Search
Close
Search
Advanced Search
Search Menu
AI Discovery Assistant

Abstract

Aims

We aimed to evaluate the efficacy of the measured effective orifice area (EOA)/reference EOA ratio in discriminating mechanical prosthetic aortic valve (PAV) obstruction.

Methods and results

This is a retrospective study of 193 mechanical PAV patients with an elevated mean transprosthetic pressure gradient (PG) over 20 mmHg or peak velocity over 3 m/s. Of those, 143 patients were objectively proven PAV obstruction with cardiac computed tomography or surgical inspection. The EOA was measured using the continuity equation, and the reference EOA values were obtained from previous guidelines. The measured/reference EOA ratio was significantly lower in the obstruction group (0.63 ± 0.18 vs. 0.86 ± 0.17; P < 0.001). The EOA ratio added incremental value for discriminating obstruction from the conventional parameters recommended in the guidelines. Receiver operating characteristic curve analysis revealed that the measured/reference EOA ratio discriminated PAV obstruction from those without obstruction [area under the curve (AUC), 0.840; 95% confidence interval, 0.783–0.898; P < 0.001]. A cutoff of 0.71 had 73.4% sensitivity and 82.0% specificity. The novel diagnostic algorithm adding the EOA ratio had similar accuracy to previous guideline algorithms, including reference EOA, and conventional Doppler parameters (AUC, 0.763 vs. 0.731; P = 0.309). In patients with a large PAV (≥23 mm), the novel algorithm had higher accuracy than the previous algorithm (AUC, 0.788 vs. 0.642; P = 0.019).

Conclusion

The ratio of measured/reference EOA adds incremental value over conventional Doppler parameters and might be helpful for distinguishing PAV obstruction.

Prosthetic aortic valves, without or with obstruction (A), novel algorithm for diagnosing pannus (B), and receiver operating characteristic curves of novel and previous guideline algorithm for detecting pannus (C). *Reference effective orifice area, †measured effective orifice area, ‡pannus formation.
Graphical Abstract

Prosthetic aortic valves, without or with obstruction (A), novel algorithm for diagnosing pannus (B), and receiver operating characteristic curves of novel and previous guideline algorithm for detecting pannus (C). *Reference effective orifice area, †measured effective orifice area, ‡pannus formation.

Open in new tabDownload slide
prosthetic aortic valve dysfunction, pannus, diagnostic algorithm, reference effective orifice area

Introduction

Valvular heart disease affects more than 100 million people worldwide with an increasing burden.1 Despite improved treatment strategies, valve replacement does not provide a definitive cure and prosthetic valve diseases persist.2 Among these, prosthetic aortic valve (PAV) obstruction derived from a pannus or thrombus is a life-threatening complication (Graphical abstract A).3 Doppler transthoracic echocardiography (TTE) is the standard method for assessing prosthetic valve function because it is non-invasive, radiation-free, and cost-effective.4 Several TTE parameters such as Doppler velocity index (DVI), acceleration time (AT), ejection time (ET) ratio, transprosthetic jet contour, and effective orifice area (EOA) are used in PAV assessment.5–8 However, the presence of PAV induces acoustic shadowing and reverberation artefacts, leading to suboptimal examination for PAV function. Because mechanical valves have more severe artefacts, evaluating PAV obstruction is challenging.7 In addition, a high transprosthetic pressure gradient (PG) or velocity alone is not proof of PAV obstruction. This phenomenon may be secondary to prosthesis–patient mismatch (PPM), a high-flow condition, PAV regurgitation, or localized high jet velocity.4 Previous guidelines recommend the additional use of computed tomography (CT), transesophageal echocardiography (TEE), or fluoroscopy.9 However, cardiac CT is associated with iodine contrast and radiation exposure, and TEE is semi-invasive. In particular, when prosthetic valve dysfunction is caused by pannus, the diagnosis is challenging using echocardiography.10 Recent recommendations from the European Association of Cardiovascular Imaging have described the use of reference EOA values, but clinical field validation is lacking.11 This study aimed to evaluate the efficacy of reference EOA values for discriminating PAV obstruction and construct the novel diagnostic algorithm using reference EOA values.

Methods

Study population

This study was a retrospective study from single tertiary hospital. A total of 996 patients with mechanical PAV and elevated transprosthetic velocity (>3.0 m/s) or mean systolic PG (≥20 mmHg) were identified. Patients with bioprosthetic AV were not included in this study. Among them, 223 underwent CT or cardiac surgery within 3 months. Patients with active endocarditis involving PAV, without reference EOA values or PAV information, with poor image quality for calculating Doppler parameters were excluded. After exclusion, 193 patients with PAV were analyzed; of them, 143 were diagnosed with PAV obstruction and 50 were diagnosed have intact PAV function, classified as no PAV obstruction group (Figure 1). Diagnosis for PAV obstruction was made by surgical inspection or cardiac CT findings. In case of discordance between surgical inspection and CT, we considered surgical inspection first. CT scan and referring patients for surgery were performed at discretion of the responsible physician. Patient data were obtained from medical records. The reference EOA values for each aortic valve (AV) prosthesis were obtained from previous guidelines and studies (see Supplementary data online, Table S1).2,9,11,12 Then, projected indexed EOA values were calculated as the reference EOA from the guidelines/body surface area (BSA), which was described in recent guideline from European Association of Cardiovascular Imaging.11 Patients with a projected indexed EOA ≤ 0.65 [≤0.55 in body mass index (BMI) ≥ 30 kg/m2] were defined as having severe PPM and 0.66–0.85 (0.56–0.70 in BMI ≥ 30 kg/m2) as moderate PPM. We performed sensitivity analysis to test if our findings were robust. First, we analyzed patients diagnosed with PAV obstruction based only on surgical findings. Second, we defined the PAV obstruction as in case of limited motion of the PAV leaflet by CT or surgical inspection. The Institutional Review Board of Severance Hospital approved this study, which was conducted in compliance with the Declaration of Helsinki. The requirement for informed consent was waived.

Study flow. MPG; mean pressure gradient, PAV; prosthetic aortic valve.
Figure 1

Study flow. MPG; mean pressure gradient, PAV; prosthetic aortic valve.

Open in new tabDownload slide

Echocardiography

All echocardiograms were analyzed in a blinded fashion by trained cardiologists. According to current guidelines,11,13 the stroke volume was calculated using the following formula: annulus diameter2 × 0.785 × velocity-time integral of LVOT (VTILVOT). The transprosthetic peak velocity, peak, and mean transprosthetic PG were measured using continuous-wave Doppler VTI. The EOA was measured as stroke volume/Velocity-time integral of aortic valve (VTIAV) by continuity equation. The ratio of measured/reference EOA was calculated using mean reference EOA value of each prosthetic AV from previous guidelines.9,11 The transprosthetic jet contour was qualitatively estimated and classified into 3 categories (triangular and early peaking, intermediate, rounded, and symmetric).9,11 The DVI was calculated as VTILVOT/VTIAV. AT was defined as the time interval from the beginning of systolic flow to the peak velocity. ET was measured as the time from onset to the end of systolic flow across the PAV.7 All measurements represent an average of three cardiac cycles for patients with sinus rhythm and an average of five cardiac cycles for patients with atrial fibrillation.11 Intraclass correlation coefficients (ICCs) were calculated to assess the inter- and intra-observer variability of the measured/reference EOA ratio and the difference between the measured LVOT diameter and reference annulus diameters of the PAV. A total of 20 samples (included 10 without PAV obstructions, and 10 PAV obstructions) were assessed by the same observer on different occasions in a random order and by another observer. We then constructed novel hierarchical diagnostic algorithm models for PAV obstruction by pannus and compared the novel algorithm with previously recommended algorithms (see Supplementary data online, Figure S1).9,11 Fluoroscopic data and EOA changes during follow-up, which are described in 2016 recommendation, were not included for the purpose of this study to evaluate incremental value of single echocardiographic parameter, measured/reference EOA ratio. We performed a subgroup analysis of patients expected to have difficulties in the accurate assessment of PAV function: large-size PAV (≥23 mm), atrial fibrillation, double valve replacement (DVR), and lower transprosthetic PG (<40 mmHg).

Cardiac CT

All CT scans were performed with a dual-source CT scanner (SOMATOM Definition Flash; Siemens Healthcare, Forchheim, Germany). Scans were performed in retrospective ECG-gated data acquisition mode using the triple-phase injection method (70 mL of iopamidol followed by 30 mL of 30% blended iopamidol with saline and 20 mL of saline at 5 mL/s). Images were generated using iterative reconstruction (sinogram-affirmed iterative reconstruction). Image reconstruction was performed with a medium kernel (I36f), and the reconstruction slice thickness was 0.75 mm with 0.5-mm increments. For all patients, the datasets of transverse plane image were reconstructed every 10% of the cardiac cycle. Image analysis was performed using 3D software (Aquarius iNtuition, Ver. 4.4.11, TeraRecon, San Mateo, CA, USA). The presence of subprosthetic soft tissue mass with low-attenuation (pannus) was assessed.14 When CT attenuation of a subvalvular mass was lower than that of the interventricular septum, thrombus was favoured over pannus. Limited motion of a mechanical PAV was considered present when motion of a leaflet or leaflets was restricted, with an opening angle which was reduced more than 4° compared to the manufacturer’s value.15 All CT analyses were independently performed by radiologists blinded to clinical information and echocardiographic results.

Statistical analysis

Continuous variables are presented as mean ± standard deviation (SD) and compared using Student’s t-test. If the variables were not normally distributed, they were presented as median (interquartile range) and compared using Wilcoxon rank-sum test. Categorical variables are presented as number (percentage and compared using the χ2 test. The incremental value of the measured/reference EOA ratio over conventional Doppler parameters for discriminating PAV obstruction was assessed using logistic regression models. The R-squared and global χ2 scores for each model were calculated. Prognostic values were compared using the likelihood-ratio test. A receiver operating characteristic (ROC) curve was plotted to determine the accuracy of each Doppler parameter and the best cutoff values for identifying PAV obstruction. We also developed a hierarchical algorithm model using the reference EOA value and compared it with previous diagnostic algorithm models.9,11 The area under the ROC curve (AUC) was calculated for each algorithm. The accuracy of the diagnostic algorithm was compared using DeLong’s test.16 All statistical analyses were performed with R version 4.1.0 (The R Foundation for Statistical Computing; www.R-project.org).

Results

Baseline characteristics and echocardiographic parameters

The patients’ baseline characteristics are presented in Table 1. The details of the PAVs are presented in Supplementary data online, Table S2. The patients with PAV obstruction were older than those of no PAV obstruction group [median age: 65.0 (54.0–69.0) vs. 58.0 (52.0–67.0) years; P = 0.036], and female was more common (76.2 vs. 40.0%; P < 0.001). Blood pressure, heart rate, BMI, heart rhythm, and the proportion of projected PPM did not differ significantly between the groups. The method of diagnosis for PAV obstruction was mainly based on surgical inspection in PAV obstruction group (65.7 vs. 26.0%), but CT in no PAV obstruction group (34.3 vs. 74.0%, P < 0.001). Both CT and surgical inspection were performed in 91 (47.2%) of 213 patients. Three of 91 patients showed discordant findings; CT was negative, but pannus was detected in surgical inspection. There was one case of PAV thrombi and it was combined with pannus. In no PAV obstruction group, 13 of 50 (26.0%) patients underwent surgery. Indications of surgery in patients without PAV obstruction were paravalvular regurgitation in 2, and severe dysfunction of mitral or tricuspid valves in 11. In PAV obstruction group, 94 of 143 (65.7%) patients underwent cardiac surgery whereas 49 (34.3%) patients did not.

Table 1
Open in new tab

Baseline characteristics

No PAV obstruction (n = 50)PAV obstruction (n = 143)P value
Age, years58.0 (52.0–67.0)65.0 (54.0–69.0)0.036
Sex<0.001
ȃMale30 (60.0%)34 (23.8%)
ȃFemale20 (40.0%)109 (76.2%)
Height, cm165.1 ± 8.8158.5 ± 8.2<0.001
Weight, kg59.5 (54.0–72.0)55.0 (50.0–62.5)0.002
BSA, m21.67 (1.54–1.84)1.55 (1.46–1.68)<0.001
BMI, kg/m222.5 (20.4–25.7)22.3 (20.3–24.4)0.503
ȃ ≥ 302 (4.0%)4 (2.8%)>0.999
ȃ < 3048 (96.0%)139 (97.2%)>0.999
Systolic blood pressure, mmHg121.0 ± 17.8121.3 ± 16.10.926
Diastolic blood pressure, mmHg70.9 ± 9.268.7 ± 10.50.193
Heart rate, beats per minute68.0 (60.0–80.0)67.0 (60.0–76.0)0.326
Rhythm0.178
ȃSinus rhythm20 (40.0%)42 (29.4%)
ȃAtrial fibrillation27 (54.0%)97 (67.8%)
ȃPaced3 (6.0%)4 (2.8%)
Prosthetic valve size21.0 (19.0–23.0)21.0 (19.0–21.0)0.032
ȃRange19–2519–25
ȃLarge size (≥23 mm)18 (36.0%)16 (11.2%)<0.001
Prosthetic valve type
ȃBileaflet49 (98.0%)137 (95.8)0.783
ȃMonoleaflet1 (2.0%)6 (4.2%)0.783
ȃSupra-annular type6 (12.0%)12 (8.4%)0.636
Time interval from previous AVR, years11.9 ± 9.219.3 ± 7.8<0.001
Reference EOA1.50 (1.30–1.70)1.40 (1.20–1.50)0.005
Projected indexed EOA, cm2/m20.90 (0.75–0.98)0.91 (0.76–0.99)0.991
