Since 1990, there has been a steady increase in cardiac

Transcription

1 Outcomes of Patients Undergoing Concomitant Aortic and Mitral Valve Surgery in Northern New England Bruce J. Leavitt, MD; Yvon R. Baribeau, MD; Anthony W. DiScipio, MD; Cathy S. Ross, MS; Reed D. Quinn, MD; Elaine M. Olmstead, BA; Donato Sisto, MD; Donald S. Likosky, PhD; Richard P. Cochran, MD; Robert A. Clough, MD; Richard A. Boss, Jr, MD; Robert S. Kramer, MD; Gerald T. O Connor, PhD, ScD; for the Northern New England Cardiovascular Disease Study Group Background Concomitant aortic (AV) and mitral (MV) valve surgery accounts for 4% of all valve procedures in northern New England. We examined in-hospital and long-term mortality. Methods and Results This is a report of a prospective study of 1057 patients undergoing concomitant AV and MV surgery from 1989 to The Social Security Administration Death Master File was used to assess long-term survival. Kaplan Meier and log-rank tests were performed. In-hospital mortality was 15.5% (11.0% for patients 70 years, 18.0% for 70- to 79-year-olds, and 24% for those 80 years). Overall median survival was 7.3 years. Median survival without coronary artery bypass grafting was 9.5 years and with coronary artery bypass grafting was 5.7 years (P 1). Survival in women was worse than in men (7.3 versus 9.3, years, P 0.033). Median survival by age was 11.0 years for patients 70 years, 5.4 years for 70- to 79-year-olds, and 4.8 years for those 80 years. Median survival was not significantly different for patients 80 years compared with those who were 70 to 79 years old (P 0.245). Conclusions Double-valve surgery has a high in-hospital mortality rate and a median survival of 7.3 years. After patients have survived surgery, long-term survival is similar between men and women, smaller and larger patients, and those receiving MV repair or replacement. Survival continues to decline after surviving surgery for patients 70 years old and those who undergo concomitant coronary artery bypass grafting. In patients 70 years, either mechanical valves in both positions or a tissue AV and mitral repair have the lowest in-hospital mortality and the best long-term survival. In patients 70 years, tissue valves in both positions have the best in-hospital and long-term survival. (Circulation. 2009;120[suppl 1]:S155 S162.) Key Words: cardiopulmonary bypass mortality revascularization survival valves Since 1990, there has been a steady increase in cardiac valve surgery performed for aortic and mitral disease in northern New England. 1,2 The frequency of double-valve surgery (DVS; concomitant mitral [MV] and aortic [AV] valve surgery) remains low (3% to 14%) compared with isolated valve surgery for aortic and mitral pathologies. 3 5 For these patients, surgery remains the sole and definitive treatment for relieving symptoms and extending life. Studies involving concomitant AV and MV surgery are limited compared with those for isolated AV and MV surgery. Previous reports of in-hospital mortality rates for DVS range from 9.6% to 14%. 3 5 Astor et al 6 reported an in-hospital odds ratio for mortality of 1.75 (P 1) in patients having DVS. Several studies have documented increased in-hospital mortality in older patients undergoing DVS. 3 5 Many studies have also documented that concomitant coronary artery bypass graft (CABG) surgery increases the mortality of isolated valvular heart surgery. Therefore, it is not surprising that the addition of CABG to DVS increases the in-hospital mortality rate compared with DVS alone. In the New York State Cardiac Surgery Registry, mortality was 18.7% in CABG/DVS patients compared with 9.6% for those who underwent DVS alone. 4 Other studies have documented an increased mortality in DVS patients who have a CABG procedure done concomitantly. 3,7 The long-term survival of DVS patients has not been well investigated, and there are limited data for different patient subgroups. In 1992, Galloway et al 7 from New York City documented a 78% 5-year survival in 513 patients (mean follow-up, 3.1 years) who underwent DVS. In 2003, Hamamoto et al 8 from Japan published a 15-year survival of 81% for 379 patients who had aortic (AVR) and mitral (MVR) valve replacement and of 79% for 80 patients who had AVR and MV repair (MVP). In 2007, Kuwaki et al, 9 also from Japan, reported a 12-year freedom from cardiac-related death of 81% for 47 patients who had AVR and MVR and of 79% for 80 patients who had AVR and MVP. From Fletcher Allen Health Care (B.J.L.), Burlington, Vt; Catholic Medical Center (Y.R.B.), Manchester, NH; Dartmouth-Hitchcock Medical Center (A.W.D.), Lebanon, NH; Dartmouth Institute for Health Policy and Clinical Practice (C.S.R., G.T.O.), Dartmouth College, Lebanon, NH; Maine Medical Center (R.D.Q., R.S.K.), Portland, Me; Dartmouth Medical School (E.M.O., D.S.L.), Lebanon, NH; Portsmouth Regional Hospital (D.S.), Portsmouth, NH; Central Maine Medical Center (R.P.C.), Lewiston, Me; Eastern Maine Medical Center (R.A.C.), Bangor, Me; and Concord Hospital (R.A.B. Jr), Concord, NH. Presented in part at American Heart Association Scientific Sessions 2008, November 8 12, 2008, New Orleans, La. Correspondence to Bruce J. Leavitt, MD, Fletcher Allen Health Care, 111 Colchester Ave, Burlington, VT bruce.leavitt@vtmednet.org 2009 American Heart Association, Inc. Circulation is available at DOI: /CIRCULATIONAHA S155

