A better method to evaluate the reliability of echocardiography for assessment of pulmonary hypertension: comparison of tricuspid regurgitant spectrum quality grading and tricuspid valve regurgitation degree

Abstract

Background: Transthoracic echocardiography (TTE) is recommended as the most important noninvasive screening tool for the diagnosis of pulmonary hypertension (PH), sonographers usually measure the volume of regurgitant flow rather than evaluating the spectral quality, so physicians will determine whether the ultrasound measurements of pulmonary arterial systolic pressure (US-PASP) are reliable based on the volume of tricuspid regurgitation (TR). Therefore, for the first time, we grade the quality of TR spectrum (TRS) based on its integrity and clarity, aiming to assess clinical application value of different tricuspid regurgitant spectrum quality grades (TR-SQG), and investigate whether the accuracy of US-PASP is more trustworthy than TR.

Methods: We retrospectively analyzed 108 patients with chronic thromboembolic PH (CTEPH) to compare the correlation and agreement between US-PASP and right heart catheterization measurements of PASP (RHC-PASP). TR area (TRA) and TRS were measured in each patient, and TR-SQG was performed.

Results: The correlation coefficients between US-PASP and RHC-PASP were r=0.622 (P<0.001), r=0.754 (P<0.001), r=0.595 (P<0.001) in mild, moderate, severe TR, and r=0.301 (P=0.135), r=0.747 (P<0.001), r=0.739 (P<0.001), r=0.828 (P<0.001) in TR-SQG I-IV, respectively. Bland-Altman analysis revealed the mean biases of 5.05, 3.06, 7.62 mmHg in mild, moderate, severe TR, and -16.47, -8.07, 1.82, 6.09 mmHg in TR-SQG I-IV, respectively. In mild TR with the TR-SQG III and IV, the correlation coefficients between US-PASP and RHC-PASP were r=0.779 (P<0.001), intraclass correlation coefficient (ICC) =0.774, paired t-test P=0.160, respectively; and the consistency was significantly higher than that of mild TR without considering TR-SQG. In moderate TR with the TR-SQG III and IV, the r=0.749, ICC =0.746, paired t-test P=0.298 between US-PASP and RHC-PASP.

Conclusions: The US-PASP with TR-SQG III or IV is trustworthy, and its accuracy and consistency are better than those predicted by the traditional severity of TR. The establishment of the ultrasound evaluation system of TR-SQG helps clinicians to judge whether the US-PASP is accurate, credible, and reliable.

Keywords: Pulmonary hypertension (PH); echocardiography; right heart catheterization (RHC); tricuspid regurgitant spectrum; tricuspid regurgitation velocity (TRV).

Figure 1

Classification of the TR signal quality using CW Doppler. (A) TR-SQG I; (B) TR-SQG II; (C) TR-SQG III; (D) TR-SQG IV. TR, tricuspid regurgitation; CW, continuous-wave; TR-SQG, tricuspid regurgitant spectrum quality grade.

Figure 2

Comparison of PASP as estimated by TTE and RHC in mild, moderate, and severe TR. (A-C) Paired t-tests in severity TR. ns, no significance; **, P<0.01. (D-F) Linear regression analysis plots of invasive and noninvasive values of PASP. r, correlation coefficient (Pearson). (G-I) Bland-Altman plots of US-PASP measured using TTE and RHC-PASP measured using RHC. The dotted lines indicate the mean biases, whereas the whole lines represent the upper and lower 95% limits of agreement. TR, tricuspid regurgitation; PASP, pulmonary arterial systolic pressure; US-PASP, ultrasound measurements of PASP; RHC-PASP, right heart catheterization measurements of PASP; TTE, transthoracic echocardiography; RHC, right heart catheterization.

Figure 3

Comparison of PASP as estimated by TTE and RHC in TR-SQG I–IV. (A-D) Paired t-tests in TR-SQG I–IV. ns, no significance; **, P<0.01; ***, P<0.001. (E-H) Linear regression analysis plots of invasive and noninvasive PASP values. r, Pearson correlation coefficient. (I-L) Bland-Altman analysis reveal mean biases (dashed lines) and 95% limits of agreement ranges (full lines). PASP, pulmonary arterial systolic pressure; US-PASP, ultrasound measurements of PASP; RHC-PASP, right heart catheterization measurements of PASP; TTE, transthoracic echocardiography; RHC, right heart catheterization; TR-SQG, tricuspid regurgitant spectrum quality grade.

Figure 4

Distribution of US-PASP measurements. TR, tricuspid regurgitation; TR-SQG, tricuspid regurgitant spectrum quality grade; US-PASP, ultrasound measurements of pulmonary arterial systolic pressure.

Figure 5

Comparison of PASP as estimated by TTE and RHC in severe TR [(A-C) before BPA; (D-F) after BPA] as well as TR-SQG IV [(G-I) before BPA; (J-L) after BPA], including paired t-tests, Pearson analysis and Bland-Altman plots (dashed lines reveal mean biases, full lines reveal 95% limits of agreement ranges). ns, no significance; *, P<0.05; **, P<0.01. PASP, pulmonary arterial systolic pressure; TR, tricuspid regurgitation; BPA, balloon pulmonary angioplasty; US-PASP, ultrasound measurements of PASP; RHC-PASP, right heart catheterization measurements of PASP; TR-SQG, tricuspid regurgitant spectrum quality grade; TTE, transthoracic echocardiography; RHC, right heart catheterization.