Hypothesis

Tea garden population is an agrarian population of Assam, mostly associated with tea plantation. Assam is a state in the north eastern part of India which is a major tea producer state. A total of 803 tea gardens are present in Assam out of which 177 are present in the Dibrugarh district alone. State of Assam accounts for more than 50% of the tea production of India. Dibrugarh is famously known as “Tea City of India”, and is surrounded by the tea estates. It also has a high burden of tuberculosis especially in the people working in tea gardens (Nikshay portal, Govt of India).

Hence, this cross-sectional study was performed from the tea garden community of Dibrugarh district with the following aim and objective:

1. Detection of Aspergillus fumigatus IgG antibody in patients with chronic respiratory symptoms.
2. Detection of Chronic pulmonary aspergillosis in this population.

Ethics statement

Ethics approval was obtained from the Institutional Ethics Committee (Human) with registration number ECR/2003/Inst/AS/2024 issued under New Drugs and Clinical Trial Rules, New Delhi, India, 2019. Approval to conduct this study by IHEC was granted on 16^th February 2023. Informed consents (in the form of signatures or thumb prints) were obtained from each and every individual in their own language (Assamese).

Study design

This was a community-based cross-sectional study that included four tea gardens and associated tea garden hospitals (3 tea garden hospitals and 2 referral hospitals) situated within a 15 km radius from a teaching hospital and a tertiary care centre, Assam Medical College and Hospital, where this study was conducted.

Study setting

Each tea garden hospital was visited for sample and data collection until the required sample size was fulfilled. Every subsequent patient with significant chronic respiratory symptoms of more than 3 months duration was taken up for the study, and house-to-house sample collection was done as and when required. Radiological test data were collected. Patients in whom chest X-ray/HRCT was unable to be performed were also included in the study. Before initiating the study, a meeting was arranged with the in-charge doctors, nursing staff, Accredited Social Health Activist (ASHA) workers, and other staff of the tea garden hospitals. The details, such as the purpose of the study, samples that will be required, and the type of patients who will be included, were discussed in the meeting in lay terms in their local language.

Sample size

The sample size was calculated to be 138 using Fisher’s formula (“n = Z² 4pq ÷ d²”) with a prevalence of 10% taken from the latest Indian study [1]. The coefficient interval was 95% with a margin of error of 0.05. A total of 128 samples were collected during the study period. Adults with chronic respiratory symptoms as mentioned in the above specific criteria were included.

Case definition

Chronic pulmonary aspergillosis can be diagnosed on the basis of four required categories as given by the ‘Global Action Fund for Fungal Infections’ (GAFFI) and used by the CDC (Centre for Disease Control) as follows [2]:

1. Clinically, major required symptoms are persistent cough with or without haemoptysis and weight loss for a minimum period of 3 months. Others commonly present are fatigue/tiredness, productive sputum, dyspnoea, and chest pain (but not a required criteria).
2. Microbiologically, seropositivity of Aspergillus-specific IgG antibody and/or microscopy of sputum samples showing (dichotomous branching and septate hyphae) hyphae of Aspergillus and/or growth of Aspergillus species in more than 2 sputum samples or other respiratory samples such as bronchoalveolar lavage (BAL), lung biopsy, and fluid aspirated from cavity (if present) or from pleural effusion is required.
3. Radiologically, chest imaging that shows either cavitation or a fungal ball inside the cavity or thickening of the pleura or fibrosis or infiltrates around the cavities.
4. A negative sputum smear, GeneXpert, and/or mycobacterial culture in order to rule out mycobacterial infection is necessary.

Proven CPA: Patients with chronic respiratory symptoms of cough and/or haemoptysis, weight loss, and/or other respiratory symptoms with or without underlying significant respiratory conditions, Aspergillus IgG seropositivity (>27 mgA/l in Immunocap testing), and radiological evidence from HRCT/chest X-ray were taken as ‘Proven CPA cases’ in our study.

Possible CPA: Cases with risk factors and significant respiratory symptoms of cough and/or haemoptysis, weight loss, and/or other respiratory symptoms with Aspergillus fumigatus-specific IgG seropositivity where radiological test records were not found were included as ‘Possible CPA’ in our study.

The following figure (Fig 1) depicts the diagnostic guidelines followed for the detection of proven and possible CPA cases in this study.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Diagnostic chart for detection of proven and possible CPA cases.

https://doi.org/10.1371/journal.pntd.0012756.g001

For concurrent CPA and pulmonary tuberculosis infections, Aspergillus seropositivity, chest X-ray/Computed Tomography (CT) scan showing any of the defining features of CPA as described by GAFFI, and mycobacterial positivity (a positive sputum smear, GeneXpert) were taken in our study.

CPA can either present as a co-infection with TB or as a complication after anti-TB therapy [3]. Hence, the following periods of estimation can be considered while evaluating cases of CPA as proposed by Denning and Ray in their study [4]:

• Early (2–6 months during ATT).
• Late (7–12 months after starting ATT).
• 2–5 years after completion of ATT.
• More than 5 years after diagnosis of TB.

In our study, we have labelled the patients belonging to the early period as “CPA with active TB” and those belonging to the rest of the periods as “CPA with post TB.”

Data collection tools

Laboratory diagnosis

Prevalence of chronic pulmonary aspergillosis

Twenty-two out of 41 (53.6%) serologically and clinically significant patients could be followed up with HRCT/chest-X ray and all of them (100%) had radiological as well as clinical features consistent with CPA. As this study was done in a resource- and time-limited setting of 1 year, the radiology follow-up was limited. The rest of the 19 (46.3%) serologically significant cases with clinically significant respiratory symptoms without radiological data were categorized as possible CPA [2]. The total population of the four tea gardens (namely Greenwood, Maijan, Bokel and Chabua tea gardens) included in the study was 20,903; where 14 cases of CPA were detected (out of 66 clinically suspected patients from the four tea gardens), which gives us a prevalence of 60 per 100 000, which is higher than the global prevalence reported. The remaining 62 clinically suspected patients of the tea population were from two referral hospitals, from whom 27 CPA cases were detected. Global prevalence of CPA is reported to be 42 per 100 000 [5], while in India it was estimated at 0.02% in 2011 by Denning et al [6]. Revised estimation of 5-year prevalence of CPA in India related to TB was reported as 1.5 million cases [7]. In the US and Europe, the prevalence is <1 per 100 000 (0.6) and 42.9 per 100 000 in the African nations of the Democratic Republic of Congo and Nigeria [5]. In another study done in France by Maitre T et al., the prevalence was estimated to be 5.02 per 100 000 in 2018 [8].

A total of 41 out of 128 patients (32.0%) were categorized as CPA (22 Proven CPA 17.1%, 95% CI 10.7% −26.0% and 19 Possible CPA 14.8%, 95% CI 8.9%−23.1%) out of these patients. In our study, the prevalence of Aspergillus IgG seropositivity in post-TB population with chronic respiratory symptoms attending garden hospitals is 48.9% (24 out of 49 post-TB patients). One systematic meta-analytic review from Asia has reported the prevalence of Aspergillus antibodies in post-TB patients as 14.7% [9]. A study from North India by Sehgal et al. showed a prevalence of CPA as 10.3%, and another from New Delhi by Jha et al. showed a prevalence of 21.5% in post-tuberculosis patients [1,10]. A recently published review, which gives the revised burden from India, determines the rate of developing CPA after pulmonary TB diagnosis to be 10% in the first year and 1.5% in the subsequent 2–5 years [4]. Other studies from Sierra Leone, Kenya, and Ghana showed prevalence in the tuberculosis population of 20.8%, 19.7%, and 50%, respectively [11–13].

The tea garden community of Assam is an agrarian community, the adult population of which works closely in tea plantations. Many dependents of this population also work outside the tea gardens, engaged in various supportive occupations for the society, like masons, plumbers, carpenters, manual labourers, etc. This study is done for the first time in the tea population of Eastern Assam to get an insight into the role of Aspergillus species, commonly Aspergillus fumigatus, in the causation of respiratory illness. The Aspergillus fumigatus-specific IgG antibody level is helpful in diagnosing several respiratory disorders, mainly CPA (chronic pulmonary aspergillosis), ABPA (allergic bronchopulmonary aspergillosis), chronic Aspergillus sinusitis/fungal ball, hypersensitivity pneumonitis/extrinsic allergic (broncho) alveolitis, and Aspergillus bronchitis (in cystic fibrosis and bronchiectasis) [2]. Out of 128 patients, 44 (34.4%) patients had antibody levels of more than 21 mgA/l while 41 (32.0%) patients had levels more than 27 mgA/l. In a recent study in Taiwan, the cutoff value of significant antibody was determined to be as low as 20 mgA/L [14]. In a couple of latest Indian studies from North India by Agarwal et al. and Sehgal et al., it was determined as 27 mgA/l for diagnosis of CPA [15,16].

In our study, Aspergillus fumigatus-specific IgG antibody was found to be significantly high (>27 mgA/l) in 29 out of 76 patients (38.1%) with a history of tuberculosis (p-value: 0.0355). Amongst them, seropositivity with active TB was in 5 out of 27 patients (18.5%), and with post TB it was in 24 out of 49 patients (48.9% & p-value: 0.0070). The range of significant antibody levels found in post-tuberculosis patients was 27 mgA/l to 216 mgA/l. In a recently published study by Jha et al. from New Delhi, seropositivity rate was 38.9% in patients treated for TB [10]. In another study from Sierra Leone in Africa, the Aspergillus seropositivity is reported to be 20.8% as against 19.7% from Kenya in the post-tuberculosis patients [11,12]. In two studies from Indonesia, a total seropositivity of 43% was reported in patients treated for TB, 26.9% in bacteriologically confirmed TB patients, and 18.2% in those without bacteriological confirmation respectively [17,18].

In our study, in patients without a history of TB, the range of Aspergillus fumigatus-specific IgG antibody observed was 2.38 mgA/l to 105 mgA/l and significantly high antibody levels (>27 mgA/l) were found in 8 out of 42 patients (19.0%). The antibody levels were found to be statistically significantly high in the population with a history of tuberculosis in comparison with the non-tuberculosis group (p-value: 0.0355).

The limitation of the study is the lack of radiological evidence in the 19 cases. Hence, the prevalence of proven CPA (22 out of 128 clinically significant patients) in our study is 17.2%. If the chest radiography was included, the true prevalence in this population would have been considerably greater. Also, because of various socioeconomic reasons in this population, the cases may not report to the hospitals early enough for diagnosis.

The male to female ratio in the study was 1.6:1 with a male preponderance; the same was observed in the CPA cases as well. Similar findings were seen in studies done in France, Sierra Leone and Ghana [8,11,13]. However, the Kenya study showed a female preponderance [12].

In our study, patients in the age group of 21–50 years were predominant, and 31.9% of them had CPA [mean age 41.89 (±15.69)]. Similar findings were seen in Ghana (mean age 41.5 years) and Indonesian (mean age 44.4 years) studies [13,17]. However, in the Sierra Leone study, patients aged more than 45 years were having a higher prevalence of CPA [11]. In our study, 34.6% of the married people and 19.0% of the singles had CPA. However, the Sierra Leone study had 56.9% married patients in their study [11].

Almost half of the subjects in our study were working in the tea gardens or tea factories, while the other half of them were dependents of the tea garden workers who were students, unemployed, or employed in the same district doing various jobs (Mason, carpenter, plumber, manual labourers etc., or unemployed). Though there was no statistically significant difference in the occurrence of CPA between these two groups. In our study, CPA was more commonly seen in smokers (39.3%) than in non-smokers (29.5%) and did not have a statistical significance. CPA was predominant in non-smokers from another study from Kenya, in which 23.9% were smokers and 76.1% were non-smokers [12].

In our study, besides TB, other respiratory illnesses such as COPD and asthma were not found to be statistically significant as independent predictors of Aspergillus seropositivity. CPA was found in only 1 out of 13 (7.7%) asthma patients in our study. But the number of asthma patients in our study was also quite low. Studies elsewhere have shown a prevalence of 9.8% from Sierra Leone, 10.3% from Manchester and 3.9% from the Ghana study [11,13,19]. Out of 6 COPD patients, 3 had CPA in our study, which is 7.3% of the total CPA cases reported (3 out of 41 CPA cases). COPD was found to be a risk factor for CPA in 43.9% in a France study, whereas 2.4% and 5.8% were estimated in a study from Sierra Leone and Ghana, respectively [8,11,13].

In our study, 1.5% of the enrolled patients had chronic pulmonary aspergillosis with HIV, which is much lower than 6% as reported in a literature review from India by Denning et al. while in a Nigerian study, 10% of HIV-infected patients had CPA [4,20]. In another Mozambican study by Salzer et al. no CPA cases were found in HIV-positive TB patients [21].

Clinical symptoms most commonly associated with CPA were haemoptysis and fatigue/tiredness with a high statistical significance (p-values: 0.0086 and 0.0098, respectively). This is consistent with case definition given by GAFFI [2]. In a similar study done by Jha et al. the most common symptoms in association with CPA were dyspnoea and haemoptysis (p-value: 0.066 and 0.011) [10]. In a study by Setianingram et al. cough, haemoptysis, and fatigue were significantly associated. Productive cough and dyspnoea were commonly present in patients with CPA in a Vietnamese study [22]. In African studies from Ghana, Sierra Leone, and Kenya, the symptoms most commonly found were cough and chest pain, followed by fatigue, weight loss, and night sweats [11–13].

The most significant radiological findings in CPA patients in our study were single and multiple cavity formations in the lungs followed by fungal balls. Nodules and consolidations were the most common findings in a study done by Jha et al [10]. However, the APICAL study states cavitation, pleural thickening, and infiltrates to be the most common radiological findings in CPA patients. While one study from Vietnam also states multiple cavities and pleural thickening as the common radiological findings [17,22]. Cavitation, para-cavitary fibrosis, and fungal ball were seen commonly in CPA patients in Ocansey et al.’s study from Ghana [13]. Infiltrates, cavitary changes, pleural effusion, pleural thickening, and fibrotic changes were significantly associated with CPA cases in Lakoh et al.’s study from Sierra Leone [11].

In our study, chronic cavitary aspergillosis (CCPA) was the most common type of CPA, followed by chronic fibrosing pulmonary aspergillosis and simple aspergilloma. Similarly, CCPA (88.9% & 3.9%) was the most common finding in Page et al. & Denning et al.’s studies [19,23].

Though the sample size is less in our study, we included cases from 3 tea garden hospitals and 2 referral hospitals. Therefore, the estimations of the prevalence of CPA in tea garden populations with chronic respiratory illness approximate its true prevalence.