Skip to content
Open access

Neutrophil-to-lymphocyte ratio is associated with in-hospital mortality in patients with cirrhosis and pneumonia: a multicenter retrospective study

Oct 2026 · Frontiers in Medicine · Vol 13 · 0 citations · 47 references
Medicine

Abstract

Background Patients with cirrhosis complicated by pneumonia face high in-hospital mortality, yet reliable prognostic tools remain limited. The neutrophil-to-lymphocyte ratio (NLR) is a readily available inflammatory marker with established prognostic value across various diseases. However, multicenter evidence on the independent association of log10NLR with mortality in this specific population is lacking. This study therefore investigated the association between log10NLR and in-hospital mortality in patients with cirrhosis and pneumonia, and evaluated its discriminative performance. Methods This multicenter study included 851 patients with cirrhosis and pneumonia from five Chinese centers (2016–2025). Admission log10NLR was the exposure and in-hospital mortality the outcome. Multivariable logistic regression with multiple imputation, restricted cubic splines, and ROC analysis was used; the incremental value of log10NLR beyond the Child-Pugh and CURB-65 scores was assessed by the DeLong test, calibration curves, and decision-curve analysis. Results Among 851 patients, in-hospital mortality was 14.9%. In the fully adjusted model, a 1-unit increase in log10NLR (a 10-fold increase in NLR) was associated with higher odds of in-hospital death (OR: 3.82, 95% CI: 1.97–7.38; P < 0.001). Across tertiles, the odds increased progressively (second: OR: 2.81, 95% CI: 1.44–5.47; third: OR: 3.35, 95% CI: 1.65–6.78; P for trend = 0.001), and restricted cubic spline analysis showed a linear relationship (P for overall association = 0.003; P for non-linearity = 0.431). The AUC for log10NLR was 0.707 (95% CI: 0.660–0.754), numerically higher than for Child-Pugh (0.647) and CURB-65 (0.658), though not significantly (DeLong P > 0.05). Adding log10NLR to each score significantly improved discrimination (ΔAUC = 0.090 and 0.076; both P < 0.001) and net benefit. An exploratory cutoff of approximately 5 (sensitivity 74.0%, specificity 57.6%) requires independent validation. Conclusions Log10NLR was independently associated with in-hospital mortality in cirrhosis with pneumonia, with a linear dose–response relationship. Its discrimination alone was comparable to the Child-Pugh and CURB-65 scores; however, adding log10NLR significantly improved discrimination (ΔAUC = 0.090 and 0.076; both P < 0.001) and net benefit, suggesting a complementary role in risk stratification.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.