Patients with irregular heart rhythm (especially those who also have narrowing of the valve between the two left heart chambers) have higher risks of brain attack and require blood-thinning medications to prevent them. Direct oral anticoagulants (DOACs) and warfarin are commonly used for this purpose. In patients without narrowing or replacement of heart valve, DOACs are currently preferred over warfarin since it is safer and more effective. However, individuals with narrowing of heart valve were largely excluded from randomised clinical trials of DOACs, thus whether DOACs are safe and effective compared to warfarin for these individuals remains unclear. This study aims to compare DOACs versus warfarin on the risks of brain attack and bleeding in people with irregular heart rhythm and also narrowing of heart valve. The findings will help clinicians to make medication management decision on whether to prescribe or switch to DOACs for preventing brain attack in these patients.
This study aims to compare DOACs versus warfarin on the risks of stroke and bleeding in patients with atrial fibrillation and mitral stenosis. Patients aged ≥18 years with mitral stenosis and received DOACs or warfarin in 2010-2022 will be included. Target trial emulation with prevalent new user design will be adopted. At each calendar month during the study period (i.e., index date), eligible patients without prior use of DOACs are included. Those receiving a prescription of any DOAC (or warfarin) during the index month are categorized as the DOAC (or warfarin) group. Each patient in the DOAC group will be matched with four patients in the warfarin group according to the duration of prior warfarin exposure, and followed up till outcome occurrence, death, or end of study (31 December 2022). Outcomes of interest include ischemic stroke, systemic embolism, intracranial haemorrhage, gastrointestinal bleeding, and all-cause mortality.
Inverse probability of treatment weighting (IPTW) with propensity score will be used to minimise potential confounding. Covariates including age, sex, ethnicity, smoking status, deprivation index, CHA2DS2-VASC score, comorbidities, concomitant chronic medications, and clinical parameters will be adjusted for. Weighted pooled logistic regression will be used to estimate the hazard ratio and absolute risk differences. Additionally, patients will be censored on treatment switch or discontinuation during per-protocol analyses, and stabilised inverse probability of censoring weighting will be used to account for the potential bias.
Subgroup analyses will be conducted by age group, sex, CHA2DS2-VASC score, prior stroke or systemic embolism, prior bleeding, PPI/H2RA exposure. Sensitivity analyses restricting to incident new users without prior warfarin exposure in the DOAC group, and positive control analyses in patients with other valvular heart diseases will be conducted. Findings shall provide evidence on whether DOACs are safe and effective compared to warfarin for stroke prevention in AF-MS population.
Primary outcomes: composite of ischemic stroke or systemic embolism
Secondary outcomes: ischemic stroke, systemic embolism, intracranial haemorrhage, gastrointestinal bleeding, all-cause mortality
Li Wei - Chief Investigator - University College London ( UCL )
Chengsheng Ju - Corresponding Applicant - University College London ( UCL )
Kenneth Man - Collaborator - University College London ( UCL )
Vincent Ka Chun Yan - Collaborator - University College London ( UCL )
Kenneth Man - Collaborator - University College London ( UCL )
Vincent Ka Chun Yan - Collaborator - University College London ( UCL )
HES Accident and Emergency;HES Admitted Patient Care;HES Outpatient;ONS Death Registration Data;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation