Chromoblastomycosis caused by Fonsecaea: clinicopathology, susceptibility and molecular identification of seven consecutive cases in southern China

Authors


  • The first two authors contributed equally to this study.

Corresponding author: Y.-M. Fan, Department of Dermatology, Affiliated Hospital of Guangdong Medical College, Zhanjiang, Guangdong 524001, China.

E-mail: ymfan1963@163.com

Abstract

The clinicopathological and microbiological features of chromoblastomycosis caused by Fonsecaea pedrosoi or Fonsecaea monophora are summarized. Four F. monophora and three F. pedrosoi strains were isolated from seven consecutive chromoblastomycosis patients and identified by genetic analysis between 2004 and 2012 in a teaching hospital in southern China. Six strains were sensitive to voriconazole, itraconazole and terbinafine using E-test and Neo-Sensitabs. Six patients healed after oral itraconazole or terbinafine, and one was lost to follow up. Internal transcribed spacer sequence is sufficient for species delimitation of Fonsecaea, and the Neo-Sensitabs test and E-test are comparable in their susceptibility testing. Itraconazole and/or terbinafine may be the preferred treatment for this chromoblastomycosis.

Introduction

Chromoblastomycosis is characterized by polymorphic skin lesions and muriform cells in the granulomatous and suppurative tissue. Several dematiaceous fungi including the genera Fonsecaea, Phialophora and Cladophialophora can lead to chromoblastomycosis, of which Fonsecaea pedrosoi and Cladophialophora carrionii are most frequent [1, 2]. More than 500 cases of chromoblastomycosis have been documented in mainland China since it was first reported in 1952, and the predominant pathogens are C. carrionii in northern China and F. pedrosoi in southern China [1]. The genus Fonsecaea comprises three species that can cause human infections: F. pedrosoi, F. monophora and F. nubica; F. pedrosoi and F. nubica are strictly associated with chromoblastomycosis, whereas F. monophora is also involved in phaeohyphomycosis of brain and other organs [3]. Fonsecaea multimorphosa was isolated from a feline cerebral abscess [4].

The skin lesions of chromoblastomycosis include nodular, tumorous, verrucous, plaque and cicatricial types, and their severity is divided into mild, moderate and severe forms [2]. Definite diagnosis of chromoblastomycosis is based on clinicopathological data, especially recognition of the muriform cells in tissue and skin scrapings [2, 5]. Although morphological distinction of Fonsecaea species is difficult, PCR-based methods have been developed for their separation [3, 5]. Chromoblastomycosis is still a therapeutic challenge because of the marked chronic lesions, dense dermal fibrosis and oedema, and there is no specified treatment of choice [2, 6].

Because the clinicopathological and microbiological data of chromoblastomycosis caused by F. pedrosoi or F. monophora are rarely compared, we aimed to delineate the clinical, pathological and microbiological findings of seven consecutive patients diagnosed by molecular identification in our teaching hospital between 2004 and 2012.

Patients and Methods

Sample collection

Clinical, pathological and microbiological data of seven consecutive patients with culture-proven chromoblastomycosis caused by Fonsecaea were retrospectively reviewed at our department from 2004 to 2012. Skin biopsies were performed in all cases, and the specimens were fixed in 10% neutral buffered formalin, embedded in paraffin, sectioned and stained with haematoxylin & eosin (H&E), periodic acid-Schiff (PAS) and argyrophilic nucleolar organizer region (AgNOR) stains. Informed consents were obtained from all patients, and this study was also approved by the ethics committee of our hospital.

Mycological examination

Smears of skin scrapings and crushed tissues were observed in 10% potassium hydroxide, and cultured on Sabouraud dextrose agar to isolate the agents. The isolates were identified by slide-culture microscopy.

DNA sequence analysis

Genomic DNA was extracted using a fungal DNA Mini Kit (Omega Bio-Tek Inc., Norcross, GA, USA). The internal transcribed spacer (ITS) and D1/D2 domain of rDNA were amplified by ITS1 (5′-GTC GTA ACA AGG TTT CCG TAG GTG-3′) and ITS4 (5′-TCC TCC GCT TAT TGA TAT GC-3′), and NL1 (5′-GCA TAT CGG TAA GCG GAG GAA AAG-3′) and NL4 (5′-GGT CCG TGT TTC AAG ACG G-3′), respectively. Amplification products underwent direct sequencing, and the sequences were aligned with those in GenBank by BLAST analysis.

Antifungal susceptibility testing

In vitro antifungal susceptibility testing of six Fonsecaea strains was performed using Neo-Sensitabs (Rosco, Taastrup, Denmark) and E-test (AB bioMérieux, Askim, Sweden) following the manufacturers' instructions. Inoculum suspensions were prepared by homogenizing moulds in 0.85% NaCl to achieve 0.5 McFarland turbidity. Modified Shadomy medium plates with 90-mm diameter (Guangzhou Detgerm Microbiology Technology Co., Ltd, Guangzhou, China) were used. Inoculum suspension (a 0.5-mL sub-sample) was added to each plate, and smeared evenly on the agar surface with a disposable sterile swab. After drying at 35°C for 15 min, E-test strips or Neo-Sensitabs tablets were gently placed on the lawn cultures. The MICs and inhibition zone diameter were determined after 5 days of incubation at 27°C.

Results

The clinical data of four men and three women are summarized in Table 1. There were six farmers and one worker, with ages ranging from 45 to 84 years (mean of 63.3 years). All patients were immunocompetent except for case 7 who had untreated diabetes. The mean duration before presentation was 49.3 months (range, 6–120 months). Only three cases had a history of local injury caused by chicken or wood. All skin lesions involved the extremities. Five patients had a solitary plaque, one had three plaques, and one had a well-demarcated ulcer. All but one had pain and itching. Of seven patients, five achieved a complete resolution of skin lesions with itraconazole (0.2–0.4 g/day) for an average duration of 4.1 months (range, 3–8 months) (Fig. 1a,b), and no relapse for a mean follow-up of 43.4 months (range, 13–91 months); one underwent a combined treatment with itraconazole (0.2 g/day) for 5 months and terbinafine (0.25 g/day) for 1 month, and his skin lesions healed substantially (Fig. 2a,b); one patient was lost to follow up.

Table 1. Clinical and fungal results of seven cases of Fonsecaea-induced chromoblastomycosis
CaseDateAge (years)/sexClinical features/siteDuration (m)Treatment regimenFollow up/responsePathogenGenBank accession number
  1. F, female; M, male; ITZ, itraconazole; TBF, terbinafine; ITS, internal transcribed spacer;

  2. a

    History of local trauma.

  3. b

    Associated with diabetes.

12004.965/F4.5 × 3.5 cm plaque/right ankle12ITZ (0.2 g/day) for 3 months91 months/cure F. pedrosoi

ITS: JN418758

D1/D2: JN418759

22006.251/M5 × 3 cm plaque/right wrist72ITZ (0.2 g/day) for 8 months69 months/cure F. monophora

ITS: JQ906791

D1/D2: JQ906792

3a2008.576/F4 × 2.5 cm ulcer/left leg9NoNo F. pedrosoi

ITS: JQ906789

D1/D2: JQ906790

4a2010.563/F4 × 4 cm plaque/left hand6ITZ (0.2 g/day) for 3.5 months22 months/cure F. monophora

ITS: JN418760

D1/D2: JN418761

5a2010.559/M5 × 4 cm plaque/right hand6ITZ (0.4 g/day) for 3 months22 months/cure F. monophora

ITS: JQ906793

D1/D2: JQ906794

62011.345/M5 × 3 cm plaque/right wrist120ITZ (0.2 g/day) for 3 m13 m/cure F. monophora

ITS: JN418762

D1/D2: JN418763

7b2012.284/M10 × 5 cm, 4.5 × 2.5 cm and 1 × 0.8 cm plaques/right knee120ITZ (0.2 g/day) for 5 months, TBF (0.25 g/day) for 1 monthsIn treatment F. pedrosoi

ITS: JX134567

D1/D2: JX134568

Figure 1.

