early tracheotomy vs prolonged intubation after cardiac surgery

Early Percutaneous Tracheotomy Versus Prolonged Intubationof Mechanically Ventilated Patients After Cardiac SurgeryA Randomized Trial*Jean-Louis Trouillet, MD; Charles-Edouard Luyt, MD, PhD; Marguerite Guiguet, PhD; Alexandre Ouattara, MD, PhD; Elisabeth Vaissier, MD;Ralouka Makri, MD; Ania Nieszkowska, MD; Pascal Leprince, MD, PhD; Alain Pavie, MD; Jean Chastre, MD; and Alain Combes, MD, PhD

Background: Whether early percutaneous tracheotomy in patientswho require prolonged mechanical ventilation can shorten mechan-ical ventilation duration and lower mortality remains controversial.

Objective: To compare the outcomes of severely ill patients whorequire prolonged mechanical ventilation randomly assigned toearly percutaneous tracheotomy or prolonged intubation.

Design: Prospective, randomized, controlled, single-center trial(ClinicalTrials.gov registration number: NCT00347321).

Setting: Academic center.

Patients: 216 adults requiring mechanical ventilation 4 or moredays after cardiac surgery.

Intervention: Immediate early percutaneous tracheotomy or pro-longed intubation with tracheotomy 15 days after randomization.

Measurements: The primary end point was the number ofventilator-free days during the first 60 days after randomization.Secondary outcomes included 28-, 60-, or 90-day mortality rates;durations of mechanical ventilation, intensive care unit stay, andhospitalization; sedative, analgesic, and neuroleptic use; ventilator-associated pneumonia rate; unscheduled extubations; comfort andease of care; and long-term health-related quality of life (HRQoL)and psychosocial evaluations.

Results: There was no difference in ventilator-free days during thefirst 60 days after randomization between early percutaneous tra-cheotomy and prolonged intubation groups (mean, 30.4 days [SD,

22.4] vs. 28.3 days [SD, 23.7], respectively; absolute difference, 2.1days [95% CI, �4.1 to 8.3 days]) nor in 28-, 60-, or 90-daymortality rates (16% vs. 21%, 26% vs. 28%, and 30% vs. 30%,respectively). The durations of mechanical ventilation and hospital-ization, as well as frequencies of ventilator-associated pneumoniaand other severe infections, were also similar. However, early per-cutaneous tracheotomy was associated with less intravenous seda-tion; less time of heavy sedation; less haloperidol use for agitation,delirium, or both; fewer unscheduled extubations; better comfortand ease of care; and earlier resumption of oral nutrition. After amedian follow-up of 873 days, between-group survival, psychoso-cial evaluations, and HRQoL were similar.

Limitation: The prolonged intubation group had more ventilator-free days during days 1 to 60 than what was hypothesized (mean,23.0 days [SD, 17.0]).

Conclusion: Early tracheotomy provided no benefit in terms ofmechanical ventilation and length of hospital stay, rates of mortalityor infectious complications, and long-term HRQoL for patients whorequire prolonged mechanical ventilation after cardiac surgery.However, the well-tolerated procedure was associated with lesssedation, better comfort, and earlier resumption of autonomy.

Primary Funding Source: French Ministry of Health.

Ann Intern Med. 2011;154:373-383. www.annals.orgFor author affiliations, see end of text.* The Early Percutaneous Tracheotomy for Cardiac Surgery Trial.

Tracheotomy is a common procedure for patients whorequire prolonged mechanical ventilation (1, 2). The

development of the percutaneous dilatation technique,which allows easy and rapid bedside placement of thetracheal cannula by trained intensivists, has markedlyincreased its implementation in intensive care units(ICUs) over the past decade (3, 4). Tracheotomy may bebeneficial by lowering airway resistance, improving oralhygiene, and improving pulmonary toilet and airwaysecurity (4 – 6). It might also be associated with lesssedative administration (6, 7), less time of heavy seda-tion (6, 7), enhanced patient autonomy and comfort,and fewer pulmonary infections (6, 8). However, trans-laryngeal intubation avoids local tracheotomy complica-tions (for example, bleeding, perforation, and pneumo-thorax) and preserves cough strength. The effect of earlytracheotomy on hospital mortality, mechanical ventila-tion duration, and ventilator-associated pneumoniarates of long-term ventilated ICU patients remains hotlydebated (2, 6, 9 –14).

To assess whether early percutaneous tracheotomy re-ally improves the outcomes of patients who require long-term mechanical ventilation, it must be tested on patientswith prolonged mechanical ventilation (exceeding morethan 14 to 21 days) whose confounding factors that might

See also:

PrintEditors’ Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374Editorial comment. . . . . . . . . . . . . . . . . . . . . . . . . . 434Summary for Patients. . . . . . . . . . . . . . . . . . . . . . . I-38

Web-OnlyAppendixAppendix TablesAppendix FiguresCME quizConversion of graphics into slides

Annals of Internal Medicine Original Research

© 2011 American College of Physicians 373

affect mechanical ventilation duration, for example, seda-tion and mechanical ventilation weaning policies, aretightly controlled. Prolonged mechanical ventilation aftercardiac surgery is becoming more and more common (15–17) because patients referred for heart surgery have everpoorer risk profiles with more comorbid conditions, andpatients still on mechanical ventilation for more than 3 to4 days after cardiac operations are infrequently weanedwithin the following 7 to 10 days (16, 18). As a result, thisgroup of critically ill patients might represent the pop-ulation of choice for testing potential early tracheotomybenefit.

This prospective, randomized, controlled studycompared the outcomes of cardiac surgery patients whostill require mechanical ventilation 4 or more days aftersurgery who are randomly assigned to immediate earlytracheotomy or prolonged intubation with possibly latetracheotomy.

METHODS

DesignThis prospective, randomized, controlled, single-

center trial, conducted from June 2006 to March 2009,included patients who required prolonged mechanical ven-tilation after heart surgery and were randomly assigned toimmediate early tracheotomy or prolonged intubation withtracheotomy only when mechanical ventilation exceededday 15 after randomization. Sedation and mechanicalventilation weaning policies were rigorously controlled.The local ethics committee approved the protocol. Pa-

tients’ relatives provided written informed consent be-fore randomization.

PatientsPatients were eligible if they had undergone cardiac

surgery; were still on mechanical ventilation 4 days there-after; had not successfully passed a mechanical ventilationweaning screening test or spontaneous breathing trial onthe day of randomization, according to the Ely protocol(19); and were expected to require mechanical ventilationfor 7 or more days (16). Exclusion criteria included personswho were younger than 18 years, were pregnant, were pre-viously enrolled in this or other trials evaluating morbidityor mortality, had received more than 48 hours of mechan-ical ventilation preoperatively, had previous tracheotomywithin 6 months, had received an artificial heart device,had a prothrombin time greater than 1.5 times the upperlimit of normal or platelet count less than 50 �109 cells/Ldespite replacement, had an irreversible neurologic disor-der, had a Simplified Acute Physiology Score II greaterthan 80 (20), or had decided to limit care. Patients withsoft-tissue neck infections or anatomical deformities, mak-ing percutaneous tracheotomy unsafe, or those who hadconcomitant neck or carotid surgery were also excluded.

Demographic, preoperative and perioperative physio-logic and radiographic (21, 22) features, coexisting condi-tions (23, 24), and mechanical ventilation characteristics atrandomization were recorded. Preoperative predicted oper-ative mortality was calculated by using the European Sys-tem for Cardiac Operative Risk Evaluation (25).

Randomization and InterventionAfter screening for inclusion and exclusion criteria on

the day of randomization, we used an independent,computer-generated randomization sequence (Unite deRecherche Clinique, Pitie-Salpetriere Hospital, Paris,France) to assign patients, in a 1:1 ratio, to either earlytracheotomy (before the end of calendar day 5 after sur-gery) or prolonged intubation. Randomization was strati-fied (minimization) by the Simplified Acute PhysiologyScore II (either �45 or �45) calculated on the dayof randomization; the randomization procedure waspassword-protected and accessed by the principal inves-tigators or study coordinator after the patient had metselection criteria and the surrogate gave consent. Thepatient’s initials were entered, and treatment allocationwas assigned.

Tracheotomies were done bedside by experienced in-tensivists in the ICU using the Ciaglia percutaneous tech-nique (Ciaglia Blue Rhino, Cook Critical Care, Blooming-ton, Illinois). Any side effect or technical problem thatoccurred during the procedure was recorded. All patientswere managed with goal-directed sedation, guided by theRichmond Agitation Sedation Scale (RASS) (26). Only 2sedatives (propofol or midazolam) and 1 narcotic (sufen-tanil) were used. Sedation was monitored and evaluated bynurses 8 times daily to maintain that the patient was calm

Context

It is not clear whether early tracheotomy improves theoutcome of patients who are expected to require pro-longed mechanical ventilation.

Contribution

In this randomized trial of patients who still requiredmechanical ventilation 4 days after cardiac surgery, imme-diate tracheotomy did not decrease mortality or length ofintensive care unit or hospital stay or increase the numberof days off the ventilator compared with waiting 2 weeksto consider tracheotomy. Early tracheotomy did, however,reduce the use of sedatives, ease nursing care, and im-prove patient comfort.

Caution

Patients in the control group were free of mechanicalventilation sooner than anticipated, possibly limiting theability to detect a benefit from early tracheotomy.

Implication

More study is required to determine whether routine earlytracheotomy is beneficial.

