Impact of early nutrition and feeding route on outcomes of mechanically ventilated patients with shock: a post hoc marginal structural model study

Few data are available about optimal nutrition modalities in mechanically ventilated patients with shock. Our objective was to assess associations linking early nutrition (< 48 h after intubation), feeding route and calorie intake to mortality and risk of ventilator-associated pneumonia (VAP) in patients with invasive mechanical ventilation (IMV) and shock. In the prospective OutcomeRea database, we identified adults with IMV > 72 h and shock (arterial systolic pressure < 90 mmHg) within 48 h after intubation. A marginal structural Cox model was used to create a pseudo-population in which treatment was unconfounded by subject-specific characteristics. We included 3,032 patients. Early nutrition was associated with lower day-28 mortality [HR 0.89, 95 % confidence interval (CI) 0.81-0.98, P = 0.01] and day-7 mortality (HR 0.76, CI 0.66-0.87, P < 0.001) but not with lower day-7 to day-28 mortality (HR 1.00, CI 0.89-1.12, P = 0.98). Early nutrition increased VAP risk over the 28 days (HR 1.08, CI 1.00-1.17, P = 0.046) and until day 7 (HR 7.17, CI 6.27-8.19, P < 0.001) but decreased VAP risk from days 7 to 28 (HR 0.85, CI 0.78-0.92, P < 0.001). Compared to parenteral feeding, enteral feeding was associated with a slightly increased VAP risk (HR 1.11, CI 1.00-1.22, P = 0.04) but not with mortality. Neither mortality nor VAP risk differed between early calorie intakes of a parts per thousand yen20 and < 20 kcal/kg/day. In mechanically ventilated patients with shock, early nutrition was associated with reduced mortality. Neither feeding route nor early calorie intake was associated with mortality. Early nutrition and enteral feeding were associated with increased VAP risk.