The initiation of routine screening of blood donors has also led to the identification of many persons with anti-HCV who are asymptomatic, have no history of liver disease, and deny any risk factors for exposure to viral hepatitis. These blood donors who test positive for anti-HCV are informed by blood bank personnel of their serologic status and are advised to consult their private physicians for evaluation [7]. However, the clinical significance of finding anti-HCV in these otherwise healthy persons is unclear. Previous studies have shown that 60% to 80% of blood donors with anti-HCV have elevated serum aminotransferase levels; in most cases, these elevations are persistent, indicating the presence of chronic hepatitis. Furthermore, most anti-HCV-positive blood donors can be shown by molecular techniques to harbor HCV RNA in serum and can transmit hepatitis C regardless of whether serum aminotransferases are elevated [8-14]. Studies of liver histologic findings in healthy blood donors with anti-HCV indicate that most of these donors have chronic hepatitis, despite the absence of symptoms or abnormal serum aminotransferase levels [15-22].

Thus, the clinical significance of the presence of anti-HCV in volunteer, apparently healthy blood donors remains uncertain. To better define the natural history of chronic hepatitis C in this population, we have initiated a long-term study of volunteer blood donors found to have anti-HCV on routine testing [23]. An initial cohort of these donors has had extensive medical evaluation, including liver biopsy.

Methods

Top Methods Results Discussion Author & Article Info References

Participants

Volunteer blood donors identified as positive for anti-HCV by first- or second-generation enzyme immunoassay (EIA-1 or EIA-2 according to availability; Ortho Diagnostics Co., Raritan, New Jersey) on routine screening by the Greater Chesapeake and Potomac Region of the American Red Cross were referred to the Department of Transfusion Medicine at the National Institutes of Health for participation in a long-term study of the natural history of HCV infection [23]. Donors who agreed to enter the study and signed an informed consent form had a clinical history, physical examination, and blood tests. These participants were followed at 1- to 4-month intervals with repeat testing for serum aminotransferase levels and anti-HCV. After they had been followed for at least 6 months, participants in whom serum anti-HCV was detected by EIA-2 and confirmed by a second-generation immunoblot assay (RIBA-2; Ortho Diagnostics Co.) were contacted by a study coordinator and asked to participate in our current study. Participants were divided into three groups on the basis of the maximal serum alanine amino-transferase (ALT) level measured on multiple determinations (mean number of determinations, 5; range, 2 to 9) done during the 8- to 30-month screening period. Participants in group I had normal serum ALT levels ( 0.68 µkat/L) on all occasions before admission for liver biopsy; participants in group II had elevated ALT levels that were never more than twice the upper limit of the normal range (0.7 to 1.37 µkat/L); and participants in group III had abnormal serum ALT levels, at least one of which was greater than twice the upper limit of the normal range (> 1.37 µkat/L). These groups were categorized before the study began to separate participants with completely normal serum aminotransferase levels from those with either mildly abnormal or more significantly abnormal levels. The first 20 persons in each group who agreed to participate in the study were evaluated. All participants gave written informed consent to participate. Our study was approved by the Institutional Review Board of the National Institute of Diabetes and Digestive and Kidney Diseases.

Clinical Evaluation

Participants were hospitalized for 2 to 3 days at the National Institutes of Health Clinical Center for evaluation. A detailed clinical history was obtained, and the presence and severity of symptoms of liver disease were assessed using a standardized symptom questionnaire. Alcohol use was evaluated using a short form of the Michigan alcoholism screening test (MAST) [24], and the total amount of alcohol consumed during the participant's lifetime was estimated by a lifetime drinking history, in which consumption of more than 3.5 kg of alcohol per kg body weight is considered to be associated with a high risk for developing alcohol-related liver injury [25, 26].

Laboratory investigations included a complete blood count, measurement of prothrombin time, serum biochemical profile, iron studies, and measurement of ceruloplasmin and 1-antitrypsin levels. Serologic studies included measurement of rheumatoid factor, antinuclear antibody, immunoglobulin, and cryoglobulin levels. Virologic testing included tests for serum anti-HCV, hepatitis B surface antigen (HBsAg), antibody to HBsAg, antibody to hepatitis B core antigen, and antibody to human immunodeficiency virus (HIV) type 1 by commercial immunoassays (Abbott Laboratories, North Chicago, Illinois). Serum was tested for HCV RNA by nested polymerase chain reaction (PCR) using primers complementary to the 5'-untranslated region (5'-UTR) [27]. Serum HCV RNA was quantitated by a branched-DNA assay (Chiron Corp., Emeryville, California) [28]. The HCV genotypes were determined by a line-probe assay, which uses hybridization of 5'-UTR nested PCR products with genotype-specific oligonucleotide probes immobilized on membrane strips (Innogenetics N.V., Belgium) [29].

