Factors associated with pathological fractures among patients with multiple myeloma: a single-center retrospective analysis of 692 cases

Multiple myeloma (MM) is the second most prevalent hematologic malignancy, and pathological fracture represents the most severe complication of MM-related osteolytic bone disease, which significantly impairs patient survival and quality of life. While most existing studies have focused on Western populations, data on Chinese MM patients remains limited. This study aimed to investigate the incidence and associated factors of pathological fractures in patients with multiple myeloma (MM). This research is a single-center retrospective observational study. We retrospectively reviewed patients with newly diagnosed MM between 2007 and 2023 at a tertiary teaching hospital in Changsha, China and included 692 patients in the final analysis. All patients underwent imaging examinations (X-ray, CT, or MRI) to confirm pathological fracture diagnosis. We calculated the overall fracture incidence, and performed binary logistic regression to identify associated factors. Overall, the incidence of pathological fractures among the included MM patients was 48.4%, the most common fracture sites were the spine (72.2%), ribs (30.0%), and limbs (6.9%), respectively. Binary logistic regression identified six associated factors for pathological fracture. Factors positively associated with pathological fractures included overweight status (BMI ≥ 24), hypoalbuminemia (serum albumin < 35 g/L), hyperglobulinemia (serum globulin > 40 g/L), and worse Eastern Cooperative Oncology Group (ECOG) performance status. Meanwhile, glucocorticoid and bisphosphonate use was inversely associated with pathological fractures. Pathological fractures are highly prevalent among Chinese MM patients. Indicators including elevated BMI, hypoalbuminemia, hyperglobulinemia and high ECOG performance‑status score may assist clinicians in identifying patients likely to present with pathological fractures during clinical evaluation. Of note, the observed associations cannot establish temporal or causal relationships, given that clinical variables were assessed at the time of fracture event. Whether interventions targeting these factors can reduce fracture rates warrants further prospective investigation.