Spectrum of diffuse parenchymal lung diseases with special reference to idiopathic pulmonary fibrosis and connective tissue disease: An eastern India experience.

OBJECTIVE: To evaluate the clinical spectrum of diffuse parenchymal lung diseases (DPLD) encountered in the Indian setting and to compare idiopathic pulmonary fibrosis (IPF) and connective tissue disease associated DPLD (CTD-DPLD), the two commonest aetiologies.
MATERIALS AND METHODS: A prospective study of clinical, imaging and laboratory parameters of patients diagnosed as DPLD and followed up in the Pulmonary Medicine Department of a tertiary-care teaching institution in eastern India was conducted over a period of one year.
RESULTS: 92 patients of DPLD were diagnosed in the study period with IPF (n = 35, 38.04%), CTD-DPLD (n = 29, 31.5%), hypersensitivity pneumonitis (n = 10, 10.9%), sarcoidosis (n = 5, 5.4%) and silicosis (n = 5, 5.4%) being the common causes. The CTD-DPLD group had a lower mean age (39.5 ± 1.86 vs 56.9 ± 1.12 years), a longer duration of symptoms (3.5 ± 0.27 vs 2.5 ± 0.26 years), more extra pulmonary manifestations, significantly more base line FVC and 6-minute-walk-distance than the IPF patients. 19 patients of IPF (54%) opted for treatment. All the IPF patients had a significant fall in FVC after six months (mean change -0.203 ± 0.01 litres) compared to the CTD-DPLD group (mean change - 0.05 ± 0.04 litres.).
CONCLUSION: CTD-DPLD patients belong to a younger age group, with longer duration of symptoms, more extrapulmonary features, better physiological parameters and better response to therapy than IPF patients. Larger prospective epidemiological studies and enrolment in clinical trials are necessary for better understanding of the spectrum of diffuse parenchymal lung disorders and their therapeutic options.

INTRODUCTION

Diffuse parenchymal lung diseases (DPLD) comprise a wide spectrum of disorders, with varied presentations and prognosis. Foremost among the known causes is the connective tissue disease (CTD), the others being occupational or environmental exposure and drug-induced pneumopathies. Among the disorders of unknown cause, idiopathic pulmonary fibrosis (IPF) has the worst outcome.[1] Most publications in the 1980s and 1990s from this country dealt with CTD-associated DPLD (CTD-DPLD)[23] and there were some early reports on fibrosing alveolitis.[45] With subsequent characterization of idiopathic interstitial pneumonias (IIPs),[16] an increase in the burden of IPF is being reported from various centers in India.[789] Although high-resolution computed tomography scan (HRCT) pattern and pathological findings in IIPs are now being applied to CTD-DPLD, there are differences in the treatment and prognosis, making it imperative to differentiate the two conditions.[10]

In this study, we prospectively evaluated the clinical spectrum of DPLDs encountered in the Indian setting and compared the two leading causes of DPLD namely IPF and CTD-DPLD.

MATERIALS AND METHODS

Study design

A prospective observational study was conducted inthe Pulmonary Medicine Department of a tertiary care teaching institution in eastern India over a period of one year, on cases of diffuse parenchymal lung disease. Written informed consent was taken from all the patients and the study was cleared by the institute's Ethics Committee.

The inclusion criteria were:

Patients having clinical features/pulmonary function test suggestive of, and HRCT consistent with DPLD

Age 12 years and above, either sex.

The exclusion criteria were:

Patients with prior corticosteroid therapy, for more than one month

Patients not willing to follow the study protocol

DPLD cases diagnosed to have tuberculosis were excluded from the final analysis.