ȃModerate patient–prosthesis mismatch12 (24.0%)36 (25.2%)>0.999
ȃSevere patient–prosthesis mismatch4 (8.0%)14 (9.8%)0.927
DVR status33 (66.0%)119 (83.2%)0.018
Limited motion of PAV leaflet0 (0%)84 (58.7%)<0.001
Concomitant aortic valve regurgitation0.091
ȃNo, mild45 (90.0%)139 (97.2%)
ȃModerate or severe5 (10.0%)4 (2.8%)
Method of diagnosis<0.001
ȃSurgical inspection13 (26.0%)94 (65.7%)
ȃCardiac CT37 (74.0%)49 (34.3%)
Indication for reoperation<0.001
ȃPAV obstruction0 (0%)94 (65.7%)
ȃPAV paravalvular regurgitation2 (4.0%)0 (0%)
ȃMitral or tricuspid valvular dysfunction11 (22.0%)29 (20.3%)
Redo aortic valve replacement2 (4.0%)94 (65.7%)<0.001
Echocardiography
ȃEOA, cm21.26 (1.00–1.50)0.80 (0.68–0.99)<0.001
ȃMeasured/reference EOA ratio0.86 ± 0.170.63 ± 0.18<0.001
ȃEOA/BSA, cm2/m20.76 (0.64–0.90)0.51 (0.43–0.64)<0.001
ȃLVEF, %62.0 (57.0–68.0)66.0 (61.0–71.0)0.009
ȃȃ < 45%5 (10.0%)9 (6.3%)0.580
ȃLVOT diameter, mm21.7 (19.9–23.3)20.5 (19.0–21.2)<0.001
ȃLVOT diameter difference from valve size0.45 (−0.30–1.00)−0.10 (−1.25–0.20)<0.001
ȃLVOT VTI, cm22.7 ± 4.822.7 ± 4.60.994
ȃPeak LVOT velocity, m/s1.09 (0.97–1.29)1.08 (0.93–1.21)0.296
ȃStroke volume, mL83.2 (72.0–103.1)69.2 (60.7–83.4)<0.001
ȃStroke volume index, mL/m249.4 (41.4–60.1)44.7 (38.6–51.7)0.019
ȃDynamic LVOT obstruction0 (0%)8 (5.6%)0.195
ȃAV VTI, cm66.1 (57.6–75.3)88.0 (72.1–101.3)<0.001
ȃPeak transprosthetic velocity, m/s3.33 (3.09–3.67)4.10 (3.63–4.71)<0.001
ȃPeak transprosthetic pressure, mmHg44.5 (38.0–54.0)67.0 (52.0–88.2)<0.001
ȃMean transprosthetic pressure, mmHg25.5 (21.0–30.0)37.4 (29.3–50.5)<0.001
ȃDoppler velocity index0.34 (0.29–0.41)0.26 (0.23–0.30)<0.001
ȃAcceleration time of aortic velocity, msec83.4 ± 22.598.9 ± 23.0<0.001
ȃEjection time, msec282.8 ± 44.2293.5 ± 35.10.128
ȃAcceleration time/ejection time ratio0.30 ± 0.070.34 ± 0.06<0.001
ȃCW Doppler jet contour<0.001
ȃȃTriangle, early peaking11 (22.0%)17 (11.9%)
ȃȃIntermediate38 (76.0%)78 (54.5%)
ȃȃRound, symmetrical1 (2.0%)48 (33.6%)
No PAV obstruction (n = 50)PAV obstruction (n = 143)P value
Age, years58.0 (52.0–67.0)65.0 (54.0–69.0)0.036
Sex<0.001
ȃMale30 (60.0%)34 (23.8%)
ȃFemale20 (40.0%)109 (76.2%)
Height, cm165.1 ± 8.8158.5 ± 8.2<0.001
Weight, kg59.5 (54.0–72.0)55.0 (50.0–62.5)0.002
BSA, m21.67 (1.54–1.84)1.55 (1.46–1.68)<0.001
BMI, kg/m222.5 (20.4–25.7)22.3 (20.3–24.4)0.503
ȃ ≥ 302 (4.0%)4 (2.8%)>0.999
ȃ < 3048 (96.0%)139 (97.2%)>0.999
Systolic blood pressure, mmHg121.0 ± 17.8121.3 ± 16.10.926
Diastolic blood pressure, mmHg70.9 ± 9.268.7 ± 10.50.193
Heart rate, beats per minute68.0 (60.0–80.0)67.0 (60.0–76.0)0.326
Rhythm0.178
ȃSinus rhythm20 (40.0%)42 (29.4%)
ȃAtrial fibrillation27 (54.0%)97 (67.8%)
ȃPaced3 (6.0%)4 (2.8%)
Prosthetic valve size21.0 (19.0–23.0)21.0 (19.0–21.0)0.032
ȃRange19–2519–25
ȃLarge size (≥23 mm)18 (36.0%)16 (11.2%)<0.001
Prosthetic valve type
ȃBileaflet49 (98.0%)137 (95.8)0.783
ȃMonoleaflet1 (2.0%)6 (4.2%)0.783
ȃSupra-annular type6 (12.0%)12 (8.4%)0.636
Time interval from previous AVR, years11.9 ± 9.219.3 ± 7.8<0.001
Reference EOA1.50 (1.30–1.70)1.40 (1.20–1.50)0.005
Projected indexed EOA, cm2/m20.90 (0.75–0.98)0.91 (0.76–0.99)0.991
ȃModerate patient–prosthesis mismatch12 (24.0%)36 (25.2%)>0.999
ȃSevere patient–prosthesis mismatch4 (8.0%)14 (9.8%)0.927
DVR status33 (66.0%)119 (83.2%)0.018
Limited motion of PAV leaflet0 (0%)84 (58.7%)<0.001
Concomitant aortic valve regurgitation0.091
ȃNo, mild45 (90.0%)139 (97.2%)
ȃModerate or severe5 (10.0%)4 (2.8%)
Method of diagnosis<0.001
ȃSurgical inspection13 (26.0%)94 (65.7%)
ȃCardiac CT37 (74.0%)49 (34.3%)
Indication for reoperation<0.001
ȃPAV obstruction0 (0%)94 (65.7%)
ȃPAV paravalvular regurgitation2 (4.0%)0 (0%)
ȃMitral or tricuspid valvular dysfunction11 (22.0%)29 (20.3%)
Redo aortic valve replacement2 (4.0%)94 (65.7%)<0.001
Echocardiography
ȃEOA, cm21.26 (1.00–1.50)0.80 (0.68–0.99)<0.001
ȃMeasured/reference EOA ratio0.86 ± 0.170.63 ± 0.18<0.001
ȃEOA/BSA, cm2/m20.76 (0.64–0.90)0.51 (0.43–0.64)<0.001
ȃLVEF, %62.0 (57.0–68.0)66.0 (61.0–71.0)0.009
ȃȃ < 45%5 (10.0%)9 (6.3%)0.580
ȃLVOT diameter, mm21.7 (19.9–23.3)20.5 (19.0–21.2)<0.001
ȃLVOT diameter difference from valve size0.45 (−0.30–1.00)−0.10 (−1.25–0.20)<0.001
ȃLVOT VTI, cm22.7 ± 4.822.7 ± 4.60.994
ȃPeak LVOT velocity, m/s1.09 (0.97–1.29)1.08 (0.93–1.21)0.296
ȃStroke volume, mL83.2 (72.0–103.1)69.2 (60.7–83.4)<0.001
ȃStroke volume index, mL/m249.4 (41.4–60.1)44.7 (38.6–51.7)0.019
ȃDynamic LVOT obstruction0 (0%)8 (5.6%)0.195
ȃAV VTI, cm66.1 (57.6–75.3)88.0 (72.1–101.3)<0.001
ȃPeak transprosthetic velocity, m/s3.33 (3.09–3.67)4.10 (3.63–4.71)<0.001
ȃPeak transprosthetic pressure, mmHg44.5 (38.0–54.0)67.0 (52.0–88.2)<0.001
ȃMean transprosthetic pressure, mmHg25.5 (21.0–30.0)37.4 (29.3–50.5)<0.001
ȃDoppler velocity index0.34 (0.29–0.41)0.26 (0.23–0.30)<0.001
ȃAcceleration time of aortic velocity, msec83.4 ± 22.598.9 ± 23.0<0.001
ȃEjection time, msec282.8 ± 44.2293.5 ± 35.10.128
ȃAcceleration time/ejection time ratio0.30 ± 0.070.34 ± 0.06<0.001
ȃCW Doppler jet contour<0.001
ȃȃTriangle, early peaking11 (22.0%)17 (11.9%)
ȃȃIntermediate38 (76.0%)78 (54.5%)
ȃȃRound, symmetrical1 (2.0%)48 (33.6%)

