorporated a customer service emphasis that can be challenging for staff. The research team interviewed nurses to explore their perceptions of the benefits and challenges in providing FCC. They found that • Visitation 24 hours a day allows nurses to know families better but also means dividing nurses’ time between the child and the family. • Change to individual patient rooms (vs 1 open space) provides families with privacy and protection from infection and noise but facilitates nurse-nurse isolation, creating possible issues with patient safety and senior-to-junior role modeling. While acknowledging the benefits of FCC, Coats and colleagues recom-mend that (1) support systems be available for nurses dealing alone with families, (2) new nurses have a longer mentorship to become comfortable with providing care independently in private rooms, and (3) nurses be involved in the implementation of FCC practices.

See Article, pp 52-58

Comparison of Forehead Core Temperatures With Rectal and Bladder Temperatures

Body temperature readings provide vital information for accurate diagnoses and required therapies. However, standard

practices for measuring body temperature in intensive care units (ICUs) vary and often require invasive methods with limited accuracy in dif-ferent types of patients. Schell-Chaple and colleagues evaluated a new, noninvasive monitoring system that is applied to the lateral forehead, called SpotOn. They compared temperatures taken with the SpotOn system with rectal and bladder temperatures in febrile adult patients. They found the following: • Most of the noninvasive forehead tempera-tures and the rectal and bladder temperatures were within ±0.5ºC. • No signs of skin irritation were noted after 4½ hours of forehead temperature monitoring. The SpotOn method was accurate. Although the authors recommend further research to test for longer periods, they suggest this noninvasive system be considered for continuous monitoring of core temperature in ICU patients.


©2018 American Association of Critical-Care Nurses, doi:https://doi.org/10.4037/ajcc2018491

Nurses’ Sedation Practices in Mechanical Ventilation Weaning

Patients who require mechanical ventilation often receive sedative therapy to reduce anxiety and optimize com-fort and safety. Although various sedation protocols

have been successfully used to facilitate quick weaning from mechanical ventilation, there is no standardized practice despite guidelines from organizations such as the Society of Critical Care Medicine. Borkowska and colleagues surveyed more than 300 nurses from an adult intensive care unit and asked them to evaluate their daily sedation practices. They found the following: • Sedation protocols were routinely used by 62% of the nurses, although only 44% stated that a protocol was in place. • Barriers to daily interruption of sedation included con-cern about the patient’s comfort, respiratory deterioration, and possible self-removal of medical devices. • The attending physician was the clinician considered most involved in decision-making about sedation practice. Nurses’ perceptions, attitudes, and knowledge are vital to the success of any sedation practice. Organizational and multi-disciplinary collaboration is needed to improve sedation prac-tices at regional, national, and international levels.


Long-term Kidney Outcomes in Patients With Acute Respiratory Distress Syndrome Due to H1N1

The influenza A(H1N1) pandemic of 2009 to 2010 caused thousands of deaths across the world and left survivors with significant functional limitations and

morbidities. Many patients had acute renal failure requiring continuous renal replacement therapy (CRRT); however, the long-term consequences are unknown. Napolitano and colleagues examined the long-term recovery of adult patients in a surgical intensive care unit (ICU) with acute respiratory distress syndrome (ARDS) who tested positive for H1N1. They found the following: • About 95% had the most severe stage of acute kidney injury, with 67% requiring CRRT. • Patients requiring CRRT had a 53% mortality rate, higher illness severity, more frequent vasopressor use, and longer ICU stays than did patients not needing CRRT. • Of the patients who required CRRT and survived, 94% regained full renal function and were dialysis independent. This study is the largest to date to report renal outcomes in ARDS-H1N1 patients. The findings provide clinicians with valuable information needed to optimize future resource use.




Background Alarm fatigue threatens patient safety by

delaying or reducing clinician response to alarms, which

can lead to missed critical events. Interventions to reduce

alarms without jeopardizing patient safety target either

inaccurate or clinically irrelevant alarms, so assessment

of alarm accuracy and clinical relevance may enhance the

rigor of alarm intervention studies done in clinical units.

Objectives To (1) examine approaches used to measure

accuracy and/or clinical relevance of physiological mon-

itor alarms in intensive care units and (2) compare the

proportions of inaccurate and clinically irrelevant alarms.

Methods An integrative review was used to systemati-

cally search the literature and synthesize resulting articles.

Results Twelve studies explicitly measuring alarm accu-

racy and/or clinical relevance on a clinical unit were identi-

fied. In the most rigorous studies, alarms were annotated

retrospectively by obtaining alarm data and parameter

waveforms rather than being annotated in real time. More

than half of arrhythmia alarms in recent studies were inac-

curate. However, contextual data were needed to deter-

mine alarms’ clinical relevance. Proportions of clinically

irrelevant alarms were high, but definitions of clinically

irrelevant alarms often included inaccurate alarms.

Conclusions Future studies testing interventions on clin-

ical units should include alarm accuracy and/or clinical

relevance as outcome measures. Arrhythmia alarm accu-

racy should improve with advances in technology. Clinical

interventions should focus on reducing clinically irrele-

vant alarms, with careful consideration of how clinical

relevance is defined and measured. (American Journal

of Critical Care. 2018; 27:11-21)

MEASUREMENT OF PHYSIOLOGICAL MONITOR ALARM ACCURACY AND CLINICAL RELEVANCE IN INTENSIVE CARE UNITSBy Halley Ruppel, RN, MS, Marjorie Funk, RN, PhD, and Robin Whittemore, PhD, APRN

1.0 HourC EThis article has been designated for CE contact hour(s). See more CE information at the end of this article.

Patient Safety Issues


E RBEvidence-Based Review on pp 22-23

Alarm fatigue results from the multitude of alarms produced by medical devices in acute care settings and is a widely recognized patient safety concern.1-3 Alarm fatigue is desensitization to alarms3,4 associated with delayed or absent response to alarms.5,6 When nurses and other staff experience alarm fatigue, critical patient events may go unnoticed and preventable patient harm may occur. From 2009 to

2012, patient deaths were associated with 80 of the 98 alarm-related sentinel events voluntarily reported to the Joint Commission.7 The Joint Commission now requires hospitals to implement policies to address unnecessary alarms.2

More rigorous research on alarm

management is needed.


Clinicians, engineers, and hospital administra-tors are working to meet the Joint Commission requirements. Interventions to manage alarms can be technological, such as intelligent alarm algorithms implemented within the monitoring system, or clini-cal, such as alarm settings customized by nurses at the bedside. Technological interventions can be tested in existing databases of alarms,8 but the effectiveness of clinical interventions is more difficult to mea-

sure. Published interventions imple-mented in clinical settings have been primarily quality improvement proj-ects, often bundling several compo-nents together.9-13 Although these projects have resulted in a reduction of 43% to 89% of total alarms on the clinical units where they were

implemented,9-13 it was generally not possible to determine which component of the intervention was most effective. Moreover, the projects often lacked statistical evaluation and were not generalizable beyond the setting.9

More rigorous alarm intervention research is clearly needed.9 One additional method to enhance rigor is measurement of outcomes beyond reduction in the total number of alarms. Because physiological monitor alarm reduction strategies typically target either inaccurate or clinically irrelevant alarms,14,15 understanding which is more problematic for a clin-ical unit would not only lead to more specific inter-vention selection but would also be a more precise assessment of intervention effectiveness.16 In this review, we examine approaches to classifying

physiological monitor alarms by their accuracy (whether they reflect the patient’s actual condition) and their clinical relevance (whether they inform patient care).

Many terms have been ascribed to the concepts of alarm accuracy and clinical relevance; our use of these terms is presented in Figure 1. An accurate physiological monitor alarm condition corresponds to the patient’s actual status or to a technical prob-lem related to the monitor. An example of an inac-curate alarm is one for “ventricular tachycardia” that is caused by artifact from the patient moving. A clin-ical intervention such as a daily change of electrocar-diogram electrodes may reduce inaccurate alarms by decreasing artifact. A clinically relevant alarm is not only accurate but also useful to patient care. An alarm for an actual transient oxygen desaturation to 89% that requires no intervention is accurate but clinically irrelevant. An intervention such as having nurses customize alarms for a patient’s particular condition would reduce clinically irrelevant alarms.

Assessing alarm accuracy and clinical relevance requires time-consuming evaluation of each alarm in context but would result in more appropriate intervention selection and assessment of interven-tion effectiveness. Although intervention studies implemented on clinical units have not used pro-portions of inaccurate and clinically irrelevant alarms as outcome measures, the proportions of these types of alarms have been assessed in descrip-tive studies.9,17 The methods used in descriptive studies to measure alarm accuracy and clinical rele-vance may inform the design of future research that tests the effectiveness of interventions on clinical units. Therefore, the purposes of this integrative review were to (1) examine the approaches that have been used to measure physiological monitor alarm accuracy and/or clinical relevance in intensive care units (ICUs) and (2) compare proportions of inac-curate and clinically irrelevant alarms reported in the studies. We chose to focus our inquiry on ICUs, where physiological monitors are ubiquitous.

About the AuthorsHalley Ruppel is a doctoral candidate, Marjorie Funk is the Helen Porter Jayne and Martha Prosser Jayne pro-fessor of nursing, and Robin Whittemore is a professor at Yale School of Nursing, West Haven, Connecticut.

Corresponding author: Halley Ruppel, PO Box 27399, West Haven, CT 06516-7399 (email: [email protected]).


Methods We conducted this review using the integrative

review method described by Whittemore and Knafl.18 A medical librarian helped determine controlled vocabulary (Medical Subject Heading terms) and keywords related to the following: alarms AND physiological monitors (eg, monitor, telemetry, electrocardiography, oximetry) AND the clinical setting (eg, hospital, critical care, intensive care). One author (H.R.) searched 5 databases and adjusted search strategies according to syntax appropriate for each database. We limited the search to English-language articles but did not limit by publication date. We performed our search in April and May 2016 and confirmed it in October 2016. See Figure 2 for a flow diagram of the results.

One author (H.R.) screened titles and abstracts and read full texts, eliminating articles that did not meet the inclusion criteria. Two authors (H.R. and M.F.) discussed articles in question and reviewed full texts of all studies ultimately included in the review. We included studies if they met the following criteria: (1) explicitly sought to gather alarm data and anno-tate accuracy and/or clinical relevance of alarms (we broadly defined annotation as any attempt to catego-rize alarms by accuracy and/or clinical relevance, as defined by the researchers), (2) were full-text studies in peer-reviewed journals, and (3) were conducted with patients receiving continuous physiological monitoring using equipment standard for that ICU. We excluded studies if they used preexisting data-bases because we were interested in approaches used to collect, as well as to annotate, alarm data.

We entered data extracted from the articles into matrices to compare the studies across various fea-tures,18 including conceptual and operational definitions of accuracy and/or clinical relevance, measurement method for accuracy and/or clinical relevance, and proportions of inaccurate and/or clinically irrelevant alarms found. Because denomi-nators for alarm rates varied widely across studies (eg, alarms per unit per day or alarms per bed per hour), we opted not to include rates of alarms. Instead, we compared proportions of inaccurate and/or clini-cally irrelevant alarms (percentage of total annotated alarms). We examined studies in chronological order to determine evolution in methods over time, given the likely influence of improvements to monitoring technology and available software.

Results After removing duplicates, we reviewed 1787

records from the 5 databases. Twelve studies

Alarm

Clinically irrelevantb,d (eg, “low SpO2” that self-resolves and

does not inform patient care)

Clinically relevant (eg, “asystole” indicating

cardiac arrest)

Physiological Technical (eg, “leads off”)b

Inaccurateb,c (eg, artifact)

Accuratea

Figure 1 Schema for alarms contributing to alarm fatigue.

Abbreviation: SpO2, oxygen saturation.a Also referred to as “true” alarms.b Alarm types that may contribute to alarm fatigue. c Also referred to as “false” alarms.d Also referred to as “nonactionable” or “nuisance” alarms.

Figure 2 Results flow diagram.

Adapted from the PRISMA Group (Moher et al19).

Records identified through database searching:Ovid MEDLINE: 684

PubMed: 704CINAHL: 439Scopus: 415Embase: 382

Records after duplicates removed:

1787(837 duplicates)

Full-text articles assessed for eligibility:

26

Studies included in integrative review

12(+2 associated

methods articles)

Records excluded by title and abstract

1761

Full-text articles excluded: Not intensive care unit (5)

Did not report/unable to calculate proportions of accurate/relevant

alarms (6)Used data collected in a study

included in this review (1)


The proportion of patients’ alarms

requiring action by a clinician was

consistently low.

An integrative review method was used for this article.

ultimately met the inclusion criteria (Figure 2). Included studies were published from 1986 through 2015. Four of the studies were conducted in pediat-ric populations5,20-22 and 7 were conducted in adult populations.23-29 One study included both children and adults.30 Additionally, in 1 study,5 alarms in

both a pediatric general care area and a pediatric ICU were anno-tated; for this review, we used only pediatric ICU data. Most studies originated in the United States,5,20-23,26,29 with 1 study each from England,30 Finland,28 France,27

Germany,25 and Japan.24 All studies used observa-tional/descriptive designs; no intervention studies met inclusion criteria. Studies are presented in Table 1 in chronological order by publication year.

Approaches Used to Measure Alarm Accuracy and Clinical Relevance

The studies used 2 main approaches to measure alarm accuracy and clinical relevance: real-time annotation and retrospective annotation (Table 2). In studies published from 1986 through 1994, nurses and other staff working on the clinical units were asked to annotate alarms in real time.21,28-30 In studies published from 1997 through 2009, research personnel observed and annotated alarms in real time.22,26,27 In real-time annotation studies, alarms were primarily annotated on the basis of clinician response21,26-30 rather than whether an intervention was objectively warranted. Some of these studies included other devices (eg, infusion pumps),26,30 which do not produce alarms that can be categorized by clinical relevance. In 1 study, alarms were also annotated on the basis of corre-spondence to a life-threatening incident.30

In 2010, researchers began publishing on extracted and stored electronic alarm data and/or physiological parameter waveforms and retrospec-

tive alarm annotation.5,20,23-25 In 1 study,32 researchers obtained wave-forms from video recordings of monitor screens. Researchers used alarm data captured from the moni-toring system to determine the time of alarm and then reviewed the cor-responding video recording. In other studies, actual digitized wave-forms were stored for review.20,23,24

By evaluating waveforms to identify artifact during an alarm condition, researchers assessed alarms for accuracy. Some researchers also used video

recordings in patients’ rooms to annotate alarms’ clinical relevance by assessing patient status and interventions at the time of the alarm.24,31,32

Retrospective annotation of alarms offered important validity and reliability advantages over real-time annotation. In several studies using retro-spective annotation, more than 10 000 alarms were annotated,20,23,24 as compared with fewer than 4000 in real-time annotation studies.21,22,26-30 The use of multiple raters to verify annotations was reported in most studies using retrospective methods5,20,23,25 but not in studies using real-time annotation. Retro-spective annotation also allowed for more represen-tative samples because data from all patients during a specified time period could be reviewed. In contrast, not all alarms could be annotated in real time. Real-time annotation may also have been biased by selec-tion criteria such as “monitors likely to alarm.”22

Proportions of Inaccurate and Clinically Irrelevant Alarms

Overall, the studies indicated consistently low proportions of patient alarms requiring clinician action (5%-13%).5,26 However, it was difficult to parse out proportions of inaccurate and clinically irrelevant alarms because they were not often mea-sured as distinct concepts and were inconsistently operationalized. In half of the studies, inaccurate and clinically irrelevant alarms were combined, often because the measurement approach did not permit distinction.5,21,24,26-28 Examples of the various terms and definitions are presented in Table 3.

In real-time annotation studies, where alarms were often annotated on the basis of clinician response, proportions of inaccurate and clinically irrelevant alarms varied widely. For example, in 2 studies, clinicians annotated alarms as they responded to them. In one of these studies,28 27% of alarms were considered “artifact”; in the other,21 68% of alarms were considered “false.” When researchers annotated alarms by direct observation based solely on clinician response, 72% to 77% of alarms resulted in no response.26,27 However, not responding to an alarm does not necessarily indicate that the alarm is inaccurate or clinically irrelevant. In another study22 in which alarms were annotated by researcher observation but not on the basis of clinician response, 86% of alarms were considered inaccurate and only 6% were considered clinically irrelevant.

In the studies using retrospective annotation methods, operationalization of alarm accuracy was clearer. Arrhythmia alarms had the highest propor-tion of inaccurate alarms. In 2 studies20,23 annotating

Year, author, country

Table 1Articles measuring alarm accuracy and/or clinical relevance in intensive care units (chronological order by publication year)

1986O’Carroll30

England

1990 Koski et al28 Finland

1994Lawless21

United States

1994 Klaas and Cheng29

United States

1997 Tsien and Fackler22

United States

Nurses caring for the patients anno-tated alarm source, reason, action, and whether the alarm was associated with a life-threatening incident.

Nurses caring for the patients annotated the alarms in real time.

The staff person who silenced the alarm annotated the alarm in real time.

Nurse or respira-tory therapist who responded to alarm annotated alarm in real time.

Single observer from research team annotated alarms in real time.

Continued

Corresponded to a life- threatening incident: also called “potentially serious problem” but not specifi-cally defined

False patient monitor alarm: caused by removal of elec-trocardiography leads by nurse or patient

Artifact: a technical artifact alarm or an unimportant alarm < 15 seconds long

Undue: alarm > 15 seconds long but nurse took no action except briefly check-ing monitoring system and silencing alarm

Significant: patient’s condition checked; treatment implemented

False: technical monitor-ing device failure and accu-rate alarms in response to physiological changes that resolved before medical intervention necessary

Induced: by staff manipulation of patient but not judged to be clinically important

Significant: occurring either spontaneously or with manipulation of patient, resulting in a change in therapy

False: probe was off or malpo-sitioned, or signal was poor

True: stable correlation between oximeter pulse rate and electrocardio-graphic data

Led to change in treatment: eg, oxygen therapy, encour-aging cough/deep breathing, bronchodilator, diuretic

False positive: alarm was inap-propriate given input data value

True positive, clinically rele-vant: alarm was appropriate given the actual data and patient’s condition required prompt attention

True positive, clinically irrel-evant: alarm was appropri-ate given the input data but patient’s condition had not changed in a way that required additional medical attention

Provide data on the frequency and origins of alarms

Determine fre-quency and sig-nificance of audible limit alarms

Examine the pre-dictive value of alarms in iden-tifying clinically important events

Investigate how often the central alarm was detected before the alarm at the bedside monitor was noticed by staff in the inten-sive care unit

Determine positive predictive value of patient moni-toring alarms

Patient monitor alarms, ventilator alarms, infusion device alarms

Patient monitor limit alarms

Patient monitor alarms, ventilator alarms

Continuous pulse oximetry alarms

Patient monitor alarms

1455

1307

2176

74

2942

0.5b

75

27

60

12

68

27

6

35

65

47

86

8

6

Annotation method%Terminology and

definitionsa Primary aim or purposea

Types of alarms annotated

No. of alarms annotated


Table 1Continued

1999Chambrin et al27 France

2009Gorges et al26

United States

2010Siebig et al25,31

Germany

2013Inokuchi et al24

Japan

2013 Rosman et al20

United States

Research team (nurses) observed and annotated alarms and clini-cians’ responses in real time.

Research team (non-clinician) observed and annotated alarms and clini-cians’ responses in real time.

Retrospective anno-tation by physi-cians using video recording at the patient’s bed-side and numeri-cal measurements, alarms, and physi-ological parameter waveforms collected in 1-second intervals

Retrospective annotation by nurses and physi-cians using video recording at the patient’s bedside and numerical measurements, alarms, and physi-ological parameter waveforms saved continuously

Retrospective anno-tation by physicians using stored rhythm alarms, physio-logical parameter waveforms, and the patient’s medi-cal record

Continued

True positive: followed by an action whatever the reason, including technical problems solved by staff

False positive: led to “no action”

Effective: alarm-related action performed within 5 minutes of end of alarm (technical or patient intervention)

Ineffective: care provider did not enter room in response to alarm or was not present during alarm

Ignored: care provider in room during alarm but did not perform alarm-related action within 5 minutes of end of alarm

Technically false: does not correctly identify patient’s condition

Technically true: correctly reflects patient’s condition or technical problem

Alarm relevant: followed by diagnostic or therapeutic decision or solution of tech-nical problem

Alerting: technically true alarm that did not require immediate action but was judged helpful25

Not alarm relevant: not followed by diagnostic or therapeutic decision

Technically false: waveform was obviously artifact produced by movement or procedure

Technically true: alarms that did not meet the technically false criteria

Relevant: immediate clini-cal assessment and diag-nostic/therapeutic decision necessary

Helpful but not relevant: required clinical assessment but no diagnostic or thera-peutic decision

Irrelevant: not defined

False: artifactual

Assess the rele-vance of moni-toring alarms as a warning system

Observe alarms to identify methods for reducing false alarms

Generate a data-base of anno-tated alarms and report rate of alarms and their clinical validity

Determine if the proportion of clinically relevant alarms differs by monitoring device type and clinical severity of the patient

Test whether rou-tine daily review of rhythm alarms would reveal unrecognized but clinically impor-tant arrhythmias

Patient monitor alarms, ventilator alarms

Patient monitor alarms, ventila-tor alarms, intra-venous infusion device alarms, feeding pump alarms



Patient monitor: arrhythmia alarms only

3188

1214

5820

11 591

15 771

28

72

23

36

41

40

58

15

38

46

21

71

6

33

61

55c

Annotation method %Terminology and

definitionsa Primary aim or purposea




ability to determine alarm accuracy. In retrospective studies in which other monitor alarms (eg, thresh-old alarms) were also annotated, the proportion of inaccurate alarms was lower (21%-40%).5,24,25 In all 3 studies using video, alarms that were inaccurate were also considered to be clinically irrelevant, so proportions of “nonactionable” (87%),5 “not alarm

more than 12 000 arrhythmia alarms, inaccurate alarms accounted for 55% to 89% of alarms. In one of the first studies to employ retrospective annota-tion using stored waveforms, 104 arrhythmia alarms were annotated, but only 11% were inaccurate.25 However, methods for storing waveforms were less sophisticated, which may have affected researchers’


Table 1Continued

2014Drew et al23

United States

2015, 2014Bonafide et al5,32

United States

Retrospective anno-tation by nurses using stored alarms and physio-logical param eter waveforms

Retrospective anno-tation by research assistants and physician using video record-ing of patient’s room and monitor for physiological parameter wave-forms and stored alarm data

False positive: although not conceptually defined, a detailed protocol described criteria for true-positive and false-positive alarms on the basis of reflection of patient’s actual physiological status

Nonactionable: any alarm that does not lead to (or should not have led to) intervention or bedside consultation, including invalid alarms

Actionable: valid alarm that leads to clinical interven-tion or bedside consultation or that should have led to intervention or consultation but was missed by staff

Valid: correctly identifies the physiological status of the patient

Invalid: does not correctly identify physiological status of patient

Investigate the fre-quency, types, and accuracy of physiological monitor alarms

Determine associ-ation between nurses’ exposure to nonactionable physiological monitor alarms and response time to future alarms

Patient monitor: 6 arrhythmia alarms (asystole, ventricular fibril-lation, ventricu-lar tachycardia, accelerated ven-tricular rhythm, pause, ventricular bradycardia)


12 671

1014

89

87

13

76

24

Annotation method%Terminology and

definitionsaPrimary aim or purposea



a Terms are directly quoted from studies. Primary aim or purpose and definitions were paraphrased as little as possible from the studies to shorten their length but maintain the original meaning.

b Calculated from raw data provided by authors (8 of 1455 alarms corresponded to a potentially serious problem).c Calculated from raw data provided by authors. Authors also reported 72% false critical arrhythmia alarms and 50% false noncritical arrhythmia alarms.

Unit nurses or staff

Table 2Studies by alarm annotation method

O’Carroll30 (1986)

Koski et al28 (1990)

Lawless21 (1994)

Klaas and Cheng29 (1994)

Siebig et al25,31 (2010)

Inokuchi et al24 (2013)

Bonafide et al5,32 (2015, 2014)a

Rosman et al20 (2013)

Drew et al23 (2014)

Tsien and Fackler22 (1997)

Chambrin et al27 (1999)

Gorges et al26 (2009)

Stored alarms, physiological parameter waveforms, and video

recording at patient’s bedside

Retrospective annotationReal-time annotation

Stored alarms and physiological parameter waveforms

On-unit observers from research team

a Used video recording of patient’s monitor and patient’s room.


but also in the ways in which these terms were defined and measured.

Methods for annotating alarm accuracy and clinical relevance have progressed from real-time alarm annotation to retrospective annotation using stored alarm data, waveforms, and video. Large alarm datasets, including waveform data, are increasingly accessible as software becomes available to aggregate these data8; however, these datasets alone cannot be used to determine an alarm’s clinical relevance.34 We found reported proportions of clinically irrele-vant alarms particularly misleading because of con-ceptual inconsistencies and overlap with inaccurate alarms, highlighting the difficulty of measuring the clinical relevance of alarms.

Determining clinical relevance requires under-standing alarm context, including the clinical status of the patient. In the studies in this review, real-time observation and video of the patient were the primary methods used to assess alarm context. In the first study using video annotation, Siebig et al31 detailed the benefits of video annotation of alarms over real-time observation, including reduced Hawthorne effect and the ability to review an event multiple times. Subsequently, 2 studies24,32 incorporated video meth-ods to annotate clinical relevance. However, retro-spective annotation does not allow for real-time clarification of the clinical relevance of an alarm.31 Researchers using qualitative methods have demon-strated that observation provides a richer

relevant” (46%),25 and “irrelevant” (61%)24 alarms reported in these studies included both inaccurate and clinically irrelevant alarms.

Finally, although clinical relevance of an alarm was primarily a dichotomous variable (relevant or not), 2 studies included a third category of clinical relevance: “alerting” (38%)25 or “helpful but not relevant” (33%).24 These represented alarms that either “did not require an immediate action but nevertheless [were] judged helpful”25 or “required clinical examination but did not require a diagnostic or therapeutic decision.”24

Discussion In this review, we synthesized major approaches

that have been used to measure accuracy and clinical relevance of physiological monitor alarms in ICU settings and considered how these approaches

influenced the resulting proportions of inaccurate and clinically irrelevant alarms. Predictably, mean-ingful comparison of pro-portions of inaccurate and clinically irrelevant alarms was difficult because of the wide variation in operation-

alization of concepts. Others have noted variation in alarm terminology across studies.9,31,33 We explored variation not only in the terms themselves

Term

Table 3Examples of terms and definitions used for inaccurate and/or clinically irrelevant alarmsa

Definition

Inaccurate Technically false24

Technically false25

False21

False20

False positive22

False positive27

Artifact28

Clinically irrelevant Not alarm relevant25

Ineffective26

True positive, clinically irrelevant22

Induced21

Undue alarm28

Nonactionable5

Waveform that was obviously artifact produced by movement/procedureAlarm that does not correctly identify patient’s conditionTechnical monitoring device failure alarms and accurate alarms in response to physiological changes that

resolved before medical intervention necessaryArtifactualAlarm that was inappropriate given input data valueAlarm that led to “no action”Technical artifact alarm or an unimportant alarm less than 15 seconds long

Alarm not followed by diagnostic or therapeutic decisionCare provider did not enter the room in response to alarm/was not present during alarmAlarm that was appropriate given the input data but patient’s condition had not changed in a way that

required additional medical attentionAlarm that was induced by staff manipulation of patient but not judged to be clinically importantAlarm longer than 15 seconds but nurse took no other action except briefly checking monitoring system and

silencing alarmAlarm that does not lead to (or should not have led to) intervention or bedside consultation, including those

caused by motion artifact

a Terms are directly quoted from studies. Definitions were paraphrased as little as possible from the studies to shorten their length but maintain the origi-nal meaning.

Clinical interventions should be focused on

the problem of clinically irrelevant alarms.


understanding of the relevance of alarms to nursing staff,35 but this method may be impractical for anno-tating the large number of alarms required to detect the effectiveness of an intervention.

To measure clinical relevance, one should give careful consideration to what makes an alarm “clinically irrelevant.” An alarm that is deemed irrelevant by one clinician may be relevant to another. For example, a patient requiring endotra-cheal tube suctioning may generate an alarm indi-cating the need for routine nursing care, but typically this alarm would not change the patient’s diag-nosis or treatment plan. Especially in ICUs, nurses may use alarms to help them track changes in a patient’s condition.36 In 2 included studies,24,25 close to one-third of alarms fell into the category of alarms that were alerting but did not warrant intervention. These same alarms may have been considered clinically irrelevant in studies in which relevance was defined by the need for action or by alarms that corresponded to a life-threatening inci-dent. Clinical relevance needs to be considered from the perspective of all clinicians and should reflect alarms that may be informative even if not immediately actionable. Importantly, having clini-cians annotate alarms as they respond to them and annotating alarms by observing clinicians’ response are not useful measures of alarm relevance because clinicians affected by alarm fatigue may indiscrimi-nately reduce or delay response to alarms.

Methods of defining clinical relevance have important implications for determining which alarms are targeted for elimination on a clinical unit. Suggested interventions such as widening default alarm parameter settings for an entire unit may reduce the number of alarms but may have unintended consequences for patient safety if, for example, nurses’ awareness of their patients’ status is reduced. An intervention such as nurse-driven customization of alarms would allow nurses to maintain control over important alarms.

For assessing the accuracy of alarms, more recent studies using retrospective annotation methods were more rigorous. Within these studies, fewer than half of physiological monitor alarms were inaccurate, with the exception of arrhythmia alarms. Although this result indicates the need to address arrhythmia alarm accuracy, Drew et al23 found that inaccurate arrhyth-mia alarms were typically caused not by poor signal quality but rather by brief periods of artifact, such as that caused by the patient moving. Therefore, rou-tine electrode changes may not address these high proportions of inaccurate arrhythmia alarms.23

Engineers and vendors are focused on improving algorithms to enhance the accuracy of alarm sys-tems,8 and specific recommendations for improving monitoring systems are available.23

Directions for Future ResearchThis review builds on previous recommenda-

tions for more rigorous research on alarm interven-tions9 and for a focus on alarms’ informativeness16 (ie, accuracy and clinical relevance). Arrhythmia alarms appear to be most problematic from an accuracy perspective and should continue to be the focus of technical interventions, but they may not need to be the primary focus of clinical inter-ventions to address alarm fatigue. Alarm accuracy should continue to improve as technology improves. Clinical interven-tions ought to be focused on the problem of clini-cally irrelevant alarms. Several interventions that address clinical relevance have been discussed but need more rigorous evaluation. For example, an intervention such as changing default settings for an entire unit, although potentially effective, must be assessed for safety.9 Having nurses customize alarm settings at the patient’s bedside could reduce clinically irrelevant alarms, but that intervention has received little attention.

To enhance the rigor of future research that tests clinical alarm interventions, outcome measures should include not only the total rate of alarms but also the proportions of inaccurate and clinically irrelevant alarms. These outcome measures will help determine if an intervention has enhanced the accu-racy and clinical relevance of alarms on the unit. For measuring alarm accuracy, retrospective review of alarm data and parameter waveforms is more rigor-ous than direct observation because of the number of alarms that can be annotated by multiple review-ers. To determine the clinical relevance of an alarm retrospectively, researchers need additional informa-tion about the clinical context of the alarm, which is best obtained through video recording at the patient’s bedside and review of the medical record. The medi-cal record can provide insight into baseline and condition-specific factors that would make an alarm clinically relevant to the patient. Authors of several studies31,32 included in this review provided in-depth descriptions of their methods for annotating alarm accuracy and clinical relevance, which can provide guidance for future research. Researchers must clearly

Alarm accuracy should improve as technology advances.


justify how alarm accuracy and clinical relevance are defined and measured in their work.

