Commonly prescribed medications were associated with variable effects on fracture healing in adult patients, with most evidence focused on fracture risk rather than union or time to healing, according to a narrative review by lead study author Yash Sharma, MBBS, Core Surgical Trainee in Trauma and Orthopaedics of Royal Victoria Hospital, Belfast, and colleagues.
The review synthesized studies identified through PubMed/MEDLINE and reference screening, including randomized trials, observational studies, systematic reviews, and meta-analyses in adult patients with fractures. Medication classes evaluated included nonsteroidal anti-inflammatory drugs, corticosteroids, anticoagulants, psychotropic medications, proton pump inhibitors, antidiabetic therapies, and bone-active agents. Outcomes of interest included delayed union, nonunion, and time to union, although these endpoints were inconsistently reported across studies.
Across medication classes, findings were heterogeneous and depended on timing, duration, fracture characteristics, and comorbidity burden. Most evidence addressed fracture incidence or bone health rather than fracture healing specifically, limiting direct conclusions about fracture repair.
Nonsteroidal anti-inflammatory drugs were the most extensively studied class. Meta-analyses of randomized trials suggested short-term use had limited or difficult-to-detect effects on healing, whereas observational studies more often reported associations with delayed union or nonunion. These associations were most consistently observed with prolonged exposure, higher cumulative dose, and early-phase use, with variation across drug subclasses.
Systemic corticosteroids were associated with reduced bone formation and altered remodeling through effects on osteoblast and osteocyte activity, but direct evidence on fracture healing outcomes was limited. Most data addressed bone mineral density and fracture risk rather than union-specific endpoints.
Vitamin K antagonists were associated with increased fracture risk in population studies, while direct oral anticoagulants showed lower fracture rates in comparative analyses. However, studies evaluating fracture healing outcomes in anticoagulant-exposed patients were limited, with few reporting union or time-to-healing endpoints.
Selective serotonin reuptake inhibitors were consistently associated with increased fracture risk across meta-analyses and cohort studies, though evidence on fracture healing outcomes remained sparse. Proton pump inhibitors showed modest and inconsistent associations with fracture risk and no clear evidence of impaired fracture healing.
Metabolic disease, particularly diabetes, was associated with delayed healing and higher complication rates, supported by mechanistic evidence involving hyperglycemia and microvascular dysfunction. Polypharmacy was described as a marker of multimorbidity and treatment complexity rather than a direct causal determinant of impaired fracture healing.
Limitations
The narrative design did not include quantitative pooling or formal risk-of-bias assessment. Heterogeneity in study design, exposure definitions, and outcome measures limited comparability. For many medication classes, evidence was indirect and based on fracture risk rather than healing-specific outcomes, and observational findings were subject to confounding by indication and comorbidity burden.
Conclusion
Medication effects on fracture healing were not uniform and should be interpreted within clinical context, particularly with respect to exposure timing, dose, and patient characteristics.
“Fracture healing occurs within a complex clinical context shaped by injury characteristics, comorbidity burden, and cumulative medication exposure,” the researchers wrote.
Disclosures: The researchers reported no conflicts of interest.
Source: Cureus
