Late liver-related mortality from complications of transfusion-acquired hepatitis C

Authors


  • Potential conflict of interest: Nothing to report.

Abstract

Although several cohort studies have been reported in individuals with chronic hepatitis C virus (HCV) infection, little is known about liver-related mortality among the elderly. We conducted a cohort study in 302 patients with tuberculosis sequelae who had received a blood transfusion at a young age and had subsequently been treated at a chest clinic. The cohort consisted of 147 patients with antibody to HCV (anti-HCV), of whom 81% were positive for HCV RNA, and 155 without anti-HCV. The cohort was followed for a mean duration of 5.7 years. There were no differences between the two groups in the mean age of the patients at the time of transfusion (31 vs. 34 years) or at the time of entry into the study (65 vs. 66 years). The outcome of 143 patients with, and 145 without, anti-HCV could be traced; 92 (64%) and 82 (57%) had died, respectively. The main cause of death was tuberculosis sequelae in 61 (42%) and 66 (46%) patients, respectively. Eight (6%) of the 143 patients with anti-HCV died of liver disease (hepatocellular carcinoma: seven; rupture of varices: one). The average annual mortality from liver disease from study entry in the patients with anti-HCV was 9.8 per 1,000 person-years. The patients with anti-HCV had a significantly lower cause-specific survival probability for liver disease (92% vs. 100% at 10 years, P < .005). In conclusion, in our study, liver-related mortality appeared to be high among elderly HCV-infected individuals. (HEPATOLOGY 2005;41:819–825.)

The clinical features and prognosis of individuals with chronic hepatitis C virus (HCV) infection vary widely. Some persons suffer from chronic progressive liver disease, which may eventually develop into cirrhosis and hepatocellular carcinoma (HCC).1–5 Others show persistently normal levels of serum alanine aminotransferase (ALT) and remain without clinical symptoms.6, 7 It is difficult to determine the prognosis of persons with chronic HCV infection.. Although a limited number of cohort studies have been reported,8–14 most have been conducted in subjects whose average age was 45 years or younger8–13; the prognosis for individuals in the older population remains unclear.

To evaluate the impact of HCV infection on the liver-related mortality in the older population, we conducted a study in a cohort of patients aged 52 years and older, diagnosed with tuberculosis sequelae and positive for antibody to HCV (anti-HCV), who had received a blood transfusion at a young age. As a control, we used a group of tuberculosis sequelae patients who were negative for anti-HCV. The condition tuberculosis sequelae refers to any complication related to pulmonary tuberculosis that has appeared after the infection has been cured, such as pulmonary dysfunction, cor pulmonale, or pulmonary mycosis. In general, these conditions are much more serious in patients who have undergone chest surgery than in those who have not. In Japan, most blood for transfusion was obtained commercially between 1951 and 1967, which is known to have frequently caused hepatitis.15, 16

Abbreviations

HCV, hepatitis C virus; HCC, hepatocellular carcinoma; ALT, alanine aminotransaminase; EIA, enzyme-linked immunosorbent assay; RIBA, recombinant immunoblot assay.

Patients and Methods

Patients and HCV Infection.

A total of 328 tuberculosis sequelae patients, receiving care at the Department of Respiratory Disease in the Tokyo National Hospital (previously called the National Tokyo Sanatorium) between July 1989 and June 1995 had received surgery for pulmonary tuberculosis between 1946 and 1990. Of them, 306 patients who had received a blood transfusion at that time were enrolled in the study; four of the enrolled patients had serum positive for hepatitis B surface antigen and were excluded from the study.

All 302 patients were tested for anti-HCV by enzyme-linked immunosorbent assay (EIA), and HCV RNA was determined in those patients who tested positive for anti-HCV. Then, in patients who tested negative for HCV RNA, anti-HCV was measured by recombinant immunoblot assay (RIBA). We regarded the patients as truly possessing anti-HCV if they were positive for HCV RNA or anti-HCV in serum was detected by RIBA. These patients were considered to be currently infected with HCV at high or low level, or to have been infected in the past; they were named HCV-infected patients. Mortality was compared between 147 patients with anti-HCV and 155 patients without anti-HCV (they were named HCV non-infected patients). Demographic characteristics of these two patient groups are shown in Table 1.

Table 1. Characteristics of Patients With Tuberculosis Sequelae With or Without Anti-HCV
FeaturesAnti-HCV(+) (n = 147)Anti-HCV(−) (n = 155)Differences
  • NOTE: Anti-HCV(+): One hundred nineteen HCV RNA–positive patients and 28 anti-HCV by recombinant immunoblot assay (RIBA)–positive patients. Anti-HCV(−): Patients who were negative for anti-HCV by ELISA or RIBA in serum. Mean ± SD is shown for continuous variables.

  • *

    Both commercial and donated blood was used for transfusion.

  • Most blood was obtained commercially.

  • Almost all blood was provided by volunteer donor.

  • §

    The prevalence of anti-HCV was significantly higher in patients who received transfusion between 1951 and 1967 than in those receiving it after 1967 (54% (128/235) vs. 24% (10/41), P < .0005).

  • ALT values were within normal limits (≤34 IU/L) in 59 (49%) of 119 HCV RNA-positive patients.

Male98 (67%)88 (57%) 
Age (years) (range)65.2 ± 6.2 (52–83)65.7 ± 7.4 (44–85) 
 40–4904 (3%) 
 50–5923 (16%)20 (13%) 
 60–6988 (60%)86 (54%) 
 70–7934 (23%)38 (25%) 
 80–892 (1%)7 (5%) 
Age at transfusion (range)31.4 ± 8.1 (15–62)33.5 ± 11.5 (13–64) 
Period after transfusion until study entry (years) (range)33.7 ± 6.5 (8–47)32.1 ± 10.2 (2–48) 
Year of transfusion (range)1959 ± 6.3 (1946–1982)1961 ± 9.9 (1946–1990) 
 <1950*9 (6%)17 (11%) 
 1951–1967§128 (87%)107 (69%) 
 >1968§10 (7%)31 (20%) 
Alcohol consumption (>50 g/day)6 (4%)5 (3%) 
ALT (IU/L)38.8 ± 35.316.3 ± 15.1P < .0001
 ≤3485 (58%)146 (94%) 
 35–6841 (28%)7 (5%) 
 ≥6921 (14%)2 (1%) 
Albumin (g/100 mL) (range)4.04 ± 0.40 (2.8–4.9)4.10 ± 0.43 (2.4–4.9) 
Platelets (×104/mm3) (range)18.6 ± 5.9 (7.2–35.1)22.7 ± 5.9 (11.2–43.2)P < .0001

Entry.

To be eligible for the study, patients had to have tuberculosis sequelae and have received a blood transfusion at the time of surgery for pulmonary tuberculosis. Clinical records retained at the clinic were used to make the initial selection of tuberculosis sequelae patients who had had an operation for pulmonary tuberculosis. Subsequent clinical interviews ascertained whether the selected patients had received a blood transfusion at the time of their operation, later, or not at all. The amount of alcohol a patient consumed was also noted. Patients with a history of 10 years or more of drinking more than 50 g of alcohol per day were defined as heavy drinkers. The entry into the study was defined as the time when the assessments of these covariates had been completed.

Of the 147 patients with anti-HCV, 12 had already been receiving care at the liver clinic in our hospital at the time of their entry into the study. All 12 had been referred to the chest clinic prior totheir care at the liver clinic. An additional 44 patients were referred to the liver clinic at least once after enrollment.

Outcome.

Patients were followed-up until October 2002, when mortality was compared between the patients with (N = 147) and without (N = 155) anti-HCV. Prognosis and cause of death could be confirmed for 288 patients by clinical records (261 patients), by telephone questionnaires to them or their family members (19 patients, of whom 10 had died), or by telephone questionnaires to the last physicians in charge of these patients, who were asked to confirm clinical records (8 patients, all of whom had died). Information could not be obtained for the remaining 14 patients; the end points for the observation of them were the last days they saw the doctors in our hospital, and at that time they were censored. We questioned family members of the 10 patients on cause of death recorded on a death certificate. Regarding the 27 patients, or their closest kin, on whom information was obtained through telephone questionnaires, they all understood the purpose of the study and gave their informed consent. The cause of death was classified into three categories: (a) tuberculosis sequelae, (b) liver disease, and (c) other diseases. Tuberculosis sequelae that led to death consisted mainly of respiratory failure, and to a lesser extent suffocation by hemoptysis. Liver disease that caused death was HCC and severe complications of cirrhosis. Two patients with anti-HCV received interferon therapy 2 or 4 years after the entry: one was a nonresponder, and another was a sustained virological responder. No other patients received antiviral treatments.

Both the overall and the cause-specific survival curves were calculated. In calculating the cause-specific survival curve, for which the end point was liver-related death, deaths of tuberculosis sequelae or other diseases were censored.

Transfusion History.

Fifty patients had received an additional transfusion 1 year or more after their first, mostly because of repeated surgery. The mean period between the first and the last transfusion was 8.5 ± 6.1 years (1-26 years), and for these patients the period after transfusion was measured from the year in which they received their first transfusion. Of these 50 patients, 40 (80%) received their first transfusion between 1951 and 1967.

Markers of HCV Infection.

Anti-HCV was tested by EIA(Abbott HCV EIA 2nd Generation, Abbott Japan, Tokyo, Japan) and RIBA (Chiron HCV RIBA test 3rd Generation, Ortho-Clinical Diagnostics, Tokyo, Japan), and hepatitis B surface antigen by passive hemagglutination (Mycell, Institute of Immunology Co., Ltd., Tokyo, Japan). Serum samples, which were kept frozen at −20°C, were tested for HCV RNA within 3 years after collecting the blood. Nucleic acids were extracted from 100 μL of serum by the guanidine-thiocyanate-phenol-method17 and reverse-transcribed to complementary DNA, which was then amplified by a two-stage polymerase chain reaction with nested primers deduced from the well-conserved 5′-noncoding region of the HCV genome.18

Statistical Analysis.

Differences in baseline characteristics between the groups were evaluated by the chi-square test for dichotomous variables and by the Wilcoxon rank-sum test for continuous variables. Kaplan–Meier survival curves were calculated and compared using the log-rank test. P values less than .05 were considered to be significant.

Results

HCV Infection in Patients With Tuberculosis Sequelae.

Anti-HCV by EIAwas detected in 162 (54%) of the 302 patients, of which 119 were positive for HCV RNA. Of the 43 anti-HCV–positive patients who were positive by EIA but HCV RNA negative, 28 were positive for anti-HCV by RIBA. Therefore, 147 (49%) of the 302 patients were positive for HCV RNA or anti-HCV by RIBA, and considered to truly possess anti-HCV antibody. Of the 147 patients, 119 (81%) were positive for HCV RNA; this corresponded to 39% of all patients. The remaining 155 patients were negative for anti-HCV by EIA or RIBA. No differences were observed between the anti-HCV–positive (n = 147) and –negative patients (n = 155) with respect to age, sex, age at first transfusion, year of transfusion, time after transfusion until study entry, or in the percentage of heavy drinkers (Table 1). In the anti-HCV–positive patients, the mean ALT value was significantly higher than in the anti-HCV–negative patients (38.8 IU/L ± 35.3 vs. 16.3 IU/L ± 15.1, P < .0001); nevertheless, values were within normal limits (≤34 IU/L) in 58% of the anti-HCV–positive patients and 49% of the HCV RNA–positive patients. There were no differences between groups in mean albumin value; the platelet count was significantly lower in the patients with anti-HCV than in those without (P < .0001).

Seventy-eight percent of 302 patients (87% of anti-HCV–positive patients and 69% of anti-HCV–negative patients) received transfusions between 1951 and 1967. The prevalence of anti-HCV was significantly higher in patients who had received transfusions before 1967 than in those who had received it after 1967 (54% (128/235) vs. 24% (10/41), P < .0005).

Outcome of the Patients at the End of Follow Up.

Outcome at the end of follow-up was compared between the patients with and without anti-HCV; 143 (97%) patients with and 145 (94%) without anti-HCV completed the follow up. The duration of follow-up (measured from date of entry until death or lost to follow-up, or October 2002, whichever came first) did not differ significantly between the two groups (Table 2). Of the 288 patients who completed follow-up, 92 patients (62%) with anti-HCV and 82 (53%) without had died. Among the anti-HCV–positive patients, the cause of death was tuberculosis sequelae in 61 (42%), whereas 23 (16%) died of other diseases. Among the anti-HCV–negative patients, 66 (46%) died of tuberculosis sequelae and 16 (11%) of other diseases. Eight (6%) patients with anti-HCV died of liver disease, which accounted for 26% of the deaths not related to tuberculosis sequelae; none of the anti-HCV–negative patients died of liver disease. The average annual mortality from liver disease from study entry in anti-HCV–positive patients was 9.8 per 1,000 person-years.

Table 2. Outcomes of Patients With Tuberculosis Sequelae With or Without Anti-HCV
 Anti-HCV(+) (n = 147)Anti-HCV(−) (n = 155)
  1. NOTE: Anti-HCV(+): One hundred nineteen HCV RNA–positive patients and 28 anti-HCV by recombinant immunoblot assay (RIBA)–positive patients. Anti-HCV(−): Patients who were negative for anti-HCV by ELISA or RIBA in serum.

Duration of follow-up (years)
 Mean ± SD (range)5.5 ± 3.3 (0.2–13.3)5.9 ± 3.4 (0.1–13.3)
Alive51 (36%)63 (43%)
Dead92 (64%)82 (57%)
 Due to tuberculosis sequelae61 (42%)66 (46%)
 Due to liver disease8 (6%)0
 Other causes23 (16%)16 (11%)
Unknown4 (3%)10 (6%)

There was no significant difference between the 2-, 5-, and 10-year overall survival probabilities from study entry of the patients with anti-HCV (84%, 60% and 35%, respectively), compared with those of the patients without anti-HCV (85%, 66% and 44%, respectively) (P =.12) (Fig. 1). The patients with anti-HCV, however, had significantly lower cause-specific survival probabilities for liver disease from study entry than did those without it: 99%, 96%, and 92%, at 2, 5, and 10 years, respectively (P < .005) (Fig. 2).

Figure 1.

The cumulative overall survival curves from study entry in anti-HCV–positive patients (solid line) and the anti-HCV–negative controls (dotted line). The 2-, 5-, and 10-year overall survival probabilities for the subjects were 84%, 60%, and 35%, respectively, and those for the controls were 85%, 66%, and 44%, respectively. The differences between groups was not significant (P =.12). Abbreviation: anti-HCV, antibody to HCV; NOTE: Anti-HCV(+): HCV RNA positive or anti-HCV by recombinant immunoblot assay (RIBA) positive patients. Anti-HCV(−): Patients who were negative for anti-HCV by EIA or RIBA in serum.

Figure 2.

The cause-specific survival curves for liver disease from study entry in the anti-HCV–positive patients (solid line) and the anti-HCV–negative controls (dotted line). The anti-HCV–positive patients showed significantly lower cause-specific survival probabilities for liver disease (P < .005), 99%, 96%, and 92% at 2, 5, and 10 years, respectively, than the controls. Abbreviation: anti-HCV, antibody to HCV; NOTE: Anti-HCV(+): HCV RNA positive or anti-HCV by recombinant immunoblot assay (RIBA) positive patients. Anti-HCV(−): Patients who were negative for anti-HCV by EIA or RIBA in serum.

Of the liver-related deaths in the eight patients with anti-HCV (six men, two women), seven were caused by HCC and one by massive bleeding from esophageal varices (Table 3). All of the patients were positive for HCV RNA and showed abnormal ALT values at entry. None had a history of heavy alcohol use. They had received their transfusions at the ages of 25 to 41 years, had been diagnosed with HCC at 65 to 75 years, and died at 66 to 76 years. The period between transfusion and diagnosis of HCC was 33 to 45 years. Of the seven patients with HCC, two died within 1 year of diagnosis, two within 2 years, and two within 3 years. The remaining patient with HCC, who had been treated successfully with percutaneous ethanol injection therapy, died of recurrent HCC after 7 years. Seven of the eight patients who died of liver disease had been diagnosed with cirrhosis by histological and/or clinical examinations.

Table 3. Patients With Anti-HCV Who Died of Liver Disease
Case No.SexAge atHCV RNAALT (IU/L)Esophageal VaricesLiver HistologyCauses of Death
TransfusionDiagnosis of HCCDeath
  1. NOTE: ALT at the entry is shown.

  2. Abbreviations: F, Female; M, Male; HCV RNA, hepatitis C virus ribonucleic acid; HCC, hepatocellular carcinoma.

1F256566(+)72(+)UnknownHCC
2F267173(+)59(+)CirrhosisHCC
3M276666(+)37(+)UnknownHCC
4M316471(+)71(+)UnknownHCC
5M336870(+)68(−)CirrhosisHCC
6M357576(+)139(−)CirrhosisHCC
7M3670(+)95(+)CirrhosisVarices rupture
8M417575(+)41UnknownUnknownHCC

Details of the causes of death of patients from “other diseases” are shown in Table 4.

