Refractory Post-Operative Wound Infection by Arthrographis Kalrae: A Case Report and Review of Literature

Arthrographis kalrae (A. Kalrae) infections, although rare, have often demonstrated an extended clinical course. We report a case of refractory post-operative wound infection with A. Kalrae after radical soft tissue and femur resection following total femoral replacement. Identification of A. Kalrae was done based on detailed microscopic and phenotypic characteristics and biochemical tests. Additionally, it was confirmed with sequencing. Although antifungal therapy was started on the basis of the breakpoints observed, clinical failure was met. However, wound infection was subsequently controlled with extensive tissue debridement and a combination of antifungal drug therapy over a period of 33 weeks. The rarity of global data about such unusual fungal infections adds to diagnostic and therapeutic challenges.

Introduction

Arthrographis kalrae (A. kalrae) was first isolated in 1938 in northern France from an onychomycotic lesion. Kochet described it as a new species, and it was named as A. langeroni [1]. In India, Tewari and Macpherson reported the first case in 1960 from the sputum sample of a patient with an indistinct respiratory illness [2]. They had reported it as Oidiodendron kalrai [2]. Subsequently, A. langeroni and O. kalrae were accommodated in A. kalrae [3]. Although a diversified group of fungi is established to cause cutaneous, subcutaneous, or systemic infections in immunocompetent as well as immunocompromised individuals, sub-cutaneous hyalohyphomycosis caused by A. kalrae is unusual. To the best of our knowledge, we describe the first case of subcutaneous infection by A. kalrae in an immunocompromised young adult in India.

Case Description

A 17-year-old male presented left thigh swelling and pain, with associated redness and skin pigmentation for the past four months. Upon investigation, he was diagnosed with osteosarcoma (High Grade, Stage II) of left femur. No metastases were noted in lungs following chest CT scan, however multiple deep venous thromboses were present in left lower limb, for which he was put on antiplatelet drugs. Patient was treated with adjuvant chemotherapy for osteosarcoma. He also underwent left thigh radical soft tissue and femur resection following total femoral replacement one month prior (Day 1). However, wound was non-healing and was associated tissue necrosis. Two rounds of irrigation and debridement were done for limb salvage, and specimens of wound cultures were taken (Day 25 and Day 32). Despite this, wound healing was poor with progressive necrosis.

Six surgical swabs were sent to the laboratory for Gram and ZN staining, aerobic and anaerobic bacteria culture, acid-fast bacteria culture, and fungal stain and culture (Day 32). Gram staining showed pus cells, even so bacteria or yeast was not detected. ZN stain smear was negative for acid-fast bacilli. However, few hyaline hyphae were appreciated in 10% KOH stain and calcofluor white stain. The patient was immediately started on itraconazole (100 mg OD) (Day 33). Specimens were inoculated on Blood agar, Chocolate agar, MacConkey agar, and 4 sets of Sabouraud’s Dextrose agar (SDA) with chloramphenicol (0.05 mg/ml). No growth was observed at the end of 24 hours in any bacterial culture plates, which were re-incubated for another 24 hours. On day 3 of inoculation (Day 35), a handful of creamy, moist colonies was seen on the SDA plates kept at 37 °C. However, growth was not detected on the SDA plates kept at 25 °C. On day 6 of incubation (Day 38), growth was observed in plates kept at 25 °C. The initial colonies were cream-coloured and glabrous. It gradually turned velvety, maize in colour, and eventually formed a yellowish powdery colony with soft and frangible texture after 7 days (Day 39; see Figs. 1 and 2).

Fig 1. Initial cream-coloured moist colonies of A. kalrae (Inoculation Day 3).

Fig. 2. Maize Velvety colonies of A. kalrae (Inoculation Day 7).

