Ann Thorac Surg 2005;80:204-209
© 2005 The Society of Thoracic Surgeons
Original article: Cardiovascular
Long-Term Results of the Cloth-Covered Starr-Edwards Ball Valve
Motomi Shiono, MD, PhD*,
Yukiyasu Sezai, MD, PhD,
Akira Sezai, MD, PhD,
Mitsumasa Hata, MD, PhD,
Mitsuru Iida, MD, PhD,
Nanao Negishi, MD, PhD
Department of Cardiovascular Surgery, Nihon University School of Medicine, Tokyo, Japan
Accepted for publication February 1, 2005.
* Address reprint requests to Dr Shiono, Department of Cardiovascular Surgery, Nihon University School of Medicine, 30-1 Oyaguchi-kamimachi, Itabashi-ku, Tokyo 173-8610, Japan (Email: mshiono{at}med.nihon-u.ac.jp).
 |
Abstract
|
|---|
BACKGROUND: Between June 1968 and March 1977, Starr-Edwards cloth-covered ball valves were exclusively implanted, in a total of 74 patients at our institution, on a routine basis.
METHODS: The retrospective postoperative follow-up rate was 84.5%, for up to 36 years in mitral valve patients (982.5 patient-years); and 81.3%, for up to 34 years in aortic valve patients (282.0 patient-years). Among 66 operative survivors, 20 patients required reoperation due to valve-related complications. Mortality and morbidity after valve replacement was reviewed, and excised valves were examined.
RESULTS: Survival rates after 10, 20, and 30 years were 74.6%, 64.1%, and 31.2%, respectively, after mitral valve replacement, and 62.5%, 50.0%, and 43.8%, respectively, after aortic valve replacement. Freedom from all valve-related complications, respectively after 10, 20, and 30 years, was 70.5%, 55.9%, and 46.4% after mitral valve replacement, and 56.2%, 37.5%, and 31.2% after aortic valve replacement. Cloth wear or pannus overgrowth was observed in all excised prostheses. Remarkable orifice tear was observed in mitral valves that were more than 20 years old. Pannus overgrowth on the studs contributed to prosthetic regurgitation in the older valves. Concomitant valve procedures were frequently required for valve-related complications or other aggravated valve lesions in the mitral position during the follow-up period.
CONCLUSIONS: Early diagnosis of valve dysfunction and the decision to reoperate are important to improve the long-term results for surviving patients who have received a cloth-covered Starr-Edwards valve, especially in the mitral position.
|
Introduction
|
|---|
After the first successful prosthetic replacement of a damaged mitral valve using the Starr-Edwards caged-ball valve [1], surgical treatment for heart valve disease advanced dramatically with introduction of the modern era of heart valve replacement. Although caged-ball valves are less widely used nowadays, the Starr-Edwards ball valve has played a significant role in the past decades. Several engineering modifications were made to enhance hemodynamic performance and fixation. The fabric covering was extended to the inflow orifice and cage, and a silastic ball was changed to a stellite ball, which resulted in a group of "cloth-covered Starr-Edwards ball valves" [2]. Between June 1968 and March 1977, the cloth-covered ball valves were exclusively implanted at our institution for mitral and/or aortic valve replacement, on a routine basis. Long-term results with the Starr-Edwards ball valve have been reported, showing satisfactory results with reliable durability and safety, and it used to be the standard in mechanical valve replacement, until recent prostheses could show significant improvement in long-term results. Studying the clinical outcome with earlier valve models is an important aspect in determining how to improve next-generation valves and the prognosis of valvular heart disease. However, there have been only a few articles reporting long-term results with exclusive use of the cloth-covered Starr-Edwards valve. This study reports long-term results in patients with the cloth-covered model of the Starr-Edwards ball prostheses.
|
Patients and Methods
|
|---|
Patient Population and Initial Operation
Valve replacements using the cloth-covered Starr-Edwards valve were performed on a total of 74 patients (41 males and 33 females), with a mean age of 32.6 ± 10.6 years (range, 10 to 58 years). Mitral valve replacement was performed in 58 patients (group M), and aortic valve replacement was performed in 16 (group A). Within the implantation time frame at our institution, several modifications of the valve were made by the manufacturer (Edwards Laboratories, Santa Ana, CA). Table 1 describes the implanted models, implantation dates, the number of each model implanted at our institution, and the modified characteristics. Only patients who required isolated mitral or aortic valve replacement were included in the study; those with double valve replacement were excluded.