AVR, aortic valve replacement; CW, continuous wave; LVEF, left ventricular ejection fraction; LVOT, left ventricular outflow tract; VTI, velocity-time integral.

Table 1
Open in new tab

Baseline characteristics

No PAV obstruction (n = 50)PAV obstruction (n = 143)P value
Age, years58.0 (52.0–67.0)65.0 (54.0–69.0)0.036
Sex<0.001
ȃMale30 (60.0%)34 (23.8%)
ȃFemale20 (40.0%)109 (76.2%)
Height, cm165.1 ± 8.8158.5 ± 8.2<0.001
Weight, kg59.5 (54.0–72.0)55.0 (50.0–62.5)0.002
BSA, m21.67 (1.54–1.84)1.55 (1.46–1.68)<0.001
BMI, kg/m222.5 (20.4–25.7)22.3 (20.3–24.4)0.503
ȃ ≥ 302 (4.0%)4 (2.8%)>0.999
ȃ < 3048 (96.0%)139 (97.2%)>0.999
Systolic blood pressure, mmHg121.0 ± 17.8121.3 ± 16.10.926
Diastolic blood pressure, mmHg70.9 ± 9.268.7 ± 10.50.193
Heart rate, beats per minute68.0 (60.0–80.0)67.0 (60.0–76.0)0.326
Rhythm0.178
ȃSinus rhythm20 (40.0%)42 (29.4%)
ȃAtrial fibrillation27 (54.0%)97 (67.8%)
ȃPaced3 (6.0%)4 (2.8%)
Prosthetic valve size21.0 (19.0–23.0)21.0 (19.0–21.0)0.032
ȃRange19–2519–25
ȃLarge size (≥23 mm)18 (36.0%)16 (11.2%)<0.001
Prosthetic valve type
ȃBileaflet49 (98.0%)137 (95.8)0.783
ȃMonoleaflet1 (2.0%)6 (4.2%)0.783
ȃSupra-annular type6 (12.0%)12 (8.4%)0.636
Time interval from previous AVR, years11.9 ± 9.219.3 ± 7.8<0.001
Reference EOA1.50 (1.30–1.70)1.40 (1.20–1.50)0.005
Projected indexed EOA, cm2/m20.90 (0.75–0.98)0.91 (0.76–0.99)0.991
ȃModerate patient–prosthesis mismatch12 (24.0%)36 (25.2%)>0.999
ȃSevere patient–prosthesis mismatch4 (8.0%)14 (9.8%)0.927
DVR status33 (66.0%)119 (83.2%)0.018
Limited motion of PAV leaflet0 (0%)84 (58.7%)<0.001
Concomitant aortic valve regurgitation0.091
ȃNo, mild45 (90.0%)139 (97.2%)
ȃModerate or severe5 (10.0%)4 (2.8%)
Method of diagnosis<0.001
ȃSurgical inspection13 (26.0%)94 (65.7%)
ȃCardiac CT37 (74.0%)49 (34.3%)
Indication for reoperation<0.001
ȃPAV obstruction0 (0%)94 (65.7%)
ȃPAV paravalvular regurgitation2 (4.0%)0 (0%)
ȃMitral or tricuspid valvular dysfunction11 (22.0%)29 (20.3%)
Redo aortic valve replacement2 (4.0%)94 (65.7%)<0.001
Echocardiography
ȃEOA, cm21.26 (1.00–1.50)0.80 (0.68–0.99)<0.001
ȃMeasured/reference EOA ratio0.86 ± 0.170.63 ± 0.18<0.001
ȃEOA/BSA, cm2/m20.76 (0.64–0.90)0.51 (0.43–0.64)<0.001
ȃLVEF, %62.0 (57.0–68.0)66.0 (61.0–71.0)0.009
ȃȃ < 45%5 (10.0%)9 (6.3%)0.580
ȃLVOT diameter, mm21.7 (19.9–23.3)20.5 (19.0–21.2)<0.001
ȃLVOT diameter difference from valve size0.45 (−0.30–1.00)−0.10 (−1.25–0.20)<0.001
ȃLVOT VTI, cm22.7 ± 4.822.7 ± 4.60.994
ȃPeak LVOT velocity, m/s1.09 (0.97–1.29)1.08 (0.93–1.21)0.296
ȃStroke volume, mL83.2 (72.0–103.1)69.2 (60.7–83.4)<0.001
ȃStroke volume index, mL/m249.4 (41.4–60.1)44.7 (38.6–51.7)0.019
ȃDynamic LVOT obstruction0 (0%)8 (5.6%)0.195
ȃAV VTI, cm66.1 (57.6–75.3)88.0 (72.1–101.3)<0.001
ȃPeak transprosthetic velocity, m/s3.33 (3.09–3.67)4.10 (3.63–4.71)<0.001
ȃPeak transprosthetic pressure, mmHg44.5 (38.0–54.0)67.0 (52.0–88.2)<0.001
ȃMean transprosthetic pressure, mmHg25.5 (21.0–30.0)37.4 (29.3–50.5)<0.001
ȃDoppler velocity index0.34 (0.29–0.41)0.26 (0.23–0.30)<0.001
ȃAcceleration time of aortic velocity, msec83.4 ± 22.598.9 ± 23.0<0.001
ȃEjection time, msec282.8 ± 44.2293.5 ± 35.10.128
ȃAcceleration time/ejection time ratio0.30 ± 0.070.34 ± 0.06<0.001
ȃCW Doppler jet contour<0.001
ȃȃTriangle, early peaking11 (22.0%)17 (11.9%)
ȃȃIntermediate38 (76.0%)78 (54.5%)
ȃȃRound, symmetrical1 (2.0%)48 (33.6%)
No PAV obstruction (n = 50)PAV obstruction (n = 143)P value
Age, years58.0 (52.0–67.0)65.0 (54.0–69.0)0.036
Sex<0.001
ȃMale30 (60.0%)34 (23.8%)
ȃFemale20 (40.0%)109 (76.2%)
Height, cm165.1 ± 8.8158.5 ± 8.2<0.001
Weight, kg59.5 (54.0–72.0)55.0 (50.0–62.5)0.002
BSA, m21.67 (1.54–1.84)1.55 (1.46–1.68)<0.001
BMI, kg/m222.5 (20.4–25.7)22.3 (20.3–24.4)0.503
ȃ ≥ 302 (4.0%)4 (2.8%)>0.999
ȃ < 3048 (96.0%)139 (97.2%)>0.999
Systolic blood pressure, mmHg121.0 ± 17.8121.3 ± 16.10.926
Diastolic blood pressure, mmHg70.9 ± 9.268.7 ± 10.50.193
Heart rate, beats per minute68.0 (60.0–80.0)67.0 (60.0–76.0)0.326
Rhythm0.178
ȃSinus rhythm20 (40.0%)42 (29.4%)
ȃAtrial fibrillation27 (54.0%)97 (67.8%)
ȃPaced3 (6.0%)4 (2.8%)
Prosthetic valve size21.0 (19.0–23.0)21.0 (19.0–21.0)0.032
ȃRange19–2519–25
ȃLarge size (≥23 mm)18 (36.0%)16 (11.2%)<0.001
Prosthetic valve type
ȃBileaflet49 (98.0%)137 (95.8)0.783
ȃMonoleaflet1 (2.0%)6 (4.2%)0.783
ȃSupra-annular type6 (12.0%)12 (8.4%)0.636
Time interval from previous AVR, years11.9 ± 9.219.3 ± 7.8<0.001
Reference EOA1.50 (1.30–1.70)1.40 (1.20–1.50)0.005
Projected indexed EOA, cm2/m20.90 (0.75–0.98)0.91 (0.76–0.99)0.991
ȃModerate patient–prosthesis mismatch12 (24.0%)36 (25.2%)>0.999
ȃSevere patient–prosthesis mismatch4 (8.0%)14 (9.8%)0.927
DVR status33 (66.0%)119 (83.2%)0.018
Limited motion of PAV leaflet0 (0%)84 (58.7%)<0.001
Concomitant aortic valve regurgitation0.091
ȃNo, mild45 (90.0%)139 (97.2%)
ȃModerate or severe5 (10.0%)4 (2.8%)
Method of diagnosis<0.001
ȃSurgical inspection13 (26.0%)94 (65.7%)
ȃCardiac CT37 (74.0%)49 (34.3%)
Indication for reoperation<0.001
ȃPAV obstruction0 (0%)94 (65.7%)
ȃPAV paravalvular regurgitation2 (4.0%)0 (0%)
ȃMitral or tricuspid valvular dysfunction11 (22.0%)29 (20.3%)
Redo aortic valve replacement2 (4.0%)94 (65.7%)<0.001
Echocardiography
ȃEOA, cm21.26 (1.00–1.50)0.80 (0.68–0.99)<0.001
ȃMeasured/reference EOA ratio0.86 ± 0.170.63 ± 0.18<0.001
ȃEOA/BSA, cm2/m20.76 (0.64–0.90)0.51 (0.43–0.64)<0.001
ȃLVEF, %62.0 (57.0–68.0)66.0 (61.0–71.0)0.009
ȃȃ < 45%5 (10.0%)9 (6.3%)0.580
ȃLVOT diameter, mm21.7 (19.9–23.3)20.5 (19.0–21.2)<0.001
ȃLVOT diameter difference from valve size0.45 (−0.30–1.00)−0.10 (−1.25–0.20)<0.001
ȃLVOT VTI, cm22.7 ± 4.822.7 ± 4.60.994
ȃPeak LVOT velocity, m/s1.09 (0.97–1.29)1.08 (0.93–1.21)0.296
ȃStroke volume, mL83.2 (72.0–103.1)69.2 (60.7–83.4)<0.001
ȃStroke volume index, mL/m249.4 (41.4–60.1)44.7 (38.6–51.7)0.019
ȃDynamic LVOT obstruction0 (0%)8 (5.6%)0.195
ȃAV VTI, cm66.1 (57.6–75.3)88.0 (72.1–101.3)<0.001
ȃPeak transprosthetic velocity, m/s3.33 (3.09–3.67)4.10 (3.63–4.71)<0.001
ȃPeak transprosthetic pressure, mmHg44.5 (38.0–54.0)67.0 (52.0–88.2)<0.001
ȃMean transprosthetic pressure, mmHg25.5 (21.0–30.0)37.4 (29.3–50.5)<0.001
ȃDoppler velocity index0.34 (0.29–0.41)0.26 (0.23–0.30)<0.001
ȃAcceleration time of aortic velocity, msec83.4 ± 22.598.9 ± 23.0<0.001
ȃEjection time, msec282.8 ± 44.2293.5 ± 35.10.128
ȃAcceleration time/ejection time ratio0.30 ± 0.070.34 ± 0.06<0.001
ȃCW Doppler jet contour<0.001
ȃȃTriangle, early peaking11 (22.0%)17 (11.9%)
ȃȃIntermediate38 (76.0%)78 (54.5%)
ȃȃRound, symmetrical1 (2.0%)48 (33.6%)

AVR, aortic valve replacement; CW, continuous wave; LVEF, left ventricular ejection fraction; LVOT, left ventricular outflow tract; VTI, velocity-time integral.

Regarding echocardiographic parameters, patients in the PAV obstruction group had a lower measured EOA [0.80 (0.68–0.99) vs. 1.26 (1.00–1.50) cm2; P < 0.001] and measured/reference EOA ratio (0.63 ± 0.18 vs. 0.86 ± 0.17; P < 0.001). The PAV obstruction group had smaller PAV size, reference EOA value [1.40 (1.20–1.50) vs. 1.50 (1.30–1.70) cm2; P = 0.005], and lower stroke volume index [SVi, 44.7 (38.6–51.7) vs. 49.4 (41.4–60.1) mL/m2, P = 0.019]. Patients with DVR (with mitral prosthesis) showed lower SVi than isolated AVR [44.9 (38.2–51.8) vs. 50.1 (40.6–60.1) mL/m2, P = 0.018]. The patients with surgical inspection had significantly higher mean transprosthetic PGs [40.0 (29.1–53.0) vs. 29.0 (23.0–35.8) mmHg, P < 0.001] and lower measured/reference EOA ratio (0.65 ± 0.18 vs. 0.73 ± 0.22, P = 0.007). Conventional Doppler parameters, including DVI, AT, AT/ET ratio, and Doppler jet contour differed significantly between the two groups (Table 1). The ICC between the LVOT diameter and reference annulus diameter of the PAV was not reliable (ICC = 0.566).

Diagnostic accuracy of Doppler parameters

The Doppler parameters and ROC analysis are described in Table 2. All conventional Doppler parameters could discriminate between PAV obstruction patients and no PAV obstructions, except for an AT/ET ratio over 0.37. The largest area under the ROC curve was observed with the measured EOA [0.849; 95% confidence interval (CI), 0.788–0.910; P < 0.001], followed by the measured/reference EOA ratio (0.840; 95% CI, 0.783–0.898; P < 0.001). The best cutoff value was 1.07 for the measured EOA and 0.71 for the measured/reference EOA ratio. The best cutoff value of measured EOA was different according to PAV size [1.25 for large PAV (≥23 mm), 0.80 for small PAV]. However, those of measured/reference EOA ratio was consistent as 0.71. When the measured/reference EOA ratio cutoff was set to 0.70 for convenience, the sensitivity and specificity were 68.5 and 84.0%, respectively. The measured/reference EOA ratio added incremental value over conventional Doppler parameters, such as mean transprosthetic PG, DVI, and AT/ET ratio (Figure 2). The inter- (ICC = 0.981; P < 0.001) and intra-observer variabilities (ICC = 0.984, P < 0.001) for the measured/reference EOA ratio showed excellent agreement (see Supplementary data online, Figure S2). The mean difference between the observers was −0.014 ± 0.039. The mean difference for the same observer was −0.002 ± 0.038.

Incremental value of measured/reference effective orifice ration for discriminating prosthetic aortic valve obstruction.
Figure 2

Incremental value of measured/reference effective orifice ration for discriminating prosthetic aortic valve obstruction.

Open in new tabDownload slide
Table 2
Open in new tab

Receiver operating characteristic curve analysis for PAV obstruction of Doppler parameters

AUC (95% CI)P ValueBest cutoff valueSensitivity, %Specificity, %Positive predictive value, %Negative predictive value, %
Measured EOA/reference EOA ratio0.840 (0.783–0.898)<0.0010.7173.482.092.952.5
ȃMeasured EOA/reference EOA ratio <0.70.763 (0.699–0.827)<0.001Positive68.584.092.548.3
Measured EOA vs. reference EOA
ȃMeasured EOA < reference EOA—1SD0.733 (0.658–0.807)<0.001Positive82.564.086.856.1
ȃMeasured EOA < reference EOA—2SD0.691 (0.635–0.747)<0.001Positive46.292.094.337.4
Measured EOA, cm20.849 (0.788–0.910)<0.0011.0783.270.088.859.3
ȃMeasured EOA <1.1 cm20.756 (0.684–0.828)<0.001Positive83.268.088.158.6
ȃMeasured EOA <0.8 cm20.728 (0.679–0.778)<0.001Positive49.796.097.340.0
Doppler velocity index0.766 (0.687–0.846)<0.0010.318164.086.554.2
ȃDoppler velocity index <0.350.644 (0.570–0.717)<0.001Positive88.740.080.855.6
ȃDoppler velocity index < 0.250.637 (0.576–0.698)<0.001Positive39.488.090.333.8
Rounded, symmetrical CW jet contour0.658 (0.614–0.701)<0.001Positive33.698.098.034.0
Acceleration time, msec0.691 (0.607–0.776)<0.0018866.466.084.840.7
ȃAcceleration time ≥80 msec0.615 (0.537–0.693)0.001Positive76.946.080.341.1
ȃAcceleration time >100 msec0.637 (0.570–0.705)<0.001Positive45.582.087.834.5
Acceleration time/LV ejection time ratio0.668 (0.575–0.762)<0.0010.3355.976.087.037.6
ȃAcceleration time/LV ejection time ratio >0.320.637 (0.561–0.713)<0.001Positive57.370.084.536.5
ȃAcceleration time/LV ejection time ratio >0.370.529 (0.465–0.593)0.200Positive23.882.079.127.3
AUC (95% CI)P ValueBest cutoff valueSensitivity, %Specificity, %Positive predictive value, %Negative predictive value, %
Measured EOA/reference EOA ratio0.840 (0.783–0.898)<0.0010.7173.482.092.952.5
ȃMeasured EOA/reference EOA ratio <0.70.763 (0.699–0.827)<0.001Positive68.584.092.548.3
Measured EOA vs. reference EOA
ȃMeasured EOA < reference EOA—1SD0.733 (0.658–0.807)<0.001Positive82.564.086.856.1
ȃMeasured EOA < reference EOA—2SD0.691 (0.635–0.747)<0.001Positive46.292.094.337.4
Measured EOA, cm20.849 (0.788–0.910)<0.0011.0783.270.088.859.3
ȃMeasured EOA <1.1 cm20.756 (0.684–0.828)<0.001Positive83.268.088.158.6
ȃMeasured EOA <0.8 cm20.728 (0.679–0.778)<0.001Positive49.796.097.340.0
Doppler velocity index0.766 (0.687–0.846)<0.0010.318164.086.554.2
ȃDoppler velocity index <0.350.644 (0.570–0.717)<0.001Positive88.740.080.855.6
ȃDoppler velocity index < 0.250.637 (0.576–0.698)<0.001Positive39.488.090.333.8
Rounded, symmetrical CW jet contour0.658 (0.614–0.701)<0.001Positive33.698.098.034.0
Acceleration time, msec0.691 (0.607–0.776)<0.0018866.466.084.840.7
ȃAcceleration time ≥80 msec0.615 (0.537–0.693)0.001Positive76.946.080.341.1
ȃAcceleration time >100 msec0.637 (0.570–0.705)<0.001Positive45.582.087.834.5
Acceleration time/LV ejection time ratio0.668 (0.575–0.762)<0.0010.3355.976.087.037.6
ȃAcceleration time/LV ejection time ratio >0.320.637 (0.561–0.713)<0.001Positive57.370.084.536.5
ȃAcceleration time/LV ejection time ratio >0.370.529 (0.465–0.593)0.200Positive23.882.079.127.3

AVR, aortic valve replacement; CW, continuous wave; LVEF, left ventricular ejection fraction; LVOT, left ventricular outflow tract; VTI, velocity-time integral.

Table 2
Open in new tab

Receiver operating characteristic curve analysis for PAV obstruction of Doppler parameters

AUC (95% CI)P ValueBest cutoff valueSensitivity, %Specificity, %Positive predictive value, %Negative predictive value, %
Measured EOA/reference EOA ratio0.840 (0.783–0.898)<0.0010.7173.482.092.952.5
ȃMeasured EOA/reference EOA ratio <0.70.763 (0.699–0.827)<0.001Positive68.584.092.548.3
Measured EOA vs. reference EOA
ȃMeasured EOA < reference EOA—1SD0.733 (0.658–0.807)<0.001Positive82.564.086.856.1
ȃMeasured EOA < reference EOA—2SD0.691 (0.635–0.747)<0.001Positive46.292.094.337.4
Measured EOA, cm20.849 (0.788–0.910)<0.0011.0783.270.088.859.3
ȃMeasured EOA <1.1 cm20.756 (0.684–0.828)<0.001Positive83.268.088.158.6
ȃMeasured EOA <0.8 cm20.728 (0.679–0.778)<0.001Positive49.796.097.340.0
Doppler velocity index0.766 (0.687–0.846)<0.0010.318164.086.554.2
ȃDoppler velocity index <0.350.644 (0.570–0.717)<0.001Positive88.740.080.855.6
ȃDoppler velocity index < 0.250.637 (0.576–0.698)<0.001Positive39.488.090.333.8
Rounded, symmetrical CW jet contour0.658 (0.614–0.701)<0.001Positive33.698.098.034.0
Acceleration time, msec0.691 (0.607–0.776)<0.0018866.466.084.840.7
ȃAcceleration time ≥80 msec0.615 (0.537–0.693)0.001Positive76.946.080.341.1
ȃAcceleration time >100 msec0.637 (0.570–0.705)<0.001Positive45.582.087.834.5
Acceleration time/LV ejection time ratio0.668 (0.575–0.762)<0.0010.3355.976.087.037.6
ȃAcceleration time/LV ejection time ratio >0.320.637 (0.561–0.713)<0.001Positive57.370.084.536.5
ȃAcceleration time/LV ejection time ratio >0.370.529 (0.465–0.593)0.200Positive23.882.079.127.3
AUC (95% CI)P ValueBest cutoff valueSensitivity, %Specificity, %Positive predictive value, %Negative predictive value, %
Measured EOA/reference EOA ratio0.840 (0.783–0.898)<0.0010.7173.482.092.952.5
ȃMeasured EOA/reference EOA ratio <0.70.763 (0.699–0.827)<0.001Positive68.584.092.548.3
Measured EOA vs. reference EOA
ȃMeasured EOA < reference EOA—1SD0.733 (0.658–0.807)<0.001Positive82.564.086.856.1
ȃMeasured EOA < reference EOA—2SD0.691 (0.635–0.747)<0.001Positive46.292.094.337.4
Measured EOA, cm20.849 (0.788–0.910)<0.0011.0783.270.088.859.3
ȃMeasured EOA <1.1 cm20.756 (0.684–0.828)<0.001Positive83.268.088.158.6
ȃMeasured EOA <0.8 cm20.728 (0.679–0.778)<0.001Positive49.796.097.340.0
Doppler velocity index0.766 (0.687–0.846)<0.0010.318164.086.554.2
ȃDoppler velocity index <0.350.644 (0.570–0.717)<0.001Positive88.740.080.855.6
ȃDoppler velocity index < 0.250.637 (0.576–0.698)<0.001Positive39.488.090.333.8
Rounded, symmetrical CW jet contour0.658 (0.614–0.701)<0.001Positive33.698.098.034.0
Acceleration time, msec0.691 (0.607–0.776)<0.0018866.466.084.840.7
ȃAcceleration time ≥80 msec0.615 (0.537–0.693)0.001Positive76.946.080.341.1
ȃAcceleration time >100 msec0.637 (0.570–0.705)<0.001Positive45.582.087.834.5
Acceleration time/LV ejection time ratio0.668 (0.575–0.762)<0.0010.3355.976.087.037.6
ȃAcceleration time/LV ejection time ratio >0.320.637 (0.561–0.713)<0.001Positive57.370.084.536.5
ȃAcceleration time/LV ejection time ratio >0.370.529 (0.465–0.593)0.200Positive23.882.079.127.3

