Clinico-mycological analysis and antifungal resistance pattern in human immunodeficiency virus-associated candidiasis: An Indian perspective.

OBJECTIVES: Candidiasis is a common human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome-associated opportunistic mycoses. The present study ascertained the species spectrum of Candida strains recovered from different clinical samples from symptomatic HIV-positive individuals and determined the antifungal susceptibility profile of the isolates.
MATERIALS AND METHODS: A variety of specimens were collected from 234 symptomatic HIV seropositive individuals depending on their clinical manifestations and subjected to direct microscopic examination. Blood samples were inoculated in biphasic blood culture medium and all other specimens on Sabouraud dextrose agar with chloramphenicol and incubated at 35°C-37°C. Species identification of the recovered Candida isolates was attempted on the basis of germ tube production, micromorphology on corn meal agar, color and morphology on HiCrome Candida Differential agar, and carbohydrate fermentation and assimilation tests. Susceptibility testing of the isolates was performed employing the VITEK 2 system.
RESULTS: A total of 167 Candida isolates were obtained; Candida albicans (136), Candida tropicalis (13), Candida krusei (8), Candida parapsilosis (5), Candida glabrata (4), and Candida kefyr (1). Fluconazole resistance was more frequent among nonalbicans species, and significantly higher 5-fluorocytosine resistance compared to C. albicans was also observed. Eight Candida strains (six C. krusei, one C. kefyr, and one C. albicans) were multidrug resistant.
CONCLUSION: Although C. albicans continues to be the leading etiological agent of candidiasis, the incidence of nonalbicans species among HIV-positive Indian individuals is rising. Antifungal resistance was higher among nonalbicans Candida species. Another issue of therapeutic concern is the possible emergence of multidrug-resistant Candida strains among these patients.

INTRODUCTION

Acquired immunodeficiency syndrome (AIDS) caused by human immunodeficiency virus (HIV) is a major public health challenge of the present century. The virus selectively destroys the CD4 population of T cells, the declining numbers of which herald the onset of a spectrum of opportunistic infections (OIs).[1] These OIs are a major cause of mortality and morbidity among the HIV infected.[2] Tuberculosis is the most common OI seen in HIV/AIDS patients in India, followed by candidiasis.[3]

Candida species are the most common cause of fungal infections, with the spectrum of manifestations ranging from asymptomatic colonization to severe clinical forms. They can cause a panorama of clinical infections including the protean manifestations of oropharyngeal candidiasis, esophageal candidiasis, vulvovaginal disease, cutaneous candidiasis, and systemic or invasive forms such as candidemia.[45] A state of cell-mediated immunodeficiency and T-cell dysfunction as seen in HIV infection is pivotal in creating a state of host susceptibility to Candida infections.[6] HIV/AIDS patients show increased rates of colonization by Candida species compared to HIV-negative individuals.[78] Oropharyngeal colonization by Candida species is seen in nearly 90% of HIV/AIDS patients at some point during the course of HIV disease, and a CD4 count <200 cells/μl is associated with a significantly higher frequency of Candida species colonization as well as subsequent development of oropharyngeal candidiasis.[891011] This implies that a timely diagnosis of this condition, that is often the sentinel event indicative of progression to AIDS, and the mere initiation of highly active antiretroviral therapy in these patients could obviate the need for specific antifungal treatment.[12]

A gradual change in the distribution of Candida species causating candidiasis has been noted in recent years, with several nonalbicans Candida species being implicated with a greater frequency.[1314] Furthermore, the common practice of treating mucosal and deep-seated forms of candidiasis with fluconazole, as well as the extensive chemoprophylactic use of azole antifungals in HIV-infected patients, has led to an increase in the number of Candida isolates that are azole-resistant.[15] There have been previous published reports of azole-resistant isolates of Candida albicans being recovered with increasing frequency in HIV/AIDS patients.[16] In addition, nonalbicans Candida species such as Candida krusei and Candida glabrata that exhibit intrinsic reduced susceptibility to the commonly used azole antifungals are isolated in high frequency among HIV-positive patients.[1718]

Mycological cultures and antifungal susceptibility testing of Candida species are not routinely performed in Indian laboratories. Further, there is a paucity of literature on the yield and diversity of Candida species recovered from sites other than oropharynx in HIV-positive individuals. The present study was, therefore, designed to define the epidemiology of Candida infection in HIV/AIDS patients including its correlation with CD4 counts; to identify and characterize the species of Candida isolates and ascertain their distribution in various specimen types; and to determine the in vitro susceptibility profile of the prevalent isolates to the four common antifungal agents-fluconazole, voriconazole, amphotericin-B, and 5-fluorocytosine.

