Projecting future
complications of chronic hepatitis C in the United States
|
| |
| |
Summary: Chronic hepatitis C virus (HCV)
infection is common and often results in slowly progressive
liver disease. Although acute hepatitis C is now uncommon,
most patients with acute infection have developed chronic
hepatitis, and, therefore, the pool of infected patients is
large. We used a modification of a previously described
natural history model for HCV infection to project the
number of cases of HCV infection, cirrhosis, and liver
failure over the next 40 years. The model estimated the
prevalence of HCV infection in the United States was 3.07 x
106 (3 million) in 1993 (compared with an adjusted National
Health and Nutrition Evaluation Survey (NHANES) III estimate
of 2.8 to 3.5 x 106). A gradual decline in the prevalence of
infection should occur by year 2040 because of aging and
natural deaths among the infected pool. However, as the
duration of infection increases in the surviving cohort, the
proportion with cirrhosis will increase from 16% to 32% by
2020 in an untreated population. Complications of cirrhosis
also will increase dramatically over the next 20 years:
hepatic decompensation (up 106%), hepatocellular carcinoma
(up 81%), and liver-related deaths (up 180%). Although
current treatment regimens eradicate HCV in over 50% of
cases, many more patients would need to be treated to
significantly impact disease progression. Identification and
treatment of every case of HCV infection (with or without
cirrhosis) would reduce the number of cases of decompensated
cirrhosis by almost half after 20 years. Despite the
declining incidence of acute HCV infection, chronic
hepatitis C is common. The prevalence of cirrhosis and the
incidence of its complications will increase over the next
10 to 20 years, because the duration of infection increases
among those with chronic hepatitis C. These data emphasize
the need for greater access to transplantation by expansion
of the donor pool, increasing use of split livers and living
donors, and novel options such as xenotransplantation.
(Liver Transplantation April 2003;9:331-338.)
editorial note: although the parargraph above refers only to
the need for greater access to transplantation, without
adequate access to testing, treatment, and care deaths and
sickness will similarly increase. Right now there is little
funding to support HCV testing, alcohol counseling,
counseling, education for doctors & patients, public
awareness, and expanding the care infrastructure to absorb
patients.
A population survey a decade ago estimated that nearly 4
million Americans have detectable antibody to the hepatitis
C virus (HCV). In fact, this number may be a significant
underestimate of the true prevalence because certain high
prevalence groups, such as prisoners and other
institutionalized persons, were not included in the survey.2
Most of those infected with HCV acquired the disease 10 to
20 years ago, before identification of the virus and the
availability of screening tests. The recognition of
potential risk factors and the improved safety of the blood
supply have led to a dramatic decrease in the incidence of
new HCV infections in recent years. However, the overall
prevalence of chronic infection has not fallen because most
acutely infected patients develop chronic infection. Because
liver disease caused by hepatitis C progresses slowly and
does not result in major morbidity for many years, we are
only now beginning to see the magnitude of the consequences
of chronic infection.
Patients with chronic hepatitis use health care resources in
obvious and direct ways such as clinic visits, diagnostic
tests, drug therapy, hospitalization for management of
complications of cirrhosis, and liver transplantation. Also,
there are indirect costs related to lost work time and
impaired quality of life. Although the current economic
impact of chronic hepatitis C is substantial, there are few
published studies. Cirrhosis caused by chronic hepatitis C
currently accounts for 8,000 to 12,000 deaths per year in
the United States and it is the leading indication for liver
transplantation.
The current burden placed on the health care system by HCV
infection is relatively small considering the prevalence of
infection. However, as the large pool of currently infected
patients age and their disease has time to progress, more
patients may develop complications of liver disease and the
burden on the health care system will increase as a result.
Thus, the purpose of this study was to use a previously
described and validated mathematical model of the natural
history of chronic hepatitis C to project the future
prevalence of chronic hepatitis C, the incidence of
complications related to cirrhosis, and the potential impact
of treatment on these events.
The previously described projections of HCV-related disease
and its complications assume that no treatment was provided.
Therefore, we examined the effect of treating some of the
patients with chronic hepatitis C. It is not known what
proportion of HCV-infected patients currently receive or
could be identified to be considered for therapy or would be
acceptable for treatment. One survey estimated that only 30%
to 40% of infected patients in a Veterans Hospital
population would be optimal candidates. Given these
uncertainties, we modeled the effects of treating various
proportions between 10% (a minimum estimate) and 70%
(perhaps the maximum proportion of potentially treatable
candidates) of infected patients. Model projections show
that identification and treatment of patients with chronic
hepatitis C reduces the number of cases of decompensated
cirrhosis in nearly direct proportion to the proportion of
the cohort identified and treated.
Treating 10%, 50%, or 70% of all hepatitis patients with
compensated liver disease would decrease complications of
cirrhosis after 20 years by 5%, 24%, and 34%, respectively.
Initial experience with interferon monotherapy showed that
SVR was unusual in patients with cirrhosis, and, therefore,
many physicians chose not to treat them. However, recent
data with combination therapy suggests that patients with
fibrosis or cirrhosis respond nearly as well as others.
Therefore, we assessed the long-term effect on complications
of limiting combination treatment to those with or without
cirrhosis. Treatment of all patients was the optimal
strategy for reducing both decompensation and hepatic deaths
(Table 2).
