Method-Dependent Epidemiological Cutoff Values for Detection of Triazole Resistance in Candida and Aspergillus Species for the Sensititre YeastOne Colorimetric Broth and Etest Agar Diffusion Methods.

Although the Sensititre Yeast-One (SYO) and Etest methods are widely utilized, interpretive criteria are not available for triazole susceptibility testing of Candida or Aspergillus species. We collected fluconazole, itraconazole, posaconazole, and voriconazole SYO and Etest MICs from 39 laboratories representing all continents for (method/agent-dependent) 11,171 Candida albicans, 215 C. dubliniensis, 4,418 C. glabrata species complex, 157 C.guilliermondii (Meyerozyma guilliermondii), 676 C. krusei (Pichia kudriavzevii), 298 C.lusitaniae (Clavispora lusitaniae), 911 C.parapsilosissensu stricto, 3,691 C.parapsilosis species complex, 36 C.metapsilosis, 110 C.orthopsilosis, 1,854 C.tropicalis, 244 Saccharomyces cerevisiae, 1,409 Aspergillus fumigatus, 389 A.flavus, 130 A.nidulans, 233 A.niger, and 302 A.terreus complex isolates. SYO/Etest MICs for 282 confirmed non-wild-type (non-WT) isolates were included: ERG11 (C. albicans), ERG11 and MRR1 (C. parapsilosis), cyp51A (A. fumigatus), and CDR2 and CDR1 overexpression (C. albicans and C. glabrata, respectively). Interlaboratory modal agreement was superior by SYO for yeast species and by the Etest for Aspergillus spp. Distributions fulfilling CLSI criteria for epidemiological cutoff value (ECV) definition were pooled, and we proposed SYO ECVs for S. cerevisiae and 9 yeast and 3 Aspergillus species and Etest ECVs for 5 yeast and 4 Aspergillus species. The posaconazole SYO ECV of 0.06 µg/ml for C. albicans and the Etest itraconazole ECV of 2 µg/ml for A. fumigatus were the best predictors of non-WT isolates. These findings support the need for method-dependent ECVs, as, overall, the SYO appears to perform better for susceptibility testing of yeast species and the Etest appears to perform better for susceptibility testing of Aspergillus spp. Further evaluations should be conducted with more Candida mutants.

Posaconazole MIC Distributions for Aspergillus fumigatus Species Complex by Four Methods: Impact of cyp51A Mutations on Estimation of Epidemiological Cutoff Values.

Estimating epidemiological cutoff endpoints (ECVs/ECOFFS) may be hindered by the overlap of MICs for mutant and nonmutant strains (strains harboring or not harboring mutations, respectively). Posaconazole MIC distributions for the Aspergillus fumigatus species complex were collected from 26 laboratories (in Australia, Canada, Europe, India, South and North America, and Taiwan) and published studies. Distributions that fulfilled CLSI criteria were pooled and ECVs were estimated. The sensitivity of three ECV analytical techniques (the ECOFFinder, normalized resistance interpretation [NRI], derivatization methods) to the inclusion of MICs for mutants was examined for three susceptibility testing methods (the CLSI, EUCAST, and Etest methods). The totals of posaconazole MICs for nonmutant isolates (isolates with no known cyp51A mutations) and mutant A. fumigatus isolates were as follows: by the CLSI method, 2,223 and 274, respectively; by the EUCAST method, 556 and 52, respectively; and by Etest, 1,365 and 29, respectively. MICs for 381 isolates with unknown mutational status were also evaluated with the Sensititre YeastOne system (SYO). We observed an overlap in posaconazole MICs among nonmutants and cyp51A mutants. At the commonly chosen percentage of the modeled wild-type population (97.5%), almost all ECVs remained the same when the MICs for nonmutant and mutant distributions were merged: ECOFFinder ECVs, 0.5 µg/ml for the CLSI method and 0.25 µg/ml for the EUCAST method and Etest; NRI ECVs, 0.5 µg/ml for all three methods. However, the ECOFFinder ECV for 95% of the nonmutant population by the CLSI method was 0.25 µg/ml. The tentative ECOFFinder ECV with SYO was 0.06 µg/ml (data from 3/8 laboratories). Derivatization ECVs with or without mutant inclusion were either 0.25 µg/ml (CLSI, EUCAST, Etest) or 0.06 µg/ml (SYO). It appears that ECV analytical techniques may not be vulnerable to overlap between presumptive wild-type isolates and cyp51A mutants when up to 11.6% of the estimated wild-type population includes mutants.

Prospective multicenter international surveillance of azole resistance in Aspergillus fumigatus.

To investigate azole resistance in clinical Aspergillus isolates, we conducted prospective multicenter international surveillance. A total of 3,788 Aspergillus isolates were screened in 22 centers from 19 countries. Azole-resistant A. fumigatus was more frequently found (3.2% prevalence) than previously acknowledged, causing resistant invasive and noninvasive aspergillosis and severely compromising clinical use of azoles.

Interlaboratory variability of Caspofungin MICs for Candida spp. Using CLSI and EUCAST methods: should the clinical laboratory be testing this agent?

Although Clinical and Laboratory Standards Institute (CLSI) clinical breakpoints (CBPs) are available for interpreting echinocandin MICs for Candida spp., epidemiologic cutoff values (ECVs) based on collective MIC data from multiple laboratories have not been defined. While collating CLSI caspofungin MICs for 145 to 11,550 Candida isolates from 17 laboratories (Brazil, Canada, Europe, Mexico, Peru, and the United States), we observed an extraordinary amount of modal variability (wide ranges) among laboratories as well as truncated and bimodal MIC distributions. The species-specific modes across different laboratories ranged from 0.016 to 0.5 µg/ml for C. albicans and C. tropicalis, 0.031 to 0.5 µg/ml for C. glabrata, and 0.063 to 1 µg/ml for C. krusei. Variability was also similar among MIC distributions for C. dubliniensis and C. lusitaniae. The exceptions were C. parapsilosis and C. guilliermondii MIC distributions, where most modes were within one 2-fold dilution of each other. These findings were consistent with available data from the European Committee on Antimicrobial Susceptibility Testing (EUCAST) (403 to 2,556 MICs) for C. albicans, C. glabrata, C. krusei, and C. tropicalis. Although many factors (caspofungin powder source, stock solution solvent, powder storage time length and temperature, and MIC determination testing parameters) were examined as a potential cause of such unprecedented variability, a single specific cause was not identified. Therefore, it seems highly likely that the use of the CLSI species-specific caspofungin CBPs could lead to reporting an excessive number of wild-type (WT) isolates (e.g., C. glabrata and C. krusei) as either non-WT or resistant isolates. Until this problem is resolved, routine testing or reporting of CLSI caspofungin MICs for Candida is not recommended; micafungin or anidulafungin data could be used instead.

Rifampicin-resistant and rifabutin-susceptible Mycobacterium tuberculosis strains: a breakpoint artefact?

OBJECTIVES: It has long been assumed that some rifampicin-resistant Mycobacterium tuberculosis strains are susceptible to, and thus treatable with, rifabutin. However, clinical breakpoints for susceptibility testing of rifabutin as well as the evidence for a clinical effect of rifabutin in rifampicin-resistant strains remains poorly defined. The objective of this study was to re-evaluate the breakpoint for rifabutin in relation to its MIC wild-type distribution and the presence of mutations in rpoB. METHODS: The MIC in 7H10 Middlebrook medium was determined for clinical isolates of M. tuberculosis (n = 95), where a majority were multidrug resistant. Additionally, all strains were screened for rpoB mutations by sequencing and the GenoType MTBDRplus assay. RESULTS: Rifampicin resistance was confirmed by genotypical and/or phenotypical tests in 73 isolates (76.8%). Nineteen isolates, defined as rifampicin resistant and rifabutin susceptible according to the present breakpoint, exhibited significantly higher MICs of rifabutin (0.064-0.5 mg/L) than rifabutin-susceptible isolates without any detectable mutations in rpoB (P < 0.001). These 19 isolates were clearly resistant to rifampicin (MIC 2-256 mg/L) and all but one had mutations in rpoB, with 9 (47.4%) specifically in Asp516Val. CONCLUSIONS: Our results indicate that rifampicin-resistant but rifabutin-susceptible isolates according to the present breakpoints harbour rpoB mutations and have a rifabutin MIC significantly higher than strains without any detectable mutations in rpoB. So far there are no clinical, pharmacological or microbiological data to confirm that such isolates can be treated with rifabutin and we suggest a revision of the current breakpoints.

Candidaemia in Sweden: a nationwide prospective observational survey.

A prospective observational nationwide investigation was performed from September 2005 to August 2006 to study the epidemiology of candidaemia in Sweden. From 385 patients, 403 isolates were recovered, yielding an incidence of 4.2 cases per 100 000 inhabitants. Candida albicans was the most common species (61%), followed by Candida glabrata (20%) and Candida parapsilosis (9%). The rates of resistance to fluconazole were ≤ 1% in C. albicans and 6-29% in non-albicans species other than C. glabrata and Candida krusei. Resistance to voriconazole was rare, except for C. glabrata and C. krusei. Only three isolates had reduced susceptibility to amphotericin B, and one had reduced susceptibility to caspofungin.

Occurrence of yeasts in faecal samples from Antarctic and South American seabirds.

During an expedition to the Southern Argentinean town of Ushuaia, the Antarctic Peninsula, Antarctic Islands and the Falkland Islands, we collected 94 faecal specimens from wild birds to screen for yeast within the different bird species. The yeast species were identified by morphological features and commercial characterisation kits. From 54% of the specimens, we isolated 122 strains representing 29 yeast species. Debaryomyces hansenii, Candida lambica and Candida krusei were the most frequently isolated species. We found a plethora of yeasts in birds living in proximity to humans, whereas birds living in more remote areas were colonised with a lower number of fungal species.

Wild-type distributions of seven oral second-line drugs against Mycobacterium tuberculosis.

OBJECTIVES: To determine wild-type minimum inhibitory concentration (MIC) distributions for Mycobacterium tuberculosis, as the background data for defining susceptibility breakpoints are limited. METHODS: We determined wild-type MIC distributions of M. tuberculosis using a 96-stick replicator in Middlebrook 7H10 (7H10) medium for ethionamide (ETH), prothionamide, thiacetazone, cycloserine, rifabutin (RFB), clofazimine and linezolid in consecutive susceptible clinical isolates (n = 78). RESULTS: Tentative epidemiological wild-type cut-offs (ECOFF) were determined for all investigated drugs where World Health Organization recommended critical concentrations for 7H10 are lacking, except for ETH. As the ECOFF was closely related to the non-wild-type strains for ETH and thiacetazone, the use of an intermediary (I) category in drug susceptibility testing could increase reproducibility. The cross-resistance between ETH and isoniazid was 21%. Applying 0.5 mg/l as a breakpoint for RFB classified two non-wild type and rpoB mutated isolates as susceptible for RFB and resistant against rifampicin. CONCLUSIONS: We propose that wild-type MIC distributions should be used as a tool to define clinical breakpoints against second-line drugs. This is increasingly important considering the rapid emergence of drug resistance.

Wild-type MIC distributions for aminoglycoside and cyclic polypeptide antibiotics used for treatment of Mycobacterium tuberculosis infections.

The aminoglycosides and cyclic polypeptides are essential drugs in the treatment of multidrug-resistant tuberculosis, underscoring the need for accurate and reproducible drug susceptibility testing (DST). The epidemiological cutoff value (ECOFF) separating wild-type susceptible strains from non-wild-type strains is an important but rarely used tool for indicating susceptibility breakpoints against Mycobacterium tuberculosis. In this study, we established wild-type MIC distributions on Middlebrook 7H10 medium for amikacin, kanamycin, streptomycin, capreomycin, and viomycin using 90 consecutive clinical isolates and 21 resistant strains. Overall, the MIC variation between and within runs did not exceed +/-1 MIC dilution step, and validation of MIC values in Bactec 960 MGIT demonstrated good agreement. Tentative ECOFFs defining the wild type were established for all investigated drugs, including amikacin and viomycin, which currently lack susceptibility breakpoints for 7H10. Five out of seven amikacin- and kanamycin-resistant isolates were classified as susceptible to capreomycin according to the current critical concentration (10 mg/liter) but were non-wild type according to the ECOFF (4 mg/liter), suggesting that the critical concentration may be too high. All amikacin- and kanamycin-resistant isolates were clearly below the ECOFF for viomycin, and two of them were below the ECOFF for streptomycin, indicating that these two drugs may be considered for treatment of amikacin-resistant strains. Pharmacodynamic indices (peak serum concentration [Cmax]/MIC) were more favorable for amikacin and viomycin compared to kanamycin and capreomycin. In conclusion, our data emphasize the importance of establishing wild-type MIC distributions for improving the quality of drug susceptibility testing against Mycobacterium tuberculosis.

Wild-type MIC distributions of four fluoroquinolones active against Mycobacterium tuberculosis in relation to current critical concentrations and available pharmacokinetic and pharmacodynamic data.

OBJECTIVES: To describe wild-type distributions of the MIC of fluoroquinolones for Mycobacterium tuberculosis in relation to current critical concentrations used for drug susceptibility testing and pharmacokinetic/pharmacodynamic (PK/PD) data. METHODS: A 96-stick replicator on Middlebrook 7H10 medium was used to define the MICs of ciprofloxacin, ofloxacin, moxifloxacin and levofloxacin for 90 consecutive clinical strains and 24 drug-resistant strains. The MICs were compared with routine BACTEC 460 susceptibility results and with MIC determinations in the BACTEC MGIT 960 system in a subset of strains using ofloxacin as a class representative. PK/PD data for each drug were reviewed in relation to the wild-type MIC distribution. RESULTS: The wild-type MICs of ciprofloxacin, ofloxacin, moxifloxacin and levofloxacin were distributed from 0.125 to 1, 0.25 to 1, 0.032 to 0.5 and 0.125 to 0.5 mg/L, respectively. The MIC data correlated well with the BACTEC 960 MGIT and BACTEC 460 results. PD indices were the most favourable for levofloxacin, followed by moxifloxacin, ofloxacin and ciprofloxacin. CONCLUSIONS: We propose S (susceptible)

Extensive transmission of an isoniazid-resistant strain of Mycobacterium tuberculosis in Sweden.

SETTING: City of Stockholm, Sweden. BACKGROUND: The incidence of tuberculosis (TB) in Sweden increased by 40% between 2003 and 2005. The spread of a unique TB strain resistant to isoniazid (INH) contributed to this increase. OBJECTIVE: To describe outbreaks of TB caused by this single strain, elucidate possible causes for its extensive spread and identify shortcomings of the TB control programme in Sweden. RESULTS: We identified a cluster consisting of 102 culture-confirmed TB cases with identical DNA fingerprints and 26 epidemiologically related cases, not confirmed by culture, all diagnosed between 1996 and 2005. Five partly separate outbreaks of this strain were discovered. Epidemiological links were established for 56% of the culture-confirmed cases and for all cases not confirmed by culture. Three patients died while receiving treatment, four became failures and eight defaulted or were lost to follow-up. Only eight patients received directly observed treatment (DOT) up to a period of 3 months, although 40% had poor adherence. CONCLUSIONS: Shortcomings of the national TB programme were revealed. Improved contact tracing and case holding, including DOT, is crucial to reduce TB transmission in Sweden.

Histoplasmosis in Europe: report on an epidemiological survey from the European Confederation of Medical Mycology Working Group.

The purpose of this survey was to systematically collect data on individuals with histoplasmosis in Europe over a 5-year period (from January 1995 to December 1999). This included information on where and how the infection was acquired, the patient's risk factors, the causative organism, how the infection was diagnosed and what therapy the patients received. Data were sent on a standardized survey form via a national convenor to the coordinator. During the survey, 118 cases were reported, with 62 patients having disseminated disease, 31 acute pulmonary infection, chronic pulmonary infection in 6 and localized disease in 2 patients. For 17 patients, the diagnosis of histoplasmosis was incidental, usually secondary to investigations for lung cancer. Most patients had travelled to known endemic areas, but 8 patients (from Italy, Germany and Turkey) indicated that they had not been outside their countries of origin and hence these cases appear to be autochthonous. Notable observations during the survey were the reactivation of the disease up to 50 years after the initial infection in some patients and transmission of the infection by a transplanted liver. Itraconazole was the most commonly used therapy in both pulmonary and disseminated disease. The observation of autochthonous cases of disease suggests that the endemic area of histoplasmosis is wider than classically reported and supports continued surveillance of the disease throughout Europe.

Performance of commercial latex agglutination tests for the differentiation of Candida dubliniensis and Candida albicans in routine diagnostics.

Candida dubliniensis is phenotypically similar to Candida albicans and may therefore be underdiagnosed in the clinical microbiology laboratory. The performance of Bichro-Dubli latex agglutination test for rapid species identification of C. dubliniensis was prospectively evaluated on 111 vaginal and 118 respiratory isolates. These had presumptively been identified as C. albicans/C. dubliniensis by their green colonies on CHROMagar Candida plates. Bichro-Dubli test identifed 2 (1.8%) vaginal and 6 (5.1%) respiratory isolates as C. dubliniensis. The test was also positive for 37 C. dubliniensis control strains characterised by 18S-28S DNA-sequencing. Bichro-Dubli test is thus a sensitive and accurate tool for rapid diagnostics in routine laboratories.

Multicentre determination of quality control strains and quality control ranges for antifungal susceptibility testing of yeasts and filamentous fungi using the methods of the Antifungal Susceptibility Testing Subcommittee of the European Committee on Antimicrobial Susceptibility Testing (AFST-EUCAST).

A multicentre study involving seven laboratories was performed using techniques recommended by the Antifungal Susceptibility Testing Subcommittee of the European Committee on Antimicrobial Susceptibility Testing (AFST-EUCAST) to evaluate and propose quality control ranges and strains for susceptibility testing of fermentative yeasts and filamentous fungi. Participating laboratories tested the susceptibilities of a panel of 12 encoded isolates to amphotericin B, flucytosine, fluconazole, itraconazole, voriconazole and posaconazole. In total, 15 lots of assay medium were tested, with one lot being common to all laboratories, and 18 144 MIC values were determined. Intra- and inter-laboratory agreements and intra-class correlation coefficients (ICCs) of the results for each drug/strain/lot combination were calculated. An average value of 85% agreement was selected for validation purposes. The average percentage of intra-laboratory agreement was 90-95%, with ICC values of 0.90-0.95 (p <0.01). Inter-laboratory reproducibility was also high, with 92% agreement and an ICC of 0.97 (p <0.01). The reproducibility was somewhat better with the common lot of assay medium (96% agreement) than with the different lots (91% agreement), but this difference was not significant. Two isolates that showed trailing growth had agreement percentages below the 85% limit selected for validation purposes and were therefore excluded from the panel of quality control strains. The recommended EUCAST methodologies were found to be highly reproducible and reliable for susceptibility testing of yeasts and filamentous fungi. Ten isolates are proposed for use as quality control strains with these EUCAST procedures.

Tacrolimus ointment vs steroid ointment for eyelid dermatitis in patients with atopic keratoconjunctivitis.

AIMS: The main objective of this explorative study was to evaluate if tacrolimus ointment could be safer than corticosteroid ointment, with special reference to the intraocular pressure in the treatment of eyelid eczema in patients with atopic keratoconjunctivitis (AKC). Secondary aims were to compare the effects of the treatments on eyelid eczema and their potential impact on ocular surface inflammation. METHODS: Tacrolimus 0.1% ointment and clobetasone butyrate 0.05% ointment were compared in a double-masked explorative crossover study. In total, 25 AKC patients were included. Each ointment was applied twice daily for 3 weeks, with 2 weeks of washout before, between, and after treatments. Efficacy was determined by eye examination and the patients' own symptom scoring. Cytology and cytokine measurements were performed on tear samples. Safety parameters were intraocular pressure, presence of bacteria and fungi, and the patients' reports of adverse events. The validity of the crossover design was explored with analysis of variance, and the effect of each medication was calculated with paired t-test and Wilcoxon paired test. RESULTS: A total of 20 patients completed the study. Both treatments were effective in reducing signs and symptoms of eyelid eczema, with a near superior benefit for tacrolimus in terms of eczema (total skin score) signs (P=0.05). No serious adverse events occurred and interestingly, intraocular pressure was not evidently affected by either treatment. CONCLUSION: Tacrolimus 0.1% ointment is a promising alternative therapy for eyelid eczema in AKC patients. Long-term studies are needed to further determine the value of tacrolimus in this patient group.

Comparison of the EUCAST-AFST broth dilution method with the CLSI reference broth dilution method (M38-A) for susceptibility testing of posaconazole and voriconazole against Aspergillus spp.

The susceptibilities of 40 clinical isolates of Aspergillus spp. (Aspergillus fumigatus, Aspergillus flavus, Aspergillus niger, Aspergillus terreus) were determined for posaconazole and voriconazole by the CLSI M38-A and EUCAST-AFST broth dilution methods. Where a discrepancy was observed between the methods, the EUCAST method tended to give higher MIC values. Overall, the level of agreement was 92.5% and the intra-class correlation coefficient was > 0.90.

Clinico-mycological profile of dermatophytosis in a reference centre for leprosy and dermatological diseases in Addis Ababa.

