ObjectiveTo determine the effect of zoledronic acid (ZA) on the risk of total knee replacement (TKR) in patients with symptomatic knee osteoarthritis and without severe joint space narrowing (JSN).MethodsWe included 222 participants (mean age 62 years, 52% female) from the 2‐year ZAP2 trial (113 received 5mg of ZA annually and 109 received placebo) conducted between November 2013 and October 2017. Primary TKR were identified until 22 Feb 2022. The effect of ZA on TKR risk was evaluated using Cox proportional hazard regression models. Because the treatment effect failed the proportional hazards assumption, a time‐varying coefficients analysis for treatment was conducted by splitting the study into 2 periods (i.e. within and after 2 years of randomisation).ResultsOver a mean follow‐up of 7 years, 39% and 30% of participants had any TKR in the ZA and placebo groups, and 28% and 18% had TKR in the study knee, respectively. Use of ZA was associated with a higher risk of TKR in any knee (hazard ratio [HR] 4.2, 95% confidence interval [CI] 1.2 to 14.7) and showed a trend in the study knee (HR 6.8, 95%CI 0.9 to 53.9) during the trial. In the post‐trial period, the risk of TKR was similar in the ZA and the placebo groups for any knee (HR 1.2, 95%CI 0.5 to 1.8) and the study knee (HR 1.4, 95%CI 0.5 to 2.2).ConclusionThese results suggest that ZA is not protective against TKR in patients with symptomatic knee osteoarthritis and without severe JSN.image