View this table:
[in this window]
[in a new window]
|
Table 1. Implanted Models, Implant Dates, the Number of Implantations at Our Institution, and the Modified Characteristics of the Cloth-Covered Starr-Edwards Valves
|
|
At the initial operation, a median sternotomy was performed as the approach for operative exposure. Cardiopulmonary bypass was implemented through the femoral artery and right atrium in the earlier years; later, the ascending aorta was used for arterial perfusion. Hypothermia was employed, lowering the patient’s systemic temperature to 26–28°C. Topical cooling of the heart was achieved with cold saline solution. Cardioplegic arrest had not yet been applied, and the intracardiac maneuver was performed during ventricular fibrillation. After resection of the valve leaflets, an interrupted method using braided polyester sutures was employed without pledgets. An oral anticoagulation regimen was administered after surgery and follow-up was conducted at our outpatient clinic or a local general practitioner’s clinic.
Reoperation and Previously Implanted Valves
Among 66 operative survivors who underwent an isolated mitral valve replacement (52 patients) or aortic valve replacement (14 patients), 20 patients (11 males and 9 females) required reoperation due to valve dysfunction, thromboembolic complication, paravalvular leakage, hemolytic anemia, and/or prosthetic endocarditis. Reoperation was carried out at a mean age of 15.9 ± 9.8 years (range, 2.3 to 34.2 years) after initial valve replacement. The Starr-Edwards valves were replaced 14 times in the mitral position (13 patients), and 8 times in the aortic position (7 patients); the mean patient age was 44.7 ± 13.9 years (range, 14 to 66 years). Table 2 describes the models implanted at the initial operation, the number of reimplantations, and the causes of reoperation.
Reoperations were performed by the same group of surgeons. The heart was completely mobilized through a repeat median sternotomy, and cardiopulmonary bypass was implemented through the femoral artery or the aorta for arterial perfusion, and the right atrium for venous drainage. The patients’ systemic temperature was cooled to 32°C, and the heart was arrested using St Thomas’ cold crystalloid cardioplegic solution. Care was taken during the prosthesis excision not to remove any structures from the prosthetic surfaces to avoid artificial changes on the prosthesis. The prosthetic valve was replaced using an interrupted method with braided polyester sutures or pledgeted polyester sutures at the supraannular position.
Excised valves were carefully examined, and morbidity after initial and repeat surgery was defined according to the guidelines for reporting morbidity and mortality after cardiac valve operations as published by Edmunds and colleagues (3).
Follow-Up
Attempts were made to find and review the patients by contacting, finally, known treating physicians. The patients that could not be found were recorded as being lost to follow-up. The retrospective postoperative follow-up rate was 84.5% for up to 36 years in group M (982.5 patient-years); and 81.3% for up to 34 years in group A (282.0 patient-years).
Statistical Analyses
Continuous data were reported as mean ± standard deviation. Freedom curves from various events were obtained using the Kaplan-Meier method. Statistical analysis was carried out using StatView software version 5.0 (SAS Institute, Inc, Cary, NC).
|
Results
|
|---|
Operative Mortality and Survival
Six mitral valve patients and 2 aortic valve patients died within the first 30 days after the valve replacement, for total operative mortalities of 10.3% (M) and 12.5% (A), respectively. Causes of death were heart failure (3 patients), multiple organ failure (2 patients), bleeding (1 patient), left ventricular rupture (1 patient), and uncontrollable ventricular arrhythmia (1 patient).
The actuarial long-term survival curves, including early mortality of both mitral valve patients and aortic valve patients, are presented in Figure 1. The survival rates were 74.6% (M) and 62.5% (A) after 10 years, 64.1% (M) and 50.0% (A) after 20 years, and 31.2% (M) and 43.8% (A) after 30 years.