AVR, aortic valve replacement; CW, continuous wave; LVEF, left ventricular ejection fraction; LVOT, left ventricular outflow tract; VTI, velocity-time integral.

Accuracy of novel diagnostic algorithm using measured/reference EOA ratio

The diagnostic algorithm models and their ROC analyses are presented in Table 3 and Graphical abstract B and C. All diagnostic models showed significant accuracy for detecting PAV obstruction. The novel model had the highest AUC value (0.763, P < 0.001), 68.5% sensitivity, and 84.0% specificity. Compared with the 2009 algorithm, which did not include reference EOA values, the novel model had significantly higher AUC values (0.763 vs. 0.642; P = 0.003). The accuracies of the novel and 2016 algorithms were not significantly different (AUC, 0.763 vs. 0.731; P = 0.309). Despite its simplicity, the novel algorithm had a significantly higher AUC value than the 2016 recommendation (0.788 vs. 0.642; P = 0.019) in patients with a large PAV (≥23 mm) (Table 3). The sensitivity analyses with patients performed surgical inspection only and patients with limited leaflet motion showed similar results to those of the main findings (see Supplementary data online, Table S3).

Table 3
Open in new tab

Receiver operating characteristic curve analysis of diagnostic algorithm for PAV obstruction in specified subgroups

AUC (95% CI)P valueP value (vs. novel)Sensitivity, %Specificity, %Positive predictive value, %Negative predictive value, %
Overall study population, n = 193
ȃNovel algorithm0.763 (0.699–0.827)<0.00168.584.092.548.3
ȃ2016 recommendation0.731 (0.658–0.804)<0.0010.30976.270.087.950.7
ȃ2009 recommendation0.642 (0.584–0.700)<0.0010.00338.590.091.733.8
Large prosthetic AV (≥23 mm), n = 34
ȃNovel algorithm0.788 (0.649–0.927)<0.00168.888.984.676.2
ȃ2016 recommendation0.642 (0.478–0.807)0.0510.01956.272.264.365.0
ȃ2009 recommendation0.510 (0.378–0.643)0.4480.00318.883.350.053.6
Low gradient (mean PG <40 mmHg), n = 129
ȃNovel algorithm0.668 (0.592–0.744)<0.00150.083.783.350.6
ȃ2016 recommendation0.688 (0.606–0.771)<0.0010.59166.271.479.156.5
ȃ2009 recommendation0.572 (0.512–0.632)0.0180.03322.591.881.842.1
DVR status, n = 152
ȃNovel diagnostic algorithm0.758 (0.679–0.836)<0.00169.781.893.342.9
ȃ2016 recommendation0.724 (0.634–0.814)<0.0010.36478.266.789.445.8
ȃ2009 recommendation0.649 (0.578–0.721)<0.0010.03242.087.992.629.6
Aortic valve replacement only, n = 41
ȃNovel diagnostic algorithm0.754 (0.627–0.880)<0.00162.588.288.262.5
ȃ2016 recommendation0.716 (0.574–0.857)0.0040.58266.776.580.061.9
ȃ2009 recommendation0.575 (0.474–0.676)0.0970.01420.894.183.345.7
Atrial fibrillation, n = 124
ȃNovel algorithm0.724 (0.631–0.817)<0.00167.077.891.539.6
ȃ2016 recommendation0.662 (0.558–0.767)<0.0010.16073.259.386.638.1
ȃ2009 recommendation0.620 (0.543–0.697)0.0080.06835.188.991.927.6
AUC (95% CI)P valueP value (vs. novel)Sensitivity, %Specificity, %Positive predictive value, %Negative predictive value, %
Overall study population, n = 193
ȃNovel algorithm0.763 (0.699–0.827)<0.00168.584.092.548.3
ȃ2016 recommendation0.731 (0.658–0.804)<0.0010.30976.270.087.950.7
ȃ2009 recommendation0.642 (0.584–0.700)<0.0010.00338.590.091.733.8
Large prosthetic AV (≥23 mm), n = 34
ȃNovel algorithm0.788 (0.649–0.927)<0.00168.888.984.676.2
ȃ2016 recommendation0.642 (0.478–0.807)0.0510.01956.272.264.365.0
ȃ2009 recommendation0.510 (0.378–0.643)0.4480.00318.883.350.053.6
Low gradient (mean PG <40 mmHg), n = 129
ȃNovel algorithm0.668 (0.592–0.744)<0.00150.083.783.350.6
ȃ2016 recommendation0.688 (0.606–0.771)<0.0010.59166.271.479.156.5
ȃ2009 recommendation0.572 (0.512–0.632)0.0180.03322.591.881.842.1
DVR status, n = 152
ȃNovel diagnostic algorithm0.758 (0.679–0.836)<0.00169.781.893.342.9
ȃ2016 recommendation0.724 (0.634–0.814)<0.0010.36478.266.789.445.8
ȃ2009 recommendation0.649 (0.578–0.721)<0.0010.03242.087.992.629.6
Aortic valve replacement only, n = 41
ȃNovel diagnostic algorithm0.754 (0.627–0.880)<0.00162.588.288.262.5
ȃ2016 recommendation0.716 (0.574–0.857)0.0040.58266.776.580.061.9
ȃ2009 recommendation0.575 (0.474–0.676)0.0970.01420.894.183.345.7
Atrial fibrillation, n = 124
ȃNovel algorithm0.724 (0.631–0.817)<0.00167.077.891.539.6
ȃ2016 recommendation0.662 (0.558–0.767)<0.0010.16073.259.386.638.1
ȃ2009 recommendation0.620 (0.543–0.697)0.0080.06835.188.991.927.6

AVR, aortic valve replacement; CW, continuous wave; LVEF, left ventricular ejection fraction; LVOT, left ventricular outflow tract; VTI, velocity-time integral.

Table 3
Open in new tab

Receiver operating characteristic curve analysis of diagnostic algorithm for PAV obstruction in specified subgroups

AUC (95% CI)P valueP value (vs. novel)Sensitivity, %Specificity, %Positive predictive value, %Negative predictive value, %
Overall study population, n = 193
ȃNovel algorithm0.763 (0.699–0.827)<0.00168.584.092.548.3
ȃ2016 recommendation0.731 (0.658–0.804)<0.0010.30976.270.087.950.7
ȃ2009 recommendation0.642 (0.584–0.700)<0.0010.00338.590.091.733.8
Large prosthetic AV (≥23 mm), n = 34
ȃNovel algorithm0.788 (0.649–0.927)<0.00168.888.984.676.2
ȃ2016 recommendation0.642 (0.478–0.807)0.0510.01956.272.264.365.0
ȃ2009 recommendation0.510 (0.378–0.643)0.4480.00318.883.350.053.6
Low gradient (mean PG <40 mmHg), n = 129
ȃNovel algorithm0.668 (0.592–0.744)<0.00150.083.783.350.6
ȃ2016 recommendation0.688 (0.606–0.771)<0.0010.59166.271.479.156.5
ȃ2009 recommendation0.572 (0.512–0.632)0.0180.03322.591.881.842.1
DVR status, n = 152
ȃNovel diagnostic algorithm0.758 (0.679–0.836)<0.00169.781.893.342.9
ȃ2016 recommendation0.724 (0.634–0.814)<0.0010.36478.266.789.445.8
ȃ2009 recommendation0.649 (0.578–0.721)<0.0010.03242.087.992.629.6
Aortic valve replacement only, n = 41
ȃNovel diagnostic algorithm0.754 (0.627–0.880)<0.00162.588.288.262.5
ȃ2016 recommendation0.716 (0.574–0.857)0.0040.58266.776.580.061.9
ȃ2009 recommendation0.575 (0.474–0.676)0.0970.01420.894.183.345.7
Atrial fibrillation, n = 124
ȃNovel algorithm0.724 (0.631–0.817)<0.00167.077.891.539.6
ȃ2016 recommendation0.662 (0.558–0.767)<0.0010.16073.259.386.638.1
ȃ2009 recommendation0.620 (0.543–0.697)0.0080.06835.188.991.927.6
AUC (95% CI)P valueP value (vs. novel)Sensitivity, %Specificity, %Positive predictive value, %Negative predictive value, %
Overall study population, n = 193
ȃNovel algorithm0.763 (0.699–0.827)<0.00168.584.092.548.3
ȃ2016 recommendation0.731 (0.658–0.804)<0.0010.30976.270.087.950.7
ȃ2009 recommendation0.642 (0.584–0.700)<0.0010.00338.590.091.733.8
Large prosthetic AV (≥23 mm), n = 34
ȃNovel algorithm0.788 (0.649–0.927)<0.00168.888.984.676.2
ȃ2016 recommendation0.642 (0.478–0.807)0.0510.01956.272.264.365.0
ȃ2009 recommendation0.510 (0.378–0.643)0.4480.00318.883.350.053.6
Low gradient (mean PG <40 mmHg), n = 129
ȃNovel algorithm0.668 (0.592–0.744)<0.00150.083.783.350.6
ȃ2016 recommendation0.688 (0.606–0.771)<0.0010.59166.271.479.156.5
ȃ2009 recommendation0.572 (0.512–0.632)0.0180.03322.591.881.842.1
DVR status, n = 152
ȃNovel diagnostic algorithm0.758 (0.679–0.836)<0.00169.781.893.342.9
ȃ2016 recommendation0.724 (0.634–0.814)<0.0010.36478.266.789.445.8
ȃ2009 recommendation0.649 (0.578–0.721)<0.0010.03242.087.992.629.6
Aortic valve replacement only, n = 41
ȃNovel diagnostic algorithm0.754 (0.627–0.880)<0.00162.588.288.262.5
ȃ2016 recommendation0.716 (0.574–0.857)0.0040.58266.776.580.061.9
ȃ2009 recommendation0.575 (0.474–0.676)0.0970.01420.894.183.345.7
Atrial fibrillation, n = 124
ȃNovel algorithm0.724 (0.631–0.817)<0.00167.077.891.539.6
ȃ2016 recommendation0.662 (0.558–0.767)<0.0010.16073.259.386.638.1
ȃ2009 recommendation0.620 (0.543–0.697)0.0080.06835.188.991.927.6

AVR, aortic valve replacement; CW, continuous wave; LVEF, left ventricular ejection fraction; LVOT, left ventricular outflow tract; VTI, velocity-time integral.

Discussion

The main study findings were as follows: (i) patients with PAV obstruction had smaller measured EOA values and smaller PAV sizes than without PAV obstruction; (ii) the measured/reference EOA ratio had an incremental value over the conventional Doppler parameters; (iii) diagnostic models using reference EOA values had higher accuracy than previous model based on the 2009 recommendations;9 and (iv) for large valves, the novel diagnostic algorithm model using the measured EOA/reference EOA ratio had a higher AUC than the previous algorithm based on the 2016 recommendation.11 Using reference EOA values, a more accurate identification of PAV dysfunction may be possible.

Demographic and echocardiographic characteristics of patients with PAV obstruction

In this study, more patients with PAV obstruction were female, had smaller PAV sizes, and more DVR. Most of PAV obstruction was derived from pannus. The detailed mechanism of pannus formation has not been fully demonstrated, but previous studies reported young age, small sized PAV, and DVR were the risk factors for pannus formation and described the inflammatory reaction and changes of subaortic hemodynamics as possible explanation.17,18

Similar to native valve stenosis, most conventional Doppler parameters are applicable to PAV, but some inherent caveats and pitfalls caused by the nature of prosthetic valves exist.19,20 Even a normally functioning PAV has some degree of blood flow obstruction, resulting in an elevated transprosthetic PG. Among all parameters, measured EOA and measured/reference EOA ratio had the highest AUC in this study. The ROC curve–derived optimal cutoff values for each Doppler parameter were similar to previous recommendations.9,11 The AUC of the AT/ET ratio was relatively low compared with that reported by Zekry et al.7 Our study included more patients with atrial fibrillation and a larger study population. Even multiple cardiac cycles were averaged; atrial fibrillation impacted ET and caused inaccuracies.

As the LVOT diameter is squared in the EOA measurement by the CE, small measurement errors are significant in the EOA calculation. In this study, there was no significant difference in the VTILVOT, but the LVOT diameter was significantly lower in the PAV obstruction group. There was significant difference in LVOT diameter between reference size and measurement. As previously reported, substitution of the LVOT diameter with the labelled prosthesis size was found to be inappropriate.21 Circumferential or focal pannus formation may lead to small LVOT size and lower EOA value, without leaflet motion limitations.

Incremental value of reference EOA

Two-dimensional TTE remains the method of choice for evaluating PAV function, but there are unmet needs to be overcome. Measured EOA is a single reliable parameter for detecting PAV obstruction, but it has different cutoff values according to size of PAV. On the other hand, measured/reference EOA ratio had consistent cutoff value, regardless of PAV size. Therefore measured/reference EOA ratio was considered to be more clinically useful. A recent recommendation described the efficacy of comparing the measured EOA to the reference EOA rather than using fixed cut-off values of measured EOA.2,4,12 Use of the reference EOA value is theoretically correct as every PAV has its own hemodynamic characteristics, but there are a few validation studies in real-world practice. As the recent expert consensus for optimal valve prosthesis selection insisted on presenting each prosthesis’s hemodynamic performance uniformly,22 the use of reference values for evaluating PAV performance is expected to increase. Our study showed incremental values of using reference EOA over conventional Doppler parameters. Furthermore, diagnostic algorithm models that included reference EOA was more accurate than the 2009 recommendation, which did not include the reference EOA. In addition, our algorithm showed a higher AUC value for large-sized PAV than 2016 recommendations. Larger-sized PAV had larger reference EOA values and higher standard deviations and might result in wider reference values and might have reduced the diagnostic accuracy in 2016 recommendations. These results probably related with the higher SD values in larger sized PAV.

Limitations

The present study has some limitations. First, it was a retrospective study from a single tertiary center, which inherently limits the generalizability of its results. However, this study had a larger population than previous studies and various subgroups, and the sensitivity analysis showed consistent results. Second, there may be errors in the EOA calculation using the continuity equation. Measuring the LVOT diameter may be challenging in patients with prosthetic valves. Since our study had a high proportion of double-valve replacements, these errors might be magnified. To overcome this limitation, we performed intra-and interobserver variability analyses and demonstrated consistency in the EOA values. However, inter- and intraobserver reproducibility was studied from repeated measurement of same images rather than repeated acquisition and measurement. Therefore, variability might be underestimated. Third, some subgroup analyses included a small number of subjects. However, as we had a larger population, even the small subgroup had similar numbers compared with previous studies. Fourth, although surgical confirmation of the presence of pannus would be the gold standard, not all patients were surgically confirmed in our study. In addition, patients with strong suspicion of PAV obstruction had undergone CT or surgery, which might lead to ascertainment bias. Although the accuracy of CT was validated in previous studies,11 there might be false positives or negatives. However, there were only small numbers of false positive or negative in CT, compared with surgical finding. We also performed a sensitivity analysis using only surgical data, which showed similar results to the main findings.

Conclusion

The ratio of measured/reference EOA has incremental value over conventional Doppler parameters. This parameter might be helpful for distinguishing PAV obstruction. The simple novel diagnostic algorithm model using the measured/reference EOA ratio had higher accuracy than the algorithm from previous recommendations, especially for large-sized PAV.

Supplementary data

Supplementary data are available at European Heart Journal - Cardiovascular Imaging online.

Acknowledgements

MID (Medical Illustration & Design), a part of the Medical Research Support Services of Yonsei University College of Medicine, for providing excellent support with medical illustration.

Funding

None declared.

Data availability

Data would be available upon reasonable request.

References

1

Nkomo
VT
,
Gardin
JM
,
Skelton
TN
,
Gottdiener
JS
,
Scott
CG
,
Enriquez-Sarano
M
.
Burden of valvular heart diseases: a population-based study
.
Lancet
2006
;
368
:
1005
–
11
.

