Very Low Incidence Rates of Community-Acquired Hepatitis C Virus Infection in Company Employees, Long-Term Inpatients, and Blood Donors in Japan

To assess the contemporary rate of hepatitis C virus (HCV) propagation in Japanese community, we conducted a sentinel study on three groups of people: company employees, long-term inpatients, and blood donors. A total of 3079 company employees negative for antiHCV were followed thereafter, and seroconversion to anti-HCV was found in 5 per 5786 personyears. None of them was positive for HCV RNA. In the group of 703 long-term inpatients (25 of whom were initially positive for anti-HCV), none showed seroconversion to anti-HCV per 2712 person-years. As for 114266 repeated blood donors who were initially antibody-negative, 227 became anti-HCV positive later. Of these seroconverted donors, 83 were found to have anti-HCV with titers of 26 or greater. HCV RNA was positive in only 3 of them. Thus, the incidence rates of acquired HCV viremia in these three groups were 0 for both company employees and long-term inpatients, and 1.78 (95% C.l.: 0.37-5.19) per 100000 person-years in blood donors. These results suggest that community-acquired HCV infection is now rare in Japan, and that even if it occurred it hardly leads to persistent viremia. J Epidemiol, 1996 ; 6 : 198-203.

To assess the contemporary rate of hepatitis C virus (HCV) propagation in Japanese community, we conducted a sentinel study on three groups of people: company employees, long-term inpatients, and blood donors.A total of 3079 company employees negative for anti-HCV were followed thereafter, and seroconversion to anti-HCV was found in 5 per 5786 personyears.None of them was positive for HCV RNA.In the group of 703 long-term inpatients (25 of whom were initially positive for anti-HCV), none showed seroconversion to anti-HCV per 2712 person-years.
As for 114266 repeated blood donors who were initially antibody-negative, 227 became anti-HCV positive later.Of these seroconverted donors, 83 were found to have anti-HCV with titers of 26 or greater.HCV RNA was positive in only 3 of them.Thus, the incidence rates of acquired HCV viremia in these three groups were 0 for both company employees and long-term inpatients, and 1.78 (95% C.l.: 0.37-5.19) per 100000 person-years in blood donors.These results suggest that community-acquired HCV infection is now rare in Japan, and that even if it occurred it hardly leads to persistent viremia.J Epidemiol, 1996 ;6 : 198-203.
hepatitis C virus (HCV), epidemiology, community-acquired infection, asymptomatic carriers, person-time method Hepatitis C virus (HCV), as is hepatitis B virus (HBV), is a blood-borne virus that infects humans transiently or persistently and causes a wide spectrum of liver diseases ranging from acute hepatitis to hepatocellular carcinoma.The morbidity and mortality due to persistent HCV infection is an urging problem particularly in Japan, because more than half of Japanese patients with chronic hepatitis, liver cirrhosis, or hepatocellular carcinoma are infected with HCV, whereas in most other Asian countries HBV outweighs HCV in the etiology of such liver diseases 1).Under the current absence of vaccines or hepatitis immune globulins against HCV, only one preventive measures now being actively conducted is blood screening for anti-HCV to prevent transfusion-transmitted infection.Blood screening for this purpose was set forth by Japanese Red Cross blood centers using the first-generation anti-HCV assay in November, 1988: the first nationwide HCV-screening in the world.Indeed, it has resulted in great reduction of incidence rate of post-transfusion non-A, non-B hepatitis 2,3).On the other hand, however, community-acquired HCV infection has been left un-intervened by now, inevitably due to unavailability of appropriate countermeasures.
Since HCV is a blood-borne virus, it is highly prevalent in high-risk groups such as intravenous drug abusers who are frequently exposed to the virus via contaminated needles or syringes.Epidemiology of HCV in high-risk groups has thus been much appreciated 4-10)However, because majority of general population are spending their life without engaging in such high-risk behaviors, we need to know also what happens on "low-risk" groups.To predict whether HCV is prevailing or diminishing in one community as a whole, we must focus on the rates of HCV acquisition by ordinary individuals leading a standard life-style in the community.For this purpose, we selected three groups of people from Japanese population, and assessed the rates of HCV acquisition by person-time method.

Company Employees
Traditionally, Japanese "salary-man" does not drift among different companies, but, instead, tends to work stable for more than a decade (mostly until his retirement at age 55 or 60) in the company where he found his first job in his life ("salarywoman" is more flexible in this context).This was favorable for us to conduct a sentinel study there: rate of drop-out from follow-up is very low.During the period from April, 1992 to March, 1995, a big company in Hiroshima (NTT: Nippon Telegraph & Telephone Corporation, Hiroshima Administration Center, Hiroshima, Japan) conducted annual health check on all the employees older than 40 years of age and on those who were younger than 40 years old but wanted for the health check.Sera obtained from them were stored frozen at -20 or lower temperature for later analysis of viral markers.A total of 3079 employees were found to be negative for anti-HCV at the first testings and were enrolled in this study.Their ages distributed from 19 to 84 years (male/female ratio was about 3.7).The total person-years observed for this group were 5786.

Long-Term Inpatients
Another site of our sentinel study was Tenryu Kosei-kai Hospital, Shizuoka, Japan, where mentally or physically handicapped people spend a long-term hospitalization for treatment.At the beginning of our sentinel study in 1988, there were 678 inpatients (male 342, female 336) who were negative for anti-HCV, together with 25 antibody-positive patients.Their ages were 20 to 90 (mean: 49 17) years.All of them remained still hospitalized even four years later at the end of our study there in 1992, when anti-HCV was determined again.The personyears observed for this group of patients were 2712.

Blood Donors
During the three years from February, 1992 to January, 1995, a total of 555292 blood units were donated by 258290 voluntary blood donors at Japanese Real Cross Hirmshima Blood Center, Hiroshima, Japan.Of these, 114266 were repeated donors and they were found toy be negative for anti-HCV at the time of first donation.Their ages were 16 to 64 (mean: 33 14) years, and male / female ratio was 1.3.Most sera from seroconverted donors were stored frozen for later analyses.The total person-years for this group were 168479.

Detection of Anti-HCV and HCV RNA
Anti-HCV was detected qualitatively by 2nd generation ELISA (Ortho Diagnostics, Tokyo) and /or quantitatively by passive hemagglutination (PHA) method (Dainabot, Tokyo) with procedures indicated by the manufacturers.Freeze-stored sera were used for detection of HCV RNA in seroconverted cases.HCV RNA was detected by nested RT-PCR using primers deduced from 5' untranslated region of HCV genome according to previously reported method 11).HCV RNA was genotyped by PCR using genotype-specific core regionderived primers by procedures described previously 12).Nomenclature of HCV genotypes was based on both Okamoto et al's 12) and Simmonds et al's 13).

RESULTS
As shown in Table 1, in the company employees survey, 5 showed seroconversion from anti-HCV negative to anti-HCV positive, suggesting that they acquired HCV infection during the study period.However, the infection seemed to be tran-  Titers of anti-HCV were determined by passive hemagglutination (PHA) method.N.T.: not tested.
Table 3. Determination of anti-HCV titers and detection of HCV RNA in seroconverted blood donors.
Cut-off for the anti-HCV PHA is 25.N.T.: not tested.
sient and silent in all of them: there were no symptoms and signs of hepatitis claimed subjectively from them; serum ALT levels tested normal over time; anti-HCV titers were generally low; and HCV RNA was undetectable by PCR (Table 2).
In the survey of long-term inpatients, all of the 678 patients who were found to be negative for anti-HCV in 1988 remained still hospitalized in 1992.None of them had seroconverted to anti-HCV in 1992 (Table 1).
In the survey of repeated blood donors, 227 out of 114266 donors showed seroconversion to anti-HCV during the observation period (Table 1).Of the 227 seroconverted donors, 200 donors were further examined for anti-HCV antibody titers by passive hemagglutination (PHA) method (remaining 27 donors were not examined because their sera had not been stored).Distribution of anti-HCV titers in a total of 306 sera from the 200 donors is shown in Table 3. Antibody titers in this group of sera were generally low, and HCV RNA was detected in only 3 sera (from 3 individuals) that showed relatively higher titers of anti-HCV (Table 3).Changes in serum ALT levels in these 3 blood donors over time suggested that at least one of them (case #1 in Table 4) contracted clinical hepatitis by acquiring HCV infection during the observation period.Remaining two cases (cases #2 & #3 in Table 4) also showed an elevation of serum ALT levels, but only within normal range.

DISCUSSION
Of the three groups we selected from Japanese population for the present study, company employees and blood donors may represent general population of Japanese community, because their ordinary daily lives are different one by one, with various life styles as well as with various social activities in the community.On the other hand, long-term inpatients spend 24 hours a day in a special circumstance with little contacts with outer world.Therefore, it is not surprising if there exists a difference in the risk for HCV infection between the company employees / blood donors and long-term inpatients.Indeed, seroconversion to anti-HCV was observed more or less in company employees (86.4 per 10 5 person-years) and in blood donors (49.3 per 10 5 person-years), but in none of long-term inpatients.However, the rate of becoming viremic with HCV was very low irrespective of the three groups.In total, only 1.70 per 101 person-years were found to become viremic with HCV during the study period (Table 1).This suggests that expansion of HCV reservoir in Japanese community as a whole has already ceased at present.Thus, with the next to zero occurrence of transfusion-transmitted HCV infection 2,3) taken together, it is highly possible to speculate that total number of HCV carriers in Japan has already begun decreasing 11).
It is noteworthy that the anti-HCV titers in seroconverted individuals in the present study were generally low (Tables 2  & 3), and only a few of them were positive for HCV RNA.This suggests that most cases of the community-acquired HCV infection are transient and clinically silent and hardly lead to persistent infection.Similar observation was previously reported for HCV infections associated with needle-stick accidents 15).By contrast, high proportion of transfusion-transmitted HCV infections develops persistent viremia 2).It is, therefore, conceivable that exposure to very small amount of virus, as in community-acquired infection or in needlestick exposure, may be insufficient to establish persistent infection.
In our present study, only 3 among 232 seroconverted individuals were also positive for HCV RNA, and they had relatively higher titers of anti-HCV than those who were positive for antibody but not for RNA (Table 3).This is compatible with previous findings that antibody titers correlate with HCV viremia [16][17][18].In this context, it is also noteworthy that most of the seroconverted cases in the present study had anti-HCV of very low titers: as many as 162 sera from blood donors had anti-HCV at the titer as low as 2 5-6(cut-off level of the assay system was 2 5) (Table 3).This suggests that many of the seroconverted cases in this study are in fact not a true case of seroconversion, but a case where the titers of pre-existing anti-HCV were fluctuating around the threshold levels of assay systems used.Therefore, the rate of acquisition of HCV infection calculated on the basis of seroconversion in this study may still be an over-estimation.In conclusion, our present results suggest that healthy people leading a standard life-style in Japan now seldom become HCV carriers by community-acquired infection, and that intra-hospital HCV transmission is also a rare event.This means that there is almost no gain in the number of HCV-infected individuals in Japan.
Then, how about a loss?According to a nationwide statistics of disease mortality of Japan 19), a total of 44645 died of chronic liver disease (27765 of hepatocellular carcinoma and 16880 of the other liver disease) in 1993.If we assume 70% of these were HCV-carriers, 31252 individuals might have died of HCV-related liver disease in 1993.This leads to an estimation that a loss of HCV-carriers from Japanese community by their death is occurring at a rate of 25.2 per 105 person-years (this rate is an underestimation because a small fraction of those dying of apparently HCV-unrelated diseases may also be HCV carriers).Additional loss of HCV-carriers is also being achieved by the interferon treatment in the therapy of chronic hepatitis C. If 20% of interferon-treated patients achieve sustained response, and 40000 patients are being treated by interferon annually, the rate of decrease in the number of HCV carriers by means of interferon therapy is calculated to be 6.5 per 105 person-years.Thus, with the above figures combined, the loss of HCV carriers is estimated to be occurring in Japan at a rate of 31.7 + a per 105 person-years (Table 5).
Finally, before comparing loss and gain, we may need to consider yet another factor for the gain: vertical transmission of HCV from infected mothers to their babies although its rate has been controversial.If we assume that the rate for infants bom to HCV RNA positive mothers to become HCV carriers is 2.3% 20), that the HCV RNA positive rate in women of reproductive ages is 0.5%, and that 1000000 babies are bom yearly (actually 1188282 babies were born in 1993), then 115 babies may become infected with HCV annually.Since the total population of Japan in 1993 was 123,788,000, the gain in the HCV carriers through vertical transmission would be occurring at the rate of only about 0.1 per 10 5 person-years.
This gain is almost negligible compared to the gain by communityacquired infection revealed by the current study: 1.78 per 10 5 person-years.
Table 5. Balance between estimated loss and gain in the number of HCV carriers in Japanese population.
Alpha represents a rate due to the death of HCV-carners dying of apparently HCV-unrelated diseases.
Thus, the significant unbalance between the loss and gain obtained through the above estimations, 31.7 + a versus 1.9 per 105 person-years (Table 5), clearly indicates that the number of HCV carriers in Japan is now decreasing at a considerably high rate.

Table 1 .
Synopsis of the study results.

Table 2 .
Seroconverted cases in the company employees survey.

Table 4 .
Seroconverted cases in repeated donors.
III, III, and IV are Okamoto et al's 12) and lb, 2a, and 2b are Simmonds et al's 13) nomenclatures of HCV genotypes.Titers of anti-HCV were determined by passive hemagglutination (PHA) methood.N.T.: not tested.