What were you able to do in your daily life? Performance status for the critically ill patient
Mortality, which is the most robust outcome in critically ill patients, depends on risk factors such as comorbid conditions, age and severity of the acute illness. Whilst these risk factors are widely accepted in severity scores for predicting hospital mortality, there is a lack of information about the baseline performance status (PS) of the patients when being admitted to the intensive care unit (ICU). To allow an adequate clinical decision-making process and further discussion with patients or relatives, accurate prognostication of a patient’s outcome seems to be mandatory, however difficult and emotionally charged the task is for the physician.
Zampieri et al., in an article recently published in Intensive Care Medicine, add some insight into the value of PS in the outcome . This is a secondary analysis of a multicentre retrospective cohort study of 59,693 patients admitted to 78 Brazilian ICUs in 2013 . The authors  aimed to evaluate the impact of PS, assessed by the Eastern Cooperative Oncology Group (ECOG) scale, on the hospital mortality of 59,693 ICU patients. PS impairment was absent/minor in 75.8 %, moderate in 17.3 % and severe in 6.9 % of the patients. Increasing PS impairment was associated with an increase of mortality according to severity of illness, comorbidities, age, and admission type. The logistic regression analysis showed that PS was independently associated with hospital mortality. Adding the PS assessment to the calculated SAPS3 score slightly improved the discrimination, with small changes in Bier score and without visual impact on the SAPS3 calibration curve. This may happen because the short-term prognosis after intensive care is most likely the result of the interplay between illness severity, baseline patient characteristics (comorbidities, PS) and the quality of care. As Zampieri et al.  pointed out, “the full picture of a patient’s chronic health status” is not entirely captured by variables commonly included in illness severity score, such as age and major comorbidities. Interestingly, another recent study  found that duration of ICU stay prior to death in critically ill patients was longer in older patients but comorbidities did not represent a significant risk factor for the length of stay prior to death. This has important implications in current clinical practice and suggests using a PS index or a frailty index instead of only comorbidities when discussing futility of support care.
Credit must be given to the authors who measured a variable that intensivists, generally, include in the clinical evaluation but do not quantify. The large population involved and the robust statistical analyses are the major strengths of the study. This complex analysis included the imputation of the few missing values (5.8 %). There are also some weaknesses. The assessment of PS referred to 1 week before hospital admission. We cannot exclude that, for some patients, the poor PS in the week before hospital admission could be related to the clinical deterioration due to the illness responsible for the hospital and ICU admission. Therefore, the full picture of chronic health status or frailty in the study patients may be overrepresented, as correctly stated by the authors. One concern from the study lies in the use of only three categories to classify PS impairment with a grading system that comprised absent/minor (ECOG 0-1), moderate (ECOG 2) or severe (ECOG 3-4). However, the three categories seem to allow for the increase of PS impairment in the groups, because odds ratio for adjusted hospital mortality was 1.96 (95 % CI 1.63–2.35) for moderate and 4.22 (95 % CI 3.32–5.35) for severe impairment. However, we need to verify the generalizability of the results in countries with national health systems. Although, on the basis of its current gross national income, Brazil is an “upper middle income” country, most of the study patients (92 %) were from private hospitals . We cannot exclude that intensivists working in private hospitals may be more willing to accept patients with poor PS in ICU when requested by ward physicians or families. Conversely, ICUs in public hospitals may not admit some patients with very poor PS as a result of resource constraints. The patient severity of illness (SAPS3 43.0 ± 14.9) was similar to that reported in Spain (46.29 ± 14.34) , but lower than in Italy (65.8 ± 17.1) , and the percentage of mechanically ventilated patients was low (19 %). A weakness of the study  is the lack of report of end-of-life decision practices. This aspect needs to be considered for future clinical implementation of PS.
Characterization of the PS on ICU admission will become an immediate need in the near future for the whole ICU population. Since intensivists have more and more accessible and complex technologies to provide life support, performance status information will improve the clinical decision-making process for the entire ICU population. Nevertheless, several questions must be debated before integrating PS in severity tools or in our daily ICU admission process: Which measure can we use? What time period before hospital admission do we have to consider? Who must report the PS? Future studies should address these questions.
The authors thank Susan Paris for her help in English editing.
Compliance with ethical standards
Conflicts of interest
On behalf of all authors, the corresponding author states that there is no conflict of interest.
- 4.Bagshaw SM, Stelfox HT, McDermid RC, Rolfson DB, Tsuyuki RT, Baig N, Artiuch B, Ibrahim Q, Stollery DE, Rokosh E et al (2014) Association between frailty and short- and long-term outcomes among critically ill patients: a multicentre prospective cohort study. CMAJ 186:E95–E102CrossRefPubMedPubMedCentralGoogle Scholar