BRIEF RESEARCH COMMUNICATION

Abstract

Paradoxical Low-Flow Low-Gradient Aortic Stenosis: Effect of Low Transvalvular Flow Conditions on Indexed Stroke Volume after Transcatheter Aortic Valve Replacement

Paradoxical low-flow low-gradient aortic stenosis (P-LFLG AS) is a high-risk yet incompletely understood subset of severe AS,¹ for which studies have not uniformly noted improved outcomes after aortic valve replacement.^2,3 Several vascular, valvular, and myocardial conditions contribute to low transvalvular flow (LF) across stenosed aortic valves.⁴ The cumulative burden of LF conditions may affect hemodynamic improvement after transcatheter aortic valve replacement (TAVR) and explain the aforementioned varying survival benefits. In this cohort of P-LFLG AS patients who underwent TAVR, we assessed whether the cumulative burden of LF conditions present at the time of TAVR was associated with change in indexed stroke volume (SVi) post-TAVR.

This was a retrospective review of echocardiogram and electronic medical record data for 66 consecutive patients with P-LFLG AS (calculated aortic valve area < 1.0 cm² or < 0.6 cm²/m², left ventricular ejection fraction [LVEF] > 45%, a mean aortic gradient < 40 mm Hg, and SVi < 35 mL/m²) treated with TAVR. Baseline and 30-day post-TAVR transthoracic echocardiograms were assessed. Left ventricular outflow tract diameter and Doppler waveforms (averaged over three cardiac cycles for patients in sinus rhythm and five for patients in atrial fibrillation [AFib]) were independently reviewed by a board-certified echocardiographer. Consistent with prior reports,⁴ hypertension (HTN), AFib, a low-normal LVEF (≤50%), ≥moderate mitral stenosis (MS), ≥moderate mitral regurgitation (MR), ≥moderate tricuspid regurgitation (TR), pulmonary HTN (defined as right ventricular systolic pressure ≥ 30 mm Hg, pulmonary HTN), a small left ventricular cavity (left ventricular end-diastolic volume < 34 mL/m² for male patients and <29 mL/m² for female patients),⁵ diastolic abnormalities (DA; E/e′ > 14 or E/A > 2), and bundle branch blocks (right or left) or paced rhythms were designated LF conditions. Patients were divided into low (1–2), intermediate (3), or high (4–5) groups based on the cumulative number of LF conditions. To minimize the risk of misclassification due to measurement error, an increase in SVi after TAVR was defined as a ≥5% change.

Baseline characteristics of patients are shown in Table 1. Patients were female (n = 27, 40.9%), elderly (mean age, 81 ± 8 years), overweight (median body mass index, 28.5, IQR 8.4), and smokers (n = 42, 63.6%). They had a mean LVEF of 56% ± 6.7% and elevated median valvuloarterial impedance (5.3 mm Hg/mL/m², IQR 1.1). Of LF conditions, HTN (n = 59, 89.4%), AFib (n = 40, 60.6%), DA (n = 24, 36.6%), pulmonary HTN (n = 27, 40.9%), bundle branch block/paced rhythms (n = 19, 28.8%), and low-normal EF (n = 15, 22.7%) were the most prevalent. Tricuspid regurgitation (n = 13, 19.7%) was the most common valvular (nonaortic) pathology followed by MR (n = 7, 10.6%). Thirty-nine (59%) patients had three or more LF conditions. Median SVi was 29.8 mL/m² (IQR 6.1) pre-TAVR and 33 mL/m² (IQR 10.4) post-TAVR. Post-TAVR SVi varied widely (Figure 1A). Thirty-seven (56.1%) patients had improvement in SVi post-TAVR. The SVi for 26 patients (39.4% of the overall cohort) improved to ≥35 mL/m². Mean increases in SVi in the low (0–1 conditions), intermediate (2–3 conditions), and high (≥4 conditions) groups were 7.2 ± 10.8, 4.6 ± 7.4, and 1.9 ± 8.3 mL/m², respectively. The percentage of patients with improvement in SVi post-TAVR was significantly higher in the low group (n = 18, 66.7%) compared with the intermediate (n = 11, 61.1%) and high (n = 8, 38.1%) groups (P = .05; Figure 1B).

