Daily Sedation Vacation’: Long-Term Adverse Effects?

Abstract & Commentary

Synopsis: This single-center study reveals that, compared to continuous sedation, daily sedative interruption is not associated with adverse psychological effects after 6-21 months.

Source: Kress JP, et al. Am J Respir Crit Care Med. 2003;168:1457-1461.

Effects of sedation on long-term psychological functioning of critically ill patients are not well described. Although there is good evidence that daily sedation interruption (the "sedation vacation") offers considerable short-term benefits when compared to continuous sedation,1 there are concerns that this may negatively effect long-term psychological outcomes.2 Kress and associates set out to find evidence that daily sedative interruption may be associated with long-term psychological harm. A secondary aim was to determine whether daily sedative interruption may instead be associated with improved long-term psychological outcomes.

The study was conducted at the University of Chicago medical ICU. Study subjects included participants from a prior randomized, controlled trial of daily sedative interruption (intervention) vs continuous sedation (control)1 plus contemporaneous subjects on the same protocol but not in that study. Demographics, severity of illness, hospital and ICU lengths of stay, mechanical ventilation duration, new and prior diagnoses (including psychiatric), and related treatments were recorded. Patients were contacted at least 6 months after discharge. Those who agreed to participate were evaluated by clinical psychologists who were blinded to the sedation protocol used. Evaluation included a structured interview, the Impact of Events Scale (IES, measuring signs of post-traumatic stress disorder [PTSD]), consideration of PTSD diagnosis according to DSM-IV criteria, the Medical Outcomes Study 36 item short-form health survey (SF-36, assessing overall perception of health and well-being), standard anxiety and depression questionnaires (State-Trait Anxiety Inventory [STAI] and Beck Depression Inventory-2 [Beck]), and the Psychological Adjustment to Illness score (PAIS).

Of the patients who were screened, 105 survived to hospital discharge. Only 35 (30%) could be contacted, and of these, 32 (91%) agreed to participate. There were 19 control and 13 intervention subjects, evaluated at 6-21 months post-discharge. As expected, there was a trend toward decreased ICU and hospital lengths of stay in the intervention group but otherwise no significant differences in baseline characteristics. Most patients (69%) in both groups remembered being in the ICU; none in the intervention arm recalled awakening there. The intervention group had significantly better IES scores, and none had PTSD per DSM-IV criteria; 6 patients in the control arm had PTSD. There were no significant differences between the 2 groups in terms of SF-36, STAI, Beck, or PAIS scores, although in almost every case the trend was toward better scores in the intervention group.

Kress et al conclude that daily sedative interruption is not psychologically harmful to patients in the long term. They highlight the trend toward decreased PTSD in the intervention group and suggest that daily sedative interruption may improve some psychological outcomes.

Comment by Saadia R. Akhtar, MD, MSc

Kress et al have produced a novel and valuable report on a topic that deserves much greater attention. This is a long-term follow-up of a small cohort of subjects from a randomized, controlled trial of 2 sedation protocols (continuous vs daily interruption) combined with non-randomized subjects on the same protocol. Kress et al are concise but thoughtful and complete in their presentation. They make reasonable attempts to address the inherent limitations of this study design and are quite clear and appropriately cautious in their discussion and conclusions. This is a well-done and well-written important pilot study.

The study has some limitations due to its size and methodology, as Kress et al point out nicely in their discussion. For instance, the lack of difference in psychological outcome between the control and intervention arms could be due to the small study sample size: this greatly limits the power of the study to detect differences. The differences in duration of mechanical ventilation and length of ICU stay, key and significant short-term outcomes in the original daily sedative interruption study, were not statistically significant in this smaller cohort. Thus, we must also interpret psychological outcomes data with some caution. Furthermore, follow-up/recruitment was very low at 30%, and the final study patients are a mix of 2 separate cohorts (one from a randomized, controlled study). Kress et al did compare known baseline characteristics of the cohorts, as well as those of recruited subjects vs those deceased or lost to follow-up and did not find significant differences; however, considerable unmeasured differences may still exist and may greatly bias the results. For example, there are no available data on baseline psychiatric diagnoses, subsequent medical or psychiatric diagnoses and related treatments, or reason for death or loss to follow-up of the unavailable subjects.

Taking these issues into account, Kress et al’s results still represent the first reasonable evidence of relative safety of daily sedative interruption with respect to long-term psychological outcomes. They have set the groundwork for this area of research.

However, it is not enough to stop at equivalence to current outcomes, as it is clear that mental health and quality of life post-ICU stay is quite poor for many patients.3,4 For this reason, Kress et al’s observations of fewer PTSD symptoms and diagnoses, as well as a trend toward improved results in other psychological measures in the intervention group, are particularly intriguing and require further study and clarification. Is daily sedative interruption actually better and less psychologically stressful for patients? If so, what is the underlying mechanism? Does it relate to the sedative agent or dose, the medical staff attention given in order to conduct daily sedative interruption, the reduced time of ventilation or ICU stay, the underlying disease processes leading to ICU admission, the differences in physiology of sleep that may occur with interrupted vs continuous sedation, changes in memory (conscious or unconscious) of the ICU experience, or other factors? Additional aspects of long-term mental health such as neuropsychological functioning also must be examined in future studies.5

With continued investigation, we may someday achieve the ultimate goal of providing truly adequate and appropriate sedation at the lowest-effective doses and without adverse outcomes for all patients. For now, we have some good evidence to guide us. All medical ICUs should be implementing daily sedative interruption with the knowledge that it improves short-term outcomes and appears to be equivalent to continuous sedation in terms of patients’ long-term psychological functioning.


1. Kress JP, et al. N Engl J Med. 2000:342:1471-1477.
2. Heffner JE. N Engl J Med. 2000:342:520-1522.
3. Nelson BJ, et al. Crit Care Med. 2000:28:3626-3630.
4. Herridge MS, et al. N Engl J Med. 2003:348:683-693.
5. Jackson JC, et al. Crit Care Med. 2003:31:1226-1234.

Saadia R. Akhtar, MD, MSc, Pulmonary and Critical Care Medicine, Yale University School of Medicine, is Associate Editor of Critical Care Alert.