The effect of augmented protein intake on nitrogen balance: A nested cohort study of the TARGET protein trial

Abstract

Purpose The impact of augmenting enteral protein delivery on nitrogen balance, urea generation, and kidney function in critically ill patients remains poorly defined. This study aimed to investigate these effects in a nested cohort within the TARGET Protein trial. Methods We conducted a single-center study nested within a cluster randomized crossover trial. Consecutive patients were enrolled during either the usual protein intake period ( n = 52) or the augmented protein intake period ( n = 55). Nitrogen balance, its components, plasma urea concentrations, and urea nitrogen accumulation were compared between groups. Multivariable analyses identified factors independently associated with nitrogen balance and incident acute kidney injury (AKI). A causal mediation analysis was performed to assess whether the effect of augmented protein delivery on urea concentrations was mediated through incident AKI. Results From day 1 to day 7, mean protein intake was 0.84 ± 0.73 g·kg −1 ·day −1 in the usual-protein group and 1.35 ± 0.94 g·kg −1 ·day −1 in the augmented-protein group. Patients receiving augmented protein achieved a less negative nitrogen balance than those receiving usual protein (mean difference, +5 g per day; 95% confidence interval [CI], 0.4 to 10; P = 0.033), driven by greater nitrogen intake (mean difference, +7 g per day; 95% CI, 5 to 10; P < 0.001). Independent predictors of nitrogen balance were ideal body weight (−0.18 g.kg −1 ; 95% CI, −0.35 to −0.01; P = 0.040), AKI at ICU admission (+4.7 g; 95% CI, 0.10 to 9.3; P = 0.046), and allocation to augmented protein therapy (+5.1 g; 95% CI, 0.43 to 9.7; P = 0.033). Compared with usual protein delivery, augmented protein delivery resulted in higher plasma urea concentrations (median, 14 vs. 11 mmol/L) and greater urea nitrogen accumulation (+3 g; 95% CI, 1.0 to 6.0; P = 0.035). Approximately 60% of the additional positive nitrogen balance was accounted for by urea generation. Among patients without AKI at ICU admission, augmented protein delivery was independently associated with incident AKI (odds ratio, 4.79; 95% CI, 1.14 to 26.4; P = 0.046). In mediation analyses, most of the increase in urea concentrations associated with augmented protein delivery was attributable to a direct effect rather than mediation through incident AKI. Conclusions In critically ill patients, augmented protein delivery resulted in a less negative nitrogen balance but substantially increased urea generation and plasma urea concentrations. These findings suggest that a substantial proportion of additional protein intake may be directed toward nitrogen waste production rather than net anabolic utilization.