Table 1.

Baseline characteristics of patients with P-LFLG AS, compared by groups with and without improvement in SVi

	All (N = 66)	Improvement in SVi (n = 37)	No improvement in SVi (n = 29)
Demographics
Sex, female	27, 40.9	17, 45.9	10, 34.5
Age, years	81 ± 8	80 ± 7	81 ± 8
Body mass index, kg/cm2	28.5 (8.4)	27.5 (5.8)	29.5 (9.5)
Body surface area, m2	2.0 ± 0.2	1.9 ± 0.2	2.0 ± 0.3
Comorbidities
HTN	59, 89.4	33, 89.2	26, 89.7
Diabetes	23, 34.8	12, 32.4	11, 37.9
AFib	40, 60.6	22, 59.5	18, 62.1
Coronary artery disease	37, 56.1	21, 56.8	16, 55.2
Peripheral arterial disease	12, 18.2	8, 21.6	4, 13.8
Cerebrovascular accident	9, 13.6	7, 18.9	2, 6.9
Hyperlipidemia	48, 72.7	25, 67.6	23, 79.3
Chronic obstructive pulmonary disease	18, 27.3	12, 32.4	6, 20.7
Ever smokers	42, 63.6	26, 70.3	16, 55.2
Systemic arterial hemodynamics and LV afterload
SBP, mm Hg	132 ± 18	132 ± 20	132 ± 17
DBP, mm Hg	69 ± 12	71 ± 12	68 ± 12
Valvuloarterial impedance, mm Hg/mL/m2	5.3 (1.1)	5.4 (1.3)	5.2 (1.1)
LV geometry
LVEF	56.0 ± 6.7	56.3 ± 6.8	55.6 ± 6.7
Low-normal ejection fraction	15, 22.7	7, 18.9	8, 27.6
LV internal diameter systolic, cm	3.0 ± 0.7	2.9 ± 0.6	3.2 ± 0.7
LV end-diastolic internal diameter, cm	4.2 ± 0.7	4.2 ± 0.6	4.2 ± 0.8
LV septal wall thickness, cm	1.2 ± 0.2	1.2 ± 0.2	1.3 ± 0.2
LV posterior wall thickness, cm	1.2 ± 0.2	1.2 ± 0.2	1.2 ± 0.2
LV mass index, g/m2	95.0 ± 27.3	93.2 ± 26.0	103.1 ± 26.5
LV end-diastolic volume, mL*	92.5 ± 31.8	91.1 ± 32.1	94.4 ± 32.2
LV end-systolic volume, mL*	41.1 ± 17.2	38.9 ± 13.7	44.2 ± 21.2
Indexed LV end-diastolic volume, mL/m2*	47.8 ± 15.0	47.4 ± 15.2	48.4 ± 15.1
Small cavity	6, 10.7	2, 6.1	4, 17.4
AS severity
Aortic valve area, cm2†	0.8 ± 0.1	0.7 ± 0.1	0.8 ± 0.1
Indexed aortic valve area, cm2/m2†	0.4 ± 0.1	0.4 ± 0.1	0.4 ± 0.1
Mean aortic gradient, mm Hg	27.2 ± 6.8	27.1 ± 6.9	27.2 ± 6.8
Extra-aortic valvular/cardiac dysfunction
MS	3, 4.5	3, 8.1	0, 0
MR	7, 10.6	3, 8.1	4, 13.8
TR	13, 19.7	7, 18.9	6, 20.7
Pulmonary HTN	27, 40.9	15, 40.5	12, 41.4
LV diastolic function
E/e′‡	13.4 (6.9)	13.1 (10.2)	15.0 (4.8)
E/e′ >14‡	19, 48.7	9, 47.4	10, 50.0
E/A§	0.89 (0.75)	0.84 (0.60)	0.93 (1.41)
E/A >2§	8, 19.5	3, 13.0	5, 27.8
Any DA in entire cohort¶	24, 36.4	11, 29.7	13, 48.1
Any DA in cohort with data	24, 53.3	11, 47.8	13, 59.1
LV systolic function
Stroke volume by left ventricular outflow tract pre-TAVR, mL	57.8 (13.2)	56.4 (12.4)	60.5 (15.1)
SVi pre-TAVR, mL/m2	29.8 (6.1)	29.8 (6.3)	30.5 (4.7)
SVi post-TAVR, mL/m2∥	33.0 (29.8–40.2)
Change in SVi mL/m2∥	3.3 (−2.3–10.2)
Electrocardiogram parameters
Heart rate, bpm	75.0 (19.8)	76.0 (17.0)	71.0 (27.0)
PR interval, msec	184 (44)	184.0 (36.0)	180.0 (86.0)
QRS duration, msec	104 (34)	102.0 (26.0)	112.0 (42.0)
Bundle branch block or paced	19, 28.8	9, 24.3	10, 34.5
Laboratory values
Cr mg/dL	1.1 (0.4)	1.1 (0.6)	1.2 (0.4)
GFR mL/m2	60 (32)	55.0 (38.0)	58.0 (28.0)
Hgb, g/dL	12.2 ± 1.8	12.0 ± 1.9	12.5 ± 1.7

Cr, Creatinine; DBP, diastolic blood pressure; GFR, glomerular filtration rate; Hgb, hemoglobin; LV, left ventricle/ventricular; SBP, systolic blood pressure.

Values are expressed as n, % for categorical variables and mean ± SD or median (interquartile range) for numerical variables.

^*Data available in 56 of the 66 patients.

^†Data available 65 of the 66 patients.

^‡Data available in 39 of the 66 patients.

^§Data available in 41 of the 66 patients.

^¶Data available in 45 of the 66 patients.

^∥Data available in 62 of the 66 patients, four in-hospital deaths excluded (hence, with no subgroup comparison).

Figure 1.

(A) Pre- to post-TAVR change in SVi in P-LFLG AS patients and (B) proportion of P-LFLG AS patients with improvement in SVi post-TAVR by groups of conditions causing LF, with the number of LF conditions in parentheses.

These observations are hypothesis generating and suggest that P-LFLG AS may represent a diverse set of pathophysiologic conditions that explain why some patients do not have hemodynamic (or symptomatic) improvement after treatment of AS. The results of our study demonstrate that there is less improvement in SVi post-TAVR as the cumulative burden of LF conditions increases.

In this unselected cohort of patients with P-LFLG AS treated with TAVR there was a heterogeneous distribution of conditions contributing to LF. Hypertension was common and is known to be associated with a paradoxical increase in afterload post-TAVR and higher mortality in P-LFLG AS.⁶ This, along with an elevated median valvuloarterial impedance, suggests a persistent excess ventricular afterload following TAVR that may represent a separate therapeutic target for certain patients after TAVR. Atrial fibrillation and DA were also commonly seen in this cohort and may have variable courses after TAVR, depending on the timing of intervention and whether or not significant remodeling has occurred.⁷ We included patients with nonaortic valvular pathologies treated at our institution as they are often encountered in clinical practice. We noted that TR and MR were common. To our knowledge, this is the first report suggesting that in routine practice TR is a more common cause of LF than MR. Taken together, these data suggest that P-LFLG AS is not a single disease state but instead represents a heterogeneous group of conditions with different responses to TAVR.

To our knowledge this is the largest series of serial echocardiograms for unselected P-LFLG AS patients treated with TAVR. While this is a comprehensively described cohort, the numbers are modest and the analysis looks at changes in SVi as a surrogate for clinical improvement post TAVR.⁸ With a larger population, future studies can evaluate the relative weights of conditions contributing to LF and associations with clinical outcomes and ultimately identify patients with P-LFLG AS who are more (or less) likely to benefit from TAVR.