CLINICAL STUDY: DISEASE OF THE AORTIC VALVE OR AORTA
Clinical efficacy of Doppler-echocardiographic indices of aortic valve stenosis:a comparative test-based analysis of outcome
Javier Bermejo, MD*,*,
Rodolfo Odreman, MD*,
Josefina Feijoo, MD*,
M. Mar Moreno, MD*,
Paz Gómez-Moreno, RDCS* and
Miguel A. García-Fernández, MD*
* Department of Cardiology, Hospital General Universitario Gregorio Marañón, Madrid, Spain
Manuscript received March 3, 2000;
revised manuscript received April 29, 2002,
accepted August 30, 2002.
* Reprint requests and correspondence: Dr. Javier Bermejo, Laboratory of Echocardiography, Department of Cardiology, Hospital General Universitario Gregorio Marañón, Dr. Esquerdo 46. 28007 Madrid, Spain.
javbermejo{at}jet.es
 |
Abstract
|
|---|
OBJECTIVES: This study was designed to assess which hemodynamic index best accounts for clinical severity of aortic stenosis (AS) and to analyze the value of low-dose dobutamine testing.
BACKGROUND: Pressure gradient and valve area are suboptimal because they depend on flow rate, correlate poorly with symptoms, and provide limited prognostic information. Recently, new indices and low-dose inotropic stimulation have been introduced, but their clinical value remains uncertain.
METHODS: A total of 307 consecutive patients with AS were included in an ambispective study design (71 ± 12 years old; peak jet velocity: 3.7 ± 1.1 m/s). Clinical and Doppler-echocardiographic data were obtained, as well as results of low-dose dobutamine infusion (47 patients). Using receiver-operator-characteristic curve analysis, we evaluated jet velocity, pressure gradient, valve area, resistance, stroke-work loss (SWL), and dobutamine-induced increase in area for predicting 1) symptomatic status at entry, 2) early ( 3 months) cardiovascular death or aortic valve replacement, and 3) long-term outcome. Logistic regression and Cox models were designed multivariate and adjusted by bootstrapping.
RESULTS: Only 28% of patients were alive without valve replacement at the end of the follow-up period (22 ± 4 months). The decision for valve replacement was made by the referring physician, blinded to the SWL, valve resistance, and dobutamine results. Non–flow-corrected indices performed better than valve area and valve resistance. Among them, SWL best predicted the defined end points. Odds/hazard ratios associated with a SWL  = 17% were 5.14 for presenting AS symptoms, 4.68 for early events, and 2.31 for late outcome. A cutoff value of SWL >25% best discriminated clinical end points. Other independent predictors of prognosis were symptomatic status and left ventricular ejection fraction. Dobutamine testing added no value to baseline models.
CONCLUSIONS: Non–flow-corrected indices show the highest clinical efficacy in aortic stenosis. Among these, SWL best predicts symptomatic status and outcome and therefore should be incorporated to aid patient management in unclear situations.
|
Abbreviations and Acronyms
| | AS | | aortic valve stenosis | | AVA | | aortic valve area | | AVR | | aortic valve resistance | | CI | | confidence interval | | EF | | ejection fraction | | LV | | left ventricular | | ROC | | receiver-operator characteristic | | SBP | | systolic blood pressure | | SWL | | stroke-work loss | | Vmax | | peak transaortic jet velocity | | P | | mean transvalvular systolic pressure gradient |
|
Therapeutic decisions in adult aortic stenosis (AS) are based on the definition of symptomatic status and of hemodynamic severity (1). Accurate evaluation of these two issues is mandatory, because prognosis of asymptomatic patients is not improved by surgery (2) and because poor outcome follows replacement of nonseverely stenotic valves (3,4). However, establishing symptomatic status and severity of AS is not always straightforward. Assessment of subjective symptoms and functional capacity is frequently ambiguous, particularly in the elderly (5). Also, establishing disease significance can be challenging because a gold standard of valve obstruction is unavailable (6). Current hemodynamic assessment of severity relies on indices such as transvalvular pressure gradient (P) and aortic valve area (AVA) that are obtained by cardiac catheterization or, more frequently today, by Doppler echocardiography (1). It is recognized that these indices are suboptimal because they correlate poorly with patients’ symptoms (7), provide little prognostic information (8), and depend on flow rate (9–11). Furthermore, the appropriate cutoff values of AVA and P for establishing severity are unclear (12). On the basis of different fluid-dynamic assumptions, aortic valve resistance (AVR) and left ventricular (LV) stroke-work loss (SWL) have been proposed as alternative indices of AS (13–15), but their clinical value remains to be ascertained. Also, it has been suggested that stress echocardiography may play a prognostic role in AS (16); by unmasking the reserve of valve orifice enlargement available when cardiac output increases, stress echocardiography may predict patients’ symptoms and anticipate valve replacement (16).
Evaluation of diagnostic tests should be performed with a hierarchical framework in which clinical outcome efficacy represents the final objective of diagnostic imaging (17–19). Clinical outcome efficacy summarizes correlation with symptom severity, functional outcome, patient utility assessment, quality of life, morbidity avoided, and mortality rate (17,19). Previous longitudinal studies have aimed to identify natural-history predictors among classic indices of AS and have focused on selected patient groups such as asymptomatic (2,20–25) or moderate disease (26). Hence, a test-based comparative evaluation including conventional and alternative indices is lacking. The present study was designed to compare the clinical outcome efficacy of severity indices and of low-dose inotropic stress testing, in terms of 1) symptom correlation, 2) identification of critical disease, and 3) prediction of long-term prognosis.
|
Methods
|
|---|
Study population.
