The evidence for optimal antithrombotic therapy after isolated surgical mitral valve repair (MVr) is limited resulting in variable use of antithrombotic regimens in practice. This study aimed to characterize what antithrombotic regimens are being used and elucidate rationale. We conducted a retrospective descriptive study of adult patients after isolated surgical MVr at two cardiac surgery centers to characterize the type of antithrombotics used, timing of initiation, and duration. We then conducted semi-structured interviews of cardiac surgeons across Canada to elucidate post-MVr antithrombotic prescribing practices along with their rationale. Of 365 MVr patient screened, 89 had isolated MVr without a concomitant indication for anticoagulation and majority were prescribed dual antiplatelet therapy (DAPT) (40%) or warfarin (36%). Other regimens included antiplatelet (13%), warfarin plus acetylsalicylic acid (ASA) (6%), or apixaban (5%). The intended duration of therapy was 3 months in 85% of cases, and after completion of initial antithrombotic therapy most cases (80%) were followed by indefinite ASA thereafter. Of the 15 surgeons interviewed, preferred regimens included ASA (47%), warfarin plus ASA (20%), warfarin (13%), apixaban (13%), apixaban plus ASA (13%), or DAPT (13%). Surgeons identified convention or clinical experience as rationale for choice of therapy. Those preferring anticoagulation cited high rates of post-operative atrial fibrillation and concern for preventable strokes while those preferring antiplatelets cited minimal risk of stroke and high risks of bleeds. In conclusion, this retrospective descriptive study and semi-structured interview found substantial heterogeneity in the antithrombotic regimens used and prescribing patterns following isolated mitral valve repair.
The burden of mitral valve regurgitation is increasing worldwide, for which surgical mitral valve repair (MVr) is the preferred treatment. ,, Following isolated MVr, defined as surgical MVr without other indications for anticoagulation or concomitant procedures, patients have a thromboembolic complication risk of 1.6% to 2.1% at 6 months while on ASA or warfarin. The optimal antithrombotic regimen in this setting is unknown due to a lack of randomized evidence. Furthermore, the benefits of TE risk reduction must be weighed against the potential harms of major bleeding, especially cardiac tamponade. ,,
Guideline recommendations conflict in either recommending warfarin or antiplatelet therapy for 3 months (low to very-low level of evidence). ,, Recommendations stem from retrospective observational studies comparing ASA, warfarin, combinations of both, or no anticoagulation (antiplatelet use not consistently reported) with varying conclusions favoring warfarin, no anticoagulation (antiplatelet use not reported), antithrombotic therapy over nothing, or finding no difference between treatment arms. ,,,,,,, Two observational studies and 1 small randomized controlled trial (RCT) comparing direct acting oral anticoagulants (DOAC) to warfarin after MVr found no difference in thromboembolic nor bleeding outcomes but were underpowered. ,, Systematic reviews on optimal antithrombotics after MVr are limited by the significant confounding and heterogeneity of these studies. ,,
Due to the heterogeneity in the available evidence, the objective of this study was to characterize the antithrombotic regimens used after isolated surgical MVr in patients without other indications for anticoagulation and highlight provider rationale to inform further research in this area.
Methods
We conducted a two-part study: a retrospective descriptive study and a semistructured interview study of cardiac surgeons across Canada. This study was approved by the University of British Columbia—Providence Health Care Clinical Research Ethics Board (H24-03439) and Behavioral Research Ethics Board (H24-03800).
Part 1: Descriptive study of antithrombotic prescribing patterns
Eligibility criteria
We screened all patients after surgical MVr admitted to two tertiary referral centers for cardiac surgery in Vancouver, Canada. Patients were screened if they had a surgical MVr between November 16, 2019 and January 7, 2025 at St. Paul’s Hospital or between November 5, 2022 and January 7, 2025 at Vancouver General Hospital. These dates were chosen for complete capture of cases after electronic medical record (EMR) implementation. The EMR list was verified with a list obtained from the Cardiac Services of British Columbia.
Inclusion/Exclusion
We included patients with isolated surgical MVr defined as having no other indication for anticoagulation and no other concomitant procedures. We excluded those with repeat mitral valve surgery or complex congenital heart disease anatomy.
Outcomes
The primary outcome was the proportion of antithrombotic regimens at discharge in patients with isolated surgical MVr. Secondary outcomes included timing of antithrombotic initiation, differences in baseline characteristics between those prescribed anticoagulants, antiplatelets, or both.
Data collection
We collected baseline demographics (medications and comorbidities), intraoperative variables (indication for surgery, anonymized surgeon, and operative details) and postoperative variables (clinical outcomes during index hospitalization including thromboembolism, bleeds as per BARC criteria, length of stay, and discharge antithrombotic). Data was collected using REDCap electronic data capture tools hosted at University of British Columbia by one researcher (JJ), with a second researcher (EC) independently auditing 10% of charts included.
Statistical analysis
Differences between baseline characteristics of patients discharged on anticoagulant, antiplatelet, or both were assessed using Chi-squared or Fisher’s exact tests for categorical variables, and one-way ANOVA for normally distributed continuous variables or the Kruskal–Wallis test for non-normally distributed continuous variables. Statistical significance was based on a 2-sided significance level of 0.05. All analyses were performed in R version 4.4.3.
Part 2: Interviews of cardiac surgeons
We conducted semistructured interviews of cardiac surgeons across Canada who currently perform surgical MVr. Surgeons were identified and contacted via email through snowball sampling of cardiac surgeons and pharmacists. Participant written informed consent was attained prior to the interview and interviews were conducted and recorded over Microsoft Teams video teleconferencing. Reporting is done according to the COREQ checklist.
Interview process and data collection
Pilot testing was performed with a cardiac surgeon who was part of the study team (EP). One investigator (JJ) conducted the interviews, which were 10 to 20 minutes in length. This investigator was a female pharmacist (BSc, PharmD, ACPR1), guided by members of the research team with experience in qualitative study design and semistructured interviews (EP, RDT), who did not have any relationship with participants prior to the study. Participants did not know the investigator and only participants and researchers were present for the interviews. Guided by an interview guide, the interviewer asked participants demographic details, choice of antithrombotic therapy after MVr, and rationale in patients with isolated MVr ( Supplemental Appendix S1 ). There were no repeat interviews. Interviews were recorded and transcribed verbatim. Transcripts were not returned to participants.
Statistical analysis
Transcripts were analyzed in NVivo version 1.7.2 (QSR International, Burlington, MA) using thematic analysis chronologically and in tandem with the interview process by one investigator (JJ). An inductive analytic approach was used whereby categories and themes were derived directly from participant data. One investigator (JJ) coded the deidentified interview transcriptions for common themes. A thematic coding scheme was developed based on the interview guides and the emerging themes.
Results
Part 1: Descriptive study
Baseline characteristics
From 865 patients screened, 364 patients had surgical MVr ( Figure 1 A). Of these 364 patients, 123 (34%) had pre-existing indications for anticoagulation and 200 of 364 (55%) had postoperative indications for anticoagulation ( Supplemental Table S1 ). After exclusion criteria, we included 89 patients with isolated MVr without a concomitant indication for anticoagulation (24% of MVr procedures). The median age was 61 years, 24 (27%) were female, 10 (11%) had coronary artery disease, 8 (9%) had a history of stroke or transient ischemic attack, and 14 (16%) were on an antiplatelet at baseline ( Table 1 ).
Descriptive cohort study flow chart.
CSBC = Cardiac Services of BC; EMR = electronical medical record; MV = mitral valve; MVr = mitral valve repair.
Panel A. Descriptive cohort study flowchart. Panel B: semistructured interview study of cardiac surgeons flowchart.
Table 1
Baseline characteristics
| Variable | Isolated MVr (n = 89) |
|---|---|
| Age (years), median (IQR) | 61 (51, 67) |
| Female (%) | 24 (27%) |
| Hypertension | 30 (34%) |
| Coronary artery disease | 10 (11%) |
| Diabetes | 5 (5.6%) |
| Chronic kidney disease stage 3 to 5 | 1 (1%) |
| Heart failure with reduced ejection fraction | 2 (2.2%) |
| Atrial fibrillation | 1 (1.1%) |
| History of stroke/TIA | 8 (9.0%) |
| Baseline Medications | |
| On antiplatelet at baseline | 14 (16%) |
| ASA | 14 (16%) |
| Clopidogrel | 1 (1.1%) |
| Baseline Surgical Characteristics | |
| Indication for surgery | |
| Primary MR | |
| Mitral valve prolapse | 77 (87%) |
| Infective endocarditis | 10 (11%) |
| Mitral annular dilation | 1 (1.1%) |
| Secondary MR | |
| Ischemic cardiomyopathy | 1 (1.1%) |
| Other | 0 |
| Urgency of Procedure | |
| Elective | 79 (89%) |
| Inpatient | 10 (11%) |
| MV Repair Details | |
| Annuloplasty ring/band | 89 (100%) |
| Leaflet resection or plasty | 62 (70%) |
| Neochords | 62 (70%) |
The isolated MVr column refers to patients that had isolated MVr and no other indication for anticoagulation.
AF = atrial fibrillation; CABG = coronary artery bypass graft; CHA 2 DS 2 -VASc = Congestive Heart Failure, Hypertension, A ge (≥ 75 Years) (doubled), D iabetes Mellitus, S troke (doubled), V ascular Disease, A ge (65-74) Years, S ex C ategory (Female); DOAC = direct oral anticoagulant; GI bleed = gastrointestinal bleed; HAS-BLED = H ypertension, A bnormal Renal/Liver Function, S troke, B leeding History or Predisposition, L abile INR, E lderly [> 65 Years], D rugs/Alcohol Concomitantly; HFrEF = heart failure with reduced ejection fraction; IQR = interquartile range; LAA = left atrial appendage; PFO = patent foramen ovale; PUD = peptic ulcer disease; SD = standard deviation; TIA = transient ischemic attack.
