Regression of Hypertrophy After Carpentier-Edwards Pericardial Aortic Valve Replacement

Regression of Hypertrophy After Carpentier-Edwards Pericardial Aortic Valve Replacement Steven S. Khan, MD, Robert J. Siegel, MD, Michele A. DeRobertis, RN, Carlos E. Blanche, MD, Robert M. Kass, MD, Wen Cheng, MD, Gregory P. Fontana, MD, and Alfredo Trento, MD Divisions of Cardiology and Cardiothoracic Surgery, Cedars-Sinai Medical Center, Los Angeles, California Background. The purpose of this study was to determine whether significant regression of left ventricular hypertrophy is seen after implantation of small sizes (19 to 23 mm) of the Carpentier-Edwards (CE) pericardial valve, a stented pericardial valve. Methods. Echocardiograms and electrocardiograms (ECGs) were performed at least 1 year after surgery (mean 18 months) in patients with 19-, 21-, and 23-mm CE pericardial aortic valves and compared with preoperative echocardiograms and ECGs. Results. A total of 41 patients, mean age 79 9 years (range 46 to 93 years), were studied, including 7 19-mm, 22 21-mm, and 12 23-mm patients. The mean postoperative gradient was 22 7 mm Hg for 19-mm valves, 18 5 mm Hg for 21-mm valves, and 16 4 mm Hg for 23-mm valves. The postoperative valve areas were 1.1 0.3 cm 2 for the 19-mm, 1.3 0.3 cm 2 for the 21-mm, and 1.5 0.4 cm 2 for the 23-mm valves. Left ventricular end diastolic diameter, end systolic diameter, septal thickness, and posterior wall thickness all decreased significantly (p <0.05) postoperatively. The proportion of patients with significant left ventricular hypertrophy on ECG decreased from 63% to 47% (p 0.001). Left ventricular mass decreased significantly by echocardiography from 265 g preoperatively to 208 g postoperatively (p 0.004). Left ventricular mass decreased for each valve size, and the greatest absolute reduction in mass occurred in the 19-mm valve recipients. Conclusions. Implantation of the 19-, 21-, and 23-mm CE pericardial valves results in significant reductions in left ventricular mass. These findings suggest that stented pericardial valves can be used in the small aortic root without the need for aortic root enlargement procedures. (Ann Thorac Surg 2000;69:531 5) 2000 by The Society of Thoracic Surgeons In aortic stenosis, the high preoperative pressure gradient across the aortic valve causes significant hypertrophy of the left ventricle. A primary purpose of aortic valve replacement surgery is to relieve this high pressure gradient and allow regression of the ventricular hypertrophy. The importance of regression of left ventricular hypertrophy has been highlighted by recent epidemiologic data showing that the degree of left ventricular hypertrophy is an important prognostic indicator both for survival [1 3] and for the risk of stroke [4] in the general population. In at least one study, the degree of left ventricular hypertrophy was a stronger predictor of prognosis than the presence of coronary disease [2]. Thus, regression of left ventricular hypertrophy may be an important parameter in the assessment of the benefit of aortic valve replacement surgery. A major concern with small stented aortic tissue valves is that significant residual gradients may prevent regression of left ventricular hypertrophy after valve replacement. Early studies of the hemodynamics of tissue valves Accepted for publication July 28, 1999. Address reprint requests to Dr Khan, Division of Cardiothoracic Surgery, Cedars-Sinai Medical Center, 8700 Beverly Blvd, Room 6215, Los Angeles, CA 90048; e-mail: khan@csmc.edu. demonstrated that small stented porcine tissue valves may have significant residual gradients and small valve areas [5, 6]. Valve areas in series of 19-mm stented porcine valves have been reported in the range of 0.80 to 0.85 cm 2 [6, 7], and at least one author has stated that 19-mm and 21-mm tissue valves should not be implanted due to their high residual gradients [8]. This has led to surgeons performing aortic root enlargement procedures to allow implantation of larger size prostheses [9, 10]. Studies of the effect of different types of prosthetic valves on reduction of left ventricular hypertrophy have been mixed. In a study comparing stented porcine valves with homograft valves and stentless valves, Jin and colleagues [11] found no significant regression of left ventricular hypertrophy after implantation of stented tissue valves or mechanical valves, although unstented valves and homografts did allow significant regression. Gonzalez-Juanatey and coworkers [12] found no significant reduction in hypertrophy with 19-mm tissue valves Dr Khan has received honoraria for giving lectures at Baxter-Edwards-sponsored symposia, and Baxter-Edwards has provided partial funding for this study. Dr Trento owns stock in Baxter-Edwards. 2000 by The Society of Thoracic Surgeons 0003-4975/00/$20.00 Published by Elsevier Science Inc PII S0003-4975(99)01389-2

532 KHAN ET AL Ann Thorac Surg PERICARDIAL AVR AND LVH REGRESSION 2000;69:531 5 and less significant reductions with 21-mm valves than with larger valves. Thus, available data suggest that smaller sizes of stented tissue valves may not demonstrate significant reduction of left ventricular hypertrophy. Despite these data, the Carpentier-Edwards (CE) pericardial valve, a stented tissue valve made of bovine pericardium, is widely used in patients with small aortic roots. We therefore undertook this study to determine if patients who have a 19-, 21-, or 23-mm CE pericardial aortic valve implanted demonstrate significant regression of hypertrophy after valve replacement. Material and Methods Patient Population All patients who had received an isolated aortic CE pericardial valve at Cedars-Sinai Medical Center were identified in the Cardiac Surgery Database. Patients who had received a prosthetic valve at any other location were excluded, but patients undergoing concomitant mitral valve repair or coronary bypass surgery were included. Patients who were 1 year or more from surgery were contacted and asked to participate in this study. Patients were provided transportation to the hospital if necessary and evaluated by both a research nurse and at least one physician. The predominant lesion was aortic stenosis in all cases. No patient had more than trace or mild aortic regurgitation, and all preoperative valve areas were 1.1 cm 2 or less. Patient Characteristics A total of 41 patients operated on between 1992 and 1994 were studied. The average age of the patients was 79 9 years (range 46 to 93 years). The majority of the patients were female (54% [22 of 41]). Coronary risk factors were common: 49% (20 of 41) had a history of smoking, 15% (6 of 41) had diabetes, and 49% (20 of 41) had a history of hypertension. However, only 12% (5 of 41) had a history of prior myocardial infarction. A significant majority of the patients were Caucasian (93% [38 of 41]), with small percentages of Hispanic (5% [2 of 41]) and African- American patients (2.4% [1 of 41]). The majority of patients had degenerative aortic valve disease (82% [32 of 39]) with smaller numbers having rheumatic disease. The mean body surface area (BS) was 1.69 0.2 m 2 (range 1.42 to 2.07 m 2 ). Overall, 73% (n 30) underwent concomitant coronary bypass surgery, 5% (n 2) had concomitant mitral valve repair procedures, and 8% (n 3) had tricuspid valve repair procedures. The mean pump time was 159 41 minutes (range 92 to 260 minutes), and the mean crossclamp time was 120 36 minutes (range 63 to 214 minutes). The average length of stay was 13.7 5 days (range 6 to 32 days). Echocardiographic Measurements Preoperative echocardiographic studies were performed either at the Cedars-Sinai Medical Center Cardiac noninvasive laboratory or in the patient s private physician s Fig 1. Peak and mean aortic valve gradients before and after aortic valve replacement. The open bar in each set represents the preoperative gradient, while the shaded bar is the postoperative gradient across the prosthetic valve measured at least 1 year after surgery. Both mean and peak gradients were significantly reduced after surgery. office. All postoperative studies were performed at Cedars-Sinai Medical Center by the same echocardiographic technician and the consensus of two echocardiographers (R.J.S., S.S.K.). Images were stored on tape for later off-line analysis. Peak and mean velocities were recorded across the aortic valve with continuous-wave Doppler, and outflow tract velocities were recorded with pulsed Doppler. Peak and mean valve gradients were calculated using the modified Bernoulli equation, which we have shown accurately reflect catheter gradients in bioprosthetic valves [13]. Aortic valve areas were calculated using the continuity equation [14, 15], which has been validated in bioprosthetic valves by our laboratory and others [16, 17]. Left ventricular diameter and thickness were measured using M-mode echocardiography. Left ventricular mass was calculated using the Penn formula described by Devereux [18]. Patient height and weight were used to calculate body surface area. Left ventricular mass was indexed by dividing left ventricular mass (in grams) by body surface area. Data Analysis Data analysis was performed using the BMDP (BMDP Statistical Software Inc) or SAS (SAS Institute, Inc, Cary, NC) statistical packages. Continuous variables were compared between valve sizes using analysis of variance. Preoperative and postoperative measurements within a patient were compared using a paired t test. Categorical variables were compared using 2 or Fisher s exact test as appropriate. A p value of 0.05 or less was considered statistically significant for all comparisons. Results Hemodynamic Measurements The preoperative and postoperative peak and mean gradients for all valve sizes combined are shown in Figure 1. The average peak gradient preoperatively was 81 35 mm Hg (range 24 to 177 mm Hg). The mean preoperative gradient was 49 24 mm Hg (range 13 to

