Pulmonary hypertension in idiopathic pulmonary fi brosis

1687-8426 © 2014 Egyptian Journal of Bronchology DOI: 10.4103/1687-8426.137348 Introduction Idiopathic pulmonary fi brosis (IPF), the most common form of idiopathic interstitial pneumonias, is a chronic, progressive, irreversible, and usually lethal lung disease of unknown cause with an estimated survival of 2.5–5 years. IPF occurs in middle-aged and elderly adults, is limited to the lungs, and is associated with a histopathological or radiological pattern typical of usual interstitial pneum( UIP) [1–3].


Introduction
Idiopathic pulmonary fi brosis (IPF), the most common form of idiopathic interstitial pneumonias, is a chronic, progressive, irreversible, and usually lethal lung disease of unknown cause with an estimated survival of 2.5-5 years. IPF occurs in middle-aged and elderly adults, is limited to the lungs, and is associated with a histopathological or radiological pattern typical of usual interstitial pneum( UIP) [1][2][3].
Th e diagnosis of IPF often requires a multidisciplinary approach, involving pulmonologists, radiologists, and pathologists experienced in the fi eld of interstitial lung diseases [4].
Pulmonary hypertension (PH) is a hemodynamic and pathophysiological condition defi ned as an increase in mean pulmonary arterial pressure (mPAP) to at least 25 mmHg at rest as assessed by right heart catheterization (R HC) [5,6]. PH often complicates the course of IPF and may even be found in patients with preserved lung function. PH in IPF patients is associated with decreased exercise capacity and a worse prognosis. Although pulmonary artery pressures (PAPs) and other cardiac hemodynamic parameters can be accurately assessed by RHC, a simple, reliable, and noninvasive method to estimate PAPs and diagnosis of PH in patients with advanced lung disease is preferable. Doppler echocardiography has gained popularity in the last two decades for noninvasive estimation of systolic pulmonary artery pressure (S PAP) from the peak velocity of a tricuspid regurgitant jet. Doppler echocardiography is a useful tool for the detection of PH, which also provides additional information regarding associated cardiac abnormalities [7].

Aim
Th e aims of this study were to: (1) Determine the prevalence of PH among patients with IPF.
(2) Detect the correlation between pulmonary function tests (PFTs), echocardiographic parameters, radiological fi ndings, and mPAP in these patients. (3) Compare the eff ect of oxygen therapy ± drugs in the form of corticosteroids, diuretics, and immunosuppressive drugs on PH among patients with IPF.

Patients and methods
Th is study was performed on 60 patients with IPF who were randomly selected from outpatient clinics of chest and internal medicine departments at the Minia University Hospital from April 2010 to June 2013.
Th is study was approved by the ethics committee of Faculty of Medicine, Minia University, and informed consent was obtained from patients and controls.
Patients with IPF were diagnosed according to the following criteria: (1) Clinical diagnosis of IPF [8] in the form of progressive d yspnea, dry cough, and bilateral basal or widespread crackles. A restrictive ventilatory defect with a reduction in total lung capacity (T LC) and forced vital capacity (F VC) less than 80% predicted [2]. (2) High resolution computed tomography (CT) criteria for UIP pattern in the form of subpleural, basal predominance, reticular abnormality, and honeycombing with or without traction bronchiectasis.
Th e absence of features listed as inconsistent with UIP pattern are: Upper or mid-lung predominance, peribronchovascular predominance, extensive ground glass abnormality, profuse micronodules, discrete cysts (multiple, bilateral, away from areas of honeycombing), diff use mosaic attenuation/air-trapping (bilateral, in three or more lobes), and consolidation in bronchopulmonary segment(s)/lobe(s) [3].
Patients with occupational exposure to a fi brogenic agent, those on medications that cause interstitial pulmonary fi brosis, and those with a history of connective tissue diseases were excluded.
All patients were subjected to the following: (1) Complete history taking and physical examination.
(2) Routine laboratory investigations and collagen profi le. CT pattern and extent of IPF were determined either to be reticulonodular or honeycombing, basal subpleural or diff use. Th e main pulmonary artery diameter (MP AD) was measured at its widest dimension; at this same level, the widest ascending aorta diameter (AD) was measured and the MPAD/AD ratio was calculated (Figs 1 and 2).
Idiopathic pulmonary fibrosis on multidetector chest computed tomography.

