Hepatitis B and C infection at a large public sector hospital clinic: is it a burden?

BACKGROUND: Co-infections have become significant causes of morbidity and mortality in Human Immunodeficiency Virus (HIV) infected people. Due to shared routes of transmission, co-infection of HIV with Hepatitis B (HBV) and/or Hepatitis C (HCV) should be expected. In Zimbabwe, screening for both viruses in HIV infected people prior to treatment is not routinely practised despite the World Health Organisation (WHO) guidelines (2013) prioritising treatment where these co-infections exist. OBJECTIVE: To determine the prevalence of HBV and HCV infection in HIV infected adults at a public sector HIV clinic in Zimbabwe and to determine risk factors associated with these infections. DESIGN AND SETTING: An analytical cross-sectional survey carried out among systematically randomly sampled HIV infected patients coming for treatment between March and July 2012 at Parirenyatwa Hospital Opportunistic Infection Clinic. MATERIALS AND METHODS: Blood samples were tested for hepatitis B surface antigen (HBsAg) and hepatitis C virus antibodies (anti-HCV). Demographic data and exposure to risk factors were collected. RESULTS: 228 antiretroviral therapy (ART) naive adults were enrolled. 7.9% (18/228) were HBsAg positive and 0.9% (2/228) were anti-HCV positive. None of the participants were infected with both viruses. CONCLUSIONS: The prevalence of HBV has not changed during this HIV era and there is no significant HCV infection in this public sector clinic which serves quite a large sector of the population that lives in Harare, Zimbabwe. Based on these results, there is no need for HCV screening but HBV screening prior to ART initiation may be required. -

Glucose tolerance study in low and normal birth weight young adults.

Objective: To determine blood glucose levels by conducting an oral glucose tolerance test in low and normal birth weight young black adults. Design: Acase control study was done. Seventy students in the College of Health Sciences who had neonatal clinic cards as proof of birth weight were recruited into the study. Blood glucose levels were measured before, during and after the oral glucose tolerance test. Setting: Department of Physiology, University of Zimbabwe, College of Health Sciences, Harare, Zimbabwe. Main Outcome Measures and Results: A total of 70 young adult participants, 47(67%) females and 23(33%)males with mean age 20.28±0.19 years were recruited. 30 had Low Birth Weight (LBW, 21 females and 9 males) and 40 had Normal Birth Weight (NBW,26 females and 14 males).LBW individuals had significantly elevated (p<0.05) mean blood glucose levels at 30minutes(9.41±0.91 for LBW and 7.24±0.28 for NBW, p=0.029) and 60 minutes (9.22±0.75 for LBW and 7.57±0.36 for NBW, p=0.035) after the oral glucose tolerance test. Oral glucose tolerance testing detected 1case of type II diabetes (LBW individual), 13cases of impaired glucose tolerance (9 LBW and 4 NBW individuals)and 1 case of impaired fasting glucose (LBW individual).LBW was associated with an odds ratio of 3.1 for impaired glucose tolerance and it was statistically significant, p<0.05 (p=0.027). Conclusion: Low birth weight was associated with glucose intolerance and significantly higher mean blood glucose levels at 30 and 60 minutes after glucose loading in young adults.

Clinical laboratory test prices in Zimbabwe: A case of profiteering?

OBJECTIVE: To compare the prices charged for clinical laboratory tests in Zimbabwean institutions with those of similar institutions abroad. DESIGN: An online analytical cross sectional study was conducted. SETTING: An online survey. SUBJECTS: We did an online survey of clinical laboratories that published prices of the tests offered on their websites. We also extracted price information from documents published by fees regulatory authorities. MAIN OUTCOME MEASURES: Laboratory test prices for independent institutions, Laboratory test prices for State institutions. RESULTS: Overally for all countries, laboratory test prices were lower in state laboratories compared to the independent laboratories. In Zimbabwe, state laboratories generally charged about 50% of the independent laboratory tariff for most tests. However prices from both Zimbabwean institutions were generally much higher than those of the comparison countries (United Kingdom, South Africa, India, United States of America and New Zealand). CONCLUSION: Prices of laboratory tests are indeed higher in Zimbabwean institutions compared to other centres abroad. These higher prices could be attributed to challenges in consumable procurement logistics. We also present measures that could be put in place to reduce the costs and therefore prices.

Circulating cytokines in pulmonary tuberculosis according to HIV status and dietary iron content.

OBJECTIVE: To evaluate human immunodeficiency virus (HIV) serology, dietary iron and serum concentrations of markers of T-helper type (Th) 1 and Th-2 immune pathways in the setting of tuberculosis (TB). METHODS: A total of 49 patients with pulmonary TB in rural Zimbabwe, 32 of whom were HIV-positive, were evaluated at presentation and over 10 weeks of anti-tuberculosis treatment. RESULTS: Interleukin (IL) 12 and neopterin, Th-1 markers, were both elevated at presentation in 92% of the subjects. In contrast, only 23% had elevation of the Th-2 marker, IL-4. Neopterin and IL-6 concentrations decreased over 10 weeks of treatment (P

Iron overload in Africans and African-Americans and a common mutation in the SCL40A1 (ferroportin 1) gene.

The product of the SLC40A1 gene, ferroportin 1, is a main iron export protein. Pathogenic mutations in ferroportin 1 lead to an autosomal dominant hereditary iron overload syndrome characterized by high serum ferritin concentration, normal transferrin saturation, iron accumulation predominantly in macrophages, and marginal anemia. Iron overload occurs in both the African and the African-American populations, but a possible genetic basis has not been established. We analyzed the ferroportin 1 gene in 19 unrelated patients from southern Africa (N = 15) and the United States (N = 4) presenting with primary iron overload. We found a new c. 744 C-->T (Q248H) mutation in the SLC40A1 gene in 4 of these patients (3 Africans and 1 African-American). Among 22 first degree family members, 10 of whom were Q248H heterozygotes, the mutation was associated with a trend to higher serum ferritin to amino aspartate transferase ratios (means of 14.8 versus 4.3 microg/U; P = 0.1) and lower hemoglobin concentrations (means of 11.8 versus 13.2 g/dL; P = 0.1). The ratio corrects serum ferritin concentration for alcohol-induced hepatocellular damage. We also found heterozygosity for the Q248H mutation in 7 of 51 (14%) southern African community control participants selected because they had a serum ferritin concentration below 400 microg/L and in 5 of 100 (5%) anonymous African-Americans, but we did not find the change in 300 Caucasians with normal iron status and 25 Caucasians with non-HFE iron overload. The hemoglobin concentration was significantly lower in the African community controls with the Q248H mutation than in those without it. We conclude that the Q248H mutation is a common polymorphism in the ferroportin 1 gene in African populations that may be associated with mild anemia and a tendency to iron loading.

Lipoprotein(a) concentrations and apolipoprotein(a) isoform distribution in a Zimbabwean population.

OBJECTIVE: To determine serum lipoprotein(a)[Lp(a)] concentrations and apolipoprotein(a)[apo(a)] phenotypes in a Zimbabwean population. DESIGN: Cross sectional study. SETTING: Blood Transfusion Services, Harare, Zimbabwe. SUBJECTS: 84 black and 40 white blood donors. MAIN OUTCOME MEASURES: Lp(a) concentrations and apo(a) phenotypes. RESULTS: The mean and median values for Lp(a) concentrations were 506 and 350 mg/L for the black subjects and 278 and 142 mg/L for the white subjects (p < 0.005). The frequency distributions of Lp(a) concentrations for both populations were skewed to the right. The frequency distribution of apo(a) size, expressed as the number of kringle IV repeats, was determined. Comparison of the frequency distribution plots showed very similar isoform distributions between the two groups. The documented inverse relationship between apo(a) size and Lp(a) concentration was observed in the white population. CONCLUSION: The Lp(a) levels in the black population were two to three fold higher than in the white population whilst no differences in apo(a) phenotype distribution were noted. This suggests that environmental and metabolic factors may be responsible for the elevated Lp(a) levels observed in blacks. Thus different pathological thresholds may have to be established for elevated serum Lp(a) levels to be used as a risk marker for coronary heart disease in black populations.

Clinical and genetic heterogeneity in hereditary haemochromatosis: association between lymphocyte counts and expression of iron overload.

To identify a new marker of expression of disease, independent of HFE genotype in patients with hereditary haemochromatosis (HHC), the total peripheral blood lymphocyte counts were analysed according to iron status in two groups of subjects with HFE mutations. The groups consisted of 38 homozygotes for C282Y, and 107 heterozygotes for the C282Y or compound heterozygotes for C282Y and H63D. For control purposes, total lymphocyte counts and iron status were also examined in 20 index patients with African dietary iron overload, a condition not associated with HFE mutations, and in 144 members of their families and communities. Mean lymphocyte numbers were lower in C282Y homozygous HHC index subjects with cirrhosis and higher iron stores than in those without cirrhosis and with lower iron burdens [(1.65 +/- 0.43) x 10(6)/mL vs. (2.27 +/- 0.49) x 10(6)/mL; p = 0.008]. Similarly, mean lymphocyte counts were significantly lower in C282Y heterozygotes and C282Y/H63D compound heterozygotes with iron overload and increased serum ferritin concentrations compared to those with normal serum ferritin concentrations (p < 0.05). Statistically significant negative correlations were found, in males, between lymphocyte counts and the total body iron stores, either in C282Y homozygous HHC patients (p = 0.031 in a multiple regression model dependent on age) and in C282Y heterozygotes or C282Y/H63D compound heterozygotes with iron overload (p = 0.029 in a simple linear model). In contrast, lymphocyte counts increased with increasing serum ferritin concentrations among the index subjects with African iron overload (r = 0.324, not statistically significant) and among the members of their families and communities (r = 0.170, p = 0.042). These results suggest that a lower peripheral blood lymphocyte count is associated with a greater degree of iron loading in HFE haemochromatosis but not in African iron overload, and they support the notion that the lymphocyte count may serve as a marker of a non-HFE gene that influences the clinical expression of HFE haemochromatosis.

Association of pulmonary tuberculosis with increased dietary iron.

To determine whether increased dietary iron could be a risk factor for active tuberculosis, dietary iron history and human immunodeficiency virus (HIV) status were studied in 98 patients with pulmonary tuberculosis and in 98 control subjects from rural Zimbabwe. Exposure to high levels of dietary iron in the form of traditional beer is associated with increased iron stores in rural Africans. HIV seropositivity was associated with a 17.3-fold increase in the estimated odds of developing active tuberculosis (95% confidence interval [95% CI], 7.4-40.6; P<.001), and increased dietary iron was associated with a 3.5-fold increase (95% CI, 1.4-8.9; P=.009). Among patients treated for tuberculosis, HIV seropositivity was associated with a 3.8-fold increase in the estimated hazard ratio of death (95% CI, 1.0-13.8; P=.046), and increased dietary iron was associated with a 1.3-fold increase (95% CI, 0.4-6.4; P=.2). These findings are consistent with the hypothesis that elevated dietary iron may increase the risk of active pulmonary tuberculosis.

Carbohydrate-deficient transferrin and chronic alcohol ingestion in subjects with transferrin CD-variants.

Carbohydrate-deficient transferrin (CDT) is widely accepted as screening test for excessive alcohol consumption. However, results from subjects with transferrin variants must be interpreted with caution since chromatography-based methods may give false-positive results. Furthermore, due to the co-elution in HPLC or the co-migration in capillary zone electrophoresis (CZE) of the di- and trisialylated C transferrins with the tetrasialylated D peak, exact measurement of CDT is impossible in CD-variants. Therefore, in this study, we tried to offer a different solution, including only the asialo-D, asialo-C, monosialo-D, monosialo-C, disialo-D and trisialo-D transferrins in the CDT calculation and referring to a different cut-off value for CDT in transferrin CD-variants. Comparison of alcohol consumers with teetotalers demonstrated area under the receiver operating characteristic curve of 0.79 and 0.76 for carbohydrate-deficient transferrin, 0.71 and 0.71 for mean corpuscular volume and 0.51 and 0.68 for gamma-glutamyltransferase in 43 subjects with transferrin CD-variants and 225 subjects with CC-phenotypes, respectively. Since false-positive carbohydrate-deficient transferrin results due to a transferrin CD-variant have major social implications, capillary electrophoresis-based or similar methods (HPLC, FPLC) should be preferred in populations carrying a high D-allele frequency.

Transferrin polymorphism influences iron status in blacks.

