- Email: [email protected]
Aortic Valve Replacement With Mitral Valve Repair Compared With Combined Aortic and Mitral Valve Replacement
Sachin Talwar, MCh, Ankit Mathur, MS, Shiv Kumar Choudhary, MCh, Rajvir Singh, PhD, and Arkalgud Sampath Kumar, MCh Cardiothoracic Centre and Department of Biostatistics, All India Institute of Medical Sciences, New Delhi, India
Background. Double valve replacement is associated with reduced long-term survival. This study investigates aortic valve replacement with mitral valve repair as an alternative to double valve replacement in patients with rheumatic heart disease (RHD). Methods. Between January 1995 and December 2005, 369 patients with RHD underwent combined aortic and mitral valve procedures. In 76 patients (20.6%), mitral valve repair with aortic valve replacement (group 1) was done. The remaining 293 patients (79.4%) underwent double valve replacement (group 2). A total of 351 patients (95%)—73 (96%) in group 1 and 278 (94.8%) in group 2—were 50 years of age or younger. Results. There was no difference in early mortality in the groups (4 in group 1 versus 25 in group 2, p ⴝ 0.35). The median follow-up was 60 months (range, 6 to 132 months) and 96% complete in group 1 and 92% in group 2. Actuarial survival was 90.5% ⴞ 3.4% in group 1 and
81.60% ⴞ 2.4% in group 2 at 60 months (p ⴝ 0.07). Event-free survival at 60 months was 78.3% ⴞ 5.1% in group 1 and 48.4 % ⴞ 3.2% in group 2 (p < 0.001). Reoperation-free survival was 92.5% ⴞ 0.4% in group 1 and 99.5% ⴞ 0.05% in group 2 (p ⴝ 0.014). Conclusions. Mitral valve repair with aortic valve replacement provides significantly better event-free survival than double valve replacement without a better actuarial survival. Reoperation rates are higher in the mitral valve repair and aortic valve replacement group, whereas thromboembolic complications were more in the double valve replacement group. Better event-free survival in patients undergoing mitral valve repair and aortic valve replacement still argues in favor of repair of the mitral valve whenever possible.
C
tic valve replacement (group 1). The remaining 293 (79.4%) underwent DVR (group 2). A total of 351 patients (95%) —73 in group 1 (96%) and 278 in group 2 (94.8%)— were 50 years of age or younger. Only RHD patients who had mechanical valve were included in this study. We did not include patients who underwent combined aortic and mitral valve repair and those who required additional procedures such as tricuspid valve repair, closure of atrial septal defect, or coronary artery bypass grafting, or those who underwent an emergency procedure. This study was approved by the hospital ethics committee, and informed consent was obtained from all patients. Preoperative transthoracic echocardiography was performed in all patients, and cardiac catheterization was performed in older patients to rule out coronary artery disease. Aortic and mitral valve disease was graded by angiography or Doppler echocardiography, or both, in some patients according to previously published criteria [9, 10]. Intraoperative transesophageal echocardiography (TEE) was used to assess the annulus of aortic and mitral valves, thickness and mobility of the valve cusps, presence of calcification, regurgitant jets, commisural and chordal fusion, and thickness and length of chordae
ombined aortic valve and mitral valve disease is not uncommon in patients with rheumatic heart disease (RHD) [1– 4]. For these patients, double valve replacement (DVR) has been the most commonly performed operation [5]. Some groups have limited experience with combined aortic and mitral valve repair [6], whereas others have demonstrated survival benefits of an approach of aortic valve replacement with mitral valve repair over the traditional strategy of DVR [7, 8]. In this report, we discuss the differences in outcomes in patients undergoing either DVR or aortic valve replacement with mitral valve repair during a 10-year period.
Patients and Methods Between January 1995 and December 2005, 369 patients with RHD underwent combined aortic and mitral valve surgery with replacement of aortic valve and repair or replacement of the mitral valve (Table 1). Of these, 76 patients (20.6%) underwent mitral valve repair with aorAccepted for publication April 27, 2007. Address correspondence to Dr Kumar, Department of Cardiothoracic and Vascular Surgery, Cardiothoracic Centre, All India Institute of Medical Sciences, Ansari Nagar, New Delhi, 110 029, India; e-mail: asampath_ [email protected].
© 2007 by The Society of Thoracic Surgeons Published by Elsevier Inc
(Ann Thorac Surg 2007;84:1219 –25) © 2007 by The Society of Thoracic Surgeons
0003-4975/07/$32.00 doi:10.1016/j.athoracsur.2007.04.115
CARDIOVASCULAR
Aortic Valve Replacement With Mitral Valve Repair Compared With Combined Aortic and Mitral Valve Replacement
1220
TALWAR ET AL AV REPLACEMENT, MV REPAIR VS DVR
Ann Thorac Surg 2007;84:1219 –25
Table 1. Preoperative Patients Undergoing Aortic and Mitral Valve Procedures Variablea CARDIOVASCULAR
Age (years) Sex Male Female Lesion MS MR MS ⫹ MR As/As ⫹ AR AR NYHA II III IV AF LVEF CHF PAH Mild Moderate Severe
AVR ⫹ MVR (n ⫽ 76)
DVR (n ⫽ 293)
p Value
30.3 ⫾ 10.4
32.5 ⫾ 10.7
0.10
53 (69.7) 23 (30.3)
211 (71.5) 82 (27.8)
0.88
31 (40.8) 12 (15.8) 33 (43.4) 56 (73.7) 20 (26.3)
131 (44.4) 39 (13.2) 125 (42.4) 111 (37.6) 183 (62)
19 (25) 48 (63.2) 9 (11.8) 37 (48.7) 0.593 ⫾ 0.101 13 (17.1)
83 (28.1) 189 (64.1) 23 (7.8) 113 (38.3) 0.526 ⫾ 0.151 54 (18.3)
2 (2.6) 50 (65.8) 22 (28.9)
30 (10.2) 179 (60.3) 84 (28.5)
0.50
0.10 0.10 0.651 0.077
a Categoric variables are expressed as number (%); continuous variables as mean ⫾ standard deviation.
AF ⫽ atrial fibrillation; AR ⫽ aortic regurgitation; As ⫽ aortic stenosis; AVR ⫽ aortic valve replacement; CHF ⫽ congestive heart failure; DVR ⫽ double valve replacement; LVEF ⫽ left ventricular ejection fraction; MR ⫽ mitral regurgitation; MS ⫽ mitral stenosis; MV ⫽ mitral valve; NYHA ⫽ New York Heart Association; PAH ⫽ pulmonary arterial hypertension.
tendinea of the mitral valve. Criteria for possible mitral valve repair were minimal or no calcification of the mitral valve leaflets and some degree of mobility of the cusps. The patients were evenly matched for age and sex. The pattern of mitral valve disease was similar in both groups, with a preponderance of mixed mitral stenosis and regurgitation. However, combined aortic stenosis and regurgitation was more common in patients undergoing aortic valve replacement with mitral valve repair compared with the group undergoing DVR, in whom aortic regurgitation was more common. Mitral valve calcification was present in 183 patients (62%) in group 2 compared with 20 (26.3%) in group 1 (p ⬍ 0.01). Previous closed mitral valvotomy was performed in 4 patients (5.3%) in group 1 compared with 31 (10.5%) in group 2 (p ⫽ 0.16). Preoperatively, infective endocarditis was present in 8 patients (10.5%) in group 1 and in 16 (5.4%) in group 2 (p ⫽ 0.1).