Google Scholar

Crossref
Search ADS
PubMed

 
2

Pibarot
P
,
Dumesnil
JG
.
Prosthetic heart valves: selection of the optimal prosthesis and long-term management
.
Circulation
2009
;
119
:
1034
–
48
.

Google Scholar

Crossref
Search ADS
PubMed

 
3

Deviri
E
,
Sareli
P
,
Wisenbaugh
T
,
Cronje
SL
.
Obstruction of mechanical heart valve prostheses: clinical aspects and surgical management
.
J Am Coll Cardiol
1991
;
17
:
646
–
50
.

Google Scholar

Crossref
Search ADS
PubMed

 
4

Pibarot
P
,
Dumesnil
JG
.
Doppler echocardiographic evaluation of prosthetic valve function
.
Heart
2012
;
98
:
69
–
78
.

Google Scholar

Crossref
Search ADS
PubMed

 
5

Zoghbi
WA
,
Farmer
KL
,
Soto
JG
,
Nelson
JG
,
Quinones
MA
.
Accurate noninvasive quantification of stenotic aortic valve area by Doppler echocardiography
.
Circulation
1986
;
73
:
452
–
9
.

Google Scholar

Crossref
Search ADS
PubMed

 
6

Chafizadeh
ER
,
Zoghbi
WA
.
Doppler echocardiographic assessment of the st. Jude medical prosthetic valve in the aortic position using the continuity equation
.
Circulation
1991
;
83
:
213
–
23
.

Google Scholar

Crossref
Search ADS
PubMed

 
7

Ben Zekry
S
,
Saad
RM
,
Ozkan
M
,
Al Shahid
MS
,
Pepi
M
,
Muratori
M
et al.
Flow acceleration time and ratio of acceleration time to ejection time for prosthetic aortic valve function
.
JACC Cardiovasc Imaging
2011
;
4
:
1161
–
70
.

Google Scholar

Crossref
Search ADS
PubMed

 
8

Rothbart
RM
,
Castriz
JL
,
Harding
LV
,
Russo
CD
,
Teague
SM
.
Determination of aortic valve area by two-dimensional and Doppler echocardiography in patients with Normal and stenotic bioprosthetic valves
.
J Am Coll Cardiol
1990
;
15
:
817
–
24
.

Google Scholar

Crossref
Search ADS
PubMed

 
9

Zoghbi
WA
,
Chambers
JB
,
Dumesnil
JG
,
Foster
E
,
Gottdiener
JS
,
Grayburn
PA
et al.
Recommendations for evaluation of prosthetic valves with echocardiography and Doppler ultrasound: a report from the American Society of Echocardiography's guidelines and standards committee and the task force on prosthetic valves, developed in conjunction with the American College of Cardiology Cardiovascular Imaging Committee, cardiac imaging committee of the American Heart Association, the European association of echocardiography, a registered branch of the European Society of Cardiology, the Japanese Society of Echocardiography and the Canadian Society of Echocardiography, endorsed by the American College of Cardiology Foundation, American Heart Association, European Association of Echocardiography, a registered branch of the European Society of Cardiology, the Japanese Society of Echocardiography, and Canadian Society of Echocardiography
.
J Am Soc Echocardiogr
2009
;
22
:
975
–
1014
;
quiz 82–4
.

Google Scholar

Crossref
Search ADS
PubMed

 
10

Kim
JY
,
Suh
YJ
,
Han
K
,
Kim
YJ
,
Choi
BW
.
Diagnostic value of advanced imaging modalities for the detection and differentiation of prosthetic valve obstruction: a systematic review and meta-analysis
.
JACC Cardiovasc Imaging
2019
;
12
:
2182
–
92
.

Google Scholar

Crossref
Search ADS
PubMed

 
11

Lancellotti
P
,
Pibarot
P
,
Chambers
J
,
Edvardsen
T
,
Delgado
V
,
Dulgheru
R
et al.
Recommendations for the imaging assessment of prosthetic heart valves: a report from the European Association of cardiovascular imaging endorsed by the Chinese Society of Echocardiography, the inter-American society of echocardiography, and the Brazilian department of cardiovascular imaging
.
Eur Heart J Cardiovasc Imaging
2016
;
17
:
589
–
90
.

Google Scholar

Crossref
Search ADS
PubMed

 
12

Rajani
R
,
Mukherjee
D
,
Chambers
JB
.
Doppler echocardiography in normally functioning replacement aortic valves: a review of 129 studies
.
J Heart Valve Dis
2007
;
16
:
519
–
35
.

Google Scholar

PubMed
OpenURL Placeholder Text

 
13

Lang
RM
,
Badano
LP
,
Mor-Avi
V
,
Afilalo
J
,
Armstrong
A
,
Ernande
L
et al.
Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging
.
J Am Soc Echocardiogr
2015
;
28
:
1
–
39 e14
.

Google Scholar

Crossref
Search ADS
PubMed

 
14

Suh
YJ
,
Lee
S
,
Im
DJ
,
Chang
S
,
Hong
YJ
,
Lee
HJ
et al.
Added value of cardiac computed tomography for evaluation of mechanical aortic valve: emphasis on evaluation of pannus with surgical findings as standard reference
.
Int J Cardiol
2016
;
214
:
454
–
60
.

Google Scholar

Crossref
Search ADS
PubMed

 
15

Symersky
P
,
Budde
RP
,
de Mol
BA
,
Prokop
M
.
Comparison of multidetector-row computed tomography to echocardiography and fluoroscopy for evaluation of patients with mechanical prosthetic valve obstruction
.
Am J Cardiol
2009
;
104
:
1128
–
34
.

Google Scholar

Crossref
Search ADS
PubMed

 
16

DeLong
ER
,
DeLong
DM
,
Clarke-Pearson
DL
.
Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach
.
Biometrics
1988
;
44
:
837
–
45
.

Google Scholar

Crossref
Search ADS
PubMed

 
17

Choi
JW
,
Hwang
HY
,
Lee
Y
,
Sohn
SH
,
Kim
KH
,
Kim
KB
et al.
Incidence and risk factors of pannus after mechanical aortic valve replacement
.
Thorac Cardiovasc Surg
2022
;
70
:
182
–
188
.

Google Scholar

PubMed
OpenURL Placeholder Text

 
18

Teshima
H
,
Hayashida
N
,
Yano
H
,
Nishimi
M
,
Tayama
E
,
Fukunaga
S
et al.
Obstruction of St Jude medical valves in the aortic position: histology and immunohistochemistry of pannus
.
J Thorac Cardiovasc Surg
2003
;
126
:
401
–
7
.

Google Scholar

Crossref
Search ADS
PubMed

 
19

Burstow
DJ
,
Nishimura
RA
,
Bailey
KR
,
Reeder
GS
,
Holmes
DR
Jr,
Seward
JB
et al.
Continuous wave Doppler echocardiographic measurement of prosthetic valve gradients. A simultaneous Doppler-catheter correlative study
.
Circulation
1989
;
80
:
504
–
14
.

Google Scholar

Crossref
Search ADS
PubMed

 
20

Bach
DS
.
Echo/Doppler evaluation of hemodynamics after aortic valve replacement: principles of interrogation and evaluation of high gradients
.
JACC Cardiovasc Imaging
2010
;
3
:
296
–
304
.

Google Scholar

Crossref
Search ADS
PubMed

 
21

Pibarot
P
,
Honos
GN
,
Durand
LG
,
Dumesnil
JG
.
Substitution of left ventricular outflow tract diameter with prosthesis size is inadequate for calculation of the aortic prosthetic valve area by the continuity equation
.
J Am Soc Echocardiogr
1995
;
8
:
511
–
7
.

Google Scholar

Crossref
Search ADS
PubMed

 
22

Durko
AP
,
Pibarot
P
,
Atluri
P
,
Bapat
V
,
Cameron
DE
,
Casselman
FPA
et al.
Essential information on surgical heart valve characteristics for optimal valve prosthesis selection: expert consensus document from the European Association for Cardio-Thoracic Surgery (EACTS)-the Society of Thoracic Surgeons (STS)-American Association for Thoracic Surgery (AATS) valve labelling task force
.
Ann Thorac Surg
2021
;
111
:
314
–
26
.

Google Scholar

Crossref
Search ADS
PubMed

 

Author notes

Conflict of interest: None declared.

© The Author(s) 2022. Published by Oxford University Press on behalf of the European Society of Cardiology. All rights reserved. For permissions, please email: [email protected].
This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://dbpia.nl.go.kr/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Topic:
Issue Section:
Original Paper
Download all slides
  • Supplementary data

    • Supplementary data

    jeac206_Supplementary_Data - docx file