Detailed demographical and clinical parameters including age, smoking history, environmental, occupational, and drug exposure, duration and severity of breathlessness, clubbing, end-inspiratory crepitations, and extrapulmonary features (rashes, arthritis, Raynaud's phenomenon, dysphagia, oral ulcers), were assessed. HRCT of the thorax was evaluated for ground-glass opacities, reticular shadows, subpleural involvement, septal thickening, nodular lesions, and honeycombing/traction bronchiectasis. Contrast was given to patients suspected to have mediastinal lymphadenopathy. Spirometry, six-minute walk test (6MWT), according to the ATS guidelines,[11] and echocardiography was done in all cases at the baseline. Tricuspid jet velocity >3.4 m/s and estimated pulmonary artery systolic pressure >50 mmHg[12] were considered to be pulmonary hypertension. The antinuclear antibody (ANA) profile, Rheumatoid factor (RF), anti-cyclic citrullinated peptide (anti-CCP), antinuclear cytoplasmic antiboby (ANCA), and serum angiotensin converting enzyme (SACE) were estimated. Bronchoalveolar lavage (BAL) study, transbronchial lung biopsy (TBLB), and biopsies of lymph node, skin lesions or kidney were done as needed.

Diagnosis

A multidisciplinary approach involving a radiologist, pathologist, and rheumatologist was taken, to ensure accuracy of the diagnosis. A case was labeled as IPF as per the current guidelines of the American Thoracic Society/the European Respiratory Society/the Japanese Respiratory Society/the Latin American Thoracic Association (ATS/ERS/JRS/ALAT),[13] that is, exclusion of known causes and the usual interstitial pneumonia (UIP) pattern on HRCT. Diagnosis of CTD and sarcoidosis was based on their respective diagnostic criteria, while a history of exposure, HRCT patterns, and BAL fluid lymphocytosis, were taken into account for diagnosing hypersensitivity pneumonitis.[14]

Management

Cases of IPF were counseled and those consenting for treatment were prescribed a regimen of oral steroids (prednisolone 0.5 mg/kg/day tapering over three months to 0.125mg/kg/day), azathioprine (2 mg/kg/day to a maximum of 150 mg/day) + N-acetylcysteine (600 mg, orally, thrice a day) with modification, if necessary, due to comorbidities.[61315]

Scleroderma patients were put on low-dose corticosteroid (10 mg) plus cyclophosphamide (2 mg/kg body weight),[16] rheumatoid arthritis cases on disease modifying drugs including methotrexate and corticosteroids.[17] Symptomatic sarcoidosis patients with stage 2 or 3 disease or disabling extrapulmonary features were prescribed corticosteroids.[18] The vasculitis group was treated with prednisolone (1 mg/kg) and cyclophosphamide (2 mg/kg),[19] the hypersensitivity pneumonitis patients were treated chiefly by environmental management, and in the severely symptomatic patients a short course of corticosteroids was used.[14]

Follow-up

These patients were followed-up at the DPLD clinic based on clinical, spirometry, and 6MWT tests, every six months. Those on immunosuppressives were monitored monthly for hemogram and renal parameters. A patient with a fall in the forced vital capacity (FVC) of 10% or more from the baseline and/or a fall in the oxygen saturation below 88% before or during the 6MWT, with reduction in the six-minute walk distance, at six months, was considered to be with unstable disease.[13]

Statistical analysis

The statistical analysis was performed using the SPSS version 10.0 (SPSS Inc, Chicago, IL) software for MS-Windows. Cases of IPF and CTD–DPLD were compared with respect to the clinical, radiological, and physiological parameters. Descriptive frequencies were expressed in terms of mean ± standard error of mean (SEM). The P value was calculated using the Fisher's exact test for categorical variables and a student t-test was used for continuous variables and a P < 0.05 was considered to be significant.

RESULTS

Pattern of diagnosis

In this study, a total number of 112 patients with diffuse parenchymal lung diseases were enrolled initially. In 13 patients, a diagnosis of tuberculosis was made, and subsequently in two, lung carcinoma. Another five patients had clinicoradiological features suggestive of idiopathic non-specific interstitial pneumonia (idiopathic-NSIP), but they did not consent to a lung biopsy. These cases were excluded from the study, bringing the final population to 92.

Idiopathic pulmonary fibrosis (IPF) was found to be the most common variety (n = 35, 38.04%) followed by CTD (n = 29, 31.5%) [Table 1]. Of the 29 cases of CTD, systemic sclerosis (n = 17, 58.6%) and rheumatoid arthritis (RA) (n = 8, 27.5%) were the most common followed by systemic lupus erythematosus (SLE) (two cases), polymyositis, and ankylosing spondylitis (one case each). Small vessel vasculitis was diagnosed in four cases, of which two were granulomatous polyangitis, one was Churg-Strauss syndrome, and one microscopic polyangitis.

Table 1

Etiological distribution of diffuse parenchymal lung disease cases

Clinical parameters

In contrast to IPF, a majority of patients of CTD–DPLD belonged to a younger age group (n = 25, 86.2% in 20-50 years) [Table 2]. Although there was not much difference in the pulmonary symptoms and signs between IPF and CTD–DPLD [Table 3] among the extrapulmonary features, clubbing was distinctly more common in IPF (92%), while the Raynaud's phenomenon, dysphagia, and digital ulceration were seen in systemic sclerosis in 82, 58, and 35% of the cases, respectively. Joint deformity was noted in seven out of eight patients of RA, skin rashes and oliguria were seen in the SLE- and ANCA-associated vasculitis group, peripheral neuropathy in those with the Churg-Strauss syndrome, and one case of Wegener's granulomatosis.

Table 2

Demographic parameters

Table 3

Clinical parameters

Clinicophysiological parameters

Compared to CTD–DPLD, IPF patients had significantly less baseline FVC and six-minute walk distance [Table 4]. Pulmonary hypertension was seen in eight cases (47%) of systemic sclerosis (SSc), two cases of RA and one case of polymyositis.

Table 4

Physiological parameters

Imaging

All cases of IPF (n = 35, 100%) showed a classical UIP pattern on the HRCT thorax [Table 5]. The HRCT findings differed among the various subgroups of CTD–DPLDs. A majority of RA (n = 8) patients had a UIP pattern with predominant lower lobe (87%) subpleural (75%) involvement, with septal thickening (75%), and a minority had an NSIP pattern (12%). The SSc patients (n = 17) showed an NSIP pattern in 58.8% and a UIP pattern in 47%. One patient with SLE-DPLD showed a lower lobe, subpleural, reticular shadow, with honeycombing, and the other, bilateral diffuse ground-glass opacity, suggestive of diffuse alveolar hemorrhage. Polymyositis-interstitial pneumonia showed an NSIP pattern, while the case of ankylosing spondylitis had bilateral upper lobe fibrosis.

Table 5

High-resolution computed tomography thorax

Serum markers

Serum autoantibodies were helpful in differentiating CTD–DPLD from the IPF [Table 6]. In RA, The rheumatoid factor and anti-CCP antibody were found to be positive in 100 and 88% of the cases respectively, while in SSc, the anti Scl-70 and anti-centromere antibodies were positive in 88 and 12% of the cases.

Table 6

Serum markers

Spectrum of diffuse parenchymal lung diseases

Besides IPF and CTD–DPLD, ten cases were diagnosed as hypersensitivity pneumonitis (HP). Nine of these cases had exposure to pigeons and presented with chronic symptoms. Ground-glass opacities in the upper lobes with cystic changes and BAL fluid lymphocytosis (>30% of the total cell count) were seen in all the ten patients of HP.

Serum ACE was elevated in all the five patients of sarcoidosis (mean value of 98 U/L). The serum calcium and 24-hour urine calcium were elevated in three patients. The Mantoux test was negative with 10 TU in all five cases. The chest X-ray showed Stage 2 disease in three and Stage 3 disease in two. HRCT showed upper lobar involvement, with peribronchovascular thickening (100%), centrilobular nodules (60%), septal thickening, and ground-glass opacities (40%). Transbronchial lung biopsy (TBLB) showed non-caseating granuloma consistent with sarcoidosis in four cases. Cervical lymph nodes biopsy demonstrated non-caseating granuloma in the other case.

All five cases of silicosis were of workers involved in sandblasting, with upper zone involvement and mediastinal adenopathy (80%) including egg-shell calcification in one case. A case of pulmonary Langerhans cell histiocytosis (PLCH) presented with pneumothorax and characteristic HRCT features [Figure 1], while the case of alveolar microlithiasis had its unique imaging features [Figure 2].

Figure 1

Pulmonary Langerhans cell histiocytosis presenting with pneumothorax

Figure 2

Alveolar microlithiasis

Follow up

All cases were counseled for treatment options and followed up in the DPLD clinic. Nineteen out of 35 patients in the IPF group voluntarily agreed to treatment with the standardized triple drug regimen. Cases of CTD–DPLDs were put on appropriate therapy. Although 47% of the patients in the IPF group receiving the therapy experienced subjective improvement of dyspnea, on spirometry, all cases showed more than a 10% fall in FVC at six months [Table 7]. Eighty-four percent of the treated patients developed systemic hypertension, 63% diabetes, and 52% had at least one exacerbation (mainly infection) requiring hospitalization. Of the remaining 16 patients of IPF, who did not opt for triple drug therapy, after six months, 75% of them expressed subjective improvement of dyspnea, but all had a significant fall in FVC and an oxygen desaturation of <88% after the six-minute walk test. Improvement in the 6MWD, after six months, was seen in three patients (15.7%) in the treatment group and in seven patients (43.7%), who did not receive the triple drug therapy. Therefore, the triple drug therapy for the IPF group did not show any positive outcome. Two patients (on therapy) died within six months due to respiratory failure during exacerbation. Both patients had advanced disease, with a baseline dyspnea of grade 4 MMRC, FVC <1.25 liters, post six-minute walk test, and oxygen saturation <88%. On the contrary, in the CTD–DPLD group, 62.1% of the patients showed subjective improvement of dyspnea, 13 (44.8%) patients showed improvement in 6MWD, with 31% of the cases showing 6MWD of more than 300 m. Mean (± SEM) value of 6MWD and FVC after six months of treatment were 209.14 ± 9.72 m and 1.01 ± 0.03 l in the IPF group, and 267.5 ± 12.7 m and 1.17 ± 0.05 l, respectively, in the CTD–DPLD group.

Table 7

Treatment outcome after six months

DISCUSSION

Our study showed that IPF was the most common entity (38.04%) among DPLD cases [Table 1] followed by CTD–DPLD (31.5%), hypersensitivity pneumonitis (10.9%), sarcoidosis (5.4%), and silicosis (5.4%). The frequency of IPF varied from 28.6 to 46% in two studies from north India[45] and from 43 to 45% in two studies from the south,[79] while CTD–DPLD ranged from 18 to 50.8%.[57] Sarcoidosis presenting with DPLD was reported in 9.6 and 22% among the DPLD cases.[79]

On account of high prevalence of tuberculosis in India, many patients with interstitial lung disorders are misdiagnosed to be with tuberculosis, and receive antitubercular drugs (ATDs).[20] In our study, 64 patients had a prior history of intake of ATDs for more than one month, although none had tuberculosis. Ten percent of the cases of cryptogenic organizing pneumonia[21] and 37% of sarcoid patients[22] from Mumbai were initially diagnosed with TB and received ATDs. On the contrary, tuberculosis may mimic DPLDs, as 13 cases in our study, initially suspected to be DPLD, were subsequently found to be suffering from tuberculosis with the help of a BAL/TBLB study. This was also the finding of others.[7] Awareness about a varied DPLD spectrum and differentiation from tuberculosis is a very important issue in the Indian context.

The mean age of IPF cases [Table 2] in our series was 56.8 ± 8.3 years, similar to other Indian studies (53 ± 10[9] and 50.6 ± 11.9 years[8]), but less than that quoted in western literature (two-third of the cases over 60 and mean age at diagnosis of 66 years).[6] The preponderance of males and smokers in the IPF group in this study is similar to the Indian and western literature.[613] The median time between the onset of symptoms and diagnosis in the IPF group, studied by Johnston et al.,[23] was 12 months. Mean duration of illness before diagnosis was 2.5 years in our IPF group and 2.25 years in the group studied by Subhash et al.[9] Low awareness of the disease, over-reliance on the chest radiograph,[7] and high TB burden is probably responsible for the delayed diagnosis compared to western literature.

Pulmonary symptoms and signs among CTD-DPLD were similar to that of IPF, except clubbing, which was uncommon in this group. The observations of Rajasekaran et al.[24] were similar. Pulmonary hypertension was demonstrated in 22% of our IPF cases and 34.5% of the CTD-DPLD cases [Table 4]. It had been reported in 30% among DPLD cases by Subhash et al.[9] and in a third of the patients with advanced IPF (right heart catheterization) in western literature.[25] The prevalence of isolated pulmonary arterial hypertension and pulmonary hypertension with interstitial lung disease in systemic sclerosis were similar and ranged between 18 and 22% in various studies.[26]

Previously the diagnosis of IPF was based on fulfilling the major and minor criteria in the absence of surgical lung biopsy (SLB).[6] According to the current guidelines, a classical UIP pattern on HRCT, in an appropriate clinical setting, is sufficient to make a diagnosis of IPF.[13]

The HRCT patterns in patients with CTD-DPLD mostly correlate with the lung pathology.[27] Of the four usual patterns described in RA-associated DPLD[17] our cases predominantly showed UIP followed by NSIP, but none had an organizing pneumonia or bronchiolitis pattern. Although, as with western literature, NSIP was the most common radiographic subtype among SSc-associated DPLD,[17] there was also a significant number of our SSc patients with a UIP pattern. Therefore, in our series, CTD associated DPLD had more of a UIP pattern (n = 16, 55.2%) than NSIP (n = 12, 41.4%), although traditionally NSIP was known to be more frequent.

Decline in the FVC of 10% or more at six months and decreased 6MWD with desaturation have been shown to be measure of disease progression and surrogate markers of mortality.[13282930] In our study, 54% (n = 19) of the cases of IPF, who opted for the standardized triple drug regimen, with prednisolone, azathioprine, and acetylcysteine, showed no benefit (in fact, there were more drug side effects) compared to the 46% (n = 16) who opted for only supportive treatment. The triple therapy (prednisolone, azathioprine, and NAC) arm had to be discontinued due to excess morbidity and mortality compared to the placebo arm and the NAC arm in PANTHER-IPF study.[3132]

The SSc-associated DPLD patients treated with oral cyclophosphamidewere found to have less change in FVC and functional ability compared to the placebo-treated patients in the North American Scleroderma Lung study.[16] However, a recent meta-analysis found no significant improvement of pulmonary function with cyclophosphamide treatment.[36] Unlike in the IIPs, there appears to be no difference in survival between those with an NSIP and those with a UIP pattern.[37]

Treatment of RA-DPLD is essentially empirical in the absence of randomized controlled trials, and the best response has been reported with RA-associated organizing pneumonia.[17] In their study, Rajasekaran et al.,[24] found patients with RA-DPLD to have a better prognosis than those with IPF, with median survival rate of 60 months versus 27 months, respectively. However, no distinction regarding prognosis was made between the UIP and NSIP patterns in their RA-DPLD group.

Park et al.[35] reported a better prognosis of the CTD-DPLD group, not only due to the higher prevalence of NSIP in this group, but also due to a better prognosis of CTD–DPLD patients, even those with a histological pattern of UIP, compared with patients with IPF. However, some studies quote a poor prognosis and similar mortality rates between the two groups.[3334]

In this study, as a whole, the CTD–DPLD group fared better than the IPF group (both the treated and observation arms) at six months of therapy [Table 7].

Our study is probably one of the few studies from India that has looked at the spectrum of DPLD prospectively and has attempted a comparison between two of its largest contributors. However, the study has certain limitations, in that, the diffusing capacity of the lung could not be done for monitoring the disease progress; triple drug therapy versus supportive therapy in the IPF group was not randomized, the CTD–DPLD groups were not age- or sex-matched, and surgical biopsy was not feasible.

To conclude, the picture of diffuse parenchymal lung disease revealed in our study is similar to some of the retrospective studies from India. IPF seems to have presented a decade earlier in our country compared to the West. Both the burden of tuberculosis and its role as a ‘mimicker’ of DPLD caused a significant delay in the diagnosis of IPF in our country. Compared to IPF, the CTD–DPLD patients belonged to a younger age group, with a longer duration of symptoms, more extrapulmonary features, better physiological parameters, a mix of NSIP and UIP patterns on HRCT, and a better response to therapy. Larger prospective epidemiological studies and enrollment in clinical trials are necessary for a better understanding of the spectrum of diffuse parenchymal lung disorders and their therapeutic options.