A Case of Primary Invasive Aspergillus Colitis Masquerading as Clostridium difficile Infection

To the Editor:

A62-year-old female presented with abdominal pain and diarrhea. Her background included seronegative inflammatory arthritis and previous Clostridium difficile infection (CDI). She was immunosuppressed on prednisolone, methotrexate, and tocilizumab. A computed tomography (CT) scan of her abdomen revealed colitis of the descending colon. A stool sample was positive for Clostridium difficile by polymerase chain reaction (PCR), so oral vancomycin and intravenous metronidazole were commenced for presumed severe CDI. Colonoscopy revealed edema of the bowel wall but no ulceration, ischemia, or pseudomembranes. Biopsies demonstrated acute cryptitis with a differential diagnosis of infectious colitis, inflammatory bowel disease, or drug-related colitis.

She deteriorated further, and a CT scan showed a perforation of the descending colon necessitating urgent laparotomy and subtotal colectomy. She remained unwell with pyrexia and leukocytosis despite broad-spectrum antibiotics and was commenced on liposomal amphotericin B. She defervesced quickly, and her leukocytosis reversed. This dramatic response raised the question of whether her presentation was caused by a fungal colitis and the positive C. difficile PCR result reflected colonization rather than CDI. A pan-fungal PCR followed by sequencing was carried out on a paraffin block of colonic tissue and detected Aspergillus fumigatus. A diagnosis of primary invasive Aspergillus colitis was made based on the clinical picture, her immunosuppressed state, her response to liposomal amphotericin B, and a strongly positive fungal polymerase chain reaction (PCR) result.

This case highlights two important points. First, isolated invasive extra-pulmonary Aspergillus infection should be considered in immunosuppressed patients presenting with severe sepsis. It carries a mortality rate as high as 80% [1]. Although the diagnosis usually is confirmed by histologic examination, we used PCR to detect Aspergillus in this patient's colon, which is a relatively new method for diagnosing invasive aspergillosis. A systematic review of 16 studies involving 1,618 patients found the test to have a sensitivity and specificity for a single positive sample of 0.88 (95% confidence interval 0.75–0.94) and 0.75 (95% confidence interval 0.63–0.84), respectively [2].

The second point is the potential for misinterpretation of Clostridium difficile test results. Culture for toxigenic C. difficile takes 24–48 h to perform. To avoid delays, many hospitals use nucleic acid amplification testing such as PCR, which detects the toxin tcdA/tcdB genes. Because of the high sensitivity of these tests, toxigenic C. difficile is identified in asymptomatic patients also [3]. A positive toxin gene PCR informs the clinician that the patient is either colonized or infected with a toxigenic strain of C. difficile. The distinction is made from the presence or absence of symptoms. The difficulty arises when patients are experiencing symptoms consistent with colitis and test positive for the C. difficile toxin gene by PCR but have an alternative etiology for their presentation. These patients may inadvertently be labeled as having CDI when they are actually just colonized. In these cases, a negative toxin enzyme immunosassay (EIA) may help corroborate a diagnosis of C. difficile colonization as opposed to infection. The clinician should suspect an alternative diagnosis in patients who fail to respond to standard treatment regimens for CDI.