Clinicopathological and fungal examination of Fonsecaea monophora-induced chromoblastomycosis. (a) Solitary erythematous plaque on the dorsum of left hand. (b) Complete healing with hypopigmented scarring at 22-month follow up. (c) Skin scrapings showing muriform cells (10% KOH; original magnification, ×400). (d) Microculture on potato dextrose agar at 27°C on day 16 showing branched septate hyphae with conidiophores bearing short conidial chains (original magnification, ×400). (e) Histopathology revealing three brown acropetal septate chlamydospores (haematoxylin & eosin; original magnification, ×1000). (f) Histopathology revealing muriform cells (periodic acid-Schiff; original magnification, ×1000).

Figure 2.

Clinicopathological and fungal examination of Fonsecaea pedrosoi-induced chromoblastomycosis. (a) Three erythematous plaques with dark brown crusts on the right knee. (b) Nearly complete healing with hypopigmented scarring after 5 months of treatment. (c) Skin scrapings showing brown muriform cells and yellowish-brown septate hyphae and acropetal chlamydospores (10% KOH; original magnification, ×400). (d) Microculture on potato dextrose agar at 27°C on day 12 showing branched septate hyphae with conidiophores bearing short conidial chains (original magnification, ×400). (e). Histopathology revealing brown muriform cells (haematoxylin & eosin; original magnification, ×1000). (f). Histopathology revealing dark brown muriform cells (argyrophilic nucleolar organizer region; original magnification, ×1000).

Direct smears revealed brown muriform cells and yellowish-brown septate hyphae and acropetal chlamydospores (Figs 1c and 2c). Fungal culture yielded dark olivaceous to black fuzzy colonies at 27°C on Sabouraud dextrose agar after 2 weeks. Microscopic examination of slide culture showed pale olivaceous septate hyphae, and loosely branched conidiophores bearing ellipsoidal conidia in short chains (Figs 1d and 2d). Histopathology of biopsy specimens in seven cases demonstrated hyperkeratosis, pseudo-epitheliomatous hyperplasia and mixed granulomatous inflammation and neutrophil microabscesses in the dermis. Staining with HE, PAS and AgNOR revealed the muriform cells in granulomas, microabscesses or giant cells (Figs 1e–f and 2e–f).

The ITS and D1/D2 sequences of cases 1, 3 and 7 displayed 100% identity to a few strains of F. pedrosoi in GenBank except the D1/D2 sequence of case 3 had one base difference. The sequences of cases 2, 4, 5 and 6 showed 100% homology to F. monophora (Table 1).

E-test results showed that the mean MICs of the six Fonsecaea isolates were 0.024 (0.012–0.047) mg/L against voriconazole, 0.125 mg/L against itraconazole, >32 mg/L against amphotericin B and caspofungin, and >256 mg/L against fluconazole. The mean MICs were 0.025 (0.012–0.047) mg/L against voriconazole for three F. monophora strains and 0.024 (0.012–0.047) mg/L for three F. pedrosoi strains. Neo-Sensitabs results revealed that the six Fonsecaea isolates were sensitive to voriconazole, itraconazole, ketoconazole and terbinafine, but resistant to fluconazole and amphotericin B, except for one strain of F. pedrosoi that was susceptible to amphotericin B (Table 2).

Table 2. Minimal inhibitory concentration and inhibition zone diameter for six Fonsecaea isolates determined by E-test and Neo-Sensitabs
CasePathogenMIC (mg/L)/IZD (mm)
VCZITZAMBFLZCASKTZTBF
  1. IZD, inhibition zone diameter; VCZ, voriconazole; ITZ, itraconazole; AMB, amphotericin B; FLZ, fluconazole; CAS, caspofungin; KTZ, ketoconazole; TBF, terbinafine; S, susceptible; R, resistant; –, not done.

