Infection with Devriesea agamarum and Chrysosporium guarroi in an inland bearded dragon (Pogona vitticeps)

Introduction

Devriesea agamarum is part of the oral microbiota of clinically healthy bearded dragons and can also cause dermatitis.1 Chrysosporium guarroi has been isolated in cases of dermatomycosis in green iguanas and bearded dragons.2, 3 This case report describes clinical signs, diagnostic steps and treatment of a case of deep dermatitis associated with D. agamarum and C. guarroi.

Case report

A 4-year-old male inland bearded dragon (Pogona vitticeps), weighing 497 g, was presented with dysecdysis and chronic skin lesions. It had previously been treated orally (p.o.) with 10 mg/kg enrofloxacin (Baytril^®; Bayer Austria GmbH, Vienna, Austria) given once a day for 6 days, then with miconazole nitrate, polymyxin B sulfate, prednisolone acetate (Surolan^®; Janssen Pharmaceutica NV, Beerse, Belgium) and topical miconazole (Daktarin^® 2% – Oral Gel; Janssen-Cilag Pharma, Vienna, Austria) for 10 days; however, the clinical signs deteriorated during this therapy. The dragon lived alone in a 190 cm × 70 cm × 90 cm enclosure (photoperiod 12.5 h, temperature 26–50°C, humidity 40–70%, cage substrate sand) and was fed a mixed diet of crickets and salad. Physical examination revealed weight loss (from a guide weight of 680 g compared with 497 g) and retained slough on the beard. Skin lesions were brown to black, erosive to ulcerative, 0.5–2 mm in diameter, in confluent areas up to 1 cm in diameter, and multifocally distributed all over the body (Figure 1). Differential diagnoses included bacterial, fungal or viral dermatitis.

Exudative material from a skin wound on the right mandible was examined microscopically using Ziehl Neelsen staining. No acid-fast organisms were observed; bacilli, cocci and hyphal elements were found during microscopic examination. A swab was taken from the skin lesions. Blood was collected from the tail vein for haematological and plasma chemical analysis (see Table S1 in Supporting information).

The dragon was hospitalized and treated with sodium chloride and glucose infusion (Fresenius Kabi, Graz, Austria), 20 mL/kg subcutaneously (s.c.), once a day for 20 days, and wound bathing with chamomile (Kamillosan^®-Drops; Meda Pharma, Vienna, Austria) and povidone-iodine solution (Betaisodona^® Standardized Solution; Mundipharma GmbH, Vienna, Austria) for 20 days. Ceftiofur hydrochloride (Eficur^® 50 mg/mL; Laboratories Hipra SA, Amer, Spain) 5 mg/kg s.c. was given for 12 days, meloxicam (Metacam^® 1.5 mg/mL; Boehringer Ingelheim Vetmedica, Ingelheim/Rhein, Germany) 0.1–0.5 mg/kg was given p.o. once a day for 5 days, and voriconazole (Vfend^®; Pfizer Ltd, Vienna, Austria) 10 mg/kg was administered p.o. once a day for 14 days. The dragon was force fed with a herbal-based critical care diet for the first 3 days. The skin lesions were dry, clean and healing on re-evaluation on days 28, 42, 58 and 77.

On day 106, the owner represented the lizard with newly formed skin lesions. A swab for culture was taken from the new lesions. At the next visit (day 132), the lizard's condition had worsened. Therapy was based on s.c. fluid infusions, metronidazole (25 mg/mL suspension, magistral preparation) 25 mg/kg given p.o. once a day, voriconazole, topical iodine ointment (Betaisodona^® Wound Gel; Mundipharma GmbH, Limburg/Lahn, Germany) and force feeding. On day 135, a blood sample was taken for haematological and plasma chemical analysis (see Table S1 in Supporting information). Despite treatment, the animal's physical condition deteriorated and it had to be euthanized.

Microbiology and histopathology

For bacteriological analysis, the skin sample was incubated on blood agar BD Columbia Agar with 5% sheep blood (BA; Becton Dickinson, Heidelberg, Germany) in aerobic/microaerophilic (5% CO₂)/anaerobic conditions, and BD MacConkey II (Becton Dickinson) at 37°C for 24–48 h. After 24 h of incubation, the predominant colonies (bacterial isolate 2886/12) on BA were small, showing round, mucoid, semi-transparent colony morphology. After 48 h, colonies were surrounded by a narrow zone of haemolysis. Further characterization included Gram reaction, oxidase and catalase activity as described by Loncaric et al.4 For species characterization of the 2886/12 isolate, 16S rRNA gene sequence analysis was performed as described previously.5 The isolate was identified using the EzTaxon-e server (http://eztaxon-e.ezbiocloud.net/).6 The 16S rRNA gene sequence was submitted to Genbank (accession number KF647330) and was >99.9% similar to the type strain of D. agamarum IMP2^T (EU009865). The high 16S rRNA gene sequence similarity between 2886/12 and the type strain of D. agamarum suggested species affiliation.

