Thirty-Seven-Year Durability of a Starr-Edwards Aortic Prosthesis

Abstract

We report the case of a patient who was diagnosed with prosthetic valve endocarditis caused by Staphylococcus aureus. He had undergone aortic valve replacement with a Starr-Edwards prosthesis 37 years earlier. Because of uncontrolled infection despite antibiotic treatment, the patient underwent successful surgical replacement of the prosthetic valve.

The Starr-Edwards valve is one of the oldest cardiac valve prostheses in the world. The 1st aortic valve replacement (AVR) with a Starr-Edwards metal cage and silicone ball valve was performed in 1961.¹ Follow-up studies have documented the excellent durability of the Starr-Edwards valve in the aortic position.^2–4 We report the case of a patient who had undergone AVR with a Starr-Edwards aortic prosthesis 37 years earlier and developed prosthetic valve endocarditis caused by Staphylococcus aureus. He eventually required valve replacement for acute prosthetic valve endocarditis.

Case Report

In June 2003, a 58-year-old man was admitted to the Siyami Ersek Thoracic and Cardiovascular Education and Research Hospital with a 7-day history of fever. He had undergone AVR with a model 1000-9A Starr-Edwards aortic valve prosthesis for rheumatic aortic insufficiency at age 22, in April 1967, at this same hospital. Physical examination revealed right hemiparesis. His blood pressure was 110/70 mmHg; heart rate, 95 beats/min and regular; and temperature, 38 °C. Cardiac auscultation revealed a grade 2/6 systolic murmur that was audible at the left sternal border. His lungs were clear on auscultation.

A complete blood count analysis revealed anemia and leukocytosis. The white blood cell count was 12,200 mm³; hemoglobin, 10.7 g/dL; hematocrit, 33%; and erythrocyte sedimentation rate, 65 mm/h. A chest radiograph showed a mildly enlarged heart and pulmonary congestion (Fig. 1). Electrocardiography revealed 1st-degree atrioventricular block. Transesophageal echocardiography (TEE) showed a large echo-dense vegetation (20 × 6 mm) on the prosthetic valve. Three blood cultures produced S. aureus. The organism was sensitive to rifampicin, vancomycin, and gentamicin. The patient initially received intravenous vancomycin (1 g, twice per day) and gentamicin (80 mg, 3 times per day), and oral rifampicin (300 mg, 3 times per day).

Fig. 1 Chest radiograph reveals implanted prosthesis. The ball of the valve is non-radiopaque.

On the 10th day of treatment, TEE was repeated and showed large vegetations on the prosthetic valve; these were more mobile than those seen on the 1st day. In July 2003, the patient underwent urgent surgery through a median sternotomy. After the initiation of cardiopulmonary bypass (CPB), aortic cross-clamping was performed, mild systemic hypothermia was achieved, and the heart was then arrested with cold blood cardioplegia. An ascending aortotomy was performed with an S-shaped incision. Multiple vegetations were removed from the prosthetic valve, and the infected prosthetic valve was excised (Fig. 2). A 23-mm biological (bovine) valve was implanted. No perivalvular leak was detected. The aortic arteriotomy was closed with 4-0 propylene sutures and CPB was terminated. Pathologic and microbiologic examination of the materials obtained at surgery confirmed the findings of the preoperative blood cultures. The organism was sensitive to cefazolin; therefore, the medical treatment was switched to intravenous cefazolin (2 g, 4 times per day) for 6 weeks. The postoperative clinical course was excellent. At the 18-month follow-up visit, the patient was in a good state of health, in New York Heart Association functional class II.

Fig. 2 The aortic valve prosthesis after removal.

According to the patient's medical records regarding his initial 1967 valve replacement, oxygenation during extracorporeal circulation was supplied by a bubble oxygenator, with the level of general hypothermia at 28 °C. Left ventricular apical venting and aortic cross-clamping were performed, and basic antegrade coronary perfusion was induced for the intracardiac procedure. The patient was discharged with long-term anticoagulation medication. In 1983, he experienced a left-sided cerebral infarction with hemiparesis despite anticoagulation with phenprocoumon. At that time, the prothrombin time was 35% (international normalized ratio, 2.5). The patient remained in sinus rhythm at his annual examinations.

Discussion

To date, approximately 14,000 Starr-Edwards model 1000 valves have been implanted in patients worldwide. Despite complications with the Starr-Edwards model 1000, such as a high pressure gradient, a less effective orifice, and absence of central blood flow, long-term results up to 30 years after replacement with this valve have occasionally been reported in the medical literature.^4–6 Herein, we document the case of a patient in whom a Starr-Edwards aortic prosthesis functioned for 37 years. To our knowledge, this is the longest survival period for such a patient in Turkey.

Gödje and colleagues⁴ performed a study of patients who had received aortic or mitral valve Starr-Edwards prostheses and found that thromboembolism and structural valve deterioration were higher in the aortic valve group. Grunkemeier and Starr⁷ found evidence of ball variance (“poppet damage owing to fatty infiltration of the silicone rubber ball”) for the Starr-Edwards model 1000. However, in our patient, at the time of valve explantation because of infective endocarditis, there was no structural prosthetic valve deterioration. The valve was in its original shape and was functioning well. Gödje's group⁵ reported a case of cerebral infarction in a patient who had undergone double-valve replacement with Starr-Edwards prostheses. Those authors suggested that if prothrombin time had been lower than 28%, the cerebral infarction might not have occurred. In our patient, a cerebral infarction occurred 16 years after initial valve replacement, despite anticoagulation and a prothrombin time of 35% (international normalized ratio, 2.5). We presume that the patient had emboli because of inadequate anticoagulation. The 1995 guidelines of the European Society of Cardiology recommended an INR of 3.0 to 4.5 after AVR with the 1st-generation Starr-Edwards valves.⁸

Transthoracic echocardiography (TTE) is useful for the detection of vegetation and the evaluation of prosthetic valve function; however, because of acoustic shadowing, this method is limited in the demonstration of infective lesions. Transesophageal echocardiography is a better method for showing the left ventricular outflow tract (where vegetations of a prosthetic aortic valve may be present) and is considered to be more accurate than TTE in detecting valvular vegetations.⁹ We performed TEE for diagnosis of endocarditis and for follow-up of the prosthetic valves.

Few patients are reported to have lived with a Starr-Edwards prosthesis for 30 years or longer.4–6,10–12 Therefore, survival for 37 years with a Starr-Edwards aortic prosthesis is noteworthy.