Intensive care researchers have tested whether a muscle-related biomarker measured on the second day in the ICU can single out patients who would benefit from early supplemental parenteral nutrition, and the answer, according to a new secondary analysis of the EPaNIC randomized trial, is no. The study, published in Critical Care on 26 October 2026 as an open-access Brief Report, examined the creatinine production rate, an integrative indicator of muscle status that had been suggested as a tool for identifying patients upon ICU admission who might respond to certain treatments.
The context for the question is well established in the trial record itself. In the original EPaNIC randomized controlled trial, which enrolled 4640 patients, initiating supplemental parenteral nutrition early to reach caloric targets harmed ICU patients compared with omitting parenteral nutrition for one week. Because the creatinine production rate had been proposed as an integrative marker of muscle status that could guide treatment selection, the investigators asked whether it could identify patient subgroups who could benefit from early parenteral nutrition rather than being harmed by it.
The analysis focused on profiles of the creatinine production rate and a body-size-corrected CPR index from ICU day 2 through day 14. The researchers investigated whether the day-2 creatinine production rate or CPR index statistically interacted with the randomized intervention in its effect on the duration of mechanical ventilation, which was used as a surrogate for muscle weakness, with adjustment for baseline risk factors.
Day-2 creatinine production rate data were available for 1322 patients randomized to early parenteral nutrition and 1382 patients randomized to late parenteral nutrition. In the early-parenteral-nutrition group, both the creatinine production rate and the CPR index were higher than in the late-nutrition group from day 4 through day 11 of the ICU stay, indicating that the intervention itself altered the biomarker profiles over the first week and a half of critical care.
The statistical interaction tests came close to, but did not cross, the conventional threshold of significance. The day-2 creatinine production rate interacted with randomization for its impact on mechanical ventilation with a P value of 0.067, and the CPR index showed an interaction P value of 0.058, both evaluated against the prespecified interaction threshold of P less than 0.1 used in the analysis plan, while standard significance was set at P less than 0.05.
When the researchers stratified patients by whether their day-2 values were above or below the group mean, a consistent pattern emerged among patients with high values. In patients with a high creatinine production rate, early parenteral nutrition was associated with a lower likelihood of earlier live ventilator weaning compared with late parenteral nutrition, with an adjusted hazard ratio of 0.871 (95 percent confidence interval 0.778 to 0.981, P = 0.022). The same held for patients with a high CPR index, where the adjusted hazard ratio was 0.858 (95 percent confidence interval 0.764 to 0.963, P = 0.0095).
By contrast, in patients with a low day-2 creatinine production rate or a low CPR index, early parenteral nutrition showed no statistically detectable effect on the likelihood of earlier live ventilator weaning, with P values of 0.60 and 0.73 respectively. Notably, early parenteral nutrition affected the creatinine production rate and CPR index profiles only in the high-value group, reinforcing that the biomarker trajectories differed between the strata but did not point toward benefit from the intervention in either.
The authors’ conclusion is unambiguous: the day-2 creatinine production rate and CPR index did not identify ICU patients who could benefit from intensive nutritional therapy with early parenteral nutrition. If anything, the subgroup analysis suggested that patients with higher biomarker values experienced a lower likelihood of earlier ventilator weaning when given early parenteral nutrition, mirroring the overall harm signal seen in the parent trial rather than revealing a subgroup in whom the therapy was advantageous.
Several features of the study design are important for interpreting these findings. This was a secondary analysis of an existing randomized controlled trial rather than a new trial, and it excluded patients on renal replacement therapy, which limits the applicability of the results to that population. Mechanical ventilation duration served as a surrogate outcome for muscle weakness, and the interaction tests were interpreted against a relaxed threshold of P less than 0.1, so the near-significant interaction values should be read with appropriate caution.
The study was registered as ClinicalTrials.gov NCT00512122, registered on July 31, 2007, and the published version is an open-access Brief Report in Critical Care carrying a permanent DOI, released early as a citable, peer-reviewed accepted manuscript subject to further editorial edits. For clinicians and researchers hoping that a single admission-day biomarker could personalize nutritional strategy in the ICU, this analysis delivers a sobering message: the creatinine production rate, at least as measured on day 2 in this large trial population, could not be used to select patients for early parenteral nutrition.
Subject of Research: Creatinine production rate as a biomarker for individualizing nutrition in critically ill ICU patients
Article Title: Assessment of creatinine production rate as a predictive biomarker to individualize nutrition in the ICU: a secondary analysis of the EPaNIC-RCT
Article References: Vanhorebeek, I., Gunst, J., Yamamoto, N., & Van den Berghe, G. (2026). Assessment of creatinine production rate as a predictive biomarker to individualize nutrition in the ICU: a secondary analysis of the EPaNIC-RCT. Critical Care. https://doi.org/10.1186/s13054-026-06415-w
Image Credits: AI Generated
DOI: 10.1186/s13054-026-06415-w
Keywords: creatinine production rate, ICU, parenteral nutrition, EPaNIC trial, muscle status, critical care, biomarker, mechanical ventilation, nutritional therapy, secondary analysis, Assessment, creatinine
News Source: Daisy Hatcher. (October 28, 2026). Creatinine production rate fails to identify ICU patients who benefit from early parenteral nutrition. Scienmag.
