| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Interact CardioVasc Thorac Surg 2007;6:458-461. doi:10.1510/icvts.2006.135533 © 2007 European Association of Cardio-Thoracic Surgery
Edwards Mira bileaflet prosthesis in aortic position: midterm results of a prospective multi-centre study
a Department of Thoracic and Cardiovascular Surgery, University Hospital, Rennes, France Received 27 April 2006; received in revised form 16 January 2007; accepted 22 January 2007
*Corresponding author. CTCV, Hôpital Ponchaillou, 2 rue Henri Le Guilloux, 35033 Rennes, France. Tel.: +33 299282473; fax: +33 299282496.
This prospective multicentre study assessed the midterm clinical and haemodynamic results of the Edwards Mira curved bileaflet prosthesis in aortic position. From June 1998 to October 2000, 117 patients, mean age 64 years (31–78 years) underwent aortic valve replacement with the Edwards Mira valve in three institutions. Clinical status, haemodynamic performance and valve related complications were assessed. Serial echocardiographic examinations were performed at discharge and at least two years follow-up. Operative mortality was 1.7% (n=2). Follow-up was 100% complete (594.1 patient-years). Actuarial survival at one, three and five years was 96.5±1.7%, 93.9±2.2% and 88.4±3.0%, respectively. Freedom from thrombosis was 99.1±0.9%, from embolic events: 96.2±1.9%, from bleeding events: 96.4±1.7%, and from non-structural dysfunction 97.2±1.6%. There was no structural dysfunction. The peak gradient at discharge was 22.13±8.1 mmHg down to 20.8±8 mmHg at 28 months. The mean gradient at discharge was 12.7±4.5 mmHg at discharge down to 10.8±4.2 mmHg at 28 months. The permeability index was 53.3±10% at 28 months. The Edwards Mira aortic valve showed excellent midterm haemodynamic performance, good midterm survival and low valve related complications rate. Long term follow-up remains to be assessed.
Key Words: Aorta; Mechanical valve; Aortic valve replacement; Echocardiography; Outcome
Important technical progresses have been made to enhance mechanical valve profile, in order to optimise haemodynamic performances and reduce levels of anticoagulation. The Edwards Mira valve became available in Europe in early 1998. It differs from a previous mechanical valve by the curved shape of its wings covered by pyrolytic carbon. Curved wings allow less energy loss waste during systole and a predominating central flow. The valve hinge has a bearing system with no slip effect between the concave and convex areas. The silicone insert is designed to enhance the sewing ring's ability to be adapted to irregular or calcified annuli. The goal of this prospective and multicentre study was to evaluate mid-term clinical performances and haemodynamic profile of the Mira valve in aortic position.
2.1. Patients Between June 1998 and October 2000, 117 patients, 89 males (76.1%) and 28 females (23.9%), underwent an aortic replacement with standard Edwards Mira Prostheses (EMPs). Fifty were performed in Rennes, 42 in Nancy and 25 in Bordeaux. This study has been conducted according to the review board requirements of each institutional centre. Written informed consent was obtained from all patients. Mean age at implantation was 64±9 years (range 31–78 years). The mean body surface area (BSA) was 1.87±0.18 m2. Mean NYHA class was 2.5±0.5. Mean left ventricular ejection fraction (LVEF) was 59.2±13.2%, including seven patients with LVEF  30%. Follow-up period ranged from 2–6.7 years (median 5.4 years).
Underlying valvular pathology was: calcification in 64 patients (54.5%), bicuspid in 18 patients (15.3%), Marfan's disease in 16 patients (13.6%), rheumatism in seven patients (5.9%), endocarditis in seven patients (5.9%), and other in five patients (4.3%). Concomitant surgery was performed in 40 cases (34%), most commonly consisting of coronary artery bypass grafting (16 patients, 13.6%), and replacement of ascending aorta (12 patients, 8.5%) for ascending aortic aneurysm. Mean labelled valve size was 23.6±1.9 mm. Preoperative electrocardiograms showed normal sinus rhythm in 105 patients (89%) and atrial fibrillation in 10 patients (8.5%). All prostheses were inserted using separated stitches in intra-annular position, with the leaflet axis perpendicular to the septum. Mean aortic cross-clamping time was 47±18.9 min and mean cardiopulmonary bypass time was 64.7±31.1 min. 2.3. Postoperative care and follow-up Subcutaneous calcium heparin was started 4 h postoperatively, and continued until the International Normalised Ratio (INR) had reached 2–3 with warfarin, started 72 h postoperatively. For patients with atrial fibrillation, left ventricular dysfunction or a history of thromboembolic events, the INR target was raised at 2.5–3.5 ACC/AHA guidelines. Follow-up information, consisting of a clinical and echocardiographic evaluation, was obtained at discharge, and at 28 months. Patients were followed using direct contact, hospital health record and official death registry. Mortality and morbidity were used to assess the occurrence and classification of postoperative events.2.4. Haemodynamic and haematological investigations Echocardiography was performed at discharge for all patients and at 28 months for 110 patients. Data were collected in accordance with the American Food and Drug Administration heart valve criteria. Echocardiography was performed using transthoracic colour flow Doppler. The following data were assessed: trans-prosthetic pressure mean and peak gradients, permeability index (PI) calculated from pulsed and continuous aortic valve and outflow tract Doppler recording, effective orifice area (EOA) calculated using the continuity equation, and left ventricular ejection fraction. PI is defined as the calculation of the ratio of the maximal velocity recorded in the left ventricular outflow tract (pulsed Doppler) over the highest velocity of the aortic valve recorded using continuous wave Doppler. In contrast to transaortic valve gradients, the PI reflects the functional properties of the aortic valve independently of loading condition and left ventricular function. The examinations were conducted by the same cardiologist in each centre.Haematological assessment at 28 months included red cell count measurement, white cell count measurement, haematocrit, haemoglobin, serum lactate dehydrogenase (LDH) and serum haptoglobin. The Skoularigis criteria [1] were used. Patient data were analysed with (SPSS/PC+) version 12.5. Percentage format was chosen to describe qualitative data, while mean±standard deviation was used for quantitative data. Time-related events were analysed by the Kaplan–Meier method. Percentages of freedom from events are presented with standard deviation. Wilcoxon non-parametric test was used to compare series' averages. Statistical significance is defined by P-value<0.05.
Follow-up was 100% complete (594.1 patient-years), with a mean of 5.1 years and median 5.4 years, ranging from 2–6.7 years. There was no intra-operative death but two patients died within 30 days (1.7%) and twelve additional patients during the follow-up period. Two late-deaths were valve related, including an alcoholic patient who did not take his warfarin on a regular basis. He developed massive valve thrombosis six years after implantation. The second case was a brain haemorrhage one year after the operation. The five-year freedom from valve related death was 99.1±0.9%. Survival including early death was 96.5±1.7%, 93.9±2.2% and 88.4±3% at one, three and five years, respectively. At the end of the follow-up period, 92% of the patients were in NYHA functional class I or II. Two cases of valvular thrombosis were observed. The first case, with a thrombus size of 5 mm, was discovered following an echocardiographic evaluation on an alcoholic patient with cirrhosis. This patient whose INR was 1.2 when hospitalised, had been treated with heparin anticoagulation followed by warfarin. The second case occurred in a patient with poorly controlled anticoagulation treatment, who suffered two strokes during the six years following the operation. He died following these complications. Actuarial freedom from thrombosis was 99.1±0.9% at five years and valve thrombosis linear rate was 0.34% patient-year. Six patients had an embolic event (in spite of a good INR), resulting in a linearised rate of 1% patient-year. One patient died following an embolic event. Of the five patients who had an ischaemic event with no subsequent neurological deficits, three patients had normal brain computed tomography (CT) scan of the brain, and in the fourth patient an echo-Doppler exam, performed after an episode of amaurosis fugax, showed no abnormality. Actuarial freedom from thromboembolic event was 99.1±0.9%, 98.2±1.3% and 96.2±1.9% at one, three and five years, respectively. Haemorrhage event linearised rate was 0.67% patient years with the following actuarial freedom: 98.2±1.2%, 98.2±1.3% and 96.4±1.7% at one, three and five years, respectively. Three patients presented with gastro-intestinal haemorrhage, requiring surgical intervention in one patient. INR was 2.9, 2.5 and 2, respectively. The fourth patient died after cerebral haemorrhage with INR 3.5. 3.3. Non-structural dysfunction In three patients (2.6%), a para-valvular leak was observed, without clinical consequenses. Actuarial freedom from non-structural dysfunction was 100%, 97±1.6% and 97±1.6% at one, three and five years, respectively. Linearised rate for para-valvular leak is 0.5% patient year.One patient presented Staphylococcus aureus endocarditis two years after implantation, followed by successful reoperation. 3.4. Haemodynamic and haematological profile The valve size ranged from 19–27 mm, including 41% of 23 mm and 29% of 25 mm. Echocardiography was performed at discharge for 115 patients (98%) and at 28 months in 110 patients (94%). Haemodynamic data are reported in Table 1.
A significant decrease in mean and peak pressure gradient was observed between discharge and follow-up at 28 months with P=0.03 and P=0.001, respectively. In the 21-mm size valve the gradient remained stable between 13.6 mmHg and 13.9 mmHg. Permeability index was 53.3±10%, mean EOA was 2.1 cm2 (ranging from 1.1 cm2 and 3.2 cm2) and mean indexed EOA (IEOA) was 1.1 cm2/m2 at 28 months. Haematological data are reported in Table 2. In seven patients (6%) sub-clinical haemolysis was observed.
This prospective multicentre study presents for the first time detailed mid-term results for the Mira prosthesis in aortic position. The hospital mortality was low (1.7%), in accordance with an earlier study by Remadi et al. [2]. Actuarial survival at five years was 88.4±3.0% (linearised rate of 2.3% patient-year). Five-year freedom from thromboembolism was 96.2±1.9% (linearised rate of 1% patient-year) in our study. At seven-year follow-up, the Vitale study considering the Sorin Bicarbon valve reported 94.8% freedom from thromboembolism [3]. These results, especially our observation in regard to the Mira mechanical valve implanted in the aortic position, were different and better than those published with the Carbomedics valve [4] (87.1% freedom from thromboembolism at four years). Actuarial freedom from haemorrhagic event at five years is 96.4% in our study (linearised rate of 0.67% patient-year). The four patients who suffered from this complication were 52, 67, 67 and 74 years old, which is in accordance with Jamieson et al. conclusions [5]. Whitaker et al. [6] presented a low rate of thromboembolic events with the Sorin Bicarbon valve (0.28% patient-year). Emery et al. [7] reported a thromboembolic event linearised rate of 2.7% patient-year with the SJM valve while Lund et al. [8] reported a rate of 2.24% patient-year. In our series, three cases of non-structural dysfunction occurred (non-infectious para-valvular leaks, two with grade 1 and one with grade 2 leaks), resulting in an actuarial freedom at five years of 97% (linearised rate of 0.5% patient-year). This result compares favourably to Vitale et al. results [3] with the Sorin Bicarbon valve (actuarial freedom from non-structural dysfunction event at five years=94.7%). The Edwards Mira valve could be an interesting alternative in conditions of a severely calcified aorta or endocarditis, due to its curved and compliant silicone sewing ring. We propose to evaluate the PI instead of effective orifice area, because in our experience the left ventricular outflow tract diameter is often difficult to assess. We report a mean PI of 53.3% at 28 months, which indicates acceptable haemodynamic features of the Edwards Mira valve in aortic position. For instance, Lesbre et al. [10] previously reported a PI ranging from 31% to 46% for a series of SJM prostheses. The Edwards Mira valve has demonstrated haemodynamic performances similar to other mechanical prostheses (Tables 3 and 4).
Prostheses–patient mismatch (PPM) can be avoided if the IEOA is 0.85 cm2/m2. In our study, 30% of patients had IEOA <0.85 cm2/m2. Gradients and IEOA were largely correlated. As a consequence, PPM generates high postoperative gradients. According to Pibarot and Dumnestil [11], negative effects of PPM can be proved beyond eight-year follow-up only; this is therefore not possible in this study. The low rate of subclinical haemolysis in our cohort, seven patients (6%), is probably related to the single aortic valve position as Josa et al. have suggested [15].
In summary, the Edwards Mira mechanical valve in aortic position has shown good mid-term haemodynamic results. The permeability index and gradients were acceptable and comparable to valves classically elected prosthetic valves.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
Top
Abstract
1. Introduction