Patients and setting

Participants were recruited at the Muhimbili National Hospital (MNH) HIV clinic in Dar es Salaam, Tanzania from January 2006 to September 2007. Patients were eligible for the study if they were HIV-positive, 18 years of age and above with clinical picture of OPC, characterized by creamy, white, curd like patches, or by typical erythematous lesions on the oral or/and pharyngeal mucosa. Patients were included after obtaining their written informed consent. Patients were interviewed using a standard structured questionnaire and their hospital records were also used to determine the previous episodes of OPC, use of antifungal agents and past medical conditions. A standard oral examination method recommended by WHO was used [25]. Blood was collected in EDTA tubes from each patient for enumeration of CD4+ and CD8+ T cells using a fluorescent activated cell sorter (FACS) count machine after staining patients' blood with monoclonal antibodies [26]. Primary OPC was defined as the one that occur for the first time and recurrent OPC as the second episode and above during the course of HIV disease.

Oral isolates and species identification

Specimens were obtained by firmly swabbing the lesion site with sterile cotton wool swab [27]. The swabs were cultured on sabouraud dextrose agar (SDA) (Oxoid Ltd, Hampshire, England) supplemented with 0.02% chloramphenicol and incubated aerobically at 30°C for 5 days. Isolates were identified to species level by colonial morphology on SDA and microscopic morphology. The germ tube production test was done. To distinguish between C. albicans and C. dubliniensis; all germ tube positive isolates were subcultured on SDA and incubated at 45°C aerobically and an isolate which did not grow at 45°C, was identified using AUXACOLOR 2 (Bio-Rad, France). Germ tube negative yeasts were identified using AUXACOLOR 2 (Bio-Rad, France) and the auxanographic method [28]. In the presence of multiple colony morphologies, all were identified to species level. After the final identification, isolates were stored at -80°C in 50% glycerol until susceptibility tests were performed.

In vitro antifungal susceptibility testing

The in vitro activities of fluconazole (Pfizer, Kent, UK), itraconazole, miconazole (Janssen-Pharmaceutica N.V., Beerse, Belgium), clotrimazole (Sigma Aldrich, Steinheim, Germany), amphotericin B (Bristol-Myers Squibb, Woerden, The Netherlands) and nystatin (Sigma Aldrich, Steinheim, Germany) were assessed. The minimum inhibitory concentration (MIC) was determined using a broth microdilution format (M27-A2) according to the guidelines of the Clinical and Laboratory Standard Institute (CLSI) [29], and all the testing done in duplicates. The respective manufacturers provided the antifungal agents as standard powders. Fluconazole was dissolved in sterile distilled water and itraconazole, miconazole, clotrimazole, amphotericin B and nystatin were dissolved in dimethyl sulfoxide (DMSO) to make stock solutions. RPMI 1640 medium with L-glutamine, without sodium bicarbonate (GIBCO BRL, Life Technologies, Woerden, The Netherlands) and buffered with morpholinepropanesulfonic acid (MOPS) at 0.165 M (Sigma-Aldrich Chemie GmbH, Steinheim, Germany) was used as a test medium. Serial twofold dilutions and final dilutions of each antifungal agent were prepared in RPMI 1640 medium [29]. The range of fluconazole concentrations tested was 0.063–64 μg/ml and for itraconazole, miconazole, clotrimazole, amphotericin B and nystatin the range was 0.016–16 μg/ml. Aliquots of 100 μl of each antifungal agent at a concentration two times the targeted final concentration were dispensed in the wells of flat bottom 96-well microtiter plates (Costar, Corning, N.Y). Prior to testing, isolates were subcultured on SDA at 30°C under aerobic conditions for 24 hours. Yeast inocula were prepared spectrophotometrically and further diluted in RPMI 1640 medium in order to obtain a final concentration of 0.5 to 2.5 × 10³ CFU/ml [29]. A constant volume (100 μl) of the innoculum was added to each microdilution well containing 100 μl of the serial dilution of the antifungal agents to reach final concentrations. The microtiter plates were incubated at 35°C for 48 hours.

The minimum inhibitory concentrations (MICs) were determined after 48 hours, after which readings were performed spectrophotometrically with a microplate reader (Anthos ht III; Anthos Labtec Instruments, Salzburg, Austria) at 405 nm using modified broth microdilution format (M27-A2) as described previously [21, 30]. The optical densities (ODs) of the blanks were subtracted from the ODs of the inoculated plates and the percentages of growth for each well were calculated. The MICs of polyene antifungals (amphotericin B and nystatin) were defined as the lowest concentrations that inhibit growth by 100%. The MICs of the azoles (fluconazole, itraconazole, clotrimazole and miconazole) were defined as the lowest concentrations that inhibit growth by 50%. Antifungal activity was expressed as the MIC of each isolate to the antifungal agent. The MIC values were the average of MICs of the first and duplicate measurement. The MIC values for fluconazole and itraconazole were compared to the CLSI interpretative guidelines on antifungal susceptibility testing. MICs of ≤ 8 μg/ml considered as susceptible, 16 – 32 μg/ml as susceptible dose dependent (SSD) and ≥ 64 μg/ml as resistant for fluconazole and for itraconazole, ≤ 0.125 μg/ml as susceptible, 0.25 – 0.5 μg/ml as SDD and ≥ 1 μg/ml as resistant [29]. American Type Culture Collection (ATCC) strains recommended by CLSI, C. parapsilosis (ATCC 22019) and C. krusei (ATCC 6258) were used as controls for each test.

Ethical issues

The Ethics Committees of the Muhimbili University of Health and Allied Sciences (MUHAS) and the Muhimbili National Hospital, Dar es salaam, Tanzania approved the study. All patient information and test results were confidentially kept.

Data and statistical analysis

The geometric mean of MICs, MIC₅₀ and MIC₉₀ were calculated. The MIC₅₀ and MIC₉₀ values were calculated as the concentrations of antifungals that were able to inhibit 50% and 90% of the isolates, respectively. The high and low off-scale MICs were included in the analysis by conversion to the next higher and lower drug concentrations, respectively. An isolate was considered resistant to fluconazole and itraconazole if average MICs were greater or equal to their respective CLSI breakpoints [29]. Data were analyzed using the SPSS version 14.0. Antifungal susceptibility of isolates from patients with primary OPC was compared with those of isolates from patients with recurrent OPC. Similarly, the antifungal susceptibility of isolates from patients who had not received an antifungal agent was compared with those of isolates from patients who had previously received antifungal therapy. Comparison of species distribution and antifungal susceptibility rates was performed using Fisher's exact test and Mann-Whitney U-test. A P-value of < 0.05 was considered significant.

Patients' characteristics

Two hundred and ninety two (292) HIV infected patients with OPC, 164 with primary OPC and 128 with recurrent OPC, participated in the study. Pseudomembranous candidiasis was the most predominant type of OPC 194 (66.4%) followed by combinations of pseudomembranous and erythematous 66 (22.6%), pseudomembranous and angular cheilitis 12 (4.1%), pseudomembranous and hyperplastic candidiasis 9 (3.1%), hyperplastic candidiasis 7 (2.4%) and erythematous candidiasis were the least 4 (1.4%). Patients' ages ranged from 18 to 75 years, median 34 years of whom 218 (74.7%) were females and 74 (25.3%) were males. The majority of the patients with recurrent OPC had a history of one previous episode of OPC (n = 92, 71.8%) and the range of recurrences was 1 to 4 episodes. Their CD4+ T cells values ranged from 1 to 699 cells/mm³ with a median count of 93 cells/mm³ and 229 patients (78.4%) had CD4+ T cells count below 200 cells/mm³. One hundred and seven patients (36.6%) were under HAART, predominantly (72.9%) a combination of stavudine, lamivudine and nevirapine and 65 patients (60.7%) were on HAART for duration below six months.

Distribution of isolates

In vitrosusceptibility of isolates

Correlation between MIC and recurrence of OPC

There were significant differences in the MIC distributions for isolates from patients with primary and recurrent OPC for amphotericin B (P = 0.041), itraconazole (P = 0.008), miconazole (P = 0.045), and clotrimazole (P = 0.013). No significant difference in MIC distribution between isolates was observed for nystatin (P = 0.608) and fluconazole (P = 0.133). Based on CLSI interpretative breakpoints, fluconazole resistance for species isolated from patients with primary OPC and recurrent OPC was noted in 9 isolates (5.5%) and 6 isolates (4.6%), respectively, which was not different (P = 0.581). However for itraconazole, more isolates cultured from patients with recurrent OPC exhibited reduced susceptibility compared with those cultured from patients with primary OPC (P = 0.03).

However, when the correlation between in vitro susceptibility and previous exposure to antifungal agents was determined, it was found that all azole compounds tested were less active against Candida isolates cultured from patients who had been previously exposed to any antifungal drug compared with those cultured from the treatment naïve group (fluconazole, P = 0.037; itraconazole, P = 0.002; miconazole, P = 0.015; clotrimazole, P = 0.007). No significant differences in MICs of amphotericin B and nystatin between the two groups of isolates were noted (P = 0.087 and P = 0.117), respectively.

In the present study, two of the five C. albicans isolates that were SDD to fluconazole were also SDD to itraconazole and three isolates were resistant to itraconazole. Itraconazole resistance was in 25 (8.4%) of all tested isolates, among which 17 isolates (68%) had MICs of ≥ 1 μg/ml for clotrimazole.