Apart from D. agamarum, a moderate amount of Enterococcus sp. was isolated; these organisms were Gram positive, catalase negative, had typical colonial appearance on BA and grew well on BD Enterococcosel Agar (Becton Dickinson). For the second bacterial examination (day 106), there were large numbers of Clostridium sp. (they were Gram positive; some cells appeared Gram variable, spore forming and grew in strictly anaerobic conditions, but without the double zone of haemolysis typical for Clostridium perfringens). For mycological analysis, samples were cultivated on BD Sabouraud Agar with gentamicin and chloramphenicol (Becton Dickinson Austria GmbH, Schwechat, Austria) at 30°C for 10 days. Fungal colony characteristics and microscopic features resembled Chrysosporium sp.

Postmortem histopathological findings in the region of the skin lesions confirmed a severe granulomatous dermatitis (Figure 2) with associated fungal hyphae (Figure 3). The liver showed a severe diffuse fatty degeneration and, in some areas, a severe granulomatous hepatitis (Figure 4) with intralesional hyphae (Figure 3, inset). In these skin and liver lesions, bacteria and acid-fast organisms were not found with special stains including haematoxylin and eosin, Ziehl Neelsen, Gram and Brown & Brenn. For identification of fungal species, DNA was extracted from pure cultures isolated from skin samples as well as from formalin-fixed, paraffin-embedded liver tissues using GenElute^TM Mammalian Genomic DNA Miniprep Kit (Sigma-Aldrich Handels GmbH, Vienna, Austria) or BiOstic FFPE Tissue DNA Isolation Kit (MO BIO Laboratories, Carlsbad, CA, USA) according to the manufacturer's instructions. PCR amplification of ITS1, the 5.8S rRNA gene, and ITS2 was performed as described previously and products were sequenced at LGC Genomics, Berlin.7 BLASTn with default settings (http://www.ncbi.nlm.nih.gov/BLAST) was used to compare both obtained sequences (culture from skin sample and the paraffin-embedded liver sample) with database sequences. The two sequences of ITS1-5.8S-ITS2 obtained (from skin and liver samples) were 527 bp and identical. A search on GenBank database using BLAST revealed that the sequences had 100% similarity to a C. guarroi CCFVB CH11 (EU018452.1) sequence. The sequence was submitted to Genbank (accession number KF647329). According to phylogenetic studies, C. guarroi has recently been placed in the genus Nannizziopsis.3

Discussion

Devriesea agamarum is a facultative pathogenic bacterium that is able to cause dermatitis and skin lesions;1 recommended treatments include ceftiofur hydrochloride at a dose of 5 mg/kg once daily.11 Large numbers of Clostridium sp. were isolated at the second bacterial examination (day 106) and were treated with metronidazole. The Enterococcus sp. isolate was considered to be nonpathogenic.

Fungal hyphae were observed during the microscopic examination of the wound exudate, leading to a tentative diagnosis of Aspergillus sp.; however, no fungi were isolated after culture of a swab sample. Later, Chrysosporium sp. was diagnosed in the same animal. It is possible that Chrysosporium sp. was microscopically misinterpreted as Aspergillus sp. and not found because the sample was only incubated routinely for 3 days. Colonies of C. guarroi are slow growing and reach 15–19 mm on Sabouraud dextrose agar after 14 days at 25°C.2 Clinical signs and histopathological findings in this case resembled those described for C. guarroi in green iguanas.12 Voriconazole, the recommended safe systemic therapy of Chrysosporium anamorph of Nannizziopsis vriesii (CANV) in bearded dragons,13 was chosen to treat suspected Aspergillus sp., and it appeared to be clinically effective. Treatment time would have been longer if C. guarroi had been identified at the first visit. After Chrysosporium sp. was isolated in the skin sample, voriconazole was started again. Unfortunately, the animal was already in a poor condition and suffering from a systemic fungal infection.

Samples of suspected Chrysosporium sp. should be cultured at 30°C for 10–14 days. Immediate antifungal treatment is necessary to avoid systemic and potentially fatal infection with C. guarroi. In a previous study, elimination of CANV by voriconazole was achieved on average after 47 days.13 Primary causes or predisposing factors of co-infection remain questionable. Husbandry seemed to be adequate, and other bearded dragons from the same owner, housed separately, were not affected.