Ann Thorac Surg KHAN ET AL 2000;69:531 5 PERICARDIAL AVR AND LVH REGRESSION 533 Table 1. Postoperative Valvular Hemodynamics 19 mm 21 mm 23 mm p n 7 22 12 BSA (m 2 ) 1.59 0.14 1.66 0.15 1.80 0.15 Peak gradient 36 11 30 8 26 7 0.07 (mm Hg) Mean gradient 22 7 18 5 16 4 0.06 (mm Hg) V1/V2 ratio 0.41 0.1 0.42 0.07 0.46 0.1 0.45 Valve area (cm 2 ) 1.1 0.3 1.3 0.3 1.5 0.4 0.05 The p values reflect differences between valve sizes. 108 mm Hg). The postoperative measurements were performed at a mean of 18 months after surgery (range 12 to 29 months). Overall, the postoperative average peak gradient had decreased significantly to 30 9mmHg ( p 0.0001), with a range of 14 to 52 mm Hg, and 90% of peak gradients were under 40 mm Hg. The mean gradient had decreased to 18 5mmHg(p 0.0001), with a range from 8 to 33 mm Hg, and 90% of the mean gradients were under 30 mm Hg. The average aortic valve area before surgery was 0.64 0.24 cm 2, and increased to 1.3 0.3 cm 2 (range 0.7 to 2.3 cm 2, 90% of areas were over 1.1 cm 2 ) after valve replacement ( p 0.0001). Postoperative hemodynamic measurements broken down by valve size are shown in Table 1. As would be expected, the larger valve sizes had lower postoperative mean and peak gradients and larger valve areas, but the differences were of borderline significance ( p 0.06 and p 0.07, respectively). The valve areas were 1.1 0.3 cm 2 for the 19-mm valve, 1.3 0.3 cm 2 for the 21-mm valve, and 1.5 0.4 cm 2 for the 23-mm valve, and did differ between the valve sizes ( p 0.049). Left Ventricular Mass and Geometry Measurements of the preoperative and postoperative mass and left ventricular geometry are listed in Table 2 by valve size. The preoperative mean left ventricular Table 2. Changes in Left Ventricular Geometry 19 mm 21 mm 23 mm LV mass pre (g) 326 96 233 45 277 147 LV mass post (g) 228 47.5 187 63 238 126 LV mass index pre 212 77 135 25 170 101 (g/m 2 ) LV mass index post 145 21 114 35 143 83 (g/m 2 ) Post wall pre (cm) 1.38 0.39 1.19 0.24 1.26 0.15 Post wall post (cm) 1.14 0.089 1.11 0.15 1.11 0.11 Septum pre (cm) 1.40 0.25 1.21 0.24 1.31 0.25 Septum post (cm) 1.22 0.18 1.11 0.15 1.24 0.35 LV EDD pre (cm) 4.8 0.52 4.73 0.59 4.6 0.44 LV EDD post (cm) 4.5 0.63 4.3 0.69 4.2 0.58 LVEF pre 58 8% 63 1.4% 60 12% LVEF post 70 12% 65 9% 63 12% EDD end diastolic diameter; LV left ventricle; LVEF left ventricular ejection fraction. Fig 2. Left ventricular mass before (open bar) and 1 year after (shaded bar) aortic valve replacement surgery. Left ventricular mass was significantly reduced at the follow-up study. mass was 265 83 g. Postoperatively, the average left ventricular mass decreased significantly ( p 0.004) to 208 70 g (Fig 2). As shown in Figures 3 and 4, this decrease in mass was due to reductions in posterior wall thickness (P 0.02), septal thickness ( p 0.02), and left ventricular end diastolic dimension ( p 0.01). Importantly, the reduction in left ventricular mass after implantation of the 19-mm valve was at least as great as with the 21-mm and 23-mm valves. In fact, the 19-mm valve demonstrated the greatest magnitude of reduction in left ventricular mass after surgery. In addition to echocardiographic assessment of left ventricular hypertrophy, electrocardiographic assessment was also performed. The percent of patients with left ventricular hypertrophy on electrocardiogram (ECG) decreased from 63% to 47% (n 32, p 0.001). Thus, there was evidence for significant echocardiographic and electrocardiographic regression of hypertrophy. Left ventricular systolic function is usually evaluated using either the left ventricular ejection fraction or the left ventricular end systolic dimension. Preoperatively, patients had normal systolic function with a mean preoperative ejection fraction of 60 10% and a mean left ventricular end systolic dimension of 3.0 0.47 cm (normal 5.5 cm). Postoperatively, left ventricular ejection fraction improved slightly (from 60% to 65%, mean increase of 5%), but the differences were not statistically Fig 3. Left ventricular septal and posterior wall thickness before (open bars) and after (shaded bars) surgery. Both posterior wall and septal thickness are significantly reduced after surgery.

534 KHAN ET AL Ann Thorac Surg PERICARDIAL AVR AND LVH REGRESSION 2000;69:531 5 Fig 4. Left ventricular end diastolic (LVEDD) and end systolic (LVESD) diameters before (open bars) and after (shaded bars) surgery. Both LVEDD and LVESD are significantly reduced after surgery. significant ( p 0.35). However, end systolic dimension, an important indicator of intrinsic left ventricular systolic function, decreased from 3.0 0.5 to 2.7 0.6 cm (mean reduction, 0.3 0.6 cm, p 0.05), suggesting an improvement in left ventricular systolic function (Fig 4). Comment Our findings demonstrate that significant regression of left ventricular hypertrophy occurs after implantation of the smaller sizes of the CE pericardial valve, a stented tissue valve. Reductions in left ventricular mass were seen in all three valve sizes studied: 19, 21, and 23 mm. In addition to left ventricular mass, left ventricular wall thickness decreased and end systolic dimension, an important predictor of intrinsic ventricular systolic function, decreased. These findings suggest significant clinical benefit occurred after implantation of stented pericardial valves. These valves may therefore be a viable alternative to stentless valves, aortic homografts, or aortic enlargement procedures in selected patients. Postoperative Hemodynamics The mean gradients and aortic valve areas observed in this study are consistent with those reported previously. For example, Salomon and associates [19] reported mean valve areas of 1.1 cm 2 for the 19-mm CE pericardial valve, 1.3 cm 2 for the 21-mm valve, and 1.5 cm 2 for the 23-mm valve, which were nearly identical to our value of 1.1, 1.3, and 1.5 cm 2, respectively. As expected, valve area increases and gradients decrease with progressively increasing valve size with the pericardial valve. Regression of Hypertrophy Although hemodynamic parameters are important, the effect of the new aortic valve on ventricular mass may be of even greater importance. Although there are no data on the prognostic importance of regression of hypertrophy in patients with aortic stenosis, recent data demonstrate that left ventricular hypertrophy is a significant predictor of future survival [1, 2] and of the risk of stroke [4] in the general population. Left ventricular hypertrophy may be a stronger predictor of survival than the presence of coronary disease, particularly in African- Americans. The prognostic importance of left ventricular mass may be even greater in women than in men [20]. Importantly, reduction of left ventricular mass in hypertensive patients has been associated with an improvement in prognosis [21]. The reduction in mass observed in this study ranged from 90 g in the 19-mm valve patients to 39 g in the 23-mm valve patients. Liao and associates have shown that an increase in left ventricular mass of 45 g/m 2 is associated with an increase in all-cause mortality of 40% in men and 70% in women [20]. Thus, the magnitude of reduction in left ventricular mass seen in this study is both statistically significant and clinically significant. Although the reduction in left ventricular mass seen in these patients was significant, it should be noted that the average postoperative left ventricular mass remained above the normal range. A similar observation has been made by De Paulis and colleagues, who also noted significant regression of left ventricular mass with small (19 to 21-mm) mechanical valves [22], but observed persistent elevations in mass compared with controls, which were due to persistent septal thickening. There are several possible reasons for incomplete regression of hypertrophy in these patients. First, it should be noted that the preoperative left ventricular mass in these patients is markedly elevated, even compared with reported series of hypertensive patients. In a metaanalysis of regression of left ventricular mass with antihypertensives [23], the mean pretreatment left ventricular mass was 131 g/m 2 for 19 studies of calcium