From the aNational AIDS Reference Laboratory, National Center for AIDS/STD Control and Prevention, Beijing, China
bHIV Confirmatory Central Laboratory, Chongqing Municipal Center for Disease Control and Prevention, Chongqing, China
cHIV Confirmatory Central Laboratory, Xinjiang Uigyur Autonomous Region Center for Disease Control and Prevention, Urumqi, Xinjiang, China
dHIV Confirmatory Laboratory, Dehong Dai and Jingpo Autonomous Prefecture Center for Disease Control and Prevention, Dehong, Yunnan, China.
Correspondence to Yan Jiang, National AIDS Reference Laboratory, National Center for AIDS/STD Control and Prevention, China CDC, Nanwei Road No 27, Beijing 100050, China. Tel: +86 10 63024898; fax: +86 10 63024898; e-mail: email@example.com
The HIV pandemic continues to spread with nearly 40 million people living with HIV worldwide. Globally, approximately 4.1 million new infections were believed to have occurred in 2005 . In China, there are approximately 650 000 individuals currently living with HIV/AIDS and approximately 70 000 new HIV infections are estimated to have occurred in 2005. Injection drug use represents the largest single cause of HIV transmission in China, accounting for 44.3% of infections at the end of 2005 . The Ministry of Public Security data suggest that the number of registered drug users was approximately 1.16 million in 2005. The total number, including unregistered drug users, was thought to be much higher, with one estimate suggesting 3.5 million . As the HIV epidemic is spreading from high-risk groups to the general population, the control of HIV in injection drug users (IDU) can help prevent further spread in China .
Estimation of HIV-1 incidence is a key component of monitoring the HIV-1 epidemic in various populations around the world. It is important to understand the current status of transmission dynamics, identify high-risk populations, monitor prevention efforts and target resources at programmes that are most effective in reducing transmissions . The identification of recently infected individuals (generally within 6 months of infection) and accurate estimations of incidence are, however, difficult and have relied traditionally on following a prospective cohort of people at risk.
In the past few years, several different laboratory methods have been proposed to estimate HIV-1 incidence from cross-sectional surveys. In 1998, Janssen and colleagues  described a serological testing algorithm for recent HIV seroconversion approach to detect recent infections and to estimate HIV-1 incidence in various populations. Since then there has been more interest in using laboratory methods to estimate HIV incidence. In 2002, Parekh and colleagues  developed a new assay, the HIV-1 subtypes B, E, and D IgG-capture enzyme immunoassay (BED-CEIA), which is based on a gradual increase in the proportion of HIV-1-specific IgG to total IgG after seroconversion, and showed similar sensitivity in detecting HIV-1-specific antibodies among multiple HIV-1 subtypes. This assay was introduced to the National AIDS Reference Laboratory (NARL) in the China CDC in 2005; the performance of BED-CEIA in China is excellent . The objectives of the present study were to use the BED-CEIA to detect recent HIV-1 infections among HIV-1-positive IDU in three cities and to estimate local HIV-1 incidence.
Materials and methods
Sample collection and serological testing
Blood samples were collected from HIV confirmatory laboratories of three cities labelled C, D, and E. The study populations included all drug users in HIV surveillance sentinel sites. Two IDU sentinel sites in city C, six IDU sentinel sites in city D, and one IDU sentinel site in city E were included. The data of site A in city C in 2003 were not available because of programmatic disruptions as a result of severe adult respiratory syndrome.
Samples were first screened for HIV-1 antibody using HIV enzyme immunoassay (Beijing Kinghawk, Beijing, China), and then confirmed with a Western blot assay (HIV blot 2.2; Genelab Technologies, Inc., Singapore). Some HIV confirmation tests were carried out by using repeat enzyme-linked immunosorbent assays in city D. All serological tests were performed following the manufacturer's instructions strictly at the three cities.
A total of 17 213 samples were tested for HIV, and 1733 (10.1%) were HIV-1 positive; 1560 (90.1%) were available for BED-CEIA testing (Table 1). We excluded all samples from individuals with AIDS-defining conditions as well as those who were using antiretroviral therapy. HIV-1-positive serum samples were stored at −20°C in the three confirmatory laboratories and refrozen no more than three times.
BED capture enzyme immunoassay
The assay procedure has been described in detail elsewhere [6–10]. In brief, plates coated with goat-antihuman IgG were used to capture both HIV-specific and non-HIV IgG in the test sera. HIV-specific IgG was detected by a branched multisubtype gp41 peptide (BED) labelled with biotin. Incubation with streptavidin-peroxidase followed by tetramethylbenzidine substrate allowed the colorimetric detection of HIV IgG. Appropriate calibrator and negative control, low positive control and high positive control were run in triplicate for every plate. The optical density (OD) values of test specimens were normalized by a ratio using a calibrator (specimen OD/calibrator OD) to minimize interrun variations. Those samples with normalized OD of 1.2 or less were tested again in triplicate and the median values were used for evaluation. Samples with normalized OD of 0.8 or less were considered to be from individuals with recent infection (≤ 155 days). All Calypte HIV-1 BED-CEIA incidence enzyme immunoassay kits were provided by Calypte Biomedical Corporation (Lake Oswego, New York, USA).
The three HIV confirmatory laboratories and NARL participated in a proficiency testing programme for BED incidence test organized by the United States Centers for Disease Control and Prevention (CDC) in Atlanta, Georgia, USA, from 2006. Most of the BED-CEIA tests for this study were performed at NARL, some tests were performed at city C and city D confirmatory laboratories under the direction of the operator from NARL. All the results were accepted only when the tests were valid, or they were repeated.
Calculation of incidence and prevalence
HIV-1 incidence was calculated using the following consensus formula to correct for under or overestimations, as determined by US CDC expert consultants and CDC staff :
Equation (Uncited)Image Tools
where I is the annual HIV-1 incidence, w is the window period. In China, w = 155 days. Ninc is the number of recent HIV infections as determined by BED-CEIA, Nneg is the total number of HIV-seronegative subjects.
The 95% confidence interval (CI) for the incidence estimate is given by the following formula:
Equation (Uncited)Image Tools
where I equals incidence, SQRT of Ninc is the square root of the number of recent HIV infections.
Equation (Uncited)Image Tools
when Npos is the number of HIV-1 infections, N is the total number of the study population.
If there are HIV-1-positive samples missing, the Nneg should be adjusted accordingly.
Descriptive statistics were generated using the statistical analysis system software version 9.1 (SAS, Inc., Cary, North Carolina, USA).
HIV-1 incidence trends in the two IDU sentinel surveillance sites in city C are shown in Table 1 and Fig. 1. There was a significant increase between 2001 and 2002, from 0.57 to 0.93% annualized (P < 0.05). From 2004 to 2006, the HIV-1 incidence remained stable at approximately 1.0% (P > 0.05 comparing 2004, 2005, and 2006, Fig. 1). The HIV-1 prevalence has been stable near 2% at IDU surveillance sentinel site A and near 9% at IDU surveillance sentinel site B in city C.
In city D, the HIV-1 incidence among drug users as estimated by the BED-CEIA assay was 2.1% in 2005; among IDU it was 9.6%. Individuals aged 15 and 35 years accounted for 88% (22/25) of BED-CEIA-identified newly infected cases. Annualized HIV-1 prevalence rates among all drug users and among IDU in city D were 13.4 and 32.0%, respectively.
Among IDU in city E, a slight decreasing trend in estimated annualized HIV-1 incidence rates were observed between 2000 and 2003 (Fig. 2), from 9.2 and 9.8% in 2000 and 2001 to 8.2 and 7.9% in 2002 and 2003. Prevalence among IDU hovered at approximately 20% during this period.
To our knowledge, this is the first application of the BED-CEIA testing strategy to estimate HIV-1 incidence in IDU in China. We have compared the incidence measured from a prospective cohort study of IDU in Xinjiang Uigyur autonomous region with the incidence estimated by adjusted BED-CEIA in a subset of 225 IDU samples from Xinjiang; each yielded comparable results .
The prevalence of HIV-1 infection gives a snapshot of the magnitude of the disease burden in public health; it represents new and pre-existing cases alive on a certain date. Incidence is, however, the more fundamental, timely, and effective marker of the success or failure of programmes aimed at preventing transmission. Incidence rates reflect new cases of a condition diagnosed during a given period of time . The prevalence of HIV is a function of both the incidence of the disease and the survival of the patient.
Table 1 showed that the HIV-1 incidence between IDU surveillance sentinel sites A and B of city C were very close in the past 3 years, although the prevalence differed (site A at 2% and B at 9%). We can also see the HIV-1 incidence is at a relatively steady level in IDU in city C from 2004 to 2006, and the HIV-1 incidence and prevalence showed a similar trend.
The Chinese central government budget for HIV/AIDS prevention and care has increased greatly from 100 million RMB (US$12.5 million) in 2001 to 800 million RMB (US$100 million) in 2005 [2,3]. Mass screening has been carried out among key populations such as IDU, such that many long-standing HIV-1 infections and AIDS cases have been detected in recent years, especially in 2005 . This may explain the sharp increases in HIV-1 prevalence in city C, especially the curve peak of HIV-1 prevalence and incidence in 2005.
In city D, the estimated HIV-1 incidence in general drug users was 2.1% annualized, whereas the incidence in IDU was 9.6%. This incidence is close to the 9.0% measured in a prospective cohort of IDU in Bangkok, Thailand , which is geographically proximate to city D compared with the other Chinese cities. The HIV-1 prevalence in IDU of city D observed in 2005 was lower than that seen in 2003 (46) reported by Hu et al.  and Lu and colleagues , close to that in Bangkok and higher than that in city C. As reported, the first indigenous HIV/AIDS infection cases in China were confirmed among drug users in city D in 1989, near China's southwest border . City D lies along the 500 km-long border with Myanmar, just near the famous golden triangle. A census in January 2005 showed that there were 25 000 drug addicts in city D, 87.3% of whom lived in rural areas, where drug abuse is rampant. Large quantities of China's illegal drugs are distributed through this region. All these factors may lead to the difference between cities C and D in HIV prevalence and incidence.
The HIV-1 incidence was quite similar to IDU in sentinel surveillance sites between city D and city E. There was a decline in HIV-1 incidence, although the prevalence shows an increasing trend. The trends were not, however, statistically significant.
Sharing of needles by IDU remains the primary mode of HIV transmission in all three cities, although a flourishing sex/tourism trade may make heterosexual sex the primary transmission mode within a few years. The government should promptly implement needle exchange and methadone maintenance treatment programmes to control the spread of HIV/AIDS. Other adopted measures, such as behavioural therapy via social marketing and peer education to treat drug addicts and contain HIV/AIDS, are necessary .
BED-CEIA misclassification may occur because some HIV-infected individuals who receive combination antiretroviral therapy that includes a protease inhibitor early in the course of their infection may have a declining antibody concentration , which results in a false-positive test by BED-CEIA. The false-positive results will cause a higher estimated HIV-1 incidence than reality, although this can be modelled and adjusted for. Moreover, it must be kept in mind that the BED-CEIA is still in development; this assay can only be used at the population level in estimating HIV-1 incidence, and should not be used for individual diagnosis . Notwithstanding these important caveats, we have found BED-CEIA-based incidence surveillance to be helpful in identifying those cities at greater or lesser need of intervention among IDU.
The authors would like to thank Professor Bharat Parekh and Dr Xin Liu from the United States Centers for Disease Control and Prevention for their help with the BED-CEIA training and technology instruction in China.
Conflicts of interest: None.
1. Joint United Nations Programme on HIV/AIDS (UNAIDS). 2006 Report on the global AIDS epidemic
. Geneva, Switzerland: UNAIDS; 2006.
2. Ministry of Health, People's Republic of China, Joint United Nations Programme on HIV/AIDS, World Health Organization. 2005 Update on the HIV/AIDS epidemic and response in China
. Beijing: Ministry of Health; 24 January 2006.
3. Wu ZY, Sullivan SG, Wang Y, Rotheram-Borus MJ, Detels R. Evolution of China's response to HIV/AIDS. Lancet 2007; 369:679–690.
4. Parekh BS, McDougal JS. Application of laboratory methods for estimation of HIV-1 incidence. Indian J Med Res 2005; 12:510–518.
5. Janssen RS, Satten GA, Stramer SL. New testing strategy to detect early HIV-1 infection for use in incidence estimates and for clinical and prevention purposes. JAMA 1998; 280:42–48.
6. Parekh BS, Kennedy MS, Dobbs T, Pau CP, Byers R, Green T, et al
. Quantitative detection of increasing HIV type 1 antibodies after seroconversion: A simple assay for detecting recent HIV infection and estimating incidence. AIDS Res Hum Retroviruses 2002; 18:295–307.
7. Wang MJ, Jiang Y, Han M, Li YL, Xiao Y, Feng JG, et al
. An assessment of the performance of BED-CEIA, an assay to detect recent HIV-1 infection. Chin J AIDS STD 2007; 4:305–308.
8. Dobbs T, Kennedy S, Pau CP, McDougal JS, Parekh BS. Performance characteristics of the immunoglobulin G-capture BED enzyme immunoassay, an assay to detect recent human immunodeficiency virus type 1 seroconversion. J Clin Microbiol 2004; 42:2623–2628.
9. Nesheim S, Parekh BS, Sullivan K, Bulterys M, Dobbs T, Lindsay M, et al
. Temporal trends in HIV type 1 incidence among inner-city childbearing women in Atlanta: use of the IgG-Capture BED-Enzyme Immunoassay. J Acquir Immune Defic Syndr 2005; 21:537–544.
10. Xiao Y, Jiang Y, Feng JG, Xu WY, Wang MJ, Vermund SH, et al
. Measuring recent human immunodeficiency virus type 1 infections in China. J Clin Microbiol 2007; 14:1384–1386.
11. Calypte HIV-1 BED Incidence EIA. Enzyme immunoassay for population estimates of HIV-1 incidence
. Calypte Biomedical Corporation; Five Centerpoint Drive, Suite 400, Lake Oswego, OR 97035 2006.
12. Saphonn V, Parekh BS, Dobbs T, Mean C, Bun LH, Ly SP, et al
. Trends of HIV-1 seroincidence among HIV-1 sentinel surveillance groups in Cambodia, 1999-2002. J Acquir Immune Defic Syndr 2005; 39:587–592.
13. Hu DJ, Vanichseni S, Mock PA, Young NL, Dobbs T, Byers R, et al
. HIV type 1 incidence estimates by detection of recent infection from a cross-sectional sampling of injection drug users in Bangkok: use of the IgG capture BED enzyme immunoassay. AIDS Res Hum Retroviruses 2003; 19:727–730.
14. Lu L, Jia MH, Zhang XB, Luo HB, Ma YL, Fu LR, et al
. Analysis for epidemic trend of acquired immunodefieiency syndrome in Yunnan Province of China. Chin J Prev Med 2003; 38:309–312.
15. Rawal BD, Hecht FM, Kahn JO, Chesney MA, Rosenberg ES, Busch MP. Impact of HAART on virus load and anti-HIV-1 antibody in acute HIV-1 infection
. In: 7th Conference on Retroviruses and Opportunistic Infections
. San Francisco, CA, 30 January–2 February 2000 [Abstract 554].
© 2007 Lippincott Williams & Wilkins, Inc.