Global epidemiology of hepatitis C virus infection

Summary

Hepatitis C virus (HCV) is a major cause of liver disease worldwide and a potential cause of substantial morbidity and mortality in the future. The complexity and uncertainty related to the geographic distribution of HCV infection and chronic hepatitis C, determination of its associated risk factors, and evaluation of cofactors that accelerate its progression, underscore the difficulties in global prevention and control of HCV. Because there is no vaccine and no post-exposure prophylaxis for HCV, the focus of primary prevention efforts should be safer blood supply in the developing world, safe injection practices in health care and other settings, and decreasing the number of people who initiate injection drug use.

Introduction

Since its discovery in 1989, hepatitis C virus (HCV) has been recognised as a major cause of chronic liver disease worldwide. The most recent WHO estimate of the prevalence of HCV infection is 2%, representing 123 million people.¹ HCV is the leading cause of liver transplantation in developed countries, and the most common chronic bloodborne infection in the USA.

Most descriptions of global HCV epidemiology rely heavily upon HCV seroprevalence studies. These studies are typically cross-sectional in design and are done in select populations—eg, blood donors or patients with chronic liver disease—which are not representative of the community or region in which they reside. Population-based studies representative of an entire community are far more useful, but this kind of study is not feasible in most parts of the world.

Nonetheless, for several years WHO has reported data on the worldwide prevalence of HCV infection, based on both published studies and submitted data (figure). Although HCV is endemic worldwide, there is a large degree of geographic variability in its distribution. Countries with the highest reported prevalence rates are located in Africa and Asia; areas with lower prevalence include the industrialised nations in North America, northern and western Europe, and Australia. Populous nations in the developed world with relatively low rates of HCV seroprevalence include Germany (0·6%),² Canada (0·8%),³ France (1·1%),⁴ and Australia (1·1%).5, 6 Low, but slightly higher seroprevalence rates have been reported in the USA (1·8%),⁷ Japan (1·5–2·3%),8, 9, 10 and Italy (2·2%).¹¹

There is a wide range of prevalence estimates among developing countries, and generally less data available to validate assumptions about the burden of disease than in the developed world. This range in prevalence is reflected in reviewing the estimates from developing countries that are among the world's most populous nations (table).¹⁷ China, whose citizens account for one-fifth of the world's population, has a reported seroprevalence of 3·2%.¹² In India, which holds an additional one-fifth of the world's population, one community-based survey reported an overall rate of 0·9%.¹³ Indonesia's rate is 2·1%, but is based on serosurveys of voluntary blood donors.¹⁴ More thorough data exist on the seroprevalence in Pakistan, where most reported rates range between 2·4% and 6·5%.16, 18, 19, 20 Egypt, with an estimated population of 73 million,²¹ has the highest reported seroprevalence rate, 22%.²²

Although HCV infection has both acute and chronic forms, most of the morbidity associated with infection is realised through the development of chronic liver disease in a subset of infected people years after initial acquisition of the infection. Thus, a major determinant of the future burden of disease is the past and present incidence of infection.²³ However, establishing the incidence of HCV infection is difficult because most infections are initially asymptomatic and available assays do not distinguish acute from chronic or resolved infection. Acute disease reporting systems can underestimate the incidence of HCV infection, even in countries with well-established surveillance systems.²⁴

Because the direct measurement of HCV infection incidence is impractical, researchers have relied upon mathematical models to infer trends in incidence. These undertakings have occurred primarily in developed countries where population-based age-specific seroprevalence data are available, and rely on the assumption that current prevalence reflects the cumulative risk of acquiring infection.

In the USA, the Centers for Disease Control and Prevention (CDC) has modelled trends in HCV incidence using age-specific reported cases of acute disease and data from a cross-sectional national survey done from 1988 to 1994 that provided nationally representative seroprevalence estimates.7, 25 This model revealed a period of low incidence (0–44 per 100 000) before 1965, a transitional period of increasing incidence between 1965 and 1980, and a period of high incidence in the 1980s (100–200 per 100 000).²⁶ A model of HCV burden in France, which used death rates from hepatocellular carcinoma in addition to cross-sectional seroprevalence studies to estimate past incidence, showed a similar trend of increasing incidence through the 1980s.²⁷

An alternate approach to modelling disease burden in Australia showed a steady increase in new HCV infections in that country from 1961 to 2001.⁵ By contrast, the incidence of HCV infection in the USA dropped sharply and steadily through the 1990s, based on data from the CDC's Sentinel Counties Study.²⁸ The rate of new HCV infections also declined in Italy in the 1990s according to analysis of acute disease reporting data.²⁹ Differences in 1990s incidence trends notwithstanding, all published models predict that the incidence of HCV-related sequelae will rise in their respective countries in the coming decades.