Surgical Technique All patients were operated on by the senior author (ASK). Surgery was performed through a mid sternotomy with aortobicaval cannulation and normothermic cardiopulmonary bypass (CPB). Cold blood cardioplegia delivered directly into the coronary ostia and topical ice slush were used for myocardial protection.
After the cardioplegia was delivered, the mitral valve was approached through an incision just behind the interatrial groove. Depending upon the morphology of the mitral valve, a decision was first taken to repair or replace the valve. If the valve was heavily calcified with shrunken leaflets and severe subvalvular fusion and was deemed unsuitable for repair, it was replaced using a mechanical valve with full chordal perseveration. If mitral valve repair was performed, no more than mild mitral regurgitation was accepted. A variety of techniques described by us previously were used for mitral valve repair [11–16]. These include commissurotomy in 56 patients, cuspal thinning in 68, chordal shortening in 12, cusp excision/plication in 5, cleft suture in 7, decalcification in 26, chordal transfer in 5, and neochordal construction in 2. Mitral valve annuloplasty (n ⫽ 42) was performed using a C-shaped ring of polytetrafluoroethylene felt as described earlier [13]. There was significant difference in both the groups in amount of calcification at the mitral valve (p ⬍ 0.05), and patients in group 2 had more severe mitral valve calcification. A decalcification procedure was done in 29 patients in group 1, but the calcification was typically mild and mostly located at posterior mitral leaflet tip, whereas the rest of the valve was thickened but pliable. After completing mitral valve repair, the mitral valve was tested by injecting cold saline into the left ventricular cavity to observe coaptation of the leaflets. However, the final assessment of the repair was made only by TEE after terminating CPB. After mitral valve repair, the aortic valve was replaced with mechanical valve using standard surgical techniques. Before termination of CPB, the left atrial appendage was routinely ligated in all patients. After termination of CPB, the valve function was assessed in all patients undergoing mitral valve repair. Before discharge from the hospital, transthoracic echocardiography was done in all patients and was repeated at 6-month intervals.
Anticoagulation Protocol All patients received acenocoumarin and life-long aspirin, starting on the first postoperative day. The dose of Table 2. Causes of Early and Late Death AVR ⫹ MVR (n)
DVR (n)
Early deaths Arrhythmia Bleeding Sepsis CHF Late deaths CVA Bleeding CHF Stuck valve
4 1 1 1 1 5 1 1 3 —
25 5 4 4 12 26 4 6 8 8
AVR ⫽ aortic valve replacement; CVA ⫽ cerebrovascular accident; MVR ⫽ mitral valve repair.
CHF ⫽ congestive heart failure; DVR ⫽ double valve replacement;
TALWAR ET AL AV REPLACEMENT, MV REPAIR VS DVR
1221
Fig 1. Actuarial survival in patients undergoing either aortic valve replacement (AVR) with mitral valve (MV) repair (group 1, hash marks) or double valve replacement (DVR; group 2, triangles).
acenocoumarin was titrated to maintain an international normalized ratio (INR) of 2.5 to 3.5 in patients undergoing DVR and 2.0 to 3.0 for patients undergoing aortic valve replacement with mitral valve repair according to previously published guidelines for antithrombotic therapy in patients with prosthetic heart valves [17]. In follow-up, these measurements were obtained after discharge from the hospital at 1 week, 1 month, 3 months, and then at 6-month intervals at our hospital, if the INR was stable and in the desired range. Because the patients come from distant regions from all over India, we insisted that these patients get routine measurements of INR near their place of residence at least once a month, and more often if indicated.
Statistical Analysis Data for continuous and categoric variables were collected. Mean and median were calculated for the continuous variables, and frequency distribution with percentage were calculated for categoric variables. The Student t test and Mann Whitney U test were applied to see significant difference between categoric and continuous variables, and the 2 test was applied to see association between categoric variables. To see significant difference in survival according to specific categories, Kaplan-Meier survival analysis was performed. A large number of factors were analyzed as predictors of early and late mortality, reoperation, and event-free survival. These included age, sex, type of mitral valve lesion, (mitral regurgitation or mixed), preoperative New York Heart Association (NYHA) functional class, preoperative congestive heart failure, atrial fibrillation, left ventricular dysfunction (ejection fraction of less than 0.50), CPB and aortic cross-clamp time, and
mitral valve repair versus replacement. All these variables were subjected to a stepwise multiple logistic regression analysis to determine the independent predictors of early and late mortality, reoperation, and eventfree survival. The Cox proportional hazard model was used to determine the prognostic factors for development of significant mitral regurgitation. The SPSS 11.5 statistical package (SPSS Inc, Chicago, IL) was used for the analysis. A value of p ⬍ 0.05 was considered as significant. A valve-related event was defined according to established guidelines [18], and these are summarized in the Appendix.
Results Mortality There were 29 (7.9%) early deaths: four (5.3%) in group 1 and 25 (8.5%) in group 2 (p ⫽ 0.35). There were five late deaths in group 1 and 26 late deaths in-group 2. The causes of deaths are listed in Table 2. Actuarial survival at 60 months was 90.5% ⫾ 3.4% in group 1 and 81.60% ⫾ 2.3% in group 2 (log-rank, 3.2; p ⫽ 0.07; Fig 1).
Follow-Up The mean follow-up was 69 ⫾ 41 months in-group 1 and 52 ⫾ 36.9 months (range, 6 to 132 months) in group 2 (p ⬎ 0.05). Follow-up was 96% complete in group 1 and totaled 414 patients years and was 92% in group 2 and totaled 1161.3 patients years. Among survivors in group 1, 33 (45.8%) were followed up for 8 years or more, 45 (62.5%) for 5 years or more, 58 (80.5%) for 3 years or more, 66 (91.6%) for 2 years or more, and 72 (96%) for 1 year or more. In group 2, 56 (20%) were followed up for 8 years or
CARDIOVASCULAR
Ann Thorac Surg 2007;84:1219 –25
1222
TALWAR ET AL AV REPLACEMENT, MV REPAIR VS DVR
Ann Thorac Surg 2007;84:1219 –25
CARDIOVASCULAR
Fig 2. Reoperation-free survival in patients undergoing aortic valve replacement (AVR) with mitral valve (MV) repair (group 1, hash marks) or double valve replacement (DVR; group 2, triangles).
more, 134 (49%) for 5 years or more, 201(73%) for 3 years or more, 232 (85%) for 2 years or more, and 270 (92%) for 1 year or more.
Thromboembolism A thromboembolic complication developed in 8 patients (10.5%) in group 1 and in 62 (21%) in group 2 (p ⫽ 0.037). Four were in atrial fibrillation (AF) in group 1 and 16 were in atrial fibrillation in group 2. In group 1, 7 patients (9.7%) had minor events, including peripheral embolism, transient ischemic event, and syncope. All of these patients were alive at the last follow-up. One patient had dense hemiparesis and died. Fig 3. Event-free survival in patients undergoing aortic valve replacement (AVR) with mitral valve (MV) repair (group 1, hash marks) or double valve replacement (DVR; group 2, triangles).
In group 2, 52 patients (17 %) had minor events. Major events occurred in 10 patients (3.4%), and 4 patients died. Of the remaining 6 patients, 2 recovered, and the remaining 4 patients continue to have hemiparesis. No obvious cause (eg, left atrial clot, atheroma) was found in these patients.
Hemorrhage Bleeding complications developed in 29 patients (38.2%) in group 1 and in 148 (50.3%) in group 2 that directly related to anticoagulation (p ⫽ 0.138). In group 1, 3 patients (4.1%) had major bleeding episodes, which included gastrointestinal hemorrhage in
2. One patient died, and 1 patient had epistaxis but recovered uneventfully. Minor hemorrhage occurred in 26 other patients (36%) in this group, but no intervention was required. In group 2, there were 12 (4.1%) major bleeding episodes: 3 patients had gastrointestinal hemorrhage, 2 had hemoptysis, 4 had epistaxis, and 3 had intracranial hemorrhage. Six patients died, and the rest were successfully managed conservatively.
Infective Endocarditis A total of 8 patients had postoperative infective endocarditis: 2 in group 1 and 6 in group 2 (p ⫽ 0.107). All except 3 patients in group 2 had evidence of preoperative endocarditis. All these patients were treated conservatively.
Reoperation Eight patients underwent reoperation: 7 (9.2%) in group 1 and 1 (0.3%) in group 2 (p ⬍ 0.01). In group 1, 4 patients requiring reoperation had documented evidence of recurrent rheumatic fever with progressive annular dilatation and progressive mitral regurgitation. Three patients who underwent reoperation within 24 months were found to have dehiscence of the annuloplasty felt and annular dilatation without evidence of rheumatic activity. Seven reoperations in group 1 were for significant mitral regurgitation (n ⫽ 6) and severe mitral stenosis (n ⫽ 1) at a mean of 65 months after the initial operation (range, 12 to 108 months). One patient died after reoperation of low cardiac output syndrome. The rest recovered uneventfully. The reoperation in group 2 was required for significant mitral prosthesis paravalvular leak 8 months after the initial operation. Actuarial freedom from reoperation was 92.5% ⫾ 0.4% in group 1 and 99.5% ⫾ 0.5% in group 2 at a median follow-up of 60 months (log-rank ⫽10.22; p ⫽ 0.014; Fig 2).
Valve Function Of the 268 patients in group 2 who survived the initial operation, 1 patient required reoperation for significant mitral paravalvular leak. Prosthetic valve thrombosis was encountered in 36 patients (13.4%) and affected the mitral valve in 28, and the aortic valve in 8. Eleven patients had prosthetic valve thrombosis within 1 year of the operation, 11 between 13 and 24 months, 8 between 25 and 48 months, 4 between 49 to 72 months, and 2 between 73 and 84 months. The INR was suboptimal in 34 of these patients and was in the normal range in 2 patients. All these patients underwent thrombolysis, after which 28 patients recovered, but 8 died awaiting reoperation. The patients who died had prosthetic valve thrombosis in the mitral position. In group 1, thrombosis of the prosthetic aortic valve was encountered in 3 patients at 44, 48, and 72 months after surgery. These 3 patients had suboptimal INR and responded to thrombolysis. Of the 72 survivors in group 1, 48 (67%) had no significant mitral regurgitation, 17 (24%) had moderate, and 6 (9%) had severe mitral regur-
TALWAR ET AL AV REPLACEMENT, MV REPAIR VS DVR
1223
gitation. Patients with moderate mitral regurgitation are in NYHA class II and are being followed up. All the patients with severe mitral regurgitation have undergone reoperation as described, and 1 has died. In addition to these patients, 1 patient underwent reoperation for significant mitral stenosis. Freedom from mitral regurgitation at 60 months was 89.9% ⫾ 0.2% in group 1.
Functional Class In group 1, 26 patients (34.7%) were in NYHA class I, 32 (44.4%) were in class II, and 14 (19.4%) were in class III at follow-up. In group 2, 128 patients (47.4%) were in class I, 111 (41.1%) were in class II, and 31 (11.5%) were in class III/IV. Event-free survival (freedom from all major and minor events) at 60 months was 78.3% ⫾ 5.1% in group 1 and 48.4% ⫾ 3.2% in group 2 (p ⬍ 0.001). The patients were free of events for a mean 17.03 ⫾ 26 months in group 1 and 10.55 ⫾ 17.06 months in group 2 (Fig 3). Multivariate analysis found no specific predictors of early or late mortality or reoperation; however, mitral valve replacement was associated with a poorer event free survival (log-rank, 14.86; p ⫽ 0.0001).
Comment The two groups in this retrospective study were comparable in age, sex, etiology, distribution of lesions, congestive cardiac failure, left ventricular ejection fraction, atrial fibrillation, infective endocarditis, and NYHA class (p ⬎ 0.05). All the patients in our series had RHD. There were no significant differences in number and cause of early death between the two groups. The difference in late deaths (4 in group 1 versus 26 in group 2) appears striking but is not statistically significant; thus, our results do not seem to indicate that mitral valve repair with aortic valve replacement is better in terms of actuarial survival compared with DVR. The reoperation rates after mitral valve repair have been reported at 1% to 18% [7, 11–14, 16]. In this study, the higher reoperation rate in group 1was due to a combination of recurrent episodes of rheumatic fever and technical issues. In group 1, 4 patients requiring reoperation had documented evidence of recurrent rheumatic fever with progressive annular dilatation and progressive mitral regurgitation. Three patients who underwent reoperation within 24 months were found to have dehiscence of the annuloplasty felt and annular dilatation, without evidence of rheumatic activity. Therefore, this may not be in agreement with other studies that have shown freedom from reoperation to be as low as 5% [19] to 65% on follow-up [5, 20] for the mitral valve due to recurrent rheumatic fever and have justified routine mitral valve replacement. Group 2 had a lower reoperation rate than group 1. There were no episodes of structural deterioration, and only 1 patient had periprosthetic leak. Eight more patients were definite candidates for reoperation, but they died before reoperation could be done. Thus, the significantly smaller number of reoperations in group 2 could be fallacious.
