Gut Aspergillus Infections in Leukemia and Stem Cell Transplant Patients: A Rare and Unrecognized Disease.

Invasive aspergillosis is a life-threatening infection in immunocompromised patients, mainly in acute leukemia and stem cell transplant (SCT) patients. Aspergillosis mainly involves the lower respiratory tract and sinuses. Other sites of infection are rarely observed, except in the setting of aspergillus dissemination. Gut aspergillosis has been exceptionally reported in the literature, sometimes diagnosed only at autopsy. The goal of this retrospective study was to report on cases of gut aspergillosis in hematology patients, clinical presentation, risk factors, and outcome.

This is a multicenter retrospective study in French, Swiss and Belgium hematology centers. We collected data from patients with hematologic disease and proven gut aspergillosis. Patients with only liver or spleen lesions were excluded. As the EORTC-MSG criteria do not propose a diagnosis of probable gut aspergillosis on the basis of gut imaging and indirect microbiological data, we only collected patients with biopsy, autopsy, or tissue-culture proven gut aspergillosis. The data were collected from the medical charts through a specific questionnaire. Segal et al. criteria were used to assess the response to treatment at 12 weeks after the diagnosis (CID 2008;47:674). Patients: 20 patients from 9 centers were included. The underlying disease was acute leukemia in 12, acceleration of CML in 2, lymphoma in 2, myeloma in 3, and myelodysplasia in 1. Nine patients had received an allogeneic SCT, The mean age was 48 years. Eleven out of 20 (55%) were neutropenic at time of gut aspergillosis.

Four patients had upper gut localization of aspergillosis (oesophagus: 2; stomach: 2), 11 had lower localizations (small bowel, colon, rectum) and 5 patients had concomitant upper and lower localizations. On the basis of imaging, aspiration, biopsy or autopsy, 12 patients had disseminated aspergillosis, and 8 had localized gut infection. Diagnosis was made by autopsy in 5 patients, endoscopic biopsy in 6, and biopsy through laparoscopy in 9 patients. All gut biopsies showed filaments evocative of aspergillus sp. Additionally, among 12 of them who were cultured, 9 grew A fumigatus, 1 grew A flavus and 2 were negative in culture. Clinical presentation was poorly specific. In lower gut lesions, we noted febrile diarrhea in 10 patients, abdominal pain in 16, gut hemorrhage in 7, intestinal occlusion in 5, and perforation in 1. These symptoms mostly led to endoscopy when lesions were accessible. However, only surgery or laparoscopy could get small-bowel lesions, mostly in emergency. In upper gut lesions, dysphagia and odynophagia usually led to endoscopy and the diagnosis was easier than for lower gut lesions. For 19 patients where the data was available, 15 had a positive galactomannan (GM) test (> 0.5 ng/mL). Among them, all the patients who got a negative GM test under treatment survived at 12 weeks except one. Five of the 11 operated patients survived, one died during surgery, from septic shock with small bowel perforation. All patients diagnosed before death received antifilamentous drugs, mostly voriconazole. At 12 weeks after the diagnosis, excluding the 5 patients diagnosed at autopsy, 8/15 (53%) were alive, with 6 in complete response, 1 in partial response, and 1 not evaluable. It is noteworthy that 4 patients were either from Africa or Pacific Ocean islands, one had an African diet, and one had lived in Tahiti two years before diagnosis. We hypothesize that a food enriched in spices may favor aspergillus gut colonization and increase the risk of developing aspergillosis during immunodepression. Conclusion: The diagnosis of gut aspergillosis is a difficult one in the absence of surgery, due to the poor specificity of the symptoms and the absence of characteristic imaging. The mortality is high. Patients with a positive GM unexplained by respiratory lesions should be suspected of gut aspergillosis each time there is gut symptoms. Hematologists have to be aware of this complication whose frequency is likely underestimated. In most cases, it seems logical to associate medical and surgical therapy. However, the optimal therapy is so far not determined.

Herbrecht:Pfizer: Research Funding, Speakers Bureau; Gilead: Consultancy, Speakers Bureau; MSD: Consultancy, Speakers Bureau; Schering-Plough/Essex: Consultancy, Speakers Bureau.