The risk factors most frequently cited as accounting for the bulk of HCV transmission worldwide are blood transfusions from unscreened donors, injection drug use, unsafe therapeutic injections, and other health-care-related procedures. Most developed countries have accumulated evidence that the predominant source of new HCV infections within their borders over the past few decades is injection drug use. In the developing world, unsafe therapeutic injections and transfusions are likely to be the major modes of transmission, especially in countries where age-specific seroprevalence rates suggest ongoing increased risk of HCV infection.³⁰ In developed countries with high seroprevalence in older age groups, unsafe therapeutic injections probably had a substantial role in HCV transmission 30–50 years ago, and may persist as an important cause of transmission in isolated, hyperendemic areas.31, 32, 33

Injection drug use is the primary mode of transmission for HCV infection in the developed world. In countries such as the USA and Australia, where the highest seroprevalence is among middle-aged people, injection drug use has been the dominant mode of transmission for more than 30 years, and accounts for 68% and 80% of current infections, respectively.34, 35 The prevalence of HCV infection among long-term injection drug users is high—64–94% among those with a duration of injecting of 6 years or more.36, 37 HCV infection is thought to occur rapidly after initiating injecting behaviour, based on a seroprevalence of 65% observed in the late 1980s among injection drug users with less than 1 year of injecting.³⁸ More recent studies among young injection drug users with 5 years or fewer of injecting have reported HCV seroprevalence rates of 20–46%.39, 40 Fewer sharing partners are necessary to sustain HCV transmission than are necessary for other bloodborne viruses,⁴¹ and indirect drug sharing and preparation practices—eg, backloading (injecting with a syringe filled with drugs that were first mixed or measured in someone else's syringe), and sharing cotton, cooker (containers used to mix and heat drugs), and rinse water—have been associated with HCV transmission.⁴²

Several European countries have also identified injection drug use as the dominant risk factor for HCV infection within their borders. In Norway, 67% of prevalent cases of HCV infection reported a history of injection drug use.⁴³ In Italy, injection drug use was the most commonly reported risk factor among incident cases of acute hepatitis C from 1994 to 1996, and was reported by 60% of patients aged 15–24 years.⁴⁴ In England and Wales, injection drug use was the most commonly reported risk factor for people with positive antibody to HCV (anti-HCV) results tested at seven public-health laboratories over a 3 month period in 1997.⁴⁵ Among anti-HCV-positive voluntary blood donors in France, the most commonly reported risk factor for HCV infection was injection drug use.⁴⁶ Very little data exist regarding the prevalence of injection drug use and its contribution to HCV transmission in the developing world.

In the developed world, the relative contribution of health-care-related transmission of HCV infection to overall HCV infection transmission is difficult to quantify, but likely small, despite numerous recent outbreaks stemming from lapses in aseptic techniques and infection control practices.47, 48, 49, 50 However, in many developing countries, supplies of sterile syringes may be inadequate or non-existent, non-professionals often give injections outside the medical setting, and injections are often given to deliver medications that could otherwise be delivered by the oral route.⁵¹ In this environment people may receive multiple contaminated injections over the course of a lifetime, incurring a substantial cumulative risk of HCV infection.

Contaminated injection equipment appears to be the major risk factor for HCV infection in several countries, including several of the most populous nations in the world. In Egypt, the country with the highest reported seroprevalence in the world, transmission has been attributed to contaminated glass syringes used in nationwide schistosomiasis treatment campaigns from 1960 to 1987.²² In India, seroprevalence of HCV infection among patients receiving multiple injections to treat kala-azar was 31·1%—well above the seroprevalence among hospitalised and community controls.⁵² Two 2003 studies among populations in different regions of India found substantial associations between prevalent HCV infection and frequent visits to “freelance” or unlicensed practitioners of medicine, as well as a history of therapeutic injections using reusable syringes.13, 53 Similarly, a case control study in a community in Pakistan found that HCV-infected cases were more likely to report five or more injections per year from a health-care provider in the past 10 years than were controls.²⁰ In Taiwan, a study involving consecutive anti-HCV-positive patients at a medical practice in a rural agricultural community showed that anti-HCV-positive patients were substantially more likely to report receiving frequent medical injections (six injections per year for the past 2 years) and visiting “freelance” practitioners (vs doctors, pharmacists, and non-medical staff under physician supervision) than were consecutive anti-HCV-negative patient controls from the same practice.⁵⁴ Therapeutic injections were similarly associated with HCV infection in prevalence studies among both paediatric and elderly Taiwanese populations.55, 56

Prompted by evidence for ongoing transmission of HCV and other bloodborne viruses via unsafe therapeutic injection practices, WHO has coordinated the Safe Injection Global Network (SIGN), a coalition of governments, international health agencies, corporations, and individuals that advocate for safer injection practices worldwide.⁵⁷ WHO has sponsored assessments of injection practices in countries suspected of having excessive health-care-related transmission of bloodborne viruses.⁵⁸ WHO models estimate that unsafe injections accounted for 2 million new HCV infections in 2000.⁵¹ As part of the 2000 update of WHO's global burden of disease study,⁵⁹ WHO and collaborating epidemiologists estimated the global burden of disease attributable to contaminated injections in the health-care setting.⁶⁰ They found the highest reported rates of needle reuse in the middle east, southeast Asia, and the western Pacific. In most countries in these areas, studies with the power to examine potential associations between needle reuse and prevalent HCV infection have not been done.