CARDIOVASCULAR
Ann Thorac Surg 2007;84:1219 –25
1224
TALWAR ET AL AV REPLACEMENT, MV REPAIR VS DVR
CARDIOVASCULAR
Gillinov and colleagues [21] clearly demonstrated a survival advantage of mitral valve repair in patients with double valve procedures, presumably due to the younger age of the RHD patients. Their study, however, did not take into consideration other variables such as thromboembolism, infective endocarditis, bleeding, and other valve-related events. Similar results have also been reported by others [7, 19, 22], who have related higher complications in the DVR group to higher doses of anticoagulation required in these patients. Event-free survival in our series was significantly better in group 1 than in group 2. We observed a higher incidence of thromboembolic complications in group 2. Most of these events were minor, and no obvious cause was detected. Such high rates of thromboembolism were also quoted by Armenti and colleagues [23], because they also recorded all major and minor events. Most of the other reported series have lower rate of thromboembolism because only major thromboembolic episodes were recorded. The incidence of prosthetic valve dysfunction was significantly different in the two groups. All the patients in group 1 with thrombosis of the prosthetic aortic valve responded to thrombolysis. Of 36 patients in group 2 who had a stuck valve, 28 occurred at the mitral position, and 8 of these died. The rest of the patients recovered with thrombolysis. A higher incidence of mitral valve thrombosis with higher mortality was noted in group 2. All deaths in these patients were related to mitral prosthesis, which argues in favor of mitral valve repair if otherwise feasible. This is in agreement with previous experience that thromboembolic complications and prosthetic valve thrombosis are increased with mitral valve replacement compared with repair and this reduces the survival [9]. A higher INR is recommended for patients undergoing mitral valve replacement or DVR compared with isolated aortic valve replacement [17, 24]. It is also well known that long-term survival after prosthetic valve replacement is directly related to the hazards of anticoagulation. Also, the actuarial survival is not reduced despite a higher reoperation rate in these patients (Fig 3), because in the present era, the reoperations are safe and do not significantly increase the risk of death [11]. Thus, in few subset of patients who are unable to adhere to a more intensive anticoagulation regimen and do not come for regular follow-up or require lower INR, there may be a survival advantage of mitral valve repair and aortic valve replacement compared with DVR, despite a higher reoperation rate. Although more than 50% of patients in group 2 had a bleeding complication, the difference was not statistically significant. We also did not find any significant difference in the incidence of infective endocarditis between the two groups, because all except 2 patients had evidence of preoperative endocarditis. This study is of retrospective nature and the patients were not randomized to either group. Intraoperative evaluation of the mitral valve on TEE or operative findings were used to decide whether the mitral valve should be repaired or replaced. This itself introduces a bias, wherein the repair was offered to patients who had
Ann Thorac Surg 2007;84:1219 –25
pliable, noncalcified or minimally calcified mitral valves. Replacement was undertaken for calcified nonpliable mitral valves where repair was not considered an option. Considering the benefits of valve repair over replacement, it may not be ethically justifiable to conduct a prospective randomized study of these patients. Mitral valve repair with aortic valve replacement is superior to DVR in terms of better event-free survival, but there is no survival benefit. We observed a higher incidence of thromboembolic complications in the DVR group and higher reoperation rates in the mitral valve repair and aortic valve replacement group. The better event-free survival in patients undergoing mitral valve repair and aortic valve replacement still argues in favor of repair of mitral valve when ever possible.
References 1. Nakano K, Koyanagi H, Hashimoto A, et al. Twelve year’s experience with the St Jude Medical valve prosthesis. Ann Thorac Surg 1994;57:697–703. 2. Baudet EM, Puel V, McBride JT, et al. Long-term results of valve replacement with the St. Jude Medical prosthesis. J Thorac Cardiovasc Surg 1995;109:858 –70. 3. Kinsley RH, Antunes MJ, Colsen PR. St Jude Medical valve replacement: an evaluation of valve performance. J Thorac Cardiovasc Surg 1986;92:349 – 60. 4. Hartz RS, Fisher EB, Finkelmeier B, et al. An eight-year experience with porcine bioprosthetic cardiac valves. J Thorac Cardiovasc Surg 1986;91:910 –7. 5. Grossi EA, Galloway AC, Miller JS, et al. Valve repair versus replacement for mitral insufficiency: when is a mechanical valve still indicated? J Thorac Cardiovasc Surg 1998;115:389 –96. 6. Gillinov AM, Blackstone EH, White J, et al. Durability of combined aortic and mitral valve repair. Ann Thorac Surg 2001;72:20 –7. 7. Kaul P, John S, John CN, et al. Aortic valve replacement with concomitant open mitral valvotomy— early results and longterm follow up in 72 consecutive patients. Indian Heart J 1993;45:113–5. 8. Szentpetery S, Rich JB, Azar H, Newton JR, Tenzer MM. Mitral valve repair combined with aortic valve replacement. J Heart Valve Dis 1997;6:32– 6. 9. Elmcke F, Nanda NC, Hsiung MC, et al. Color Doppler assessment of mitral regurgitation with orthogonal planes. Circulation 1987;75:175– 83. 10. Grossman W. Profiles in valvular heart disease. In: Baim DS, Grossman W, eds. Cardiac catheterization, angiography, and intervention. 5th ed. Baltimore, MD: Willams & Wilkins; 1996:742– 4. 11. Choudhary SK, Talwar S, Dubey B, Chopra A, Saxena A, Kumar AS. Mitral valve repair in a predominantly rheumatic population. Long-term results. Tex Heart Inst J 2001;28:8 –15. 12. Duran CG, Revuelta JM, Gaite L, Alonso C, Fleitas MG. Stability of mitral reconstructive surgery at 10–12 years for predominantly rheumatic valvular disease. Circulation 1988;78:I91–6. 13. Kumar AS, Rao PN, Saxena A. Results of mitral valve reconstruction in children with rheumatic heart disease. Ann Thorac Surg 1995;60:1044 –7. 14. Kumar AS, Kumar RV, Shrivastava S, Venugopal P, Sood AK, Gopinath N. Mitral valve reconstruction: early results of a modified cooley technique. Tex Heart Inst J 1992;19:107–11. 15. Kumar AS, Rao PN. Restoration of pliability to the mitral leaflets during reconstruction. J Heart Valve Dis 1995;4:251–3. 16. Kumar AS, Bhan A, Kumar RV, Shrivastava S, Sood AK, Gopinath N. Cusp-level chordal shortening for rheumatic mitral regurgitation: early results. Tex Heart Inst J 1992;19:47–50.
Ann Thorac Surg 2007;84:1219 –25
1225
24. Kulik A, Rubens FD, Wells PS, et al. Early postoperative anticoagulation after mechanical valve replacement: a systematic review. Ann Thorac Surg 2006;81:770 – 81.
Appendix Events Considered for Calculating Event-Free Survival ● ● ● ● ● ● ● ● ● ● ● ● ●
Early death Late death Structural valve deterioration Significant mitral stenosis or regurgitation Prosthetic valve thrombosis Paravalvular leak Hemolytic anaemia related to valve lesions Any new temporary or permanent focal or global neurologic deficit Transient ischemic attack Peripheral embolic event Bleeding events: All major and minor hemorrhage Infective endocarditis Reoperation
CARDIOVASCULAR
17. Stein PD, Alpert JS, Dalen JE, Horsekotte D, Turpie AG. Antithrombotic therapy in patients with mechanical and biologic prosthetic heart valves. Chest 1998;114:602:S-610. 18. Edmunds LH Jr, Clark RE, Cohn LH, Grunkemeier GL, Miller DC, Weisel RD. Guidelines for reporting morbidity and mortality after cardiac valvular operations. J Thorac Cardiovasc Surg 1996;112:708 –11. 19. Hamamoto M, Bando K, Kobayashi J, et al. Durability and outcome of aortic valve replacement with mitral valve repair versus double valve replacement. Ann Thorac Surg 2003;75:28–34. 20. Kuwaki K, Kawaharada N, Morishita K, et al. Mitral valve repair versus replacement in simultaneous mitral and aortic valve surgery for rheumatic disease. Ann Thorac Surg 2007; 83:558 – 63. 21. Gillinov AM, Blackstone EH, Cosgrove DM, et al. Mitral valve repair with Aortic valve replacement is superior to double valve replacement. J Thorac Cardiovasc Surg 2003; 125:1372– 87. 22. Tri Ho HQ, Nguyen VP, Phan KP, Pham NV. Mitral valve repair with Aortic valve replacement in rheumatic heart disease. Asian Cardiovasc Thorac Ann 2004;12:341–5. 23. Armenti F, Stephenson LW, Edmunds LH Jr. Simultaneous implantation of St. Jude Medical aortic and mitral prostheses. J Thorac Cardiovasc Surg 1987;94:733–9.
