Enteral Nutrition Delivery Methods and Clinical Outcomes in Critically Ill Patients: A Retrospective Cohort Study

BackgroundAdequate enteral nutrition is essential in critical illness, yet optimal delivery strategies remain debated. Rate-based and volume-based are commonly used, with prior studies suggesting improved caloric delivery with volume-based. However, real-world clinical outcomes, particularly with mixed protocols and in COVID-19 patients, are unclear. We hypothesized that volume-based would improve caloric delivery versus rate-based or mixed, and that critically ill COVID-19 patients would achieve a lower goal intake.MethodsThis single-center retrospective cohort study included ICU patients (2020-2023) receiving ≥48 h of enteral nutrition via rate-based, volume-based, or mixed strategies. The primary outcome was the percentage of prescribed intake achieved. Secondary outcomes included hospital and ICU length of stay, ventilator days, hospital-acquired pressure injuries and infections. Baseline illness severity was assessed using standard comorbidity and organ failure scores.ResultsAmong 324 patients, 24% achieved ≥80% of caloric goal, with no difference across feeding strategies (P = .30). Mixed feeding was associated with longer hospital and ICU stays, more ventilator days, and longer enteral nutrition duration (all P < .01). Mixed feeding was associated with higher pressure injury compared with rate-based (30% vs 16%, P = .03), with no association with hospital-acquired infections. Among patients achieving ≥80% prescribed intake, associations with ICU length of stay and ventilator days were no longer significant.COVID-19 status did not modify these associations, with longer stays in COVID-positive patients receiving mixed protocol.ConclusionsVolume-based did not improve total energy delivery. Mixed feeding was associated with worse clinical outcomes, likely reflecting greater illness complexity rather than causation. Consistency of nutrition delivery may be more important than protocol type. Prospective multicenter studies are needed.