| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |


*
Department of Health Science, Tokyo Medical and Dental University,
Bunkyo-ku, Tokyo, Japan.
Department of Blood Purification, Tokyo Medical and Dental University,
Bunkyo-ku, Tokyo, Japan.
Second Department of Internal Medicine, Tokyo Medical and Dental
University, Bunkyo-ku, Tokyo, Japan.
Correspondence to Dr. Chifumi Sato, Department of Health Science, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8519, Japan. Phone: 81-3-5803-5335; Fax: 81-3-5803-0152; E-mail: c.sato.ns{at}tmd.ac.jp
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Development of hepatocellular carcinoma and liver cirrhosis is a serious complication of HCV infection (5,6). Interferon therapy is effective for chronic hepatitis C, but at best, 20% of patients are cured (7). A recent study in patients with normal renal function showed that interferon treatment lowered the rate of the occurrence of hepatocellular carcinoma by a factor of 2 (8), whereas another study reported that the therapy improved the survival but did not reduce the risk of liver disease in chronic hepatitis C (9).
The therapy has also been applied to hemodialyzed patients (10,11), but several limitations compromise its usefulness for these patients: Interferon is expensive (12), liver biopsy for definite diagnosis is reported to be risky for hemodialyzed patients because of their tendency of bleed (13), the therapy may cause serious adverse effects (14), and, most important, the natural history of HCV infection in hemodialyzed patients is unclear. Therefore, we conducted a prospective cohort study to assess whether HCV infection influences the mortality of patients who are on chronic hemodialysis.
| Materials and Methods |
|---|
|
|
|---|
Using that cohort population, we designed a prospective cohort study of patients who were on chronic hemodialysis. Among the 23 dialysis centers, 2 were excluded because agreements of cooperation could not be obtained, 2 were excluded because they were located too far from the metropolitan area to permit efficient collection of precise data, and 3 were excluded because the basal demographic data were not complete. Finally, 1470 patients (68.9% of the original study) from 16 centers were enrolled in the present study. Natural histories and various medical data on the patients were obtained from each dialysis center by referring the survey sheets of each dialysis center to the Statistic Survey Committee of the Japanese Society for Dialysis Therapy.
The end point of the study was death of the patient. The development of hepatocellular carcinoma or liver cirrhosis was also investigated.
Laboratory Methods and Medical Diagnosis
Serum anti-HCV antibodies and anti-hepatitis B surface (HBs) antibodies
were determined using the second generation anti-HCV assay kit (Abbott RIA
kit, Abbott Laboratories, IL; or Ortho EIA kit, Ortho Diagnostic Systems,
Ratian, NJ) and an anti-HBs assay kit (Abbott RIA kit), respectively. Blood
samples for assays of urea nitrogen creatinine, aspartate aminotransferase
(AST), alanine aminotransferase (ALT), and albumin were obtained before
routine hemodialysis procedures. Routine screening of liver disease was
performed using ultrasonographic examination as previously reported
(16) at 6-mo intervals.
Diagnosis of liver cirrhosis was based on physical findings (nodular liver,
esophageal varix, ascites), ultrasonographic findings (right lobe atrophy and
swelling of the left lobe, dullness of the edge and irregularity of the
surface, patchy patterns of the parenchyma, presence of ascites, splenomegaly
and/or extrahepatic shunts), and computed tomographic findings (irregularity
of the surface, presence of ascites, splenomegaly and/or extrahepatic shunts).
When serum
-fetoprotein levels were increased or a mass lesion was
observed by ultrasonography, the diagnosis of hepatocellular carcinoma was
confirmed either by dynamic computed tomography (CT), angiography,
CT-angiography, and CT-portography or by needle liver biopsy.
Causes of death were classified in accordance with the cause of death code used in the annual survey of the Japanese Society for Dialysis Therapy, a coding system that recently added hepatocellular carcinoma as a category (17).
Statistical Analyses
In the comparison between anti-HCV-antibody-positive and -negative
patients, the t test was used for continuous variables and the
Pearson's test and Fisher's exact test were used for discrete variables.
