Clinical Strains of Mycobacterium tuberculosis Representing Different Genotype Families Exhibit Distinct Propensities to Adopt the Differentially Culturable State.

Growing evidence points to the presence of differentially culturable tubercle bacteria (DCTB) in clinical specimens from individuals with active tuberculosis (TB) disease. These bacteria are unable to grow on solid media but can resuscitate in liquid media. Given the epidemiological success of certain clinical genotype families of Mycobacterium tuberculosis, we hypothesize that different strains may have distinct mechanisms of adaptation and tolerance. We used an in vitro carbon starvation model to determine the propensity of strains from lineages 2 and 4 that included the Beijing and LAM families respectively, to generate DCTB. Beijing strains were associated with a greater propensity to produce DCTB compared to LAM strains. Furthermore, LAM strains required culture filtrate (CF) for resuscitation whilst starved Beijing strains were not dependent on CF. Moreover, Beijing strains showed improved resuscitation with cognate CF, suggesting the presence of unique growth stimulatory molecules in this family. Analysis of starved Beijing and LAM strains showed longer cells, which with resuscitation were restored to a shorter length. Cell wall staining with fluorescent D-amino acids identified strain-specific incorporation patterns, indicating that cell surface remodeling during resuscitation was distinct between clinical strains. Collectively, our data demonstrate that M. tuberculosis clinical strains from different genotype lineages have differential propensities to generate DCTB, which may have implications for TB treatment success.

Assessing potential drug-drug interactions between clofazimine and other frequently used agents to treat drug-resistant tuberculosis.

Clofazimine is included in drug regimens to treat rifampicin/drug-resistant tuberculosis (DR-TB), but there is little information about its interaction with other drugs in DR-TB regimens. We evaluated the pharmacokinetic interaction between clofazimine and isoniazid, linezolid, levofloxacin, and cycloserine, dosed as terizidone. Newly diagnosed adults with DR-TB at Klerksdorp/Tshepong Hospital, South Africa, were started on the then-standard treatment with clofazimine temporarily excluded for the initial 2 weeks. Pharmacokinetic sampling was done immediately before and 3 weeks after starting clofazimine, and drug concentrations were determined using validated liquid chromatography-tandem mass spectrometry assays. The data were interpreted with population pharmacokinetics in NONMEM v7.5.1 to explore the impact of clofazimine co-administration and other relevant covariates on the pharmacokinetics of isoniazid, linezolid, levofloxacin, and cycloserine. Clofazimine, isoniazid, linezolid, levofloxacin, and cycloserine data were available for 16, 27, 21, 21, and 6 participants, respectively. The median age and weight for the full cohort were 39 years and 52 kg, respectively. Clofazimine exposures were in the expected range, and its addition to the regimen did not significantly affect the pharmacokinetics of the other drugs except levofloxacin, for which it caused a 15% reduction in clearance. A posteriori power size calculations predicted that our sample sizes had 97%, 90%, and 87% power at P < 0.05 to detect a 30% change in clearance of isoniazid, linezolid, and cycloserine, respectively. Although clofazimine increased the area under the curve of levofloxacin by 19%, this is unlikely to be of great clinical significance, and the lack of interaction with other drugs tested is reassuring.

Characteristics of infections with ancestral, Beta and Delta variants of SARS-CoV-2 in the PHIRST-C community cohort study, South Africa, 2020-2021.

BACKGROUND: Data on the characteristics of individuals with mild and asymptomatic infections with different SARS-CoV-2 variants are limited. We therefore compared the characteristics of individuals infected with ancestral, Beta and Delta SARS-CoV-2 variants in South Africa. METHODS: We conducted a prospective cohort study in a rural and an urban site during July 2020-August 2021. Mid-turbinate nasal swabs were collected twice-weekly from household members irrespective of symptoms and tested for SARS-CoV-2 using real-time reverse transcription polymerase chain reaction (rRT-PCR). Differences in demographic and clinical characteristics, shedding and cycle threshold (Ct) value of infection episodes by variant were evaluated using multinomial regression. Overall and age-specific incidence rates of infection were compared by variant. RESULTS: We included 1200 individuals from 222 households and 648 rRT-PCR-confirmed infection episodes (66, 10% ancestral, 260, 40% Beta, 322, 50% Delta). Symptomatic proportion was similar for ancestral (7, 11%), Beta (44, 17%), and Delta (46, 14%) infections (p=0.4). After accounting for previous infection, peak incidence shifted to younger age groups in successive waves (40-59 years ancestral, 19-39 years Beta, 13-18 years Delta). On multivariable analysis, compared to ancestral, Beta infection was more common in individuals aged 5-12 years (vs 19-39)(adjusted odds ratio (aOR) 2.6, 95% confidence interval (CI)1.1-6.6) and PCR cycle threshold (Ct) value <30 (vs >35)(aOR 3.2, 95%CI 1.3-7.9), while Delta was more common in individuals aged <5 (aOR 6.7, 95%CI1.4-31.2) and 5-12 years (aOR 6.6 95%CI2.6-16.7)(vs 19-39) and Ct value <30 (aOR 4.5, 95%CI 1.3-15.5) and 30-35 (aOR 6.0, 95%CI 2.3-15.7)(vs >35). CONCLUSIONS: Consecutive SARS-CoV-2 waves with Beta and Delta variants were associated with a shift to younger individuals. Beta and Delta infections were associated with higher peak viral loads, potentially increasing infectiousness.

Rapid Emergence and Evolution of SARS-CoV-2 Variants in Advanced HIV Infection.

Previous studies have linked the evolution of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) genetic variants to persistent infections in people with immunocompromising conditions1-4, but the evolutionary processes underlying these observations are incompletely understood. Here we used high-throughput, single-genome amplification and sequencing (HT-SGS) to obtain up to ~103 SARS-CoV-2 spike gene sequences in each of 184 respiratory samples from 22 people with HIV (PWH) and 25 people without HIV (PWOH). Twelve of 22 PWH had advanced HIV infection, defined by peripheral blood CD4 T cell counts (i.e., CD4 counts) <200 cells/µL. In PWOH and PWH with CD4 counts ≥200 cells/µL, most single-genome spike sequences in each person matched one haplotype that predominated throughout the infection. By contrast, people with advanced HIV showed elevated intra-host spike diversity with a median of 46 haplotypes per person (IQR 14-114). Higher intra-host spike diversity immediately after COVID-19 symptom onset predicted longer SARS-CoV-2 RNA shedding among PWH, and intra-host spike diversity at this timepoint was significantly higher in people with advanced HIV than in PWOH. Composition of spike sequence populations in people with advanced HIV fluctuated rapidly over time, with founder sequences often replaced by groups of new haplotypes. These population-level changes were associated with a high total burden of intra-host mutations and positive selection at functionally important residues. In several cases, delayed emergence of detectable serum binding to spike was associated with positive selection for presumptive antibody-escape mutations. Taken together, our findings show remarkable intra-host genetic diversity of SARS-CoV-2 in advanced HIV infection and suggest that adaptive intra-host SARS-CoV-2 evolution in this setting may contribute to the emergence of new variants of concern (VOCs).

Incidence and transmission of respiratory syncytial virus in urban and rural South Africa, 2017-2018.

Data on respiratory syncytial virus (RSV) incidence and household transmission are limited. To describe RSV incidence and transmission, we conducted a prospective cohort study in rural and urban communities in South Africa over two seasons during 2017-2018. Nasopharyngeal swabs were collected twice-weekly for 10 months annually and tested for RSV using PCR. We tested 81,430 samples from 1,116 participants in 225 households (follow-up 90%). 32% (359/1116) of individuals had ≥1 RSV infection; 10% (37/359) had repeat infection during the same season, 33% (132/396) of infections were symptomatic, and 2% (9/396) sought medical care. Incidence was 47.2 infections/100 person-years and highest in children <5 years (78.3). Symptoms were commonest in individuals aged <12 and ≥65 years. Individuals 1-12 years accounted for 55% (134/242) of index cases. Household cumulative infection risk was 11%. On multivariable analysis, index cases with ≥2 symptoms and shedding duration >10 days were more likely to transmit; household contacts aged 1-4 years vs. ≥65 years were more likely to acquire infection. Within two South African communities, RSV attack rate was high, and most infections asymptomatic. Young children were more likely to introduce RSV into the home, and to be infected. Future studies should examine whether vaccines targeting children aged <12 years could reduce community transmission.

Bacterial microbiome and host inflammatory gene expression in foreskin tissue.

The penile epithelial microbiome remains underexplored. We sequenced human RNA and a segment of the bacterial 16S rRNA gene from the foreskin tissue of 144 adolescents from South Africa and Uganda collected during penile circumcision after receipt of 1-2 doses of placebo, emtricitabine + tenofovir disoproxil fumarate, or emtricitabine + tenofovir alafenamide to investigate the microbiome of foreskin tissue and its potential changes with antiretroviral use. We identified a large number of anaerobic species, including Corynebacterium acnes, which was detected more frequently in participants from South Africa than Uganda. Bacterial populations did not differ by treatment received, and no differentially abundant taxa were identified between placebo versus active drug recipients. The relative abundance of specific bacterial taxa was negatively correlated with expression of genes downstream of the innate immune response to bacteria and regulation of inflammation. Our results show no difference in the tissue microbiome of the foreskin with short-course antiretroviral use but that bacterial taxa were largely inversely correlated with inflammatory gene expression, consistent with commensal colonization.

Reducing HIV-associated post-hospital mortality through home-based care in South Africa: a randomized controlled trial.

BACKGROUND: Twenty-three percent of people with HIV (PWH) die within 6-months of hospital discharge. We tested the hypothesis whether a series of structured home visits could reduce mortality. METHODS: We designed a disease neutral home visit package with up to 6 home visits starting 1-week post-hospitalization and every 2 weeks thereafter. The home visit team used a structured assessment algorithm to evaluate and triage social and medical needs of the participant and provide nutritional support. We compared all-cause mortality 6-months following discharge for the intervention compared to usual care in a pilot randomized trial conducted in South Africa. To inform potential scale-up we also included and separately analyzed a group of people without HIV (PWOH). RESULTS: We enrolled 125 people with HIV and randomized them 1:1 to the home visit intervention or usual care. Fourteen were late exclusions because of death prior to discharge or delayed discharge leaving 111 for analysis. The median age was 39 years, 31% were men; and 70% had advanced HIV disease. At six months among PWH 4 (7.3%) in the home visit arm and 10 (17.9%) in the usual care arm (p = 0.09) had died. Among the 70 PWOH enrolled overall 6-month mortality was 10.1%. Of those in the home visit arm, 91% received at least one home visit. CONCLUSIONS: We demonstrated feasibility of delivering post-hospital home visits and demonstrated preliminary efficacy among PWH with a substantial, but not statistically significant, effect size (59% reduction in mortality). COVID-19 related challenges resulted in under-enrollment.

Ex vivo challenge models for infectious diseases.

Traditionally, molecular mechanisms of pathogenesis for infectious agents were studied in cell culture or animal models but have limitations on the extent to which the resulting data reflect natural infection in humans. The COVID-19 pandemic has highlighted the urgent need to rapidly develop laboratory models that enable the study of host-pathogen interactions, particularly the relative efficacy of preventive measures. Recently, human and animal ex vivo tissue challenge models have emerged as a promising avenue to study immune responses, screen potential therapies and triage vaccine candidates. This approach offers the opportunity to closely approximate human disease from the perspective of pathology and immune response. It has advantages compared to animal models which are expensive, lengthy and often require containment facilities. Herein, we summarize some recent advances in the development of ex vivo tissue challenge models for COVID-19, HIV-1 and other pathogens. We focus on the contribution of these models to enhancing knowledge of host-pathogen interactions, immune modulation, and their value in testing therapeutic agents. We further highlight the advantages and limitations of using ex vivo challenge models and briefly summarize how the use of organoids provides a useful advancement over current approaches. Collectively, these developments have enormous potential for the study of infectious diseases.

Transient increased risk of influenza infection following RSV infection in South Africa: findings from the PHIRST study, South Africa, 2016-2018.

BACKGROUND: Large-scale prevention of respiratory syncytial virus (RSV) infection may have ecological consequences for co-circulating pathogens, including influenza. We assessed if and for how long RSV infection alters the risk for subsequent influenza infection. METHODS: We analysed a prospective longitudinal cohort study conducted in South Africa between 2016 and 2018. For participating households, nasopharyngeal samples were taken twice weekly, irrespective of symptoms, across three respiratory virus seasons, and real-time polymerase chain reaction (PCR) was used to identify infection with RSV and/or influenza. We fitted an individual-level hidden Markov transmission model in order to estimate RSV and influenza infection rates and their interdependence. RESULTS: Of a total of 122,113 samples collected, 1265 (1.0%) were positive for influenza and 1002 (0.8%) positive for RSV, with 15 (0.01%) samples from 12 individuals positive for both influenza and RSV. We observed a 2.25-fold higher incidence of co-infection than expected if assuming infections were unrelated. We estimated that infection with influenza is 2.13 (95% CI 0.97-4.69) times more likely when already infected with, and for a week following, RSV infection, adjusted for age. This equates to 1.4% of influenza infections that may be attributable to RSV in this population. Due to the local seasonality (RSV season precedes the influenza season), we were unable to estimate changes in RSV infection risk following influenza infection. CONCLUSIONS: We find no evidence to suggest that RSV was associated with a subsequent reduced risk of influenza infection. Instead, we observed an increased risk for influenza infection for a short period after infection. However, the impact on population-level transmission dynamics of this individual-level synergistic effect was not measurable in this setting.

The prevalence of multimorbidity in virally suppressed HIV-positive patients in Limpopo.

Background: Non-communicable diseases (NCDs) are an emerging global public health problem. Objectives: To assess the prevalence of NCDs and their risk factors among adults on antiretroviral therapy (ART). Method: This was a cross-sectional study (July 2019 - January 2020) in Limpopo, South Africa. Patients were enrolled if they were ≥ 40 years, HIV-positive, and virologically suppressed on ART. Data were analysed descriptively, and a binomial regression model was used to identify risk factors for NCDs. Results: The majority of participants (65%; 319/488) were women. Most (83%; 405/488) were aged 40-59 years; 60% (285/472) were overweight or obese. Based on self-report, 22% (107/488) were currently smokers. Almost half (44%) 213/488) reported daily consumption of vegetables and 65% (319/488) exercised regularly and 39% (190/488) reported treatment for another chronic disease. The leading comorbid conditions were hypertension (32%; 158/488) and diabetes mellitus (5%; 24/488). Risk factors for hypertension included age 60 years and older (relative risk [RR]: 1.72; 95% confidence interval [CI]: 1.29-2.30) diabetes (RR: 1.42; 95% CI: 1.08-1.87), overweight (RR: 1.32; 95% CI: 1.03-1.69) and obesity (RR: 1.69; 95% CI: 1.32-2.17). Conclusion: There is a high prevalence, both of risk factors for NCDs and multimorbidity (> 1 chronic disease) in patients who are ≥ 40 years and virologically suppressed on ART.

The performance of tongue swabs for detection of pulmonary tuberculosis.

Introduction: Oral and/or tongue swabs have demonstrated ability to detect Mycobacterium tuberculosis (Mtb) in adults with pulmonary tuberculosis (TB). Swabs provide useful alternative specimens for diagnosis of TB using molecular assays however, the diagnostic pickup by culture requires further improvement and development. Several studies identified the presence of differentially culturable tubercle bacilli (DCTB) populations in a variety of clinical specimens. These organisms do not grow in routine laboratory media and require growth factors in the form of culture filtrate (CF) from logarithmic phase cultures of Mtb H37Rv. Methods: Herein, we compared the diagnostic performance of sputum and tongue swabs using Mycobacterial Growth Indicator Tube (MGIT) assays, Auramine smear, GeneXpert and DCTB assays supplemented with or without CF. Results: From 89 eligible participants, 83 (93%), 66 (74%) and 79 (89%) were sputum positive by MGIT, smear and GeneXpert, respectively. The corresponding tongue swabs displayed a lower sensitivity with 39 (44%), 2 (2.0%) and 18 (20%) participants respectively for the same tests. We aimed to improve the diagnostic yield by utilizing DCTB assays. Sputum samples were associated with a higher positivity rate for CF-augmented DCTB at 82/89 (92%) relative to tongue swabs at 36/89 (40%). Similarly, sputum samples had a higher positivity rate for DCTB populations that were CF-independent at 64/89 (72%) relative to tongue swabs at 26/89 (29%). DCTB positivity increased significantly, relative to MGIT culture, for tongue swabs taken from HIV-positive participants. We next tested whether the use of an alternative smear stain, DMN-Trehalose, would improve diagnostic yield but noted no substantial increase. Discussion: Collectively, our data show that while tongue swabs yield lower bacterial numbers for diagnostic testing, the use of growth supplementation may improve detection of TB particularly in HIV-positive people but this requires further interrogation in larger studies.

Quantitative Chest X-ray Radiomics for Therapy Response Monitoring in Patients with Pulmonary Tuberculosis.

Tuberculosis (TB) remains the second leading cause of death globally from a single infectious agent, and there is a critical need to develop improved imaging biomarkers and aid rapid assessments of responses to therapy. We aimed to utilize radiomics, a rapidly developing image analysis tool, to develop a scoring system for this purpose. A chest X-ray radiomics score (RadScore) was developed by implementing a unique segmentation method, followed by feature extraction and parameter map construction. Signature parameter maps that showed a high correlation to lung pathology were consolidated into four frequency bins to obtain the RadScore. A clinical score (TBscore) and a radiological score (RLscore) were also developed based on existing scoring algorithms. The correlation between the change in the three scores, calculated from serial X-rays taken while patients received TB therapy, was evaluated using Spearman's correlation. Poor correlations were observed between the changes in the TBscore and the RLscore (0.09 (p-value = 0.36)) and the TBscore and the RadScore (0.02 (p-value = 0.86)). The changes in the RLscore and the RadScore had a much stronger correlation of 0.22, which is statistically significant (p-value = 0.02). This shows that the developed RadScore has the potential to be a quantitative monitoring tool for responses to therapy.

Brief Report: Changing Characteristics Among In-Hospital HIV Deaths: An 11-Year Retrospective Review of a Regional Hospital in South Africa.

BACKGROUND: Elevated HIV-associated mortality persists, despite a notable decline with the expansion of antiretroviral therapy (ART). In South Africa, the relative majority of deaths occur in health facilities, providing an opportunity to track decedent characteristics. SETTING: We analyzed data from 14,870 adult patients who died between 2008 and 2018 at Klerksdorp/Tshepong Hospital Complex in South Africa. METHODS: Recorded data included demographics, causes of death, HIV status, ART, and tuberculosis (TB) history. We present summary statistics and results from linear, log-binomial, and multinomial regressions to quantify changes over time. RESULTS: Over the study period, the median age of decedents with HIV in the hospital increased from 39.3 to 43.4 years, and there was a switch to male predominance (46%-54%). Those who died at a younger age (<40 years) remained more likely to be HIV-positive than the older age group, despite the overall proportion of HIV-positivity decreasing over time. The proportion of decedents with HIV ever started on ART increased from 21% to 67%. The proportion of HIV patients dying from TB and AIDS-defining illnesses decreased from 31% to 22%. CONCLUSIONS: We noted a shift in deaths over time to more men and older individuals, whereas the burden of HIV was heaviest on the younger age groups. Advanced HIV disease remained an important cause of mortality. We also observed an increase in less-traditional opportunistic illnesses among those with HIV, including malignancy, cardiovascular disease, and kidney disease. The high proportion of patients on ART who died prematurely requires further research and interventions.

Knowledge, attitudes, practices and intention to get vaccinated against COVID-19: results from a cross-sectional survey in three peri-urban communities in South Africa.

Introduction: South Africa has the largest number of confirmed cases of COVID-19 in Africa. Data to inform public health strategies to mitigate the spread of new variants and severity of disease is needed, including information on knowledge, attitudes and practices (KAP) regarding COVID-19, factors associated with intention to get vaccinated, and viewpoints on reliable sources of data. Methods: we investigated these topics as part of the COVID-19 healthcare utilization and seroprevalence (HUTS) cross-sectional survey in three communities in South Africa: Mitchell´s Plain (Western Cape Province), Pietermaritzburg (KwaZulu-Natal Province) and Klerksdorp (North West Province) during and after the second wave of COVID-19 prior to vaccine availability. Results: primary caregivers from 5799 households participated in the study, 41.1% from Pietermaritzburg, 34.2% from Klerksdorp and 24.7% from Mitchells Plain. Two-thirds and 94.7% of respondents had correct knowledge on the cause and spread of COVID-19, respectively. Knowledge measures were significantly associated with age less than 65 years, the highest level of education and site (Mitchells Plain). Desired preventive behaviors were associated with higher socio-economic status. While 64.7% of people intended to get vaccinated, those over 64 years of age were more likely to intend to vaccinate (aOR: 1.25, 95% CI: 1.06-1.47). Vaccine intention related to protection of self (58.4%) and family (40.0%). The most trusted source of COVID-19 information was television (59.3%) followed by radio (20.0%). Conclusion: these data can be used to design targeted public health campaigns for the current COVID-19 and future epidemics, ensuring that socio-economic constraints and preference for trusted information are considered.

Evaluation of a human mucosal tissue explant model for SARS-CoV-2 replication.

With the onset of COVID-19, the development of ex vivo laboratory models became an urgent priority to study host-pathogen interactions in response to the pandemic. In this study, we aimed to establish an ex vivo mucosal tissue explant challenge model for studying SARS-CoV-2 infection and replication. Nasal or oral tissue samples were collected from eligible participants and explants generated from the tissue were infected with various SARS-CoV-2 strains, including IC19 (lineage B.1.13), Beta (lineage B.1.351) and Delta (lineage B.1.617.2). A qRT-PCR assay used to measure viral replication in the tissue explants over a 15-day period, demonstrated no replication for any viral strains tested. Based on this, the ex vivo challenge protocol was modified by reducing the viral infection time and duration of sampling. Despite these changes, viral infectivity of the nasal and oral mucosa was not improved. Since 67% of the enrolled participants were already vaccinated against SARS-CoV-2, it is possible that neutralizing antibodies in explant tissue may have prevented the establishment of infection. However, we were unable to optimize plaque assays aimed at titrating the virus in supernatants from both infected and uninfected tissue, due to limited volume of culture supernatant available at the various collection time points. Currently, the reasons for the inability of these mucosal tissue samples to support replication of SARS-CoV-2 ex vivo remains unclear and requires further investigation.

A cross-sectional study of attitudes and behaviors of snuff use and cessation among people with and without HIV in South Africa.

Background: Despite a high (48%) prevalence of snuff use among women with HIV in South Africa, little is known of the attitudes and behaviors of use, strategies for cessation, and potential health risks. Methods: In a cross-sectional study, a questionnaire was administered to adults (≥18 years) with (HIV+) and without HIV (HIV-) who self-reported current snuff use to collect information on demographics, snuff use and cessation attempts, preferred strategies for cessation, other substance use, history of respiratory illness, and mental health. Results: 150 (74 HIV+, 76 HIV-) participants were enrolled; 115 (77%) were daily snuff users, 6 (4%) were current smokers, and 17 (11%) former smokers. Top reasons for current snuff use included improving health (n = 48, 32%), reducing stress (n = 26, 16%), and "being a habit" (n = 38, 25%). Participants believed snuff use to have mostly positive (n = 68, 46%) or no (n = 54, 36%) health impacts, and 57 (38%) participants believed snuff cures headaches. 103 (69%) participants reported a previous quit attempt, and 110 (73%) indicated high interest in quitting snuff. Although 105 (70%) participants indicated that advice from a healthcare provider would aid them in quitting snuff, only 30 (20%) reported ever receiving that advice. A majority of participants (n = 141, 94%) suffer from moderate to high levels of perceived stress, and overall few differences were seen by HIV status. Conclusions: Education on negative impacts of snuff, advice to quit from healthcare providers, and nicotine replacement therapy should be considered in the development of a snuff cessation program.

Healthcare utilization during the first two waves of the COVID-19 epidemic in South Africa: A cross-sectional household survey.

Healthcare utilization surveys contextualize facility-based surveillance data for burden estimates. We describe healthcare utilization in the catchment areas for sentinel site healthcare facilities during the first year of the COVID-19 pandemic. We conducted a cross-sectional healthcare utilization survey in households in three communities from three provinces (KwaZulu-Natal, Western Cape and North West). Field workers administered structured questionnaires electronically with the household members reporting influenza-like illness (ILI) in the past 30 days or severe respiratory illness (SRI) since March 2020. Multivariable logistic regression was used to identify factors associated with healthcare utilization among individuals that reported illness. From November 2020 through April 2021, we enrolled 5804 households and 23,003 individuals. Any respiratory illness was reported by 1.6% of individuals; 0.7% reported ILI only, 0.8% reported SRI only, and 0.1% reported both ILI and SRI. Any form of medical care was sought by 40.8% (95% CI 32.9% - 49.6%) and 71.3% (95% CI 63.2% - 78.6%) of individuals with ILI and SRI, respectively. On multivariable analysis, respiratory illness was more likely to be medically attended for individuals at the Pietermaritzburg site (aOR 3.2, 95% CI 1.1-9.5, compared to Klerksdorp), that were underweight (aOR 11.5, 95% CI 1.5-90.2, compared to normal weight), with underlying illness (aOR 3.2, 95%CI 1.2-8.5), that experienced severe illness (aOR 4.8, 95% CI 1.6-14.3) and those with symptom duration of ≥10 days (aOR 7.9, 95% CI 2.1-30.2, compared to <5 days). Less than half of ILI episodes and only 71% of SRI episodes were medically attended during the first two COVID-19 waves in South Africa. Facility-based data may underestimate disease burden during the COVID-19 pandemic.

Investigating linkage to care between hospitals and primary care clinics for people with TB in rural South Africa.

People with tuberculosis (TB) are often lost to follow-up during treatment transition to another facility. These losses may result in substantial morbidity and mortality but are rarely recorded. We conducted a record review on adults diagnosed with TB at 11 hospitals in Limpopo, South Africa, who were subsequently transferred to a local clinic to initiate or continue treatment. We then performed in-depth record reviews at the primary care clinic to which they were referred and called participants who could not be identified as starting treatment. Between August 2017 and April 2018, we reviewed records of 778 individuals diagnosed with TB in-hospital and later referred to local clinics for treatment. Of the 778, 88 (11%) did not link to care, and an additional 43 (5.5%) died. Compared to people without cough, those with cough had higher odds of linking to care (aOR = 2.01, 95% CI: 1.26-3.25, p = 0.005) and were also linked more quickly [adjusted Time Ratio (aTR) = 0.53, 95% CI:0.36-0.79, p<0.001], as were those diagnosed microbiologically (aOR = 1.86, 95% CI: 1.16-3.06, p = 0.012; aTR = 0.58, 95% CI: 0.34-0.98, p = 0.04). People diagnosed with TB in hospitals often disengage following referral to local clinics. Interventions to identify and re-engage people who do not present to local clinics within days of referral might close an important gap in the TB treatment cascade.

Introducing a secondary segmentation to construct a radiomics model for pulmonary tuberculosis cavities.

PURPOSE: Accurate segmentation (separating diseased portions of the lung from normal appearing lung) is a challenge in radiomic studies of non-neoplastic diseases, such as pulmonary tuberculosis (PTB). In this study, we developed a segmentation method, applicable to chest X-rays (CXR), that can eliminate the need for precise disease delineation, and that is effective for constructing radiomic models for automatic PTB cavity classification. METHODS: This retrospective study used a dataset of 266 posteroanterior CXR of patients diagnosed with laboratory confirmed PTB. The lungs were segmented using a U-net-based in-house automatic segmentation model. A secondary segmentation was developed using a sliding window, superimposed on the primary lung segmentation. Pyradiomics was used for feature extraction from every window which increased the dimensionality of the data, but this allowed us to accurately capture the spread of the features across the lung. Two separate measures (standard-deviation and variance) were used to consolidate the features. Pearson's correlation analysis (with a 0.8 cut-off value) was then applied for dimensionality reduction followed by the construction of Random Forest radiomic models. RESULTS: Two almost identical radiomic signatures consisting of 10 texture features each (9 were the same plus 1 other feature) were identified using the two separate consolidation measures. Two well performing random forest models were constructed from these signatures. The standard-deviation model (AUC = 0.9444 (95% CI, 0.8762; 0.9814)) performed marginally better than the variance model (AUC = 0.9288 (95% CI, 0.9046; 0.9843)). CONCLUSION: The introduction of the secondary sliding window segmentation on CXR could eliminate the need for disease delineation in pulmonary radiomic studies, and it could improve the accuracy of CXR reporting currently regaining prominence as a high-volume screening tool as the developed radiomic models correctly classify cavities from normal CXR.