All participants had percutaneous liver biopsy; most of each biopsy specimen was routinely processed for histologic evaluation, and the rest was snap-frozen in liquid nitrogen for direct immunofluorescence staining for HCV antigen [30]. All routine biopsy specimens were read under code by a hepatic pathologist (DK) using conventional criteria and calculation of the histologic activity index [31, 32].

Statistical Analysis

Continuous variables were analyzed by one-way analysis of variance with post hoc Student t-tests. Discrete variables were analyzed by the chi-square test with Yates' correction and by the Fisher exact test.

Results

Top Methods Results Discussion Author & Article Info References

Among 618 062 volunteer blood donors screened between 1991 and 1993, 3653 (0.6%) were anti-HCV positive and were asked to participate in our long-term follow-up study. Four hundred sixty of these donors agreed to participate. Testing of serum by RIBA-2 showed that 207 donors were reactive; 197 of these (54 in group I, 85 in group II, and 58 in group III) were contacted for the current study. The first 20 persons in each group who agreed to participate were evaluated.

Demographic and Clinical Characteristics

The demographic and clinical characteristics of the three groups are shown in Table 1. Participants with normal serum ALT levels tended to be older and were more likely to be female than those with abnormal levels. Other features were similar; however, because the number of patients studied was relatively small, only major differences could have been detected. The duration of disease, which could be estimated in 39 participants who had a history of an exposure, ranged from 6 to 37 years; the average was the same (16 to 20 years) among the three groups. The suspected sources of hepatitis C were parenteral drug abuse in 21 participants (35%), transfusions in 12 (20%), and medical care occupation in 7 (12%). All donors who admitted to using parenteral drugs had not used the drugs for several years. The donors who had a history of blood transfusion were not aware of having developed hepatitis after any of the transfusions.

Symptoms

Although all participants were volunteer blood donors and were thus considered asymptomatic, many participants reported mild symptoms on the self-administered symptom questionnaire. The most common symptoms were fatigue (61%), headaches (54%), anxiety (54%), drowsiness (53%), muscle aches (46%), itching (34%), depression (29%), and excessive sweating (29%). Fever, dizziness, weight loss, and vomiting were uncommon (range, 5% to 10%), and only one donor reported a history of jaundice. The distribution, average number, and symptom score were similar among the three groups, and most participants reported that the symptoms were mild and did not interfere with daily activities or work. None of these participants was seeing a physician for these symptoms, and none had a concurrent disease or were regularly taking any medications.

Alcohol Use

Only 29 of the 60 participants admitted that they currently used alcohol, and only 3 of these drank more than one drink per day on average. Eight participants had never consumed alcohol. In addition, most participants scored low on the MAST questionnaire; only 11 participants (18%) answered 3 or more of the 13 questions affirmatively, a score suggestive of alcohol abuse. Estimation of lifetime alcohol intake indicated that 9 participants (2 in group I, 1 in group II, and 6 in group III) had a level in the range that is considered to represent a high risk for alcohol-associated illness. The average and distribution of MAST scores and lifetime drinking history estimations were similar among the three groups. Alcohol consumption and the serum levels of ALT and aspartate aminotransferase (AST) were not correlated (data not shown). A correlation was seen between heavy alcohol use and a history of illicit drug use. The average estimated lifetime alcohol intake of the 21 participants with a history of injection drug use was 2.8 kg of alcohol/kg body weight (range, 0 to 7.9 kg/kg body weight) compared with 1.1 kg/kg body weight (range, 0 to 6.6 kg/kg body weight) among participants without such a history (P = 0.012).

