Ann Thorac Surg 2009;88:752-756. doi:10.1016/j.athoracsur.2009.05.025
© 2009 The Society of Thoracic Surgeons
Original Articles: Adult Cardiac
Survival Benefit of Aortic Valve Replacement in Patients With Severe Aortic Regurgitation and Pulmonary Hypertension
Sumit Khandhar, DO,
Padmini Varadarajan, MD,
Rami Turk, MD,
Unnati Sampat, MD,
Reena Patel, MD,
Ashvin Kamath, BS,
Ramdas G. Pai, MD*
Division of Cardiology, Loma Linda University Medical Center, Loma Linda, California
Accepted for publication May 7, 2009.
* Address correspondence to Dr Pai, Division of Cardiology, 11234 Anderson St, Room 4414, Loma Linda University Medical Center, Loma Linda, CA 92354 (Email: ramdaspai{at}yahoo.com).
 |
Abstract
|
|---|
Background: Severe pulmonary hypertension occurs in approximately 10% of patients with severe aortic regurgitation (AR). The potential survival benefit of aortic valve replacement (AVR) in these patients is not known, and was analyzed in a large cohort of patients.
Methods: Our echocardiographic data was screened for severe AR patients with severe pulmonary hypertension defined as pulmonary artery systolic pressure of 60 mm Hg or greater. Chart reviews were performed for clinical, pharmacologic, and surgical details, and survival data were analyzed as a function of AVR.
Results: Of the 506 patients with severe AR and measurable pulmonary artery pressures by echocardiography, 83 had severe pulmonary hypertension defined as a pulmonary artery systolic pressure of 60 mm Hg or greater. Severe pulmonary hypertension was associated with lower left ventricular ejection fraction (47% ± 22% versus 53% ± 19%, p = 0.006), larger left ventricular size (p = 0.03), and higher grades of mitral regurgitation (2.7 ± 1.2 versus 1.7 ± 1.1, p < 0.0001). Of the 83 patients with severe pulmonary hypertension, 32 underwent AVR, which was associated with better survival compared with patients who did not (1-year survival 90% versus 58% and 5-year survival 62% versus 22%, respectively; p = 0.004). After adjusting for comorbidities, AVR remained an independent predictor of better survival (hazard ratio 0.45, 95% confidence interval: 0.22 to 0.92, p = 0.03). This survival benefit of AVR was further supported by propensity score analysis.
Conclusions: Severe pulmonary hypertension occurs in approximately 16% of patients with severe AR and is associated with left ventricular enlargement with dysfunction and resultant mitral regurgitation. Aortic valve replacement is associated with an independent survival benefit in these patients.
|
Introduction
|
|---|
Pulmonary hypertension (PHT) in patients with severe aortic regurgitation (AR) is not uncommon, and these patients are frequently denied aortic valve surgery fearing a high surgical risk. In small observational studies, PHT has been shown to be reflective of the late stages of aortic valve disease, and to reflect high left ventricular (LV) end-diastolic pressure [1]. In a series of 151 patients with isolated aortic valve disease, 17 patients (11%) had severe PHT defined as pulmonary arterial systolic pressure greater than 60 mm Hg [1]. Although aortic valve replacement (AVR) can be performed with reasonable mortality in these patients, it is not known if it results in a survival benefit [2]. We analyzed the incidence and risk factors for severe PHT, as well as the effect of AVR in a large cohort of severe AR patients comprehensively characterized in terms of clinical comorbidities and pharmacologic therapy, along with echocardiographic and surgical details.
|
Patients and Methods
|
|---|
Patient Population
This retrospective cohort study is from a large university medical center. This study was approved by our Institutional Review Board, which waived the need for patient consent because of the retrospective nature of the study. As described previously, our echocardiographic database between the years 1993 and 2007 was searched for patients with severe AR, defined by a jet height ratio of greater than 60% or prominent holodiastolic flow reversal in the distal aortic arch or abdominal aorta [3]. This yielded a total of 785 patients of whom 506 had measurable tricuspid regurgitation velocity signals to measure the pulmonary artery systolic pressure. From this cohort, 83 patients had severe PHT, defined as pulmonary artery systolic pressure of 60 mm Hg or greater. Complete clinical, echocardiographic, and pharmacologic data were compiled from comprehensive chart reviews. Of the 83 patients with severe PHT, 32 underwent AVR, and the remaining 51 patients were treated medically. Operative details were obtained for these patients. Decision for AVR was made by treating physicians in conjunction with the patients, and not by the investigators.
Clinical Variables
Systemic hypertension was defined as blood pressure greater than 140/90 mm Hg, or a history of hypertension and receiving antihypertensive medications. Diabetes mellitus was defined as fasting blood sugar greater than 125 mg/dL or receiving antidiabetic agents. Coronary artery disease was deemed to be present if any of the following were evident: a history of myocardial infarction, angiographic evidence of coronary artery disease, coronary intervention, or coronary artery bypass surgery. Heart failure was diagnosed using Framingham criteria [4].
Pharmacologic Data
Pharmacotherapy at the time of echocardiography was recorded. This was broadly categorized into beta-blockers, dihydropyridine calcium-channel blockers, nondihydropyridine calcium-channel blockers, angiotensin-converting enzyme inhibitors, aspirin, and statins.
Echocardiographic Data
All patients had complete two-dimensional echocardiographic examinations. Left ventricular ejection fraction was assessed visually by a level 3-trained echocardiographer (a cardiologist with at least 1 year of training in echocardiography and qualified to direct the echocardiographic laboratory) and entered into a database at the time of the examination. That has been shown to be reliable and has been validated against contrast and radionuclide left ventricular angiography [5, 6]. Anatomic and Doppler examinations and measurements were performed according to the recommendations of the American Society of Echocardiography [7]. The pulmonary artery systolic pressure was calculated from the tricuspid regurgitation velocity signal using the simplified Bernouli equation and estimated right atrial pressure.
Mortality Data
The primary endpoint of the study was all-cause mortality. Mortality data were obtained from the National Death Index using the social security numbers.
Statistical Analysis
The data were imported into the Stat View 5.01 (SAS Institute, Cary, NC) program for statistical analysis. Group comparisons were made using the Student t test for continuous variables and the 
2 test for categorical variables. Survival analysis was performed using various statistical tools such as Kaplan-Meier analysis, Cox regression models (including time varying Cox regression), and propensity score analysis, as discussed later in the results section. A p value of 0.05 or less was considered statistically significant.
|
Results
|
|---|
Baseline Patient Characteristics
The baseline features of the 83 patients with severe PHT are summarized in Table 1, along with the 423 patients without PHT. The two groups were comparable in terms of age, sex, LV wall thickness, and comorbidities such as hypertension, diabetes mellitus, and coronary artery disease. However, patients with PHT had greater LV end-diastolic dimensions (5.9 ± 1.2 cm versus 5.6 ± 1.0 cm, p = 0.03) and end-systolic dimensions (4.2 ± 1.3 cm versus 3.9 ± 1.2 cm, p = 0.02), lower LV ejection fraction (47% ± 22% versus 53% ± 19%, p = 0.006), and higher grades of mitral regurgitation (MR [2.7 ± 1.2 versus 1.7 ± 1.1, p < 0.0001]). Atrial fibrillation was more frequent in patients with PHT (43 versus 30%, p = 0.01).
Aortic Valve Replacement
Of the 83 patients with PHT, 32 had AVR—mechanical in 24 and bioprosthetic in 8. The concomitant surgical procedures included aortic root replacement in 4, coronary artery bypass surgery in 8, mitral valve repair in 8, and mitral valve replacement in 8. Table 2
summarizes the characteristics of patients undergoing AVR and those treated medically. Patients who underwent AVR were significantly younger than those patients who were managed medically (53 ± 16 years versus 71 ± 14 years, p = 0.0001).
AVR and Survival
As shown in Figure 1, patients receiving AVR had 1-year survivals of 90% compared with 58% for patiens who did not receive AVR, indicating a rapid attrition without surgical therapy. The 5-year survivals were 62% and 22%, respectively (p = 0.004). The operative mortality was 3%.
View larger version (17K):
[in this window]
[in a new window]
|
Fig 1. Survival curves of severe aortic regurgitation patients with severe pulmonary hypertension with and without aortic valve replacement (AVR).
|
|
Cox Regression Analysis
Different Cox regression models were used. In the first one, adjustment was made for group differences between the AVR and no AVR groups. After adjusting for age, sex, LV ejection fraction, and LV end-diastolic dimension, AVR remained an independent predictor of reduced mortality (hazard ratio 0.40, 95% confidence interval: 0.22 to 0.85, p = 0.02). The other group differences, including age and ejection fraction, were nonsignificant. In another model, all the covariates (clinical, echocardiographic, and pharmacologic) listed in Table 2 were used as confounders: AVR remained an independent predictor of reduced mortality (hazard ratio 0.35, p = 0.01). The other predictor of outcome was renal insufficiency, which increased the risk of mortality.
Propensity Score Analysis
The propensity score for an individual is the probability of receiving a treatment only on the basis of an individual's covariate values [8–10]. Given the observational nature of this study, the intent of propensity score analysis is to balance the difference in pretreatment covariates to reduce the impact of treatment bias [8–10]. A propensity score for receiving AVR was calculated using logistic regression analysis based on clinically relevant covariate imbalances, age, sex, rhythm, coronary disease, diabetes, renal insufficiency, and hypertension. After adjusting for the propensity score using the Cox regresson model, AVR carried a mortality hazard ratio of 0.36 (95% confidence interval: 0.18 to 0.71, p = 0.004).
Sensitivity Analysis
To analyze the impact of nonproportional early hazard in the medically treated patients and further reduce the impact of treatment bias of less sick patients, sensitivity analysis was performed by serially eliminating observations with duration of follow-up of less than 30 days, less than 90 days, and less than 6 months (Fig 2). Survival benefit of AVR was evident even with elimination of observation of less than 90 days, further supporting the robustness of the survival benefit of AVR. Table 3
compares the patients included in Figure 2C and illustrates that both AVR and non-AVR groups were similar, apart from a difference in age.
View larger version (13K):
[in this window]
[in a new window]
|
Fig 2. Results of sensitivity analysis sequentially eliminating patients with durations of follow-up of less than 1 (A), 3 (B), and 6 (C) months, respectively. (AVR = aortic valve replacement.)
|
|
Survival Benefit With AVR as Function of PHT
As shown in Figure 3, AVR was associated with a survival benefit in both patients with and without PHT. Pulmonary hypertension was associated with a higher mortality rate in both the surgically and medically treated patients. The lowest mortality was seen among patients without PHT who underwent AVR, and the highest mortality was among patients with PHT who did not undergo AVR.
View larger version (20K):
[in this window]
[in a new window]
|
Fig 3. Survival with aortic valve replacement (AVR) as a function of presence or absence of severe pulmonary hypertension (PHT).
|
|
Effect of AVR on Pulmonary Artery Pressure and MR
Of the 32 patients undergoing AVR, 20 had follow-up echocardiograms performed at our institution. The pulmonary artery pressure dropped from 66 ± 7 mm Hg to 31 ± 5 mm Hg in these patients. Of these patients, 12 had 3+ or 4+ MR preoperatively and only 1 had 3+ MR on the postoperative echocardiogram.
|
Comment
|
|---|
This observational study confirms that AVR can be performed with an acceptable operative mortality rate (3%) and that it may confer a survival benefit to severe AR patients with associated severe PHT. Our series is larger than all other reported series and has, additionally, comprehensive echocardiographic and pharmacologic data that have potential prognostic implications.
Frequency and Mechanism of PHT
In a study of 151 patients with severe aortic valve disease, Basu and colleagues [1], in 1978, found a 11% incidence of severe PHT and showed it to be related to increased pulmonary arteriolar resistance secondary to raised left atrial pressure. Naidoo and associates [6], in 1991, reviewed the hemodynamic data of 139 patients with severe, chronic AR and found severe PHT in 34 patients (24%) consequent to raised LV end-diastolic pressure. Our study included 506 severe AR patients with measured pulmonary artery pressure, with 83 patients (16%) having severe PHT. Although we do not have measured left atrial pressures, association of PHT with LV enlargement and systolic dysfunction resulting in MR lends support to left-side heart failure as the basis of PHT in these patients.