Two patient cohorts constitute the basis for this study. Forty-seven patients (15%) were prospectively enrolled on the basis of the following criteria: a diagnosis of valvular AS, a suitable echocardiographic window, New York Heart Association functional class I to III, absence of ventricular arrhythmias, and willingness to participate. Some of these patients have been the basis of a previous report (10), and the clinical efficacy of dobutamine testing was assessed in this cohort. A total of 275 patients were retrospectively recruited among all inpatients referred for a Doppler-echocardiographic examination from 4/97 to 4/98 in whom AS was diagnosed. Patients with any >2+ valvular regurgitation or more than mild mitral stenosis were excluded. Follow-up data were unavailable for 15 patients in the retrospective group; no patient from the prospective cohort was lost. The combination of the prospective and retrospective cohorts constitutes the basis for this report (Group A, n = 307). Group B was derived as the subgroup of Group A in whom all indices of AS were available (n = 154). Group B was used for selecting the models with best predictivity among different indices of AS (model building). The performance of these models was then measured in Group A (model testing and validation).
The study was approved by the institutional review board and written informed consent was obtained from all patients in the prospective cohort; informed consent was obtained from the patients of the retrospective cohort at the time of follow-up interview. Clinical characteristics of the study population are summarized in Table 1.
Clinical data
In the prospective cohort, a full clinical interview was performed at entry. Patients were carefully screened for the presence of dyspnea, angina, or syncope using New York Heart Association functional classification. Patients manifesting angina, syncope, congestive heart failure, or exercise dyspnea class  2 were classified as presenting symptoms attributable to AS. Comorbidity was assessed by patient anamnesis followed by careful review of medical records, and was coded using a well-validated score (27). Identical criteria were used for collecting clinical data in the retrospective cohort. Blood pressure was measured at the time of the echocardiographic examination in the prospective cohort and taken from the patient’s clinical record within a <5-h interval from the index exam in the retrospective group. An ad hoc study performed in our laboratory (40 consecutive inpatients) demonstrated limits of agreement between values of systolic blood pressure (SBP) measured in the laboratory and values written in the clinical records in the hospitalization wards to be +4 to +11 mm Hg (95% confidence interval [CI]).
Doppler-echocardiographic examination
Examinations were performed using phase-array ultrasound devices (Acuson Sequoia [Mountain View, California] or Hewlett-Packard [Andover, Massachusetts] Sonos 1500 and 2500) with 2.5/2.0 MHz duplex and 1.9 MHz (Pedoff) transducers. All exams were either performed (prospective cohort) or reviewed by one of the investigators. Images and spectrograms were obtained from parasternal, apical, subcostal, and suprasternal views. Left ventricular volumes and ejection fraction (EF) were measured as recommended (28). Aortic and mitral regurgitation were graded on the basis of color flow imaging. Mitral valve area was calculated using two-dimensional planimetry and the pressure half-time method. In patients in sinus rhythm, LV diastolic function was coded according to transmitral pulsed-wave Doppler spectrograms as 1) restrictive, if the E/A ratio was 2 or E/A ratio was between 1 and 2 and E-wave deceleration time was 140 ms (29); 2) prolonged relaxation, if the E/A ratio was less than the 95% CIs of normal age-matched subjects or the deceleration time was above 95% CI limits (30); and 3) normal, if neither restrictive or prolonged relaxation criteria were met.
In the retrospective cohort, measurements of outflow tract diameter and pulsed-wave spectrogram required to calculate AVA were performed when peak jet velocity was >3 m/s or LV function was abnormal. This represents current clinical practice in our laboratory and follows cost-efficacy recommendations (31). Patients in whom outflow tract diameter and spectrograms were not recorded were included in Group A but not in Group B. Measurements were averaged from 4 to 6 beats of patients in sinus rhythm and from 6 to 10 consecutive beats in patients in atrial fibrillation. Valve area and P were calculated using the continuity and Bernoulli equations, respectively. Valve resistance was calculated as
(10). Left ventricular SWL was expressed as percentage and obtained as
(9,15). Reproducibility values of indices of AS in our laboratory have been reported (10).
In the prospective group, a low-dose dobutamine infusion protocol was begun after the baseline study at 5 µg/kg body weight/min up to 20 µg/kg/min, titrated upwards at steps of 5 µg/kg/min every 5 min (10). Doppler spectrograms of LV outflow tract and AS jet velocity were obtained within the last 2 min of each dose. Blood pressure was monitored using automatic sphygnomanometry. All concomitant vasodilator and beta-blocker therapy was suspended 24 h before the index examination. Dobutamine-induced aortic orifice enlargement was calculated as the absolute difference between values of AVA observed at peak flow rate and at baseline.
Patient outcome and definition of end points
Clinical and Doppler echocardiogram patient follow-up was performed on a yearly basis in the prospective cohort. At the end of the four-year follow-up period, medical charts from the whole study group were reviewed and patients alive were directly interviewed. The composite outcome end point was defined as cardiac death or aortic valve replacement during the follow-up period. Death was classified as either cardiac or noncardiac according to patient medical records, information provided by the referring physician, and autopsy data when available. Identical criteria were used in the retrospective cohort. Aortic stenosis was defined in clinical terms as critical when the patient’s aortic orifice was small enough to cause symptoms and require valve replacement or induce cardiac death during the three months following patient enrollment. This three-month criterion was established at the time of study design because it is the maximum time recommended for valve replacement for patients who develop symptoms (25); also, three months is above the time for performing valve replacement once indicated at our institution. The decision to indicate valve surgery was made by the patient’s physician, according to his own interpretation of symptomatic status. Valve area and P were reported to the referring physician, whereas AVR, SWL, and the response to dobutamine were withheld. Patients dying from noncardiac causes were censored at the date of death, and patients still alive without valve replacement were censored at the end of the follow-up period.
Statistical analysis
Variables are described as mean ± SD. Association among hemodynamic indices was assessed by cluster analysis based on Spearman’s correlation coefficient. All predictive models were designed multivariate. Logistic regression models were used to assess symptom correlation and identification of critical AS, whereas long-term outcome was analyzed by Cox proportional-hazards survival analysis. Five AS indices (peak transaortic jet velocity [Vmax], P, AVA, AVR, and SWL) as well as dobutamine-induced increase in AVA were compared. For covariable testing, a composite severity score was computed as the first term of the principal components analysis based on the correlation matrix of all indices. This score therefore synthesizes average prediction capability of the five indices, providing equal weight to each of them. Factors previously reported to correlate with disease symptoms or outcome (age, comorbidity, rhythm, symptomatic status, functional class, LV mass, volumes, diastolic profile, and EF) were tested as covariables and entered in full saturated models for each prediction, including only the composite score as index of severity. Then, relevant covariables were selected by backward stepwise variable selection based on Alkaike’s information criterion (32). Next, each severity index was consecutively substituted in place of the composite severity score. Finally, dobutamine-induced increase in AVA was forced in the model of the index showing best baseline prediction (prospective cohort). Models were validated by resampling 400 bootstrap replications to exclude overfitting. Agreement between each dataset (training sample) and the original data (test sample) was excellent (slope shrinkage factor >0.9; maximum absolute error in predicted probability <0.02) (32). Efficacy was compared computing bias-corrected and adjusted areas under the receiver-operator characteristic (ROC) curves for models containing each of the severity indices (33). Using these models in Group A, cutoff values were obtained using classification-and-regression trees (34), followed by the log-rank test when appropriate. All analyses were performed using S-Plus software (MathSoft, version 2000), expanded by public-domain libraries "survcart" (35), and "design" and "hmisc" (32,36). Statistical significance was assumed at p < 0.05.
|
Results
|
|---|
Correlation among hemodynamic indices.
Figure 1 shows the association among the five indices of severity. Non–flow-corrected indices (Vmax, P, and SWL) correlated closely into a single cluster, whereas flow-corrected indices (AVA and AVR) were grouped in another one. Correlation between these two clusters was fair (rho = 0.37).
View larger version (23K):
[in this window]
[in a new window]
|
Figure 1 Association among hemodynamic indices of aortic stenosis. Results of the cluster analysis of variables in Group B show an initial split into two groups, flow-corrected and non–flow-corrected indices, with a fair correlation between them (rho2 = 0.37). Stroke work loss closely correlates with P, as expected from its calculation formula. Also, there was a close association between AVA and AVR.
|
|
Patient outcome
The follow-up period was 14 ± 16 months (range 10 days to 55 months, median 6 months). For patients without early events, the follow-up period was 22 ± 14 (median 21) months. Outcome of Groups A and B is summarized in Figure 2. There were fewer symptomatic and critical AS patients in Group A than in Group B (71% vs. 81%, p = 0.02; and 39% vs. 55%, p = 0.002, respectively). A very high event rate was observed in the study population, resulting in the fact that only 28% were alive without valve replacement at the end of the follow-up period. Probability for performing valve replacement in symptomatic patients was related (multivariate logistic regression) to younger age (95% CI odds ratio [OR] for 14 years: 0.15 to 0.47; p < 0.0001), higher P (95% CI OR 34 mm Hg: 3.70 to 13.6; p < 0.0001), and less comorbidity (95% CI OR 2 score points: 0.34 to 0.85; p = 0.007).
View larger version (23K):
[in this window]
[in a new window]
|
Figure 2 Follow-up data of the study population. Values are shown for patients in Group A and Group B, separated by a slash. AS = aortic valve stenosis.
|
|
The ROC analysis of clinical efficacy end points is shown in Figure 3, best unadjusted cutoff values for hemodynamic indices to predict outcome are summarized in Table 2, and the results of multivariate prediction models are shown in Table 3.
View larger version (50K):
[in this window]
[in a new window]
|
Figure 3 Clinical efficacy of aortic stenosis indices. Prediction capability is shown for the clinical objectives of predicting: 1) symptomatic status, 2) critical aortic stenosis (AS), 3) late AS events, 4) early or late AS events in patients with depressed left ventricular function, and 5) any-cause mortality in unoperated patients. Clinical efficacy is assessed by receiver-operator characteristic analysis for populations derived from Group B (see methods section). EF = ejection fraction.
|
|
Prediction of symptomatic status
No covariables were retained in the models of symptomatic status prediction. Box plots of severity indices according to symptomatic status demonstrated significant overlap of values of AVA and AVR between symptomatic and asymptomatic patients; among non–flow-corrected indices, the discriminative power of SWL was highest (Figs. 3 and 4). Sensitivity, specificity, and positive and negative predictive values (95% CI) for the SWL criterion of 23% vs. >23% were 0.54 (0.48 to 0.59), 0.81 (0.74 to 0.87), 0.89 (0.83 to 0.93), and 0.42 (0.36 to 0.48), respectively. The dobutamine test did not increase predictive accuracy to the model based on SWL in the prospective cohort (Wald chi-squared = –1.30; p = 0.2; area under the ROC curve = 0.80 vs. 0.80 with and without dobutamine increase in AVA).