MATERIALS AND METHODS

Study setting and design

The present descriptive, cross-sectional study was conducted prospectively over a period of 3 years from October 2008 to September 2011, at the Department of Microbiology of a tertiary care health center in New Delhi, India. A total of 234 symptomatic HIV seropositive patients with varied clinical manifestations and presenting to the antiretroviral center or admitted in the ward were recruited for the study.

Enrollment and evaluation of patients

Before enrollment and specimen collection, a written informed consent was obtained from each participant which was duly explained to him in the Hindi language. HIV serostatus of the participants was confirmed employing the algorithm as described under strategy III of National AIDS Control Organization guidelines.[19] Specific, structured, and predesigned questionnaires were used to gather information regarding sociodemographic profile, risk factors, and clinical history of the patients. A detailed clinical examination was performed by the physicians, and routine laboratory investigations including chest radiographs were obtained. CD4 counts of all enrolled HIV-positive patients were measured using the fluorescence-activated cell sorting count system, Becton Dickinson, Singapore, BD. The institutional ethics committee approved the protocol for the study.

Specimen collection

Specimens were collected depending on the clinical manifestations of the patients. These included oral swabs, sputum, blood, cerebrospinal fluid, stool, urine (midstream), vaginal swabs and skin scrapings. Standard precautions and strict asepsis were followed throughout sample collection, and the specimens were immediately transported to the laboratory for further processing.

Isolation and identification of Candida species

An aliquot of the specimens was transferred to a clean, grease-free microscope slide with a drop of 10% potassium hydroxide, and a coverslip was placed over it. The preparation was slightly warmed and examined under ×40 objective of the microscope for fungal elements. In addition, smears were Gram-stained and examined under oil immersion objective of the microscope for the presence of Gram-positive budding yeast cells and pseudohyphae.

Blood sample for fungal culture as collected by venipuncture was inoculated in biphasic blood culture medium (brain heart infusion agar and broth) and incubated at 37°C. All other specimens were inoculated on sabouraud dextrose agar (SDA) with chloramphenicol and incubated at 35°C–37°C for 48 h. Culture tubes were examined after 48 h of incubation and those with no growth were further incubated for a week before they were reported as sterile. Any pasty, opaque, and cream-colored colonies on SDA were identified by Gram stain. Once the Gram-stain confirmed the identification of the isolate as Gram-positive budding yeast cells, speciation of Candida isolates was done by their ability to produce germ tubes, micromorphology on corn meal agar with Tween 80 (Hi-Media, India), color and morphology on HiCrome Candida Differential agar (Hi-Media, India); and biochemical profile including carbohydrate fermentation and assimilation tests using yeast nitrogen base agar (Difco, Becton Dickinson, India). All tests were performed as per standard recommended procedures.[2021] The following strains were used as controls: C. albicans ATCC90028, Candida parapsilosis ATCC22019, C. krusei ATCC6258, C. glabrata ATCC90030, and Candida tropicalis ATCC750.

In vitro antifungal susceptibility testing

Antifungal susceptibility testing of isolates was performed employing the VITEK 2 system. For each strain, a standardized suspension was prepared in sterile saline and its turbidity adjusted to a 2.0 McFarland standard. Each inoculum was placed in a VITEK 2 cassette along with a sterile polystyrene test tube and an antifungal susceptibility test card. The cassettes were then loaded into the VITEK 2 instrument, and dilution of yeast suspensions, as well as filling, incubation, and reading of cards was performed automatically by the instrument. With each run of cards, appropriate quality control strains were also included in this study.

Data analysis

All the data were entered in a Microsoft Excel sheet and analyzed at a 95% confidence interval using the Epi-info software, version 3.5.3, CDC, Atlanta, GA, USA. Quantitative data were summarized as arithmetic mean and standard deviation for central tendencies and median for nonnormal/skewed distributions. Categorical variables were expressed in proportions and percentages. Chi-square and Fisher's exact test were applied to analyze the differences between proportions. A two-tailed P value < 0.05 was considered as statistically significant.

RESULTS

A summary of the sociodemographic and immunological profile of study individuals is depicted in Table 1. The various clinical manifestations seen in these patients were as follows: weight loss (183; 78.2%), oral ulcers (175; 74.8%), fever (157; 67.1%), anorexia (103; 44%), cough (93; 39.7%), diarrhea (66; 28.2%), dyspnea (63; 26.9%), neck rigidity (62; 26.5%), night sweats (37; 15.8%), dysphagia (36; 15.4%), generalized lymphadenopathy (24; 10.3%), burning micturition (20; 8.5%), vaginal discharge (3; 1.3%), and cutaneous manifestations (2; 0.9%). A history of past tubercular disease or evidence of coexisting active tuberculosis was discernible in 116 (49.6%) of 234 enrolled individuals.

