“Allergic Bronchopulmonary Aspergillosis: Radiological and Microbiological Prole of Patients Presented in an Outpatient Pulmonary Clinic of a Developing Country”.

Background: There is limited data available about allergic bronchopulmonary aspergillosis (ABPA) from Pakistan. The aim of the study was to describe the radiological and microbiological prole of ABPA patients presenting to outpatient pulmonary clinic of a tertiary care hospital, Karachi, Pakistan. Methods A retrospective study was conducted on ABPA patients who presented to pulmonary outpatient clinic, Aga Hospital, Pakistan from January 2017 to December 2018. Data was collected on microbiology and radiology features on predesigned proforma.


Conclusion
Bronchiectasis was frequently observed in our cohort of patients with ABPA. Pseudomonas aeruginosa was found to be common among bacterial pathogens. Isolation of fungus is not uncommon in these patients.

Background
Allergic bronchopulmonary aspergillosis (ABPA) is a hypersensitivity disease caused by an immunological reaction to the Aspergillus species, with Aspergillus fumigatus being the most commonly implicated pathogen [1] . The occurrence of ABPA is commonly seen among asthma and cystic brosis patients with a prevalence of 12.9% and 8.9% respectively [1] .
Previous (2,3) diagnostic criteria has been modi ed by the international society of Human and animal mycology (ISHAM) (4) since APBA may be present without bronchiectasis and all criteria previously proposed are not required to establish the diagnosis of ABPA. Radiological ndings of ABPA vary and include, eeting pulmonary in ltrates, centrilobular nodules characterizing dilated and opaci ed bronchioles, bronchiectasis and mucoid impaction leading to bronchocoele formation (the nger in glove sign) [5] . ABPA is often misdiagnosed as tuberculosis (TB) or pneumonia, due to similar radiological presentations [6,7] , leading to a considerable delay in the provision of appropriate treatment especially in a TB endemic country like Pakistan.
Clinically ABPA is divided into ve stages, stage I with acute are while stage v is advanced brotic disease. It is further classi ed into two categories, according to Patterson et al, namely ABPA-S called seropositive only in the absence of bronchiectasis and ABPA-CB (central bronchiectasis) if bronchiectasis is present. ABPA-CB is the more aggravated form of the disease [8] .
A Japanese study (9) found isolation of Aspergillus spp. in sputum in 59% patients, including A. fumigatus (33%), A. niger (6%), A. terreus (4%), unspeci ed Aspergillus spp. (16%) and Schizophyllum commune was identi ed in 6%. Sputum microscopy and culture was positive in 63% of patients in Indian study (10) but overall data is limited on microbiology for both fungi and bacterial pathogens among patients with ABPA. However, the most frequently isolated organisms reported in patients with non-cystic brosis CF bronchiectasis are Haemophilus in uenzae, Pseudomonas aeruginosa, and Moraxella catarrhalis while Pseudomonas aeruginosa was reported with more advanced bronchiectasis. (11) Another complication of ABPA is development of chronic pulmonary aspergillosis (CPA), which has an estimated burden of 411,000 patients, out of 4,837,000 ABPA patients. CPA is associated with signi cant morbidities, including potentially fatal hemoptysis [12] .
A study estimates that, annually, 1661 cases of ABPA develop in the Pakistani population [13] but data on radiological and microbiological pro le of these patients are very limited. The objective of this study is to determine the microbiological and radiographic pro le of ABPA patients. We believe that an understanding of radiological and microbiological patterns among ABPA patients in a TB endemic country, may considerably improve diagnosis and prognosis in these patients.

Methods
This retrospective study was conducted on patients diagnosed to have ABPA at the adult outpatient pulmonary clinic of Aga Khan University Hospital, (AKUH) Karachi, Pakistan from January 2017 to December 2018. AKUH is the one of the largest tertiary care facilities of Karachi with 650-beds, and largest outpatient department that receives patients from wide socio-economic backgrounds from all over the city and outside from different cities of Pakistan as well. The study was approved by the Ethical Review Committee of Aga Khan University Hospital.
The patients of ABPA were initially identi ed through outpatient pulmonary clinic database. Medical records of all patients who were labelled as ABPA in database were reviewed. Those patients who ful lled the inclusion criteria were then recruited for study. The inclusion criteria for ABPS was adopted from ISHAM 2013 (4) and included: 1) Age ≥ 18 years with underlying asthma, 2) Serum IgE level ≥ 1000 IU/ml, Results A total of 7759 asthmatic patients presented at outpatient pulmonology clinic during study period, 245 (3.15%) were labeled as ABPA and 167 /245(68.16%) patients ful lled the inclusion criteria. The mean age of patients was 41.9 ±13.0 years, and 91 (54.5%) were females. All patients had a long-standing history of asthma with a mean duration of 17.7 ± 13.1 years before they were diagnosed with ABPA. Out of 167 patients, 104 (62.3%) patients had ABPA-CB and 63/167 (37.2%) had ABPA-S. All ve stages of ABPA were observed among the patients and 87 (52.1%) had stage 3.Cough (78.4%), dyspnea (56.9%), wheezing (25.1%), chest pain (25.1%) and hemoptysis (18.6%) were the predominant presenting symptoms at clinic. All patients received systemic and inhaled corticosteroids while 135 (80.8%) patients were on itraconazole ( Table 1).
Sputum culture results were available in 103 (61.7%) patients with bacterial growth in 66 (64.1%) of the cases. The microbiological pro le is presented in (Table 3). The common pathogens identi ed were Pseudomonas aeruginosa (31.1%) and Hemophilus in uenzae (15.5%). One or more fungi were isolated 3) Peripheral eosinophilia ≥ 500 cells/µL in steroid naïve patients, 4) Radiographic pulmonary opacities consistent with ABPA, 5) A positive type I Aspergillus skin test (immediate cutaneous hypersensitivity to Aspergillus antigen), 6) Active TB and other infections were ruled out by by sputum microscopy, culture and Xpert MTB/Rif. We did not use Aspergillus skin test in all of our patients.. We classi ed patients into ve stages, stage 1 patients with acute are, stage 2 patients who underwent remission after treatment and remain asymptomatic, stage 3 patients with recurrent exacerbations, stage 4 patients who were steroids dependent and stage 5 patients with advanced brotic lung disease. Sputum results were obtained during exacerbations and worsening of symptoms. Chest imaging was reviewed by two independent pulmonologists and imaging reports issued by radiologist were also reviewed. Data was collected on predesigned proforma which included demographics, co-morbid, duration of asthma and ABPA, stages of ABPA, presenting clinical symptoms, smoking status, serum IgE and eosinophilia levels and radiological and microbiology ndings. Patients with an incomplete record were excluded from this study.

Discussion
This is the rst study from Pakistan that has highlighted radiographic and the microbiological features associated with ABPA. We have found hyperin ation on chest X-ray and central bronchiectasis on HRCT to be the most common radiological ndings. Microbiologically, we found Pseudomonas aeruginosa to be the most common bacterial infection among ABPA-CB patients. Isolation of fungi, such as Aspergillus fumigatus was also seen in patients with ABPA.
In our study, more patients had ABPA-CB than ABPA-S which is in concordance with an another study conducted in India [14] .On radiological evaluation, bilateral chest x-ray and HRCT observations were noted in 84.4% and 92.1% of the patients respectively. The most common of the HRCT observation was bronchiectasis which was seen in 77.8% of the patients. This observation was found to be consistent with similar studies carried out in China and India, which also reported bronchiectasis as the most common observation [15,16] . Agarwal et al [15] reported 36.6% of patients with a normal HRCT whereas in our study none of the patients had normal HRCT this may be due to delayed diagnosis leading to advanced disease or referral bias of a tertiary care hospital. Additionally, patients with minimal disease did not undergo HRCT chest as usually at our center, HRCT is performed when underlying parenchymal abnormality is suspected. On the other hand, study from China [16] reported mucus plugging as a common radiographic nding, among many other observations, similar to our study. However, hyperin ation was not reported. The study also described the presenting clinical symptoms of cough, dyspnea, sputum, wheezing, chest pain, fever and hemoptysis in the diagnosed patients consistent with our study.
To the best of our knowledge, there have been very limited data available on chest X-ray ndings associated with ABPA, except for an article which states that during early stages, chest X-ray images are usually normal or mimicking asthma [5] . In our study, X-ray showed eeting ltrations and hyperin ation to be present in more than half of the patients while lung consolidations, cavitation, brosis, nodules and cystic changes were also seen.
Laboratory testing showed an increase in mean serum IgE which was comparable with published data [17] . Normally a blood eosinophil level>1000 cells/mm³ is used as a diagnostic criterion [18] . However, 25% of ABPA patients are also reported to have an absolute eosinophil count of <500 cells/mm³ [17] , which may possibly be attributed to the use of drugs prior to diagnosis which may have reduced blood eosinophil levels. In our study, we measured the differential eosinophil count which was found to have a mean value of 11.2 ± 7.7%.
The concept of the ve stages of ABPA was rst proposed by Patterson et al [19] . The brotic stage represents the most advanced stage of the disease with complications and a poor disease prognosis. It has been reported by Greenberger et al [20] that patients frequently present in stage 3 or the exacerbation stage of the disease which is characterized by the presence of chest in ltrates on radiography.
Furthermore, the same study [20] reported that the total serum IgE concentration in stage 3 patients is normally two times as high as the baseline IgE levels, while in the remainder of the stages the IgE levels may be elevated or even normal. More than half (52%) of our patients were also presented in stage 3 of the disease, while 50% of the patients were seen to have eeting in ltrates on radiography. The total mean serum IgE levels in our study were elevated most likely because the majority of the patients were in stage 3 of ABPA.
The fungal isolation in ABPA is known to act as both the infecting organism as well as an allergen, both of which are responsible for eliciting the symptoms of ABPA [21] . These allergens lead to an immune response only in already immunocompromised patients. Furthermore, fungal hyphae are responsible for the release of chemicals which further trigger the secretion of mediators such as certain interleukins and cytokines and in ammatory cells, including eosinophils [17] . Aspergillus species have been previously reported in other studies like Shah et al [22] to be a major cause for ABPA, however data on bacterial spectrum in ABPA patients is not available.

Limitations
A few limitations of our study included (1) the small sample size that was taken into account and the fact that the ABPA patients from only one tertiary care hospital were included which might have created a referral bias in our results. Therefore, we suggest for a large-scale study involving multiple hospitals throughout the country to be conducted. (2) HRCT chest was not done in all patients and was only done where suspicious of bronchiectasis was high so may be percentage of bronchiectasis is under reported.
(3) We did not use Aspergillus skin test or precipitating and IgG antibodies to Aspergillus in all of our patients due to limited availability.

Conclusion
Hyperin ation and eeting in ltrates on chest X-rays and bronchiectasis on HRCT was the most common radiological nding in our ABPA patients. Pseudomonas aeruginosa and A. fumigatus was identi ed as a most common bacterial and fungal pathogens isolated respectively on sputum samples during exacerbation particularly with ABPA-CB. We suggest that both radiographic evaluation and microbiological pro le should be identify early to reduce the extent of lung damage.

Declarations
Ethics approval and consent to participate The study was approved by the ethical review board of Aga Khan University Hospital. It's a retrospective chart review study so no consent was required.

Consent for publication: Not applicable
Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Competing interests: None to declare Funding: None Authors' contributions NI has made contributions in has made contributions to conception and design, interpretation of data, drafting the manuscript and revising it critically for important intellectual content.
MI has made contributions in conception and design, interpretation of data, drafting the manuscript and revising it critically for important intellectual content.
MBAZ has made contributions to conception and design, interpretation of data, drafting the manuscript and revising it critically for important intellectual content.
MA has made contributions in conception and design, interpretation of data, drafting the manuscript and revising it critically for important intellectual content.
SA has made a contribution in statistical analysis and interpretation of data.
KJ has made contributions in conception and design, interpretation of data, drafting the manuscript and revising it critically for important intellectual content.  (100) ABPA-S = allergic bronchopulmonary aspergillosis seropositive, ABPA-CB = allergic bronchopulmonary aspergillosis central bronchiectasis