BACKGROUND: Genetic variants of human transferrin (TF) have been described, but little is known about their functional differences. We studied iron status according to TF phenotype in a healthy Zimbabwean population and in subjects at risk of African iron overload. METHODS: The study population consisted of 483 nondrinkers, 31 drinking spouse pairs, and 5 family pedigrees (n = 88) with index cases of iron overload. TF phenotypes were determined using starch gel electrophoresis. To evaluate iron status, serum iron, total iron-binding capacity (TIBC), ferritin, and soluble TF receptors were measured, and the percentage of saturation and the serum iron:TF ratio were calculated. The binding of the TF variants was studied by equilibrium dialysis. RESULTS: The reference population was characterized by a high TF D allele frequency (0.050) and a complete absence of homozygous TF DD individuals. Similar allele frequencies were observed in subjects at risk of African iron overload. In the reference population, male TF CD heterozygotes had significantly lower (P <0.01) values for serum iron, TIBC, TF saturation, and serum iron:TF ratio than the TF CC homozygotes; in females, only TIBC was significantly different. Overall red blood cell indices did not differ according to TF phenotype. In the population at risk of African iron overload, only serum iron:TF ratio was consistently significantly lower in TF CD phenotypes (P <0.05). After equilibrium dialysis, the amount of iron bound by TF was significantly lower (P <0.01) in TF CD individuals. CONCLUSIONS: The present data demonstrate a functional difference between TF phenotypes in blacks.

Haptoglobin polymorphism and mortality in patients with tuberculosis.

SETTING: A rural Zimbabwean hospital and the surrounding community. OBJECTIVES: To determine whether a particular haptoglobin phenotype is associated with increased susceptibility to clinical pulmonary tuberculosis, and to determine the outcome of treatment for pulmonary tuberculosis according to haptoglobin phenotype. DESIGN: A case-control study, and a prospective cohort study. RESULTS: We studied 98 consecutive patients with sputum-positive pulmonary tuberculosis and 98 sex- and age-matched controls. The haptoglobin (Hp) phenotype distributions did not differ significantly between the tuberculosis patients and controls (P = 0.5). During the 18-month follow-up period after the start of tuberculosis treatment, 6/18 (33%) cases with Hp 2-2 phenotype died compared to 9/47 (19%) with Hp 2-1 and 3/31 (10%) with Hp 1-1. In a logistic regression model, the odds of dying were 6.1-fold greater with Hp 2-2 than with Hp 1-1 (95%CI 1.04-35.1, P = 0.04). CONCLUSION: Our results suggest that there is equal susceptibility to clinical pulmonary tuberculosis disease amongst different haptoglobin phenotypes. Nonetheless, tuberculosis patients with Hp 2-2 phenotype had a higher risk of mortality.

African iron overload.

African iron overload has been recognised in sub-Saharan Africa for seventy years. The condition is distinct from the well-characterised HLA-linked haemochromatosis described in Caucasians. Increased dietary iron intake predisposes to the condition. Recent evidence suggest that African iron overload may be caused by an interaction between increased dietary iron and a genetic defect not associated with the HLA-locus. Iron deposition is prominent both in macrophages and in hepatic parenchymal cells. Iron overload is distinct from alcoholic liver disease, although the excess dietary iron is derived from a traditional beverage that contains alcohol. African iron overload has clinical consequences. It is a cause of hepatic fibrosis and cirrhosis, and associations with diabetes mellitus, peritonitis, scurvy and osteoporosis have been described. African iron overload may be a cause of hepatocellular carcinoma. The disorder is associated with a poor outcome in tuberculosis, an infection that is highly prevalent in sub-Saharan Africa.

Reference range of serum haptoglobin is haptoglobin phenotype-dependent in blacks.

Reference values for serum haptoglobin (Hp), were established in a Black Zimbabwean population. The upper limit (2.15 g/l) is comparable to the one in Caucasians, but the lower limit (0.12 g/l) is much lower than the proposed interim international reference limit (0.3 g/l). Subjects that typed as Hp 0-0 by starch gel electrophoresis technique were retyped using high performance gel permeation chromatography. This resulted in a 32% decrease in the frequency of Hp 0-0, but an increase in Hp 2-2 and Hp 2-1M phenotype frequencies. In the Zimbabwean Blacks, the Hp 0-0 frequency was estimated to be 2.9%. Haptoglobin reference values were found to be Hp phenotype-dependent; highest values were found in Hp 1-1 (median 0.88 g/l; range 0.31-1.69 g/l) and in Hp 2-1 (median 0.90 g/l; range 0.31-2.22 g/l) and lower values (median 0.66 g/l; range 0.13-1.79 g/l) in Hp 2-2 subjects. The Hp 2-1M phenotype was characterized by low reference values (0.18-1.25 g/l) (P<0.05). In three cases of the rare variant Hp Johnson, high Hp concentrations were found (median 1. 57 g/l; range 0.98-1.57 g/l).

A traditional beverage prevents iron deficiency in African women of child bearing age.

OBJECTIVE: To determine if a traditional item in the diet might be useful in preventing iron deficiency in African women of child-bearing age. DESIGN: In a prospective study, the iron status of women who did and did not drink traditional beer high in iron and folic acid, was compared. Iron status was determined by a combination of haemoglobin, serum ferritin and transferrin saturation. SETTING: The study was conducted amongst rural villagers in the Murehwa and Zaka districts of Zimbabwe and in Mpumalanga Province, South Africa. SUBJECTS: 112 women aged between 12 and 50 y from a population of 425 rural people participating in on-going family genetic studies. RESULTS: Women who consumed traditional beer had significantly higher serum ferritin concentrations and transferrin saturations compared to non-drinkers (P = 0.0001 and 0.03 respectively). Iron deficiency anaemia was not present in drinkers but the prevalence in non-drinkers was 13%. Forty seven percent of the non-drinkers and only 14% of the drinkers had evidence of iron deficiency (P = 0.002). Six (21%) of the drinkers and none of the non-drinkers had evidence of iron overload (transferrin saturation > 55% and serum ferritin > 400 ug/l). CONCLUSION: We conclude that the consumption of traditional beer, rich in iron, protects women against iron deficiency. While the use of an alcoholic beverage is not ideal, our findings suggest that indigenous cultural practices might be successfully employed or adapted for promoting iron nutrition.

Urinary iodine concentrations and thyroid function in adult Zimbabweans during a period of transition in iodine status.

BACKGROUND: In 1993 the compulsory iodization of salt was introduced in Zimbabwe, a country that was previously an area of severe iodine deficiency. OBJECTIVE: The objective of this study was to document urinary iodine excretion and biochemical thyroid function in seemingly healthy, community-dwelling adults after the introduction of iodization. DESIGN: A multistage, random sampling method was used in rural and urban settings to identify households from which the senior household member (aged >35 y) was recruited (alternating male and female recruits). Demographic data were collected for each subject and urinary and venous blood samples were taken. Urinary iodine excretion and serum thyroid hormone status (thyrotropin and total thyroxin) were evaluated according to age, sex, and area of residence. RESULTS: A total of 736 adults were recruited (253 men; mean age: 64 y). Urinary iodine concentrations were high [median (first and third quartiles): 4.41 (2.84, 6.78) micromol/L, or 560 (360, 860) microgram/L] and were significantly higher in rural areas than in urban areas [4.73 (3.07, 7.14) micromol/L, or 600 (390, 906) microgram/L, compared with 3.47 (2.05, 4.73) micromol/L, or 440 (260, 600) microgram/L; P < 0.001]. Urinary iodine excretion declined significantly with increasing age (r = -0.29, P < 0.001). Serum thyroid status suggested that the prevalence of biochemical hyperthyroidism in the study was 3%, with 13 of 415 cases in rural and 3 of 149 cases in urban subjects. CONCLUSION: This study reaffirms the need to continuously monitor iodine replacement programs to ensure efficacy.

Iron overload in urban Africans in the 1990s.

BACKGROUND: In a previously described model, heterozygotes for an African iron loading locus develop iron overload only when dietary iron is high, but homozygotes may do so with normal dietary iron. If an iron loading gene is common, then homozygotes with iron overload will be found even in an urban population where traditional beer, the source of iron, is uncommon. AIMS: To determine whether iron overload and the C282Y mutation characteristic of hereditary haemochromatosis are readily identifiable in an urban African population. METHODS: Histological assessment, hepatocellular iron grading, and dry weight non-haem iron concentration were determined in post mortem tissue from liver, spleen, heart, lungs, and skin. DNA of subjects with elevated hepatic iron indexes was analysed for the C282Y mutation. Iron concentrations in other tissues were compared. RESULTS: A moderate increase (>30 micromol/g) in hepatic iron concentrations was found in 31 subjects (23%; 95% confidence interval 15.9 to 30.1%), and they were considerably elevated (>180 micromol/g) in seven subjects (5.2%; 95% confidence interval 1.5 to 8.9%). Appreciably elevated hepatic iron concentrations were associated with heavy iron deposition in both hepatocytes and macrophages, and either portal fibrosis or cirrhosis. All were negative for the C282Y mutation. Very high concentrations were uncommon in subjects dying in hospital. Concentrations of iron in spleen, heart, lung, and skin were significantly higher in subjects with elevated hepatic iron. CONCLUSIONS: Iron overload is readily identified among urban Africans and is associated with hepatic damage and iron loading of several tissues. The condition is unrelated to the genetic mutation found in hereditary haemochromatosis.

Iron and alcohol content of traditional beers in rural Zimbabwe.

OBJECTIVES: To determine the concentrations of iron and alcohol in traditional beer, as well as how these may be related to the brewing process. DESIGN: Cross sectional study. SETTING/SUBJECTS: Rural communities living in four of Zimbabwe's nine provinces. MAIN OUTCOME MEASURES: Ionic iron concentration and alcohol concentration in 94 different types of alcoholic beverages prepared in rural areas, and 18 commercially produced beers. RESULTS: The commonest types of traditional beer were a seven day beverage called 'doro rematanda', a by-product of this seven day beer called 'muchaiwa,' and a one-day beverage called 'chikokiyana'. Methods of preparation were similar in the four provinces. Median (Q1, Q3) ionic iron concentrations were 52 (31 to 75) mg/L for the seven-day beer (n = 51), 24 (18 to 36) mg/L for muchaiwa (n = 30) and 21 (17 to 63) mg/L for chikokiyana (n = 13). In contrast, ionic iron concentrations in 12 samples of commercially prepared clear beers were 0.1 mg/L and in commercial opaque beer were 3.6 mg/L. Mean (SD) alcohol concentration in traditional beer was 4.1 g/100 ml (+/- 0.873) compared to 2.8 g/100 ml +/- 1.394) in the muchaiwa and 3.6 g/100 ml (+/- 1.445) in the one day brew, chikokiyana. Mean alcohol concentrations in the three commercial beers are reportedly 3.5 g/100 ml in the opaque beer (Scud), and 4.7 to 5.0 g/ml in clear beer (Zambezi and Castle lagers). CONCLUSIONS: Several preparation methods lead to traditional fermented beverages with very high iron concentrations. Measures to prevent dietary iron overload should include all of these beverages in their scope.

Serum transferrin receptors are decreased in the presence of iron overload.

To test the hypothesis that the quantities of circulating transferrin receptors are reduced in iron overload, we studied serum transferrin receptors and indirect measures of iron status in 150 subjects from rural Zimbabwe. We found significant inverse correlations between serum concentrations of transferrin receptors and ferritin, the ratio of ferritin to aspartate aminotransferase, and transferrin saturation (r > or = 0.44; P < 0.001). The mean +/- SD concentration of serum transferrin receptors in 23 subjects classified as having iron overload (ferritin > 300 microg/L and transferrin saturation > 60%) was 1.55 +/- 0.61 mg/L, significantly lower than the 2.50 +/- 0.62 mg/L in 75 subjects with normal iron stores (ferritin 20-300 microg/L and transferrin saturation 15-55%; P < 0.0005) and the 2.83 +/- 1.14 mg/L in 8 subjects with iron deficiency (ferritin < 20 microg/L; P = 0.001). In keeping with the regulation of transferrin receptor expression at the cellular level, our findings suggest that serum transferrin receptors are decreased in the presence of iron overload.

Pleural tuberculosis in Harare, Zimbabwe: the relationship between human immunodeficiency virus, CD4 lymphocyte count, granuloma formation and disseminated disease.

OBJECTIVE: To elucidate the relationship between HIV, CD4+ count and pleural TB. METHOD: In a prospective study, 94 patients presenting at two large Harare hospitals with clinically suspected pleural TB were enrolled over a 10-month period. All underwent standardized evaluation, closed pleural aspiration and biopsy. Patients receiving directly observed anti-TB therapy were followed-up. RESULTS: Pleural TB was diagnosed in 90 individuals (median age 33 years; range 18-65; 64 males); the seroprevalence of HIV was 85%. HIV-positive patients were older than HIV-negative individuals (median age 33 vs 23 years, P = 0.013) and had a significantly lower median CD4+ count (191 vs 1106 x 10(6)/l respectively, P = 0.004). A CD4+ count of <200 x 10(6)/l was associated with a length of illness >30 days (65% vs 37%; P = 0.05), a positive pleural fluid smear (37% vs 0%; P = 0.0006) and a positive pleural biopsy Ziehl-Neelsen stain (35% vs 7%; P = 0.021). However, a relationship between CD4+ count and either pleural granuloma formation or radiological evidence of disseminated disease was not observed. CONCLUSION: In sub-Saharan Africa, TB pleural effusions have become associated with older age, a chronic onset, and an increased mycobacterial load. These data emphasize the complex relationship between pleural TB, HIV infection and a low CD4+ count.