Laboratory Test Results

Average results of routine laboratory tests for the three groups are shown in Table 2. Except for the serum ALT and AST levels, the test results did not significantly differ among the three groups. Participants were assigned to each group on the basis of serial determinations done during the screening period; in some participants, ALT levels fluctuated in and out of the normal range. On the day of liver biopsy, serum ALT levels were elevated in 33 participants (including 3 patients in group I for the first time), AST levels were elevated in 26, alkaline phosphatase levels were elevated in 8, and -glutamyl transpeptidase levels were elevated in 14. Thus, only 10 of the 20 participants in group I and none of the participants in groups II and III had normal values for all serum enzymes on all occasions. Results of tests of hepatic synthetic functions were usually normal. The serum bilirubin level was 21 µmol/L or less in all participants; the serum albumin level was 35 g/L or greater in all but two participants; and the prothrombin time was mildly prolonged (14.5 seconds) in only one participant.

Rheumatoid factor was detected in 53% of participants, antinuclear antibodies in 23%, and cryoglobulins (levels greater than 3%) in none. Ceruloplasmin and 1-antitrypsin levels were normal in all participants. Ferritin levels were elevated in three participants, but only one participant (in group III) was found to have genetic hemochromatosis with a hepatic iron index of 3 mmol/g dry weight per year. One participant had increased uroporphyrin levels in urine but no signs or symptoms of porphyria; another had a history of porphyria cutanea tarda and was in remission. No participant had antibody to HIV or HBsAg in serum, and a similar percentage in each group had antibodies to hepatitis B virus (data not shown).

Virologic Test Results

Multiple serum specimens from all 60 participants tested positive for anti-HCV by EIA-2 and RIBA-2 assays. Hepatitis C virus RNA was detectable by PCR in 51 participants (85%) and by bDNA assay in 45 participants (75%) (Table 3). For the nine participants without HCV RNA, several specimens were tested by PCR (range of specimens, 2 to 7) in two separate laboratories. These participants were also negative by the bDNA assay. Hepatitis C virus RNA was detected by PCR more frequently among participants in groups II and III (both 95%) than among those in group I (65%). The average amount of HCV RNA as assessed by bDNA assay was similar in the three groups. Genotyping of HCV RNA was done on HCV RNA-positive specimens, and a genotype was assigned in 50 participants: Twenty-three (38%) had genotype 1a; 20 (33%) had 1b; 1 (2%) had 2a; and 4 (7%) had 2b. One participant had a mixture of genotypes 1a and 2, and a second had genotype 1, which could not be resolved further into subtypes. One HCV RNA-positive sample could not be genotyped, probably because of low levels of HCV RNA. Remarkably, only two participants in group I (10%) had genotype 1a compared with 50% to 55% of participants in groups II and III, respectively.

Liver Histologic Findings

The liver histologic results are summarized in Table 4. Only three biopsy specimens were read as completely normal, and three more had minimal nonspecific reactive changes. Examination of biopsy specimens showed chronic hepatitis with mild activity in 33 (55%) and chronic hepatitis with moderate to severe activity in 20 (33%); cirrhosis was seen in only 1 specimen. Thus, 54 donors (90%) had some histologic evidence of chronic hepatitis.

Participants with normal serum ALT levels tended to have milder histologic forms of liver disease, as assessed by conventional diagnoses and by the histologic activity index. The six biopsy specimens that were considered to be normal or that had nonspecific changes were obtained from participants in group I. All these patients had normal serum ALT, AST, -glutamyl-transpeptidase, and alkaline phosphatase levels. In addition, the four features of the histologic activity index were all milder in participants in group I than those in participants in groups II and III. This difference was most marked for lobular activity and to a lesser but still significant extent for fibrosis. Thus, only 4 participants (20%) in group I had some degree of fibrosis on liver biopsy compared with 8 (40%) in group II and 14 (70%) in group III. Six participants had bridging fibrosis (1 in group I, 2 in group II, and 3 in group III), and 1 participant (group III) had active cirrhosis. Histologic features were more severe in participants in group III than in those in group II, but these differences did not reach statistical significance. None of the biopsy specimens showed characteristic features of alcoholic liver disease, including specimens from participants with a high estimated lifetime intake of alcohol. The presence of steatosis was similar in all three groups, was marked in only one participant, and did not correlate with alcohol history (data not shown). Nevertheless, the mean histologic activity index score was higher among participants with a lifetime alcohol consumption greater than 3.5 kg/kg body weight (mean score, 9; range, 6 to 15) than that among those who consumed less alcohol (mean score, 7; range, 1 to 16) (P = 0.038).

The one participant with histologically proven cirrhosis had no symptoms of liver disease but had markedly elevated serum ALT (3.08 µkat/L) and AST (3.68 µkat/L) levels, decreased serum albumin levels (30.0 g/L), borderline elevated serum bilirubin levels (21 µmol/L), and a prolonged prothrombin time (14.5 seconds). This participant consumed little alcohol and had no evidence of other forms of liver disease.

