|
|
||||||||
| ABSTRACT |
|
|
|---|
| INTRODUCTION |
|
|
|---|
In previous studies of CSWs, sexual transmission of HCV was proposed to play an important role in infection, and, although it has not been evaluated in CSWs who are not intravenous drug users (IDUs), evidence suggested that sexual transmission may be the second most common risk factor for HCV infection.12 However, whether sexual transmission is an independent risk factor for HCV transmission is debatable because of the frequently confounding effect of concomitant intravenous drug use, which is one of the most important risk factors for HCV infection.6,12
We performed a cross-sectional seroepidemiologic study to estimate the prevalence of HCV infection and its potential risk factors among female CSWs who were not IDUs and who were seronegative for human immunodeficiency virus (HIV). We also evaluated the independent contribution of sexual activity to HCV transmission.
| MATERIALS AND METHODS |
|
|
|---|
Serologic testing. Blood samples were collected into jelly vacuum tubes under aseptic conditions and were subsequently tested for syphilis and HIV at each health center laboratory. The tubes were sent to the Central Laboratory, Jeollanam-do Institute of Health and Environment on the same day. Sera were isolated by centrifugation and stored at 70°C. The sera were tested for HCV and hepatitis B virus (HBV) in the same laboratory. Third-generation enzyme-linked immunosorbent assays (ELISAs) were performed to detect antibodies to HCV (Genedia HCV ELISA 3.0; Green Cross Sang-A Corp., Chungcheongbuk-do, South Korea) and hepatitis B virus surface antigen (Genedia hepatitis B surface antigen [HBsAg] ELISA 3.0; Green Cross Sang-A Corp.) using an ELISA plate processor (Behring ELISA Processor III; Dade Behring, Marburg, Germany). Third-generation tests for antibodies to HCV contain antigens from HCV core protein and non-structural (NS) 3, NS4, and NS5 gene proteins. Previous infection with HIV was determined by ELISA (Genedia HIV 1/2 ELISA 3.0; Green Cross Sang-A Corp.) using an ELISA plate processor (Plato 3300; Rosys, Hombrechtikon, Switzerland) and confirmed by Western blotting (HIV blot 2.2; Genelabs Diagnostics, Singapore). Syphilis serologic status was screened by the rapid plasmid reagin (RPR) test or the Venereal Disease Research Laboratory (VDRL) test and was confirmed by the Treponema pallidum hemagglutination assay (TPHA).
Definitions. To determine the serologic status of syphilis and HIV, we traced the laboratory data of each public health center and the registered data of the National Communicable Diseases Surveillance System in Korea, 20022004. Subjects with reactive TPHA and previous positive VDRL or RPR test results were defined as syphilis seropositive. The presence of diabetes mellitus was confirmed by a self-report that the subject was taking anti-hyperglycemic medication because almost none of the study subjects had been fasting. Subjects who had been exposed to potential risk factors, such as acupuncture, on more than one occasion were considered positive for that risk factor. A family history of hepatitis C was defined as the presence of HCV in an immediate family member (e.g., parent or full siblings). A non-paying sexual partner was defined as a spouse or male cohabitant.
Statistical methods. Continuous variables, such as age and years spent as a CSW, were grouped and treated as categorical variables. Comparisons for the associations of HCV status were conducted using the chi-square test, Students t-test, or Fishers exact test as appropriate. The analysis of risk factors was conducted using multivariate logistic regression. Previously identified risk factors that were significant at the P < 0.1 level in an unadjusted model (age group, acupuncture, diabetes, time spent as a CSW, and history of hemodialysis) were included in the model. Data processing and statistical analyses were performed with SPSS version 11.5 (SPSS Inc., Chicago, IL).
| RESULTS |
|
|
|---|
5 years, and the remaining 8.8% had been CSWs for > 5 years; 966 subjects (63.3%) worked in seaside cities and 561 (36.7%) worked in inland cities. Most of the subjects (86.6%) reported a history of sexual contact with non-paying partners (husband or cohabitant). A family history of HCV or HBV was infrequent and was reported by only 21 (1.4%) or 45 (2.9%) of the subjects, respectively.
|
|
Independent risk factors for HCV infection.
