Aortic valve replacement improves survival for patients with low-gradient aortic valve stenosis, but there is a paucity of data on postoperative quality of life for this population.
In a single-center retrospective analysis of 304 patients with severe aortic valve stenosis who underwent transcatheter aortic valve replacement, patients were divided into 4 groups based on mean pressure gradient, left ventricular ejection fraction, and stroke volume index. Using the Kansas City Cardiomyopathy Questionnaire-12, quality of life was assessed immediately before and 1 month after transcatheter aortic valve replacement.
Most patients in the low-flow, low-gradient group were men; this group had higher relative rates of cardiovascular disease and type 2 diabetes than the paradoxical low-flow, low-gradient group; the normal-flow, low-gradient group; and the high-gradient group. All-cause mortality did not differ significantly among the groups at 1 month after surgery, and all groups experienced a significant improvement in quality-of-life scores after surgery. The mean improvement was 27 points in the low-flow, low-gradient group, 25 points in the paradoxical low-flow, low-gradient group, 30 points in the normal-flow, low-gradient group, and 30 points in the high-gradient group (all P < .001).
Quality of life improves significantly across all subgroups of aortic valve stenosis after trans-catheter aortic valve replacement, regardless of flow characteristics or aortic valve gradients.
Abbreviations and Acronyms
aortic valve stenosis
aortic valve area
body mass index
coronary artery bypass graft
high-gradient aortic valve stenosis
Kansas City Cardiomyopathy Questionnaire-12
low-flow,low-gradient aortic valve stenosis
length of stay
left ventricular ejection fraction
mean pressure gradient
normal-flow, low-gradient aortic valve stenosis
New York Heart Association
peripheral arterial disease
percutaneous coronary intervention
paradoxical low-flow, low-gradient aortic valve stenosis
quality of life
Society of Thoracic Surgeons
transcatheter aortic valve replacement
University of California, Los Angeles
Aortic valve stenosis (AS) is the most common degenerative valvular disease in developed countries and is becoming a growing healthcare burden as the population ages. The prevalence of AS steadily increases with age, with 1 estimate noting a prevalence of 0.2% in individuals 50 to 59 years of age, increasing to 9.8% in those 80 to 89 years of age.1 High-gradient AS (HG-AS)—defined as an aortic valve area (AVA) of 1.0 cm2 or less and a mean pressure gradient (MPG) of 40 mm Hg or higher—constitutes the majority of AS.2 Although it has been well established that aortic valve replacement prolongs survival in patients with symptomatic HG-AS, the approach to low-gradient AS is less-clearly defined.3,4
Patients with low-gradient, severe AS—defined as an AVA of 1.0 cm2 or less and an MPG of less than 40 mm Hg—present with a complex set of physiologic findings. Although one of the most common etiologies of the low-flow state is reduced left ventricular ejection fraction (LVEF), other pathologic findings—such as diastolic dysfunction, other valvular disease, pulmonary hypertension, right ventricular failure, and rhythm disturbances—may also lead to low-flow states. This heterogeneous group of patients with low-gradient AS is divided into those with low-flow, low-gradient AS (LFLG-AS; LVEF <50%), paradoxical low-flow, low-gradient AS (pLFLG-AS; LVEF ≥50%, stroke volume index [Svi] <35 mL/m2), or normal-flow, low-gradient AS (NFLG-AS; LVEF ≥50%, Svi ≥35 mL/m2). The presence of a low gradient often leads to discrepancies in valvular evaluation, with uncertainty about the severity of stenosis and the perceived benefits of intervention.
Patients with low-gradient AS, like their peers with high gradients, have a dismal prognosis with conservative management. However, their estimated 2-year survival rate of 50% increases to 80% with aortic valve replacement.5 Previous studies have evaluated the mortality benefits of transcatheter aortic valve replacement (TAVR) for patients with low-gradient AS,6–9 highlighting improved event-free survival compared with surgical aortic valve replacement, and postoperative outcomes that are comparable to those of patients with HG-AS.10–12 However, there remains a paucity of data on quality of life (QOL) after valve replacement in this population. This study aims to compare post-TAVR outcomes among patients with LFLG-AS, pLFLG-AS, NFLG-AS, and HG-AS to help inform appropriate clinical decision-making for patients with different subgroups of AS.
Patients and Methods
This study was approved by the institutional review board of the University of California, Los Angeles (UCLA). All investigations were performed in compliance with human-studies guidelines for this institution and with the guidelines of the US Food and Drug Administration.
Records were obtained from a database of patients with severe AS who underwent TAVR at UCLA Ronald Reagan Medical Center between January 2016 and December 2018. Patients with a prior TAVR or surgical aortic valve replacement were excluded, as were those for whom any essential information was not available, including QOL scores (n = 16), identifying information (n = 1), or AS gradient (n = 1). Subgroups were created based on the findings of the most recent pre-TAVR transthoracic echocardiogram.
Patients were assessed at a preoperative visit, at 1 month after surgery, and at 1 year after surgery. The Kansas City Cardiomyopathy Questionnaire-12 (KCCQ-12) was used to assess QOL across 4 domains (QOL, social limitations, physical limitations, and symptom frequency). Each domain was scored on a scale of 0 (worst) to 100 (best), and the total combined KCCQ-12 score was also scaled from 0 to 100. Trans thoracic echocardiography was performed before TAVR and at 1 month after surgery.
Continuous parametric variables were expressed as the mean (SD), and categorical variables were expressed as the number and percentage. Statistical differences between groups were analyzed using analysis of variance, and post hoc analysis of between-group comparisons was performed using a Bonferroni test. Categorical variables were compared using the χ2 test unless there were very few numbers, in which case they were compared using Fisher exact test. Comparison of pre- and post-TAVR variables—including KCCQ-12 scores, LVEF, and New York Heart Association (NYHA) classification—was done using a paired t test. Comparison of NYHA class pre- and post-TAVR was done using the McNemar-Bowker test. A 2-sided P < .05 was considered statistically significant. Data were analyzed using SPSS software, version 25.0 (IBM).
Of the 304 patients who met the study inclusion criteria, 49 had LFLG-AS, 80 had pLFLG-AS, 45 had NFLG-AS, and 130 had HG-AS. The AS subgroups were similar with respect to patient age, body mass index, presence of hypertension, smoking status, pacemaker history, and stroke history. The LFLG-AS group had a smaller proportion of female patients (27%) than did the pLFLG-AS group (49%) and the HG-AS group (53%) (P < .001; Table I). Women represented 67% of the NFLG-AS group. Ischemic disease was more prominent in patients with LFLG-AS, who had a higher prevalence of prior percutaneous coronary intervention (45%; P = .011) and prior coronary artery bypass graft (31%; P = .002). The LFLG-AS group had a higher relative prevalence of diabetes mellitus in particular and more comorbidities in general, as suggested by a significantly higher Society of Thoracic Surgeons (STS) short-term cardiac surgery risk score (9.1 [5.5]; P < .001) compared with the other 3 groups.
Transthoracic echocardiography was performed immediately before and 1 month after TAVR. At baseline, the mean (SD) AVA had significant variation and was lowest in the HG-AS group at 0.65 (0.18) (P < .001; Table II). As expected, the high-gradient group had the largest mean (SD) gradient at 51 (9.4) (P < .001), and the lower-gradient subgroups had similar mean (SD) gradients (LFLG-AS, 27 [7.8]; pLFLG-AS, 28 [6.6]; NFLG-AS, 31 [6.4]). The LFLG-AS and HG-AS groups showed improvements in the mean (SD) LVEF after TAVR (LFLG-AS, 5% [10.1%]; P = .002; HG-AS, 5% [11.5%]; P < .001), but the LVEF in the pLFLG-AS and NFLG-AS groups remained unchanged. As expected, the pLFLG-AS group had a mean (SD) Svi of 28 (5), and the NFLG-AS group had a mean (SD) Svi of 43 (6).
Baseline overall KCCQ-12 scores were similar across all 4 AS categories, and there were significant improvements in each group at 1 month. The mean (SD) improvements were 27 (31) points in the LFLG-AS group, 25 (24) points in the pLFLG-AS group, 30 (28) points in the NFLG-AS group, and 30 (28) points in the HG-AS group (P < .001; Table III). In addition to the clear improvement in overall KCCQ-12 scores, each of the 4 AS subgroups showed substantial improvement at 1 month in almost all individual QOL domains.
Of the 263 patients who completed the KCCQ-12 at 1 month, 197 (75%) were subsequently reassessed at 1 year. The remaining 66 patients (25%) were either lost to follow up, had limited follow up, or had died by the time of the 1-year appointment. For the patients who completed KCCQ-12 testing at 1 year, the mean overall scores for all 4 AS subgroups were similar to those obtained at the 1-month assessment. The mean (SD) 1-year KCCQ-12 scores for each group were: LFLG-AS, 78 (23); pLFLG-AS, 81 (23); NFLG-AS, 86 (14); and HG-AS, 86 (18) (Supplementary Table I).
The NYHA classification improved after TAVR for all AS subgroups (P < .001; Fig. 1). At pre-TAVR baseline, most patients were classified as class II and class III (Table IV). Consistent with the KCCQ-12 data, most patients improved postoperatively, falling mostly into class I and class II. Only 2 patients were categorized as class IV after TAVR, compared with 18 before TAVR.
The patients in the NFLG-AS group tended to have the shortest hospital stays, with a mean (SD) length of stay of 3.0 (3.4) days compared with 6.4 (6.0) days for the LFLG-AS group and 7.9 (14.3) days for the pLFLG-AS group (P = .007).
At the 1-year mark, there remained no difference in mortality across the 4 groups (LFLG-AS, 14.0%; pLFLG-AS, 11.4%; NFLG-AS, 6.8%; and HG-AS, 10.2%; P = .674).
Understanding the benefits of TAVR within various AS subgroups is of critical importance. The prevalence of AS is substantially higher in older adult patients, many of whom have comorbid cardiovascular disease and poor functional status. Although post-TAVR mortality benefit has been well studied, the effect of TAVR on QOL remains largely unexplored.4,11,13 Not only has there been controversy about the diagnosis of true severe AS across the physiologic spectrum, the data for QOL and how it is affected by TAVR has not been well studied. In this analysis, changes in QOL across the spectrum of hemodynamic presentations of AS were studied based on AS flow states and gradients.
The baseline KCCQ-12 scores were similar across all 4 subgroups of AS, and TAVR resulted in a statistically significant improvement in overall KCCQ-12 scores, regardless of AS subgroup. There was substantial improvement across all 4 domains of the questionnaire, including the “QOL” and “social limitations” categories. In this population, the KCCQ-12 reveals important social benefits that may not be fully captured by other functional assessments.
Although there is a paucity of data on QOL outcomes for AS subgroups, a recent study did compare KCCQ-12 overall scores after TAVR in patients with low-flow and normal-flow AS and found improved QOL scores at 1 month.14 Previous work has defined poor outcomes after TAVR as either death or a decrease in KCCQ-12 score by 10 points or more.15 By this metric, the mortality data and increases in KCCQ-12 scores seen in the present study further confirm that TAVR yields meaningful benefit for patients in all 4 AS subgroups. The present study also showed an overall improvement in NYHA classification, affirming the observed benefit of TAVR for physical capacity and functional status. This shift was striking: only a small fraction of patients remained in class IV after TAVR, with the overwhelming majority falling into class I or II. The available data suggest that improvements in KCCQ-12 may be preserved at the 1-year mark.
There were no noted differences in 1-month or 1-year mortality among the 4 groups. These results are consistent with those of previous studies that showed similar 30-day mortality among AS subgroups.12,16–19 Researchers have shown that 1-year mortality for pLFLG-AS and LFLG-AS is significantly higher than for HG-AS, but the present analysis may be underpowered to detect this difference.11,20,21
The presence of both normal flow and a low gradient in the NFLG-AS group is seemingly contradictory; however, this finding is purportedly a product of discrepancies in the criteria for severe AS. One study of 333 cardiac catheterizations for severe AS found that using an AVA cutoff of 1 cm2 may correspond more closely to an MPG of 30 mm Hg rather than the standard 40 mm Hg.22 Of note, the NFLG-AS group in the present study had the largest mean AVA and MPG of the 3 lower-gradient subgroups. This group also had generally favorable outcomes, a finding that is consistent with those of prior studies showing that these patients have less advanced disease and better survival.2,9,23,24
Ultimately, the decision to proceed with TAVR for patients in the various subgroups of severe AS should be primarily based on underlying physiology and surgical risk. However, when the risk profile or mortality benefit is equivocal, expected QOL improvement can help inform the decision to proceed with TAVR.
This study is based on data from a single center, which may limit the generalizability of the conclusions to the general population. The analysis is retrospective, non-randomized, and does not include comparisons to patients managed with surgery or conservative therapy. Because of the nature of the intervention and the absence of a true control group (which would involve a sham procedure), some degree of placebo effect cannot be ruled out.
Regardless, this retrospective analysis shows a clinically significant QOL benefit for patients in all severe AS subgroups who are managed with TAVR. However, QOL information at 1 year is missing for some of the patients in this study.
Future work should expand on this analysis and include a larger number of patients from multiple sites. Extending the follow-up QOL assessment to 1 year or longer will reveal information about the durability of the initial QOL improvement. As the TAVR procedure itself improves in terms of safety, technical ability, and device technology, repeat analysis may reveal even more substantial benefit.
Patients with severe AS experience substantially improved QOL after TAVR, as assessed by the KCCQ-12 score, regardless of their baseline valve gradient.
Conflict of Interest Disclosure: Dr Murray H. Kwon and Dr Richard J. Shemin are paid consultants of Edwards Lifesciences.