Revisiting RIF/PZA: Toxicity reports give docs the jitters
Analysis of incidents still under way
Some TB clinicians are backing away from rifampin/pyrazinamide (RIF/PZA), the new short-course regimen recently approved for treating latent TB infection (LTBI). On top of a death and an episode of serious hepatotoxicity already chronicled
in the April 2001 Morbidity and Mortality Weekly Report (MMWR), unpublished reports of two more deaths have surfaced, plus other episodes of severe hepatotoxicity. The recent events have left many practitioners feeling decidedly jittery about the regimen.
By late June, some TB experts had decided to meet in Baltimore prior to the National TB Controllers’ Association annual conference so they could compare notes on the adverse effects that seemed to be cropping up with use of the regimen.
At the Centers for Disease Control and Prevention in Atlanta, the Division of TB Elimination (DTBE) has asked jurisdictions to report any incidents of toxicity that appear to be linked to either the use of RIF/PZA or to standard isoniazid (INH) prophylaxis. DTBE also is preparing to begin on-site investigations of the incidents. An update on the situation is expected to be published in the MMWR by late summer.
What’s known so far suggests that the incidence of serious side effects associated with RIF/PZA may far exceed that for INH prophylaxis, says John Sbarbaro, MD, professor of medicine at the University of Colorado Health Sciences Center in Denver. From the start, Sbarbaro has been a vocal critic of the new regimen, predicting that RIF/PZA toxicity might prove unacceptably high.
So far, the deaths from RIF/PZA seem to have occurred among a denominator of perhaps 5,000 patients who have used the new regimen. If that denominator is correct, the rate clearly exceeds that for INH prophylaxis, estimated conservatively by the CDC at about one death for every 10,000 times that regimen is used in TB clinics, Sbarbaro says.
Regardless of how the data on frequency eventually pan out, the current problems seem to fall into two types, says Naomi Bock, MD, an epidemiologist at the DTBE and an assistant professor at Emory University School of Medicine.
First, using the high end of the dose range suggested for PZA "seems [sometimes] to lead to immediate toxicity," she notes. In such instances, patients beginning the regimen "complain of stomachaches, and then they stop taking the medications," which puts an end to the problem.
That type of experience has already spurred some clinicians to begin using the lower range of the suggested dosage for PZA, she adds.
In the second kind of scenario, problems seem to develop more gradually, with some patients experiencing liver inflammation after taking the regimen for a month or so.
Bock says three issues still need to be sorted out:
• whether either (or both) of the problems is dose-related;
• whether serious toxicity is associated with an underlying history of liver disease;
• whether clinicians will be able to steer clear of trouble by taking steps such as monitoring hepatic enzymes or adjusting the PZA dose.
Underlying liver disease may be implicated
In Massachusetts, the Boston city TB clinic
and other state providers have treated about 650 patients with RIF/PZA, with three instances of toxicity severe enough to require hospitalization, says Edward Nardell, MD, medical advisor to the Massachusetts health department TB control program and professor of pulmonary medicine at Cambridge (MA) University Hospital.
As a result of the three episodes, Nardell has called a temporary halt to use of the regimen in the state. Signs and symptoms in the Massachusetts cases have included vomiting, edema, and high hepatic enzymes in the absence of symptoms.
All three episodes, it turns out, have been linked to an underlying history of liver damage, Nardell adds — though in two cases, patient histories had failed to turn up any clues that pre-existing damage might be present.
"It seems that underlying liver disease may be more of a problem with RIF/PZA than with INH," he says. Getting laboratory baselines might have made a difference in two of the three cases, he adds. Current guidelines call for laboratory baselines only if patient histories suggest the presence of either liver damage or regular alcohol use.
Evidence from one of the Massachusetts cases suggests that even when baseline hepatic measurements are obtained, things can still go awry. In one instance, Nardell says, a test revealed that enzymes were elevated; but because the patient was a close contact who was judged at risk for developing disease, the decision was made to go ahead and treat. Active symptoms never emerged, but subsequent laboratory tests showed hepatic enzymes were continuing to rise, prompting the physician in the case to stop treatment and hospitalize the patient.
One reason routine lab work is no longer even recommended for prophylaxis is the realization that episodes of so-called "silent liver injury" are findings not infrequently associated with INH prophylaxis, says John Jereb, MD, an epidemiologist at DTBE. "Usually, these patients not only never have symptoms, they have complete resolution of the problem," he says. An estimated 10% to 15% of patients receiving INH prophylaxis probably undergo such episodes, almost always with no symptoms and no lasting effects, Jereb adds.
Patients fail to report symptoms
With RIF/PZA, the evidence so far seems to suggest that when trouble develops, it’s because patients experience symptoms but fail to report them, meanwhile continuing to take the medication, says John Bernardo, MD, TB control officer of Boston. Even stern warnings about reporting symptoms and stopping treatment occasionally fall on deaf ears, Bernardo adds. "Some people will deny or ignore symptoms even when they’ve been explicitly warned about them,"
he says.
When the short-course regimen was tested
in trials on HIV-negative patients in Haiti, few problems with toxicity were reported. If the regimen does prove to be more troublesome for HIV-negative patients, one reason could have to do with PZA’s mechanism of action, says Sbarbaro.
"PZA doesn’t kill bugs," he says. Instead, it blocks an action whereby the TB microbe converts fats in order to construct a rugged lipid wall around itself. PZA’s prevention of the wall’s construction leaves the microbe vulnerable to assault by anti-TB drugs. But PZA can only do its work in an acidic environment, Sbarbaro adds. It could be that among patients with competent immune systems — namely, those who are HIV-negative — most circulating TB microbes have been engulfed by macrophages, which lack the acidic environment the drug requires to do its work.
On a more elementary level, Sbarbaro argues that because PZA isn’t active in the continuation phase of therapy, there’s little reason to think it will be helpful for prophylaxis of latent disease. "It’s just common sense," he declares.
Sbarbaro adds that instead of a retooled version of RIF/PZA, he’d prefer to see more emphasis on four months of RIF, another short-course regimen already available as an option for LTBI treatment. "You can’t have any deaths associated with a preventive regimen and expect people to use it," he says.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content