Ann Thorac Surg 2003;76:689-693
© 2003 The Society of Thoracic Surgeons
Original article: cardiovascular
Pulmonary autografts in patients with severe left ventricular dysfunction
Stephen C. Gauthier, MDa,
Joshua G. Barton, MDa,
Mary M. Lane, PhDb,
Ronald C. Elkins, MDb*
a Section of General Surgery, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma, USA
b Section of Thoracic and Cardiovascular Surgery, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma, USA
* Address reprint requests to Dr Elkins, University of Oklahoma Health Sciences Center, WP2230, P.O. Box 26901, Oklahoma City, OK 73190, USA.
e-mail: ronald-elkins{at}ouhsc.edu
Presented at the Forty-ninth Annual Meeting of the Southern Thoracic Surgical Association, Miami Beach, FL, Nov 7–9, 2002.
 |
Abstract
|
|---|
BACKGROUND: Performing a Ross operation in patients with severe left ventricular dysfunction is controversial. The objective in this retrospective study was to determine the outcome of 15 patients with aortic valve disease (11 had aortic insufficiency and 4 had aortic insufficiency and aortic stenosis) associated with reduced left ventricular function (ejection fraction < 40%) treated with a pulmonary autograft.
METHODS: We identified 15 patients with severe left ventricular dysfunction from 226 consecutive pulmonary autograft procedures done between age 18 and 50 years from 1986 to 2001. Patients had documented preoperative ejection fraction less than 40% and were in New York Heart Association class III or IV. Preoperative ejection fraction ranged from 18% to 37% (mean, 31% ± 6.5%). Transthoracic echocardiograms obtained preoperatively and at 1-week, 6-month, and 1- and 2-year intervals were reviewed. Records were evaluated for survival, clinical status, left ventricular function, and valve function.
RESULTS: There were no operative deaths, late deaths, or reoperations. All patients had follow-up examinations within the past year and are clinically well (67% > 2 years follow-up). Ten of 15 patients (67%) had substantially improved ventricular function (> 20% increase). The average ejection fraction increased from 31% ± 7% preoperatively to 51% ± 11% at 2 years, and the increase is significant from 1 week on (p < 0.02). Average left ventricular mass index decreased by 41% at 6 months (n = 10; p = 0.009) and by 44% at 2 years (n = 9; p = 0.02). Mean Z values for left ventricular mass decreased from 7.6 to 3.6 after more than 2 years (p = 0.007).
CONCLUSIONS: The Ross operation is an appropriate option in adults age 50 or younger in the presence of decreased left ventricular ejection fraction. Neither operative mortality nor postoperative sequelae were identified in our subset of patients. Excellent survival and ventricular recovery are predicted.
|
Introduction
|
|---|
| Dr Elkins discloses that he has a financial relationship with CryoLife, Inc.
|
For patients with aortic valve disease requiring surgical replacement, there is an array of aortic valve substitutes. Among them the pulmonary autograft has several proven benefits. Advantages such as valve viability, propensity for repair, ability to grow, and avoidance of anticoagulation have been well documented [1–3]. Normalization of left ventricular function and left ventricular mass index after a Ross operation have been demonstrated [4, 5].
In patients who would normally be considered for a Ross operation, the surgeon faced with aortic valve disease combined with left ventricular dysfunction often avoids a pulmonary autograft. The difficult dissection, double valve replacement, and length of the procedure often lead to other types of aortic valve substitutes being utilized. The objective in this retrospective study is to summarize our experience in young adults with aortic valve disease combined with significant left ventricular dysfunction who have had a Ross operation.
|
Patients and methods
|
|---|
Patient characteristics
The medical database of the 480 Ross operations at the University of Oklahoma Health Sciences Center was reviewed, and of the 226 consecutive adult patients aged 18 to 50 years, we identified 15 patients with aortic valve disease and severe left ventricular dysfunction, ie, left ventricular (LV) ejection fraction (EF) less than 40% by preoperative echocardiography. Inclusion of a patient in this series required concurrence of assessment of LV EF by echocardiography and LV angiography, when performed (one patient was excluded because of significant difference between the angiographic and echocardiographic assessment of LV function). The hospital records and clinical follow-up were reviewed retrospectively. The patients were operated on between August 1986 and June 2001, and the primary surgeon for all patients was the senior author.
The mean age of the 13 men and 2 women was 36 ± 10 years. One patient had a prior aortic valve replacement with an aortic homograft. All patients had aortic insufficiency; 11 had severe aortic insufficiency, and 4 had moderate aortic insufficiency with associated aortic stenosis, which was moderate in 3 and severe in 1. Preoperative LV EF ranged from 18% to 37% (mean 31% ± 6.5%). Five of the patients had documentation of severe reduction in their EF (25% or less) for 2 or more years, with class 3 or 4 symptoms of congestive heart failure. Nine had documentation of a reduced EF for at least 1 year. In one patient with active endocarditis, the reduced EF was believed to be of recent onset. At the time of operation, 10 bicuspid aortic valves and 5 tricuspid aortic valves were identified. Patient characteristics are shown in Table 1.
The Ross operations were all done as root replacements, and concomitant procedures were done in 14 patients. Twelve patients had aortic annular reduction and fixation, seven required ascending aortic replacement or vertical aortoplasty, one had a mitral commissurotomy and ring annuloplasty, and one required debridement of an aortic annular abscess for active bacterial endocarditis. None of the patients was thought to have Marfan syndrome or aortic valve disease secondary to immune complex disease, except 1 patient with rheumatic heart disease. The median cardiopulmonary bypass time was 156 minutes (range, 137 to 226 minutes). Median aortic cross-clamp time was 113 minutes (range, 83 to 162 minutes). The operative technique has been described previously [6].
Echocardiographic data
Preoperative transthoracic echocardiograms were obtained on all patients within 1 week of their operation and confirmed by transesophageal echocardiogram immediately before operation. Postoperative transthoracic echocardiograms were scheduled at 1-week, 6-months, and 1- and 2-year intervals. Ejection fraction, left ventricular end-diastolic dimension, left ventricular end-systolic dimension, left ventricular posterior wall thickness, and interventricular septal thickness were evaluated and documented. Left ventricular mass [7], as well as left ventricular mass index (LVMI) and Z values [8] were calculated from the echocardiographic data. Echocardiograms of the individual patients were obtained during routine clinical follow-up of the patients and then reviewed by our cardiologist for adequacy of data collection.
Follow-up
Follow-up data were obtained by retrospective review of each patient’s clinical reports and echocardiographic data. Improvement in a patient’s clinical status was determined from subjective reports as well as objective examination data. Follow-up was greater than 2 years in 10 of 15 (67%) patients. Median follow-up was 3.6 years with a maximum follow-up of 8 years and minimum of 1 year. Records were reviewed for survival, clinical status, EF, left ventricular dimensions, and valve function at each time period.
Statistical analysis
All analyses were performed using SAS software, version 8.1 (SAS Institute, Cary, NC). Differences in continuous variables were tested using analysis of variance, with repeated-measures techniques for changes over time. The 
2 or Fisher exact methods were used to test differences between proportions. P values of less than 0.05 were considered significant, and all p values are reported. Mean values are reported ± one standard deviation.
|
Results
|
|---|
There were no operative deaths, late deaths, or reoperations in the 15 patients with low preoperative LV function. No patients required postoperative intraaortic balloon pump for hemodynamic instability. The average length of hospital stay was 9.3 days (range, 5 to 30 days). All patients had follow-up within the past year, and all are clinically well. Sixty-seven percent of patients have more than 2 years of follow-up.
All 15 patients had improved left ventricular function. Fourteen of 15 patients (93%) had substantially improved left ventricular function, with a greater than 20% increase in EF from 6 months to 2 years postoperatively. The mean EF preoperatively was 31.1% ± 6.5% and at 2 years was 51.2% ± 10.2%. The increase in EF was statistically significant at all time intervals (p < 0.02) (Fig 1).
The left ventricular end-diastolic dimension decreased from a mean of 68 ± 8 mm (range, 58 to 86 mm) preoperatively to a mean of 55 ± 6 mm (range, 44 to 56 mm) in the 15 patients with echocardiograms at 6 months. In the 10 patients with 2-year postoperative echocardiograms, the left ventricular end-diastolic dimension was a mean of 52 ± 6 mm with a range of 43 to 61 mm (Fig 2).
The left ventricular end-systolic dimension was a mean of 52 ± 10 mm with a range of 36 to 75 mm preoperatively. At 6 months postoperatively, the left ventricular end-systolic dimension had decreased to a mean of 39 ± 7 mm with a range of 26 to 51 mm (Fig 3). The mean LVMI preoperatively was 256 ± 107 g/m2 with a range of 144 to 505 g/m2. The mean LVMI decreased by 39% at 6 months (n = 10, p = 0.009) and by 42% at 2 years (n = 5, p = 0.21). The mean LVMI at 2 years was 144 ± 44 g/m2 with a range of 72 to 206 g/m2 (Fig 4).
Mean Z values for LV mass decreased from 7.2 to 3.0 at more than 2 years (p = 0.036 to 0.186). The Z value for LV mass decreased in all patients and was less than 2.0 in 3 of 10 patients at 2 years.
View larger version (12K):
[in this window]
[in a new window]
|
Fig 1. Left ventricular ejection fraction at time of operation and postoperatively at 1 week, 6 months, 1 year, and 2 years. Upper and lower lines indicate one standard error of the mean.
|
|
View larger version (9K):
[in this window]
[in a new window]
|
Fig 2. Left ventricular end-diastolic dimension in millimeters at time of operation and postoperatively at 1 week, 6 months, 1 year, and 2 years. Upper and lower lines indicate one standard error of the mean.
|
|
View larger version (10K):
[in this window]
[in a new window]
|
Fig 3. Left ventricular end-systolic dimension in millimeters at time of operation and postoperatively at 1 week, 6 months, 1 year, and 2 years. Upper and lower lines indicate one standard error of the mean.
|
|
View larger version (11K):
[in this window]
[in a new window]
|
Fig 4. Left ventricular mass index in grams per meter squared at time of operation and postoperatively at 1 week, 6 months, 1 year, and 2 years. Upper and lower lines indicate one standard error of the mean.
|
|
|
Comment
|
|---|
Guidelines for the treatment of patients with aortic valve disease continue to evolve as the effect of operation and patient factors are better understood, with the primary purpose of valve replacement being to relieve symptoms and abnormal hemodynamics. The traditionally accepted view has been that operative risk and long-term survival in patients with aortic valve disease is worse if combined with severe left ventricular dysfunction. Only recently has it been shown that survival is significantly improved by aortic valve replacement in patients with severe LV dysfunction [9]. At operation, the surgeon and patient have several options, including valve repair, prosthetic or bioprosthetic valve replacement, allograft valve replacement, or Ross operation. Each of these options has significant advantages and disadvantages. A prosthetic valve replacement could be favored for its ease of insertion and durability; however, its disadvantages include the need for anticoagulation, the constant risk of thromboembolism, anticoagulant-related bleeding, and endocarditis. Also, the abnormal flow pattern of a prosthetic valve may not be conducive to similar LV remodeling seen with stentless valve replacements. A stented bioprosthetic aortic valve replacement has a similar ease of insertion, does not require long-term anticoagulation in most patients, and has a limited risk of anticoagulant-related bleeding. It retains a risk of endocarditis and abnormal flow patterns, but most importantly in patients aged 16 to 50 years it is associated with accelerated structural valve degeneration. Stentless bioprosthetic valves are relatively easy to insert, do not require anticoagulation, but their durability in this age group has not been proven to be significantly longer than that of a stented bioprosthetic valve. Valve repair has all of the advantages of an allograft and an autograft but is technically demanding and requires a near-perfect hemodynamic result for long-term durability. It also is not applicable for many patients with aortic valve disease. The allograft valve replacement is less demanding than the autograft, with perhaps shorter ischemic times and similar hemodynamic results; however, in patients 30 years old and younger, structural valve degeneration is accelerated. The autograft valve is almost uniformly applicable, but its major deterrent is the technical difficulty with relatively long ischemic times. Long-term durability of the root replacement in a large series of patients has yet to be proven. Its normal hemodynamics at rest and with extreme exercise, the avoidance of anticoagulation, and low risk of endocarditis are appealing characteristics.
Recent studies have focused on the impact of aortic valve replacement on ventricular recovery. By correcting the primary valve lesion, the hypertrophic stimulus is removed, which allows for left ventricular remodeling to occur [9]. This is reflected in decreasing echocardiographic dimensions and LVMI, as well as improved EF. A recent study from the Mayo Clinic concluded (1) that the only predictor of postoperative survival after aortic valve replacement was improvement in postoperative functional class and (2) that functional improvement occurred in most patients postoperatively (74% showed improvement in their series) [10]. This improvement in functional status was thought to be the result of ventricular remodeling. In a recent retrospective study of aortic valve replacement and severe left ventricular dysfunction from the Cleveland Clinic, the authors concluded that even with severe dysfunction, left ventricular dilation, and New York Heart Association class III and IV symptoms, chronic aortic valve disease should be treated with valve replacement. Good early and late survival is to be expected along with improvement in ventricular function [11].
The type of aortic valve replacement has been the focus of recent attention. Stentless aortic valves with more normal aortic valve hemodynamics have been suggested to allow greater resolution of LV hypertrophy and perhaps improved long-term survival. Some centers have shown greater regression of LV hypertrophy with the use of stentless aortic valves compared with stented bioprostheses or prosthetic aortic valves [12–15]. Others have shown no difference in the performance of a stentless bioprosthesis versus a stented bioprosthesis in terms of reduction of LV mass and functional status between the two groups [16]. More importantly, a recent study by Medalion and colleagues [17] demonstrated that moderate mismatch of the patient’s prosthesis did not affect long-term survival after aortic valve replacement. That study involved prosthetic, bioprosthetic, and allograft aortic valves but did not assess the effect of valve choice or patient prosthesis mismatch on aortic valve replacement in patients with severe LV dysfunction.
Intuitively, the use of a tissue valve to maximize the hemodynamic benefit of an aortic valve replacement in patients with left ventricular failure and aortic valve disease has led to the increased use of stentless tissue valves (xenograft or allograft valves). Grocott-Mason and associates [18] concluded that the long-term outcome of homograft aortic valve replacement in patients with congestive heart failure is acceptable, with a reduced rate of serious complications and death compared with other valve substitutes. In their series of 301 patients, homografts independently decreased the rate of serious complications and cardiac death compared with mechanical valve or xenograft replacement. However, they determined that further improvements in ventricular recovery would be expected if the rate of primary tissue failure was reduced. No similar study has been performed in patients who had a Ross operation.
Results of the pulmonary autograft for aortic valve replacement have been excellent, with rare development of late autograft dysfunction or dilation. In 1999, Elkins [19] reviewed his 12-year experience with the Ross operation and documented excellent hemodynamic results, with limited incidence of reoperation and replacement of the autograft valve. Legarra and coworkers [5] reviewed the outcomes of 30 patients with aortic valve disease who underwent a Ross procedure; they found normalization of left ventricular dimensions and improvement of left ventricular function early in the postoperative period, as is seen with other replacements for the aortic valve. Thus, the use of the autograft valve in patients with aortic valve disease and severe left ventricular dysfunction would be justified if there is no increase in perioperative complications or late mortality rate.
Those findings prompted a review of our experience with the Ross operation in patients with severe left ventricular dysfunction. Of 226 consecutive pulmonary autografts in patients aged 18 to 50 years, 15 patients were identified with LVEF less than 40%. In this small subset of patients, we have shown excellent early and midterm results. The absence of operative or late deaths suggests that there is not an increased risk associated with the use of the pulmonary autograft for aortic valve replacement in patients with ventricular dysfunction when the operative procedure is performed by an experienced surgeon. All patients had improved functional status with an increase in EF that was statistically significant at all time intervals. Echocardiographic measurements of left ventricular dimensions documented decreasing systolic and diastolic measurements at all time intervals, with a decrease in LVMI indicative of ventricular remodeling. Of the patients in this series who had both aortic insufficiency and a preoperative left ventricular end-systolic dimension of more than 55 mm—a group with generally poor ventricular recovery [20]—all 4 patients had substantial recovery of ventricular function, with echocardiographic measurements reflecting ventricular remodeling. The benefits of the Ross operation in terms of durability and longevity have been identified in our group of patients [19, 21]. We conclude that the surgical results, early clinical outcomes, and valve function are satisfactory after aortic valve replacement with a pulmonary autograft in the presence of left ventricular dysfunction in adults 50 years of age or less. Longer follow-up and careful evaluation in additional patients will define its role in this challenging clinical problem.
|
Discussion
|
|---|
DR JOHN D. OSWALT (Austin, TX): As everybody knows, Dr Elkins is the master at doing the Ross procedure in the United States, and certainly this paper deserves congratulations. This obviously shows that you do excellent myocardial protection during the procedure, and I too have found that by having proper myocardial protection you can do these patients who have an ejection fraction as low as 15%. We have two in our series that were at 15%. These were in young 18- and 14-year-old children who just had huge hearts with congenital aortic stenosis, and these patients both did well, and, again, secondary to good myocardial perfusion.
I also think the autograft is a superb choice because of the low gradient that you leave the patient with, so the ventricle does not have to have a 20 or even 15 mm gradient to work against. If you put a mechanical valve in these younger people, there will be a gradient, and the ventricle may struggle to recover initially. I think it is the best choice to help that sick ventricle, and I think that also adds to your excellent results.
|
References
|
|---|
- Ross D., Jackson M., Davies J. The pulmonary autograft–a permanent aortic valve. Eur J Cardiothorac Surg 1992;6:113-117.[Abstract]
- Ross D. Pulmonary valve autotransplantation (the Ross operation). J Cardiac Surg 1988;3:313-319.[Medline]
- Elkins R.C., Knott-Craig C.J., Ward K.E., McCue C., Lane M.M. Pulmonary autograft in children: realized growth potential. Ann Thorac Surg 1994;57:1387-1394.[Abstract]
- Santangelo K., Elkins R.C., Stelzer P., et al. Normal left ventricular function following pulmonary autograft replacement of the aortic valve in children. J Cardiac Surg 1991;6:633-637.[Medline]
- Legarra J.J., Concha M., Casares J., Merino C., Munoz I., Alados P. Left ventricular remodeling after pulmonary autograft replacement of the aortic valve (Ross operation). J Heart Valve Dis 2001;10(1):43-48.[Medline]
- Elkins R.C. Pulmonary autograft. In: Franco K.L., Verrier E.D., eds. Advanced therapy in cardiac surgery. Hamilton, Ontario: B.C. Decker, Inc, 1999:183-194.
- Devereux R.B., Reichek N. Echocardiographic determination of left ventricular mass in man. Anatomic validation of the method. Circulation 1977;55:613-618.[Abstract/Free Full Text]
- Kirklin J.W., Barratt-Boyes B.G. Cardiac surgery, 2nd ed New York, NY: Churchill Livingstone, 1993.
- Pereira J.J., Lauer M.S., Bashir M., et al. Survival after aortic valve replacement for severe aortic stenosis with low transvalvular gradients and severe left ventricular dysfunction. J Am Coll Cardiol 2002;39(8):1356-1363.[Abstract/Free Full Text]
- Powell D.E., Tunick P.A., Rosenzweig B.P., et al. Aortic valve replacement in patients with aortic stenosis and severe left ventricular dysfunction. Arch Intern Med 2000;160(9):1337-1341.[Abstract/Free Full Text]
- McCarthy P.M. Aortic valve surgery in patients with left ventricular dysfunction. Semin Thorac Cardiovasc Surg 2002;14(2):137-143.[Medline]
- Walther T., Falk V., Langebartels G., et al. Prospectively randomized evaluation of stentless versus conventional biological aortic valves: impact on early regression of left ventricular hypertrophy. Circulation 1999;100(19 Suppl):II6-II10.
- Walther T., Falk V., Langebartels G., et al. Regression of left ventricular hypertrophy after stentless versus conventional aortic valve replacement. Semin Thorac Cardiovasc Surg 1999;11(4 Suppl 1):18-21.[Medline]
- Gelsomino S., Morocutti G., Frassani R., et al. Early recovery of left ventricular function after stentless versus stented aortic valve replacement for pure aortic stenosis and severe cardiac dysfunction. Semin Thorac Cardiovasc Surg 2001;13(4 Suppl 1):120-128.[Medline]
- Collinson J., Henein M., Flather M., Pepper J.R., Gibson D.G. Valve replacement for aortic stenosis in patients with poor left ventricular function: comparison of early changes with stented and stentless valves. Circulation 1999;100(19 Suppl):II1-II5.
- Cohen G., Christakis G.T., Joyner C.D., et al. Are stentless valves hemodynamically superior to stented valves? A prospective randomized trial. Ann Thorac Surg 2002;73(3):767-775.[Abstract/Free Full Text]
- Medalion B., Blackstone E.H., Lytle B.W., White J., Arnold J.H., Cosgrove D.M. Aortic valve replacement: is valve size important?. J Thorac Cardiovasc Surg 2001;119:963-974.
- Grocott-Mason R.M., Lund O., Elwidaa H., et al. Long-term results after aortic valve replacement in patients with congestive heart failure. Homografts vs prosthetic valves. Eur Heart J 2000;21(20):1698-1707.[Abstract/Free Full Text]
- Elkins R.C. The Ross operation - a 12 year experience. Ann Thorac Surg 1999;68:S14-S18.
- Corti R., Binggeli C., Turina M., Jenni R., Luscher T.F., Turina J. Predictors of long-term survival after valve replacement for chronic aortic regurgitation: is M-mode echocardiography sufficient?. Eur Heart J 2001;22(10):866-873.[Abstract/Free Full Text]
- Elkins R.C., Lane M.M., McCue C., Ward K.E. Pulmonary autograft root replacement: mid-term results. J Heart Valve Dis 1999;8:499-506.[Medline]
Top
Abstract
Introduction
