Estimates of Prevalence of Pulmonary Hypertension according to Different International Definitions

Background Pulmonary hypertension is a dreaded disease associated with considerable morbidity and mortality. The pulmonary hypertension developing due to chronic respiratory disease is a unique subset with symptoms often getting masqueraded by the underlying respiratory condition. The importance of early detection of this complication has been realized worldwide, and recently, the definition of pulmonary hypertension was revised to set the cutoff of mean pulmonary artery pressure (mPAP) at 20 mmHg instead of 25 mmHg at rest. In our study, we have tried to estimate the difference this new definition brings to the prevalence of pulmonary hypertension among interstitial lung disease patients at our centre. Methods This was a cross-sectional study in which all the patients of ILDs (n = 239) attending the outdoor and indoor Department of Respiratory Medicine, King George's Medical University, India, for the duration of one year were subjected to transthoracic echocardiography along with measurement of serum pro-B-type natriuretic peptide (BNP) and troponin T values. The data were analyzed using the different definitions, and the prevalence was compared. Result Incidence of pulmonary hypertension among ILD patients at mPAP cutoff ≥ 25 was 28.9%, while that at value ≥20 mmHg, incidence of pulmonary hypertension increased to 46.0%. An increment of 15–20% in incidence of pulmonary hypertension was observed among different types of ILD when cutoff of mPAP was changed. Conclusion The new definition helps in a significant increase in the detection of pulmonary hypertension, which certainly helps in earlier detection and better management of patients.


Introduction
Pulmonary hypertension is a morbid condition that gained recognition in 1973 when the World Health Organization organized its first symposium. e hemodynamic definition was given as mean pulmonary artery pressure > 25 mmHg at rest measured by right heart catheterization [1]. e second symposium was marked by the classification of PH into five groups based on common pathophysiological findings. e association between pulmonary diseases and PH was thus formally established and assigned the third group in the classification. Dana Point Classification was given during the 4 th symposium and the treatment algorithm updated [2,3]. In 2018, the 6 th World Symposium on Pulmonary Hypertension was a landmark as a proposal was made to revise hemodynamic definition and set mPAP threshold at a lower value to ≥20 mmHg [4,5]. is proposal comes in light of the fact that normal mPAP is 14 mmHg ± 3.3 mmHg. us, the values from 20 to 25 mmHg remained a grey zone. Studies have proven that patients with mPAP 20-25 mmHg are likely to progress to ≥25 mmHg and are at higher risk for morbidity [4]. In our study, we aim to study the effect of the new definition on the prevalence of pulmonary among the patients of ILDs at our tertiary care centre. diagnosed with interstitial lung diseases who provided written informed consent were included in the study. Approval was obtained from Institutional Ethics Committee. Patients who did not provide consent or had left-sided heart disease were excluded. 239 subjects were enrolled in the study. All were subjected to transthoracic echocardiography, and the following parameters were recorded: tricuspid regurgitation (TR) velocity, right ventricular systolic pressure (RVSP), pulmonary artery acceleration time, right atrial area, right atrial and right ventricular enlargement, tricuspid annular plane systolic excursion (TAPSE), fractional area change (FAC), interventricular septum (IVS) flattening, pericardial effusion, and left ventricular ejection fraction. Right atrial enlargement was said to occur when right atrial area > 18 cm 2 . TAPSE indicated the longitudinal function of right ventricle. Its normal value was >1.6 cm [6]. Mean pulmonary artery pressure was calculated using Mahan's equation: mPAP � 79 − (0.45 * PAT) [7]. RVSP was calculated using TR velocity as RVSP � 4 V 2 + right atrial pressure, where V is the TR velocity. e severity was defined as mild if mPAP was 20-40 mmHg in new definition or 25-40 mmHg in the previous one and moderate and severe when mPAP was 41-55 mmHg and >55 mmHg, respectively [8-10]. Diameter of right pulmonary artery > 16 mm on chest radiograph was used to define its enlargement [11]. Pro-B-type natriuretic peptide (pro-BNP) and troponin T levels were measured in all the subjects. e statistical analysis was done using SPSS (Statistical Package for Social Sciences) Version 21.0 statistical analysis software. e values were represented in number (%) and mean ± SD. e statistical tools employed were mean, standard deviation, chi-square test, Student' ttest, analysis of variance (ANOVA), and logistic regression.

Results
239 subjects were enrolled which comprised of females and had a mean age of years. Most were resident of rural region. Other baseline characteristics are given in Table 1. Majority of the patients fell under the diagnosis of hypersensitivity pneumonitis (HSP) and connective tissue disease-related interstitial lung disease (CTD-ILD). Rest were cases of idiopathic pulmonary fibrosis (IPF), non-IPF idiopathic interstitial pneumonias (IIPs), sarcoidosis, and others. e other ILDs included cases such as combined pulmonary fibrosis and emphysema (CPFE), Langerhans' cell histiocytosis (LCH), occupational lung disease, radiation-induced fibrosis, and unclassified. e distribution of population according to the ILD diagnosis is given in Table 2. When compared gender wise, IPF was more prevalent in males. All the subjects were subjected to transthoracic echocardiography, and statistics was applied using different cutoffs of mPAP. e prevalence of PH rose from 28.9% to 46.0% when mPAP > 20 mmHg was used to define PH (Tables 3  and 4). While the prevalence was higher in females, no statistical significance was seen using either definition. Among those with PH, nearly two-thirds were in mild category. e number of severe cases was only two; hence, it was combined with the moderate category while applying statistics. e largest number of cases was seen in HSP subjects, as it was the most prevalent group. Among the different types of ILDs, non-IPF IIPs had the maximum prevalence though the number of subjects was too small (Tables 5 and 6). Echocardiographic parameters were significantly deranged in the moderate-to-severe group only where mean tricuspid regurgitation velocity was 2.53 m/s 2 , mean RVSP was 54.7 mmHg, and mean diameter of IVC was 21.57 mm (Tables 7 and 8 and Figure 1). Chest radiograph findings were not so specific as reticulations are seen in ILDs as well. Pulmonary artery was enlarged only in moderate-tosevere cases. Right ventricular hypertrophy and p-pulmonale on ECG were also seen in moderate-severe category. In the lung function test, no significant correlation with was seen with forced vital capacity. Mean pro-BNP was 8582.48 pg/ml using the older definition group, while it decreased to 5716.68 pg/ml applying the new cutoff (Figure 2). We ascertained the level of pro-BNP and Trop-T above which all the subjects had mPAP ≥ 20 mmHg. Pro-BNP ≥ 535.5 pg/ml was 87.3% sensitive and 79.1% specific, while Trop-T ≥ 0.0295 ng/ml had sensitivity of 73.6% and specificity of 73.6% (Table 9 and Figure 3).

Discussion
e present study was carried out in the Department of Respiratory Medicine, King George's Medical University, Lucknow, for a duration of one year. e prevalence of PH was estimated among outdoor/indoor ILD patients according to both new and old definition. e lower cutoff of the new definition raised the prevalence by 17% among our patients.
Among the 239 ILD patients, hypersensitivity pneumonitis (n � 77) and CTD-ILD (n � 75) comprised most of the study group. Prevalence of other major groups is shown in Figure 1. e higher prevalence of these two subtypes is in accordance with ILD India registry that was the largest study on ILD in the country. e study population had an average age of 52.4 years with female predominance (Table 1). e prevalence of IPF was significantly higher in males (Table 2) as has been previously documented in different international as well as Indian studies [12][13][14]. While connective tissue disorders are more common in females, associated ILD has been reported more frequently in males [15,16]. However, our study had a nearly equal prevalence that is probably due to an overall higher female proportion of the study population. When the older definition of PH was applied, 28.9% of the subjects surpassed the diagnostic cutoff (Table 3). Most fell in the range of mild PH, and only two qualified for the severe PH criteria. Applying the new criteria, the prevalence went up to comprise nearly half of the study population (Table 4). A higher prevalence in Indian scenario could possibly be attributed to a delay in diagnosis of ILD in the peripheral and rural areas that results in unchecked progression of the disease [17,18]. Overall, the prevalence is higher in females, which is in accordance with the existing literature, and the values were not significant though [19,20]. Female sex hormones have been implicated in the probable mechanism for a higher prevalence. Females have a higher number of estrogen receptors in all the tissues including blood vessels. In general, estrogen has the effect of boosting immunity and promotion of remodeling and fibrosis [21,22]. e β type of estrogen receptors increases the arterial tone and raises the pressure, while the α type has a protective role and prevents fibrosis. e balance between these two types of receptors seems to be responsible for female preponderance [23,24]. Both the subjects of severe PH are males; however, the number is too small to consider it relevant. e prevalence seems higher among IIPs other than IPF and other ILDs as the number of study subjects with these diagnosis was quite low. Nearly 18.2% of HSP patients had mild PH that rose to 37.7% with the revised definition (Tables 5 and 6) Although previous studies have utilized the old definition of PH to make the diagnosis, a high prevalence has been noted by many. Several studies have reported variable prevalence of pulmonary        [29][30][31]. In our study, out of 18 patients with CTD-ILD, 10 (55.55%) had systemic sclerosis.
A high prevalence has been reported previously in CPFE and other rare ILDs such as LCH and lymphangioleiomyomatosis in advanced cases [32][33][34]. Oliviera RK and Koschel DS have reported a fairly common prevalence of pulmonary hypertension in chronic hypersensitivity pneumonitis ranging from 19% to 50% [35,36]. e current study has the highest number of HSP subjects which explains the higher number of PH cases in    [40,41]. In sarcoidosis, pulmonary hypertension has been reported in up to 20% of patients, and the mechanisms involved are different. Often, the ones with advanced disease are involved [42][43][44]. Concomitant vasculopathy and inflammatory milieu may also contribute to PH. In the present study, the prevalence of PH among sarcoidosis subjects varied from 33.4% (older definition) to 46.7% (new definition). Most patients report to our tertiary centre in fibrocystic stage in hypoxia, which can account for a higher prevalence of elevated pulmonary artery pressure. e mean TR velocity in mild cases was 1.83 m/s, which decreased to 1.63 m/s on applying the new definition ( Figure 2). Echocardiography and chest radiography showed significant aberration in moderate to severe PH patients (Tables 7 and 8). Previous studies vary on correlation of imaging findings with mean pulmonary artery pressure [45,46]. Indeed, the diameter of the artery varies not just with the pressure but with blood flow as well. Mean pro-BNP and troponin T values were elevated in the subjects with PH (Table 9 and Figure 3). is is in accordance with American Heart Association guidelines [47]. We admit the drawback in our study that the gold standard investigation right heart catheterization was not done. ere have been studies that have established the reliability of transthoracic echocardiography as a noninvasive diagnostic modality [48][49][50].
us, in our study, we could conclude that there were a considerable number of subjects with interstitial lung diseases who had their mPAP in the range of 20-25 mmHg, and with application of the new definition, the pulmonary hemodynamics got the attention earlier which is likely to impact the management and overall quality of life of such patients.

Conclusion
We have ascertained the prevalence of pulmonary hypertension among patients of interstitial lung diseases utilizing two different international definitions. e new definition certainly unmasks a large number of subjects who previously fell in the range of elevated pulmonary artery pressure but could not qualify the criteria for diagnosis.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that they have no conflicts of interest.    Canadian Respiratory Journal