Finding ‘Just the Right Moment’ for Operative Intervention in the Asymptomatic Patient With Moderate to Severe Aortic Regurgitation
The decision to recommend operative intervention to the patient with valve disease is the most difficult when the valve disease is severe chronic aortic regurgitation and the patient is asymptomatic or minimally symptomatic. Because aortic valve replacement is almost always necessary and because the biological and mechanical valves still have problems resulting in significant mortality and morbidity, the decision to recommend surgery in an asymptomatic patient must wait until there are indications that changes are occurring that will predict an increased risk of death or congestive heart failure even after valve replacement.
The goals of operative intervention in valve disease are to (1) decrease or diminish symptomatology and thus to improve the quality of life for the patient; (2) prevent “catastrophes” or problems that are irreversible, such as myocardial infarction, left ventricular (LV) fibrosis, decreasing myocardial contractility, thromboemboli, and sudden death; and (3) prolong life. There is little doubt that the first goal is achievable, that of decreasing symptomatology. Every study reported of aortic valve replacement succeeds in that goal. If the patient survives, the chances of improvement in symptoms and functional classification are excellent.
Success of the third goal, that of prolonging life, is difficult to prove. The problem is that the mortality rate in patients with aortic regurgitation, especially those who are asymptomatic or only mildly symptomatic, is very low, and it would take large numbers of patients followed for a long period of time to randomize for study. This is not likely to be done. To randomize patients who are symptomatic would be unethical because surgery would improve the quality of life, even if it did not prolong survival.
Consequently, the most difficult task for the cardiologist who is following an asymptomatic or minimally symptomatic patient with moderate to severe aortic regurgitation is to choose which parameters to watch that will predict a less-than-optimal result if the patient is operated on. The time to replace the valve is, therefore, just before that happens. In this issue of Circulation, Borer and colleagues1 offer evidence that they have found “just the right moment” to recommend surgery in the asymptomatic or minimally symptomatic patient with chronic severe aortic regurgitation and normal LV performance.
There have been many attempts to define the preoperative clinical or hemodynamic parameters that would be predictive of cardiac death, poor left ventricular function, or congestive heart failure—that is, predictive of less-than-optimal results after valve replacement. Ideally, we would like to look at the state of myocardial contractility. Conventional wisdom says that once contractility has decreased, it will remain impaired even after valve replacement, and many patients will continue to progress to early congestive heart failure and premature cardiac death in the years after valve replacement. As physicians, we can follow LV performance but have difficulty finding a way of detecting a decrease in myocardial contractility as distinct from changes in LV preload and afterload. Most valuable was the work of Suga and Sagawa2 in 1974 in which they described the end-systolic pressure-volume relationship as being independent of preload and afterload. With decreasing myocardial contractility, the slope of this relationship decreases and moves to the right. In practical terms, this rightward shift and change in slope of the end-systolic pressure-volume relationship are detected by following the LV systolic and end-diastolic volumes. An increasing end-systolic and end- diastolic LV volume and decreasing ejection fraction indicate a probable decrease in myocardial contractility.
In patients with aortic regurgitation, this principle has been applied in a number of studies, most notably a study by Henry and colleagues3 4 in 1980, in which LV end-systolic diameter by M-mode echocardiography of >55 mm and fractional shortening of <25% predicted operative mortality and postoperative late mortality and congestive heart failure. This report stimulated other studies that threw doubt on the findings, notably a study by Daniel and colleagues,5 in which the same M-mode end-systolic diameter or percent fractional shortening was not predictive of late mortality. The reason for this discrepancy is possibly explained by the fact that many of the patients of Henry and colleagues at the National Institutes of Health were seen early in the era of aortic valve replacement, when myocardial preservation during surgery was poor. It is not surprising that the larger the heart before surgery, the more likely is intraoperative myocardial damage and late postoperative mortality, and the chance of developing congestive heart failure.
Subsequently, other studies considered symptoms and other measures of LV performance, mainly resting LV ejection fraction (LVEF) and exercise capacity, and all showed varying success in predicting postoperative improvement in ventricular performance and even survival. Combinations of several of these factors made prediction of survival even better. For instance, in the study of Greves and colleagues,6 patients operated on for aortic insufficiency had a low 5-year cardiac mortality if resting LVEF was >50% regardless of the degree of symptomatology. Even patients with LVEF of <50% did well if they were relatively asymptomatic. Only those with New York Heart Association (NYHA) functional class III and IV symptoms and a decreased LVEF had a significantly poorer 5-year survival. Bonow and colleagues7 showed in symptomatic patients with severe aortic regurgitation that those who had good exercise capacity on the treadmill had no operative mortality and 100% survival at 3 years and had a greater return toward normal of the LV performance than did those with poor exercise capacity.
Gaasch and colleagues8 used an index of preoperative resting peak LV wall stress ([LV systolic pressure×LV end-diastolic radius]/LV wall thickness) combined with LV end-systolic dimension to predict late postoperative survival. Carabello and colleagues9 looked at afterload-corrected end-systolic volume index (end-systolic stress/end-systolic volume index) before surgery and showed that a high ratio predicted a significantly poor postoperative outcome, high mortality, and persistent symptoms. However, they had only nine patients with aortic regurgitation.
In this issue of Circulation, Borer and colleagues1 made an important contribution to the evidence that stress-normalized LV performance is an important predictor of the development of symptomatology, LV dysfunction, and even sudden death in asymptomatic or mildly symptomatic patients with aortic regurgitation. It is remarkable, indeed, that in this era in which the rush to publication is so frantic, Borer and colleagues at New York Hospital-Cornell Medical Center have persisted in performing a prospective study over a period of 15 years. Starting in 1979, they systematically collected clinical information on patients with asymptomatic or mildly symptomatic aortic regurgitation. It was a complicated study requiring rest and exercise radionuclide angiograms and a resting echocardiogram and then follow-up yearly to document the development of cardiac end points: symptoms justifying valve replacement, cardiac death, or the development of resting LV dysfunction as defined by an LVEF below their laboratory’s normal limits.
They had a total of 104 patients: 83 asymptomatic patients and 21 with “early NYHA II.” Patients without valve replacement were followed for a mean of 7.3 years (range, 1 to 15 years). There were 39 patients who developed a cardiac end point: 4 died suddenly, 22 developed symptoms leading to surgery, and 13 developed subnormal LV performance with or without symptoms. The annual rate of progression was 6.2%, which is similar to the rate of progression to aortic valve replacement in other studies by Bonow et al (5%)10 and Siemienczuk et al (4%).11
Univariate analysis of potential LV function or volume predictors of adverse outcomes showed the only variables significantly predicting all cardiac end points were ΔLVEF, rest-to-exercise alone, and ΔLVEF rest-to-exercise normalized for Δend-systolic stress (ESS), rest-to-exercise. Significant associations were present between at least one, but not all, end points for other predictors by analysis constructing a multivariate Cox proportional hazards model. The only independent objective predictor of risk of end points was ΔLVEF/ΔESS. However, the much simpler ΔLVEF alone was almost as efficient in predicting the risk of developing operable symptoms and LV dysfunction with or without symptoms.
The authors are to be congratulated for such a difficult study, which took a very long time to accumulate the sufficient number of cardiac end points to develop the statistical power to identify clinical and ventricular performance variables that were independent predictors of these cardiac end points.
There are several problems, however, that must be considered in interpreting the results of the study and the recommendations for use of this complicated, load-adjusted performance variable for following patients.
