Common errors and pitfalls in observational studies examining the association of medical nutrition therapy with outcomes in critically ill patients

Relevance of Protein Intake for Weaning in the Mechanically Ventilated Critically Ill: Analysis of a Large International Database

OBJECTIVES

The association between protein intake and the need for mechanical ventilation (MV) is controversial. We aimed to investigate the associations between protein intake and outcomes in ventilated critically ill patients.

DESIGN

Analysis of a subset of a large international point prevalence survey of nutritional practice in ICUs.

SETTING

A total of 785 international ICUs.

PATIENTS

A total of 12,930 patients had been in the ICU for at least 96 hours and required MV by the fourth day after ICU admission at the latest.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

We modeled associations between the adjusted hazard rate (aHR) of death in patients requiring MV and successful weaning (competing risks), and three categories of protein intake (low: < 0.8 g/kg/d, standard: 0.8-1.2 g/kg/d, high: > 1.2 g/kg/d). We compared five different hypothetical protein diets (an exclusively low protein intake, a standard protein intake given early (days 1-4) or late (days 5-11) after ICU admission, and an early or late high protein intake). There was no evidence that the level of protein intake was associated with time to weaning. However, compared with an exclusively low protein intake, a standard protein intake was associated with a lower hazard of death in MV: minimum aHR 0.60 (95% CI, 0.45-0.80). With an early high intake, there was a trend to a higher risk of death in patients requiring MV: maximum aHR 1.35 (95% CI, 0.99-1.85) compared with a standard diet.

CONCLUSIONS

The duration of MV does not appear to depend on protein intake, whereas mortality in patients requiring MV may be improved by a standard protein intake. Adverse effects of a high protein intake cannot be excluded.