Extensive Mycobacterium tuberculosis circulation in a highly endemic prison and the need for urgent environmental interventions.

Aimed at assessing the circulation of Mycobacterium tuberculosis in a highly endemic prison, this 13-month prospective study was performed on strains isolated from tuberculosis (TB) cases detected passively and actively. We used X-ray screening of newly admitted inmates and mass screening at the beginning of the study and again 1 year later. Of the 94 strains genotyped by restriction fragment-length polymorphism, 79 (84·0%) belonged to one of the 12 identified clusters (2-21 strains each), including two main clusters (18 and 21 cases, respectively). A history of TB treatment was reported in 22/79 (27·8%) clustered cases. Time-space distribution of clustered cases was predominantly consistent with transmission, in micro-epidemics. Given the dominant pattern of exogenous infection and the extensive strain circulation, effective TB control should emphasize reduction of overcrowding and improvement of environmental measures as a complement to detection and treatment of cases.

International institutions, global health initiatives and the challenge of sustainability: lessons from the Brazilian AIDS programme.

The sustainability of successful public health programmes remains a challenge in low and middle income settings. These programmes are often subjected to mobilization-demobilization cycle. Indeed, political and organizational factors are of major importance to ensure this sustainability. The cooperation between the World Bank and the Brazilian AIDS programme highlights the role of international institutions and global health initiatives (GHI), not only to scale up programmes but also to guarantee their stability and sustainability, at a time when advocacy is diminishing and vertical programmes are integrated within health systems. This role is critical at the local level, particularly when economic crisis may hamper the future of public health programmes. Political and organizational evolution should be monitored and warnings should trigger interventions of GHI before the decline of these programmes.

Malaria risk perception, knowledge and prophylaxis practices among travellers of African ethnicity living in Paris and visiting their country of origin in sub-Saharan Africa.

An observational prospective cohort study assessed malaria risk perception, knowledge and prophylaxis practices among individuals of African ethnicity living in Paris and travelling to their country of origin to visit friends or relatives (VFR). The study compared two groups of VFR who had visited a travel clinic (TC; n=122) or a travel agency (TA; n=69) before departure. Of the 47% of VFR citing malaria as a health concern, 75% knew that malaria is mosquito-borne and that bed nets are an effective preventive measure. Perception of high malaria risk was greater in the TA group (33%) than in the TC group (7%). The availability of a malaria vaccine was mentioned by 35% of VFR, with frequent confusion between yellow fever vaccine and malaria prevention. Twenty-nine percent took adequate chemoprophylaxis with complete adherence, which was higher among the TC group (41%) than the TA group (12%). Effective antivector protection measures used were bed nets (16%), wearing long clothes at night (14%) and air conditioning (8%), with no differences between the study groups except in the use of impregnated bed nets (11% of the TC group and none of the TA group). Media coverage, malaria chemoprophylaxis repayment and cultural adaptation of preventive messages should be improved to reduce the high rate of inadequate malaria prophylaxis in VFR.

Benefit of amoxicillin in differentiating between TB suspects whose initial AFB sputum smears are negative.

SETTING: Out-patient dispensary in Conakry, Guinea, West Africa. OBJECTIVE: To differentiate between pulmonary tuberculosis (PTB) and non-PTB diseases among 204 acid-fast bacilli (AFB) smear-negative adult TB suspects. DESIGN: We derived scores from clinical, serological and radiological findings among PTB suspects aged > or = 15 years who, after having had three AFB-negative smears, were treated for 10 days with amoxicillin (AMX, 1.5 g/day). RESULTS: At the selected cut-off score from model 1 (clinical), sensitivity for PTB was 95%, specificity 40%, negative predictive value (NPV) 84%, and positive predictive value (PPV) 69%. Comparable values from model 2 (clinical + serological + radiological) were: sensitivity 99%, specificity 45%, NPV 97%, and PPV 71%. Results from AMX were better: sensitivity 92%, specificity 93%, NPV 94%, and PPV 91%. Of the 117 suspects who failed to respond clinically and radiographically to AMX and remained AFB smear-negative, 110 (94%) had PTB, confirmed either by positive culture (73 patients) or response to anti-tuberculosis treatment (37 patients). CONCLUSION: The clinical and radiographic response to AMX is better than derived scores at differentiating between PTB and non-PTB in TB suspects presenting to a dispensary in Guinea, a low HIV-seroprevalence country.

Forecasting the development of primary hepatocellular carcinoma by the use of risk factors: studies in West Africa.

An association between hepatitis B virus (HBV) and primary hepatocellular carcinoma (PHC) has been found in several studies in Africa, Asia, and elsewhere. In this paper we considered the interrelations between several events related to HBV infection, which include the presence of: 1) hepatitis B surface antigen (HBsAg), 2) antibody to hepatitis B core antigen (anti-HBc), 3) antibody to the surface antigen (anti-HBs), 4) chronic liver disease, 5) elevated alpha-fetoprotein, and 6) PHC. With the use of preliminary epidemiologic data, risk factors related to these events were calculated. We suggested that the interactions between these events and HBV infection in parents be used to estimate the risk of PHC for an individual in this environment.

Antibodies to HTLV-I p24 in African and Portuguese populations.

Using a radioimmunoassay to detect HTLV-I protein antibodies of molecular weight 24,000, we screened populations from Algeria (140 subjects), Tunisia (442), Mali (69), Senegal (415), Uganda (135), the Central African Republic (77), the Congo (360), and Madagascar (193). Only four subjects were positive (1 from Senegal, 1 from Uganda, 2 from the Congo). This is a much lower figure than that found by others in Africa by the enzyme-linked immunosorbent assay technique. In addition, 319 Portuguese blood donors (46 of whom have lived in Angola or Mozambique) were screened using the same radioimmunoassay. All were negative.

Prevalence of pulmonary tuberculosis and comparative evaluation of screening strategies in a Brazilian prison.

SETTING: A prison (1171 male inmates) in Rio de Janeiro, Brazil. OBJECTIVES: To determine the prevalence of active pulmonary tuberculosis (TB) and to assess the performance of several screening strategies. DESIGN: In a cross-sectional study, all inmates underwent chest radiographic screening. Subjects with abnormal findings underwent sputum smear examination and sputum culture. Taking this strategy as the reference, we assessed three targeted screening strategies to identify TB suspects: Strategy 1: cough >3 weeks; Strategy 2: WHO score > or = 5; Strategy 3: presence of at least one potentially TB-related symptom. RESULTS: The prevalence of TB cases was 4.6% (48/1052) and 2.7% for definite TB cases. If TB suspects identified by targeted screening had sputum smear examination alone, 37 (86.0%) of the 43 cases would have been missed by Strategy 1, 34/43 (79.1%) by Strategy 2 and 34/43 (79.1%) by Strategy 3. If TB suspects had both sputum smear examination and, for smear-negative subjects, chest radiography, respectively 28/43 (65.1%), 18/43 (41.9%) and 13/43 (30.2%) of cases would have been missed. CONCLUSION: All three targeted screening strategies were unreliable. Given the importance of early TB diagnosis in overcrowded and highly endemic settings, routine radiography-based screening may be warranted.

Bacillus Calmette-Guérin immunization in infants born to HIV-1-seropositive mothers.

During the prospective follow-up of 64 babies at risk for perinatal HIV-1 infection because their mothers were seropositive, and of 130 control babies whose mothers were seronegative, we studied the occurrence of complications of bacillus Calmette-Guérin (BCG) immunization and its ability to induce cutaneous reactivity to tuberculin. Babies born both to HIV-1-positive and HIV-1-negative mothers received BCG immunization during their first month of life according to the Expanded Programme on Immunization (EPI) recommendations. Local and regional complications of BCG vaccine were looked for at 3, 6 and 9 months after inoculation. A tuberculin skin test was performed at 6 or 9 months of age. Most babies born to HIV-1-positive mothers were later classified as infected or uninfected according to their clinical condition and/or serological status at 18 months of age. The mean duration of the follow-up was 36 months (range 30-40 months). No chronic or deep ulcerations at the site of injection or disseminated forms of BCG infection were observed. The frequency of BCG-related lymphadenitis in the group of HIV-1-infected children (24%) did not differ significantly from the group of uninfected children (19%; Fisher test: P = 0.73). In contrast, the tuberculin skin test responses were positive less often in the group of HIV-1-infected children (33%) than in the uninfected group (83%; Fisher test: P = 0.007). Because BCG vaccine appears to be safe--even when given to perinatally infected babies--continuation of the BCG immunization policies of the EPI is justified, especially in view of the growing incidence of tuberculosis as a complication of HIV infection.

Mother-child transmission of HIV-1 and infant survival in Brazzaville, Congo.

The aim of this study was to compare the probability of survival of infants born to anti-HIV-1-positive and anti-HIV-1-negative mothers. One thousand, eight hundred and thirty-three pregnant women, recruited sequentially in two mother-child clinics in Brazzaville, were screened for anti-HIV-1 (by enzyme-linked immunosorbent assay with confirmation by Western blot). Each seropositive mother (71 out of 1833, 3.9%) was matched for age, presumed date of delivery and place of residence with two seronegative mothers. Sixty-four babies born to anti-HIV-1-positive mothers and 130 control babies born to anti-HIV-1-negative mothers were followed up for 12-22 months (mean, 18 months). The probabilities of survival were estimated by the Kaplan-Meier method. At birth, the two groups of babies did not differ with regard to rate of stillbirths, gestational age, sex ratio and weight. Among babies born to seropositive mothers, the probability of survival was 0.87 (s.d. 0.04) at 3 months, 0.71 (s.d. 0.06) at 6 months, 0.68 (s.d. 0.06) at 9 months and 0.61 (s.d. 0.06) at 12.5 months. In the controls the probability of survival was 0.98 (s.d. 0.01) at 3 months and 0.97 (s.d. 0.02) at 12 months. The excess of mortality in the babies born to anti-HIV-1-positive mothers is highly significant (P less than 0.001). The deaths occurred more frequently and earlier than in similar cohort studies performed in developed countries.

Epidemiological and molecular characteristics of HIV infection in Gabon, 1986-1994.

OBJECTIVE: To describe trends in the prevalence of HIV-1 infection in different populations in Gabon, and the molecular characteristics of circulating HIV strains. METHODS: Data were collected on HIV prevalence through sentinel surveillance surveys in different populations in Libreville (the capital) and in Franceville. In Libreville, a total of 7082 individuals (hospitalized patients, tuberculosis patients, pregnant women, asymptomatic adults, prisoners) were recruited between 1986 and 1994. In Franceville, we tested 771 pregnant women and 886 healthy asymptomatic adults (1986-1988). Sera were screened for HIV antibodies by enzyme-linked immunosorbent assay (ELISA) and confirmed by Western blot or line immunoassay (LIA). Reactive samples in ELISA were tested for the presence of antibodies to HIV-1 group O viruses by ELISA using V3 peptides from HIV-1 ANT-70 and HIV-1 MVP-5180 followed by confirmation by LIA and a specific Western blot. Seventeen HIV-1 strains were isolated (1988-1993) and a 900 base-pair fragment encoding the env region containing V3, V4, V5 and beginning of gp41 was sequenced and a phylogenetic tree was constructed. RESULTS: HIV prevalence was relatively low and remained stable (0.7-1.6% in pregnant women, 2.1-2.2% in the general population). The prevalence was also stable among prisoners (2.1-2.6%). Among hospitalized and tuberculosis patients prevalence was higher and increased (1.8-12.7% and 1.5-16.2%, respectively). Only three sera had antibodies to HIV-1 group O. The 17 HIV-1 strains represent six different genetic subtypes including type O. CONCLUSION: Our data from 1986 to 1994 show a stable and low HIV prevalence in Gabon, and a high genetic diversity of HIV-1 strains. This, also observed in Cameroon, is in contrast to that found elsewhere in Africa. Differences in rate of spread of HIV infection are probably explained by interplay between numerous factors. The role of different HIV subtypes in the dynamics of the HIV epidemic should be examined further.

Characteristics associated with HIV-1 infection in pregnant women in Brazzaville, Congo.

Maternal human immunodeficiency virus type 1 (HIV-1) infection in sub-Saharan Africa is a major public health concern because of the high prevalence among women of childbearing age and the poor prognosis for perinatally infected children. Characteristics associated with HIV seroprevalence were studied in a population of 1,833 pregnant women seen in two large mother-child clinics in Brazzaville, Congo. The prevalence of HIV infection was 3.9% (95% confidence interval, 3.0-4.9%) and differed significantly according to the district of residence, marital status, duration of the relationship with the current partner, number of sexual partners in the year prior to pregnancy, number of living and dead children, and history of blood transfusion and/or hospitalization. Logistic regression analysis identified six significant factors independently associated with seropositivity; age, history of blood transfusion and/or hospitalization, district of residence, duration of the relationship, number of living children, and number of decreased children. However, the predictive value of the model was poor: while 80% of the truly positive women were correctly predicted positive by the model, 50% of the truly negative women were misclassified. Among pregnant women attending these clinics it is therefore difficult to identify a subgroup at risk toward which specific actions could be targeted.

Blood transfusion as a major risk factor for HTLV-I infection among hospitalized children in Gabon (Equatorial Africa).

To study the risk factors for HTLV-I infection of African infants living in a high seroprevalence area, a survey was conducted among hospitalized children and their mothers in Franceville, southern Gabon. A total of 610 children (6 months to 14 years old) from 555 families and 434 mothers participated in the study. HTLV-I seroprevalence was 7.1% in the mothers and 2.8% in the children. No increase by age was observed in the children. Significantly more children with sickle cell anemia (11 of 57, or 19.2%) were infected than others (1%) (Fisher's exact test p < 0.001). Of the 13 seropositive infants (C+) whose mothers were tested, six (43%) had a seropositive mother (M+) [p < 0.001, prevalence ratio (PR) 10.12, 95% CI 3.40-30.35, attributable risk (AR) 0.55]. Risk factors for nonmaternally acquired HTLV-I infection were previous blood transfusion (Fisher's exact test p < 0.001, PR 15.74, CI 5.20-47.60, AR 0.71) and hospital admission. In a hospitalized pediatric population in Gabon mother-to-child transmission and blood transfusion (because of sickle cell disease) were equally implicated in HTLV-I infection. Control of HTLV-I infection in children in Africa requires prevention of both vertical and transfusion-associated transmission, which may be exceedingly difficult in developing countries.

Seroepidemiological survey of hepatitis C virus among commercial sex workers and pregnant women in Kinshasa, Democratic Republic of Congo.

BACKGROUND: Studies conducted mainly in industrialized countries have shown that the transmission of hepatitis C virus (HCV) is mainly parenteral, and have emphasized the role of nosocomial transmission. In Equatorial Africa, the respective contributions of parenteral and non-parenteral routes of transmission are unknown. The potential role of sexual transmission in this area of high HCV endemicity, where sexually transmitted infections (STI) are frequent, is suggested by the fact that HCV infection is rare in infants and young adolescents, but increases thereafter with age. The present study, conducted in Democratic Republic of Congo, was designed to determine the prevalence of HCV infection and associated sexual risk factors in two female populations with different sexual behaviour. METHODS: Cross-sectional studies conducted among commercial sex workers (CSW; n = 1144) and pregnant women (n = 1092) in the late 1980s in Kinshasa showed a high frequency of at-risk sexual behaviour, STI and human immunodeficiency virus (HIV) infection, particularly among CSW. We screened samples collected during these epidemiological studies for antibodies to HCV using a second-generation ELISA with confirmation by a third-generation LIA. We also assessed sociodemographic variables, medical history, STI markers and sexual behaviour, and their potential association with HCV infection. RESULTS: The overall prevalence of anti-HCV was 6.6% (95% CI : 5.2-8.2) among CSW and 4.3% (95% CI : 3.2-5.7) among pregnant women (age-adjusted OR = 1.5, 95% CI : 1.0-2.1, P = 0.05). Multivariate analysis showed that the presence of anti-HCV among CSW was independently associated with a previous history of blood transfusion (P < 0.001), age >30 years (P < 0.001) and the presence of at least one biological marker of STI (P < 0.03). No such links were found among pregnant women (although the history of blood transfusions was not investigated in this group). Anti-HCV was not associated with sociodemographic variables or sexual behaviour in either group, or with individual markers of STI. Despite the high-risk sexual behaviour and the higher prevalence of STI in CSW, the difference in HCV seroprevalence between CSW and pregnant women (6.6% versus 4.3%) was small, particularly when compared with the difference in the seroprevalence of HIV (34.1% versus 2.8%). CONCLUSION: The role of sexual transmission in the spread of HCV seems to be limited. Parenteral transmission (including blood transfusion and injections), possibly related to the treatment of STI, probably plays a major role.

Seroepidemiology of hepatitis B in a population of children in central Tunisia.

The prevalence rate of HBV markers in the overall population was 18.0%. The prevalence rates of HBsAg and anti-HBs were 3.3% and 12.2% respectively. 2.5% of the children were negative for these markers but positive for anti-HBc alone. Of the 17 HBsAg positive children, 10 were HBeAg positive and four were anti-HBe positive. The statistical study using multifactorial correspondence analysis and the chi 2 test showed a positive relationship between the prevalence rate of anti-HBs and age, and a negative relationship between the prevalence rate of 'absence of markers' and age. The male sex and urban dwelling were related to higher prevalence rates of HBsAg and anti-HBc alone and to a lower prevalence rate of 'absence of markers'. No relationship was found between geographical zone, father's education and HBV markers. The implications of these results in terms of prevention are discussed.

Demographic factors associated with HTLV-1 infection in a Gabonese community.

An epidemiological study of human T-lymphotropic virus type 1 (HTLV-1) and syphilis has been carried out in a multiethnic community of seven neighbouring villages located in eastern Gabon on 1240 subjects over 5 years old (82.7% of the population in this age range). Antibodies to HTLV-1 (anti-HTLV-1) were detected by ELISA with confirmation by Western Blot and antibodies to syphilis by Venereal Diseases Research Laboratory assay with confirmation by the Treponema pallidum haemaglutination assay. The prevalence rate of anti-HTLV-1 was 8.5% and increased from 3.7% in the 5-14 years age group to 23.8% in the over 60 years age group. Logistic regression showed that the positivity for anti-HTLV-1 was associated with age, ethnic group and sex (higher prevalence in females). The seroprevalence rate of syphilis was 8.2%. Seropositivity for syphilis and HTLV-1 were related but age was a confounding variable in this relationship. This study showing a highly heterogeneous distribution of HTLV-1 in a geographically limited area suggests the role of environmental and behavioural factors in HTLV-1 transmission.

Seroprevalence of hepatitis E virus in an adult urban population from Burundi.

The prevalence of antibodies to the hepatitis E virus (HEV) was measured in a group of 129 adults from Bujumbura, Burundi, using an ELISA. The prevalence of anti-HEV IgG was 14%, much lower than that of hepatitis A virus (HAV) (97.7%). In addition to the lability of antibodies to HEV, this difference might be explained by the extensive availability of good-quality drinking water in the city. The presence of serologic markers of HBV (77.6%), HCV (27.1%), and human immunodeficiency virus (30.2%) was not associated with that of anti-HEV.

PIP: The seroprevalence of hepatitis E virus (HEV) was measured through use of data from a 1992-93 case-control study of patients with chronic liver diseases conducted at Kamenge University Hospital in Bujumbura, Burundi. 97.7% of subjects were anti-hepatitis A virus (HAV)-positive. In contrast, the seroprevalence of anti-HEV IgG was only 14%. Hepatitis B virus (HBV) markers were as follows: HBV surface antigen, 4.7%; antibody to HBV surface antigen, 55.8%; and antibody to HBV core antigen, 65.1%. The prevalence for all 3 HBV markers combined was 77.6%. No seropositivity was found for anti-hepatitis D virus among subjects positive for HBV surface antigen (4.7%) or for antibody to HBV core antigen (17.1%). 27.1% were anti-hepatitis C virus-positive. The prevalence of HIV was 30.2%. The presence of serologic markers of hepatitis A, B, and C virus was not associated with that of antibody to hepatitis E or HIV. Previous studies have found high rates of HEV in areas that have experienced high rainfall and flooding. The relatively low rate of HEV recorded in this study may reflect the fact that most Bujumbura residents use drinking water pumped from the middle of Lake Tanganyika and piped to taps near homes.

Serologic survey of human hydatid disease in high risk populations from central Tunisia. Preliminary results.

The annual surgical incidence rate of hydatid disease in central Tunisia varies from 11.1 to 30.1 per 100,000 population depending on the district. In order to obtain more comprehensive data, we conducted a serologic survey in the district with the highest rate, using hemagglutination-inhibition to test 480 workers from a state farm and 190 inhabitants of villages in the neighborhood. The overall prevalence rate was 1.3% (9/670). Echotomography revealed liver cysts in all cases and the diagnosis was confirmed in the eight subjects who underwent surgery. If these preliminary results are confirmed by further surveys, mass screening might be considered in high risk districts.

Echotomographic and serologic screening for hydatidosis in a Tunisian village.

Echotomographic and serologic screening for hydatidosis was carried out in 1,650 subjects aged over 5 years in a village located in a high risk region of central Tunisia. Echotomography detected liver cysts in 6 subjects (prevalence rate: 3.6 per 1,000). These 6 cases were among the 7 positive cases detected by ELISA. In all 6, the diagnosis of liver cysts was confirmed. No other localization was found in any of the subjects including one with a positive ELISA and a normal echotomographic examination. This work demonstrates the high feasibility of echotomography in the field and confirms the high prevalence rate of hydatidosis in the community.

Absence of a relationship between antibodies to hepatitis C virus and hepatocellular carcinoma in Mozambique.

To study the relationship between hepatocellular carcinoma (HCC) and hepatitis C virus (HCV), sera from 178 patients with HCC and 194 blood donors from Maputo, Mozambique were tested for antibodies to HCV (anti-HCV) using a second generation enzyme-linked immunosorbent assay and a confirmatory test with six synthetic peptides as reagents. The presence of hepatitis B surface antigen (HBsAg) was tested using an enzyme immunoassay. The prevalence of anti-HCV was higher in patients with HCC than in controls, but the difference was not significant after adjustment for age. Therefore, this difference reflected a difference in the age structure between the two groups. The prevalence of HBsAg was higher in patients with HCC than in controls. There was a negative association between anti-HCV and HBsAg in patients with HCC that was not significant after adjustment for age. These serologic results, which contrast with previous reports, show the need for further studies on the relationship between HCC and HCV using second generation serologic tests and molecular biology techniques.

Prevalence and genetic variants of hepatitis GB-C/HG and TT viruses in Gabon, equatorial Africa.

The distribution of Hepatitis GB-C/HG (GB-C/HG) and TT viruses (TTV) infections was investigated in selected populations from Gabon using Polymerase Chain Reaction (PCR) and Enzyme Linked Immunosorbent Assay (ELISA) for anti-Envelop 2 (anti-E2) GBV-C/HGV antibodies. Among pregnant women, 29 of 229 (12.6%) were Hepatitis GB virus-C and Hepatitis G virus (GBV-C/HGV) RNA positive (+) and 32 of 81 (39.5%) anti-E2 + versus 8 of 39 (20.5%) TTV DNA +. Among sickle cell anemia patients, 9.7% (3/31) were GBV-C/HGV RNA + versus 22.5% (7/31) TTV DNA +. For tuberculosis patients, the figures were 11.5% (4/35) and 0%. A study of hepatocellular carcinoma cases (n = 27) versus controls (n = 66) did not show significant differences for GBV-C/HGV RNA (10.7% versus 12.1%) and TTV DNA (44.4% versus 30.3%). According to phylogenetic analysis, the 15 GBV-C/HGV strains investigated clustered in group 1, the most common in sub-Saharan Africa whereas TTV sequences (n = 4) mostly clustered in genotypes G1 and one close to genotype G3. In the Gabonese populations investigated, GBV-C/HGV and TTV infections were highly endemic. These data are consistent with the low pathogenicity of these agents.