Complications
Among 66 survivors, thromboembolic events after valve replacement were observed in 8 mitral valve patients and 3 aortic valve patients. Freedom from thromboembolic events after the valve replacement was 90.9% (M) and 78.6% (A) after 10 years, 78.3% (M) and 78.6% (A) after 20 years, and 78.3% (M) and 78.6% (A) after 30 years, with linearized rates of 1.35% per pt-y (M) and 1.06% per pt-y (A) (Fig. 2). Prosthetic endocarditis was diagnosed in 1 mitral valve patient and 2 aortic valve patients, with freedom rates of 97.6% (M) and 85.7% (A) after 30 years. A bleeding event was observed in 1 mitral valve patient, with a linearized rate of 0.11% per pt-y (M).
Reoperation was performed in a total of 20 patients (30.3% of the operative survivors). During exclusive use of the Starr-Edwards valve, 1 mitral valve patient and 1 aortic valve patient underwent reoperation with the Starr-Edwards cloth-covered model again, for pannus stenosis and paravalvular leakage, respectively. Structural valve dysfunction was found in all 20 patients, 21 times, excluding 1 aortic valve patient who underwent reoperation for paravalvular leakage. The reasons for reoperation were valve dysfunction (9 patients), thromboembolism (8 patients), paravalvular leakage (2 patients), prosthetic endocarditis (1 patient), hemolytic anemia (1 patient), and pannus stenosis (1 patient) (Table 2). In the preoperative catheterization and/or transthoracic-transesophageal echocardiographic study before reoperation, prosthetic regurgitation was more frequently observed in the mitral position compared with the aortic. Table 3 describes concomitant valve procedures at reoperation. Tricuspid valve operation was frequently required for secondary massive tricuspid regurgitation. Two patients died due to bleeding and low output syndrome after reoperation, and the 30-day mortality for repeat implantation was 10.0%. The freedom from reoperation was 94.3% (M) and 75.0% (A) after 10 years, 83.0% (M) and 62.5% (A) after 20 years, and 78.9% (M) and 56.2% (A) after 30 years (Fig. 3). The freedom from all valve-related complications, including early mortality, was 70.5% (M) and 56.2% (A) after 10 years, 55.9% (M) and 37.5% (A) after 20 years, and 46.4% (M) and 31.2% (A) after 30 years (Fig. 4).
Excised Valve Findings
Cloth wear-tear or pannus formation was observed in all excised valves, as summarized in Table 3. As postoperative time passed, cloth degradation was markedly present at the orifice area of the prostheses. The cloth inside the struts was worn through and the cage metal exposed, while remnant cloth surrounding the outside was detached from struts and loosened. Orifice cloth was markedly worn in mitral valves compared with aortic valves, and tearing was usually observed in the older valves (more than 20 years of age; Fig 5). Pannus overgrowth was observed on the studs of the orifice and at the struts area around the orifice where the struts arose in the older valves, resulting in regurgitation through the orifice. The stellite ball could not contact the studs around the orifice because of pannus overgrowth.
View larger version (165K):
[in this window]
[in a new window]
|
Fig 5. Excised 34.2-year-old mitral valve (model 6310). Significant cloth wear and dislodgement were observed inside the struts and the orifice, showing bare metal struts and metal orifice. Pannus overgrowth was observed behind the struts and over the studs at the strut orifice. The stellite ball could not contact the studs around the orifice.
|
|
|
Comment
|
|---|
Implantations of Starr-Edwards valves, with several modifications, exceed 200,000 cases, and long-term results with these valves have been reported (1, 2, 4), showing satisfactory results with reliable durability and safety. The cloth-covered model of the Starr-Edwards stellite-ball valve was introduced to eliminate the larger metal surface and ball variance in the original silastic ball prosthesis, aiming to reduce with thromboembolism. The fabric extension was extended from the inflow orifice to the cage (model 2300 and model 6300), and a composite seat prosthesis with small metallic studs around the orifice (model 2310 and model 6310) was introduced, resulting in protection of the orifice cloth on valve closure. Subsequent models with a tapered cage (model 2320 and model 6320) were introduced, in which the ball was situated on the composite seat along the cage. The final ball valves, called "track valves" with a bare stellite track inside each strut (model 2400 and model 6400), were introduced in which the poppet contacted only the metal track and studs, protecting struts and orifice cloth [5]. However, it became evident that the initially observed improvement was a function of the time frame of implantation, and that the cloth covering offered no overall advantage for reduced thromboembolism [6]. Because of a high incidence of cloth wear, these models were later discontinued and became historic prostheses. However, there are still surviving patients with these valves who require strict follow-up, and studying the long-term results is an important aspect in determining how to improve next-generation valves and the prognosis of the target population.
