The natural history of chronic hepatitis C concerning liver fibrosis progression has been vigorously studied using liver biopsy specimens. The extent of liver fibrosis is usually evaluated as categorical stages. For example, METAVIR Score uses five stages, F0-F4, for fibrosis evaluation[26
]. The fibrosis progression in hepatitis C patients, calculated by using paired liver biopsy, was reported to be 0.1-0.133 Unit per year[12,13
]. Liver stiffness measured by transient elastography is now widely accepted as a surrogate marker of liver fibrosis[27
]. Liver stiffness is expressed as a continuous variable in kPa unit. The cut-off for cirrhosis is reportedly 13-17 kPa, and the upper limit of measurement is currently 75 kPa. Thus LSM has a wider dynamic range than histological staging, and the rate of fibrosis progression may be more accurately analyzed with LSM.
In the present study, the increase rate of liver stiffness was positively correlated with the age at blood transfusion, as shown by the steeper slopes of approximation curves when patients received BTF at older ages. The cause of this phenomenon is not clear but age-related changes in immunity may be involved. If this is the case, the increase rate is likely to become higher in the same patient with age. Indeed, each approximation curve in the figure apparently becomes steeper with age, suggesting age-related acceleration. This is to be confirmed in future longitudinal studies.
LSM is useful not only as a surrogate of liver biopsy but also as a risk indicator of HCC development. Indeed, in the present study, liver stiffness is high in patients with HCC regardless of duration of infection. The patients who developed HCC with short duration of infection received blood transfusion at an older age and were older at the time of LSM, male dominant, and showed higher liver stiffness than patients without HCC with similar duration of infection. The difference between patients with and without HCC became smaller with longer duration of infection, as the average liver stiffness in patients with HCC became lower and that in patients without HCC became higher. We speculated that patients with high liver stiffness who received blood transfusion in the early period have already died of HCC or liver failure and were eliminated from the study population. Another possibility is that HCC may develop in patients with relatively low liver stiffness when infection has lasted a long time.
In the present study, the median increase in liver stiffness was calculated as 0.275 kPa per year. Using 13.01 kPa as a cut-off for cirrhosis[28
], it will take around 40 years on average to develop cirrhosis, which is consistent with previous reports based on liver biopsy[29
]. Admittedly, the present study is basically cross-sectional, and prospective longitudinal LSM will be obviously superior in understanding the natural course of liver fibrosis progression. However, the estimated average increase rate of liver stiffness indicates that such studies will require repeated LSM at an interval of at least five years.
Age at viral infection, alcohol consumption, and male gender were reported to be associated with accelerated fibrosis progression[8-11
]. In the present study, we performed subgroup analysis and indeed found that blood transfusion at an age older than 40, male gender, and alcohol consumption more than 50 g ethanol/d were significantly associated with rapid increase in liver stiffness. There is consensus that heavy alcohol consumption is associated with fibrosis progression[30
]. Alcohol, which by itself can cause liver disease and fibrosis, may affect liver stiffness and worsen fibrosis in hepatitis C[31
]. We did not find a difference in liver stiffness increase rate between HCV genotypes 1 (mostly 1b) and 2 (2a/2b), although we could not evaluate genotypes 1a, 3 or 4.
This study has some limitations. First, since this is a cross-sectional study performed after LSM became available, patients with more rapid disease progression may have died and been excluded from the study. Second, because transfusion-associated HCV infection has been virtually eliminated in Japan since 1992, we could not include patients with shorter duration of infection. Lastly, we did not perform paired LSM but assumed that liver stiffness was normal at the time of infection. Longitudinal observation is now on-going but will take several years to obtain results.
In conclusion, although liver stiffness gradually increases over time from the onset of infection in general, HCC develops in patients with high liver stiffness regardless of the duration of infection. Patients who acquired HCV infection at older ages showed higher increase rate of liver stiffness and probably more rapid disease progression.