The fungal isolate could not be identified by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI TOF MS) (bioMérieux, France). The isolate was, therefore, sent for genomic sequencing, and further morphological and biochemical tests were performed simultaneously. The fungus was found to be weak urease positive, cycloheximide resistant, nitrate non-assimilator, and showed growth at 42°C. The isolate on slide cultures, grown on SDA and potato dextrose agar (PDA), appeared to have irregular or non-branched hyphae with dendritic conidiophores accompanied with chains of rectangular, single-celled arthroconidia. Occasional hyaline sessile conidia were also observed (Fig. 3). The isolate was presumptively identified to belong to the genus Arthrographis.

Fig. 3. LPCB Mount of A. kalrae showing dendritic conidiophores and rectangular arthroconidia (Magnification 40X).

Pan fungal DNA polymerase chain reaction (PCR) and sequencing were used for the molecular identification of the isolate. The ribosomal DNA (rDNA), genes (18S, 28S, and 5.8S), and the intervening internal transcribed spacer (ITS) regions (ITS1 and ITS2) were targeted. The 3500 DX Analyzer (Thermo Fisher Scientific, Massachusetts, USA) was used. For species identification, sequences were run through GenBank Basic Local Alignment Search Tool (BLAST) searches [4], [5]. BLAST searches confirmed the isolate as A. kalrae with 99.64% identification (GenBank accession no. KM588309.1) of ITS1, ITS2, 5.8S gene, and rRNA gene (Day 47). Antifungal susceptibility test (AFST) was done by the CLSI recommended broth micro-dilution (BMD) method for azoles (fluconazole, voriconazole, posaconazole and itraconazole), amphotericin B, echinocandins (caspofungin, micafungin and anidulafungin) and terbinafine. Breakpoints observed were fluconazole = 16 µg/mL, voriconazole = 0.5 µg/mL, posaconazole = 0.25 µg/mL, itraconazole = 0.5 µg/mL, amphotericin B = 1 µg/mL, caspofungin = 16 µg/mL, micafungin= 8 µg/mL anidulafungin = 8 µg/mL and terbinafine = 0.015 µg/mL (Day 48).

Treatment and Clinical Course

The patient was put on itraconazole (100 mg OD) initially for 2 weeks (Day 33–Day 48). Following the AFST result, liposomal amphotericin B (200 mg thrice a week) was added and continued for 4 weeks (Day 49–Day 80). Despite in vitro sensitive breakpoints, there was no significant wound healing, and the cultures remained positive (Day 70). The patient again underwent left thigh irrigation and debridement (Day 71). AFST was repeated by BMD, and the breakpoints observed were similar to those obtained in the first AFST (Day 80). The patient was put on intravenous liposomal amphotericin B (300 mg daily) along with posaconazole (400 mg OS) and terbinafine (500 mg OD) (Day 80) for 6 weeks (Till Day 122). Vacuum-assisted closure (VAC) dressing was applied. Wound healing started, and the culture of A. kalrae became negative (Day 115). Posaconazole prophylaxis was continued till the patient was operated on for osteosarcoma (Day 138) and then after in a post-operative period with adjuvant chemotherapy for 18 weeks with close follow-up (Day 264). No repeat infection by A. kalrae was noted during the follow-up period.

Discussion

A. kalrae belongs to Genus Arthrographis, a saprophytic fungus that is found in compost and soil globally [6]. A surge in clinical cases of invasive infections caused by A. kalrae has been noted [6]. Thus far, one case with Arthrographis spp. and 20 cases with A. kalrae have been reported with diverse clinical presentations and global distribution (see Table I). This includes seven cases of keratitis [8]–[14], one of which was associated with sinusitis [11], five pulmonary infections [6], [19]–[22], three arthritic infections [15]–[17], and two cases of onychomycoses [1], [7], and one each of endocarditis [18], meningitis [23] and stroke [24].

Table I. Characteristics of Previously Published Cases of Infections with A. Kalrae

Reference	Age	Gender	Country	Pathology	Risk factor	Surgery	Antifungal drugs	Outcome
[1]	63	M	Japan	Onychomycosis	NA	NA	Terbinafine + Miconazole	Relapse
[6]	19	F	France	Pulmonary	Cystic Fibrosis + Lung Transplant	Yes	Caspofungin + Amphotericin B	Death (8 months later)
[7]	NA	NA	Slovakia	Onychomycosis	NA	NA	NA	NA
[8]	23	F	USA	Keratitis	Contact lens	No	Itraconazole + Amphotericin B	Recovery
[9]	42	F	Germany	Keratitis	Contact lens	Yes	Voriconazole	Chronicity
[10]	52	M	Malaysia	Keratitis	Trauma	Yes	Fluconazole + Amphotericin B	Chronicity
[11]	39	M	China	Keratitis, Sinusitis	Trauma	Yes	Itraconazole + Amphotericin B	Recovery
[12]	23	F	USA	Keratitis	Contact lens	Yes	Itraconazole + Amphotericin B	Chronicity
[13]	52	M	Malaysia	Keratitis	Trauma	Yes	Amphotericin B + Fluconazole	Chronicity
[14]	38	F	Spain	Keratitis	Excimer laser keratomileusis	No	Natamycin + Voriconazole	Chronicity
[15]	33	M	Australia	Knee joint infection	Injury	Yes	Voriconazole + Fluconazole + Posaconazole	Chronicity
[16]	68	M	Italy	Knee joint infection	Diabetes mellitus, injury	No	Voriconazole	Recovery
[17]	75	M	Australia	Knee joint infection	Diabetes mellitus, TKR	Yes	Voriconazole + Terbinafine	NA
[18]	50	F	Spain	Endocarditis	ASD Repair, AVR	Yes	Posaconazole	Relapse
[19]	61	M	Netherlands	Pulmonary	Radiotherapy	Yes	Itraconazole	Recovery
[20]	64	M	USA	Pulmonary	NHL, Allogenic HSCT	No	Voriconazole	Death (3 months later)
[21]	51	F	India	Pulmonary	AIDS	No	NA	NA
[22]	7	M	Italy	Pulmonary	Cystic fibrosis	No	Itraconazole	Recovery
[23]	33	M	USA	Meningitis	AIDS	No	Fluconazole	Death
[24]	39	M	France	Fungal stroke	Malnutrition	No	No	Death (2 days later)
Present case	17	M	India	Sub-cutaneous wound infection	Osteosarcoma, previous surgeries	Yes	Itraconazole + Amphotericin B + Posaconazole + Terbinafine	Chronicity

Nine cases were reported in Europe, followed by four in the US, two in Malaysia and Australia, and one in India, Japan, and China each. To the best of our knowledge, the present case is the first reported case of post-operative non-healing wound infection caused by A. kalrae. Three of the seven patients with keratitis were soft contact lens wearers; two out of three with knee joint infections had an injury contaminated with soil. Predisposing infection factors, such as AIDS, allogeneic hematopoietic stem cell transplant, malnutrition, radiotherapy, and systemic corticosteroids, were noted in the rest of the patients (see Table I). A. kalrae is believed to enter the human body through the respiratory tract or by inoculation due to trauma [17], [20]. However, the source of A. kalrae infection remained unrecognized in the majority of reported cases. In the present case, the patient had predisposing factors like an immunocompromised state and previous multiple surgeries. There is a possibility of incidental inoculation in the post-operative period. However, it is contingent, and the source of infection remains undetermined.

The presence of uni-celled, cylindrical, hyaline, smooth-walled arthroconidia is a characteristic of the Arthrographis spp. These arthroconidia are formed either directly by fragmentation of undifferentiated hyphae or in new cultures by disjunction and segmentation of fertile hyaline branches. Upon maturation, arthroconidia become large and elongated. Additionally, uni-celled, smooth, spherical hyaline blastoconidia develop directly on the sides of either non-differentiated hyphae or on pedicles [15], [17]. Nonetheless, identification based on microscopy alone is difficult. A. kalrae can be easily confused with Candida albicans because of its primal ivory-colored yeast-like appearance [14]. Furthermore, the presence of abundant arthroconidia can lead to its misidentification as Geotrichum spp. or Trichosporon spp [25]. In previously published reports, phenotypic characteristics and microscopic morphology based on de Hoog and Sigler, as well as Carmichael descriptions, helped in diagnoses [3], [26]. In most cases, species-level identification was confirmed by either MALDI TOF or ITS, D1/D2 sequencing. In the present report, detailed microscopic and phenotypic characteristics and biochemical tests help us reach genus-level identification, which was confirmed by sequencing. MALDI TOF could not identify the causal fungi.

There is no definitive treatment for A. kalrae infections. Thus far, scarcely any reports of A. kalrae infection have covered in vitro AFST data (Table II).

Table II. Antifungal Drugs Breakpoints of A. kalrae Isolates in Previously Published Case Reports

References	Sugiura et al. [1]	Denis et al. [6]	Boan et al. [15]	Sainaghi et al. [16]	Chin-Hong et al. [23]	de Hoog et al. [26]	Present study AFST 1	Present study AFST 2
No. of isolates	01	01	04	01	01	22	01	01
Antifungal drug	Breakpoints (µg/mL)
Fluconazole	>64	>32	8–16	1	8	–	16	16
Voriconazole	–	0.125	0.06–0.5	0.030	–	0.06–1	0.5	1
Posaconazole	–	0.50	0.25–0.5	0.060	–	<0.03–0.5	0.25	0.25
Itraconazole	8	–	0.25–0.5	0.120	<0.015	<0.03–2	0.25	0.5
Amphotericin B	0.5–1	0.75	0.25–0.5	0.5	0.25	1 – 4	1	1
Caspofungin	–	–	–	0.250	–	0.5–>8	16	16
Micafungin	–	0.012	–	1	–	<0.015–>8	8	8
Anidulafungin	>16	0.047	–	0.5	–	<0.015–>8	8	8
Terbinafine	0.015	–	–	–	–	<0.004–0.06	0.015	0.015
Flucytosine	>64	>32	>64	0.060	–	–	–	–

de Hoog et al. [26] studied in vitro AFST patterns of 22 clinical isolates of A. kalrae and terbinafine was found to be the best therapeutic option, followed by the azoles (Posaconazole > Itraconazole/Voriconazole). However, echinocandins showed almost no activity with higher minimum effective concentrations (MECs). Our AFST findings are in accordance with the previously published findings. In the present report, AFST was performed twice using the BMD method with reproducible breakpoints. Despite in vitro sensitive breakpoints, initially, the patient did not respond to the amphotericin B and itraconazole combination. Low levels of drugs in serum and tissues could have possibly contributed to initial clinical failure. However, this cannot be said definitively, as therapeutic drug monitoring was not done. Additionally, the patient was severely immunocompromised, a condition where antifungal agents alone, including high-dose fungicidal agents, are known to be unable to eradicate the fungi [27]. Nonetheless, increased dosages of amphotericin B (concentration-dependent activity), along with combination therapy of terbinafine (BMQ) and posaconazole, helped in wound healing. Molecular characterization of isolated strains was not done for any point mutations. Generally, a treatment period of 3–12 months is advocated [20]. In the present case, treatment with antifungal drugs was given for 33 weeks. Additionally, surgical debridement and VAC dressing were done, which proved to be effective, and no repeat infection by A. kalrae was observed.

Conclusion

We report the first case of a sub-cutaneous A.kalrae infection in India. A. kalrae infections are rare, though they are likely to be under-recognised and under-diagnosed. Knowledge of morphological characteristics and confirmation by molecular investigations are deemed important. Terbinafine appears to be the drug of choice, followed by the azoles; however, further clinical evidence is needed to advocate these deductions. Despite combination therapy of antifungal drugs with sensitive breakpoints along with aggressive treatment, A.kalrae infection can be refractory. Such unusual mycotic infections with a limited corpus delicti add significantly to diagnostic and therapeutic challenges.