The Role of Nutritional Support in the Physical and Functional Recovery of Critically Ill Patients

A Narrative Review

Danielle E. Bear; Liesl Wandrag; Judith L. Merriweather; Bronwen Connolly; Nicholas Hart; Michael P. W. Grocott


Crit Care. 2017;21(226) 

In This Article


Whilst it is acknowledged that outcomes research is a priority for survivors of critical illness, no consensus exists on the most appropriate outcomes. There is considerable current activity in relation to core outcome sets in studies of physical rehabilitation[67] and long-term follow-up following acute respiratory failure[68] but no such initiative is ongoing for nutrition. Core outcome sets enable the combination and comparison of data from different studies of similar interventions and are urgently required in this field. Indeed, a recent scoping review of outcome measurement in ICU survivorship research from 1970–2013 found that 250 unique measurement instruments have been used across 425 studies.[69] Furthermore, only 31 RCTs included post-discharge outcomes and half of these had sample sizes of less than 100 subjects. Recent large trials undertaken in critical care nutrition have continued this pattern and utilised a variety of primary outcome measures from mortality to infectious complications and length of stay.[2,3,4,7,40,41,70] Whilst meta-analysis of such data is possible, the number of different outcome measures used profoundly limits the validity of any conclusions.

None of the large nutrition RCTs has used functional or health-related quality of life (HRQOL) measures as primary outcomes, but rather they have been included as secondary outcomes. This seems surprising, given that these are likely to be the outcomes where nutrition may show the most benefit,[71] but reflects the current uncertainty regarding the most appropriate measure to use across the continuum of critical illness and recovery.[14] It is common that a significant number of patients are effectively excluded from recording of physical and functional ability due to either the volitional nature of the measure[72] or due to logistical issues with returning to follow-up appointments. Overall, these outcome measures can be labour intensive and expensive for the researcher, which may also impact on the choice of outcomes used for each study, the time points chosen to measure and the number of patients able to be followed up long term. Accounting for logistics and costs associated with measuring long-term outcomes is essential to the success of future trials.