Douschan P, Kovacs G, Avian A, Foris V, Gruber F, Olschewski A, Olschewski H. Am J Respir Crit Care Med. 2018 Feb 15;197(4):509-516

Background: Pulmonary hypertension (PH) is defined as a mean pulmonary artery pressure (mPAP) ≥25mmHg on right heart catheterization (RHC). The normal mPAP is 14.0 ± 3.3mmHg and there has been recent interest in patients with “borderline elevations” in mPAP (i.e. above normal but below the diagnostic threshold for PH). Studies of patients with borderline PH in high-risk groups (e.g. scleroderma, IPF) and in a large VA population have demonstrated an association with poor outcomes. The current study sought to describe the prognostic significance of borderline elevations in mPAP in a population referred for suspected PH.

Methods: Patients were included if they were referred for RHC at the Medical University of Graz due to unexplained dyspnea and suspicion for PH. Patients with severe lung or heart disease were not considered for RHC, unless there were signs of decompensated right heart failure. Baseline differences and outcomes were assessed between groups in two pre-defined analyses: 1) pre-defined thresholds (mPAP≤17.3=lower-normal; mPAP 17.4-20.6=upper-normal; mPAP 20.6-24.9=borderline; mPAP ≥25=manifest PH) and 2) unbiased approach with a classification and regression tree (CART) analysis to identify optimal cut-points in mPAP as a risk factor for mortality.

Results: 547 patients had RHC and were included in this analysis. Using the pre-defined thresholds, higher mPAP group patients were older with higher NT-proBNP levels. Lung function, 6-minute walk distance, and peak VO2 all declined as mPAP group increased. Five-year survival rates were 92%, 79% 71%, and 58% in the lower-normal, upper-normal, borderline, and manifest PH groups. Compared to the lower-normal mPAP group, borderline (HR 2.37) and manifest PH (HR 5.05) patients had significantly worse mortality, after adjustment for age and number of cardiopulmonary comorbidities. CART analysis revealed three groups with significantly different survival, with optimal cut-points of 17mmHg and 26mmHg. Compared to patients with mPAP<17mmHg, the mortality risk was increased 3-fold if the mPAP was 17-26mmHg and 7-fold if mPAP was >26mmHg.

Conclusions: In a cohort of patients referred for RHC due to unexplained dyspnea and suspicion for PH, outcomes were worse as mPAP increases, with borderline and manifest PH having significantly worse survival compared to patients with lower-normal mPAP.

Commentary: Pulmonary hypertension (PH) is one of many common cardiopulmonary diseases defined by a single continuous variable. The current criterion for diagnosing PH requires a mean pulmonary artery pressure (mPAP) ≥25 mmHg measured using right heart catheterization (RHC). However, this definition is largely historical, based on consensus opinion in the absence of sufficient clinical data. By contrast, retrospective data from large RHC registries suggests a continuous relationship between mPAP and mortality that begins at mPAP ~19 mmHg and, thus, at levels considered currently to be normal. Prospective studies assessing the prognostic relevance of mPAP <25 mmHg on hard clinical end-points are positioned to address remaining uncertainty on the spectrum of clinical risk related to mPAP, but were lacking until recently.

Indeed, the work by Douschan and colleagues provides critical insight clarifying this knowledge gap. In their study, which included a sizeable cohort of patients enrolled prospectively, mPAP 20-25 prognosticated a significant increase in risk for mortality and declining exercise capacity. Furthermore, an increase in clinical events was observed shortly following RHC. This finding implies that opportunity may exist to offset adverse outcome in patients with mPAP ~19-24 mmHg immediately following diagnosis. Findings from this study also identify a new unmet need in the field: understanding the pathobiological and pathophysiological mechanisms by which mPAP levels above but near normal affect clinical trajectory. Overall, this study adds important depth to data suggesting that updating the definition of PH to mPAP >19 is required to capture the full gamut of at-risk patients.

Article summary by: Matthew Lammi MD, MSCR; Associate Professor of Medicine, Louisiana State University Health Sciences Center; Co-Director, Comprehensive Pulmonary Hypertension Center-University Medical Center

Expert commentary by: Bradley Maron MD; Assistant Professor of Medicine, Brigham and Women’s Hospital, Harvard Medical School and the Department of Cardiology, Boston VA Healthcare System.