View larger version (24K):
[in this window]
[in a new window]
|
Figure 4 Boxplots of indices of aortic stenosis according to the symptomatic status. Distributions are shown for peak jet velocity (A), mean transvalvular pressure gradient (B), stroke-work loss (C), valve area (D), and valve resistance (E), for patients in Group B. Boxes represent the interquartile distance, whereas the white line represents the median. The shaded zone represents the 95% confidence interval for the median and the whiskers represent the limits of each distribution. Significant differences between groups are ascertained by the lack of overlap of the shaded zones. The highest difference between asymptomatic (Asympt.) and symptomatic (Sympt.) patients is observed for stroke-work loss.
|
|
Identification of critical aortic stenosis (early cardiac events)
Other than AS severity, symptomatic status and EF at enrollment were identified as independent covariables related to critical disease. Non–flow-corrected indices of AS performed better than AVA and AVR and, among them, SWL was the index showing highest area under the ROC curve (Fig. 3); also in the validation group, SWL showed the best predictive accuracy. Figure 5A illustrates the regression tree for predicting a critical AS in Group A. Entering dobutamine test information added no prognostic information to the previous model in the prospective cohort (Wald chi-squared = 0.97, p = 0.35; area under the ROC curve: 0.94 vs. 0.95 for models with and without the dobutamine effect, respectively).
View larger version (20K):
[in this window]
[in a new window]
|
Figure 5 Validation of clinical efficacy models in Group A. (A) Regression tree for calculating the probability of suffering critical aortic stenosis (AS) (prob) according to stroke-work loss (SWL), presence or absence of symptoms, and ejection fraction (EF). (B) Cumulative probability of long-term outcome according to SWL and symptomatic status. Patients with critical AS have been excluded. (C) Probability of early and late AS events in patients with impaired left ventricular function (EF < 0.45). p = probability of long-rank test for comparisons between categories.
|
|
Prediction of long-term outcome
Outcome of patients in Group B was slightly worse than of patients in Group A. Figure 6 shows unadjusted long-term survival plots according to each index of severity in group B. For the analysis of long-term outcome efficacy, patients with critical AS were excluded. Best stratification was observed for non–flow-corrected indices, and again, SWL was best (Figs. 3 and 6). Survival without AS events was almost identical for groups with AVA < 0.75 cm2 and patients with AS in the 0.75 to 1 cm2 range. The only independent predictor other than AS severity in the Cox model analysis of long-term outcome was symptomatic status at entry. Outcome of all patients with SWL 25% was very poor, and >95% of symptomatic patients with SWL above this value had suffered cardiac death or had undergone valve replacement by three years (Fig. 5B). In the prospective cohort, dobutamine added no relevant information to SWL and symptomatic status (Wald chi-squared = 2.16, p = 0.15; area under the ROC curve = 0.79 vs. 0.78 for the models with and without dobutamine information, respectively).
View larger version (19K):
[in this window]
[in a new window]
|
Figure 6 Unadjusted cumulative probability of suffering aortic stenosis (AS) events (valve replacement or cardiac death). Plots are shown for peak jet velocity (Vmax) (A), P (B), stroke-work loss (SWL) (C), aortic valve area (AVA) (D), and aortic valve resistance (AVR) (E), for patients in Group B after exclusion of patients with critical AS (n = 70). There is a remarkable overlap of the outcome of the group with AVA < 0.75 cm2 and the group in the 0.75 to 1 cm2 range. Again, best discriminative power was observed for SWL.
|
|
Patients with abnormal systolic function
Twenty-nine patients in Group B (19%) and 48 patients in Group A (16%) had LV systolic dysfunction (EF < 0.45). In Group B, 27 of these patients had AS events: 14 cardiac deaths and 13 aortic valve replacements. Multivariate outcome analysis ruled out prognostic significance of the symptomatic status and EF in this group of patients. Again, SWL was the index showing highest area under the ROC curves. All patients with SWL 25% and EF <0.45 had had an AS event in less than one year (Fig. 5C), irrespective of their symptomatic status.
Overall mortality of patients that did not undergo surgery
There were 60 and 176 patients not operated on in Groups B and A, respectively, 36 (64%) and 91 (52%) of whom died during follow-up, respectively. Clinical efficacy to predict overall mortality was again maximal for SWL (Fig. 3). Variables included in the proportional-hazards model for assessing the risk of cardiac death were SWL (95% CI HR 13%: 1.13 to 1.92), comorbidity (95% CI HR 2 points: 1.02 to 1.26), age (95% CI HR 14 years: 1.35 to 2.52), presence of AS symptoms (95% CI HR 1.09 to 2.68) and EF (95% CI HR 15% 0.52 to 0.86).
The superiority of non–flow-corrected indices over AVA and AVR was corroborated in patients in sinus rhythm for the three major efficacy end points.
|
Discussion
|
|---|
Onset of dyspnea, angina, or syncope identifies a critical point in the natural history of AS (37). Once symptoms appear, average survival rapidly falls and prompt valve replacement is warranted. Because the risk of sudden death while patients remain asymptomatic does not outweigh the complications of artificial prostheses, valve replacement should be withheld until patients become symptomatic (1). Although they are well established, it is recognized that these guidelines are based on retrospective studies that defined the spontaneous course of mostly congenital and rheumatic disease (20,22,37,38). Today’s clinical spectrum of AS is different. Calcific-degenerative has become the most frequent etiology, and AS is now a highly prevalent disease in the elderly (39,40). These changes are frequently responsible for uncertainty regarding the definition of optimal timing of valve replacement in individual patients. First, defining symptomatic status is particularly difficult in the elderly (5). Second, whether mild exertion dyspnea should be considered an index symptom in order to recommend surgery also remains controversial (25). Third, degenerative AS is known to be related to widespread cardiovascular involvement (41), and even when symptoms are clear, it may be impossible to discern whether they are related to the disease or to associated ischemic or hypertensive cardiomyopathy. Finally, concomitant obstructive lung disease, a common disease in the elderly, may account for patients’ dyspnea instead of AS (5).
A robust measurement of disease severity is mandatory in these contexts. It is recognized that clinical outcome is the only end point available for defining severity, owing to the lack of any hemodynamic gold standard (2). This is the first study to compare clinical efficacy of different indices of AS and demonstrates the superiority of non–flow-corrected indices in a broad population. Multiple theoretical and technical studies have demonstrated that AVA is the index that best characterizes the severity of the outflow obstruction caused by a stenotic valve (42,43). However, best technical efficacy does not necessarily imply superior clinical outcome efficacy (18). Although they are limited to measuring AS severity from a fluid-dynamic viewpoint, non–flow-corrected indices account for additional factors such as LV and systemic response to the outflow obstruction.
Among non–flow-corrected indices, SWL showed highest clinical efficacy, even in patients with impaired LV function. The present study demonstrates that if SWL is >23%, probability of showing symptoms attributable to AS is >80% at the time of the echo-Doppler exam; if SWL is 25%, median cardiac survival without valve replacement is close to one year, or slightly better the patient is asymptomatic; and if SWL is >26% probability of events is >30% in the following three months, increasing to 87% if the patient is symptomatic. Although it may be argued that valve replacement is a "soft" end point that is arbitrarily determined during the natural history of the disease, it is noteworthy that referring physicians based their decision to operate on AVA and P, unaware of the values of SWL. Furthermore, 67 out of 198 AS events (34%) were cardiac deaths, thereby allowing assessment of the natural history of the disease in a relatively large number of patients.
Although SWL was proposed more than 30 years ago (15), little attention has been paid to its clinical application. From a fluid-dynamic basis, this index represents the amount of energy the LV dissipates as heat because of outflow obstruction (15,44). Invasive studies have demonstrated an inverse and quadratic correlation between SWL and AVA, and values >30% predict a critically narrow orifice (44); this value is close to the cutoff values identified in the present study. A number of reasons may account for the highest outcome efficacy of this index. According to its formula, SWL can be interpreted as a blood-pressure normalization of P. Low systolic blood pressure is a hallmark of severe AS (45), and SWL accounts for such effect: higher values are obtained as blood pressure falls for a given P. Blood pressure response to AS is known to be subject to a number of factors, particularly age. Therefore the findings of our study, demonstrating maximal prognostic value of SWL in a cohort with a majority of elderly patients, justify further investigation on peripheral adaptation to AS. Also, even though flow-normalized indices such as AVA are theoretically the most robust measurements of severity, follow-up studies of unselected patients have found prognostic value to be highest for non–flow-corrected indices such as Vmax and P (8,25,46). Placed above the other diagnostic goals, increased clinical outcome efficacy is probably the consequence of improved technical and diagnostic-accuracy efficacy (19). Stroke-work loss is based only on pressure estimates, without the need of measuring flow rate. Because the latter is the most important source of error and variability, both in Doppler and cardiac catheterization-derived hemodynamic calculations (10,46), SWL can be more accurately obtained than AVA and AVR, particularly in patients in atrial fibrillation. Also, P has shown to correlate with symptomatic status (47), and its combination with blood pressure has been demonstrated to powerfully predict survival in unoperated patients (48). Stroke-work loss mathematically incorporates both P and systolic blood pressure, and therefore adds their individual statistical value.
The demonstration of flow-mediated changes in AVA led some authors to postulate that stress interventions may unmask the opening reserve available for patients to increase their cardiac output (9,16,49). As valves become stiffer, valvular reserve would decline and limit the capability of the ventricle to increase output. Once valvular reserve reaches a critical point, patients would develop characteristic exercise symptoms. Because valve stiffness is not directly related to baseline AVA, stress interventions are required to assess it (10,50). However, both the present and a previous (2) study have failed to find any additive clinical value of stress testing in cohorts of patients with mostly normal LV function.
Study limitations.
Stroke-work loss in the retrospective cohort was calculated using non-simultaneous measurements of P and SBP. As stated earlier, systolic blood pressure in the echocardiography laboratory was higher than values registered in the clinical records. However, the impact of this small bias on calculated SWL is negligible: an increase of SBP of 11 mm Hg (upper 95% CI of bias) translates to only a 1% absolute decrease in SWL (23% to 22%, for average values of P and SBP in Group A).
The purpose of the study was to assess clinical performance of AS diagnostic tests and not to analyze natural-history predictors. Hence, the goal was to recruit a representative sample of most subjects in whom clinical indices are employed to guide clinical management, as recommended for test evaluation designs (51). Furthermore, the data analysis strategy was designed to minimize the number of variables tested, and the stability of the final models was demonstrated by resampling techniques. Using these methods, overfitting is remarkably reduced, allowing validation populations to include groups used for model definition (32).
Conclusions
The present study demonstrates that non–flow-corrected indices of AS are the most clinically efficient in terms of predicting symptomatic status and outcome. Among them, SWL is the Doppler-echocardiographic index that best accounts for clinical severity of adult AS. Patients showing values >25% are likely to be symptomatic and have an adverse outcome, irrespective of symptomatic status. Thus, SWL should be incorporated in clinical assessment of AS and used to aid patient management in unclear situations. Inotropic stimulation is of no prognostic value in unselected patients.
|
Acknowledgments
|
|---|
We are indebted to Catherine M. Otto for her suggestions on data presentation.
|
References
|
|---|
1. Bonow RO, Carabello B, de Leon AC Jr, et al. Guidelines for the management of patients with valvular heart disease: executive summary A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Management of Patients with Valvular Heart Disease). Circulation. 1998;98:1949–1984[Free Full Text]
2. Otto CM. Aortic stenosis: echocardiographic evaluation of disease severity, disease progression, and the role of echocardiography in clinical decision making. Otto CM. The Practice of Clinical Echocardiography. Philadelphia, PA: W.B. Saunders Company; 1997. p. 405–432
3. Carabello BA. Do all patients with aortic stenosis and left ventricular dysfunction benefit from aortic valve replacement? Cathet Cardiovasc Diagn. 1989;17:131–132[Medline]
4. Smucker ML, Manning SB, Stuckey TD, Tyson DL, Nygaard TW, Kron IL. Preoperative left ventricular wall stress, ejection fraction, and aortic valve gradient as prognostic indicators in aortic valve stenosis. Cathet Cardiovasc Diagn. 1989;17:133–143[Medline]
5. Horan MA. Presentation of disease in old age. Tallis R, Fillit H, Brocklehurst JC. Brockelhurst’s Textbook of Geriatric Medicine and Gerontology. 5th ed. New York, NY: Churchill Livingstone; 1996. p. 201–206
6. Otto CM. The difficulties in assessing patients with moderate aortic stenosis. Heart. 1999;82:5–6[Free Full Text]
7. Lombard JT, Selzer A. Valvular aortic stenosis. A clinical and hemodynamic profile of patients. Ann Intern Med. 1987;106:292–298[Abstract/Free Full Text]
8. Otto CM, Burwash IG, Legget ME, et al. Prospective study of asymptomatic valvular aortic stenosis. Clinical, echocardiographic, and exercise predictors of outcome. Circulation. 1997;95:2262–2270[Abstract/Free Full Text]
9. Burwash IG, Pearlman AS, Kraft CD, Miyake-Hull C, Healy NL, Otto CM. Flow dependence of measures of aortic stenosis severity during exercise. J Am Coll Cardiol. 1994;24:1342–1350[Abstract]
10. Bermejo J, Garcia-Fernandez MA, Torrecilla EG, et al. Effects of dobutamine on Doppler echocardiographic indexes of aortic stenosis. J Am Coll Cardiol. 1996;28:1206–1213[Abstract]
11. Burwash IG, Thomas DD, Sadahiro M, et al. Dependence of Gorlin formula and continuity equation valve areas on transvalvular volume flow rate in valvular aortic stenosis. Circulation. 1994;89:827–835[Abstract/Free Full Text]
12. Carabello BA. Clinical practice. Aortic stenosis. N Engl J Med. 2002;346:677–682[Free Full Text]
13. Ford LE, Feldman T, Chiu YC, Carroll JD. Hemodynamic resistance as a measure of functional impairment in aortic valvular stenosis. Circ Res. 1990;66:1–7[Abstract/Free Full Text]
14. Rodrigo FA, Snellen HA. Estimation of valve area and valvular resistance. Am Heart J. 1953;45:1–12[CrossRef][Medline]
15. Tobin JR Jr, Rahimtoola SH, Blundell PE, Swan HJ. Percentage of left ventricular stroke work loss. A simple hemodynamic concept for estimation of severity in valvular aortic stenosis. Circulation. 1967;35:868–879[Abstract/Free Full Text]
16. Otto CM, Pearlman AS, Kraft CD, Miyake-Hull CY, Burwash IG, Gardner CJ. Physiologic changes with maximal exercise in asymptomatic valvular aortic stenosis assessed by Doppler echocardiography. J Am Coll Cardiol. 1992;20:1160–1167[Abstract]
17. Fryback DG, Thornbury JR. The efficacy of diagnostic imaging. Med Decis Making. 1991;11:88–94[Abstract/Free Full Text]
18. Thornbury JR, Eugene W. Caldwell Lecture. Clinical efficacy of diagnostic imaging: love it or leave it. Am J Roentgenol. 1994;162:1–8[Abstract/Free Full Text]
19. Pearl WS. A hierarchical outcomes approach to test assessment. Ann Emerg Med. 1999;33:77–84[CrossRef][Medline]
20. Chizner MA, Pearle DL, deLeon AC Jr. The natural history of aortic stenosis in adults. Am Heart J. 1980;99:419–424[CrossRef][Medline]
21. Rosenhek R, Binder T, Porenta G, et al. Predictors of outcome in severe, asymptomatic aortic stenosis. N Engl J Med. 2000;343:611–617[Abstract/Free Full Text]
22. Turina J, Hess O, Sepulcri F, Krayenbuehl HP. Spontaneous course of aortic valve disease. Eur Heart J. 1987;8:471–483[Abstract/Free Full Text]
23. Horstkotte D, Loogen F. The natural history of aortic valve stenosis. Eur Heart J. 1988;9(Suppl E):57–64[Abstract]
24. Kelly TA, Rothbart RM, Cooper CM, Kaiser DL, Smucker ML, Gibson RS. Comparison of outcome of asymptomatic to symptomatic patients older than 20 years of age with valvular aortic stenosis. Am J Cardiol. 1988;61:123–130[CrossRef][Medline]
25. Pellikka PA, Nishimura RA, Bailey KR, Tajik AJ. The natural history of adults with asymptomatic, hemodynamically significant aortic stenosis. J Am Coll Cardiol. 1990;15:1012–1017[Abstract]
26. Kennedy KD, Nishimura RA, Holmes DR Jr., Bailey KR. Natural history of moderate aortic stenosis. J Am Coll Cardiol. 1991;17:313–319[Abstract]
27. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373–383[CrossRef][Medline]
28. American Society of Echocardiography Committee on Standards, Subcommittee on Quantitation of Two-Dimensional Echocardiograms. Recommendations for quantification of the left ventricle by two-dimensional echocardiography. J Am Soc Echocardiogr. 1989;2:361–367
29. Nijland F, Kamp O, Karreman AJ, van Eenige MJ, Visser CA. Prognostic implications of restrictive left ventricular filling in acute myocardial infarction: a serial Doppler echocardiographic study. J Am Coll Cardiol. 1997;30:1618–1624[Abstract]
30. Sakata K, Kashiro S, Hirata S, Yanagisawa A, Ishikawa K. Prognostic value of Doppler transmitral flow velocity patterns in acute myocardial infarction. Am J Cardiol. 1997;79:1165–1169[CrossRef][Medline]
31. Otto CM, Pearlman AS. Doppler echocardiography in adults with symptomatic aortic stenosis. Diagnostic utility and cost-effectiveness. Arch Intern Med. 1988;148:2553–2560[Abstract/Free Full Text]
32. Harrell FE. Predicting Outcomes: Applied Survival Analysis and Logistic Regression. 5th ed. Charlottesville, VA: University of Virginia Press; 1997.
33. Carlos ME, Smart SC, Wynsen JC, Sagar KB. Dobutamine stress echocardiography for risk stratification after myocardial infarction. Circulation. 1997;95:1402–1410[Abstract/Free Full Text]
34. Breiman L, Friedman JH, Olshen RA, Stone CJ. Classification and Regression Trees. 5th ed. Pacific Grove, CA: Wadsworth and Brooks/Cole; 1984.
35. George H, Mansmann U. Survcart S and C code for classification and regression trees with survival data. Public-domain software available from StatLib (http://lib.stat.cmu.edu/S/), 1995
36. Harrell FE. Hmisc and Design S-Plus function library. Public-domain software available from http://hesweb1.med.virginia.edu/biostat/s/splus.html, 2001
37. Ross J Jr, Braunwald E. Aortic stenosis. Circulation. 1968;38:61–67[Medline]
38. Olesen KH, Warburg E. Isolated aortic stenosis—the late prognosis. Acta Med Scand. 1957;160:437–446
39. Lindroos M, Kupari M, Heikkila J, Tilvis R. Prevalence of aortic valve abnormalities in the elderly: an echocardiographic study of a random population sample. J Am Coll Cardiol. 1993;21:1220–1225[Abstract]
40. Stewart BF, Siscovick D, Lind BK, et al. Clinical factors associated with calcific aortic valve disease. Cardiovascular Health Study. J Am Coll Cardiol. 1997;29:630–634[Abstract]
41. Otto CM, Lind BK, Kitzman DW, Gersh BJ, Siscovick DS. Association of aortic-valve sclerosis with cardiovascular mortality and morbidity in the elderly. N Engl J Med. 1999;341:142–147[Abstract/Free Full Text]
42. Voelker W, Reul H, Nienhaus G, et al. Comparison of valvular resistance, stroke work loss, and Gorlin valve area for quantification of aortic stenosis. An in vitro study in a pulsatile aortic flow model. Circulation. 1995;91:1196–1204[Abstract/Free Full Text]
43. Bermejo J, Antoranz JC, Burwash IG, et al. In-vivo analysis of the instantaneous transvalvular pressure difference-flow relationship in aortic valve stenosis: implications of unsteady fluid-dynamics for the clinical assessment of disease severity. J Heart Valve Dis. 2002;11:557–566[Medline]
44. Sprigings DC, Chambers JB, Cochrane T, Allen J, Jackson G. Ventricular stroke work loss: validation of a method of quantifying the severity of aortic stenosis and derivation of an orifice formula. J Am Coll Cardiol. 1990;16:1608–1614[Abstract]
45. Wood P. Aortic stenosis. Am J Cardiol. 1958;1:553–571[CrossRef][Medline]
46. Otto CM, Pearlman AS, Gardner CL. Hemodynamic progression of aortic stenosis in adults assessed by Doppler echocardiography. J Am Coll Cardiol. 1989;13:545–550[Abstract]
47. Archer SL, Mike DK, Hetland MB, Kostamo KL, Shafer RB, Chesler E. Usefulness of mean aortic valve gradient and left ventricular diastolic filling pattern for distinguishing symptomatic from asymptomatic patients. Am J Cardiol. 1994;73:275–281[CrossRef][Medline]
48. Hammermeister KE, Cantor AB, Burchfiel CM, Sethi GK, Hong DM. Clinical, haemodynamic and angiographic predictors of survival in unoperated patients with aortic stenosis. The VA Cooperative Study on Valvular Heart Disease. Eur Heart J. 1988;9(Suppl E):65–69[Abstract/Free Full Text]
49. Lester SJ, Heilbron B, Gin K, Dodek A, Jue J. The natural history and rate of progression of aortic stenosis. Chest. 1998;113:1109–1114[CrossRef][Medline]
50. Shively BK, Charlton GA, Crawford MH, Chaney RK. Flow dependence of valve area in aortic stenosis: relation to valve morphology. J Am Coll Cardiol. 1998;31:654–660[Abstract/Free Full Text]
51. Reid MC, Lachs MS, Feinstein AR. Use of methodological standards in diagnostic test research. Getting better but still not good. JAMA. 1995;274:645–651[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
A. N. Azadani, N. Jaussaud, P. B. Matthews, L. Ge, T. S. Guy, T. A.M. Chuter, and E. E. Tseng
Energy Loss Due to Paravalvular Leak With Transcatheter Aortic Valve Implantation
Ann. Thorac. Surg.,
December 1, 2009;
88(6):
1857 - 1863.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
H. Baumgartner, J. Hung, J. Bermejo, J. B. Chambers, A. Evangelista, B. P. Griffin, B. Iung, C. M. Otto, P. A. Pellikka, and M. Quinones
Echocardiographic assessment of valve stenosis: EAE/ASE recommendations for clinical practice
Eur J Echocardiogr,
January 1, 2009;
10(1):
1 - 25.