Table 4. Details of the Reasons for Death in Patients Who Died of “Other Causes”
Cause of DeathAnti-HCV(+)Anti-HCV(−)
  1. NOTE: Anti-HCV(+): HCV RNA positive or anti-HCV by recombinant immunoblot assay (RIBA) positive patients. Anti-HCV(−): Patients who were negative for anti-HCV by ELISA, or RIBA in serum.

  2. Abbreviation: anti-HCV, antibody to HCV.

Cerebral vascular diseases43
Myocardial infarction11
Rupture of aortic aneurysm11
Renal failure21
Myelodysplastic syndrome10
Multiple organ failure01
Sudden death of unknown etiology20
Senile decay03
Accident10
Brain tumor10
Malignancy  
 Thyroid cancer01
 Lung cancer20
 Esophageal cancer10
 Gastric cancer21
 Colon cancer22
 Gallbladder cancer01
 Pancreatic cancer10
 Multiple myeloma10
 Malignant lymphoma11

Discussion

In Japan, the surgical treatment of pulmonary tuberculosis, such as thoracoplasty or lobectomy, was common in the 1950s until the early 1960s; more than 20,000 patients per year underwent such surgeries during this time.19 These operations required a large volume of blood by transfusion.15 Between 1951 and 1967, such blood was obtained commercially. It has been documented that hepatitis was acquired posttransfusion in up to 50% to 80% of patients who had undergone a major operation that involved the use of commercial blood, including those undergoing surgery for pulmonary tuberculosis.15, 16 We could not examine other risk factors for HCV infection than transfusion, so we are unable to completely discount potential routes such as sexual exposure, medical procedures, reused needles, or intravenous drug use. However, judging from the results that 78% of the patients received transfusion between 1951 and 1967, and the prevalence of anti-HCV was significantly higher in patients who received transfusion at that time than in those receiving it after 1967, it is reasonable to assume that, most of the patients were infected with HCV via a blood transfusion. Furthermore, it is not surprising that 49% of our patients with tuberculosis sequelae were positive for anti-HCV antibody and 39% for HCV RNA; the prevalence of the HCV-RNA–positive patients might be underestimated, because their samples were stored only at −20°C.

In the United States, patients with chronic HCV infection die more frequently of decompensated cirrhosis than of HCC.20 By contrast, in Japan, HCC has been the major cause of death in this patient group for a long time. This trend is growing increasingly stronger, and the proportion of death by HCC among all causes of death in patients with cirrhosis and chronic HCV infection has reached 81%.21 Hamada et al.22 reported that the age of the patient and duration of infection were independent risk factors affecting the development of HCC, and of them, age was the more significant. On average, approximately 30 years pass between the time of transfusion and the time of diagnosis of HCC.1, 3, 22 According to a nationwide survey in Japan,23 the mean age at such diagnosis was 63.0 years for males and 66.5 years for females (75% of all patients were anti-HCV–positive). Ninety-two percent of HCC patients were 60 years or older at the time they were diagnosed with HCC.22 In the west of Scotland, age-specific incidence of HCC in men aged over 55 years increased dramatically between 1975 and 1985, particularly among those aged 75 to 84 years, but not in those younger than 55 years; the major etiology was considered to be HCV infection.24 The patients in our study survived an average of 34 years after having received blood transfusion in early adulthood (at a mean age of 31 years); at the time of entry into the study they had become relatively elderly (average age, 65 years). Accordingly, they were considered to be at high risk for carcinogenesis in view of age and duration of infection.

When investigating the prognosis of liver disease in HCV-infected persons, it would be preferable to start studying patients at the time of infection and continue until they reach old age several decades later, when the risk becomes high for developing HCC. However, it is difficult to carry out such a prospective study. A retrospective cohort study on older HCV-infected persons, for whom a considerable period has elapsed between the presumed time of their infection and their enrollment into the study, and who were selected with as little bias as possible, could serve as a good alternative to clarify, relatively accurately, the eventual outcome of their liver disease, although the study would miss liver-related events that occurred before entry. Previous retrospective studies identified patients referred for liver disease.1–5 The patients in our study were recognized at a chest clinic while receiving treatment of sequelae to pulmonary tuberculosis, which had placed them at risk for HCV infection through transfusion. Thus, the advantage of this study was that selection bias would be less likely. Moreover, only two patients received antiviral treatment.

In our study, eight of the anti-HCV–positive patients and none of the anti-HCV–negative patients died of liver disease. The average annual mortality from liver disease from study entry of HCV-infected patients was 9.8 per 1000 person-years, and the cause-specific survival probabilities for liver disease from study entry were 99%, 96% and 92% at 2, 5 and 10 years, respectively. The mortality from liver disease might be underestimated, because chronic illnesses, including tuberculosis sequelae, might confound recognition of liver disease complication, and in some patients the cause of death was obtained from death certificates.

We showed by univariate analysis that the cause-specific survival probability for liver disease was significantly lower in patients with anti-HCV than in those without anti-HCV antibody. Subsequently, we should examine survival by using a Cox proportional hazards regression analysis. Nevertheless, in our study, this analysis was inappropriate, because none of the patients without anti-HCV died of liver disease. However, the patients in both groups were selected on the same conditions, and the background was not different between the two groups, as shown in Table 1. We could therefore conclude that HCV infection was probably an independent risk factor for the death of liver disease. Then, we showed that the overall mortality from study entry was not significantly different between the two patient groups. Because the overall mortality was extremely high owing to a large number of deaths from tuberculosis sequelae in both groups, the impact on mortality of liver-related deaths resulting from HCV infection would have been underestimated. In fact, of the deaths unrelated to tuberculosis sequelae, death from liver disease accounted for 26%.

Many of the patients in our study who died of liver disease died due to HCC, as the previous report from Japan had shown.21 All the patients were positive for HCV RNA. None of the patients positive for anti-HCV and negative for HCV RNA died of liver disease, and there was no excess liver-related mortality in those patients. Although ALT values at entry were within normal limits in 49% of the HCV RNA–positive patients, no patient who died of liver disease showed normal ALT values.

Several studies reporting on the prognosis of liver disease in cases of chronic HCV infection have looked at patients infected in childhood (average age, 0-8 years)8, 9 or early adulthood (average age, 19-28 years).10–13 At a median or average of 14 to 35 years after infection, clinical or histological cirrhosis was found in 0% to 8% of patients,8–13 and no or mild histological fibrosis in 81% to 87%.8, 9, 11, 12 End-stage liver disease developed with an incidence of 3.1 per 1,000 person-years,13 and liver-related mortality was 0 to 0.4%,8, 10 with a much better prognosis than was seen in our study. The difference is likely to be attributable to the difference in age of the patients at the time of the investigations. The patients in these studies were young, with an average age of 20 to 45 years. Our study showed that the eight patients who died of liver disease had apparently been infected in early adulthood (25-41 years) and died after they had reached old age (66-76 years). These results suggest that for patients who were infected in early adulthood, the long-term prognosis of liver disease, once they reach old age, is not good. It has been reported that fibrosis begins to accelerate at 50 years of age25 and that the evolution from chronic hepatitis to cirrhosis occurs more frequently and rapidly in patients aged 50 years or older than in those younger than 50 years.26 Accordingly, young individuals infected with HCV, and having a favorable course, may undergo rapid progression of the fibrosis once they reach middle age and then develop severe liver disease, including HCC, when they reach old age.

Seeff et al.14 reported the long-term mortality over approximately 25 years in 222 patients with posttransfusion hepatitis C, with an average age of 49 years, of whom approximately 77% were considered to have been positive for HCV RNA in serum. Their report is the only study that includes a control group and also deals with individuals who had reached old age at the end point of observation. Liver-related mortality was significantly higher among the cases than among the control group of matched, transfused, and nonhepatitis patients (4.1% vs. 1.3%, respectively). However, the all-cause mortality was not different between the two groups (67.1% vs. 65.0%, respectively). Furthermore, the mortality attributed to chronic hepatitis C infection was only about 3%, and liver disease was considered to be a relatively minor cause of patient death. Our study showed distinctly high mortality from liver disease compared with that observed in their study, although both studies dealt with older patients. The reason for the differences in the prognosis between their study and ours is unclear. It might be related to the difference in duration of infection, which was apparently longer in our patients than in theirs. In our patients, past tuberculosis infection, past treatment for tuberculosis, or tuberculosis sequelae might contribute to more rapid progression of liver disease. Another influencing factor might be differences in race, and this aspect warrants further study.

In conclusion, for the 147 HCV-infected patients (average age, 65 years), of whom 81% were positive for HCV RNA, with tuberculosis sequelae who had received blood transfusion at a younger age, liver-related mortality from study entry was high at 9.8 per 1,000 person-years. Among the deaths unrelated to tuberculosis sequelae, death of liver disease was the most frequently reported cause.

Ancillary

Advertisement