Candida parapsilosis and Candida guillermondii: emerging pathogens in nail candidiasis.

BACKGROUND: Onychomycosis of the fingernails and toenails is generally caused by dermatophytes and yeasts. Toenail mycoses involve mainly dermatophytes but when Candida is also involved, the strain most commonly isolated worldwide is C. albicans. AIMS: To determine Candida strains prevailing in onychomycosis.
MATERIALS AND METHODS: A retrospective, observational and descriptive study of fungal cultures retrieved from the registry of the microbiology laboratory of the Pontificia Universidad Católica was performed. Specimens obtained from patients attending the healthcare network between December 2007 and December 2010 was analyzed. STATISTICAL ANALYSIS: A descriptive statistical analysis was performed.
RESULTS: Candida was retrieved from 467 of 8443 specimens (52% fingernails and 48% toenails). Cultures were negative in 5320 specimens (63.6%). Among Candida-positive cultures, parapsilosis was the most commonly isolated strain with 202 cases (43.3%). While isolates of Candida guillermondii were 113 (24.2%), those of Candida albicans were 110 (23.6%), those of spp. were 20 (4.3%) and there were 22 cases of other isolates (4.71%). Among the 467 patients with positive cultures for Candida, 136 (29,1%) were men and 331 (70,9%) were women. All patients were older than 18 years old. Clinical files were available for only 169 of the 467 patients with positive cultures for Candida. For those, age, gender, underlying illnesses and use of immunossupresive agents during the trial was reviewed.
CONCLUSIONS: The present study shows that both C. parapsilosis as well as C. guillermondii appear as emerging pathogens that would be in fact taking the place of C. albicans as the most commonly isolated pathogen in patients with Candida onychomycosis. The relative percentage of C parapsilosis increases every year. Identification of Candida strains as etiological agents of nail candidiasis becomes relevant to the management both nail as well as systemic candidiasis, in view of the resistance to conventional treatments readily reported in the literature.

What was known?

Toenail mycoses involve mainly dermatophytes but when Candida is also involved, the strain most commonly isolated worldwide is C albicans.

Introduction

Approximately 20% of individuals aged between 40 and 60 have onychomycosis.[12] Three main groups of fungi are involved: dermatophytes, yeasts of the Candida genus and molds.

The role of Candida species in the production of onychomycosis has been largely discussed. Over years, nail candidiasis was considered only as those cases related to hand paronychia. Progress in mycology has finally recognized candida's ability to invade the nail plate and cause nail disorders indistinguishable from those generated by dermatophytes.

Candida albicans and Candida parapsilosis, among others, may be commensals and be part of the skin normal flora; however in immunodeficiency settings they are able to cause severe systemic infections. Systemic infections caused by Candida albicans are widely known, but Candida parapsilosis is emerging as culprit for severe infections among inpatients.[3] In a 10-year-study conducted in Texas, C. glabrata was associated to candidemia in patients with hematological malignancies, favored by neutropenia. C. parapsilosis was associated to the use of central venous catheters, a finding in accordance with the reports of Warnock.[45] C. tropicalis is emerging as an important pathogen in neutropenic patients and C. parapsilosis is doing so among patients receiving transparentheral nutrition. C. glabrata and C. krusei have rapidly developed resistance to fluconazole.[5]

In general it is accepted that patients undergoing nail candidiasis must have an intercurrent disease able to cause immunosuppression. Among the latter are HIV/AIDS, diabetes, circulation disorders, connective tissue diseases and prolonged immunosuppressive therapy with cytostatics or steroids. However, various recent studies have demonstrated the role of Candida in the generation of onychomycosis without referring to underlying causes, where Candida has been responsible for a very variable percentage of onychomycosis, between 3,5 and 58,5%.[16789]

When nail candidiasis associates to fingernail paronychia, particularly in women who keep their hands under soapy water for long periods of time, clinical diagnosis is simple and mycological exam just confirms a clinical suspicion. The case of toenail onychomycoses, either distal-lateral subungual or total dystrophic forms, is different. In such cases, the finding of Candida in a mycologic test must meet certain requirements for Candida to be considered the causal agent.

Recently, Candida is coming into sight as an important cause of onychomycosis, as confirmed by studies from India and Spain, where their authors are proving that Candida is the main cause of onychomycosis in their geographical location.[210] In an American study involving eleven states between 1997 and 2002, Candida was responsible for 70% of onychomycoses, especially in cases affecting fingernails.[8]

Internationally, there is not one common criterion to attribute the etiology of onychomycoses to Candida. The analysis of various works shows that methods differ in the number of inocula required to be positive. There are also differences in whether the repetition of the test twice or more is required to attribute the cause of the nail affection to a non-dermatophyte fungus.

Walshe and English criteria for the diagnosis of molds in onychomycoses, proposed in 1966, refer to the number of nail fragments positive for the suspected fungus. Five or more than twenty inocula in the absence of dermatophytes and a compatible direct test would be enough to make an etiologic diagnosis.[11] However, in 2001 Gupta et al. demonstrated that such method issued the correct diagnosis only in 23.2% of the cases.[7]

The prevalence of Candida onychomycoses and the species more commonly involved are a key issue to assess. As previously stated, the review of current works shows that the epidemiology of Candida onychomycosis is different depending on the geographical location. Undoubtedly, the most commonly reported species is albicans, both among adults and children.[112131415] however, the emergence of other species in various sites of the world is pointing towards a change in epidemiological behavior of the disease: C. krusei, C. glabrata, C. parapsilosis and C. tropicalis in India and Singapore.[216] C. tropicalis and C. guillermondii in Malta.[17] C. parapsilosis, C. tropicalis and C. guillermondii in Brazil.[18] C. parapsilosis, C. albicans and C. glabrata in Mexico,[19] C. guillermondii and C. albicans in Germany,[20] and C. tropicalis, C. albicans, C. glabrata and C. krusei in Turkey.[21] In Sao Paulo nail candidiasis outnumbered nail dermatophytosis, with C. albicans and C. parapsilosis being found more commonly, in the same order, particularly in fingernail onychomycosis.[22] Among the isolated species of Candida there are some resistant to various commonly used antifungals, such as Candida glabrata which is resistant to imidazoles.[16]

Our objective was to evaluate the presence of various Candida species in Candida onychomycosis, and to assess their relative significance, in order to either confirm the need to modify our management protocols or to continue with therapies currently used.

Materials and Methods

A retrospective, observational and descriptive study of 8443 fungal cultures was performed. Mycological cultures taken within the health network of Pontificia Universidad Católica and concentrated at the Laboratory of the Centro Médico San Joaquín were reviewed. The three year analysis period included specimens from December 2007 to December 2010. Specimens of nails were collected scrapping the surface of the nail plates to perform a direct microscopy study after treatment with 20% potassium hydroxide and were seeded on Sabouraud glucose agar with the addition of chloramphenicol and lactrimel agar, for 25 days incubation at 30°C, with periodic revision and observation. Yeast identification was achieved with the germinative tube test and through commercial tests of the API ID 32C system, BioMerieux (France), as per manufacturer's specifications.

The strains ATCC 90028 (Candida albicans), ATCC 9968 (Candida tropicalis) and ATCC 64677 (Candida glabrata) were used as controls.[2324]

Wherever available, files of patients who had a positive culture for Candida were analyzed for clinical information like their age, gender, underlying illnesses and/or use of immunosuppressive agents during the trial.

A descriptive statistical analysis was performed, because we have included all the specimens obtained from patients attending the healthcare network between December 2007 and December 2010. We have not excluded any of them from the study.

Results

A total of 8443 nail specimens were analyzed. Of the total of specimens analyzed, 3123 were positive (36.9%). Candida grew in 467 specimens (15%), of which 223 were from toenails (47.8%) and 244 from fingernails (52.2%) [Table 1]. The isolated Candida species were mainly C. parapsilosis, in 202 cases (43.3%), C. guillermondii in 113 (24,2%) and C. albicans in 110 cases (23,6%). Together the 3 strains represented 91.1% of all Candida detections.

