Sabine Hermans

TB as a cause of hospitalization and in-hospital mortality among people living with HIV worldwide: a systematic review and meta-analysis

Despite significant progress in improving access to antiretroviral therapy over the past decade, substantial numbers of people living with HIV (PLHIV) in all regions continue to experience severe illness and require hospitalization. We undertook a global review assessing the proportion of hospitalizations and in‐hospital deaths because of tuberculosis (TB) in PLHIV.

A century of tuberculosis epidemiology in the northern and southern hemisphere: the differential impact of control interventions

Background Cape Town has one of the highest TB burdens of any city in the world. In 1900 the City of Cape Town, New York City and London had high mortality of tuberculosis (TB). Throughout the 20th century contemporaneous public health measures including screening, diagnosis and treatment were implemented in all three settings. Mandatory notification of TB and vital status enabled comparison of disease burden trajectories. Methods TB mortality, notification and case fatality rates were calculated from 1912 to 2012 using annual TB notifications, TB death certifications and population estimates. Notification rates were stratified by age and in Cape Town by HIV status (from 2009 onwards). Results Pre-chemotherapy, TB mortality and notification rates declined steadily in New York and London but remained high in Cape Town. Following introduction of combination chemotherapy, mean annual case fatality dropped from 45–60% to below 10% in all three settings. Mortality and notification rates subsequently declined, although Cape Town notifications did not decline as far as those in New York or London and returned to pre-chemotherapy levels by 1980. The proportional contribution of childhood TB diminished in New York and London but remained high in Cape Town. The advent of the Cape Town HIV-epidemic in the 1990s was associated with a further two-fold increase in incidence. In 2012, notification rates among HIV-negatives remained at pre-chemotherapy levels. Conclusions TB control was achieved in New York and London but failed in Cape Town. The TB disease burden trajectories started diverging before the availability of combination chemotherapy in 1952 and further diverged following the HIV epidemic in 1990. Chemotherapy impacted case fatality but not transmission, evidenced by on-going high childhood TB rates. Currently endemic TB results from high on-going transmission, which has been exacerbated by the HIV epidemic. TB control will require reducing transmission, which is inexorably linked to prevailing socio-economic factors.

Tuberculosis in Cape Town: An age-structured transmission model

BACKGROUND Tuberculosis (TB) is the leading cause of death in South Africa. The burden of disease varies by age, with peaks in TB notification rates in the HIV-negative population at ages 0-5, 20-24, and 45-49 years. There is little variation between age groups in the rates in the HIV-positive population. The drivers of this age pattern remain unknown. METHODS We developed an age-structured simulation model of Mycobacterium tuberculosis (Mtb) transmission in Cape Town, South Africa. We considered five states of TB progression: susceptible, infected (latent TB), active TB, treated TB, and treatment default. Latently infected individuals could be re-infected; a previous Mtb infection slowed progression to active disease. We further considered three states of HIV progression: HIV negative, HIV positive, on antiretroviral therapy. To parameterize the model, we analysed treatment outcomes from the Cape Town electronic TB register, social mixing patterns from a Cape Town community and used literature estimates for other parameters. To investigate the main drivers behind the age patterns, we conducted sensitivity analyses on all parameters related to the age structure. RESULTS The model replicated the age patterns in HIV-negative TB notification rates of Cape Town in 2009. Simulated TB notification rate in HIV-negative patients was 1000/100,000 person-years (pyrs) in children aged <5 years and decreased to 51/100,000 in children 5-15 years. The peak in early adulthood occurred at 25-29 years (463/100,000 pyrs). After a subsequent decline, simulated TB notification rates gradually increased from the age of 30 years. Sensitivity analyses showed that the dip after the early adult peak was due to the protective effect of latent TB and that retreatment TB was mainly responsible for the rise in TB notification rates from the age of 30 years. CONCLUSION The protective effect of a first latent infection on subsequent infections and the faster progression in previously treated patients are the key determinants of the age-structure of TB notification rates in Cape Town.

Temporal trends in TB notification rates during ART scale-up in Cape Town: an ecological analysis

Although antiretroviral therapy (ART) reduces individual tuberculosis (TB) risk by two‐thirds, the population‐level impact remains uncertain. Cape Town reports high TB notification rates associated with endemic HIV. We examined population trends in TB notification rates during a 10‐year period of expanding ART.

Impact of Anti-Retroviral Treatment and Cotrimoxazole Prophylaxis on Helminth Infections in HIV-Infected Patients in Lambaréné, Gabon

Background Foci of the HIV epidemic and helminthic infections largely overlap geographically. Treatment options for helminth infections are limited, and there is a paucity of drug-development research in this area. Limited evidence suggests that antiretroviral therapy (ART) reduces prevalence of helminth infections in HIV-infected individuals. We investigated whether ART exposure and cotrimoxazole preventive therapy (CTX-P) is associated with a reduced prevalence of helminth infections. Methodology and Principal Findings This cross-sectional study was conducted at a primary HIV-clinic in Lambaréné, Gabon. HIV-infected adults who were ART-naïve or exposed to ART for at least 3 months submitted one blood sample and stool and urine samples on 3 consecutive days. Outcome was helminth infection with intestinal helminths, Schistosoma haematobium, Loa loa or Mansonella perstans. Multivariable logistic regression was used to assess associations between ART or CTX-P and helminth infection. In total, 408 patients were enrolled. Helminth infection was common (77/252 [30.5%]). Filarial infections were most prevalent (55/310 [17.7%]), followed by infection with intestinal helminths (35/296 [11.8%]) and S. haematobium (19/323 [5.9%]). Patients on CTX-P had a reduced risk of Loa loa microfilaremia (adjusted odds ratio (aOR) 0.47, 95% CI 0.23-0.97, P = 0.04), also in the subgroup of patients on ART (aOR 0.36, 95% CI 0.13-0.96, P = 0.04). There was no effect of ART exposure on helminth infection prevalence. Conclusions/Significance CTX-P use was associated with a decreased risk of Loa loa infection, suggesting an anthelminthic effect of antifolate drugs. No relation between ART use and helminth infections was established.

An integrated community TB-HIV adherence model provides an alternative to DOT for tuberculosis patients in Cape Town.

SETTING Cape Town, South Africa. OBJECTIVE To evaluate anti-tuberculosis treatment outcomes and rate of antiretroviral therapy (ART) initiation using weekly community-based adherence support compared to daily directly observed therapy (DOT). METHODS This was a retrospective analysis comparing two cohorts treated for tuberculosis (TB) in 70 TB clinics during 6-month periods before and after the introduction of a new adherence model comprising treatment literacy sessions during 2 weeks of DOT, followed by weekly home visits by community care workers to eligible patients managing their own treatment. Odds ratios (ORs) of treatment success and ART initiation were calculated using multivariable random effects logistic regression models. Hazard ratios (HRs) of default and death were calculated using multivariable random effects Cox regression models. RESULTS The pre-intervention cohort comprised 11 896 patients with TB and the post-intervention cohort 11 314. There was no difference in pre- and post-intervention anti-tuberculosis treatment success rates (respectively 82.8% and 82.5%, adjusted OR [aOR] 1.02, 95%CI 0.89-1.17, P = 0.76) nor an increased hazard of death (adjusted HR [aHR] 0.98, 95%CI 0.80-1.21, P = 0.87) or default (aHR 0.97, 95%CI 0.81-1.15, P = 0.69). The ART initiation rate increased from 67% to 74% (aOR 1.43; 95%CI 1.01-1.85, P < 0.01). CONCLUSION Weekly community-based adherence support was a viable alternative to daily DOT, with no deterioration in anti-tuberculosis treatment outcomes and an increase in ART initiation.

Shared locations of TB cases: places of acquisition or transmission of infection?

We read with interest the recent publication by Chamie et al. in your journal [1]. We agree with the authors that it is important to identify the places where TB transmission takes place in high-burden settings, where a high proportion of TB cases are estimated to be due to recent infection [2–4]. Chamie et al. systematically approached all new TB cases reported within a rural Ugandan population and identified all their potential contacts and locations visited in the month prior to starting TB treatment. The TB cases spent the most time in bars, workplaces, shops and the market. These locations coincide with those we identified using a 24 h diary-based survey among healthy volunteers in a high TB burden semiurban setting in Cape Town, South Africa [5], although we found that even more time was spent in other households and public transport. The authors went on to spoligotype all TB isolates. Almost two-thirds were part of six genotypic clusters, among which no cases were linked socially. Geospatial analysis then identified the locations that were visited by other cases within the cluster, which were mainly clinics but also social venues as places of worship, bars and a market. We appreciate the large amount of work that went into this study and the authors’ attempt to identify locations where these TB cases could have met in the month prior to starting TB treatment. However, we fail to understand the authors’ conclusion that their TB disease was likely acquired from these places. In HIV-uninfected persons, the median duration between infection and disease is estimated to be 4–9 months [6]. Therefore, even if this period were shorter in HIV co-infected cases (who comprised 56% of their study sample), the studied cases would have likely acquired their disease much earlier than in the time period studied. Even if one can assume that social contacts are relatively stable in this community, as the authors suggest, it seems unlikely that this would extend to the healthcare facilities that were found to be linked. From the data provided in the study, the majority of these clinics were linked by non-HIV-infected participants, making it unlikely that these participants would have visited these clinics prior to developing TB symptoms. We therefore feel that a more appropriate conclusion would be that these locations are those where the TB cases studied may have transmitted their disease to others. As the authors point out, the few weeks before starting treatment is when TB cases are at their most infectious. Therefore, these locations would nevertheless be important starting points to target prevention and screening efforts.