Mobile Edition
Print
Get the Mag
Weekly UpdatesScreen ... and rescreen!
All rheumatoid arthritis patients being treated with a tumor necrosis factor-alpha inhibitor must be screened for tuberculosis. Ideally, screening should occur before treatment begins.
Why? TB is a worldwide problem and is at its highest level in history. The World Health Organization estimates that 2 billion people worldwide are currently infected, and 2 million die annually. A full 25% of these deaths occur in developed countries.
Rheumatoid arthritis patients are already at greater risk for developing TB than is the general population, regardless of their drug regimen. A retrospective U.S. cohort study analyzed a health insurance database of 55 million people. After excluding anybody taking an anti-TNF agent and correcting for as many variables as they could, investigators found the relative risk of TB was 7.79 times greater in RA patients than in non-RA patients (Ann. Rheum. Dis. 2008;67[suppl. II]:91).
Add anti-TNF agents to the mix and the rates increase even further. One study found a rate of 90 TB cases per 100,000 person-years among patients actively treated with anti-TNFs (n = 9,882), compared with 0 for those actively receiving disease-modifying antirheumatic drugs (n = 2,883). These rates persisted among patients who were no longer on these drugs but whose most recent exposure had been anti-TNF therapy (Ann. Rheum. Dis. 2008;67[suppl. II]: 178).
Similarly, a French study concluded that while the incidence of tuberculosis was 8.7 per 100,000 patient-years among the general French population, it was 71.5 per 100,000 patient-years among those receiving monoclonal antibodies infliximab or adalimumab. Patients on etanercept showed no increase in incidence, at 6 per 100,000 patient-years (Ann. Rheum. Dis. 2008;67 [suppl. II]:52).
Screening becomes even more critical when we consider that 65%-70% of RA patients present with nonpulmonary TB. These are not easy cases to diagnose, and can affect many areas of the body.
Yet the most compelling reason to screen our patients for TB is that screening works. The adalimumab clinical trials, which were performed primarily in Europe, found that TB screening programs reduced TB by 85%. They also found that 95% of patients who screened positive and took isoniazid were successful in not reactivating TB.
One of the many interesting revelations that came from these data was that chest x-ray alone is a poor screening tool. Moreover, the 5-mm skin test proved much more sensitive than did the 10-mm skin test. Even better was a booster (a second TB skin test done 10-14 days later) combined with a 5-mm cutoff.
The Canadian Rheumatology Association's algorithm for TB screening prior to anti-TNF therapy begins with a physical examination and review of the patient's TB exposure and history, followed by a skin test and chest x-ray. If either are positive, active TB should be considered. For true cases of active TB, full TB therapy should be completed before anti-TNF therapy begins. Patients who do not have active TB should receive isoniazid; anti-TNF therapy should begin 4 weeks later.
If both the TB test and chest x-ray prove negative, consider other significant risk factors for TB. Patients without such risk factors can begin anti-TNF therapy. A two-step tuberculin skin test (TST) should be considered in patients with significant risk factors. If this test proves positive, isoniazid therapy should begin. If negative, isoniazid should still be used in some patients with a strong history of prior exposure, such as a family member with active TB.
Finally, consider the ramifications if we didn't screen and one of our patients contracted TB and had a serious consequence. I think it would be very hard to find expert opinion that would defend such a course of action.
ROBERT OFFER, M.D.
DR. OFFER, of the University of British Columbia, Penticton, is chair of the BC PharmaCare Arthritis Biologics Committee. He disclosed that he is a consultant and speaker for Abbott, Amgen, and Schering-Plough.
Screen ... but tread lightly with isoniazid.
I will be the last physician to say that we should not be screening RA patients for TB. So yes, we must screen our patients. However, isoniazid therapy should be considered and calculated very carefully because of its significant risks.
Isoniazid interferes with pyridoxine metabolism. This may cause a number of side effects, including nausea and vomiting; fatigue, headaches, and drowsiness; peripheral neuropathy; and psychosis. But the most important side effect of isoniazid treatment is hepato-toxicity, which can present as anything from asymptomatic elevated liver enzymes to fatal hepatitis.
A 2005 study estimated that isoniazid hepatotoxicity increased with age, affecting 410 per 100,000 patients aged 25-34, 850 per 100,000 aged 35-49, and 2,082 per, 000 of those older than 49 (Chest 2005;128:116-123). Other risk factors for isoniazid hepatotoxicity include preexisting elevation of liver enzymes, alcohol consumption, and possibly, coadministration of medications such as methotrexate.
Fatal hepatitis also is a concern when it comes to isoniazid administration. A 1965-1989 literature review concluded that the risk of fatal hepatitis was 14 per 100,000 when women started isoniazid prophylaxis and 23 per 100,000 after completion of isoniazid prophylaxis (Am. Rev. Respir. Dis. 1992;:494-7).
The British Thoracic Society suggests that each patient should be mathematically assessed before starting isoniazid therapy (Thorax 2005;60:800-5). If the annual risk of TB exceeds the risk of drug-induced hepatitis, prophylaxis should be given. If the risk of hepatitis is greater, the patient should be monitored.
The formula that will help you decide whether to use isoniazid prophylaxis begins with the historic annual TB rate for the patient population in question, then multiplies that figure by 5 (which reflects the increased TB risk from taking anti-TNF agents). That number then is divided by the risk of isoniazid hepatitis (calculated to be a constant 278 cases per 100,000 population). If the ratio is less than 1, you may want to observe the patient rather than give isoniazid. If it's greater than 1, then you may consider prophylaxis.
Consider the following example of a Saskatchewan Caucasian, whose historic annual TB rate is 1 per 100,000:
[1/100,000 X 5 = 5/100,000/278/100,000]
The ratio here is 0.02, meaning the risk of treatment is greater than the benefit. Thus you should observe this patient.
Even a person of Indian status in Saskatchewan--whose demographic carries a much higher risk of TB--is not a sure thing for isoniazid therapy:
[43/100,000 X 5 = 215/100,000/278/100,000]
This risk ratio is 0.8
Once you decide whether your patient should be treated for TB, the next decision is the type of treatment to administer.
Typical prophylactic therapy for TB infection comprises daily self-administered isoniazid for 9 months or isoniazid and rifampin twice weekly for 6 months.
Assuming greater than 80% compliance, daily isoniazid therapy for 6 months reduces TB risk by 69%. Increasing the treatment period to 12 months sees the TB risk reduced by 93%. However, 6 months of isoniazid and rifampin decreases tuberculosis risk by 90%. So there is evidence that we should not be giving isoniazid monotherapy, anyway.
My strong suggestion is that yes, we screen for TB, with a history, physical exam, chest x-ray, and TST. If you have any concerns, refer the patient to TB Control. But before you consider prophylactic isoniazid treatment on this patient, first consider the risks.
And remember: Instead of doing our best, we should do what is necessary for the patient.
WOJCIECH OLSZYNSKI, M.D., PH.D.
DR. OLSZYNSKI, of the University of Saskatchewan, Saskatoon, is director of the Saskatoon Osteoporosis Centre. He has no financial disclosures to make regarding this topic.
FEED