Increased protein intake may worsen quality of life in ICU patients

24 Aug 2024
Increased protein intake may worsen quality of life in ICU patients

High enteral protein provision in critically ill patients does more harm than good, being associated with poorer health-related quality of life and falling short of improving functional outcomes after discharge when compared with standard enteral protein provision, according to the results of the PRECISe trial.

PRECISe included 935 critically ill patients (64.2 percent male) who underwent invasive mechanical ventilation within 24 hours of ICU admission. None of the patients had contraindications for enteral nutrition, moribund condition, BMI <18 kg/m2, kidney failure with a no dialysis code, or hepatic encephalopathy.

The patients were randomly assigned to receive isocaloric enteral feeds that contained 1.3 kcal/mL and 0.06 g of protein/mL (standard protein, n=465) or 1.3 kcal/mL and 0.10 g of protein/mL (high protein, n=470). The study-nutrition intervention was administered only during the patients’ ICU stay in which they required enteral feeding, with a maximum of 90 days.

The primary outcome was EuroQoL 5-Dimension 5-level (EQ-5D-5L) health utility score at 30 days, 90 days, and 180 days, adjusted for baseline EQ-5D-5L health utility score. A total of 430 patients (92.5 percent) in the standard protein group and 419 (89.1 percent) in the high protein group were assessed for the primary outcome.

During 180 days, the EQ-5D-5L health utility score was consistently lower in the high protein group than in the standard protein group (mean difference, –0.05, 95 percent confidence interval [CI], –0.10 to –0.01; p=0.031).

In terms of safety, the probability of mortality during the entire follow-up was comparable between the standard and high protein groups (0.38 vs 0.42; hazard ratio, 1.14, 95 percent CI, 0.92–1.40; p=0.22). Furthermore, patients in the high protein group were more likely to experience symptoms of gastrointestinal intolerance (odds ratio, 1.76, 95 percent CI, 1.06–2.92; p=0.030).

Lancet 2024;404:659-669