Mortality was estimated by the life-table method. The log-rank test was used
to assess the cumulative survival of the patients. A Cox proportional hazard
model was used to evaluate the effects of independent predictors
(HCV-antibodies, age, duration of hemodialysis, sex, and diabetes) on
patients' survival. Computations were performed using SPSS (SPSS Inc.,
Chicago, IL). P values less than 0.05 were considered statistically
significant.
| Results |
|---|
|
|
|---|
|
Mortality
During the study period, only one patient (HCV-positive group) was lost to
follow-up. There were six patients (one in the HCV-positive group and five in
the HCV-negative group) who received a transplant and five (one in the
HCV-positive group and four in the HCV-negative group) who were changed to
continuous ambulatory peritoneal dialysis treatment. All of them were alive at
the end of the study. Cumulative survival of the HCV-positive and -negative
groups is shown in Figure 1.
During the 6-yr follow-up period, 33.0% of the HCV-positive group and 23.2% of
the HCV-negative group died (P < 0.01,
Table 2). When the mortality
was assessed by age class or duration on hemodialysis, similar results were
obtained (Table 2). A Cox
proportional hazard model performed to exclude bias of unadjusted examination,
a factor that may mask true effects, showed that diabetes (relative risk [RR],
2.00; 95% confidence interval [CI], 1.57 to 2.54), age (increase of RR per
decade, 1.71; 95% CI, 1.57 to 1.86), and the presence of anti-HCV antibodies
(RR, 1.57; 95% CI, 1.23 to 2.00) were independent risk factors of death
(Table 3).
|
|
|
The causes of death are listed in Table 4. Although most of the causes were not significantly different between the HCV-positive and -negative groups, the prevalences of hepatocellular carcinoma and liver cirrhosis were strikingly different. Hepatocellular carcinoma accounted for 5.5% of all deaths in the HCV-positive group and no deaths in the HCV-negative group (P < 0.001), whereas liver cirrhosis was present in 8.8% and 0.4% of patients who died in the HCV-positive and -negative groups, respectively (P < 0.001). Among those who died of liver cirrhosis in the HCV-positive group, one patient died of esophageal varix rupture and the other seven died of hepatic failure. The one patient who died of liver cirrhosis in the HCV-negative group was positive for HBs antigen, but the eight patients who died of liver cirrhosis in the HCV-positive group were all negative for HBs antigen. Among 13 deaths caused by infectious disease in the HCV-positive group, 1 died of sepsis, but this patient did not have liver cirrhosis.
|
At the end of this study, hepatocellular carcinoma was found in 4.8% and 0.0% in the HCV-positive and -negative groups, respectively (P < 0.001), and liver cirrhosis was found in 18.1% and 1.1% in the former and latter, respectively (P < 0.001). During the 6-yr follow-up, 8 patients (3.5%) and 0 patients (0.0%) developed hepatocellular carcinoma in the HCV-positive and -negative groups, respectively, and 30 patients (13.2%) and 5 patients (0.5%) developed liver cirrhosis in the former and latter, respectively (Table 5). Among the eight patients who developed hepatocellular carcinoma in the HCV-positive group, two were positive for anti-HBs antibodies, two were negative for both anti-HBs antibodies antihepatitis B core (HBc) antibodies, and the remaining four were negative for anti-HBs antibodies but the anti-HBc antibody status was unknown.
|
| Discussion |
|---|
|
|
|---|
Mortality during the 6-yr follow-up period was significantly higher in the HCV-positive group than in the HCV-negative group. The same difference was also obtained when the groups were further divided according to age or duration of hemodialysis. A Cox hazard examination performed to assess the possible factors other than anti-HCV antibodies that affect the mortality showed high relative risks for death in diabetic and aged patients, as well as in patients with high titers of anti-HCV antibodies. Diabetes and age are known to influence the mortality in dialysis patients (25,26). Our data are also in agreement with those of Stehman-Breen et al. (27) in their recent report that showed that the relative risk for death in anti-HCV-antibody-positive patients was significantly higher than that in anti-HCV-antibody-negative patients after adjusting for age, transplantation, duration of dialysis, and race.
We then examined whether the development of liver failure by HCV infection affects the cause of death in hemodialyzed patients. The primary causes of death of hemodialyzed patients are cardiovascular disease, cerebrovascular disease, infection, and cancer (28,29). In our study, these four diagnoses were the more frequent causes of death for 65% of patients. A previous study has shown the significance of sepsis as a cause of death in HCV-positive patients with renal transplantation (30), but only one patient died of sepsis in the present study. The discrepancy seems to be due to the difference in the number of patients who received renal transplantation. In the present study, only one patient received renal transplantation in the HCV-positive group, and he was alive. Patients who have received renal transplantation frequently are administered immunosuppressants, which places them in an immunocompromised state. In fact, a high prevalence of infection in the causes of death has been reported in patients with renal transplantation irrespective of HCV status from the same group (31). Alternatively, differences in the criteria for sepsis might affect some results. In the present study, sepsis was diagnosed when blood culture was positive.
A remarkable difference between the two groups was the percentage of hepatocellular carcinoma and liver cirrhosis as a cause of death. Approximately 15% (13 of 91 patients) of the HCV-positive group died of one of these two causes, in contrast with only 0.4% (1 of 277 patients) of the HCV-negative group. Moreover, only one patient in the negative group died of hepatic cirrhosis. At patients entry into the study, hepatocellular carcinoma and hepatic cirrhosis had already developed in more patients in the HCV-positive group (3 and 11 cases) than in the HCV-negative group (0 and 6 cases). After entry into the study, more positive patients developed these two hepatic diseases de novo. Therefore, both hepatocellular carcinoma and hepatic cirrhosis contribute to the high mortality in the HCV-positive group.
In the present study, eight subjects in the group of anti-HCV-antibody-positive patients developed hepatocellular carcinoma (0.6%/yr). This value is lower than the incidence of 1.2% per year previously found in nonhemodialyzed patients with chronic hepatitis (32). Because liver biopsy was not performed at entry of the study, it is possible that some patients in the HCV-positive group may have included healthy carriers, and this would at least partially explain this difference. Alternatively, hemodialysis patients may have died before the development of hepatocellular carcinoma. The mortality in the present study was low compared with that in a study by Stehman-Breen et al. (27). Mortality of hemodialyzed patients is reported to be low in Japan (33,34). In the United States, mortality recently decreased, presumably secondary to changes in hemodialysis treatment protocols (35). Because hepatocellular carcinoma and liver cirrhosis develop slowly, their relative frequency as causes of death may increase as overall mortality decreases. In fact, an increased prevalence of hepatocellular carcinoma in the general population has recently been reported in the United Kingdom and the United States (36,37). There is also a report that the risk of liver cancer is higher in hemodialysis patients than in the general population (38).
The present study has several limitations. First, we did not know when the patients were infected with HCV. Because no anti-HCV antibody assay system was available until 1989 (2), HCV infection could not be screened at entry into hemodialysis, which made it impossible to ascertain the period between contraction of the HCV infection to the development of hepatocellular carcinoma or liver cirrhosis, or how the duration of infection affected the mortality.
Second, because we assessed anti-HCV antibodies only at entry, anti-HCV-antibody-negative patients might have become positive during the study period, and vice versa. Kobayashi et al. (39) recently reported that 5% of initially negative patients became positive in a 5-yr follow-up period. If the same situation had occurred in our study, then the difference in the mortality between the anti-HCV-positive and -negative groups would have been amplified. They also reported that 8.1% of their initially positive patients became negative. Even though they might have become negative during the study period, we believe that our positive patients were infected with HCV before entry into the study and were different from those who had no history of HCV infection at entry.
Third, because of the prospective nature of the present study, a second-generation assay was used for anti-HCV antibodies, because third-generation anti-HCV antibody assays were not available in 1993. Although the third-generation assays are useful in detecting the antibodies in an early phase of acute infection, the sensitivity of the second-generation assays is reported to be more than 95% in the chronic phase of infection (40). Therefore, the study outcome will have been minimally affected at best.
Because of the prospective nature of the study, we were also unable to obtain information on potential confounding factors affecting hepatocarcinogenesis in patients with chronic hepatitis C and normal renal function, e.g., alcohol consumption, latent HBV infection, increased hepatic iron content. The role of alcohol on hepatocarcinogenesis caused by HCV infection is controversial; therefore, it is impossible to rule out the possibility that consumption of alcohol may have facilitated the occurrence of hepatocellular carcinoma. However, it is unlikely that only the anti-HCV-positive patients drank alcohol. Similarly, there are several controversial reports on the relation of chronic HCV infection and latent HBV infection in hepatocarcinogenesis. In the present study, two patients who developed hepatocellular carcinoma were positive for anti-HBs antibodies.