Significance of Hepatitis C Virus RNA

Among the various demographic, clinical, biochemical, serologic, and virologic features analyzed, the presence of HCV RNA in serum as detected by PCR appeared to have the most bearing on histologic findings. Findings of the 9 participants whose serum was negative for HCV RNA by PCR and the 51 participants whose serum was positive are summarized in Table 5 and in Figure 1. The presence of viremia did not correlate with the donor's age, sex, suspected source of hepatitis, or duration of infection. However, participants without detectable HCV RNA were more likely to have persistently normal serum ALT levels (78%) than those with this viral marker (20%). Indeed, the two HCV RNA-negative participants with abnormal serum ALT levels both had other possible causes for these abnormalities: One (in group II) had marked steatosis on liver biopsy, and the other (in group III) had genetic hemochromatosis.

View larger version (9K): [in this window] [in a new window]   Figure 1. Alanine aminotransferase (ALT) levels (top) and histologic activity index scores (bottom) in viremic and nonviremic volunteer blood donors. Fifty-one donors were positive on at least one occasion for hepatitis C virus RNA in serum by polymerase chain reaction [PCR], and 9 donors were repeatedly negative for hepatitis C virus RNA by PCR. Values shown are the maximal recorded during the screening period; the interrupted line indicates the upper limit of the normal range, and the horizontal bars represent the means. Analysis was done using the Student t-test). To convert alanine aminotransferase values to µkat/L, multiply by 0.01667.

Importantly, most participants who were HCV RNA negative had either normal liver histologic findings or only mild changes. All three participants with normal liver biopsy specimens and two of the three participants with nonspecific abnormalities only were HCV RNA negative by PCR. The other four HCV RNA-negative participants had chronic hepatitis with mild activity (histologic activity index scores of 5, 5, 6, and 10, respectively). Thus, the absence of detectable HCV RNA by PCR was more predictive of mild histologic changes than a normal serum ALT level. Indeed, when the 9 HCV RNA-negative patients were excluded from analysis, no statistically significant differences were seen among the three groups in liver histologic findings (mean histologic activity index scores: 7.0 in group I, 7.6 in group II, and 9.1 in group III; P = 0.122 by one-way analysis of variance).

Hepatitis C Virus Antigen in the Liver

Hepatitis C virus antigen was detected in 43 of the 52 biopsy specimens (83%) that could be evaluated, and the frequency of detectable antigen was similar in the three groups. All 43 participants in whom HCV antigen was detected in the liver had HCV RNA detectable in serum by PCR (40 were also positive by the bDNA assay). In contrast, in none of the 5 participants who were HCV RNA negative was HCV antigen detected in the liver. Thus, the presence of detectable HCV antigen in the liver appeared to correlate with the presence and level of HCV RNA in serum.

Discussion

Top Methods Results Discussion Author & Article Info References

Our study of 60 volunteer anti-HCV-positive blood donors showed that most had chronic HCV infection and some degree of ongoing chronic liver disease. Thus, 85% of the participants were viremic (that is, HCV RNA was detected in their serum), and 90% had evidence of chronic hepatitis by liver biopsy, including 70% of the donors with persistently normal serum ALT levels. These findings are similar to those reported by other groups who have studied anti-HCV-positive blood donors in Europe, Australia, and Asia [15-22].

None of the 60 participants was aware of having hepatitis before attempting to donate blood, and all answered negatively to questions about high-risk behaviors for hepatitis at the time of blood donation. On closer questioning, however, 35% of donors in this group admitted to using illicit parenteral drugs in the past, and 20% had a history of exposure to blood or blood products. Most of the 60 blood donors had no symptoms of liver disease, although questionnaires often elicited positive responses on fatigue, headaches, anxiety, depression, and muscle aches—nonspecific symptoms that are common in the general population. None of the participants studied here (including a participant with cirrhosis) had symptoms that were obviously related to liver disease. These factors underscore the silent nature of chronic hepatitis C and the absence of symptoms and history of liver problems in many patients with this disease.

The anti-HCV-positive blood donors were separated into three groups for analysis according to serial determinations of serum ALT levels. Measurement of serum ALT levels is a readily available way to assess chronic hepatitis disease activity and is believed to be a reasonably reliable indicator of severity of disease. However, it is clear from studies such as ours that the presence of normal serum ALT levels even on multiple occasions does not rule out the presence of chronic hepatitis. Indeed, 70% of participants with repeatedly normal serum ALT levels in our study had chronic hepatitis on liver biopsy. None of these participants had cirrhosis, but several had chronic hepatitis with moderate inflammatory activity; one participant had bridging fibrosis.

In our study, as in previous studies [9, 17-22], the status of HCV RNA in the serum appeared to increase the ability to predict the presence of active liver disease. Thus, patients negative for HCV RNA by PCR usually had normal serum ALT levels, and most had minimal or no evidence of liver injury even as assessed by liver biopsy. The patients who were anti-HCV positive but HCV RNA negative (on repeated samples) may actually have recovered from hepatitis C, having cleared the viral infection and then having had resolution of the hepatitis activity. Of the 60 participants studied, 5 (8%) appeared to fall into this category of having resolution of HCV infection; this was shown by normal serum levels of ALT and other hepatic enzymes, no HCV RNA in serum, and normal (n = 3) or nonspecific changes (n = 2) on liver biopsy specimens. This prevalence of resolved HCV infection fits well with reports of prospective studies in which only 5% to 15% of patients with post-transfusion or sporadic acute hepatitis C cleared the infection and had normal aminotransferase levels and no HCV RNA in serum during long-term follow-up [33, 34].

Although 90% of these 60 blood donors had histologic features of chronic hepatitis, it remains unclear whether these participants will ultimately have symptoms or disability or die of their chronic liver disease. Only one participant in this cohort had cirrhosis, and he was still asymptomatic and unaware of being ill. Although many of these participants had probably been infected with HCV for several decades, no correlation was found between the estimated duration of infection and the severity of liver injury or the amount of fibrosis. Thus, chronic hepatitis C in these blood donors was probably only slowly progressive, if at all. Only long-term natural history studies will resolve the issue of the prognosis of this chronic infection [33, 35, 36].

Until more is known about the natural history of chronic hepatitis C, it is difficult to make recommendations for managing asymptomatic, apparently healthy persons who test positive for anti-HCV. General recommendations on hygiene (blood precautions and care with handling cuts and scratches) and sexual activity (safe sexual behaviors outside of monogamous relationships) are important [7]. Alcohol use should also be discouraged in patients with chronic hepatitis C. There is ample evidence for synergy between alcohol and viral hepatitis in causing liver injury to recommend avoidance of all except ceremonial use of alcohol.

A more difficult task is advising patients who have no or minimal biochemical evidence of disease on the need for liver biopsy and therapy. Donors with normal ALT levels had a liver biopsy in our study as part of a prospective clinical research protocol that was approved by an institutional review board and was done after written informed consent was obtained. In clinical practice, liver biopsy of persons with normal ALT levels should not be considered standard or recommended. As to therapy, interferon- has been shown to be beneficial in some patients with this disease. However, a 6-month course of interferon induces long-term remissions in disease in only 15% to 25% of patients, and the efficacy of interferon in patients with normal serum ALT levels has not been proven [37, 38]. Furthermore, the natural history of HCV infection suggests that many asymptomatic carriers will have an indolent course and will not sustain long-term consequences [36]. Until more effective therapies and better ways to identify patients who are likely to benefit from treatment are available, persons with chronic HCV infection but normal or minimally elevated serum amino-transferase levels should be followed and monitored but not treated. Clearly, a better definition of the natural history of chronic hepatitis C and more effective and safer therapies for this disease are needed.

Drs. Conry-Cantilena and Alter: Department of Transfusion Medicine, National Institutes of Health, Building 10, Room 1C711, Bethesda, MD 20892.

Drs. Hayashi, Conjeevaram, and Sallie: Liver Diseases Section, National Institutes of Health, Building 10, Room 4D52, Bethesda, MD 20892-1372.

Dr. Kleiner: Laboratory of Pathology, National Cancer Institute, Building 10, Room 2N206, Bethesda, MD 20892.

Dr. Tedeschi: Laboratory of Hepatitis Research, CBER, Food and Drug Administration, Bethesda, MD 20892.

Dr. Krawczynski: Hepatitis Branch, Centers for Disease Control, 1900 Clifton Road, Atlanta, GA.

Dr. Di Bisceglie: Department of Internal Medicine, St. Louis University Health Sciences Center, 1402 South Grand Boulevard, St. Louis, MO 63104.