Table 1

Strains of Candida isolated by year

Dermatophytes were isolated from 2637 specimens and they accounted mainly for T. rubrum (88,2%) and T. mentagrophytes (11.6%) [Table 2]. Molds were isolated from 18 specimens, with Fusarium sp being the most common, accounting for 61% of mold isolates [Table 3]. When analyzing the proportion of Candida species per year, an increase of C. parapsilosis in 25.62% is evidenced [Table 1].

Table 2

Total dermatophytes

Table 3

Total molds

Among the 467 patients with positive cultures for Candida, 136 (29,1%) were men and 331 (70,9%) were women. All patients were older than 18 yrs old. Clinical files were available for only 169 of the 467 patients with positive cultures for Candida. For those patients, age, gender, underlying illnesses and use of immunossupresive agents during the trial were reviewed. Among those patients, 14 (8%) had type 2 diabetes, 11 (7%) connective tissue disorders, 9 (5%) psoriasis, 3 (2%) inflammatory bowel disease, 1 (1%) HIV, 1 (1%) hepatic cirrhosis and 1 (1%) Ewing Sarcoma. 13 (8%) of the patients who were suffering underlying illnesses where receiving immunosuppressive therapy with prednisone, methotrexate or azathioprine, for more than a year.

Discussion

The analysis of results revealed that, Candida represents 15% of all studied onychomycoses, and that it affected fingernails and toenails with similar rates.

The results of Chilean studies reported in 1981 showed that Candida accounted for 25.6% of all onychomycoses, those reported in 1987 evidenced that Candida accounted for 51% of all onychomycoses.[2526] It is worth noting that in occasions, Candida presents as the only agent and in others it is a concomitant agent together with molds or dermatophytes. The coexistence of a dermatophyte with yeasts or molds in the mycological test, is very indicative of a dermatophyte as the real culprit for the problem.[27]

Gender is also a factor that has an impact on results, since Candida is more common in women who permanently get their hands wet. Such circumstances might explain the differences seen in different studies, where proportions of specimens from women and men differ.

In our work, the most striking fact is the predominance of C. parapsilosis, accounting for 43.3% of the cases studied, and that is twice as frequent as C. guillermondii and C. albicans. That the proportion of C. parapsilosis in nails increases in a sustained manner between 2008 and 2010, while C. albicans and C. guillermondii do not follow such a growth curve and stay in relatively stable proportions, is striking [Figure 1].

Figure 1

Rate of candida isolated by year

We could also infer that most of the patients with nail candidiasis do not present comorbidities which could alter their host immunity, as has been described in the literature and that coincides with foreign publications. However, a clinical evaluation is recommended in patients with candida infection.

It is very common to see that in published reports mycology analyses of the skin and adnexae just refer to C. albicans and C. spp. Because mycosis epidemiology is a constantly evolving phenomenon, nail candidiasis caused by species other than albicans might be expected to have a role as a gateway to systemic candidiasis, thus implying an additional concern for clinical and therapeutic aspects.

An oncology institute in Slovakia reported C. krusei, C. parapsilosis, C. glabrata, C. tropicalis and C. guillermondii fungaemia over 10 years, there were 45 non-albicans C. fungaemias as compared to 75 C. albicans fungaemias.[28]

Candida fungaemia behavior has been changing in time. In Slovakia, until 1990, non-albicans Candida accounted for 10-40% of all candidemias, while between 1991-1998 it accounted for 35-65%. The most common species were C. parapsilosis, C. tropicalis and C. glabrata. Remarkably, C. krusei and C. glabrata were resistant to fluconazole and C. lusitaniae and C. rugosa were resistant to amphotericin B.[29]