—The Editors

Original Research Early Versus Late Tracheotomy After Cardiac Surgery

374 15 March 2011 Annals of Internal Medicine Volume 154 • Number 6 www.annals.org

and cooperative or lightly sedated (respective RASS scores,0, �1, and �2). Our ICU’s nurse–patient and nurse’saide–patient ratios were always 1:2.25 and 1:5 respectively.When the treating physician considered the sedation leveltoo profound (RASS score, �3 to �5) and not justified bythe severity of the patient’s respiratory distress, the studyprotocol recommended 50% reduction of the initially pre-scribed sedative dose. Sedatives, narcotics, or both wereinterrupted daily at 9 a.m. (27). Sedative infusions werestopped when the patient remained calm and cooperativeor, if agitation prevented successful weaning, were restartedat half the previous dose and adjusted according to need.After stopping intravenous sedation, oral benzodiazepines,neuroleptics, or both were prescribed as needed. In partic-ular, haloperidol (1- to 5-mg increments) was used to treatagitation, delirium, or both.

For both groups, mechanical ventilation weaning wasconducted according to a strict protocol, beginning as earlyas 1 day after randomization (Appendix, available atwww.annals.org).

Outcomes and Follow-upThe primary end point was the number of days alive

and breathing without assistance (ventilator-free days) dur-ing the first 60 days after randomization. Ventilator-freedays were counted from the last day that a patient receivedmechanical ventilation during the 60-day period. In par-ticular, patients not receiving mechanical ventilation whodied before day 60 were assigned 0 ventilator-free days, andeach day a patient required noninvasive mechanical venti-lation for more than 4 hours after extubation or decannu-lation was counted as 1 mechanical ventilation day, there-fore resetting ventilator-free days at 0. However, days withthe tracheotomy cannula in place but without mechani-cal ventilation before decannulation were counted asventilator-free days. Sensitivity analyses were also done as-signing ventilator-free days when death occurred beforeday 60 and using the exact number of days of mechanicalventilation for counting ventilator-free days (Appendix).

Secondary end points included the number ofventilator-free days at 28 and 90 days (based on datathrough 28 and 90 days); 28-, 60-, and 90-day mortalityrates; durations of mechanical ventilation and length ofICU and hospital stays; the number of endotrachealprosthesis-free days at day 60; frequencies of unscheduledextubations, decannulations (tracheal cannula removed),and reintubations or recannulations; 7-, 14-, 21-, and 28-day Sequential Organ Function Assessment (28) scores inthe ICU; and durations of vasopressor (exclusively epi-nephrine or norepinephrine) and renal replacement ther-apy. The sedatives, analgesics, and neuroleptics used andtheir cumulative daily doses during the first 15 days wererecorded. Sedation-free days at day 28 were calculated. Inaddition, RASS scores were recorded 8 times every 24hours during days 1 to 15 after randomization. Three se-dation status categories based on RASS scores were defined

a posteriori: heavily sedated (RASS score, �5, �4, or �3);calm, awake, or lightly sedated (RASS score, �2, �1, 0, or�1); and agitated (RASS score, �2, �3, or �4). Thenumber of hours per day spent in each of these conscious-ness states was calculated. If 2 consecutive scores were dif-ferent (for example, heavily sedated at 9 a.m. and awakeand calm at 12 p.m.), the patient was considered to havehad the former score 50% of the time and the latter scorefor 50% (that is, 90 minutes heavily sedated and 90 min-utes awake and calm). If the same sedation level was re-corded twice, the patient was considered to have remainedat this level during the time between these 2 assessments.Analyses of time spent in each consciousness state werebased on these daily scores.

Ventilator-associated pneumonia was suspected whena new and persistent radiographic infiltrate was accompa-nied by purulent secretions, temperature of 38.3 °C ormore, or a leukocyte count greater than 10.0 � 109 cells/L,or when 1 of these occurred in patients with baseline dif-fuse, dense infiltrates. Ventilator-associated pneumonia wasdiagnosed before administration of antibiotics by quantita-tive distal bronchoalveolar lavage cultures growing at 104

colony-forming units/mL or greater (29, 30). Other com-plicating infections (bloodstream, sternal wound, and trache-otomy stomal site, which required antibiotics, surgical de-bridement, or both) and duration of antibiotic treatment werealso recorded.

Nurses evaluated ease of care and patient comfort dailyby using subjective scores; a trained physiotherapist evalu-ated muscle strength with the Medical Research Councilscore; and laryngeal, tracheal, and other complicationswere also recorded (Appendix).

Long-Term Follow-up, Health-Related Quality of Life,and Psychosocial Evaluations

We designed a posteriori a cross-sectional study oflong-term outcomes, which was conducted from May2010 to June 2010, on all patients who were alive on day90. After explaining the study objective and asking forinformed consent during a telephone call, we administeredthe following questionnaires to survivors: activities of dailyliving (31, 32); 36-Item Short Form Health Survey (SF-36) (33, 34) evaluating health-related quality of life(HRQoL); the Hospital Anxiety and Depression Scale (35,36); and the Impact of Event Scale (37), which assessedposttraumatic stress disorder (Appendix).

Lastly, patients were asked if the endotracheal tube orcannula had caused pain or discomfort; if they had swal-lowing problems, phonation problems, or both after ICUdischarge; and if the tracheotomy scar engendered aestheticembarrassment.

Statistical AnalysisOn the basis of personal data, we predicted the pro-

longed intubation group would have a mean of 23 days(SD, 17) of ventilator-free days in the first 60 days. Todemonstrate that early tracheotomy achieved an absolute

Original ResearchEarly Versus Late Tracheotomy After Cardiac Surgery

www.annals.org 15 March 2011 Annals of Internal Medicine Volume 154 • Number 6 375

increase in ventilator-free days over 7 days, with 80%power and 5% type I error, 198 patients were required(38). To account for persons who withdrew, 216 patientswere enrolled. No interim analysis was planned, but anindependent safety monitoring board reviewed the trial’sprogress and evaluated adverse events according to treat-ment assignment. All randomly assigned patients were in-cluded in the analyses according to their randomized treat-

ment assignment, all patients received the allocatedintervention, and none was lost to follow-up during thefirst 90 days. Patients randomly assigned to the prolongedintubation group who had late tracheotomy were alwaysanalyzed in the prolonged intubation group. Binary vari-ables were compared with chi-square tests and continuousvariables with the t test, as appropriate. The between-groupdifferences were assessed by using risk difference or meandifferences, presented with 95% CIs. Post hoc adjustmentfor between-group differences in baseline characteristics(repeated operation, heart transplantation, and renal replace-ment therapy) was achieved with multivariable regressionanalyses.

Sedation administration and number of hours spentin each consciousness state (heavily sedated; calm,awake, or lightly sedated; or agitated) were evaluateddaily during days 1 to 15 after randomization. Compar-isons between groups were done using a mixed model inwhich intervention group and day were included as fixedeffects while random effects for the subject was takeninto account through the repeated measures over time.Paired t tests or Wilcoxon tests were used to comparesurvivors’ mean SF-36 scores with age- and sex-matchedFrench population normative values (33, 34). For sur-vival, follow-up started at randomization, and cumula-tive event curves were estimated with the Kaplan–Meiermethod; survival curves were compared with the log-rank test. All tests were 2-sided. We used SAS software,version 8.02 (SAS Institute, Cary, North Carolina), forstatistical analyses.

Role of the Funding SourceThis study was sponsored by an academic grant from

the French Ministry of Health (Programme Hospitalier deRecherche Clinique regional 2005). The study sponsor didnot participate in the study design, data collection, dataanalysis, data interpretation, or writing and the decision tosubmit this manuscript for publication.

RESULTS

Among the 3484 consecutive patients who had under-gone heart surgery from June 2006 to March 2009, a totalof 287 were still intubated 4 or more days after surgery,and 216 were enrolled in the study: 109 assigned to earlytracheotomy and 107 to prolonged intubation (Figure 1).None was lost to follow-up on day 90. Table 1 and Ap-pendix Table 1 (available at www.annals.org) show base-line characteristics of enrolled patients. Characteristics atrandomization were similar for the 2 groups, except forhigher rates of heart transplantation, repeated cardiac sur-gery, and renal replacement therapy in the early tracheot-omy group. Twenty-nine (27%) patients in the prolongedintubation group had late tracheotomy because of the ex-pected need for more prolonged mechanical ventilation.

There was no difference in ventilator-free days duringthe first 60 days after randomization (Table 2) between

Figure 1. Study flow diagram.

Not enrolled (n = 71)SAPS II >80: 23Considered rapidly weanable:

15Logistical reasons: 10Transferred to another

hospital: 8Associated carotid surgery: 5Declined consent: 5Artificial heart: 4With MV >48 h before

surgery: 1

Assessed for eligibility(on MV 4 d after surgery) (n = 287)

Enrolled and randomlyassigned to a treatment group

(n = 216)

Included in theprimary analysis (n = 109)Survivors at 90 d (n = 76)

Cardiac surgery patients(n = 3484)

Early percutaneoustracheotomy (n = 109)

Long-term follow-upAlive in June 2010 (n = 62)Deceased (n = 12); lost to

follow-up (n = 2)

Lost to follow-up at 90 d(n = 0)

HRQoL evaluation (n = 60)

Included in theprimary analysis (n = 107)Survivors at 90 d (n = 75)

Prolonged intubation(n = 107)

Long-term follow-upAlive in June 2010 (n = 57)Deceased (n = 17); lost to

follow-up (n = 1)

Lost to follow-up at 90 d(n = 0)

HRQoL evaluation (n = 56)

HRQoL � health-related quality of life; MV � mechanical ventilation;SAPS II � Simplified Acute Physiology Score.



early tracheotomy and prolonged intubation groups (mean,30.4 days [SD, 22.4] vs. 28.3 days [SD, 23.7], respectively;absolute difference, 2.1 days [95% CI, �4.1 to 8.3 days]),nor in 28-, 60-, or 90-day mortality rates (16% vs. 21%,26% vs. 28%, and 30% vs. 30%, respectively). TheKaplan–Meier estimates of survival probability (Figure2, top); ventilator-free days on days 28 and 90; endotra-cheal prosthesis-free days; durations of mechanical ven-tilation and ICU and hospital stays; rates of ventilator-associated pneumonia, stomal or sternal infections, orpositive blood cultures; and number of patients whowere administered catecholamine or antibiotics (Table 2and Appendix Table 2, available at www.annals.org)were also similar. Different definitions of ventilator-freedays did not alter these findings (Appendix Table 3,available at www.annals.org). However, patients in theprolonged intubation group had more frequent un-scheduled extubations and reintubations or recannula-tions. After adjustment for repeated operations, hearttransplantations, and renal replacement therapy at base-line, between-group differences in primary and second-ary outcomes remained similar (Appendix Table 4,available at www.annals.org).

