TB-related deaths among adults in Guinea-Bissau

SETTING: In an urban demographic, high TB burden surveillance site in Guinea-Bissau, most deaths occur at home, and information on cause of death (CoD) is lacking.OBJECTIVE: To examine CoD and the proportion of TB deaths in three groups: among patients examined for TB without a verified diagnosis after diagnostic workup, described as "assumed TB-negative" (aTBneg), among patients with a confirmed diagnosis of TB and in a sample of the background population.DESIGN: Verbal autopsies (VAs) were obtained for registered deaths occurring between 1 January 2010 and 15 June 2016. All deaths among aTBneg and patients with TB, and a sample of deaths in the background population were included.RESULTS: VAs were obtained from 104/112 aTBneg patients, 140/155 patients with TB, and 172/219 from the general population. The leading CoD was TB in respectively 20%, 69% and 9% of the cohorts. HIV/AIDS-related deaths were the most frequent CoD among aTBneg patients (45%) and in the background population (27%), and accounted for 9% of patients with TB.CONCLUSION: TB was shown to be a frequent CoD, not only among patients diagnosed with TB, but also among aTBneg patients and the background population. This indicates a low TB case detection rate.

A self-rated health score predicts severe disease and high mortality in patients with pulmonary TB

BACKGROUND: Health-related quality of life (HRQoL) in patients with pulmonary TB may be predictive of disease outcome; however, HRQoL instruments are often complicated and unsuitable for use in low-resource settings. A single-item self-rated health (SRH) score may represent a simpler alternative which could be used in clinical decision-making.AIMS: To evaluate internal and external validity of SRH and its correlation with TBscore/TBscoreII in a low-resource setting.METHODS: We used data from our ongoing prospective TB cohort study in Guinea-Bissau. Cohen´s d was used to assess internal validity, and receiver-operating characteristics and mortality statistics to assess external validity. Correlation between SRH and TBscore/TBscoreII was estimated using linear regression.RESULTS: SRH showed satisfactory internal validity and ability to discriminate between fatal cases at high and low scores, although not at middling scores. SRH and TBscore/TBscoreII correlated well at each examination but changes in scores did not, which may be due to ceiling/floor effects and a lag between disease severity and HRQoL.CONCLUSION: SRH shows potential as a quick and simple method to identify patients in need of intensified follow-up during treatment provision. More research is needed to assess its generalisability beyond our setting and to develop models for clinical use of SRH.

Hypertension is associated with increased mortality in patients with tuberculosis in Guinea-Bissau

OBJECTIVES

Tuberculosis (TB) is associated with a number of non-communicable co-morbidities, which can complicate treatment and impair outcome. The aim of this study was to assess the impact of hypertension on disease severity, treatment outcome and survival in a cohort of patients with TB.

METHODS

A retrospective cohort study was conducted in Guinea-Bissau. Patients newly diagnosed with TB between November 2003 and June 2016 were included. Hypertension was defined as blood pressure ≥140/90 mmHg. Disease severity was assessed using the Bandim TBscore. Survival was assessed at the end of treatment and 2 years after treatment initiation.

RESULTS

In total, 1544 patients were included in this study. Hypertension was present in 12.8% of patients at inclusion. Patients with hypertension had slightly less severe TB, but were less likely to have a successful treatment outcome and had 64% higher mortality at 2-year follow-up (adjusted hazard ratio 1.64, 95% confidence interval 1.15-2.34). Mortality rates were highest in hypertensive female patients and patients aged ≥ 45 years.

CONCLUSION

Patients with high blood pressure at the start of TB treatment had a higher mortality rate at 2-year follow-up. Mortality rates were highest in hypertensive females and patients aged ≥ 45 years.

Low prevalence of malnourishment among household contacts of patients with tuberculosis in Guinea-Bissau

SETTING

An urban demographic surveillance site in Bissau, the capital of Guinea-Bissau, West Africa.BACKGROUND We hypothesised that if previous malnutrition plays a part in acquiring active tuberculosis (TB) disease, households of TB cases would have a higher prevalence of malnutrition than those of healthy controls.

DESIGN

A cross-sectional study comparing nutritional and socio-economic status of all newly diagnosed patients with TB in 2014 with household contacts (persons residing in the household of TB cases) and random controls. Exclusion criteria were extra-pulmonary TB, age <15 years and pregnancy.

RESULTS

Prevalence of malnutrition was 5% in household contacts and healthy controls, and 51% in patients with TB. Patients with TB had 22% (95%CI 19-25) lower body weight, 22% (95%CI 20-25) lower body mass index and 22% (95%CI 19-24) lower mid-upper arm circumference than healthy controls (P < 0.001); household contacts and healthy controls were comparable for all measures. The socio-economic status of households with TB cases was lower.

CONCLUSIONS

We did not find a higher prevalence of malnourishment in households with TB cases. This finding did not support the hypothesis that malnourishment was an important causative factor for the development of active TB among patients in this study.

Impact of HIV-1, HIV-2, and HIV-1+2 dual infection on the outcome of tuberculosis

BACKGROUND

HIV-1 infection has been shown to impact the outcome of patients with tuberculosis (TB), but data regarding the impact of HIV-2 on TB outcomes are limited. The aim of this study was to assess the impact of HIV types on mortality among TB patients in Guinea-Bissau and to examine the predictive ability of the TBscoreII, a clinical score used to assess disease severity.

METHODS

In a prospective follow-up study, we examined the prevalence of HIV-1, HIV-2, and HIV-1+2 co-infection in TB patients in Guinea-Bissau, and the impact on outcomes at 12 months of follow-up. We included all adult TB patients in an observational TB cohort at the Bandim Health Project (BHP) in Guinea-Bissau between 2003 and 2013 and assessed survival status at 12 months after the start of treatment.

RESULTS

A total 1312 patients were included; 499 (38%) were female (male/female ratio 1.6). Three hundred and seventy-nine patients were HIV-infected: 241 had HIV-1, 93 had HIV-2, and 45 were HIV-1+2 dual infected. The HIV type-associated risk of TB was 6-fold higher for HIV-1, 7-fold higher for HIV-1+2 dual infection, and 2-fold higher for HIV-2 compared with the HIV-uninfected. Of the patients included, 144 (11%) died, 62 (12%) among females and 82 (9%) among males (hazard ratio (HR) 0.91, 95% confidence interval (CI) 0.64-1.30; p=0.596). Compared to male patients, female patients were younger (1 year younger, 95% CI 0.5-2; p=0.04), reported a longer duration of symptoms (14 days longer, 95% CI 4-25; p=0.003), and had a higher TBscoreII (0.5 points more, 95% CI 0.3-0.7; p<0.001). More females than males were HIV-infected (36% vs. 25%; p<0.001) and more females had a body mass index (BMI) <15 kg/m(2) (11% vs. 6%; p<0.001) and a mid upper arm circumference (MUAC) <200 mm (13% vs. 7%; p < 0.001). HIV infection increased the mortality risk, with HIV-1 infection displaying the highest HR (5.0, 95% CI 3.5-7.1), followed by HIV-1+2 (HR 4.2, 95% CI 2.2-7.8) and HIV-2 (HR 2.1, 95% CI 1.2-3.8). A TBscoreII ≥4 was associated with increased mortality (HR 2.2, 95% CI 1.5-3.1). Significantly increased HRs were found for signs of wasting; a BMI <18 kg/m(2) was associated with a HR of 1.8 (95% CI 1.3-2.6) and a MUAC <220 mm with a HR of 3.8 (95% CI 2.7-5.2).

CONCLUSION

The HIV type-associated risk of TB was much higher for HIV-1 patients and higher but less so for HIV-2 patients, compared with the HIV-uninfected. Clinical severity at presentation was also higher for HIV-infected patients, although less so for HIV-2-infected patients, and all HIV-infected patients had a poorer outcome than the uninfected; mortality was 4-5-fold higher for HIV-1 and dually infected patients and two-fold higher for HIV-2-infected patients. These differences between HIV types did not disappear after adjusting for CD4 count.