LimitationsSeveral issues limit this integrative review. Only

1 author screened titles and abstracts and extracted data from articles. We included only English-language, ICU-based studies. The description of methods in the included articles was often limited, so this review may not have captured some aspects of the methods. Proportions of inaccurate and clinically irrelevant

alarms were most likely influ-enced by variation in the fol-lowing factors across studies: patients’ characteristics, such as severity of illness5,24; the monitoring system used; and the types of alarms and

devices included in the study. We did not limit our search to a specific period of time because we wanted to capture all methods that have been used to anno-tate alarm accuracy and clinical relevance. However, monitoring technology has changed substantially in the past 30 years, with artifact reduction strategies becoming more sophisticated8 and alarm types becoming more complex. These factors most likely changed the accuracy of alarm systems during the past 30 years, but to date, alarm system technology has not affected the clinical relevance of alarms. Finally, we grouped retrospective waveform review into a single category, but the studies used different methods for collecting waveform data, which may have affected the quality of the data obtained and could limit the comparability of these studies.

Conclusion Although the effectiveness of clinical alarm inter-

ventions is time-consuming to assess, future research in this area should include measurement of inaccu-rate and/or clinically irrelevant alarms as outcome variables. These measures may improve the quality of alarm intervention studies through improved selec-tion of interventions and more precise assessment of interventions’ effectiveness. In this review, we found that using retrospective methods to annotate alarms offers many advantages over real-time observation and that the emergence of software to extract alarm data and waveform parameters has made annotation of alarm accuracy relatively straightforward. However, clinical interventions should focus on enhancing clini-cal relevance. Measures of clinically irrelevant alarms often included inaccurate alarms or were inadequately operationalized, highlighting the challenges of

measuring clinical relevance. Additional informa-tion about the patient is needed to annotate the clinical relevance of alarms, and thoughtful defini-tion of clinical relevance is necessary to improve the quality of intervention studies.

ACKNOWLEDGMENTThe authors thank Janene Batten, MLS, a reference librar-ian at Cushing/Whitney Medical Library, Yale School of Nursing, for her assistance in determining controlled vocabulary and keywords for the literature search.

FINANCIAL DISCLOSURESHalley Ruppel is a Robert Wood Johnson Foundation Future of Nursing Scholar.

SEE ALSO For more about clinical alarms, visit the Critical Care Nurse website, www.ccnonline.org, and read the article by Lukasewicz and Mattox, “Understanding Clinical Alarm Safety” (August 2015).

REFERENCES1. Top 10 health technology hazards for 2016. ECRI Institute

website. https://www.ecri.org/Pages/2016-Hazards.aspx. Published November 2015. Accessed September 16, 2017.

2. National patient safety goals effective January 1, 2016. The Joint Commission website. http://www.jointcommission.org /assets/1/6/2016_NPSG_HAP.pdf. Published 2016. Accessed September 16, 2017.

3. Sendelbach S, Funk M. Alarm fatigue: a patient safety con-cern. AACN Adv Crit Care. 2013;24(4):378-386.

4. Cvach M. Monitor alarm fatigue: an integrative review. Biomed Instrum Technol. 2012;46(4):268-277.

5. Bonafide CP, Lin R, Zander M, et al. Association between exposure to nonactionable physiologic monitor alarms and response time in a children’s hospital. J Hosp Med. 2015; 10(6): 345-351.

6. Bliss JP, Gilson RD, Deaton JE. Human probability match-ing behaviour in response to alarms of varying reliability. Ergonomics. 1995;38(11):2300-2312.

7. Joint Commission. Medical device alarm safety in hospitals. Sentinel Event Alert. 2013;(50):1-3.

8. Clifford GD, Silva I, Moody B, et al. False alarm reduction in critical care. Physiol Meas. 2016;37(8):E5-E23.

9. Paine CW, Goel VV, Ely E, et al. Systematic review of physio-logic monitor alarm characteristics and pragmatic interven-tions to reduce alarm frequency. J Hosp Med. 2016; 11(2): 136-144.

10. Sendelbach S, Wahl S, Anthony A, Shotts P. Stop the noise: a quality improvement project to decrease electrocardio-graphic nuisance alarms. Crit Care Nurse. 2015;35(4):15-22.

11. Graham KC, Cvach M. Monitor alarm fatigue: standardizing use of physiological monitors and decreasing nuisance alarms. Am J Crit Care. 2010;19(1):28-34.

12. Dandoy CE, Davies SM, Flesch L, et al. A team-based approach to reducing cardiac monitor alarms. Pediatrics. 2014; 134(6):e1686-1694.

13. Whalen DA, Covelle PM, Piepenbrink JC, Villanova KL, Cuneo CL, Awtry EH. Novel approach to cardiac alarm management on telemetry units. J Cardiovasc Nurs. 2014; 29(5): E13-22.

14. Karnik A, Bonafide CP. A framework for reducing alarm fatigue on pediatric inpatient units. Hosp Pediatr. 2015;5(3): 160-163.

15. Schondelmeyer AC, Brady PW, Landrigan CP. Alarm fatigue: clearing the air. J Hosp Med. 2016;11(2):153-154.

16. Rayo MF, Moffatt-Bruce SD. Alarm system management: evidence-based guidance encouraging direct measurement of informativeness to improve alarm response. BMJ Qual Saf. 2015;24(4):282-286.

17. Imhoff M, Kuhls S. Alarm algorithms in critical care monitor-ing. Anesth Analg. 2006;102(5):1525-1537.

18. Whittemore R, Knafl K. The integrative review: updated methodology. J Adv Nurs. 2005;52(5):546-553.

Thoughtful definition of the clinical relevance

of alarms is needed.


1.0 Hour Category AC ENotice to CE enrollees:

This article has been designated for CE contact hour(s). The evaluation demonstrates your knowledge of the following objectives:

1. Define alarm “accuracy” and “clinical relevance.”2. Describe methodological challenges related to measuring alarm accuracy and alarm clinical relevance. 3. Discuss ways in which alarm accuracy and alarm clinical relevance can inform interventions to address alarm

fatigue.

To complete the evaluation for CE contact hour(s) for this article #A182701, visit www.ajcconline.org and click the “CE Articles” button. No CE evaluation fee for AACN members. This expires on January 1, 2021. The American Association of Critical-Care Nurses is an accredited provider of continuing nursing education by the

American Nurses Credentialing Center’s Commission on Accreditation. AACN has been approved as a provider of

continuing education in nursing by the State Boards of Registered Nursing of California (#01036) and Louisiana

(#LSBN12).

19. Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med. 2009; 151(4):264-269.

20. Rosman EC, Blaufox AD, Menco A, Trope R, Seiden HS. What are we missing? Arrhythmia detection in the pediatric intensive care unit. J Pediatr. 2013;163(2):511-514.

21. Lawless ST. Crying wolf: false alarms in a pediatric intensive care unit. Crit Care Med. 1994;22(6):981-985.

22. Tsien CL, Fackler JC. Poor prognosis for existing monitors in the intensive care unit. Crit Care Med. 1997;25(4):614-619.

23. Drew BJ, Harris P, Zègre-Hemsey JK, et al. Insights into the problem of alarm fatigue with physiologic monitor devices: a comprehensive observational study of consecutive inten-sive care unit patients. PLoS One. 2014;9(10):e110274.

24. Inokuchi R, Sato H, Nanjo Y, et al. The proportion of clinically relevant alarms decreases as patient clinical severity decreases in intensive care units: a pilot study. BMJ Open. 2013;3(9):e003354.

25. Siebig S, Kuhls S, Imhoff M, Gather U, Schölmerich J, Wrede CE. Intensive care unit alarms--how many do we need? Crit Care Med. 2010;38(2):451-456.

26. Görges M, Markewitz BA, Westenskow DR. Improving alarm performance in the medical intensive care unit using delays and clinical context. Anesth Analg. 2009;108(5):1546-1552.

27. Chambrin MC, Ravaux P, Calvelo-Aros D, Jaborska A, Cho-pin C, Boniface B. Multicentric study of monitoring alarms in the adult intensive care unit (ICU): a descriptive analysis. Intensive Care Med. 1999;25(12):1360-1366.

28. Koski EM, Mäkivirta A, Sukuvaara T, Kari A. Frequency and reliability of alarms in the monitoring of cardiac postoperative patients. Int J Clin Monit Comput. 1990;7(2):129-133.

29. Klaas MA, Cheng EY. Early response to pulse oximetry alarms with telemetry. J Clin Monit. 1994;10(3):178-180.

30. O’Carroll TM. Survey of alarms in an intensive therapy unit. Anaesthesia. 1986;41(7):742-744.

31. Siebig S, Kuhls S, Imhoff M, et al. Collection of annotated data in a clinical validation study for alarm algorithms in intensive care: a methodologic framework. J Crit Care. 2010; 25(1):128-135.

32. Bonafide CP, Zander M, Graham CS, et al. Video methods for evaluating physiologic monitor alarms and alarm responses. Biomed Instrum Technol. 2014;48(3):220-230.

33. Lukasewicz CL, Mattox EA. Understanding clinical alarm safety. Crit Care Nurse. 2015;35(4):45-57.

34. Welch J, Rayo M, Kanter B, et al. Framework for alarm management process maturity. Biomed Instrum Technol. 2016;50(3):165-179.

35. Gazarian PK, Carrier N, Cohen R, Schram H, Shiromani S. A description of nurses’ decision-making in managing elec-trocardiographic monitor alarms. J Clin Nurs. 2015; 24(1-2): 151-159.

36. Benner P, Tanner CA, Chesla CA. Appendix A. In: Expertise in Nursing Practice: Caring, Clinical Judgment, and Ethics. New York, NY: Springer Publishing Company; 1996:351-372.

To purchase electronic or print reprints, contact American Association of Critical-Care Nurses, 101 Columbia, Aliso Viejo, CA 92656. Phone, (800) 899-1712 or (949) 362-2050 (ext 532); fax, (949) 362-2049; email, [email protected].

Evidence-Based Review and Discussion PointsBy Ronald L. Hickman, RN, PhD, ACNP-BC

Evidence-Based Review (EBR) is the journal club feature in the American Journal of Critical Care. In a journal club, attend ees review and critique published research articles: an important first step toward integrating evidence-based practice into patient care. General and specific questions such as those outlined in the “Discussion Points” box aid journal club participants in probing the quality of the research study, the appropriateness of the study design and methods, the validity of the conclusions, and the implications of the article for clinical practice. When critically appraising this issue’s EBR article, “Measurement of Physiological Monitor Alarm Accuracy and Clinical Relevance in Intensive Care Units” (pp 11-21), consider the questions and discussion points outlined in the “Discussion Points” box.

A cross hospital settings, patients are moni-tored with a variety of devices as part of their care, and when these devices detect

changes in the patient’s status, an audible alarm is sounded to alert a registered nurse. In clinical areas where alarms are often triggered, nurses tend to experience a phenomenon referred to as alarm fatigue: desensitization to alarms set to alert a nurse to a change in a patient’s condition. Alarm fatigue is a notable patient safety concern. When alarms go unnoticed by registered nurses and other staff, the patient’s safety is likely to be compromised.

Despite recognition of the impact of alarm fatigue on patient safety, interventional research intended to reduce alarm fatigue remains limited. To advance interventional research in this import-ant area, the authors of this article conducted an integrative review of the literature. The purpose of the integrative review was 2-fold. The first aim was to describe approaches to measure accuracy or clinical relevance of physiological monitor alarms in intensive care units (ICUs), and the second aim was to compare the proportions of inaccurate and clinically irrelevant alarms across studies.

To conduct this integrative review, the authors used a systematic approach to identify published

studies. Studies were included if they sought to gather alarm data and anno-tated accuracy or clinical relevance of alarms, were available full-text in peer-reviewed journals, and had been con-ducted with patients undergoing continuous physiological monitoring with standard equipment for a particu-lar ICU. In total, the authors reviewed 1787 studies, and 12 studies met all of the inclusion criteria. Four of the studies were conducted in pediatric populations, 7 studies were focused on adult popula-tions, and 1 study included both chil-dren and adults. The included studies were published between 1986 and 2015. The majority (58%) of the studies took place in the United States.

The findings of this integrative review indicate a need for targeted interventions to reduce clinically irrelevant alarms. Some studies lacked contextual data necessary for determining alarm clinical relevance. Definitions of clinically irrelevant alarms sometimes included inaccurate alarms. In addition, the authors point out that proportions of clinically irrelevant alarms were high and note that more than half

Investigator SpotlightThis feature briefly describes the personal journey and background story of the EBR article’s investigators, discussing the circumstances that led them to undertake the line of inquiry represented in the research article featured in this issue.

Halley Ruppel, RN, MS, is a doctoral candidate at Yale University, School of Nursing, West Haven, Connecticut.

Ms Ruppel has nearly a decade of experience in providing care to critically ill children. She is an active member of the American Associa-tion of Critical-Care Nurses and serves on 2 national coalitions of the Associa-tion for the Advancement of Medical Instrumentation concerning alarm man-agement and safe use of complex technologies. Although Ruppel and her coauthors did not conduct primary research, Ruppel

explains that conducting a rigorous integrative review had its challenges. “I think a common misconception is that conducting a review is easier than conducting primary research,” she comments. For Ruppel, the process of syn-thesizing findings in a new and meaningful way, while staying true to the original research, took time and cre-ativity. She adds, “I felt very fortunate to have coauthors who understood the process and could guide me through it.”

©2018 American Association of Critical-Care Nurses, doi:https://doi.org/10.4037/ajcc2018822

Halley Ruppel


About the AuthorRonald L. Hickman is an associate professor and assistant dean for nursing research, Case Western Reserve University, Cleveland, Ohio.

Discussion Points

A. Description of the Study What is the significance of the problem

posed by the authors? What is the purpose of the study?B. Literature Evaluation What is the clinical significance of alarm

fatigue? What are the 3 classifications of alarms that contribute to alarm fatigue?C. Sample What were the inclusion and exclusion criteria used in this publication? What was the process for determining whether

or not a study met the inclusion criteria?D. Methods and Design Describe how data were collected for this

publication. Describe how the authors analyzed the data

collected. E. Results What were the major findings of this project? How can you use the findings of this project

to improve the quality of nursing care at your facility?

of arrhythmia alarms were inaccurate. On the basis of the results of their integrative review, the authors recommend that clinician interventions intended to reduce alarm fatigue be focused on clinically irrele-vant alarms with careful consideration of how clini-cal relevance is defined and measured.

Information From the AuthorsHalley Ruppel, RN, MS, lead author on this EBR

article, provides additional details about the inte-grative review. She comments that the review stems from her first-hand experience with rapid growth of technology in ICUs and alarm fatigue.

Ruppel comments, “Because of the extensive use of technology in the ICUs, I always considered alarms to be irritating, and sometimes overwhelming, but necessary elements of our work.” As alarm fatigue gained momentum as a patient safety issue, “I began to think more about the factors that influence alarm management,” she adds.

To better understand the complexities of mea-suring meaningful outcomes for alarm fatigue stud-ies, Ruppel felt that a critical appraisal and synthesis of the existing literature was needed. In particular, she says “As I was exposed to more research on alarm fatigue in acute care, I found that the terminology was inconsistent, and that made it difficult to synthesize findings.” For Ruppel and her coauthors, the integra-tive review was a necessary first step toward develop-ing effective interventions targeting alarm fatigue.

Implications for PracticeRuppel encourages readers of the American Journal

of Critical Care to understand the 2 categories of alarms that contribute to alarm fatigue. The first category is inaccurate alarms, which are often referred to as false alarms, and the second category is clinically

irrelevant alarms, which are described as nonaction-able alarms. “Professionals at the bedside, in partic-ular registered nurses, can help reduce alarm fatigue by customizing alarms to reflect their patient’s con-dition, which in turn decreases the frequency of clinically irrelevant alarms,” she explains. Accord-ing to Ruppel, one of the significant challenges for professionals attempting to address the problem of alarm fatigue is determining what alarm strate-gies will be most useful for their unit, and she aspires to generate evidence that will help inform clinicians and administrators in their efforts to reduce alarm fatigue and enhance patient safety.




Background Endotracheal and nasogastric tubes are

recognized risk factors for nosocomial sinusitis. The extent

to which these tubes affect the overall incidence of nos-

ocomial sinusitis in acute care hospitals is unknown.

Objective To use data for 2008 through 2013 from the

Nationwide Inpatient Sample database to compare the

incidence of sinusitis in patients with nasogastric tubes

with that in patients with an endotracheal tube alone or

with both an endotracheal tube and a nasogastric tube.

Methods Patients’ data with any of the following Inter-

national Classification of Disease, Ninth Revision, Clinical

Modification codes were abstracted from the database:

(1) 96.6, enteral infusion of concentrated nutritional sub-

stances; (2) 96.07, insertion of other (naso-)gastric tube;

or (3) 96.04, insertion of an endotracheal tube. Sinusitis

was defined by the appropriate codes. Weighted and

unweighted frequencies and weighted percentages

were calculated, categorical comparisons were made

by 2 test, and logistic regression was used to examine

odds of sinusitis development by tube type.

Results Of 1 141 632 included cases, most (68.57%) had

an endotracheal tube only, 23.02% had a nasogastric

tube only, and 8.41% had both types of tubes. Sinusitis

was present in 0.15% of the sample. Compared with

patients with only a nasogastric tube, the risk for sinus-

itis was 41% greater in patients with an endotracheal

tube and 200% greater in patients with both tubes.

Conclusion Despite the low incidence of sinusitis, a

significant association exists between sinusitis and the

presence of an endotracheal tube, especially when a

nasogastric tube is also present. (American Journal of

Critical Care. 2018; 27:24-31)

INCIDENCE OF SINUSITIS ASSOCIATED WITH ENDOTRACHEAL AND NASOGASTRIC TUBES: NIS DATABASEBy Norma A. Metheny, RN, PhD, Leslie J. Hinyard, PhD, MSW, and Kahee A. Mohammed, MD, MPH

Critical Care Evaluation


Endotracheal and nasogastric tubes are recognized risk factors for nosocomial sinus-itis1-6 and present a concern in intensive care units where both types of tubes are commonly used. For example, endotracheal intubation and mechanical ventilation was the third most commonly performed procedure in US hospitals in 2011.7 It is unclear, however, to what extent endotracheal and nasogastric tubes actually affect

the overall incidence of nosocomial sinusitis in acute care hospitals. A possible source for the desired information is the Nationwide Inpatient Sample (NIS) database, which is sponsored by the Agency for Healthcare Research and Quality.

Reports of the incidence of sinusitis in patients with endotracheal and nasogastric tubes have largely come from single-institution studies and vary con-siderably according to diagnostic criteria. Plain radiographs and computed tomography scans are commonly used to detect fluid collections in the sinuses; however, the latter is considered more sensitive.8 When a fluid collection is identified, a culture is usually performed to determine if the fluid is infected.9 The incidence of sinusitis is also affected by patient population. Most at risk are those with a facial or cranial injury.3 For example, a sinusitis incidence of 11.5% was found in a study of 208 critically ill patients with head injury10 and an incidence of 17.7% was found in a study of 111 simi-lar patients11 (computed tomography scans and pres-ence of purulent sinus fluid were used as diagnostic measures in both studies).10,11 A lower incidence of sinusitis (7.7%) was reported in a study of 340 patients in 2 medical intensive care units in a Mid-western hospital (computed tomography scan and culture of sinus fluid were used as diagnostic tests).12 Although endotracheal tubes can be inserted either nasally or orally, nasotracheal tubes present a higher risk for sinusitis by causing edema of the nasal passage and obstruction of sinus drain-age.6,13-15 In a prospective study that used plain radiographs to compare the incidence of sinusitis according to the route of endotracheal intubation in 111 patients (53 in the orotracheal group and 58 in

the nasotracheal group), sinusitis occurred more often in patients intubated by the nasal route (43.1%) than by the oral route (1.8%).16

Nasogastric tubes are commonly used for gastric decompression, enteral feedings, and medication administration in critically ill patients. As with nasally inserted endotracheal tubes, the body can perceive these devices as foreign; thus, their pres-ence may predispose to inflammation and sinus-itis, especially if the tubes are left in place for a prolonged period.17-19 Mechanical irritation of the tissues is more pro-nounced with firm, large-diameter tubes than with small-diameter, flexible tubes. In tube-fed patients, formula may enter the nasal pas-sages after regurgitation (or per-haps by a wicking effect from the tube), predisposing to microbial growth.12 Any tube in the nasal passage acts as a locus for biofilm formation, which could enhance upper airway colo-nization and predispose to sinusitis.20 In a study21 of 47 adult patients in intensive care units who had a current or recent nasogastric tube placement, 27 had a unilateral positive culture; sinusitis developed on the same side as intubation in 25 of the 27 patients.

The objectives of the present study were (1) to use information in the NIS database for the years 2008 through 2013 to compare the incidence of sinusitis in patients with nasogastric tubes with the incidence in patients with an endotracheal tube only or a combination of an endotracheal tube and a nasogastric tube; and (2) to describe the extent to which the following variables were associated with the diagnosis of sinusitis in the patients specified in the preceding objective: sex, age, race, type of admis-sion (elective or nonelective), insurance status, and died in hospital.

Methods The NIS database is the largest of all payer data-

bases in the United States and contains discharge

About the AuthorsNorma A. Metheny is a professor of nursing, Saint Louis University, St Louis, Missouri. Leslie J. Hinyard is an associate professor of health outcomes research, Saint Louis University Center for Health Outcomes Research, and assistant director, Saint Louis University Center for Interprofessional Education and Research. Kahee A. Mohammed is a resident physician, Department of Inter-nal Medicine, Saint Louis University Center for Health Outcomes Research.

Corresponding author: Norma A. Metheny, RN, PhD, Saint Louis University, 3525 Caroline Mall, St Louis, MO 63104 (email: [email protected]).

The incidence of sinusitis in intubated patients is uncertain.


abstracts of approximately 8 million annual hospital stays from about 1000 hospitals, which represents a random stratified sample of 20% of all hospital admis-sions to nonfederal hospitals in the United States. Hospitals are stratified according to geographic region, ownership, bed size, and teaching status.22 The data-base contains clinical and resource-use information. Furthermore, patients’ diagnoses were documented with International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) codes and clinically meaningful clusters of ICD-9-CM codes, termed Clinical Classification Software codes.23 The database’s sampling method permits the application of weighted variables that allow for the calculation of national estimates that have been validated against other hospital registries in the United States.

We abstracted data from the NIS database for the years 2008 through 2013 for all patients with any of the following ICD-9-CM codes in any proce-dure field: (1) 96.6, enteral infusion of concentrated nutritional substances; (2) 96.07, insertion of other

(naso-)gastric tube (intubation for decompression); or (3) 96.04, insertion of endotracheal tube. Patients whose records included procedure code 96.6 plus a diagnosis code of “V441 gastrostomy tube” were excluded in an attempt to ensure that most of the enteral feeding tubes included in the study were inserted nasally (although a few may have been inserted orally). Thus, for purposes of this study, codes 96.6 and 96.07 were used as a proxy for naso-gastric tubes. Patients with an ICD-9 code for facial trauma or fracture were excluded, as were patients younger than 18 years.

The primary independent variable was the type of tube, which was classified as nasogastric tube, endotracheal tube, or both nasogastric and endotra-cheal tubes. The primary outcome variable was diag-nosis of acute sinusitis during the hospitalization. Patients were considered to have sinusitis if they had an ICD-9-CM diagnosis code of 461, 461.0, 461.1, 461.2, 461.3, 461.8, or 461.9 in any diagnosis field. A secondary analysis examined in-hospital mortality between patients with and without sinusitis.

Unweighted and weighted frequencies, weighted percentages, and standard errors are reported for demographics of the full sample. Categorical com-parisons were made using a 2 test. Logistic regres-sion was used to examine the odds of sinusitis by type of tube (nasogastric, endotracheal, or both) adjusted for age, race, and hospital length of stay, and to examine the crude association between sinusitis and inpatient mortality. All analyses were conducted in SAS 9.4 (SAS Institute), and SAS sur-vey procedures were used to account for the sam-pling scheme of the NIS data.

Results During the period for which data were collected

(2008 through 2013), the number of hospitaliza-tions included in the database was 46 209 705. Of these, only 1 141 632 cases contained a code for an endotracheal tube, a nasogastric tube, or a combina-tion of both; met inclusion and exclusion criteria; and were included in the sample. Most cases (68.57%) had a code for an endotracheal tube only, less than a quarter (23.02%) had a code for a nasogastric tube only, and 8.41% had ICD-9-CM codes for both a nasogastric tube and an endotracheal tube. Table 1 lists the demographic characteristics of the study sample. Just less than half of the patients in the sample (47.32%) were female and slightly more than half of the patients (52.06%) were age 65 years or older. The sample was predominantly non-Hispanic white (68.03%) and most were insured (94.11%). Most

Characteristic

Table 1Frequency and weighted percentage for sample demographics (N = 1 141 632, unweighted)

Sinusitis Yes No

Type of tube Nasogastric Endotracheal Both nasogastric and endotracheal

Sex (n = 1 141 400) Female Male

Age, years 18-44 45-64 65

Race (n = 1 031 443) Non-Hispanic white Non-Hispanic black Hispanic Other

Type of admission (n = 1 138 642) Nonelective Elective

Length of stay, days 1-2 3-4 5

Insurance (n = 1 100 099) Insured Uninsured

0.15 (0.004)99.85 (0.004)

23.02 (0.04)68.57 (0.04) 8.41 (0.03)

47.32 (0.05)52.68 (0.05)

13.88 (0.03)34.06 (0.04)52.06 (0.05)

68.03 (0.04)16.68 (0.04) 8.74 (0.03) 6.55 (0.02)

89.27 (0.03)10.73 (0.03)

15.55 (0.03)12.32 (0.03)72.13 (0.04)

94.11 (0.02) 5.88 (0.02)

82385 651 378

1 302 5863 880 936

476 093

2 677 5312 980 918

785 4901 927 4922 946 633

3 481 725853 450447 320335 008

5 039 424605 536

880 115697 214

4 082 287

5 133 369321 053

16681 139 964

262 828782 86995 935

540 338601 062

158 512388 947594 173

702 530171 54590 01067 358

1 016 381122 261

177 564140 756823 312

1 035 33664 763

Weighted % (SE)

Weighted N

Unweighted N


Sinusitis was diagnosed in < 1% of the > 1 million patients in the sample.

patients (89.27%) had nonelective hospital admis-sions and hospital stays longer than 5 days (72.13%).

Sinusitis was present in 0.15% of the 1 141 632 cases. Table 2 lists bivariate comparisons of individ-uals with and without sinusitis. The proportion of patients with an endotracheal tube and the propor-tion of patients with both an endotracheal tube and a nasogastric tube were higher in the group of patients with sinusitis than in the group of patients without sinusitis ( 2 = 73.7; P < .001). Women were less represented among those with sinusitis ( 2 = 9.4; P = .002). Persons with sinusitis were younger than those without sinusitis ( 2 = 262.8; P < .001). The presence of sinusitis did not differ among race/eth-nicity groups ( 2 = 5.9; P = .12). Patients with sinus-itis had longer stays ( 2 = 44.8; P < .001), were more likely to have had a nonelective admission ( 2 = 30.1; P < .001), and were less likely to be insured ( 2 = 22.0; P < .001) than were patients without sinusitis.

The results of crude and adjusted logistic regres-sion models that examined the odds of having sinus-itis by tube type are presented in Table 3. The adjusted logistic regression model included only those vari-ables that were statistically significant in the bivari-ate analysis. In the unadjusted model, the odds of sinusitis developing in patients with an endotracheal tube were 43% greater (odds ratio [OR], 1.43; 95% CI,

1.25-1.63) than in patients with a nasogastric tube only. The odds of sinusitis developing in patients with both a nasogastric tube and an endotracheal tube were 200% greater (OR, 2.17; 95% CI, 1.81-2.60) than in patients with a nasogastric tube only. After adjustment for sex, age, length of stay, elective admission, and insurance status, the rela-tionship between tube type and sinusitis remained. In the adjusted model, the odds of sinusitis developing in patients with an endotra-cheal tube were 41% greater (OR, 1.41; 95% CI, 1.23-1.62) and those with both a nasogastric tube and an endotracheal tube were 200% greater (OR, 2.05; 95% CI, 1.70-2.48) than in those with a nasogastric tube only. In the adjusted model, sex and insurance status were not significant predictors of sinusitis.

The odds of sinusitis developing decreased as age increased, with patients ages 45 through 64 years being 34% less likely to have sinusitis (OR, 0.66; 95% CI, 0.58-0.75) and patients age 65 years and older being 63% less likely to have sinusitis (OR, 0.37; 95% CI, 0.32-0.42) compared with patients ages 18 through 44 years. Longer stay was associated with increased odds of having sinusitis. The odds of

Sinusitis No sinusitis

Table 2Bivariate comparisons

Tube type Nasogastric Endotracheal Both

Female

Race Non-Hispanic white Non-Hispanic black Hispanic Other

Age, years 18-44 45-64 65


Elective admission

Insured

< .001

.002

.12

< .001

< .001

< .001

< .001

73.7

9.4

5.9

262.8

44.8

30.1

22.2

23.02 (0.04)68.76 (0.04) 8.40 (0.03)

47.32 (0.05)

68.03 (0.05)16.68 (0.04) 8.74 (0.03) 6.55 (0.02)

13.86 (0.03)34.04 (0.04)51.09 (0.05)

15.56 (0.03)12.32 (0.03)72.12 (0.04)

10.73 (0.03)

94.12 (0.02)

16.53 (0.90)70.38 (1.12)13.08 (0.82)

43.55 (1.21)

69.98 (1.19)16.96 (0.97) 7.68 (0.69) 5.38 (0.59)

24.31 (1.06)41.73 (1.21)33.96 (1.16)

9.70 (0.73)12.14 (0.80)78.16 (1.02)

6.59 (0.60)

91.31 (0.71)

1 301 2253 875 138

475 015

2 673 945

3 476 540852 194446 751334 609

783 4881 924 0542 943 835

879 316696 214

4 075 848

604 994

5 126 281

136257981078

3586

51851257

568.81398.63

200334382798

798.8410006439

541.62

7088

262 552781 693 95 719

539 611

701 480171 291 89 894 67 278

158 107388 250593 607

177 402140 554822 008

122 149

1 033 901

2761176 216

727

1050 254 116 80

405 697 566

162 2021304

112

1435

P2Weighted

% (SE)Weighted

% (SE)Weighted

NWeighted

NUnweighted

NUnweighted

NCharacteristica

a Missing unweighted data: sex (n = 232), elective admission (n = 2990), and insured (n = 41 533).


sinusitis developing in patients with a 3- to 4-day length of stay were 64% greater (OR, 1.64; 95% CI, 1.32-2.03) and in those with a length of stay of 5 or more days were 87% greater (OR, 1.87; 95% CI, 1.57-2.22) than in patients with a 1- to 2-day length of stay.

Elective admission was associated with a 36% reduced odds of having sinusitis (OR, 0.64; 95% CI, 0.53-0.78) compared with nonelective admission.

Table 4 presents the results of the crude and adjusted models examining the odds of in-hospital mortality associated with sinusitis. In the unadjusted model, patients with sinusitis were 61% less likely to die in the hospital (OR, 0.39; 95% CI, 0.34-0.45) compared with those without sinusitis. After adjust-ment for type of tube, sex, age, length of stay, type of admission, and insurance status, patients with sinus-itis were 56% less likely to die in the hospital (OR, 0.44; 95% CI, 0.38-0.52) compared with those with-out sinusitis.