Intravenous sedative and analgesic use at randomiza-tion were similar but decreased more rapidly thereafter forearly tracheotomy patients (Figure 3). The early tracheot-omy group had lower cumulative sufentanil, propofol, andmidazolam consumption during days 1 to 15 after ran-domization and significantly more sedation-free days dur-ing days 1 to 28 (Table 2). More haloperidol was used totreat agitation, delirium, or both in the prolonged intuba-tion group. During days 1 to 15 after randomization,20 290 sedation scores were recorded. Patients in the earlytracheotomy group spent less time heavily sedated; moretime calm, awake, or lightly sedated (Appendix Figure 1,available at www.annals.org); and more days comfortableand with care deemed “easy” according to the nurses’ sub-jective scores (Table 2). Patients in early tracheotomygroup received oral nutrition and were transferred frombed to chair sooner than patients in the prolonged intuba-tion group. By day 15, more of the patients in the earlytracheotomy group were receiving oral nutrition andundergoing bed-to-chair transfers. Muscle strengthscores were similar between the 2 groups (Table 2). Onepatient in each group had a major tracheotomy or intu-bation complication, but both had favorable outcomes.The numbers and severities of other adverse events weresimilar for the 2 groups (Appendix Table 5, available atwww.annals.org).

At long-term follow-up of 873 days after randomiza-tion (interquartile range, 547 to 1201 days; range, 93 to1491 days), 29 patients had died, 3 were lost to follow-up,and 119 were long-term survivors (Figure 1). According tothe Kaplan–Meier method, between-group long-term sur-vival probability was similar (Figure 2, bottom). Nobetween-group differences were found for activities of daily

living, anxiety, depression, or posttraumatic stress disorder(Table 3). Both groups had similar HRQoL scores at long-term follow-up compared with age- and sex-matched pop-ulation normative values; however, they had worse physicalfunctioning scores (Appendix Figure 2, available at www.annals.org). Severe anxiety and depression symptoms werepresent in 16% to 30% of long-term survivors (Table 3).Only 1 survivor in each group was deemed at risk forposttraumatic stress disorder. Only 3 patients in the earlytracheotomy group and 5 in the prolonged intubationgroup remembered feeling pain or discomfort associatedwith endotracheal prosthesis. None of the long-term survi-vors had problems with swallowing, phonation, or both,and only 1 patient reported being embarrassed by the tra-cheotomy scar.

DISCUSSION

In this large, randomized, controlled trial of patientswho require prolonged mechanical ventilation after cardiacsurgery, early tracheotomy provided no benefit in terms ofmechanical ventilation duration, length of hospital stay,mortality rate, or frequency of infectious complications

Table 1. Baseline Characteristics of Participants

Characteristic EarlyPercutaneousTracheotomy(n � 109)

ProlongedIntubation(n � 107)

Preoperative or perioperativeMean age (SD), y 64.1 (13.3) 66.0 (12.4)Men, n (%) 77 (71) 66 (62)Mean Charlson comorbidity score (SD) 2.9 (1.9) 2.6 (1.8)LVEF �0.35, n (%) 33 (30) 30 (28)Repeated cardiac surgery, n (%) 32 (29) 21 (20)Timing of surgery, n (%)

Elective 43 (39) 50 (47)Urgent 39 (36) 31 (29)Emergent 27 (25) 26 (24)

Mean EuroSCORE score (SD) 8.7 (3.2) 8.6 (4.4)Type of surgery, n (%)

CABG 27 (25) 27 (25)Valve 31 (28) 34 (32)CABG � valve 14 (13) 20 (19)Other cardiovascular surgery 14 (13) 18 (17)Heart transplantation 23 (21) 8 (7)

Mean CPB duration (SD), min 115.5 (56.9) 119.6 (55.7)

At randomizationMean SAPS II (SD) 47.2 (12.4) 45.8 (11.4)Mean SOFA score (SD) 11.6 (3.5) 10.9 (3.6)Mean epinephrine or norepinephrine

dose (SD), �g/kg/min0.59 (0.24) 0.53 (0.18)

Mean Glasgow Coma Scale (SD) 10.9 (3.3) 11.4 (3.0)Mean creatinine level (SD)

�mol/L 188.6 (98.2) 172.2 (98.6)mg/dL 2.14 (1.11) 1.94 (1.11)

Renal replacement therapy, n (%) 41 (38) 18 (17)Mean lung injury score (SD) 1.7 (0.7) 1.8 (0.7)

CABG � coronary artery bypass graft; CPB � cardiopulmonary bypass; Euro-SCORE � European System for Cardiac Operative Risk Evaluation; LVEF � leftventricular ejection fraction; SAPS II � Simplified Acute Physiology Score;SOFA � Sequential Organ Function Assessment.



over prolonged intubation possibly followed by late trache-otomy. However, it was associated with less intravenoussedation; less time of heavy sedation; less haloperidol use totreat agitation, delirium, or both; fewer unscheduled extu-bations and reintubations; better comfort; and earlier oralnutrition and bed-to-chair transfers. After a medianfollow-up exceeding 2 years, survival, psychosocial vari-ables, and HRQoL did not differ between groups, with thelatter being similar to that of age- and sex-matched Frenchpopulation normative values.

Therapeutic practices decreasing sedation and improv-ing whole-body rehabilitation of patients on mechanicalventilation were recently shown to limit delirium episodes(39, 40) and shorten the duration of mechanical ventila-tion (27, 40). As previously reported by our group (7) andothers (6), early tracheotomy was associated with less sed-ative and analgesic administration; less time of heavy seda-

tion; and earlier oral nutrition, out-of-bed mobilization, orboth. Less haloperidol use might also indicate that lesssedation and analgesia led to less agitation, delirium, orboth, although this was not specifically assessed herein.However, the early tracheotomy group’s 20% significantlyhigher number of sedation-free days did not translateinto a shorter duration of mechanical ventilation andmore ventilator-free days. Possible explanations are thatpolyneuropathy acquired in the ICU might have pro-longed mechanical ventilation for both groups or thestrict mechanical ventilation weaning protocol appliedto the prolonged intubation group might have compen-sated for higher sedative administration.

Less sedative consumption has also been postulated toimprove long-term outcomes of patients surviving criticalcare (41). Despite lower use and better nurse-assessed com-fort levels during critical care, we found no difference in

Table 2. Short-Term Outcomes and Differences Between Treatment Groups

Variable Early PercutaneousTracheotomy(n � 109)

Prolonged Intubation(n � 107)

Mean Difference orAbsolute Risk Difference(95% CI)

P Value

Mean VFDs during 1–60 d (SD) 30.4 (22.4) 28.3 (23.7) 2.1 (�4.1 to 8.3) 0.50Mean VFDs during 1–28 d (SD) 10.0 (8.8) 9.2 (10.2) 0.8 (�1.7 to 3.4) 0.52Mean VFDs during 1–90 d (SD) 49.3 (36.4) 47.5 (36.6) 1.8 (�8.0 to 11.6) 0.72Mortality, n (%)

28 d 17 (16) 23 (21) �5.9 (�16.2 to 4.4) 0.3060 d 28 (26) 30 (28) �2.3 (�14.2 to 9.5) 0.7690 d 33 (30) 32 (30) 0.4 (�11.9 to 12.6) 1.00

Mean length of ICU stay (SD), d 23.9 (21.3) 25.5 (22.2) �1.5 (�7.4 to 4.3) 0.85Mean length of hospital stay (SD), d 39.0 (27.0) 37.5 (26.9) 1.5 (�5.7 to 8.8) 0.56Mean days of MV during 1�60 d (SD) 17.9 (14.9) 19.3 (16.9) �1.3 (�5.6 to 3.0) 0.55Mean endotracheal prosthesis-free days during 1–60 d (SD) 30.3 (22.5) 28.6 (24.1) 1.7 (�4.6 to 7.9) 0.60Patients with unscheduled extubation or decannulation during 1–60 d, n (%)* 3 (3) 17 (16) �13.1 (�20.7 to �5.6) �0.001Patients with reintubation or recannulation during 1–60 d, n (%)† 17 (16) 35 (33) �17.1 (�28.3 to �5.9) 0.003Patients with noninvasive ventilation �4 h/d (during 1–60 d), n (%) 11 (10) 27 (25) �15.1 (�25.1 to �5.2) 0.004

SedationMean duration of intravenous sedation (SD), d‡ 6.4 (5.9) 9.6 (7.3) �3.2 (�5.0 to �1.3) 0.007Mean sedation-free days during 1–28 d (SD) 19.0 (9.1) 15.5 (9.3) 4.5 (1.2 to 6.9) 0.005Mean cumulative sufentanil dose during 1–15 d (SD), �g/kg 4.0 (6.5) 10.2 (18.2) �6.2 (�9.9 to �2.5) 0.001Mean cumulative propofol dose during 1–15 d (SD), mg/kg 32.9 (60.2) 67.8 (116.7) �34.9 (�60.1 to �9.8) 0.004Mean cumulative midazolam dose during 1–15 d (SD), mg/kg 2.7 (4.7) 6.4 (14.3) �3.7 (�6.6 to �0.8) 0.01Mean days (during 1�15 d) of haloperidol therapy (SD) 1.9 (3.0) 3.2 (4.2) �1.3 (�2.3 to �0.3) 0.01Mean cumulative haloperidol dose during 1–15 d (SD), mg/kg 0.26 (0.51) 0.57 (0.92) �0.3 (�0.5 to �0.1) 0.002VAP after randomization, n (%) 50 (46) 47 (44) 2.0 (�11.3 to 15.2) 0.77Sternal wound infection, n (%) 14 (13) 14 (13) �0.2 (�9.2 to 8.7) 0.96Bloodstream infection, n (%) 18 (17) 16 (15) 1.5 (�8.1 to 11.3) 0.85Mean days (during 1�15 d) nurse-assessed as comfortable (SD) 11.8 (3.8) 10.4 (4.4) 1.4 (0.3 to 2.5) 0.01Mean days (during 1�15 d) with nurse-assessed easy management (SD) 12.0 (3.8) 10.8 (4.4) 1.2 (0.05 to 2.3) 0.04Received oral nutrition at 15 d, n (%) 91 (83) 57 (53) 30.2 (18.5 to 42.2) �0.001Bed-to-chair transfer at 15 d, n (%) 72 (66) 47 (44) 22.1 (9.2 to 35.1) 0.002Muscle strength assessment (SD)§