2 S156 Circulation September 15, 2009 Controversy exists involving the decision to repair or replace the MV when a concomitant AVR is being performing. In 1998, Grossi et al 10 from New York University Medical Center reported that MVR offered an improved 8-year freedom from late cardiac death compared with MVP in those patients who underwent concomitant AVR. Hamamoto et al 8 also suggested in 2003 that MVR should be the procedure of choice compared with MVP in patients also undergoing AVR. In 2003, Gillinov et al 11 from the Cleveland Clinic reported their experience with MVP and AVR compared with DVR. Late survival was documented out to 15 years and was improved in the repair group compared with the replacement group (P 0.01). In this study, we examined patients in our region who underwent concomitant AV and MV surgery from 1989 through In-hospital mortality and long-term survival were determined. In addition, in-hospital mortality and longterm survival were compared in patients having MVP versus MVR in patients undergoing AVR as well. Methods Setting and Patient Population The Northern New England Cardiovascular Disease Study Group is a voluntary research consortium representing all centers performing cardiac surgery in Vermont, New Hampshire, and Maine. The intent of the group is to foster continuous improvement in the quality of care of patients with cardiovascular disease. The registries contain data on all CABG procedures performed in the region since 1987 and on all valve procedures performed since The data are regularly validated to verify capture of all procedures and status at discharge. This is a prospective, regional, cohort study of 1057 consecutive patients who underwent concomitant AV and MV surgery in northern New England from 1989 through Data Collection Patient and procedure information was collected prospectively for all patients. Data included age ( 70, 70 to 79, and 80 years), sex, body surface area (BSA; 1.70, 1.70 to 1.99, and 2.00 m 2 ), comorbid conditions (vascular disease, diabetes with and without sequelae, chronic obstructive pulmonary disease, congestive heart failure, preoperative creatinine value 2, preoperative dialysisdependent renal failure, and preoperative atrial fibrillation), coronary artery disease, type of valve procedure, valve type, concomitant CABG, and priority at surgery. Valve procedure was classified as replacement or repair. Valve type was classified as tissue or mechanical. Data definitions can be found at The etiology of valvular pathologies, type of valve repair, and whether or not a chordal-sparing procedure was used for MVR were not collected. Study Measures and Outcomes The outcome measures for this study were in-hospital mortality (defined as death within the index admission) and all-cause mortality during a 10-year period. Long-term survival data were obtained by linkage to the Social Security Administration s Death Master File, US Department of Commerce Technology Administration. The sensitivity of the Social Security Administration s Death Master File (92.2%) is comparable to that of the National Death Index among American-born individuals (87% to 98%). 12 Probabilistic matching was performed on the basis of patient name (first, middle initial, and last), Social Security number, date of birth, sex, date last known alive, and state of last known residence. Presented data for patients who underwent DVS were stratified by age, MVR or MVP, sex, with or without CABG, and BSA. Statistical Analysis Survival analyses were conducted with Kaplan Meier techniques. Survival curves were compared with the log-rank statistic. Analyses were performed with STATA software (STATA Corp, College Station, Tex). Table 1. Patient and Disease Characteristics for DVS Variable No. of procedures 1057 Age, y, % by age group Sex, % female 44.1 BSA, m 2, % by group Comorbid disease Vascular disease, % yes 17.7 Diabetes, % yes 19.5 COPD, % yes 11.6 CHF, % yes 60.7 Creatinine 2 (no dialysis), % yes 5.3 Dialysis, % yes 2.4 Preoperative atrial fibrillation, % yes 38.0 Coronary artery disease, % yes 61.6 Valve procedure, % yes Aortic Replacement 98.1 Repair 1.9 Mitral Replacement 60.5 Repair 39.5 Valve type, if replacement, % yes Aortic Tissue 64.7 Mechanical 35.4 Mitral Tissue 51.2 Mechanical 48.8 Concomitant