Change in TB
guidelines could save lives, money
BY TIM STEPHENS
Tuberculosis screening
programs enable doctors to identify outwardly healthy
people infected with the tuberculosis bacterium and treat
the infection before they develop active disease and
spread it to others. Unfortunately, the drug used for
this preventive treatment, isoniazid, has been known to
cause fatal liver damage. For this reason, it is often
withheld from infected patients deemed at low risk of
developing active tuberculosis.
Monitoring of liver
function during treatment, however, can greatly reduce
the risk of liver damage from isoniazid. A risk-benefit
analysis by researchers from Stanford and elsewhere now
shows that isoniazid treatment coupled with
liver-function monitoring is a safe and cost-effective
way to prevent active tuberculosis.
Indeed, contrary to
current guidelines issued to U.S. physicians, the
researchers conclude that most people who test positive
on the tuberculin skin test should receive monitored
treatment with isoniazid (also known as INH). The new
findings appear in the Dec. 15 issue of Annals of
Internal Medicine.
"Monitored INH
prophylaxis is not only safe, it is inexpensive, and over
time it saves lives and money by preventing the
development of active tuberculosis and the spread of
infection to other people," said Dr. Shelley
Salpeter, who directed the new study. Salpeter is
associate chief of primary care at Santa Clara Valley
Medical Center and a clinical associate professor of
medicine at Stanford.
The new analysis suggests
that current guidelines for using isoniazid as a
preventive treatment should be changed, said Dr. Douglas
Owens, the senior author of the paper. "Our findings
suggest that most people who test positive should
probably get isoniazid prophylaxis, and that's quite
different from the current recommendations," said
Owens, a senior research associate at the Veterans
Affairs Palo Alto Health Care System and associate
professor of medicine and of health research and policy
at Stanford.
Isoniazid should be
withheld only from patients with active liver disease,
Salpeter added.
At present, the American
Thoracic Society recommends restricting the use of
isoniazid prophylaxis to tuberculin reactors (people with
positive skin tests) younger than 35 and to those older
than 35 who are at increased risk for activation of the
infection. Factors that increase the risk of activation
include diabetes, renal failure, alcoholism,
immunosuppression and abnormal chest x-rays suggesting
previously active tuberculosis.
The society issued these
recommendations in 1974 after reports of several deaths
due to isoniazid-induced hepatitis. Most of the deaths
were in patients older than 35 years who also had
underlying liver disease, Salpeter said.
Since 1983, periodic
liver-function tests have been recommended for patients
over age 35 taking isoniazid. Two recent studies looked
at the outcomes of patients placed on isoniazid
prophylaxis since 1983 and found that the risk of death
is now significantly lower than it was before the use of
liver-function monitoring. Salpeter, who performed one of
those studies, said the risk is now negligible.
"On average there is
about 1 death per 100,000 patients on INH prophylaxis,
which is about the same as the death rate from
penicillin," she said.
Tuberculosis can remain
"silent," or inactive, for many years after the
initial infection, which goes unrecognized in the vast
majority of cases. During the inactive phase, the
infected person has no symptoms and cannot transmit the
disease to others. To eliminate an inactive infection,
patients must take isoniazid every day for six months.
Because some people don't complete the full course of
treatment, studies have shown that isoniazid prophylaxis,
on average, reduces the risk of active disease by about
70 percent, Salpeter said.
In the new study, Salpeter
and her co-authors compared the health outcomes and
economic effects of prescribing or withholding isoniazid
prophylaxis for people who would not be treated under
current guidelines in other words, those older than 35
who test positive for tuberculosis but have normal chest
x-rays and are at low risk for activation of the disease.
The researchers
incorporated many variables into their analysis of risks,
benefits and costs. In addition to considering the risks
and benefits to individuals, they looked at the effects
of isoniazid treatment on TB transmission within the U.S.
population.
At the individual level,
the analysis showed a modest benefit from isoniazid
prophylaxis, Owens said. Even patients at low risk for
developing active disease are more likely to die of
tuberculosis than of isoniazid-related liver damage, the
researchers found.
"In the first year
alone, there are more lives saved by prevention of TB
deaths than are caused by INH-induced hepatitis, and once
the treatment is over, you have only benefits and no
risks," Salpeter said.
Moreover, for every case
of active tuberculosis prevented by isoniazid, the
treatment also prevents the additional cases that would
have resulted from transmission of the infection.
"The epidemiology model showed that for every case
of active tuberculosis, there are ultimately 1.2 other
cases of active tuberculosis that result from multiple
successive transmissions," Salpeter said.
To analyze the effects in
a population over time, she collaborated with her father,
Edwin Salpeter, an astrophysicist and professor of space
sciences at Cornell University. The senior Salpeter, who
received the 1997 Crafoord Prize in Astronomy from the
Royal Swedish Academy, contributed his mathematical
expertise to the development of the epidemiology model.
Gillian Sanders, a PhD
student in Stanford's medical informatics program and
another co-author of the paper, helped develop the
cost-effectiveness analysis.
The researchers found that
by expanding isoniazid treatment to include virtually all
tuberculin reactors over age 35 in the United States, the
nation could avert as many as 35,176 deaths and reduce
medical costs by $2.11 billion.
"In reality, of
course, we wouldn't be able to reach all of those people,
but even if we treated only 20 percent of them, we would
get 20 percent of those benefits," Shelley Salpeter
said.
Owens noted that this type
of analysis will be the focus of two new School of
Medicine programs aimed at improving clinical practice:
the Evidence-based Practice Center and the Stanford
Center for Evaluation of Health Practices and Advancement
of Primary Care.
"This is the kind of
focus we plan to have, and we hope that when results like
these come out, people will use them in developing
practice guidelines," said Owens, who is active in
both new programs.
The isoniazid study was
supported in part by the Santa Clara Valley Medical
Center and by a Career Development Award to Owens from
the VA Health Services Research and Development Service.
SR
|