Quite a few articles [2, 4, 7] have reported long-term results of more than 30 years after valve replacement. The survival rates, including operative mortality, reported by Goedje and colleagues [4] were 70.0% (M) and 49.2% (A) after 10 years, 39.2% (M) and 46.6% (A) after 20 years, and 18.4% (M) and 16.6% (A) after 30 years. The survival rates of our patients are higher compared with those of Goedje’s, in which two types of ball valves, noncloth-covered and cloth-covered, were used. One major reason for this better survival rate may be the mean patient age of 32.6 years in our study, which is much younger compared with Goedje’s patients (40.1 years). Because of a limited number of patients, comparison between group M and group A was not attempted in the survival rates and freedom rates in our study.
Morbidity after Starr-Edwards valve replacement, occurring with variable frequency, consisted of thromboembolism, valve thrombosis, anticoagulant-related bleeding events, paravalvular leakage, and/or endocarditis, which are the major limitations of any mechanical prosthesis. Cloth wear or tear is regarded as a unique problem of the Starr-Edwards valve. The incidence of cloth wear in surviving patients who had a cloth-covered model has been reported at less than 2.6% within 10-year follow-up [8, 9], and 6% with 12-year to 15-year follow-up [10], while actual incidence of cloth wear at over 15 years’ follow-up has not been reported. In a report by Bonchek and Starr [10], among 250 patients with the model 6310/20 aortic prosthesis, there were 14 patients who required reoperation and 10 (71%) patients were found to have strut cloth wear at reoperation. Among 171 patients with the model 6310/20 mitral prosthesis, there were 9 patients who needed reoperation, and 2 patients were confirmed to have orifice cloth tearing. In the aortic position, Lund and colleagues [7] reported a detailed follow-up on 717 patients who underwent aortic valve replacement with a silastic ball valve, a cloth-covered valve, or a track valve. In their study the three valve types did not differ as to long-term survival or freedom from complications, and there were no instances of structural failure, apart from wear of the cloth covering the cage struts of the cloth-covered valves. They concluded that the Starr-Edwards ball valves were durable through the remaining lifetime of the patients and able to secure near normal age-specific and sex-specific survival, provided valve and patient size mismatch was avoided.
The freedom from reoperation reported by Goedje and colleagues [4] was 96.4% (M) and 85.8% (A) after 10 years, 87% (M) and 80.2% (A) after 20 years, and 87% (M) and 62.3% (A) after 30 years, in the series of two types of ball valves. The freedom from reoperation of our patients is lower compared with their findings. One reason for this high reoperation rate with exclusive use of the cloth-covered model may be due to cloth wear and prosthetic regurgitation by pannus overgrowth at the orifice area. In our previous study [11] reviewing reoperation for the cloth-covered Starr-Edwards valve, there were 12 patients who required reoperation at our institute. In that study the mean interval until reoperation was 7.9 years and marked cloth wear was observed in all aortic prostheses, but only slight wear was seen in mitral valves. In this study, the mean interval until reoperation was 16.8 years, the cloth wear frequently involved the mitral position, and even cloth tearing was observed in those over 15 years follow-up. The degradation process of the cloth would be progressive and cause thromboembolism. Today, most of the still surviving patients who had Starr-Edwards valve replacement have survived at least 25 years after the operation. Some of them required reoperation for valve-related complications or other aggravated valve lesions during the follow-up period. Pannus inside the orifice and over the studs contributed to valve dysfunction through insufficient valve closing. Because of protracted prosthetic regurgitation, left atrial enlargement and secondary massive tricuspid regurgitation were frequently observed, and other valve procedures were required, especially in patients who underwent mitral valve replacement more than 20 years after initial valve replacement. Severe cloth wear-tear and pannus overgrowth in the orifice area, may contribute to thromboembolic complications and hemodynamic deterioration. In our preoperative catheterization and echocardiographic study, moderate valve regurgitation was more frequently observed in the mitral position compared with the aortic position. The majority of the surviving patients of the present study are in New York Heart Association class II. Although reimplantation should be evaluated on an individual basis, reoperation is recommended when the cardiothoracic ratio is markedly increased on chest X-ray, and moderate prosthetic regurgitation is observed, with left atrial enlargement, on echocardiography or angiography. In our recent reoperations, all of the patients have been in New York Heart Association class I after reoperation. Clinical manifestations of heart failure, such as increasing dyspnea during exercise and edema of the extremities, disappeared after reoperation, especially in mitral valve patients. Hemodynamically deteriorated conditions due to valve dysfunction could be corrected by reoperation.
