Abstract
Objective. To determine the prevalence of oral candidiasis and multiple oral Candida infections in patients with primary Sjögren’s syndrome (SS), and the clinical and drug susceptibility profile.
Methods. Thirty patients with primary SS were enrolled in our study. The diagnosis of oral candidiasis was based on the clinical manifestation, and confirmed by a concentrated rinse culture. Candida spp. assessment was accomplished using standard methods: Sabouraud dextrose agar with 50 mg/l chloramphenicol and CHROMagar were used for the rapid screening of clinical species, followed by the API 20C system for further species identification. In vitro antifungal drug susceptibility of Candida isolates was determined by the minimal inhibitory concentrations.
Results. In our study, 87% (26/30) of subjects had oral candidiasis, in which 42% (11/26) had multiple Candida spp. infection. Although C. albicans remains the predominant isolate, other rare species such as C. tropicalis, C. glabrata, C. parapsilosis, and C. krusei were present, alone or in combination. Chronic atrophic candidiasis is the most common clinical type of oral candidiasis in patients with SS. The susceptibilities of the 44 Candida isolates to 7 antifungal agents varied dramatically. The resistance to azoles was remarkable, and the phenomenon of cross-resistance between itraconazole and fluconazole was observed.
Conclusion. Patients with primary SS carry a high risk of oral candidiasis and a high frequency of multiple Candida infections. The azole resistance patterns of Candida spp. support the necessity for drug susceptibility testing as a routine procedure for patients with oral Candida infections.
Sjögren’s syndrome (SS) is a systemic autoimmune disease characterized by dysfunction and destruction of the exocrine glands associated with lymphocytic infiltrates, and immunological hyperreactivity. Salivary and lacrimal glands are the most commonly affected glands, resulting in the typical clinical presentation of dryness of mouth (hyposalivation) and eyes (keratoconjunctivitis sicca)1,2. A study by Atkinson, et al found that 88% of the SS subjects had a reduced salivary flow rate (SFR)3. And xerostomia, the subjective complaint of dry mouth, has also been reported in high percentages (75% to 92%) of patients with SS4.
Saliva has antibacterial and fungicidal properties, and plays a critical role in maintaining oral health. A continuous flow of saliva is important to prevent oral colonization by Candida. Several of the constituents of saliva are thought to inhibit fungal growth, including defensins, lysozymes, peroxidase, lactoferrin, secretory IgA, and histatin5,6. As a consequence of hyposalivation, patients with SS are prone to develop oral candidiasis. Various studies have reported a high prevalence of oral Candida carriage in patients with SS, ranging from 54.2% to 81.25%7,8,9,10.
Although it is generally accepted that patients with SS have an increased risk for oral candidiasis, little is known about multiple Candida infection and Candida spp. distribution. In addition, patients with SS are likely to get recurrent oral Candida infections, and frequent antifungal treatments are given for symptom management. As a result, development of antifungal drug resistance is a growing problem, as with antibiotics.
The aim of our study was to assess the prevalence of oral candidiasis and multiple oral Candida infection in patients with primary SS, as well as the clinical and drug susceptibility profile. This knowledge would potentially lead to an improved strategy for managing oral candidiasis in patients with SS.
MATERIALS AND METHODS
Population
The subject population consisted of 30 patients with primary SS, who were randomly selected by computer-generated numbers from all the patients with SS who had followup visits at the Oral Medicine Clinic at the Stomatological Hospital of Peking University, China. All 30 patients were diagnosed using the revised European classification criteria11. Selection criteria included subjects who had not received corticosteroid, antibiotic, or antimycotic therapy over the previous 4 weeks. Thirty sex-matched and age-matched patients with oral candidiasis but without SS or other systemic diseases were studied in parallel for oral Candida distribution.
The Ethics Committee of Peking University Health Science Center approved the study.
Oral examinations
The diagnosis of oral candidiasis was based on the clinical presentation (such as loss of papilla of the dorsal tongue, erythema and fissuring of the tongue, erythema of other mucosal surfaces, or angular cheilitis), and then verified by positive potassium hydroxide slide and/or positive culture. Whole unstimulated saliva was collected in a sterilized plastic tube for 5 min. The SFR was recorded as ml/10 min. The oral hygiene, caries, and presence/absence of dentures were also recorded. All oral examinations were performed by the same examiner.
Mycological evaluation
Saliva samples were collected and cultured using an oral rinse technique12, in which patients swished 10 ml of sterile saline solution without preservative for 30 s, and then spat into a sterile container. Cultures were performed on Sabouraud agar and CHROMagar chromogenic medium (CHROMagar Microbiology, Paris, France). The total colony-forming units per milliliter (CFU/ml) were counted, and colony color was recorded. The Candida spp. identification was based on a combination of morphological and biochemical criteria: representative colonies of each morphologic type and color were isolated, routinely Gram-stained, then identified by API 20C kits (bioMérieux, Craponne, France), complemented with germ tube formation, growth at 37°C and 45°C. The results were subsequently interpreted according to the manufacturer’s instructions.
Susceptibility testing
Seven antifungal agents (fluconazole, itraconazole, amphotericin B, flucytosine, ketoconazole, miconazole, and terbinafine) were obtained from their manufacturers. Stock solutions of each antifungal agent were prepared and stored at −20°C or lower.
The National Committee for Clinical Laboratory Standards-approved method (NCCLS M27-A)13 for fungal drug susceptibility was employed, with a broth microdilution using RPMI-1640 medium buffered to pH 7.0 with 3-(N-morpholino) propanesulfonic acid (MOPS), and an inoculum of 0.5–2.4 × 104 cells/ml. Specifically, 5 colonies of each isolate were selected and placed in the medium with MOPS, and diluted to the desired concentration using spectrophotometric techniques. Serial dilutions of drugs were made from 0.03 to 128 μg/ml, to which the yeast cell inoculum was added and incubated at 35°C for 48 h. Two quality-control standard strains, C. albicans ATCC 90028 and C. krusei ATCC 6258, were tested in parallel.
RESULTS
Subjects
The ages of the 30 patients with primary SS ranged from 26 to 75 years (mean 48.6 yrs). All were women, all had a history of oral candidiasis, and all had received prior antifungal treatment. At the time of sample collection, they had not received antibiotic, corticosteroid, or antimycotic drugs for at least 4 weeks.
Prevalence of oral candidiasis
Of the 30 patients, 26 (87%) were diagnosed with oral candidiasis, based on the clinical presentation and a positive oral rinse culture. There was a significant inverse relationship between unstimulated salivary flow and candidal infection. In subjects (4/30) with an unstimulated SFR ≥ 1 ml/min, no clinical signs of oral candidiasis were detected, and subsequent oral rinse culture ruled out their infections. Conversely, all patients (26/30) with low SFR < 1 ml/min had detectable oral candidiasis.
Clinical profile
Oral manifestations (Table 1) included erythematous candidiasis of the tongue (17 patients), angular cheilitis (11 patients), denture-related candidiasis (6 patients), and median rhomboid glossitis (3 patients). Although median rhomboid glossitis is a type of erythematous candidiasis, it occurs uniquely in the posterior midline of the dorsum and was separated here from erythematous candidiasis.
Candida spp. distribution
A total of 44 Candida isolates were identified from 26 patients. Among the isolates, C. albicans was most common, accounting for 55% (24/44) of all isolated species. The next most frequent isolate was C. tropicalis (7 strains), followed by C. glabrata (5 strains), C. gulliermodii (4 strains), C. krusei (2 strains), and C. parapsilosis (2 strains).
There is a high diversity of Candida spp. distribution in this group of the population. In our study, 58% (15/26) of the cases were infected by single species. Co-infection with other Candida spp. was found in the rest of the cases (11/26). In the 15 single-species infections, 14 were by C. albicans and 1 by C. gulliermodii. The rest of the 11 cases were infected by 2 or 3 Candida spp. In 2 cases of multiple infections, C. tropicalis and C. glabrata were each “dominant” in 1 case, although C. albicans was “dominant” in unit counts in most multiple species infections. Table 2 shows the distribution and diversity of Candida spp. identified from those with single or multiple infections.
In the control group, composed of patients with oral candidiasis but without systemic disease, 83% (25/30) were mono-infected by C. albicans, and the remaining 5 cases were co-infected by 2 species (Table 3).
Antifungal susceptibility testing
Although almost all the isolates (98%) were sensitive to flucytosine, this medication is rarely used in our clinic because of its unfavorable adverse effect profile and rapid development of resistance. For the triazoles, fluconazole and itraconazole, which are first-line medications in the management of oral candidiasis, susceptibility was 48% and 57%, respectively (Table 4). Cross-resistance between itraconazole and fluconazole was rather high, at 67% of all Candida isolates. This phenomenon was found in all isolated Candida spp. (Table 5).
The minimal inhibitory concentration range for the quality control strains C. albicans ATCC 90028 and C. krusei ATCC 6258 were within the recommended range at every test occasion.
DISCUSSION
Oral candidiasis is recognized as one of the common, painful, and often chronic complications of SS. It is important that practitioners have a thorough knowledge about the clinical and mycological profile of these infections.