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. P. Dal-Bianco, B. K. Khandheria, F. Mookadam, F. Gentile, and P. P. Sengupta
Management of Asymptomatic Severe Aortic Stenosis
J. Am. Coll. Cardiol.,
October 14, 2008;
52(16):
1279 - 1292.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. W. Akins, B. Travis, and A. P. Yoganathan
Energy loss for evaluating heart valve performance.
J. Thorac. Cardiovasc. Surg.,
October 1, 2008;
136(4):
820 - 833.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. Wachtell
Left ventricular systolic performance in asymptomatic aortic stenosis
Eur. Heart J. Suppl.,
July 1, 2008;
10(suppl_E):
E16 - E22.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
H. Turto, J. Lommi, M. Ventila, and M. Kupari
Doppler echocardiography markedly underestimates left ventricular stroke work loss in severe aortic valve stenosis
Eur J Echocardiogr,
October 1, 2007;
8(5):
341 - 345.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. H Little, K.-L. Chan, and I. G Burwash
Impact of blood pressure on the Doppler echocardiographic assessment of severity of aortic stenosis
Heart,
July 1, 2007;
93(7):
848 - 855.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J Bermejo and R Yotti
Low-gradient aortic valve stenosis: value and limitations of dobutamine stress testing
Heart,
March 1, 2007;
93(3):
298 - 302.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
Authors/Task Force Members, A. Vahanian, H. Baumgartner, J. Bax, E. Butchart, R. Dion, G. Filippatos, F. Flachskampf, R. Hall, B. Iung, et al.
Guidelines on the management of valvular heart disease: The Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology
Eur. Heart J.,
January 26, 2007;
(2007)
ehl428v1.
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. K. Stout and C. M. Otto
Indications for Aortic Valve Replacement in Aortic Stenosis
J Intensive Care Med,
January 1, 2007;
22(1):
14 - 25.
[Abstract]
[PDF]
|
|
|
|
|
|
|
C. M. Otto
Valvular Aortic Stenosis: Disease Severity and Timing of Intervention
J. Am. Coll. Cardiol.,
June 6, 2006;
47(11):
2141 - 2151.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. Blais, I. G. Burwash, G. Mundigler, J. G. Dumesnil, N. Loho, F. Rader, H. Baumgartner, R. S. Beanlands, B. Chayer, L. Kadem, et al.
Projected Valve Area at Normal Flow Rate Improves the Assessment of Stenosis Severity in Patients With Low-Flow, Low-Gradient Aortic Stenosis: The Multicenter TOPAS (Truly or Pseudo-Severe Aortic Stenosis) Study
Circulation,
February 7, 2006;
113(5):
711 - 721.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J Jimenez-Candil, J Bermejo, R Yotti, C Cortina, M Moreno, J L Cantalapiedra, and M A Garcia-Fernandez
Effects of angiotensin converting enzyme inhibitors in hypertensive patients with aortic valve stenosis: a drug withdrawal study
Heart,
October 1, 2005;
91(10):
1311 - 1318.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. Briand, J. G. Dumesnil, L. Kadem, A. G. Tongue, R. Rieu, D. Garcia, and P. Pibarot
Reduced Systemic Arterial Compliance Impacts Significantly on Left Ventricular Afterload and Function in Aortic Stenosis: Implications for Diagnosis and Treatment
J. Am. Coll. Cardiol.,
July 19, 2005;
46(2):
291 - 298.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J Bermejo
The effects of hypertension on aortic valve stenosis
Heart,
March 1, 2005;
91(3):
280 - 282.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
V. D. Bello, D. Giorgi, P. Viacava, T. Enrica, C. Nardi, C. Palagi, M. G. D. Donne, F. Verunelli, M. A. Mariani, J. Grandjean, et al.
Severe Aortic Stenosis and Myocardial Function: Diagnostic and Prognostic Usefulness of Ultrasonic Integrated Backscatter Analysis
Circulation,
August 17, 2004;
110(7):
849 - 855.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. Weerasinghe, M. Yusuf, T. Athanasiou, A. Wood, P. Magee, and R. Uppal
Role of transvalvular gradient in outcome from valve replacement for aortic stenosis
Ann. Thorac. Surg.,
April 1, 2004;
77(4):
1266 - 1271.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. E. Weyman
The year in echocardiography
J. Am. Coll. Cardiol.,
January 7, 2004;
43(1):
140 - 148.
[Full Text]
[PDF]
|
|
|
|
Top
Abstract
Methods