Table 1

Demographic characteristics and CD4 profile of study population (n=234)

A total of 395 clinical samples comprising of 232 oral swabs, 71 sputa, 32 blood cultures, 21 cerebrospinal fluids, 21 stool samples, 13 urine samples, three vaginal swabs and two skin scrapings were processed in the present study. Candida species were isolated in 137 (58.5%) patients, and a total of 167 isolates of Candida species were obtained. The specimen-wise yield of Candida species was as follows: isolation rate was highest with oral swabs (55.2%; 128/232), followed by sputum (38%; 27/71), vaginal discharge (33.3%; 1/3), urine (23.1%; 3/13), blood (15.6%; 5/32), stool (9.5%; 2 of 21) and cerebrospinal fluid (4.8%; 1 of 21). None of the two skin scrapings yielded any Candida growth.

Six different Candida species were identified, of which C. albicans (136/167; 81.4%) was the most prevalent species isolated, while nonalbicans Candida species (31) constituted 18.6% of the isolates. The nonalbicans Candida species recovered included C. tropicalis (13), C. krusei (8), C. parapsilosis (5), C. glabrata (4), and Candida kefyr (1). The distribution of Candida species in different specimen types is shown in Table 2.

Table 2

Distribution of various Candida species (n=167) in different specimen types

The majority (68 of 137; 49.6%) of HIV-positive patients with a positive culture for Candida species had CD4 counts between 201 and 500 cells/μl; 59 (43.1%) had CD4 counts below 200 cells/μl and 10 (7.3%) had CD4 counts more than 500 cells/μl. Further, the differential prevalence of Candida species isolation did not vary significantly across different CD4 categories (P = 0.62).

A graphical representation comparing susceptibility profile of C. albicans and nonalbicans Candida strains to the four antifungals is provided in Figure 1. It was observed that fluconazole resistance was significantly higher among nonalbicans Candida species versus C. albicans (29% vs. 0.7%; P < 0.0001). Likewise, significant differences were also observed between nonalbicans Candida species and C. albicans with regard to resistance to 5-fluorocytosine (6.5% vs. 0%; P = 0.03). The species-wise susceptibility profile of Candida strains to all the four antifungals is summarized in Table 3.

Figure 1

Susceptibility of albicans versus nonalbicans Candida species to various antifungal agents

Table 3

Antifungal susceptibility profile of various Candida species (n=167)

Considering an intermediate susceptibility and susceptible dose dependence to a given antifungal agent also as resistance, 4 (2.4%), 6 (3.6%), and 2 (1.2%) Candida strains were found to be resistant to one, two, and three of the antifungal agents, respectively. Eight strains of Candida (six C. krusei, one C. kefyr, and one C. albicans) were found to be multidrug resistant, that is, resistant to two or more antifungals.

DISCUSSION

Prompt and rapid diagnosis of candidiasis is essential. Past few years have witnessed a rising number of reports of Candida infections from the Indian subcontinent. High isolation rates of Candida species have been reported in HIV-infected individuals with oral ulceration, which is in agreement with our study.[2223] In addition, similar to previous reports, C. albicans was the predominant Candida species isolated from oral cavity of HIV-positive patients.[2425] However, recent literature has highlighted a shift in the distribution profile of Candida species and the emerging role of nonalbicans Candida species in the causation of oral candidiasis. Kwamin et al. have documented nearly 30.5% of Candida isolates recovered from HIV-positive patients with oropharyngeal candidiasis to be nonalbicans Candida species.[23] We also found 15.6% (20/128) of Candida isolates from oral cavities of HIV-positive individuals to be nonalbicans species. With regard to species distribution, Maninder and Usha, in their study on oropharyngeal candidiasis in HIV-infected patients, found C. tropicalis and C. krusei to be the commonest nonalbicans Candida species isolated.[26] We also observed C. tropicalis to be the leading nonalbicans Candida species isolated from oral swabs of HIV-infected individuals. Furthermore, several studies have attempted to analyze the association between candidiasis and CD4 counts. Maurya et al. reported a strong correlation between Candida colonization/candidiasis of the oral cavity and CD4 counts <200 cells/μl.[8] On the other hand, similar to our report, Costa et al. reveal contradictory findings with no significant correlation noted between CD4 cell counts and Candida isolation rates.[27]

In our study, recovery rate of Candida was highest with oral swabs, followed by sputum and vaginal swabs. The yield was lowest with cerebrospinal fluid samples and skin scrapings. While the highest yield of Candida species seen with oral swabs is consistent with other studies, blood cultures have generally yielded the lowest number of Candida isolates in HIV-positive patients.[1128] A comparative summary of specimen-wise Candida species prevalence profile (other than those from oral swabs) in our study versus similar reports from India and the Western world is summarized in Table 4.

Table 4

Studies undertaken on human immunodeficiency virus-associated candidiasis in recent years

Fluconazole is the most widely employed drug for the management of candidiasis. We report 94% of Candida strains isolated in our study to be susceptible and 6% to exhibit in vitro resistance to fluconazole. Similar figures were reported by Lattif et al. in their study on Indian HIV/AIDS patients with oropharyngeal lesions.[22] A plausible hypothesis for the low fluconazole resistance rates as documented in our study is that it was only a single time analysis. Perhaps, repeated collection and susceptibility testing of isolates from the same patient at different stages of HIV disease and at different time intervals would provide a better insight into the evolution and emergence of azole-resistant Candida strains in HIV seropositive individuals. Moreover, the extensive prophylactic and therapeutic use of fluconazole in developed countries versus the developing nations is additionally responsible for the lower incidence of fluconazole refractile Candida strains in the latter. Reports of fluconazole-resistant C. albicans from HIV seropositive individuals are common.[1638] However, we encountered only one fluconazole-resistant C. albicans strain in our study, a finding similar to that of Kumar et al., who observed none of the C. albicans strains isolated in their study to be resistant to fluconazole.[39] On the contrary, another study examining oral Candida isolates from HIV-reactive populations of South Africa and Cameroon has documented nearly half of C. albicans isolates collected by the researchers to be resistant to all the azole antifungals tested.[40] Among nonalbicans Candida species, fluconazole resistance was highest among C. krusei, a finding well-supported by the previous published studies.[17] In addition, while cross-resistance to voriconazole was documented in the single fluconazole resistant C. albicans strain recovered in our study, all the eight strains of C. krusei and the single strain of C. kefyr that we isolated exhibited 100% susceptibility to voriconazole while being fluconazole resistant. This is in contrast to a previous published paper that reports a high level of cross-resistance between azole antifungals and recommends avoiding other azole agents for the treatment of fluconazole-resistant Candida strains.[41]

Despite the use of polyene antifungals for nearly 50 years, amphotericin B resistance among Candida strains is infrequently encountered. Likewise, none of the of C. albicans strains in our study was amphotericin B resistant, and among the nonalbicans, Candida species isolated, only two strains of C. krusei exhibited a susceptible dose-dependent response to this antifungal. While several researchers have reported high amphotericin B susceptibility among Candida isolates, there are reports of emerging amphotericin B resistance published by others.[16384243]

Furthermore, all the strains of C. albicans in our study were susceptible to 5-fluorocytosine while 75% strains of C. krusei and the single strain of C. kefyr isolated showed intermediate susceptibility and 50% strains of C. glabrata were 5-fluorocytosine resistant. In a study undertaken among late presenting AIDS patients in Ethiopia, one of the 25 strains of C. albicans, one of the 25 strains of C. glabrata, and one of the 25 strains of C. tropicalis were found to be 5-fluorocytosine resistant.[43]

Mulu et al. have reported multidrug resistance in eight of the 90 Candida isolates tested for antifungal susceptibility in their study.[43] Likewise, the authors of the present study also report multidrug resistance, with majority of the multidrug-resistant strains being nonalbicans Candida species. Emergence of multidrug-resistant Candida strains could be a matter of grave therapeutic concern in India, more so in the immunocompromised populations.

The present study was undertaken at one of the largest tertiary care centers in North India that caters to a large and diverse population base. The study, however, is fraught with certain limitations. It was not a case–control analysis and thus, we could not compare the isolation rates, species diversity, and antifungal susceptibility profile of Candida strains recovered from symptomatic HIV seropositive individuals with those from healthy HIV-negative controls. Second, the sample size of the study population was small, and the study was essentially a single-center analysis. Another shortcoming was that a history or documentation of prior fluconazole exposure was not a part of our study protocol and thus, we could not describe the azole resistance in Candida strains as being intrinsic or acquired. Furthermore, the lack of a longitudinal follow-up study design precluded an analysis of species distribution profile and evolution of antifungal resistance among Candida strains with reference to correlation with HIV disease progression.

CONCLUSION

Our study has showcased the present etiological trends of candidiasis in HIV-positive individuals in an Indian setting. It points toward a gradually evolving trend of Candida infections whereby, C. albicans continues to remain the predominant Candida species isolated from HIV-positive individuals, but with the incidence of nonalbicans species on the rise. Our study also highlights the current resistance patterns of various Candida species and points toward a possible emergence of multidrug-resistant Candida strains among HIV-infected populations of India. This could pose an imminent therapeutic challenge in the future.

Financial support and sponsorship

Indian Council of Medical Research.

Conflicts of interest

There are no conflicts of interest.