Abstract

Background: Whipple disease, which is caused by infection with Tropheryma whipplei, can be treated effectively with antimicrobials. Occasionally, inflammation reappears after initial improvement; this is often interpreted as refractory or recurrent disease. However, polymerase chain reaction for T. whipplei in tissue is sometimes negative during reinflammation, indicating absence of vital bacteria, and this reinflammation does not respond to antimicrobials but does respond to steroids. Objective: To demonstrate that the immune reconstitution inflammatory syndrome (IRIS) occurs in patients treated for Whipple disease. Design: Cohort study. (International Standard Randomised Controlled Trial Number Register registration number: ISRCTN45658456) Setting: 2 academic medical centers in Germany. Methods: 142 patients treated for Whipple disease out of a cohort of 187 were observed for reappearance of inflammatory signs after effective antibiotic therapy. Definitions of IRIS in HIV infection, tuberculosis, and leprosy were adapted for application to Whipple disease. Results: On the basis of study definitions, IRIS was diagnosed in 15 of 142 patients. Symptoms included fever, arthritis, pleurisy, erythema nodosum, inflammatory orbitopathy, small-bowel perforation, and a hypothalamic syndrome. Two patients died. There was a positive correlation with previous immunosuppressive treatment and a negative correlation with previous diarrhea and weight loss. Limitations: The study was observational and thus has inherent weaknesses, such as incomplete and potentially selective data recording. Conclusion: The immune reconstitution inflammatory syndrome was diagnosed in about 10% of patients with Whipple disease in the study cohort; the outcome varied from mild to fatal. Patients who had had previous immunosuppressive therapy were at particular risk. An immune reconstitution syndrome should be considered in patients with Whipple disease in whom inflammatory symptoms recur after effective treatment. Early diagnosis and treatment with steroids may be beneficial; prospective studies are needed.