Oral Presentation 29th Australian and New Zealand Bone and Mineral Society Annual Scientific Meeting 2019

Cardiovascular events in a population-based observational cohort of bisphosphonate users and untreated control subjects linked to BMD and co-morbidity information   (#55)

Alexander J. Rodriguez 1 , Martin T. Ernst 2 , Mads Nybo 3 , Daniel Prieto-Alhambra 4 , Peter R. Ebeling 1 , Pernille Hermann 5 , Bo Abrahamsen 2
  1. Bone and Muscle Research Group, School of Clinical Sciences at Monash Health, Monash University, Melbourne, Victoria, Australia
  2. OPEN-Odense Patient Data Explorative Network, Institute of Clinical Research, University of Southern Denmark, Odense, Denmark
  3. Department of Clinical Biochemistry and Pharmacology, Odense University Hospital, Odense, Denmark
  4. Musculoskeletal Pharmaco- and Device Epidemiology, Centre for Statistics in Medicine, Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Oxford, United Kingdon
  5. Department of Endocrinology and Metabolism, Odense University Hospital, Odense, Denmark

Background. The relationship between bisphosphonate (BP) use and cardiovascular (CV) events is complex, with various reports of increased as well as decreased risk for CV events. Critically, low BMD is in itself associated with increased CV risk so information about CV risk factors and BMD is crucial when interpreting observational cohorts in individuals with osteoporosis.

 

Methodology. Individuals aged >45 years with areal Bone Mineral Density (aBMD) assessed by DXA in a single centre from 1 January 2003 to 31 December 2014 were included. Individuals were considered BP users for the purpose of this analysis if the filled at least one prescription for oral BP within two years of their aBMD assessment.

 

Outcomes. Clinical information on demographics, anthropometry, renal function, prior fractures, Charlson co-morbidity index, prescription history and hospitalisations were obtained from local and national registers. Cox-proportional hazard models [95% confidence interval (CI)] were fitted to determine the risk of CV events requiring hospitalisation and cause-specific CV hospitalisations associated with BP use controlling for age, sex and aBMD compared with non-users.

 

Results. 5,643 BP users (women, 4564 [80.9%]) were identified. Median age was 69 years [62-77] and BMI was 25.9kg/m2 [23.0, 29.3]. Alendronate accounted for 96% of prescriptions. Non-users who were significantly younger but had similar BMI and gender distribution [Table 1]. On survival analysis, BP was associated with an approximate 29% reduced risk of CV events (hazard ratio=0.71 [0.65, 0.76]). Hospitalisation for atrial fibrillation (0.52 [0.46, 0.59]); heart failure (0.51 [0.43, 0.60]); myocardial infarction (0.58 [0.46, 0.72]); cardiac arrhythmias (0.45 [0.38, 0.58]); stroke (0.53 [0.45, 0.61]); atherosclerotic vascular disease (0.51 [0.41, 0.63]); aneurysms (0.51 [0.45, 0.59]) and peripheral artery disease (0.81 [0.63, 1.04]) all showed lower risks.

 

Conclusion. In this preliminary report, BP use had lower risks of CV hospitalisations. A propensity-score matched sub-analysis of this cohort is underway.