TALWAR ET AL AV REPLACEMENT, MV REPAIR VS DVR
Background. Double valve replacement is associated with reduced long-term survival. This study investigates aortic valve replacement with mitral valve repair as an alternative to double valve replacement in patients with rheumatic heart disease (RHD). Methods. Between January 1995 and December 2005, 369 patients with RHD underwent combined aortic and mitral valve procedures. In 76 patients (20.6%), mitral valve repair with aortic valve replacement (group 1) was done. The remaining 293 patients (79.4%) underwent double valve replacement (group 2). A total of 351 patients (95%)—73 (96%) in group 1 and 278 (94.8%) in group 2—were 50 years of age or younger. Results. There was no difference in early mortality in the groups (4 in group 1 versus 25 in group 2, p ⴝ 0.35). The median follow-up was 60 months (range, 6 to 132 months) and 96% complete in group 1 and 92% in group 2. Actuarial survival was 90.5% ⴞ 3.4% in group 1 and
81.60% ⴞ 2.4% in group 2 at 60 months (p ⴝ 0.07). Event-free survival at 60 months was 78.3% ⴞ 5.1% in group 1 and 48.4 % ⴞ 3.2% in group 2 (p < 0.001). Reoperation-free survival was 92.5% ⴞ 0.4% in group 1 and 99.5% ⴞ 0.05% in group 2 (p ⴝ 0.014). Conclusions. Mitral valve repair with aortic valve replacement provides significantly better event-free survival than double valve replacement without a better actuarial survival. Reoperation rates are higher in the mitral valve repair and aortic valve replacement group, whereas thromboembolic complications were more in the double valve replacement group. Better event-free survival in patients undergoing mitral valve repair and aortic valve replacement still argues in favor of repair of the mitral valve whenever possible.
C
tic valve replacement (group 1). The remaining 293 (79.4%) underwent DVR (group 2). A total of 351 patients (95%) —73 in group 1 (96%) and 278 in group 2 (94.8%)— were 50 years of age or younger. Only RHD patients who had mechanical valve were included in this study. We did not include patients who underwent combined aortic and mitral valve repair and those who required additional procedures such as tricuspid valve repair, closure of atrial septal defect, or coronary artery bypass grafting, or those who underwent an emergency procedure. This study was approved by the hospital ethics committee, and informed consent was obtained from all patients. Preoperative transthoracic echocardiography was performed in all patients, and cardiac catheterization was performed in older patients to rule out coronary artery disease. Aortic and mitral valve disease was graded by angiography or Doppler echocardiography, or both, in some patients according to previously published criteria [9, 10]. Intraoperative transesophageal echocardiography (TEE) was used to assess the annulus of aortic and mitral valves, thickness and mobility of the valve cusps, presence of calcification, regurgitant jets, commisural and chordal fusion, and thickness and length of chordae
ombined aortic valve and mitral valve disease is not uncommon in patients with rheumatic heart disease (RHD) [1– 4]. For these patients, double valve replacement (DVR) has been the most commonly performed operation [5]. Some groups have limited experience with combined aortic and mitral valve repair [6], whereas others have demonstrated survival benefits of an approach of aortic valve replacement with mitral valve repair over the traditional strategy of DVR [7, 8]. In this report, we discuss the differences in outcomes in patients undergoing either DVR or aortic valve replacement with mitral valve repair during a 10-year period.
Patients and Methods Between January 1995 and December 2005, 369 patients with RHD underwent combined aortic and mitral valve surgery with replacement of aortic valve and repair or replacement of the mitral valve (Table 1). Of these, 76 patients (20.6%) underwent mitral valve repair with aorAccepted for publication April 27, 2007. Address correspondence to Dr Kumar, Department of Cardiothoracic and Vascular Surgery, Cardiothoracic Centre, All India Institute of Medical Sciences, Ansari Nagar, New Delhi, 110 029, India; e-mail: asampath_ [email protected].
© 2007 by The Society of Thoracic Surgeons Published by Elsevier Inc
(Ann Thorac Surg 2007;84:1219 –25) © 2007 by The Society of Thoracic Surgeons
0003-4975/07/$32.00 doi:10.1016/j.athoracsur.2007.04.115
CARDIOVASCULAR
Aortic Valve Replacement With Mitral Valve Repair Compared With Combined Aortic and Mitral Valve Replacement
1220
TALWAR ET AL AV REPLACEMENT, MV REPAIR VS DVR
Ann Thorac Surg 2007;84:1219 –25
Table 1. Preoperative Patients Undergoing Aortic and Mitral Valve Procedures Variablea CARDIOVASCULAR
Age (years) Sex Male Female Lesion MS MR MS ⫹ MR As/As ⫹ AR AR NYHA II III IV AF LVEF CHF PAH Mild Moderate Severe
AVR ⫹ MVR (n ⫽ 76)
DVR (n ⫽ 293)
p Value
30.3 ⫾ 10.4
32.5 ⫾ 10.7
0.10
53 (69.7) 23 (30.3)
211 (71.5) 82 (27.8)
0.88
31 (40.8) 12 (15.8) 33 (43.4) 56 (73.7) 20 (26.3)
131 (44.4) 39 (13.2) 125 (42.4) 111 (37.6) 183 (62)
19 (25) 48 (63.2) 9 (11.8) 37 (48.7) 0.593 ⫾ 0.101 13 (17.1)
83 (28.1) 189 (64.1) 23 (7.8) 113 (38.3) 0.526 ⫾ 0.151 54 (18.3)
2 (2.6) 50 (65.8) 22 (28.9)
30 (10.2) 179 (60.3) 84 (28.5)
0.50
0.10 0.10 0.651 0.077
a Categoric variables are expressed as number (%); continuous variables as mean ⫾ standard deviation.
AF ⫽ atrial fibrillation; AR ⫽ aortic regurgitation; As ⫽ aortic stenosis; AVR ⫽ aortic valve replacement; CHF ⫽ congestive heart failure; DVR ⫽ double valve replacement; LVEF ⫽ left ventricular ejection fraction; MR ⫽ mitral regurgitation; MS ⫽ mitral stenosis; MV ⫽ mitral valve; NYHA ⫽ New York Heart Association; PAH ⫽ pulmonary arterial hypertension.
tendinea of the mitral valve. Criteria for possible mitral valve repair were minimal or no calcification of the mitral valve leaflets and some degree of mobility of the cusps. The patients were evenly matched for age and sex. The pattern of mitral valve disease was similar in both groups, with a preponderance of mixed mitral stenosis and regurgitation. However, combined aortic stenosis and regurgitation was more common in patients undergoing aortic valve replacement with mitral valve repair compared with the group undergoing DVR, in whom aortic regurgitation was more common. Mitral valve calcification was present in 183 patients (62%) in group 2 compared with 20 (26.3%) in group 1 (p ⬍ 0.01). Previous closed mitral valvotomy was performed in 4 patients (5.3%) in group 1 compared with 31 (10.5%) in group 2 (p ⫽ 0.16). Preoperatively, infective endocarditis was present in 8 patients (10.5%) in group 1 and in 16 (5.4%) in group 2 (p ⫽ 0.1).