1 F. pedrosoi 0.047/30(S)0.125/35(S)>32/0(R)>256/0(R)>32/––/28(S)–/54(S)
3 F. pedrosoi 0.012/52(S)0.125/36(S)>32/0(R)>256/0(R)>32/––/50(S)–/42(S)
4 F. monophora 0.047/43(S)0.125/36(S)>32/0(R)>256/0(R)>32/––/37(S)–/62(S)
5 F. monophora 0.016/44(S)0.125/40(S)>32/0(R)>256/0(R)>32/––/50(S)–/49(S)
6 F. monophora 0.012/55(S)0.125/40(S)>32/0(R)>256/0(R)>32/––/40(S)–/65(S)
7 F. pedrosoi 0.012/46(S)0.125/45(S)>32/25(S)>256/0(R)>32/––/40(S)–/48(S)

Discussion

Fonsecaea species are saprobes in rotten wood, live plants and soil [7]. Human Fonsecaea infections are presumably acquired after being pricked by contaminated thorns or wood splinters, but the natural source, infection process and dispersal routes are still unknown [8, 9]. The environmental strains cannot always be linked directly to clinical cases. The available Fonsecaea strains can almost always be isolated from humans, but rarely from the environment [8]. Fonsecaea monophora and unknown Fonsecaea species were isolated from an environmental sampling performed in southern Brazil, but F. pedrosoi was not [9]. Both F. pedrosoi and F. monophora are easily misidentified in clinical practice because both have similar morphological features. However, their pathogenic potential is different. Fonsecaea pedrosoi may almost be a pathogen of chromoblastomycosis, whereas F. monophora seems to be a general opportunist [6]. In the seven subjects, all cases had lesions on exposed areas, but only three had a history of trauma and one had diabetes. Therefore, these infections mainly involve otherwise healthy individuals and might originate from minor traumatic implantation of environmental fungi.

Diagnosis of chromoblastomycosis is dependent on direct examination, fungal culture and histopathology, and the existence of muriform cells in clinical samples is mandatory [2]. In our study, sections stained with HE, PAS and AgNOR from all patients revealed mixed granulomatous response, microabscesses and muriform cells. Although multiple PCR-based methods are suitable for their identification, ITS sequence analysis and phylogenetic study are most common [3, 5, 7, 8]. In line with the above reports, our study also demonstrates that ITS sequence analysis is superior to the D1/D2 sequence in species identification of Fonsecaea.

The therapeutic outcomes of chromoblastomycosis are variable according to infection site, lesion size, pathogen and the host's health status. Several treatment options yield protracted disease, low cure rates and frequent relapse [10]. Severe lesions tend to respond slowly or even become non-responsive to antifungal drugs [2]. In this study, five patients with mild lesions had a satisfactory response to oral itraconazole. In Case 7—an 84-year-old man with diabetes and a 10-year history of multiple lesions—the skin lesions showed an insignificant response to itraconazole at the start of therapy, but disappeared almost completely after combined administration of itraconazole and terbinafine. The advanced age, associated diabetes, long-standing course, and moderate lesions could have accounted for the poor response to itraconazole monotherapy.

Our susceptibility results showed that all strains were sensitive to voriconazole and itraconazole by using E-test, and to voriconazole, itraconazole, ketoconazole and terbinafine by Neo-Sensitabs. Clinical Fonsecaea isolates including F. pedrosoi, F. monophora and F. nubica were susceptible to posaconazole, itraconazole, isavuconazole and voriconazole, and the antifungal activities of these drugs were not significantly different [10]. Itraconazole and terbinafine had a synergistic effect against the F. monophora isolates [11]. The Neo-Sensitabs method has been proven to be a quick and simple alternative to the E-test assay for evaluating the susceptibility of filamentous fungi to antifungals [12]. Our observation also suggests that the Neo-Sensitabs test is comparable to E-test, and in vitro susceptibility results of these strains are compatible with clinical responses.

In conclusion, clinicopathological and mycological features are alike in Fonsecaea-induced chromoblastomycosis, and ITS sequence analysis is sufficient for genetic delimitation of Fonsecaea. Itraconazole and terbinafine may be the antifungals of choice in the treatment of this chromoblastomycosis.

Acknowledgements

We thank Prof. Feng-Yan Bai (Systematic Mycology and Lichenology Laboratory, Institute of Microbiology, Chinese Academy of Sciences, Beijing, China) for assistance with the genetic identification.

Transparency Declaration

The authors declare no conflicts of interest. This study received no funding

Ancillary