Matthias HaenggiStephan M. JakobJukka Takala

Sedation level and prevalenceof delirium: responseto Brummel and Ely

Accepted: 28 August 2013Published online: 6 September 2013� Springer-Verlag Berlin Heidelberg andESICM 2013

This reply refers to the comment availableat: doi:10.1007/s00134-013-3083-9.

Dear Editor,We thank Dr. Brummel and his team[1] for their interest in our paper [2].We agree that the prevalence ofdelirium in awake ICU patients in ourstudy is similar to that found in earlierstudies (42.5 % in the study by Elyet al. [3], 31 % in our study). How-ever, we propose that reports ondelirium should be stratified for asedation score during the assessment,because the apparent prevalence ofdelirium is dependent on how adepressed level of consciousness aftersedation stop is interpreted. Forinstance, of 471 daily evaluationscompleted during the study by Elyet al. [3], a diagnosis by the referencestandard of delirium was made in25.2 %, stupor in 21.3 %, and comain 28.5 % of all observations. Yet, theauthors report that delirium occurredin 83.3 % of their patients. Unfortu-nately, in the validation studiesneither CAM-ICU nor ICDSCdescribed how sedated patients werehandled.

The new DSM-V criteria fordelirium published recently indicatethat non-comatose patients with areduced level of arousal of acuteonset should be classified as having

severe inattention and cognitivechange, and hence delirium. Weagree that such patients can be delir-ious. Nevertheless, it should be notedthat the same criteria—which corre-spond to a RASS-2/-3 level—can alsoapply for sedated healthy volunteers,and patients with uneventful postop-erative course during awakening fromanesthesia.

By no means do we suggest thatdelirium should not be considered insedated patients. However, we ques-tion the specificity of the assessmentstools CAM-ICU and ICDSC in seda-ted patients who cannot hold eyecontact for 10 s. In contrast to whathas been proposed for the CAM-ICU[4], we suggest to use the deliriumassessment tools after sedation stop,and to indicate the level of con-sciousness during the assessment. Anexclusive diagnosis of delirium isimpossible before effects of sedationare ruled out. Regular sedation stopshave been proven to improve out-come and reduce resource use.

We should make every effort not tomiss a diagnosis of delirium. At thesame time, misdiagnosing persistentsedation as delirium is dangerous aswell: a recent delirium trial was pre-maturely stopped because ofincreased mortality in the treatmentgroup [5]. In this study, almost 40 %of the included patients had a nega-tive RASS, half of them a RASS of-2. In our view, sedation should bemanaged with a protocolizedapproach including a sedation goaland frequent assessments, and notwith a delirium screening tool.

Conflicts of interest The Department ofIntensive Care Medicine, Bern UniversityHospital has, or has had in the past, researchcontracts with Abbott Nutrition Interna-tional, B. Braun Medical AG, CSEM SA,Edwards Lifesciences Services GmbH,Kenta Biotech Ltd, Maquet Critical CareAB, Omnicare Clinical Research AG, andOrion Pharma; and research and develop-ment/consulting contracts with EdwardsLifesciences SA, Maquet Critical Care AB,

and Nestle. The money is/was paid into adepartmental fund; neither MH nor SMJ orJT have received any personal financialgain. The department has received unre-stricted educational grants from thefollowing organizations for organizing aquarterly postgraduate educational sympo-sium, the Berner Forum for Intensive Care:Fresenius Kabi, GlaxoSmithKline, MerckSharp & Dohme, Lilly, Baxter, Astellas,AstraZeneca, B. Braun, CSL Behring, Ma-quet, Novartis, Covidien, Mycomed, andRobaPharma.

References

1. Brummel NE, Ely EW (2013) Sedationlevel and the prevalence of delirium.Intensive Care Med. doi:10.1007/s00134-013-3083-9

2. Haenggi M, Blum S, Brechbuehl R,Brunello A, Jakob SM, Takala J (2013)Effect of sedation level on the prevalenceof delirium when assessed with CAM-ICU and ICDSC. Intensive Care Med.doi:10.1007/s00134-013-3034-5

3. Ely EW, Inouye SK, Bernard GR,Gordon S, Francis J, May L, Truman B,Speroff T, Gautam S, Margolin R, HartRP, Dittus R (2001) Delirium inmechanically ventilated patients: validityand reliability of the confusionassessment method for the intensive careunit (CAM-ICU). JAMA286:2703–2710. doi:11730446

4. Ely EW (2010) Confusion assessmentmethod for the ICU (CAM-ICU) Thecomplete training manual.http://www.mc.vanderbilt.edu/icudelirium/docs/CAM_ICU_training.pdf. Accessed 15Mar 2013

5. van Eijk MM, Roes KC, Honing ML,Kuiper MA, Karakus A, van der Jagt M,Spronk PE, van Gool WA, van der MastRC, Kesecioglu J, Slooter AJ (2010)Effect of rivastigmine as an adjunct tousual care with haloperidol on durationof delirium and mortality in critically illpatients: a multicentre, double-blind,placebo-controlled randomised trial.Lancet 376:1829–1837

M. Haenggi ()) � S. M. Jakob � J. TakalaDepartment of Intensive Care Medicine,Bern University Hospital, Inselspital,Bern, Switzerlande-mail: matthias.haenggi@insel.chTel.: ?41-31-6323029

Intensive Care Med (2014) 40:136DOI 10.1007/s00134-013-3092-8 CORRESPONDENCE