In conclusion, marked cloth wear and/or pannus formation were observed in all excised prostheses, and these changes notably affected prosthetic valve function, with the most significant problem in the mitral prostheses being protracted regurgitation. Therefore, early diagnosis of valve dysfunction and the decision to repeat surgery are important to improve the long-term results for surviving patients who have received a cloth-covered Starr-Edwards valve, especially in the mitral position.
|
References
|
|---|
- Starr A, Edwards ML. Mitral replacementclinical experience with a ball-valve prosthesis. Ann Surg 1961;154:726-740.[Medline]
- Gao G, Wu Y, Grunkemeier GL, Furnary AP, Starr A. Forty-year survival with the Starr-Edwards heart valve prosthesis J Heart Valve Dis 2004;13:91-96.[Medline]
- Edmunds Jr LH, 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;112:708-711.[Free Full Text]
- Goedje OL, Fischlein T, Adelhard K, Nollert G, Klinner W, Reichart B. Thirty-year results of Starr-Edwards prostheses in the aortic and mitral position Ann Thorac Surg 1997;63:613-619.[Abstract/Free Full Text]
- Starr A, Grunkemeier GL, Lambert LE, Thomas DR, Sugimura S, Lefrak EA. Aortic valve replacementa ten-year follow-up of non-cloth-covered versus cloth-covered caged-ball prostheses. Circulation 1977;56(suppl II):133-139.[Free Full Text]
- Macmanus Q, Grunkemeier GL, Lambert LE, Teply JF, Harlan BJ, Starr A. Year of operation as a risk factor in late results of valve replacement J Thorac Cardiovasc Surg 1980;80:834-841.[Abstract]
- Lund O, Pilegaard HK, Ilkjaer LB, Nielsen SL, Arildsen H, Albrechtsen OK. Performance profile of the Starr-Edwards aortic cloth covered valve, track valve, and silastic ball valve Eur J Cardiothorac Surg 1999;16:403-413.[Abstract/Free Full Text]
- Shah A, Dolgin M, Tice DA, Trehan N. Complications due to cloth wear in cloth-covered Starr-Edwards aortic and mitral valve prostheses, and their management Am Heart J 1978;96:407-414.[Medline]
- Shapira Y, Feinberg MS, Hirsch R, Nili M, Sagie A, Fernberg MS. Echocardiography can detect cloth cover tears in fully covered Starr-Edwards valvesa long-term clinical and echocardiographic study. Am Heart J 1997;134:665-671.[Medline]
- Bonchek LI, Starr A. Ball valve prosthesescurrent appraisal of late results. Am J Cardiol 1975;35:843-854.[Medline]
- Tsukamoto S, Shiono M, Sezai Y, et al. Macroscopic aspects of cloth-covered Starr-Edwards prostheses at reoperationwhat the precursory valve teaches us. J Heart Valve Dis 1998;7:556-560.[Medline]
This article has been cited by other articles:
|
|
|
|
 
R. W. Emery, A. M. Emery, A. Knutsen, and G. V. Raikar
Aortic Valve Replacement with a Mechanical Cardiac Valve Prosthesis
Card. Surg. Adult,
January 1, 2008;
3(2008):
841 - 856.
[Full Text]
|
|
|
Top
Abstract
Introduction
= mitral valve; 
= aortic valve.
