Correlation of NLR and Procalcitonin in Pregnancy Sepsis: Toward an Affordable Maternal Mortality Prediction Tool
摘要
To evaluate the correlation between neutrophil-to-lymphocyte ratio (NLR) and procalcitonin (PCT) levels in pregnancy-associated sepsis and to assess their combined utility as potential, affordable mortality risk predictors in resource-limited settings.
MethodsA retrospective observational study was conducted from January 2024 to July 2025 at Dr. Moewardi Hospital, Surakarta, including 50 pregnant women with sepsis, divided into survivors (n = 33) and non-survivors (n = 17). Clinical, obstetric, and laboratory data were reviewed at the time of diagnosis. PCT and NLR levels were compared, their correlation assessed, and receiver operating characteristic (ROC) analysis performed to identify optimal mortality prediction cut-offs. Logistic regression was used to determine mortality risk. Due to the retrospective design, the duration of illness and serial biomarker changes during treatment were not consistently available.
ResultsNon-survivors exhibited significantly higher PCT (median 112.3 ng/mL vs. 20.0 ng/mL; p < 0.001) and NLR (24.0 vs. 13.0; p < 0.001). A strong positive correlation was observed between PCT and NLR (r = 0.80, p < 0.001). ROC analysis showed good predictive performance for PCT (AUC 0.88, 95% CI: 0.75–0.96) and moderate performance for NLR (AUC 0.80, 95% CI: 0.67–0.91). Logistic regression revealed that elevated PCT (> 50.6 ng/mL) and NLR (> 21.2) were associated with increased mortality risk (OR 7.8 and OR 5.9, respectively).
ConclusionsNLR correlates strongly with PCT and serves as a low-cost and accessible surrogate biomarker for maternal sepsis severity. Combined assessment shows promise for affordable risk stratification and early identification of high-risk patients, particularly in resource-limited settings. However, given the retrospective, single-center design and limited microbiological and temporal data, these findings should be interpreted cautiously and validated in larger multicenter studies.