Appendix

Definitions

This section summarizes terminology describing the tests used to identify patients with HCV infection, the results of these tests (Appendix Table 1), and the response to treatment. The Centers for Disease Control and Prevention (CDC) has recently published detailed guidelines for performing laboratory testing and reporting results of anti-HCV and supplemental testing.⁹⁰

Enzyme-linked immunoassay (ELISA) or enzyme immunoassay (EIA): The ELISA (also referred to as EIA) detects antibodies against recombinant HCV antigens. "First generation" ELISAs used a single antigen; later tests added additional antigens.^146-148 Second and third-generation tests are both in standard use. Because of concerns about false-positive tests, particularly in low-prevalence populations (such as blood donors or asymptomatic adults), the CDC has recommended confirming positive ELISA results with a supplemental test (recombinant immunoblot assay or polymerase chain reaction), unless the signal-to-cut-off ratio is above a predetermined threshold that has been shown to confirm positive more than 95% of the time.⁹⁰

ELISA is the least expensive diagnostic test for HCV infection, with an average charge of about $60.00.³⁴

Recombinant immunoblot assay (RIBA): RIBA is a supplemental test that also detects antibodies against HCV antigens. In these assays, multiple HCV antigens are individually displayed on a nitrocellulose strip as bands. Positive RIBA results have at least 2 reactive bands; indeterminate results have 1 reactive band. Because positive RIBA results require reactivity to more than 1 HCV antigen, they are considered more specific (but not more sensitive) than ELISA for past HCV infection, and are used to confirm positive ELISA results in low-prevalence populations.³⁴ However, RIBA is not an independent gold standard for ELISA because the 2 tests use similar antigens to detect anti-HCV antibodies.

Currently available third-generation RIBA are thought to be more specific than earlier-generation tests because they produce fewer indeterminate results.¹⁴⁹ The interpretation of indeterminate RIBA results remains uncertain.^90,150,151 The relative proportion of RIBA-positive, RIBA-indeterminate, and RIBA-negative test results in patients with positive results on ELISA varies according to the patients studied.

The RIBA is typically 2 to 3 times more expensive than ELISA; usual charges are about $140.00.³⁴

HCV core antigen testing: Recently developed tests to detect HCV core antigen may aid in diagnosing acute infection in the "window period" before HCV antibodies develop.^152,153 The role of HCV core antigen testing in screening has not yet been established.

Reverse transcription polymerase chain reaction (RT-PCR or PCR): This is a laboratory method used to detect circulating HCV RNA in blood. A PCR can be quantitative or qualitative, and under optimal conditions qualitative PCR can detect 100 international units (IU)/mL or less of circulating virus.^34,116 Because the absence of viremia in patients who test positive for anti-HCV-antibodies is associated with little or no risk for HCV infection¹⁵⁴ or complications related to chronic HCV infection, sustained PCR-detected viremia has become the gold standard for chronic HCV infection.^{6,132,155-157} In patients who have positive results on PCR, the degree of viremia correlates poorly with degree of liver damage,^157-161 although these results may help predict the likelihood of response to treatment.¹⁶²

Strict quality control is necessary for PCR testing to be reliable. False negative test results can occur because some patients with active infection have intermittent viremia, and a small portion of patients with chronic HCV infection can become non-viremic, particularly if they develop hepatocellular cancer.^163-165 For this reason, repeat PCR resting is suggested in high-risk patients who are positive for anti-HCV antibodies but have negative results on initial PCR. False-positive PCR test results may also occur due to contamination of samples (11% in 1 early quality control study) but appear to be much less frequent since standardization of assay techniques.¹⁶⁴

PCR testing is associated with charges of about $130.00 for a qualitative test and $200.00 for a quantitative test.³⁴

False-positive ELISA results: Patients who have positive results on ELISA but negative results on RIBA or negative results on both RIBA and PCR are usually considered false-positive results (that is, they have no evidence of past or current HCV infection). False positive ELISA results may occur in patients with autoimmune diseases and in neonates born to mothers with chronic HCV infection, who frequently pass on antibodies to their children but usually do not pass on the virus.^34,116

False-negative ELISA results: Patients who have negative results on ELISA but positive results on PCR are usually considered to have false-negative results. False-negatives results are probably most common very early after infection (it takes 6 to 8 weeks for third-generation-ELISAs to yield positive results vs. 2 to 3 weeks for PCR) or in patients who have an impaired immune system.³⁴

Cleared or resolved HCV infection: Patients who have positive results on ELISA and RIBA but negative results on PCR on repeated testing are generally considered to have cleared or resolved HCV infection. This is usually not considered a false-positive finding because the positive RIBA test results provides specific evidence of past exposure to HCV.¹⁶⁶ Patients who have positive results on ELISA, indeterminate results on RIBA (or no RIBA performed), and negative results on PCR may either have false-positive results or have cleared their HCV infection. False-positive tests are more common in low-prevalence settings.^167,168

Chronic or active HCV infection: Patients who have persistent positive results on PCR are said to have chronic HCV infection. Chronic infection may present with or without symptoms, abnormal aminotransferase levels, or abnormal biopsy findings. In this review, the term asymptomatic chronic HCV infection refers to patients who report no symptoms of HCV infection. Like symptomatic patients, asymptomatic patients may or may not have abnormal biopsy results or aminotransferase levels.

Liver biopsy results: The Histologic Activity Index is used to grade histologic findings. The Knodell score and the METAVIR scoring system are common methods used to report the Histologic Activity Index.^169,170 The Knodell score is a semiquantitative scoring system in which fibrosis and portal, periportal, and lobular necrotic and inflammatory components are assessed separately and their coding values added. Maximum scores vary depending on exactly how the scores are totaled.¹⁷¹ The METAVIR system reports both the inflammatory and the fibrosis scores using separate standardized scores for activity and fibrosis.¹⁷⁰

Early responders: Patients with HCV infection who receive treatment and clear their viremia (viremia load undetectable by PCR) or have a significant response (usually defined as a 2-log drop in HCV RNA level) in the first few months of treatment are referred to as early responders. People who are not early responders (usually measured at 12 weeks of therapy) have a low chance of successful treatment and may not benefit from further therapy.¹¹⁶ Normalization of aminotransferase levels (biochemical response) was reported in earlier trials of HCV treatment, but has been replaced by assessments of virologic status, which are thought to be more accurate predictors of successful treatment.

End-of-treatment responders: Patients with HCV infection who receive treatment, clear their viremia, and maintain this response until the end of treatment are referred to as end-of-treatment responders. Presence of HCV RNA at the end of treatment is highly predictive of relapse when therapy is stopped.¹¹⁶

Sustained responders or sustained virologic responders: Patients with HCV infection who receive treatment and clear their viremia and maintain this response 6 to 12 months after the completion of treatment are referred to as sustained responders or sustained virologic responders.

Nonresponders: Patients with HCV infection who do not clear their viremia during treatment are considered nonresponders.

Analytic Framework and Key Questions

The analytic framework in the Appendix Figure indicates the strategy we used to evaluate screening for HCV infection in adults without known or suspected liver disease or abnormalities on liver function tests. The key questions, which guided our literature review, were determined in conjunction with liaisons from the U.S. Preventive Services Task Force.

The analytic framework shows the target samples, interventions, and intermediate and health outcome measures we examined. We narrowed the scope of the literature review after a preliminary search. We excluded children from the review because of the low prevalence of anti-HCV antibodies (0.2% to 0.4% in those 6 to 19 years old)³ and the unclear safety and efficacy of treatment in this population.¹⁷² We also excluded pregnant women because of unclear safety of treatment and insufficient evidence regarding ability to lower vertical transmission rates (estimated at approximately 5% in mothers without HIV infection).^173-176 We excluded other specific populations, such as patients who had received transplants, HIV-infected patients, and patients receiving hemodialysis patients. In these patients, screening test characteristics and natural history of HCV infection may differ from what is observed in the general population.^23,177-180 In addition, these populations have generally been excluded from large trials of treatment and data regarding clinical outcomes are lacking. Patients with occupational exposures were also excluded because of clear consensus regarding screening after percutaneous exposures.¹

Our review evaluated the screening strategy in which a second- or third-generation HCV ELISA is the initial test, with confirmatory RIBA. These are the screening tests that are currently in standard use for the diagnosis of current or resolved HCV infection.⁹⁰ PCR testing, aminotransferase testing, and liver biopsy were considered the standard work-up to determine presence of chronic HCV infection and eligibility for treatment in patients who tested positive for anti-HCV antibodies.

