Richard Kwizera is a mycologist in the research department at the Infectious Diseases Institute, College of Health Sciences, Makerere University. He is a member of the IDI cryptococcus team investigating the outcomes of HIV patients with cryptococcal meningitis. He has a special research interest in diagnostics, epidemiology and management of pulmonary and opportunistic fungal infections. He has experience and skills in fungal disease diagnosis and management as well as participation in clinical trials. He is a member of the Pan African fungal working group and the WHO global mycetoma working group. He is attached to the research team evaluating the incidence of chronic pulmonary aspergillosis in HIV patients with past TB. He holds a masters of science in Medical mycology from the University of Manchester. He is currently a THRIVE-2 PhD fellow investigating the burden of fungal asthma in Africa attached to the African Severe Asthma Program. He has a goal of developing translational research in mycology in Uganda and across the East African region. Supervisors: Prof David Denning, Prof David Boulware, and Prof David Meya
Background Fungal-bacterial cocolonization and coinfections pose an emerging challenge among pati... more Background Fungal-bacterial cocolonization and coinfections pose an emerging challenge among patients suspected of having pulmonary tuberculosis (PTB); however, the underlying pathogenic mechanisms and microbiome interactions are poorly understood. Understanding how environmental microbes, such as fungi and bacteria, coevolve and develop traits to evade host immune responses and resist treatment is critical to controlling opportunistic pulmonary fungal coinfections. In this project, we propose to study the coexistence of fungal and bacterial microbial communities during chronic pulmonary diseases, with a keen interest in underpinning fungal etiological evolution and the predominating interactions that may exist between fungi and bacteria. Objective This is a protocol for a study aimed at investigating the metabolic and molecular ecological evolution of opportunistic pulmonary fungal coinfections through determining and characterizing the burden, etiological profiles, microbial commu...
BACKGROUND Fungal-bacterial co-colonization and co-infections is an emerging challenge among pati... more BACKGROUND Fungal-bacterial co-colonization and co-infections is an emerging challenge among patients with pulmonary tuberculosis (PTB), yet the underlying pathogenic mechanisms and microbiome interactions are poorly understood. Understanding how environmental microbes such as fungi and bacteria co-evolve and develop traits to evade host immune responses, and resist treatment is critical to controlling opportunistic co-infections. OBJECTIVE This is a protocol for a study aimed at determining and characterizing the burden, etiological profiles, microbial communities, and interaction relationships established between fungi and bacteria as implicated in co-infections in patients with presumptive PTB. METHODS This will be a laboratory-based cross-sectional study, with a sample size of 406 participants. From each participant, two sputum (an on-spot and early morning) samples will be collected. We will also collect baseline data including, demographic and clinical history using a patient-...
Background: Research and clinical use of clinical pharmacology laboratories are limited in low- a... more Background: Research and clinical use of clinical pharmacology laboratories are limited in low- and middle-income countries. We describe our experience in building and sustaining laboratory capacity for clinical pharmacology at the Infectious Diseases Institute, Kampala, Uganda.Intervention: Existing laboratory infrastructure was repurposed, and new equipment was acquired. Laboratory personnel were hired and trained to optimise, validate, and develop in-house methods for testing antiretroviral, anti-tuberculosis and other drugs, including 10 high-performance liquid chromatography methods and four mass spectrometry methods. We reviewed all research collaborations and projects for which samples were assayed in the laboratory from January 2006 to November 2020. We assessed laboratory staff mentorship from collaborative relationships and the contribution of research projects towards human resource development, assay development, and equipment and maintenance costs. We further assessed t...
<p>Data presented are the percentage, numerator/denominator, and 95% confidence interval. P... more <p>Data presented are the percentage, numerator/denominator, and 95% confidence interval. PPV = Positive predictive value, NPV = negative predictive value, C- statistic = receiver-operator characteristic (ROC) area under the curve (AUC) for sensitivity and specificity.</p
In sub-Saharan Africa, an estimated 25% of people with HIV present with advanced HIV and are at h... more In sub-Saharan Africa, an estimated 25% of people with HIV present with advanced HIV and are at high risk of opportunistic infections. Whereas histoplasmosis has occasionally been seen in Uganda, the understanding of the local risk of acute infection is limited. We sought to determine the prevalence of Histoplasma antigenuria using an enzyme immunoassay (EIA, clarus Histoplasma GM EIA, IMMY; Norman, OK, USA) in a cohort of outpatients with advanced HIV disease in Kampala, Uganda. Among the persons with positive urine Histoplasma antigen tests, we assessed their clinical presentation and outcomes. The EIA was run on stored urine samples as per the manufacturer’s instructions. Specimens ≥1 EIA units were considered positive. Among the 388 tested urine samples, 4 (1.2%) were positive for Histoplasma antigen. The histoplasmosis prevalence among participants with a CD4 < 100 cells/mcL was 2.5% (4/158). Three of the four participants with a positive Histoplasma antigen test reported sy...