Compared with those with a nasogastric tube only, patients with an endotracheal tube and patients with both an endotracheal tube and a nasogastric tube were more likely to die in the hospital (OR, 5.93 [95% CI, 5.84-6.03]; and OR, 5.79 [95% CI, 5.67-5.92], respectively). Patients in the age groups 45 to 64 years and 65 years and older were more likely to die in the hospital than those 18 through 44 years old (OR, 2.45 [95% CI, 2.41-2.50]; and OR, 4.52 [95% CI, 4.44-4.61], respectively). Elective admission was associated with 21% reduced odds of in-hospital mortality (OR, 0.79; 95% CI, 0.78-0.81), and being insured was associated with 15% increased odds of in-hospital mortality (OR, 1.15; 95% CI, 1.12-1.17).

Discussion The 0.15% incidence of sinusitis identified in our

sample of 1 141 632 patients in the NIS database is much lower than incidences of 1.8% to 43.1% found in several single-site clinical studies10-12,16 specifically designed to detect sinusitis. It is unlikely that clini-cians who provided care to the patients in the NIS database sample applied equal zeal to searching for sinusitis as a source of infection. It has been sug-gested that nosocomial sinusitis is underestimated in clinical settings because objective tests for the condition may not be performed as often as war-ranted by signs and symptoms.4,24 It is also possible that the diagnosis of sinusitis is missed in patients who have a more obvious source of infection (eg, pneumonia or a bloodstream infection, both known to be associated with sinusitis).20 Several studies have shown that nosocomial sinusitis episodes are fre-quently associated with nosocomial pneumonia with similar pathogenic organisms.16,25,26 Treatment of the more obvious infection site may lead to resolu-tion of an undiagnosed sinus infection.20

According to a National Nosocomial Infection Surveillance System report of 498 998 patients from

Model

Table 3Crude and adjusted odds of sinusitis by type of tube

Crude model Type of tube Nasogastric Endotracheal Both nasogastric and endotracheal

Adjusted model Type of tube Nasogastric Endotracheal Both nasogastric and endotracheal

Female

Age, years 18-44 45-64 65


Elective admission

Insured

Reference1.43 (1.25-1.63)2.17 (1.81-2.60)

Reference1.41 (1.23-1.62)2.05 (1.70-2.48)

0.92 (0.83-1.02)

Reference0.66 (0.58-0.75)0.37 (0.32-0.42)

Reference1.64 (1.32-2.03)1.87 (1.57-2.22)

0.64 (0.53-0.78)

0.94 (0.78-1.13)

Odds ratio (95% CI)

Model

Table 4Crude and adjusted odds of in-hospital mortality for sinusitis

Crude model Sinusitis

Adjusted model Sinusitis

Type of tube Nasogastric Endotracheal Both nasogastric and endotracheal

Female

Age, years 18-44 45-64 65


Elective admission

Insured

0.39 (0.34-0.45)

0.44 (0.38-0.52)

Reference5.93 (5.84-6.03)5.79 (5.67-5.92)

0.91 (0.90-0.92)

Reference2.45 (2.41-2.50)4.52 (4.44-4.61)

Reference0.36 (0.35-0.36)0.19 (0.18-0.19)

0.79 (0.78-0.81)

1.15 (1.12-1.17)

Odds ratio (95% CI)


Risk for sinusitis was greatest in patients with both an endo-tracheal tube and a nasogastric tube.

205 mixed medical-surgical intensive care units (1992-1998), 658 of the patients (0.13%) had a diagnosis of nosocomial sinusitis.27 This is very similar to the 0.15% incidence found in our NIS database study covering the years 2008 through 2013. However, the relationship between endotracheal and nasogas-tric tubes and sinusitis in the National Nosocomial Infection Surveillance System study cannot be deter-mined, because use of these tubes was not addressed.

Although we found 41% greater odds of sinus-itis developing in patients with endotracheal tubes, it was not possible to compare the incidence of sinusitis according to the route of tube insertion (ie, nasally or orally), because the NIS database does not provide information about insertion sites. Our finding that patients with ICD 9-CM codes for both an endotracheal and a nasogastric tube had 200% greater odds of sinusitis developing corrobo-rates findings from previously reported clinical stud-ies. For example, George et al,12 in a study of critically patients with and without sinusitis, found that nasoenteric intubation was present in 27 of the 28 patients with a endotracheal tube and sinusitis, as opposed to being present in 210 of the 338 patients without sinusitis (P < .001). In that study,12 the cumu-lative incidence of sinusitis was 11.4% among patients who underwent any nasoenteral intubation; in con-trast, the incidence was only 0.8% in patients who received no nasoenteric intubation.

Older patients in our study were at less risk for sinusitis. Others have reported that age older than 60 years was protective against worsening sinus opacification in patients in intensive care units.28 It is possible that the younger individuals were trauma victims who were more likely to require both endo-tracheal and nasoenteral intubation for a longer time than did older patients with medical conditions. It is likely that the 36% reduced odds of sinusitis devel-oping in patients who were admitted electively can be explained by a lower acuity level than in patients admitted nonelectively.

A possible explanation for the finding that patients with a hospital length of stay of 5 days or longer had 87% greater odds of sinusitis developing than patients with a length of stay of 1 to 2 days is that the former may have had endotracheal tubes and/or nasogastric tubes in place longer. As shown in multiple studies, the longer an endotracheal tube is in place, the greater is the risk for sinusitis.29-31 For example, in a study31 in which computed tomog-raphy scans were used to monitor 16 critically ill patients with nasotracheal tubes, 6 had signs of sinusitis between the second and third days, and all had signs of sinusitis by the eighth day of

intubation.Others have emphasized increased risk for sinusitis in patients who have had a nasotracheal tube for 5 days or longer.32,33

A strange finding was that patients with sinusitis in our study were less likely to die in the hospital than were patients without sinusitis. Reports vary as to the risk for death associated with sinusitis. A meta-analysis of 2 studies that reported on mortality of patients with ventilator-associated sinusitis indicated that the condition was not associated with excess mortality (when compared with patients without ventilator-associated sinusitis).20 This is in contrast to a 1989 report of 53 patients with endotracheal tubes and sinusitis; 20 had a lower respiratory tract infection develop, 10 had septicemia develop, and 6 of the 53 patients died as a result of their infection.25

Prevention of SinusitisThe orotracheal route for intubation has been

recommended to prevent sinusitis14 and ventilator-associated pneumonia.34 Removal of all nasal devices as quickly as feasible is also recommended.3 A ran-domized study of the ability of topically applied nasal decongestants and corticosteroids to reduce sinusitis in a group of critically ill trauma patients receiving mechanical ventila-tion did not show a significant effect.35 According to Riga et al,36 the clinical significance of this approach in critically ill patients has not been suffi-ciently addressed. The compre-hensive evidence-based clinical practice guidelines for preven-tion of ventilator-associated pneumonia34 make no recommendation about the use of a systematic search for sinusitis or the preven-tion of maxillary sinusitis. However, Stein and Caplan3 state that a search for sinusitis should be considered in febrile patients with risk factors for nosocomial sinusitis (or if no other source of infec-tion can be found), because doing so may decrease the patient’s risk for poor outcomes (including sep-sis and ventilator-associated pneumonia).

Strengths and LimitationsThe primary strength of our study is the inclu-

sion of more than 1 million patients from a random, stratified sample of hospitals in the United States. Several limitations exist. Use of retrospective data makes it impossible to determine cause-and-effect relationships between variables. We could not pro-vide information about the route of intubation or the types of nasogastric tubes used in the patients


(or if the endotracheal and nasogastric tubes were concurrently present). Also, we were unable to deter-mine the timeline of sinusitis and insertion of endo-tracheal and nasogastric tubes. Finally, sinusitis may have been present in some patients before admis-sion to the hospital.

Conclusion Despite the low incidence of sinusitis identified

in our sample of more than 1 million patients from the NIS database for the years 2008 through 2013, a significant association was found between sinusitis and the presence of an endotracheal tube, especially when a nasogastric tube also was present. These find-ings underscore the need to consider sinusitis as a possible site of infection in febrile patients with these types of tubes, especially when another source of infection cannot be identified. It is probable that the sinusitis incidence of 0.15% identified in our study is an underestimate; possible reasons include underper-formance of diagnostic tests for sinusitis and masking of sinusitis by more apparent concurrent infections (eg, pneumonia or bloodstream infections).

REFERENCES1. Fasquelle D, Alami M, Dumas G, Fockenier F, Sibille JP. [Epi-

demiology of sinusitis seen in hospitalized patients. Apropos of 77 episodes of sinusitis among 72 patients between 1993 and 1996]. Pathol Biol (Paris). 1998;46(10):751-759.

2. van Zanten AR, Dixon JM, Nipshagen MD, de Bree R, Girbes AR, Polderman KH. Hospital-acquired sinusitis is a common cause of fever of unknown origin in orotracheally intubated critically ill patients. Crit Care. 2005;9(5):R583-590.

3. Stein M, Caplan ES. Nosocomial sinusitis: a unique subset of sinusitis. Curr Opin Infect Dis. 2005;18(2):147-150.

4. Bert F, Lambert-Zechovsky N. Sinusitis in mechanically venti-lated patients and its role in the pathogenesis of nosocomial pneumonia. Eur J Clin Microbiol Infect Dis. 1996; 15(7):533-544.

5. Lum Cheong RS, Cornwell EE 3rd. Suppurative sinusitis in critically ill patients: a case report and review of the litera-ture. J Natl Med Assoc. 1992;84(12):1057-1059.

6. Rouby JJ, Laurent P, Gosnach M, et al. Risk factors and clin-ical relevance of nosocomial maxillary sinusitis in the criti-cally ill. Am J Respir Crit Care Med. 1994;150(3):776-783.

7. Pfuntner A, Wier LM, Stocks C. Statistical Brief #165. Most frequent procedures performed in U.S. hospitals, 2011. https://www.hcup-us.ahrq.gov/reports/statbriefs/sb165.jsp. Accessed October 3, 2017.

8. Mafee MF. Modern imaging of paranasal sinuses and the role of limited sinus computerized tomography; consider-ations of time, cost and radiation. Ear Nose Throat J. 1994;73(8):532-534, 536-538, 540-532 passim.

9. Talmor M, Li P, Barie PS. Acute paranasal sinusitis in criti-cally ill patients: guidelines for prevention, diagnosis, and treatment. Clin Infect Dis. 1997;25(6):1441-1446.

10. Humphrey MA, Simpson GT, Grindlinger GA. Clinical char-acteristics of nosocomial sinusitis. Ann Otol Rhinol Laryn-gol. 1987;96(6):687-690.

11. Grindlinger GA, Niehoff J, Hughes SL, Humphrey MA, Simpson G. Acute paranasal sinusitis related to nasotra-cheal intubation of head-injured patients. Crit Care Med. 1987;15(3):214-217.

12. George DL, Falk PS, Umberto Meduri G, et al. Nosocomial sinusitis in patients in the medical intensive care unit: a prospective epidemiological study. Clin Infect Dis. 1998; 27(3): 463-470.

13. Holzapfel L, Chevret S, Madinier G, et al. Influence of long-term oro- or nasotracheal intubation on nosocomial maxillary sinusitis and pneumonia: results of a prospective, random-ized, clinical trial. Crit Care Med. 1993;21(8):1132-1138.

14. Bach A, Boehrer H, Schmidt H, Geiss HK. Nosocomial sinus-itis in ventilated patients. Nasotracheal versus orotracheal intubation. Anaesthesia. 1992;47(4):335-339.

15. Donatelli J, Gupta A, Santhosh R, et al. To breathe or not to breathe: a review of artificial airway placement and related complications. Emerg Radiol. 2015;22(2):171-179.

16. Salord F, Gaussorgues P, Marti-Flich J, et al. Nosocomial maxillary sinusitis during mechanical ventilation: a pro-spective comparison of orotracheal versus the nasotra-cheal route for intubation. Intensive Care Med. 1990;16(6): 390-393.

17. Prabhakaran S, Doraiswamy VA, Nagaraja V, et al. Nasoenteric tube complications. Scand J Surg. 2012;101(3):147-155.

18. Blumenstein I, Shastri YM, Stein J. Gastroenteric tube feed-ing: techniques, problems and solutions. World J Gastroen-terol. 2014;20(26):8505-8524.

19. Wrenn K. The lowly nasogastric tube: still appropriate after all these years (at times). Am J Emerg Med. 1993;11(1):84-89.

20. Agrafiotis M, Vardakas KZ, Gkegkes ID, Kapaskelis A, Falagas ME. Ventilator-associated sinusitis in adults: systematic review and meta-analysis. Respir Med. 2012;106(8):1082-1095.

21. Elwany S, Helmy SA, El-Reweny EM, Harfoush R, Sobhy A. Endoscopically directed middle meatal cultures vs computed tomographic scans in the diagnosis of bacterial sinusitis in intensive care units. J Crit Care. 2012;27(3):315.e311-315.

22. HCUP NIS Database Documentation. Healthcare Cost and Utilization Project (HCUP). March 2017. Rockville, MD: Agency for Healthcare Research and Quality. https://www .hcup-us.ahrq.gov/db/nation/nis/nisdbdocumentation.jsp. Accessed October 24, 2017.

23. Papolos A, Narula J, Bavishi C, Chaudhry FA, Sengupta PP. U.S. hospital use of echocardiography: insights from the Nation-wide Inpatient Sample. J Am Coll Cardiol. 2016; 67(5): 502-511.

24. Deutschman CS, Wilton P, Sinow J, Dibbell D Jr, Konstan-tinides FN, Cerra FB. Paranasal sinusitis associated with nasotracheal intubation: a frequently unrecognized and treatable source of sepsis. Crit Care Med. 1986;14(2):111-114.

25. Guerin JM, Meyer P, Segrestaa JM, Reizine D, Levy C. [Nosocomial sinusitis and nasotracheal intubation. Pro-spective study of 53 patients]. Ann Med Interne (Paris). 1989;140(2):106-107.

26. Holzapfel L, Chastang C, Demingeon G, Bohe J, Piralla B, Coupry A. A randomized study assessing the systematic search for maxillary sinusitis in nasotracheally mechani-cally ventilated patients. Influence of nosocomial maxillary sinusitis on the occurrence of ventilator-associated pneu-monia. Am J Respir Crit Care Med. 1999;159(3):695-701.

27. Richards MJ, Edwards JR, Culver DH, Gaynes RP. Nosoco-mial infections in combined medical-surgical intensive care units in the United States. Infect Control Hosp Epidemiol. 2000;21(8):510-515.

28. Huyett P, Lee S, Ferguson BJ, Wang EW. Sinus opacification in the intensive care unit patient. Laryngoscope. 2016;126(11): 2433-2438.

29. Pedersen J, Schurizek BA, Melsen NC, Juhl B. The effect of nasotracheal intubation on the paranasal sinuses. A pro-spective study of 434 intensive care patients. Acta Anaes-thesiol Scand. 1991;35(1):11-13.

30. Aebert H, Hunefeld G, Regel G. Paranasal sinusitis and sep-sis in ICU patients with nasotracheal intubation. Intensive Care Med. 1988;15(1):27-30.

31. Fassoulaki A, Pamouktsoglou P. Prolonged nasotracheal intubation and its association with inflammation of parana-sal sinuses. Anesth Analg. 1989;69(1):50-52.

32. O’Reilly MJ, Reddick EJ, Black W, et al. Sepsis from sinusitis in nasotracheally intubated patients: a diagnostic dilemma. Am J Surg. 1984;147(5):601-604.

33. Willatts SM, Cochrane DF. Paranasal sinusitis: a complication of nasotracheal intubation: two case reports. Br J Anaesth. 1985;57(10):1026-1028.

34. Muscedere J, Dodek P, Keenan S, Fowler R, Cook D, Hey-land D; VAP Guidelines Committee and the Canadian Critical Care Trials Group. Comprehensive evidence-based clinical practice guidelines for ventilator-associated


pneumonia: diagnosis and treatment. J Crit Care. 2008; 23(1): 138-147.

35. Pneumatikos I, Konstantonis D, Tsagaris I, et al. Prevention of nosocomial maxillary sinusitis in the ICU: the effects of topically applied alpha-adrenergic agonists and corticoste-roids. Intensive Care Med. 2006;32(4):532-537.

36. Riga M, Danielidis V, Pneumatikos I. Rhinosinusitis in the intensive care unit patients: a review of the possible under-lying mechanisms and proposals for the investigation of

their potential role in functional treatment interventions. J Crit Care. 2010;25(1):171.e9-171.e14.


Write a letter to the editors of AJCC!Visit AJCC's EditorialManager website(http://www.editorialmanager.com/ajcc)

1. Click on Author Login and enter yourusername and password.

First time?Click on the blue Register Now and follow the instructions to set up your username and password.

2. Once logged in, click on Submit NewManuscript.

3. At Choose Article Type, select Letterto the Editor from the pull-down menu.

4. Follow the rest of the instructions tosubmit your letter.

Letters must be approved by thejournal's coeditors for publication.


Pulmonary Critical Care

Background Sedation and analgesia have an important

impact on the outcome of patients treated with mechan-

ical ventilation. International guidelines recommend use

of sedation protocols to ensure best patient care.

Objective To determine the sedation practice of intensive

care nurses weaning adults from mechanical ventilation.

Methods A cross-sectional survey with a self-adminis-

tered questionnaire was used to determine sedation

practices of Flemish critical care nurses during weaning.

Consensus on content validity was achieved through a

Delphi procedure among experts. Data were collected

during the 32nd Annual Congress of the Flemish Society

of Critical Care Nurses in Ghent, Belgium, December 2014.

Results A total of 342 nurses were included in the study.

Of these, 43.7% had a sedation protocol in their unit

that was used by 61.8% of the respondents. Sedation

protocols were more often available (P < .001) in aca-

demic hospitals (72%) than in general hospitals (41.5%).

Sedatives were administered via continuous infusion

with bolus doses if needed (81%). Level of sedation

was assessed every 2 hours (56%), mostly via the Rich-

mond Agitation-Sedation Scale (59.1%). Daily interrup-

tion of sedation was used by 16.5% of respondents. The

biggest barriers to daily interruption were patient com-

fort (49.4%) and fear of respiratory worsening (46.6%).

Conclusions A considerable discrepancy exists between

international recommendations and actual sedation

practices. Standardization of sedation practices across

different institutions on a regional and national level

may improve the quality of care. (American Journal of


NURSES’ SEDATION PRACTICES DURING WEANING OF ADULTS FROM

MECHANICAL VENTILATION

IN AN INTENSIVE CARE UNITBy Marta Borkowska, RN, MNSc, Sonia Labeau, RN, MA, MNSc, PhD, Tom Schepens, MD, Dominique Vandijck, RN, MSc, MA, PhD, Katrien Van de Vyver, RN, MNSc, Daphné Christiaens, RN, MSc, Christelle Lizy, RN, MNSc, Bronagh Blackwood, PhD, MSc, BSc (Hons), RGN, RNT, and Stijn I. Blot, MNSc, PhD




Nurses play a crucial role in maintaining optimal sedation of patients.

Mechanical ventilation for adults with acute respiratory failure is potentially lifesaving but is a distressing and invasive intervention and can induce pain, fear, and anxiety in patients and thus affect a patient’s outcome.1 Patients treated with mechanical ventilation generally receive some form of sedative therapy,2,3 usually various combinations of opioids and benzodiazepines.4-7

Sedation also has clear benefits, such as reduction of pain, anxiety, agitation, and stress responses; prevention of removal of monitoring devices; facilitation of ventilation; and provision of comfort and safety.2,3,8,9

Suboptimal sedation, however, is common and is associated with numerous adverse events.10 The type of agent11-13 and route of administration1,11,14-17 influence patients’ outcomes. Continuous intrave-nous infusion of a sedative agent is the most fre-quently used method of sedation, despite the method’s association with prolonged mechanical ventilation,14,15 increased risk for ventilator-associated pneumonia,16 and increased intensive care unit (ICU) and hospital lengths of stay.14,17

Optimally sedated patients should be awake, comfortable, calm, and cooperative, a situation that would enable faster discontinuation of mechanical ventilation.1,18 This situation is, however, often diffi-cult to achieve. Various strategies have been tried to reach this goal, including sedation algorithms,

guidelines,18,19 sedation protocols alone15,17,19 and in a combination with weaning protocols,20,21 and daily interruption of sedation (DIS).1,13,15,16,19,22,23 DIS con-tributes to faster weaning from mechanical ventila-tion by decreasing the use of sedatives, preventing drug accumulation, promoting a patient’s awareness, and facilitating a patient’s interaction with the envi-ronment.11,15,16,19 DIS may also prevent depression and posttraumatic stress disorder.1,11 Therefore, incorporation of DIS and target sedation goals in a sedation protocol is recom-mended.21 However, no strong evidence supports the use of sedation proto-cols and DIS, possibly because of organizational and contextual factors. Available information indicates the positive influence of sedation protocols and DIS on patients’ outcomes.1,13,15-17,19,22,23 However, addi-tional research is needed to summarize the available data.5,10

Past research has focused on identifying the most appropriate sedative and analgesic agents. The agents should have a rapid onset of action and be easy to manage. Most importantly, they should cause few complications, not accumulate in the tissues, and be affordable. Although the ideal drug does not exist yet, strategies for sedating patients have changed toward lighter sedation than before.1,8,24 The most recent guidelines of the Society of Critical Care Medi-cine18 emphasize the importance of lighter and analgesia-based sedation and recommend the use of nonbenzodiazepine agents, limited use of neuro-muscular blocking agents (NMBAs), and monitoring patients’ sedation level. Despite efforts to standard-ize sedation care worldwide, substantial variety in clinical practice remains between Europe, North Amer-ica, and Australia.6,7,25-33 A detailed inventory of cur-rent sedation practices might be useful to identify the factors contributing to this discrepancy.

The role of nurses in assessing and maintaining

About the AuthorsMarta Borkowska is a registered nurse in the surgical intensive care unit at Ghent University Hospital, Ghent, Belgium. Sonia Labeau is head of the nursing degree program, Faculty of Education, Health, and Social Work, University College Ghent, Ghent, Belgium. Tom Schep-ens is a clinical fellow, Department of Critical Care Medi-cine, Hospital for Sick Children, University of Toronto, Canada, and a doctoral student at University of Antwerp, Antwerp, Belgium. Dominique Vandijck is a professor, Faculty of Medicine and Life Sciences, Department of Patient Safety, Hasselt University, Diepenbeek, Belgium, and Faculty of Medicine and Health Sciences, Ghent Uni-versity, Ghent, Belgium, and director of quality and safety, Zorgnet-Icuro, Brussels, Belgium. Katrien Van de Vyver is an infection prevention and control nurse, AZ Maria Middelares Hospital, Ghent, Belgium. Daphné Christiaens is a study nurse, SAFE-PEDRUG study team, Pediatric Nephrology and Rheumatology, Ghent University Hospi-tal. Christelle Lizy is head nurse, Nephrology, Endocrinol-ogy, Cardiology Department, Ghent University Hospital. Bronagh Blackwood is professor of critical care, Centre for Experimental Medicine, School of Medicine, Dentistry and Biomedical Sciences, Queen’s University Belfast, Belfast, Northern Ireland. Stijn I. Blot is a research pro-fessor, Department of Internal Medicine, Ghent Univer-sity, and an honorary professor, Burns Trauma and Critical Care Research Centre, University of Queensland, Bris-bane, Australia.

Corresponding author: Marta Borkowska, RN, MNSc, Ghent University Hospital, De Pintelaan 185, Ghent, Oost Vlaan-deren 9000, Belgium (email: [email protected]).


The reported availability of a sedation protocol

and protocol adherence were alarmingly low.

optimal sedation is paramount.27,28 Nurses’ perception of sedation practice; nurses’ attitudes, knowledge, and experiences; and the level of multidisciplinary

collaboration are essential for understanding and guid-ing nurses’ behavior.27-30 Because of organizational and cultural differences, identifying the rationales of nurses’ decisions in the ICU is a challenge.6,23,24,28,29,33

Recognition of the patterns that influence these sub-stantial differences in practice is important to ensure effective, homogenous, and evidence-based care for all ICU patients.

Aim The aim of our study was to determine nurses’

sedation practices during weaning of patients from mechanical ventilation in adult ICUs in Flanders (Belgium). We constructed a questionnaire to explore the prevalence of and adherence to sedation proto-cols, use of DIS, and nurses’ perceptions of barriers toward use of DIS and use of sedation protocols. We also sought to describe the range and type of anal-gosedative agents used for sedation in the ICU and the extent of nurses’ autonomy in sedation practice.

Methods Development of the Questionnaire

We used a cross-sectional survey of ICU nurses with a self-administered questionnaire (see Appendix) to evaluate the nurses’ daily sedation practices. The questionnaire was used to gather sociodemographic data and included 18 questions based on recent literature. Of the 18 questions, 11 were closed-ended multiple-choice questions, and 4 were single-answer questions. The 3 remaining questions were to be answered by using 4-point Likert scales.34 Content validity35 was assessed via a Delphi procedure36 in which 4 independent experts in ICU nursing research and 1 ICU physician participated. The panel mem-bers remained anonymous to one another through-out the procedure.

Data Collection and AnalysisAs approved by the board of the Flemish Soci-

ety of Critical Care Nurses, the questionnaire was distributed to 640 attendants of the 32nd Annual Congress of the Flemish Society of Critical Care Nurses in Ghent, Belgium (December 12, 2014). This non-profit society from the northern, Dutch-speaking part of Belgium is one of the largest European

congresses and attracts a mean of 700 critical care nurses per year.

The chairman of the society presented the aims of the survey to the attendants in a Microsoft Power-Point presentation, and an interval of 30 minutes was provided for completing the questionnaire. Respon-dents were requested to return the questionnaire by dropping it in boxes provided at each door of the congress room.

One of us (M.B.) entered the data into the SPSS Statistics, version 21, software program (IBM SPSS) for data analysis. Only the responses of bedside nurses in adult ICUs were analyzed. Descriptive statistics were used to analyze categorical variables and are presented as frequencies and percentages. Differences in responses between nurses working in academic hospitals and nurses working in gen-eral hospitals were determined by using 2 tests for categorical variables. Significance was set at P less than .05. Statistical analysis was conducted by 2 of us (M.B. and S.B.)

Ethical ConsiderationsEthical approval was obtained from the ethics

committee at Ghent University Hospital. The respon-dents were orally informed during the congress about the purpose of the study and its voluntary character. In the introduction to the survey, a writ-ten assurance of anonymity was given, and the voluntary character of the study was emphasized. Informed consent was assumed when respondents filled out the questionnaire. The data were accessi-ble only to the research team and were stored on a password-protected computer.

Results Response Rate and Demographics

Completed questionnaires were obtained from 423 of the 640 congress delegates, for a response rate of 66.1%. Of the 423 respondents, 342 were bedside ICU nurses whose responses were included in the analysis. Table 1 gives the demographic charac-teristics of the respondents.

Sedation PracticesFor sedation for less than 24 hours, mainly

short-acting agents were used (Figure 1). For seda-tion for 24 hours or longer, both short- and long-acting agents were administered (Figure 2). Among the 342 respondents, 265 (77.5%) indicated that patient-ventilator dyssynchrony was the most fre-quent indication for use of NMBAs. Less common indications included acute lung injury or acute


respiratory distress syndrome (130 respondents, 38.0%), prevention or treatment of shivering in patients with induced therapeutic hypothermia (104, 30.4%), and instances of high intracranial pressure (76, 22.2%). Among the total sample, 277 nurses (81.0%) administered sedatives via continuous intra-venous infusion with bolus doses if needed.

High percentages of the 342 nurses reported changing sedative infusion rates (256 nurses, 74.9%) and delivering sedative bolus doses (269, 78.7%) without a physician’s order. For 341 respondents, less than half of the nurses (149, 43.7%) reported the presence of a sedation protocol in their ICU, and 35 (10.3%) were not aware of a protocol being available. Sedation protocols were reported as being available more frequently (P < .001) in academic hospitals (54 of 75, 72.0%) than in general hospi-tals (93 of 224, 41.5%).

Among the 149 nurses who answered the ques-tion about sedation protocols, protocols were used always or mostly by 61.8% (always: 12 nurses, 8.1%; mostly: 80, 53.7%). A majority of respondents (283 of 339, 83.5%) indicated that additional analgesic agents were generally administered if sedation was withdrawn (Table 2).

Slightly more than half of the 146 respon-dents who answered the relevant question had a patient-targeted protocol (78 respondents, 53.4%). Among the 149 respondents who indicated that a sedation protocol was used in their ICU, 117 (78.5%) reported that the protocols were mainly developed by ICU physicians, and 77 (51.7%) indicated that the protocols were developed by nurses. Other health care professionals (anesthetist, physiotherapist, respiratory therapist, pneumologist) were involved less frequently or not at all (psychologist) in the development. Nearly 54% of the nurses (80 respon-dents) reported that they would not use a protocol if a physician desired to work without it or if no phy-sician’s order was available (29 nurses, 19.5%). Rea-sons for not applying sedation protocols included a high workload (16 nurses, 10.7%) and the use of short-term sedation (57, 38.3%).

Among 336 respondents, 188 nurses (56.0%) reported that level of sedation was generally evalu-ated every 2 hours (Table 3). In addition, among 330 respondents, 195 nurses (59.1%) reported that the level was evaluated by using the Richmond Agitation-Sedation Scale (RASS); 155 (47.0%) reported using the Glasgow Coma Scale and 96 (29.1%) reported using the Ramsay Sedation Scale. DIS was applied variably; among 340 respondents, the frequency of using DIS was never for 93 nurses

(27.4%), rarely for 182 (53.5%), mostly for 49 (14.4%), and always for 7 (2.1%). Nine nurses (2.6%) could not estimate how often they used DIS. Among 249 respondents, 215 (86.3%) reported that DIS was used to evaluate patients’ neurological status, and 110 (44.2%) indicated that it was used to shorten the duration of mechanical ventilation. A total of 193 (77.5%) reported that DIS was not used during night shifts.

Numerous barriers to use of DIS were reported. Of the 249 respondents, 123 (49.4%) noted that patient comfort was the most frequent concern, and 116 (46.6%) indicated respiratory deteriora-tion as the most frequent concern. The most rele-vant barriers are presented in Figure 3.

Characteristic

Table 1Characteristics of the respondents

Sex (n = 340) Male Female

Academic degreea,b (n = 342) Bachelor’s in nursing Postgraduate in emergency and intensive care Master’s in nursing or equal

Years’ experience working in intensive care unitb (n = 342) 1-5 6-10 10

Hospital type (n = 334) Academic General

Type of intensive care unita,b (n = 342) General Medical Surgical Cardiosurgical

Number of beds (n = 336) 8 9-16 17

Nurse to patient ratio per shiftb

Early shift (n = 325) 1:2 1:3 Late shift (n = 320) 1:2 1:3 Night shift (n = 321) 1:2 1:3 1:4

25.9 (88)74.1 (252)

77.8 (266)88.9 (304) 8.5 (29)

19.0 (65)16.7 (57)61.7 (211)

26.9 (90)73.1 (244)

66.1 (226)13.2 (45)17.8 (61) 8.2 (28)

24.4 (82)44.9 (151)30.7 (103)

53.5 (174)40.9 (133)

35.0 (112)60.3 (193)

19.0 (61)56.7 (182)21.2 (68)

Percentage (No. of respondents)

a The number of respondents does not total 342 because more than 1 answer was possible.

b Values less than 5% are not shown.


A majority (225 nurses, 65.8%) of 342 respon-dents indicated agitation, confusion, and ICU delirium as frequently occurring problems during reduction of sedative agents. Furthermore, major concerns included patients’ comfort or pain (156 nurses, 45.6%), need for physical restraints (119, 34.8%), and patients biting on their endotracheal tube, tongue, or lips (129, 37.7%).

Respondents (n = 342) were also asked to give their perception of multidisciplinary collaboration in their ICU. A total of 323 nurses (94.4%) reported that of all team members, the attending physician was the team member most involved in decision-making on reduction of sedative infusions, cessa-tion of sedative infusions (328 respondents, 95.9%), and implementation of DIS (238, 69.6%). Second most involved were nurses; involvement in reduction of sedatives was reported by 276 respondents (80.7%), cessation of sedative infusions by 225 (65.8%), and implementation of DIS by 149 (43.6%). The involve-ment of other health care personnel such as respira-tory therapists and physiotherapists was reported as less than 2%.

Discussion Our results delineate some current sedation prac-

tices among Flemish nurses. We found alarmingly low availability of sedation protocols and level of compliance with the protocols. DIS was infrequent, and barriers to DIS were numerous. Analgosedation, particularly with short-acting agents, was mostly provided as continuous intravenous infusion with bolus doses if needed. The level of sedation was often assessed by using RASS.

Nurse-led sedation protocols have an important influence on weaning outcomes, duration of mechan-ical ventilation, and hospital and ICU lengths of stay16 and should be considered a standard practice in the ICU.11,18,19,37 However, no strong evidence indi-cates the beneficial influence of sedation protocols on patients’ outcomes.10 The available data are generally derived from nonrandomized studies, resulting in conflicting conclusions. Our results indicated that the use of sedation protocols was low (43.7%), with a significant difference (P < .001) in responses between nurses working in academic hospitals (72.0%) and nurses working in general hospitals (41.5%). Similar differences were noted in a Canadian survey of physicians.6 The reported adherence to the protocol in our study was slightly greater than 60%, indicating that a low number of patients received sedation care based on a protocol. In a Belgian survey of ICUs,30 the reported availability

Short-term sedation (< 24 hours)

Propofol + remifentanil45.9% (157)

Combined agents

Remifentanil 12.3% (42)

Benzodiazepines 11.7% (40)

Propofol54.7% (187)

Dexmedetomidine 14.3% (49)

Single agent

Figure 1 Agents used for short-term sedation (< 24 hours). Only relevant proportions (> 10% of 342 respondents) are reported.

Long-term sedation ( 24 hours)

Combined agents

Benzodiazepines + fentanyl 11.7% (40)

Benzodiazepines + sufentanil 16.7% (57)

Benzodiazepines + remifentanil 23.1% (79)

Propofol + sufentanil 10.2% (35)

Propofol + remifentanil 46.8% (160)

Dexmedetomidine + remifentanil10.5% (36)

Benzodiazepines 13.5% (46)

Propofol19.6% (67)

Dexmedetomidine 19.3% (66)

Single agent

Figure 2 Agents used for long-term sedation (≥ 24 hours). Only relevant proportions (> 10% of 342 respondents) are reported.

Benzodiazepines + morphine 19.6% (67)


of a sedation protocol was even lower (26% of the nurses surveyed); however, only 8% of nurses reported not using a protocol at all. Differences in sampling may be responsible for the variability of results in the Canadian and Belgian surveys.

Results from other surveys25,26,28,38-40 have indi-cated further deficits in availability and implementa-tion of sedation protocols and scales. Protocols are applied rather poorly in most European countries; however, a trend toward more frequent use of proto-cols has been reported.39 North American studies do not show better results in this regard. In the study by Tanios et al,29 the overall presence of a sedation protocol was higher than in our survey (64.0%), with a low response rate (7.1%). The respondents were mainly physicians.29 The results of another American survey27 (60.5% of nurse respondents) and of a Canadian observational study23 (54.9% of respondents) also were similar. In the study by Patel et al,7 the availability of a protocol was higher (71%), but the sample consisted of different profes-sionals, and nurses were a minority (23%). The inter-est in use of protocols for sedation has increased in Australia31,33; however, the benefits of use of proto-cols has not been demonstrated yet.

Organizational and contextual factors, such as nurse to patient ratio, preexisting practice and cul-ture, level of multidisciplinary collaboration, and nurses’ attitudes and knowledge may affect everyday practice in the ICU.27,28,33,41-43 The perceived level of collaboration on sedation practices in our survey was

high; physicians were most prominently involved in all decisions. We noticed that multidisciplinary collaboration applied almost exclusively to physicians and nurses; other health care personnel were rarely

Question

Table 2Use of sedation protocol and administration of additional analgesic agents during withdrawal of sedation

Use of sedation protocol (n = 149)a

Additional analgesic agents (n = 339)a

12 (8.1)

63 (18.6)

80 (53.7)

220 (64.9)

55 (36.9)

51 (15)

2 (1.3)

5 (1.5)

AlwaysMostlyRarelyNever

No. (%) of respondents

a Total number of respondents who answered that question.

Frequency of examination of adequacy of sedation level (n = 336)a

Table 3Examination of adequacy of sedation level per shift

Every hour

Every 2 hours

Every 4 hours

Every 6 hours

Every 8 hours

No examination

76 (22.6)

188 (56.0)

44 (13.1)

1 (0.3)

15 (4.5)

12 (3.6)

No. (%) of responsesb

a Total number of respondents who answered that question.b Because of rounding, percentages may not total 100.

Figure 3 Reported barriers to performing daily interruption of sedation (DIS). Only relevant proportions (> 10% of respondents) are reported.

Patient’s lack of comfort

60

50

40

30

20

10

0Respiratory worsening

Possible self-removal of medical

devices

Greater possibility

of agitation

Workload too high

Barrier

Perc

enta

ge

of

resp

on

ses

Physician never

suggested DIS

Insufficient medical

supervision

Difficult care


DIS is rarely applied by nurses, and barriers to

DIS are numerous.

A low nurse to patient ratio may decrease

quality of care for patients receiving

mechanical ventilation.

included. Also, the nurse to patient ratio in our study was lower than the ratio in other stud-ies,28,31,44,45 a finding that raises some questions about quality of care. We think that particularly in settings with a low nurse to patient ratio, the use of protocols should be recommended to ensure patients’ safety. Unfortunately, we cannot provide more insights into organizational or contextual factors.

The implementation of sedation scales into clinical practice remains problematic.24 The Ramsey Sedation Scale seems to be the most favorable scale

in North America7,27 and Europe,25,30,38,39 despite rec-ommendations to use the Sedation Agitation Scale or the RASS.18,19 Since the publi-cation of the 2013 guidelines of the Society of Critical Care

Medicine,18 use of the RASS and the Sedation Agita-tion Scale23,28 has increased somewhat. In our survey, adoption of the RASS (59.1%) was higher than in other studies.27,28,30 Although availability and use of sedation protocols in our study were low, the possi-ble adverse impact on sedation care might be lim-ited by the frequent use of valid sedation scales.

DIS was rarely used by our respondents. Similar data have been reported by others; the highest reported percentage of respondents practicing DIS was around 50% or was described as poor.23,26,28-30,38-40 Studies from the United Kingdom25 and North America7 indicated more frequent adoption of DIS (77.8% and 77%, respectively), but according to the authors,7,25 these figures might be overestimations. Despite the proven safety of DIS,1,15 multiple barriers to use of this prac-tice might be responsible for the low adoption rates. In our study, barriers to DIS included fear of respira-tory worsening, lack of patient comfort, and possible

removal of the endotracheal tube. Similar barriers have been described by other researchers.24,29,30 Some recog-nizable nurses’ concerns were reported in a qualitative study by Everingham et al,46 who reported that the rigid appli-cation of targeted care rather than individual care dis-

tressed nurses. Nurses reported having to deal with the consequences of caring for agitated patients and struggling to provide safe care. As a consequence, nurses reported a sense of failure, guilt, and lack of satisfaction in care provided. They reported not only an increasing workload associated with sedation breaks but also concerns about patients’ experiences

during the awake period.46 The barriers to DIS pro-vided by nurses in the context of questionable evidence supporting the use of DIS5 warrant fur-ther research and should be an important nursing research priority.

Strategies designed to minimize the use of seda-tives, such as sedation protocols and sedation scales, can improve patients’ outcomes.18 Limited use of NMBAs is important to achieve lighter sedation.19,47 In our survey, nurses indicated that NMBAs were reserved for instances of patient-ventilator asyn-chrony and acute lung injury or acute respiratory distress syndrome. These findings differ from those of Reschreiter et al,25 who reported that NMBAs were used by 71% of ICUs for less than 5% of the time and mainly for patients with neurological problems. The difference in findings suggests differ-ences between countries. Further exploration of the use of NMBAs in Flanders should be considered to determine why Flemish findings differ from the findings of other countries and to evaluate whether the use of NMBAs in Flanders is routine or occa-sional. Further, the choice of analgosedative regimen is crucial in daily sedation practice. Although the ideal sedative does not exist, sedation regimens based on short-acting agents have had favorable results.1,11,13,18

The respondents in our survey reported that sed-atives were frequently administered as a continuous intravenous infusion with bolus doses if necessary, a situation that could result in sedative overdose.2 The short-acting agents propofol and remifentanil were frequently used by our respondents, as suggested in the latest guidelines.18 We suppose that when use of short-acting agents is predominant, the deleteri-ous impact of continuous infusions might be lim-ited but not eliminated.

According to our respondents, use of benzodiaz-epines was mostly reserved for sedation for 24 hours or longer. The use of benzodiazepines for sedation less than 24 hours (in monotherapy or combination therapy) is limited. Internationally, the trend toward use of propofol is growing; however, a broad variation in choice of sedatives still exists.24 Some investiga-tors25,28,30,32,39 reported that propofol was the first-choice sedative agent, whereas others7,31,40 reported that its use was similar to that of benzodiazepines. Marked variations in choice of analgesic agents has been reported; morphine or fentanyl or both are the most frequently used agents.6,23,26,31,32,38 The use of remifentanil and of dexmedetomidine is rare,24 pos-sibly because of high costs and limited experience with those agents.31 In contrast, use of dexmedeto-midine and remifentanil in our study was rather high (19.3% and 12.3%, respectively, in monotherapy).


Limitations Because of the self-reporting nature of our study,

we cannot guarantee absence of response set bias. Participation in the study was on a voluntary basis, so our findings are prone to selection bias as well. Furthermore, our response rate might be adversely affected by the length of the questionnaire, thus resulting in loss of potentially important informa-tion. We recommend caution in extrapolating our data to other geographic regions because organiza-tional and contextual factors influence sedation prac-tices. Finally, questions on psychosocial aspects of sedation, ICU delirium, and pain protocols were not included in the survey, although these issues are also important.

Conclusions We found a large discrepancy between interna-

tional recommendations and actual sedation practices during weaning from mechanical ventilation, a situ-ation that might result in prolonged weaning from the therapy. Quality improvement initiatives for imple-mentation of safe and cost-effective sedation practices are recommended but are challenging. A collective engagement of governmental institutions and some nonprofit institutions is necessary to gain further insights in health care professionals’ sedation prac-tices. Promising well-designed quality improvement trials that combine simultaneously used different interventions48 should be strongly considered for improving complex health care interventions. Stan-dardization of sedation practices on a regional (Flemish) or, preferably, national level may further improve the quality of sedation care.10 Additional research on this subject, particularly on how nurses can improve sedation practices at the bedside, should receive more attention from researchers.

ACKNOWLEDGMENTDr Blot holds a research mandate of the Special Research Fund, Ghent University.

REFERENCES1. Schweickert WD, Kress JP. Strategies to optimize analgesia

and sedation. Crit Care. 2008;2(suppl 3):S6. http://ccforum .com/content/12/S3/S6. Accessed September 15, 2017.

2. Kress JP, Hall JB. Sedation in the mechanically ventilated patient. Crit Care Med. 2006;34(10):2541-2546.

3. McGrane S, Pandharipande P. Sedation in the intensive care unit. Minerva Anestesiol. 2012;78(3):369-380.

4. Shehabi Y, Bellomo R, Mehta S, Riker R, Takala J. Intensive care sedation: the past, present and the future. Crit Care. 2013; 17(3):322. http://ccforum.com/content/17/3/322. Accessed September 15, 2017.

5. Burry L, Rose L, McCullagh IJ, Fergusson DA, Ferguson ND, Mehta S. Daily sedation interruption versus no daily sedation interruption for critically ill adult patients requiring invasive mechanical ventilation. Cochrane Database Syst Rev. 2014; (7):CD009176.

6. Mehta S, Burry L, Fischer S, et al; Canadian Critical Care Trials Group. Canadian survey of the use of sedatives,

analgesics, and neuromuscular blocking agents in criti-cally ill patients. Crit Care Med. 2006;34(2):374-380.

7. Patel RP, Gambrell M, Speroff T, et al. Delirium and sedation in the intensive care unit: survey of behaviors and attitudes of 1384 healthcare professionals. Crit Care Med. 2009;37(3): 825-832.

8. Gehlbach BK, Kress JP. Sedation in the intensive care unit. Curr Opin Crit Care. 2002;8(4):290-298.

9. Hogarth DK, Hall J. Management of sedation in mechanically ventilated patients. Curr Opin Crit Care. 2004;10(1):40-46.

10. Jackson DL, Proudfoot CW, Cann KF, Walsh TS. The incidence of sub-optimal sedation in the ICU: a systematic review. Crit Care. 2009;13(6):R204. http://ccforum.com/content/13/6 /R204. Accessed September 15, 2017.

11. Patel SB, Kress JP. Sedation and analgesia in the mechanically ventilated patient. Am J Respir Crit Care Med. 2012; 185(5): 486-497.

12. Futier E, Chanques G, Cayot Constantin S, et al. Influence of opioid choice on mechanical ventilation duration and ICU length of stay. Minerva Anestesiol. 2012;78(1):46-53.

13. Fraser GL, Devlin JW, Worby CP, et al. Benzodiazepine versus nonbenzodiazepine-based sedation for mechanically ventilated, critically ill adults: a systematic review and meta-analysis of randomized trials. Crit Care Med. 2013; 41(9)(suppl 1):S30-S38.

14. Kollef MH, Levy NT, Ahrens TS, Schaiff R, Prentice D, Sherman G. The use of continuous IV sedation is associated with prolon-gation of mechanical ventilation. Chest. 1998; 114(2): 541-548.

15. Kress JP, Pohlman AS, O’Connor MF, Hall JB. Daily interrup-tion of sedative infusions in critically ill patients undergoing mechanical ventilation. N Engl J Med. 2000; 342(20): 1471-1477.

16. Schweickert WD, Gehlbach BK, Pohlman AS, Hall JB, Kress JP. Daily interruption of sedative infusions and complications of critical illness in mechanically ventilated patients. Crit Care Med. 2004;32(6):1272-1276.

17. Arias-Rivera S, Sánchez-Sánchez Mdel M, Santos-Díaz R, et al. Effect of a nursing-implemented sedation protocol on weaning outcome. Crit Care Med. 2008;36(7): 2054-2060.

18. Barr J, Fraser GL, Puntillo K, et al; American College of Crit-ical Care Medicine. Clinical practice guidelines for the man-agement of pain, agitation, and delirium in adult patients in the intensive care unit. Crit Care Med. 2013; 41(1): 263-306.

19. Celis-Rodriguez E, Birchenall C, de la Cal MA, et al. Clinical practice guidelines for evidence-based management of sedoanalgesia in critically ill adult patients. Med Intensiva. 2013;37(8):519-574.

20. Girard TD, Kress JP, Fuchs BD, et al. Efficacy and safety of a paired sedation and ventilator weaning protocol for mechan-ically ventilated patients in intensive care (Awakening and Breathing Controlled trial): a randomised controlled trial. Lancet. 2008;371(9607):126-134.

21. Ely EW, Meade MO, Haponik EF, et al. Mechanical ventilator weaning protocols driven by nonphysician health-care pro-fessionals. Chest. 2001;120(6)(suppl):454S-463S.

22. Boles JM, Bion J, Connors A, et al. Weaning from mechanical ventilation. Eur Respir J. 2007;29(5):1033-1056.

23. Burry LD, Williamson DR, Perreault MM, et al. Analgesic, sedative, antipsychotic, and neuromuscular blocker use in Canadian intensive care units: a prospective, multicentre, observational study. Can J Anaesth. 2014;61(7):619-630.

24. Mehta S, McCullagh I, Burry L. Current sedation practices: lessons learned from international surveys. Crit Care Clin. 2009;25(3):471-488.

25. Reschreiter H, Maiden M, Kapila A. Sedation practice in the intensive care unit: a UK national survey. Crit Care. 2008; 12(6):R152. http://ccforum.com/content/12/6/R152. Accessed September 15, 2017.

26. Randen I, Bjørk IT. Sedation practice in three Norwegian ICUs: a survey of intensive care nurses’ perceptions of personal and unit practice. Intensive Crit Care Nurs. 2010; 26(5): 270-277.

27. Guttormson JL, Chlan L, Weinert C, Savik K. Factors influenc-ing nurse sedation practices with mechanically ventilated patients: a US national survey. Intensive Crit Care Nurs. 2010;26(1):44-50.

28. Egerod I, Albarran JW, Ring M, Blackwood B. Sedation prac-tice in Nordic and non-Nordic ICUs: a European survey. Nurs Crit Care. 2013;18(4):166-175.

29. Tanios MA, de Wit M, Epstein SK, Devlin JW. Perceived barriers to the use of sedation protocols and daily sedation interruption: a multidisciplinary survey. J Crit Care. 2009; 24(1): 66-73.

30. Sneyers B, Laterre PF, Perreault MM, Wouters D, Spinewine A. Current practices and barriers impairing physicians’ and

Appendix Sedation practices: a questionnaire

The questions listed below are related to sedation practices in your Intensive Care Unit (ICU) and/or to your personal perception on the subject.

Generally, you can provide more than one answer unless asked otherwise.

1. What agents are being used for:Short-term sedation (< 24h)? Long-term sedation ( 24h)?

just morphine just morphinejust fentanyl just fentanyljust sufentanil (Sufenta) just sufentanil (Sufenta)just remifentanil (Ultiva) just remifentanil (Ultiva)benzodiazepines (eg, Dormicum) + morphine benzodiazepines(eg, Dormicum) + morphinebenzodiazepines (eg, Dormicum) + fentanyl benzodiazepines (eg, Dormicum) + fentanyl benzodiazepines (eg, Dormicum) + sufentanil (Sufenta) benzodiazepines (eg, Dormicum) + sufentanil (Sufenta)benzodiazepines (eg, Dormicum) + remifentanil (Ultiva) benzodiazepines(eg Dormicum) + remifentanil (Ultiva) just benzodiazepines (eg, Dormicum) just benzodiazepines (eg, Dormicum) just propofol (Diprivan/Propolipid) just propofol (Diprivan/Propolipid) propofol (Diprivan/Propolipid) + morphine propofol (Diprivan/Propolipid) + morphine propofol (Diprivan/Propolipid) + fentanyl propofol (Diprivan/Propolipid) + fentanyl propofol (Diprivan/Propolipid) + sufentanil (Sufenta) propofol (Diprivan/Propolipid) + sufentanil (Sufenta) propofol (Diprivan/Propolipid) + remifentanil (Ultiva) propofol (Diprivan/Propolipid) + remifentanil (Ultiva) just dexmedetomidine (Dexdor) just dexmedetomidine (Dexdor) dexmedetomidine (Dexdor) + morphine dexmedetomidine (Dexdor) + morphine dexmedetomidine (Dexdor) + fentanyl dexmedetomidine (Dexdor) + fentanyl dexmedetomidine (Dexdor) + sufentanil (Sufenta) dexmedetomidine (Dexdor) + sufentanil (Sufenta) dexmedetomidine (Dexdor) + remifentanil (Ultiva) dexmedetomidine (Dexdor) + remifentanil (Ultiva) other (specify): ________________ other (specify): ________________

2. What are the indications for use of neuromuscular blocking agents at your department (except for intubation)? acute lung injury (ALI)/ acute respiratory distress syndrome (ARDS) high intracranial pressure patient-ventilator dyssynchrony (patient difficult to ventilate) intra-abdominal hypertension status asthmaticus prevention/treatment of shivering during therapeutic routinely used during sedation hypothermia

not applicable (no neuromuscular blocking agents used)

3. What are the actions that you take without physician’s order? starting the sedative infusion delivering sedative bolus doses cessation of sedative infusion changing sedative infusion rate interrupting sedation none

4. How is intravenous sedation being administered in your department? only as continuous infusion only intermittent bolus doses continuous infusion with bolus doses if needed other

5. If sedation is being withdrawn, are there any additional analgesic agents administered? (with little or no sedative effect) never rarely mostly always

6. Is there a sedation protocol available in your ICU? yes no (go to question 11) I don’t know (go to question 11)

7. Is a sedation protocol being consequently used? (only 1 answer possible) never rarely mostly always

8. Which disciplines were involved in development of the sedation protocol? intensivist physiotherapist nurse pneumologist anesthetist respiratory therapist psychologist I don’t know

9. What kind of sedation protocol is being used in your ICU? patient-targeted sedation protocol daily interruption of sedative infusions (DIS) a combination of both I don’t know

10. I DON’T use a sedation protocol: if a patient needs short-term sedation (<24 hours) if a physician desires to work outside protocol out of fear of overdosing sedatives in case of absence of physician’s order out of fear of underdosing sedatives if I didn’t get any explanation concerning the protocol in case of too high workload if a protocol is difficult to find in case of shortage of nursing staff I always use a protocol

11. How frequently do you assess the adequacy of sedation during a shift? (besides on clinical indications) every hour every 4 hours every 8 hours every 2 hours every 6 hours no examination (go to question 13)

Continued



nurses’ adherence to analgo-sedation recommendations in the intensive care unit—a national survey. Crit Care. 2014;18(6):655. http://ccforum.biomedcentral.com /articles /10.1186/s13054-014-0655-1. Accessed September 15, 2017.

31. O’Connor M, Bucknall T, Manias E. Sedation management in Australian and New Zealand intensive care units: doc-tors’ and nurses’ practices and opinions. Am J Crit Care. 2010;19(3):285-295.

32. Wunsch H, Kahn JM, Kramer AA, Rubenfeld GD. Use of intravenous infusion sedation among mechanically venti-lated patients in the United States. Crit Care Med. 2009; 37(12):3031-3039.

33. Bucknall TK, Manias E, Presneill JJ. A randomized trial of protocol-directed sedation management for mechanical ventilation in an Australian intensive care unit. Crit Care Med. 2008;36(5):1444-1450.

34. Likert R. A technique for the measurement of attitudes. Arch Psychol. 1932;22(140):1-55.

35. Lynn MR. Determination and quantification of content validity. Nurs Res. 1986;35(6):382-385.

36. Powell C. The Delphi technique: myths and realities. J Adv Nurs. 2003;41(4):376-382.

37. Jacobi J, Fraser GL, Coursin DB, et al; Task Force of the American College of Critical Care Medicine (ACCM) of the Society of Critical Care Medicine (SCCM), American Society of Health-System Phar-macists (ASHP), American College of Chest Physicians. Clinical practice guidelines for the sustained use of sedatives and analge-sics in the critically ill adult [published correction appears in Crit Care Med. 2002; 30(3):726]. Crit Care Med. 2002;30(1):119-141.

38. Egerod I, Christensen BV, Johansen L. Trends in sedation practices in Danish intensive care units in 2003: a national survey. Intensive Care Med. 2006;32(1):60-66.

12. Which of the instruments listed below do you use to assess the adequacy of sedation? The Richmond Agitation Sedation Scale (RASS) Glasgow Coma Scale (GCS) Numeric Rating Scale (NRS) Riker Sedation-Agitation Score (SAS) Adaptation to the Intensive Care Environment (ATICE) The Ramsay Sedation Scale (RSS) Bispectral Index (BIS) Critical Care Pain Observation Tool Bispectral Index (BIS) + a scale other Nursing Instrument for the Communication of Sedation (NICS) none

Daily interruption of sedative infusions (DIS)

13. How frequently is DIS applied in your ICU? (only 1 answer possible) never (go to question 17) rarely mostly always I don’t know

14. What are the main reasons for performing DIS in the ICU? shortening the duration of mechanical ventilation facilitation of muscles’ spontaneous movements waking up a patient prevention of posttraumatic stress disorder (PTSD) neurological evaluation obtaining better communication with a patient reduction of need for vasopressors I don’t know prevention of accumulation of sedatives

15. When is DIS NOT implemented? in the daytime at night during the weekend

16. What are the 3 most important barriers to you performing DIS? respiratory deterioration bad experiences with DIS cardiac ischemia colleagues’ bad experiences with DIS not comfortable for the patient unexperienced team greater possibility of agitation nobody does it can lead to posttraumatic stress disorder (PTSD) no added value possibility of self-extubation/removal of drains/catheters insufficient medical supervision difficult care physician does not ever suggest it a technique difficult to control it’s not in the protocol too high workload I always perform DIS insufficient knowledge

17. Identify three of the most frequently occurring problems that you recognize in your everyday practice during reduction of the sedatives. respiratory deterioration patient’s fear patient-ventilator dyssynchrony biting on endotracheal tube/tongue/lips heart rhythm disorders self-removal of drains/infusions/catheters hemodynamic instability self-extubation neurological deterioration greater need to fixate a patient agitation/confusion/ICU delirium inconvenience for nurses patient not comfortable or in pain other

18. Indicate who is involved in making decisions about:

Appendix Continued

Physician Nurse Respiratory therapist Physiotherapist Other discipline Per protocol Not applicable

Reduction of sedative infusions

Cessation of sedative infusions

Implementation of DIS


1.0 Hour Category AC ENotice to CE enrollees:


1. Analyze availability of sedation protocols and protocol adherence.2. Explore the frequency in use of daily interruption of sedation (DIS) and barriers to nurses implementing this

in practice.3. Describe the range and type of analgosedative agents used for sedation and the extent of nurses’ autonomy

in sedation practice.




(#LSBN12).

39. Martin J, Franck M, Sigel S, Weiss M, Spies C. Changes in sedation management in German intensive care units between 2002 and 2006: a national follow-up survey. Crit Care. 2007;11(6):R124.

40. Wøien H, Stubhaug A, Bjørk IT. Analgesia and sedation of mechanically ventilated patients—a national survey of clinical practice. Acta Anaesthesiol Scand. 2012;56:23-29.

41. Aitken LM, Bucknall T, Kent B, Mitchell M, Burmeister E, Keogh SJ. Protocol-directed sedation versus non-protocol-directed sedation to reduce duration of mechanical ven-tilation in mechanically ventilated intensive care patients. Cochrane Database Syst Rev. 2015;(1):CD009771.

42. Blackwood B, Wilson-Barnett J, Patterson CC, Trinder TJ, Lavery GG. An evaluation of protocolised weaning on the duration of mechanical ventilation. Anaesthesia. 2006; 61(11): 1079-1086.

43. Rose L, Dainty KN, Jordan J, Blackwood B. Weaning from mechanical ventilation: a scoping review of qualitative studies. Am J Crit Care. 2014;23(5):e54-e70.

44. Rose L, Blackwood B, Egerod I, et al. Decisional responsibility for mechanical ventilation and weaning: an international survey. Crit Care. 2011;15(6):R295.

45. Rose L, Blackwood B, Burns SM, Frazier SK, Egerod I. International perspectives on the influence of structure

and process of weaning from mechanical ventilation. Am J Crit Care. 2011;20(1):e1-e18.

46. Everingham K, Fawcett T, Walsh T. “Targeting” sedation: the lived experience of the intensive care nurse. J Clin Nurs. 2013;23(5-6):694-703.

47. Greenberg SB, Vender J. The use of neuromuscular blocking agents in the ICU: where are we now? Crit Care Med. 2013; 41(5):1332-1344.

48. Walsh TS, Kydonaki K, Antonelli J, et al; Development and Evaluation of Strategies to Improve Sedation practice in inTensive care Study investigators. Rationale, design and methodology of a trial evaluating three strategies designed to improve sedation quality in intensive care units (DESIST study). BMJ Open. 2016;6(3):e010148. http://bmjopen.bmj .com/content/6/3/e010148.full?sid=0f889067-5b56-40b8-a054-ada94d5a94d1. Accessed September 16, 2017.




Background Methods and frequency of temperature

monitoring in intensive care unit patients vary widely.

The recently available SpotOn system uses zero-heat-flux

technology and offers a noninvasive method for con-

tinuous monitoring of core temperature of critical care

patients at risk for alterations in body temperature.

Objective To evaluate agreement between and precision

of a zero-heat-flux thermometry system (SpotOn) and

continuous rectal and urinary bladder thermometry during

fever and defervescence in adult patients in intensive

care units.

Methods Prospective comparison of SpotOn vs rectal

and urinary bladder thermometry in eligible patients

enrolled in a randomized clinical trial on the effect of

acetaminophen on core body temperature and hemo-

dynamic status.

Results A total of 748 paired temperature measurements

from 38 patients who had both SpotOn monitoring and

either continuous rectal or continuous bladder thermom-

etry were analyzed. Temperatures during the study were

from 36.6°C to 39.9°C. The mean difference for SpotOn

compared with bladder thermometry was -0.07°C (SD,

0.24°C; 95% limits of agreement, ± 0.47ºC [-0.54°C, 0.40°C]).

The mean difference for SpotOn compared with rectal

thermometry was -0.24°C (SD, 0.29°C; 95% limits of

agreement, ± 0.57ºC [-0.81°C, 0.33°C]). Most differences

in temperature between methods were within ± 0.5°C

in both groups (96% bladder and 85% rectal).

Conclusions The SpotOn thermometry system has excel-

lent agreement and good precision and is a potential

alternative for noninvasive continuous monitoring of core

temperature in critical care patients, especially when alter-

native methods are contraindicated or not available.

(American Journal of Critical Care. 2018; 27:43-50)

RECTAL AND BLADDER TEMPERATURES VS FOREHEAD

CORE TEMPERATURES

MEASURED WITH SPOTON

MONITORING SYSTEMBy Hildy M. Schell-Chaple, RN, PhD, Kathleen D. Liu, PhD, MD, Michael A. Matthay, MD, and Kathleen A. Puntillo, RN, PhD

Critical Care Techniques

This article is followed by an AJCC Patient Care Page on page 51.


Critically ill patients are at risk for alterations in body temperature, which may be related to a patient’s clinical diagnosis, therapies administered, or exposure in the intensive care unit (ICU) environment (eg, during bathing and procedures). Fre-quent monitoring of body temperature facilitates early detection of changes in clini-cal condition and a patient’s responses to therapies. For example, early detection

of a fever due to an infection or a drug reaction and/or hypothermia related to continuous renal replacement therapy or a massive blood transfusion can lead to further assessments, interven-tions, or both. However, temperature monitoring in ICUs is highly variable with regard to the site of measurement, thermometry technology used, and the frequency of measurement.1-3

A continuous, noninva-sive, accurate measure

of core temperature that is feasible for use

and comfortable for patients would be ideal.

Although 37°C is widely accepted as normal core temperature, various temperatures exist within the body at the same time. The temperatures of the core organs and large vessels (eg, brain, heart, liver, pulmonary artery, jugular vein) are influenced by the metabolic rate of the thermometry site. Core tempera-

ture is typically higher than the temperatures at periph-eral tissue and skin sites because the last 2 sites are influenced by dynamic alter-ations in vascular perfusion.4 A recent meta-analysis5 indicated that esophageal, urinary bladder, and rectal temperatures are within clinically acceptable limits of agreement (LOA) com-

pared with pulmonary artery blood temperatures and hence are widely accepted as accurate estimates of core temperature. Evidence-based guidelines6 for temperature monitoring in ICU patients also recom-mend the use of pulmonary artery, esophageal, blad-der, and rectal thermometry.

