Recommendations for Prevention and Control of Hepatitis C Virus (HCV) Infection and HCV-Related Chronic Disease
October 16, 1998
HCV infection occurs among persons of all ages, but the highest incidence of acute hepatitis C is found among persons aged 20-39 years, and males predominate slightly (5). African Americans and whites have similar incidence of acute disease; persons of Hispanic ethnicity have higher rates. In the general population, the highest prevalence rates of HCV infection are found among persons aged 30-49 years and among males (3). Unlike the racial/ethnic pattern of acute disease, African Americans have a substantially higher prevalence of HCV infection than do whites (Figure 2).
Prevalence of HCV Infection in Selected Populations in the United States
The greatest variation in prevalence of HCV infection occurs among persons with different risk factors for infection (15) (Table 1). Highest prevalence of infection is found among those with large or repeated direct percutaneous exposures to blood (e.g., injecting-drug users, persons with hemophilia who were treated with clotting factor concentrates produced before 1987, and recipients of transfusions from HCV-positive donors) (12,13,16,17,18,19,20,21,22). Moderate prevalence is found among those with frequent but smaller direct percutaneous exposures (e.g., long-term hemodialysis patients) (23). Lower prevalence is found among those with inapparent percutaneous or mucosal exposures (e.g., persons with evidence of high-risk sexual practices) (24,25,26,27,28) or among those with small, sporadic percutaneous exposures (e.g., health-care workers) (29,30,31,32,33). Lowest prevalence of HCV infection is found among those with no high-risk characteristics (e.g., volunteer blood donors) (34; personal communication, RY Dodd, Ph.D., Head, Transmissible Diseases Department, Holland Laboratory, American Red Cross, Rockville, MD, July 1998). The estimated prevalence of persons with different risk factors and characteristics also varies widely in the U.S. population (Table 1) (3,35,36,37,38,39; CDC, unpublished data).
Most risk factors associated with transmission of HCV in the United States were identified in case-control studies conducted during 1978-1986 (40,41). These risk factors included blood transfusion, injecting-drug use, employment in patient care or clinical laboratory work, exposure to a sex partner or household member who has had a history of hepatitis, exposure to multiple sex partners, and low socioeconomic level. These studies reported no association with military service or exposures resulting from medical, surgical, or dental procedures, tattooing, acupuncture, ear piercing, or foreign travel. If transmission from such exposures does occur, the frequency might be too low to detect.
Transfusions and Transplants. Currently, HCV is rarely transmitted by blood transfusion. During 1985-1990, cases of transfusion-associated non-A, non-B hepatitis declined by >50% because of screening policies that excluded donors with human immunodeficiency virus (HIV) infection and donors with surrogate markers for non-A, non-B hepatitis (5,42). By 1990, risk for transfusion-associated HCV infection was approximately 1.5% per recipient or approximately 0.02% per unit transfused (42). During May 1990, routine testing of donors for evidence of HCV infection was initiated, and during July 1992, more sensitive -- multiantigen -- testing was implemented, reducing further the risk for infection to 0.001% per unit transfused (43).
Receipt of clotting factor concentrates prepared from plasma pools posed a high risk for HCV infection (44) until effective procedures to inactivate viruses, including HCV, were introduced during 1985 (Factor VIII) and 1987 (Factor IX). Persons with hemophilia who were treated with products before inactivation of those products have prevalence rates of HCV infection as high as 90% (20-22). Although plasma derivatives (e.g., albumin and immune globulin [IG] for intramuscular [IM] administration) have not been associated with transmission of HCV infection in the United States, intravenous (IV) IG that was not virally inactivated was the source of one outbreak of hepatitis C during 1993-1994 (45,46). Since December 1994, all IG products -- IV and IM -- commercially available in the United States must undergo an inactivation procedure or be negative for HCV RNA (ribonucleic acid) before release.
Transplantation of organs (e.g., heart, kidney, or liver) from infectious donors to the organ recipient also carried a high risk for transmitting HCV infection before donor screening (47,48). Limited studies of recipients of transplanted tissue have implicated transmission of HCV only from nonirradiated bone tissue of unscreened donors (49,50). As with blood-donor screening, use of anti-HCV-negative organ and tissue donors has virtually eliminated risks for HCV transmission from transplantation.
