Prospective randomized evaluation of stentless vs. stented aortic biologic prosthetic valves in the elderly at five years

Transcription

1 doi: /icvts Interactive CardioVascular and Thoracic Surgery 8 (009) Follow-up papers - Valves Prospective randomized evaluation of stentless vs. stented aortic biologic prosthetic valves in the elderly at five years Petar S. Risteski*, Sven Martens, Amin Rouhollahpour, Gerhard Wimmer-Greinecker, Anton Moritz, Mirko Doss Department of Thoracic and Cardiovascular Surgery, Johann Wolfgang Goethe University, Theodor Stern Kai 7, 60590, Frankfurt am Main, Germany Received 6 April 008; received in revised form 1 August 008; accepted 13 August 008 Abstract Objectives: Randomized trials comparing stentless to stented bioprostheses for aortic valve replacement in elderly are scarce. The aim of this study was early and mid-term evaluation of these bioprostheses, with regards to clinical outcome and hemodynamic performance. Methods: Between September 1999 and January 001, 40 patients with aortic stenosis, over the age of 75 years, were randomly assigned to receive either the stented Perimount (ns0) or the stentless Prima Plus (ns0) bioprosthesis. Clinical outcomes, left ventricular mass regression, effective orifice area, ejection fraction and mean gradients were evaluated at discharge, six months, one year and five years after surgery. Results: At five years, there were 5y0 (5%) deaths in the stentless group and 6y0 (30%) deaths in the stented group (all non-valve-related). There was one case of endocarditis in each group, early postoperatively. Overall, a significant decrease in left ventricular mass was found five years postoperatively. However, there was no significant difference in the rate and completeness of LVmass regression between the groups (LV mass index 114"34.1 vs. 10"7.). Furthermore, hemodynamic performance of the valves (mean gradient of 9.9"4.8 mmhg vs. 10."4. mmhg) did not differ significantly between the groups. Conclusions: At five years, stentless valves were not superior to the stented valves, with regards to hemodynamic performance, regression of left ventricular mass and clinical outcome. 009 Published by European Association for Cardio-Thoracic Surgery. All rights reserved. Keywords: Aortic valve replacement; Heart valve; Bioprosthesis; Stentless valve 1. Introduction Aortic valve replacement (AVR) with biologic prosthesis is the only definitive treatment for patients older than 65 years with severe aortic valve stenosis. Despite excellent early postoperative outcome, long-term survival is not satisfactory. Reported survival rates in all age groups range between 50% and 66% w1, x and further decrease to 18% at 15 years in patients older than 75 years of age w3x. Several studies have related these poor results after AVR with the incomplete regression of the left ventricular hypertrophy w4, 5x. Left ventricular hypertrophy (LVH), a known complication of aortic stenosis, has been strongly associated with increased risk of sudden death, congestive heart failure, stroke, myocardial infarction and overall cardiovascular mortality. Incomplete regression of the LVH in patients undergoing AVR has been linked to the obstructive nature of the valve sewing ring and stent, or to patient prosthesis mismatch, which are being held responsible for persistently elevated transvalvular gradients. In the late 1980s, stentless valves were introduced with the goal of maximizing the effective orifice area for flow by eliminating the valvular stent and sewing ring, therefore *Corresponding author. Tel.: q ; fax: q address: docpsr@yahoo.com (P.S. Risteski). 009 Published by European Association for Cardio-Thoracic Surgery facilitating faster and more complete regression of LVH. Over the next decade, several groups have published their initial results; many of them indicating faster and more complete regression of left ventricular mass after stentless as compared with the stented AVR w6, 7x. However, these advantages have been obtained in the setting of nonrandomized trials. We therefore set forth to determine, if we could measure these early and mid-term postoperative improvements in older patients receiving a stentless versus a stented bioprosthetic aortic valve, in a prospective randomized setting.. Materials and methods Between September 1999 and January 001, 40 consecutive patients with severe aortic valve stenosis, older than 75 years, were eligible for inclusion in this prospective randomized trial. Patients were randomized to receive either the Carpentier Edwards Perimount (Edwards Life Sciences Inc, Irvine, CA), second-generation pericardial stented biologic prosthesis or the Prima Plus, porcine stentless biologic prosthesis (Edwards Life Sciences Inc, Irvine, CA). Prior to inclusion in this study, all patients provided written informed consent. This consent form was approved by the ethical committee of our institution.

