Socio-economic and demographic factors associated with prevalence of HIV infection among pregnant women in Dar es Salaam, Tanzania.

BACKGROUND: HIV/AIDS epidemic has become generalised in low resource settings in sub-Saharan Africa where 90% of all maternal-foetal transmission of HIV infection occurs. Global effort to scale-up pMTCT is underway, however, mechanisms to maximise screening of HIV- 1 positive women for Nevirapine treatment and other interventions, are not clear. OBJECTIVE: To identify socioeconomic and demographic characteristics associated with the prevalence of HIV- 1 infection among Tanzanian women. DESIGN: Cross-sectional study. SETTING: Four antenatal clinics in Dar es Salaam. RESULTS: HIV prevalence rate was 13.1 (95% confidence interval (CI): 12.7% - 13.5%) and it increased with increasing maternal age. Older age than 25, mid-arm circumference less than 25cm, geographic location, working in a public house, and partner's occupation were independently associated with higher prevalence of infection. Women in monogamous marriages were 77% less likely to be HIV infected compared to women with no regular partner. Similarly, women with more than five persons per household, and those who spent less on food had a significantly lower HIV prevalence. CONCLUSION: HIV infection is sufficiently widespread among women in Dar es Salaam suggesting that screening based on socioeconomic and demographic characteristics would miss a large proportion of the positives. There is need to increase facilities for counselling and testing using an opt-out approach for testing in all antenatal clinics in the city.

Evaluation of the FACScount, TRAx CD4 and Dynabeads methods for CD4 lymphocyte determination.

A study to evaluate the performance of the FACScount, TRAx CD4 and Dynabeads methods for the determination of CD4+ T lymphocyte subset levels was conducted in Tanzania as part of a World Health Organization (WHO) collaborative multicenter field evaluation of alternative methodologies for the enumeration of CD4+ T lymphocytes. The objective was to compare the performance of these alternative methods in a developing country setting, against that of flow cytometry as the reference standard. T lymphocyte subset levels were determined in 91 HIV seronegative and 98 HIV-1 seropositive adults using the three alternative methods. CD4+ and CD8+ T lymphocyte counts were determined by all methods except for TRAx CD4 enzyme linked immunosorbent assay (ELISA) which measures CD4+ T lymphocyte levels only. Linear regression analysis was done to correlate the counts obtained by the alternative methods to those obtained by flow cytometry. The overall correlation coefficients of FACScount and Dynabeads CD4+ and CD8+ T lymphocyte counts with those of flow cytometry were high (r > 0.9). A lower correlation (r = 0.631) was obtained when TRAx CD4+ ELISA counts were compared to those of the reference method. These results show that two of these alternative methodologies are suitable for the determination of CD4+ and CD8+ T lymphocyte counts with the use of African blood samples. Since the methods are simpler and cheaper than flow cytometry, they provide an alternative option for the enumeration of T lymphocyte subsets in laboratories with limited facilities.

A new human immunodeficiency virus type 1 circulating recombinant form from Tanzania.

It is becoming increasingly important to identify and to study human immunodeficiency virus type 1 (HIV-1) circulating recombinant forms (CRFs) with evidence of epidemic spread, since mosaic strains arise frequently, especially in populations where multiple subtypes cocirculate. We describe the almost complete nucleotide sequence of 3 subtype C and D recombinant viruses, selected from a pool of 13 D(gag)-D/C/D(env) perinatally infected infants from Dar es Salaam, Tanzania. All three genomes had cross-over points with approximately the same genomic localization. The subtype C-like sequences were located within pol, vif, vpr, vpu, the first exons of rev and tat, V3, and the U3-R regions of the LTR. Phylogenetic analyses of the full-length genomic sequences from these viruses showed the formation of a distinct subcluster on the HIV-1 subtype D branch. The pattern of recombination of genomes belonging to this new CRF, named CRF10_CD, might have resulted from independent recombination events occurring at high frequency or from a single source that originated earlier in this population. Future surveys will be needed to determine the potential of this CRF for epidemic spread.

Antimicrobial susceptibillty of Staphylococcus aureus strains at Muhimbili Medical Centre, Tanzania.

OBJECTIVE: To determine the antimicrobial susceptibility pattern of S. aureus isolates including the presence of methicillin resistant S. aureus strains. DESIGN: Cross-sectional study. SETTING: Department of Microbiology and Immunology, Muhimbili Medical Centre, Dar es Salaam, Tanzania between October 1997 and March 1998. PATIENTS: Two hundred and sixty patients consisting of 67 neonates, 114 children aged 18 years and below and 79 adults. MAIN OUTCOME MEASURES: Antimicrobial susceptibility to tetracycline, erythromycin, cefuroxime, methicillin and penicillin G and presence of mec A gene. RESULTS: Among the S. aureus strains, 97.3%, 68.1%, 37.3% and 6.5% were sensitive to cefuroxime, erythromycin, tetracycline and penicillin G respectively. Only one (0.4%) S. aureus isolate was resistant to methicillin using both the E test and presence of mec A gene. There was no significant difference between the sensitive S. aureus isolates from the neonates, children and adults. CONCLUSION: S. aureus strains are becoming more resistant to commonly used antimicrobial agents, the prevalence of methicillin resistant S. aureus strains in our study population is low compared with reported studies.

Immunohaematological findings in healthy and HIV-1 infected adults in Dar es Salaam, Tanzania.

In order to assess the prognostic value of lymphocyte subsets and immune activation markers in HIV-1 infected Tanzanian patients, peripheral white blood cell(WBC) count, total lymphocytes, CD4+ and CD8+ T-lymphocytes and Beta-2 microglobulin (B-2M) concentrations were determined among healthy HIV-1 seronegative Tanzanian blood donors and in infected Tanzania individuals in different clinical stages of HIV-1 infection. CD4+ T-lymphocytes, CD8+ T-lymphocyte percentages, CD4:CD8 lymphocyte ratios and the concentrations of B-2M were strongly correlated with the clinical stages of HIV-1 infection. These results suggest that B-2M could be a useful prognostic marker in HIV-1 infection in settings where T-lymphocyte subset determinations cannot be done.

PIP: Lymphocyte subsets and concentrations of beta-2 microglobulin (B2M) were determined among 119 HIV-1 seronegative and 183 HIV-1 seropositive individuals at Muhimbili Medical Center (MMC) to assess their prognostic value in HIV-1 infected Tanzanian patients. The HIV-negative individuals were blood donors at MMC, while the HIV-positive participants were blood donors, patients admitted to one medical ward, and those seen at MMC's outpatient clinic who were found to be HIV-positive during the study period. HIV-negative blood donors were of mean age 28.8 years and patients were of mean age 33.5. The measurement of peripheral white blood cell (WBC) count, total lymphocytes, CD4 and CD8 T-lymphocytes, and B2M concentrations found CD4 T-lymphocytes, CD8 T-lymphocyte percentages, CD4:CD8 lymphocyte ratios, and the concentrations of B2M to be strongly correlated with the clinical stages of HIV-1 infection. These findings suggest that B2M could be a useful prognostic marker in HIV-1 infection in settings where T-lymphocyte subset determinations cannot be made.

Detection of anti-HIV-1 IgG antibodies in whole saliva by GACELISA and Western blot assays.

The present study, based on 158 HIV seropositives and 167 HIV seronegatives, demonstrates that saliva collected with the Omni-SAL device and tested with GACELISA (an IgG antibody capture ELISA) is an effective non-invasive alternative to serum for anti-HIV IgG antibody screening. The study also shows that a conventional serum Western blot kit can be used, with slight modifications, for confirmatory testing of saliva specimens. Collecting saliva with the Omni-SAL device had a very good acceptance rate among Tanzanian subjects, and although this diagnostic method is not yet known by the general public, 65% of the study participants preferred to give saliva instead of blood for HIV testing.

PIP: In Dar es Salaam, Tanzania, health workers at Muhimbili Medical Centre collected serum and saliva samples from 135 HIV-positive persons attending the AIDS Clinical Trial Clinic, 130 people who came for voluntary HIV testing, and 60 hospital patients. Researchers aimed to assess the suitability of the Omni-SAL device in collecting saliva and the sensitivity, specificity, and feasibility of detecting HIV-1 IgG antibodies in saliva using GACELISA (an IgG capture ELISA) and Western blot assays. Laboratory personnel optimized Western blot for confirmatory testing of saliva specimens by using a biotin/avidin detection as suggested by McMahan and Hofman. All 135 patients attending the AIDS Clinical Trial Clinic, 8 (6.15%) people undergoing voluntary HIV testing, and 15 (25%) of hospital patients tested positive for HIV (total = 158). GACELISA detected all HIV-1 seropositive individuals and did not detect HIV-1 in any of the HIV-1 seronegative individuals (sensitivity 100%; specificity 100%). The saliva optical density to cut-off value for the HIV-1 seropositives was 5.26-9.82, indicating no ambiguity in the results. All saliva specimens on GACELISA reacted strongly to HIV-1 viral proteins Env, Pol, and Gag on the Western blot optimized for testing saliva specimens. It took more than 10 minutes to saturate the collecting pad (Omni-SAL) in 2% of individuals. Saturation of the collecting pad took less than 3 minutes in most cases (64%). Most individuals preferred saliva to be collected for HIV testing than serum and urine (65% vs. 23% and 12%, respectively). 96% of all individuals thought the Omni-SAL device to be easy. These findings suggest that saliva is an adequate specimen for screening and diagnosis of HIV infection. Since many saliva samples can be collected quickly, easily, and safely, Omni-SAL and GACELISA can be done under any field situation by people with minimal training.

Comparison of in-house and commercial sample preparation and PCR amplification systems for detection of human immunodeficiency virus type 1 DNA in blood samples from Tanzanian adults.

This study compared the performance of several in-house nested PCR systems and the Amplicor human immunodeficiency virus type 1 (HIV-1) PCR kit in the detection of HIV-1 DNA in Tanzanian samples prepared by two different methods. All six of the in-house primer sets evaluated had a higher sensitivity for HIV DNA detection in samples prepared by the Amplicor PCR sample preparation method than in those prepared by the Ficoll-Isopaque (FIP) density gradient centrifugation method. A sensitivity of 100% was achieved by combining two in-house primer sets. The sensitivity of the standard Amplicor HIV-1 PCR kit was only 59%, whereas a modified Amplicor HIV-1 PCR test had a sensitivity of 98%. Our data show that Tanzanian samples prepared by the Amplicor preparation method are more suitable for HIV-1 PCR testing than samples prepared by the FIP method. The modified, but not the standard, Amplicor HIV-1 PCR kit provides an alternative to the nested in-house PCR technique for the diagnosis of HIV infection.

PIP: Blood samples were collected from 73 pregnant mothers attending an antenatal clinic and from 14 adult females recruited into ongoing studies of the incidence and natural history of HIV-1 infection in Dar es Salaam, Tanzania. Study subjects were asymptomatic for HIV infection, but 65 tested HIV-positive. The authors compared the performance of several in-house nested polymerase chain reaction (PCR) systems and the Amplicor HIV-1 PCR kit in detecting HIV-1 DNA in the seropositive samples prepared by two different methods. All six of the in-house primer sets evaluated were more sensitive for HIV DNA detection in samples prepared by the Amplicor PCR sample preparation method than in those prepared by the Ficoll-Isopaque (FIP) density gradient centrifugation method. A sensitivity of 100% was achieved by combining two in-house primer sets. The sensitivity of the standard Amplicor HIV-1 PCR kit was only 59%, while a modified Amplicor HIV-1 PCR test had a sensitivity of 98%. These data indicate that Tanzanian samples prepared by the Amplicor preparation method are more suitable for HIV-1 PCR testing than samples prepared by the FIP method.

Is there an association between periodontal condition and HIV infection?

Individuals in Tanzania who have limited access to medical and dental treatment provide an opportunity to study the natural association between periodontal condition and HIV infection and the stage of infection. 119 HIV-infected adult individuals and 73 individuals with AIDS from the AIDS Clinical Trial Clinic at Muhimbili Medical Centre (MMC) in Dar-es-Salaam participated as cases. Mean age was 35.3 and 35.1 years, respectively. 156 individuals with a mean age of 28.3 years, confirmed as HIV-seronegative, served as controls. There were no significant differences in bleeding on probing, pocket formation or attachment loss among the HIV-seronegative individuals, HIV-seropositive and AIDS patients. We applied multiple logistic regression to calculate odds ratios for presence of periodontal conditions adjusting for age, gender and DMFT. Our odds ratios did not reveal any significant associations between bleeding on probing, pocket formation or attachment loss with regard to lymphocyte and CD4+ T-cell counts among the HIV-infected individuals and AIDS patients. When associations were investigated with regard to HIV-serostatus (HIV-seronegative, HIV-seropositive or AIDS), our adjusted odds ratios were insignificant, too. In fact, most odds ratios were close to 1. Thus, our study supports recent views that the presence, extent and severity of periodontal disease among HIV-infected individuals, may be less that hitherto thought.

Differences in perinatal transmission among human immunodeficiency virus type 1 genotypes.

OBJECTIVE: To determine whether genotypes from human immunodeficiency virus type 1 (HIV-1) subtypes A, C, or D or intersubtype recombinants have the same probability of being transmitted from mother to child. METHODS: We determined the HIV-1 genetic subtype and maternal risk factors of 51 matched transmitting and nontransmitting mothers from Tanzania. The HIV-1 gag (p24-p7) and env (C2-C5) nucleotide sequences were used for genotype classification, and matched logistic regression analysis was used to assess differences among genotypes. RESULTS: Mothers infected with HIV-1 subtype A (odds ratio, 3.8; 95% CI, 0.8-24.7%), HIV-1 subtype C (odds ratio, 5.1; 95% CI, 1.3-30.8%), or HIV-1 intersubtype recombinant viruses (odds ratio, 5.3; 95% CI, 1.2-33.4%) were more likely to transmit HIV-1 to their infants than mothers infected with HIV-1 subtype D. Lower CD4 cell counts at enrollment were associated with transmission, but CD4 cell counts within each genotype did not explain differences in transmission among HIV-1 genotypes. CONCLUSION: We have shown that HIV-1 genotypes might be associated with differential risk for vertical transmission. These findings provide the first evidence that HIV-1 genetic subtypes may play a role in rates of vertical transmission in an African setting.

HIV-1 LTR subtype and perinatal transmission.

Multiple subtypes of HIV-1 have been identified; however, there is little data on the relative transmissibility of viruses belonging to different subtypes. A matched case-control study addressed whether viruses with different long terminal repeat (LTR) subtypes were transmitted equally from mother to infant. The LTR subtype was determined for 45 matched cases and controls who participated in a clinical trial in Tanzania. HIV-1 subtypes A, C, and D and intersubtype recombinant sequences were identified. Exact matched logistic regression analysis showed that viruses containing subtype A or intersubtype recombinant LTRs were 3.2 and 4.8 times more likely to be transmitted from mother to infant than viruses with subtype D LTRs. Viruses containing subtype C LTRs were 6.1 times more likely to be transmitted than those with subtype D LTRs. These differences in transmission were independent of maternal CD4 at enrollment. Thus, it appears that HIV-1 subtype may be associated with differing rates of perinatal transmission in Tanzania.