The ideal monitoring system would provide a continuous, noninvasive accurate measure of core body temperature and be feasible for use, comfort-able for patients, and compatible with care interven-tions in ICU patients. Unfortunately, the standard

thermometry methods used in ICUs for continuous monitoring are invasive and have barriers to use. The pulmonary artery catheter is designed to mea-sure the temperature of blood in the pulmonary artery, and this measurement is considered the gold standard for core body temperature. Yet, pulmonary artery catheters are invasive, associated with poten-tial risks, not broadly used across ICUs, and typically used for only a short time during an ICU stay.7-10 Although strong agreement exists between tempera-tures obtained via urinary catheter and pulmonary artery catheter thermometry, urinary catheters are also invasive. Current practice standards include efforts to reduce duration of use of urinary catheters because of the risk for device-related infection of the urinary tract.11,12 Continuous esophageal probe thermometry also provides accurate measures of core temperature, but maintaining the position of the probe for more than a short time is challenging, and use of this method is typically limited to intu-bated patients.13 Continuous rectal thermometry is now used less frequently than before in ICUs for various reasons: a recent focus on prevention of device-related pressure injuries to skin and mucosa, introduction of new fecal management devices, and concerns about patients’ dignity and discomfort.3,14 Temporal artery thermometry is noninvasive, but not continuous, and is not recommended for use because of the limited precision of the method, especially in febrile patients.6,15 Thus, obtaining continuous data on core temperature in ICU patients remains a challenge.

A recently available continuous, noninvasive temperature monitoring system that uses zero-heat-flux (ZHF) technology to measure core temperature is the SpotOn system (3M Healthcare). This system consists of a control unit, a cable, and a small dispos-able sensor applied to the lateral part of the patient’s forehead. The thermal insulator and resistive warming circuit are used to eliminate flow of skin-surface heat to the environment, establishing conditions in which

About the AuthorsHildy M. Schell-Chaple is a clinical nurse specialist and an associate professor of nursing, University of Califor-nia, San Francisco Medical Center, San Francisco, Cali-fornia. Kathleen D. Liu is a professor of medicine and Michael A. Matthay is a professor of medicine and anes-thesia, University of California, San Francisco School of Medicine, San Francisco, California. Kathleen A. Puntillo is professor emeritus, University of California, San Fran-cisco School of Nursing, San Francisco, California.

Corresponding author: Hildy M. Schell-Chaple, RN, PhD, 505 Parnassus Ave, San Francisco, CA 94143-0358 (email: [email protected]).


The SpotOn system was used to measure core temperature continuously and noninvasively.

the temperature gradient should be zero between the thermistor on the skin surface and deep tissue. ZHF technology is designed to provide an estimate of core temperature by measuring the temperature from 1 to 2 cm below the skin surface via the iso-thermal tunnel created under the sensor (Figure 1). A manufacturer-sponsored trial to evaluate clinical accuracy of the device indicated excellent agreement between the forehead core SpotOn system and pul-monary artery blood temperatures, with a mean dif-ference of -0.23°C (SD, 0.42°C) between methods in perioperative cardiac surgery patients.16 The dis-tinction between temperature measurement via ZHF technology and measurement with skin sensor tech-nology with thermally sensitive resistors is import-ant. Skin temperature sensors are not insulated to ambient air and allow heat radiation from the skin. Forehead skin temperatures can be up to 3°C lower than forehead core temperatures measured at the same time by using the SpotOn system.17

Method-comparison study design is widely accepted for evaluation of technology used to mea-sure physiological variables in health care.18,19 This design is used to determine the degree of accuracy, evaluated on the basis of the mean difference between the measurement obtained with the new method and the true value or gold standard reference mea-surement. The method-comparison design may also be used when adoption of new technology is considered and new methods of measurement are compared with established methods. The mean of differences between temperatures quantifi es the degree of agreement between measurement methods in the sample. The standard deviation of the mean difference represents the degree of imprecision with repeated measurements and is used to calculate the 95% LOA, an estimate of where most differences will fall in similar subjects. The LOA and precision data help inform decisions when use of an alternative mea-surement method for clinical practice is considered.

Studies designed to evaluate the accuracy of earlier thermometry systems with ZHF technology compared with standard core thermometry methods (pulmonary artery, esophageal, nasopharyngeal, urinary bladder) used in perioperative and ICU settings have indicated that the ZHF technology is a suffi ciently accurate measure of core body tempera-ture in surgical patients.16,17,20 However, method-comparison studies of the recently available SpotOn system in diverse populations of patients across a range of temperatures are needed to determine the usefulness of this thermometry method in ICUs before the method is widely accepted in clinical

practice. Thus, the purpose of our study was to eval-uate the agreement and precision between the SpotOn temperature monitoring system and established rectal and urinary bladder continuous thermometry meth-ods during fever and defervescence in ICU patients with medical, surgical, and neurological conditions.

Methods Patients enrolled in a randomized clinical trial21

to test the effects of intravenous acetaminophen on core temperature and hemodynamic responses in febrile critically ill patients were eligible for the study. We used a method-comparison study design to compare simultaneous tempera-ture measurements obtained from the con-tinuous, noninvasive SpotOn system with measure-ments from continuous rectal or urinary bladder temperature monitoring systems. Rectal and bladder thermometry methods were selected as reference

Figure 1 3M SpotOn temperature monitoring system and isothermal tunnel formation with zero-heat-fl ux technology.

Reproduced with permission from 3M, St Paul, Minnesota.

Temperature sensor

3M™ SpotOn™control unit

Sensor connector

Sensor cable

Power supply Monitor cable


Temperatures measured with SpotOn were com-

pared with rectal and bladder temperatures.

sources of temperature; obtaining gold-standard core temperatures was not feasible because of the limited use of pulmonary artery catheters in the ICU. We hypothesized that the noninvasive SpotOn system with ZHF technology is sufficiently compara-ble to invasive continuous rectal and bladder ther-mometry methods to within a value of ± 0.5°C for agreement and precision.

A total of 41 patients enrolled from the adult medical-surgical and neuroscience ICUs at the Uni-versity of California, San Francisco Medical Center, during the period September 2013 through August 2015 were randomized to receive the study interven-tion of either intravenous acetaminophen (1 g) or placebo during a 15-minute period, and data were collected during a 4-hour period. Patients were eligi-ble for enrollment if they were 18 years or older,

weighed at least 50 kg (110 lb), had a fever (temperature ≥ 38.3°C [≥ 100.9°F]) and did not meet exclusion criteria of malignant hyperthermia or heat stroke, acute liver failure, therapeutic tem-

perature management, or extracorporeal blood circuit therapy. Among the 41 patients, 3 did not have continuous rectal or bladder thermometry and therefore were excluded from this study. Approval was obtained from the appropriate institutional review board, and signed written informed consent was obtained from patients or patients’ surrogates before data collection. Patients’ characteristics, ther-apies administered during the study period, and ambient room temperature were recorded. Physical cooling and warming interventions were not permit-ted during the study period.

Temperature MeasurementForehead Core Temperatures. Forehead core tem-

perature was measured on all patients by using the noninvasive SpotOn temperature monitoring sys-tem during the study period. The SpotOn system is accurate for a temperature range of 25°C to 43˚C (SD, 0.2°C). The disposable sensor (SpotOn Tem-perature Monitoring Sensor, model 36000) was connected to the control unit (SpotOn Temperature Monitoring System Control Unit, model 370). After the skin was cleansed with alcohol, the sensor was placed on the left lateral part of the patient’s fore-head above the orbital ridge. Time (1-4 minutes) was allotted for system equilibration of temperature from the deeper tissue to the skin surface.

Rectal Temperatures. Rectal temperatures were measured continuously by using temperature probes inserted at least 10 cm (3.9 in) into the rec-tum and connected to the Solar 8000i Bedside Monitor (GE Healthcare). The temperature probes (Level-1, Smiths Medical ASD Inc) have thermally sensitive resistors that measure temperature in the surrounding environment.

Bladder Temperatures. Bladder temperature-sensing catheters (DeRoyal, DeRoyal Industries, Inc) were used for continuous monitoring and were connected to the Solar 8000i Bedside Monitor. The bladder ther-mometers also have thermally sensitive resistors that measure the temperature of the urine flow through the catheter.

Calibration. All the thermometer systems used in the study were calibrated in accordance with manu-facturers’ recommendations. The same investigator (H.S.) applied the thermometry systems and measured temperatures to ensure fidelity to the study protocol. Simultaneous measurements of temperatures from both the SpotOn system and the rectal or bladder thermometry system were recorded at baseline, every 5 minutes for 15 minutes, and then every 15 minutes for the 4-hour study period.

Statistical AnalysisThe Bland-Altman method22 was used to ana-

lyze the comparison data to estimate the direction and extent of agreement and the precision among the measurement methods. Graphical plots of the method temperature differences were examined for patterns, and calculated estimates for agreement and precision were analyzed. Agreement is presented as the bias (mean difference) with the 95% CI. Pre-cision is presented as the 95% LOA, either a single value (± [SD × 1.96]) or as upper and lower LOA values surrounding the bias (bias ± [SD × 1.96]). For example, if the bias is 0.25 with a standard deviation of 0.15, then the LOA would be 0.15 × 1.96 = ± 0.29 or 0.25 ± 0.29 = -0.04, 0.54). In this article, both single values and upper and lower values for the 95% LOA are presented. A priori, we chose ± 0.5°C as a clinically acceptable limit for agreement and precision. Consistent with previous thermome-try method-comparison studies13,16,17,20,23,24 in popu-lations of critically ill patients, this limit was selected because differences beyond these limits are deemed clinically significant.

The proportion of temperature differences within ± 0.5°C from SpotOn and corresponding rectal and bladder temperature comparisons were also computed for both groups. Large proportions of temperature


differences beyond ± 0.5°C were interpreted as clinically relevant. Analysis of data for the subset of patients with external ventricular drainage devices was also conducted to evaluate for differences in agreement and precision.

A sample size for the method-comparison analysis was not calculated before the study because the sample was based on the number of patients eligible for the clinical trial. Yet, the sample sizes of 180 and 568 paired temperatures in the bladder and rectal comparisons, respectively, provided adequate power to detect a significant difference between the measurement methods. Data were analyzed by using SPSS, version 24, computer software (SPSS, Inc).

Results A total of 748 paired temperature measurements

from 38 patients from the clinical trial who had both SpotOn and either continuous rectal (n = 29) or bladder (n = 9) thermometry were included in this study. A total of 20 temperature pairs were recorded from all patients with the exception of 8 pairs col-lected from 1 patient who did not complete the primary study protocol. Patients’ characteristics for the sample are shown in the Table. A total of 20 patients in the sample (53%) received acetamin-ophen during the study period. Temperature ranges during the study period were from 36.9°C to 39.7°C (rectal), 36.9°C to 39.9°C (bladder), and 36.6°C to 39.4°C (forehead core SpotOn).

Figures 2 and 3 display the Bland-Altman plots with graphical presentation of agreement and preci-sion for SpotOn compared with bladder temperatures and SpotOn compared with rectal temperatures, respectively. Inspection of the data plots revealed no patterns of temperature differences between SpotOn and either rectal or bladder thermometry methods across the range of temperature values. The mean difference for SpotOn measurements compared with bladder measurements was -0.07°C (SD, 0.24°C; 95% CI, -0.04°C to -0.11°C) with 95% LOA of ± 0.47°C (-0.54°C, 0.40°C). The mean dif-ference for SpotOn measurements compared with rectal measurements was -0.24°C (SD, 0.29°C; 95% CI, -0.21°C to -0.26°C) with 95% LOA of ± 0.57°C (-0.81°C, 0.33°C).

Most differences in temperature between the methods of measurement used were within ± 0.5°C in both groups. The proportion of differences within ± 0.5°C was high at 96% (95% CI, 93%-99%) in the bladder comparison group and 85% (95% CI, 82%-88%) in the rectal comparison group.

The subset of 10 patients who had neurological injury and an external ventricular drainage device had intracranial pressure monitoring as well as rectal ther-mometry. The results of the subgroup analysis of 200 paired SpotOn temperatures with rectal temperatures in this subset were similar to the results of the larger sample: subset mean difference -0.29°C (SD, 0.33°C) with 95% LOA of ± 0.65°C (-0.95°C, 0.36°C).

No indications of skin irritation under the sensor were noted when the sensor was removed after 4.5 hours, and discomfort was not mentioned by patients who could self-report. Despite sweating in some patients during febrile periods and deferves-cence, the forehead ZHF sensors maintained their seal during the study period. Severe diaphoresis related to autonomic dysfunction in 1 neurologi-cally injured patient did not disrupt the SpotOn thermometry system, and the ZHF sensor main-tained its seal. Also, the mean difference between this patient’s 20 paired rectal measurements and SpotOn temperatures was 0.24°C (SD, 0.11°C).

Characteristic

TableCharacteristics of 38 patients and the environment

Age, mean (SD), range, y

Sex, male/female, No. (%)

Ethnicity, No. (%) Asian/Pacific Islander Black White Hispanic

Admitting diagnosis type,a No. (%) Medical Neurological Surgical

APACHE II score, mean (SD), range

Body mass index,b mean (SD), range

Mechanical ventilation, No. (%)

Cause of fever, No. (%) Infectious Neurological

Ambient ICU room temperature, mean (SD), ˚C

Temperature, mean (SD), range, ˚C Bladder comparison (n = 180) Urinary bladder SpotOn forehead core Rectal comparison (n = 568) Rectal SpotOn forehead core

57 (15), 20-78

20/18 (53/47)

10 (26)2 (5)

17 (45) 9 (24)

9 (24)23 (61) 6 (16)

24 (6), 14-43

29.8 (6.6), 18.1-48.2

26 (68)

28 (74)10 (26)

21.4 (1.2)

38.2 (0.61), 36.9-39.938.2 (0.71), 36.7-39.9

38.4 (0.53), 36.9-39.738.2 (0.59), 36.6-39.4

Value

Abbreviations: APACHE, Acute Physiology and Chronic Health Evaluation; ICU, intensive care unit.a Because of rounding, percentages may not total 100.b Calculated as weight in kilograms divided by height in meters squared.


Discussion This study is the fi rst one in which the SpotOn

system was compared with 2 established thermome-try methods used to monitor core temperature in ICU patients. Our results indicate that the SpotOn system has excellent agreement with the established clinical thermometry systems. Both comparison groups were within the a priori defi ned accuracy limits (difference of means) for both bladder and rectal groups and revealed slight underestimates of temperature by the SpotOn system compared with both rectal and bladder temperatures. The SpotOn system also had good precision in both compari-son groups even though the 95% LOA for the rectal group slightly exceeded the a priori defi ned clini-cally acceptable LOA of ± 0.5°C for precision (95% LOA: bladder, ± 0.47°C; rectal, ± 0.57°C).

Eshraghi et al16 evaluated the accuracy of the prototype SpotOn system by comparing the fore-head core (SpotOn Prototype) with pulmonary artery temperatures in 105 cardiac surgical patients during intraoperative and postoperative ICU periods. The level of agreement and precision were suffi-ciently accurate for use in intraoperative and post-operative clinical practice: mean overall difference, -0.23°C (SD, 0.42°C) with 95% LOA of ± 0.82°C (-1.06°C, 0.60°C). The levels of agreement and precision for the temperature comparisons in the postoperative ICU subset were similar to the com-bined intraoperative and postoperative estimates: mean difference, -0.32°C (SD, 0.38°C) with 95% LOA of ± 0.74°C (-1.06°C, 0.42°C). Although Eshraghi et al compared SpotOn temperatures with pulmonary artery blood temperatures, the percentage of differences (84%) within 0.5°C in the ICU subset was similar to our fi ndings.

In studies16,17,20,25 of other thermometry systems with ZHF technology, mostly prototypes, tempera-tures obtained with the ZHF systems were compared with temperatures from pulmonary artery, esopha-geal, and bladder sources. In a method-comparison study25 of esophageal and forehead core tempera-tures in which a ZHF thermometry prototype was used in patients during targeted hypothermia ther-apy and rewarming after cardiac arrest, the results indicated a good level of agreement, with a mean difference of -0.12°C, and good precision with a 95% LOA of ± 0.48°C ( -0.59°C, 0.36°C). Yet, Maki-nen et al17 found poor agreement between SpotOn and nasopharyngeal temperatures in cardiac surgery patients during induced hypothermia. A limitation of our study was the lack of hypothermic patients in our sample; that condition warrants continuous

Figure 2 Bland-Altman plot of SpotOn forehead core tempera-tures and urinary bladder temperatures (n = 180 paired measure-ments). Bias, -0.07˚C (SD, 0.24˚C); 95% limits of agreement (LOA), -0.54ºC, 0.40ºC. Bolded outline of plot points indicates multiple difference measures.

39.538.537.536.5

-0.90

-0.65

-0.40

-0.15

0.10

Upper LOA 0.40

Lower LOA -0.54

Bias -0.07

0.35

0.60

Mean: (SpotOn core + bladder temperature)/2, ºC

Dif

fere

nce

: Sp

otO

n c

ore

– b

lad

der

tem

per

atu

re, º

C

Figure 3 Bland-Altman plot of SpotOn forehead core temperatures and rectal temperatures (n = 568 paired measurements). Bias, -0.24˚C (SD, 0.29˚C); 95% limits of agreement (LOA), -0.81ºC, 0.33ºC. Bolded outline of plot points indicates multiple differ-ence measures.

39.538.537.536.5

-1.50

-1.00

-0.50

0.00

Upper LOA 0.33

Lower LOA -0.81

Bias -0.24

0.50

Mean: (SpotOn core + rectal temperature)/2, ºC

Dif

fere

nce

: Sp

otO

n c

ore

– r

ecta

l tem

per

atu

re, º

C


thermometry, especially during targeted temperature management in critically ill patients.

Makinen et al also compared the SpotOn sys-tem with esophageal thermometry in patients under-going cardiac and vascular surgery and reported excellent agreement and precision: mean difference, 0.08°C (SD, 0.16°C) and a 95% LOA of ± 0.32°C (-0.25°C, 0.40°C). Iden et al26 compared nasopha-ryngeal temperatures with forehead core tempera-tures obtained by using the SpotOn system during elective surgical procedures and also found excel-lent agreement and precision: mean difference, 0.07°C (SD, 0.21°C) and a 95% LOA of ± 0.41°C (-0.34°C, 0.48°C). The agreement and precision findings in both of these studies are similar to our findings for bladder thermometry. Esophageal and nasopharyngeal sources are considered accurate and reliable estimates of core temperature.

The 4-hour time frame of the study period is a limitation of our study. Further evaluation of sensor securement to maintain ZHF insulation and the related reliability over longer periods is warranted. Evaluation of the SpotOn system with cooling via fans and other physical interventions used in the ICU is also indicated because we did not permit physical cooling, including use of fans, during the study period. Another limitation of our study is that the SpotOn system was not evaluated in patients with spontaneous or induced mild hypothermia, a condition that warrants continuous thermometry.

Conclusion The findings of our method-comparison study

indicate that the SpotOn system is a suitable nonin-vasive alternative to established invasive methods for continuous thermometry in medical, surgical, and neurologically injured ICU patients. The SpotOn system is an appealing alternative for continuous measurement of core temperatures because of its unique noninvasive design. Further research to eval-uate agreement and precision of the SpotOn system during longer periods and during targeted tempera-ture therapies in ICU patients is warranted.

ACKNOWLEDGMENTSWe thank 3M for lending us the SpotOn device; Daniel Sessler, MD, chair of the Department of Outcomes Research, Cleveland Clinic, Cleveland, for comments and suggestions early in the project; and S. Paul, PhD, princi-pal biostatistician, University of California, San Francisco School of Nursing, for assistance in statistical analysis.This research was performed at the University of Califor-nia, San Francisco Medical Center, School of Nursing, and School of Medicine.

REFERENCES1. Egi M, Morita K. Fever in non-neurological critically ill patients:

a systematic review of observational studies. J Crit Care. 2012;27(5):428-433.

2. Johnston NJ, King AT, Protheroe R, Childs C. Body tempera-ture management after severe traumatic brain injury: meth-ods and protocols used in the United Kingdom and Ireland. Resuscitation. 2006;70(2):254-262.

3. Niven DJ, Laupland KB, Tabah A, et al; EUROBACT Investi-gators. Diagnosis and management of temperature abnor-mality in ICUs: a EUROBACT investigators’ survey. Crit Care. 2013;17(6):R289.

4. Taylor NA, Tipton MJ, Kenny GP. Considerations for the measurement of core, skin and mean body temperatures. J Therm Biol. 2014;46:72-101.

5. Niven DJ, Gaudet JE, Laupland KB, Mrklas KJ, Roberts DJ, Stelfox HT. Accuracy of peripheral thermometers for estimat-ing temperature: a systematic review and meta-analysis. Ann Intern Med. 2015;163(10):768-777.

6. O’Grady NP, Barie PS, Bartlett JG, et al; American College of Critical Care Medicine; Infectious Diseases Society of America. Guidelines for evaluation of new fever in critically ill adult patients: 2008 update from the American College of Critical Care Medicine and the Infectious Diseases Soci-ety of America. Crit Care Med. 2008;36(4):1330-1349.

7. Bignami E, Belletti A, Moliterni P, Frati E, Guarnieri M, Tri-tapepe L. Clinical practice in perioperative monitoring in adult cardiac surgery: is there a standard of care? results from a national survey. J Clin Monit Comput. 2015; 30(3): 347-365.

8. Koo KK, Sun JC, Zhou Q, et al. Pulmonary artery catheters: evolving rates and reasons for use. Crit Care Med. 2011; 39(7): 1613-1618.

9. Bussières JS. Iatrogenic pulmonary artery rupture. Curr Opin Anaesthesiol. 2007;20(1):48-52.

10. Villaverde RV, Vanhaebost J, Grabherr S, Palmiere C. Pul-monary artery rupture during Swan-Ganz catheterisation: a case report. Leg Med (Tokyo). 2014;16(2)76-80.

11. Knapik P, Rychlik W, Duda D, Golyszny R, Borowik D, Ciesla D. Relationship between blood, nasopharyngeal and urinary bladder temperature during intravascular cooling for thera-peutic hypothermia after cardiac arrest. Resuscitation. 2012; 83(2):208-212.

12. Al-Hazmi H. Role of duration of catheterization and length of hospital stay on the rate of catheter-related hospital-acquired urinary tract infections. Res Rep Urol. 2015;7:41-47.

13. Lefrant JY, Muller L, de La Coussaye JE, et al. Temperature measurement in intensive care patients: comparison of uri-nary bladder, oesophageal, rectal, axillary, and inguinal methods versus pulmonary artery core method. Intensive Care Med. 2003;29(3):414-418.

14. Hammond NE, Saxena MK, Taylor C, et al; George Institute for Global Health; Australian and New Zealand Intensive Care Society Clinical Trials Group. Temperature management of non-elective intensive care patients without neurological abnormalities: a point prevalence study of practice in Aus-tralia and New Zealand. Crit Care Resusc. 2013; 15(3):228-233.

15. Geijer H, Udumyan R, Lohse G, Nilsagård Y. Temperature measurements with a temporal scanner: systematic review and meta-analysis. BMJ Open. 2016;6(3):e009509.

16. Eshraghi Y, Nasr V, Parra-Sanchez I, et al. An evaluation of a zero-heat-flux cutaneous thermometer in cardiac surgical patients. Anesth Analg. 2014;119(3): 543-549.

17. Makinen MT, Pesonen A, Jousela I, et al. Novel zero-heat-flux deep body temperature measurement in lower extremity vascular and cardiac surgery. J Cardiothorac Vasc Anesth. 2016;30(4):973-978.

18. Bland JM, Altman DG. Measuring agreement in method com-parison studies. Stat Methods Med Res. 1999;8(2):135-160.

19. Hanneman SK. Design, analysis, and interpretation of method-comparison studies. AACN Adv Crit Care. 2008; 19(2): 223-234.

20. Langham GE, Maheshwari A, Contrera K, You J, Mascha E, Sessler DI. Noninvasive temperature monitoring in postan-esthesia care units. Anesthesiology. 2009;111(1):90-96.

21. Schell-Chaple HM, Liu KD, Matthay MA, Sessler DI, Puntillo KA. Effects of IV acetaminophen on core body temperature


and hemodynamic responses in febrile critically ill adults: a randomized controlled trial. Crit Care Med. 2017;45:1199-1207.

22. Bland JM, Altman DG. Agreement between methods of measurement with multiple observations per individual. J Biopharm Stat. 2007;17(4):571-582.

23. Kimberger O, Thell R, Schuh M, Koch J, Sessler DI, Kurz A. Accuracy and precision of a novel non-invasive core ther-mometer. Brit J Anaesth. 2009;103(2):226-231.

24. Lawson L, Bridges EJ, Ballou I, et al. Accuracy and precision of noninvasive temperature measurement in adult intensive care patients. Am J Crit Care. 2007;16(5):485-496.

25. Zeiner A, Klewer J, Sterz F, et al. Non-invasive continuous cerebral temperature monitoring in patients treated with mild therapeutic hypothermia: an observational pilot study. Resuscitation. 2010;81(7):861-866.

26. Iden T, Horn EP, Bein B, Böhm R, Beese J, Höcker J. Intraop-erative temperature monitoring with zero heat fl ux tech-nology (3M SpotOn sensor) in comparison with sublingual and nasopharyngeal temperature: an observational study [published correction appears in Eur J Anaesthesiol. 2015; 32(10):747]. Eur J Anaesthesiol. 2015;32(6):387-391.


CHOOSING WISELY FOR MONITORING PATIENTSBy Cindy Cain, RN, DNP, CNS, CCRN

AJCC Patient Care Page The AJCC Patient Care Page is a service of the American Journal of Critical Care and the American Association of Critical-Care Nurses. Designed to elaborate on AACN practice guidelines based on content in select articles, this page may be photocopied noncommer-cially for use by readers in their workplace, in continuing education programs, or for distribution to colleagues, patients, or patients’ families. To purchase bulk reprints, call (800) 899-1712.

Based on material from and published as a supplement to the article by Schell-Chaple et al, “Rectal and Bladder Temperatures vs Forehead Core Temperatures Measured With SpotOn Monitoring System” (American Journal of Critical Care. 2018;27:43-50).

Health care professionals are encouraged through campaigns like Choosing Wisely to select tests and procedures on the basis of supporting evidence, necessity, and freedom from harm. Collecting and

assessing patients’ data are core elements of nursing care. However, the imperative of collecting accurate data must be weighed against the risks that procedures may pose. For instance, pulmonary artery catheters can pro-vide information about a patient’s hemodynamic status, but they do so at a price; their insertion can contribute to poor outcomes for patients because of the increased risk for infection, injury, and bleeding. A noninvasive procedure such as a passive leg raise can guide fluid resuscitation without those risks.

Schell-Chaple and colleagues compared rectal and bladder temperatures with forehead core temperatures measured by using a noninvasive monitoring system. Their evidence suggests that the noninvasive system is as reliable as, more comfortable than, and safer than either of the invasive measurements of rectal and bladder tem-perature. The key to safety is not only the degree of risk in any given measurement method; safety also depends on the technique and accuracy of the person using the devices. As guardians of patient safety, nurses who are choosing wisely may need to switch to an equally accu-rate but less invasive strategy for collecting and assessing patients’ data.

Here’s what you can do:• Discuss evidence that less invasive monitoring strat-

egies can yield equally accurate information.• Choose the safest, least invasive method for all phys-

iological measurements, for example, monitoring a patient’s temperature.

• Make sure that you are using each physiological monitoring system according to the manufacturer’s recommendations.

• Be aware of any risks associated with use of the sys-tem and institute interventions to minimize those risks.

• Ensure that all users of measurement equipment are competent in use of the devices.

• Consider alternative practices in your unit in situations in which invasive devices are used for data collection and monitoring of patients.

Other helpful resources:• Deranged Physiology provides a comparison of the

advantages and limitations of various ways of monitoring body temperature in “Methods of Temperature Monitoring.” http://www.derangedphysiology.com/main/required-reading /cardiac-arrest-and-resuscitation/Chapter%203.5.1/methods- temperature-monitoring.

• Choosing Wisely. Twenty Things Nurses and Patients Should Question. http://www.choosingwisely.org/societies /american-academy-of-nursing/.

• View the AACN webinar: “Integrating Functional Hemo-dynamics at the Point of Care.” https://www.aacn.org/education /webinar-series/wb0014/integrating-functional- hemodynamics -at-the-point-of-care.

• AACN Practice Alert: Obtaining accurate noninvasive blood pressure measurements in adults. Crit Care Nurse. 2016; 36(3):e12-e16.

Illustration by Steve Oh





Background Family-centered care is a proposed way of

supporting family involvement with a child’s care and

decreasing distress associated with a child’s critical illness

by improving communication, helping manage stress

and coping, and decreasing conflicts. Nurses are critical

to successful implementation of family-centered care.

Objectives To describe nurses’ perceptions of the bene-

fits and challenges of providing family-centered care in

pediatric intensive care units.

Methods Semistructured interviews of 10 bedside and

charge nurses in pediatric, cardiac, and neonatal inten-

sive care units. Questions were related to 4 domains:

the intensive care unit environment and its relationship

to the structure and delivery of critical care, stressors

for nurses and families, communication challenges

and strategies, and involvement of families in care

and decision-making.

Results The main thematic finding was the nurses’

descriptions of a “balancing act” to provide quality

family-centered care. The balancing act was character-

ized by the interaction between 2 types of changes:

(1) intensive care unit policies related to visitation hours

and family presence at the bedside and (2) physical

transformations in the intensive care unit from shared

open space to individual private rooms.

Conclusions All of the nurses viewed the transition to

family-centered care as having benefits for families.

They also described how changes had created new

challenges for the delivery of nursing care in intensive

care units, particularly regarding mentorship and the

safety of patients and staff. (American Journal of Criti-

cal Care. 2018; 27:52-58)

NURSES’ REFLECTIONS ON

BENEFITS AND CHALLENGES

OF IMPLEMENTING FAMILY-CENTERED CARE IN PEDIATRIC

INTENSIVE CARE UNITSBy Heather Coats, PhD, MS, APRN, Erica Bourget, MPH, Helene Starks, PhD, MPH, Taryn Lindhorst, PhD, MSW, Shigeko Saiki-Craighill, RN, PhD, J. Randall Curtis, MD, MPH, Ross Hays, MD, and Ardith Doorenbos, RN, PhD

F amilies in Critical Care


Nurses are critical to successful imple-mentation of family-centered care.

The pediatric intensive care unit (PICU) is a stressful environment that places con-siderable demands on patients and their families and can have negative effects on their short- and long-term psychosocial outcomes.1-4 The medical complexity and seriousness of the child’s illness can be overwhelming for patients’ families and providers.5-8 A shift to family-centered care (FCC) has been proposed as a way to

support the family’s involvement with their child and decrease some of the distress by improv-ing communication, helping manage stress and coping, and decreasing conflicts.9-15

Changes to support FCC include alterations of the physical environment and modifications of poli-cies and clinical interactions. Historically, parents were allowed to visit their child in the PICU only briefly because of concern for infection control, privacy, and space. However, when research indi-cated that children were less stressed and were com-forted by having their parents present at the bedside,8 policy changes allowed longer family visits, includ-ing 24-hour access. Changes in the physical environ-ment included transitions from open floor plans to individual private rooms.

Nurses play a critical role in the successful implementation of FCC. They are often the first point of contact for patients’ families and are a con-sistent presence at the bedside, which allows nurses to assess family members’ coping abilities, emotional

states, needs, and preferences for information.8 Nurses are advocates to ensure that the entire family’s needs are met.8,16 They serve as educators by interpreting medical information and facilitating communica-tion between physicians and family.17 The purpose of this study is to describe PICU nurses’ perceptions of the benefits and challenges of providing FCC, given the policy and practice changes needed to facilitate FCC in PICUs.

Methods Overview and Setting

In this study, we used the qualitative description method described by Sandelowski.18 The goal is to produce a descriptive analysis with re-presentation of data that remains near to the original data. Although description is the key feature, analytic interpretation remains appropriate and essential to this method.18 Some of the common features of qualitative description are purposeful sampling, semistructured open-ended interviews, and concur-rent qualitative content analysis.18,19

This analysis is part of a larger study of stressors and supports for family members of children who are in a PICU for more than 1 week.20 We sought a purposeful sample of nurses with maximum varia-tion of participant cases.21 This sample included nurses from the PICU, cardiac ICU (CICU), and neonatal ICU (NICU) in an urban children’s hospital. We used this sampling strategy to ensure representation from the 3 ICUs and to include nurses with a range of years of experience and responsibilities (eg, bedside or charge nurse). The authors anticipated that a sample of nurses varying in these particular characteristics would provide a range of perspectives that could be important for analysis.22,23 All partici-pants in this study received a $20 gift card. The hospital’s institutional review board granted ethical approval for the study.

About the AuthorsHeather Coats is an assistant professor of research, Uni-versity of Colorado College of Nursing, Denver, Colorado. Erica Bourget is a research assistant at the Center for Child Health, Behavior and Development, Seattle Children’s Research Institute, Seattle, Washington. Helene Starks is an associate professor, Department of Bioethics and Humanities, University of Washington School of Medi-cine, Seattle, Washington. Taryn Lindhorst is a professor, University of Washington School of Social Work, Seattle, Washington. Shigeko Saiki-Craighill is a professor, Faculty of Nursing and Medical Care, Keio University, Tokyo, Japan. J. Randall Curtis is a professor, Department of Medicine, Division of Pulmonary and Critical Care Medicine, Uni-versity of Washington School of Medicine. Ross Hays is a professor in the Center for Child Health, Behavior and Development, Seattle Children’s Research Institute, and in the Department of Rehabilitative Medicine, Univer-sity of Washington School of Medicine. Ardith Doorenbos is a professor in the Department of Biobehavioral Nursing and Health Systems, University of Washington School of Nursing; the Department of Bioethics and Humanities, University of Washington School of Medicine; and the Department of Anesthesiology and Pain Medicine, Uni-versity of Washington School of Medicine.

Corresponding author: Ardith Doorenbos, RN, PhD, FAAN, Biobehavioral Nursing and Health Systems, University of Washington, Health Sciences Building, 1959 NE Pacific Street, Office Box 357266, Seattle, WA 98195-7266 (email: [email protected]).


Nurses saw family-centered care as having benefits for families and

creating new challenges for nurses.

Data Collection and Analysis We conducted the 1-time semistructured inter-

views by using an interview guide with questions in 4 domains of interest to the main study: the ICU environment and its relationship to the structure and delivery of critical care, stressors for nurses and families, communication challenges and strategies, and involvement of families in the care and decision-making processes. We made audio recordings of all interviews, which lasted between 30 and 60 min-utes. Interview audio recordings were professionally

transcribed verbatim and verified by the interviewer for accuracy.21

Qualitative content analysis is a primary ana-lytic approach in qualita-tive description studies. We elected to use conven-tional content analysis as described by Hsieh and

Shannon24 because the lack of existing publications specific to nurses’ experience with FCC precluded the use of a priori codes. A 4-member analysis team (E.B., H.S., A.D., and N.O.) reviewed the transcripts. Analysis began with all 4 team members inde-pendently reading and coding 2 of the transcripts to develop a coding scheme. Team members com-pared and consolidated the preliminary codes and entered them, along with definitions for each code, into a qualitative data management program (Dedoose; SocioCultural Research Consultants, LLC). E.B. then coded the remaining 8 interviews, which 1 other team member also reviewed. Agreement about the coding was consistently greater than 80%.

Following this initial coding, the team com-pared responses from the participants and identified themes that characterized the main topics across the 4 domains of interest. To gain new insights and to develop categories inductively, we looked for con-nections, overlaps, and contrasts between data ele-ments and codes.24 Later in the analysis, we grouped, synthesized, and consolidated categories to form themes that encompassed major related concepts. We compared similarities and differences in partici-pant characteristics within and between themes. We discussed these themes and example quotes in team meetings and synthesized them into 2 separate but related themes, reported in the Results.

Results Participants

Ten nurses volunteered and completed inter-views. Reflecting the composition of nurses working

in the ICUs, the nurses were all women and had 1 to 39 years of experience (mean, 20 years; SD, 14.25 years). Four worked in the PICU, 3 worked in the CICU, and 3 worked in the NICU. Three (1 from each unit) were charge nurses.

Themes All of the nurses viewed the transition to FCC and

the associated changes in the physical and policy environments as having benefits for families. How-ever, all of the nurses also described how changes had created new challenges for the delivery of nurs-ing care in PICUs. Overall, the nurses described these benefits and challenges to providing quality FCC as a “balancing act” that was characterized by the inter-action between 2 types of changes: (1) ICU policies related to visitation hours and family presence at the bedside and (2) transformation of the physical ICU environment from a shared open space to individual private rooms. The Table provides an overview of these 2 themes. Details regarding the benefits and challenges of these changes are described in the fol-lowing paragraphs with exemplar quotes. The quoted speakers are identified by ICU and years of experience.

Change in ICU Policy: 24-Hour Visitation Recent policy changes allow families to be pres-

ent 24 hours a day with their child. This increased bedside presence has created opportunities for par-ents to provide hands-on care and receive real-time information about their child’s medical condition. These changes have provided both benefits and chal-lenges for nurses caring for the child in the complex PICU environment.

It’s good for [the families] to be there all the time, [so the child can] hear their voices, talk to them, which is great, but it does make it more difficult for us. We have so much technology to deal with, like a code happens and just the hustle and bustle of the ICU. I think it’s trying to balance that. (NICU-3, 1 year)

Benefits. Having families constantly at the bed-side gave nurses more opportunities to build rela-tionships and trust and to include families in their child’s care. Nurses reported giving family members “little jobs” such as changing diapers, giving the child a bath, or helping in certain procedures, which gave family members some control, allowed them to assume parental roles again, and “put a little more normalcy into their life.” It also showed parents that the nurses were trustworthy and allowed them to feel it was safe to leave the bedside.


This constant family presence allowed nurses to get to know families better, including their back-ground, their understanding of their child’s medical condition, and their preferences for care. It also cre-ated more opportunities to attend to the family’s needs as well as their child’s. For example, 1 nurse spoke of providing “simple acts of kindness” to families:

When a family’s really exhausted, and they’ve passed out on the back couch, and they don’t have a pillow or a blanket, if you stop what you’re doing, and go and get them a pillow and a warm blan-ket, that makes a huge difference. Some of them are shocked. They are like, “Thank you so much,” a simple act of kindness for a family does a lot to establish trust in the relationship. (CICU-1, 30 years)

Challenges. The nurses also reported that having the family at the bedside 24 hours a day could be “distracting” and “exhausting,” especially when families asked a lot of questions when the nurse was in the middle of providing hands-on care to a criti-cally ill child. Family presence meant dividing their attention between the child and family, which was part of the balancing act when the nurse needed to focus on the child. As 1 nurse explained, “[The

parents are] just talking nonstop and needing to know every little thing that you’re doing. You can be just stressful busy and they are asking questions non-stop. It’s a disruption when you’re trying to think of what you’re supposed to do” (PICU-1, 34 years).

Nurses also spoke about the culture shift toward FCC in which families are empowered to be more directive in their child’s care and have more say in how things are done. One nurse described the con-cern of how “nurses can be fired now by families” on the basis of different styles of nursing care.

If a nurse does something a specific way, just because that’s her routine and that’s her way of doing it, but it comes across to the parents that’s the way it’s supposed to be done, and then another nurse does it definitely a right way but a different way, then that can be hard for the families. Also with families whose kids are chronic and have been here a really long time and they have their way at home or doing things, they’ll be really aggressive if you don’t do it the way they do it at home. (PICU-4, 5 years)

The culture shift embraces a customer-service approach that can create conflicts with the nurses’

Theme

TableThematic categories of benefits and challenges

Policy 24-hour visitation

Family involvement

Physical structure Privacy

Patient safety

Nurses “stretched” to provide care to address the needs of seriously ill child while also attempting to attend to families’ needs, stressors, and grief

Families place higher importance on small acts of kindness than on complex medical care of the child

Uncoordinated communication: multiple physicians/teams say different things to the family

Family relies on nurses to explain the big picture and help reconcile different physicians’ views

Family input seen as driving care and proceduresDisagreements over importance of child’s care needs can lead

to distressNurses responsible for the child’s outcomes (legal consequences)

Nurse-to-nurse shift handoffs lack the opportunity to provide rapid technical communication

Mentoring other nurses hindered by change to individual rooms; must call mentee into individual rooms to see skills/interaction with family

Nurse with 2 patients unable to see and hear both rooms/alarms at the same time

Isolation makes it difficult for nurse to seek out help from other nurses

Comfort for patientFamily able to receive information and ask questionsFamily participation in patient’s careFacilitation of relationship building and capacity for

advocacyFamily provides continuity across multiple shift

changes and team turnovers

Family participation in rounds for discussion of patient’s daily care plan

Nurses able to interpret and clarify information and answer questions immediately

Family as “experts” on the child can help with patient care and decision-making

Allows families to have a sense of control and assume parental roles

Comfort for family and patientNurse can discuss sensitive information at bedside

with family

Reduces infection ratesReduces noise disruption from other patients and

families

Challenges for nursesBenefits to families


sense of their purpose and mission, which they understand to be caring primarily for critically ill children. Although they want to accommodate families’ needs and preferences, the nurses’ first commitment is to ensuring high-quality patient care. One nurse spoke about this issue:

I think medicine has kind of moved away to where we’re trying to be like every other profession. I feel like [the families are] basically going to Nordstrom’s and expect a certain level of service versus coming to the hospital for care. This isn’t about how comfortable you are here. It’s about how well we can get your baby and then you go home. It’s not so you can hang out here and be like, “Oh, I loved it. I have luxuries of home here.” [Chuckles.] . . . I say that at some point I’m gonna walk around with a little tip jar and say, “Tip me for how well you think I did today.” I mean every single ounce of it is customer service, so that, I think, is killing my love for nursing, because I do love the aspect of critical thinking and saving a life, obviously. You can’t beat that, but customer service—that is not what I went in here for. (PICU-2, 10 years)

Nurses also felt the pressure of being constantly observed by families and not being able to do a thor-ough handoff to the next nurse at shift change if the family was present at bedside. This sometimes meant having to make extra time to find a private place and leave the room to share information.

It’s so family-[focused] care now to be almost extreme and the nurses are like, what about nursing care? We don’t get our half-hour to shift change anymore . . . a parent never has to leave now. Never. . . . We really need that [time] from 7:00 to 7:30 change of shift to talk not in front of [the parents]. (NICU-1, 21 years)

Changes in the Physical ICU Environment During the study period, the ICUs at the urban

children’s hospital underwent a transition in which the physical space was redesigned from a single open, multiple-bed room where nurses worked side by side to individual patient rooms where nurses now work alone with each patient and family. The nurses in this study, with a wide range of years in practice, described how neither the shared nor

private space created the perfect environment for delivering FCC.

Benefits. Individual rooms reduce the rate of patient infection, decrease stressful stimuli, and provide privacy and space for patients and their families. One nurse described the transition from the previous open design to the individual room:

The peds ICU looked entirely different and was one big, open room basically with beds coming out, radiating out from a center place. It was very noisy. No privacy at all. Rocking chair hit rock-ing chair. . . . The only way you had a single room was if you needed isolation for something. When this unit opened up and [we] moved in here, it was entirely different and you had to get used to the single rooms. (NICU-2, 26 years)

Family privacy was described as one of the sig-nificant benefits of individual rooms, “where [the parents] can be with their child,” “sleep in the room,” and “do things such as breastfeed in the room.”

Challenges. Although the private rooms increased protection from infection and provided privacy for families, they created different issues regarding patient safety, particularly when nurses needed extra help. In the open environment, nurses were able to see and hear all of the patients, monitor multiple patients at the same time, and provide backup for other nurses.

From a nursing point of view, I think there were a lot of pros to having it open. One of them was just plain old backup. There was always someone who could help you with something. There was always some-one who could answer a question. There was always someone who could watch one kid while your other kid was needing something. There was always someone there. (PICU-3, 39 years)

Another advantage of the open environment was the opportunity for experienced nurses to men-tor their junior colleagues because they could watch them work and give them immediate feedback. Nurses with less experience were able to learn more than just the task-oriented skills by seeing nurses with more experience interact with patients’ families, demonstrating the finesse that is sometimes needed to communicate well with families.

There are huge benefits [to the open room]. When you’re trying to mentor


An open, multibed room offers huge benefits when you’re trying to mentor nurses.

people—it’s very difficult to teach young people and to give them enough experi-ences in the situation we have now. . . . You had a more experienced nurse work-ing right next to you and they might be doin’ somethin’ you never thought of doing, that doesn’t get learned. That’s the kind of thing you need to mentor. All these young people are learning how to take care of a patient and turn ‘em, move around and stuff like that, but they’re not learning [what] to think and say. (PICU-1, 34 years)

With individual rooms, both patients and nurses were isolated, so nurses monitoring multiple rooms could not see more than 1 patient at the same time. In addition, nurses could not hear alarms for the patient in the room next door and therefore needed to go back and forth between rooms. The nurses also commented that some parents who became accus-tomed to the individualized attention objected to having their child left alone. This increased the par-ents’ feelings that they needed to stay at the bedside around the clock because they knew the nurse would be busy at times in the other room with another patient.

Additionally, nurses commented that they sometimes worried about their own safety when they were alone with family members who were upset, hostile, or behaving inappropriately. Because no one else was watching, they felt they had no choice but to stay in the room and handle whatever came their way, even if they felt threatened.

It’s a little bit threatening to be in a room with a family, especially an angry family. Some of them are physically and verbally threatening, and especially the men, towards the young nurses. I tell them [the other nurses], “If you feel threatened in any way, walk out of the room.” There’s lots of issues to think about when you’re isolated in a room with people who are critically stressed, and then there’s risk. It’s risk to—they’re going to take it out on somebody, and you’re the person standing right there, and you are young, and you seem kind of vulnerable. (CICU-1, 30 years)

Discussion FCC is frequently cited as an approach to care

that supports the parental role and involvement of

parents in the care of their child10,11 with beneficial effects for families, children, and health care profes-sionals.12-15 The structural design and policies of PICUs have changed to implement FCC,9-15 and nurses play a critical role in the successful imple-mentation of these practice changes.8,16

The culture shift to FCC incorporates an orien-tation toward customer service that includes caring for the entire family unit, not just the child, on a 24-hour basis. This shift was generally perceived as positive, with benefits for both families and nurses. Yet the nurses’ experiences in this study high-light some of the challenges that also need to be addressed following policy and practice changes to support FCC. The nurses described their lack of skills and knowledge about how to balance caring for the critically ill child and the family at the same time in the highly complex environment of the ICU.

They felt challenged to prioritize efficient, in-the-moment, complex care for the child over the needs of family involved in the care of the child. These challenges worsened when the prognosis was more uncertain or the child’s health rapidly deteriorated. In these situations, the nurses had to find a balance so they could perform the highly technical and physi-cal tasks needed to care for a seriously ill child. Some of these challenges, which have been noted in other studies on the implementation of FCC, include cli-nicians’ attitudes and lack of family-centered skills and knowledge.12-14,25

In the current FCC environment, nurses are interacting with more parents who have their own ideas, which may have come from unreliable Inter-net sources. The length of time dealing with fami-lies further reduces opportunities for confidential discussion, mentoring, and nurse cohesion. These changes affect nurses’ time for communication, which, ironically, adversely affects patient care.

Implications for Practice From the perspective of these nurses, the policy

and practice changes associated with supporting FCC have many positive effects, including building positive relationships and engendering trust between parents and care providers.26 Highlighting the pos-itive aspects of FCC can be included in orientation of new nurses to ICU settings in children’s hospitals. However, the shift toward private rooms has several unintended consequences that should be addressed.


One way to manage these consequences is to create more opportunities to mentor junior nurses who can no longer learn through observation. Junior nurses coming into ICUs in children’s hospitals may need to have a longer mentorship time to become comfortable with the increased independence of caring for children and their families in private rooms. In addition, more support systems are needed for nurses who feel vulnerable when left alone with family members who confront them about their style of practice or who exhibit more threatening behavior. The constant presence of families also creates the need for private spaces where nurses can share information at shift changes and have time to foster staff cohesion.

Conclusion Because of nurses’ central role in the provision

of FCC, it is important for them to be involved in decisions on how best to implement FCC. Policy and environmental changes to implement FCC within ICUs in children’s hospitals should be made with nurses’ input.

ACKNOWLEDGMENTThe authors thank Natalie Oman, compensted with fund-ing from award R01 NR011179, for her work in collecting and managing data.

FINANCIAL DISCLOSURESResearch reported in this article was supported by the following National Institutes of Health agencies: National Institute of Nursing Research, award numbers R01 NR011179 and K24 NR015340, and the National Heart, Blood, and Lung Institute, award number T32 HL125195. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

REFERENCES1. Balluffi A, Kassam-Adams N, Kazak A, Tucker M, Dominguez

T, Helfaer M. Traumatic stress in parents of children admitted to the pediatric intensive care unit. Pediatr Crit Care Med. 2004;5(6):547-553.

2. Marsac ML, Hildenbrand AK, Kohser KL, Winston FK, Li Y, Kassam-Adams N. Preventing posttraumatic stress follow-ing pediatric injury: a randomized controlled trial of a web-based psycho-educational intervention for parents. J Pediatr Psychol. 2013;38(10):1101-1111.

3. Meert KL, Thurston CS,Thomas R. Parental coping and bereavement outcome after the death of a child in the pediatric intensive care unit. Pediatr Crit Care Med. 2001; 2(4):324-328.

4. Davidson JE, Jones C, Bienvenu OJ. Family response to critical illness: postintensive care syndrome-family. Crit Care Med. 2012;40(2):618-624.

5. Colville G, Darkins J, Hesketh J, Bennett V, Alcock J, Noyes J. The impact on parents of a child’s admission to intensive

care: integration of qualitative findings from a cross-sectional study. Intensive Crit Care Nurs. 2009;25(2):72-79.

6. Aamir M, Mittal K, Kaushik JS, Kashyap H, Kaur G. Predic-tors of stress among parents in pediatric intensive care unit: a prospective observational study. Indian J Pediatr. 2014; 81(11):1167-1170.

7. Bouet KM, Claudio N, Ramirez V, García-Fragoso L. Loss of parental role as a cause of stress in the neonatal intensive care unit. Bol Asoc Med P R. 2012;104(1):8-11.

8. Meert KL, Clark J, Eggly S. Family-centered care in the pediatric intensive care unit. Pediatr Clin North Am. 2013; 60(3):761-772.

9. Hutchfield K. Family-centred care: a concept analysis. J Adv Nurs. 1999;29(5):1178-1187.

10. Malusky SK. A concept analysis of family-centered care in the NICU. Neonatal Netw. 2005;24(6): 25-32.

11. Smith VC, Steelfisher GK, Salhi C, Shen LY. Coping with the neonatal intensive care unit experience: parents’ strategies and views of staff support. J Perinat Neonatal Nurs. 2012; 26(4): 343-352.

12. Coyne I. Families and health-care professionals’ perspectives and expectations of family-centred care: hidden expectations and unclear roles. Health Expect. 2015; 18(5): 796-808.

13. Mikkelsen G, Frederiksen K. Family-centred care of children in hospital: a concept analysis. J Adv Nurs. 2011; 67(5): 1152- 1162.

14. Ramezani T, Hadian Shirazi Z, Sabet Sarvestani R, Moattari M. Family-centered care in neonatal intensive care unit: a concept analysis. Int J Community Based Nurs Midwifery. 2014;2(4):268-278.

15. Wolfer J, Gaynard L, Goldberger J, Laidley LN, Thompson R. An experimental evaluation of a model child life program. Child Health Care. 1988;16(4):244-254.

16. Harrison TM. Family-centered pediatric nursing care: state of the science. J Pediatr Nurs. 2010;25(5):335-343.

17. DeLemos D, Chen M, Romer A, et al. Building trust through communication in the intensive care unit: HICCC. Pediatr Crit Care Med. 2010;11(3):378-384.

18. Sandelowski M. What’s in a name? Qualitative description revisited. Res Nurs Health. 2010;33(1):77-84.

19. Kim H, Sefcik JS, Bradway C. Characteristics of qualitative descriptive studies: a systematic review. Res Nurs Health. 2017;40(1):23-42.

20. Starks H, Doorenbos A, Lindhorst T, et al. The Family Com-munication Study: a randomized trial of prospective pallia-tive care consultation, study methodology and perceptions of participation burden. Contemp Clin Trials. 2016;49:15-20.

21. Maxwell JA. Qualitative Research Design: An Interactive Approach. 3rd ed. Thousand Oaks, CA: Sage; 2013.

22. Guest G, Bunce A, Johnson L. How many interviews are enough? An experiment with data saturation and variability. Field Methods. 2006;18(1):59-82.

23. Starks H, Trinidad SB. Choose your method: a comparison of phenomenology, discourse analysis, and grounded the-ory. Qual Health Res. 2007;17(10):1372-1380.

24. Hsieh HF, Shannon SE. Three approaches to qualitative content analysis. Qual Health Res. 2005;15:1277-1288.

25. Smith J, Swallow V, Coyne I. Involving parents in managing their child’s long-term condition—a concept synthesis of family-centered care and partnership-in-care. J Pediatr Nurs. 2015;30(1):143-159.

26. Doorenbos A, Lindhorst T, Starks H, Aisenberg E, Curtis JR, Hays R. Palliative care in the pediatric ICU: challenges and opportunities for family-centered practice. J Soc Work End Life Palliat Care. 2012;8(4):297-315.



Background Moral distress in registered nurses causes

decreased job satisfaction, turnover in staffing, burnout,

and heightened states of psychological distress. To date,

investigation of modifiable factors, such as perceptions

of the practice environment and patient safety, among a

diverse sample of critical care nurses has been limited.

Objective To explore the relationships among the sever-

ity of moral distress, the practice environment, and patient

safety in a national sample of critical care nurses.

Methods Critical care nurses experienced in working

with adults (> 1 year of intensive care unit experience)

and who were subscribers to the American Association

of Critical-Care Nurses’ e-mail listserv and social media

sites anonymously participated in this descriptive study.

Participants completed a demographic questionnaire, the

Moral Distress Scale–Revised, and the Practice Environ-

ment Scale of the Nursing Work Index. Descriptive sta-

tistics, bivariate correlation coefficients, and a hierarchical

regression analysis were used to describe the sample

characteristics and to assess relationships among the

study variables.

Results Of a national sample of 328 critical care nurses,

56% had less than 20 years of experience as a registered

nurse. Moral distress was modestly associated with

negative perceptions of the practice environment and

patient safety. Job satisfaction, practice environment,

and the participant’s age were statistically significant

predictors of moral distress in this sample.

Conclusions Modifications of organizational factors such

as the development of healthy work environments that

promote collegial relationships could reduce moral dis-

tress among critical care nurses. (American Journal of


PREDICTORS OF MORAL DISTRESS IN A US SAMPLE OF CRITICAL CARE NURSESCatherine A. Hiler, RN, DNP, CCRN, CMC, Ronald L. Hickman, Jr, RN, PhD, ACNP-BC, Andrew P. Reimer, RN, PhD, and Kimberly Wilson, RN, DNP


Healthy Work Environments


The intensive care unit is a practice environment where life-sustaining care intersects with end-of-life care. At the forefront of care delivery in the intensive care unit are registered nurses, who often experience moral distress associated with care that is inconsistent with the patient’s or family’s preferences.1,2 Conceptually, moral dis-tress consists of a moral dilemma that elicits psychological, cognitive, and behav-

ioral modifications within individuals and alters their performance over time.1,3

Commitment to fostering healthy

practice environments and optimizing the

quality of patient care decreases moral distress.


Registered nurses who work in critical care environments are exposed to numerous physical and psychological strains that can result in prema-ture departure from the nursing profession and affect their ability to deliver high-quality care.4,5 Moral distress among critical care nurses has been investigated and recognized as a national concern.6,7 Moreover, evidence shows that personnel who work for health care organizations that are committed to

fostering healthy practice environ-ments and optimizing the quality of patient care have lower states of moral distress.8,9

The American Association of Critical-Care Nurses (AACN) emphasizes the significance of a healthy work environment with respect to patient safety, quality of care, and nurse retention. In 2005, the AACN established stan-dards for creating and sustaining healthy work environments.10

These standards encompass skilled communication, true collaboration, effective decision-making, appro-priate staffing, meaningful recognition, and authen-tic leadership.10 Recent work highlights a need to continue investigating the factors that contribute to perceptions of unhealthy work environments among critical care nurses.11

Critical care nurses provide care for patients in tenuous states of illness. The environment is fast paced and often chaotic, resulting in distractions and competing priorities, which may affect the quality of care.3,12 In 2001, a survey conducted by the American Nurses Association revealed stress as the top health and safety concern among registered

nurses.13 Most of these nurses (75%) reported that unsafe working conditions hindered their ability to provide high-quality patient care.12 In 2004, the Institute of Medicine released a report titled “Keep-ing Patients Safe: Transforming the Work Environment of Nurses” that indicated that the work environment of nurses leads to patient harm and nurse burnout.14 The consequences of moral distress on job satisfac-tion, burnout, and retention have been studied exten-sively; however, few studies have addressed the effect on the delivery of high-quality patient care.15-17

Guided by a transactional model of stress and coping, this study explores the relationships among moral distress, perceived healthiness of the practice environment, and perception of quality among criti-cal care nurses.18 The theoretical premise of Lazarus and Folkman’s transactional stress and coping the-ory,18 along with theoretical underpinnings of moral distress published by Corley and colleagues,1,5 pro-vides a conceptual basis for the exploration into the associations between moral distress and work envi-ronment among critical care nurses. Consistent with work by Corley and colleagues,1,5 we posited that the antecedent phenomena, a nurse’s perception of medi-cal futility (psychological stressor) and the subsequent enactment of primary and secondary appraisal pro-cesses (coping), can result in moral distress.

Although important, stress and coping are not directly measured in this study. Yet, moral distress is conceptualized as a manifestation of these 2 anteced-ent conditions and is presumed to be associated with a nurse’s perceptions of the work environment, which also are posited to affect a nurse’s stress perception and adaptive coping processes. Given the nature of nurses’ work, stress can pose a challenge that can further compromise a nurse’s coping skills.18-20 Recur-rent exposure to the psychological stressor can over-whelm the critical care nurse’s regulatory mechanisms and eventually cause psychological harm, such as moral distress, and alter the nurse’s perceptions of the work environment. This study is conceptually consistent with the work of McAndrew and col-leagues21 and builds on the extant theoretical and empirical literature to provide insights on the

About the AuthorsCatherine A. Hiler is an assistant professor, Ronald L. Hickman, Jr, is an associate professor, Andrew P. Reimer and Kimberly Wilson are assistant professors, Frances Payne Bolton School of Nursing, Case Western Reserve University, Cleveland, Ohio.

Corresponding author: Catherine A. Hiler, 101 Elm Ave, Roanoke, VA 24014 (email: [email protected]).


Critical care nurses’ per-ceptions of moral distress, practice environment, and care quality were assessed.

relationships among moral distress and elements of the practice environment.

Methods A descriptive correlational design was used. Sur-

vey data included demographics, the Moral Distress Scale–Revised (MDS-R), and the Practice Environment Scale of the Nursing Work Index (PES-NWI) to cap-ture moral distress, practice environment, and care quality as perceived by critical care nurses. Approval was obtained from the institutional review board before study procedures were begun.

SampleThe study cohort comprised a sample of critical

care nurses in the United States who had at least 1 year of work experience in an adult intensive care unit. Study participants were recruited for 4 weeks through the AACN e-newsletter and social media sites. An a priori power analysis with equal to .05, and a small effect size (d = 0.20) indicated that a total sample of 191 participants was needed to achieve suf-ficient statistical power (1 – ) of 0.80 for bivariate correlation coefficients among the study variables.

InstrumentsThe MDS-R and the PES-NWI, 2 valid and reliable

survey instruments, were used for this study. Permis-sion was granted for instrument use before the study was begun. A demographic survey also was included.

Moral DistressThe MDS-R is a 21-item instrument that captures

the frequency and intensity of moral distress in nurses and other health care providers across contexts of care.13 Each item of the MDS-R is measured on a Likert scale, with separate scores given for frequency (ie, how often the situation occurs) and intensity (ie, how distressing the situation is when it occurs). For this study, participants were administered all 21 items, but they were asked only to endorse how fre-quently a situation occurred, which was thoughtfully chosen to minimize participant burden and the threat of recall bias in the appraisals of the intensity of moral distress in this sample. A total frequency score was derived by summing each item: Higher scores on the MDS-R indicated greater frequency of morally distressing situations encountered by the participant. The MDS-R has been tested for content validity and reliability, and all items were considered relevant.22 Initial psychometric evaluation of the MDS-R yielded a Cronbach coefficient of 0.89 among registered nurses.22 In the present study, the internal reliability

consistency coefficient (Cronbach ) was 0.87 for the MDS-R frequency subscale.

Healthy Work Environment and Patient SafetyThe PES-NWI is a 31-item, 5-factor instrument

that measures nurses’ perceptions of the practice environment.23 The PES-NWI encompasses a com-prehensive list of attributes affecting staff nurses’ job satisfaction and perceived productivity, defined as the perception of an environment conducive to quality nursing care. The original testing for content validity and reliability revealed Cronbach coeffi-cients from 0.71 to 0.84 among registered nurses, and the mean interrater reliability was from 0.86 to 0.97.23 This measure is constructed with 5 interre-lated subscales: (1) Nurse Participation in Hospital Affairs (NPH); (2) Nursing Foundations for Quality of Care (NFQ); (3) Nurse Manager Ability, Leader-ship, and Support of Nurses (NMA); (4) Staffing and Resource Adequacy (SRA); and (5) Collegial Nurse-Physician Relations (NPR). Items are scored using a 4-point Likert scale ranging from 1 (strongly disagree) to 4 (strongly agree) and comprising 5 sub-scales. Mean subscale scores were calculated and used in the statistical analyses. A higher score indicates strong agreement that the perceived practice environment contains a specific characteristic and a mean score less than 2.5 represents dis-agreement. In the present study, 4 of the subscales applied to a healthy work environment (ie, NPH, NMA, SRA, and NPR), and patient safety was captured using the NFQ subscale. Cronbach was greater than 0.70 for all subscales in this sample of critical care nurses.

Demographic DataThe demographic questionnaire used for this

study included the following information: sex, age, race, years in practice, years in critical care, years in current unit, type of unit, educational level, geographic location, level of job satisfaction, intent to leave cur-rent position, and whether the individual was employed in an intensive care unit recognized with an AACN Beacon Award or in a Magnet-designated hospital.

Statistical AnalysisData were analyzed using the SPSS version

20.0 (IBM). For surveys with fewer than 4 omitted responses, mean substitution was used to enhance


Nurses who perceived positive quality of care indicated lower levels

of moral distress.

the statistical power of the study results. Descriptive statistics were calculated for the demographic vari-ables, frequency scores for each item on the MDS-R, and mean scores for the 5 subscales of the PES-NWI. To measure moral distress frequency, total mean fre-

quency scores were calculated. Relationships between moral distress scores and scores on the 5 subscales of the PES-NWI were analyzed by using Pearson correlations. Bivariate correlational analysis and an independent t test were per-

formed to examine frequency scores for moral dis-tress and the relationship to perceptions of the practice environment and quality of care.

A 2-step hierarchical (sequential) multiple regres-sion analysis was conducted with the mean frequency scores for moral distress as the outcome variable. According to the bivariate correlational analysis, all study variables that had statistically significant bivari-ate correlation coefficients (r > 0.10; P < .05) were eligi-ble for entry in the hierarchical regression analysis. Before the analysis was conducted, we ensured that the assumptions for this statistical test were met.

In the first step, the variables that operational-ized the participants’ perceptions of their practice environment and quality (ie, NPH, NFQ, NMA, SRA, NPR, and Beacon unit designation) were entered to assess the variables’ contribution to explained vari-ance in moral distress frequency. Additional inde-pendent variables corresponding to characteristics of the participants (ie, age, job satisfaction) were entered in the second step. Statistical significance was specified at P less than .05 for all tests.

Results A convenience sample of 461 participants accessed

the electronic surveys. For the study variables of interest, 328 participants (71%) had complete data, and 133 participants (29%) who had missing data (≥ 4 omitted responses per study measure) were not included in the analytic sample. Therefore, data from a sample of 328 critical care nurses were used to address the purpose of this research.

This cohort consisted mostly of critical care nurses who had earned baccalaureate degrees (58%; n = 189) and had practiced as registered nurses for less than 20 years (56%; n = 185). Most participants (59%; n = 192) were satisfied with their current job, but 73% (n = 238) had contemplated leaving their current job within the past 6 months. Additional demographic charac-teristics are listed in Table 1.

We found high levels of moral distress frequency when nurses deemed that the care being provided was futile. The highest level of moral distress fre-quency occurred when the wishes of the patient’s family to continue life support were followed even though the nurse believed that doing so was not in the best interest of the patient.

Beacon designation was hypothesized to repre-sent a healthy practice environment. However, in this sample, few nurses (19%) reported working in an intensive care unit with this designation. Beacon designation was not a predictor of moral distress fre-quency in this sample. The data from our study indi-cated that nurses employed in a Beacon-recognized unit experienced less moral distress.

Associations Among Healthy Work Environment, Patient Safety, and Moral Distress

The analysis of the relationship between moral distress and perceived practice environment demon-strated that scores on all 5 subscales of the PES-NWI correlated significantly (P < .001) with MDS scores (Table 2). The significant correlations indicate that as the practice environment deteriorates, the level of moral distress is higher. As nurses participated in

Characteristic

Table 1Demographic characteristics of 328 nurses in the study

Sex (n = 327) Female Male

Race (n = 327) White Asian African American Hispanic Other

Geographic region (n = 326) Midwest Northeast Southeast Southwest West

Years in current unit 1-3 4-6 7-10 >10

Beacon designation (n = 326) Yes No

Intent to leave Yes No

283 (87) 44 (13)

282 (86)16 (5) 9 (3) 9 (3)11 (3)

82 (25) 73 (22) 86 (26) 34 (10) 51 (16)

128 (39) 47 (14) 47 (14)106 (32)

60 (18)266 (82)

238 (73) 90 (27)

N (%)a

a Percentages may not total 100 because of rounding.


more hospital affairs, their moral distress decreased. Nurses who perceived positive quality of care indi-cated decreased moral distress. As nurse leadership support increased, moral distress decreased. Nurses who reported adequate staffing and resources reported less moral distress. As nurses perceived collegial relationships between themselves and physicians, they reported less moral distress.

The severity of moral distress was significantly associated with a nurse’s intention to leave his or her current position. In this sample, participants who intended to leave their current position had signifi-cantly higher states of moral distress when compared with participants who intended to stay in their cur-rent position (t

326 = 3.53; P < .001).

Predictors of Moral DistressThe hierarchical multiple regression models

established that practice environment, quality of care, and nurse characteristics were influential pre-dictors of the frequency of moral distress. The cor-relation coefficients between the multiple regression variables are reported in Table 2 and the regression statistics are listed in Table 3.

In the initial regression model (F6,319 = 17.65;

P < .001), variables that captured the participants’ perceptions of their practice environment and qual-ity of care accounted for 25% (R2) of the variation in moral distress. The introduction of the variables that operationalized participant characteristics (ie, age, job satisfaction) explained an additional 5% of the variation in moral distress, and the change in R2 was significant (F2,317 = 17.07; P < .001). Thus, the final regression model with 8 predictor variables

accounted for 30% of the explained variance in moral distress. Of the 8 independent variables entered into the final regression model, higher states of job dis-satisfaction, practice environment (SRA and NPR subscale scores), and age were statistically significant predictors of moral distress frequency. The post hoc power (N = 325) was estimated at greater than 0.99

Variable

Table 2Correlation coefficient matrix (N = 328)

NPH

NFQ

NMA

SRA

NPR

Age

Job satisfaction

Beacon designationc

MDS

Abbreviations: MDS, mean moral distress frequency score; NFQ, Nursing Foundations for Quality of Care; NMA, Nurse Manager Ability; NPH, Nurse Participa-tion in Hospital Affairs; NPR, Nurse-Physician Relations; SRA, Staffing and Resource Adequacy.

a P < .001.b P = .01.c Coded as 0 = no, 1 = yes (n = 326).d P = .02.

0.03

-0.05

-0.39a

-0.07

-0.06

-0.09

0.37a

-0.04

0.40a

-0.10

-0.41a

-0.14b

0.32a

-0.10

-0.38a

0.55a

0.37a

-0.06

0.46a

-0.13d

-0.35a

0.56a

0.55a

0.47a

-0.07

0.33a

-0.12d

-0.40a

1.00

0.74a

0.67a

0.56a

0.42a

-0.07

0.40a

-0.13d

-0.39a

Beacon designationJob satisfactionAgeSRA NPRNMANPH NFQ

Predictor variable

Table 3Predictors of moral distress (N = 328)

Step 1 NPH NFQ NMA SRA NPR Beacon designation

Step 2c

NPH NFQ NMA SRA NPR Beacon designatione

Age Job satisfaction

11.77a

-0.058-0.105-0.062

-0.226a

-0.193b

-0.041

-0.038-0.120-0.001

-0.185d

-0.181b

-0.031 -0.139b

-0.215a

0.499

0.549

0.249

0.301

0.249

0.052

17.65a

17.07a

FR2 FR2R

Abbreviations: NFQ, Nursing Foundations for Quality of Care; NMA, Nurse Man-ager Ability; NPH, Nurse Participation in Hospital Affairs; NPR, Nurse-Physician Relations; SRA, Staffing and Resource Adequacy.

a P < .001.b P = .001.c Post hoc power greater than 0.99 and the observed effect size (f 2) was 0.43 for

the 8-predictor regression model.d P = .003.e Coded as 0 = no, 1 = yes.


with an observed effect size (f2) of 0.43 for the 8-predictor regression model (Table 3).

Discussion Moral distress is recognized as a national prob-

lem that significantly contributes to job dissatisfaction, turnover, burnout, and decreases in work productiv-ity among critical care nurses. The existing evidence on the effect of moral distress on critical care nurses working with adult patients has principally been focused on perceptions of ethical climate, nurse characteristics, and, to some extent, organizational characteristics.9,15,24,25 The growing body of literature on moral distress among critical care nurses has been constrained to single-site investigations, which has limited external generalizability. Although this study is consistent with work of McAndrew and colleagues,21 the present study encompasses a more comprehensive sample.

Moral distress is associated with perceptions of a healthy practice environment and care quality.11,26 High levels of moral distress frequency were reported when nurses viewed provided care as futile. Moral distress occurred most frequently when wishes of the patient’s family to continue life support were followed even though the nurse believed that doing so was

not in the best interest of the patient. This result was not unexpected given that nurses have repeatedly identified dis-tress in this context.9,26-28 Con-sistent with research conducted by McAndrew et al,21 the high-est frequency of moral distress

occurred from “continuing to participate in care for a hopelessly ill person” (mean score, 3.74 on a Likert scale of 1 to 5).

In our study, consistent with existing literature, moral distress had significant negative effects on job satisfaction and nurses’ intentions to leave their cur-rent position.6,19,25-27 Critical care nurses’ perception that they were working in a healthy environment was associated with higher job satisfaction and lower levels of moral distress. One plausible explanation for these findings is that healthy practice environ-ments foster collegial relationships of support and trust, and such environments provide employees with resources and adequate staffing that, in turn, maintain high states of job satisfaction and less inten-tion by nurses to leave their current positions. Sig-nificant relationships were demonstrated between moral distress frequency and the perception of the practice environment. The finding that that nurses

employed in a Beacon-recognized unit experienced less moral distress is consistent with the intent of the AACN Beacon Award for Excellence, in that criti-cal care units given the award have established high standards for care quality by improving outcomes on the basis of evidence. Beacon-designated units demonstrate collegial practice environments with higher morale and increase staff retention, exempli-fying the standards of a healthy work environment.29

In this study, perceptions of job satisfaction and practice environment, as well as the age of the critical care nurse, were predictors of the severity of moral distress. Prior research has described similar relationships between job satisfaction and age. Addi-tionally, a significant proportion of the literature has been focused on perceptions of the ethical cli-mate and moral distress.6,9,11 Conceptually, there is some interrelatedness between ethical climate and the measures of healthy practice environment and care quality that further supports the results of the hierarchical regression analysis. Specifically, NMA and SRA subscale scores were statistically significant predictors of moral distress. Of particular note, per-ceptions of managerial support, staffing, and ade-quate resources have emerged as influential factors associated with moral distress across studies.11,26,30 The findings of this study underscore select charac-teristics of critical care nurses that enhance their psy-chological susceptibility to moral distress and offer evidence for preventive strategies to lower the preva-lence of moral distress.

LimitationsThis study has several limitations that affect

the external generalizability of the results. Although appropriate for the purpose of the study, the cross-sectional descriptive design does not adequately permit the identification of causal inferences among the study variables. The second limitation is the poten-tial of a biased sample, given that participants who completed the study surveys may have been moti-vated by personal experiences with moral distress. Last, moral distress frequency was assessed rather than moral distress intensity. Despite these limita-tions, this study identifies influential contributors of moral distress and provides insight for future multilevel interventions to attenuate moral distress among critical care nurses.

Conclusions This study offers insights on modifiable organi-

zational factors that contribute to moral distress. As underscored by AACN, moral distress is a pressing

Healthy work environ-ments are associated

with lower levels of moral distress.


issue affecting critical care nurses. The results of this study provide evidence for multilevel interventions for moral distress that can enhance the quality of the practice environment. On the basis of the study results, it is recommended that adoption of the AACN’s standards for creating and sustaining healthy work environments is essential.10 Additionally, the results of this study support interventions to promote collaboration between critical care nurses and physi-cians, as well as educational and emotional support for critical care nurses who provide end-of-life care.2,25,29 Research is needed to better understand how organizational factors, practice environment, and nurse characteristics converge to potentiate moral distress frequency and intensity.

FINANCIAL DISCLOSURESA.P.R. was supported by the National Center for Research Resources and the National Center for Advancing Trans-lational Sciences, National Institutes of Health, through Grant KL2TR000440; and through the National Institutes of Health Loan Repayment Program, National Center for Advancing Translational Sciences. The content is solely the responsibility of the authors and does not necessar-ily represent the official views of the National Institutes of Health.

SEE ALSO For more about moral distress, visit the Critical Care Nurse website, www.ccnonline.org, and read the article by Olmstead and Dahnke, “The Need for an Effective Process to Resolve Conflicts Over Medical Futility: A Case Study and Analysis” (December 2016).

REFERENCES1. Corley MC. Nurse moral distress: a proposed theory and

research agenda. Nurs Ethics. 2002;9(6):636-650.2. Oberle K, Hughes D. Doctors’ and nurses’ perceptions of

ethical problems in end-of-life decisions. J Adv Nurs. 2001; 33(6):707-715.

3. Jameton A. Nursing Practice : The Ethical Issues. Englewood Cliffs, NJ: Prentice-Hall; 1984.

4. Aiken LH, Clarke SP, Sochalski JA, Silber JH. Hospital nurses staffing and patient mortality, nurse burnout, and job dis-satisfaction. JAMA. 2002;288(16):1987-1993.

5. Corley MC, Elswick RK, Gorman M, Clor T. Development and evaluation of a moral distress scale. J Adv Nurs. 2001; 33(2):250-256.

6. Erlen J, Sereika S. Critical care nurses, ethical decision-making and stress. J Adv Nurs. 1997;26(5):953-961.

7. Kirwan M, Matthews A, Scott PA. The impact of the work environment of nurses on patient safety outcomes: A multi-level modelling approach. Int J Nurs Stud. 2013; 50(2):253-263.

8. Kane RL, Shamliyan TA, Mueller C, Duval S, Wilt TJ. The association of registered nurse staffing levels and patient outcomes: systematic review and meta-analysis. Med Care. 2007;45(12):1195-1204.

9. Epstein EG, Delgado S. Understanding and addressing moral distress. Online J Issues Nurs. 2010;15(3):manuscript 1.

10. American Association of Critical-Care Nurses. AACN stan-dards for establishing and sustaining healthy work envi-ronments: a journey to excellence. Am J Crit Care. 2005; 14(3):187-197.

11. Ulrich CM, Hamric AB, Grady C. Moral distress: a growing problem in the health professions? Hastings Cent Rep. 2010; 40(1):20-22.

12. Lin L, Liang BA. Addressing the nursing work environment to promote patient safety. Nurs Forum. 2007;42(1):20-30.

13. American Nurses Association. NursingWorld.org Health and Safety Survey, September 2001. www.nursingworld.org /hssurvey-2011. Accessed October 12, 2017.

14. Institute of Medicine. Keeping Patients Safe: Transforming the Work Environment of Nurses. Washington, DC: The National Academies Press; 2004.

15. de Veer AJE, Francke AL, Struijs A, Willems DL. Determinants of moral distress in daily nursing practice: A cross sectional correlational questionnaire survey. Int J Nurs Stud. 2013; 50(1):100-108.

16. Erlen JA. Moral distress: a pervasive problem. Orthop Nurs. 2001;20(2):76-80.

17. Pendry PS. Moral distress: recognizing it to retain nurses. Nurs Econ. 2007;25(4):217-221.

18. Lazarus RS, Folkman S. Stress, Appraisal, and Coping. New York, NY: Springer; 1984.

19. Wilson MA, Goettemoeller DM, Bevan NA, McCord JM. Moral distress: levels, coping and preferred interventions in critical care and transitional care nurses. J Clin Nurs. 2013; 22(9-10):1455-1466.

20. Wakim N. Occupational stressors, stress perception levels, and coping styles of medical surgical RNs: a generational perspective. J Nurs Adm. 2014;44(12):632-639.

21. McAndrew NS, Leske JS, Garcia A. Influence of moral dis-tress on the professional practice environment during prog-nostic conflict in critical care. J Trauma Nurs. 2011;18(4): 221-230.

22. Hamric AB, Borchers T, Epstein E. Development and testing of an instrument to measure moral distress in healthcare professionals. AJOB Prim Res. 2012;3(2):1-9.

23. Lake ET. Development of the practice environment scale of the Nursing Work Index. Res Nurs Health. 2002;25(3):176-188.

24. Bruce CR, Miller SM, Zimmerman JL. A qualitative study exploring moral distress in the ICU team: the importance of unit functionality and intrateam dynamics. Crit Care Med. 2015;43(4):823-831.

25. Elpern EH, Covert B, Kleinpell R. Moral distress of staff nurses in a medical intensive care unit. Am J Crit Care. 2005;14(6):523-530.

26. Browning AM. CNE article: moral distress and psychological empowerment in critical care nurses caring for adults at end of life. Am J Crit Care. 2013;22(2):143-151.

27. Calvin AO, Kite-Powell DM, Hickey JV. The neuroscience ICU nurse’s perceptions about end-of-life care. J Neurosci Nurs. 2007;39(3):143-150.

28. Neville TH, Wiley JF, Yamamoto MC, et al. Concordance of nurses and physicians on whether critical care patients are receiving futile treatment. Am J Crit Care. 2015;24(5):403-410.

29. American Association of Critical-Care Nurses. Beacon Award for Excellence. American Association of Critical-Care Nurses website. http://www.aacn.org/wd/beaconapps/content /beacon-program-overview.content?menu=beaconApps. Accessed September 23, 2017.

30. Burston AS, Tuckett AG. Moral distress in nursing: contrib-uting factors, outcomes and interventions. Nurs Ethics. 2013;20(3):312-324.



1.0 Hour Category BC E Notice to CE enrollees:


1. Identify infl uential predictors of the frequency of moral distress.2. Demonstrate the relationship between moral distress frequency and the perception of the practice

environment.3. Correlate data to explain the relationship between Beacon recognition and the experience of moral distress.




(#LSBN12).



Background The incidence and long-term outcomes of

acute kidney injury in patients with severe acute respira-

tory distress syndrome (ARDS) due to influenza A(H1N1)

pdm09 virus (pH1N1) have not been examined.

Objective To assess long-term renal recovery in patients

with acute kidney injury and severe ARDS due to pH1N1.

Methods A retrospective observational cohort study of

adults with severe pH1N1-associated ARDS admitted to a

tertiary referral center. Baseline characteristics, acute kid-

ney injury stage, continuous renal replacement therapy

(CRRT), intermittent hemodialysis, extracorporeal mem-

brane oxygenation, survival, and renal recovery (defined

as dialysis independence) were evaluated.

Results Fifty-seven patients, most with stage 3 acute kid-

ney injury, were included. The 53% mortality rate among

the 38 patients requiring CRRT was significantly higher

than the 0% mortality rate among the 19 patients not

requiring CRRT or intermittent hemodialysis. Increased

duration of CRRT was not significantly associated with

decreased survival. Fifteen CRRT patients required tran-

sition to intermittent hemodialysis. Of the CRRT patients

who survived, 94% experienced renal recovery. Extracorpo-

real membrane oxygenation was instituted in 17 patients;

15 of these patients required CRRT.

Conclusions Acute kidney injury is common in patients

with severe ARDS caused by pH1N1 infection. CRRT is a

significant risk factor for increased mortality, but most

patients who survived experienced full renal recovery.

(American Journal of Critical Care. 2018; 27:67-73)

OUTCOMES OF ACUTE KIDNEY INJURY IN PATIENTS

WITH SEVERE ARDS DUE

TO INFLUENZA A(H1N1)PDM09 VIRUSBy Christopher J. Tignanelli, MD, Arek J. Wiktor, MD, Cory J. Vatsaas, MD, Gaurav Sachdev, MD, Michael Heung, MD, MS, Pauline K. Park, MD, Krishnan Raghavendran, MD, and Lena M. Napolitano, MD

Renal Issues

The worldwide pandemic caused by influenza A(H1N1)pdm09 virus (pH1N1) resulted in more than 18 449 deaths in 214 countries.1,2 From April 2009 to April 2010, an estimated 60.8 million cases and more than 250 000 hospitalizations due to pH1N1 occurred in the United States.3 Approximately 10% to 30% of hospitalized patients required intensive care unit (ICU) admission, with most of

these patients requiring mechanical ventilation.4 Unlike previous influenza viruses, the pH1N1 virus caused severe pneumonia that often progressed to acute respiratory distress syndrome (ARDS) and acute kidney injury (AKI), resulting in high mortality and morbidity. The virus also had a propensity to affect adults between the ages of 30 and 50 years.3

The incidence of AKI and use of RRT

was high in patients with critical illness

due to pH1N1.


About the AuthorsChristopher J. Tignanelli is a surgical critical care fellow, University of Michigan Health System, Ann Arbor, Michi-gan. Arek J. Wiktor is an assistant professor of surgery, University of Colorado, Denver, Colorado. Cory J. Vatsaas is an assistant professor of surgery, Duke University Health System, Durham, North Carolina. Gaurav Sachdev is an assistant professor of surgery, Carolinas Medical Center, Charlotte, North Carolina. Michael Heung is an associate professor of medicine, Division of Nephrology, University of Michigan Health System. Pauline K. Park and Krishnan Raghavendran are professors of surgery, Univer-sity of Michigan Health System. Lena M. Napolitano is professor of surgery, division chief of acute care surgery, director of trauma and surgical critical care, and associate chair of the Department of Surgery, University of Michigan Health System.

Corresponding author: Lena M. Napolitano, MD, Univer-sity of Michigan Health System, 1500 E. Medical Drive, SPC 5033, Ann Arbor, MI 48109-5033 (email: [email protected]).

Numerous reports have documented significant functional limitations and morbidity in survivors. For example, by 1 year after hospital discharge, only 60% to 80% of patients have returned to work. Addition-ally, nearly half of the survivors experience significant exertional dyspnea and decreased diffusing capacity

of the lung for carbon monoxide at 1 year after discharge.5 Throm-boembolic complications include a significant rate (up to 86%) of de novo pulmonary embolism, arterial thrombosis, and hypercoagulabil-ity for up to 1 year after recovery.6,7 Patients also experience significant psychological disturbances and are at risk for posttraumatic stress

disorder.5 Unfortunately, few studies have examined long-term renal recovery in this population.

Acute renal failure (ARF) requiring hemodialysis develops in a significant number of patients acutely ill with pH1N1 infection. Compared with non-pandemic influenza infections, pH1N1 infection

is associated with a higher rate of AKI and use of renal replacement therapy (RRT).8 However, few studies have examined the long-term outcomes of patients with severe ARDS due to pH1N1 infection who required RRT during their critical illness.

Given this knowledge gap, our primary objec-tive was to assess the epidemiology and long-term outcomes of AKI and the need for RRT in patients with severe ARDS due to pH1N1 infection at an ARDS referral center in the United States.

Methods Study Design

We conducted a retrospective observational cohort study analyzing all adult patients with severe ARDS due to pH1N1 infection admitted to the sur-gical ICU for extracorporeal membrane oxygenation (ECMO) evaluation from June 2009 through 2011. Eligible participants were patients with severe ARDS and positive pH1N1 virus test results (polymerase chain reaction assay or viral respiratory culture). We excluded patients who had a known history of chronic kidney disease or required chronic dialysis. The patient population was drawn from a large ter-tiary referral area because our institution serves as an ARDS referral center with the ability to provide ECMO support for severe ARDS. All patients were transferred initially as possible ECMO candidates, but not all ultimately required ECMO. We obtained follow-up data from clinical visits after hospital dis-charge and from examination of medical records.

Data CollectionWe performed a retrospective chart review and

obtained all data from computerized medical records. Demographic data included age, sex, and body mass index. Acute Physiology and Chronic Health Evalua-tion (APACHE) III scores were calculated on the first day of admission to the ICU.9 Survival was defined as survival to hospital discharge or transfer back to


the referring hospital. Total ICU days, total ventilator days, and total days requiring vasopressor support (any vasopressor at any dose) were also recorded. The pH1N1 subtype was confirmed by polymerase chain reaction assay or viral culture. The Berlin definition was used to define ARDS; severe ARDS was defined as a ratio of PaO2 to fraction of inspired oxygen of 100 mm Hg or less.10

Baseline creatinine was defined as the serum creatinine level on the day of admission to our ICU. Peak creatinine was the highest value attained during the entire ICU stay. AKI was defined per the Kidney Disease: Improving Global Outcomes (KDIGO) AKI criteria and was determined by the highest creatinine level documented during the ICU stay.11 Patients who required continuous RRT (CRRT) or intermittent hemodialysis (IHD) were classified as having stage 3 AKI. Timing of CRRT initiation and dosage were at the discretion of the surgical ICU attending phy-sician. Once the patient was deemed a candidate to transition from CRRT, the consulting nephrology team determined IHD scheduling and type. The total number of days requiring CRRT and IHD were recorded, with extended CRRT defined as requiring more than 14 days of therapy. Renal recovery was defined as dialysis independence. Surgical intensiv-ists in our clinic followed up patients after their

discharge from the ICU. Long-term outcomes were also obtained from medical information provided by local physicians after patients had transitioned to health care closer to home.

Statistical Analysis Continuous variables were expressed as means

(with SD) or medians (with interquartile range). Cat-egorical variables were presented as frequency (per-centage). For statistical analyses, the 2-tailed t test was used for continuous variables and the 2 test was used for categorical variables, except where specified. A P of .05 or less was considered significant.

Results From June 2009 through December 2011, 57

adult patients underwent ECMO evaluation because of severe ARDS due to pH1N1 infection. Thirty-six (64%) of the patients were male; other characteristics of the patients are presented in the Table. According to the KDIGO criteria, all 57 patients had AKI classi-fied as either stage 2 or stage 3 (stage 2 in 3 patients [5%], stage 3 in 54 patients [95%]). The mean (SD) ICU length of stay was 26.9 (28.5) days.

All patients who underwent RRT were managed with CRRT. The timing of CRRT initiation varied, and the primary indication was ARF with associated

Characteristic

TablePatients’ characteristics

Age, mean (SD), y

Body mass index,a mean (SD)

APACHE III score Mean Median Interquartile range

Baseline creatinine level, mg/dL Mean Median Interquartile range

Peak level of creatinine, mg/dL Mean Median Interquartile range

Vasopressor use, d Mean Median Interquartile range

Survival, %

Abbreviations: APACHE, Acute Physiology and Chronic Health Evaluation; CRRT, continuous renal replacement therapy.SI conversion factor: To convert creatinine to μmol/L, multiply by 88.4.

a Calculated as weight in kilograms divided by height in meters squared.

.86

.99

.009

.02

< .001

.003

< .001

41.4 (13.1)

38 (10.6)

70.668

53.7-88.2

1.91.2

1-2.6

5.15.2

3.4-6.3

107.5

3.2-13.7

47.4 (n = 18)

42 (11.4)

38 (8.5)

50.847.5

45.2-55.7

1.21.1

0.7-1.4

2.32.2

1.7-2.6

1.91

0-3

100 (n = 19)

41.6 (12.43)

38 (9.8)

6459

47-78

1.71.2

0.9-2.4

4.13.5

2.2-5.6

7.35

1-10

65 (n = 37)

P(CRRT vs no CRRT)CRRT (n = 38)No CRRT (n = 19)Total (N = 57)


hypotension and inability to tolerate IHD. Thirty-eight patients (67%) required CRRT. The total number of CRRT days was 714 (mean, 18.8 days per patient). Compared with patients not requiring CRRT, those requiring CRRT had signifi cantly higher average APACHE III scores (mean, 70.6 vs 50.8; P = .009 and ICU length of stay (mean [SD], 32.3 [33] days vs 16.1 [10.1] days; P = .008). Seventeen patients (30% of the total cohort) required ECMO, and 15 (88%) of the ECMO patients required CRRT.

Vasopressors were required in 44 patients (77%) during their ICU admission. Compared with patients who did not require CRRT, more patients requiring CRRT received vasopressors (34 patients [89%] vs 10 patients [53%], P = .008) and received vasopressors for a longer time (mean [SD], 10 [10.2] days vs 1.9 [2.4] days; P = .003). Only 24 (65%) of the 37 patients who survived, as compared with all 20 of those who did not, required vasopressors (P = .008).

Survival to hospital discharge or transfer was 65% (n = 37). Of the 38 patients requiring CRRT, 18 sur-vived (47%); conversely, all 19 patients not requiring CRRT survived (P < .001) (Figure 1). Increased time on CRRT was not associated with decreased hospital survival (50% of patients on CRRT for ≤ 14 days

survived and 44% of patients on CRRT for > 14 days survived; P = .92) (Figure 2). Of the 18 CRRT patients who survived, 15 (83%) required transition to IHD because of persistent renal failure. The total number of IHD days in these 15 patients was 247 (mean, 16.5 days per patient). Sixteen of the 17 patients (94%) who underwent CRRT and survived had renal recov-ery to dialysis independence. The patient with per-sistent renal failure requiring dialysis notably had underlying chronic kidney disease (baseline serum creatinine level, 2.4 mg/dL). In patients who died, renal histopathological fi ndings on autopsy were typical of acute tubular necrosis.

Discussion We reported the fi rst cases of severe ARDS due

to pH1N1 virus infection in 2009.7,12 In the current study, we report that AKI was present in all patients with severe ARDS due to pH1N1 infection, with 95% of patients reaching KDIGO stage 3. The need for CRRT was associated with increased mortality, but duration of CRRT did not correlate with mortality. Vasopressor use and duration were associated with CRRT utilization and with mortality.

An important fi nding is that 94% of the patients who required CRRT and survived experienced long-term renal recovery (defi ned as dialysis independence). This fi nding is interesting in light of the well-known high mortality rate and low rate of dialysis indepen-dence associated with AKI in critically ill patients in general.13,14 Data from large observational studies of critically ill patients requiring RRT indicate a rate of dialysis independence at discharge of 13% to 29%.15 However, it is important to point out that the rate of dialysis independence does decrease over time after discharge. In one study of critically ill patients requir-ing RRT, 68% of patients were dialysis independent at discharge, and 78% were dialysis independent by 1 year.16 However, one must be cautious when com-paring our study outcomes with outcomes of studies of AKI in the critically ill because of the potential of the pH1N1 virus to infect much younger patients who have no signifi cant comorbid conditions. The popu-lation of patients with pH1N1 infection therefore has the potential for signifi cant recovery after illness. Younger age has been associated with a signifi cant reduction in mortality in critically ill patients with ARF and a signifi cant reduction in dialysis depen-dence after discharge.15,17

We observed an AKI incidence of 100% in patients with severe ARDS due to pH1N1 infection. This incidence is signifi cantly higher than the inci-dence presented in other studies (17%-61%), most

Figure 1 Survival of patients with acute kidney injury and infl u-enza A(H1N1)pdm09 infection.

57 Adult patients with H1N1 infection, acute kidney injury developed in 100%

38 (67%) Required continuous renal

replacement therapy

18 (47%) Survived to discharge

20 (53%) Died in the hospital

19 (100%) Survived to discharge

19 (33%) Did not require continuous renal

replacement therapy

Figure 2 Patient survival versus duration of continuous renal replacement therapy (CRRT).

Surv

ival

, %

No. of days requiring CRRT> 147-14< 7

60

50

40

30

20

10

0


likely because of our role as a tertiary referral center for ARDS and ECMO.18-23 In a similar fashion, our rate of CRRT utilization was significantly higher (67%) than has been reported in other studies (7% to 24%), most likely because of the severity of dis-ease in our population of patients.18-21,23,24

A significant finding of this study was that the need for RRT was associated with increased mortality in patients with severe ARDS due to pH1N1 infection, in contrast to findings from early reports of patients with this disease process.25,26 Results from small obser-vational studies conducted since these initial reports have been conflicting regarding the relationship of RRT and mortality. We identified 4 studies that showed a direct link between the need for RRT and increased mortality in patients with severe ARDS due to pH1N1 infection. A Brazilian study of patients with pH1N1 infection identified 25 patients with AKI. ARF neces-sitating CRRT developed in only 9 (36%) of these patients. Indications for CRRT in the Brazilian study were acidosis, uremia, and hypervolemia. Similar to our findings, the need for CRRT was significantly associated with increased mortality (P = .03).27 A study conducted at the Cleveland Clinic included 12 patients with pH1N1 infection who required RRT. The need for RRT was associated with a significant increase in mortality (6 of 12 patients [50%] vs 4 of 24 patients [17%], P = .05).22 A Korean study included 33 patients who required CRRT. In this study, the need for CRRT was associated with a statistically significant increase in hospital mortality (28.2% mortality vs 7.9%, P < .001).20 A study conducted in Argentina included 22 patients with pH1N1 pneumonia. ARF developed in 10 of these patients, and 4 of the 10 required CRRT.28 Three of the 4 patients who required CRRT ultimately died.

In contrast to these findings, several observa-tional studies failed to show an association between mortality and the need for RRT in patients with severe ARDS due to pH1N1 virus. An initial Canadian study24 included 50 patients with pH1N1 infection. Similar to patients in our study and previous reports, these patients were critically ill, with a mean APACHE II score of 19, and all required mechanical ventilation. Using the risk, injury, failure, loss of kidney function, and end-stage kidney disease criteria, the investiga-tors observed an AKI rate of 66.7%, with 22% of these patients requiring RRT. The researchers noted a 22% mortality rate in patients who required RRT. The need for RRT was not associated with a statisti-cally significant increase in mortality (odds ratio, 1.21; 95% CI, 0.23-5.00). However, their sample included only 11 patients who needed RRT, a significant

limitation. The development of ARF was associated with a significantly increased risk of death (odds ratio,11.29; 95% CI, 1.29-98.8), as we and others have shown.

In the larger Canadian Critical Care Trials Group H1N1 Collaborative study,18 which included 342 patients with AKI, ARF (as defined by the risk, injury, failure, loss of kidney function, and end-stage kidney disease criteria) developed in 204 patients. Eighty-two of these patients required RRT.18 Unfortunately, the investigators made no direct com-parison between mortality in patients requiring or not requiring RRT. However, ARF was associated with a 31.9% (65 of 204 patients) in-hospital mortality rate, and the need for RRT was associated with a similar mortality rate of 31% (26 of 81 patients). A Spanish study also showed no dif-ference in mortality rate according to need for RRT. In this study of 43 patients with AKI, 19 required RRT. The requirement of RRT was not associated with increased mortality (74% for CRRT vs 67% for no CRRT, P = .40).23

A significant limitation in all of these studies is the small number of patients who required RRT, resulting in insufficient power to detect significant changes. Another limitation in larger studies evalu-ating the effects of AKI in patients with pH1N1 influenza is the failure to directly compare outcomes between patients requiring RRT and those with AKI not requiring RRT. A third limitation of these studies is the lack of a standardized set of criteria for the ini-tiation of RRT. Consequently, we cannot be sure that all patients who went on to receive RRT were simi-larly ill and necessitated initiation.

To the best of our knowledge, ours is the largest study to date that has specifically analyzed the long-term outcomes of patients with severe ARDS due to pH1N1 infection who required RRT. Our results show a signifi-cantly higher mortality in patients needing RRT (53% in patients with RRT vs 0% in patients with no RRT; P < .001).

Few studies have examined long-term outcomes including renal recovery (as defined by dialysis independence) in this population. Although studies have shown dialysis independence rates of approximately 68% at dis-charge for critically ill patients with ARF, the popu-lation of ARF patients is significantly different from

Outcomes of AKI in patients with severe ARDS due to pH1N1 were assessed.

CRRT is a signifi-cant risk factor for increased mortality in patients with AKI due to pH1N1.


the population of pH1N1 influenza patients, who are younger and have fewer comorbid conditions.16,25,26 Therefore, we hypothesized that the rate of dialysis independence would be higher in pH1N1 influenza patients than in ARF patients. This hypothesis was confirmed by our finding of 94% dialysis indepen-

dence in survivors who required CRRT during their acute illness. Similar to this finding, all 9 patients requiring CRRT in the Brazilian study27 were dialysis independent at discharge. This finding was echoed in the Cana-dian study,24 which revealed a dialysis-free rate of 89% at dis-charge for survivors. This dialysis

independence rate is an important finding for prog-nostic purposes for patients with pH1N1 infection and AKI. Unfortunately, we know of no studies that have examined whether these patients would be at a higher risk for chronic kidney disease in the future.

The etiology of AKI in pH1N1-ARDS patients is most likely multifactorial. A recent autopsy study reported renal pathological findings of mild to mod-erate acute tubular necrosis, the presence of myoglo-bin, and thrombotic microangiopathy.29 In another study of patients with pH1N1 infection who had ARF and underwent kidney biopsy, various degrees of vac-uolar degenerative tubular changes were present, but no signs of acute tubular necrosis were apparent.30 A recent small study examined the profile of angio-genic and inflammatory factors in pH1N1-ARDS patients with or without AKI. The investigators determined that patients with pH1N1 infection and ARDS/AKI have an overproduction of monocyte chemoattractant protein-1, vascular endothelial growth factor, and interferon-inducible protein 10, possibly contributing to AKI and associated with a higher risk of death.31

Our data are subject to several limitations. This analysis is an observational study and therefore patients’ follow-up was not standardized, which could have resulted in underreporting of return to dialysis independence or unrecognized mortality or complications after the last follow-up. Second, ours is a tertiary referral center, so our population is not likely to reflect the general population of patients with pH1N1 infection but rather is a subset of the sickest patients.

To our knowledge, this is the largest study report-ing the epidemiology and outcomes of AKI and ARF requiring RRT in critically ill patients with severe ARDS due to pH1N1 virus infection in the United States. In conclusion, this study documents that AKI was

present in all patients with severe ARDS due to pH1N1 viral infection and was associated with a high mortality rate. The need for CRRT was an inde-pendent significant risk factor for increased mortal-ity. However, 94% of the patients requiring CRRT who survived had full renal recovery. These findings are important for clinicians and intensivists provid-ing care for pH1N1 influenza patients with ARDS and AKI and for optimal preparation of resources required for the care of these critically ill patients with multiple organ failure.


SEE ALSO For more about acute kidney injury, visit the Critical Care Nurse website, www.ccnonline.org, and read the article by Lambert et al, “Reducing Acute Kidney Injury Due to Contrast Material: How Nurses Can Improve Patient Safety” (February 2017).

REFERENCES1. Cheng VC, To KK, Tse H, Hung IF, Yuen KY. Two years after

pandemic influenza A/2009/H1N1: what have we learned? Clin Microbiol Rev. 2012;25(2):223-263.

2. Dawood FS, Iuliano AD, Reed C, et al. Estimated global mor-tality associated with the first 12 months of 2009 pandemic influenza A H1N1 virus circulation: a modelling study. Lancet Infect Dis. 2012;12(9):687-695.

3. LaRussa P. Pandemic novel 2009 H1N1 influenza: what have we learned? Semin Respir Crit Care Med. 2011;32(4):393-399.

4. Napolitano LM, Angus DC, Uyeki TM. Critically ill patients with influenza A(H1N1)pdm09 virus infection in 2014. JAMA. 2014;311(13):1289-1290.

5. Luyt CE, Combes A, Becquemin MH, et al; REVA Study Group. Long-term outcomes of pandemic 2009 influenza A(H1N1)-associated severe ARDS. Chest. 2012;142(3):583-592.

6. Dimakakos E, Grapsa D, Vathiotis I, et al. H1N1-induced venous thromboembolic events? Results of a single-institution case series. Open Forum Infect Dis. 2016; 3(4): ofw214.

7. Centers for Disease Control and Prevention (CDC). Intensive-care patients with severe novel influenza A (H1N1) virus infection: Michigan, June 2009. MMWR Morb Mortal Wkly Rep. 2009;58(27):749-752.

8. Writing Committee of the WHO Consultation on Clinical Aspects of Pandemic (H1N1) 2009 Influenza, Bautista E, Chotpitayasunondh T, Gao Z, et al. Clinical aspects of pan-demic 2009 influenza A (H1N1) virus infection. N Engl J Med. 2010;362(18):1708-1719.

9. Knaus WA, Wagner DP, Draper EA, et al. The APACHE III prog-nostic system. Risk prediction of hospital mortality for criti-cally ill hospitalized adults. Chest. 1991;100(6):1619-1636.

10. ARDS Definition Task Force, Ranieri VM, Rubenfeld GD, Thompson BT, et al. Acute respiratory distress syndrome: the Berlin definition. JAMA. 2012;307(23):2526-2533.

11. KDIGO. KDIGO Clinical Practice Guideline for Acute Kidney Injury. http://kdigo.org/guidelines/acute-kidney-injury/. Pub-lished 2012. Accessed July 28, 2017.

12. Harms PW, Schmidt LA, Smith LB, et al. Autopsy findings in eight patients with fatal H1N1 influenza. Am J Clin Pathol. 2010;134(1):27-35.

13. Bagshaw SM, Mortis G, Godinez-Luna T, Doig CJ, Laupland KB. Renal recovery after severe acute renal failure. Int J Artif Organs. 2006;29(11):1023-1030.

14. Bagshaw SM, Uchino S, Bellomo R, et al; Beginning and Ending Supportive Therapy for the Kidney (BEST Kidney) Investigators. Septic acute kidney injury in critically ill patients: clinical characteristics and outcomes. Clin J Am Soc Nephrol. 2007;2(3):431-439.

15. Rewa O, Bagshaw SM. Acute kidney injury—epidemiology, outcomes and economics. Nat Rev Nephrol. 2014;10(4): 193-207.

Of the CRRT patients who survived, 94% experienced renal

recovery and dialysis independence.


16. Bagshaw SM, Laupland KB, Doig CJ, et al. Prognosis for long-term survival and renal recovery in critically ill patients with severe acute renal failure: a population-based study. Crit Care. 2005;9(6):R700-709.

17. Uchino S, Kellum JA, Bellomo R, et al; Beginning and Ending Supportive Therapy for the Kidney (BEST Kidney) Investiga-tors. Acute renal failure in critically ill patients: a multina-tional, multicenter study. JAMA. 2005;294(7):813-818.

18. Bagshaw SM, Sood MM, Long J, Fowler RA, Adhikari NK; Canadian Critical Care Trials Group H1N1 Collaborative. Acute kidney injury among critically ill patients with pan-demic H1N1 influenza A in Canada: cohort study. BMC Nephrol. 2013;14:123.

19. Martin-Loeches I, Papiol E, Rodríguez A, et al; H1N1 SEMI-CYUC Working Group. Acute kidney injury in critical ill patients affected by influenza A (H1N1) virus infection. Crit Care. 2011;15(1):R66.

20. Jung JY, Park BH, Hong SB, et al. Acute kidney injury in critically ill patients with pandemic influenza A pneumonia 2009 in Korea: a multicenter study. J Crit Care. 2011; 26(6): 577-585.

21. Pettilä V, Webb SA, Bailey M, Howe B, Seppelt IM, Bellomo R. Acute kidney injury in patients with influenza A (H1N1) 2009. Intensive Care Med. 2011;37(5):763-767.

22. Demirjian SG, Raina R, Bhimraj A, et al. 2009 Influenza A infection and acute kidney injury: incidence, risk factors, and complications. Am J Nephrol. 2011;34(1):1-8.

23. Nin N, Lorente JA, Soto L, et al. Acute kidney injury in criti-cally ill patients with 2009 influenza A (H1N1) viral pneumo-nia: an observational study. Intensive Care Med. 2011;37(5): 768-774.

24. Sood MM, Rigatto C, Zarychanski R, et al. Acute kidney injury in critically ill patients infected with 2009 pandemic

influenza A(H1N1): report from a Canadian province. Am J Kidney Dis. 2010;55(5):848-855.

25. ANZIC Influenza Investigators, Webb SA, Pettilä V, Seppelt I, et al. Critical care services and 2009 H1N1 influenza in Australia and New Zealand. N Engl J Med. 2009;361(20): 1925-1934.

26. Louie JK, Acosta M, Winter K, et al; California Pandemic (H1N1) Working Group. Factors associated with death or hospitalization due to pandemic 2009 influenza A(H1N1) infection in California. JAMA. 2009;302(17):1896-1902.

27. Abdulkader RC, Ho YL, de Sousa Santos S, et al. Character-istics of acute kidney injury in patients infected with the 2009 influenza A (H1N1) virus. Clin J Am Soc Nephrol. 2010;5(11):1916-1921.

28. Trimarchi H, Greloni G, Campolo-Girard V, et al. H1N1 infec-tion and the kidney in critically ill patients. J Nephrol. 2010; 23(6):725-731.

29. Mauad T, Hajjar LA, Callegari GD, et al. Lung pathology in fatal novel human influenza A (H1N1) infection. Am J Respir Crit Care Med. 2010;181(1):72-79.

30. Sevignani G, Soares MF, Marques GL, et al. Acute kidney injury in patients infected by H1N1: clinical histological cor-relation in a series of cases. J Bras Nefrol. 2013;35(3):185-190.

31. Bautista E, Arcos M, Jimenez-Alvarez L, et al. Angiogenic and inflammatory markers in acute respiratory distress syndrome and renal injury associated to A/H1N1 virus infection. Exp Mol Pathol. 2013;94(3):486-492.



Don’t miss a single issue

PLUS,

Get our best deal –

SUBSCRIBE TO THE

Today!of the premier source for

evidence-based critical care practice. AJCC offers the latest advances in multidisciplinary clinical research, with features

that highlight bedside applications of select research articles.

Each issue of AJCC brings you the most current clinically relevant content at your fingertips. You will have the latest

research by leaders in critical care. you will have access to information on the journal website, including

the latest clinical references, supplemental publications and content that is only online.

$100 for 2 years ($ outside the U.S.) or $59 for 1 year ($1 0 outside the U.S.) for individuals.

(Institutional pricing on request.)Order online at: http://ajcc.aacnjournals.org/site/subscriptions/

Or call: (800) 336-6348 or (818) 487-2075

Commentary

THOUGHTS ON THE IMMIGRATION DEBATE AND HEALTH CARE: A PERSONAL STORYBy Harold A. Fernandez, MD

At age 13, I found myself on a small boat, along with my brother, who was 11, and 10 other undocumented immi-grants. Accompanied only by the dark-ness of the night and by the ferocious

power of the Atlantic Ocean, we were crossing at midnight through one of the corners of the perilous waters of the Bermuda Triangle. Our goal was to reach the coast of Miami undetected by the US Coast Guard. We all thought that we were going to die. We were all praying, and crying, and imploring the Lord to help us stay alive. Almost 40 years later, my feelings during that ordeal are still vivid in my mind. I had not seen my parents for several years, and thus I was asking God for just 1 more second, so that I could see my parents again, and touch their faces, and feel their warmth and their love. In that moment, when I thought that I was going to die, I was not thinking about the opportunities that my parents had told us we would find in America, or about the prospect of a better life and a better education. I was just begging for the simple gift to be reunited with my parents, even if it was for just 1 second. Since then, I have learned to have a deeper appreciation for the gift of being together with my family. As such, it really tears at my heart when I see families ripped apart as a result of our current immigration policies.

I am now a professor of cardiothoracic surgery for the Zucker School of Medicine at Hofstra/North-well. I have had the incredible privilege and honor to attend some of the most prestigious universities in America, including Princeton, Harvard, and the Massachusetts Institute of Technology. But what

qualifies me to write this article is something unre-lated to my educational pedigree. You see, I actually know firsthand what it feels like to live in the shad-ows of American society as an undocumented immi-grant. The fact is that behind the façade of a normal life, our days are filled with fear and the torment that someday we, or one of our loved ones, will be dis-covered by the immigration service and sent away from our family. I also know what it feels like to live away from your parents and the people that you love. I have witnessed my own family and many members of my community suffer these fears for many years.

Just to be clear, we are talking about millions of hard-working, law-abiding, decent people who love this country, but are trapped in a legal battle to become permanent residents and do not see any hope or help available. This is in fact a human trag-edy right in our own backyard, and it is one that can be resolved. I would propose a plan to enact legisla-tion that would enable the law-abiding immigrant community to come out of the shadows and begin a process of legal residency in the United States, with an eventual path to full citizenship over a period of 5 years. The people that I am talking about are not criminals or rapists; they are hard-working people who are just looking for dignity and for an opportunity to be part of the American dream. This plan is the most compassionate and pragmatic way to solve this problem. It would not be the first time in our history that we have acted with compassion and empathy to solve or improve our immigration problems.

When I was an undocumented student at Princeton University in the mid-1980s, I had the support of a Democratic senator, Bill Bradley; a Republican governor of New Jersey, Thomas Kean;

©2018 American Association of Critical-Care Nursesdoi: https://doi.org/10.4037/ajcc2018509



“ I feel that the element of compassion is missing from our

debate on immigration reform. ”

and a Republican president, Ronald Reagan. At that time, there was a true, honest, bipartisan effort to come up with a plan for immigration reform and to try to do what felt most compassionate. Currently, I feel that the element of compassion is missing from our debate on immigration reform. We have almost forgotten that we are actually talking about humans and not about aliens from another planet. Many of our political leaders have embraced rhetoric of divi-sion and have promoted fear and anger toward immi-grant communities in ways that are derogatory and divisive. Nonetheless, my love and respect for our country have not changed and will never change. I adore the United States and everything that it rep-resents. We have the most balanced democratic sys-tem in the world. Furthermore, I still feel that this is the most compassionate country in the world. I have confidence that if most of the people who oppose a compassionate solution to immigration reform actually made a personal acquaintance at a human level with an undocumented family in America, their feelings would change. Imagine that instead of hav-ing a “day without an immigrant,” we had a day where all of us who are American citizens would reach out to an undocumented family and simply listen to their story.

I feel that our discussions about immigration reform have been so polarized by extreme groups that we have lost touch with the people who are really being affected. I know, for example, that peo-ple who read my book, Undocumented: My Journey to Princeton and Harvard and Life as a Heart Surgeon,1 really start to think about immigration reform from another point of view. They truly begin to see the human side. They begin to see that we are not talking about criminals, or rapists, or terrorists. We are in fact talking about the people who labor with extreme devotion on our farms picking the fruits

and vegetables that we eat every day, in the kitchens of the restaurants that we visit, cutting the lawns of our homes, helping many families with child care, and so on. Once most Americans really meet these people in a personal way, they will change how they feel about forcing them out of our country.

So, what would I do if I could change the debate on immigration? The way I see it is very simple. It is very difficult to change how people feel concerning immigration by just talking about the facts and the data, especially now that we have been inundated with countless stories of false news. We know that for every point of information that someone on the left has about the beneficial impact of immigrants, someone on the right will have a different interpre-tation and then will passionately talk about a crime committed by a few undocumented immigrants or examples of how they may be stealing jobs from American citizens. I myself have been accused of having stolen a position at Princeton and at Harvard from a well-deserving American citizen. I have often been involved in such discussions with people who I know are good people. I get this. Nonetheless, this is a very important topic that affects all of us, espe-cially all of us in health care delivery, and we need to have an honest discussion that examines all view-points. So when our politicians come together and discuss immigration reform, I would hope that they all make an effort to visit an undocumented family at their home and listen to them. If they cannot, I hope that they will read my book; this is an invita-tion into the life of undocumented families in Amer-ica: our struggles, dreams, and desires. I know that if politicians did this, they would no longer fear undocumented immigrants, and they would make a greater effort to have an open mind and an open heart when they decide how to proceed with immi-gration reform.

One of my current activities outside my hospital duties is reaching out to young immigrant students. Despite the rhetoric of hate and division, I still advise undocumented students to be patient and to keep on dreaming. In fact, I tell them to dream big! I tell them to use all their energy to become the best stu-dents they can possibly become. There are already several great universities that will accept you even as an undocumented student if they feel that you meet their requirements. I tell them to use all their

About the AuthorHarold A. Fernandez is professor of cardiovascular and thoracic surgery, Zucker School of Medicine at Hofstra/ Northwell, Hempstead, New York. He is also chief of cardiovascular surgery at Southside Hospital, Bay Shore, New York, and the director of surgical heart failure at Northwell Health System, Great Neck, New York.

Corresponding author: Harold A. Fernandez, MD, Southside Hospital, 301 E. Main St, Bay Shore, NY 11706 (email: [email protected]).


reform would save us hundreds of millions of dol-lars so that we can concentrate on preventing diseases rather than treating the ravaging consequences of long-standing diabetes and obesity, which have become so prevalent in our community.

I honestly feel that on many levels, it is our moral and ethical obligation to support and demand a comprehensive plan for immigration reform. Because such reform will not be a simple undertaking and will take years to resolve, we have made a strong effort as a system to reach out and provide this com-munity with some peace of mind by letting them know that our doors are open and that we will con-tinue to treat everyone who needs help with respect, compassion, and human dignity.

The statements and opinions contained in this commentary are solely those of the author.


REFERENCE1. Fernandez HA. Undocumented: My Journey to Princeton

and Harvard and Life as a Heart Surgeon. Santa Monica, CA: Malevolent Books; 2017.


“ It is our moral and ethical obligation to support and

demand a comprehensive plan for immigration reform. ”

feelings of despair to fuel their passion for more edu-cation. After all, I am a big believer in the concept that only through an education can you improve your chances of finding compassion and opening up the doors to better opportunities. This is in fact how the general immigrant community feels. They know that if they are productive members of our society, they will eventually be rewarded with compassionate and fair treatment from our society.

My current clinical job is at the Northwell Health System, the largest health system in the state of New York. I am the chief of cardiothoracic surgery at South-side Hospital in Bay Shore and also the director of surgical heart failure for the system. We have made a conscious effort to bring state-of-the-art care to all the regions that we serve in our system, including the regions that have a high concentration of patients who are underinsured. Our system treats thousands of immigrant patients, both legal and undocumented. After the last election, many people in the Latino community felt marginalized by the inflammatory and divisive rhetoric initiated by our current presi-dent. Many were concerned about seeking medical assistance even if they were critically ill. Immigration reform will allow people to feel comfortable about seeking medical assistance. It will also allow people to go and see their doctors for both early diagnosis and preventive care. In the long term, immigration

PROGNOSTIC ECG CHANGES IN A PREOPERATIVE ASSESSMENTBy Teri M. Kozik, RN, PhD, CNS, CCRN, Salah S. Al-Zaiti, RN, PhD, CRNP, CCRN, Mary G. Carey, RN, CNS, PhD, and Michele M. Pelter, RN, PhD

©2018 American Association of Critical-Care Nurses doi:https://doi.org/10.4037/ajcc2018405

Teri M. Kozik is a nurse researcher at St Joseph’s Medical Center, Stockton, California. Salah S. Al-Zaiti is an assistant professor at the Department of Acute and Tertiary Care Nursing, University of Pittsburgh, Pennsylvania. Mary G. Carey is associate director for clinical nursing research, Strong Memorial Hospital, Rochester, New York. Michele M. Pelter is an assistant professor at the Department of Physiological Nursing, University of California, San Francisco, California.

ECG Puzzler A regular feature of the American Journal of Critical Care, the ECG Puzzler addresses electrocardiogram (ECG) interpretation for clinical practice. We welcome letters regarding this feature.

Yes No NA

Yes No NA

Yes No NA

Yes No NA

Yes No NA

Yes No NA

Yes No NA

Yes No NA

Scenario: This 12-lead electrocardiogram (ECG) was obtained from a 53-year-old man on a Monday morning at an outpatient surgery center as part of a preoperative assessment for an esophagogastoduode-noscopy and colonoscopy. This was part of an evalua-tion for a kidney transplant due to end-stage renal

failure. Other history includes hypertension, diabe-tes, and hypercholesteremia. His dialysis was sched-uled for 3 times per week (Monday, Tuesday, and Fridays), however, he missed his Friday appointment. He complained of muscle weakness and tiredness, but all other vital signs were within normal range.


Interpretation Questions:

1. Is the ECG properly calibrated (10 mm) and are leads properly placed? If no, interpret cautiously.2. Is this a sinus rhythm (one P wave preceding every QRS complex)?

If no, check for number of P waves in relation to QRS complexes.3. Is the heart rate (R-R interval) normal (60-100/min)?

If no, check for supraventricular or ventricular arrhythmias.4. Is the QRS complex narrow (duration ≤ 110 milliseconds [ms] in V

1)? If no, check for bundle branch blocks (BBBs), pacing, or ventricular arrhythmia.5. Is the ST segment deviated (≥ 2 mm in V2-V3, or ≥ 1 mm in other leads)?

If yes, check for similar deviations in contiguous cardiac territories.6. Is the T wave inverted in relation to the QRS (> 5 mm)?

If yes, check for ST deviation or conduction abnormalities.7. Is the QT interval lengthened (> 450 ms [men] or > 470 ms [women])?

If yes, check for ventricular arrhythmias or left ventricular hypertrophy.8. Is R- or S-wave amplitude enlarged (S wave V

1 + R wave V5 > 35 mm)?If yes, check for axis deviation or other chamber hypertrophy criteria.

I

II

III

II

Speed: 25 mm/sec Limb: 10 mm/mV Chest: 10 mm/mV

aVF

aVL

aVR V1

V2

V3

V6

V5

V4

I

II

III

II

Speed: 25 mm/sec Limb: 10 mm/mV Chest: 10 mm/mV

aVF

aVL

aVR V1

V2

V3

V6

V5

V4

InterpretationNormal sinus rhythm with early repolarization pat-

tern (verified from prior ECGs) and peaked, tentlike T waves suggestive of hyperkalemia. Early repolarization can manifest as atypical nonischemic ST-segment eleva-tion indicated by the “fishhook” pattern.

Rationale Because this patient had not had dialysis for 6 days, complained of muscle weakness, and displayed tentlike T waves, potassium level was measured urgently and was 8.2 mmol/L (normal range, 3.5-4.8 mol/L). Tall T waves can be a normal variant, especially in young men; how-ever, the T wave will typically have a broad base and a rounded apex. In the presence of hyperkalemia, as seen in this example, the T wave is tall, has a narrow base, and a characteristically rapid rise and fall, which results in a tented or pinched appearance. Other ECG findings related to hyperkalemia include low-amplitude P waves as seen in this ECG and cardiac arrhythmias including

ventricular tachycardia, ventricular fibrillation, sinus arrest, and asystole. Other signs and symptoms of hyperkalemia are muscle weakness, also present in this patient, and in some cases paralysis.

ManagementNormal potassium levels are maintained by the kidneys

or dialysis in end-stage renal disease. Because this patient missed a scheduled dialysis appointment, followed by a weekend with no scheduled dialysis, his potassium level increased to a seriously elevated level. Because this patient had muscle weakness and the ECG findings described here, he was classified as a hyperkalemic emergency. Treatment in this scenario often consists of dialysis and intravenous administration of calcium, insulin, and/or diuretics if nor-mal kidney function is present. This patient had a dialysis appointment scheduled that morning, so he was transported emergently by ambulance from the outpatient clinic to the dialysis center with continuous cardiac monitoring and was treated successfully.

Answers:1. Yes, the ECG is properly calibrated per the calibration marks.2. Yes, this is sinus rhythm.3. Yes, the heart rate is 60/min.4. Yes, the QRS interval is narrow.5. Yes, the ST segments are deviated in leads I and V

4 through V6, which have a fishhook appearance.6. No, the T waves are not inverted, but are > 5 mm (I, II, aVR, V4 to V6) and have a tall “tented” appearance.7. No, the QT interval is normal. 8. No, no evidence of ventricular hypertrophy is apparent.


FLORIDAPlantation43rd Annual Spring SeminarDate: April 14, 2018. Place: Renaissance Hotel. Address: 1230 S Pine Island Rd, Plantation, FL 33324. Keynote Speakers: Christine Schulman, AACN President, Vicki Good, Rosemary Lee, Tracey Melhuish, Janis Smith-Love, Denese Sabatino. Sponsor: Broward County Chapter of AACN. Contact: Patty Kelly. Phone: 954-722-8020. Email: pattykelly7 @att.net. Fee: Contact chapter.

HAWAIIHonoluluBreadth of Why: 2018 Acute and Critical Care SymposiumDate: February 14-15, 2018. Place: Ala Moana Hotel. Address: 410 Atkinson Dr, Honolulu, HI 96814. Keynote Speakers: Christine Schulman, AACN President, Nicole Kupchik, Eugene Mondor, Dr Hao Chi Ho, Dr Emilio Ganitano, Dr Scott Harvey, Sharon Chun. Sponsor: Hawaii Islands Chapter of AACN. Contact: Sherwin Alop. Phone: 808-372-3436. Email: [email protected]: hicaacn.nursingnetwork.com. Fee: Early bird, $150-$280 both days; after January 3, $175-$300 both days.

MONTANAMissoulaCritical Care SymposiumDate: February 26, 2018. Place: Holiday Inn Parkside, Mis-soula, MT. Keynote Speaker: Multiple Speakers. Sponsor: Bitterroot Chapter of AACN. Contact: Kati McLeod. Phone: 406-698-2346. E-mail: [email protected]. Fee: $80.

TEXASTylerCCRN/PCCN ReviewDate: March 13-14, 2018. Place: University of Texas, Health Science Center at Tyler. Address: 11937 US-271, Tyler, TX 75708. Keynote Speaker: Shirlene Soloman. Sponsor: Greater East Texas Chapter of AACN. Contact: Trey Moses. Phone: 903-352-0432. Email: [email protected]. Fee: $200 both days; $125 one day; $225 nonmembers.

WISCONSINGreen Bay2 Day CCRN ReviewDate: January 20-21, 2018. Place: Bellin College. Address: 3201 Eaton Rd, Green Bay, WI. Keynote Speaker: Kay Hoppe. Sponsor: Northeastern Wisconsin Chapter of AACN. Contact: Lisa Renn. Email: [email protected]. Phone: 920-606-6609. Fee: $150, members; $200, nonmembers.

For AJCC Education Directory submission information, phone (800) 809-2273, ext 532, or e-mail [email protected].

Education Directory

www.ajcconline.org

Your ticket to a great job offer.

15-0876

Search daily job postings

on the homepage.

Choose from the best career

opportunities in nursing.

Start now. The perfect job

may be waiting for you.

You’re admitted to AACN’s Official Career Center.

Designed as a comprehensive career

resource for nurses of all levels, the

Career Center enables you to explore

job postings by specialty area,

location, and hospital/facility.

Post your CV or résumé today.

AACNCareerCenter.org

17-N

HPr

m-0

636B

TM

/CH

9/1

7

THE DEFINITIVE RESOURCE for critical care nursing certification and practice

American Association of Critical-Care Nurses; and Tonja M. Hartjes, DNP, ACNP-BC, FNP-BC, CCRN-CSC, FAANPOctober 2017ISBN: 978-1-4557-1065-2

Visit aacn.org/store to get your copy today!

• NEW! Significantly updated content corresponds to the most recent CCRN® examination blueprint and reflects the most current literature, national and international guidelines, clinical studies, AACN/ANA scope of practice, and ECCO 3.0.

• NEW! Expanded coverage of key topics addresses the healthcare needs of today’s patients, including pain management, palliative care, and end-of-life care; nutritional support; transplantation; and key AACN initiatives.

• NEW! Clinical Pearls and Key Concepts boxes highlight important content and serve as a convenient quick reference.

• NEW! Improved navigation includes printed index tabs to help distinguish chapters and find information fast.

• Basic to advanced levels of coverage, with an emphasis on clinical nursing practice, prepare you for the full range of knowledge you will encounter on the CCRN exam and in practice.

NEW EDITION!

© 2017 Medline Industries, Inc. All rights reserved. Medline and Comfort Glide are registered trademarks of Medline Industries, Inc. MKT1790938 / 15

Comfort Glide AirComfort GlideRepositioning Sheets

Comfort Glide WedgesComfort Glide Drypad

Comfort Glide®

We’ve got your back.

Visit medline.com/go/comfortglide or call 1-800-MEDLINE

Reduce pressure injuries. Save caregivers’ backs. Repositioning patients is a best practice for optimal skin health. But it can lead to caregiver injury if not done properly. We can help.

From slings and repositioning sheets to wedges and drypads, our Comfort Glide products work together to help protect patients and caregiversfrom injury.

american journal of critical care -...

Documents