Blood transfusion is a highly effective means of transmitting HCV infection. In most of the developed world, numerous measures over the past four decades have resulted in progressive reductions in the risk of transfusion-transmitted HCV infection. These measures include adoption of an all-volunteer donor system, screening of blood donations with surrogate laboratory tests for liver disease (eg, alanine aminotransferase), screening of potential donors based upon answers to questions related to HIV risk factors, anti-HCV testing, and HCV nucleic acid testing. Blood is now so safe in many developed countries that classic methods to measure risk are no longer sensitive enough to provide meaningful estimates or document transfusion-related transmission events.⁶¹

The largest reductions in the incidence of transfusion-transmitted HCV infection have coincided with adoption of an all-volunteer donor system. In the USA, a more than threefold drop in the incidence of post-transfusion non-A, non-B hepatitis was observed in one veterans' hospital after the proportion of paid donor blood used for transfusions was reduced from 91% to 4%.⁶² Because most blood donations in the developing world do not come from voluntary, non-remunerated donors,63, 64 transfusion is probably a major source of HCV transmission throughout the developing world, much as it was in the developed world decades ago. The obstacles to creation of a nationwide system of all-volunteer blood donors in the developing world are complex and vary widely. The widespread use of paid donor blood in China has been ascribed to cultural beliefs incompatible with blood donation and inadequate efforts to recruit volunteer blood donors.⁶⁵ In India, some observers have suggested that problems with regulatory oversight of the nation's blood transfusion service led to insufficient use of volunteer-donated blood.⁶⁶ In Kenya, government hospitals outside Nairobi are responsible for their own blood donor recruitment, blood collection, and testing; budget shortfalls commonly lead to the use of family/replacement donors.⁶⁷ A study of blood banking practices in countries throughout North and South America suggested a correlation between per capita gross national product (GNP) and percentage of blood donations coming from voluntary, non-remunerated donors. Of 28 countries with per capita GNP less than US$5000, only Cuba had greater than 90% of its blood donations from voluntary, non-remunerated donors.⁶⁸

Most countries in the developing world do not screen blood donations for the presence of HCV. WHO's Global Database on Blood Safety estimates that 43% of donated blood in the developing world is not screened adequately for transfusion-transmitted infections, including HCV.⁶⁴ A review of transfusion safety in 12 Latin American countries found that half screened all blood products for HCV.⁶⁹ In India, HCV screening of blood products is mandated by law but not usually done due to financial constraints.⁶⁶ In New Delhi, among 182 anti-HCV-negative hospitalised patients studied prospectively following a blood transfusion, HCV infection developed in 5·4%.⁷⁰ In Ghana, one in 2578 donations is estimated to contain HCV.⁷¹

Transmission of HCV infection through occupational, perinatal, and sexual exposures occurs with much less efficiency compared with transmission through large or repeated percutaneous exposures. Thus, occupational, perinatal, and sexual transmission are unlikely to be major sources of new HCV infections, regardless of the population or geographic area. Occupational transmission of HCV infection is largely confined to health-care workers who have sustained a contaminated needlestick injury, and observed attack rates under these circumstances are as low as 0·3%.11, 72, 73 Acquisition of HCV infection through perinatal transmission is estimated to occur in 2·7–8·4% of infants born to HCV-infected mothers, and a higher proportion of infants born to HIV/HCV coinfected mothers.74, 75, 76, 77 Sex with an infected partner and with multiple partners have been identified as risk factors for HCV transmission,⁷⁸ but sexual transmission of HCV is far less efficient than that of other sexually transmitted viruses. Among people in long-term monogamous relationships in particular, the risk of sexual transmission of HCV is extremely low.79, 80 There are no published data sufficient to show whether specific sexually transmitted coinfections or particular sexual practices increase the likelihood of sexual transmission of HCV.

Because of the wide variety of human activities that involve the potential for percutaneous exposure to blood or blood-derived body fluids, there are many biologically plausible modes of transmission besides those with clearly demonstrated epidemiological associations with infection. These modes of transmission include cosmetic procedures and religious or cultural practices such as tattooing, body-piercing, commercial barbering, ritual scarification, circumcision, acupuncture, and cupping.

In most regions of the world, there are insufficient data to determine whether these risk factors make any measurable contribution to overall HCV transmission. In those countries where adequate studies have been done, none of these activities have been consistently associated with HCV transmission. Case control studies of acute hepatitis C in the USA failed to find a significant association with tattooing, ear piercing, or acupuncture.78, 81 Cross-sectional prevalence studies among blood donors in the UK and Australia found significant associations between anti-HCV seropositivity and a history of tattooing (p<0·00001), but not with ear piercing or acupuncture.82, 83 A community-based cross-sectional seroprevalence study in Taiwan found a significant association with acupuncture (p<0·05), but not with tattooing.⁸⁴