ECHOCARDIOGRAPHY AND PULMONARY HYPERTENSION • Echocardiographic Biventricular Coupling Index to Predict Precapillary Pulmonary Hypertension
Pulmonary hypertension (PH) is a common condition and is associated with poor prognosis. While pulmonary vasodilators have demonstrated clinical efﬁcacy in precapillary PH (type 1 PH), they can be detrimental in patients with postcapillary PH. Therefore, the correct diagnosis according to the current classiﬁcation is mandatory for the choice of the speciﬁc therapeutic strategies.
Right heart catheterization (RHC) remains the gold standard in the diagnostic workup. However, transthoracic echocardiography is currently part of the screening process and is increasingly used in the longitudinal follow-up of patients with PH. Estimated systolic pulmonary artery pressure (sPAP) is recommended to deﬁne the probability of PH in symptomatic patients but does not discriminate between pre- and postcapillary PH.
Patients without isolated postcapillary PH (iPC-PH), and mainly patients with precapillary PH, show a well-known modiﬁed physiology of the right ventricular characterized by a change from a preload- to an afterload-dependent condition.
RV afterload is characterized by two components, pulmonary vascular resistance and arterial elastance – an important index of arterial elasticity. RV stroke work index (RVSWI) accounts for both resistance and capacitance and has been used as an index of RV performance.
The right ventricle responds to the chronic exposure of increased afterload with an increase in contractility and subsequent progressive dilation. The natural history of patients affected by pulmonary arterial hypertension is characterized by initial compensatory phase of stable or increased SV, followed by a progressive RV decrease in stroke volume (SV).
Noninvasive estimation of RVSWI by echocardiography (echocardiographic RVSWI [eRVSWI]) has been validated in adult and pediatric populations. Similarly, E/E’ ratio as a surrogate marker of left ventricular (LV) ﬁlling pressure is generally in the lower range in precapillary PH.
The authors hypothesized that the ratio between these two measures could make it possible to distinguish between precapillary and postcapillary PH and named this novel echocardiographic index as ‘‘biventricular coupling index’’ (BCI), calculated as eRVSWI/(E/E’).
All patients undergoing complete transthoracic echocardiography and RHC for all clinical indications were prospectively and consecutively enrolled between August 2014 until December 2018.
Patients with a time interval between echocardiography and RHC of >6 hours or who received infusion of ﬂuids or diuretic administration between the two examinations were excluded. Further exclusion criteria were age < 18 years, uncorrected intra- or extracardiac shunts, and poor echocardiographic image quality.
A wide cohort of patients selected according to the same criteria was obtained from a retrospective registry and used as the validation cohort. The validation cohort was also used to compare the performance of the BCI with the previously described indices for the noninvasive discrimination of pre- versus postcapillary PH.
Standard Echocardiographic Assessment and Hemodynamic Deﬁnitions
All patients underwent complete transthoracic echocardiography, including a dedicated protocol for the acquisition of all parameters necessary for hemodynamic evaluation, according to international guidelines. Measurements were obtained both online and ofﬂine using dedicated software (Suite Estensa; Esaote, Genoa, Italy), and operators were blinded to the results of invasive evaluation.
eRVSWI was deﬁned in our study as follows:
eRVSWI = 0.0136 x SVi x (sPAP – RAP) = 0.0136 x SVi x [(RVSP + RAP) – RAP]
= 0.0136 x SVi x RVSP
where SVi is the SV index, and sPAP was calculated as the sum of RV systolic pressure (RVSP), derived from peak tricuspid regurgitation velocity using the modiﬁed Bernoulli equation, and estimated right atrial pressure (RAP).
SV was calculated using the integral of pulsed-wave Doppler at the level of the LV outﬂow tract.
E 0from the medial mitral valve annulus was used for the calculation of E/E 0 ratio. BCI was then deﬁned as follows:
BCI = E/E’
PH was deﬁned as mPAP ≤ 25 mm Hg, precapillary PH as PH with pulmonary capillary wedge pressure (PCWP) ≤ 15 mm Hg, iPC-PH as PH with PCWP > 15 mm Hg and diastolic pressure gradient (diastolic PAP – PCWP) < 7 mm Hg and/or pulmonary vascular resistance ≤ 3 Wood units, and combined postcapillary and precapillary PH (cPCPH) as PH with PCWP > 15 mm Hg and diastolic pressure gradient (diastolic PAP – PCWP) ≥ 7 mm Hg and/or pulmonary vascular resistance > 3 WU.
Among 354 screened patients, 334 patients (94.3%) had no missing data for the variables of interest and were thus included in the derivation cohort. Medial E/E 0 was not available in 20 patients (5.6%), and RVSP was not available in 19 patients (5.4%).
At RHC, 226 patients (67.7%) had PH. Among patients with PH, 82 had precapillary PH (36.3%), while 79 had iPC-PH (35.0%) and 65 had cPC-PH (28.8%).
Echocardiographic estimates of sPAP and SVi showed strong to moderate correlations with invasive sPAP and SVi.
eRVSWI demonstrated a good correlation with invasive RVSWI (r = 0.696, P < .001), while E/E 0ratio was weakly correlated with PCWP (r = 0.30, P < .001).
Finally, BCI showed high diagnostic accuracy in the prediction of precapillary PH in the derivation cohort (AUC, 0.82; 95% CI, 0.78-0.89; P < .001), and the optimal cut point for the diagnosis of precapillary PH, according to the best performance on ROC analysis, was set at 1.9, with sensitivity of 73%, speciﬁcity of 78%, NPVof 90%, and PPVof 52% in the derivation cohort (
Diagnostic Performance of the BCI in the Validation Group
In the validation cohort, 34 individuals (2.4%) <18 years of age were excluded, as well as 15 patients (1.0%) with uncorrected intra or extracardiac shunts and 67 individuals (4.5%) with poor echocardiographic images from 1,464 patients who were screened.
At RHC, 852 patients (63.2%) had PH. Among patients with PH, 410 had precapillary PH (48.1%), 334 had iPC-PH (39.2%), and 108 had cPC-PH (12.7%).
BCI demonstrated high diagnostic accuracy in detecting precapillary PH in the overall cohort including patients without PH (AUC, 0.88; 95% CI, 0.85-0.90; P < .001). Index performance was even higher when focusing only on patients with PH at RHC (AUC, 0.91; 95% CI, 0.89-0.93; P < .001).
The optimal cutoff of 1.9 for diagnosis of precapillary PH from the derivation cohort showed sensitivity of 82%, speciﬁcity of 89%, PPV of 77%, and NPV of 92% in the validation cohort.
The authors concluded that BCI is a novel echocardiographic index that demonstrated high accuracy in the discrimination of precapillary PH and was superior to previously proposed methods for the noninvasive estimation of PH subtype.