Abstract
Background
There are limited and conflicting data on the long-term outcomes in normal-flow low-gradient (NFLG) and paradoxical low-flow low-gradient (pLFLG) aortic stenosis (AS) following transcatheter aortic valve replacement (TAVR).
Methods
Patients who underwent TAVR for severe symptomatic native AS with preserved left ventricular ejection fraction (≥50%) from 2012 to 2022 at our center were included. The cohort was divided into 3 different hemodynamic AS groups: high-gradient aortic stenosis (HGAS) (aortic valve mean gradient [AVMG] ≥40 mm Hg, aortic valve [AV] peak velocity ≥4 m/s, and AV area ≤1 cm2), NFLG (stroke volume index ≥35 mL/m2, AVMG <40 mm Hg, AV area ≤1 cm2), and pLFLG (stroke volume index <35 mL/m2, AVMG <40 mm Hg, AV area ≤1 cm2). The primary outcome was all-cause mortality at 1 and 5 years.
Results
Totally, 1683 patients (929 HGAS, 344 pLFLG, and 410 NFLG) were included; median age was 82.8 (77.2-87.2) years. At 1 year, mortality was higher among patients with pLFLG AS (15.5%) compared with HGAS (10.8%, P = .05). Although NFLG AS had a higher mortality than HGAS at 1 year, this did not reach statistical significance (15.9% vs 10.8%, respectively; P = .06). There was no significant difference in survival between the 3 groups at 5 years post-TAVR.
Conclusions
In patients who undergo TAVR, those with a pLFLG pattern of AS have a worse 1-year survival, but similar 5-year survival when compared with traditional HGAS AS. There was no significant difference in survival in patients with NFLG AS when compared with HGAS at 1 and 5 years post-TAVR.
Keywords: high-gradient aortic stenosis, normal-flow low-gradient aortic stenosis, paradoxical low-flow low-gradient aortic stenosis, transcatheter aortic valve replacement
Introduction
Aortic stenosis (AS) remains one of the most prevalent valvular heart diseases globally, affecting up to 5% of adults over the age of 70 years.1 Transcatheter aortic valve replacement (TAVR) has evolved as a standard of care treatment option in patients with severe AS, classically defined as an aortic valve area (AVA) ≤1 cm2, mean transvalvular pressure gradient ≥40 mm Hg, and peak aortic jet velocity >4 m/s.2, 3, 4 However, up to 50% of patients with severe AS can present with discordant echocardiographic parameters, where the mean transvalvular pressure gradient <40 mm Hg and/or aortic jet velocity <4 m/s with an AVA ≤1 cm2.5 Because of these variations, severe AS is categorized into several subgroups based on multiple hemodynamic characteristics of the aortic valve. Patients with severe AS are divided into high-gradient aortic stenosis (HGAS) (mean transvalvular pressure gradient ≥40 mm Hg) or low-gradient AS (mean transvalvular pressure gradient <40 mm Hg). The latter group is further delineated based on left ventricular ejection fraction (LVEF) and transvalvular flow (stroke volume index [SVi]) into classical low-flow low-gradient AS (LVEF <50%), paradoxical low-flow low-gradient (pLFLG) AS (SVi <35 mL/m2, LVEF >50%), and normal-flow low-gradient (NFLG) AS (SVi >35 mL/m2, LVEF >50%).6
Patients with pLFLG AS represent a subset of patients who are more likely to be women, have more cardiac comorbidities, and are characterized by smaller left ventricular size, along with impaired diastolic filling and there are conflicting results about the survival outcomes of these patients after TAVR compared with the other subgroups of AS.7, 8, 9, 10, 11, 12, 13 Contrary to pLFLG AS, NFLG AS is less well understood, and there is controversy about whether this entity represents actual severe AS or moderate AS.11,14 It has been described that this subgroup may represent up to 15% to 40% of all AS patients, and shows increased mortality compared with moderate AS, and shares similar clinical outcomes as classical HGAS following TAVR.15 Depending on the subgroup, severe symptomatic AS carries a high mortality rate if left untreated, with the 5-year mortality rates ranging from 49% to 73%, underlining the importance of recognition and treatment.5
The 2020 ACC/AHA guidelines recommend aortic valve intervention for any severe symptomatic AS regardless of subtype and recognize HGAS (stage D1) and pLFLG AS (stage D3) as severe AS entities. However, NFLG AS remains in the gray area without specific guidelines and recommendations by the ACC/AHA and is often left to the physician’s discretion.6 There is divided literature regarding the degree of mortality benefit of TAVR in pLFLG AS and a lack of contemporary data for NFLG AS. Therefore, this study aimed to compare long-term clinical outcomes in patients with HGAS, pLFLG AS, and NFLG AS following TAVR at our institution.
Methods
Study design
This study was a retrospective data analysis on patients who underwent a TAVR procedure at our institution from January 2012 through August 2022. The selection of this period allowed for a 10-year study period with 1- and 5-year clinical outcome data analysis. The decision for these patients to undergo TAVR was made through a collaborative heart team approach involving interventional cardiologists, cardiovascular surgeons, and general cardiologists in our cardiovascular valve clinics. The patients included in this study were those with clinically significant symptomatic severe native AS as determined by the heart team. This study was approved by the Advocate Health Institutional Review Board. A waiver of consent and HIPAA was obtained because of the retrospective nature of this study.
Study groups
The cohort was screened to exclude those with an LVEF <50% and AVA >1 cm2. The remaining patients were separated into 1 of 3 hemodynamic groups of severe AS as follows: (1) HGAS—aortic valve (AV) mean gradient ≥40 mm Hg, AV peak velocity ≥4 m/s; (2) pLFLG AS—AV mean gradient (AVMG) <40 mm Hg and SVi <35 mL/m2; and (3) NFLG AS—AVMG <40 mm Hg and SVi ≥35 mL/m2. The study flowchart is depicted in Figure 1.
Figure 1.
Flowchart of study design. A detailed flowchart of the study design, including inclusion and exclusion criteria. AS, aortic stenosis; HGAS, high-gradient aortic stenosis; LVEF, left ventricular ejection fraction; MG, mean gradient; NFLG, normal-flow low-gradient; pLFLG, paradoxical low-flow low-gradient; SVi, stroke volume index; TAVR, transcatheter aortic valve replacement.
Study outcomes
The primary clinical end point was all-cause death at 1 and 5 years following TAVR. The secondary clinical end point was major adverse cardiac and cerebrovascular events (MACCE), defined as a composite end point of the following: (1) hospitalization for congestive heart failure, (2) hospitalization for myocardial infarction, and (3) hospitalization for transient ischemic attack (TIA) or stroke at 1 and 5 years. Additionally, the individual components of MACCE and echocardiographic hemodynamic parameters were evaluated at 1 and 5 years following TAVR. Follow-up data were obtained through review of the electronic medical record, and mortality status was supplemented with queries of the Social Security Death Index. Discrepant data were reviewed by 2 independent investigators and adjudicated by a third when necessary.
Statistical analysis
Differences between groups were assessed with χ2 tests, Fisher exact tests for categorical variables, Kruskal-Wallis tests, and Wilcoxon rank sum tests for continuous variables. The Dwass, Steel, Critchlow-Fligner correction was applied exclusively to pairwise comparisons of baseline continuous variables following a significant Kruskal-Wallis result. No type 1 error control was applied to the primary or secondary outcome analyses. Time-to-event analyses of outcomes were conducted using Kaplan-Meier curves and log-rank tests. Multivariate Cox proportional hazards regression, adjusting for age, sex, and Society of Thoracic Surgeons (STS) risk score, was utilized to confirm univariate results. Presented P values for time-to-event outcomes are adjusted. A P value <.05 was considered statistically significant without control for type 1 error. SAS version 9.4 (SAS Institute) was used for all analyses. The STS risk score was chosen as a composite covariate because it incorporates multiple clinical and comorbidity parameters, which include renal, pulmonary, and cardiovascular risk, thereby minimizing collinearity and overfitting within the model.
Results
Patient characteristics
A total of 1683 patients (929 HGAS [55.2%], 344 pLFLG [20.4%], and 410 NFLG [24.4%]) were included in this study. The median age of the entire cohort was 82.8 years (77.2-87.2) and did not significantly differ between the groups. The pLFLG cohort had the lowest percentage of women at 45.4% compared with the HGAS (54%; P < .01) and NFLG (58.3%; P < .01). The most common comorbidities were hypertension (HGAS 90.1%, pLFLG 92.2%, and NFLG 92.2%; P = .33), diabetes mellitus (HGAS 37.6%, pLFLG 43.9%, and NFLG 35.6%; P = .05), prior percutaneous coronary intervention/coronary artery bypass grafting (HGAS 34.9%, pLFLG 43.6%, and NFLG 40%; P = .01), and atrial fibrillation (HGAS 24.9%, pLFLG 47.7%, and NFLG 29.5%; P < .01). The pLFLG also had higher STS risk scores than HGAS (3.7 vs 3.4; P = .03). The BMI was the highest in the pLFLG (29.9 kg/m2) compared with HGAS (28.8 kg/m2; P = .03) and NFLG (27.5; P < .01). The HGAS had the highest aortic calcium score at 2617 Agatston units (AU), followed by PFLG at 1903 AU, and NFLG had the lowest at 1549 AU (P < .01 in all). Lastly, the pLFLG group had the highest percentage of left main (≥50%) and proximal left anterior descending (≥70%) coronary artery disease (P < .01 and P = .01, respectively). Overall, the pLFLG had the highest burden of cardiovascular comorbidities compared with the other 2 groups. Detailed patient characteristics and comorbidities are presented in Table 1.
Table 1.
Patient characteristics.
| Characteristics | HGAS (n = 929) | pLFLG (n = 344) | NFLG (n = 410) | Omnibus P value | HGAS vs pLFLG | HGAS vs NFLG | pLFLG vs NFLG |
|---|---|---|---|---|---|---|---|
| Demographic characteristics | |||||||
| Age, y | 82.3 (76.6-87.3) | 82.4 (76.4-87.3) | 83.4 (78.7-87.1) | .09 | – | – | – |
| Female sex | 502 (54) | 156 (45.4) | 239 (58.3) | <.01 | <.01 | .15 | <.01 |
| Race—White | 896 (96.4) | 339 (98.6) | 402 (98) | .07 | – | – | – |
| Race—Black or African American | 25 (2.7) | 5 (1.4) | 6 (1.5) | – | – | – | – |
| Race—other | 8 (0.9) | 0 | 2 (0.5) | – | – | – | – |
| Hispanic or Latino ethnicity | 20 (2.2) | 6 (1.7) | 11 (2.7) | .67 | – | – | – |
| Baseline comorbidities | |||||||
| Permanent pacemaker | 72 (7.8) | 50 (14.5) | 47 (11.5) | <.01 | <.01 | .03 | .21 |
| Previous ICD | 12 (1.3) | 5 (1.4) | 8 (2) | .65 | |||
| Prior PCI or CABG | 324 (34.9) | 150 (43.6) | 164 (40) | .01 | <.01 | .07 | .32 |
| Prior MI | 108 (11.6) | 49 (14.2) | 57 (13.9) | .33 | – | – | – |
| Prior stroke/TIA | 126 (13.6) | 56 (16.3) | 74 (18) | .09 | – | – | – |
| Peripheral arterial disease | 169 (18.2) | 78 (22.7) | 87 (21.2) | .15 | – | – | – |
| Current/recent smoker (within 1 y) | 41 (4.4) | 11 (3.2) | 16 (3.9) | .61 | – | – | – |
| Hypertension | 837 (90.1) | 317 (92.2) | 378 (92.2) | .33 | – | – | – |
| Diabetes mellitus | 349 (37.6) | 151 (43.9) | 146 (35.6) | .05 | .04 | .49 | .02 |
| End-stage renal disease | 32 (4) | 8 (2.8) | 8 (2.2) | .26 | – | – | – |
| Chronic lung disease | 89 (9.6) | 33 (9.6) | 63 (15.4) | <.01 | .99 | <.01 | .02 |
| Immunocompromise present | 61 (7.6) | 24 (8.4) | 37 (10.3) | .30 | – | – | – |
| Atrial fibrillation/flutter | 231 (24.9) | 164 (47.7) | 121 (29.5) | <.01 | <.01 | .08 | <.01 |
| BMI, kg/m2 | 28.8 (24.9-33.9) | 29.9 (25.9-35) | 27.5 (23.7-31.8) | <.01 | <.01 | .03 | <.01 |
| AV replacement—STS risk score | 3.4 (2.1-5.5) | 3.7 (2.4-5.7) | 3.6 (2.4-6) | .02 | .03 | .15 | .91 |
| AV calcium score | 2617 (1751-2788) | 1903 (1197-3019) | 1549 (1040-2478) | <.01 | <.01 | <.01 | <.01 |
| NYHA class within 2 wk | |||||||
| Class II | 575 (61.9) | 202 (58.7) | 258 (62.9) | .19 | – | – | – |
| Class III | 318 (34.2) | 132 (38.4) | 139 (33.9) | – | – | – | – |
| Class IV | 36 (3.9) | 10 (2.9) | 13 (3.2) | – | – | – | – |
| Preoprocedure labs and biomarkers | |||||||
| Pre-procedure hemoglobin, g/dL | 12.1 (10.7-13.2) | 12.4 (10.8-13.4) | 12.1 (10.7-13.3) | .14 | – | – | – |
| Pre-procedure creatinine, mg/dL | 1 (0.8-1.2) | 1 (0.8-1.3) | 1 (0.8-1.3) | .19 | – | – | – |
| Preprocedure angiogram | |||||||
| Left main ≥50% | 53 (8.6) | 35 (16.5) | 17 (5.8) | <.01 | <.01 | .13 | <.01 |
| Prox LAD ≥70% | 96 (15.8) | 52 (24.5) | 58 (19.7) | <.01 | <.01 | .12 | .20 |
| Preprocedure echocardiographic evaluation | |||||||
| Left ventricle ejection fraction, % | 65 (60-69) | 60 (55-65) | 63 (59-67) | <.01 | .02 | <.01 | <.01 |
| Stroke volume index, mL/m2 | 41 (34.5-47.5) | 29.4 (25.8-32.3) | 41.9 (38.1-47.1) | <.01 | <.01 | <.01 | <.01 |
| AV morphology | |||||||
| Bicuspid | 40 (4.4) | 7 (2.1) | 5 (1.2) | – | – | – | – |
| Tricuspid | 871 (95.6) | 329 (97.9) | 395 (98.7) | – | – | – | – |
| AV disease—AV annulus size, mm | 24 (22-26) | 24 (23-26) | 23 (21-25) | <.01 | .03 | .04 | .01 |
Values are median (IQR) or n (%).
AV, aortic valve; CABG, coronary artery bypass grafting; HGAS, high-gradient aortic stenosis; ICD, implantable cardioverter-defibrillator; LAD, left anterior descending; MI, myocardial infarction; NFLG, normal-flow low-gradient; PCI, percutaneous coronary intervention; pLFLG, paradoxical low-flow low-gradient; STS, Society of Thoracic Surgeons; TIA, transient ischemic attack.
Procedural and adverse events
The procedural information and in-hospitalization adverse events are listed in Table 2. Most of the valves used were the Medtronic Evolut self-expanding valve, and the device success rate was >99% for each group, with femoral being the most common approach. The HGAS and NFLG had similar rates of new atrial fibrillation (4.4% vs 4.1%; P = .80), whereas pLFLG had the lowest rate at 1.1% compared with HGAS (P = .04) and NFLG (P = .06). There was no difference in new permanent pacemaker insertion, stroke, bleeding, or major vascular complications. The remaining adverse events were comparable in all 3 groups (Table 2).
Table 2.
Procedural information and in-hospital adverse events.
| HGAS (n = 929) | pLFLG (n = 344) | NFLG (n = 410) | Omnibus P value | HGAS vs pLFLG | HGAS vs NFLG | pLFLG vs NFLG | |
|---|---|---|---|---|---|---|---|
| Valve sheath access | |||||||
| Axillary | 2 (0.3) | 0 | 0 | – | – | – | – |
| Femoral | 599 (96.6) | 204 (95.8) | 281 (94.6) | – | – | – | – |
| Subclavian | 1 (0.2) | 0 | 3 (1) | – | – | – | – |
| Transaortic | 10 (1.6) | 7 (3.3) | 11 (3.7) | – | – | – | – |
| Transapical | 5 (0.8) | 1 (0.5) | 0 | – | – | – | – |
| Transcarotid | 2 (0.3) | 1 (0.5) | 2 (0.7) | – | – | – | – |
| Other | 1 (0.2) | 0 | 0 | – | – | – | – |
| Device used | |||||||
| SAPIEN (balloon-expanding) | 145 (15.6) | 47 (13.7) | 45 (11) | – | – | – | – |
| Evolut (self-expanding) | 779 (83.8) | 295 (85.8) | 365 (89) | – | – | – | – |
| Other | 5 (0.5) | 2 (0.6) | 0 | – | – | – | – |
| Device success rate, % | 99.6 | 99.1 | 99.5 | .62 | – | – | – |
| In-hospital adverse events | |||||||
| Coronary compression or obstruction | 1 (0.1) | 1 (0.3) | 0 | – | – | – | – |
| Cardiac arrest | 18 (1.9) | 6 (1.7) | 3 (0.7) | .26 | – | – | – |
| Atrial fibrillationa | 31 (4.4) | 2 (1.1) | 10 (4.1) | .11 | .04 | .84 | .06 |
| Annular dissection | 2 (0.2) | 0 | 0 | – | – | – | – |
| Aortic dissection | 2 (0.2) | 0 | 0 | – | – | – | – |
| Perforation with or without tamponade | 8 (0.9) | 0 | 1 (0.2) | – | – | – | – |
| Transient ischemic attack | 0 | 0 | 1 (0.2) | – | – | – | – |
| Ischemic stroke | 30 (3.2) | 13 (13.8) | 7 (1.7) | .20 | – | – | – |
| Hemorrhagic stroke | 1 (0.1) | 1 (0.3) | 0 | – | – | – | – |
| Undetermined stroke | 2 (0.2) | 0 | 2 (0.5) | – | – | – | – |
| Retroperitoneal bleeding | 4 (0.4) | 0 | 1 (0.2) | – | – | – | – |
| GI bleed | 3 (0.3) | 1 (0.3) | 4 (1) | – | – | – | – |
| GU bleed | 1 (0.1) | 1 (0.3) | 0 | – | – | – | – |
| Other bleed | 9 (1) | 1 (0.3) | 5 (12) | .01 | – | – | – |
| Device Embolization to aorta | 4 (0.4) | 1 (0.3) | 2 (0.5) | – | – | – | – |
| New requirement for dialysis | 7 (0.8) | 0 | 1 (0.2) | – | – | – | – |
| Unplanned vascular surgery/intervention | 35 (3.8) | 9 (2.6) | 18 (4.4) | .43 | – | – | – |
| PCI | 0 | 0 | 1 (0.2) | – | – | – | – |
| New pacermaker/ICDa | 119 (13.9) | 46 (15.6) | 46 (12.7) | .63 | – | – | – |
| Major vascular complications | 12 (1.3) | 0 | 0 | – | – | – | – |
| Myocardial infarction | 0 | 0 | 0 | – | – | – | – |
| Death in lab/operating room | 1 (0.1) | 0 | 1 (0.2) | – | – | – | – |
Values are n (%) unless noted.
GI, gastrointestinal; GU, genitourinary; HGAS, high-gradient aortic stenosis; ICD, implantable cardioverter-defibrillator; NFLG, normal-flow low-gradient; PCI, percutaneous coronary intervention; pLFLG, paradoxical low-flow low-gradient.
Excludes those with previous conditions.
Postprocedural assessment
The immediate pre- and post-TAVR assessment is shown in Figure 2. Following the TAVR procedure, all subgroups had improvement in AV area (HGAS: 0.7 cm2 to 2.2 cm2, pLFLG: 0.7 cm2 to 2.2 cm2, NFLG: 0.8 cm2 to 2.3; P < .05 in all). The AV peak velocity was reduced in all groups with TAVR (HGAS: 4.5 m/s to 2.1 m/s, pLFLG: 3.8 m/s to 1.9 m/s, NFLG: 3.9 m/s to 1.9 m/s; P < .05 in all). The mean transvalvular pressure gradient was reduced in HGAS (46 mm Hg to 8 mm Hg), pLFLG (31 mm Hg to 7 mm Hg), and NFLG (34 mm Hg to 7 mm Hg) (P < .05 in all). There was no statistical difference in the degree of aortic regurgitation between the 3 cohorts.
Figure 2.
Aortic valve hemodynamics pre- and immediate post-transcatheter aortic valve replacement. The aortic valve (AV) mean gradient (A), AV area (B), and AV peak velocity (C) of the different groups are shown before and after transcatheter aortic valve replacement. The asterisks represent statistical significance of the comparison. HGAS, high-gradient aortic stenosis; NFLG, normal-flow low-gradient; pLFLG, paradoxical low-flow low-gradient.
Clinical outcomes
At 1 year following TAVR, all-cause mortality was significantly higher in patients with pLFLG (15.5% [95% CI, 11.9-20.0]) compared with those with HGAS (10.8% [95% CI, 8.9-13.1; P = .05]), whereas mortality was similar to that observed in patients with NFLG AS (15.9% [95% CI, 12.5-20.0]; P = .87). Patients with NFLG also demonstrated higher mortality compared with HGAS, though this did not reach statistical significance (P = .06). By 5 years, although the pLFLG group demonstrated higher all-cause mortality (61% [95% CI, 54.0-68.1]) compared with both the HGAS group (56.4% [95% CI, 51.6-61.2]; P = .08) and NFLG group (53.5% [95% CI, 46.8-60.3]; P = .11), these differences were not statistically significant. There was no significant difference in mortality risk among the HGAS, pLFLG, and NFLG groups at 5 years after TAVR (Central Illustration).
Central Illustration.
Long-term clinical outcomes in different flow patterns of severe aortic stenosis. AS, aortic stenosis; HGAS, high-gradient aortic stenosis; NFLG, normal-flow low-gradient; pLFLG, paradoxical low-flow low-gradient; TAVR, transcatheter aortic valve replacement.
At 1 year, there was no significant difference in the incidence of MACCE among the 3 groups (HGAS 8.6%, pLFLG 10.4%, NFLG 11.5%; P > .05 in all comparisons). Similarly, at 5 years, MACCE rates remained comparable across the HGAS, pLFLG, and HGAS groups (P > .05 in all comparisons). The detailed outcomes at the 1 and 5-year follow-up for both all-cause mortality and MACCE end points are presented in Figure 3, along with Kaplan-Meier estimates across 10 years.
Figure 3.
Primary and secondary outcomes. Kaplan-Meier curves with event outcome rates at 1 and 5 years are shown for each of the 3 groups for all-cause mortality (A) and for the composite major adverse cardiac and cerebrovascular events (MACCE) end point (B). MACCE is defined as a composite end point of the following: (i) hospitalizations for congestive heart failure, (ii) myocardial infarction, and (iii) stroke/transient ischemic attack. Event rates are Kaplan-Meier estimates from log-rank tests. An additional analysis was then performed using a multivariable Cox proportional hazards regression, adjusting for age, sex, and Society of Thoracic Surgeons risk score. CI, confidence interval; HGAS, high-gradient aortic stenosis; HR, hazard ratio; NFLG, normal-flow low-gradient; pLFLG, paradoxical low-flow low-gradient.
The individual components of MACCE were analyzed at the follow-up study periods. Patients with pLFLG had a higher rate of CHF hospitalization compared with HGAS at 1 year (8.5% vs 5.1%; P < .01) and 5 years (22.4% vs 19.6%; P < .01) following TAVR. The pLFLG group had comparable risks of myocardial infarction and stroke/TIA compared with HGAS at both follow-up periods. Patients with NFLG had a significantly higher rate of myocardial infarction hospitalization at the 5-year follow-up (9% vs 3.3%; P < .01), whereas they had similar rates at 1 year (1.8% vs 0.6%; P = .07) compared with those with HGAS. The individual MACCE components are shown as Kaplan-Meier graphs in Figure 4.
Figure 4.
Major adverse cardiac and cerebrovascular event components. A Kaplan-Meier curve is shown for each of the 3 groups for stroke/transient ischemic attack (TIA) (A), myocardial infarction (MI) (B), and congestive heart failure (CHF) (C) through 10 years. CI, confidence interval; HGAS, high-gradient aortic stenosis; HR, hazard ratio; NFLG, normal-flow low-gradient; pLFLG, paradoxical low-flow low-gradient.
Predictors of mortality and MACCE
In univariable analysis, factors associated with increased all-cause mortality included pLFLG AS, age, BMI, AF, diabetes, prior stroke, and ESRD (P ≤ .05 in all). After adjustment for confounding in the multivariable model, only age, BMI, atrial fibrillation, diabetes, and ESRD remained independent predictors of mortality (P < .05 in all).
In an unadjusted analysis, atrial fibrillation was the only variable associated with a higher risk of MACCE. After multivariate adjustment, age and women, in addition to atrial fibrillation, were identified as significant predictors of MACCE (P < .05). The details are presented in Table 3.
Table 3.
Predictors of mortality and MACCE.
| Mortality |
MACCE |
||||||||
|---|---|---|---|---|---|---|---|---|---|
| Variable | Univariate |
Multivariate |
Variable | Univariate |
Multivariate |
||||
| HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | ||
| HGAS | Reference | Reference | Reference | Reference | HGAS | Reference | Reference | Reference | Reference |
| pLFLG | 1.1 (1.0-1.5) | <.05 | 1.1 (0.9-1.4) | .27 | pLFLG | 1.2 (0.9-1.6) | .16 | 1.1 (0.8-1.5) | .58 |
| NFLG | 1.1 (0.9-1.3) | .58 | 1.0 (0.9-1.3) | .72 | NFLG | 1.2 (0.9-1.6) | .19 | 1.1 (0.9-1.5) | .34 |
| Age | 1.0 (1.0-1.0) | .04 | 1.0 (1.0-1.0) | <.01 | Age | 1.0 (1.0-1.0) | .06 | 1.0 (1.0-1.0) | <.05 |
| Male | Reference | Reference | Reference | Reference | Male | Reference | Reference | Reference | Reference |
| Female | 1.0 (0.8-1.1) | .68 | 1.0 (0.9-1.2) | .69 | Female | 1.2 (1.0-1.6) | .06 | 1.3 (1.0-1.7) | .02 |
| BMI | 1.0 (1.0-1.0) | <.05 | 1.0 (1.0-1.0) | <.05 | BMI | 1.0 (1.0-1.0) | .68 | 1.0 (1.0-1.0) | .52 |
| Atrial fibrillation | 1.5 (1.3-1.8) | <.01 | 1.4 (1.2-1.7) | <.01 | Atrial fibrillation | 1.8 (1.4-2.2) | <.01 | 1.7 (1.4-2.2) | <.01 |
| Hypertension | 1.2 (0.9-1.7) | .19 | 1.1 (0.8-1.5) | .49 | Hypertension | 1.6 (0.9-2.5) | .08 | 1.4 (0.9-2.3) | .19 |
| Diabetes | 1.3 (1.2-1.6) | <.01 | 1.4 (1.2-1.7) | <.01 | Diabetes | 1.2 (0.9-1.5) | .18 | 1.2 (0.9-1.5) | .22 |
| PCI/CABG | 1.0 (0.9-1.2) | .81 | 1.0 (0.8-1.2) | .97 | PCI/CABG | 1.0 (0.8-1.3) | .90 | 1.0 (0.8-1.3) | .75 |
| Stroke/TIA | 1.2 (1.0-1.5) | <.05 | 1.2 (0.9-1.4) | .16 | Stroke/TIA | 1.2 (0.9-1.6) | .31 | 1.2 (0.8-1.5) | .45 |
| ESRD | 2.2 (1.5-3.1) | <.01 | 2.4 (1.6-3.5) | <.01 | ESRD | 1.4 (0.7-2.6) | .30 | 1.6 (0.8-3.0) | .17 |
Higher HR denotes higher mortality or MACCE.
BMI, body mass index; CABG, coronary artery bypass grafting; ESRD, end-stage renal disease HGAS, high-gradient aortic stenosis; MACCE, major adverse cardiac and cerebrovascular events; NFLG, normal-flow low-gradient; PCI, percutaneous coronary intervention; pLFLG, paradoxical low-flow low-gradient; TIA, transient ischemic attack.
Echocardiographic follow-up
Figure 5 shows echocardiographic hemodynamics between the 3 cohorts at 1 and 5 years following TAVR. At 1-year post-TAVR, the HGAS had a higher AVMG (7.8 mm Hg) and AV peak velocity (2.1 m/s) compared with pLFLG (6.8 mm Hg and 1.8 m/s; P = .002 and P = .001, respectively) and NFLG (6.6 mm Hg and 1.9 m/s; P = .003 and P = .02, respectively), with no differences between the pLFLG and NFLG groups (P = .96 and P = .70, respectively). At the 5-year follow-up, there was no statistically significant difference in AVMG (HGAS 6.4 mm Hg, pLFLG 9.4 mm Hg, and NFLG 6.5 mm Hg; P = .07) and AV peak velocity (HGAS 1.8 m/s, pLFLG 2.0 m/s, and NFLG 2.0 m/s; P = .33).
Figure 5.
Aortic valve hemodynamics 1 and 5 years post-transcatheter aortic valve replacement. AV, aortic valve; HGAS, high-gradient aortic stenosis; NFLG, normal-flow low-gradient; pLFG, paradoxical low-flow low-gradient.
Discussion
This study represents the largest real-world, single-center retrospective analysis evaluating the impact of TAVR in patients with severe symptomatic AS with preserved ejection fraction in different hemodynamic subtypes. Patients were stratified into 3 groups based on AVMG and transvalvular flow: HGAS, pLFLG, and NFLG. The major findings of this study are: (1) patients with pLFLG had higher 1-year mortality compared with HGAS, whereas no significant differences were observed among subtypes at 5 years; (2) there was no difference in MACCE (composite end point of CHF, myocardial infarction, and TIA/stroke hospitalizations) at either 1 or 5 years follow-up; and (3) patients with pLFLG had the highest burden of cardiovascular comorbidities at baseline, and experienced higher rates of heart failure admissions at both 1 and 5 years compared with HGAS. In our cohort, HGAS (55.2%) was the most common pattern of severe AS, followed by a similar prevalence of NFLG (24.4%) and pLFLG (20.4%).
Paradoxical low-flow low-gradient AS
Baseline clinical characteristics varied significantly among patients with different AS subtypes. In our cohort, the pLFLG group had a significantly lower percentage of women compared with HGAS, though still within the broader spectrum of reported literature.11 The higher proportion of men in this cohort may reflect alternative competing phenotypes causing low-flow in men, such as higher prevalence of coronary artery disease, subtle impairment of contractile reserve, or other comorbid conditions that lead to LV dysfunction, despite preserved ejection fraction. Referral patterns and patient selection may also have contributed, particularly if women with concomitant diastolic dysfunction were classified differently or excluded. Consistent with prior literature, the pLFLG group demonstrated a higher burden of comorbidities, including vascular disease, diabetes, atrial fibrillation, and STS risk scores, as well as lower aortic valve calcium scores compared with HGAS.12
Several studies have reported a survival benefit of TAVR with pLFLG AS, supporting its inclusion in the 2020 AHA/ACC guidelines.6 However, the degree of benefit compared with other AS subtypes remains controversial and conflicting. In our study, we found that patients with pLFLG had a higher 1-year mortality rate compared with those with HGAS. Patients with pLFLG AS likely experience higher 1-year mortality due to a combination of smaller LV cavity size with concentric remodeling, impaired diastolic function and vascular stiffness, and a greater burden of comorbidities and frailty, which may limit early hemodynamic recovery despite relief of valvular obstruction with TAVR. Over time, however, survival converges with HGAS as evidenced by our findings at 5 years. It is worth noting that the unadjusted 5-year mortality in the pLFLG was significant; however, the association was no longer statistically significant after multivariate adjustment (P = .08).
Wagener et al,8 after analyzing the multicenter SwissTAVI registry (N = 9541), similarly observed higher mortality in pLFLG compared with HGAS at both 1 and 5 years. It is important to note that their study defined low transvalvular flow using AVMG <40 mm Hg as opposed to using the SVi as in our study, which is in accordance with established ACC/AHA guidelines.6 This approach likely incorporated patients with normal-flow AS into their pLFLG cohort. Though our study did not observe a higher risk of mortality in the NFLG group, the inclusion of these patients in Wagener et al’s8 cohort may have amplified the overall risk of mortality signal in their group. This distinction underscores the heterogeneity of low-gradient subtypes and highlights the importance of using SVi to better delineate prognosis. Similarly, Osman et al13 reported higher all-cause mortality in pLFLG compared with HGAS as early as 30 days post-TAVR.
By contrast, other studies have found comparable clinical outcomes between pLFLG and HGAS. Ullah et al7 performed a comprehensive network meta-analysis including 17,292 patients. The study demonstrated similar 1-year major adverse cardiovascular events outcomes between the 2 groups, both the composite end point (all-cause mortality, nonfatal myocardial infarction, and nonfatal stroke) and for each of the individual components. Importantly, their analysis also showed that all-cause mortality alone was not significantly higher in pLFLG compared with HGAS (OR, 1.05; 95% CI [0.81-1.37]), which contrasts our finding of increased 1-year mortality in the pLFLG group. This neutral finding may reflect the dilution of subtle mortality differences when pooling heterogeneous cohorts across multiple studies, in contrast to our single-center population, where uniform definitions may have allowed for more precise detection of mortality in pLFLG. Likewise, in a post hoc analysis of the PARTNER 2 trial and SAPIEN 3 registry, Salaun et al12 reported no significant differences in overall clinical outcomes, which included all-cause mortality, heart failure hospitalization, or stroke, through 2 years.
Outcomes for patients with pLFLG AS following TAVR have been inconsistently reported with various conclusions. The discrepant findings across studies are likely multifactorial, driven by differences in defining pLFLG, improvement in procedural technique, variations in patient selection, and the impact of earlier recognition and intervention, all of which may influence clinical outcomes. These findings underscore the need for additional contemporary investigations in this high-risk subgroup.
NFLG AS
Patients with NFLG AS continue to be in a diagnostic and treatment dilemma as the 2020 American guidelines do not provide specific recommendations, and the European guidelines suggest surveillance, based on earlier studies reporting outcomes similar to moderate AS.15,16 Dayan et al,15 in a meta-analysis, have challenged this paradigm by reporting increased mortality in patients with NFLG AS compared with moderate AS and HGAS and demonstrating survival benefits from aortic valve replacement.
In our cohort, outcomes in NFLG patients were similar to those in HGAS, with no significant differences in mortality at 1 or 5 years. This finding is consistent with the post hoc analysis from PARTNER 2, which also demonstrated no survival difference between these 2 groups at 2 years.12 Additionally, Badr et al17 showed comparable mortality rates at both 1 and 5 years post-TAVR between NFLG and HGAS. By contrast, Khaleel et al18 observed a significantly higher 1-year mortality among NFLG patients compared with HGAS (12.9% vs 5.8%, P < .001). This may be explained by the difference in cohort characteristics and analytic approach because of sample size and adjustment models. In an unadjusted model, our NFLG did have a significantly higher mortality rate compared with HGAS; however, after adjustment, it was no longer significant (adjusted P = .06). Similarly, Steffen et al19 divided their NFLG AS group based on the median AVMG of the entire cohort into higher gradient NFLG (>33 mm Hg) and lower gradient NFLG (<33 mm Hg) and showed higher 3-year mortality rates in the lower gradient NFLG, but similar rates in the higher gradient NFLG compared with HGAS. These findings highlight the inherent heterogeneity of patients with NFLG and may explain why we did not observe a higher mortality in our NFLG group, where the median AV gradient was 34 mm Hg.
The variability in published outcomes for NFLG AS following TAVR likely reflects the differences in definitions and an absence of noninterventional control arms in most cohorts.
Conclusion
The present study shows that patients with symptomatic pLFLG AS had worse survival at 1 year and experienced higher rates of heart failure hospitalizations at both 1 and 5 years compared with traditional HGAS, whereas no survival differences were observed at 5 years following TAVR. In contrast, NFLG AS had survival and clinical outcomes similar to HGAS at both 1 and 5 years. These findings illustrate pLFLG AS as a particularly high-risk subgroup following TAVR, while also raising questions about the clinical value of nuanced hemodynamic classification if long-term clinical outcomes ultimately converge across severe AS subtypes in an increasingly older population. A better prognostic indicator and future direction of AS regarding TAVR may lie in investigating nonechocardiographic parameters, such as cardiac biomarkers, global longitudinal strain, and cardiac MRI metrics.20,21
Limitations
This study has a few important limitations. First, this was a retrospective, single-center analysis spanning 10 years, which inherently introduces the potential for selection bias, unmeasured confounding, and variability in clinical practice over time. Second, classification into pLFLG, NFLG, and HGAS subgroups was based on echocardiographic parameters and may have been subject to measurement variability, particularly imaging technology, protocol, and operator expertise that evolved during the study period. Third, baseline characteristics differed between groups, including sex distribution, comorbidity burden, and echocardiographic features, which may have independently influenced outcomes. Notably, the lower proportion of women in the pLFLG group may reflect cohort-specific differences or selection bias. Fourth, the majority of patients (>80%) were treated with self-expanding Medtronic valves, which may limit generalizability. Lastly, although all-cause mortality was used as an end point, cardiovascular-specific mortality was not available, which may have provided additional insight into outcome differences following TAVR. The study team also acknowledges that there may have been covariates that were not adjusted for that may have influenced outcomes.
Despite these limitations, this study represents one of the largest real-world analyses comparing clinical characteristics and outcomes across hemodynamic subtypes of severe AS undergoing TAVR, with flow states classified using SVi.
Acknowledgments
The authors thank the following from Aurora Cardiovascular and Thoracic Services: Sarah Kennedy and Jennifer Pfaff for editorial preparation of the manuscript and Brian Schurrer for assistance with the figures.
Declaration of competing interest
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding sources
This work was not supported by funding agencies in the public, commercial, or not-for-profit sectors. A portion of Prabhjot Hundal and Zaid Abood’s time is supported by the Colton Scholarship, and Niloufar Javadi is supported by the Crivello scholarship, which had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.
Ethics statement and patient consent
This study was approved by the Advocate Health Institutional Review Board. A waiver of consent and HIPAA was obtained because of the retrospective nature of this study.
Footnotes
Previously presented at SCAI 2025 Scientific Sessions in Washington, DC, May 1-3, 2025.
References
- 1.Ambrosy A.P., Go A.S., Leong T.K., et al. Temporal trends in the prevalence and severity of aortic stenosis within a contemporary and diverse community-based cohort. Int J Cardiol. 2023;384:107–111. doi: 10.1016/j.ijcard.2023.04.047. [DOI] [PubMed] [Google Scholar]
- 2.Mack M.J., Leon M.B., Smith C.R., et al. 5-year outcomes of transcatheter aortic valve replacement or surgical aortic valve replacement for high surgical risk patients with aortic stenosis (PARTNER 1): a randomised controlled trial. Lancet. 2015;385(9986):2477–2484. doi: 10.1016/s0140-6736(15)60308-7. [DOI] [PubMed] [Google Scholar]
- 3.Leon M.B., Smith C.R., Mack M.J., et al. Transcatheter or surgical aortic-valve replacement in intermediate-risk patients. N Engl J Med. 2016;374(17):1609–1620. doi: 10.1056/NEJMoa1514616. [DOI] [PubMed] [Google Scholar]
- 4.Mack M.J., Leon M.B., Thourani V.H., et al. Transcatheter aortic-valve replacement with a balloon-expandable valve in low-risk patients. N Engl J Med. 2019;380(18):1695–1705. doi: 10.1056/NEJMoa1814052. [DOI] [PubMed] [Google Scholar]
- 5.Snir A.D., Ng M.K., Strange G., Playford D., Stewart S., Celermajer D.S. Prevalence and outcomes of low-gradient severe aortic stenosis—from the National Echo Database of Australia. J Am Heart Assoc. 2021;10(22) doi: 10.1161/jaha.121.021126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Writing Committee Members, Otto C.M., Nishimura R.A., et al. 2020 ACC/AHA guideline for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. J Am Coll Cardiol. 2021;77(4):e25–e197. doi: 10.1016/j.jacc.2020.11.018. [DOI] [PubMed] [Google Scholar]
- 7.Ullah W., Sana M.K., Mukhtar M., et al. Transcatheter aortic valve replacement across hemodynamic subtypes of severe aortic valve stenosis: a network meta-analysis. J Soc Cardiovasc Angiogr Interv. 2024;3(3Part A) doi: 10.1016/j.jscai.2023.101255. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Wagener M., Reuthebuch O., Heg D., et al. Clinical outcomes in high-gradient, classical low-flow, low-gradient, and paradoxical low-flow, low-gradient aortic stenosis after transcatheter aortic valve implantation: a report from the SwissTAVI Registry. J Am Heart Assoc. 2023;12(12) doi: 10.1161/jaha.123.029489. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hachicha Z., Dumesnil J.G., Bogaty P., Pibarot P. Paradoxical low-flow, low-gradient severe aortic stenosis despite preserved ejection fraction is associated with higher afterload and reduced survival. Circulation. 2007;115(22):2856–2864. doi: 10.1161/circulationaha.106.668681. [DOI] [PubMed] [Google Scholar]
- 10.Mehrotra P., Jansen K., Flynn A.W., et al. Differential left ventricular remodelling and longitudinal function distinguishes low flow from normal-flow preserved ejection fraction low-gradient severe aortic stenosis. Eur Heart J. 2013;34(25):1906–1914. doi: 10.1093/eurheartj/eht094. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Clavel M.A., Magne J., Pibarot P. Low-gradient aortic stenosis. Eur Heart J. 2016;37(34):2645–2657. doi: 10.1093/eurheartj/ehw096. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Salaun E., Clavel M.A., Hahn R.T., et al. Outcome of flow-gradient patterns of aortic stenosis after aortic valve replacement: an analysis of the PARTNER 2 trial and registry. Circ Cardiovasc Interv. 2020;13(7) doi: 10.1161/circinterventions.119.008792. [DOI] [PubMed] [Google Scholar]
- 13.Osman M., Ghaffar Y.A., Foster T., et al. Meta-analysis of outcomes of transcatheter aortic valve implantation among patients with low gradient severe aortic stenosis. Am J Cardiol. 2019;124(3):423–429. doi: 10.1016/j.amjcard.2019.05.006. [DOI] [PubMed] [Google Scholar]
- 14.Guzzetti E., Pibarot P., Clavel M.A. Normal-flow low-gradient severe aortic stenosis is a frequent and real entity. Eur Heart J Cardiovasc Imaging. 2019;20(10):1102–1104. doi: 10.1093/ehjci/jez211. [DOI] [PubMed] [Google Scholar]
- 15.Dayan V., Vignolo G., Magne J., Clavel M.A., Mohty D., Pibarot P. Outcome and impact of aortic valve replacement in patients with preserved LVEF and low-gradient aortic stenosis. J Am Coll Cardiol. 2015;66(23):2594–2603. doi: 10.1016/j.jacc.2015.09.076. [DOI] [PubMed] [Google Scholar]
- 16.Vahanian A., Beyersdorf F., Praz F., et al. 2021 ESC/EACTS guidelines for the management of valvular heart disease. Eur Heart J. 2022;43(7):561–632. doi: 10.1093/eurheartj/ehab395. [DOI] [PubMed] [Google Scholar]
- 17.Badr A., Suppah M., Awad K., et al. Reevaluating normal-flow low-gradient severe aortic stenosis: clinical phenotypes and outcomes in severe aortic stenosis among transcatheter aortic valve replacement patients. J Am Soc Echocardiogr. 2025;38(4):310–319. doi: 10.1016/j.echo.2024.12.010. [DOI] [PubMed] [Google Scholar]
- 18.Khaleel I., Harris A.W., Seth M., et al. Quality of life after transcatheter aortic valve replacement in normal-flow, low-gradient aortic stenosis. JACC Adv. 2023;2(9) doi: 10.1016/j.jacadv.2023.100641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Steffen J., Andreae D., Nabauer M., et al. TAVI for patients with normal-flow low-gradient compared to high-gradient aortic stenosis. Int J Cardiol. 2023;371:299–304. doi: 10.1016/j.ijcard.2022.10.143. [DOI] [PubMed] [Google Scholar]
- 20.Stens N.A., van Iersel O., Rooijakkers M.J.P., et al. Prognostic value of preprocedural LV global longitudinal strain for post-TAVR-related morbidity and mortality: a meta-analysis. JACC Cardiovasc Imaging. 2023;16(3):332–341. doi: 10.1016/j.jcmg.2023.01.005. [DOI] [PubMed] [Google Scholar]
- 21.Hecht S., Giuliani C., Nuche J., et al. Multimarker approach to improve risk stratification of patients undergoing transcatheter aortic valve implantation. JACC Adv. 2023;3(2) doi: 10.1016/j.jacadv.2023.100761. [DOI] [PMC free article] [PubMed] [Google Scholar]