Even with such a long-term prospective study, the number of end points was small. The most frequent end point was the development of symptoms that the patient’s physician believed sufficient to justify valve replacement. There were too few hard end points (sudden death in 4 patients and decreased LV function in 13 patients) to justify multivariate analysis. Symptoms, as such, are a soft end point in that they are subjective. Patients have a remarkable ability to rationalize what they feel, overlook the importance of the sensation, and therefore fail to inform the physician about the symptom. This is especially true if a patient recognizes that the consequence of reporting the feeling is an operation. It is also questionable whether any test is needed to predict the development of symptoms that the patient will report to the physician. It is interesting that in those who initially were “mildly symptomatic” or had “early” NYHA II symptoms, merely the presence of the symptoms by multivariate analysis was an independent predictor of cardiac endpoints. This is not surprising because a principle end point was the development of symptoms of late NYHA II or greater.
For the hard end points of cardiac death and the development of decreased LV function, the univariate analysis showed that the ΔLVEF rest-to-exercise was as good as the more complicated ΔLVEF rest-to-exercise/ΔESS rest-to-exercise. The complicated variable of ΔLVEF/ΔESS required a rest and exercise radionuclide angiogram and a resting echocardiogram. The formulae used to derive LVESS were from M-mode measurements and required a complicated set of assumptions to calculate ESS with exercise. Using M-mode echocariography, there were no measurements of LV length. Whether in these patients with moderate to severe aortic regurgitation there is a change in sphericity of the LV with exercise is unknown. In a study by Tischler and colleagues,12 it was found that in patients with diseases other than aortic regurgitation causing LV dysfunction, with exercise there is a change in shape of the LV, with the sphericity increasing. This strongly correlated with the patient’s ability to exercise, whereas the conventional descriptors of LV function, such as LVED volume, LVES volume, and LVEF, were poor predictors of exercise capacity. Whether the patients with moderate to severe aortic regurgitation in this study changed the sphericity of the LV with exercise is unknown because the length of the LV was not measured.
Furthermore, the mass of the LV was assumed to be unchanged from rest to exercise. However, the blood in capillaries and arterioles constitute a major component of wall mass, and this may change with exercise in a patient with aortic regurgitation. Finally, peak indirectly measured arterial systolic pressure was used as near-equivalent to end-systolic pressure. Although this is reasonable in normal individuals, in the patient with severe chronic aortic regurgitation, the peak-systolic pressure and the end-systolic pressure are markedly different.
Prediction of risk of cardiac end points was made on the basis of studies done on entrance of the patient into the study. The end point in most cases occurred months or years later. The significant finding of this study is that when first studied, if the ΔLVEF/ΔESS is abnormal, even with normal resting LVEF and no symptoms, the patient who is likely to have future problems is already identified.
There are several reasons why a study done at one point in time might not predict future problems. The most common identified etiology for aortic regurgitation seen in this study was non–Marfan’s idiopathic aortic root dilatation, which was present in 26 patients, and infective endocarditis, which was present in 10 patients. Progressive aortic regurgitation, as a result of aortic root dilatation, fibrosis, or tearing of the valve, cannot be predicted by any clinical or LV performance variable done at the time the patient is first seen. Some valves develop progressive calcification, as occurred in three of the patients in this study, which increases LV afterload, and again is not predicted by the initial evaluation. Finally, the indication for surgery may not be symptoms or LV dysfunction but instead the presence of progressive aneurysmal dilatation of the root of the aorta.
The most important problem with this load-adjusted index of LV performance (ΔLVEF rest-to-exercise/ΔESS rest-to-exercise) is that it is too impractical to be clinically useful. The unadjusted ΔLVEF rest-to-exercise, which was recognized by Greenberg and colleagues13 as an indicator of probable increase in LV end-systolic wall stress and decreased ventricular reserve, is, however, useful and had almost the same power to predict the development of symptoms and/or decrease in LV function justifying surgery. Finally, we are not certain of what the prophylactic use of afterload-reduction drugs will do to the predictive value of any of these indices. To the authors’ credit, most of these problems are recognized in the report.
This study supports the practice of following the asymptomatic patient with chronic aortic regurgitation by evaluating the LV systolic performance. Common clinical practice is to repeat echocardiograms periodically in asymptomatic patients with moderate-to-severe chronic aortic regurgitation to detect a decrease in LVEF and an increase in end-systolic volume and, if confirmed by a second study repeated shortly thereafter, to recommend surgery. There are numerous studies, albeit without as long a follow-up period as the present study, that support this approach for predicting good postoperative survival without heart failure. It is not clear whether this approach is inferior to the more complicated study proposed by Borer and colleagues.1
Finally, we should remember that there is no proof when such an abnormal stress LV performance is found that prophylactic valve replacement is beneficial in preventing a less-than-optimal post–valve replacement result. In dealing with a patient with valve disease, no matter how severe, it is always easier to recommend surgery to the symptomatic patient who recognizes that the risks taken are balanced by the benefits of symptom relief. Remember, it is never possible to make a really asymptomatic patient feel better through surgery. To advise surgery in an asymptomatic patient, there must be unequivocal evidence that the future benefits make the present real risks of surgery worthwhile.
The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.
- Copyright © 1998 by American Heart Association
Borer JS, Hochreiter C, Herrold EMcM, Supino P, Aschermann M, Wencker D, Devereux RB, Roman MJ, Szule M, Kligfield P, Isom OW. Prediction of indications for valve replacement among asymptomatic or minimally symptomatic patients with chronic aortic regurgitation and normal left ventricular performance. Circulation. 1998;97:525–534.
Suga H, Sagawa K. Instantaneous pressure-volume relationships and their ratio in the excised supported canine left ventricle. Circ Res. 1974;35:117.
Henry WL, Bonow RO, Borer JS, Ware JH, Kent KM, Redwood DR, McIntosh CL, Morrow AG, Epstein SE. Observations on the optimum time for operative intervention for aortic regurgitation, I: evaluation of the results of aortic valve replacement in symptomatic patients. Circulation. 1980;61:471–483.
Henry WL, Bonow RO, Rosing DR, Epstein SE. Observations on the optimum time for operative intervention for aortic regurgitation, II: serial echocardiographic evaluation of asymptomatic patients. Circulation. 1980;61:484–492.
Daniel WG, Hood WP Jr, Siart A, Hausmann D, Nellessen U, Oelert H, Lichtlen PR. Chronic aortic regurgitation: reassessment of the prognostic value of preoperative left ventricular end-systolic dimension and fractional shortening. Circulation. 1985;71:669–680.
Bonow RO, Borer JS, Rosing DR, Henry WL, Pearlman AS, McIntosh CL, Morrow AG, Epstein SE. Preoperative exercise capacity in symptomatic patients with aortic regurgitation as a predictor of postoperative left ventricular function and long-term prognosis. Circulation. 1980;62:1280–1290.
Carabello BA, Williams H, Gash AK, Kent R, Belber D, Maurer A, Siegel J, Blasius K, Spann JF. Hemodynamic predictors of outcome in patients undergoing valve replacement. Circulation. 1986;74:1309–1316.
Bonow RO, Lakatos E, Maron BJ, Epstein SE. Serial long-term assessment of the natural history of asymptomatic patients with chronic aortic regurgitation and normal left ventricular systolic function. Circulation. 1991;84:1625–1635.
Siemienczuk D, Greenberg B, Morris C, Massie B, Wilson RA, Topic N, Bristow JD, Cheitlin M. Chronic aortic insufficiency: factors associated with progression to aortic valve replacement. Ann Intern Med. 1989;110:587–592.
Greenberg B, Massie B, Thomas D, Bristow JD, Cheitlin M, Broudy D, Szlachcic J, Krishnamurthy G. Association between the exercise ejection fraction response and systolic wall stress in patients with chronic aortic insufficiency. Circulation. 1985;71:458–465.