+ Direct Access
July 28, 2015

Intensive Care Medicine journal




Intensive care medicine’s natural short term perspective in the acute treatment phase has often been reflected in the time-span chosen for follow-up studies, with most studies using up to one year as the follow-up time but not more. This study, with it’s 5-year follow-up span adds an important component to the scarce data written about patients quality of life over a longer period of time. 

Hofhuis et al. [1] performed a prospective study between 2000 and 2007 of all patients admitted for more than 48 hours in a Dutch mixed-ICU. Patients were evaluated with the SF-36 questionnaire on admission (albeit mostly proxies answered), on ICU discharge, hospital discharge, and after 1, 2 and 5 years. No other exclusion criteria than less than 48 hours admission and unability to communicate or impaired self-awareness were used. Of 3775 screened patients, 749 were included in the first part and thereafter a natural decline in the number of participants followed (378 followed-up after 1 year, 301 after 2 years and 234 after 5 years). Analysis was not only performed within the group but also with an age-matched control group in the general population, although this was only made using the global question in the SF-36; ”In general would you say your health is excellent, very good, good, fair or poor?”


The main finding from this long term study was that although the effect size is small, even after 5 years, ICU survivors are not back to their pre-admission HRQOL (in the physical functioning, social functioning and general health dimensions). Additionally, investigators found a significant lower pre-admission vitality and mental health score in 5-year survivors compared with a general population suggesting that the ICU patient is not fully representative of the general population. Also, perhaps not surprisingly, the pre-admission score of 5-year survivors were significantly higher than in non-survivors.

Study Strengths and Limitations

This is a nicely performed prospective study which adds a number of missing pieces to the puzzle of long term HRQOL after ICU-care. The 48 hour cut-off filters out low-risk patients and makes the study relatively generalisable for the group of patients with the highest risk of post-ICU syndrome. Also, in their analysis, Hofhuis et al. have compensated for natural decline in HRQOL using the natural decline of the Dutch normal population. Finally, the authors should be applauded for a thorough analysis of the missing data in the supplementary material, indicating the amount of work they have put into the extensive follow-up.

The main limitation of the study would be the use of proxies for the premorbid HRQOL. There are but a few studies performed regarding the accuracy with divergent results, but this is a dilemma not only for this study but for all critical care quality-of-life studies. Also, the 7 years inclusion-period could be regarded as an advantage as well as a drawback, considering the period’s development in sedatives, sedation protocols etc. Lastly, although the SF-36 might provide a convenient way of collecting HRQOL data, the results are difficult to translate into the actual burden for the individual patient. To this day though, no better instrument exists.


The differences in HRQOL between pre-admission and 5-years follow-up are small enough to question whether an even longer follow-up is meaningful or even feasible. The trajectories in HRQOL shown in this study suggests that a longer follow-up time than 5 years might be of little value. The study beautifully demonstrates a horizon in time on the post-ICU symptoms and evokes the question on what factors differentiates the “non-recoverers” from the rest.

This article review was submitted by ESICM Journal Review Club member Johan A. Malmgren on behalf of the Ethics section.


1.    Hofhuis et al. ICU survivors show no decline in health-related quality of life after 5 years. Intensive Care Med (2015) 41:495-504 DOI 10.1007/s00134-015-3669-5

Comment on this news