Using a multivariate logistic regression model, we analyzed all factors associated at the P < 0.1 level with HCV serologic status. There were too few subjects with some of the traditional risk factors, such as blood transfusion and HBsAg carrier status, to perform a multivariate analysis for those factors. In the multivariate logistic model, HCV was significantly associated with diabetes (odds ratio [OR] = 11.2, 95% confidence interval [CI] = 2.6347.8, P = 0.001) and acupuncture (OR = 3.30; 95% CI = 1.169.34, P = 0.024), whereas time spent as a CSW (
3 years) did not reach statistical significance (OR = 2.21, 95% CI = 0.905.41, P = 0.08). A history of hemodialysis and age (
30 years) were not associated with HCV serologic status (Table 3
).
|
| DISCUSSION |
|
|
|---|
Women who work in the commercial sex industry differ from the general population in their increased vulnerability to infection and their potential for increased rates of STDs and blood-borne virus infections, including HCV. CSWs play a specific role as a bridge population between reservoirs of infectious disease and the general populations; therefore, they require intensive health care. Several studies have reported a high prevalence of HCV infection in CSWs. Kim and others reported that HCV prevalence in CSWs in South Korea was 10.3%, and this rate was three times higher than that of voluntary recipients for STD testing in public health centers.11 Despite a small sample size and selection bias, these findings suggest the possibility of sexual transmission of HCV in South Korea. Terraut estimated the average seroprevalence of antibodies to HCV among female sex workers to be at 6% in the United States;6 Laurent and others reported a seroprevalence of 6.6% in the Congo,13 and Ishi and others reported a seroprevalence of 3.2% in Japan.14 The HCV seroprevalence of 1.4% in female CSWs in our study is lower than that in most previous reports except another for a report from Japan,15 and is lower than that in the general female population in South Korea.16 In addition, CSWs in other countries show a higher seroprevalence than in our study,8,13,17 despite some differences in the study designs. We suggest that this difference in HCV seroprevalence is attributable to differences in the pattern of intravenous drug use in the study populations. Illicit intravenous drug use in CWSs ranged form 3%18 to 9.1%17 in previous studies and was 4.4% in low-income, young American women.8 Our sample consisted of a large number of female CSWs, but none were current or prior IDUs. IDUs are very rare in South Korea even among CSWs because illicit drug use is strictly banned in South Korea. Montella and others suggested that illicit intravenous drug use is a major transmission route for HCV infection in younger people and that iatrogenic transmission via non-sterile syringes is a major route of transmission in older age groups.19 Given that three of four of our study subjects were less than 30 years of age, it is likely that the absence of IDUs accounts for the lower HCV seroprevalence in our study compared with that in other studies.
The 6.2% prevalence of HBsAg carriers among female CSWs in our study is higher than that reported in previous studies of female CSWs in Japan,14 Uruguay,18 and South Korea,11 and in male CSWs in Australia.17 It is also higher than that in the general female population of South Korea,1,3,20 in which prevalence ranges from 1.4% to 5%. This may be because South Korea is endemic for HBV and sexual transmission of HBV is well established, unlike that of HCV. The HBV vaccination coverage rate in our study subjects was low, which may be related to the relatively high fee that must be paid by an applicant in the Korean vaccination system.
HIV was not detected in our study subjects. The total number of patients with HIV/acquired immunodeficiency syndrome (AIDS) in South Korea at the end of 2003 was approximately estimated 8,300, according to the World Health Organization,21 HIV/AIDS occurred predominantly in men,22 and the prevalence rate in CSWs was 1.75 per 10,000 person. Therefore, HIV was not likely to be detected in a study of our size.
Our results suggest that acupuncture is an important risk factor for HCV infection, as reported in previous studies in Asia.2325 Acupuncture is a common health practice in South Korea, as reflected in our study subjects. Despite the frequent use of acupuncture, it is conducted in non-medical areas and commonly by unqualified persons with non-sterile equipment. Therefore, acupuncture should be considered a mode of HCV transmission in South Korea3,23 and the rest of the world.24,2629 Shin and others reported that 55% of female HCV transmission is attributable to acupuncture and suggested that safer acupuncture practices should be a priority for HCV prevention in South Korea.3 Shimoyama and others reported that acupuncture accounts for
20% of HCV cases in Japan.25 Of the 21 subjects positive for antibodies to HCV in the present study, only eight reported a prior surgery or hemodialysis, and their only other likely risk factor for HCV transmission, apart from being CSWs, was the use of acupuncture. In addition, a strong association between the presence of antibodies to HCV and acupuncture remained in our regression model after adjusting for other factors. These results suggest that acupuncture plays a major role in HCV transmission among female CSWs who are not IDUs in South Korea, as in other Asian countries.
None of the variables associated with sexual activity, such as time spent as a CSW, syphilis, or HBV serologies, and non-paying sexual contact with HCV-positive partners, that were analyzed in the present study were associated with the presence of antibodies to HCV. Time spent as a CSW was shown to be strongly associated with the presence of antibodies to HCV by univariate analysis, but it did not remain significantly associated by multivariate analysis. Only 2 of 21 subjects positive for antibodies to HCV were free from potential risk factor exposures except those related to sexual activity. Together with the relatively low prevalence of antibodies to HCV previously described, these results indicate that HCV was transmitted mainly by the more common route, such as acupuncture, rather than by sexual transmission, which appears to play a minor role even in CSWs. However, the possibility of sexual transmission is not ruled out by our study because indicators that may more directly gauge sexual activity, such as condom use or number of daily sexual partners, were not analyzed. Although sexual activity seems to have a minimal effect on HCV transmission, the frequency of sexual activity in most adults necessitates further consideration of the potential for sexual transmission of HCV.
An interesting finding in our study was the positive association between diabetes and positivity for antibodies to HCV, which is consistent with the findings of most studies of general populations30,31 and of patients with diabetes or HCV,32 with the exception of one study from Greece.33 However, advanced liver disease, such as cirrhosis and HCC, has been shown to cause glucose intolerance and insulin resistance, so it is still not clear whether mild or severe HCV infection is actually associated with diabetes. In the present study, diabetes was the factor most strongly associated with HCV positivity. Furthermore, none of the subjects positive for antibodies to HCV in our study had been clinically diagnosed with liver disease including cirrhosis and HCC. Thus, despite the small number of subjects with diabetes and the possibility of an inaccurate definition of diabetes, diabetes may still be a risk factor for HCV infection. In addition, none of the persons with diabetes had used needles for insulin injection.
The transfusion of blood or blood products carries an extremely high risk of HCV infection worldwide. Therefore, patients with hemophilia have a very high incidence of HCV infection (4690%).3436 Hemodialysis is also an effective transmission route for HCV. Recent epidemiologic data show that the risk of HCV infection among patients undergoing dialysis and staff at dialysis units has increased.37,38 However, in our study, there were too few subjects with a history of transfusion or hemodialysis to establish an association with the presence of antibodies to HCV.
Our study has some limitations. First, direct indicators of sexual activities, such as condom use and daily numbers of sex partners, and some potential risk factors, such as intranasal drug use, tattooing, and body piercing, were not used in the analysis because of insufficient data. Second, CSWs can be classified into two groups that show differences in behaviors and sexual activities: 1) direct CSWs who live in brothels in red-light areas, and 2) indirect CSWs who are nonbrothel-based. However, it was not possible to classify our study subjects into these two groups. Third, the subjects recruited for this study may not have been representative of the female CSWs in the target area; unregistered CSWs who were not recruited may be direct CSWs with more risky behavior than registered CSWs. However, it is expected that there were only a small number of unregistered direct CSWs and that these CSWs would soon need to become registered to remain CSWs. Therefore, we believe that the effects of this selection bias and its implications could be ignored. Fourth, a misclassification bias concerning the definition of diabetes and under-reporting of intravenous drug use may have occurred. Lastly, our study has low statistical power because of the small number of positive results for antibodies to HCV, diabetes, and syphilis serologic status.
Our findings suggest that the prevalence of antibodies to HCV in female CSWs without HIV or a history of intravenous drug use was no higher than that in the general population of South Korea or that previously reported from other countries. The higher prevalence rates reported in other studies may be related to the prevalence of intravenous drug use in study populations. Although we cannot rule out the role of sexual activity as an independent risk factor for HCV infection, we suggest that sexual activity is less of a risk factor than are acupuncture or diabetes in female CSWs who are not IDUs. This remains to be quantified in a prospective study.
Received November 2, 2005. Accepted for publication January 11, 2006.
Financial support: This work was supported by the Jeollanam-do Institute of Health and Environment and the Korean Center for Disease Control and Prevention.
* Address correspondence to Sun-Seog Kweon, Department of Preventive Medicine, Seonam University College of Medicine, 720, Gwanchi-dong, Namwon, Jeonbuk 590-711, South Korea. E-mail: sskweonx{at}netian.com ![]()
Authors addresses: Sun-Seog Kweon and Min-Ho Shin, Department of Preventive Medicine, Seonam University College of Medicine, 720, Gwanchi-dong, Namwon, Jeonbuk 590-711, South Korea, Telephone: 82-63-620-0373, Fax: 82-63-620-0375, E-mails: sskweonx{at}netian.com and mhshinx{at}orgio.net. Hyeon-Je Song and Doo-Young Jeon, Microbiology Division, Jeollanam-do Institute of Health and Environment, 291-1, Nongsung-dong, Gwangju 502-200, South Korea, Telephone: 82-62-360-5338, Fax: 82-62-366-7413, E-mails: songha1{at}dreamwiz.com and jeondy{at}hanmail.net. Jin-Su Choi, Department of Preventive Medicine, Chonnam National University College of Medicine, Chonnam National University Research Center of Medical Sciences, 5, Hak-dong, Dong-gu, Gwangju 501-746, South Korea, Telephone: 82-62-220-4172, Fax: 82-62-233-0305, E-mail: jschoix{at}chonnam.ac.kr.
| REFERENCES |
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |