Construction and validation of a simple, scoreable model for predicting infection risk in patients with multiple myeloma: a retrospective single-center study.

Objective: This study aims to identify the risk factors for infection in patients with multiple myeloma (MM) and to develop a predictive model for infection. Methods: We retrospectively analyzed the clinical data of 180 multiple myeloma patients with MM who underwent chemotherapy at the First Affiliated Hospital of Jishou University from January 2017 to December 2022. A predictive model for infection events was constructed based on these data. Results: In the modeling group, 34 out of 90 patients (37.78%) experienced infections, whereas in the validation group, 40 out of 90 patients (44.44%) had infections. Binary logistic regression analysis showed that the levels of C-reactive protein level, fasting blood glucose level, lactate dehydrogenase level, Eastern Cooperative Oncology Group (ECOG) score, and the percentage of bone marrow plasma cell percentage were independent risk factors for infection in patients with MM (P < 0.05). The infection prediction model developed using these variables demonstrated good accuracy, with an area under the ROC curve of 0.827 (95% CI: 73.66%-91.78%) in the modeling group and for the validation group being 0.760 (95% CI: 65.97%-85.93%) in the validation group. Conclusion: This study confirms that C-reactive protein level, fasting blood glucose level, lactate dehydrogenase level, ECOG score, and the percentage of bone marrow plasma cell percentage are significant risk factors for infection in patients with MM. Clinical significance: This infection prediction model offers substantial clinical value by enabling a shift from reactive management to proactive, preventive intervention for infections in this vulnerable population.