Article Text

Correspondence on ‘Warfarin use and risk of knee and hip replacements’
  1. Na He1,2,
  2. Zhenwei Fang3,
  3. Xiaotong Li1,4,
  4. Suodi Zhai1,4
  1. 1Department of Pharmacy, Peking University Third Hospital, Beijing, China
  2. 2Department of Pharmacy Administration and Clinical Pharmacy, School of Pharmaceutical Science, Peking University, Beijing, China
  3. 3Department of Pharmacy, Beijing An Zhen Hospital, Beijing, China
  4. 4Institute for Drug Evaluation, Peking University Health Science Centre, Beijing, China
  1. Correspondence to Suodi Zhai, Department of Pharmacy, Peking University Third Hospital, Beijing 100191, China; zhaisuodi{at}

Statistics from

Warfarin, the traditional oral anticoagulant (OAC), functions by inhibiting the biological activity of vitamin K, and thus inhibiting the carboxylation of coagulation factors. Meanwhile, warfarin also interacts with the formation of active vitamin K-dependent bone and cartilage proteins, including osteocalcin, matrix Gla protein and Gla-rich protein.1 2 Therefore, warfarin can be hypothesised to be associated with adverse musculoskeletal events.2 In contrast, novel OACs (NOACs) act independently of vitamin K.

We have read with great interest the article by Ballal et al3 who conducted a nested case–control study and raised the concern that warfarin was associated with significantly higher risk of hip or knee replacements compared with NOACs (adjusted OR 1.59, 95% CI 1.31 to 1.92). Another cohort study also indicated that vitamin K antagonist was associated with an increased risk of osteoarthritis (OA) incidence and progression (OR 2.50, 95% CI 1.94 to 3.20).4 As OA is a relatively rare event and current evidence on the OAC-associated OA risk was limited, we aimed to examine the association between OACs and OA risk using the Food and Drug Administration Adverse Event Reporting System (FAERS) data.

We queried FAERS Database using OpenVigil V.2.1 ( Reports of OA events related to OACs from 2004 Q1 to 2020 Q4 were retrieved and analysed. We used Medical Dictionary for Regulatory Activities to identify cases with adverse events of interest. The primary outcome was OA. Secondary outcomes included OA-related composite event, osteoarthropathies and hip or knee replacements. The search terms for adverse events of interest are detailed in online supplemental table S1. We set three pairs of comparison: warfarin versus all other drugs, warfarin versus NOACs (dabigatran, rivaroxaban, apixaban and edoxaban) and NOACs versus all other drugs. Both reporting OR (ROR) and information component (IC) (Bayesian Confidence Propagation Neural Network analysis) were calculated to detect the potential safety signal. A signal was defined as a 95% CI of ROR or exp (IC) >1.6 The sensitivity analysis was also performed by restricting analysis to reports with patients aged 18 years or older. Analysis was performed in R (V.4.0.3).

We obtained 99 938 reports with warfarin and 207 513 reports with NOACs overall, and OA event was involved in 192 and 170 of them, respectively. The characteristics of warfarin and NOAC-associated OA reports are presented in online supplemental table S2. The reporting of OA was significantly higher with warfarin than all other drugs (1.92 vs 1.35 per 1000 cases; ROR 1.42, 95% CI 1.23 to 1.64; exp (IC) 1.64, 95% CI 1.33 to 2.01) and NOACs (1.92 vs 0.82 per 1000 cases; ROR 2.35, 95% CI 1.91 to 2.89; exp (IC) 2.01, 95% CI 1.56 to 2.58). Warfarin was also reported at disproportionately higher rate regarding OA-related composite event, osteoarthropathies and hip or knee replacement event (table 1). Meanwhile, no signals of disproportionate reporting of OA or joint replacements with NOACs were identified (table 1). When we restricted the analysis to reports of patients aged 18 years or older, no inconsistency with primary analysis was found (online supplemental table S3).

Table 1

Disproportionality analysis using different measures to identify potential association between different OACs and the risk of OA and joint replacement

The data mining of FAERS Database found consistent results with recent observational studies,3 4 which could augment the evidence of warfarin-associated OA and hip or knee replacement risk to some extent. Our study also has several limitations. First, analysis on spontaneous reports is prone to reporting bias, such as under-reporting and Weber effect. Apart from this, data on detailed comorbidities, co-medications and the treatment details are often missing and we could not control confounding bias properly.

To our knowledge, OA cannot be reversed and management on modifiable risk factors is essential. Except for the well-known disadvantage in thromboembolic event prevention, bleeding risk and drug–drug interactions,7 warfarin may be also associated with higher risk of OA and joint replacements compared with NOACs. The comparison of warfarin versus NOACs on this issue further supports NOACs recommended as the first-line therapy, especially among elderly patients who are also vulnerable to ageing-related musculoskeletal disorders. Notably, although recent correspondence on this topic had a discussion on the use of analgesics between different OAC groups,8 9 limited attention has been paid to patient-reported outcome (ie, joint pain) in previous studies. Also, further large population-based studies are warranted to investigate the duration-response of warfarin-associated OA risk, which plays an important role in making causal inference.10

In conclusion, no signals of disproportionate reporting of OA with NOACs were identified, while significantly disproportionate association was found for warfarin. Warfarin’s chronic adverse effect by interfacing with off-target vitamin K-dependent proteins deserves health professionals’ attention.

Ethics statements

Ethics approval

As the FAERS Database is open to the public, this study was exempt from institutional review board approval.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Contributors SZ and NH conceived the study and completed study design. NH collected data, performed statistical analysis and drafted the initial manuscript. ZF, XL and SZ revised the manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; internally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles