Ours was a retrospective study carried out at the Hospital Universitario Son Espases. Using our database of echocardiographic results EchoPAC® (General Electric Healthcare, Waukesha, WI, United States) we reviewed all tests performed between January 2007 and February 2010, selecting patients with severe AS (AVA<1cm2) and preserved EF (EF≥50%). Patients with a suboptimal test result, moderate-severe aortic insufficiency, or mitral valvulopathy were excluded from the study. The follow-up protocol involved personal communications in medical consultations, review of digital medical histories, medical reports, and telephone interviews. This study was approved by the hospital research committee.

The clinical information compiled included age, sex, tobacco use, body mass index, diagnosis of hypertension, diabetes mellitus, hypercholesterolemia, atrial fibrillation (AF), and coronary disease (defined as angina, myocardial infarction, angiographic evidence of coronary disease, percutaneous coronary intervention, and myocardial revascularization surgery).

The Doppler echocardiographic indexes of AS severity included maximum transvalvular velocity, maximum gradient, mean gradient, and aortic jet velocity-time integral. AVA was obtained using the standard continuity equation and was adjusted for body surface area. Systolic volume index (SVI) is the SV adjusted for body surface area and measured using Doppler in the left ventricular outflow tract.

Systemic vascular resistance, systemic arterial compliance, and valvuloarterial impedance were calculated using their respective formulas.5 In 16 patients, systolic blood pressure was not measured at the time of the echocardiogram. Left ventricle (LV) dimensions were measured in accordance with the guidelines of the American Society of Echocardiography (ASE) and adjusted for body surface area. Telediastolic and telesystolic volumes were calculated using the Teichholz volume method.15 LV mass and relative wall thickness were measured using the ASE-recommended formulas. Left ventricular ejection fraction (LVEF) was obtained in all patients using the Teichholz method and through visual estimation. Midwall fractional shortening and stroke work of the LV were calculated using standard formulas.15,16

A coronary angiogram was performed in 259 patients (71.3%). Coronary disease was defined as the presence of stenosis≥50% of the left main coronary artery and/or stenosis≥70% of the epicardial coronary arteries.

The cohort of patients was divided into 4 groups based on the presence of normal or reduced transaortic flow (SVI>35mL/m2 or<35 mL/m2) and high or low gradient (mean gradient>40mmHg or<40mmHg), in accordance with current clinical practice guidelines. The following groups were thus defined: group I: normal flow, high gradient; group II: normal flow, low gradient; group III: low flow, high gradient; group IV: low flow, low gradient.

The results are expressed as means and 95% confidence intervals (95%CI), median and quartiles, or percentages. Differences among patient groups were analyzed using ANOVA or Kruskal-Wallis tests for continuous variables, and chi-square tests or Fisher's exact tests for categorical variables. The primary end-point of this study was overall mortality. Survival was assessed using Kaplan-Meier curves for each group. Differences among groups in terms of risk factors were determined using log rank tests. The effects of clinical and echocardiographic variables on survival were determined using Cox proportional hazards models, including aortic valve replacement as a time-dependent variable. All possible associated variables and those that resulted in a P-value<.2 in the univariate analysis were included in the multivariate analysis. We used SPSS statistical software for Windows®, version 15.0, for all statistical analyses.

Our study showed that 27% of patients had low flow AS, which is in agreement with previous publications.4,6,13 In our sample, 56% of patients with low flow had a high gradient (group III), which confirms that low flow is not synonymous with low aortic gradient. Our results reaffirm the previous hypothesis4–10 that these patients with severe AS and low flow have a greater afterload, resulting from the combination of valvular obstruction and increased peripheral resistance, which would produce increased concentric ventricular remodelling and myocardial dysfunction.

As in past studies,14 we observed 2 clearly differentiated groups of low aortic flow; one with less severe cases of stenosis and low gradients (group IV) and another with a greater severity of valvular obstruction and high gradients (group III). In both groups, we observed a decrease in SV and a worse prognosis than in other patients with severe AS and normal flow. In this context, both low-flow groups had a higher prevalence of AF than did patients with normal flow rates. AF may reflect more advanced disease, given its predictive value in the multivariate analysis, in addition to the fact that the loss in atrial contraction can reduce ventricular filling and contribute to reducing cardiac output. Globally, patients with AF had a lower SVI (38 vs 43). In a previous study, AF was associated with decreased survival following surgery in patients with severe AS and ventricular dysfunction.17 In a more recent study,18 asymptomatic patients with severe AS and normal EF (and exclusion of patients with AF), those with low flow had a significantly worse prognosis during follow-up. In our study, AF acted as an independent risk factor for mortality.

Group IV patients (low flow and low gradient) who underwent surgery had a lower survival rate than group III patients undergoing surgery (low flow and high gradient). Group III patients had critical AS, and the decrease in contractile function and lower SV may be correlated with the severity of valvular obstruction, whereas the results in group IV (with less severe stenosis) suggest the coexistence of other factors in addition to valvular obstruction. In our study population, and similar to the study by Clavel et al.,19 coronary arterial disease was significantly more prevalent in this group. Group IV constituted a more heterogeneous group with greater comorbidity (age, diabetes, ischemic heart disease, AF...), with a tendency for a higher overall mortality rate, fewer indications for surgery, and inferior surgery results compared with the remaining groups. On a possibly related topic, Hermann et al.20 described a greater degree of myocardial fibrosis in the subendocardium and decreased longitudinal function (calculated using 2-dimensional strain) in patients with low flow and a low gradient, anomalies which were irreversible following the aortic valve replacement, and which were considered to be the cause of the worse postoperative results in these patients. Recently, in a prospective study, Adda et al.21 also observed a severe deterioration in longitudinal function of the left ventricle using 2-dimensional strain in patients with a low gradient and low flow.

In our study, 39% of patients had severe low gradient AS. Most of our patients with low gradients had normal aortic flow (70%; group II). These patients had a less severe degree of stenosis and reduced ventricular remodelling with better indexes of heart function, and consequently a better prognosis, than other patients from the remaining groups. In patients with severe AS and low gradient, 30% had low flow (group IV, 12% of all patients with AS) and had a worse prognosis than those with a low gradient and normal flow (group II) and had a similar prognosis to patients with high gradients. These findings are in agreement with those of Dumesnil et al.14 and differ from those of a study by Jander et al.,22 in which patients with severe AS, low gradient, and normal or low aortic flow had a similar prognosis to that of patients with moderate AS.In our study, as in previous publications, the surgery rate in patients with a low gradient was lower,10,14,22 possibly due to the clinical perception of less severe AS. Our results corroborate the finding that a low aortic gradient does not exclude the possibility of severe AS, as well as the benefits of surgery in these patients, regardless of the aortic gradient value.10,14,18,23-25

Our study shows that severe AS with low aortic flow results in a worse prognosis than severe AS with normal flow. Another important finding was the higher prevalence of AF in patients with low flow, which also has prognostic value.

This study confirms the benefits of valve replacement surgery in all symptomatic patients with severe stenosis (AVA<1cm2, according to current guidelines) regardless of aortic gradient. Importantly, most of the patients in this study had an AVA<0.8cm2, and only group II patients (low gradient and normal flow) tended to have an AVA>0.8cm2 (Table 2).

Group II had less severe stenosis and a better prognosis than the remaining groups in our study. Several publications on this topic have proposed a readjustment of the cut-off values for defining the severity of stenosis,13,22 suggesting that these patients in particular be classified as having moderate/severe stenosis.26 However, as our study has shown, this group also benefits from valve replacement surgery. This is a borderline group that is the subject of much debate. In these patients, careful clinical evaluation of symptoms and/or patient stratification using additional tests such as ergometry and aortic calcification score are essential.

The primary limitation of our study is its retrospective design and the lack of randomization for surgical treatment, although all patients were referred to surgery due to the presence of symptoms following a medical/surgical evaluation.

The patients that did not undergo surgery formed a heterogeneous group. Surgery was not considered in many of these patients due to the presence of comorbidities that may not have been properly reflected in our study (such as advanced age, quality of life, and cognitive deterioration). In other patients, the symptoms may have derived from a noncardiac origin, or the AS may have been considered nonsevere. Some patients also refused surgery when offered. Nevertheless, we believe that these limitations reflect the reality of daily clinical practice and the management of patients with severe AS in our environment.

Finally, we would like to highlight the importance of correctly measuring aortic gradient (which may be underestimated in group II and IV patients) and SVI, and therefore, of careful measurement of the left ventricular outflow tract and velocity-time integral using Doppler echocardiography,27 which is even more important in patients with AF, given the different R-R interval.28 The retrospective design of our study implies that the values may have been obtained with these potential errors. However, we believe that the difference in prognosis between groups II and IV would rule out the possibility that this is an artificial calculation based on improperly calculated SV.