Our study aimed to assess the risk of all fractures and hip fractures in patients with atrial fibrillation (AF) who took non-vitamin K antagonist oral anticoagulants (NOACs) compared to warfarin.

We searched PubMed, Embase, and Cochrane Library and Clinical Trials.gov Website. Reviewed related researches up to January 31, 2020, to identify studies with more than 12 months of follow-up data. The protocol for this systematic review and meta-analysis has been registered in the International Prospective Register of Systematic Reviews (PROSPERO Number: CRD42020156893).

We included five RCT studies, and five observational studies that contained a total of 326,846 patients in our meta-analysis. Our meta-analysis showed that patients taken NOACs had no significant all fracture risk (RR = 0.91, 95% CI [0.81–1.01]) and hip fracture risk (RR = 0.92, 95% CI [0.82–1.03]) compared with those taken warfarin. Subanalysis showed that the risk of all fractures and hip fractures treated by NOACs were significant lower compared with warfarin in observational studies compared with RCT studies. Also, a subanalysis across the duration of anticoagulation showed the NOACs users have lower all fracture risk than warfarin users when the duration of anticoagulation ≤2 years (RR = 0.89, 95% CI [0.80–0.99]). Further analysis, significant lower all fracture risk in the rivaroxaban therapy (RR = 0.81; 95% CI [0.76–0.86]) compared with warfarin but no statistical significance in hip fracture. There were no significant difference of all fracture risk and hip fracture risk in dabigatran, apixaban, and edoxaban therapy compared with warfarin.

The meta-analysis demonstrated that NOACs associated with a significantly lower all fracture risk compared with warfarin when the duration of anticoagulation more than 2 years. Rivaroxaban users had lower risk of all fracture than warfarin users in AF patients. But there was no evidence to verify apixaban, edoxaban, and dabigatranin could decrease all fracture and hip fracture risk compared with warfarin.