Invasive fungal diseases (IFDs) are of huge concern in resource-limited settings, particularly in... more Invasive fungal diseases (IFDs) are of huge concern in resource-limited settings, particularly in Africa, due to the unavailability of diagnostic armamentarium for IFDs, thus making definitive diagnosis challenging. IFDs have non-specific systemic manifestations overlapping with more frequent illnesses, such as tuberculosis, HIV, and HIV-related opportunistic infections and malignancies. Consequently, IFDs are often undiagnosed or misdiagnosed. We critically reviewed the available literature on IFDs in Africa to provide a better understanding of their epidemiology, disease burden to guide future research and interventions. Cryptococcosis is the most encountered IFD in Africa, accounting for most of the HIV-related deaths in sub-Saharan Africa. Invasive aspergillosis, though somewhat underdiagnosed and/or misdiagnosed as tuberculosis, is increasingly being reported with a similar predilection towards people living with HIV. More cases of histoplasmosis are also being reported with re...
BackgroundFungal lung diseases are global in distribution and require specific tests for diagnosi... more BackgroundFungal lung diseases are global in distribution and require specific tests for diagnosis. We report a survey of diagnostic service provision in Africa.MethodsA written questionnaire was followed by a video conference call with each respondent(s) and external validation. To disseminate the questionnaire, a snowball sample was used.ResultsData were successfully collected from 50 of 51 African countries with populations >1 million. The questionnaire was completed by respondents affiliated with 72 health facilities. Of these 72 respondents, 33 (45.8%) reported data for the whole country while others reported data for a specific region/province within their country. In the public sector, chest X-ray and computed tomography are performed often in 49 countries (98%) and occasionally in 37 countries (74%), and less often in the private sector. Bronchoscopy and spirometry were done often in 28 countries (56%) and occasionally in 18 countries (36%) in the tertiary health faciliti...
Histoplasmosis, tuberculosis and HIV are all highly prevalent in sub-Saharan Africa (SSA). Co-occ... more Histoplasmosis, tuberculosis and HIV are all highly prevalent in sub-Saharan Africa (SSA). Co-occurrence of two or more of these infections has been reported in several populations of patients, especially those with advanced HIV infection where these opportunistic infections contribute to a significant morbidity and mortality. With a high burden of pulmonary tuberculosis (PTB) secondary to HIV in SSA, histoplasmosis is commonly misdiagnosed as smear-negative PTB in HIV patients due to similar clinical and radiological presentations. This is also partly the result of the lack of trained clinical and laboratory personnel to make a definite diagnosis of histoplasmosis. There is a low index of clinical suspicion for histoplasmosis, and cases are mostly discovered accidently and documented through case reports and case series. Similarly, the high cost and lack of fungal diagnostics in most SSA countries makes it difficult to make a diagnosis. There is a need to build local capacity for m...
Data set for an autopsy sub-study nested under a prospective observational cohort study conducted... more Data set for an autopsy sub-study nested under a prospective observational cohort study conducted in Uganda among inpatients on the infectious diseases, pulmonary, and emergency wards at Kiruddu National Referral Hospital from February 2017 to August 2020
Background Cryptococcal meningitis is a leading cause of mortality in advanced HIV disease. A pos... more Background Cryptococcal meningitis is a leading cause of mortality in advanced HIV disease. A positive cerebrospinal fluid cryptococcal antigen (CrAg) test defines cryptococcal meningitis. Herein, we present a patient with serum and cerebrospinal fluid CrAg negative cryptococcal meningitis, despite a positive cerebrospinal fluid India ink examination and quantitative culture. Case Details A 56-year-old HIV-positive Ugandan woman, with an undetectable HIV RNA viral load and CD4+ T-cell count of 766 cells per microlitre presented with signs and symptoms consistent with cryptococcal meningitis. Her serum and cerebrospinal fluid CrAg tests were negative despite having a positive cerebrospinal fluid India ink and quantitative culture. On day 1, she was commenced on intravenous amphotericin B deoxycholate (1mg/kg) for 3 days (considering 10 CFU growth of Cryptococcus spp) in combination with oral flucytosine (100mg/kg) for 7 days and then fluconazole 1200mg once daily for the next 11 days...
Background Fungal-bacterial cocolonization and coinfections pose an emerging challenge among pati... more Background Fungal-bacterial cocolonization and coinfections pose an emerging challenge among patients suspected of having pulmonary tuberculosis (PTB); however, the underlying pathogenic mechanisms and microbiome interactions are poorly understood. Understanding how environmental microbes, such as fungi and bacteria, coevolve and develop traits to evade host immune responses and resist treatment is critical to controlling opportunistic pulmonary fungal coinfections. In this project, we propose to study the coexistence of fungal and bacterial microbial communities during chronic pulmonary diseases, with a keen interest in underpinning fungal etiological evolution and the predominating interactions that may exist between fungi and bacteria. Objective This is a protocol for a study aimed at investigating the metabolic and molecular ecological evolution of opportunistic pulmonary fungal coinfections through determining and characterizing the burden, etiological profiles, microbial commu...
BACKGROUND Fungal-bacterial co-colonization and co-infections is an emerging challenge among pati... more BACKGROUND Fungal-bacterial co-colonization and co-infections is an emerging challenge among patients with pulmonary tuberculosis (PTB), yet the underlying pathogenic mechanisms and microbiome interactions are poorly understood. Understanding how environmental microbes such as fungi and bacteria co-evolve and develop traits to evade host immune responses, and resist treatment is critical to controlling opportunistic co-infections. OBJECTIVE This is a protocol for a study aimed at determining and characterizing the burden, etiological profiles, microbial communities, and interaction relationships established between fungi and bacteria as implicated in co-infections in patients with presumptive PTB. METHODS This will be a laboratory-based cross-sectional study, with a sample size of 406 participants. From each participant, two sputum (an on-spot and early morning) samples will be collected. We will also collect baseline data including, demographic and clinical history using a patient-...
Background: Research and clinical use of clinical pharmacology laboratories are limited in low- a... more Background: Research and clinical use of clinical pharmacology laboratories are limited in low- and middle-income countries. We describe our experience in building and sustaining laboratory capacity for clinical pharmacology at the Infectious Diseases Institute, Kampala, Uganda.Intervention: Existing laboratory infrastructure was repurposed, and new equipment was acquired. Laboratory personnel were hired and trained to optimise, validate, and develop in-house methods for testing antiretroviral, anti-tuberculosis and other drugs, including 10 high-performance liquid chromatography methods and four mass spectrometry methods. We reviewed all research collaborations and projects for which samples were assayed in the laboratory from January 2006 to November 2020. We assessed laboratory staff mentorship from collaborative relationships and the contribution of research projects towards human resource development, assay development, and equipment and maintenance costs. We further assessed t...
<p>Data presented are the percentage, numerator/denominator, and 95% confidence interval. P... more <p>Data presented are the percentage, numerator/denominator, and 95% confidence interval. PPV = Positive predictive value, NPV = negative predictive value, C- statistic = receiver-operator characteristic (ROC) area under the curve (AUC) for sensitivity and specificity.</p
In sub-Saharan Africa, an estimated 25% of people with HIV present with advanced HIV and are at h... more In sub-Saharan Africa, an estimated 25% of people with HIV present with advanced HIV and are at high risk of opportunistic infections. Whereas histoplasmosis has occasionally been seen in Uganda, the understanding of the local risk of acute infection is limited. We sought to determine the prevalence of Histoplasma antigenuria using an enzyme immunoassay (EIA, clarus Histoplasma GM EIA, IMMY; Norman, OK, USA) in a cohort of outpatients with advanced HIV disease in Kampala, Uganda. Among the persons with positive urine Histoplasma antigen tests, we assessed their clinical presentation and outcomes. The EIA was run on stored urine samples as per the manufacturer’s instructions. Specimens ≥1 EIA units were considered positive. Among the 388 tested urine samples, 4 (1.2%) were positive for Histoplasma antigen. The histoplasmosis prevalence among participants with a CD4 < 100 cells/mcL was 2.5% (4/158). Three of the four participants with a positive Histoplasma antigen test reported sy...
Invasive fungal diseases (IFDs) are of huge concern in resource-limited settings, particularly in... more Invasive fungal diseases (IFDs) are of huge concern in resource-limited settings, particularly in Africa, due to the unavailability of diagnostic armamentarium for IFDs, thus making definitive diagnosis challenging. IFDs have non-specific systemic manifestations overlapping with more frequent illnesses, such as tuberculosis, HIV, and HIV-related opportunistic infections and malignancies. Consequently, IFDs are often undiagnosed or misdiagnosed. We critically reviewed the available literature on IFDs in Africa to provide a better understanding of their epidemiology, disease burden to guide future research and interventions. Cryptococcosis is the most encountered IFD in Africa, accounting for most of the HIV-related deaths in sub-Saharan Africa. Invasive aspergillosis, though somewhat underdiagnosed and/or misdiagnosed as tuberculosis, is increasingly being reported with a similar predilection towards people living with HIV. More cases of histoplasmosis are also being reported with re...
BackgroundFungal lung diseases are global in distribution and require specific tests for diagnosi... more BackgroundFungal lung diseases are global in distribution and require specific tests for diagnosis. We report a survey of diagnostic service provision in Africa.MethodsA written questionnaire was followed by a video conference call with each respondent(s) and external validation. To disseminate the questionnaire, a snowball sample was used.ResultsData were successfully collected from 50 of 51 African countries with populations >1 million. The questionnaire was completed by respondents affiliated with 72 health facilities. Of these 72 respondents, 33 (45.8%) reported data for the whole country while others reported data for a specific region/province within their country. In the public sector, chest X-ray and computed tomography are performed often in 49 countries (98%) and occasionally in 37 countries (74%), and less often in the private sector. Bronchoscopy and spirometry were done often in 28 countries (56%) and occasionally in 18 countries (36%) in the tertiary health faciliti...
Histoplasmosis, tuberculosis and HIV are all highly prevalent in sub-Saharan Africa (SSA). Co-occ... more Histoplasmosis, tuberculosis and HIV are all highly prevalent in sub-Saharan Africa (SSA). Co-occurrence of two or more of these infections has been reported in several populations of patients, especially those with advanced HIV infection where these opportunistic infections contribute to a significant morbidity and mortality. With a high burden of pulmonary tuberculosis (PTB) secondary to HIV in SSA, histoplasmosis is commonly misdiagnosed as smear-negative PTB in HIV patients due to similar clinical and radiological presentations. This is also partly the result of the lack of trained clinical and laboratory personnel to make a definite diagnosis of histoplasmosis. There is a low index of clinical suspicion for histoplasmosis, and cases are mostly discovered accidently and documented through case reports and case series. Similarly, the high cost and lack of fungal diagnostics in most SSA countries makes it difficult to make a diagnosis. There is a need to build local capacity for m...
Data set for an autopsy sub-study nested under a prospective observational cohort study conducted... more Data set for an autopsy sub-study nested under a prospective observational cohort study conducted in Uganda among inpatients on the infectious diseases, pulmonary, and emergency wards at Kiruddu National Referral Hospital from February 2017 to August 2020
Background Cryptococcal meningitis is a leading cause of mortality in advanced HIV disease. A pos... more Background Cryptococcal meningitis is a leading cause of mortality in advanced HIV disease. A positive cerebrospinal fluid cryptococcal antigen (CrAg) test defines cryptococcal meningitis. Herein, we present a patient with serum and cerebrospinal fluid CrAg negative cryptococcal meningitis, despite a positive cerebrospinal fluid India ink examination and quantitative culture. Case Details A 56-year-old HIV-positive Ugandan woman, with an undetectable HIV RNA viral load and CD4+ T-cell count of 766 cells per microlitre presented with signs and symptoms consistent with cryptococcal meningitis. Her serum and cerebrospinal fluid CrAg tests were negative despite having a positive cerebrospinal fluid India ink and quantitative culture. On day 1, she was commenced on intravenous amphotericin B deoxycholate (1mg/kg) for 3 days (considering 10 CFU growth of Cryptococcus spp) in combination with oral flucytosine (100mg/kg) for 7 days and then fluconazole 1200mg once daily for the next 11 days...
Background: Individuals with cryptococcal antigenemia are at high risk for developing cryptococca... more Background: Individuals with cryptococcal antigenemia are at high risk for developing cryptococcal meningitis (CM) if left untreated. It is unknown if meningitis precedes the detection of cryptococcal antigen (CrAg) in cerebrospinal fluid (CSF). We present a sub-population of individuals with symptomatic cryptococcal antigenemia, but negative CSF CrAg. Methods: We evaluated serum and CSF CrAg among 1,236 HIV-seropositive individuals presenting with suspected meningitis in Kampala and Mbarara, Uganda between August 2013 and May 2017. Baseline characteristics and clinical outcomes of individuals with symptomatic cryptococcal antigenemia with negative CSF CrAg were compared to individuals with confirmed CM. Results: We found 45% (559/1,236) of individuals screened had CM and 4% (54/1,236) had symptomatic cryptococcal antigenemia with negative CSF CrAg. Of those presenting with negative CSF CrAg, 6% (3/54) subsequently grew Cryptococcus on CSF culture. Of those without evidence of cryptococcal meningitis by CSF CrAg or culture, 12% (6/51) had confirmed tuberculous meningitis (TBM). The median CD4 count was higher for individuals with negative CSF CrAg compared to those with CM (p=0.15). Headache was a less common symptom for individuals with negative CSF CrAg compared to those with CM (82% vs 97%; p<0.01). Individuals with a negative CSF CrAg presented more frequently with normal CSF WBC (p<0.01) and had lower opening pressure [median 118 (95% CI 90-150) vs 265 (180-390)mmH2O; p<0.01]. There were no cases of bacterial meningitis on CSF PCR or culture, and most with symptomatic CrAg antigenemia had no other known meningitis etiology. Among those with recorded in-hospital outcome, mortality was similar between those with symptomatic antigenemia with negative CSF CrAg and those with CM (33% vs 32%; p=0.85). Conclusions: Symptomatic cryptococcal antigenemia with a negative CSF CrAg is a relatively common presentation. It is unclear whether this represents early cryptococcal meningitis. We observed similar high in-hospital mortality in individuals with CrAg antigenemia regardless of CSF CrAg positivity. Further studies to better understand the clinical course and management of this subset of individuals are warranted. Abstract Etiology of Symptomatic CrAg antigenemia
Purpose:
CPA is estimated to affect 3 million people globally. The most common underlying conditi... more Purpose: CPA is estimated to affect 3 million people globally. The most common underlying condition in CPA is treated pulmonary tuberculosis. Our recent cross-sectional survey identified CPA in 8.2% of 400 Ugandan adults with previously treated pulmonary tuberculosis (Union Lung Conference 2015, abstract PC-957-05). CPA diagnosis is often made years after completing tuberculosis treatment, but no prospective surveys have defined the natural history. Co-infection with atypical mycobacteria frequently occurs in CPA. Co-infection with Mycobacterium tuberculosis and Aspergillus has been described in several case reports. Subacute invasive aspergillosis (SAIA) has been documented in 3-11% of autopsies of patients who died of AIDS, in the absence of pulmonary tuberculosis. This raises the possibility that onset of CPA may occur during active pulmonary tuberculosis, especially in patients with AIDS. We aimed to estimate the prevalence of aspergillosis in patients with active pulmonary tuberculosis. Methods: Stored sera were available from 57 adult patients admitted to Mulago Hospital, Kampala between March 2010 and March 2011. All patients had between 2 weeks and 6 months cough and were diagnosed with pulmonary tuberculosis based on culture or GeneXpert PCR testing of sputum and/or broncho-alveolar fluid. We measured Aspergillus-specific IgG in these samples using the Siemens Immulite assay, which has a specificity of 98% and sensitivity of 96% for the diagnosis of chronic pulmonary aspergillosis (Journal of Infection, in press). Results: 46 (81%) patients and 2% controls were HIV positive. Mean CD4 count in patients with HIV was 99 cells/mL (range 2 - 581 cells/mL). 35 (61%) patients had CD4 count <100 cells/mL and 24 (42%) patients had CD4 count <50 cells/mL. Mean Aspergillus-specific IgG levels were 5 mg/L in healthy controls and 11mg/L in pulmonary tuberculosis cases (p – 0.000). Raised levels were found in 2 (2%) of healthy controls and 27 (47%) pulmonary tuberculosis patients (p – 0.000). There was no significant difference in 2-month mortality in tuberculosis patients with normal or raised levels of Aspergillus-specific IgG. Conclusion: These results cannot differentiate Aspergillus colonization from active disease. This is a select group of tuberculosis patients requiring emergency hospital admission and results may not be generalisable to other patients with tuberculosis. However, given the diagnostic accuracy of the Siemens Immulite assay active Aspergillus co-infection is probably present in many of those with positive results. This possibility should be considered in patients who fail to improve or clinically relapse in spite of appropriate tuberculosis therapy. The frequency of raised Aspergillus-specific IgG levels is four times higher in these patients with active pulmonary tuberculosis than the rate of CPA found in patients with treated tuberculosis in our earlier survey. It is possible that many patients develop Aspergillus colonization during active pulmonary tuberculosis, which then resolves as the tuberculosis is cured. Alternatively, our earlier survey may have failed to identify those who died of CPA soon after completing tuberculosis treatment and so underestimated the frequency of CPA complicating pulmonary tuberculosis. Prospective studies are now needed to record the outcome of patients with pulmonary tuberculosis and raised Aspergillus-specific IgG and define the natural history of CPA secondary to pulmonary tuberculosis.
Abstract:
Recent ISHAM guidance includes raised Aspergillus-specific IgG as an alternative to pr... more Abstract: Recent ISHAM guidance includes raised Aspergillus-specific IgG as an alternative to precipitins testing as one of the “other criteria” for the diagnosis of allergic bronchopulmonary aspergillosis (ABPA). Many commercial Aspergillus-specific IgG kits exist, but their diagnostic efficacy for ABPA is poorly described. The studies upon which manufacturers’ recommended diagnostic cut-offs are based are either tiny or unpublished. We aimed to define the optimal diagnostic cut offs for Aspergillus-specific IgG in ABPA and to describe the diagnostic efficacy of different methods in this context. Methods: We acquired stored sera from the following groups; 1 – 80 patients with known ABPA, not complicated by chronic pulmonary aspergillosis (CPA) who were not on antifungal treatment at the time of sampling, 2 – 100 healthy controls, 3 – 100 diseased controls with asthma, but not ABPA, 4 - 242 samples from patients with CPA. Aspergillus-specific IgG was measured by ImmunoCAP, Siemens Immulite (Germany), Serion (Germany), Omega (UK) and Dynamiker (China) assays, plus precipitins by counterimmunoelectrophoresis using Microgen Aspergillus antigens (UK). Results: Receiver operating characteristic curve results are shown in the table. Diagnostic cut-offs with a specificity of ≥95% against severe asthmatic controls were identified. The Immulite cut-off (40 mg/L) has 50% sensitivity. Dynamiker cut-off (500 AU/ml) has 20% sensitivity. Serion cut-off (100 AU/ml) has 27% sensitivity. Genesis cut-off (20 U/ml) has 38% sensitivity. Precipitins testing produced specificity of 97% and sensitivity of 4%. Alternative diagnostic cut-offs with a specificity of 98% against healthy controls were also assessed. The Immulite cut-off (10 mg/L) has 81% sensitivity. ImmunoCAP cut-off (20mg/L) has 77% sensitivity. Dynamiker cut-off (65 AU/ml) has 66% sensitivity. Serion cut-off (35 AU/ml) has 62% sensitivity. Genesis cut-off (15 U/ml) has 46% sensitivity. We considered whether Aspergillus-specific IgG could be used to identify cases of CPA complicating ABPA. ImmunoCAP levels of over 150 mg/L had 95% specificity and 37% sensitivity and Immulite levels of over 125 mg/L had 95% specificity and 65% sensitivity. Positive precipitins had 96% specificity and 59% sensitivity. Other assays performed poorly Conclusion: All the ELISAs performed substantially better than precipitins for the diagnosis of ABPA. The asthmatic controls used in this retrospective study are severe asthmatics attending a tertiary referral clinic. These patients were not specifically screened for aspergillosis and so may include undiagnosed cases of CPA, ABPA or Aspergillus bronchitis resulting raised levels of Aspergillus-specific IgG. Cut-offs defined in relation to this group may not be optimal for use in other asthmatics. An alternative option is to use a cut off with high specificity against healthy controls. Such a method does not allow Aspergillus-specific IgG to be used as a single test to differentiate ABPA from asthma, but Aspergillus-specific IgG has never been used in this manner in ABPA diagnosis. Using healthy controls allows the identification of patients with ‘abnormally’ high levels of Aspergillus-specific IgG, to be used as a single aspect of a composite diagnostic criteria. These cut-offs are also optimal for use in the diagnosis of CPA and other forms of aspergillosis and probably represent the best substitute for the role precipitins has traditionally played in ABPA diagnosis.
Abstract:
Objectives
Chronic pulmonary aspergillosis (CPA) is a fatal condition that is estimate... more Abstract: Objectives Chronic pulmonary aspergillosis (CPA) is a fatal condition that is estimated to effect 2-3 million people worldwide. Measurement of Aspergillus-specific IgG is central to the diagnosis of CPA, alongside clinical and radiological features. In-house assays are in use in some referral centers, but cannot be easily reproduced in other laboratories. Various assays are commercially produced. Such assays might provide consistently reliable results that are reproducible across many areas. Our major aim was to compare the efficacy of five commercial assays for the diagnosis of CPA in an untreated population. We also noted that the diagnostic cut-offs for these assays were largely determined in mixed groups of aspergillosis patients and may not be optimal for the diagnosis of CPA. Hence, we aimed reconsider the appropriateness of existing diagnostic cut-offs for these assays in our patient population. Methods We tested sera from 250 patients with untreated CPA at the UK National Aspergillosis Center. Diagnosis of CPA was taken from clinical notes and required appropriate symptoms, radiological changes and microbiological evidence of infection. 100 sera from adolescent Ugandan blood donors were tested as controls. Intra-assay variability was measured for each assay by repeating a single test 20 times. Aspergillus-specific IgG was measured on all sera by five methods. ImmunoCAP was also performed during routine diagnostic work-up. We performed ROC curve analysis to compare the efficacy of each assay for CPA diagnosis. We suggest new diagnostic cut-offs for each assay and describe sensitivity and specificity with these cut-offs. Results Co-efficient of variation (CoV), Receiver-operator curve area under curve (ROC AUC) and sensitivity and specificity for each test are shown in table 1. ROC curves are shown in figure 1. Conclusions The Siemens Immunolite assay has statistically superior efficacy to all other assays where ROC analysis was possible, for the diagnosis of CPA. It also has superior sensitivity to the ImmunoCAP assay, which is widely used in Europe. Its high sensitivity and specificity mean that it can be used with confidence for the diagnosis of CPA, when clinical and radiological features are also present.
Background: Current diagnostics for Cryptococcal meningitis (CM) rely on India ink, CSF culture, ... more Background: Current diagnostics for Cryptococcal meningitis (CM) rely on India ink, CSF culture, or cryptococcal antigen (CRAG) latex agglutination test. India ink is relatively insensitive. Culture and CRAG require lab infrastructure. Recently, a novel point-of-care lateral flow assay (LFA) has been FDA-approved for diagnosis of C. neoformans in serum samples. The aim of this study was to evaluate the LFA performance in plasma and CSF samples. Methods: ART-naïve subjects with and without cryptococcosis were enrolled in two prospective cohorts in Kampala, Uganda. Initial CM diagnosis was by CRAG, India ink, and quantitative CSF culture. Plasma and CSF were frozen at-80C. Quantitative CRAG titers and LFA testing with titers were conducted on stored CSF, plasma, and serum. Plasma samples were tested in 112 subjects (n=61 with CM, n=51 with no clinical OI). CSF was tested in 161 subjects with suspected meningitis (105 with CM) from 2006-2009 and on 102 (58 with CM) in 2011. LFA results were compared for concordance with CRAG latex agglutination, India ink, and CSF culture. Results: The LFA was highly sensitive and specific with CSF and plasma samples compared to both culture and CRAG. CRAG and LFA assays had high Kappa concordance in plasma and CSF samples, 0.93 and 0.91, respectively. LFA titers were 2.5 – 10-fold higher than CRAG titers in CSF and in plasma, respectively; however high correlation existed between the two assays (Spearman's ρ=0.90 for plasma and 0.93 for CSF; p<0.001 for each). Culture and India ink had much lower concordance (Kappa 0.69 and 0.72, respectively) with LFA results (due to better LFA sensitivity), and titers were less correlated with quantitative culture results (ρ=0.71, p<0.001). Similar low concordance with culture and India ink were also seen with CRAG titers; suggesting the LFA results were not false-positives but more sensitive in detecting cryptococcal infection. Conclusions: The LFA assay has excellent concordance with CRAG latex agglutination and titers in both plasma and CSF samples and is more sensitive than standard India ink and CSF culture. Because of the wide availability of plasma samples for other clinical assessments (e.g. CD4 testing), the excellent agreement with the established CRAG assay, and the ease of use, the cryptococcal LFA holds great promise for rapid diagnosis of CM as well as the potential for cryptococcal screening prior to ART initiation, particularly in resource-limited areas.
Uploads
CPA is estimated to affect 3 million people globally. The most common underlying condition in CPA is treated pulmonary tuberculosis. Our recent cross-sectional survey identified CPA in 8.2% of 400 Ugandan adults with previously treated pulmonary tuberculosis (Union Lung Conference 2015, abstract PC-957-05).
CPA diagnosis is often made years after completing tuberculosis treatment, but no prospective surveys have defined the natural history. Co-infection with atypical mycobacteria frequently occurs in CPA. Co-infection with Mycobacterium tuberculosis and Aspergillus has been described in several case reports. Subacute invasive aspergillosis (SAIA) has been documented in 3-11% of autopsies of patients who died of AIDS, in the absence of pulmonary tuberculosis. This raises the possibility that onset of CPA may occur during active pulmonary tuberculosis, especially in patients with AIDS. We aimed to estimate the prevalence of aspergillosis in patients with active pulmonary tuberculosis.
Methods:
Stored sera were available from 57 adult patients admitted to Mulago Hospital, Kampala between March 2010 and March 2011. All patients had between 2 weeks and 6 months cough and were diagnosed with pulmonary tuberculosis based on culture or GeneXpert PCR testing of sputum and/or broncho-alveolar fluid. We measured Aspergillus-specific IgG in these samples using the Siemens Immulite assay, which has a specificity of 98% and sensitivity of 96% for the diagnosis of chronic pulmonary aspergillosis (Journal of Infection, in press).
Results:
46 (81%) patients and 2% controls were HIV positive. Mean CD4 count in patients with HIV was 99 cells/mL (range 2 - 581 cells/mL). 35 (61%) patients had CD4 count <100 cells/mL and 24 (42%) patients had CD4 count <50 cells/mL.
Mean Aspergillus-specific IgG levels were 5 mg/L in healthy controls and 11mg/L in pulmonary tuberculosis cases (p – 0.000). Raised levels were found in 2 (2%) of healthy controls and 27 (47%) pulmonary tuberculosis patients (p – 0.000). There was no significant difference in 2-month mortality in tuberculosis patients with normal or raised levels of Aspergillus-specific IgG.
Conclusion:
These results cannot differentiate Aspergillus colonization from active disease. This is a select group of tuberculosis patients requiring emergency hospital admission and results may not be generalisable to other patients with tuberculosis. However, given the diagnostic accuracy of the Siemens Immulite assay active Aspergillus co-infection is probably present in many of those with positive results. This possibility should be considered in patients who fail to improve or clinically relapse in spite of appropriate tuberculosis therapy.
The frequency of raised Aspergillus-specific IgG levels is four times higher in these patients with active pulmonary tuberculosis than the rate of CPA found in patients with treated tuberculosis in our earlier survey. It is possible that many patients develop Aspergillus colonization during active pulmonary tuberculosis, which then resolves as the tuberculosis is cured. Alternatively, our earlier survey may have failed to identify those who died of CPA soon after completing tuberculosis treatment and so underestimated the frequency of CPA complicating pulmonary tuberculosis. Prospective studies are now needed to record the outcome of patients with pulmonary tuberculosis and raised Aspergillus-specific IgG and define the natural history of CPA secondary to pulmonary tuberculosis.
Recent ISHAM guidance includes raised Aspergillus-specific IgG as an alternative to precipitins testing as one of the “other criteria” for the diagnosis of allergic bronchopulmonary aspergillosis (ABPA). Many commercial Aspergillus-specific IgG kits exist, but their diagnostic efficacy for ABPA is poorly described. The studies upon which manufacturers’ recommended diagnostic cut-offs are based are either tiny or unpublished. We aimed to define the optimal diagnostic cut offs for Aspergillus-specific IgG in ABPA and to describe the diagnostic efficacy of different methods in this context.
Methods:
We acquired stored sera from the following groups; 1 – 80 patients with known ABPA, not complicated by chronic pulmonary aspergillosis (CPA) who were not on antifungal treatment at the time of sampling, 2 – 100 healthy controls, 3 – 100 diseased controls with asthma, but not ABPA, 4 - 242 samples from patients with CPA.
Aspergillus-specific IgG was measured by ImmunoCAP, Siemens Immulite (Germany), Serion (Germany), Omega (UK) and Dynamiker (China) assays, plus precipitins by counterimmunoelectrophoresis using Microgen Aspergillus antigens (UK).
Results:
Receiver operating characteristic curve results are shown in the table. Diagnostic cut-offs with a specificity of ≥95% against severe asthmatic controls were identified. The Immulite cut-off (40 mg/L) has 50% sensitivity. Dynamiker cut-off (500 AU/ml) has 20% sensitivity. Serion cut-off (100 AU/ml) has 27% sensitivity. Genesis cut-off (20 U/ml) has 38% sensitivity. Precipitins testing produced specificity of 97% and sensitivity of 4%.
Alternative diagnostic cut-offs with a specificity of 98% against healthy controls were also assessed. The Immulite cut-off (10 mg/L) has 81% sensitivity. ImmunoCAP cut-off (20mg/L) has 77% sensitivity. Dynamiker cut-off (65 AU/ml) has 66% sensitivity. Serion cut-off (35 AU/ml) has 62% sensitivity. Genesis cut-off (15 U/ml) has 46% sensitivity.
We considered whether Aspergillus-specific IgG could be used to identify cases of CPA complicating ABPA. ImmunoCAP levels of over 150 mg/L had 95% specificity and 37% sensitivity and Immulite levels of over 125 mg/L had 95% specificity and 65% sensitivity. Positive precipitins had 96% specificity and 59% sensitivity. Other assays performed poorly
Conclusion:
All the ELISAs performed substantially better than precipitins for the diagnosis of ABPA. The asthmatic controls used in this retrospective study are severe asthmatics attending a tertiary referral clinic. These patients were not specifically screened for aspergillosis and so may include undiagnosed cases of CPA, ABPA or Aspergillus bronchitis resulting raised levels of Aspergillus-specific IgG. Cut-offs defined in relation to this group may not be optimal for use in other asthmatics.
An alternative option is to use a cut off with high specificity against healthy controls. Such a method does not allow Aspergillus-specific IgG to be used as a single test to differentiate ABPA from asthma, but Aspergillus-specific IgG has never been used in this manner in ABPA diagnosis. Using healthy controls allows the identification of patients with ‘abnormally’ high levels of Aspergillus-specific IgG, to be used as a single aspect of a composite diagnostic criteria. These cut-offs are also optimal for use in the diagnosis of CPA and other forms of aspergillosis and probably represent the best substitute for the role precipitins has traditionally played in ABPA diagnosis.
Objectives
Chronic pulmonary aspergillosis (CPA) is a fatal condition that is estimated to effect 2-3 million people worldwide. Measurement of
Aspergillus-specific IgG is central to the diagnosis of CPA, alongside clinical and radiological features. In-house assays are in use in some
referral centers, but cannot be easily reproduced in other laboratories. Various assays are commercially produced. Such assays might
provide consistently reliable results that are reproducible across many areas.
Our major aim was to compare the efficacy of five commercial assays for the diagnosis of CPA in an untreated population. We also noted
that the diagnostic cut-offs for these assays were largely determined in mixed groups of aspergillosis patients and may not be optimal for
the diagnosis of CPA. Hence, we aimed reconsider the appropriateness of existing diagnostic cut-offs for these assays in our patient
population.
Methods
We tested sera from 250 patients with untreated CPA at the UK National Aspergillosis Center. Diagnosis of CPA was taken from clinical
notes and required appropriate symptoms, radiological changes and microbiological evidence of infection. 100 sera from adolescent
Ugandan blood donors were tested as controls.
Intra-assay variability was measured for each assay by repeating a single test 20 times. Aspergillus-specific IgG was measured on all sera
by five methods. ImmunoCAP was also performed during routine diagnostic work-up. We performed ROC curve analysis to compare the
efficacy of each assay for CPA diagnosis. We suggest new diagnostic cut-offs for each assay and describe sensitivity and specificity with
these cut-offs.
Results
Co-efficient of variation (CoV), Receiver-operator curve area under curve (ROC AUC) and sensitivity and specificity for each test are shown
in table 1.
ROC curves are shown in figure 1.
Conclusions
The Siemens Immunolite assay has statistically superior efficacy to all other assays where ROC analysis was possible, for the diagnosis of
CPA. It also has superior sensitivity to the ImmunoCAP assay, which is widely used in Europe. Its high sensitivity and specificity mean that it
can be used with confidence for the diagnosis of CPA, when clinical and radiological features are also present.