The platelet-to-white blood cell ratio as an independent risk factor for invasive intervention in tubo-ovarian abscess

Tubo-ovarian abscess (TOA) is a severe, potentially life-threatening complication of pelvic inflammatory disease. This study aimed to evaluate the predictive value of the platelet-to-white blood cell ratio (PWR) for the need for invasive intervention in patients with TOA. This retrospective cohort study included patients diagnosed with TOA at our institution between January 2014 and April 2026. Patients were stratified into two groups based on treatment modality: a non-invasive group (managed with antibiotics alone) and an invasive group (undergoing surgical drainage or aspiration). PWR was calculated as platelet count divided by white blood cell count. Multivariate logistic regression analysis was performed to identify independent risk factors for invasive intervention. Model performance was evaluated using receiver operating characteristic (ROC) curve analysis, decision curve analysis (DCA), and calibration plots. Among 174 patients, 89 (51.1%) required invasive intervention. Multivariate analysis showed that fever (odds ratio [OR] = 2.575; P < 0.05), age (OR = 1.056; P < 0.05), lesion diameter (OR = 1.425; P < 0.05), PWR (OR = 0.953; P < 0.05), and systemic immune-inflammation index (SII) (OR = 1.001; P < 0.05) were independently associated with invasive intervention. The area under the ROC curve (AUC) for PWR was 0.669, with an optimal cutoff of 28.2 (sensitivity, 67.4%; specificity, 61.2%). The area under the AUC for SII was 0.679, with an optimal cutoff of 1821.43 (sensitivity, 60.7%; specificity, 74.1%). When both models included the same other independent risk factors, the PWR-augmented model demonstrated predictive performance comparable to that of the SII-augmented model (AUC: 0.825 vs. 0.819, with overlapping 95% confidence intervals, greater net benefit, and improved calibration). A lower PWR at admission is independently associated with an increased risk of requiring invasive intervention in patients with TOA. Adding PWR to conventional clinical variables (lesion diameter, fever, age) improves early risk identification. The PWR-augmented model offers a simple, cost-effective tool for identifying high-risk patients, with predictive performance comparable to that of the SII-augmented model.