2 450 P.S. Risteski et al. / Interactive CardioVascular and Thoracic Surgery 8 (009) Inclusion and exclusion criteria Eligible patients included those with isolated aortic valve stenosis (maximum transvalvular gradient 50 mmhg or aortic valve area -0.8 cm ) in which preoperative evaluation indicated the need for an isolated AVR. Only patients expected to survive the surgery and expected to be available to return for all follow-up examinations were included in the study. Patients who specifically chose to have a mechanical valve substitute were not suitable for enrolment. Patients that required repair or replacement of an additional heart valve and those that had prior implantation of a bioprosthetic, mechanical valve or annuloplasty device were excluded from the study. Other exclusion criteria included subaortic stenosis requiring septal resection, active endocarditis, emergency operation and history of myocardial infarction. Intraoperatively, patients were excluded from the study if their aortic root and surrounding tissues had severe calcifications that could not be completely removed surgically and if their valve anatomy indicated an abnormally dilated aortic root or would require excessive trimming of the bioprosthesis... Echocardiography Preoperative and all postoperative echocardiographic studies were performed by two experienced operators using a single echo machine (Vingmed System 5, GE Medical Systems, Wisconsin, USA). All data collected were entered in a central database. In addition to the standard imaging views, preoperative echocardiographic assessments also included the measurement of the diameter of the annulus and the size of the native aorta at the level of the sinotubular junction. We found this necessary in order to identify a possible mismatch between the annulus and the sinotubular junction, which in turn would make the patients unsuitable for the implantation of a stentless valve. Left ventricular mass index (LVMI) was calculated using the formula postulated by Devereux and Reichek, as follows: w 3 LVMI (gym )s(1.05= (EDDqPWTdqIVSTd) EDD 3x 13.6)yBSA where the EDD is the LV end-diastolic diameter (cm), the PWTd is the LV postero-lateral diastolic wall thickness (cm), the IVSTd is the interventricular septum diastolic thickness (cm), and the BSA is the body surface area of the patient w8x..3. Surgical technique All operations were performed by four senior surgeons. Intraoperatively, all patients had retrograde cold blood cardioplegia and carbon dioxide insufflation into the operative field at a rate of lymin for optimal organ protection. The aortic valve was approached through a hockey stick aortotomy. After complete resection of the native aortic valve and debridement of the aortic annulus, accurate sizing was carried out using the respective Carpentier Edwards sizers for the Prima Plus stentless and the Perimount stented valves. The Prima Plus stentless valves were implanted in the subcoronary position. The commissures were positioned 108 apart, with the muscular shelf corresponding to the right coronary sinus. Care was taken to suture the base of the valve subannularly, to ensure that the coaptation line of the leaflets was at the height of the native annulus. Single interrupted unpledgeted 4-0 braided polyester sutures were used for the proximal end, and the rims of the valve commissures were sutured to the native aorta using 4-0 polypropylene running sutures. For the Carpentier Edwards Perimount stented valve implantation, interrupted mattressed pledgeted -0 braided polyester sutures were placed circumferentially from below the annulus. The valves were implanted in the supra-annular position, with the valvular stent positioned so as not to interfere with the coronary ostia..4. Anticoagulation regime Patients in both groups underwent the same anticoagulation regime postoperatively. It included subcutaneous low molecular heparin for the first days and parallel oral anticoagulation with vitamin K antagonists. As soon as the International Normalized Ratio (INR) levels reached the therapeutic range of.5 3.5, heparin was stopped. Oral anticoagulation was continued for three months. The INR levels were monitored by the patient s general practitioner. After three months oral anticoagulation was terminated..5. Follow-up Follow-up examinations were scheduled for discharge from the hospital, at 6, 1 months and 5 years following the operation. All patients had evaluation of their clinical status including New York Heart Association (NYHA) classification, blood analysis including signs of hemolysis and coagulation profile, occurrence of early and late complications and echocardiographic data. Follow-up was 100% complete. We focused our emphasis on the evaluation of left ventricular (LV) mass regression. Both completeness and rate of LV mass regression were assessed. Additional endpoints were changes in LV function and hemodynamics including effective orifice area (EOA) and changes in postoperative transvalvular gradients..6. Statistical analysis Data were compiled and analyzed using Microsoft Access, Microsoft Excel (Redmond, WA) and Stat view (Cary, NC). The baseline characteristics and hospital outcomes for the two groups were compared using x or Fisher s exact test for categorical data and unpaired t-tests for continuous variables. Results, unless otherwise indicated, are reported as mean"s.d. in text and tables. Statistical significance was defined as a P-value of Results All preoperative clinical characteristics including age, gender, body surface area, hypertension, NYHA functional

