Introduction
Chronic pulmonary aspergillosis is a progressively destructive lung disease caused by
Aspergillus species, mainly
A.fumigatus. The fungi can infect both immunocompetent and immunocompromised population [
1]. The disease is common among individuals with previous or underlying pulmonary disorders such as tuberculosis, chronic obstructive pulmonary disease, sarcoidosis, atypical mycobacterial infection, emphysema and pneumothorax [
2-
4] and infrequently, among HIV, lung cancer and diabetes mellitus patients [
2,
3].
Worldwide, CPA is estimated to affect about 3 million people, with associated high morbidity and mortality. Left untreated CPA mortality can be up to 80% within 5 years of infection [
4,
5]. Pulmonary tuberculosis infection is recognised as the most important risk factor for the global burden of CPA. A recent global report estimates about 1.2 million patients with CPA as a sequel to PTB annually [
6]. In developing countries with high TB burden, more than half of the cases of CPA are reported as a complication of post TB treatment. Pulmonary tuberculosis causes structural lung damages with residual cavitary lesions often present in 20–40% of patients. The cavitary lesions post TB infections serves as a good medium for infective fungal spore colonization preceding CPA progression [
7,
8]. Unfortunately, most clinical and radiological features of pulmonary fungal infections mimic those of PTB or other pulmonary disorders. This results in misdiagnosis of PTB with patients being enrolled for retreatment without microbiological evidence [
9].
Chronic pulmonary aspergillosis and PTB can co –exist, complicating diagnosis. In Africa and Asia, a co-infection rate of 15.4% among patients with PTB is reported. The two diseases are difficult to distinguish by clinical and radiological features alone[
10,
11].
Sputum cultures are of low sensitivity with culture positivity of less than 80% of samples [
4,
9,
12]. However, cultures of lung fluids of patients with CPA are usually positive for the infective
Aspergillus species and other co-infecting pathogens [
4,
8,
13]. Positive culture results are important for identifying the
Aspergillus species causing infection and for performing drug susceptibility testing. However, fungal diagnostics are often not available in this settings a field that is somewhat ignored. Previous studies indicate misdiagnosis of CPA as PTB smear negative in 35% of patients due to fungal diagnostic challenges in resource limited settings [
9,
12,
14].
Clinically, the Global Action Fund for Fungal Infections (GAFFI) defines CPA as an illness lasting 3 months with all of the following conditions: (1) weight loss, persistent cough, and/or haemoptysis; (2) chest images showing progressive cavitary infiltrates and/or a fungal ball and/or pericavitary fibrosis or infiltrates or pleural thickening and (3) a positive
Aspergillus IgG assay result or other evidence of
Aspergillus infection. This definition has been recommended for use in resource limited settings [
9,
15].
Kenya ranks among the 30 high burden TB/HIV countries. As at 2020, the estimated TB incidence was 259 cases per 100,000 population [
11]. Pulmonary TB is the fifth leading cause of death and a common complication of HIV infections in the country [
16,
17]. While, PTB diagnosis has evolved considerably, there is unlikelihood that a MTB infected patient is misdiagnosed based on clinical, radiological, microscopy and now molecular assays including GeneXpert MTB/RIF®. Unfortunately, patients with worsening clinical symptoms but negative on MTB diagnostic algorithm are considered MTB relapse and retreated [
11,
14,
16]. Approximately, 10% of TB sputum cultures reports fungal contamination [
8,
18]. Unfortunately, fungi growing on TB cultures are usually considered culture contaminants without any consideration as a possible pathogen.
Due to diagnostic challenges of advanced imaging requirement, a positive
Aspergillus IgG assay with clinical symptoms that mimic PTB relapse is a CPA case definition, after exclusion of alternate diagnoses (e.g. pulmonary TB or NTM). Such case is considered an evidence of
Aspergillus infection and warrant fungal investigations and antifungal management in resource limited settings [
9,
14]. Recently, a validated point-of-care lateral flow device,
Aspergillus IgG/IgM LFD (LDBio Diagnostic, Lyon, France) with a sensitivity and specificity of 91.6% and 98.0% respectively, was introduced for the rapid detection of
Aspergillus-specific immunoglobulins [
19]. However, the use of commercial enzyme-linked immunosorbent assay (ELISA) for detection of
Aspergillus IgG serology has been reported to be superior in diagnosis of CPA and is considered the test of choice in clinical practice [
20,
21].
Although the exact prevalence of CPA remains unclear in Kenya, a recent population based estimate put CPA burden following PTB infections at 10,848 cases with an incidence rate of 32 per 100,000 persons. However, this estimate is based on an actuarial approach and need to be treated with caution [
22].
The lack of technical and infrastructural capabilities to diagnose and treat fungal infections and low index for clinical suspicion of CPA among clinicians, remains a key challenge in the diagnosis and management of respiratory fungal diseases. This is likely to result in significant morbidity and mortality among high risk patients in resource constrained settings. The study was aimed at providing serological evidence of CPA among PTB patients as a cause of smear/GeneXpert negative TB and /or TB treatment failures using A. fumigatus specific immunoglobulin G serology in four TB referral clinics in Kenya.
Discussion
Tuberculosis remains a significant cause of death and a major public health problem worldwide. In 2021, a WHO report indicates that 10 million people were infected with TB with 1.5 million deaths. In Africa, TB accounts for about 80% mortality among people living with HIV/AIDS [
11,
26]. Chronic Pulmonary Aspergillosis is a well-established complication of PTB [
5,
8,
9,
26,
27].
The current study evaluated current and post pulmonary tuberculosis patients for CPA as a cause of smear/GeneXpert negative TB and /or TB treatment failure using sputum fungal culture and A. fumigatus specific immunoglobulin G serology in four TB referral clinics in Kenya. The overall seroprevalence of A. fumigatus IgG antibody among patients in our setting was 19.7%. This finding is comparable with findings from similar studies in Nigeria 14.5% [
20], Iran 13.7% [
28] and Japan 16.7% [
29]. In the Iran and Nigeria studies, HIV infected PTB patients were excluded in the analysis due to a markedly lower CPA seroprevalence rates compared to HIV negative patients. The low CPA seroprevalence among HIV positive patients is thought to possibly be due to poor immune response during
Aspergillus infections in this population, particularly in patients with low CD4 counts [
20].
In 2020, Kenya reported overall tuberculosis infections of 140,000 patients, of which 120,400 (86%) were pulmonary tuberculosis cases with 6.3% mortality among PTB [
17]. If we apply the CPA seroprevalence rate obtained in this study, we could anticipate an annual incidence of approximately 22,225 CPA cases among pulmonary tuberculosis patients. This is substationally greater than the population estimate of 10,848 CPA cases [
22]. This finding indicates that CPA is a significant co-morbidity in PTB patients in our setting and warrant fungal investigations.
We analysed specific antibodies against
A. fumigatus, because, it is the most common fungi implicated in chronic pulmonary infections worldwide. In addition, A
spergillus IgG is particularly sensitive and positive in over 90% of patients with CPA [
1,
19,
21,
26]. However, other
Aspergillus species including
A. niger,
A. flavus have been detected in CPA patients. Antibody assays for
A.fumigatus may have low sensitivity for detection of infection with non
fumigatus Aspergillus. According to Oladele and co-workers, ImmunoCAP® automated system (ThermoFisher Scientific, USA) was unable to detect
A.niger and
A.flavus species in CPA confirmed positive sputum cultures [
20]. In our study, a significant proportion, 74.4% of the population were
A.fumigatus IgG negative. In this population, twelve sputum samples obtained were positive for non
fumigatus pathogens, with predominantly
A.niger species, 9.09%. Our findings is comparable to findings from other studies among CPA patients in which
A.niger, was second most isolated fungi from sputum cultures [
18,
20,
28,
30].
The ELISA antibody assay (Bordier Affinity Products SA) used in this study has sensitivity and specificity of 97% and 90.3% respectively. It has superior performance compared to other commercial ELISA
A.fumigatus IgG tests and only comparable to automated fluorescent ELISA assays [
21,
31].
Due to low sensitivity of sputum culture and serology [
9,
21] it is likely that the actual burden of pulmonary aspergillosis among pulmonary tuberculosis patients in our settings is under represented, considering that over 40% of the clinical Aspergillus species are non fumigatus Aspergillus. Aspergillus niger, A. terreus and A. flavus are common human pathogens while A.clavatus and A. candidus and other species are known potent allergens [
32].
Geographical and climatic variation in
Aspergillus IgG seroprevalence and fungal recovery has been reported elsewhere [
28,
33]. In the present study, Nairobi county, a warmer region with annual temperature average of 18.8 °C reported low seroprevalence compared to the other regions in the study with higher average annual temperatures [
34].
In the present study, more female than male patients were seropositive for A.fumigatus IgG antibody
and reported worsening clinical symptoms. There was a significant difference between the two groups, Z = 3.457;
p = 0.00, with females 3 times likely to test positive for
A.fumigatus IgG antibodies than males (OR 3.24; 95% CI 1.664–6.305) (Table
2). In a similar study in Taiwan, being female gender and previous pulmonary TB were found to be independent predictors of
A.fumigatus IgG positivity [
33]. In addition to gender, mouldy habitation and smoking were important demographic risk factors for
A.fumigatus IgG seropositivity,
p = 0.00 and
p = 0.00 respectively. However, age and occupation did not show significant association with
A.fumigatus IgG seroprevalence (Table
2).
The most common clinical features of the patients positive for
Aspergillus infections were productive cough, fever, chest pain and night sweat and less frequently haemoptysis. These symptoms are consistent with previous studies and suggest an acute clinical presentation due to chronic pulmonary aspergillosis and/or coinfection with bacteria or other pathogens [
10,
14,
35,
36]. There was a significant difference in clinical presentations between the gender, Z = 2.239;
p = 0.01, with
A.fumigatus IgG seropositive females 4 times likely to experience difficulty in breathing than males (OR 4.22; 95% CI 1.197–14.896) (Table
3).
About 20% of patients positive for
A.fumigatus specific IgG antibodies presented with haemoptysis (bloody sputum). Haemoptysis is a well know presentation of pulmonary aspergilloma and is associated with high morbidity and mortality [
20,
37,
38]. In a study in Nigeria, aspergilloma associated haemoptysis accounted for over 30% mortality, despite surgical resection and treatment [
39]. Implying that, early diagnosis and treatment is crucial to improve patient outcomes, prevent disease progression and lower morbidity and mortality.
Kenya introduced GeneXpert MTB/RIF® in 2011, as a presumptive diagnostic test for suspected TB infections. However, smear microscopy is still the gold standard used in many peripheral health facilities where GeneXpert® is not available. In both cases, conventional CXR is often the initial imaging method for evaluation of pulmonary infections [
9,
14,
40]. However, conventional CXR is of low sensitivity and specificity [
10]. In the absence of advanced imaging techniques in resource limited settings, a positive
Aspergillus IgG assay with CPA case definition after exclusion of alternative diagnoses (e.g., pulmonary TB or NTM) is an important evidence of
Aspergillus spp. infections in patients misdiagnosed with PTB in Uganda [
9,
14].
Nearly half, 47.8% of patients were being managed for PTB disease despite GeneXpert negative results. This finding is comparable with findings from similar studies in which 35% of CPA patients were misdiagnosed and managed with anti-TB medication based on clinical suspicion without evidence of mycobacterium infection [
12]. In this settings, patients negative on TB diagnostic algorithms and presenting with PTB like symptoms are often considered TB relapse and retreated [
14,
17,
20].
In our study sites, no laboratory was performing fungal culture or
Aspergillus serology for patients with clinical symptoms that mimic PTB relapse following GeneXpert negative results, neither were patients on antifungal treatment. Lack of diagnostic capacity for fungal infections and low index of suspicion for fungal respiratory infections among clinicians is thought to be a challenge to CPA diagnosis and management in resource limited settings [
7,
41].
Mycological investigations of sputum samples confirmed fungi with potential to cause respiratory infections and invasive disease particularly among individuals with immunosuppression and those with respiratory disorders [
7,
26]. Previous study in Kenya by Tonui and co-workers [
30] isolated pathogenic and opportunistic fungi from sputum samples of smear negative PTB retreatment patients attending TB clinics. These studies highlight the significance of pulmonary fungal colonization and infection among bacteriolocally negative patients presenting with clinical symptoms suggestive of pulmonary TB. Microbial colonization of respiratory tract has been associated with poor treatment outcomes among PTB patients [
18,
42].
In the present study, nearly half 47.8% of patients positive for
A.fumigatus IgG, reported exposure to fungal spores through mouldy habitation. These contaminated indoor and outdoor environments [
22,
43] serve as a source of inhalation of infective fungal spores and yeasts and may explain the diversity of fungi recovered from the patient’s sputum.
Management of CPA requires long term antifungal therapy of over 6 months, essentially to prevent haemoptysis and relapse [
7,
44,
45].Therefore, early diagnosis and treatment is important to lower morbidity and hence reduce the apparent high mortality for PTB patients in the country.
The national diagnostic and treatment algorithms for GeneXpert negative patients with PTB symptoms are not clear. To our knowledge, no study has been done to evaluated CPA incidences or seroprevalence among pulmonary TB patients in Kenya.
The study excluded patients with extra pulmonary TB and those on second line anti-TB treatment. The HIV status of the patients were also not done, which could introduce some bias and study limitation.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.