Non-albicans Candida incidence has proved to be higher than that of C. albicans (54.4% vs. 45.6%). While C. krusei relates to the prior use of antifungals, hematological malignancies, neutropenia, stem-cell transplant and steroid therapy,[30] C. parapsilosis has been associated to endocarditis in intravenous drug abusers and nosocomial infections such as septic arthritis, peritonitis and endophthalmitis and invasive medical instrumentation. Between 1989 and 1992, C. krusei was responsible for 5% of fungaemias in Houston, and between 1993-1996 such figure increased to 10%. It presented especially among patients with leukemia and C. albicans related to catheter use. In neutropenic patients, C. krusei was associated to higher mortality rates (49%) than C. albicans (28%).[31]

The frequency of C. parapsilosis in nosocomial infections related to skin colonization, particularly the subungual space and oral cavity, has increased. In Taiwan, from a group of 323 healthy volunteers, 15.2% had oral cavity colonization by Candida, with C. albicans accounting for 57.7% and C. parapsilosis accounting for 15.4% of cases.[32]

The sensitivity of various Candida species varies in different countries. A decrease in the susceptibility of C. parapsilosis to amphotericin B, 5-fluorocytosine but a good susceptibility to imidazole drugs have been reported.[3334]

A study conducted in Venezuela with inpatients, revealed different Candida species have different susceptibility to the different antimycotic drugs. C. albicans showed sensitivity to fluconazole in 92.6%, C. tropicalis in 87.1% and C. glabrata in 56.7%. Also, all the studied strains were sensitive to amphotericin B and voriconazole.[35] Another study reveals that C. krusei and C. glabrata are poorly sensitive to fluconazole and the resistance increase is suggested to be a consequence of the frequent prophylactic use of fluconazole.[36] The association between non-albicans species has been observed in Japan in fluconazole-related candidemias. While C. tropicalis, C. parapsilosis and C. guillermondii were sensitive to therapy, C. glabrata and C. krusei were less sensitive. Moreover, C. parapsilosis and C. guillermondii had a high minimal inhibitory concentration (MIC) for micafungin.[37]

In Kuwait, the species most commonly isolated from blood cultures was C. albicans (39,5%) as opposed to C. parapsilosis that was isolated from 30,6% of the cases. Interestingly, all the strains were sensitive to voriconazole and resistance to amphotericin B and fluconazole was persistently low, in spite of the routine use of such drugs in systemic candidiasis.[38] Authors suggest the use of voriconazole to treat infections caused by fluconazole and itraconazole – resistant yeast.[39]

With regard to cell wall inhibitors such as echinocandins, C. albicans has also been revealed as the most sensitive species, while C. glabrata, C. tropicalis and C. krusei are less sensitive; C. parapsilosis and C. guillermondii are intrinsically resistant by virtue of specific mutations.[40]

A study published in Spain revealed the best MIC of amphotericin B for C. albicans, higher than the MIC obtained for C. tropicalis, C. glabrata and C. parapsilosis.[41]

In vitro sensitivity of various Candida species to various polyene formulations (amphotericin, nystatin) and itraconazole revealed seldom differences between themselves. However, sensitivity to polyenes is superior than sensivity to fluconazole.[42]

There is evidence that terbinafine has good therapeutic results in C. parapsilosis, but not in other species of Candida; this could lead to the conclusion that terbinafine may be useful in the treatment of nail candidiasis adding a new therapeutic tool.[4344]

To date there are no evidences of a literature reference showing the results presented herein concerning the proportional increase of C. parapsilosis in nail candidiasis. It is worth noting that the works published do not express rates but simple percentages since studies do not involve a defined assigned population. Laboratories receive specimens from all centers without considering specific populations. Moreover, they refer to the consulting population, thus inducing a bias.

In the future, we should assess the prevalence of onychomycosis in the general population to really appreciate the magnitude of this public health issue.

What is new?

The present study shows that both as well as appear as emerging pathogens that would be in fact taking the place of as the most commonly isolated pathogen in patients with Candida onychomycosis.

Identification of Candida strains as etiological agents of nail candidiasis becomes relevant to the management both nail as well as systemic candidiasis, in view of the resistance to conventional treatments readily reported in the literature.