14 d (n � 76, 68) 156.9 (87.0) 134.9 (92.8) 22.0 (�7.7 to 51.6) 0.1528 d (n � 36, 36) 164.0 (86.1) 176.9 (85.6) �12.9 (�53.3 to 27.5) 0.5242 d (n � 21, 21) 170.1 (86.4) 195.4 (67.5) �25.3 (�73.6 to 23.1) 0.3056 d (n � 8, 11) 149.7 (70.4) 185.4 (76.0) �35.7 (�108.0 to 36.6) 0.31

ICU � intensive care unit; MV � mechanical ventilation; VAP � ventilator-associated pneumonia; VFD � ventilator-free day.* Overall, from 1 to 60 d, 4 unscheduled decannulations were observed in the early tracheotomy group and 20 unscheduled extubations or decannulations were observed inthe prolonged intubation group.† Overall, from 1 to 60 d, the early tracheotomy group had 19 reintubations or recannulations and the prolonged intubation group had 42 reintubations or recannulations.‡ Mean number of days among persons with intravenous sedation (103 patients in the early tracheotomy group and 106 patients in the prolonged intubation group).§ Muscle strength was evaluated by using the Medical Research Council score. Numbers reported with each time point are patients in the early percutaneous tracheotomygroup and the prolonged intubation group, respectively.



long-term, between-group HRQoL. However, we cannotexclude the possibility that early tracheotomy might pro-vide short-term HRQoL and emotional health benefits be-cause evaluations were conducted long after ICU dis-charge. Psychosocial variables and HRQoL did notdiffer from those of French population normative val-ues, confirming that the long-term outcomes of patientswho require prolonged mechanical ventilation after car-

diac surgery are better than those of other ICU patients,probably because of the surgical cure of severe coronaryor valve diseases (17, 18, 42). However, although thefrequency of posttraumatic stress disorder was negligi-ble, 15% to 30% of long-term survivors had signs ofanxiety, depression, or both and therefore might benefitfrom strategies aimed at attenuating emotional and psy-chological distress.

Figure 2. Kaplan–Meier estimates of survival probability.

Prop

orti

on A

live

at 9

0 D

ays

Time Since Randomization, d

EPT

PILog-rank P = 0.96

Persons at risk, n

EPT 109 94 91 87 81 78 76

PI 107 94 83 82 77 75 75

0 15 30 45 60 75 900

0.2

0.4

0.6

0.8

1.0

Prop

orti

on A

live

on 3

0 Ju

ne 2

010


Log-rank P = 0.49

Persons at risk, n

EPT 109 70 70 55 46 34 21

PI 107 67 63 51 38 26 17

0 200 400 600 800 1000 12000

0.2

0.4

0.6

0.8

1.0

No significant difference (log-rank test) was found between patients assigned to the EPT or PI group (hazard ratio, 1.01 [95% CI, 0.62 to 1.65]). EPT �early percutaneous tracheotomy; PI � prolonged intubation.



Another potential benefit attributed to early tracheot-omy is a lower ventilator-associated pneumonia rate (6, 43,44), which might result from a combination of factors, forexample, easier oral hygiene and bronchopulmonary toiletor less time spent deeply sedated. Although extreme vigi-lance for ventilator-associated pneumonia was maintainedthroughout our study, early tracheotomy did not modify

the frequency of this infection. Four other trials (45–48)and the randomized Early Versus Late Tracheotomy Trial,whose primary end point was ventilator-associated pneu-monia (14), obtained similar results. Early tracheotomyalso did not affect the rates of other infectious complica-tions after cardiac surgery (for example, bloodstream infec-tion or mediastinitis).

The effect of early tracheotomy on the outcomes ofpatients requiring prolonged mechanical ventilation hasbeen the matter of heated debates, with fewer ICU andhospital deaths after tracheotomy observed in some retro-spective, cohort studies (9, 10, 13, 49, 50) but not others(44–48). A prospective, randomized study on a mixedpopulation of 120 ICU patients established that patientswho had early tracheotomy had lower mortality rates in theICU and hospital and spent less time on mechanical ven-tilation (6). However, a recent study of 419 mechanicallyventilated, mixed medical and surgical ICU, adult patientsdemonstrated that early tracheotomy did not lower thefrequency of ventilator-associated pneumonia, affect thelength of hospital stay or 28-day and 1-year mortality, orinfluence the need for care at a long-term health facilitybut did increase the number of ICU-free days andventilator-free days (14). Our analysis of 216 patients withprojected prolonged mechanical ventilation also found nomortality benefit for early tracheotomy. The results of an-other randomized, controlled trial of early tracheotomy inICU patients should help to further clarify the role of tra-cheotomy in critically ill patients (51).

Several limitations of our study should be noted. First,it was conducted within a single center, and hence, ourresults may not be applicable to patients who receive me-chanical ventilation in other centers with different casemixes, sedation practices, mechanical ventilation weaningstrategies, and ICU discharge policies. Conversely, thismonocentric design might be an advantage, with minimi-zation of biases, because of the population’s relative homo-geneity and our application of strict protocols for thosepractices. Second, according to personal data, we had basedour study sample calculation on the hypothetical mean of23 ventilator-free days (SD, 17) during the first 60 daysafter randomization for the prolonged intubation groupand an expected benefit of 30% more ventilator-free daysfor the early tracheotomy group, which was considered ameaningful clinical difference. The prolonged intubationgroup had a higher mean number of ventilator-free daysduring the first 60 days after randomization than pre-dicted, close to the expected benefit in the early tracheot-omy group. Although chance cannot be excluded, anotherpossible explanation is the implementation of modifica-tions of our ICU practices during the 2 years between thepilot study and this trial’s start. The higher number ofventilator-free days observed might also reflect the stan-dardized weaning and sedation strategies for mechanicalventilation and more focused attention on sedation reduc-tion and early identification of mechanical ventilation

Figure 3. Mean daily doses of midazolam, propofol, orsufentanil administered from 24 hours before randomizationthrough days 1 to 15 thereafter.

Sufe

ntan

il, g/kg

/d


Early percutaneous tracheotomyProlonged intubation

–1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 150

0.5

1.0

1.5

Prop

ofol

, mg/kg

/d

0

6

4

2

8

10

Mid

azol

am, m

g/kg

/d

0

0.8

0.6

0.4

0.2

1.0

1.2

µµ

Patients assigned to early percutaneous tracheotomy received less mida-zolam (mean difference, �0.31 mg/kg/d [95% CI, �0.53 to �0.09mg/kg/d]), propofol (mean difference, �2.87 mg/kg/d [CI, �4.76 to�0.98 mg/kg/d]), and sufentanil (mean difference, �0.48 �g/kg/d [CI,�0.77 to �0.19 �g/kg/d) than patients assigned to prolonged intuba-tion. The bars indicate 95% CIs.



weaning criteria for patients included in this trial. Third,27% of patients in the prolonged intubation group hadlate tracheotomy, and we cannot exclude that it might havediluted the overall early tracheotomy effect on the entirecohort. Lastly, because tracheal examinations were notdone systematically in the weeks after extubation or decan-nulation, we perhaps missed laryngotracheal complicationsrelated to tracheotomy (or possibly intubation). However,rates of tracheal stenosis, tracheal symptoms, or both (forexample, dysphonia or swallowing disorders) were low andequally distributed in patients with or without tracheot-omy in previous studies (6, 47).

In conclusion, early tracheotomy in patients who re-quire prolonged mechanical ventilation after cardiac sur-gery provided no benefit in terms of mechanical ventilationduration, length of hospital stay, and mortality and infec-tious complication rates but was associated with dimin-ished sedative, analgesic, and neuroleptic consumption andbetter comfort, ease of care, and earlier oral nutrition andbed-to-chair transfers. Future studies might explore the po-tential contribution of later (after mechanical ventilationday 10 to 15) tracheotomy.

From Institut de Cardiologie, Hopital de la Pitie–Salpetriere, AssistancePublique–Hopitaux de Paris, Universite Pierre et Marie Curie, InstitutNational de la Sante et de la Recherche Medicale, Paris, France.

Acknowledgment: The authors thank Agnes Gaubert for her data col-lection and excellent technical assistance and Chloe Djiniadhis for per-forming the muscle strength tests.

Grant Support: The Early Percutaneous Tracheotomy for Cardiac Sur-gery Trial received a research grant from the French Ministry of Health,Department de la Recherche Clinique et du Developpement (Pro-gramme Hospitalier de Recherche Clinique regional P051013–Appeld’Offre Regionale 0511, institutional review board authorization num-ber, Comite Consultatif de Protection des Personnes dans la RechercheBiomedicale 6-06).

