Population-based studies on risk of fracture in patients with neurological disorders

Abstract

Introduction Patients with neurological disorders may be at an increased risk of fracture via multiple causal pathways, including increases in the risk of falls, changes in bone mineral density and quality of bone microarchitecture. Risk of fracture may be increased by the disease itself, by comorbidities and by their treatment. Concomitant treatment with antidepressants, antipsychotics and proton pump inhibitors (PPIs) may cause an additional increase of fracture risk in these patients. Therefore, the main objectives of this thesis are to determine the risk of fracture in patients exposed to PPIs, antidepressants and antipsychotic and to determine fracture risk in patients after stroke, in patients with Parkinson’s disease (PD), myasthenia gravis (MG), muscular dystrophy (MD), Charcot-Marie-Tooth (CMT) and Guillain-Barré Syndrome (GBS). Finally, a fracture risk prediction model will be developed for PD patients. Methods The Dutch PHARMO Record Linkage System (RLS) database was used to determine the risk of hip/femur fracture for patients who used PPIs, antidepressants, antipsychotics and those who experienced a stroke. The United Kingdom Clinical Practice Research Datalink (CPRD) database was used to determine fracture risks in patients with PD, MG, MD, CMT and GBS. All patients were compared with population-based control patients. Results Current users of PPIs had a 1.2-fold increased risk of hip/femur fracture. Fracture risk attenuated with increasing durations of use. Risk of hip/femur fracture was 2.4-fold increased with current use of selective serotonin reuptake-inhibitor (SSRIs) and 1.8-fold for tricyclic antidepressants (TCAs). The risk declined rapidly after discontinuation of antidepressant use. The 1.7-fold increased risk for current antipsychotic users was significantly greater than the 1.3-fold increased risk for past use. A 2.0-fold increased risk of hip/femur fracture was observed in patients who experienced a stroke at any time before their fracture. The risk was highest among patients who sustained a stroke within 3 months after their stroke diagnosis. Risk of any fracture was 1.4-fold increased in MD patients. PD patients had a 1.9-fold increased risk for any fracture and we developed a fracture risk prediction model for PD patients with C-statistics (parameter to test internal validity of the model) of 0.69 for osteoporotic and 0.73 for hip fracture. There was no increased risk observed in patients with MG or GBS for any fracture. For CMT patients risk of non-osteoporotic fracture was 1.5-fold increased, whereas risk of any and osteoporotic fracture did not reach statistical significance. Discussion In conclusion, our findings demonstrate that patients with a diagnosis of the neurological disorders stroke, PD, MD, CMT and patients currently exposed to PPIs, antidepressants and antipsychotics have an increased risk of fracture. Falls and bone fragility are the principle determinants for an increased risk of fracture. Therefore, fall prevention programs to reduce the risk of falls and bisphosphonate use to reduce bone fragility may be recommended in those patients with the neurological disorders stroke, PD, MD and CMT and those patients exposed to antidepressants and antipsychotics. For these patients a specific fracture risk score may facilitate to identify those patients at increased risk of fracture.