RESUMEN
This prospectively registered systematic review and meta-analysis examines whether exercise (EX) training has an additive effect to osteoanabolic and/or antiresorptive pharmacological therapy (PT) in people with osteoporosis on bone mineral density (BMD), bone turnover markers (BTMs), fracture healing, and fractures. Four databases (inception to 6 May 2022), 5 trial registries, and reference lists were searched. Included were randomized controlled trials comparing the effect of EX + PT vs. PT with regard to BMD, BTM, fracture healing, and fractures. Risk of bias was assessed using the Cochrane RoB2 and certainty of evidence by the GRADE approach. Random-effects meta-analysis with Hartung-Knapp-Sidik-Jonkman adjustment was used to estimate standardized mean differences and 95% confidence intervals. Out of 2593 records, five RCTs with 530 participants were included. Meta-analysis showed with very low certainty evidence and wide confidence intervals that EX + PT compared to PT had larger effect sizes for BMD at 12 months at the hip (SMD [95%CI]: 0.18 [- 1.71; 2.06], n = 3 studies), tibia (0.25 [- 4.85; 5.34], n = 2), lumbar spine (0.20 [- 1.15; 1.55], n = 4), and forearm (0.05 [- 0.35; 0.46], n = 3), but not femoral neck (- 0.03 [- 1.80; 1.75], n = 3). Furthermore, no improvement was revealed for BTM such as bone ALP (- 0.68 [- 5.88; 4.53], n = 3), PINP (- 0.74 [- 10.42; 8.93], n = 2), and CTX-I (- 0.69 [- 9.61; 8.23], n = 2), but with very wide confidence intervals. Three potentially relevant ongoing trials were identified via registries. No data were found for fracture healing or fracture outcomes. It remains unclear whether EX has an additive impact to PT in people with osteoporosis. High-quality, adequately powered, targetted RCTs are required. PROTOCOL REGISTRATION: PROSPERO CRD42022336132.