Diagnosis and 10-Year Follow-Up Of a Community-Based Hepatitis C Cohort

Only people who were residents of Olmsted County for at least 1 year before being diagnosed with hepatitis C are included. This step was necessary to ensure that all subjects were community members. Besides its large liver transplantation service, Olmsted County has several inpatient and outpatient chemical dependency treatment programs and halfway houses that may bring patients with hepatitis C to the community for short periods of time. Inclusion of these people would have skewed the community-based focus of the study. Prisoners incarcerated in local facilities were excluded as well.

Measures

For each subject in the cohort, we reviewed all medical records from the Mayo Clinic and hospitals, the Olmsted Medical Center and hospital, and all other care providers in the county. Data collected included information on the initial diagnostic process as well as on HCV-related follow-up; specifically, all aspartate aminotransferase/amino alanine transferase (AST/ALT) testing and all HCV treatment given. All diagnoses of cirrhosis, ascites, gastrointestinal (GI) bleeding, encephalopathy, jaundice, and hepatocellular carcinoma were recorded. Data on risk factors as well as on comorbid conditions believed to influence the progression of HCV-related liver disease (eg, alcoholism, chronic heavy alcohol ingestion, hepatitis B, and HIV disease) were noted.

Data analysis

We summarized demographic information and data on risk factors, comorbid conditions, the pattern of laboratory test follow-up, and HCV treatment and, when appropriate, stratified these data by date of diagnosis. We used logistic regression models to look for associations among personal, demographic, and clinical factors associated with continued AST/ALT follow-up 1 or more years after initial HCV diagnosis.

Results

Of the 355 subjects with a diagnosis of hepatitis C between January 1, 1990, and December 31, 1999, 136 (38%) were women and 219 (62%) were men. The mean age at diagnosis was 42.6 years (Figure 1). The rate of new diagnoses of hepatitis C varied only slightly by year (Figure 2). After the period 1990–92, when HCV testing first became available, the difference in rates of new diagnoses is not statistically significant.

Complete follow-up data from the date of diagnosis until December 31, 1999, or the subject’s death were available in 78% of subjects with mean follow-up of 3.6 years, median 3.0 years, and range 0 to 9.8 years. Other subjects were lost to follow-up after they moved from the community; however, vital status (dead or alive) was obtained in 85% of all subjects as of January 1, 2000.

IDU was documented in 177 cases (50% of subjects) (Table 1) with the mean duration of 9.6 years (SD 7.9 years, range single use to 34 years). A single risk factor was recorded for 89 subjects (69 who had had a blood transfusion before 1992 and 20 health care workers with possible exposure to blood products or body fluids, including 5 with documented needlesticks). Sexual exposure and IDU were frequent coexisting risk factors.

All subjects had a positive anti-hepatitis C antibody test; 304 (86%) had a positive RIBA; 13 (4%) had an indeterminate RIBA with risk factors; 14 (3.9%) had PCR tests used in the diagnostic process; and the rest (n = 24) had only positive serology plus risk factors. Overall, 202 people (60%) were seen by a GI or hepatology specialist at least once after the diagnosis of HCV had been made. Confirmatory liver biopsies or PCR tests were used at some time in the follow-up of 157 subjects (44%), usually before the consideration of treatment or after referral to a hepatologist.

Among subjects, 21 (no gender differences) had hepatic decompensation, defined as cirrhosis with ascites, encephalopathy, or jaundice, or hepatocellular carcinoma identified either before or within 1 month of the hepatitis C diagnosis. These findings suggest that HCV evaluation was based on the presence of advanced liver disease. Thirty-seven (10%) of the patients, including 5 who died within days to weeks of the initial HCV diagnosis, died during the observation period.

At or around the time (±1 month) of diagnosis, serum albumin (n = 215, 61%), bilirubin (n = 265, 75%), and ALT or AST (n = 308, 87%) tests were commonly done. Albumin and bilirubin levels were normal in almost all cases (99% and 85%, respectively). The majority of the initial serum ALT/AST levels were elevated (262/308, 85%). Although the elevation was often modest, levels of 119 of the 246 initial tests (48%) were less than 2 times the upper limit of normal.

Follow-up of initial AST/ALT testing was not universal. Among subjects, 51% had one or more rechecks of liver function tests (LFTs) during the first year after diagnosis; 55%, 1 to 2 years after diagnosis; 56%, 2 to 3 years after diagnosis; and 45%, 3 to 4 years after diagnosis, based on the number of subjects not lost to follow-up for 1 to 4 years. Some subjects lost to hepatitis C follow-up had periods of active alcohol or drug abuse that appeared to disrupt hepatitis C care. Variations in rates of continued monitoring of AST/ALT, however, were not associated with risk factors such as IDU or transfusion nor with demographic factors such as age. Long-term follow-up (3 to 4 years after diagnosis) was associated with AST/ALT levels more than 2 times normal at diagnosis (P = .03) and a diagnosis of cirrhosis (P= .03). Women were more likely to have a repeat evaluation in the first year, but no gender differences were seen after that.