Some recent reports described a relationship between progression of HCV-related liver disease and hepatic iron content. Patients who are on chronic hemodialysis may have an excessive iron load as a consequence of frequent blood transfusions, but there are no good biochemical markers for hepatic iron content. Serum ferritin concentrations are not good markers of hepatic iron content because ferritin is then released from hepatocellular stores in patients with liver injury (41), including patients who have HCV infection and are on chronic hemodialysis (42).
More recent, an international cohort study has shown various cancer risks in patients who are on dialysis for end-stage renal disease (43). Deaths caused by hepatocellular carcinoma accounted for only 0.6 to 2.4% of total cancer deaths. In that study, however, 82 to 90% of the patients were on dialysis for less than 5 yr, a much shorter period than the patients in the present study. Furthermore, their average observation period was 2.5 yr, which was much shorter than the 6-yr observation period in the present study. It is widely known that hepatocellular carcinoma develops 20 to 30 yr after infection. Therefore, the frequency of hepatocellular carcinoma in that cohort may increase with a longer observation period.
In conclusion, HCV infection increases the risk of death in patients who are on chronic hemodialysis, and hepatocellular carcinoma and liver cirrhosis are responsible for this increase.
| Acknowledgments |
|---|
Drs. Yoshinori Sakai (Tsuchiura Kyodo Hospital) and Yuji Hoshino (Yokosuka Kyosai Hospital) are also acknowledged for their generous help during this study.
| References |
|---|
|
|
|---|
Hepatocellular Carcinoma Study
Group: Effect of interferon-
on progression of cirrhosis to
hepatocellular carcinoma: A retrospective cohort study.
Lancet 351:1535
-1539, 1998[Medline]
interferon treatment of non-A, non-B
(type C) hepatitis: Review of studies and recommendations for treatment.
J Hepatol 11:S72
-S77, 1990
This article has been cited by other articles:
![]() |
J. E. Trevizoli, R. de Paula Menezes, L. F. Ribeiro Velasco, R. Amorim, M. B. de Carvalho, L. S. Mendes, C. J. Neto, J. R. de Deus Macedo, F. de Assis, and R. Neves Hepatitis C Is Less Aggressive in Hemodialysis Patients than in Nonuremic Patients Clin. J. Am. Soc. Nephrol., September 1, 2008; 3(5): 1385 - 1390. [Abstract] [Full Text] [PDF] |
||||
![]() |
C-H Liu, C-C Liang, J-W Lin, S-I Chen, H-B Tsai, C-S Chang, P-H Hung, J-H Kao, C-J Liu, M-Y Lai, et al. Pegylated interferon {alpha}-2a versus standard interferon {alpha}-2a for treatment-naive dialysis patients with chronic hepatitis C: a randomised study Gut, April 1, 2008; 57(4): 525 - 530. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. van Leusen, R. P. R. Adang, R. A. de Vries, T. T. Cnossen, C. J. A. M. Konings, S. W. Schalm, and A. C. I. T. L. Tan Pegylated interferon alfa-2a (40 kD) and ribavirin in haemodialysis patients with chronic hepatitis C Nephrol. Dial. Transplant., February 1, 2008; 23(2): 721 - 725. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. A. Terrault and D. B. Adey The Kidney Transplant Recipient with Hepatitis C Infection: Pre- and Posttransplantation Treatment Clin. J. Am. Soc. Nephrol., May 1, 2007; 2(3): 563 - 575. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Kalantar-Zadeh, R. D. Kilpatrick, C. J. McAllister, L. G. Miller, E. S. Daar, D. W. Gjertson, J. D. Kopple, and S. Greenland Hepatitis C Virus and Death Risk in Hemodialysis Patients J. Am. Soc. Nephrol., May 1, 2007; 18(5): 1584 - 1593. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Kalantar-Zadeh, C. J. McAllister, and L. G. Miller Clinical characteristics and mortality in hepatitis C-positive haemodialysis patients: a population based study Nephrol. Dial. Transplant., August 1, 2005; 20(8): 1662 - 1669. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Jadoul, J.-L. Poignet, C. Geddes, F. Locatelli, C. Medin, M. Krajewska, G. Barril, E. Scheuermann, S. Sonkodi, and P. Goubau The changing epidemiology of hepatitis C virus (HCV) infection in haemodialysis: European multicentre study Nephrol. Dial. Transplant., April 1, 2004; 19(4): 904 - 909. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |