Prevalence and epidemiology of chronic hepatitis C among prisoners of Mato Grosso do Sul State , Brazil

In Mato Grosso do Sul state, Brazil, the number of prisoners has increased in the recent years and the control of hepatitis C virus (HCV) has become more complex. The aim of the present study was to estimate the prevalence and identify the genotypes of HCV in prisoners as well as the factors associated with this infectious disease. Thereby, 443 men and 243 women from prisons were interviewed and subjected to blood collection. Anti-HCV reactive samples were analyzed by RT-PCR and genotyped. The overall seroprevalence of HCV infection was 4.8% (95%CI: 3.4 to 6.8%). Furthermore, the prevalence was higher in: men, injecting drug users, tattooed persons, those who were more than 50 years old, individuals who have been arrested multiple times, people with previous history of sexually transmitted disease (STD), persons who received blood transfusions or those with HIV/AIDS. The prevalence of RNA HCV by PCR was 3.0% (95%CI: 1.7 to 4.2%). Moreover, the coinfection of HIV and HCV was 33.3%. In addition, genotype 1 was the most frequent (85%) followed by genotype 3 (15%). The screening strategy for HCV and other infectious diseases in inmates is important as it establishes an early diagnosis, opportunity for treatment and allows the breaking of the transmission chain.


INTRODUCTION
Hepatitis C virus (HCV) is responsible for the infection of approximately 170 million people worldwide.Incarcerated individuals are particularly affected since they are at higher risk of being infected due to tattooing, piercing and use of injecting drugs (1)(2)(3)(4)(5)(6)(7)(8).The living conditions of these people outside and inside prisons contribute to the transmission of infectious diseases.Their lifestyles, combined with worsened drug abuse inside prisons, reflect precarious levels of health care compared to the general population (3,4).
HCV is classified into six genotypes and multiple subtypes.This variability has epidemiologic importance and is directly related to diagnosis as well as therapeutic response (20,21).
The aim of the current study was to estimate the prevalence of HCV as well as to identify the factors associated with the infection among prison inmates, therefore determining the circulating genotypes of HCV.

PATIENTS AND METHODS
In 2009, from a total of 3,418 men and 395 women, a non-probability sample of 686 prisoners was obtained by using an estimated prevalence of 9.6% (± 2.6) for men and 3.6% (± 2.2) for women.It was difficult to have access to all prison pavilions and cells due to security reasons.The total sample size included 443 men and 243 women, with a significance level of 5%.All individuals were interviewed and their blood samples were collected.The samples were tested by enzyme linked immunosorbent assay (ELISA) and by the immunodot method (rapid test), for the detection of an anti-HCV marker (Bioelisa HCV 4.0®, BioKit, Spain and ImmunoComb II HCV® kit, Orgenics, Germany) and anti-HIV (ELISA and Western blot).
Weakly reactive samples to anti-HCV (OD/cutoff < 3.0) were tested again by line immunoassay (INNO-LIA® HCV Ab III, Innogenetics, Belgium) and anti-HCV positive ones were tested for viral RNA detection by reverse transcriptionpolymerase chain reaction (RT-PCR) with primers complementary to the conserved area of the 5' NC region of HCV, essentially as previously described (32).Positive samples were also submitted to genotyping method by line probe assay (INNO-LiPA®, Innogenetics, Belgium) using biotinylated primers complementary to the 5' NC region of HCV genome.To confirm HCV genotypes and subtypes, positive samples were amplified with primers complementary to the region NS5B of HCV under the same conditions as described by Sandres-Sauné et al. (33).Chi-square, chi-square for trend and Fisher's exact tests were employed for statistical analyses and the prevalence ratios were calculated, with a 95% confidence interval.In order to estimate adjusted prevalence ratios, the Cox regression was chosen (time interval equal to one unit), and variables of significance less than 5% were selected.The present study was approved by the Research Ethics Committee of Federal University of Mato Grosso do Sul.
HCV was more prevalent in men, injecting drug users, tattooed individuals or people who had been subjected to blood transfusion as well as individuals coinfected with HIV.Higher prevalence in persons aged 50 years or more, inmates with multiple arrests and with history of sexually transmitted diseases (STDs was significant in the bivariate analysis (Tables 1, 2  and 3).
Only 23 prisoners reported having undergone acupuncture and all of them were anti-HCV negative.History of hemodialysis was reported by two prisoners, none with hepatitis C. Piercing was reported by only two individuals in the anti-HCV positive group.

DISCUSSION
This investigation is the first study on the epidemiology of HCV infection in Campo Grande, Mato Grosso do Sul State, and has important public health implications.The prevalence of HCV found among prisoners was higher than that observed in the general population (1.42-1.5%)(34,35).
When compared with international studies, the prevalence of HCV infection among inmates No information -11 100.0 -PR: prevalence ratio; if p ≤ 0.05, there is significant statistical difference between values; blank information was excluded; (1) chisquare test; (2) chi-square for trend test.PR: prevalence ratio; if p ≤ 0.05, there is significant statistical difference between values; blank information was excluded; (1) chisquare test; (2) Fisher´s exact test. in Campo Grande was lower than those found in northwestern Spain (47.9%),Italy (38%), Canada (26.1%), United States (23.1%), and Ghana (18.7%) (9,11,(36)(37)(38).On the other hand, it was similar to values found in Iran (8.1%), France (4.9%), and higher than in Venezuela (1.5%) (10, 13, 39).The higher prevalence among men in Brazil differs from data from Canada, where women had a greater exposure to injecting drugs (37).In the female incarcerated population of Campo Grande, the prevalence of HCV infection is lower than in Butantã Prison in São Paulo (16.2%) (17).A study carried out in the Central-West region of Brazil found a prevalence of 6.9% for anti-HCV (95%CI: 5.2-9.2%)among drug users, which is similar to the present findings (40).The frequent use of injecting drugs by women was also observed in Thailand (41).
Low education level, lack of knowledge about hepatitis C, multiple arrests and long periods in prison have been identified by other studies as factors associated with higher occurrences of the infection (9-11, 16, 17, 40).The current study found that less than half of the studied population had steady partners, few individuals used condoms on a regular basis and approximately 1/5 of the studied subjects had a history of homosexual relationship.Although hepatitis C presents low sexual transmission rate, all these factors are connected with high exposure to other STDs (9-11, 16, 19, 39).
The prevalence of HCV was significantly higher among individuals who received blood transfusions and derivates as described in Venezuela in a period prior to 1994 (13).
Tattooing was significantly higher in anti-HCV positive groups, which agrees with previous research (10,11,18).Despite these other factors, the highest prevalence of hepatitis C in prisons is related to injecting drug use, which was clear in both the bivariate and multivariate analyses, as described in other studies (7,13,16,18,(36)(37)(38)(39).The prevalence of anti-HCV was higher among older individuals, which was most probably due to risky behaviors throughout life, a fact observed in Campo Grande and in other incarcerated populations and groups of injecting drug users (11,37,40).The use of injecting drugs and tattooing was not significantly higher among blood donors with hepatitis C, probably due to the peculiarities of this population (42).
In the present study, from a total of 29 subjects who underwent testing for HCV RNA, 20 were positive, showing a viremia of 69%, which was lower than values found in injecting drug users in the Central-West region of Brazil (85.4%) and in individuals from Belo Horizonte, Minas Gerais state (98.6%) (40,43).Such finding agrees with the rate of viremia found in the Northeast region of the country (65.4%) (30).
Genotypes 1 and 3 are more prevalent in Brazil (24-31, 40, 43).The high frequency of genotype 1 and its subtypes was reported among injecting drug users, blood donors, hemophilic persons and patients subjected to hemodialisys (26,40,43).In the current work, the subtype 1a was more common, similarly to other studies in the Central-West region, which differs from studies with equivalent subtypes 1a and 1b or predominance of 1b (27, 29-31, 40, 43).The finding of mixed genotype 1a/b (5%) has been described and may indicate a re-infection by different genotypes (27,30,43).
The rate of co-infection of HIV and HCV was high (33.3%)and can be attributed to common factors associated with the risks of these types of infections, such as injecting drug behavior and tattooing (36,37).Among co-infected patients, only HCV genotype 1 was isolated in the present study, which differs from another work that reported the association of genotype 3 with coinfection and injecting drug users (44).
Finally, the current findings demonstrate that early diagnosis, prevention programs and therapeutic interventions are necessary in order to minimize risks involved in the epidemic spread of HCV inside prison systems.

Figure 1 .
Figure 1.Distribution of blood samples obtained from prisoners according to HCV genotypes in Campo Grande, Mato Grosso do Sul state, Brazil, 2010 (n = 20).

Table 1 .
Distribution of data from prisoners according to factors associated with HCV infection in Campo Grande, Mato Grosso do Sul, Brazil in 2010 (n = 686)

Table 2 .
Distribution of data from prisoners according to other factors associated with HCV, Campo Grande, Mato Grosso do Sul, Brazil in 2010 (n = 686)

Table 3 .
Variables analyzed by multivariate analysis on the prevalence of HCV among inmates in Campo Grande, Mato Grosso do Sul, Brazil in 2010 Cox regression: if p ≤ 0.05, then there is significant statistical difference between variables categories.