Drug-Induced Lupus Associated with Rifabutin

Abstract

Drug-induced lupus (DIL) is a rare adverse reaction to medications with features resembling idiopathic systemic lupus erythromatosis. Rifabutin/rifamycins have only rarely been reported as a cause of DIL, and no cases have been reported in blacks. A 55-year-old African American woman with HIV presented with severe generalized arthralgias and recurrent oral ulcers while receiving treatment for tuberculous meningitis. Arthralgias, which began in her knees after 5 weeks of antituberculous therapy, progressed to involve the joints in the ankles, wrists, and hands. She had no associated fever or rash. When she had these symptoms her antinuclear antibody (ANA) was 1:1280 homogenous pattern, antidouble stranded DNA was negative, antihistone antibody was strongly positive, anti-smith and antiribonucleoprotein (anti-RNP) were negative. Her symptoms resolved within 2 months of stopping rifabutin while continuing other antituberculous medications and her ANA titer started to decrease. We review the existing literature on this subject.

Case Report

A 55-year-old woman with HIV (CD4 487/8%, nadir CD4 = 249 cells/mm³, HIV RNA 50 000 copies/mL) presented with intermittent headache, weakness, and lightheadedness for 2 months. Headaches were stabbing, frontotemporal, nonradiating, 8 out of 10 in intensity, occurring daily, worse with standing, and relieved partially by rest and ibuprofen. She had photophobia and phonophobia during the headaches but never lost consciousness and had no associated neck stiffness, vertigo, tinnitus, fever, or night sweats. She had self-discontinued her antiretroviral (ARV) medications 3 months prior to presentation for unclear reasons and reported 15 pounds weight loss since then. Her physical examination was notable for word finding difficulty, positive Romberg sign, and poor tandem gait. Laboratory studies revealed a white blood cell count (WBC) of 9400/µL, neutrophils 22%, lymphocytes 66%, hemoglobin 14 g/dL, and hematocrit 41%. Lumbar puncture revealed an opening pressure of 21 mm Hg and a closing pressure of 13 mm Hg, cerebrospinal fluid (CSF) WBC 200/mm³ with 99% lymphocytes, glucose of 60 mg/dL, and protein of 526 mg/dL. Other CSF studies including, herpes simplex virus, cytomegalovirus, Epstein-Barr virus, JC virus, BK virus polymerase chain reaction (PCR), west nile virus, measles, mumps, varicella zoster, toxoplasma, and Cryptococcus serologies were all negative. The CSF acid-fast bacilli smear and mycobacteria PCR were also negative. The magnetic resonance imaging showed possible leptomeningeal enhancement. Repeat CSF studies after 5 days showed rising WBC to 711/mm³ and persistently elevated protein of 438 mg/dL. Tuberculous skin test 8 months earlier was negative but quantiferon in the current admission was positive. She was started empirically on antituberculous medications using standard dosing of rifabutin, isoniazid, pyrazinamide, and ethambutol. Rifabutin was utilized because she was planning to resume her HIV antiretroviral therapy that included ritonavir (RTV). She improved remarkably with resolution of her CNS symptoms after 2 months of treatment. Her clinical course was complicated by elevated transaminases after 4 months of treatment which interrupted treatment for 10 weeks. This was attributed to isoniazid hepatotoxicity. Rifabutin 150 mg orally 3 times a week, ethambutol 1600 mg oral 3 times a week, and pyrazinamide 2500 mg oral 3 times a week were started after transaminases improved.

Four weeks after restarting treatment, she presented with severe generalized arthralgias and recurrent oral ulcers. Arthralgias which began in her knees later progressed to involve the joints in the ankle, wrists, and hands. She had no associated fever or rash. When she had these symptoms, her antinuclear antibody (ANA) was 1:1280 homogenous pattern, antidouble stranded DNA was negative, antihistone antibody was strongly positive, and anti-smith and antiribonucleoprotein (anti-RNP) were negative. Rifabutin random level was within the normal limits. A prior ANA obtained when she initially presented 5 months earlier was negative. Her arthralgias resolved within 1 month of stopping rifabutin, while continuing ethambutol and pyrazinamide. The oral ulcers resolved after 2 months of stopping rifabutin and her ANA titer continues to trend down now 1:640.