3 P.S. Risteski et al. / Interactive CardioVascular and Thoracic Surgery 8 (009) Table 1 Preoperative patient characteristics Perimount (ns0) Prima Plus (ns0) P Male Mean age (years) 79"4.3 78" BSA (m ) 1.85" " Hypertension (mmhg) Atrial fibrillation Mean gradient (mmhg) 50.9" " IVS thickness (mm) 19.1" " LVW thickness (mm) 19.8" " LVEDD (mm) 46"3.0 48" LVESD (mm) 35" 3"3 0.8 LV EF (%) 66.6" " NYHA Class III IV BSA, body surface area; IVS, interventricular septum; LVW, left ventricular wall; LVEDD, left ventricular end-diastolic diameter; LVESD, left ventricular end-systolic diameter; LV EF, left ventricular ejection fraction; NYHA, New York Heart Association. class as well as the echocardiographic parameters were comparable between the two groups (Table 1). The intraoperative outcomes as well as early and late morbidity and mortality are summarized in Table. The aortic cross-clamp and cardiopulmonary times were significantly longer in the group receiving stentless biologic prosthesis, however, without association to excessive morbidity and mortality in this group. In both groups, for any given annular diameter, a slightly larger valve prosthesis was implanted, averaging 1.9 mm in the Perimount group and 1.6 mm in the Prima Plus group. The cumulative rates of valve- and procedure-related complications were low and similar in both groups. Early and late mortality included two and three patients in the Perimount group and none and six patients in the Prima Plus group, respectively. All deaths were not related to the valve. All patients displayed continuous clinical improvement after the operation; at five years, all of the survivors were in New York Heart Association class I or II. Mean transvalvular gradients (Fig. 1a) have remained consistently low throughout the follow-up with neither clinical nor statistical relevance in the differences between the groups at any of the given time points. Also noted was the lack of significant difference in the follow-up values of the effective orifice areas (Fig. 1b) of both prostheses, although a tendency toward increase of the same in both groups was obvious early in follow-up (at 1 months with regards to 6 months) only to disappear at the 5-year followup examination. The left ventricular ejection fraction (Fig. 1c) did not change over the time of follow-up. At 6 and 1 months, as well as at 5 years it did not differ between the groups. The left ventricular mass index (LVMI, Fig. 1d) did display a continuous rate of decrease in the first years after the surgery; however, this tendency was lost after the first year as the mean LVMI at 5 years was almost the same to that at 1 months. Finally, the index failed to reach the normal range in both groups. At all time points, the difference between the groups did not reach statistical significance. 4. Discussion Stentless aortic valve prostheses for AVR were introduced almost two decades ago with the aim to provide a superior hemodynamic profile and eventually account for improved survival in older patients. Several retrospective studies in the late 1990s have depicted a faster and more complete regression of LV hypertrophy in patients undergoing stentless AVR. Del Rizzo and co-workers have shown a survival benefit for stentless over stented valves, with nearly a 5- fold greater probability of death at 5 years in patients younger than 60 years of age who received a stented valve prosthesis w9x. With respect to the current literature on this issue, older patients have rarely been a subject of research in comparing stentless with stented AVR. Several groups, as well as ours w10x, have initiated prospective randomized trials to compare these two types of biologic valves. Motivated by the lack of available literature on this topic in elderly, we focused on octogenarians, over 75 years of age and looked on the effects of the specific valve substitutes primarily upon the regression of the LVH as well as upon the hemodynamics and clinical outcome. Table Intraoperative outcomes, early and late morbidity and mortality Perimount (ns0) Prima Plus (ns0) P Cross-clamp time (min) 79"17 108" CPB time (min) 105"3 130"7 0.0 Implantation technique Supra-annular Subcoronary Prosthesis size (mm) Prosthesis oversizing (mm) 1.9" " Permanent neurologic deficit 1 (5%) 1 (5%) 0.9 Anticoagulation-related hemorrhage 1 (5%) Prosthetic valve thrombosis 0 0 Paravalvular leak 0 0 Structural valve deterioration 0 0 Reoperation for early prosthetic valve endocarditis 1 (5%) 1 (5%) 0.9 Early mortality (10%) Late mortality 3 (15%) 6 (30%) 0.13 CPB, cardiopulmonary bypass.

