|
|
||||||||
Ann Thorac Surg 2006;82:847-852
© 2006 The Society of Thoracic Surgeons
a Division of Cardio-Thoracic Surgery, University Hospital Basel, Basel, Switzerland
b Institute of Social and Preventive Medicine, University of Basel, Basel, Switzerland
Accepted for publication April 18, 2006.
* Address correspondence to Dr Baykut, Division of Cardio-Thoracic Surgery, University Hospital Basel, Spitalstrasse 21, CH-4031 Basel, Switzerland (Email: baykutd{at}uhbs.ch).
| Abstract |
|---|
|
|
|---|
METHODS: Between January 1993 and December 2003, 601 ATS valves (393 aortic valve replacement [AVR], 168 mitral valve replacement [MVR], 20 aortic plus mitral valve replacement [DVR]) were implanted in 581 patients (377 male, 204 female; mean age, 63.7 years; range 18 to 89). Preoperatively, 47 (8%) were New York Heart Association class I, 212 (36.5%) were class II, 267 (46%) were class III, and 55 (9.5%) were class IV. Preoperative comorbidities were coronary artery disease, 167 (29%); diabetes mellitus, 72 (12%); cerebrovascular disease, 2 (0.3%); endocarditis, 53 (9%); and atrial fibrillation, 115 (20%).
RESULTS: Follow-up is 99% complete. Data represent 2,500 cumulative patient-years. Mean follow-up is 4.3 ± 2.6 years (range, 0.1 to 11.6). Overall hospital mortality is as follows: AVR, 3.8% (15 patients); MVR, 1.8% (3 patients); DVR, 10% (2 patients). Structural valve failure was not encountered. Overall survival at 10 years is AVR, 84.7% ± 3.1%, AVR plus coronary artery bypass, 67.5% ± 8.2%; MVR, 59.8% ± 7.1%, MVR plus coronary artery bypass, 39% ± 27.8%; and DVR, 74.3% ± 10%. Freedom from valve-related death at 10 years is AVR, 99.2%; MVR, 94.6%; and DVR, 100%. Linearized rates for postoperative complications are paravalvular leak, 0.6% per patient-year; valve thrombosis, 0.04% per patient-year; thromboembolism, 1.1% per patient-year; major bleeding, 0.5% per patient-year; and de novo prosthesis endocarditis, 0.1% per patient-year. Postoperative mortality risk was significantly elevated by diabetes (p < 0.01), but not by other comorbidities.
CONCLUSIONS: Our 11-year experience demonstrates low rates of adverse events and valve-related complications with the ATS Open Pivot heart valve.
| Introduction |
|---|
|
|
|---|
| Patients and Methods |
|---|
|
|
|---|
Patient age ranged from 18 to 89 years, with a mean of 63.7 years. There were 377 males and 204 females. Before surgery, all patients received cardiac catheterization including coronary angiography. Preoperatively, 47 (8%) patients were classified in New York Heart Association (NYHA) class I, 212 (36.5%) in class II, 267 (46%) in class III and 55 (9.5%) in class IV. Elective surgery was performed in 97% of patients, and 3% underwent urgent or emergent surgery. In 167 patients (29%), concomitant coronary artery disease with significant stenosis of at least one vessel was diagnosed. Seventy-two patients (12%) presented with concomitant diabetes mellitus. Fifty-three patients (9%) had a history of preoperative endocarditis; none of these patients had active endocarditis at time of surgery. Preoperative atrial fibrillation was observed in 115 cases (20%): 10% of AVR patients, 42% of MVR patients, and 30% of DVR patients. Demographic patient data are presented in Table 1. Of the 393 AVR patients, predominant aortic stenosis was diagnosed in 72%, aortic regurgitation in 23%. Combined aortic stenosis and regurgitation were observed in 5% of the patients. Of the 168 MVR patients, predominant mitral stenosis was present in 5%, mitral regurgitation in 91%. Combined mitral stenosis and regurgitation were observed in 4% of the patients.
|
Patient Follow-Up
Current patient health status and outcome events were ascertained by telephone contact with the patients themselves or family physicians if not collected on a routine basis through our Quality Management Database (Bundesgeschaeftsstelle Qualitaetssicherung, Germany; and Patient Analysis and Tracking System, United Kingdom). Reported events were corroborated through medical records. A standardized data collection form for patient self-reporting of adverse events and quality of life was used. Additional information about relocated and deceased patients was provided by the Swiss Federal Bureau of Statistics, whenever possible. Only 6 of 581 patients (1%) were lost to follow-up.
Statistical Analysis
Score and quantitative variables were summarized in terms of means and standard deviations. For comparison between ATS subgroups, the Wilcoxon-Mann-Whitney test, Fisher's exact test, and the
2 test were used. Linearized rates are expressed in percent per patient-year. Time to event variables with censored values were described by Kaplan-Meier statistics, and differences between groups assessed with the log-rank test. The defined endpoints were early mortality, overall mortality, and valve-related early and late complications. Analyses were performed using SAS version 8.2 [8].
| Results |
|---|
|
|
|---|
Follow-up is 99% complete, with 2,500 cumulative patient years represented. Mean follow-up is 4.3 ± 2.6 years (range, 0.1 to 11.6).
Mortality
Overall hospital mortality was 3.4%: AVR, 3.8% (15); MVR, 1.8% (3); and DVR, 10% (2). The postoperative mortality risk was significantly increased only in patients with diabetes mellitus (p < 0.01) but not in patients with further concomitant disorders, including coronary artery disease. Overall survival at 10 years was 84.7% ± 3.1% for isolated AVR and 67.5% ± 8.2% for AVR + CABG (p = 0.048); 59.8% ± 7.1% for isolated MVR and 39% ± 27.8% for MVR + CABG (p = 0.469); and 74.3% ± 10% for DVR (Fig 1). No patient underwent DVR + CABG. Freedom from valve-related death at 10 years was 99.2%, 94.6%, and 100% for AVR, MVR, and DVR, respectively (Fig 2). Causes of valve-related deaths (7 patients) are found in Table 2.
|
|
|
|
Nonstructural Dysfunction
Paravalvular leak was observed in 16 patients (AVR, 10; MVR, 6). Among these patients, 3 had a history of endocarditis. Six patients required reoperation. Two valves were repaired in the second operation, and 4 were replaced. In 8 cases, minor leaks with no clinical consequences were observed. The linearized rate for paravalvular leak was 0.6% per patient-year for the entire group (0.6% per patient-year for AVR, and 0.7% per patient-year for MVR). The actuarial freedom from paravalvular leak at 10 years was 96.1% ± 1.3%, 92.4% ± 3.4%, and 100% for AVR, MVR, and DVR, respectively.
Valve Thrombosis
Valve thrombosis was encountered in only 1 patient: an 84-year-old woman with MVR + CABG at 5 years after surgery. Before the event, the anticoagulation with coumarin was discontinued and acetylsalicylic acid treatment was initiated. No reason for this change was documented. The INR level at the time of the event was unknown. The subsequent mitral valve thrombosis generated a severe cerebral embolization. The patient died in hospital 4 months after the event. The linearized rate for valve thrombosis was 0.04% per patient-year for the entire group (0% per patient-year for AVR, and 0.1% per patient-year for MVR). The actuarial freedom from valve thrombosis at 10 years was 100%, 98.8% ± 1.2%, and 100% for AVR, MVR, and DVR, respectively.
Embolism
Thromboembolic events including neurologic complications were observed in 28 patients (AVR, 17; MVR, 11). Thirteen of these patients were rehospitalized, 1 of whom died after a stroke. Two patients were identified with transient and 1 with permanent neurologic deficits. In 14 cases, no additional consequence of the thromboembolic complications occurred. One of these patients died during further follow-up of unknown cause. The linearized rate for thromboembolic events was 1.1% per patient-year for the entire group: 1.1% per patient-year for AVR and 1.3% per patient-year for MVR (Fig 4).
|
|
| Comment |
|---|
|
|
|---|
In the mitral position, Westaby and colleagues [11] found that a particular orientation is not essential in the majority of patients. The main goal is to preserve the subvalvular apparatus and to provide free motion of the leaflets [9, 17, 18]. With the low profile of the ATS valve housing, minimal leaflet exposure and appropriate clearance of the posterior wall is guaranteed when preserving the subvalvular mitral apparatus.
Our overall early mortality rate of 3.4% was comparable to previously published data on other bileaflet valve series of similar size and follow-up including St. Jude Medical, CarboMedics, On-X, or Sorin Bicarbon valves [14, 1928].
Ten-year actuarial overall survival rates in AVR and DVR are comparable with those of other bileaflet valve models [14, 2729]. The freedom from valve-related mortality was 99.2% in the AVR group, 94.6% in the MVR group, and 100% in the DVR group. Our retrospective analysis demonstrated that the actuarial freedom from valve-related mortality with ATS valves was at least comparable with the results of similar bileaflet prostheses in any position. In a 10-year follow-up of CarboMedics valves, Tominaga and coworkers [14] reported 92.7%, 85.4%, and 94.7% in the AVR, MVR, and DVR groups respectively. Borman and colleagues [27] presented 95.5% overall valve-related mortality with Sorin Bicarbon valve after 8 years. Similarly, in the 18-year follow-up of Lund and associates [29] with standard St. Jude Medical prostheses, the 10-year freedom from valve-related deaths was 96%, with an overall survival rate of 58% in all positions. In our MVR group, the 10-year survival rate was 59.8% for isolated MVR and 39.0% for MVR and CABG, resulting in an overall 10-year survival for the entire MVR group of 50.1%.
An interesting observation in our study was that MVR + CABG patients demonstrated a slightly greater survival benefit the first 5 years after surgery with an overall survival rate of 83.9% ± 6.1%, compared with 79.6% ± 3.9% of patients with isolated MVR. We suspect this due to the fact that in the second half of the 10-year postoperative period, the concomitant coronary artery disease becomes responsible for increased mortality in this subgroup. In fact, the MVR patient population at our institution is composed of patients who are not candidates for mitral valve repair. These patients were in a more critical clinical stage than AVR patients at the time of the operation: 76% of MVR patients were in NYHA class III or higher compared with only 47% in the AVR group (p < 0.001).
The linearized rate for thromboembolic events of 1.1% per patient-year represents very good results compared with other experience with ATS valves: Emery and coworkers [1, 2] reported 2.7% per patient-year in a 5-year follow-up with 342 patients, and in a 6.5-year follow-up 1.85% and 3.19% per patient-year for AVR and MVR, respectively. Tominaga and associates [14] observed a linearized thromboembolism rate of 1.45% to 1.78% per patient-year in AVR and MVR patients, respectively, with CarboMedics valves. Ikonomidis and colleagues [28] reported a linearized rate for thromboembolic events of 2% per patient-year for AVR and 3.4% per patient-year for MVR patients with St. Jude Medical valves. In the study of Borman and colleagues [27] with the Sorin Bicarbon prostheses, this variable was given as 1.2% per patient-year [27] which was comparable with our results. One possible explanation of our low rates is the careful and thorough follow-up surveillance by general practitioners in this region of Switzerland, and the use of self-management, leading to better targeting of the anticoagulation corridor, which may therefore present additional health benefits for this patient cohort.
Paravalvular leak and prosthesis endocarditis are rare complications with modern bileaflet prostheses. Our linearized rates of freedom from both these complications were very high and comparable with those reported in other studies [2, 2729]. Our patient group demonstrated no evidence of any valve-related complication originating from a structural defect of the implanted prosthesis.
An interesting observation was that, after an initially high morbidity and mortality, the midterm outcomes of our DVR patients were better than those published by other authors [14, 28]. Even though the total number of patients in our DVR group is low, the results allow the conclusion that patients surviving the high perioperative risk of the double-valve procedure have a good probability of midterm survival. Another reason for better midterm results of DVR patients in our follow-up is that this group had a lower mean age than others at the time of the operation (55.3 years versus 64 in AVR and 63.9 years in MVR patients, p = 0.021).
In conclusion, based on our current experience with 601 prostheses implanted since 1993, we believe that the ATS Open Pivot valve provides a very low incidence of adverse events and valve-related complications.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Sezai, M. Hata, T. Niino, I. Yoshitake, Y. Kasamaki, A. Hirayama, and K. Minami Fifteen years of experience with ATS mechanical heart valve prostheses J. Thorac. Cardiovasc. Surg., June 1, 2010; 139(6): 1494 - 1500. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Schulenberg, P. Antonitsis, A. Stroebel, and S. Westaby Chronic Atrial Fibrillation Is Associated With Reduced Survival After Aortic and Double Valve Replacement Ann. Thorac. Surg., March 1, 2010; 89(3): 738 - 744. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Spiliopoulos, A. Haschemi, P. Parasiris, and B. M. Kemkes Sorin BicarbonTM bileaflet valve: a 9.8-year experience. Clinical performance of the prosthesis after heart valve replacement in 587 patients Interact CardioVasc Thorac Surg, February 1, 2009; 8(2): 252 - 259. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |