Evaluation of a point-of-care test for the diagnosis of Taenia solium neurocysticercosis in rural southern Tanzania: a diagnostic accuracy study.

BACKGROUND: Neurocysticercosis is a common cause of epilepsy in Taenia solium-endemic areas in sub-Saharan Africa but is often undiagnosed because of an absence of affordable diagnostic tools. This study evaluated the diagnostic accuracy of a T solium cysticercosis antibody-detecting lateral-flow point-of-care assay (TS POC test) for the neuroimaging-based diagnosis of neurocysticercosis. METHODS: Patients with epileptic seizures or severe progressive headache were recruited consecutively from three hospitals in southern Tanzania. All patients were tested with the TS POC test. All patients positive for cysticercosis on the TS POC test and every tenth patient who was negative for cysticercosis received a brain CT examination and underwent reference testing for T solium cysticercosis (ie, rT24H-EITB, LLGP-EITB, and antigen ELISA). The primary outcome of the study was the sensitivity of the TS POC test for the diagnosis of neurocysticercosis. FINDINGS: Of the 601 recruited participants, 102 (17%) tested positive for cysticercosis with the TS POC test. Overall, 48 (62%) of the 77 patients positive for cysticercosis and five (17%) of the 29 patients negative for cysticercosis on the TS POC test had CT-confirmed neurocysticercosis. The TS POC test yielded a sensitivity of 49% (uncertainty interval [UI] 41-58) for neurocysticercosis. Sensitivity was similar to that of the rT24H-EITB (44%, UI 37-51) and the antigen ELISA (50%, 43-56). For the subset of neurocysticercosis cases with at least one active (ie, vesicular) lesion, sensitivity was above 98% for the TS POC test, the rT24H-ETIB, and the antigen ELISA. INTERPRETATION: The TS POC test showed promising results for the diagnosis of neurocysticercosis in patients with vesicular lesions, which need to be confirmed in a larger study. This test could be considered to support policies on screening patients with suspected neurocysticercosis in clinical settings, which would allow appropriate referral for neuroimaging and early treatment. FUNDING: German Federal Ministry of Education and Research and the European & Developing Countries Clinical Trials Partnership. TRANSLATION: For the Swahili translation of the abstract see Supplementary Materials section.

Knowledge and misconceptions about epilepsy among people with epilepsy and their caregivers attending mental health clinics: A qualitative study in Taenia solium endemic pig-keeping communities in Tanzania.

OBJECTIVE: Taenia solium (T. solium) neurocysticercosis (NCC) affects the central nervous system and is associated with 30% of acquired epilepsy in some endemic areas. Epilepsy is a stigmatizing disease in many societies and people with epilepsy (PWE) and their families experience discrimination. This study aimed to explore the knowledge, perceptions, and experiences of epilepsy among PWE and their caregivers attending mental health clinics. METHODS: In T. solium endemic areas of Tanzania, PWE and their caregivers attending mental health clinics were identified and their informed consent was sought prior to study participation. In-depth interviews were conducted in Swahili language and analyzed thematically. The coding was undertaken by two independent researchers using NVivo (Version 12, QSR International). RESULTS: Thirty-eight participants were interviewed. Three themes were identified during the analysis, namely, knowledge about epilepsy; perception of epilepsy; and experience with epilepsy among PWE and their caregivers. Participants commonly defined epilepsy as a "falling disease," perceived to be caused by witchcraft, and were unaware of the association between T. solium and epilepsy. Stigmatization of epilepsy was reported as a problem. Reported treatment patterns after the initial onset of epilepsy varied widely; however, patients usually began treatment with traditional healing methods, and only later opted for biomedical treatment. Patients had generally poor adherence to antiseizure medication, which could be caused by inadequate knowledge or irregular supply. SIGNIFICANCE: Level of knowledge about epilepsy was low, and NCC was not mentioned as a cause of epilepsy among participants. Epilepsy was generally perceived to be the result of witchcraft, evil spirits, or curses. Health education is needed, including an explanation of the model of T. solium transmission and the insistence on hygiene measures. This could reduce the number of new infections with T. solium, improve access to prompt biomedical treatment, and improve the lives of PWE.

Correction: Characteristics of people with epilepsy and Neurocysticercosis in three eastern African countries-A pooled analysis.

[This corrects the article DOI: 10.1371/journal.pntd.0010870.].

Epidemiological, clinical and radiological characteristics of people with neurocysticercosis in Tanzania-A cross-sectional study.

BACKGROUND: Neurocysticercosis (NCC) is common among people with epilepsy in low-resource settings. Prevalence of NCC and radiological characteristics of patients with NCC vary considerably even within small areas but differences have been poorly characterized so far. METHODS: We conducted a cross-sectional study between August 2018 and April 2020 in three district hospitals in southern Tanzania (Ifisi, Tukuyu and Vwawa). Patients with and without epileptic seizures were included in this study. All patients were tested with a novel antibody-detecting point-of-care test for the diagnosis of Taenia solium cysticercosis. All test positives and a subset of test negatives had a further clinical work-up including medical examination and computed tomography of the brain. NCC was defined according to the Del Brutto criteria. We assessed epidemiological, clinical and radiological characteristics of patients with NCC by presence of epileptic seizures and by serology status. RESULTS: In all three district hospitals, more than 30% of all people with epileptic seizures (PWE) had NCC lesions in their brain (38% in Vwawa, 32% in Tukuyu and 31% in Ifisi). Most PWE with NCC had multiple lesions and mostly parenchymal lesions (at least 85%). If patients were serologically positive, they had in the median more lesions than serologically negative patients (15 [interquartile range 8-29] versus 5 [1.8-11]), and only serologically positive patients had active stage lesions. Furthermore, serologically positive PWE had more lesions than serologically positive people without epileptic seizures (10.5 [7-23]), and more often had active lesions. PWE diagnosed with NCC (n = 53) were older, and more commonly had focal onset seizures (68% versus 44%, p = 0.03) and headache episodes (34% versus 14%, p = 0.06), which were also stronger than in PWE without NCC (p = 0.04). CONCLUSION: NCC is common among PWE. A combination of clinical and serological factors could help to establish an algorithm to identify patients potentially suffering from active NCC, who benefit from further clinical investigation including neuroimaging.

Characteristics of people with epilepsy and Neurocysticercosis in three eastern African countries-A pooled analysis.

BACKGROUND: Neurocysticercosis (NCC), a zoonotic disease caused by the pork tapeworm T. solium, represents one of the most common causes of secondary epilepsy but remains often undiagnosed due to lack of awareness and diagnostic facilities. METHODOLOGY: We pooled data from four cross-sectional studies on epilepsy and NCC in eastern Africa. Study sites were in Uganda, Malawi and in Tanzania (Dar es Salaam and Haydom). The study in Uganda and Malawi were community-based, the two studies in Tanzania were hospital-based. The same questionnaire was used for assessment of clinical characteristics of patients with epilepsy. Computed tomography (CT) scans and serological testing were performed in order to diagnose NCC. RESULTS: Overall, 1,179 people with epilepsy were included in our analysis. Of those, 941 PWE underwent CT scanning and were pooled for NCC analysis. Seventy patients were diagnosed with NCC, but NCC prevalence differed considerably between sites ranging from 2.0% (95%CI 0.4% to 3.6%) in Dar es Salaam to 17.5% (95%CI 12.4% to 22.6%) in Haydom. NCC prevalence did not show any association with sex but increased with age and was higher in rural than urban settings. In addition, being a farmer, non-Muslim, eating pork and living with pigs close by was associated with a higher NCC prevalence. PWE with NCC experienced their first epileptic seizure around 3 years later in life compared to PWE without NCC and their epileptic seizures seemed to be better controlled (p<0.001). There was no difference between focal onset seizures and focal signs on neurological examination in both groups (p = 0.49 and p = 0.92, respectively). The rT24H-EITB had a sensitivity for the detection of NCC of 70% (95% confidence interval [CI] 51 to 84%), the LLGP of 76% (95%CI 58 to 89%) and the antigen ELISA of 36% (95% CI 20 to 55%). CONCLUSIONS: NCC is prevalent among PWE in eastern Africa, although it may not be as common as previously stated. Demographic characteristics of PWE with NCC differed from those without NCC, but semiological characteristics and results on neurological examination did not differ compared to PWE without NCC. Interestingly, seizures seemed to be less frequent in PWE with NCC. Being aware of those differences and similarities may help triaging PWE for neuroimaging in order to establish a diagnosis of NCC.

Characteristics of people with epilepsy in three Eastern African countries - a pooled analysis.

BACKGROUND: Epilepsy is one of the most common neurological disorders worldwide. Yet, its treatment gap is large in some areas and especially in sub-Saharan Africa data on clinical, radiological and semiological characteristics, as well as on treatment of persons with epilepsy (PWE) are still scarce. METHODS: We pooled data from four cross-sectional studies on epilepsy in eastern Africa. Two studies from Malawi and Uganda were community-based; two studies in Tanzania (urban Dar es Salaam and rural Haydom) were hospital-based. Clinical characteristics of PWE were assessed by the same questionnaire. Additionally, data on treatment were collected and computed tomography (CT) scans were performed. RESULTS: Overall, 1179 PWE were included in our analysis (581 (49.3%) female, median age 22 years (IQR 15-32 years)). Up to 25% of the patients had focal onset seizures. Those showed a higher rate of remarkable CT scan findings, with especially post-ischaemic and neurocysticercosis-associated lesions, compared to PWE with generalized onset seizures (35.1% vs. 20%). The majority of the patients experienced tonic-clonic seizures (70-85%). Only 67-78% of PWE received anti-seizure medication (ASM) treatment in the community-based studies, mostly monotherapy with phenobarbital, phenytoin or carbamazepine. Yet, underdosage was frequent and a large proportion of PWE received alternative non-ASM treatment consisting of herbal treatment (up to 83%) and/or scarification (up to 20%). CONCLUSIONS: Epilepsy is common in sub-Saharan Africa, often caused by neurocysticercosis or ischaemic strokes. PWE suffer from high seizure rates and subsequent injuries, as well as from socio-economic consequences due to insufficient ASM treatment. This pooled analysis illustrates the need for structural programmes for adequate identification, education, assessment and treatment of PWE in sub-Saharan Africa.

Impact of Isoniazid Preventive Therapy on Tuberculosis incidence among people living with HIV: A secondary data analysis using Inverse Probability Weighting of individuals attending HIV care and treatment clinics in Tanzania.

BACKGROUND: Information on how well Isoniazid Preventive Therapy (IPT) works on reducing TB incidence among people living with HIV (PLHIV) in routine settings using robust statistical methods to establish causality in observational studies is scarce. OBJECTIVES: To evaluate the effectiveness of IPT in routine clinical settings by comparing TB incidence between IPT and non-IPT groups. METHODS: We used data from PLHIV enrolled in 315 HIV care and treatment clinic from January 2012 to December 2016. We used Inverse Probability of Treatment Weighting to adjust for the probability of receiving IPT; balancing the baseline covariates between IPT and non-IPT groups. The effectiveness of IPT on TB incidence was estimated using Cox regression using the weighted sample. RESULTS: Of 171,743 PLHIV enrolled in the clinics over the five years, 10,326 (6.01%) were excluded leaving 161,417 available for the analysis. Of the 24,800 who received IPT, 1.00% developed TB disease whereas of the 136,617 who never received IPT 6,085 (4.98%) developed TB disease. In 278,545.90 person-years of follow up, a total 7,052 new TB cases were diagnosed. Using the weighted sample, the overall TB incidence was 11.57 (95% CI: 11.09-12.07) per 1,000 person-years. The TB incidence among PLHIV who received IPT was 10.49 (95% CI: 9.11-12.15) per 1,000 person-years and 12.00 (95% CI: 11.69-12.33) per 1,000 person-years in those who never received IPT. After adjusting for other covariates there was 52% lower risk of developing TB disease among those who received IPT compared to those who never received IPT: aHR = 0.48 (95% CI: 0.40-0.58, P<0.001). CONCLUSION: IPT reduced TB incidence by 52% in PLHIV attending routine CTC in Tanzania. IPTW adjusted the groups for imbalances in the covariates associated with receiving IPT to achieve comparable groups of IPT and non-IPT. This study has added evidence on the effectiveness of IPT in routine clinical settings and on the use of IPTW to determine impact of interventions in observational studies.

Neurocysticercosis and HIV/AIDS co-infection: A scoping review.

OBJECTIVES: Neurocysticercosis (NCC) and human immunodeficiency virus (HIV) have a high disease burden and are prevalent in overlapping low- and middle-income areas. Yet, treatment guidance for people living with HIV/AIDS (PLWH/A) co-infected with NCC is currently lacking. This study aims to scope the available literature on HIV/AIDS and NCC co-infection, focusing on epidemiology, clinical characteristics, diagnostics and treatment outcomes. METHODS: The scoping literature review methodological framework, and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines were followed. A total of 16,969 records identified through database searching, and 45 additional records from other sources were reduced to 52 included studies after a standardised selection process. RESULTS: Two experimental studies, ten observational studies, 23 case series/case reports and 17 reviews or letters were identified. Observational studies demonstrated similar NCC seroprevalence in PLWH/A and their HIV-negative counterparts. Of 29 PLWH/A and NCC co-infection, 17 (59%) suffered from epileptic seizures, 15 (52%) from headaches and 15 (52%) had focal neurological deficits. Eighteen (62%) had viable vesicular cysts, and six (21%) had calcified cysts. Fifteen (52%) were treated with albendazole, of which 11 (73%) responded well to treatment. Five individuals potentially demonstrated an immune-reconstitution inflammatory syndrome after commencing antiretroviral therapy, although this was in the absence of immunological and neuroimaging confirmation. CONCLUSIONS: There is a paucity of evidence to guide treatment of PLWH/A and NCC co-infection. There is a pressing need for high-quality studies in this patient group to appropriately inform diagnostic and management guidelines for HIV-positive patients with NCC.

Lifetime prevalence of epilepsy in urban Tanzania - A door-to-door random cluster survey.

OBJECTIVES: Epileptic seizures and epilepsy in urban settings of low-income and middle-income countries (LMIC) are largely under-researched, but their prevalence is necessary for good healthcare planning. This study aimed to determine the lifetime prevalence of epileptic seizures and epilepsy in urban Dar es Salaam. METHODS: Nearly 50,000 people in former Kinondoni district, Dar es Salaam, were screened for epileptic seizures using a set of nine questions. Answers to these nine questions were categorized into generalized, focal, and unspecified seizures. Screening positivity rates were adjusted for questionnaire inaccuracy using two scenarios to analyse true epilepsy prevalences. RESULTS: Overall, 1085 (2.23%) people fulfilled the criteria for ever having had at least one type of epileptic seizure. Two-hundred-ninety-six (0.60%) people screened positive for generalized seizures, 986 (2.02%) for focal seizures, and 32 (0.07%) for unspecified seizures. Women more commonly screened positive than men (2.61% versus 1.72%, p < 0.001), particularly for focal seizures (p < 0.001). Adjusting for different degrees of accuracy of the screening questionnaire yielded true lifetime prevalences for epilepsy of any type between 1.59% and 2.41%. We furthermore observed a considerable variation of screening positivity rates between wards in Kinondoni district (p < 0.001). CONCLUSION: The prevalence of epilepsy, based on a questionnaire survey in urban Tanzania, was higher than previously observed, probably due to the screening questionnaire, which contained questions specifically designed to identify focal seizures. Further studies on epileptic seizures/epilepsy are needed for urban settings in LMIC, preferably with an integrated follow-up of positive cases.

Red meat consumption and its association with hypertension and hyperlipidaemia among adult Maasai pastoralists of Ngorongoro Conservation Area, Tanzania.

BACKGROUND: Red meat is an important dietary source of protein and other essential nutrients. Its high intake has been associated with an increased risk of cardiovascular morbidity and mortality, including hypertension (HTN) and hyperlipidaemia (HLP). Despite being physically active, the Maasai at Ngorongoro Conservation Area (NCA) depend heavily on animals' products as their staple food with fewer intakes of vegetables or fruits due to restriction from carrying out agricultural activities within the NCA. This study aimed at determining the prevalence of HTN and HLP and their association with red meat consumption among adult Maasai of NCA. METHODS: A community-based cross-sectional study was conducted in October 2018 using multistage sampling technique. Eight hundred and ninety-four (894) participants enrolled from seven villages in three wards within NCA Data were collected using a modified WHO NCDs-STEPS tool. Anthropometric measurements, blood pressure (BP) measurements, and blood samples for glucose and cholesterol tests were obtained from the study participants. Crude and adjusted prevalence ratio (PR) for factors associated with HTN and HLP were estimated using Ordinal and Bayesian logistic regression models, respectively. RESULTS: The prevalence of HLP was 23.7 percent. The levels were higher among males than were among the females (29.0% vs. 20.1%, p = 0.002). The prevalence of HTN and pre-HTN (elevated BP) were 9.8 and 37.0 percent, respectively. Both HTN and elevated BP were higher among males than were among females (hypertensive [10.9% vs. 9.0%]; elevated BP [44.0% vs. 32.1%], p<0.001). The prevalence of HLP was significantly associated with level II (PR = 1.56, 95%CrI: 1.10-2.09) and level III (PR = 1.64, 95%CrI: 1.08-2.41) of red meat consumption as opposed to level I. CONCLUSION: The prevalence of hyperlipidaemia and elevated BP were high among NCA Maasai. We found a significant association between red meat consumption and hyperlipidaemia. Further follow-up studies are warranted to establish a temporal relationship between red meat consumption and both conditions.

Determinants of isoniazid preventive therapy completion among people living with HIV attending care and treatment clinics from 2013 to 2017 in Dar es Salaam Region, Tanzania. A cross-sectional analytical study.

BACKGROUND: Tuberculosis (TB) disease is a common opportunistic infection among people living with HIV (PLHIV). WHO recommends at least 6 months of isoniazid Preventive Therapy (IPT) to reduce the risk of active TB. It is important to monitor the six-month IPT completion since a suboptimal dose may not protect PLHIV from TB infection. This study determined the six-month IPT completion and factors associated with six-month IPT completion among PLHIV aged 15 years or more in Dar es Salaam region, Tanzania. METHODS: Secondary analysis of routine data from PLHIV attending 58 care and treatment clinics in Dar es Salaam region was used. PLHIV, aged 15 years and above, who screened negative for TB symptoms and initiated IPT from January, 2013 to June, 2017 were recruited. Modified Poisson regression with robust standard errors was used to estimate prevalence ratios (PR) and 95% confidence interval (CI) for factors associated with IPT completion. Multilevel analysis was used to account for health facility random effects in order to estimate adjusted PR (APR) for factors associated with IPT six-month completion. RESULTS: A total of 29,382 PLHIV were initiated IPT, with 21,808 (74%) female. Overall 17,092 (58%) six-month IPT completion, increasing from 42% (773/1857) in year 2013 to 76% (2929/3856) in 2017. Multilevel multivariable model accounting for health facilities as clusters, showed PLHIV who were not on ART had 46% lower IPT completion compared to those were on ART (APR: 0.54: 95%CI: 0.45-0.64). There was 37% lower IPT completion among PLHIV who transferred from another clinic (APR: 0.63: 95% CI (0.54-0.74) compared to those who did not transfer. PLHIV aged 25-34 years had a 6% lower prevalence of IPT completion as compared to those aged 15 to 24 years (APR:0.94 95%CI:0.89-0.98). CONCLUSION: The IPT completion rate in PLHIV increased over time, but there was lower IPT completion in PLHIV who transferred from other clinics, who were aged 25 to 34 years and those not on ART. Interventions to support IPT in these groups are urgently needed.

High prevalence of tuberculosis diagnosed during autopsy examination at Muhimbili National Hospital in Dar es Salaam, Tanzania.

The primary aims of tuberculosis (TB) control programmes is early diagnosis and prompt treatment of infectious cases to limit transmission. Failure to diagnose and adequately treat TB could lead to premature death and unrecognized transmission of Mycobacterium tuberculosis. The proportion of missed TB cases has not been reported in Tanzania. The objective of this study was to quantify the number of cases of TB identified by autopsy. Deceased morbid bodies from Muhimbili National Hospital were involved. Retrieval of admission, diagnostic and other important records used to manage the patient after admission was done. Demographic information, site and type of disease, past medical history, chest x-ray report, clinical diagnosis and cause of death reported upon death certification were recorded. Lung tissues, lymphnodes and blood clots for HIV testing were collected. Biopsy tissues were processed through Ziehl Nielsen staining and examined by microscopy. The study involved 74 deceased individuals where 56 (75.7%) were males. Information for duration of seeking health care before death was available for 41 (55.4%) subjects. Thirty-four (45.9%) cases received diagnosis before death. The main diagnoses were pneumonia 10 (13.5%), heart failure 6 (8.1%), AIDS-related illnesses 6 (6.8%) and malaria 5 (6.8%). The main clinical findings were wasting (51/74 (68.9%)) and abnormal fluid collection in different body cavities, 61 (50.8%). In 24 out of 71 (33.8%) biopsies acid fast bacilli (AFB) were detected. Records of lymphnodes examination were available in 63 cases and 22 of them had AFB. Twenty-two (34.9%) from the paratracheal and hilar lymphnodes were observed to have AFB. HIV was detected by ELISA in 19 (33.3%) out of 57 deceased, and 12 (63.2%) of the HIV positive deceased were co-infected with TB. Out of the 22 cases positive for AFB on tissue-biopsies 12 (54.5%) were HIV positive. There is a high number of TB cases diagnosed after death that could not be detected before they died. There is a need for increased awareness and to include post-mortem data in the annual statistics of TB for precise reporting of the magnitude of the TB burden in the country.

Immunohaematological reference values in human immunodeficiency virus-negative adolescent and adults in rural northern Tanzania.

BACKGROUND: The amount of CD4 T cells is used for monitoring HIV progression and improvement, and to make decisions to start antiretroviral therapy and prophylactic drugs for opportunistic infections. The aim of this study was to determine normal reference values for CD4 T cells, lymphocytes, leucocytes and haemoglobin level in healthy, HIV negative adolescents and adults in rural northern Tanzania. METHODS: A cross sectional study was conducted from September 2006 to March 2007 in rural northern Tanzania. Participants were recruited from voluntary HIV counselling and testing clinics. Patients were counselled for HIV test and those who consented were tested for HIV. Clinical screening was done, and blood samples were collected for CD4 T cell counts and complete blood cell counts. RESULTS: We enrolled 102 participants, forty two (41.2%) males and 60 (58.8%) females. The mean age was 32.6 +/- 95% CI 30.2-35.0. The mean absolute CD4 T cell count was 745.8 +/- 95% CI 695.5-796.3, absolute CD8 T cells 504.6 +/- 95% CI 461.7-547.5, absolute leukocyte count 5.1 +/- 95% CI 4.8-5.4, absolute lymphocyte count 1.8 +/- 95% CI 1.7-1.9, and haemoglobin level 13.2 +/- 95% CI 12.7-13.7. Females had significantly higher mean absolute CD4 T cell count (p = 0.008), mean absolute CD8 T cell count (p = 0.009) and significantly lower mean haemoglobin level than males (p = 0.003) CONCLUSION: Immunohaematological values found in this study were different from standard values for western countries. Females had significantly higher mean CD4 T cell counts and lower mean haemoglobin levels than males. This raises the issue of the appropriateness of the present reference values and guidelines for monitoring HIV/AIDS patients in Tanzania.

Pulmonary tuberculosis among people living with HIV/AIDS attending care and treatment in rural northern Tanzania.

BACKGROUND: Tuberculosis is the commonest opportunistic infection and the number one cause of death in HIV/AIDS patients in developing countries. To address the extent of the tuberculosis HIV coinfection in rural Tanzania we conducted a cross sectional study including HIV/AIDS patients attending care and treatment clinic from September 2006 to March 2007. METHODS: Sputum samples were collected for microscopy, culture and drug susceptibility testing. Chest X-ray was done for those patients who consented. Blood samples were collected for CD4+ T cells count. RESULTS: The prevalence of tuberculosis was 20/233 (8.5%). Twenty (8.5%) sputum samples were culture positive. Eight of the culture positive samples (40%) were smear positive. Fifteen (75%) of these patients neither had clinical symptoms nor chest X-ray findings suggestive of tuberculosis. Nineteen isolates (95%) were susceptible to rifampicin, isoniazid, streptomycin and ethambutol (the first line tuberculosis drugs). One isolate (5%) from HIV/tuberculosis coinfected patients was resistant to isoniazid. No cases of multi- drug resistant tuberculosis were identified. CONCLUSION: We found high prevalence of tuberculosis disease in this setting. Chest radiograph suggestive of tuberculosis and clinical symptoms of fever and cough were uncommon findings in HIV/tuberculosis coinfected patients. Tuberculosis can occur at any stage of CD4+T cells depletion.

Predictors of mortality in HIV-infected patients starting antiretroviral therapy in a rural hospital in Tanzania.

BACKGROUND: Studies of antiretroviral therapy (ART) programs in Africa have shown high initial mortality. Factors contributing to this high mortality are poorly described. The aim of the present study was to assess mortality and to identify predictors of mortality in HIV-infected patients starting ART in a rural hospital in Tanzania. METHODS: This was a cohort study of 320 treatment-naïve adults who started ART between October 2003 and November 2006. Reliable CD4 cell counts were not available, thus ART initiation was based on clinical criteria in accordance with WHO and Tanzanian guidelines. Kaplan-Meier models were used to estimate mortality and Cox proportional hazards models to identify predictors of mortality. RESULTS: Patients were followed for a median of 10.9 months (IQR 2.9-19.5). Overall, 95 patients died, among whom 59 died within 3 months of starting ART. Estimated mortality was 19.2, 29.0 and 40.7% at 3, 12 and 36 months, respectively. Independent predictors of mortality were severe anemia (hemoglobin <8 g/dL; adjusted hazard ratio [AHR] 9.20; 95% CI 2.05-41.3), moderate anemia (hemoglobin 8-9.9 g/dL; AHR 7.50; 95% CI 1.77-31.9), thrombocytopenia (platelet count <150 x 109/L; AHR 2.30; 95% CI 1.33-3.99) and severe malnutrition (body mass index <16 kg/m2; AHR 2.12; 95% CI 1.06-4.24). Estimated one year mortality was 55.2% in patients with severe anemia, compared to 3.7% in patients without anemia (P < 0.001). CONCLUSION: Mortality was found to be high, with the majority of deaths occurring within 3 months of starting ART. Anemia, thrombocytopenia and severe malnutrition were strong independent predictors of mortality. A prognostic model based on hemoglobin level appears to be a useful tool for initial risk assessment in resource-limited settings.