Abstract
BACKGROUND: Urinary uromodulin (uUMOD) is one of the novel biomarkers for predicting AKI. However, currently available publications showed inconsistent results. We designed this meta-analysis to evaluate the potential association between uUMOD and AKI. METHODS: We searched research articles with no language restriction in Medline, Web of Science, Cochrane Library, Embase, and 3 Chinese datasets from inception to February 2021. We used random-effects models to estimate the standardized mean difference (SMD) between patients with AKI or not, while the leave-one-out method and random-effects meta-regression to evaluate the sensitivity and the impact of potential confounders such as age and surgery. RESULTS: The meta-analysis comprising 3148 subjects from 11 studies showed that the uUMOD of the AKI group is significantly lower than the non-AKI group (SMD: - 0.71; 95% confidence interval (CI), - 1.00, - 0.42, P < 0. 001, I(2) = 78.8%). Subgroup analysis revealed the difference is also significant in a different age, surgery condition, and assay time but not acute rejection (AR) group, especially in children (SMD: - 1.21, 95% CI: - 1.80, - 0.61; P < 0.001) and patients undergoing surgery (SMD: - 1.03, 95% CI: - 1.75, - 0.30; P < 0.001). Lower uromodulin is associated with higher odds for AKI incidence (odds ratio = 2.47, 95% CI: 1.12, 5.47; P < 0.001, I(2) = 89%). Meta-reggression found that age was associated with the SMD of uUMOD. The study outcome was reliably confirmed by the sensitivity analysis. CONCLUSION: The present study suggested a negative association between uUMOD and AKI especially in children and surgical patients.