Inflammation Course and Skeletal Muscle Wasting Correlation During the First Week in ICU: A New Approach for Personalised Feeding of Critically Ill Patients?

Background: Critically ill patients undergo rapid and clinically significant skeletal muscle loss during the first week of ICU admission, driven by a complex interplay of systemic inflammation, neuroendocrine dysregulation, and accelerated protein catabolism. While CRP is an established marker of the inflammatory response, its temporal relationship with early muscle wasting-specifically whether the initial inflammatory peak or its subsequent persistence most strongly determines muscle loss-remains poorly characterised. This study investigated the serial dynamics of inflammatory biomarkers and their correlation with skeletal muscle changes during the first seven ICU days. Methods: This is a post hoc analysis of the NUTRITI prospective observational cohort, conducted at a single academic ICU in Udine, Italy (Ethics Committee approval: CEUR-2019-Os-17). Sixty-six adult critically ill patients with an anticipated ICU stay exceeding 72 h and requiring artificial nutritional support were included; patients on renal replacement therapy or with contraindications to bioelectrical impedance analysis (BIA) were excluded. Body composition-skeletal muscle mass (MM, kg) and phase angle (PA°)-was assessed by single-frequency BIA (50 kHz) on Day 1 and Day 7. The primary outcome was ΔMM%, the percentage change in muscle mass between admission and Day 7, calculated as ΔMM% = [(MMd - MMa)/MMa] × 100. Daily inflammatory biomarkers-CRP (mg/L), total WBC (×103/mm3), and lymphocyte count (×103/mm3)-were collected throughout. Spearman rank correlations between ΔMM% and serial biomarkers were computed for each day. Given the large number of tests (42 total), Bonferroni false discovery rate corrections were applied. Results: The cohort had a median age of 68.5 years (IQR 61-77.8), was predominantly male (71.2%), with median APACHE II 21.5 and SOFA 7. Median muscle mass declined significantly from 34.3 kg (IQR 29.9-39.5) at Day 1 to 30.6 kg (IQR 26.5-34.9) at Day 7 (p < 0.001), corresponding to a median MM% of -8.45% (IQR -14.2% to -1.52%). Phase angle also declined significantly (4.9° to 4.5°; p < 0.01). CRP showed no significant correlation with ΔMM% at Days 1 or 2, but a negative correlation emerged at Day 3 (ρ = -0.297; p = 0.020) and peaked at Day 4 (ρ = -0.355; p = 0.006), attenuating thereafter. CRP at Days 5 and 6 correlated with phase angle changes (p = 0.017 and p = 0.022, respectively). WBC showed no significant correlations at any time point. Day-7 lymphocyte count was nominally correlated with ΔMM% (ρ = -0.314; p = 0.040). Neither APACHE II nor SOFA at admission correlated with ΔMM%. No test survived correction for multiple comparisons. Conclusions: The kinetics of CRP-rather than its initial intensity-seem to be associated with early skeletal muscle catabolism in critically ill patients, with a temporally specific signal emerging at Days 3-4 of ICU admission. Although these findings are exploratory and did not survive multiple testing correction, their biological plausibility-grounded in the known kinetics of ubiquitin-proteasome activation and NF-κB signalling-and their alignment with the emerging concept of inflammation-guided nutritional timing both support their value as a hypothesis-generating observation. Serial CRP monitoring may represent a pragmatic candidate biomarker to identify the optimal window for nutritional escalation, pending prospective validation in adequately powered trials.