Does adherence to empirical recommendations for serial evaluations of aortic stenosis improve clinical outcomes?
In this cohort study of 300 patients with asymptomatic severe aortic stenosis seen in an ambulatory clinic, there was poorer survival and higher heart failure hospitalization for patients who were not adherent to current recommendations for close monitoring over a median follow-up period of 4.5 years.
Guideline adherence for serial evaluations in aortic stenosis is associated with improved clinical outcomes, and further initiatives to improve such adherence are warranted.
For patients with asymptomatic severe aortic stenosis and normal left ventricular function, current practice guidelines empirically recommend serial evaluations every 6 to 12 months. The benefit of this clinical monitoring is unknown.
To determine the association of guideline adherence with clinical outcomes in patients with asymptomatic severe aortic stenosis.
Design, Setting, and Participants
This retrospective cohort study involved 300 patients with asymptomatic severe aortic stenosis who were seen in the ambulatory Minneapolis Heart Institute at Abbott Northwestern Hospital. Rates of survival and adverse clinical events, including myocardial infarction, stroke, and heart failure hospitalization, were compared between patients who adhered to serial evaluation guidance and those who did not. Medical records were reviewed from July 25, 2007, to December 6, 2012. Data analysis took place from February 4, 2017, to July 10, 2017.
Main Outcomes and Measures
All-cause mortality, heart failure hospitalization, and major adverse clinical events during follow-up.
The study population of 300 comprised 143 men (47.7%) and had a mean (SD) age of 78.6 (11.5) years. There were no differences in age, race/ethnicity, sex, comorbidities, insurance status, left ventricular function, and aortic stenosis severity between patients with (n = 202) and patients without (n = 98) guideline adherence. Aortic valve replacement (surgical or catheter based) was performed more frequently (54.0% vs 19.4%; P < .001) and the median (interquartile range) time for this performance was earlier (2.2 [1.2-3.6] years vs 3.5 [2.0-5.8] years; P < .001) in patients with guideline adherence. All-cause mortality was higher for nonadherent patients (hazard ratio [HR], 1.57; 95% CI, 1.07-2.30; P < .001), and these patients also had a higher rate of hospital admission for heart failure decompensation in follow-up (HR, 1.66; 95% CI, 1.27-2.18; P < .001). Four-year survival that is free from death and heart failure hospitalization was higher for adherent patients than for nonadherent patients (38.7% vs 23.3%; P < .001), and this difference remained significant in models adjusted for baseline variables (adjusted HR, 1.54; 95% CI, 1.04-2.29; P = .03).
Conclusions and Relevance
The findings of this study support the need for close monitoring of patients with asymptomatic severe aortic stenosis and help to validate current guidelines for serial evaluations. These findings also support initiatives to improve guideline adherence in clinical practice.
Natural history studies of patients with asymptomatic severe aortic stenosis have demonstrated a low risk of sudden cardiac death, but most of these patients will develop symptoms or have cardiac events within 4 years.1-3 Current practice guidelines empirically recommend serial evaluations every 6 to 12 months for patients with asymptomatic severe aortic stenosis and normal left ventricular function, yet the benefit of this close monitoring is unknown.4 Guideline adherence for serial evaluations has relevance not only for clinical outcomes but also for resource utilization and educational initiatives for promoting disease awareness and penetrance of therapy in patients with severe aortic stenosis who may benefit from surgical intervention.5
Accordingly, we undertook this investigation to examine the association of guideline adherence with survival, major adverse clinical events, and heart failure hospitalization in patients with asymptomatic severe aortic stenosis.
Using the electronic medical record, we retrospectively sampled 300 patients who had been diagnosed with asymptomatic severe aortic stenosis at the Minneapolis Heart Institute at Abbott Northwestern Hospital, Minneapolis, Minnesota. Inclusion criteria for this investigation were age 18 years or older; severe aortic stenosis, according to current practice guidelines; asymptomatic status, defined as the absence of dyspnea, angina, presyncope, or syncope; no prior aortic valve surgical or catheter-based intervention; no clinical indication for cardiac surgery; and clinical evaluation before December 31, 2012, to enable adequate follow-up duration. The Allina Institutional Review Board approved this study. All patients provided written informed consent for the use of their medical record for this research. Medical records were reviewed from July 25, 2007, to December 6, 2012. Data analysis took place from February 4, 2017, to July 10, 2017.
Data Collection and Definitions
The electronic medical record was reviewed for baseline patient characteristics, including age, sex, morbidities, surgical history, and echocardiographic findings. The date of the clinical evaluation corresponding to the first diagnosis of asymptomatic severe aortic stenosis was noted. Qualifying as an appropriate subsequent evaluation of aortic stenosis required the following documentation: a comprehensive clinical evaluation that included a description of the presence or absence of cardiac symptoms as well as a cardiopulmonary physical examination and a 2-dimensional and Doppler echocardiogram that included an assessment of left ventricular function and the hemodynamic severity of aortic stenosis, with documentation of the aortic valve area and either the peak aortic velocity or mean aortic valve gradient. Guideline adherence was defined as a serial evaluation occurring every 12 (±6) months until aortic valve replacement or death during the follow-up period.
Rates of mortality, heart failure hospitalization, and major adverse clinical events during clinical follow-up were compared between patients who adhered to serial evaluation guidance and those who did not. To help account for differences in baseline characteristics, multivariate modeling was performed for the end point of death using the following variables: age (per 5 years), male sex (categorical), coronary artery disease (categorical), atrial fibrillation (categorical), diabetes mellitus (categorical), peak aortic velocity (per 1 m/s), mean aortic valve gradient (per 5 mm Hg), aortic valve area (per 0.1 cm2), prior percutaneous coronary intervention (categorical), left ventricular ejection fraction (per 10%), and guideline adherence (categorical). Descriptive statistics are displayed as means and SDs for continuous variables; number and percentage with characteristic are given for categorical variables. When continuous variables had skewed distributions (creatinine and echo parameters), data were summarized with medians and interquartile ranges (IQRs).
Categorical variables were analyzed using Pearson χ2 or the Fisher exact test. Continuous variables were analyzed using the unpaired, 2-tailed t test for normally distributed variables or the Kruskal-Wallis test for continuous variables with nonnormal distribution. Time-to-event analyses were performed using the Kaplan-Meier method. Comparisons between groups were assessed using the log-rank test. Two-sided P < .05 was considered significant, and P values were 2-sided where possible. All statistical calculations and plots were done with Stata, version 14.1 (StataCorp LLC).
The Table lists the baseline characteristics for the study population (n = 300), which comprised 143 men (47.7%) and had a mean (SD) age of 78.6 (11.5) years. Morbidities were common, including diabetes (77 patients [25.7%]), chronic obstructive pulmonary disease (46 [15.3%]), atrial fibrillation (110 [36.7%]), and coronary artery disease (143 [47.7%]). All patients met the criteria for severe aortic stenosis and preserved left ventricular function. Concomitant valvular heart disease included aortic regurgitation in 42 patients (15.1%) (moderate, 41 [14.7%]; severe, 1 [0.4%]), mitral regurgitation in 70 (23.9%) (moderate, 60 [20.5%]; severe, 10 [3.4%]), and tricuspid regurgitation in 60 (20.7%) (moderate, 38 [13.1%]; severe, 22 [7.6%]). In comparisons between patients with (n = 202) and patients without (n = 98) guideline adherence for serial clinical evaluations, there were no differences in age, race/ethnicity, sex, comorbidities, insurance status, left ventricular function, and severity of aortic stenosis.
Median (IQR) follow-up time for the entire cohort was 4.5 (2.8-6.5) years and did not differ between adherent and nonadherent patients (4.5 [2.9-6.5] vs 4.3 [2.3-6.2] years; P = .10). In the entire study population, deaths occurred in 110 patients (36.7%) during the follow-up evaluation, with a median (IQR) time to death of 3.4 (2.4-4.6) years.
Patients without guideline adherence had higher rates of death (hazard ratio [HR], 1.57; 95% CI, 1.07-2.30; P < .001), myocardial infarction (HR, 1.87; 95% CI, 1.00-3.49; P = .04), and stroke (HR, 1.94; 95% CI, 1.02-3.71; P = .04). Figure 1A shows survival that is free of all-cause mortality for adherent vs nonadherent patients. Four-year survival was poorer for nonadherent than for adherent patients (67.5% vs 81.3%; P < .001). The association of guideline nonadherence and all-cause mortality remained significant after multivariate adjustment for baseline variables (adjusted HR, 1.54; 95% CI, 1.04-2.29; P = .03). Hospitalization for heart failure also was more common for nonadherent than for adherent patients (HR, 1.66; 95% CI, 1.27-2.18; P < .001), occurring in 220 patients (73.3%) at a median (IQR) follow-up time of 2.2 (1.2-3.5) years (Figure 1B). For the combined end point of death and heart failure hospitalization, 4-year survival was greater for adherent than for nonadherent patients (38.7% vs 23.3%; P < .001; Figure 1C).
In the entire study cohort, 128 patients (42.7%) underwent surgical or catheter-based aortic valve replacement. Patients without adherence had a lower rate of aortic valve replacement than did adherent patients (19.4% vs 54.0%; P < .001). The median (IQR) time to aortic valve replacement was 2.2 (1.2-3.6) years for adherent patients vs 3.5 (2.0-5.8) years for nonadherent patients (P < .001) (Figure 2).
To our knowledge, the present investigation is the first to demonstrate a survival benefit associated with adherence to guideline recommendations for serial clinical evaluations in patients with asymptomatic severe aortic stenosis. By helping to validate current guideline recommendations for closely monitoring patients with asymptomatic severe aortic stenosis, our findings support the efforts to improve guideline adherence, with the ultimate goal of improving clinical outcomes for these patients.
Several studies on the natural history of asymptomatic severe aortic stenosis have demonstrated the occurrence of mortality, need for surgery, or symptom onset in most patients within 5 years of diagnosis. The rates of these events range from approximately 10% to 40% per year and are influenced by age, severity of stenosis, calcification, and accompanying morbidities such as renal insufficiency and vascular disease.2,3,6-8 Given the risk of these events, close monitoring of patients with asymptomatic severe aortic stenosis is warranted, with guideline recommendations of serial clinical evaluations every 6 to 12 months. These recommendations, nonetheless, are empirical and are, thus far, not based on clinical outcomes observed with their implementation. In this study, we observed greater survival, lower heart failure hospitalization, and lower rates of myocardial infarction and stroke for patients who were monitored closely within guideline recommendations.
Several explanations for the clinical improvement with guideline adherence are plausible. Patients with greater adherence may benefit from early identification of indications for surgical intervention, which is well known to be a life-saving therapy for severe aortic stenosis. Indeed, a higher frequency of patients received aortic valve replacement in the adherent group than in the nonadherent group (54% vs 19%; P < .001), and the adherent patients who had aortic valve replacement were treated earlier than the nonadherent patients (2.2 vs 3.5 years; P < .001). Alternatively, a healthy user effect may have been present for the adherent patients, whereas palliative approaches might have been undertaken more commonly in those who were nonadherent. Certainly, the nonrandomized nature of this study creates the potential for selection bias. Notably, there were no significant differences in the clinical characteristics between the 2 groups. Moreover, the overall population, by exclusion criteria, had no indications for surgical intervention at the initial evaluation. Of importance, in multivariate analyses, which help to account for potential selection bias, the association of guideline adherence with survival remained significant.
Our findings support initiatives to improve adherence to practice guidelines that prescribe close monitoring of patients with asymptomatic severe aortic stenosis, with the ultimate goal of pursuing surgical intervention as early as possible when indications are present. Aortic stenosis is a degenerative disease entity, and progression toward heart failure is inevitable for many patients. The principal indication for surgery is the onset of symptoms or left ventricular dysfunction, but disease progression is usually insidious and therefore can be challenging to manage without close follow-up evaluations. For patients with valvular heart disease, significant practice gaps in care optimization have been described, even in centers with focused expertise.5 Previous studies also have documented poor penetrance of life-saving therapy in patients who may be surgical candidates.5,9,10 These findings are particularly noteworthy because the risk of aortic valve replacement therapy continues to decrease, especially for those who may be transcatheter candidates, for whom the operative mortality may be less than 3%.11
Close monitoring of patients with asymptomatic severe aortic stenosis is warranted, and the current guidelines of serial evaluations performed annually are beneficial. More initiatives to promote guideline adherence are warranted, with the ultimate goal of timely intervention for patients who may benefit from aortic valve replacement.
Corresponding Author: Mario Gössl, MD, PhD, Valve Science Center, Minneapolis Heart Institute Foundation, 920 E 28th St, Ste 300, Minneapolis, MN 55407 (email@example.com).
Accepted for Publication: July 12, 2017.
Published Online: September 6, 2017. doi:10.1001/jamacardio.2017.2952
Author Contributions: Dr Gössl had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Sorajja, Farivar, Gössl.
Acquisition, analysis, or interpretation of data: Ahmed, Garberich, Harris, Gössl.
Drafting of the manuscript: Ahmed, Sorajja, Garberich, Farivar, Gössl.
Critical revision of the manuscript for important intellectual content: Ahmed, Farivar, Harris, Gössl.
Statistical analysis: Sorajja, Garberich.
Administrative, technical, or material support: Ahmed.
Study supervision: Sorajja, Farivar, Harris, Gössl.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Farivar reported receiving personal fees for consulting work from Abbott Vasculars, Edwards Lifesciences, and Medtronic. No other disclosures were reported.
AJ. The natural history of adults with asymptomatic, hemodynamically significant aortic stenosis. J Am Coll Cardiol
. 1990;15(5):1012-1017.PubMedGoogle ScholarCrossref
et al. Predictors of outcome in severe, asymptomatic aortic stenosis. N Engl J Med
. 2000;343(9):611-617.PubMedGoogle ScholarCrossref
et al. Natural history of very severe aortic stenosis. Circulation
. 2010;121(1):151-156.PubMedGoogle ScholarCrossref
et al; American College of Cardiology/American Heart Association Task Force on Practice Guidelines. 2014 AHA/ACC guideline for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol
. 2014;63(22):e57-e185.PubMedGoogle ScholarCrossref
et al. Practice gaps in the care of mitral valve regurgitation: insights from the American College of Cardiology mitral regurgitation gap analysis and advisory panel. Am Heart J
. 2016;172:70-79.PubMedGoogle ScholarCrossref
et al. Outcome of 622 adults with asymptomatic, hemodynamically significant aortic stenosis during prolonged follow-up. Circulation
. 2005;111(24):3290-3295.PubMedGoogle ScholarCrossref
RS. Comparison of outcome of asymptomatic to symptomatic patients older than 20 years of age with valvular aortic stenosis. Am J Cardiol
. 1988;61(1):123-130.PubMedGoogle ScholarCrossref
C, de la Morena
et al; Influencia del Diagnóstico de Estenosis Aórtica Severa—Influence of the Severe Aortic Stenosis Diagnosis (IDEAS) Investigators. Comparison of 1-year outcome in patients with severe aorta stenosis treated conservatively or by aortic valve replacement or by percutaneous transcatheter aortic valve implantation (data from a multicenter Spanish registry). Am J Cardiol
. 2016;118(2):244-250.PubMedGoogle ScholarCrossref
et al. Reasons for nonadherence to guidelines for aortic valve replacement in patients with severe aortic stenosis and potential solutions. Am J Cardiol
. 2010;105(9):1339-1342.PubMedGoogle ScholarCrossref
SK. Evaluation of patients with severe symptomatic aortic stenosis who do not undergo aortic valve replacement: the potential role of subjectively overestimated operative risk. Circ Cardiovasc Qual Outcomes
. 2009;2(6):533-539.PubMedGoogle ScholarCrossref
et al. Two-year outcomes in patients with severe aortic valve stenosis randomized to transcatheter versus surgical aortic valve replacement: the all-comers Nordic Aortic Valve Intervention randomized clinical trial. Circ Cardiovasc Interv
. 2016;9(6):e003665.PubMedGoogle ScholarCrossref