CABG, % yes 46.9 Priority at surgery, % Emergency 4.0 Urgent 50.4 Elective 45.6 COPD indicates chronic obstructive pulmonary disease; CHF, congestive heart failure. Statement of Responsibility The authors had full access to the data and take responsibility for their integrity. All authors have read and agreed to the manuscript as written. The patient consent for use of these data was obtained, and the use of the data was approved by each center s internal review board committee. Results Patient and disease characteristics for patients who underwent DVS are presented in Table 1. There were 1057 patients: 45.1% 70 years old, 41.0% 70 to 79 years old, and 13.9% 80 years old. Forty-four percent were female. BSA was categorized into 3 groups: 21.4% had a BSA 1.70 m 2, 44.0% had a BSA of 1.70 to 1.99 m 2, and 34.6% had a BSA

3 Leavitt et al Outcome After Concomitant Aortic/Mitral Valve Surgery S157 Table 2. In-Hospital and Long-Term Mortality Among AVR/MV Surgery Patients Subgroup In-Hospital Mortality, % P Value ( 2 ) Deaths/100 Patient- if Discharged Alive, % P Value (Log-Rank Test) All patients Age group, y Sex Male Female BSA, m 2 Median by sex Median by sex Concomitant CABG No Yes Mitral procedure Repair Replacement Valve type Aortic tissue/mitral tissue Aortic mechanical/mitral mechanical Aortic tissue/mitral repair of 2.00 m 2. The majority of patients presented with congestive heart failure (60.7%). Other comorbid diseases were as follows: 17.7% had vascular disease; 19.5%, diabetes with and without sequelae; 11.6%, chronic obstructive pulmonary disease; and 5.3%, a creatinine value 2.0. Few had dialysisdependant renal failure (2.4%), but 38.0% had preexisting atrial fibrillation. Most patients who had DVS had MVR instead of MVP (60.5% versus 39.5%); of those who had MVR, 51.2% received a tissue valve and 48.8% received a mechanical valve. Most AVs were tissue valves (64.7%, compared with 35.4% mechanical valves). Concomitant CABG was performed in 46.9% of DVS patients. The priority of most patients was urgent (50.4%), 45.6% were elective, and 4.0% were emergent. Median follow-up was 3.5 years. In-hospital mortality and median survival are shown in Table 2. Overall in-hospital mortality was 15.5%, and the annual incidence rate of death was 7.4 deaths/100 patientyears. In-hospital mortality and annual incidence of death were stratified by age group. The in-hospital mortality was 10.9% in patients 70 years, 17.8% for those between 70 and 79 years, and 23.8% in those who were 80 years (P 1). The annual incidence of death was 5.0 in patients 70 years, 9.9 in those 70 to 79 years, and 11.3 in those 80 years. Both in-hospital mortality and annual incidence of death were statistically significant between age groups. There was no significant difference in in-hospital mortality and annual incidence of death between sexes. When BSA was less than the median for both men and women, in-hospital mortality was 18.3% and the annual incidence of death was This difference was significantly greater than inhospital mortality and annual incidence of death for those with a BSA greater than or equal to their sex-specific median (in-hospital mortality, 12.4%, P 0.025; annual incidence of death, 7.1, P 8). Patients who underwent concomitant CABG had a significantly greater in-hospital mortality (18.8% versus 12.7%, P 7) and annual incidence of death (9.9 versus 5.7, P 1). MVP was associated with a significantly lower in-hospital mortality than MVR (11.1% versus 18.2%, P 2). However, the annual incidence of death was not significantly different (6.2 versus 8.2, P 0.061). To examine mortality and annual incidence of death by valve type, we analyzed the groups according to age category, 70 and 70 years. Then the cohort was separated into 3 groups: aortic tissue/mitral tissue, aortic tissue/mvp, and aortic mechanical/mitral mechanical. In patients 70 years, in-hospital mortality and annual incidence of death were highest in the patients who had tissue valves in both positions (23.2% and 11.0, respectively). In patients with tissue aortic valves and MVP, in-hospital mortality and annual incidence of death were 8.1% and 5.3, respectively. In patients with aortic and mitral mechanical valves, in-hospital mortality and annual incidence of death were 9.0% and 4.4, respectively. These differences were significant (P 3 and P 1). In patients 70 years of age or older, the results were different. Patients with both mechanical valves had an in-hospital mortality of 27.1% and an annual incidence of death of Patients with tissue valves in both positions had an in-hospital mortality of 22.3% and an annual incidence of death of Patients who had a tissue AV and MVP had the lowest in-hospital mortality of 12.8% and also the lowest annual incidence of death of 7.1. These differences were also significant (P and P 3).