Antithrombotic regimen for isolated MVr with no indication for anticoagulation
For the primary outcome, 48 (54%) were discharged on an antiplatelet, 36 (41%) on an anticoagulant, and 5 (6%) on both an anticoagulant and antiplatelet ( Table 2 , Central Illustration ). Specifically, the discharge antithrombotic regimens included dual antiplatelet therapy (DAPT) with ASA and clopidogrel (40%), warfarin alone (36%), ASA alone (11%), warfarin and ASA (6%), apixaban (4%), and clopidogrel (2%). Of the 36 patients discharged on an anticoagulant, 15 patients were discharged on warfarin and ASA with the plan to discontinue ASA when a therapeutic INR was reached. These patients were classified as discharged on anticoagulant monotherapy. Majority (85%) of the prescribed antithrombotic regimens, except for ASA monotherapy, were for an intended 3-month duration ( Table 2 ). Of those that were not on ASA monotherapy at discharge, 80% were to switch to ASA monotherapy indefinitely after completing their initial antithrombotic prescribed course. The remaining 20% of patients did not have a documented plan.
Table 2
Antithrombotic regimen at discharge
| Antithrombotic type | n (%) | Median [IQR] time to initiate, post-operative day | Planned duration 3 months, n (%) | Plan for indefinite ASA following early regimen, n (%) |
|---|---|---|---|---|
| Antiplatelet | 48 (54) | |||
| ASA + clopidogrel | 36 (75) | Clopidogrel: 4 [3, 6]* | 34 (94) | 34 (100) |
| ASA | 10 (21) | 1 [1, 1] | 0 (0) | 10 (100) |
| Clopidogrel | 2 (4) | 1 [1, 1] | 1 (50) | 0 (0) |
| Anticoagulant | 36 (40) | |||
| Warfarin | 32 (89) | 2 [2, 2] | 32 (100) | 22 (69) |
| Apixaban | 4 (11) | 5 [5, 5.75] | 4 (100) | 2 (50) |
| Combination of anticoagulant and antiplatelet | 5 (6) | |||
| Warfarin + ASA | 5 (100) | Warfarin: 2 [2, 2] * | 5 (100) | 5 (100) |
ASA = acetylsalicylic acid; INR = international normalized ratio.
Antithrombotic date of initiation
The median date of initiation of warfarin and DOAC was on postoperative day 2 and 5, respectively. In those discharged on DAPT, clopidogrel was initiated on median postoperative day 4 while ASA was started on postoperative day 1.
Differences in baseline characteristics between antithrombotic type at discharge
For the secondary outcome, there was a statistically significant difference in procedure site and patients that received neochords across discharge antithrombotic groups (p <0.001) ( Supplemental Table S2 ). As highlighted in Figure 2 , surgeons 2 and 4 at the same site favored anticoagulant therapy. Conversely, surgeons 1, 3, and 5 from the other site had varying prescribing patterns favoring either DAPT or antiplatelet monotherapy. Between surgeons, there was also a statistically significant difference in the use of neochords (p <0.001).
Discharge antithrombotic type by operating surgeon.
In patients with isolated MVr with no other indication for anticoagulation, n = 89.
Clinical outcomes during index hospitalization
There were no thromboembolic events noted during index hospitalization ( Table 3 ). Bleeding events primarily occurred prior to or the day of antithrombotic initiation. Of the 5 excluded patients that died, cause of death was unrelated to antithrombotic therapy (fungal infective endocarditis, pneumosepsis, aspergillus pneumonia, multiorgan failure).
Table 3
In-hospital clinical outcomes in isolated MVr without concomitant indication for anticoagulation
| ASA (n = 10) | Clopidogrel (n = 2) | DAPT (n = 36) | Warfarin (n = 32) | DOAC (n = 4) | Warfarin + ASA (n = 5) | |
|---|---|---|---|---|---|---|
| Thrombo-embolism | 0 | 0 | 0 | 0 | 0 | 0 |
| Major bleed | 0 | 0 | 2 (6%) | 3 (9%) | 0 | 0 |
| Any bleed | 0 | 0 | 2 (6%) | 5 (16%) | 0 | 1 (20%) |
| POD of bleed, median | N/A | N/A | 0 [0, 0] | 1 [1, 3] | N/A | 4 [4, 4] |
| Length of stay (days), median (IQR) | 4.5 [4, 6] | 4 [3.5, 4.5] | 6 [5, 7.25] | 6 [5, 7] | 7.5 [5.5, 11.5] | 5 [4, 6] |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
