A Simple Derived Prediction Score for the Identification of an Elevated Pulmonary Artery Wedge Pressure Using Precatheterization Clinical Data in Patients Referred to a Pulmonary Hypertension Center

Machine learning to differentiate pulmonary hypertension due to left heart disease from pulmonary arterial hypertension

Background and aims

Pulmonary hypertension due to left heart disease (PH-LHD) is the most frequent form of PH. As differential diagnosis with pulmonary arterial hypertension (PAH) has therapeutic implications, it is important to accurately and noninvasively differentiate PH-LHD from PAH before referral to PH centres. The aim was to develop and validate a machine learning (ML) model to improve prediction of PH-LHD in a population of PAH and PH-LHD patients.

Methods

Noninvasive PH-LHD predictors from 172 PAH and 172 PH-LHD patients from the PH centre database at the University Hospitals of Leuven (Leuven, Belgium) were used to develop an ML model. The Jacobs score was used as performance benchmark. The dataset was split into a training and test set (70:30) and the best model was selected after 10-fold cross-validation on the training dataset (n=240). The final model was externally validated using 165 patients (91 PAH, 74 PH-LHD) from Erasme Hospital (Brussels, Belgium).

Results

In the internal test dataset (n=104), a random forest-based model correctly diagnosed 70% of PH-LHD patients (sensitivity: n=35/50), with 100% positive predicted value, 78% negative predicted value and 100% specificity. The model outperformed the Jacobs score, which identified 18% (n=9/50) of the patients with PH-LHD without false positives. In external validation, the model had 64% sensitivity at 100% specificity, while the Jacobs score had a sensitivity of 3% for no false positives.

Conclusions

ML significantly improves the sensitivity of PH-LHD prediction at 100% specificity. Such a model may substantially reduce the number of patients referred for invasive diagnostics without missing PAH diagnoses.

Corrected MRI Pulmonary Transit Time for Identification of Combined Precapillary and Postcapillary Pulmonary Hypertension in Patients With Left Heart Disease

BACKGROUND

The identification of combined precapillary and postcapillary pulmonary hypertension (CpcPH) in patients with pulmonary hypertension (PH) due to left heart disease (LHD) can influence therapy and outcome and is currently based on invasively determined hemodynamic parameters.

PURPOSE

To investigate the diagnostic value of MRI-derived corrected pulmonary transit time (PTTc) in PH-LHD sub-grouped according to hemodynamic phenotypes.

STUDY TYPE

Prospective observational study.

POPULATION

A total of 60 patients with PH-LHD (18 with isolated postcapillary PH [IpcPH] and 42 with CpcPH), and 33 healthy subjects.

FIELD STRENGTH/SEQUENCE

A 3.0 T/balanced steady-state free precession cine and gradient echo-train echo planar pulse first-pass perfusion.

ASSESSMENT

In patients, right heart catheterization (RHC) and MRI were performed within 30 days. Pulmonary vascular resistance (PVR) was used as the diagnostic "reference standard." The PTTc was calculated as the time interval between the peaks of the biventricular signal-intensity/time curve and corrected for heart rate. PTTc was compared between patient groups and healthy subjects and its relationship to PVR assessed. The diagnostic accuracy of PTTc for distinguishing IpcPH and CpcPH was determined.

STATISTICAL TESTS

Student's t-test, Mann-Whitney U-test, linear and logistic regression analysis, and receiver-operating characteristic curves. Significance level: P < 0.05.

RESULTS

PTTc was significantly prolonged in CpcPH compared with IpcPH and normal controls (17.28 ± 7.67 vs. 8.82 ± 2.55 vs. 6.86 ± 2.11 seconds), and in IpcPH compared with normal controls (8.82 ± 2.55 vs. 6.86 ± 2.11 seconds). Prolonged PTTc was significantly associated with increased PVR. Furthermore, PTTc was a significantly independent predictor of CpcPH (odds ratio: 1.395, 95% confidence interval: 1.071-1.816). The area under curve was 0.852 at a cut-off value of 11.61 seconds for PTTc to distinguish between CpcPH and IpcPH (sensitivity 71.43% and specificity 94.12%).

DATA CONCLUSION

PTTc may be used to identify CpcPH. Our findings have potential to improve selection for invasive RHC for PH-LHD patients.

EVIDENCE LEVEL

3 TECHNICAL EFFICACY: Stage 2.

Predicting Group II pulmonary hypertension: diagnostic accuracy of the H2FPEF and OPTICS scores in Scotland

OBJECTIVE

Group II pulmonary hypertension (PH) can be challenging to distinguish from Group I PH without proceeding to right heart catheterisation (RHC). The diagnostic accuracy of the H2FPEF and OPTICS scores was investigated in Scotland.

METHODS

Patients were included in the study if they were referred to the Scottish Pulmonary Vascular Unit between 2016 and 2020 and subsequently diagnosed with Group II PH or Group I PH which was either idiopathic, heritable or pulmonary veno-occlusive disease. The established cut offs for the H2FPEF and for the OPTICS scores were applied retrospectively to predict the presence of Group II PH. The diagnosis from the scores were compared with the MDT consensus diagnosis following RHC.

RESULTS

107 patients with Group I PH and 86 patients with Group II PH were included. Retrospective application of the OPTICS score demonstrated that pretest scoring would detect 28% of cases with Group II PH yet at the cost of misdiagnosing 4% of patients with Group I as Group II PH (specificity 0.96). The H2FPEF score had a far greater sensitivity (0.70) yet reduced specificity (0.91), leading to misdiagnosis of 9% of Group I PH cases.

CONCLUSION

While the specificity of these scores was high, the lack of perfect specificity limits their utility as it results in missed patients with Group I PH. As a consequence, they cannot replace RHC as the means of diagnosing the aetiology of PH in their current form. The scores may still be used to support clinical judgement or to indicate the advisability for further provocative testing at RHC.

Future perspective in diabetic patients with pre- and post-capillary pulmonary hypertension

Pulmonary hypertension is a clinical syndrome that may include multiple clinical conditions and can complicate the majority of cardiovascular and respiratory diseases. Pulmonary hypertension secondary to left heart disease is the prevalent clinical condition and accounts for two-thirds of all cases. Type 2 diabetes mellitus, which affects about 422 million adults worldwide, has emerged as an independent risk factor for the development of pulmonary hypertension in patients with left heart failure. While a correct diagnosis of pulmonary hypertension secondary to left heart disease requires invasive hemodynamic evaluation through right heart catheterization, several scores integrating clinical and echocardiographic parameters have been proposed to discriminate pre- and post-capillary types of pulmonary hypertension. Despite new emerging evidence on the pathophysiological mechanisms behind the effects of diabetes in patients with pre- and/or post-capillary pulmonary hypertension, no specific drug has been yet approved for this group of patients. In the last few years, the attention has been focused on the role of antidiabetic drugs in patients with pulmonary hypertension secondary to left heart failure, both in animal models and in clinical trials. The aim of the present review is to highlight the links emerged in the recent years between diabetes and pre- and/or post-capillary pulmonary hypertension and new perspectives for antidiabetic drugs in this setting.

Deep Learning for Detection of Elevated Pulmonary Artery Wedge Pressure Using Standard Chest X-Ray

BACKGROUND

To accurately diagnose and control heart failure (HF), it is important to carry out a simple assessment of elevated pulmonary arterial wedge pressure (PAWP). The aim of this study was to develop and validate an objective method for detecting elevated PAWP by applying deep learning (DL) to a chest x-ray (CXR).

METHODS

We enrolled 1013 consecutive patients with a right-heart catheter between October 2009 and February 2020. We developed a convolutional neural network to identify patients with elevated PAWP (> 18 mm Hg) as the actual value of PAWP to be used in the dataset for training. In the prospective validation dataset used to detect elevated PAWP, the area under the receiver operating characteristic curve (AUC) was calculated using the DL model that evaluated the CXR.

RESULTS

In the prospective validation dataset, the AUC of the DL model with CXR was not significantly different from the AUC produced by brain natriuretic peptide (BNP) and the echocardiographic left-ventricular diastolic dysfunction (DD) algorithm (DL model: 0.77 vs BNP: 0.77 vs DD algorithm: 0.70; respectively; P = NS for all comparisons); it was, however, significantly higher than the AUC of the cardiothoracic ratio (DL model vs cardiothoracic ratio [CTR]: 0.66, P = 0.044). The model based on 3 parameters (BNP, DD algorithm, and CTR) was improved by adding the DL model (AUC: from 0.80 to 0.86; P = 0.041).

CONCLUSIONS

Applying the DL model based on a CXR (a classical, universal, and low-cost test) is useful for screening for elevated PAWP.

Impact of post-capillary pulmonary hypertension on mortality in interstitial lung disease

BACKGROUND

Pulmonary hypertension (PH) influences mortality in patients with interstitial lung disease (ILD). Almost all studies on patients with ILD, have focused on the clinical impact of pre-capillary PH on survival. Therefore, little is known about the influence of post-capillary PH. We aimed to assess the prevalence of post-capillary PH and its clinical impact on survival in patients with ILD, followed by comparison with pre-capillary PH.

METHODS

This retrospective study enrolled 1152 patients with ILD who were diagnosed with PH using right heart catheterization between May 2007 and December 2015. We analyzed the demographics and composite outcomes (defined as death from any cause or lung transplantation) of patients with post-capillary PH and compared them with patients with pre-capillary PH.

RESULTS

Thirty-two (20%) of the 157 patients with ILD-PH were diagnosed with post-capillary PH. Patients with post-capillary PH had significantly lower modified Medical Research Council scores, higher diffusion capacity for carbon monoxide, higher resting PaO₂, lower pulmonary vascular resistance (PVR), and higher lowest oxygen saturation during the 6-min walk test compared to those with pre-capillary PH. Cardiovascular diseases were associated with a higher risk of mortality in patients with post-capillary PH. Multivariate Cox proportional hazards analysis demonstrated no significant difference between the composite outcomes in pre-capillary and post-capillary PH, while PVR and the ILD Gender-Age-Physiology Index were significantly associated with the composite outcome.

CONCLUSIONS

We found that approximately one-fifth of patients with ILD-PH were diagnosed with post-capillary PH, and that PVR and not post-capillary PH was associated with mortality.

Prediction of pulmonary hypertension in older adults based on vital capacity and systolic pulmonary artery pressure

Objective

Right heart catheterization (RHC) is associated with a higher procedural risk in older adults, but non-invasive estimation of pulmonary hypertension (PH) is a challenge. We aimed to elaborate a non-invasive prediction model to estimate PH.

Methods and design

We retrospectively analysed 134 older adults (70.0 years ±12.3; 44.9% males) who reported to our clinic with unclear dyspnea between 01/2015 and 01/2020 and had received RHC as a part of their diagnostic workup. Lung function testing, analysis of blood gas samples, 6 min walk distance and echocardiography were performed within 24 hours of RHC.

Main outcome measures

In a stepwise statistical approach by using an in/exclusion algorithm (using the AIC criterion) we analysed non-invasive parameters to test their value in predicting PH (defined as mean pulmonary artery pressure, PA_mean, >25mmHg). Discrimination capability of the final model was measured by the AUC (area under curve) from an ROC (receiver operating characteristics) analysis.

Results

We yielded a sensitivity of 87.2% and a specificity of 62.5% in a combinatorial logistical model with systolic pulmonary artery pressure (sPAP) and forced vital capacity (VC_max), the discrimination index was 86.7%. The odds ratios for an increase of 10 mmHg of sPAP were 2.99 (2.08–4.65) and 1.86 (1.11–3.21) for a 1 l decrease in VC_max. On their own, VC_max proved to be specific (83.3%), while sPAP was a sensitive (79.1%) predictor for PH.

Conclusions

We provide a combinatorial model to predict PH from sPAP and VC_max in older adults, which may help to avoid invasive procedures.

Noninvasive Prediction of Elevated Wedge Pressure in Pulmonary Hypertension Patients Without Clear Signs of Left-Sided Heart Disease: External Validation of the OPTICS Risk Score

Background

Although most newly presenting patients with pulmonary hypertension (PH) have elevated pulmonary artery wedge pressure, identification of so‐called postcapillary PH can be challenging. A noninvasive tool predicting elevated pulmonary artery wedge pressure in patients with incident PH may help avoid unnecessary invasive diagnostic procedures.

Methods and Results

A combination of clinical data, ECG, and echocardiographic parameters was used to refine a previously developed left heart failure risk score in a retrospective cohort of pre‐ and postcapillary PH patients. This updated score (renamed the OPTICS risk score) was externally validated in a prospective cohort of patients from 12 Dutch nonreferral centers the OPTICS network. Using the updated OPTICS risk score, the presence of postcapillary PH could be predicted on the basis of body mass index ≥30, diabetes mellitus, atrial fibrillation, dyslipidemia, history of valvular surgery, sum of SV1 (deflection in V1 in millimeters) and RV6 (deflection in V6 in millimeters) on ECG, and left atrial dilation. The external validation cohort included 81 postcapillary PH patients and 66 precapillary PH patients. Using a predefined cutoff of >104, the OPTICS score had 100% specificity for postcapillary PH (sensitivity, 22%). In addition, we investigated whether a high probability of heart failure with preserved ejection fraction, assessed by the H₂FPEF score (obesity, atrial fibrillation, age >60 yrs, ≥2 antihypertensives, E/e' >9, and pulmonary artery systolic pressure by echo >35 mmHg), similarly predicted the presence of elevated pulmonary artery wedge pressure. High probability of heart failure with preserved ejection fraction (H₂FPEF score ≥6) was less specific for postcapillary PH.

Conclusions

In a community setting, the OPTICS risk score can predict elevated pulmonary artery wedge pressure in PH patients without clear signs of left‐sided heart disease. The OPTICS risk score may be used to tailor the decision to perform invasive diagnostic testing.

Pulmonary arterial hypertension and heart failure with preserved ejection fraction: are they so discordant?

Heart failure with preserved ejection fraction (HFpEF) and pulmonary arterial hypertension (PAH) are two emerging diseases focusing the attention of numerous researchers. In the last PAH guideline, there is a crossroad between the two diseases and pulmonary hypertension (PH) due to heart failure (HF) is categorized as subtype 2. In order to assess the correct diagnosis and management, it should be better understood the points of convergence and divergence of two diseases. Although, risk factors, demographic characteristics and haemodynamics are different, we report several similarities regarding vascular alterations, some aspects of cardiac remodelling, and clinical presentation. This model suggests HFpEF and PAH as two comparable conditions, with different cardiac adaptation and trajectories, linked to the intrinsic properties of either right and left ventricles. In both diseases the early pathophysiological mechanisms appear to begin from peripheral vasculature and to be backward transmitted to the larger arterial vascular district, and eventually to the myocardial structure. In this paper we would propose a simple approach to recognize the concordances and, all at once, distinguish the peculiarities of the two diseases.

Pathophysiology and Diagnosis of Pulmonary Hypertension Due to Left Heart Disease

Pulmonary hypertension due to left heart disease (PH-LHD) is the most common type of pulmonary hypertension, although an accurate prevalence is challenging. PH-LHD includes PH due to systolic or diastolic left ventricular dysfunction, mitral or aortic valve disease and congenital left heart disease. In recent years a new and distinct phenotype of “combined post-capillary and pre-capillary PH,” based on diastolic pulmonary gradient and pulmonary vascular resistance, has been recognized. The roles of right ventricular dysfunction and pulmonary vascular compliance in PH-LHD have also been elucidated recently and they appear to have significant clinical implications. Echocardiography continues to play a seminal role in diagnosis of PH-LHD and heart failure with preserved LV ejection fraction, as it can identify valve disease and help to distinguish PH-LHD from pre-capillary PH. Right, and occasionally left heart catheterization, remains the gold-standard for diagnosis and phenotyping of PH-LHD, although Cardiac Magnetic Resonance Imaging is emerging as a useful alternative tool in non-invasive diagnostic and prognostic assessment of PH-LHD. In this review, the latest evidence for more recent advances will be discussed, including the role of fluid challenge and exercise during cardiac catheterization to unravel occult post-capillary and the role of vasoreactivity testing. The use of many or all of these diagnostic techniques will undoubtedly provide key information about sub-groups of patients with PH-LHD that might benefit from medical therapy previously considered to be only suitable for pulmonary arterial hypertension.

Pulmonary hypertension due to left heart disease

Pulmonary hypertension due to left heart disease, also known as group 2 pulmonary hypertension according to the European Society of Cardiology/European Respiratory Society classification, is the most common cause of pulmonary hypertension. In patients with left heart disease, the development of pulmonary hypertension favours right heart dysfunction, which has a major impact on disease severity and outcome. Over the past few years, this condition has been considered more frequently. However, epidemiological studies of group 2 pulmonary hypertension are less exhaustive than studies of other causes of pulmonary hypertension. In group 2 patients, pulmonary hypertension may be caused by an isolated increase in left-sided filling pressures or by a combination of this condition with increased pulmonary vascular resistance, with an abnormally high pressure gradient between arteries and pulmonary veins. A better understanding of the conditions underlying pulmonary hypertension is of key importance to establish a comprehensive diagnosis, leading to an adapted treatment to reduce heart failure morbidity and mortality. In this review, epidemiology, mechanisms and diagnostic approaches are reviewed; then, treatment options and future approaches are considered.