In our study, most of the patients had severe dry mouth symptoms and low SFR. Patients with low SFR were at a higher risk for developing oral candidiasis. Interestingly, of the 4 subjects in our study with flow rates ≥ 1 ml/min, none had clinical signs of oral candidiasis or infections detectable by Candida culture. This finding is consistent with other studies that reported an inverse relationship between salivary flow and Candida carriage7, which highlights the critical role that saliva plays as a protective mechanism.
It was reported that oral Candida carriage in patients with primary SS, secondary SS, and xerostomia was 81%, 67%, and 71%, respectively7,8,9,10. The high prevalence of oral candidiasis in patients with primary SS in our study may be due to the detection method and the population studied. An oral rinse technique (> 300 CFU/ml confirmed the oral candidiasis) was applied in our study, and it is reported to provide a better Candida detection sensitivity compared to the traditional culture method, especially for subjects with a low SFR.
As with previous observations7,8, C. albicans was the predominant species, and was isolated from 25/26 patients diagnosed with oral candidiasis. However, only 56% of these patients (14/25) had a mono-C. albicans infection. In the remaining cases, multiinfections with either 1 or 2 Candida spp. were identified (Table 2).
For patients with SS who had candidiasis, our findings differed from previous observations — a remarkably high proportion (11 out of 26) had multispecies oral Candida infections. Although oral candidiasis has been reported with great frequency, reports of multiple Candida spp. isolation and distribution in patients with SS are rare7,8,9,10. This might be due to conventional isolation and methods used to identify infections, which detect mostly single causative species instead of multiple infections. There are some methods, such as polymerase chain reaction methods, or those that use a differential isolation medium containing chromogenic substrate14, that are able to detect multiple Candida spp. We used CHROMagar Candida chromogenic medium because of its ease of use. We found that in Beijing there was a rather high frequency of mixed colonization, with 2 or 3 Candida spp. simultaneously. On the other hand, non-C. albicans spp. (C. tropicalis, C. glabrata, C. parapsilosis, C. krusei, and C. gulliermodii) accounted for slightly under half (20/44) of the isolates.
This phenomenon is clinically relevant, since susceptibility to antifungal medication differs significantly among Candida spp. For example, C. krusei has a natural resistance to fluconazole, a standard antifungal agent commonly used in the clinic. Another example is C. glabrata, which possesses a low-level intrinsic resistance to azoles. These treatments are not effective in 15%–20% of C. glabrata infection cases. Therefore, Candida spp. other than C. albicans warrant more research. Undoubtedly, multiple-species infections will make antifungal treatment far more difficult, requiring increased diagnostic testing. It also provides a possible answer to the frustrating clinical question of why many patients with SS are so resistant to antifungal treatment. Antifungal susceptibility testing prior to prescribing medication is therefore highly recommended. It has been universally accepted that in vitro standardized susceptibility testing can provide predictive utility for clinical outcome and a reliable reference for decision-making15.
In susceptibility testing, 43% of Candida isolates were not responsive to fluconazole, a firstline prescription antifungal drug. While some have said that the organism possesses an innate immunity to the drugs, it is more likely that the organism possesses an evolved resistance to the drugs. More importantly, we found a significant cross-resistance between fluconazole and itraconazole. This means that clinically, when 1 triazole antifungal medication has failed in treating Candida infection, a switch to another azole agent may also be ineffective. In addition to azoles, polyene antimycotics such as amphotericin B and nystatin are alternatives, because some strains are still highly vulnerable to amphotericin B. When mixed Candida infection and diverse susceptible strains are present, a combination of medications might be required to achieve the best effect. We also advise considering non-azoles and topical antifungal agents as part of the therapy, especially when species such as C. krusei and C. glabrata are involved.
Another finding of our study is a high rate of resistance to triazole agents in C. albicans, which is typically susceptible to fluconazole and other azoles. This could be primary or, more likely, secondary resistance. The population group in our study had a recurrent candidal infection history, which is common in patients with SS. Accordingly, the common resistance to fluconazole more likely developed through repeated treatment. There are reports about C. albicans resistance to azoles, especially in hosts with the human immunodeficiency virus (HIV) who have undergone repeated courses of antifungal therapy16,17. Such repeated treatments make it important clinically to apply drug susceptibility testing (especially to azoles) to patients with SS who have oral candidiasis. We also advise considering non-azole and topical antifungal agents in patients with extensive prior azole exposure, as well as medication to enhance salivation.
Our study supports the requirement for regular stomatologic surveillance of patients with SS, and underlines the clinical importance of identifying mixed Candida spp. and their antifungal susceptibility. The detection of multiple Candida spp. will lead to early and better antifungal management of oral candidiasis in SS.
Acknowledgment
We thank Lixin Liu and Prof. Lakshman P. Samaranayake for their technical help.
- Accepted for publication June 27, 2011.