To assess the clinical and fungal species spectrum of dermatophyte infection in a reference centre in Addis Ababa, 539 dermatological patients with signs of dermatophytosis were investigated. Seventy-one percent were female and 29% male, aged 2-66 years (median 9). Four hundred-fifteen (77%) had at least one skin lesion. Tinea capitis was diagnosed in 138/155 males (89%) as compared to 214/384 females (40%) (p < 0.05). T. capitis was diagnosed in 69% of the 374 children. Fingernails were affected in 132/145 (91%) of onychomycosis, 118 (90%) of these patients were females and 14 males (p < 0.05). Tinea corporis was observed in 45, and other types of tinea in 12 patients. Thirty-six percent of all patients had also other skin lesions, mostly impetigo. Of 490 cultured samples 364 (74%) grew dermatophytes: Trichophyton violaceum in 84%, Trichophyton verrucosum in 9.6%, Trichophyton tonsurans in 1.4% and T. rubrum in 0.5%. Additionally, 15 isolates were identified as white variants of T. violaceum, in 3 cases confirmed by sequencing of the rDNA ITS 2 region. T. capitis in young males and T. unguium of fingernails in females were the most common manifestations of dermatophytosis in Addis Ababa, usually caused by T.violaceum.

White variants of Trichophyton violaceum isolated in Ethiopia.

Certain dermatophytes are geographically restricted and endemic in particular parts of the world, while other species may have a sporadic but worldwide distribution. Trichophyton violaceum is one of the most common dermatophytes causing tinea capitis, and is the predominant cause of tinea in Africa, South America and the Indian subcontinent. Among 1187 dermatophyte isolates collected from Ethiopian patients with various types of tinea, 32 isolates had uncharacteristic phenotypic features. Based on conventional methods complemented by sequence analysis of the rDNA ITS2 region, these isolates were identified as white variants of T. violaceum. This is the first time that white isolates of T. violaceum have been identified in Ethiopia.

Prevalence of tinea capitis in Ethiopian schoolchildren.

The prevalence of dermatophytosis and the spectrum of dermatophyte species were determined in children attending two schools in Addis Ababa, Ethiopia. Demographic and clinico-dermatological data were collected. Specimens were taken for microscopy and culture from all suspected lesions. Dermatophyte species were identified by morphology and biochemical tests, supplemented by sequencing of the rDNA ITS 2 region in selected isolates. From the Biruh Tesfa Elementary School (BTES) 824 students, and from Mount Olive Academy (MOA) all 124 students, were included. In BTES 513 (62.3%) students were clinically diagnosed with dermatophytosis, 463 (90.3 %) of them with tinea capitis. In 200 consecutive samples from BTES, and in 66 from MOA, 75 and 62%, respectively, contained fungal elements at microscopy. From BTES, 163/496 (33%) samples were culture-positive, of which 149 (91.4%) grew with dark purple colonies identified as Trichophyton violaceum, while 244 (49.4%) samples were contaminated. A few strains grew slowly developing white to cream colonies, two were identified as T. verrucosum, and 12 as white T. violaceum. From MOA 44 (66.7%) of samples were culture-positive, 38 (87%) were identified as T. violaceum, and one (2.3%) as T. verrucosum, while 33% showed no growth. Four white isolates of T.violaceum were confirmed by DNA-sequencing. Dermatophytosis was thus diagnosed in 55-62% of children screened at two schools of different socioeconomic standards in the Ethiopian capital. Trichophyton violaceum constituted 87-90% of all isolates. White variants of T. violaceum were diagnosed in 16 cases.

No apparent association between periocular and ocular microcolonization and the degree of inflammation in patients with atopic keratoconjunctivitis.

BACKGROUND: The cause of the chronic inflammation in atopic keratoconjunctivitis (AKC), the ocular manifestation of atopic eczema/dermatitis syndrome, is largely unknown. OBJECTIVE: To investigate the possibility that microorganisms may be important in the inflammatory activity in AKC. METHODS: Fifteen patients with AKC participated in the study. The presence of aerobic bacteria and fungi was related to the severity of clinical signs, the numbers of inflammatory cells in tears and conjunctival biopsies, and the concentration of various cytokines in tears. In addition, serological evidence for IgE sensitization to Staphylococcus aureus B antigen and Malassezia sympodialis antigen was investigated. Twelve healthy subjects were included for control purposes. RESULTS: The patients exhibited moderate clinical signs of AKC. No relation was found between the severity of AKC and the presence of microorganisms, despite the fact that S. aureus was frequently isolated. AKC patients showed significantly higher levels of IFN-gamma, TNF-alpha (tumour necrosis factor-alpha), IL-2, IL-4, IL-5 and IL-10 than controls. An association was found between conjunctival signs and the levels of all cytokines except IL-5. CONCLUSION: We found no evidence to suggest that periocular and ocular microcolonization are related to inflammatory parameters in AKC. However, confirmation of the present results in a longitudinal study with repeated clinical examinations and samplings in the same individual is required before the contribution of S. aureus to on-going inflammation in AKC can be dismissed.