Borderline pulmonary hypertension is associated with exercise intolerance and increased risk for acute exacerbation in patients with interstitial lung disease

We conducted a retrospective review of consecutive patients with ILD who underwent RHC between November 2013 and October 2016. RHC was performed in subjects with suspected PH based on clinical criteria. The diagnosis of ILD was made based on the clinical findings, serologic findings, and findings of high-resolution computed tomography (HRCT). The diagnosis of IPF was established using standard criteria [14]. Although a surgical lung biopsy was not required for confirmation, in some patients, the diagnosis was established by lung biopsy. We included all patients with CTD-ILD who were diagnosed on the basis of each criteria, however, in whom CTD-associated pulmonary arterial hypertension was excluded based on clinical judgment/multidisciplinary discussion. The exclusion criteria were as follows: (1) patients in whom the initial evaluation was performed using supplemental oxygen; (2) patients suffering from left heart disease, pulmonary arterial thromboembolism, chronic liver disease and obstructive sleep apnea syndrome, all of which could cause secondary PH other than CTD-ILD; (3) PAWP > 15 mmHg on RHC; (4) patients with orthopedic impairment(s) that could interfere with the performance in the 6MWT. According to a previous definition based on the 5th WSPH [1], ILD patients were divided into 3 groups according to the mPAP as measured by RHC: mPAP ≤20 mmHg (No-PH group), 20 < mPAP < 25 mmHg (Bo-PH group), and mPAP ≥25 mmHg (PH group). This study was reviewed and approved by the Ibarakihigashi National Hospital Institutional Review Board (IRB number 2017–022).

We recorded the patients’ characteristics, pulmonary function parameters, KL-6, PaO₂, the results of the 6MWT, echocardiography and hemodynamics retrospectively. All the patients underwent spirometry (CHESTAC-8900; Chest, Tokyo, Japan) based on the American Thoracic Society (ATS) recommendations for acceptability and reproducibility [15]. Single-breath DLco was also measured (CHESTAC-8900). The values of VC, FEV₁, TLC and DLco were expressed as percentages of the predicted normal values. The 6MWT was performed in all patients, in accordance with the ATS statement [16]. SpO₂ was monitored continuously with a wireless pulse oximeter during the 6MWT. If the SpO₂ fell below 85%, the test was stopped for safety concerns. The distance that the patients could walk in the test was recorded as the 6-min walk distance (6MWD), and the ΔSpO₂ (initial SpO₂ – lowest SpO₂ on 6MWT) was calculated. The right ventricular systolic pressure (RVSP) obtained by echocardiography was measured in all the patients. RHC was performed at rest using a Swan-Ganz catheter inserted percutaneously via either the jugular vein or the femoral vein. Cardiac output was measured by a thermodilution method.

We also evaluated the 1-year incidence of AEs and the 1-year survival after the initial RHC. AE of IPF was defined as previously described [17]. In brief, patients had a previous diagnosis of IPF, and presented with acute worsening of dyspnea within a 1-month period and newly developed bilateral ground-glass opacities and/or consolidation on HRCT, with no evidence of cardiac failure or fluid overload to explain the respiratory deterioration [17]. On the other hand, there is no existing official definition of AE for patients with ILDs other than IPF. However, AEs in patients with ILDs other than IPF resembles the AEs noted in patients with IPF [18]; therefore, we applied the definition of AE in patients with IPF to AE in patients with any kind of ILD.

Data are expressed as means ± standard deviation (SD) or median (interquartile range), as appropriate. Distribution of continuous measurements was evaluated using the Shapiro-Wilk test. To assess the differences among the 3 groups of patients classified as above, we performed one-way analysis of variance (ANOVA), the Kruskal-Wallis test, and the χ² test for parametric continuous, nonparametric continuous, and categorical measurements, respectively. The differences among the groups in the 6MWD and ΔSpO₂ were evaluated using one-way ANOVA, followed by the Tukey-Kramer post hoc test. Using the Kaplan-Meier method and the log-rank test, the impacts of the mPAP on the 1-year incidence of AEs and the 1-year survival were estimated. p values of < 0.05 were considered as being indicative of statistical significance. Analysis of all data was performed using SPSS version 24 for Windows (SPSS Inc., Chicago).

A total of 80 ILD patients were included in our analysis, 23 of whom had IPF (Table 1). Subjects with other ILDs included patients with CTD-ILD (n = 15), CPFE (n = 8), idiopathic non-specific interstitial pneumonia (n = 8), chronic hypersensitivity pneumonia (n = 8), pleuroparenchymal fibroelastosis (n = 7), pneumoconiosis (n = 2), microscopic polyangiitis-ILD (n = 1), and unclassifiable ILD (n = 8). The numbers of CTD-ILD patients with polymyositis/dermatomyositis (PM/DM), systemic sclerosis (SSc), rheumatoid arthritis (RA) and Sjögren’s syndrome (SjS) were 5 (33.3%), 4 (26.7%), 3 (20.0%) and 3 (20.0%), respectively.

The mPAP values as assessed by RHC in the 80 patients with ILD were as follows: mPAP ≤20 mmHg (No-PH group; 56 patients (70%)); 20 < mPAP < 25 mmHg (Bo-PH group; 18 patients (22.5%)), and mPAP ≥25 mmHg (PH group; 6 patients (7.5%)). None of the patients were receiving corticosteroids, immunosuppressive agents or vasodilator therapy at the time of the RHC. The clinical characteristics of each groups classified according to the mPAP are summarized in Table 1. There were no differences among the 3 groups in the age, BMI, smoking status, distribution of underlying diseases other than IPF, KL-6, PaO₂, or measures of pulmonary function. On the other hand, the 6MWD, ΔSpO₂, RVSP, mPAP, PVR and cardiac index were significantly different among the 3 groups. During the follow-up period after initial evaluation on room air, only 2 patients (Bo-PH group, n = 1; PH group, n = 1) received vasodilator therapy (tadalafil alone); 18 patients (No-PH group, n = 14; Bo-PH group, n = 2; PH group, n = 2) received anti-fibrotic drugs for IPF and 19 patients (No-PH group, n = 10; Bo-PH group, n = 8; PH group, n = 1) received corticosteroid and/or immunosuppressive therapy; 5 patients (belonging only to the PH group) received long-term oxygen therapy.

Comparisons of the 6MWD and ΔSpO₂ among the three groups are shown in Fig. 1. The 6MWD was significantly lower in both the Bo-PH and PH groups than in the No-PH group (p < 0.001 for both), however, there was no difference between the Bo-PH group and PH group. Although there were no significant differences in the ΔSpO₂ among the groups, the ΔSpO₂ was higher in the Bo-PH group as compared to the No-PH group (p = 0.059).

Kaplan-Meier analysis showed that the 1-year incidence of AE was significantly higher in both the Bo-PH and PH groups as compared to the No-PH group (p < 0.001, p = 0.023, respectively) (Fig. 2). In contrast, the Kaplan-Meier analysis revealed no significant differences in the 1-year survival among the 3 groups (Fig. 3).