Previous fracture and subsequent fracture risk: a meta-analysis to update FRAX
John А. Kanis, Helena Johansson, Eugène McCloskey, Enwu Liu, Kristina Åkesson, Frederick A. Anderson, Rafael Azagra, Cecilie L. Bager, Charlotte Beaudart, Heike A. Bischoff‐Ferrari, Emmanuel Biver, Olivier Bruyère, Jane A. Cauley, Roland Chapurlat, Claus Christiansen, Cyrus Cooper, Carolyn Crandall, Steven R. Cummings, José António Pereira da Silva, Bess Dawson‐Hughes, Adolfo Díez‐Pérez, Alyssa B. Dufour, John A. Eisman, Petra J. M. Elders, Serge Ferrari, Yuki Fujita, Saeko Fujiwara, C-C Glüer, Inbal Goldshtein, David Goltzman, Vilmundur Guðnason, J. Hall, Didier Hans, Mari Hoff, Rosemary Hollick, Martijn Huisman, Masayuki Iki, Sophia Ish‐Shalom, Graeme Jones, Magnus K. Karlsson, Sundeep Khosla, Douglas P. Kiel, W.-P. Koh, Fjorda Koromani, Mark A. Kotowicz, Heikki Kröger, Timothy Kwok, Olivier Lamy, Arnulf Langhammer, Bagher Larijani, Kurt Lippuner, Dan Mellström, Thomas Merlijn, Anna Nordström, Peter Nordström, Terence W O’Neill, Barbara Obermayer‐Pietsch, Claes Ohlsson, Eric Orwoll, Julie A. Pasco, Fernando Rivadeneira, Anne‐Marie Schott, Eric J. Shiroma, Kristín Siggeirsdóttir, Eleanor M. Simonsick, Elisabeth Sornay‐Rendu, Reijo Sund, Karin M. A. Swart, Paweł Szulc, Junko Tamaki, David Torgerson, Natasja M. van Schoor, Tjeerd van Staa, Joan Vila, Nicholas J. Wareham, Nick Wright, Noriko Yoshimura, M. Carola Zillikens, Marta Zwart, Liesbeth Vandenput, Nicholas C. Harvey, Mattias Lorentzon, William D. Leslie
Abstract
A large international meta-analysis using primary data from 64 cohorts has quantified the increased risk of fracture associated with a previous history of fracture for future use in FRAX. INTRODUCTION: The aim of this study was to quantify the fracture risk associated with a prior fracture on an international basis and to explore the relationship of this risk with age, sex, time since baseline and bone mineral density (BMD). METHODS: We studied 665,971 men and 1,438,535 women from 64 cohorts in 32 countries followed for a total of 19.5 million person-years. The effect of a prior history of fracture on the risk of any clinical fracture, any osteoporotic fracture, major osteoporotic fracture, and hip fracture alone was examined using an extended Poisson model in each cohort. Covariates examined were age, sex, BMD, and duration of follow-up. The results of the different studies were merged by using the weighted β-coefficients. RESULTS: A previous fracture history, compared with individuals without a prior fracture, was associated with a significantly increased risk of any clinical fracture (hazard ratio, HR = 1.88; 95% CI = 1.72-2.07). The risk ratio was similar for the outcome of osteoporotic fracture (HR = 1.87; 95% CI = 1.69-2.07), major osteoporotic fracture (HR = 1.83; 95% CI = 1.63-2.06), or for hip fracture (HR = 1.82; 95% CI = 1.62-2.06). There was no significant difference in risk ratio between men and women. Subsequent fracture risk was marginally downward adjusted when account was taken of BMD. Low BMD explained a minority of the risk for any clinical fracture (14%), osteoporotic fracture (17%), and for hip fracture (33%). The risk ratio for all fracture outcomes related to prior fracture decreased significantly with adjustment for age and time since baseline examination. CONCLUSION: A previous history of fracture confers an increased risk of fracture of substantial importance beyond that explained by BMD. The effect is similar in men and women. Its quantitation on an international basis permits the more accurate use of this risk factor in case finding strategies.