Fig. 1
Measurements of main pulmonary artery diameter (mPA) and ratio of the diameters of main pulmonary artery to ascending aorta. (8) Doppler echocardiography was performed using conventional clinical echocardiographic equipment (GE Vivid 3; General Electric) with 2.5 or 3.5 mHz transducers. Tricuspid regurgitant fl ow was identifi ed by co lor fl ow Doppler techniques, and the maximum jet velocity was measured by continuous wave Doppler without the use of intravenous contrast. Right ventricular (RV) systolic pressure was estimated on the basis of the modifi ed Bernoulli equation and was considered to be equal to the SPAP in the absence of RV outfl ow obstruction: SPAP (mmHg) = RV systolic pressure = (4 × TR V 2 + RAP) where TRV is the tricuspid regurgitation velocity in m/s and RAP is the right atrium pressure [11,12]. RAP was estimated to be 5, 10, or 15 mmHg on the basis of the variation in the size of inferior vena cava with inspiration as follows [13]: complete total collapse, RAP = 5 mmHg; partial collapse, RAP = 10 mmHg; and no collapse, RAP = 15 mmHg. (a) mPAP was calculated as follows: mPAP = (0.61 × SPAP + 2 mmHg) [14]. (b) RV systolic function using right ventricular fractional area change (RV -FAC) was obtained from a four-chamber view, where the right ventricular end-diastolic area (RV -ED A) and right ventricular end-systolic area (RV -ES A) were measured, and the RV -FA C was calculated as follows: RV-FAC (%) = (RV-EDA-RV-ESA)/RV-EDA × 100. Normal value for RV-FAC is above 35% [15]. RV-FAC is a measure of RV systolic function that has been shown to correlate with RV ejection fraction by MRI [16,17]. (c) Tei index was calculated using tissue Doppler imaging. Tei index = isovolumic contraction time+isovolumic relaxation time/ejection time. It is an index of RV performance; in adults, values for the RV less than 0.4 are considered normal [18].
PH is defi ned as mPAP of at least 25 mmHg at rest [5].
According to this defi nition the 60 patients of IPF were divided into two groups: Group A: included 14 patients without PH.
Group B: included 46 patients with PH.
Th e study also included 15 healthy volunteers as a control group (group C) matched in age and sex with the studied patients.
According to the line of treatment they received as inpatients, group B was divided into two subgroups: group B1 included 24 patients who received continuous oxygen therapy (15-20 h/day) only by venturi mask ranging from 28 to 31% and group B2 included 22 patients who received prednisolone 1 mg/kg/day, frusemide diuretics 60 mg/day, and azathioprine 50 mg/day increased by 25 mg every week; they also received continuous oxygen therapy by venturi mask ranging from 28 to 31%.
Follow-up echocardiography was performed after 1 month among both B1 and B2 subgroups.

Statistical analysis
Statistical analysis was performed using the statistical package for social science (S PSS, version 16; SPSS Inc., Chicago, Il linois, USA) software on a personal computer. Statistical analysis was performed using the χ 2 -test and the independent sample t-test to assess diff erences between proportions. Diff erences were considered statistically signifi cant when P value was less than 0.05. Bivariate correlation using Spearman's test was used. Table 1 shows that the prevalence of PH in IPF patients was 76.7%. On studying the general characteristics of the studied groups, it was found that there was no diff erence in the mean age in group A versus group B. Th ere was also no diff erence between both groups of IPF regarding duration of the disease. Considering GAP index stage, there was no signifi cant diff erence in GAP stage among those with or without PH.

Results
With respect to PFTs, there was a very highly signifi cant decrease in FVC% predicted, FEV 1 % predicted, and TLC% predicted and a signifi cant increase in CPI in group B compared with group A. In contrast, there was no signifi cant diff erence between both groups of IPF regarding DLCO% predicted. Th ere was a signifi cant diff erence in the mean PaO 2 among group B versus group A (Table 2). Table 3 shows echocardiographic fi ndings among the studied groups. It was found that group B had a highly signifi cant higher level of SPAP and mPAP than group A and the healthy control group (group C). Besides, there was a statistically signifi cant diff erence between group A and group B regarding RV-FAC% and Tei index. Figure 3 shows that there is a highly signifi cant negative correlation between mPAP and FVC% predicted (r = −0.630, P = 0.0001). Besides, there was also a highly signifi cant negative correlation between mPAP, TLC% predicted, DLCO% predicted (r = −0.502, P = 0.0001), and PaO 2 (r = −0.584, P = 0.001). In contrast, there was highly signifi cant positive correlation between mPAP and CPI (r = 0.550, P = 0.0001).
It was found that there was a signifi cant positive correlation between mPAP, age, duration of illness, Modifi ed Medical Research Council (MMRC) scale, GAP index stage, PA diameter, and its ratio with the ascending aorta (PA/A). In contrast, there was no correlation between mPAP and both pattern and extent of IPF in MDCT (Table 4).   performed TTE on 88 patients with IPF who were evaluated for lung transplantation. Eighty-four percent had PH (defi ned as an estimated SPAP > 35 mmHg at rest). Studies by Agarwal et al. [24] and Laz and Ahmad [25] found that the prevalence of PH in IPF patients was 36 and 33.3%, respectively, as detected by TTE.
PH develops over time in patients with IPF, as Figure 4 reveals that there is a signifi cant positive correlation between mPAP and RV T ei index.
Regarding correlation of RV systolic function as represented by RV-FAC% and PFT parameters, there was a signifi cantly positive correlation between RV-FAC% and DLCO% predicted. In contrast, there was a signifi cantly negative correlation between RV-FAC% and CPI. Th ere was a signifi cantly negative correlation between Tei index and all PFTs parameters and a signifi cantly positive correlation between Tei index and CPI. In contrast, there was no signifi cant correlation between RV-FAC% and Tei index and PaO 2 ( Table 5).

Discussion
PH is a common accompaniment of IPF and has a signifi cant impact on outcomes [8,19]. Some studies have investigated transthoracic echocardiography (TTE) as a noninvasive means of detecting PH and have demonstrated that elevated estimated SPAP is associated with reduced survival using thresholds of 40-50 mmHg [8,20].
In our study, SPAP is estimated by echocardiography as it is a simple, noninvasive, and nonexpensive method. Th e prevalence of PH in IPF patients in the present study was 76.7% on the basis of mPAP.
Most studies on PH in IPF were in patients referred for lung transplantation, in which the reported prevalence of PH is 32-46% [9,[21][22][23]. Th ese studies determine PH by RHC as a routine investigation before lung transplantation. Nadrous et al. [8]   Correlation coeffi cient between mean pulmonary artery pressure and the right ventricular (RV) Tei index among the studied groups.

Fig. 4
Correlation coeffi cient between mean pulmonary artery pressure and the forced vital capacity (FVC) % predicted in all studied groups. demonstrated by the rise in prevalence of PH in IPF patients awaiting transplantation from 33% at initial assessment up to 85% immediately before transplantation [8].
Th is wide range in the prevalence of PH reported likely refl ects the timing of the measurement during the course of the patient's disease, with patients who are later in their disease course manifesting more evidence of pulmonary arterial hypertension (PAH). It is possible that the high prevalence of PH in our study could be due to the fact that the studied patients as a whole were in relatively late stages of IPF when they underwent the study, as 67% of them had diff use honeycombing pattern in MDCT and the mean duration of illness was 4.5 years.
Th e onset of PH correlates with the magnitude of reduction in lung volumes and DLCO. In the present study, it was found the mPAP had an inverse correlation with all parameters of pulmonary functions (FVC, TLC, and DLCO) and PaO 2 and had direct correlation with CPI. In addition, it was found that 22 patients of the 27 (81.48%) patients with FVC less than 50% predicted and 24 (88.8%) patients with DLCO less than 40% predicted had PH (Table 6).
It was also found that seven of the 13 patients (53.8%) with FVC%/DLCO% greater than 1.5 had PH (Table 6). Nadrous et al. [8] and Hamada et al. [26] found a negative correlation between mPAP and DLCO% predicted and PaO 2 . In contrast, Nathan et al. [9] failed to demonstrate a signifi cant relationship between various PFTs and mPAP. Somewhat better correlation was noted with DLCO% less than 30 having a two-fold higher prevalence of PH (56.4%) compared with DLCO% more than 30 (28.6%). Th ey also found that a FVC%/DLCO% greater than 1.5 was associated with a nearly two-fold increased risk for PH.
Our study revealed that there was a positive signifi cant correlation between mPAP, age, duration of IPF, MMRC scale, and GAP index. Laz and Ahmad [25] observed signifi cant positive correlation between dyspnea scale and SPAP in patients with IPF-PH (r = 0.67, P = 0.017).
It has been reported that easily accessible CT parameters, such as the MPAD, correlate well with hemodynamic measurements on RHC and can therefore be used to assess the probability of PH [27][28][29].
In the present study, it was found that patients with IPF-PH had a signifi cantly higher PA diameter than those without PH and healthy controls (mean value was 27.9 ± 4.8 vs. 24.16 ± 2.16 and 23.7 ± 1.9 mm, respectively, P = 0.02). We also found that the ratio PA/A was signifi cantly higher among both controls and IPF patients without PH (0.97 ± 0.16 vs. 0.85 ± 0.1 and 0.88 ± 0.12, respectively, P = 0.01).
We found that there was a signifi cant positive correlation between both PA diameter and ratio PA/A on MDCT and mPAP in IPF patients (r = 0.58, P = 0.001 and r = 0.3, P = 0.01, respectively). Nayar et al. [35] found that the main PA greater than 29 mm on CT scan compared with RHC mPAP had a correlation coeffi cient of 0.56 (P<0.0001), and CT prediction of MPAD/AA diameter greater than 1 had a coeffi cient of 0.42 (P = 0.004) in patients with advanced lung disease. Another study by Lang et al. [31] found that the MPAD on CT correlated with mPAP (r = 0.496, P < 0.001) and pulmonary vascular resistance (r = 0.445, P < 0.001), and could predict borderline PH.
In contrast, Zisman et al. [36] found that there was no signifi cant correlation between mPAP and any of the chest CT-determined measures such as extent of pulmonary fi brosis, MPAD, and the ratio of the PA to AD.
RV systolic dysfunction has been identifi ed as a key element in determining prognosis of patients with chronic PH [37,38]. Th erefore, identifi cation of early RV dysfunction is of utmost clinical importance because as many as two-thirds of the deaths in patients with chronic PH may be attributed to RV failure [39]. It was documented that IPF patients exhibited impairment of both systolic and diastolic RV function. In the current study, we assessed one of the RV systolic functions through detection of RV-FAC% and it was found that IPF patients with PH had a statistically signifi cant diff erence than those without PH and healthy controls (44.5 ± 1.1 vs. 53.3 ± 1.2 and 53.8 ± 1.1, respectively, P = 0.01). Papadopoulos et al. [40] also found that IPF patients with PH had a worse RV-FAC% than controls (42 ± 5 vs. 57 ± 6, P < 0.001).
Th e myocardial performance index, also known as Tei index, incorporates elements of both systolic and diastolic phases in the assessment of global ventricular function [18]. Th e Tei index can be estimated for both the left ventricle and the RV. Th e RV Tei index is a candidate to increase the noninvasive diagnosis of PH because it refl ects the RV function, is easy to assess, and can be estimated in the same session as the echocardiographic pulmonary artery systolic pressure (PASP). Th e normal value of the RV Tei index is less than 0.4 [18]. An elevated RV Tei index should be the result of either diastolic dysfunction or PH [41,42].
Our study showed that the mean RV Tei index in patients with IPF-PH increased substantially than normal controls and those with IPF and without PH. Th e RV Tei index was elevated in 56.5% of the patients with PH, suggesting that they had RV dysfunction. Th ere was also a signifi cant correlation between RV Tei index and mPAP.
Vonk et al. [43] found a signifi cant correlation between the RV Tei index and the catheterization parameters such as SPAP, diastolic PAP, and mPAP among patients with PH due to systemic sclerosis.
Th e current therapeutic options are quite limited for IPF, and even more so for IPF-associated PH. Th e major problem in treating both fi brosis and pulmonary vascular disease is the amount of organized scar tissue inside the fi brotic lung. Th ese areas represent regions of fi nal, nonreversible damage, not only to the interstitium, but also very likely to the pulmonary vasculature [44].
Finally, we compared the eff ect of oxygen therapy alone versus oxygen with drugs in the form of corticosteroids, diuretics, and immunosuppressive drugs on PH among patients with IPF and PH.
Th ere was no signifi cant diff erence between both groups of patients who received oxygen therapy alone or when oxygen was augmented by drugs regarding mPAP before and after treatment (P > 0.05) ( Table 7).
Previous study was performed on 18 patients with idiopathic lung fi brosis [45]. Douglas et al. [46] found that there was no signifi cant diff erence in survival between those patients of IPF treated with colchicine or prednisone and those on no therapy, and no diff erence between those on oxygen therapy and those without oxygen.
Cochrane review on the use of corticosteroids in IPF [47] showed that no existing evidence supported the effi cacy of corticosteroids for treatment of IPF patients. In 2010, the Cochrane meta-analysis on the use of corticosteroids for IPF was updated and showed that there was still no evidence to support the effi cacy of corticosteroids in the management of IPF, but there was also no evidence to rule out the use of corticosteroids in IPF.
Th e present study had some limitations. First of these limitations was the diffi culty in measuring mPAP by cardiac catheterization, as it is more invasive and expensive technique. Second, the comorbid conditions that may contribute to PH were not included in this study. Finally, other shortcoming of this study was that the duration of treatment was limited to 1 month. Th is was because admission of these patients was not feasible for longer periods.
In conclusion, PH is common in IPF. Our study revealed a signifi cant negative correlation between PFTs (FVC%, TLC%, DLCO%), PaO 2 , and PH. Th ere was also a signifi cant positive correlation of mPAP with both CPI and RV Tei index. Age, duration of IPF, MMRC scale, and GAP index stage also had a signifi cant positive correlation with PH. In contrast, both the extent and pattern of IPF on CT had no correlation with PH. Measurement of MPAD and its ratio to the ascending aorta as seen in CT chest can be easily performed by the clinician to discover the early presence of PH. Evaluation of the presence of PH in patients with IPF is useful in determining prognosis and eff ect of therapy.
Further studies are required to validate our fi ndings and to evaluate therapy directed to prevent or treat this complicating comorbidity.