Injecting and Other Illegal Drug Use. Although the number of cases of acute hepatitis C among injecting-drug users has declined dramatically since 1989, both incidence and prevalence of HCV infection remain high in this group (51,52). Injecting-drug use currently accounts for most HCV transmission in the United States, and has accounted for a substantial proportion of HCV infections during past decades (2,5,53). Many persons with chronic HCV infection might have acquired their infection 20-30 years ago as a result of limited or occasional illegal drug injecting. Injecting-drug use leads to HCV transmission in a manner similar to that for other bloodborne pathogens (i.e., through transfer of HCV-infected blood by sharing syringes and needles either directly or through contamination of drug preparation equipment) (54,55). However, HCV infection is acquired more rapidly after initiation of injecting than other viral infections (i.e., hepatitis B virus [HBV] and HIV), and rates of HCV infection among young injecting-drug users are four times higher than rates of HIV infection (19). After 5 years of injecting, as many as 90% of users are infected with HCV. More rapid acquisition of HCV infection compared with other viral infections among injecting-drug users is likely caused by high prevalence of chronic HCV infection among injecting-drug users, which results in a greater likelihood of exposure to an HCV-infected person.
A study conducted among volunteer blood donors in the United States documented that HCV infection has been independently associated with a history of intranasal cocaine use (56). (The mode of transmission could be through sharing contaminated straws.) Data from NHANES III indicated that 14% of the general population have used cocaine at least once (CDC, unpublished data). Although NHANES III data also indicated that cocaine use was associated with HCV infection, injecting-drug use histories were not ascertained. Among patients with acute hepatitis C identified in CDC's sentinel counties viral hepatitis surveillance system since 1991, intranasal cocaine use in the absence of injecting-drug use was uncommon (2). Thus, at least in the recent past, intranasal cocaine use rarely appears to have contributed to transmission. Until more data are available, whether persons with a history of noninjecting illegal drug use alone (e.g., intranasal cocaine use) are likely to be infected with HCV remains unknown.
Nosocomial and Occupational Exposures. Nosocomial transmission of HCV is possible if infection-control techniques or disinfection procedures are inadequate and contaminated equipment is shared among patients. Although reports from other countries do document nosocomial HCV transmission (57,58,59), such transmission rarely has been reported in the United States (60), other than in chronic hemodialysis settings (61). Prevalence of antibody to HCV (anti-HCV) positivity among chronic hemodialysis patients averages 10%, with some centers reporting rates >60% (23). Both incidence and prevalence studies have documented an association between anti-HCV positivity and increasing years on dialysis, independent of blood transfusion (62,63). These studies, as well as investigations of dialysis-associated outbreaks of hepatitis C (64), indicate that HCV transmission might occur among patients in a hemodialysis center because of incorrect implementation of infection-control practices, particularly sharing of medication vials and supplies (65).
Health-care, emergency medical (e.g., emergency medical technicians and paramedics), and public safety workers (e.g., fire-service, law-enforcement, and correctional facility personnel) who have exposure to blood in the workplace are at risk for being infected with bloodborne pathogens. However, prevalence of HCV infection among health-care workers, including orthopedic, general, and oral surgeons, is no greater than the general population, averaging 1%-2%, and is 10 times lower than that for HBV infection (29-33). In a single study that evaluated risk factors for infection, a history of unintentional needle-stick injury was the only occupational risk factor independently associated with HCV infection (66).
The average incidence of anti-HCV seroconversion after unintentional needle sticks or sharps exposures from an HCV-positive source is 1.8% (range: 0%-7%) (67,68,69,70), with one study reporting that transmission occurred only from hollow-bore needles compared with other sharps (69). A study from Japan reported an incidence of HCV infection of 10% based on detection of HCV RNA by reverse transcriptase polymerase chain reaction (RT-PCR) (70). Although no incidence studies have documented transmission associated with mucous membrane or nonintact skin exposures, transmission of HCV from blood splashes to the conjunctiva have been described (71,72).
The risk for HCV transmission from an infected health-care worker to patients appears to be very low. One published report exists of such transmission during performance of exposure-prone invasive procedures (73). That report, from Spain, described HCV transmission from a cardiothoracic surgeon to five patients, but did not identify factors that might have contributed to transmission. Although factors (e.g., virus titer) might be related to transmission of HCV, no methods exist currently that can reliably determine infectivity, nor do data exist to determine threshold concentration of virus required for transmission.
Percutaneous Exposures in Other Settings. In other countries, HCV infection has been associated with folk medicine practices, tattooing, body piercing, and commercial barbering (74,75,76,77,78,79,80,81). However, in the United States, case-control studies have reported no association between HCV infection and these types of exposures (40,41). In addition, of patients with acute hepatitis C who were identified in CDC's sentinel counties viral hepatitis surveillance system during the past 15 years and who denied a history of injecting-drug use, only 1% reported a history of tattooing or ear piercing, and none reported a history of acupuncture (41; CDC, unpublished data). Among injecting-drug users, frequency of tattooing and ear piercing also was uncommon (3%).
Although any percutaneous exposure has the potential for transferring infectious blood and potentially transmitting bloodborne pathogens (i.e., HBV, HCV, or HIV), no data exist in the United States indicating that persons with exposures to tattooing and body piercing alone are at increased risk for HCV infection. Further studies are needed to determine if these types of exposures and settings in which they occur (e.g., correctional institutions, unregulated commercial establishments), are risk factors for HCV infection in the United States.
Sexual Activity. Case-control studies have reported an association between exposure to a sex contact with a history of hepatitis or exposure to multiple sex partners and acquiring hepatitis C (40,41). In addition, 15%-20% of patients with acute hepatitis C who have been reported to CDC's sentinel counties surveillance system, have a history of sexual exposure in the absence of other risk factors. Two thirds of these have an anti-HCV-positive sex partner, and one third reported >2 partners in the 6 months before illness (2).
In contrast, a low prevalence of HCV infection has been reported by studies of long-term spouses of patients with chronic HCV infection who had no other risk factors for infection. Five of these studies have been conducted in the United States, involving 30-85 partners each, in which average prevalence of HCV infection was 1.5% (range: 0% to 4.4%) (56,82,83,84,85). Among partners of persons with hemophilia coinfected with HCV and HIV, two studies have reported an average prevalence of HCV infection of 3% (83,86). One additional study evaluated potential transmission of HCV between sexually transmitted disease (STD) clinic patients, who denied percutaneous risk factors, and their steady partners (28). Prevalence of HCV infection among male patients with an anti-HCV-positive female partner (7%) was no different than that among males with a negative female partner (8%). However, female patients with an anti-HCV-positive partner were almost fourfold more likely to have HCV infection than females with a negative male partner (10% versus 3%, respectively). These data indicate that, similar to other bloodborne viruses, sexual transmission of HCV from males to females might be more efficient than from females to males.
Among persons with evidence of high-risk sexual practices (e.g., patients attending STD clinics and female prostitutes) who denied a history of injecting-drug use, prevalence of anti-HCV has been found to average 6% (range: 1%-10%) (24-28,87). Specific factors associated with anti-HCV positivity for both heterosexuals and men who have sex with men (MSM) included greater numbers of sex partners, a history of prior STDs, and failure to use a condom. However, the number of partners associated with infection risk varied among studies, ranging from >1 partner in the previous month to >50 in the previous year. In studies of other populations, the number of partners associated with HCV infection also varied, ranging from >2 partners in the 6 months before illness for persons with acute hepatitis C (41), to >5 partners/year for HCV-infected volunteer blood donors (56), to >10 lifetime partners for HCV- infected persons in the general population (3).
Only one study has documented an association between HCV infection and MSM activity (28), and at least in STD clinic settings, the prevalence rate of HCV infection among MSM generally has been similar to that of heterosexuals. Because sexual transmission of bloodborne viruses is recognized to be more efficient among MSM compared with heterosexual men and women, why HCV infection rates are not substantially higher among MSM compared with heterosexuals is unclear. This observation and the low prevalence of HCV infection observed among long-term spouses of persons with chronic HCV infection have raised doubts regarding the importance of sexual activity in transmission of HCV. Unacknowledged percutaneous risk factors (i.e., illegal injecting-drug use) might contribute to increased risk for HCV infection among persons with high-risk sexual practices.
Although considerable inconsistencies exist among studies, data indicate overall that sexual transmission of HCV appears to occur, but that the virus is inefficiently spread through this manner. More data are needed to determine the risk for, and factors related to, transmission of HCV between long-term steady partners as well as among persons with high-risk sexual practices, including whether other STDs promote transmission of HCV by influencing viral load or modifying mucosal barriers.
Household Contact. Case-control studies also have reported an association between nonsexual household contact and acquiring hepatitis C (40,41). The presumed mechanism of transmission is direct or inapparent percutaneous or permucosal exposure to infectious blood or body fluids containing blood. In a recent investigation in the United States, an HCV-infected mother transmitted HCV to her hemophilic child during performance of home infusion therapy, presumably when she had an unintentional needle stick and subsequently used the contaminated needle in the child (88).
Although prevalence of HCV infection among nonsexual household contacts of persons with chronic HCV infection in the United States is unknown, HCV transmission to such contacts is probably uncommon. In studies from other countries of nonsexual household contacts of patients with chronic hepatitis C, average anti-HCV prevalence was 4% (15). Although infected contacts in these studies reported no other commonly recognized risk factors for hepatitis C, most of these studies were done in countries where exposures commonly experienced in the past from contaminated equipment used in traditional and nontraditional medical procedures might have contributed to clustering of HCV infections in families (75,76,79).
Perinatal. The average rate of HCV infection among infants born to HCV-positive, HIV-negative women is 5%-6% (range: 0%-25%), based on detection of anti-HCV and HCV RNA, respectively (89,90,91,92,93,94,95,96,97,98,99,100,101). The average infection rate for infants born to women coinfected with HCV and HIV is higher -- 14% (range: 5%-36%) and 17%, based on detection of anti-HCV and HCV RNA, respectively (90,96,98-101,102,103,104). The only factor consistently found to be associated with transmission has been the presence of HCV RNA in the mother at the time of birth. Although two studies of infants born to HCV-positive, HIV-negative women reported an association with titer of HCV RNA, each study reported a different level of HCV RNA related to transmission (92, 93). Studies of HCV/HIV-coinfected women more consistently have indicated an association between virus titer and transmission of HCV (102).
Data regarding the relationship between delivery mode and HCV transmission are limited and presently indicate no difference in infection rates between infants delivered vaginally compared with cesarean-delivered infants. The transmission of HCV infection through breast milk has not been documented. In the studies that have evaluated breastfeeding in infants born to HCV-infected women, average rate of infection was 4% in both breastfed and bottle-fed infants (95,96,99,100,105,106). Diagnostic criteria for perinatal HCV infection have not been established. Various anti-HCV patterns have been observed in both infected and uninfected infants of anti-HCV-positive mothers. Passively acquired maternal antibody might persist for months, but probably not for >12 months. HCV RNA can be detected as early as 1 to 2 months.
Persons with No Recognized Source for Their Infection. Recent studies have demonstrated that injecting-drug use currently accounts for 60% of HCV transmission in the United States (2). Although the role of sexual activity in transmission of HCV remains unclear, <20% of persons with HCV infection report sexual exposures (i.e., exposure to an infected sexual partner or to multiple partners) in the absence of percutaneous risk factors (2). Other known exposures (occupational, hemodialysis, household, perinatal) together account for approximately 10% of infections. Thus, a potential risk factor can be identified for approximately 90% of persons with HCV infection. In the remaining 10%, no recognized source of infection can be identified, although most persons in this category are associated with low socioeconomic level. Although low socioeconomic level has been associated with several infectious diseases and might be a surrogate for high-risk exposures, its nonspecific nature makes targeting prevention measures difficult.
This article was provided by U.S. Centers for Disease Control and Prevention. It is a part of the publication Morbidity and Mortality Weekly Report. Visit the CDC's website to find out more about their activities, publications and services.