Natural History of Moderate Aortic Stenosis

Transcription

1 JACC Vol. 17. No CLINICAL STUDIES Natural History of Moderate Aortic Stenosis KELLEY D. KENNEDY, MD, RICK A. NISHIMURA, MD, FACC, DAVID R. HOLMES, JR., MD, FACC, KENT R. BAILEY, PHD Rochester, Minnesota The natural history of severe, symptomatic aortic stenosis has been well documented. However, the natural history of moderate aortic stenosis remains poorly understood. Therefore, a group of 66 patients was identified who had a diagnosis of moderate aortic stenosis at the time of cardiac catheterization (aortic valve area.7 to 1.2 cm 2 ) and who did not have surgical therapy during the 1st 18 days after cardiac catheterization. During a mean follow.up period of 3S months, 14 patients died of causes attributed to aortic stenosis and 21 underwent aortic valve replacement. The estimated probability for remaining free of any complication of aortic stenosis at the end of the first 4 years was 59%. Symptomatic patients with decreased ejection fraction or hemodynamic evidence of left ventricular decompensation were at greater risk for these complications. It is concluded that patients with moderate aortic stenosis are at significant risk for the development of complications. (J Am Coli CardioI1991;17:313-9) The natural history of severe symptomatic aortic stenosis has been well documented -5). The natural history of mild or moderate aortic stenosis remains poorly understood. Clinical studies (6-12) performed before the availability of cardiac catheterization attempted to determine the natural history of aortic stenosis but lacked hemodynamic information. Subsequent studies (4,5,13-2) performed after the availability of cardiac catheterization and aortic valve replacement have been difficult to interpret because the natural history of many patients with aortic stenosis has been interrupted by valve replacement. The treatment of patients with aortic stenosis, including the timing of valve replacement, requires accurate knowledge of the natural history of the condition so that patients are not exposed prematurely to the recognized complications of aortic valve replacement. This study was performed to characterize the natural history of moderate aortic stenosis by examining the outcome of a large group of patients who had been found to have moderate aortic stenosis at cardiac catheterization. Methods Patient selection. Patients for this study were selected from those undergoing cardiac catheterization at the Mayo From the Division of Cardiovascular Diseases and Internal Medicine and the Section of Biostatistics, Mayo Clinic and Mayo Foundation. Rochester, Minnesota. This study was presented in part at the 61st Annual Scientific Sessions of the American Heart Association, Washington, D.C., November Manuscript received February 1, 199: revised manuscript received August 8, 199, accepted August 28, 199. Address for reprints: Rick A. Nishimura, MD, Mayo Clinic, 2 First Street SW, Rochester, Minnesota Clinic between 198 and 1985 on the basis of having moderate aortic stenosis, defined as an aortic valve area of.7 to 1.2 cm 2 as determined at cardiac catheterization. Patients were excluded if any of the following were present: congenital heart disease other than congenital bicuspid aortic valve; primary disease of the mitral, tricuspid or pulmonary valve; aortic regurgitation of more than a moderate degree; bacterial endocarditis or previous valve surgery. Because this was a natural history study, patients were excluded from analysis if they underwent aortic valve replacement during the 1st 18 days after cardiac catheterization. One hundred thirty patients met the initial criteria; of these, 64 underwent aortic valve replacement for clinical reasons in the 1st 18 days after cardiac catheterization. The remaining 66 patients formed the final study group. Retrospective record review was carried out for each of these patients. Historical information included age when the murmur was first detected, history of prior myocardial infarction, presence or absence of other diseases including hypertension, diabetes mellitus. cerebrovascular disease and cancer, impression of the severity of the aortic stenosis before cardiac catheterization and onset and severity (by New York Heart Association functional class) of dyspnea, chest pain and syncope if present. Left ventricular hypertrophy on the electrocardiogram (ECG) (21) and cardiomegaly (cardiothoracic ratio.5) on the chest radiograph were recorded. Patient groups were subdivided into those with significant symptoms (class III or IV) at baseline study and those with minimal symptoms. Cardiac catheterization. All 66 patients had undergone cardiac catheterization by standard femoral percutaneous or brachial cut-down techniques. Before the introduction of contrast material, left ventricular end-diastolic pressure and 1991 by the American College of Cardiology /91/$3.5