channel blockers, 132 g/m 2 for 21 studies of beta-blockers, 138 g/m 2 for 13 studies of diuretics, and 143 g/m 2 for 18 studies of ACE inhibitors, compared with a pretreatment left ventricular mass of 170 g/m 2 in our 23-mm patients and 212 g/m 2 in our 19-mm patients. It is possible that the greater left ventricular mass in aortic stenosis patients may preclude complete regression of left ventricular hypertrophy. It is also possible that although significant regression of left ventricular mass had occurred, it may be a continuing process and further reductions of left ventricular mass might occur at 2 and 3 years postoperatively. Another possibility is that the aortic valve procedure itself resulted in an increase in left ventricular mass that partially offset the benefits of the lower aortic gradient. It has been demonstrated that thoracotomy results in increases in left ventricular mass after coronary artery bypass surgery [24]. In this study, left ventricular mass progressively increased from 109 23 g/m 2 before surgery to 131 23 g/m 2 by 7 months postoperatively. Limitations Several limitations of our study should be pointed out. One limitation is the small number of patients in each group. In particular, greater numbers of patients would be required to demonstrate that a significant reduction in left ventricular mass occurred with each specific valve size. However, it is clear from the data that the magni-

Ann Thorac Surg KHAN ET AL 2000;69:531 5 PERICARDIAL AVR AND LVH REGRESSION 535 tude of reduction in left ventricular mass was no less with the 19-mm valve than with the 21-mm and 23-mm valve sizes. A second limitation is the use of M-mode echocardiography to measure left ventricular mass. The use of two-dimensional or three-dimensional echocardiography or magnetic resonance imaging may provide more accurate measurements of left ventricular mass; however, these techniques are more difficult to perform, more expensive, and their prognostic significance is not as well studied. Most of the major epidemiologic studies of the prognostic importance of left ventricular mass have been performed using the M-mode measurements used in our studies. Thus, although the M-mode technique has drawbacks, it allows comparison with a wealth of existing prognostic data on survival and the risk of stroke. Another concern is that evaluation of patients at 1 year or more postoperatively creates a potential for bias. It is possible that patients who have survived 1 year or longer are those with lower gradients and greater regression of left ventricular hypertrophy. Finally, it must be recognized that the patient population studied here is relatively elderly, with a mean age of 79 years. It is possible that younger, more active patients may have higher gradients and less regression of left ventricular hypertrophy. Conclusions The small-size (19 to 23 mm) CE pericardial valves are associated with good postoperative hemodynamics and statistically significant regression of left ventricular hypertrophy. These findings suggest that aortic root enlargement procedures may not be necessary in selected patients who can receive a 19- to 23-mm pericardial valve. We would like to thank Peter Iverson for performing the echocardiograms. This study was supported in part by a grant from Baxter-Edwards Corporation. References 1. Koren MJ, Devereux RB, Casale PN, Savage DD, Laragh JH. Relation of left ventricular mass and geometry to morbidity and mortality in men and women with essential hypertension. Ann Intern Med 1991;114:345 52. 2. Liao Y, Cooper RS, McGee DL, Mensah GA, Ghali JK. The relative effects of left ventricular hypertrophy, coronary artery disease, and ventricular dysfunction on survival among black adults. JAMA 1995;273:1592 7. 3. Levy D, Garrison RJ, Savage DD, Kannel WB, Castelli WP. Prognostic implications of echocardiographically determined left ventricular mass in the Framingham Heart Study. N Engl J Med 1990;322:1561 6. 4. Bikkina M, Levy D, Evans JC, et al. Left ventricular mass and the risk of stroke in an elderly cohort: the Framingham Heart Study. JAMA 1994;272:33 6. 5. Cohn LH, Sanders JH, Jr, Collins JJ, Jr. Aortic valve replacement with the Hancock porcine xenograft. Ann Thorac Surg 1976;22:221 7. 6. Bojar RM, Rastegar H, Payne DD, Mack CA, Schwartz SL. Clinical and hemodynamic performance of the 19-mm Carpentier-Edwards porcine bioprosthesis. [Review]. Ann Thorac Surg 1993;56:1141 7. 7. Khan SS, Mitchell RS, Derby GC, Oyer PE, Miller DC. Differences in Hancock and Carpentier-Edwards porcine xenograft aortic valve hemodynamics. Effect of valve size. Circulation 1990;82(Suppl 5):117 24. 8. Jones EL, Craver JM, Morris DC, et al. Hemodynamic and clinical evaluation of the Hancock xenograft bioprosthesis of aortic valve replacement (with emphasis on management of the small aortic root). J Thorac Cardiovasc Surg 1978;75: 300 8. 9. Manouguian S, Seybold-Epting W. Patch enlargement of the aortic valve ring by extending the aortic incision into the anterior mitral leaflet: new operative technique. J Thorac Cardiovasc Surg 1979;78:402 12. 10. Rittenhouse E, Sauvage LR, Stamm SJ, Mansfield PB, Hall DG, Herndon PS. Radical enlargement of the aortic root and outflow tract to allow valve replacement. Ann Thorac Surg 1979;27:367 73. 11. 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;62:683 90. 12. Gonzalez-Juanatey JR, Garcia-Acuna JM, Fernandez MV, Cendon AA, Fuentes VC, Garcia-Bengoechea JB, de la Pena MG. Influence of the size of aortic valve prostheses on hemodynamics and change in left ventricular mass: implications for the surgical management of aortic stenosis. J Thorac Cardiovasc Surg 1996;112:273 80. 13. Baumgartner H, Khan SS, DeRobertis M, Czer L, Maurer G. Effect of prosthetic aortic valve design on the Dopplercatheter gradient correlation: an in vitro study of normal St. Jude, Medtronic-Hall, Starr-Edwards and Hancock valves. J Am Coll Cardiol 1992;19:324 32. 14. Otto CM, Pearlman AS, Comess KA, Reamer RP, Janko CL, Huntsman LL. Determination of the aortic valve area in adults using Doppler echocardiography. J Am Coll Cardiol 1986;7:509 17. 15. Zoghbi WA, Farmer KL, Soto JG, Nelson JG, Quinones MA. Accurate noninvasive quantification of stenotic aortic valve area by Doppler echocardiography. Circulation 1986;73: 452 59. 16. Baumgartner H, Khan SS, DeRobertis M, Czer LS, Maurer G. Doppler assessment of prosthetic valve orifice area: an in vitro study. Circulation 1992;85:2275 83. 17. Rothbart RM, Castriz JL, Harding LV, Russo CD, Teague SM. Determination of aortic valve area by two-dimensional and Doppler echocardiography in patients with normal and stenotic bioprosthetic valves. J Am Coll Cardiol 1990;15: 817 24. 18. Devereux RB, Reichek N. Echocardiographic determination of left ventricular mass in man: anatomic validation of the method. Circulation 1977;55:613 8. 19. Salomon N, Okies J, Krause A, Page U, Bigelow J, Colburn L. Serial follow-up of an experimental bovine pericardial aortic bioprosthesis. Circulation 1991;84(Suppl III):140 4. 20. Liao Y, Cooper RS, Mensah GA, McGee DL. Left ventricular hypertrophy has a greater impact on survival in women than in men. Circulation 1995;92:805 10. 21. Koren MJ, Ulin RJ, Laragh JH, Devereux RB. Reduction of left ventricular mass during treatment of essential hypertension is associated with improved prognosis [Abstract]. Am J Hypertens 1991;4:1. 22. De Paulis R, Sommariva L, De Matteis GM, et al. Extent and pattern of regression of left ventricular hypertrophy in patients with small size CarboMedics aortic valves. J Thorac Cardiovasc Surg 1997;113:901 9. 23. Schmieder RE, Martus P, Klingbeil A. Reversal of left ventricular hypertrophy in essential hypertension. A metaanalysis of randomized double-blind studies. JAMA 1996; 275:1507 13. 24. Tischler MD, Rowan M, LeWinter MM. Increased left ventricular mass after thoracotomy and pericardiotomy. A role for relief of pericardial constraint? Circulation 1993;87: 1921 7.