For treatment of chronic HCV infection, we focused on evidence regarding efficacy and safety of pegylated interferon with ribavirin, the treatment regimen found in good-quality trials to have the highest efficacy. Because this treatment regimen has been available for evaluation only a short time, we also reviewed evidence regarding the effect of other interferon-based treatment regimens on long-term clinical outcomes. Ribavirin alone, amantadine, and corticosteroids were not included because they have not been found to be efficacious.^1,98,181

For outcomes, we were particularly interested in reviewing any literature on the benefit of early antiviral treatment of chronic HCV infection in asymptomatic patients. Clinical outcomes that we evaluated were mortality, end-stage liver disease, cirrhosis, and hepatocellular cancer. Quality of life outcomes were also evaluated. Intermediate outcomes were loss of detectable viremia, improvement in histologic findings, and normalization of aminotransferase levels. We also reviewed adverse outcomes from screening and treatment including side effects from treatment, adverse events from liver biopsy, and effects of diagnosing chronic HCV infection on quality of life.

Other reasons for screening for HCV infection might be to prevent spread of the disease or to identify those who might benefit from hepatitis A or B vaccination, alcohol cessation counseling, or other interventions. We performed an additional literature search and review to identify potential benefits from screening that leads to these types of interventions in patients with chronic HCV.

Methods

Search Strategy

We searched the topic of HCV in the MEDLINE® (1989 to July 2002, update search in February 2003) and the Cochrane Clinical Trials Registry (2002, Issue 2). We originally performed 3 MEDLINE® searches, one for screening for HCV infection, one for work-up of HCV infection, and one for treatment of HCV infection. For screening, the medical subject headings (MeSH®) hepatitis C and hepacivirus were combined with the terms mass screening, hepatitis C antibodies, predictive value of tests, and sensitivity and specificity, and the text words antibody testing. For work-up, the MeSH® headings hepatitis C and hepacivirus were combined with the terms ultrasonography, liver function tests, liver biopsy, and viral load. For treatment, the MeSH® headings antiviral agents, interferons, and ribavarin were combined with the terms hepatitis C and hepacivirus. We conducted a search for controlled studies of treatment of HCV infection in the Cochrane Library databases, using the phrase hepatitis C in title, abstract, or keywords combined with terms for clinical trials. We retrieved the complete reference list from a recent Agency for Healthcare Research and Quality evidence report commissioned by the National Institutes of Health to update its consensus statement on management of HCV infection.³³ Periodic hand searching of hepatology, gastroenterology, and major medical journals; review of the reference lists of retrieved articles; and suggestions from expert reviewers supplemented the electronic searches.

We performed an additional MEDLINE® search in February 2003 on counseling on alcohol use, immunizations, and risky behaviors in patients with HCV infection. For this search, we combined the MeSH® headings hepatitis C, hepacivirus, or hepatitis C, chronic with the MeSH® headings patient education, counseling, alcohol drinking, viral hepatitis vaccines, hepatitis A, or vaccination.

One reader reviewed all English-language abstracts. Papers were selected for full review if they were about HCV infection, were relevant to key questions in the analytic framework, and met other inclusion criteria specific to the key questions. Reviews, policy statements, and other papers with contextual value were also obtained from the searches. Studies published as abstracts were not included in the search; although pertinent abstracts may be referred to in the text they are not included in evidence tables.

Inclusion Criteria

For all key questions, articles were limited to those that evaluated the general adult population with chronic HCV infection. We excluded studies that focused only on patients with end-stage liver disease, cirrhosis, or hepatocellular cancer. Although the population of interest was asymptomatic adults with chronic HCV infection who would be identified by screening, we included studies of patients with a broad spectrum of chronic HCV disease to get a picture of the benefits and adverse effects of screening and treatment in patients with different degrees of liver disease. Studies on persons with HCV who had undergone transplantation were excluded, as were studies of pregnant patients, children, or those with end-stage renal disease or HIV infection. Studies of non-human subjects were also excluded, and studies had to include original data. Foreign language papers were considered if they were clinical trials and an abstract was available in English. We searched for relevant systematic reviews for all key questions.

For individual key questions, additional inclusion criteria were as follows:

For key question 1, articles were included if they were clinical trials or observational studies that evaluated clinical outcomes in patients screened and not screened for HCV infection.

For key question 2, we included large observational studies that used appropriate statistical methods to assess associations between various risk factors and the presence of HCV infection. Representative smaller observational studies were also reviewed.

For key questions 3a, 3b, and 4, we included observational studies and systematic reviews that evaluated third-generation ELISA (the most recent generation) and used a credible, current reference standard (third-generation RIBA or PCR). We did not include studies that evaluated third-generation ELISA only in relationship to an earlier generation ELISA or performed the reference standard only in "discordant" samples from 2 screening tests. We also included data from large, good-quality observational studies on diagnostic test characteristics of second-generation ELISA.

For key question 5a, we included studies that evaluated the ability of blood tests to predict liver biopsy results, and performed liver biopsy as the reference standard.

For key question 5b, we included clinical trials and observational studies that reported the number of patients referred or considered for HCV treatment after a positive HCV antibody test, and that also provided detailed information about the reasons patients were considered ineligible for treatment.

For key question 6, we included observational studies that reported complications from percutaneous liver biopsy specifically in patients with chronic HCV infection. We also included representative large higher-quality observational studies of complications from percutaneous liver biopsy performed for a variety of indications.

For key questions 7a and 7b, we included controlled trials of antiviral treatment that evaluated relevant intermediate or clinical outcomes in treatment-naive samples. We included studies that evaluated pegylated interferon with or without ribavirin versus another treatment or placebo, and studies that evaluated non-pegylated interferon plus ribavirin compared to interferon alone or placebo. For question 7b, controlled trials of non-pegylated interferon without ribavirin were also included if they had more than 5 years of post-treatment followup and evaluated clinical or histologic outcomes. We reviewed clinical trials that were previously included in good-quality systematic reviews to ensure accuracy and reproducibility of the findings of the systematic reviews.

For key question 7c, we included controlled trials and observational studies that evaluated the effectiveness of counseling and immunizations in patients with HCV for improving clinical outcomes related to hepatitis A or B infection, alcohol use, or preventing spread of disease.

For key question 8, we included controlled antiviral trials and observational studies that reported adverse events in treatment-naive samples. We included studies of pegylated interferon with or without ribavirin versus another treatment or placebo and studies of nonpegylated interferon plus ribavirin versus another treatment or placebo.

For key question 9 we included controlled antiviral trials and observational studies in which long-term outcomes were stratified by intermediate responses to treatment.

For all key questions, we reviewed meta-analyses and systematic reviews when available.

Data Extraction

We used predefined criteria from the USPSTF to assess the internal validity of included systematic reviews, trials and observational studies, which we rated as good, fair, or poor. We also rated the applicability of each study to the population that would be identified by screening. The rating system was developed by the USPTF and is described in detail elsewhere.³⁵ For included trials and systematic reviews, we also abstracted information about setting, patients, interventions, and outcomes. For clinical trials, when possible we recorded the difference between the probability of a response in the treatment and control groups for each outcome studied. We evaluated the applicability of reviewed studies to the population likely to be identified by screening. We developed evidence tables for those key questions related to antiviral treatment of HCV infection (key questions 7a and 7b). We rated the overall body of evidence for each key question using the system developed by the USPTF.³⁵

Return to Contents