Outcomes With AVR
The only study addressing the issue of AVR in this population is that of Naidoo and coworkers [2]. Their study showed that severe PHT does not affect the early surgical outcome after AVR, and that pulmonary artery systolic pressure reverted to normal in the majority of patients after AVR. Our study has addressed the question of whether AVR results in a survival benefit in these patients, and it shows that AVR can be performed with a low mortality rate (3%) in these patients. Based on the best statistical tools available to analyze observational data, our study also shows that AVR may offer a large survival benefit to severe AR patients with PHT. This is borne out by Cox regression models adjusting for group differences as well as propensity score analysis, which may remove 85% to 90% of the bias associated with observational data sets [8–10]. As there have been no randomized trials addressing the question of AVR in severe AR patients with PHT and none is likely in the near future, we believe the results of this study should support AVR in such patients. One should also be vigilant for the appearance of any functional MR, which seems to be a harbinger of PHT. The 5-year survival was 62% in the surgically treated group compared with 22% in the medically treated group. As this survival is lower than for patients without PHT, we suggest that AVR should be offered at a much earlier stage in the natural history of severe AR before pulmonary hypertension develops.
Study Limitations
This is an observational study with all its inherent limitations. But we have used the best available statistical tools to address the treatment bias, which, however, can not be completely be eliminated. Only a randomized study can prove unequivocally whether AVR is beneficial for these patients.
In conclusion, our study confirms that AVR can be performed with an acceptable mortality rate in severe AR patients with severe PHT. For these patients, AVR is associated with an impressive survival benefit. Medically treated patients have a dismal prognosis. There is a general reluctance to offer AVR to such patients but AVR should be offered before the appearance MR or PHT. Our study confirms that the presence of severe pulmonary hypertension should not constitute a contraindication to AVR.
|
References
|
|---|
- Basu B, Cherian G, Krishnaswami S, Sukumar IP, John S. Severe pulmonary hypertension in advanced aortic valve disease Br Heart J 1978;40:1310-1313.[Abstract/Free Full Text]
- Naidoo DP, Mitha AS, Vythilingum S, Chetty S. Pulmonary hypertension in aortic regurgitation: early surgical outcome Q J Med 1991;80:589-595.[Abstract/Free Full Text]
- Sampat U, Varadarajan P, Turk R, Kamath A, Khandhar S, Pai RG. Effect of beta blocker therapy on survival in patients with severe aortic regurgitation: results from a cohort of 756 patients J Am Coll Cardiol 2009;54:458-459.[Free Full Text]
- McKee PA, Castelli WP, McNamara PM. The natural history of congestive heart failure: the Framingham Study N Engl J Med 1971;85:1441-1446.
- van Royen N, Jaffe CC, Krumholz HM, et al. Comparison and reproducibility of visual echocardiographic and quantitative radionuclide left ventricular ejection fractions Am J Cardiol 1996;77:843-850.[Medline]
- Amico AF, Lichtenberg GS, Reisner SA, Stone CK, Schwartz RG, Meltzer RS. Superiority of visual versus computerized echocardiographic estimation of radionuclide left ventricular ejection fraction Am Heart J 1989;118:1259-1265.[Medline]
- Schiller NB, Shah PM, Crawford M, et al. Recommendations for quantitation of the left ventricle by two-dimensional echocardiography J Am Soc Echocardiogr 1989;2:358-367.[Medline]
- D'Agastino RB. Propensity score methods for bias reduction in the comparison of a treatment group to a nonrandomized control group Stat Med 1998;17:2265-2281.[Medline]
- Cochran W. Planning and analysis of observational studiesNew York: Wiley & Sons; 1983.
- D'Agastino RB. Propensity scores in cardiovascular research Circulation 2007;115:2340-2343.[Free Full Text]
Related Article
-
Invited Commentary
- Ryan K. Kaple
Ann. Thorac. Surg. 2009 88: 757.
[Extract]
[Full Text]
[PDF]
This article has been cited by other articles:
|
|
|
|
 
R. K. Kaple
Invited Commentary
Ann. Thorac. Surg.,
September 1, 2009;
88(3):
757 - 757.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