Potential Conflicts of Interest: Dr. Trouillet: Grants received (money toinstitution): Unite de Recherche Clinique, Pitie–Salpetriere Hospital, In-stitut National de la Sante et de la Recherche Medicale U943. Dr. Luyt:Grants received/pending (money to institution): Pfizer, Kalobios, Janssen-Cilag; Payment for lectures including service on speakers bureaus: Brahms,Merck Sharp & Dohme, bioMerieux. Dr. Guiguet: Grants received(money to institution): Assistance Publique–Hopitaux de Paris. Dr.Ouattara: Consultancy (money to institution): Endotis, Abbott. Dr. Niesz-kowska: Grants received (money to institution): Pitie–Salpetriere Hospital,Institut National de la Sante et de la Recherche Medicale U943. Dr.Chastre: Grants received (money to institution): French Ministry ofHealth, Department de la Recherche Clinique et du Developpement;Board membership: Pfizer, Astellas, Sanofi-Aventis, Nektar-Bayer; Pay-ment for lectures including service on speakers bureaus: Pfizer, Astellas,Sanofi-Aventis, Nektar-Bayer. Dr. Combes: Grants received (money toinstitution): Programme Hospitalier de Recherche Clinique regional.Disclosures can also be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum�M10-1771.

Reproducible Research Statement: Study protocol: Available at www.reamedpitie.com. Statistical code: Available from Dr. Guiguet (e-mail,[email protected]). Data set: Certain portions of the ana-lytic data set are available to approved individuals through written agree-ments with the authors.

Table 3. Long-Term Outcomes Between Treatment Groups*


ProlongedIntubation (n � 56)

Mean Difference orAbsolute Risk Difference(95% CI)

P Value

Mean BADL Scale score (SD) 5.8 (0.8) 5.9 (0.5) �0.1 (�0.3 to 0.2) 0.52BADL Scale score �6, n (%) 3 (5) 3 (5) �0.4 (�7.0 to 6.0) 0.96Mean IADL Scale score (SD) 7.0 (2.1) 7.0 (2.0) �0.1 (�0.8 to 0.8) 0.90IADL Scale score �8, n (%) 14 (23) 15 (27) �3.5 (�18.5 to 12.0) 0.63Mean SF-36 domains (SD)

Physical functioning 65 (29) 68 (28) �3.2 (�14.3 to 7.9) 0.57Role—physical 74 (40) 71 (40) 2.9 (�12.8 to 15.5) 0.72Bodily pain 77 (29) 73 (28) 4.9 (�6.2 to 16.0) 0.39General health 56 (22) 64 (20) �7.8 (�16.0 to 0.4) 0.06Vitality 52 (22) 56 (20) �4.3 (�12.5 to 3.9) 0.38Social functioning 74 (37) 84 (30) �10.4 (�23.4 to 2.7) 0.12Role—emotional 75 (40) 72 (40) 2.7 (�13.0 to 18.4) 0.73Mental health 71 (21) 71 (21) �0.4 (�8.8 to 7.9) 0.98

Mean SF-36 component score (SD)Physical 45 (8) 45 (9) �0.4 (�3.7 to 2.9) 0.80Mental 48 (10) 49 (11) �1.4 (�5.6 to 2.3) 0.50

Mean HAD scale score (SD) 10.4 (6.6) 10.1 (5.9) 0.4 (�2.1 to 2.8) 0.97Mean HAD subscale A score (SD) 5.7 (3.7) 5.8 (3.8) �0.2 (�1.7 to 1.3) 0.87Mean HAD subscale D score (SD) 4.8 (3.9) 4.2 (3.5) 0.6 (�0.9 to 2.0) 0.45HAD subscale A score �8, n (%) 16 (27) 17 (30) �3.7 (�19.5 to 12.4) 0.96HAD subscale D score �8, n (%) 13 (22) 9 (16) 5.6 (�8.5 to 18.2) 0.22Mean IES score (SD) 5.4 (8.3) 5.0 (7.4) 0.4 (�2.7 to 3.6) 0.78IES score �30, n (%) 1 (2) 1 (2) �0.1 (�2.9 to 2.7) 0.99

BADL � basic activities of daily living; HAD � Hospital Anxiety and Depression; IADL � instrumental activities of daily living; IES � Impact of Event Scale; SF-36 �36-Item Short Form Health Survey.* �90 days after randomization.



Requests for Single Reprints: Jean-Louis Trouillet, MD, Service deReanimation, Institut de Cardiologie, Groupe Hospitalier Pitie–Salpetriere,47, boulevard de l’Hopital, 75651 Paris Cedex 13; e-mail, [email protected].

Current author addresses and author contributions are available atwww.annals.org.

References1. De Leyn P, Bedert L, Delcroix M, Depuydt P, Lauwers G, Sokolov Y, et al;Belgian Association of Pneumology and Belgian Association of CardiothoracicSurgery. Tracheotomy: clinical review and guidelines. Eur J Cardiothorac Surg.2007;32:412-21. [PMID: 17588767]2. Blot F, Melot C; Commission d’Epidemiologie et de Recherche Clinique.Indications, timing, and techniques of tracheostomy in 152 French ICUs. Chest.2005;127:1347-52. [PMID: 15821214]3. Antonelli M, Michetti V, Di Palma A, Conti G, Pennisi MA, Arcangeli A,et al. Percutaneous translaryngeal versus surgical tracheostomy: A randomizedtrial with 1-yr double-blind follow-up. Crit Care Med. 2005;33:1015-20.[PMID: 15891330]4. Higgins KM, Punthakee X. Meta-analysis comparison of open versus percu-taneous tracheostomy. Laryngoscope. 2007;117:447-54. [PMID: 17334304]5. Diehl JL, El Atrous S, Touchard D, Lemaire F, Brochard L. Changes in thework of breathing induced by tracheotomy in ventilator-dependent patients. AmJ Respir Crit Care Med. 1999;159:383-8. [PMID: 9927347]6. Rumbak MJ, Newton M, Truncale T, Schwartz SW, Adams JW, HazardPB. A prospective, randomized, study comparing early percutaneous dilationaltracheotomy to prolonged translaryngeal intubation (delayed tracheotomy) incritically ill medical patients. Crit Care Med. 2004;32:1689-94. [PMID:15286545]7. Nieszkowska A, Combes A, Luyt CE, Ksibi H, Trouillet JL, Gibert C, et al.Impact of tracheotomy on sedative administration, sedation level, and comfort ofmechanically ventilated intensive care unit patients. Crit Care Med. 2005;33:2527-33. [PMID: 16276177]8. Astrachan DI, Kirchner JC, Goodwin WJ Jr. Prolonged intubation vs. tra-cheotomy: complications, practical and psychological considerations. Laryngo-scope. 1988;98:1165-9. [PMID: 3185069]9. Frutos-Vivar F, Esteban A, Apezteguıa C, Anzueto A, Nightingale P, Gonza-lez M, et al; International Mechanical Ventilation Study Group. Outcome ofmechanically ventilated patients who require a tracheostomy. Crit Care Med.2005;33:290-8. [PMID: 15699830]10. Freeman BD, Borecki IB, Coopersmith CM, Buchman TG. Relationshipbetween tracheostomy timing and duration of mechanical ventilation in criticallyill patients. Crit Care Med. 2005;33:2513-20. [PMID: 16276175]11. Griffiths J, Barber VS, Morgan L, Young JD. Systematic review and meta-analysis of studies of the timing of tracheostomy in adult patients undergoingartificial ventilation. BMJ. 2005;330:1243. [PMID: 15901643]12. Kollef MH, Ahrens TS, Shannon W. Clinical predictors and outcomes forpatients requiring tracheostomy in the intensive care unit. Crit Care Med. 1999;27:1714-20. [PMID: 10507588]13. Combes A, Luyt CE, Nieszkowska A, Trouillet JL, Gibert C, Chastre J. Istracheostomy associated with better outcomes for patients requiring long-termmechanical ventilation? Crit Care Med. 2007;35:802-7. [PMID: 17255861]14. Terragni PP, Antonelli M, Fumagalli R, Faggiano C, Berardino M, Pal-lavicini FB, et al. Early vs late tracheotomy for prevention of pneumonia inmechanically ventilated adult ICU patients: a randomized controlled trial. JAMA.2010;303:1483-9. [PMID: 20407057]15. Murthy SC, Arroliga AC, Walts PA, Feng J, Yared JP, Lytle BW, et al.Ventilatory dependency after cardiovascular surgery. J Thorac Cardiovasc Surg.2007;134:484-90. [PMID: 17662794]16. Trouillet JL, Combes A, Vaissier E, Luyt CE, Ouattara A, Pavie A, et al.Prolonged mechanical ventilation after cardiac surgery: outcome and predictors. JThorac Cardiovasc Surg. 2009;138:948-53. [PMID: 19660336]17. Combes A, Costa MA, Trouillet JL, Baudot J, Mokhtari M, Gibert C, et al.Morbidity, mortality, and quality-of-life outcomes of patients requiring � or �14 days of mechanical ventilation. Crit Care Med. 2003;31:1373-81. [PMID:12771605]18. Engoren M, Buderer NF, Zacharias A. Long-term survival and health status

