How do the top 10 known offending drug classes associated with decreased bone mineral density affect total shoulder arthroplasty outcomes?

Abstract

Many regularly prescribed classes of drugs are known to negatively impact bone health. However, it is unclear if perioperative use of these drugs impacts total shoulder arthroplasty (TSA) outcomes. The purpose of this study was to analyze the impact of perioperative use of 10 drug classes with known negative effects on bone health on prosthesis-related outcomes of TSA. Patients who underwent primary TSA were retrospectively identified in the PearlDiver database. Within this population, patients prescribed proton pump inhibitors (PPIs), thiazolidinediones (TZDs), loop diuretics, glucocorticoids, aromatase inhibitors, calcineurin inhibitors, selective serotonin reuptake inhibitors (SSRIs), antiepileptic drugs (AEDs), first-generation antipsychotics (FGAs), and second-generation antipsychotics (SGAs) within 6 months before or 6 months after primary TSA were identified (n=23,748). These patients were propensity score matched 1:1 with controls (n=23,748) on age, sex, and several comorbidities. After matching, patients with perioperative drug exposure were divided into 10 subgroups (ie, 1 for each drug class). Rates of prosthesis-related complications among patients taking each medication class vs. controls were compared with multivariable logistic regression. Relative to controls, SGA exposure was associated with significantly higher rates of all-cause revision (odds ratio [OR] 1.68) and aseptic revision (OR 1.57). Loop diuretic exposure was associated with significantly higher rates of all-cause revision (OR 1.44) and aseptic revision (OR 1.43). Glucocorticoid exposure was associated with significantly higher rates of all-cause revision (OR 1.32) and aseptic revision (OR 1.30). SSRI exposure was associated with significantly higher rates of all-cause revision (OR 1.27) and aseptic revision (OR 1.24). Periprosthetic fracture, aseptic loosening, and septic revision was comparable for all drug cohorts compared to matched controls (all P>.05). Patients with perioperative exposure to PPIs, TZDs, FGAs, AEDs, aromatase inhibitors, and calcineurin inhibitors displayed comparable rates of all queried complications compared with controls (all P>.05). Compared with matched controls, patients with perioperative exposure to SGAs, loop diuretics, glucocorticoids and SSRIs exhibited significantly higher rates of all-cause and aseptic revisions following primary TSA. Several other medications that are risk factors for osteoporosis and fragility fractures did not demonstrate significant associations with any complications, including periprosthetic fracture. These results highlight the need for a thorough review of patients' medical history and current medication usage prior to preoperative risk counseling for patients seeking TSA.