Discussion

Drug-induced lupus (DIL) is a rare syndrome occurring as an adverse reaction to certain medications with features which are similar to spontaneous systemic lupus erythromatosis (SLE) but distinct in many ways.

Overall, it is estimated that 15 000 to 30 000 cases of DIL occur in the United States every year. ¹ Whites have been reported to be affected up to 6 times more frequently than blacks and may have more severe manifestations, ² although 1 review actually reported no incidence in blacks. ³ DIL has been recognized as a side effect of therapy with over 80 drugs. ^2,3 Drugs definitely known to be associated with DIL include hydralazine, procainamide, isoniazid, methyldopa, quinidine, minocycline, and chlorpromazine; others are possible or probable causes. ⁴

Rifabutin, a semisynthetic ansamycin antibiotic, has been used extensively worldwide as an antimycobacterial agent for over 35 years. To date, only 3 cases of rifabutin-induced lupus have been reported and none has been reported in blacks. Rifampin, another rifamycin antituberculous agent, has also been implicated and reported as a cause of DIL only in 4 whites. ⁵

Symptoms usually include fever, myalgia, rash, arthritis, arthralgias, and serositis. Severe reactions involving the kidney, liver, and central nervous system are rare. ^6,7 In general, the clinical and laboratory features of DIL are similar to those of SLE except that in DIL, patients recover fully after the offending agent has been discontinued. The ANA which can be used as a screening tool when considering DIL is equally present in both DIL and idiopathic SLE occurring in about 95% of the patients. In DIL, ANA has been found to be consistently homogeneous, in contrast to homogenous or speckled pattern in idiopathic SLE. Antihistone antibodies are present in 95% of DIL cases compared with 60% to 80% of idiopathic SLE cases. Anti-double stranded DNA antibodies, anti-Smith, and anti-RNP (which were negative in our patient) are usually rare in DIL. ^{1 –3,8} The pathogenesis of DIL is not fully understood but several theories have been developed which are beyond the scope of this discussion. ⁹

To date, the American College of Rheumatology has no diagnostic criteria for DIL but Borchers et al ¹ proposed that the patient should have sufficient and continuing exposure to the inciting drug, should have at least 1 symptom compatible with SLE, should have no history suggestive of SLE before starting the drug, and should have resolution of symptoms within weeks (sometimes months) after discontinuation of the offending drug. Our patient meets all of these criteria.

In the 3 previously reported cases of rifabutin-induced lupus, all the patients received either clarithromycin or ciprofloxacin concomitantly as part of their treatment. These antibiotics are known to inhibit hepatic cytochrome P450 enzymes; hence, DIL was attributed to increased rifabutin levels. Unfortunately, rifabutin levels were not measured for these patients. ⁵ In another study in which the dose-limiting toxicity of rifabutin in patients with AIDS was studied, 16 patients with AIDS were treated for 4 to 66 weeks with stepwise increasing rifabutin dose from 300 to 2400 mg/d. A reversible syndrome of arthritis/arthralgia was seen in 9 of 10 patients who received >1050 mg/d of rifabutin. Uveitis and aphthous stomatitis developed at 1800 mg in 2 of the 9 patients. Unfortunately, autoimmune serologies to confirm DIL were not obtained for any of these patients. ¹⁰

Interestingly, our patient was also concomitantly taking rifabutin and RTV, another potent inhibitor of cytochrome P450. However, at the peak of her symptoms a rifabutin level was within the normal limits. Although it remains unclear why our patient had this adverse reaction to rifabutin, her symptoms cannot be attributed to high rifabutin levels as suggested by the previous reports.

The updated rifabutin package insert from January 2010 makes no mention of DIL as an adverse reaction to this medication. ¹¹ This case highlights the importance of considering the diagnosis of DIL in the setting of rifabutin treatment and changing the therapy to reverse the morbidity associated with this adverse reaction. This is particularly relevant in the highly active antiretroviral therapy era, where rifabutin is favored over conventional rifampin as part of antituberculosis therapy because of drug–drug interactions with protease inhibitors.

Conclusion

This is the first report of a case of rifabutin causing DIL in an African American. A high index of suspicion is required to make the diagnosis by obtaining the appropriate laboratory tests. Discontinuation of the medication is required to relieve the symptoms and confirm the diagnosis.