2 314 KENNEDY ET AL. lacc Vol. 17. No.2 transvalvular gradient were obtained by the left ventricularaortic root pullback method or simultaneous monitoring of left ventricular and aortic root pressures. All patients had cardiac output determinations by the dye dilution method alone or in combination with the Fick method. Transvalvular gradient and cardiac output were used to calculate the aortic valve area by the formula of Hakki et al. (22) for all 66 patients. The aortic valve area index was calculated from body surface area recorded at the time of cardiac catheterization. Biplane left ventriculography was performed in 64 patients, allowing quantitation of ejection fraction, stroke volume index and left ventricular end-systolic and enddiastolic volume indexes (23). Cineventriculograms were used to determine the presence and degree of mitral regurgitation (24). Coronary arteriography was performed in all 66 patients. Significant coronary artery disease was defined as ;::::7% luminal diameter narrowing by visual estimation. Patients were considered to have mild, moderate or severe coronary artery disease if significant intraluminal obstruction was present in one, two or three major epicardial vessels. Patient outcome. Follow-up information for all 66 patients was obtained by telephone interview or review of medical records. This information included vital status and the presence and degree of symptoms in living patients. In those patients who underwent subsequent aortic valve replacement, the timing and indication or indications for operation were obtained. For those patients known to have died, the cause of death was obtained from autopsy records. death certificates, physician records or conversations with the patient's physician. Patients were considered to have died as a result of aortic stenosis (referred to as aortic stenosis-related death) if they died from congestive heart failure or sudden death in the absence of documented myocardial ischemia or infarction. A subset of patients referred to as having aortic stenosis complications included all patients who either underwent aortic valve replacement 18 days after cardiac catheterization or had an aortic stenosis-related death. Statistical analysis. Descriptive baseline characteristics were summarized by frequencies and percentages or by mean values and standard deviation. Intergroup comparisons of baseline characteristics were based on two-sample t tests or on the chi-square for two by two tables. Year by year cumulative event-free probabilities were estimated by the Kaplan-Meier survival method and compared between groups by using Greenwood's formula. The analysis of predictors of development of aortic stenosis complications was based on univariate and multivariate proportional hazards models as well as stratified Kaplan Meier curves and log-rank tests for two-group comparisons. Specific p values are reported; p <.5 was used informally as the criterion of significance. Results Clinical characteristics (Table 1). Of the 66 patients, 51 (77%) were men. The mean (± SD) age at entry into the study was 67 ± 1 years and the mean age at recognition of the murmur of aortic stenosis was 52 ± 21 years. Only a minority (18%) of the patients were asymptomatic; 62% had dyspnea and 42% were in New York Heart Association functional class III or IV. Syncope was present in only nine patients (14%). Thirty-three percent had left ventricular hypertrophy on the ECG and 23% had cardiomegaly on their chest radiograph. Most patients (8%) were thought to have moderate aortic stenosis on the basis of clinical examination before cardiac catheterization. The clinician overestimated more frequently than undere5timated (14% versus 6%) the severity of aortic stenosis a~ defined by our valve area criterion. Cardiac catheterization (Table 2). The mean calculated aortic valve area was.92 ±.13 cm 2, with a mean peak to peak transvalvular gradient of 33 ± 11 mm Hg. At the time of cardiac catheterization, the mean heart rate was 79 beats/min (range 58 to 14). In two patients the heart rate was <6 beats/min and in two it was 1 beats/min. Mean left ventricular end-diastolic pressure and end-systolic and end-diastolic volume indexes were moderately increased, and the mean cardiac output, ejection fraction and stroke volume were normal. Mild mitral regurgitation was found more frequently than moderate mitral regurgitation; mild and moderate aortic insufficiency were similar in frequency. Coronary artery disease was present in 38 patients; of 18 patients with two or more disl~ased vessels, 9 (5%) had a decreased ejection fraction «5%). Among the 41 patients with one or no vessel diseas{:, 12 (29%) had a decreased ejection fraction. Patient outcome. Follow-up data for all 66 patients were obtained at a mean of 35 months (maximum 7.2 years). At the end of each of the first 4 years of follow-up study, 58, 44, 36 and 2 patients, respectivel). were available for analysis. During the follow-up period, 14 patients (21%) died from causes attributed to aortic stenosis. During the same period, 21 patients (32%) underwent aortic valve replacement, 13 (62%) because of symptomatic progression and 8 (38%) because of symptomatic progreision and documented hemodynamic progression of disease. Aortic stenosis complicatiojlls (Tables 3 and 4). These complications consisted of dehth from aortic stenosis and aortic valve replacement. The estimated probability of remaining free of aortic stenosis complications was 9%, 76%, 66% and 59% at the end of each of the 1st 4 years of follow-up, respectively (Fig. l) Univariate and multivariate predictors of the development of complications related to aortic stenosis are listed in Table 3. Statistically significan:: univariate predictors were ejection fraction. aortic valve area index, left ventricular end-diastolic pressure and left ventricular end-systolic and end-diastolic volume indexes. By forward stepwise propor-

3 JACC Vol. 17, NO.2 KENNEDY ET AL. 315 February 1991 :313-9 Table 1. Clinical Characteristics of 66 Patients at Entry Into Study By Symptoms Total Group Symptomatic Minimal Symptoms (n = 66) (n = 38) (n = 28) No. % No. % No. % p Value* Mean age (yr) 67 ± 1 69 ± 7 66 ± 12 NS Gender Male NS Female Age when murmur first recorded (yr) 52 ± ± ± 19 NS Asymptomatic Symptomatic Symptoms Dyspnea None Class I or II II 39 Class III or IV Chest pain None Class I or II Class III or IV Syncope Absent Present Left ventricular hypertrophyt Absent NS Present Cardiomegaly:j: Absent Present II Clinical impression of severity None I 2 I 4 NS Mild II Moderate Severe I 4 *Comparison of symptomatic patients and patients with minimal symptoms; tby electrocardiography; :j:by chest radiography. Class = New York Heart Association functional class. tional hazards regression analysis, ejection fraction, left ventricular end-diastolic pressure and aortic valve area index were independently predictive of aortic stenosis complications. For a better description of the association between aortic stenosis complication rates and baseline left ventricular enddiastolic pressure, ejection fraction, left ventricular endsystolic and end-diastolic volume indexes and aortic valve area index, we dichotomized each variable and compared the hazard rate of aortic stenosis complications in the groups falling above and below each threshold. For example, ejection fraction was dichotomized at 5%, left ventricular end-diastolic pressure at 18 mm Hg, left ventricular end-systolic volume index at 5 mum 2, left ventricular end-diastolic volume index at 1 mum 2 and aortic valve area index at.5 cm 2 /m 2 The relative hazard and 95% confidence intervals are shown in Table 4. Each of these variables was slightly associated with increased risk of aortic stenosis complications. The relative risk was on the order of 2 to 3 in each case. Symptomatic versus minimally symptomatic patients. The patients were classified into two groups on the basis of the New York Heart Association functional classification: 1) 38 patients who were symptomatic with a history of syncope or class III or IV dyspnea or chest pain, and 2) 28 patients who were minimally symptomatic without a history of syncope and with class, I or II dyspnea or chest pain. The clinical characteristics (Table 1) and findings at cardiac catheterization (Table 2) were compared for the patients in the symptomatic and minimally symptomatic groups. There was a statistically significant difference (p =.3) in the severity of coronary artery disease between the patients in the symptomatic and minimally symptomatic groups. With this exception, there were no statistically significant differences in clinical characteristics or findings at cardiac catheterization between these two groups. The actuarial rates of survival free of aortic stenosis complications for these two subgroups are presented in Figure 2. The cumulative event rates were higher in the symptomatic group throughout the 4 years, achieving statistical significance (p =.22) at year 2. The actuarial rates of survival free of aortic stenosis

4 316 KENNEDY ET AL. lacc Vol. 17, No.2 Table 2. Results of Cardiac Catheterization in 66 Patients at Entry Into Study Left ventricular end-diastolic pressure (mmhg) Peak to peak transvalvular gradient (mmhg) Ejection fraction (%) Ventricular end-volume index (milm 2 ) Systolic Diastolic Stroke volume index (milm 2 ) Mitral regurgitation Not available None Mild Moderate Severe Cardiac output (liters/min) Aortic valve area (Hakki formula) (cm 2 ) Aortic valve area index (cm 2 /m 2 ) Coronary artery disease None Mild Moderate Severe Aortic insufficiency No murmur None Mild Moderate Severe Total Group (n = 66) 22 ± 9 (range 8-48) 33 ± 11 (range 7-58) 55 ± 17 (range 26-84) 48 ± 31 (range ) 12 ± 34 (range 45-28) 5 ± 16 (range 17-84) 2 (3%) 42 (64%) 2 (3%) 2 (3%) 5.2 ± 1.1 (range ).92 ±.13 (range.7-1.2) 28 (42%) 19 (29%) 1 (15%) 9 (14%) 53 (8%) 2 (3%) 6(9%) 5 (8%) *Comparison of symptomatic patients and patients with minimal symptoms. By Symptoms Symptomatic Minimal Symptoms (n = 38) (n = 28) p Value* 23 ± 9 (range 8-48) 22 ± 8 (range 1-47) NS 33 ± 11 (range 7-58) 34 ± 12 (range 16-58) NS 54 ± 16 (range 26-84) 58 ± 17 (range 26-8) NS 52 ± 35 (range ) 44 ± 25 (range 18-17) NS 13 ± 4 (range 45-28) 11 ± 28 (range 5-164) NS 47 ± 13 (range 28-81) 53± 19 (range 17-84) NS 1 (3%) 1 (4%) 23 (61%) 19 (68%) NS 12 (32%) 8 (29%) NS 2 (5%) (%) NS 5.2 ± 1.1 (range ) 5.3 :t 1.1 (range ) NS.92 ±.13 (range.7-1.2).94 :t.14 (range ) NS.48 ±.8.5 :t.7 NS 14 (37%) 14 (5%).3 9 (24%) 1 (36%).3 8 (21%) 2 (7%).3 7 (14%) 2 (7%).3 31 (82%) 21 (78%) NS 1 (3%) 1 (4%) 4 (11%) 2 (7%) NS 2 (5%) 3 (11%) NS complications are presented in Figure 3 for patients with an ejection fraction ;:::5% and <5%. A significant difference (p ==.4) in aortic stenosis complication rates, higher in patients with ejection fraction <5%, was observed at 4 years and a marginally significant difference (p ==.63) was observed at 1 year. The actuarial survival rates free of aortic stenosis complications for patients with symptoms or ejection fraction <5% and for the remaining patients are shown in Figure 4. A significant difference (p ==.3) was observed at 2 years and less significant differences were seen at 3 and 4 years (p ==.65 and.66, respectively). Aortic stenosis death. Table 5 gives the I, 2, 3 and 4 year probabilities of avoiding death for symptomatic and minimally symptomatic patients. The cumulative death rates were significantly different (p <.5) at 1, 2 and 3 years. However, because of the experience beyond 4 years, the overall log-rank test attained only marginal significance (p ==.16). In a similar comparison for patients with either symptoms or low ejection fraction «5%) (Table 6), the aortic stenosis death rate was significantly different at 2, 3 and 4 years. The overall log-rank test was less significant (p ==.5) because of events occurring after 4 years. Table 3. Predictors of Complications Related to Aortic Stenosis in Total Study Group (n = 66) Predictor Univariate Gender Syncope Left ventricular hypertrophyt Cardiomegaly~ Left ventricular end-diastolic pressu re Ejection fraction Left ventricular volume index End-systolic End-diastolic Stroke volume index Cardiac output Transvalvu1ar gradient Aortic valve area (Hakki formula) Aortic valve area index Severe coronary artery disease Multivariate Ejection fraction Left ventricular end-diastolic pressure Aortic valve area index p Value* *Comparison of symptomatic patients and patients with minimal sympturns; tby electrocardiography; ~by chest radiography.

5 JACC Vol. 17. No.2 KENNEDY ET AL. 317 Table 4. Analysis of Predictor Variables Related to Aortic Stenosis in the Total Study Group (n == 66) Predictor Dichotomy Relative Risk 95% CI Ejection fraction (%) <5/2: Left ventricular end-diastolic 18/:; pressure (mm Hg) Left ventricular end-systolic 5/:; volume index (mlfm 2 ) Left ventricular end-diastolic 1/:; volume index (mllm 2 ) Aortic valve area index <.5/2: (cm 2 /m 2 ) CI '" confidence interval. Discussion Natural history of aortic stenosis. In clinical studies (6 12) of the natural history of aortic stenosis before the availability of cardiac catheterization, patients with symptomatic aortic stenosis were found to have a high mortality rate, but these studies lacked objective hemodynamic information. Studies (4,5,13-2,25) performed after the availability of cardiac catheterization and aortic valve replacement attempted to define the natural history of aortic stenosis, but have been criticized for containing only small groups of highly selected patients. In those reports, the rate of progression of aortic stenosis was variable and unpredictable, probably as a result of patient selection bias or lack of complete data from cardiac catheterization. Furthermore, those studies are even less applicable to current patients because of a marked change in the cause of aortic stenosis in recent years-from rheumatic heart disease to bicuspid and senile calcific degeneration (26). Since the availability of aortic valve replacement, a true prospective study of the natural history of aortic stenosis has not been possible because this natural history is interrupted by aortic valve replacement in many patients. Because of the limited number of patients available for analysis, some studies (4,13-2) retrospectively identified the natural history of patients who refused or were denied aortic valve 1 9 '#. 8 (ij.~ 7 6 :J CIl Ql 5 ~ 4 ~ Ql 3 W Symptomatic 3 4 Figure 2. Survival free of aortic stenosis complications by symptom group in 66 patients. replacement. The natural history of this small subset of patients with aortic stenosis probably would not represent the natural history of all patients with aortic stenosis. Definition of severity of aortic stenosis. A problem that has arisen in clinical practice is the definition of "severe" aortic stenosis. In the natural history studies (6-12) performed before the routine use of cardiac catheterization, the diagnosis of the severity of aortic stenosis was made on the basis of clinical examination. These natural history studies may have included patients with catheter-defined less severe stenosis in the group defined as having severe aortic stenosis. As cardiac catheterization evolved and provided an objective measurement of the severity of aortic stenosis by means of the "aortic valve area," the criterion for severe aortic stenosis was set as an aortic valve area <.7 cm 2 (2,25), and it is now common to offer aortic valve replacement to symptomatic patients who meet this criterion (27). However, it remains unclear how to treat patients with a valve area larger than this arbitrary value. The data presented here support the concept that the valve area should not be used as a single final criterion in determining who should be a candidate for aortic valve replacement. The overall mortality rate in patients with Figure 1. Survival free of aortic stenosis complications, aortic valve replacement and aortic stenosis death in 66 patients. 1 9 <f- 8 tti.2: 7 ~ 6 :J CIl 5 Ql ~ 4 'E 3 Ql W 2 1 Figure 3. Survival free of aortic stenosis complications by left ventricular ejection fraction (LVEF). ~ (ij.~ :J ,._ '"Q) 5 ~ 4 'E 3 Q) W

