Prevalence of HIV and syphilis among high-risk groups in Bangladesh

Published information regarding the prevalence of HIV in Bangladesh is limited. Surveys conducted in the past few years [1–5] show that the prevalence of HIV in Bangladesh is still low compared to its surrounding countries [6]. In order to determine the impact of the HIV epidemic on Bangladesh, a sentinel surveillance system was organized among individuals exhibiting high-risk behaviour. This is a report of the results of the first year of this surveillance. The population groups included in the surveillance were brothel-based and floating sex workers, men with symptoms or signs of sexually transmitted diseases (STD), truck drivers and their helpers (truckers), men who have sex with men and injecting drug users (IDU). All groups were from urban areas in central Bangladesh, mainly the capital city, Dhaka. STD patients were also sampled from the north east, north west and south east regions of Bangladesh, and included men with genital ulcers, inguinal buboes or urethral discharge attending the skin and venereal disease departments of government hospitals. Brothel-based sex workers were recruited from clinics serving two brothels in central Bangladesh. IDU were sampled from detoxification clinics and included drug users with a history of injecting drug use in the past year. Truckers, floating sex workers and men who have sex with men were accessed through clinics to which they are usually motivated to attend by out-reach workers in the communities. Sex workers were all female, one IDU was female whereas the rest were male and other groups were all male. The sample size was calculated to be 380 with an estimation of the HIV prevalence rate of 1% with a 1% precision and 95% confidence level. Comparisons between groups were carried out using the Fisher's exact test, and differences were considered to be significant at a level of P < 0.05. Statistical analysis was carried out using Epi Info (version 6; USD, Stone Mountain, GA, USA). Serum was separated from blood collected by venepuncture into sterile, plain vacutainers (Becton Dickinson, Rutherford, NJ, USA). Serum samples were transported to the International Centre for Diarrhoeal Disease Research Laboratory, Bangladesh, by maintaining the cold chain. Samples collected from Dhaka were stored at −70°C, whereas those from sites outside Dhaka were stored at −20°C until testing. Testing for syphilis was linked and results were reported back to the clinics within 1 week for treatment purposes. Syphilis was tested by the rapid plasma reagin test (Organon Teknika, Turnhout, Belgium) and the Treponema pallidum haemagglutination assay (Organon Teknika). Samples positive for both assays were considered to be positive. HIV testing was unlinked, anonymous with de-linking done at the clinics. For HIV, samples were initially tested by a commercial enzyme-linked immunosorbent assay (ELISA) kit (Organon Teknika), and positive results were confirmed by a line immunoassay (LIA; Organon Teknika). An indeterminate result by LIA was considered to be negative. Quality control was carried out using standard quality control sera for HIV, which were provided by the National Reference Centre for HIV/AIDS, India, at the Christian Medical College and Hospital, Vellore, India. Twenty-two serum samples stored at –70°C that were positive for HIV by ELISA were transported to the Department of Medical Microbiology, University of Edinburgh Medical School, UK, on dry ice for characterization. HIV was characterized by polymerase chain reaction and sequencing as described previously [7]. The survey was conducted between June 1998 and March 1999. The prevalence of HIV and syphilis are shown in Table 1. A total of 17 people were confirmed to be HIV positive. Although HIV prevalence rates were significantly different (P = 0.027) in sex workers from the two brothels, the syphilis rates were similar. Among individual groups, the HIV prevalence rate was highest in IDU. IDU are at particular risk because of their needle-sharing behaviour. The consequence of continued high-risk behaviour in IDU is exemplified by the rapid rise in HIV rates among IDU in Manipur, India, where the prevalence rates have risen from 8% in 1989 to 89% in 1994 [8]. Alongside this sero-surveillance, a behaviour surveillance was conducted among similar groups of people, the findings of which show that needle sharing is highly prevalent among IDU in Bangladesh (Jenkins, unpublished observation). The low rate of syphilis in truckers is in contrast to truckers of India who exhibit high-risk behaviour and have high rates of syphilis [9]. Results from the concomitant behaviour surveillance suggest that high-risk behaviour among truckers is confined to less than half the truckers sampled (Jenkins, unpublished observation).Table 1: Prevalence of HIV and syphilis. HIV characterization showed that the virus was HIV-1 subtype C. Subtype C is the dominant HIV subtype in India [10], and has also been reported before from Bangladesh [3]. The high prevalence rates for syphilis among sex workers and the relatively high HIV prevalence rate in IDU suggests that Bangladesh is at the beginning of an HIV epidemic, which has the potential of spreading rapidly. Rapidly employed targeted interventions at this stage could have a beneficial effect by aborting the epidemic. Acknowledgements The authors are grateful to the following institutions and departments for their active participation in providing access to and samples from the risk groups: Skin and Venereal Disease Department of Dhaka Medical College Hospital, Sir Salimullah Medical College Hospital, MAG Osmani Medical College Hospital, Rajshahi Medical College Hospital; Microbiology Departments of Dhaka Medical College, Sir Salimullah Medical College, MAG Osmani Medical College, Rajshahi Medical College; Agrabad Central Skin and Social Hygiene Centre; Bangladesh Women's Health Coalition; CARE, Bangladesh; Marie Stopes Clinic Society; Bandhu Social Welfare Society; Central Drug Treatment Centre; Mukti Lawrence Foundation; Paricharja. The authors would also like to thank the Technical Advisory Group for their invaluable support and advice throughout the surveillance. Tasnim Azima Mohammed N. Islamb Jozef Bogaertsa Mohammed A. H. Mianc Mohammed S. Sarkera Kazi R. Fattaha Peter Simmondsd Carol Jenkinse Mahmud R. Choudhuryf Vadakenadayil I. Mathana