after prolonged mechanical ventilation after cardiac surgery. Crit Care Med.2000;28:2742-9. [PMID: 10966245]19. Ely EW, Baker AM, Dunagan DP, Burke HL, Smith AC, Kelly PT, et al.Effect on the duration of mechanical ventilation of identifying patients capable ofbreathing spontaneously. N Engl J Med. 1996;335:1864-9. [PMID: 8948561]20. Le Gall JR, Lemeshow S, Saulnier F. A new Simplified Acute PhysiologyScore (SAPS II) based on a European/North American multicenter study. JAMA.1993;270:2957-63. [PMID: 8254858]21. Murray JF, Matthay MA, Luce JM, Flick MR. An expanded definition ofthe adult respiratory distress syndrome. Am Rev Respir Dis. 1988;138:720-3.[PMID: 3202424]22. Weinberg PF, Matthay MA, Webster RO, Roskos KV, Goldstein IM, MurrayJF. Biologically active products of complement and acute lung injury in patients withthe sepsis syndrome. Am Rev Respir Dis. 1984;130:791-6. [PMID: 6497161]23. McCabe WR, Jackson GG. Gram-negative bacteremia. Arch Intern Med.1962;110:847-64.24. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method ofclassifying prognostic comorbidity in longitudinal studies: development and val-idation. J Chronic Dis. 1987;40:373-83. [PMID: 3558716]25. Nashef SA, Roques F, Michel P, Gauducheau E, Lemeshow S, Salamon R.European system for cardiac operative risk evaluation (EuroSCORE). Eur J Car-diothorac Surg. 1999;16:9-13. [PMID: 10456395]26. Sessler CN, Gosnell MS, Grap MJ, Brophy GM, O’Neal PV, Keane KA,et al. The Richmond Agitation-Sedation Scale: validity and reliability in adultintensive care unit patients. Am J Respir Crit Care Med. 2002;166:1338-44.[PMID: 12421743]27. Kress JP, Pohlman AS, O’Connor MF, Hall JB. Daily interruption ofsedative infusions in critically ill patients undergoing mechanical ventilation. NEngl J Med. 2000;342:1471-7. [PMID: 10816184]28. Ferreira FL, Bota DP, Bross A, Melot C, Vincent JL. Serial evaluation of theSOFA score to predict outcome in critically ill patients. JAMA. 2001;286:1754-8. [PMID: 11594901]29. Chastre J, Fagon JY. Ventilator-associated pneumonia. Am J Respir CritCare Med. 2002;165:867-903. [PMID: 11934711]30. Chastre J, Wolff M, Fagon JY, Chevret S, Thomas F, Wermert D, et al;PneumA Trial Group. Comparison of 8 vs 15 days of antibiotic therapy forventilator-associated pneumonia in adults: a randomized trial. JAMA. 2003;290:2588-98. [PMID: 14625336]31. Katz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW. Studies of illnessin the aged. The index of ADL: a standardized measure of biological and psycho-social function. JAMA. 1963;185:914-9. [PMID: 14044222]32. Lawton MP, Brody EM. Assessment of older people: self-maintaining andinstrumental activities of daily living. Gerontologist. 1969;9:179-86. [PMID:5349366]33. Leplege A, Ecosse E, Verdier A, Perneger TV. The French SF-36 HealthSurvey: translation, cultural adaptation and preliminary psychometric evaluation.J Clin Epidemiol. 1998;51:1013-23. [PMID: 9817119]34. Leplege A, Ecosse E, Pouchot J, Coste J, Perneger TV. Le QuestionnaireMOS SF-36, Manuel de l’utilisateur et Guide d’interpretation es Scores. Paris:Estem; 2001.35. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. ActaPsychiatr Scand. 1983;67:361-70. [PMID: 6880820]36. Lautrette A, Darmon M, Megarbane B, Joly LM, Chevret S, Adrie C, et al.A communication strategy and brochure for relatives of patients dying in theICU. N Engl J Med. 2007;356:469-78. [PMID: 17267907]37. Horowitz M, Wilner N, Alvarez W. Impact of Event Scale: a measure ofsubjective stress. Psychosom Med. 1979;41:209-18. [PMID: 472086]38. Casagrande JT, Pike MC. An improved approximate formula for calculatingsample sizes for comparing two binomial distributions. Biometrics. 1978;34:483-6. [PMID: 719125]39. Riker RR, Fraser GL. Adverse events associated with sedatives, analgesics, andother drugs that provide patient comfort in the intensive care unit. Pharmaco-therapy. 2005;25:8S-18S. [PMID: 15899744]40. Schweickert WD, Pohlman MC, Pohlman AS, Nigos C, Pawlik AJ, Es-brook CL, et al. Early physical and occupational therapy in mechanically venti-lated, critically ill patients: a randomised controlled trial. Lancet. 2009;373:1874-82. [PMID: 19446324]41. Kress JP, Gehlbach B, Lacy M, Pliskin N, Pohlman AS, Hall JB. Thelong-term psychological effects of daily sedative interruption on critically ill pa-tients. Am J Respir Crit Care Med. 2003;168:1457-61. [PMID: 14525802]



42. Nielsen D, Sellgren J, Ricksten SE. Quality of life after cardiac surgerycomplicated by multiple organ failure. Crit Care Med. 1997;25:52-7. [PMID:8989176]43. Nseir S, Di Pompeo C, Jozefowicz E, Cavestri B, Brisson H, Nyunga M,et al. Relationship between tracheotomy and ventilator-associated pneumonia: acase control study. Eur Respir J. 2007;30:314-20. [PMID: 16870667]44. Rodriguez JL, Steinberg SM, Luchetti FA, Gibbons KJ, Taheri PA, FlintLM. Early tracheostomy for primary airway management in the surgical criticalcare setting. Surgery. 1990;108:655-9. [PMID: 2218876]45. Saffle JR, Morris SE, Edelman L. Early tracheostomy does not improve outcomein burn patients. J Burn Care Rehabil. 2002;23:431-8. [PMID: 12432320]46. Barquist ES, Amortegui J, Hallal A, Giannotti G, Whinney R, Alzamel H,et al. Tracheostomy in ventilator dependent trauma patients: a prospective, ran-domized intention-to-treat study. J Trauma. 2006;60:91-7. [PMID: 16456441]47. Blot F, Similowski T, Trouillet JL, Chardon P, Korach JM, Costa MA,

et al. Early tracheotomy versus prolonged endotracheal intubation in unselectedseverely ill ICU patients. Intensive Care Med. 2008;34:1779-87. [PMID:18592210]48. Bouderka MA, Fakhir B, Bouaggad A, Hmamouchi B, Hamoudi D, HartiA. Early tracheostomy versus prolonged endotracheal intubation in severe headinjury. J Trauma. 2004;57:251-4. [PMID: 15345969]49. Cox CE, Carson SS, Holmes GM, Howard A, Carey TS. Increase in tra-cheostomy for prolonged mechanical ventilation in North Carolina, 1993-2002.Crit Care Med. 2004;32:2219-26. [PMID: 15640633]50. Scales DC, Thiruchelvam D, Kiss A, Redelmeier DA. The effect of trache-ostomy timing during critical illness on long-term survival. Crit Care Med. 2008;36:2547-57. [PMID: 18679113]51. Tracheostomy Management in Critical Care (TracMan). ISRCTN28588190.Accessed at www.controlled-trials.com/mrct/trial/486617/tracman on 19 January2011.



Current Author Addresses: Drs. Trouillet, Luyt, Nieszkowska, Chastre,and Combes: Service de Reanimation, Institut de Cardiologie, GroupeHospitalier Pitie-Salpetriere, 47, boulevard de l’Hopital, 75651 ParisCedex 13, France.Dr. Guiguet: Institut National de la Sante et de la Recherche MedicaleU943 and Universite Pierre et Marie Curie Unite Mixte de Recherche enSante-943, 56 boulevard Vincent Auriol, Paris F75013, France.Dr. Ouattara: Service d’Anesthesie-Reanimation II, Hopital du Haut-Leveque, Avenue Magellan, 33600 Pessac, France.Drs. Vaissier and Makri: Departement d’Anesthesie et Reanimation, In-stitut de Cardiologie, Groupe Hospitalier Pitie-Salpetriere, 47, boulevardde l’Hopital, 75651 Paris Cedex 13, France.Dr. Leprince and Pavie: Service de Chirurgie Thoracique et Cardio-vasculaire, Institut de Cardiologie, Groupe Hospitalier Pitie-Salpetriere,47, boulevard de l’Hopital, 75651 Paris Cedex 13, France.

Author Contributions: Conception and design: J.L. Trouillet, J. Chas-tre, A. Combes.Analysis and interpretation of the data: J.L. Trouillet, M. Guiguet, J.Chastre, A. Combes.Drafting of the article: J.L. Trouillet, M. Guiguet, J. Chastre, A.Combes.Critical revision of the article for important intellectual content: J.L.Trouillet, C.E. Luyt, J. Chastre, A. Combes.Final approval of the article: J.L. Trouillet, C.E. Luyt, M. Guiguet, A.Ouattara, E. Vaissier, R. Makri, P. Leprince, A. Pavie, J. Chastre, A.Combes.Provision of study materials or patients: A. Ouattara, J. Chastre, A.Combes.Statistical expertise: A. Combes.Obtaining of funding: J.L. Trouillet, A. Combes.Administrative, technical, or logistic support: C.E. Luyt, A. Ouattara.Collection and assembly of data: J.L. Trouillet, E. Vaisser, R. Makri, A.Nieszkowska, P. Leprince, A. Pavie, J. Chastre, A. Combes.

52. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. Prognosis in acuteorgan-system failure. Ann Surg. 1985;202:685-93. [PMID: 4073980]

APPENDIX

MethodsMechanical Ventilation Weaning Protocol

For both groups, mechanical ventilation weaning was con-ducted according to a strict protocol, beginning as early as thefirst day after randomization. Weaning criteria were defined asregression or resolution of the underlying cause of acute respira-tory failure; oxygen saturation greater than 90%, with a fractionof inspired oxygen (FIO2) less than 40% and a positive end-expiratory pressure of 5 cm H2O or more; correction of electro-lyte anomalies; a good level of consciousness; no further need forhigh doses of vasoactive and sedative agents, and respiratory fre-quency to tidal volume ratio less than 105 after 1 minute ofspontaneous breathing (19). Patients fulfilling these criteria had a1-hour spontaneous breathing trial (T-piece trial), and those whosuccessfully passed the trial were extubated within 6 hours ortemporarily left with the tracheotomy cannula for 24 or 48 hourswithout mechanical ventilation before decannulation. If they ful-filled clinical poor tolerance criteria (respiratory rate �35breaths/min or increased by 50%, use of accessory respiratorymuscles, diaphoresis, heart rate, or blood pressure increasedby �20% or altered consciousness), confirmed by blood gas anal-

ysis, assist control ventilation was reintroduced. T-piece trialswere repeated every day until extubation or decannulation. Whena patient developed signs of respiratory failure after extubation ordecannulation (respiratory rate �30 breaths/min, oxygen satura-tion as measured by pulse oximetry �90%, or hypercapnia),noninvasive mechanical ventilation was applied by using the pres-sure support method for 1- to 2-hour sessions.

Outcomes and Follow-upVentilator-free days were counted from the last day that a

patient received mechanical ventilation during the 60-day period.In particular, patients who did not receive mechanical ventilationand died before day 60 were assigned 0 ventilator-free days, andeach day a patient required noninvasive mechanical ventilationfor more than 4 hours after extubation or decannulation wascounted as 1 mechanical ventilation day, therefore resettingventilator-free days at 0. Sensitivity analyses with 3 other defini-tions of ventilator-free days were also done. Definition 2: count-ing ventilator-free days from the last day that a patient receivedmechanical ventilation and assigning ventilator-free days whendeath occurred before day 28, 60, or 90. Definition 3: using theexact numbers of days on mechanical ventilation to calculateventilator-free days and assigning 0 ventilator-free days whendeath occurred before day 28, 60, or 90. Definition 4: using theexact numbers of days on mechanical ventilation to calculateventilator-free days and assigning ventilator-free days when deathoccurred before day 28, 60, or 90.

Ease of Care, Patient Comfort, and Muscle StrengthAssessment

Nurses evaluated the ease of care daily by using a subjectivescore specifically designed for this trial. Results are expressed ac-cording to a numerical scale ranging from very easy (score � 1)to very difficult (score � 4). Two ease-of-care categories weredefined a posteriori: easy (score � 1 or 2) and difficult (score �3 or 4). Nurses also assessed the patient’s comfort with a similarscale specifically designed for this trial ranging from very com-fortable (score � 1) to very uncomfortable (score � 4). Similarly,2 patient comfort categories were defined a posteriori: comfort-able (score � 1 or 2) and uncomfortable (score � 3 or 4). Inaddition, durations of enteral feeding through a nasogastric tube,central venous catheter, and bladder catheterization and the timesto the first bed-to-chair transfer, the first oral nutrition, definitiveurinary catheter removal, and central venous catheter removalwere recorded.

Muscle strength was evaluated by using the Medical Re-search Council score. A trained physiotherapist assessed musclegroups of the upper (shoulder, elbow, wrist, and hand) and lower(hip, knee, and ankle) limbs. Each muscle group score rangesfrom 0 (paralysis) to 5 (normal muscle strength). The globalscore, obtained by adding all the muscle group scores, rangesfrom 0 to 280. In an attempt to limit interexaminer variability,only 1 investigator (the same qualified physiotherapist) made85% of the assessments. This score was determined the day ofinclusion and then every 14 days until the ICU discharge.

Annals of Internal Medicine

www.annals.org 15 March 2011 Annals of Internal Medicine Volume 154 • Number 6 W-121

Adverse EventsLate laryngeal and tracheal complications were systemati-

cally evaluated during fibroscopy by the intensivist 1 to 5 daysafter extubation or decannulation for the first 29 patients. For thesubsequent patients, this exploration was done only when un-usual clinical symptoms or an evident complication developedand completed with an examination by an ear, nose, and throatspecialist, when necessary. All serious adverse events not relatedto tracheotomy were also recorded.

Long-Term Follow-up, HRQoL, and Psychosocial EvaluationsA cross-sectional study of long-term outcomes, designed a

posteriori, was conducted from May 2010 to June 2010 on allpatients who were alive on day 90. After explaining the purposeof the study and asking for informed consent during a telephonecall, we administered the following questionnaires to survivors.

First, activities of daily living were assessed with the basicand instrumental scales (31, 32). The basic scale assesses the levelof difficulty and receipt of assistance with bathing, dressing, toi-leting, getting in or out of bed or chairs, controlling bowel andbladder continence, and eating, and its total score ranges from 0(major) to 6 (no) disability. The instrumental scale contains 8questions (ability to use the telephone, shop, prepare food, main-tain a home, do laundry, drive a car or take a bus or a taxi, beresponsible for own medications, and handle finances), and itstotal score ranges from 0 (major) to 8 (no) disability. Then,HRQoL was assessed with the French version of the SF-36 (33).Its 36 items are combined to evaluate 8 domains (physical func-tioning, role—physical, bodily pain, general health, vitality, so-cial functioning, role—emotional, and mental health). The ag-gregate physical and mental component summary measures werethen computed, as recommended by developers. Individual com-ponent and overall physical and emotional domain scores rangefrom 0 (poor) to 100 (excellent). Our patients’ mean SF-36 levels

were compared with age and sex-matched French populationnormative values (34).

Anxiety and depression symptoms were assessed with theHospital Anxiety and Depression scale (35), which contains 14questions—7 to assess anxiety (subscale A) and 7 to assess depres-sion (subscale D)—and uses a 4-point Likert scale (range, 0 to 3),giving a possible score of 0 (none) to 21 (severe) for each of the2 subscales. Subscale scores of 8 or more indicated clinically sig-nificant anxiety or depression (35, 36).

Symptoms related to posttraumatic stress disorder were as-sessed with the Impact of Event Scale (37), which includes 15questions divided into 2 subscales: intrusion (7 items) and avoid-ance (8 items). The total score ranges from 0 (no symptoms) to75 (severe symptoms). In agreement with previous reports, pa-tients with a total score of 30 points or more were considered athigh risk for symptoms of posttraumatic stress disorder (36).

Results: Adverse EventsOnly 1 major complication related to tracheotomy (that is,

perforation of the posterior tracheal wall) occurred in the earlypercutaneous tracheotomy group. The procedure was stopped;the patient was temporarily left intubated; and a tracheotomy wasdone the next day by an ear, nose, and throat surgeon. Alterna-tively, a large tracheal tear during the initial intubation for car-diac surgery was diagnosed in the prolonged intubation group 5days after randomization. The endotracheal tube was left inplace, and a percutaneous tracheotomy was done on day 16 afterrandomization without any subsequent complication. Both pa-tients with initial tracheal complications had favorable local out-comes and were alive on day 90. Only 1 stomal infection and nosignificant hemorrhage were related to the tracheotomy proce-dure. Laryngeal symptoms (for example, swallowing disorders ordysphonia) and abnormalities at laryngeal examination were rareand similar for the 2 groups. In addition, the numbers and se-verities of other adverse events did not differ between the 2groups (Appendix Table 4).

W-122 15 March 2011 Annals of Internal Medicine Volume 154 • Number 6 www.annals.org

Appendix Table 1. Baseline Characteristics of Participants

Characteristic Early PercutaneousTracheotomy (n � 109)

Prolonged Intubation(n � 107)

Preoperative, n (%)Smoker 28 (26) 27 (25)Hypertension 54 (50) 58 (54)History of myocardial infarction 48 (44) 45 (42)Leg arteriopathy 16 (15) 17 (16)Cerebrovascular event 14 (13) 10 (9)Chronic respiratory insufficiency 12 (11) 12 (11)Diabetes mellitus 26 (24) 27 (25)Required insulin therapy at home 10 (9) 8 (7)Chronic renal insufficiency 17 (16) 8 (7)Hepatic insufficiency 1 (1) 0McCabe–Jackson score of 3 84 (77) 76 (71)NYHA classification

I 18 (17) 19 (18)II 25 (23) 31 (29)III 36 (33) 31 (29)IV 30 (28) 26 (24)

On MV (�48 h) before surgery 12 (11) 10 (9)

PerioperativeMean aortic clamping duration (SD), min 84.1 (51.9) 91.1 (44.8)IABP, n (%) 8 (7) 13 (12)ECMO, n (%) 16 (15) 21 (20)Postoperative extubation, n (%) 26 (24) 40 (37)Mean cardiac troponin I levels on postoperative day 1 (SD), �g/L 25.9 (73.5) 20.0 (38.8)Reintervention, n (%) 16 (15) 21 (20)Patients requiring blood transfusion, n (%) 75 (69) 67 (63)Mean blood units transfused (SD) 3.2 (3.7) 2.8 (3.4)Severe immediate complications, n

Cardiac arrest 1 5Cerebral injury 2 0Pneumothorax 0 0Other 5 7

At randomizationMean PaO2–FIO2 ratio (SD) 209.9 (100.6) 192.4 (86.1)Mean platelet count (SD), � 109 cells/L 111.0 (61.8) 127.0 (79.1)Mean bilirubin level (SD)

�mol/L 44.8 (51.2) 42.1 (51.8)mg/dL 2.14 (1.11) 1.94 (1.11)

Mean daily diuresis (SD), L 1865 (1733) 1824 (1499)Mean OSF score (SD) 2.5 (0.8) 2.3 (0.7)

Respiratory, n (%) 99 (91) 98 (92)Circulatory, n (%) 109 (100) 107 (100)Neurologic, n (%) 9 (8) 7 (7)Renal, n (%) 46 (42) 24 (22)Hematologic, n (%) 4 (4) 3 (3)Hepatic, n (%) 6 (6) 6 (6)

Mean radiologic score (SD) 5.0 (2.7) 5.0 (2.6)Renal replacement therapy, n (%) 41 (38) 18 (17)Mean duration of MV (SD), h 73.9 (18.1) 66.6 (20.6)

ECMO � extracorporeal membrane oxygenation; IABP � intra-aortic balloon pump; MV � mechanical ventilation; NYHA � New York Heart Association; OSF � organsystem failure (52).


Appendix Table 2. Main Outcome Variables



Mean Difference orAbsolute RiskDifference (95% CI)

P Value

Mean SOFA score (SD)*1 d (n � 109, 107) 11.6 (3.5) 10.9 (3.6) 0.6 (�0.3 to 1.6) 0.197 d (n � 102, 93) 6.1 (4.5) 6.7 (4.1) �0.5 (�1.7 to 0.7) 0.4114 d (n � 71, 63) 4.5 (3.7) 6.1 (4.2) �1.6 (�3.0 to �0.3) 0.0221 d (n � 48, 50) 4.4 (4.0) 5.0 (3.9) �0.6 (�2.1 to 1.0) 0.4828 d (n � 30, 34) 3.6 (3.3) 3.9 (3.1) �0.3 (�1.9 to 1.3) 0.73

Patients given epinephrine or norepinephrine (1–60 d), n (%) 98 (90) 101 (94) �4.5 (�11.6 to 2.7) 0.31Mean days with epinephrine or norepinephrine (SD)† 7.4 (7.5) 9.3 (9.0) �1.9 (�4.2 to 0.4) 0.11Patients given renal replacement therapy (1–60 d), n (%)‡ 61 (56) 46 (43) 13.0 (�0.2 to 26.2) 0.06Mean days with renal replacement therapy (SD)† 4.9 (8.4) 4.6 (8.8) 0.3 (�3.9 to 4.2) 0.82Received antibiotics, n (%) 107 (98) 106 (99) �0.9 (�4.0 to 2.2) 1.0Duration of antibiotic treatment (SD), d 14.3 (11.6) 15.0 (13.3) �0.6 (�4.0 to 2.7) 0.70Mean days with nasogastric tube (SD) 14.7 (12.9) 17.7 (14.4) �3.0 (�6.7 to 0.6) 0.10Mean days with bladder catheter (SD) 21.1 (16.0) 22.3 (16.2) �1.2 (�5.6 to 3.0) 0.56Mean days with central venous catheter (SD) 16.1 (13.3) 17.3 (12.5) �1.2 (�4.7 to 2.2) 0.48Mean days until first oral nutrition (SD)† 5.7 (4.5) 11.1 (7.5) �5.4 (�7.1 to �3.6) �0.001Mean days until first bed-to-chair transfer (SD)† 10.8 (8.5) 13.3 (9.3) �2.5 (�5.2 to 0.3) 0.08

SOFA � Sequential Organ Function Assessment.* Numbers reported with each time point are patients in the early percutaneous tracheotomy group and the prolonged intubation group, respectively.† Mean number of days among patients with the event.‡ At randomization, 41 patients in the early percutaneous tracheotomy group and 18 patients in the prolonged intubation group had renal replacement therapy; 20 and 28additional patients in the respective groups required renal replacement therapy after randomization.

Appendix Table 3. Sensitivity Analyses for VFD Calculation

Mean VFDs (SD) Early PercutaneousTracheotomy(n � 109)



P Value

VFD definition 2*1–60 d 32.1 (21.5) 29.1 (23.1) 3.0 (�2.9 to 9.0) 0.321–28 d 10.0 (8.8) 9.4 (10.1) 0.6 (�1.9 to 3.2) 0.621–90 d 52.2 (34.0) 48.9 (35.4) 3.3 (�6.0 to 12.6) 0.48

VFD definition 3†1–60 d 30.9 (22.5) 28.8 (23.7) 2.1 (�4.1 to 8.3) 0.501–28 d 10.4 (8.7) 9.5 (10.2) 0.9 (�1.7 to 3.4) 0.501–90 d 50.8 (35.6) 49.3 (35.6) 1.5 (�8.0 to 11.1) 0.75

VFD definition 4‡1–60 d 32.6 (21.3) 30.0 (22.7) 2.7 (�3.2 to 8.6) 0.371–28 d 10.4 (8.7) 9.9 (10.0) 0.5 (�2.0 to 3.0) 0.691–90 d 54.1 (32.4) 51.1 (33.8) 3.0 (�5.8 to 11.9) 0.50

VFD � ventilator-free day.* VFD definition 2 is counting VFDs from the last day that a patient received mechanical ventilation and assigning VFDs when death occurred before 28, 60, or 90 d.† VFD definition 3 is using the exact numbers of days on mechanical ventilation to calculate VFDs and assigning 0 VFDs when death occurred before 28, 60, or 90 d.‡ VFD definition 4 is using the exact numbers of days on mechanical ventilation to calculate VFDs and assigning VFDs when death occurred before 28, 60, or 90 d.


Appendix Table 4. Effect of Early Percutaneous Tracheotomy Versus Prolonged Intubation: Crude Findings and Analyses Adjustedto Repeated Cardiac Surgery, Heart Transplantation, and Renal Replacement Therapy at Randomization

Variable Crude Mean Difference orCrude Odds Ratio (95% CI)

Adjusted Mean Difference orAdjusted Odds Ratio (95% CI)

Mean VFDs during 1–60 d (SD) 2.1 (�4.1 to 8.3) 4.8 (�1.4 to 11.1)Mean VFDs during 1–28 d (SD) 0.8 (�1.7 to 3.4) 1.4 (�1.2 to 4.0)Mean VFDs during 1–90 d (SD) 1.8 (�8.0 to 11.6) 5.9 (�4.1 to 15.9)Mortality

28 d 0.67 (0.34 to 1.35) 0.58 (0.27 to 1.25)60 d 0.88 (0.48 to 1.62) 0.71 (0.36 to 1.37)90 d 1.02 (0.57 to 1.82) 0.84 (0.44 to 1.57)

Mean days on MV before 60 d (SD) �1.3 (�5.6 to 3.0) �2.0 (�6.5 to 2.6)Mean days in ICU �1.5 (�7.4 to 4.3) �3.4 (�9.6 to 2.7)Mean days in the hospital 1.5 (�5.7 to 8.8) �0.02 (�7.8 to 7.7)Mean endotracheal prosthesis–free days during 1–60 d 1.7 (�4.6 to 7.9) 4.4 (�1.9 to 10.7)

ICU � intensive care unit; MV � mechanical ventilation; VFD � ventilation-free day.


Appendix Figure 1. Hours per day spent heavily sedated or calm, awake, or lightly sedated during the 15 days after randomization.

Tim

e C

alm

, Aw

ake,

or

Ligh

tly

Seda

ted,

h/d

Day

Early percutaneous tracheotomy

Prolonged intubation

1 2 3 4 5 6 7 8 9 10 11 12 13 14 150

8

4

20

16

12

24

Tim

e H

eavi

ly S

edat

ed, h

/d

0

12

8

4

20

16

24

The early percutaneous tracheotomy group spent less time heavily sedated (P � 0.03) and more time calm, awake, or lightly sedated (P � 0.001) thanthe prolonged intubation group. The bars indicate 95% CIs.


Appendix Table 5. Adverse Events Occurring After Randomization




P Value

Cardiovascular, n (%)Dysrhythmia 14 (13) 14 (13) �0.2 (�9.2 to 8.7) 1.0Shock* 11 (10) 10 (9) 0.7 (�7.1 to 8.6) 1.0Pericardial effusion 5 (5) 7 (7) �1.9 (�8.1 to 4.2) 0.57Hemorrhagic shock 3 (3) 3 (3) �0.1 (�4.4 to 4.3) 1.0Cardiac arrest 0 4 (4) �3.7 (�7.3 to �0.1) 0.06Venous thromboembolism 1 (1) 1 (1) 0 (�2.6 to 2.5) 1.0

Pulmonary, n (%)VAP after randomization 50 (46) 47 (44) 2.0 (�11.3 to 15.2) 0.77Patients with unscheduled extubation or decannulation 3 (3) 17 (16) �13.1 (�20.7 to �5.6) �0.001Patients with reintubation or recannulation 17 (16) 35 (33) �17.1 (�28.3 to �5.9) 0.004Pleural effusion 6 (6) 7 (7) �1.0 (�7.4 to 5.3) 0.78Pneumothorax 2 (2) 6 (6) �3.8 (�8.8 to 1.3) 0.17ARDS 3 (3) 7 (7) �3.8 (�9.4 to 1.8) 0.21Atelectasis 3 (3) 8 (8) �4.7 (�10.5 to 1.1) 0.13Diaphragmatic dysfunction 3 (3) 6 (6) �2.9 (�8.2 to 2.5) 0.33

Infectious, n (%)Sternal wound 14 (13) 14 (13) �0.2 (�9.2 to 8.7) 0.96Bloodstream 18 (17) 16 (15) 1.5 (�8.1 to 11.3) 0.85

Neurologic, n (%)Stroke 10 (9) 11 (10) �1.1 (�9.0 to 6.8) 0.82Seizure 1 (1) 2 (2) �1.0 (�4.1 to 2.2) 0.62

Gastrointestinal, n (%)Gastrointestinal bleeding 5 (5) 4 (4) 0.9 (�4.5 to 6.2) 1.0Intestinal ischemia 0 3 (3) �2.8 (�5.9 to 0.3) 0.12Pseudomembranous colitis 2 (2) 1 (1) 0.9 (�2.2 to 4.0) 1.0

Hematologic, n (%)Thrombocytopenia 10 (9) 7 (7) 2.6 (�4.5 to 9.8) 0.61Neutropenia, pancytopenia 4 (4) 1 (1) 2.7 (�1.2 to 6.7) 0.37

Any adverse event, n (%) 89 (82) 91 (85) �3.4 (�13.3 to 6.5) 0.58

ARDS � acute respiratory distress syndrome; VAP � ventilator-associated pneumonia.* Other than hemorrhagic shock.


Appendix Figure 2. Comparison of SF-36 scores obtained for early percutaneous tracheotomy and prolonged intubation inlong-term survivors with age- and sex-matched French population normative values.

SF-3

6 Sc

ore

*

Early percutaneous tracheotomy

Prolonged intubation

Normative values

PhysicalFunctioning

Role—Physical

BodilyPain

GeneralHealth

Vitality SocialFunctioning

Role—Emotional

MentalHealth

0

60

70

30

40

50

80

90

100*

The bars indicate 95% CIs. SF-36 � 36-Item Short Form Health Survey.* P � 0.05 for early percutaneous tracheotomy and prolonged intubation vs. French population normative values.


early tracheotomy vs prolonged intubation after cardiac surgery

Documents