6 318 KENNEDY ET AL. lacc Vol. 17, No.2 *' ~ 6 ::J II) CD 5 ~ 4 3 CD ~ 2 1 Symptoms (n = 38) 2 3 Figure 4. Survival free of aortic stenosis complications by presence of symptoms or left ventricular ejection fraction (LVEF) <5%. moderate aortic stenosis who already have symptoms approaches that in symptomatic patients with severe aortic stenosis. Therefore, aortic valve replacement should be offered to these patients just as it would be to patients with an aortic valve area <.7 cm 2 Use of aortic valve area index. Analysis using the aortic valve area alone did not predict development of aortic stenosis complications in this study. There are important ramifications of body size with regard to calculated aortic valve area in an individual patient with aortic stenosis, just as cardiac index provides more information than does cardiac output in many cases. In the patients described here, the aortic valve area index, but not the aortic valve area, predicted the development of aortic stenosis complications. Relation of symptoms to outcome. From our results, symptomatic patients had a greater incidence of events related to progressive aortic stenosis as reflected by a lower actuarial survival free of aortic stenosis complications and aortic stenosis death during the first 4 years of follow-up study. Others (28) have demonstrated, in studies including those using Doppler echocardiography, the significant risk of cardiac complications related to symptomatic aortic stenosis. However, it is important to note that 31%of the patients in our minimally symptomatic group had aortic stenosis complications in the first 4 years of follow-up study. In addition, at the end of each of the first 3 years of follow-up study, there was a statistically significant (p <.5) differ- Table S. Probability of Avoiding Death From Aortic Stenosis in Symptomatic and Minimally Symptomatic Groups Minimal symptoms (n = 28) Year No. Probability SE No. Probability SE p Value' I % 'Comparison of symptomatic patients and patients with minimal symptoms. Table 6. Probability of AvoidinE; Death From Aortic Stenosis in Patients With Symptoms or Low «5%) Ejection Fraction (EF) and in Remaining Patients Probability Patients With EF <5% or Remaining Year Symptoms Patients p Value' I.886 ± ± ± ± om 'Comparison of patients in the ty,o groups shown. ence between the rates of aortic stenosis death in the symptomatic and minimally symptomatic groups. Therefore, our data suggest that the presence of symptoms increases the likelihood of aortic stenosis complications, but the absence of moderate or severe symptoms does not necessarily ensure a favorable prognosis for patients with moderate aortic stenosis. Left ventricular dysfunction. Apparent pathophysiologic consequences of aortic stenosis occurred in our patients, including moderate increases in mean left ventricular enddiastolic pressure and left ventricular end-systolic and endsystolic volume indexes. Despite the presence of these compensatory responses to left ventricular outflow obstruction, the mean ejection fraction and cardiac output were normal in both groups. Altholgh aortic stenosis complications developed in 31% of patients with an ejection fraction 2':5%, the rate of such complications was higher in those with an ejection fraction <5%. Patients with an abnormally low ejection fraction at initial cardiac catheterization had a 64% chance of aortic stenosis complications developing during the 1st 4 years of follow-up study. Although most of our patients had normal or near normal left ventricular function at entry into the study, those with an ejection fraction ::;5% had a worse prognosis. Limitations. Patients were eligible for entry into the current study if moderate aortic stenosis was demonstrated at cardiac catheterization between 198 and As with other retrospective studies, there was the potential for selection bias, and patients would not have been included in this study if moderate aortic ~;tenosis had not been documented at cardiac catheterization. It is reassuring that 8% of our patients were suspected to have moderate aortic stenosis before cardiac catheterization and, when this clinical diagnosis was incorrect, the clinician more often overestimated the severity of aortic stenosis. The formula of Hakki et al. (22) may be less accurate than the Gorlin formula for the calculation of aortic valve area at the extremes of heart rate (29,3). At the time of carciac catheterization, only 6% of the patients had a heart rate <6 or 1 beats/min. This finding suggests that misclassification of patients because of inaccuracies in aortic valve area calculated by using heart rate occurred infrequently.

7 JACC Vol. 17, No.2 KENNEDY ET AL. 319 We could not eliminate the possibility that coronary artery disease rather than aortic stenosis may have played a role in the left ventricular dysfunction or caused the symptoms, Patients with significant coronary artery disease had more severe left ventricular dysfunction as well as a higher incidence of symptoms. The presence of coronary artery disease was significantly associated with the overall mortality rate during follow-up (p =.4). However, by itself, coronary artery disease was not even remotely associated with the primary end point of aortic stenosis-related death or valve replacement in these patients. Conclusions. Patients with moderate aortic stenosis are at significant risk of complications in the short-term follow-up period. Those with a subnormal ejection fraction and symptoms are at an even higher risk, but the absence of these findings does not necessarily ensure a favorable prognosis. Therefore, valve area should not be used as the sole criterion in determining who is a candidate for valve operation. References I. Ross J Jr, Braunwald E. Aortic stenosis. Circulation 1%8;38(suppl V):V Turina J, Hess, Sepulcri F, Krayenbuehl HP. Spontaneous course of aortic valve disease. Eur Heart J 1987;8: Wood P. Aortic stenosis. Am J CardioI1958;1: Frank S, Johnson A, Ross J Jr. Natural history of valvular aortic stenosis. Br Heart J 1973;35: Rapaport E. Natural history of aortic and mitral valve disease. Am J CardioI1975;35: Olesen KH, Warburg E. Isolated aortic stenosis: the late prognosis. Acta Med Scand 1958;16: Anderson MW. The clinical course of patients with calcific aortic stenosis. Proc Staff Meet Mayo Clin 1961 ;36: Takeda J, Warren R, Holzman D. Prognosis of aortic stenosis. Arch Surg 1%3;87: Baker C, Somerville J. Clinical features and surgical treatment of fifty patients with severe aortic stenosis. Guy Hosp Rep 1959;18: Kiloh GA. Pure aortic stenosis. Br Heart J 195;12: Mitchell AM, Sackett CH, Hunzicker WJ, Levine SA. The clinical features of aortic stenosis. Am Heart J 1954;48: Bergeron J, Abelmann WH, Vazquez-Milan H, Ellis LB. Aortic stenosis: clinical manifestations and course of the disease. Arch Intern Med 1954:94: Ng ASH, Holmes DR Jr, Smith HC, et al. Hemodynamic progression of adult valvular aortic stenosis. Cathet Cardiovasc Diagn 1986;12: Larsen VH, Jensen BS. Spontaneous progression of valvular aortic stenosis. Dan Med Bull 1985;(1)32: Nestico PF, DePace NL, Kimbiris D, et al. Progression of isolated aortic stenosis: analysis of 29 patients having more than 1 cardiac catheterization. Am J Cardiol 1983;52: Gradman AH. Hancock EW. Progressive aortic stenosis (abstr). Circulation 1976;54(suppl 11): Cheitlin MD, Gertz EW, Brundage BH, Carlson CJ, Quash JA, Bode RS Jr. Rate of progression of severity of valvular aortic stenosis in the adult. Am Heart J 1979;98: Wagner S. Selzer A. Patterns of progression of aortic stenosis: a longitudinal hemodynamic study. Circulation 1982;65: Bogart DB, Murphy BL, Wong BYS, Pugh DM, Dunn MI. Progression of aortic stenosis. Chest 1979;76: Chizner MA, Pearle OL, deleon AC Jr. The natural history of aortic stenosis in adults. Am Heart J 198;99: Romhilt OW, Estes EH Jr. Apoint-score system for the ECG diagnosis of left ventricular hypertrophy. Am Heart J 1968;75: Hakki AH, Iskandrian AS, Bemis CE, et al. Asimplified valve formula for the calculation of stenotic cardiac valve areas. Circulation 1981 ;63: Bove AA, Kreulen TH, Spann JF. Computer analysis of left ventricular dynamic geometry in man. Am J Cardiol 1978;41: Grossman WA. Profiles in valvular heart disease. In: Grossman WA, ed. Cardiac Catheterization and Angiography. 3rd ed. Philadelphia: Lea & Febiger, 1986: Selzer A. Changing aspects of the natural history of valvular aortic stenosis. N Engl J Med 1987;317: Passik CS, Ackermann OM, Pluth JR, Edwards WO. Temporal changes in the causes of aortic stenosis: a surgical pathologic study of 646 cases. Mayo Clin Proc 1987;62: Schwarz F, Baumann P, Manthey J, et al. The effect of aortic valve replacement on survival. Circulation 1982;66: Kelly TA, Rothbart RM, Cooper CM, Kaiser DL, Smucker ML, Gibson RS. Comparison of outcome of asymptomatic to symptomatic patients older than 2 years of age with valvular aortic stenosis. Am J Cardiol 1988;61: Angel J, Soler-Soler J, Anivarro 1, Oomingo E. Hemodynamic evaluation of stenotic cardiac valves. II. Modification of the simplified valve formula for mitral and aortic valve area calculation. Cathet Cardiovasc Diagn 1985;11: Hillis LO, Winniford MO. The simplified formula for calculation of aortic valve area: potential inaccuracies in patients with bradycardia or tachycardia. Cathet Cardiovasc Diagn 1987;13:31-3.