Author + information
- Received May 20, 2014
- Revision received August 17, 2014
- Accepted September 8, 2014
- Published online January 6, 2015.
- Christophe Tribouilloy, MD, PhD∗,†∗ (, )
- Dan Rusinaru, MD, PhD∗,‡,
- Sylvestre Maréchaux, MD, PhD§,
- Anne-Laure Castel, MD∗,
- Nicolas Debry, MD§,
- Julien Maizel, MD, PhD†,
- Romuald Mentaverri, PharmD, PhD†,
- Said Kamel, PharmD, PhD†,
- Michel Slama, MD, PhD† and
- Franck Lévy, MD∗,†
- ∗Department of Cardiology, University Hospital Amiens, Amiens, France
- †INSERM U-1088, Jules Verne University of Picardie, Amiens, France
- ‡Department of Cardiology, Hospital of Saint Quentin, Saint Quentin, France
- §Groupement des Hôpitaux de l'Institut Catholique de Lille/Faculté libre de médecine, Université Lille Nord de France, Lille, France
- ↵∗Reprint requests and correspondence:
Dr. Christophe Tribouilloy, INSERM, U-1088, University Hospital Amiens, Department of Cardiology, Avenue René Laënnec, 80054 Amiens Cedex 1, France.
Background Severe low-gradient, low-flow (LG/LF) aortic stenosis with preserved left ventricular ejection fraction (EF) has been described as a more advanced form of aortic stenosis. However, the natural history and need for surgery in patients with LG/LF aortic stenosis remain subjects of intense debate.
Objectives We sought to investigate the outcome of LG/LF aortic stenosis in comparison with moderate aortic stenosis and with high-gradient (HG) aortic stenosis in a real-world study, in the context of routine practice.
Methods This analysis included 809 patients (ages 75 ± 12 years) diagnosed with aortic stenosis and preserved EF (≥50%). Patients were divided into 4 groups: mild-to-moderate aortic stenosis; HG aortic stenosis; LG/LF aortic stenosis; and low-gradient, normal-flow (LG/NF) aortic stenosis.
Results Compared with mild-to-moderate aortic stenosis patients, LG/LF aortic stenosis patients had smaller valve areas and stroke volumes, higher mean gradients, and comparable degrees of ventricular hypertrophy. Under medical management (22.8 months; range 7 to 53 months), compared with mild-to-moderate aortic stenosis patients, HG aortic stenosis patients were at higher risk of death (adjusted hazard ratio [HR]: 1.47; 95% confidence interval [CI]: 1.03 to 2.07), whereas LG/LF aortic stenosis patients did not have an excess mortality risk (adjusted HR: 0.88; 95% CI: 0.53 to 1.48). During the entire (39.0 months; range 11 to 69 months) follow-up (with medical and surgical management), the mortality risk associated with LG/LF aortic stenosis was close to that of mild-to-moderate aortic stenosis (adjusted HR: 0.96; 95% CI: 0.58 to 1.53), whereas the excess risk of death associated with HG aortic stenosis was confirmed (adjusted HR: 1.74; 95% CI: 1.27 to 2.39). The benefit associated with aortic valve replacement was confined to the HG aortic stenosis group (adjusted HR: 0.29; 95% CI: 0.18 to 0.46) and was not observed for LG/LF aortic stenosis (adjusted HR: 0.75; 95% CI: 0.14 to 4.05).
Conclusions In this study, the outcome of severe LG/LF aortic stenosis with preserved EF was similar to that of mild-to-moderate aortic stenosis and was not favorably influenced by aortic surgery. Further research is needed to better understand the natural history and the progression of LG/LF aortic stenosis.
A variable proportion of patients with aortic stenosis and preserved (>50%) left ventricular ejection fraction (EF) that are classified as “severe” by echocardiography (aortic valve area [AVA] <1 cm² or index AVA <0.6 cm2/m2) (1,2) have lower peak aortic velocity (<4 m/s) and/or a lower mean Doppler gradient (MDG) (<40 mm Hg) (3–7). Besides inconsistencies related to small body surface area and errors in measurement of AVA or Doppler parameters, this discordance may reflect a low stroke volume (SV) index, despite a normal EF (paradoxical “low-flow”) (8,9). Patients with low-gradient/low-flow (LG/LF) aortic stenosis have been reported to have small ventricular cavities (3,8), severe concentric hypertrophy (3,8), increased afterload (10,11), restrictive physiology (8), subtle systolic dysfunction (12), and increased subendocardial myocardial fibrosis (13). These features have been interpreted as markers of a more advanced disease, leading to poor prognosis under conservative therapy. Survival analyses of limited patient numbers suggest that LG/LF aortic stenosis is associated with greater mortality risk than high-gradient (HG) severe aortic stenosis (3–5,7,8), and that surgery might be beneficial in this subset of patients. Therefore, guidelines recommend (class IIa recommendation) aortic valve replacement (AVR) in symptomatic patients with LG/LF aortic stenosis when documented valvular obstruction is the most probable cause of symptoms (2).
The view that LG/LF aortic stenosis with preserved EF represents a more advanced form of the disease with pejorative outcome has been questioned by a subanalysis of the SEAS (Simvastatin and Ezetimibe in Aortic Stenosis) trial (14) and by a recent single-center European cohort study (15). Consistently, magnetic resonance data in LG aortic stenosis show larger AVAs, less hypertrophy, and similar focal fibrosis compared with HG aortic stenosis (16). These recent results raise doubts about whether LG/LF aortic stenosis with preserved EF truly represents a severe form of aortic stenosis with poor prognosis and increased risk of death when treated medically. Moreover, the impact of AVR in LG/LF aortic stenosis needs to be better defined to avoid unnecessary and potentially dangerous procedures.
The basis for the present study is the consecutive experience with aortic stenosis at the echocardiography laboratories of 2-F tertiary centers (Amiens and Lille) between 2000 and 2012. The aims of this analysis are 3-fold: 1) to establish the relation between LG/LF aortic stenosis and outcome, regardless of clinical management; 2) to compare the outcomes of LG aortic stenosis, HG aortic stenosis, and moderate aortic stenosis; and 3) to understand the impact of AVR in these subsets of patients.
Consecutive patients ≥18 years of age diagnosed with mild or more than mild aortic stenosis (aortic valve calcification with reduction in systolic movements and AVA <2 cm²) and an EF of ≥50% (1) who were managed medically for at least 3 months after diagnosis were prospectively identified and included in an electronic database. We excluded patients with the following: 1) more than mild aortic and/or mitral regurgitation; 2) prosthetic valves, congenital heart disease, supravalvular or subvalvular aortic stenosis, or dynamic left ventricular outflow tract obstruction; 3) an EF <50%; and 4) patients who denied authorization for research participation. We enrolled 898 patients. Subsequently, 89 were excluded because of missing data (n = 83) or absence of follow-up (n = 6). Patients were retrospectively classified into 3 groups: mild-to-moderate aortic stenosis (AVA ≥1 cm² or indexed AVA ≥0.6 cm², and MDG <40 mm Hg; n = 420); LG aortic stenosis (AVA <1 cm², indexed AVA <0.6 cm², and MDG <40 mm Hg; n = 142); and HG aortic stenosis (AVA <1 cm², indexed AVA <0.6 cm², and MDG ≥40 mm Hg, n = 247). LG aortic stenosis was further divided according to the SV index in LG/LF aortic stenosis (n = 57) when the index SV was <35 ml/m², and low-gradient, normal-flow (LG/NF) (n = 85) aortic stenosis when index SV was ≥35 ml/m² (3,4).
An index summating the patient’s individual comorbidities was calculated (17). Coronary artery disease (CAD) was defined by the presence of a documented history of acute coronary syndromes, CAD previously confirmed by coronary angiography, or history of coronary revascularization.
We obtained institutional review board authorizations before conducting the study. The study was conducted in accordance with institutional policies, national legal requirements, and the revised Declaration of Helsinki.
All patients underwent a comprehensive Doppler echocardiographic study, using commercially available ultrasound systems. Peak aortic velocity was recorded using continuous-wave Doppler in several acoustic windows (apical 5-chamber view, right parasternal, suprasternal, epigastric). AVA was calculated by the continuity equation and indexed for body surface area. SV was calculated by multiplying the area of the left ventricular outflow tract by the outflow tract time–velocity integral (3). When patients were in sinus rhythm, 3 cardiac cycles were averaged for all measures. For patients in atrial fibrillation, 5 cardiac cycles were averaged. EF was calculated using Simpson’s biplane method. Left ventricular mass was estimated by the formula on the basis of linear measurements and indexed for body surface area.
Clinical decision and follow-up
After initial medical management, treatment was either conservative or surgical, as deemed appropriate by the patient’s personal physician. The majority of patients were followed by clinical consultation and echocardiography in the outpatient clinics of the 2 tertiary centers. Others were followed in public hospitals or private practices by referring cardiologists working with the tertiary centers. Information on follow-up was retrospectively obtained by direct patient interview or by repeated follow-up letters and questionnaires. Median follow-up with medical management was 22.8 months (range: 7 to 53 months). Median overall follow-up was 39.0 months (range: 11 to 69 months). The endpoint of the study was overall survival after diagnosis starting at baseline echocardiography; the endpoint was analyzed under medical management, and under medical and surgical management.
Continuous variables were expressed as mean ± 1 SD or medians and interquartile ranges, and categorical variables were summarized as numbers and frequency percentages. The relationship between baseline continuous baseline variables and the 4 groups was explored using 1-way analysis of variance (for normally distributed variables) or Kruskal-Wallis tests (for non-normally distributed variables). Pearson’s chi-square statistic or Fisher’s exact test were used to examine the association between the 4 groups and baseline categorical variables. The significance between mild-to-moderate aortic stenosis (referent group) and the other groups was examined if there was a significant difference across categories. Individual differences were compared with Mann-Whitney U tests (with Bonferroni correction for multiple comparisons) and Tukey tests for normally distributed data.
For analysis of outcomes under medical management, data were censored at the time of cardiac surgery (if performed). The entire follow-up was used to analyze outcomes under conservative and surgical treatments. The effect of surgery on outcome was analyzed as a time-dependent covariate using the entire follow-up (18). Survival rates ± 1 SE of the 4 groups were estimated according to the Kaplan-Meier method and compared with 2-sided log-rank tests. Multivariable analyses of all-cause mortality were performed using Cox proportional hazards models. The risk of death in the 3 severe aortic stenosis groups (LG/LF aortic stenosis, LG/NF aortic stenosis, and HG aortic stenosis) was estimated versus the referent group. We did not use model-building techniques, and entered covariates were considered of potential prognostic impact on an epidemiological basis in the models. These covariates were age, sex, body surface area, the comorbidity index, symptoms at baseline, CAD, history of atrial fibrillation, and EF. The proportional hazards assumption was confirmed using statistics and graphs on the basis of Schoenfeld residuals. All p values were results of 2-tailed tests. Data were analyzed with SPSS version 13.0 (SPSS Inc., Chicago, Illinois) and STATA (version 12, StataCorp LP, College Station, Texas). The authors had full access to the data and take responsibility for its integrity. All authors read and agreed to this paper as written.
The baseline demographic and clinical characteristics of the 809 patients with aortic stenosis are displayed in Table 1. Patients with LG aortic stenosis were older than those with mild-to-moderate aortic stenosis, were more often women, and had lower body surface areas (Table 1). The proportion of patients with New York Heart Association functional class III or IV symptoms at baseline was comparable between LG/LF aortic stenosis and moderate aortic stenosis. The comorbidity index was higher in LF/LG aortic stenosis patients compared with mild-to-moderate aortic stenosis patients, and lower in HG aortic stenosis patients (Table 1).
Baseline echocardiography data are summarized in Table 2. Compared with mild-to-moderate aortic stenosis, patients with LG/LF aortic stenosis had higher MDG and peak aortic velocity, and smaller SV, and less cardiac output. Left ventricular mass was comparable between LG/LF aortic stenosis and mild-to-moderate aortic stenosis (Table 2). HG aortic stenosis was associated with a significantly greater degree of ventricular hypertrophy than mild-to-moderate aortic stenosis (Table 2).
Outcome under medical management
In 588 patients (73%), management was solely medical. Eighty seven percent of LG/LF aortic stenosis patients presenting with symptoms at baseline were treated medically. Overall, crude 4-year mortality with medical management was similar for the 4 study groups: 28 ± 3% for mild-to-moderate aortic stenosis, 34 ± 8% for LG/LF aortic stenosis, 29 ± 7% for LG/NF aortic stenosis, and 31 ± 5% for HG aortic stenosis (Figure 1A).
On multivariable analysis (Table 3, Figure 2A), compared with the reference category (mild-to-moderate aortic stenosis), the 2 LG groups did not exhibit excess mortality risk with medical treatment (adjusted hazard ratio [HR]: 0.88; 95% confidence interval [CI]: 0.53 to 1.48 for LG/LF aortic stenosis vs. mild-to-moderate aortic stenosis, and adjusted HR: 1.06; 95% CI: 0.66 to 1.71 for LG/NF aortic stenosis vs. mild-to-moderate aortic stenosis). In contrast, HG aortic stenosis patients were at higher risk of death than those with mild-to-moderate aortic stenosis (adjusted HR: 1.47; 95% CI: 1.03 to 2.07). When only patients without symptoms at baseline were considered, the same mortality risk pattern was observed: patients with mild-to-moderate aortic stenosis, LG/LF aortic stenosis, and LG/NF aortic stenosis had comparable risks of death (adjusted HR: 0.89; 95% CI: 0.53 to 1.49 for LG/LF aortic stenosis vs. mild-to-moderate aortic stenosis, and adjusted HR: 1.06; 95% CI: 0.66 to 1.70 for LG/NF aortic stenosis vs. mild-to-moderate aortic stenosis). HG aortic stenosis patients were at significantly higher risk of death than patients with mild-to-moderate aortic stenosis (adjusted HR: 1.44; 95% CI: 1.02 to 2.04). Among symptomatic patients, patients with HG aortic stenosis had excess mortality under medical management compared with mild-to-moderate aortic stenosis (adjusted HR: 1.58; 95% CI: 1.02 to 2.48), whereas patients with LF/LG aortic stenosis and LG/NF aortic stenosis exhibited similar risks of death compared with mild-to-moderate aortic stenosis (adjusted HR: 1.29; 95% CI: 0.63 to 2.63 for LG/LF aortic stenosis, and adjusted HR: 1.08; 95% CI: 0.54 to 2.16 for LG/NF aortic stenosis). The outcome of symptomatic LF/LG aortic stenosis was not significantly different compared with asymptomatic moderate aortic stenosis (adjusted HR: 1.01; 95% CI: 0.50 to 2.04).
Outcome with medical and surgical management
AVR was eventually performed in 221 patients (27%), according to the decision of the patient’s physician. Aortic bioprostheses were used in 79% of cases (n = 174), and 67 patients had at least 1 associated bypass graft at the time of surgery. Compared with patients in whom management was exclusively medical, patients who underwent AVR were younger (p < 0.001), had lower Charlson indexes (p < 0.001), smaller AVAs (p < 0.001), higher MDGs (p < 0.001), and greater indexed left ventricular masses (p = 0.023). Surgical patients were more often men (p = 0.039) and more frequently had CAD (p < 0.001). Although 53% (n = 130) of patients with HG aortic stenosis had AVR during follow-up, this decreased to 15% (n = 63) in the mild-to-moderate aortic stenosis group. Only 12% (n = 7) of the LG/LF aortic stenosis group and 25% (n = 21) of the LG/NF aortic stenosis group underwent surgery.
The crude 4-year estimates of mortality under medical and surgical treatment were 28 ± 2% for mild-to-moderate aortic stenosis, 35 ± 7% for LG/LF aortic stenosis, 29 ± 6% for LG/NF aortic stenosis, and 29 ± 3% for HG aortic stenosis (Figure 1B). On multivariable analysis (Table 3, Figure 2B), the outcome of HG aortic stenosis was poorer than that of mild-to-moderate aortic stenosis (adjusted HR: 1.74; 95% CI: 1.27 to 2.39), whereas the outcome of LG/LF aortic stenosis and LG/NF aortic stenosis did not differ significantly from that of mild-to-moderate aortic stenosis (Table 3, Figure 2B). When the analysis was limited to asymptomatic patients, HG aortic stenosis was associated with greater risk than mild-to-moderate aortic stenosis (adjusted HR: 2.52; 95% CI: 1.48 to 4.30). LG/LF aortic stenosis (adjusted HR: 0.78; 95% CI: 0.38 to 1.62) and LG/NF aortic stenosis (adjusted HR: 1.73; 95% CI: 0.94 to 3.17) did not show significant outcome differences compared with mild-to-moderate aortic stenosis. Among patients with symptoms, patients with LG/LF aortic stenosis and LG/NF aortic stenosis had similar outcomes compared with patients with mild-to-moderate aortic stenosis (adjusted HR: 1.30; 95% CI: 0.43 to 3.94 for LG/LF aortic stenosis, and adjusted HR: 0.51; 95% CI: 0.10 to 3.87 for LG/NF aortic stenosis), whereas patients with HG aortic stenosis had excess mortality risk (adjusted HR: 1.54: 95% CI: 1.03 to 2.30). Under medical and surgical management, symptomatic LF/LG aortic stenosis showed no excess mortality risk compared with asymptomatic moderate aortic stenosis (adjusted HR: 1.13; 95% CI: 0.59 to 2.14).
Effect of surgery
Using surgery as a time-dependent variable, the benefit of surgery in the total study population was important (adjusted HR: 0.49; 95% CI: 0.35 to 0.68). Figure 3 depicts the overall benefit of surgery in patients with severe aortic stenosis. On multivariable analysis, there was a significant interaction between the aortic stenosis group classification and the magnitude of survival benefit after surgery (p for interaction = 0.022). The benefit associated with AVR was confined to the HG aortic stenosis group and was not significant in the other groups (Figure 4).
Our analysis of a large cohort of aortic stenosis patients with long-term follow-up puts into perspective the various flow-gradient patterns with regard to natural history and management. Mild-to-moderate aortic stenosis, LG aortic stenosis, and HG aortic stenosis display comparable long-term crude mortality with conservative therapy, as well as with medical and surgical management. After taking age, comorbidity, and other factors with prognostic implications into account, LG/LF aortic stenosis has a rather benign outcome under medical management, despite the symptomatic status. The natural history of LG/LF aortic stenosis appears similar to that of moderate aortic stenosis (Central Illustration). In contrast, regardless of management, HG aortic stenosis shows excess mortality compared with mild-to-moderate aortic stenosis. Finally, AVR is associated with improved outcome in HG aortic stenosis, but it does not significantly affect mortality in LG aortic stenosis.
The entity of “paradoxical” LG/LF aortic stenosis despite preserved EF was first described in 2007 (3). LG/LF aortic stenosis is diagnosed in patients presenting with an AVA of <1 cm², an indexed AVA of <0.6 cm²/m², an EF of >50%, and a MDG of <40 mm Hg in the setting of a low-flow (SV index <35 ml/m²) state (3,8). Carefully conducted invasive hemodynamics studies subsequently confirmed the existence of this entity (19,20). Its prevalence varies widely (between 3% and 24%) in published reports (3–5,7). Patients with LG/LF aortic stenosis are typically women, elderly, and have more severe concentric left ventricular hypertrophy, small cavity size, impaired ventricular filling, and an increased global hemodynamic load (3,4,8,10). For Dumesnil et al. (8,21), LG/LF aortic stenosis represents the most advanced stage of aortic stenosis with preserved EF. Nevertheless, more advanced disease at presentation does not necessarily imply a temporal relationship, for example, that HG aortic stenosis progresses to LG/LF aortic stenosis. Subsequently, severe LG aortic stenosis was frequently observed among patients with a preserved EF and NF (SV index ≥35 ml/min) and was interpreted mainly in relation to measurement errors or inconsistencies in the guidelines criteria (6,20). A new 4-group classification of aortic stenosis according to MDG and flow criteria was proposed in 2010 (8).
The prognosis of LG/LF aortic stenosis is the subject of an ongoing debate. Several studies have shown that patients with LG/LF aortic stenosis have a pejorative survival compared with both moderate aortic stenosis and HG aortic stenosis (3–5,7,22). Among 150 asymptomatic patients with severe aortic stenosis, LF/LG was identified as independent determinant of reduced event-free survival (4). Clavel et al. (5) used retrospective matching to compare the outcomes of LG/LF aortic stenosis, moderate aortic stenosis, and HG aortic stenosis, and reported a 1.71-fold increase in the adjusted risk of all-cause death and a 2.09-fold increase in the adjusted risk of cardiovascular death for LG/LF aortic stenosis compared with a pooled population of moderate aortic stenosis and HG aortic stenosis. Ozkan et al. (22) reported a 53% mortality rate for symptomatic LG aortic stenosis patients who were managed medically, with similar outcomes for patients with LG/LF aortic stenosis and LG/NF aortic stenosis who were under medical therapy. In this series, the outcome of LG aortic stenosis significantly improved with AVR (22). Finally, Eleid et al. (7) showed that the LG/LF pattern significantly increased the risk of death, whereas AVR had a markedly protective effect.
This perspective was questioned by a report from the SEAS trial that compared the long-term outcome of 2 asymptomatic subsets of aortic stenosis patients: 184 patients with moderate aortic stenosis and 435 patients with LG severe aortic stenosis (14). Outcomes, with regard to valve-related events, major cardiovascular events, or cardiac death during long-term follow-up, did not significantly differ between the 2 groups (14). Both groups had better prognoses than patients with HG aortic stenosis. This report (14) was criticized for analyzing a low-risk population that did not display typical features of LG/LF aortic stenosis, and was considered flawed by measurement errors, use of nonindexed AVA, and inconsistency of guideline criteria (21). Despite these arguments, we believe that these randomized trial data are not biased by confounding factors, such as poor functional status, symptoms and comorbidities, which play a major role when discussing surgery for LG/LS severe aortic stenosis. Recently, in a cohort of patients with severe aortic stenosis and preserved EF, Maes et al. reported better outcomes under medical management for patients with LG aortic stenosis compared with patients with HG stenosis, and observed that >80% of patients with LG aortic stenosis exhibited significant increases in MDG during follow-up (15).
Our results showed that the long-term outcome of the LG/LF pattern is similar to that of mild-to-moderate aortic stenosis, regardless of the management type and symptomatic status; therefore, LG/LF aortic stenosis does not represent a more advanced stage of the disease. This result should be interpreted with some important considerations kept in mind. First, we separated LG aortic stenosis into LG/LF and LG/NF aortic stenosis, included the indexed AVA to define severe LG/LF aortic stenosis, calculated the index SV according to guideline criteria (1,2), and in survival analyses, we systematically took body surface area into account. Second, we aimed to investigate the natural history of LG/LF aortic stenosis by allowing a medical management period of at least 3 months after diagnosis (15). In our opinion, this approach reduced surgery-related biases by eliminating patients with severe aortic stenosis who, because of severe symptoms, were operated on immediately after diagnosis. Third, older retrospective series of LG/LF aortic stenosis did not report the symptomatic status of the patients (3), whereas more recent series included either asymptomatic individuals (4,14) or patients with severe symptoms (22). Our study included the entire symptomatic spectrum of aortic stenosis, and the similar outcomes for LG/LF aortic stenosis and mild-to-moderate aortic stenosis observed in the overall population were confirmed in the subgroup of patients who were asymptomatic at baseline. Fourth, in contrast to previous studies that used composite endpoints (4,14), our study focused on all-cause death and analyzed AVR with a time-dependent methodology. We believe that AVR is not an appropriate endpoint in this type of analysis, because it is mainly the result of the physician’s assessment of the severity of the disease. For all survival analyses, we took into account age and comorbidities that significantly affected outcome and played major roles in the decision for surgery. Finally, aortic stenosis severity might be underestimated when the right parasternal window is not used for recording the continuous Doppler flow across the aortic valve. On the other hand, underestimation of the diameter of the left ventricular outflow tract might lead to an overestimation of the severity of the disease (20). In our study, we systemically performed careful measurements of the outflow tract diameter in zoomed parasternal views and multiple acoustic windows for continuous-wave Doppler, including the right parasternal window. Nevertheless, despite rigorous echocardiographic evaluation and use of currently accepted definitions for LG/LF severe aortic stenosis, inherent errors in measurement might have led to some misclassification of mild-to-moderate aortic stenosis into LG/LF severe aortic stenosis.
Our survival data were in accordance with recent anatomical and morphological data (16,23,24). In a population of patients with severe aortic stenosis and preserved EFs, magnetic resonance imaging showed larger AVAs, a lower degree of ventricular hypertrophy, and similar focal fibrosis in the LG/LF subgroup compared with the HG aortic stenosis subgroup (15). Moreover, compared with HG aortic stenosis, the aortic valves of patients with LG/LF aortic stenosis had lower weights (23) and showed less severe calcification (24).
With regard to surgery, previous series (5,7,14) suggested that LG/LF aortic stenosis has a pejorative prognosis when treated conservatively, and that AVR significantly improves the outcome. In our series, although AVR was associated overall with marked mortality reduction in the population with severe aortic stenosis, we did not observe any significant beneficial effect in the subset with LG/LF aortic stenosis.
Strengths and limitations
Our study had the inherent limitations of retrospective analyses. However, cardiologists with expertise in valvular disease performed diagnosis and follow-up, and surgical decisions were made by the heart team with the approval of the patients’ physicians, in accordance with current practice guidelines. The specific indications for surgery during follow-up were not collected in our database. Our study was not a prevalence study, because patients were recruited at the echocardiography laboratory, and we excluded patients who were operated on during the first 3 months after diagnosis. We acknowledge that the relatively small number of patients with LF/LG aortic stenosis represented a limitation, and that the relationship between surgery and outcome in the LG subgroups should be interpreted with caution. The multivariable model that investigated the effect of surgery in LG/LF aortic stenosis might be overfitted, and the result should be viewed as exploratory. Further studies with larger numbers of LG/LF aortic stenosis patients are needed to definitively establish the impact of surgery in LG/LF aortic stenosis. Censoring the patients who were initially managed conservatively at the time of surgery did not account for the benefit and/or risk of waiting. Despite systematic adjustment for CAD in multivariable analyses, CAD severity and revascularization status might increase the heterogeneity of the population and affect outcomes. Finally, this study did not use hemodynamic parameters, which might have allowed more precise classification of the flow-gradient patterns.
Across the spectrum of aortic stenosis severity, LG/LF aortic stenosis does not represent the final stage of the disease, and has a more favorable outcome compared with HG aortic stenosis. In our series, despite the management type and symptomatic status, the mortality risk of LG/LF aortic stenosis was similar to that of mild-to-moderate aortic stenosis and did not appear to be significantly improved by AVR. According to current guidelines (25), the surgical decision in LG/LF aortic stenosis should be discussed only for symptomatic patients. Such decisions should be made cautiously on a case-by-case basis when there is evidence that symptoms are most likely due to valve stenosis (1,25) and after documenting the severity of the stenosis and significant valvular calcifications. However, symptoms are often difficult to interpret, because these patients are elderly, and have CAD, atrial fibrillation, and important comorbidities. Measuring brain natriuretic peptide levels may be helpful in difficult cases. Because of the intense debate about the indication for AVR in patients with LG/LF aortic stenosis with preserved EF, the natural history and the impact of AVR in LG/LF aortic stenosis deserve future, carefully designed, large, prospective studies.
COMPETENCY IN MEDICAL KNOWLEDGE: Irrespective of management, LG/LF severe aortic stenosis with preserved EF is associated with outcomes similar to those of moderate aortic stenosis.
COMPETENCY IN MEDICAL CARE: Patients with asymptomatic severe LG/LF aortic stenosis with preserved EF must be followed carefully and closely for symptoms due to the valve lesions, and surgery must be discussed when these symptoms develop.
TRANSLATIONAL OUTLOOK: Prospective studies are needed to clarify the benefit of surgery for patients with LG/LF severe aortic stenosis with preserved EF and whether the natural history and approach to management vary for patients on the basis of specific flow-gradient patterns associated with valvular aortic stenosis.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- aortic valve area
- aortic valve replacement
- coronary artery disease
- ejection fraction
- mean Doppler gradient
- stroke volume
- Received May 20, 2014.
- Revision received August 17, 2014.
- Accepted September 8, 2014.
- American College of Cardiology Foundation
- Bonow R.O.,
- Carabello B.A.,
- Chatterjee K.,
- et al.
- Joint Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology (ESC),
- European Association for Cardio-Thoracic Surgery (EACTS),
- Vahanian A.,
- Alfieri O.,
- Andreotti F.,
- et al.
- Hachicha Z.,
- Dumesnil J.G.,
- Bogaty P.,
- et al.
- Lancellotti P.,
- Magne J.,
- Donal E.,
- et al.
- Clavel M.A.,
- Dumesnil J.G.,
- Capoulade R.,
- et al.
- Minners J.,
- Allgeier M.,
- Gohlke-Baerwolf C.,
- et al.
- Eleid M.F.,
- Sorajja P.,
- Michelena H.I.,
- et al.
- Dumesnil J.G.,
- Pibarot P.,
- Carabello B.
- Pibarot P.,
- Dumesnil J.G.
- Adda J.,
- Mielot C.,
- Giorgi R.,
- et al.
- Herrmann S.,
- Störk S.,
- Niemann M.,
- et al.
- Jander N.,
- Minners J.,
- Holme I.,
- et al.
- Maes F.,
- Boulif J.,
- Pierard S.,
- et al.
- Barone-Rochette G.,
- Piérard S.,
- Seldrum S.,
- et al.
- Venables W.N.,
- Ripley B.D.
- Lauten J.,
- Rost C.,
- Breithardt O.A.,
- et al.
- Minners J.,
- Allgeier M.,
- Gohlke-Baerwolf C.,
- et al.
- Ozkan A.,
- Hachamovitch R.,
- Kapadia S.R.,
- et al.
- Clavel M.A.,
- Cote N.,
- Mathieu P.,
- et al.
- Clavel M.A.,
- Messika-Zeitoun D.,
- Pibarot P.,
- et al.
- Nishimura R.A.,
- Otto C.M.,
- Bonow R.O.,
- et al.,
- American College of Cardiology/American Heart Association Task Force on Practice Guidelines