This prospective study was approved by the ethics board of the Aristotle University of Thessaloniki, Greece, and informed consent was obtained by all subjects. A consecutive population of 40 Caucasian patients presenting with pulmonary fibrosis, either idiopathic or secondary to connective tissue disease (CTD)15-17 and met the inclusion/exclusion criteria, were evaluated between May 2018 and March 2020 at the Pulmonary Hypertension Outpatient Clinic of the Department of Respiratory Failure, in a tertiary university hospital. Diagnosis of the underlying etiology of the fibrotic lung disease was established at the Interstitial Lung Disease Outpatient Clinic of the University Respiratory Department, according to international guidelines.15-17 Patients underwent cardiopulmonary evaluation, including clinical examination, lung function tests, 6-minute walking test (6MWT), heart ultrasonography, and cardiopulmonary exercise test (CPET), unless contraindicated, during their initial visit. A second visit was scheduled one to two weeks after the first, to perform RHC. Resting hemodynamic evaluation was followed by the exercise protocol proposed by Herve et al, using a bedside cycle ergometer in the supine position.8 Patients with resting postC PH were excluded from the exercise protocol.

Exclusion criteria were as follows: 1) functional vital capacity (FVC) <50% of predicted values, 2) presence of radiologic evidence of emphysema and/or spirometric evidence of airway obstruction, 3) contraindications for exercise tests, such as unstable angina, symptomatic arrythmia, severe hypoxemia, musculoskeletal disease etc., 4) recent acute myocardial infarction or pulmonary embolism in the past year, 5) presence of moderate or severe valvular disease or left ventricular ejection fraction < 50%, 6) treatment with specific PH agents.

Continuous monitoring of vital signs was initiated prior to RHC, including electrocardiograms, pulse oximetry and blood pressure measurements by a cuff sphygmomanometer at frequent intervals. Supine RHC was performed by percutaneously inserting a balloon-tipped, triple-lumen, fluid-filled Swan-Ganz pulmonary arterial catheter via the internal jugular vein. Complete hemodynamic and oxygen profiles were obtained at rest. Zero level was set at midthoracic level.18 Resting pressures were measured at the end of expiration.19 CO was measured by thermodilution and expressed as the mean of three measurements at rest, and of two during exercise.

During exercise RHC, patients were instructed to cycle at a rate of 60 revolutions per minute until first occurrence of any of the following: exhaustion, discomforting dyspnea or chest ache (flagged by the patient), or an observed arterial oxygen saturation of <80%. Measurements of systolic, diastolic and mean pulmonary artery pressure (sPAP, dPAP, mPAP, respectively), PAWP and CO were obtained at unloaded pedaling (0 W) and at constant workload increments of 10 W. Each stage lasted 3 minutes and values were obtained on the second minute of each stage. Pressure waveforms were averaged over three respiratory cycles, and vital signs were also recorded for each stage. Total pulmonary resistance (TPR) was defined as the mPAP to CO ratio at maximal exercise and expressed as Wood units (WU).8,20 Pulmonary vascular compliance (PVC) was calculated by the ratio of stroke volume (SV) to (sPAP-dPAP). The ratio of PAWP to CO (ΔPAWP/ΔCO) was calculated as follows: PAWP(peak)−PAWP(rest)CO(peak)−CO(rest)9 .

Resting PH was defined by mPAP ≥25 mmHg; please note that the revised PH definition of mPAP ≥20 mmHg arose after the initiation of the study.21 PAWP ≤15 mmHg defined preC-PH at rest, whereas postC-PH was defined as PAWP>15 mm Hg.3 ePH was defined as mPAP >30 mmHg and TPR > 3 mmHg·min·L− 1 at maximal exercise.8 ΔPAWP/ΔCO ≤2 mmHg/L per minute determined a preC etiology, while ΔPAWP/ΔCO >2 mmHg/L per minute a postC etiology.9

Statistical analysis was performed using SPSS software, version 23 (IBM SPSS Statistics, IBM Corporation). Values are reported as mean ± standard deviation (SD) or median [interquartile range]. Normality of distribution was assessed by the Shapiro-Wilk test. Normally distributed continuous variables were compared between groups by using a one-way analysis of variance (ANOVA) with a Scheffe post hoc analysis. Nonnormally distributed continuous variables were compared among groups by using the Kruskal-Wallis test with Dunn's post hoc test. Categorical variables were compared between groups by using a chi-squared test. Alterations of hemodynamic variables from rest to exercise were evaluated using paired t-tests. Correlations between variables were assessed using Pearson's r or Spearman's rho correlation coefficients, depending on the data distribution. P values of < 0.05 were considered statistically significant.

Forty (N = 40) IFLD patients were evaluated. Mean age was 68±11 years, and pulmonary restriction was variable, from mild to moderate. Patients were equally distributed across NYHA classes II, III and IV. Idiopathic pulmonary fibrosis (IPF) was diagnosed in 75% of the patients (n = 30), with lung fibrosis secondary to CTDs among the remaining 25% (systemic sclerosis, n = 6, rheumatoid arthritis, n = 3, systemic lupus erythematosus, n = 1). All CTD patients and 11 of the IPF patients were not receiving an antifibrotic agent, while use of nintedanib was reported by 10 of the IPF patients and use of pirfenidone by 7. Radiologic severity of fibrosis22,23 was mostly moderate (50%, n = 20), with mild fibrosis documented for 35% and severe for 15%. Patients’ characteristics grouped by initial hemodynamic allocation are summarized in Table 1.

Initial hemodynamic evaluation allocated patients into four groups: (a) normal group (n = 4, 10%), presenting normal hemodynamics at rest and exercise, (b) ePH group (n = 11, 27.5%), presenting normal resting hemodynamics and ePH, (c) preC-PH group (n = 15, 37.5%), presenting preC-PH at rest, and (d) postC-PH group (n = 10, 25%); presenting post-C PH at rest. The postC-PH group did not undergo further exercise hemodynamic evaluation.

Secondary hemodynamic evaluation involved detection of abnormal PAWP increase at peak exercise in relation to CO (ΔPAWP/ΔCO> 2mm Hg/L per minute), as stated in Methods, thus dividing the ePH and preC-PH groups into the following subgroups: (i) preC-ePH group (n = 7), presenting preC etiology of ePH, (ii) postC-ePH group (n = 4), presenting postC etiology of ePH, (iii) preC-PH-PAWP (n = 9), presenting preC etiology of PH, (iv) preC-PH+PAWP (n = 6), presenting postC etiology of PH. A cut-off value of 24mmHg for peak PAWP differentiating preC and postC etiology was driven by the implementation of the main stated criterion (ΔPAWP/ΔCO> 2mm Hg/L per minute). The hemodynamic allocation of the study cohort is illustrated with a tree diagram in Fig. 1.

Figure 1.

Hemodynamic allocation of patients based on rest and exercise hemodynamic values.

(0.26MB).

Mean values of non-invasive variables did not display statistically significant differences between hemodynamic groups, except for diffusion capacity markers (DLCO, KCO and FVC%/DLCO%), which were lower in both ePH and PH groups (p < 0.05). FVC%/ DLCO% ratio has been evaluated as a noninvasive predictor for the development of PH in patients with interstitial lung disease and the cutoff value of 1.39 has been measured to offer 96% sensitivity and 65% specificity for this predictive purpose.24 Non-invasive variables grouped by initial hemodynamic allocation are summarized in Table 1.

All patients in the postC-PH group had at least one documented cardiac comorbidity (arterial hypertension, coronary artery disease) and echocardiographic evidence of diastolic dysfunction with preserved ejection fraction.25 Cardiac comorbidities were more common in the postC-ePH and preC-PH+PAWP subgroups, compared to preC-ePH and preC-PH-PAWP subgroups respectively [69% vs 42%, (p = 0.03) and 72% vs 47%, (p = 0.04)]. Comparisons of spirometric values among the four subgroups indicated milder pulmonary restriction in the subgroups with abnormal PAWP elevations than those without. Specifically, mean FVC% was 79.4±17.28% in the postC-ePH subgroup versus 61.5±12.46% in the preC-ePH subgroup (p < 0.05), and 83.85±21.13% in the preC-PH+PAWP subgroup versus 64.31±12.28 in the preC-PH-PAWP subgroup (p < 0.05).

Statistical analysis of variables obtained from CPET did not produce statistical differences between the various groups, possibly due to small study sample, as only 22 patients performed CPET. Main reason of CPET exclusion was resting hypoxia of patients receiving long-term oxygen treatment (LTOT). Peak oxygen consumption (VO2) at anaerobic threshold (AT) and at peak exercise was reduced in the majority of patients. Ventilatory reserve was abnormal (<11lt) in 7 patients (2 preC, 2 postC, 2 ePH, and 1 normal). The ratio of minute ventilation to carbon dioxide output at anaerobic threshold (VE/VCO2 at AT), which has been validated as a detector of pulmonary vasculopathy,26 was normal in all patients of the normal group and abnormal (>34) in all patients of the ePH and PH groups.

Resting values of CO, cardiac index (CI), stroke volume (SV) and pulmonary vascular compliance (PVC) were significantly impaired in the ePH, preC-PH and postC-PHs groups when compared to the normal group. However, these variables were similar between the ePH, preC-PH and postC-PH groups. Mean and p values of hemodynamic variables are summarized in Table 2 (for resting supine RHC) and Table 3 (for exercise supine RHC).

Among the strengths of this study are the prospective nature of data collection, the use of the gold standard RHC and the fact that the same team of physicians conducted all evaluations, minimizing technique and interpretation variability.

Conversely, a study limitation is the small monocentric study sample, largely due to the rarity of the disease, which impacts the statistical power of the analysis. The normal group is much smaller from the other groups, though we included patients shortly after diagnosis. A selection bias might exist as patients with limited symptoms tend to be less willing to undergo further evaluation that includes an invasive examination.

Age-related limits to hemodynamic normality for the diagnosis of ePH might also be reasonably questioned. Although TPR appears to have high specificity and sensitivity for age groups both below and above 50 years of age, diagnosis of ePH is problematic for those aged above 70 years, as abnormal exercise hemodynamics could be the result of normal ageing.8,46 Given our study population has a mean age approximating this cut-off value, our results should be interpreted with caution.

Finally, precision in measurements of pressures during RHC is globally challenging, as movement artifacts and intrathoracic pressure swings can affect hemodynamic waveforms, especially during exercise.19 That was the reason behind excluding patients with radiologic evidence of emphysema and/or airflow limitation of any degree, to avoid overestimation of pressures due to significant air trapping.47,48