Correlates of Plasma Citrulline, a Potential Marker of Enterocyte Mass, among Children with Stunting: A Cross-Sectional Study in Uganda

Background: Environmental enteric dysfunction (EED) is associated with stunting. Citrulline, produced in mature enterocytes, may be a valuable biomarker of small intestinal enterocyte mass in the context of EED. Objectives: We aimed to explore the correlates of plasma citrulline (p-cit) in children with stunting. Methods: In a cross-sectional study using baseline data from the community-based MAGNUS (milk affecting growth, cognition and the gut in child stunting) trial (ISRCTN13093195), we explored potential correlates of p-cit in Ugandan children with stunting aged 12–59 mo. Using linear regression in univariate and multivariate models, we explored associations with socioeconomics, diet, micronutrient status, and water, sanitation, and hygiene characteristics. The influence of covariates age, fasting, and systemic inflammation were also explored. Results: In 750 children, the mean ± standard deviation age was 32.0 ± 11.7 mo, and height-for-age z-score was –3.02 ± 0.74. P-cit, available for 730 children, differed according to time fasted and was 20.7 ± 8.9, 22.3 ± 10.6 and 24.2 ± 13.1 μmol/L if fasted <2, 2–5 and >5 h, respectively. Positive correlates of p-cit were age [0.07; 95% confidence interval (CI): 0.001, 0.15 μmol/L] and log₁₀ serum insulin-like growth factor-1 (8.88; 95% CI: 5.09, 12.67 μmol/L). With adjustment for systemic inflammation, the association with serum insulin-like growth factor-1 reduced (4.98; 95% CI: 0.94, 9.03 μmol/L). Negative correlates of p-cit included food insecurity, wet season (–3.12; 95% CI: –4.97, –1.26 μmol/L), serum C-reactive protein (–0.15; 95% CI: –0.20, –0.10 μmol/L), serum α₁-acid glycoprotein (–5.34; 95% CI: –6.98, –3.70 μmol/L) and anemia (–1.95; 95% CI: –3.72, –0.18 μmol/L). Among the negatively correlated water, sanitation, and hygiene characteristics was lack of soap for handwashing (–2.53; 95% CI: –4.82, –0.25 μmol/L). Many associations attenuated with adjustment for inflammation. Conclusions: Many of the correlates of p-cit are characteristic of populations with a high EED prevalence. Systemic inflammation is strongly associated with p-cit and is implicated in EED and stunting. Adjustment for systemic inflammation attenuates many associations, reflecting either confounding, mediation, or both. This study highlights the complex interplay between p-cit and systemic inflammation.