4 S158 Circulation September 15, 2009 Median survival = 7.3 years Figure 1. Long-term survival of DVS patients. In Figures 1 through 8, median survival is shown in Kaplan Meier plots. The number of patients at risk by year is shown at the bottom of each figure. Figure 1 depicts the survival of all patients who underwent DVS. There were 1057 patients, and median survival was 7.3 years. Figure 2 illustrates the significant differences in median survival (P 1) in patients age 70 years (11.0 years), 70 to 79 years (5.4 years), and 80 years or older (4.8 years). Figure 3 shows data for patients who had an MVP compared with MVR (regardless of type of AV prosthesis). Patients with log rank test p<1 MVP had a significantly longer survival than those who had MVR (9.1 versus 6.3 years, P 1). Figure 4 depicts survival by sex. Male median survival of 9.3 years was similar to the 7.3 years for female patients (P 0.033). Figure 5 illustrates median survival by BSA based on sex-specific medians. The median BSA for men was 2.0 m 2 and for women, 1.7 m 2. A significant difference in survival was noted in those patients with a larger BSA. Their survival was 7.7 years compared with that for smaller patients, who had a survival of 5.4 years. This difference was significant Median survival <70 = 11.0 yrs Median survival = 5.4 yrs Median survival >=80 = 4.8 yrs < >=80 <70 yrs yrs >=80 yrs Figure 2. Long-term survival of DVS patients by age.

5 Leavitt et al Outcome After Concomitant Aortic/Mitral Valve Surgery S159 Median survival by MV repair = 9.1 yrs Median survival by MV replacement = 6.3 yrs log rank test p< MV repair MV replacement MV Replacement MV Repair Figure 3. Long-term survival of DVS patients by MVP vs MVR. (P 1). Figure 6 shows that the addition of CABG to the DVS procedure significantly affected median survival (P 1). Patients with concomitant CABG had a decreased survival (5.7 years) compared with 9.5 years when CABG was not performed at the time of the DVS procedure. Figure 7 depicts survival of patients 70 years stratified by surgical procedure type. Patients with tissue valves in both positions had the lowest survival compared with patients with mechanical valves in both positions and patients with a tissue AV and MVP. Figure 8 illustrates survival in patients 70 years of age. Patients with an aortic tissue valve and MVP had the best long-term survival compared with patients with mechanical and tissue prostheses in both positions. Median survival for males = 9.3 yrs Median survival for females = 7.3 yrs log rank test p= male female Male Female Figure 4. Long-term survival of DVS patients by sex.

6 S160 Circulation September 15, 2009 Median survival BSA<median by sex* = 5.4 yrs Median survival BSA>=median by sex* = 7.7 yrs log rank test p< BSA<median by sex * BSA>=median by sex * * Median BSA for men is 2.0; median BSA for women is 1.7 Figure 5. Long-term survival of DVS patients by BSA. At procedure BSA<median BSA>=median Discussion Our prospective, regional, cohort study of 1057 patients showed that patients who had undergone a concomitant valve surgery procedure had a high in-hospital mortality rate (overall 15%) and a diminished long-term survival (median survival of 7.3 years). In-hospital mortality was higher for women, patients with a smaller BSA, those who had concomitant CABG, and those who MVR instead of MVP. For patients 70 years old, in-hospital mortality was less than that for those between 70 and 79 years and even less than for those 80 years. For patients who were alive at discharge, there were significant differences in long-term survival based on age, BSA, concomitant CABG, and type of valve procedure. Sex and whether the mitral valve was repaired or replaced did not affect long-term survival. Our in-hospital mortality rates are similar to those published by Hannan et al, 4 Galloway et al, 7 and Hellgren et al 5 but are higher than those reported by Gillinov et al 11 from the Cleveland Clinic. Older age clearly affects both in-hospital and long-term survival in patients undergoing DVS. We showed that patients 70 years had a significantly lower in-hospital mortality rate Median survival without CABG = 9.5 yrs Median survival with CABG = 5.7 yrs log rank test p< no CABG with CABG No CABG With CABG Figure 6. Long-term survival of DVS patients with or without concomitant CABG.

7 Leavitt et al Outcome After Concomitant Aortic/Mitral Valve Surgery S Figure 7. Long-term survival by valve type and age 70 years. AVR tissue/mvr tissue AVR tissue/mv repair AVR mech/mvr mech AV tissue/mv At procedure tissue AV tissue/mv repair AV mech/mv mech (11%) than those patients between 70 and 79 years (18%) and those who were 80 years (24%). However, survival was not significantly different for 80-year-old patients versus those 70 to 79 year old patients if they survived surgery. Patients who underwent AVR and MVP had a lower inhospital mortality compared with those with AVR and MVR (11% versus 18%). Median survival was shorter in the MVR group, although after being discharged alive, the survival between MVP and MVR patients was similar (P 0.061). Our study results are different from those of Gillinov et al 11 that MVP had improved long-term survival compared with MVR in patients who underwent concomitant AVR. Hamamoto et al, 8 Kuwaki et al, 9 and Grossi et al 10 concluded that MVR provided greater long-term survival compared with MVP. Our current study had patient numbers similar to those of Gillinov et al (813 patients) but more than Kuwaki et al (128 patients), Hamamoto et al (459 patients), and Grossi et al (337 patients). In-hospital and median survival rates were better in male compared with female patients in this investigation. Inhospital mortality rate for males was 13% compared with 18% for females. Median survival in males was 9.3 years compared with 7.3 years in females, but for patients who were Figure 8. Long-term survival by valve type and age 70 years. AVR tissue/mvr tissue AVR tissue/mv repair AVR mech/mvr mech AV tissue/mv At procedure tissue AV tissue/mv repair AV mech/mv mech

8 S162 Circulation September 15, 2009 at alive discharge, survival was similar for men and women (P 0.033). Four previous reports of DVS documented that sex was nonsignificant for in-hospital mortality. 4,5 CABG affects both in-hospital and median survival rates. The in-hospital mortality for AV and MV surgery alone was 13% compared with 19% with CABG. These data are similar to the short-term mortality that the addition of CABG provides as documented by Hannan et al 4 and Jamieson et al 3 in these large cardiac surgery databases. Median survival was 5.7 years in the DVS and CABG group compared with 9.5 years in the group who did not have CABG performed with the AV and MV surgery. In addition, the Kaplan Meier curves diverged, showing the continuous negative survival effects of concomitant CABG surgery (P 1). BSA also appears to play a role in both in-hospital and long-term survival of patients undergoing AV and MV surgery. Patients with a BSA lower than their sex-specific median (median BSA of men was 2.0 m 2 and for women was 1.7 m 2 ) had an in-hospital mortality rate of 12% compared with 18% in those patients who had a higher sex-specific BSA. Median survival was similarly affected, with patients having a smaller BSA having a median survival of 5.4 years compared with 7.7 years for patients with a higher BSA. For patients who survived the surgery, there was still a survival advantage for larger patients (P 8). To date, there have not been any studies published on the effect of BSA, DVS, and long-term survival. Differences also were seen in both in-hospital mortality and long-term survival by type of valve procedure and age of the patient at the time of the procedure. The groups were separated at 70 years of age. In those 70 years of age, 211 patients received a mechanical valve in both positions, 87 had a tissue AV and MVP, and 69 patients had a tissue valve in both positions. Younger patients with a tissue valve in both positions had high mortality (23%) and diminished long-term survival. In-hospital mortality and long-term survival were not different in those patients with both mechanical valves replaced or those who had an aortic tissue valve and MVP. In patients age 70 years, 229 had a tissue valve in both positions, 188 had a tissue AVR and MVP, and 48 had mechanical prostheses in both positions. In-hospital mortality was best in those with an MVP and tissue AV (13%) compared with an in-hospital mortality in those having tissue valves in both positions (22%) or mechanical valves in both positions (27%). Long-term survival appears best in patients with a tissue AV and MVP compared with those with either both tissue or both mechanical prostheses. Our study has several limitations. First, the etiology of both the AV and MV pathologies was not examined. Other studies have shown a difference between rheumatic and nonrheumatic long-term survival in patients undergoing AV and MV surgery. 10,11 Second, we did not differentiate the specific type of MVP that was performed. Third, we did not evaluate whether a chordal-sparing procedure was performed at the time of MVR, nor do we have data on cause of death. This was an observational study, and 1 or more factors that we did not adjust for may have affected survival. Conclusions Concomitant AV and MV surgery had a high in-hospital mortality rate (15%) with a median survival of 7.3 years. In-hospital mortality was higher in patients 70 years of age, of female sex, with a smaller BSA, with concomitant CABG surgery, and an MVR compared with an MVP. After surviving surgery, long-term survival was similar between men and women, smaller and larger patients, and MVP and MVR groups. Survival continued to decline after surviving surgery for patients 70 years of age and for those patients who underwent concomitant CABG. In patients 70 years of age, either both mechanical valves or a tissue AV and MVP appear to be associated with the best in-hospital mortality and long-term survival. In patients 70 years, a tissue AV and MVP have the lowest hospital mortality and best long-term survival. This study will help inform patients who are presented with the surgical option of AV and MV surgery by their cardiologist and/or cardiac surgeon. Source of Funding This study was funded by the member centers of the Northern New England Cardiovascular Disease Study Group. Disclosures None. References 1. Nowicki ER, Birkmeyer NJ, Weintraub RW, Leavitt BJ, Sanders JH, Dacey LJ, Clough RA, Quinn RD, Charlesworth DC, Sisto DA, Uhlig PN, Olmstead EM, O Connor GT. Northern New England Cardiovascular Disease Study Group and the Center for Evaluative Clinical Sciences, Dartmouth Medical School. Multivariable prediction of in-hospital mortality associated with aortic and mitral valve surgery in Northern New England. Ann Thorac Surg. 2004;77: Birkmeyer NJ, Marrin CA, Morton JR, Leavitt BJ, Lahey SJ, Charlesworth DC, Hernandez F, Olmstead EM, O Connor GT. Decreasing mortality for aortic and mitral valve surgery in Northern New England. Northern New England Cardiovascular Disease Study Group. Ann Thorac Surg. 2000;70: Jamieson WR, Edwards FH, Schwartz M, Bero JW, Clark RE, Grover FL. Risk stratification for cardiac valve replacement. National Cardiac Surgery Database Committee of the Society of Thoracic Surgeons. Ann Thorac Surg. 1999;67: Hannan EL, Racz MJ, Jones RH, Gold JP, Ryan TJ, Hafner JP, Isom OW. Predictors of mortality for patients undergoing cardiac valve replacements in New York State. Ann Thorac Surg. 2000;70: Hellgren L, Kvidal P, Ståhle E. Improved early results after heart valve surgery over the last decade. Eur J Cardiothorac Surg. 2002;22: Astor BC, Kaczmarek RG, Hefflin B, Daley WR. Mortality after aortic valve replacement: results from a nationally representative database. Ann Thorac Surg. 2000;70: Galloway AC, Grossi EA, Baumann FG, LaMendola CL, Crooke GA, Harris LJ, Colvin SB, Spencer FC. Multiple valve operation for advanced valvular heart disease: results and risk factors in 513 patients. J Am Coll Cardiol. 1992;19: Hamamoto M, Bando K, Kobayashi J, Satoh T, Sasako Y, Niwaya K, Tagusari O, Yagihara T, Kitamura S. Durability and outcome of aortic valve replacement with mitral valve repair versus double valve replacement. Ann Thorac Surg. 2003;75:28 33; discussion Kuwaki K, Kawaharada N, Morishita K, Koyanagi T, Osawa H, Maeda T, Higami T. Mitral valve repair versus replacement in simultaneous mitral and aortic valve surgery for rheumatic disease. Ann Thorac Surg. 2007;83: Grossi EA, Galloway AC, Miller JS, Ribakove GH, Culliford AT, Esposito R, Delianides J, Buttenheim PM, Baumann FG, Spencer FC, Colvin SB. Valve repair versus replacement for mitral insufficiency: when is a mechanical valve still indicated? J Thorac Cardiovasc Surg. 1998;115: ; discussion Gillinov AM, Blackstone EH, Cosgrove DM III, White J, Kerr P, Marullo A, McCarthy PM, Lytle BW. Mitral valve repair with aortic valve replacement is superior to double valve replacement. J Thorac Cardiovasc Surg. 2003;25: Schisterman EF, Whitcomb BW. Use of the Social Security Administration Death Master File for ascertainment of mortality status. Popul Health Metr. 2004;2:2.