4 45 P.S. Risteski et al. / Interactive CardioVascular and Thoracic Surgery 8 (009) Fig. 1. Hemodynamic results. (a) Mean transvalvular gradients. (b) Mean effective orifice area. (c) Left ventricular ejection fraction. (d) Left ventricular mass index. At five years after the operation, we have witnessed a significant regression of left ventricular mass, achieved by both prostheses, without significant differences in the rate and completeness of this regression between the groups. In addition, no significant differences in the hemodynamic performance and postoperative left ventricular function were observed. Survival of the patients was not related to the nature of the biologic valve. The slope of the observed rate of regression of left ventricular mass was steepest during the first 6 months after the operation, and continued thereafter in the first 1 months although with somewhat less impact, only to remain nearly unchanged at mid-term. Although the myocardial cellular hypertrophy as well as the myocardial collagen fibrosis may take more than a year to regress after the AVR w11x, it seems that one year of follow-up may suffice to assess the impact of the valve substitute upon the regression of the LVH. Others have come to the same conclusion w6, 1x. The incomplete regression of left ventricular mass at midterm observed in both groups only confirms the inevitable influence of other factors like age, hypertension and gender upon this process. Since all these factors did not differ between the groups, they could not have influenced the outcomes. Over the years of follow-up, the transvalvular gradients have not displayed any clinically relevant or significant difference between the groups. We chose to assess the mean and not peak transvalvular gradients in our study since they better correlate with the left ventricular hypertrophy. Peak gradients are more factor-related and display large variations with physical exertion which on the other hand is rarely observed in a population of patients in their eight or ninth decade, like those subjected in this study. Other prospective randomized studies that emerged in the meantime have come to the same conclusion. Ali and co-workers have randomized 161 patients to receive a stented or stentless aortic valve substitute, and found neither statistically nor clinically relevant difference in the mean transvalvular gradients as well as the regression of the LV mass w13x. The ASSERT investigators concordantly concluded no difference in LV mass regression at both 6 and 1 months after replacement of the aortic valve with a stentless or stented valve substitute in 00 patients w14x. The study conducted by Walther et al., however, did show differences in peak gradients and LV mass index at 6 months after the operation w15x. They postulate that a difference of 3.4 mmhg in peak gradients is accountable for the reduced LV mass regression in the stentled valve group. Mean gradients have not been reported in their study. Worthy of mention is that they did, however, compare a stentless valve with a first-generation stented valve, the Carpentier Edwards porcine valve. However, this cause and effect relationship with the lower peak gradients was lost at mid-term follow-up w16x. Marquez and colleagues have tested the hydrodynamic performance of six commercially available bioprosthetic valves with various sizes in vitro at various flow rates w17x. Considerable differences in pressure gradients have been found among tested valves at each size. The Carpentier Edwards Perimount valve showed superior performance at all sizes tested including lowest gradients compared to the other valve substitutes for each size. Clearly, no patientrelated factors could have influenced the hydrodynamic performance of each valve in vitro. This prosthetic valve (Carpentier Edwards Perimount) has only recently been compared in vivo to another stentless substitute (Toronto stentless porcine valve) on a group of patients somewhat

5 P.S. Risteski et al. / Interactive CardioVascular and Thoracic Surgery 8 (009) younger than that investigated in this study w18x. They report of no significant differences in hemodynamic function or clinical events between both biologic valves neither in the early postoperative period nor at 5 years. To our surprise, we could not support previously reported findings, which showed faster and more complete regression of LV-hypertrophy as well as superior hemodynamics of stentless valves over stented bioprostheses. One of the possible explanations may be that those reports studied mainly first-generation stented aortic bioprostheses w15x. Modern bioprosthetic valves have been shown to have excellent hemodynamic performance w16, 19x and this may account for the equally good performance as seen with the stentless valves. As no significant differences in transvalvular gradients were detected in this study, it is not surprising that no differences in the rate and completeness of LV mass regression resulted. Overall, the complexity of stentless valve implantation with its prolonged cross-clamping times might not be justifiable under these circumstances, if as we found, the same results can be achieved with a standard stented bioprosthesis. Our results are in concordance with the major prospective randomized studies that emerged in the meantime w13, 14, 18x. References w1x David TE, Ivanov J, Armstrong S, Feindel CM, Cohen G. Late results of heart valve replacement with the Hancock II bioprosthesis. J Thorac Cardiovasc Surg 001;11: wx Hammermeister K, Sethi GK, Henderson WG, Grover FL, Oprian C, Rahimtoola SH. Outcomes 15 years after valve replacement with a mechanical versus bioprosthetic valve: final report of the Veterans Affairs randomised trial. J Am Coll Cardiol 000;36: w3x Jamieson WR, Burr LH, Tyers GF, Munro AI. Carpentier-Edwards standard and supraannular porcine bioprosthesis: 10 year comparision of structural valve deterioration. J Heart Valve Dis 1994;3: w4x Rossi A, Tomaino M, Golia G, Anselmi M, Fuca G, Zardini P. Echocardiographic prediction of clinical outcome in medically treated patients with aortic stenosis. Am Heart J 000;140: w5x Levy D. Clinical significance of left ventricular hypertrophy: insights from the Framingham Study. J Cardiovasc Pharmacol 1991;17(Suppl ): S1 S. w6x Jin XY, Zhang ZM, Gibson DG, Yacoub MH, Pepper JR. Effects of valve substitute on changes in left ventricular function and hypertrophy after aortic valve replacement. Ann Thorac Surg 1996;6: w7x Thomson HL, O Brian MF, Almeida AA, Tesar PJ, Davison MB, Burstow DJ. Haemodynamics and left ventricular mass regression: a comparision of the stentless, stented and mechanical sortic valve replacement. Eur J Cardiothorac Surg 1998;13: w8x Devereux RB, Reichek N. Echocardiographic determination of left ventricular mass in man: anatomic validation of the method. Circulation 1977;55: w9x Del Rizzo DF, Abdoh A, Cartier P, Doty D, Westaby S. Factors affecting left ventricular mass regression after aortic valve replacement with stentless valves. Semin Thorac Cardiovasc Surg 1999;8: w10x Doss M, Martens S, Wood JP, Aybek T, Kleine P, Wimmer Greinecker G, Moritz A. Performance of stentless versus stented aortic valve bioprostheses in the elderly patient: a prospective randomized trial. Eur J Cardiothorac Surg 003;3: w11x Krayenbuehl HP, Hess OM, Monrad ES, Schneider JS, Mall G, Turina M. Left ventricular myocardial structure in aortic valve disease before, immediately, and late after aortic valve replacement. Circulation 1989;79: w1x De Paulis R, Sommariva L, Colagrande L, De Matteis GM, Fratini S, Tomai F, Bassano C, Penta de Peppo A, Chiariello L. Regression of left ventricular hypertrophy after aortic valve replacement for aortic stenosis with different valve substitutes. J Thorac Cardiovasc Surg 1998;116:590. w13x Ali A, Halstead JC, Cafferty F, Sharples L, Rose F, Lee E, Rusk R, Dunning J, Argano V, Tsui S. Early clinical and hemodynamic outcomes after stented and stentless aortic valve replacement: results from a randomised controlled trial. Ann Thorac Surg 007;83: w14x Perez de Arenaza D, Lees B, Flather M, Nugara F, Husebye T, Jasinski M, Cisowski M, Khan M, Henein M, Gaer J, Guvendik L, Bochenek A, Wos S, Lie M, Van Nooten G, Pennell D, Pepper J; ASSERT (Aortic Stentless versus Stented valve assessed by Echocardiography Randomized Trial) Investigators. Randomised comparison of stentless versus stented valves for aortic stenosis: effects on left ventricular mass. Circulation 005;11: w15x Walther T, Falk V, Langebartels G, Krüger M, Bernhardt U, Diegeler A, Gummert J, Autschbach R, Mohr FW. Prospectively randomised evaluation of stentless versus conventional biological aortic valves. Circulation 1999;100(Suppl II):II-6 II-10. w16x Lehmann S, Walther T, Kempfert J, Leontjev S, Rastan A, Falk V, Mohr FW. Stentless versus conventional xenograft aortic valve replacement: midterm results of a prospectively randomized trial. Ann Thorac Surg 007;84: w17x Marquez S, Hon RT, Yoganathan AP. Comparative hydrodynamic evaluation of bioprosthetic heart valves. J Heart Valve Dis 001;10: w18x Chambers JB, Rimington HM, Hodson F, Rajani R, Blauth CI. The subcoronary Toronto stentless versus supra-annular Perimount stented replacement aortic valve: early clinical and hemodynamic results of a randomized comparison in 160 patients. J Thorac Cardiovasc Surg 006; 131: w19x Dellgren G, David TE, Raanani E, Armstrong S, Ivanov J, Rakowski H. Late hemodynamic and clinical outcomes of aortic valve replacement with the Carpentier-Edwards Perimount pericardial bioprosthesis. J Thorac Cardiovasc Surg 00;14: