Jinyu Qiu
Preoperative NLR, leukocyte count, and platelet count were associated with postoperative urosepsis in older adults with T2DM and ureteral calculi. Their combination showed improved discrimination within the matched sample; however, the sample-derived cut-off values and performance measures should not be interpreted as universal screening thresholds without external validation.
BACKGROUND: To investigate the associations of preoperative neutrophil-to-lymphocyte ratio (NLR), leukocyte count, and platelet count with postoperative urosepsis in older adults with type 2 diabetes mellitus (T2DM) and ureteral calculi, and to explore their within-sample discriminative performance.
METHODS: This retrospective matched case-control study included 85 older adults with T2DM and ureteral calculi who developed urosepsis after laser lithotripsy between January 2021 and December 2025. From 1530 eligible surgical patients without postoperative urosepsis, 85 controls were selected by 1:1 matching on age and sex. Conditional logistic regression was used to evaluate factors associated with urosepsis. Receiver operating characteristic (ROC) analysis, pairwise DeLong tests, and 1000-resample bootstrap validation were used to assess the discriminative performance of NLR, leukocyte count, platelet count, and a combined indicator derived from a binary logistic regression model including all three markers.
RESULTS: Age and sex were balanced between the two groups after matching (both p > 0.05). Compared with the control group, the urosepsis group had higher NLR, leukocyte count, platelet count, stone computed tomography (CT) value, and urinary leukocyte positivity, and a higher prevalence of staghorn calculi (all p < 0.05). Conditional logistic regression showed that stone CT value, NLR, leukocyte count, platelet count, and staghorn calculus were associated with postoperative urosepsis (all p < 0.05). The areas under the curve (AUCs) for NLR, leukocyte count, platelet count, and the combined indicator were 0.742, 0.811, 0.681, and 0.872, respectively. The AUC of the combined indicator was larger than that of each individual marker by pairwise DeLong testing (p = 0.028 for leukocyte count; p < 0.001 for the other comparisons), and its optimism-corrected bootstrap AUC was 0.854.
CONCLUSIONS: Preoperative NLR, leukocyte count, and platelet count were associated with postoperative urosepsis in older adults with T2DM and ureteral calculi. Their combination showed improved discrimination within the matched sample; however, the sample-derived cut-off values and performance measures should not be interpreted as universal screening thresholds without external validation.