Surgical Technique All patients were operated on by the senior author (ASK). Surgery was performed through a mid sternotomy with aortobicaval cannulation and normothermic cardiopulmonary bypass (CPB). Cold blood cardioplegia delivered directly into the coronary ostia and topical ice slush were used for myocardial protection.
After the cardioplegia was delivered, the mitral valve was approached through an incision just behind the interatrial groove. Depending upon the morphology of the mitral valve, a decision was first taken to repair or replace the valve. If the valve was heavily calcified with shrunken leaflets and severe subvalvular fusion and was deemed unsuitable for repair, it was replaced using a mechanical valve with full chordal perseveration. If mitral valve repair was performed, no more than mild mitral regurgitation was accepted. A variety of techniques described by us previously were used for mitral valve repair [11–16]. These include commissurotomy in 56 patients, cuspal thinning in 68, chordal shortening in 12, cusp excision/plication in 5, cleft suture in 7, decalcification in 26, chordal transfer in 5, and neochordal construction in 2. Mitral valve annuloplasty (n ⫽ 42) was performed using a C-shaped ring of polytetrafluoroethylene felt as described earlier [13]. There was significant difference in both the groups in amount of calcification at the mitral valve (p ⬍ 0.05), and patients in group 2 had more severe mitral valve calcification. A decalcification procedure was done in 29 patients in group 1, but the calcification was typically mild and mostly located at posterior mitral leaflet tip, whereas the rest of the valve was thickened but pliable. After completing mitral valve repair, the mitral valve was tested by injecting cold saline into the left ventricular cavity to observe coaptation of the leaflets. However, the final assessment of the repair was made only by TEE after terminating CPB. After mitral valve repair, the aortic valve was replaced with mechanical valve using standard surgical techniques. Before termination of CPB, the left atrial appendage was routinely ligated in all patients. After termination of CPB, the valve function was assessed in all patients undergoing mitral valve repair. Before discharge from the hospital, transthoracic echocardiography was done in all patients and was repeated at 6-month intervals.
Anticoagulation Protocol All patients received acenocoumarin and life-long aspirin, starting on the first postoperative day. The dose of Table 2. Causes of Early and Late Death AVR ⫹ MVR (n)
DVR (n)
Early deaths Arrhythmia Bleeding Sepsis CHF Late deaths CVA Bleeding CHF Stuck valve
4 1 1 1 1 5 1 1 3 —
25 5 4 4 12 26 4 6 8 8
AVR ⫽ aortic valve replacement; CVA ⫽ cerebrovascular accident; MVR ⫽ mitral valve repair.
CHF ⫽ congestive heart failure; DVR ⫽ double valve replacement;
TALWAR ET AL AV REPLACEMENT, MV REPAIR VS DVR
1221
Fig 1. Actuarial survival in patients undergoing either aortic valve replacement (AVR) with mitral valve (MV) repair (group 1, hash marks) or double valve replacement (DVR; group 2, triangles).
acenocoumarin was titrated to maintain an international normalized ratio (INR) of 2.5 to 3.5 in patients undergoing DVR and 2.0 to 3.0 for patients undergoing aortic valve replacement with mitral valve repair according to previously published guidelines for antithrombotic therapy in patients with prosthetic heart valves [17]. In follow-up, these measurements were obtained after discharge from the hospital at 1 week, 1 month, 3 months, and then at 6-month intervals at our hospital, if the INR was stable and in the desired range. Because the patients come from distant regions from all over India, we insisted that these patients get routine measurements of INR near their place of residence at least once a month, and more often if indicated.
Statistical Analysis Data for continuous and categoric variables were collected. Mean and median were calculated for the continuous variables, and frequency distribution with percentage were calculated for categoric variables. The Student t test and Mann Whitney U test were applied to see significant difference between categoric and continuous variables, and the 2 test was applied to see association between categoric variables. To see significant difference in survival according to specific categories, Kaplan-Meier survival analysis was performed. A large number of factors were analyzed as predictors of early and late mortality, reoperation, and event-free survival. These included age, sex, type of mitral valve lesion, (mitral regurgitation or mixed), preoperative New York Heart Association (NYHA) functional class, preoperative congestive heart failure, atrial fibrillation, left ventricular dysfunction (ejection fraction of less than 0.50), CPB and aortic cross-clamp time, and
mitral valve repair versus replacement. All these variables were subjected to a stepwise multiple logistic regression analysis to determine the independent predictors of early and late mortality, reoperation, and eventfree survival. The Cox proportional hazard model was used to determine the prognostic factors for development of significant mitral regurgitation. The SPSS 11.5 statistical package (SPSS Inc, Chicago, IL) was used for the analysis. A value of p ⬍ 0.05 was considered as significant. A valve-related event was defined according to established guidelines [18], and these are summarized in the Appendix.
Results Mortality There were 29 (7.9%) early deaths: four (5.3%) in group 1 and 25 (8.5%) in group 2 (p ⫽ 0.35). There were five late deaths in group 1 and 26 late deaths in-group 2. The causes of deaths are listed in Table 2. Actuarial survival at 60 months was 90.5% ⫾ 3.4% in group 1 and 81.60% ⫾ 2.3% in group 2 (log-rank, 3.2; p ⫽ 0.07; Fig 1).
Follow-Up The mean follow-up was 69 ⫾ 41 months in-group 1 and 52 ⫾ 36.9 months (range, 6 to 132 months) in group 2 (p ⬎ 0.05). Follow-up was 96% complete in group 1 and totaled 414 patients years and was 92% in group 2 and totaled 1161.3 patients years. Among survivors in group 1, 33 (45.8%) were followed up for 8 years or more, 45 (62.5%) for 5 years or more, 58 (80.5%) for 3 years or more, 66 (91.6%) for 2 years or more, and 72 (96%) for 1 year or more. In group 2, 56 (20%) were followed up for 8 years or
CARDIOVASCULAR
Ann Thorac Surg 2007;84:1219 –25
1222
TALWAR ET AL AV REPLACEMENT, MV REPAIR VS DVR
Ann Thorac Surg 2007;84:1219 –25
CARDIOVASCULAR
Fig 2. Reoperation-free survival in patients undergoing aortic valve replacement (AVR) with mitral valve (MV) repair (group 1, hash marks) or double valve replacement (DVR; group 2, triangles).
more, 134 (49%) for 5 years or more, 201(73%) for 3 years or more, 232 (85%) for 2 years or more, and 270 (92%) for 1 year or more.
Thromboembolism A thromboembolic complication developed in 8 patients (10.5%) in group 1 and in 62 (21%) in group 2 (p ⫽ 0.037). Four were in atrial fibrillation (AF) in group 1 and 16 were in atrial fibrillation in group 2. In group 1, 7 patients (9.7%) had minor events, including peripheral embolism, transient ischemic event, and syncope. All of these patients were alive at the last follow-up. One patient had dense hemiparesis and died. Fig 3. Event-free survival in patients undergoing aortic valve replacement (AVR) with mitral valve (MV) repair (group 1, hash marks) or double valve replacement (DVR; group 2, triangles).
In group 2, 52 patients (17 %) had minor events. Major events occurred in 10 patients (3.4%), and 4 patients died. Of the remaining 6 patients, 2 recovered, and the remaining 4 patients continue to have hemiparesis. No obvious cause (eg, left atrial clot, atheroma) was found in these patients.
Hemorrhage Bleeding complications developed in 29 patients (38.2%) in group 1 and in 148 (50.3%) in group 2 that directly related to anticoagulation (p ⫽ 0.138). In group 1, 3 patients (4.1%) had major bleeding episodes, which included gastrointestinal hemorrhage in
2. One patient died, and 1 patient had epistaxis but recovered uneventfully. Minor hemorrhage occurred in 26 other patients (36%) in this group, but no intervention was required. In group 2, there were 12 (4.1%) major bleeding episodes: 3 patients had gastrointestinal hemorrhage, 2 had hemoptysis, 4 had epistaxis, and 3 had intracranial hemorrhage. Six patients died, and the rest were successfully managed conservatively.
Infective Endocarditis A total of 8 patients had postoperative infective endocarditis: 2 in group 1 and 6 in group 2 (p ⫽ 0.107). All except 3 patients in group 2 had evidence of preoperative endocarditis. All these patients were treated conservatively.
Reoperation Eight patients underwent reoperation: 7 (9.2%) in group 1 and 1 (0.3%) in group 2 (p ⬍ 0.01). In group 1, 4 patients requiring reoperation had documented evidence of recurrent rheumatic fever with progressive annular dilatation and progressive mitral regurgitation. Three patients who underwent reoperation within 24 months were found to have dehiscence of the annuloplasty felt and annular dilatation without evidence of rheumatic activity. Seven reoperations in group 1 were for significant mitral regurgitation (n ⫽ 6) and severe mitral stenosis (n ⫽ 1) at a mean of 65 months after the initial operation (range, 12 to 108 months). One patient died after reoperation of low cardiac output syndrome. The rest recovered uneventfully. The reoperation in group 2 was required for significant mitral prosthesis paravalvular leak 8 months after the initial operation. Actuarial freedom from reoperation was 92.5% ⫾ 0.4% in group 1 and 99.5% ⫾ 0.5% in group 2 at a median follow-up of 60 months (log-rank ⫽10.22; p ⫽ 0.014; Fig 2).
Valve Function Of the 268 patients in group 2 who survived the initial operation, 1 patient required reoperation for significant mitral paravalvular leak. Prosthetic valve thrombosis was encountered in 36 patients (13.4%) and affected the mitral valve in 28, and the aortic valve in 8. Eleven patients had prosthetic valve thrombosis within 1 year of the operation, 11 between 13 and 24 months, 8 between 25 and 48 months, 4 between 49 to 72 months, and 2 between 73 and 84 months. The INR was suboptimal in 34 of these patients and was in the normal range in 2 patients. All these patients underwent thrombolysis, after which 28 patients recovered, but 8 died awaiting reoperation. The patients who died had prosthetic valve thrombosis in the mitral position. In group 1, thrombosis of the prosthetic aortic valve was encountered in 3 patients at 44, 48, and 72 months after surgery. These 3 patients had suboptimal INR and responded to thrombolysis. Of the 72 survivors in group 1, 48 (67%) had no significant mitral regurgitation, 17 (24%) had moderate, and 6 (9%) had severe mitral regur-
TALWAR ET AL AV REPLACEMENT, MV REPAIR VS DVR
1223
gitation. Patients with moderate mitral regurgitation are in NYHA class II and are being followed up. All the patients with severe mitral regurgitation have undergone reoperation as described, and 1 has died. In addition to these patients, 1 patient underwent reoperation for significant mitral stenosis. Freedom from mitral regurgitation at 60 months was 89.9% ⫾ 0.2% in group 1.
Functional Class In group 1, 26 patients (34.7%) were in NYHA class I, 32 (44.4%) were in class II, and 14 (19.4%) were in class III at follow-up. In group 2, 128 patients (47.4%) were in class I, 111 (41.1%) were in class II, and 31 (11.5%) were in class III/IV. Event-free survival (freedom from all major and minor events) at 60 months was 78.3% ⫾ 5.1% in group 1 and 48.4% ⫾ 3.2% in group 2 (p ⬍ 0.001). The patients were free of events for a mean 17.03 ⫾ 26 months in group 1 and 10.55 ⫾ 17.06 months in group 2 (Fig 3). Multivariate analysis found no specific predictors of early or late mortality or reoperation; however, mitral valve replacement was associated with a poorer event free survival (log-rank, 14.86; p ⫽ 0.0001).
Comment The two groups in this retrospective study were comparable in age, sex, etiology, distribution of lesions, congestive cardiac failure, left ventricular ejection fraction, atrial fibrillation, infective endocarditis, and NYHA class (p ⬎ 0.05). All the patients in our series had RHD. There were no significant differences in number and cause of early death between the two groups. The difference in late deaths (4 in group 1 versus 26 in group 2) appears striking but is not statistically significant; thus, our results do not seem to indicate that mitral valve repair with aortic valve replacement is better in terms of actuarial survival compared with DVR. The reoperation rates after mitral valve repair have been reported at 1% to 18% [7, 11–14, 16]. In this study, the higher reoperation rate in group 1was due to a combination of recurrent episodes of rheumatic fever and technical issues. In group 1, 4 patients requiring reoperation had documented evidence of recurrent rheumatic fever with progressive annular dilatation and progressive mitral regurgitation. Three patients who underwent reoperation within 24 months were found to have dehiscence of the annuloplasty felt and annular dilatation, without evidence of rheumatic activity. Therefore, this may not be in agreement with other studies that have shown freedom from reoperation to be as low as 5% [19] to 65% on follow-up [5, 20] for the mitral valve due to recurrent rheumatic fever and have justified routine mitral valve replacement. Group 2 had a lower reoperation rate than group 1. There were no episodes of structural deterioration, and only 1 patient had periprosthetic leak. Eight more patients were definite candidates for reoperation, but they died before reoperation could be done. Thus, the significantly smaller number of reoperations in group 2 could be fallacious.
CARDIOVASCULAR
Ann Thorac Surg 2007;84:1219 –25
1224
TALWAR ET AL AV REPLACEMENT, MV REPAIR VS DVR
CARDIOVASCULAR
Gillinov and colleagues [21] clearly demonstrated a survival advantage of mitral valve repair in patients with double valve procedures, presumably due to the younger age of the RHD patients. Their study, however, did not take into consideration other variables such as thromboembolism, infective endocarditis, bleeding, and other valve-related events. Similar results have also been reported by others [7, 19, 22], who have related higher complications in the DVR group to higher doses of anticoagulation required in these patients. Event-free survival in our series was significantly better in group 1 than in group 2. We observed a higher incidence of thromboembolic complications in group 2. Most of these events were minor, and no obvious cause was detected. Such high rates of thromboembolism were also quoted by Armenti and colleagues [23], because they also recorded all major and minor events. Most of the other reported series have lower rate of thromboembolism because only major thromboembolic episodes were recorded. The incidence of prosthetic valve dysfunction was significantly different in the two groups. All the patients in group 1 with thrombosis of the prosthetic aortic valve responded to thrombolysis. Of 36 patients in group 2 who had a stuck valve, 28 occurred at the mitral position, and 8 of these died. The rest of the patients recovered with thrombolysis. A higher incidence of mitral valve thrombosis with higher mortality was noted in group 2. All deaths in these patients were related to mitral prosthesis, which argues in favor of mitral valve repair if otherwise feasible. This is in agreement with previous experience that thromboembolic complications and prosthetic valve thrombosis are increased with mitral valve replacement compared with repair and this reduces the survival [9]. A higher INR is recommended for patients undergoing mitral valve replacement or DVR compared with isolated aortic valve replacement [17, 24]. It is also well known that long-term survival after prosthetic valve replacement is directly related to the hazards of anticoagulation. Also, the actuarial survival is not reduced despite a higher reoperation rate in these patients (Fig 3), because in the present era, the reoperations are safe and do not significantly increase the risk of death [11]. Thus, in few subset of patients who are unable to adhere to a more intensive anticoagulation regimen and do not come for regular follow-up or require lower INR, there may be a survival advantage of mitral valve repair and aortic valve replacement compared with DVR, despite a higher reoperation rate. Although more than 50% of patients in group 2 had a bleeding complication, the difference was not statistically significant. We also did not find any significant difference in the incidence of infective endocarditis between the two groups, because all except 2 patients had evidence of preoperative endocarditis. This study is of retrospective nature and the patients were not randomized to either group. Intraoperative evaluation of the mitral valve on TEE or operative findings were used to decide whether the mitral valve should be repaired or replaced. This itself introduces a bias, wherein the repair was offered to patients who had
Ann Thorac Surg 2007;84:1219 –25
pliable, noncalcified or minimally calcified mitral valves. Replacement was undertaken for calcified nonpliable mitral valves where repair was not considered an option. Considering the benefits of valve repair over replacement, it may not be ethically justifiable to conduct a prospective randomized study of these patients. Mitral valve repair with aortic valve replacement is superior to DVR in terms of better event-free survival, but there is no survival benefit. We observed a higher incidence of thromboembolic complications in the DVR group and higher reoperation rates in the mitral valve repair and aortic valve replacement group. The better event-free survival in patients undergoing mitral valve repair and aortic valve replacement still argues in favor of repair of mitral valve when ever possible.
References 1. Nakano K, Koyanagi H, Hashimoto A, et al. Twelve year’s experience with the St Jude Medical valve prosthesis. Ann Thorac Surg 1994;57:697–703. 2. Baudet EM, Puel V, McBride JT, et al. Long-term results of valve replacement with the St. Jude Medical prosthesis. J Thorac Cardiovasc Surg 1995;109:858 –70. 3. Kinsley RH, Antunes MJ, Colsen PR. St Jude Medical valve replacement: an evaluation of valve performance. J Thorac Cardiovasc Surg 1986;92:349 – 60. 4. Hartz RS, Fisher EB, Finkelmeier B, et al. An eight-year experience with porcine bioprosthetic cardiac valves. J Thorac Cardiovasc Surg 1986;91:910 –7. 5. Grossi EA, Galloway AC, Miller JS, et al. Valve repair versus replacement for mitral insufficiency: when is a mechanical valve still indicated? J Thorac Cardiovasc Surg 1998;115:389 –96. 6. Gillinov AM, Blackstone EH, White J, et al. Durability of combined aortic and mitral valve repair. Ann Thorac Surg 2001;72:20 –7. 7. Kaul P, John S, John CN, et al. Aortic valve replacement with concomitant open mitral valvotomy— early results and longterm follow up in 72 consecutive patients. Indian Heart J 1993;45:113–5. 8. Szentpetery S, Rich JB, Azar H, Newton JR, Tenzer MM. Mitral valve repair combined with aortic valve replacement. J Heart Valve Dis 1997;6:32– 6. 9. Elmcke F, Nanda NC, Hsiung MC, et al. Color Doppler assessment of mitral regurgitation with orthogonal planes. Circulation 1987;75:175– 83. 10. Grossman W. Profiles in valvular heart disease. In: Baim DS, Grossman W, eds. Cardiac catheterization, angiography, and intervention. 5th ed. Baltimore, MD: Willams & Wilkins; 1996:742– 4. 11. Choudhary SK, Talwar S, Dubey B, Chopra A, Saxena A, Kumar AS. Mitral valve repair in a predominantly rheumatic population. Long-term results. Tex Heart Inst J 2001;28:8 –15. 12. Duran CG, Revuelta JM, Gaite L, Alonso C, Fleitas MG. Stability of mitral reconstructive surgery at 10–12 years for predominantly rheumatic valvular disease. Circulation 1988;78:I91–6. 13. Kumar AS, Rao PN, Saxena A. Results of mitral valve reconstruction in children with rheumatic heart disease. Ann Thorac Surg 1995;60:1044 –7. 14. Kumar AS, Kumar RV, Shrivastava S, Venugopal P, Sood AK, Gopinath N. Mitral valve reconstruction: early results of a modified cooley technique. Tex Heart Inst J 1992;19:107–11. 15. Kumar AS, Rao PN. Restoration of pliability to the mitral leaflets during reconstruction. J Heart Valve Dis 1995;4:251–3. 16. Kumar AS, Bhan A, Kumar RV, Shrivastava S, Sood AK, Gopinath N. Cusp-level chordal shortening for rheumatic mitral regurgitation: early results. Tex Heart Inst J 1992;19:47–50.
Ann Thorac Surg 2007;84:1219 –25
1225
24. Kulik A, Rubens FD, Wells PS, et al. Early postoperative anticoagulation after mechanical valve replacement: a systematic review. Ann Thorac Surg 2006;81:770 – 81.
Appendix Events Considered for Calculating Event-Free Survival ● ● ● ● ● ● ● ● ● ● ● ● ●
Early death Late death Structural valve deterioration Significant mitral stenosis or regurgitation Prosthetic valve thrombosis Paravalvular leak Hemolytic anaemia related to valve lesions Any new temporary or permanent focal or global neurologic deficit Transient ischemic attack Peripheral embolic event Bleeding events: All major and minor hemorrhage Infective endocarditis Reoperation
CARDIOVASCULAR
17. Stein PD, Alpert JS, Dalen JE, Horsekotte D, Turpie AG. Antithrombotic therapy in patients with mechanical and biologic prosthetic heart valves. Chest 1998;114:602:S-610. 18. Edmunds LH Jr, Clark RE, Cohn LH, Grunkemeier GL, Miller DC, Weisel RD. Guidelines for reporting morbidity and mortality after cardiac valvular operations. J Thorac Cardiovasc Surg 1996;112:708 –11. 19. Hamamoto M, Bando K, Kobayashi J, et al. Durability and outcome of aortic valve replacement with mitral valve repair versus double valve replacement. Ann Thorac Surg 2003;75:28–34. 20. Kuwaki K, Kawaharada N, Morishita K, et al. Mitral valve repair versus replacement in simultaneous mitral and aortic valve surgery for rheumatic disease. Ann Thorac Surg 2007; 83:558 – 63. 21. Gillinov AM, Blackstone EH, Cosgrove DM, et al. Mitral valve repair with Aortic valve replacement is superior to double valve replacement. J Thorac Cardiovasc Surg 2003; 125:1372– 87. 22. Tri Ho HQ, Nguyen VP, Phan KP, Pham NV. Mitral valve repair with Aortic valve replacement in rheumatic heart disease. Asian Cardiovasc Thorac Ann 2004;12:341–5. 23. Armenti F, Stephenson LW, Edmunds LH Jr. Simultaneous implantation of St. Jude Medical aortic and mitral prostheses. J Thorac Cardiovasc Surg 1987;94:733–9.
TALWAR ET AL AV REPLACEMENT, MV REPAIR VS DVR