Table 2. Proportion of treatment-related reduction in
hepatic decompensation accounted for by different subgroups
of patients.
|
|
 |
| |
However, because hepatic
decompensation occurs only in patients with cirrhosis,
treatment responses in patients with cirrhosis account for
the early reduction in disease complications and hepatic
death (an average of 88% of the reduction observed during
the first 5 years). Therefore, excluding patients with
cirrhosis from treatment would reduce considerably the
potential benefit on disease morbidity and mortality that
could be observed during the first 10 years after treatment.
The benefits of treating patients without cirrhosis do not
accrue until much later and do not match those gained from
treating patients with cirrhosis until 11 years after
initial therapy. Treatment of patients with mild chronic
hepatitis would have little impact on the incidence of
hepatic decompensation over the next decade, but would have
a substantial impact thereafter, accounting for only 11.0%
of the reduction after 10 years, but 31.9% of the reduction
after 20 years and 58.5% of the decrease after 40 years
(Table 2). Treatment of patients with persistently normal
ALT levels would reduce disease complications very little
(2.0% during the next 10 years; 6.7% after 40 years).
Treatment of patients with persistently normal ALT levels
would decrease the number of patients with cirrhosis by only
2.1% after 20 years compared with no treatment.
Discussion by authors
Using data derived from annual incidence rates, our model
estimated that there were approximately 3.07 million people
infected with HCV in the United States in 1994, a figure
similar to the estimate of the NHANES III survey.
Mathematical modeling estimates that the total number of HCV
infected cases will decrease gradually in coming years as
the incidence of new infection falls and currently infected
patients die from nonhepatic, age-related causes. However,
the large pool of surviving patients remains at risk of
progressive liver disease as the duration of their infection
increases. Indeed, the model predicts that an expanding
proportion of the patients with chronic hepatitis C will
develop cirrhosis over the next 2 to 3 decades, doubling the
current percentage by the year 2020. As a result, there will
be a dramatic increase in the number of cases with
complications of liver failure, hepatocellular carcinoma,
and death caused by liver disease. In fact, these changes
may already be occurring. Although our model did not show a
significant increase in disease complications in the last
decade of the twentieth century, others have reported an
increase in hepatocellular carcinoma and liver failure in
patients with chronic hepatitis C.
Interferon-based treatment regimens with the combination of
pegylated interferon and ribavirin result in sustained loss
of HCV in approximately half of treated cases.31,32 Viral
eradication is associated with reduction in hepatic
fibrosis, and, therefore, successful treatment might be
expected to reduce future disease complications. However,
despite the impressive results in clinical trials with
pegylated interferons and ribavirin, many patients are
unaware of their infection, are not candidates for
treatment, are unable to complete the course of treatment,
or fail to respond. Considering these factors and assuming
that at most perhaps half of patients could be identified
and treated, we predict that the most we could expect from
aggressive treatment at the optimal doses and duration of
medication would be a 24% reduction in the incidence of
decompensated cirrhosis after 20 years. This goal is best
accomplished by giving highest priority to treating patients
with moderately severe inflammation or fibrosis. Our model
did not consider that patients who do not permanently clear
virus might benefit from treatment with a reduction in the
rate of fibrosis as has been reported by some investigators.
This theory remains unproven and is currently being
evaluated in long-term clinical trials. Thus, our
projections might underestimate the potential long-term
benefit of treatment. Nevertheless, our data show that
current therapy and practice patterns that identify and
treat a relative minority of infected patients will not be
sufficient to control the future complications of this
infection.
Identification and treatment of a larger proportion of
infected patients, education about the importance of
abstinence from alcohol (the most important risk factor for
disease progression), and development of better tolerated
therapies may help to achieve a more meaningful impact on
the morbidity and mortality of this disease. Altough this
may support recent calls for look-back programs for
transfusion recipients or large scale screening programs,
such strategies have large up-front costs.37 However, the
costs of screening and treatment may well be offset by
reducing the later costs of treating the complications of
cirrhosis, which have been estimated to exceed 1 billion
dollars per year. Indirect costs, including productivity
loss and other societal losses, could reach 7 to 8 billion
dollars per year.38 Several studies have shown that
treatment of chronic hepatitis C with either interferon
alone or in combination with ribavirin is highly
cost-effective.
Despite the effectiveness of current antiviral regimens, our
model clearly shows that the majority of cases of cirrhosis
are not prevented. Thus, the complications of cirrhosis will
continue to increase over the next 20 to 30 years. The need
for liver transplantation will continue to far exceed the
capacity of transplant centers to handle the load. This
clearly emphasizes the urgency in developing and enacting
measures to increase the availability of transplant services
through increased organ donation, use of living donors, or
splitting cadaveric livers. Research in xenotransplantation
and stem cell technology may provide future options for
these patients.
Obviously, projections provided by mathematical models are
limited by the accuracy of the model conditions and
assumptions. We used conservative assumptions that were
intentionally biased to underestimate disease complications
whenever possible. However, our use of a general-population,
age-adjusted, all-cause mortality rate may have resulted in
an overestimation of disease complications by as much as
22%. Some evidence suggests that the nonhepatic mortality
rate is higher in chronic hepatitis patients because of
comorbid conditions and risk factors for other diseases.
Furthermore, the model might have underestimated
nonvirologic benefits of treatment, if they occur, because
only SVR rates were used. Finally, we did not consider the
likely availability of more effective therapies or increased
transplant volume in the future. Nevertheless, the message
is clear that an aggressive, proactive approach is needed to
first identify, educate, and treat patients with HCV
infection and, second, to increase transplant resources.
|
|
| |
| |
|
|
|
