Annals of Diagnostic Pathology 17 (2013) 37–40

Contents lists available at SciVerse ScienceDirect

Annals of Diagnostic Pathology

Candidal infection in oral leukoplakia: a clinicopathologic study of 396 patients fromeastern China

Lan Wu DDS a,1, Jinqiu Feng DDS b,1, Linjun Shi DDS a, Xuemin Shen DDS a,Wei Liu MD c,⁎, Zengtong Zhou DDS a,⁎a Department of Oral Mucosal Diseases, Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, Chinab Department of Preventive Dentistry, Shanghai Municipal Hospital for Oral Health, Shanghai, Chinac Shanghai Key Laboratory of Stomatology, Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China

⁎ Corresponding authors. Tel.: +86 21 23271699; faxE-mail addresses: liuweb@hotmail.com (W. Liu), zho

(Z. Zhou).1 L Wu and J Feng contributed equally to this work.

1092-9134/$ – see front matter. Crown Copyright © 20http://dx.doi.org/10.1016/j.anndiagpath.2012.05.002

a b s t r a c t

a r t i c l e i n f o

Keywords:

CandidaCandidal leukoplakiaOral leukoplakiaDysplasiaClinicopathologic feature

Previous studies have suggested a link between the presence of Candida invasion and oral premalignantlesion. The objective of the current study was to investigate the clinicopathologic features of candidalinfection in biopsies of a large retrospective cohort of patients with premalignant oral leukoplakia (n = 396)from eastern China and assess the clinical implications. Candidal hyphae were detected with periodic acid-Schiff staining of the biopsy samples. The results showed that 59 patients (15.9%) with oral leukoplakia wereinfected by Candida. The average age of the patients with candidal leukoplakia was 60.7 years with equal sexratio. The tongue was the predominant site (66.1%). Epithelial hyperplasia and dysplasia were involved in44.1% and 55.9% of patients, respectively. Multivariate analysis revealed that patient older than 60 years (oddratio [OR], 2.28; P = .005), lesion located at the tongue (OR, 1.89; P = .038), and presence of dysplasia (OR,2.02; P = .018) were significant risk factors of candidal infection in oral leukoplakia. Collectively,clinicopathologic features of candidal leukoplakia in eastern China were elucidated. A point to highlightwas that we identified a subpopulation that was more liable to candidal infection. Elderly patients with oraltongue leukoplakia with epithelial dysplasia had much higher risk of candidal infection. Antifungal therapywas further recommended to be routine treatment of this subpopulation.

Crown Copyright © 2013 Published by Elsevier Inc. All rights reserved.

1. Introduction

Oral Candida is a commensal colonizer of mucous membranes thatcan become an opportunistic pathogen. Epithelial changes of the oralmucosa, suchas trauma,maceration, atrophy, hyperplasia, anddysplasia,may compromise the mucosal barrier and predispose to candidalinfection [1,2]. Histopathologically, oral Candida presents epithelialhyperplasia, hyperparakeratosis, superficial microabscess formation,and various degrees of chronic inflammation in the lamina propria.Candidal hyphaemay be seen invading the epithelium at right angles tothe surface, which associatedwith the collections of polymorphonuclearleukocytes [3,4]. Histologically, detectable Candida invasion was bestdone by periodic acid-Schiff (PAS) reagent staining [4].

Oral leukoplakia is defined as “a white plaque of questionable riskhaving excluded (other) known diseases or disorders that carry noincreased risk for cancer,”which is the best-known potentiallymalignantlesion of the oral mucosa [5,6]. Epithelial hyperplasia and dysplasia are 2of the key histologic features of leukoplakia. Candida invasion has been

: +86 21 63087076.uzengtong@hotmail.com

13 Published by Elsevier Inc. All rig

suggested to be a significant risk factor of malignant transformation oforal leukoplakia [7]. Histopathologically, candidal leukoplakia presentshyperorthokeratinized or hyperparakeratinized and various degrees of achronic inflammatory cell infiltrate seen in the lamina propria; theparakeratinized surface epithelium may show irregular separation.Candidal hyphae may be seen invading the epithelium at right angles tothe surface, which associated with the collections of polymorphonuclearleukocytes forming microabscesses [4].

Previous reports from the United Kingdom have determined thefrequency of candidal infection in biopsies of oral mucosal diseasesand shown that candidal infection correlatedwith the presence of oraldysplasia and neoplasia [8,9]. However, the issue of the prevalenceand distribution of candidal infection in oral leukoplakia has seldombeen addressed in China. We, therefore, retrospectively reviewed alarge series of patients with oral leukoplakia (n = 396) from easternChina to investigate the epidemiological and clinicopathologicfeatures of candidal infection in oral leukoplakia and to assess theclinical implications in the current hospital-based study.

2. Materials and methods

All the medical records of patients with the clinical and pathologicdiagnosis of oral leukoplakia from January 2005 to December 2007

hts reserved.

Fig. 1. Age distribution of patients with oral leukoplakia. Group 1, general leukoplakia.Group 2, candidal leukoplakia.

38 L. Wu et al. / Annals of Diagnostic Pathology 17 (2013) 37–40

were retrospectively reviewed in a standard computerized databaseof Ninth People's Hospital, Shanghai Jiao Tong University School ofMedicine. All the patients enrolled in this study underwent biopsy.The biopsywas fixed in formalin, embedded in paraffin, and processedfor routine histopathologic examination. The World Health Organi-zation criteria [10] for oral leukoplakia and epithelial dysplasia wereused when examining the histopathology of the sections. Thediagnosis of candidal leukoplakia was accorded to the Sitheeque-Samaranayake criteria [4]. Histopathologic diagnosis was given by theoral pathologists on duty from hematoxylin-eosin (HE)–stainedslides, and candidal hyphae were detected with PAS staining of thebiopsy samples.

Information regarding age, sex, lesion site, and histologic status atthe time of the initial diagnosis of oral leukoplakia was alldocumented in detail. To analyze the candidal infection in oralleukoplakia, all the study participants were classified into group 1,general leukoplakia, which was not infected by Candida, and group 2,candidal leukoplakia, which was infected by Candida. This study wasapproved by the local institutional review board.

A descriptive analysis was performed on clinicopathologic param-eters. Statistical analysis was carried out with the χ2 test and Fisherexact test among qualitative variables and the Student t test amongquantitative variables. Logistic regression was applied to evaluateodds ratios (ORs) for indicative of infection risk. Odds ratios with 95%confidence interval (CI) and P values were reported. All the tests were2 sided, and P b .05 was considered statistically significant.

3. Results

3.1. Characteristics of candidal leukoplakia

The baseline characteristics of 396 patients with oral leukoplakiaare summarized in Table 1. A total of 59 patients (15.9%)with candidalleukoplakia were identified for this study, ranging from 11 to 90 yearswith an average age of 60.7 years at the time of diagnosis. The percentdistribution of the patient age was shown in Fig. 1. The peak incidence(27.1%) of the patients with candidal leukoplakia was the fifth decadeof life. These patients with candidal leukoplakia were 29 females and30 males. The tongue was affected in 39 patients (66.1%), followed bythe buccal mucosa (20.3%). Epithelial hyperplasia and dysplasia wereinvolved in 44.1% and 55.9% of patients, respectively. Representativehistopathologic photographs of HE and PAS staining of candidalleukoplakia with epithelial hyperplasia and dysplasia were shown inFig. 2.

Table 1Baseline characteristics of 396 patients with oral leukoplakia

Characteristic, n (%) General leukoplakia Candidal leukoplakia

Total 337 (84.1) 59 (15.9)Age (y)Mean (SD) 54.6 (13.3) 60.7 (15.4)Range 7-87 11-90Not available 11 –

SexFemale 170 (50.4) 29 (49.2)Male 167 (49.6) 30 (50.8)SiteTongue 162 (48.1) 39 (66.1)Buccal mucosa 101 (30.0) 12 (20.3)Gingiva 45 (13.4) 4 (6.8)Palate 12 (3.6) 3 (5.1)Mouth floor 10 (3.0) –

Lip 6 (1.8) –

Oropharynx 1 (0.3) 1 (1.7)HistologyHyperplasia 224 (66.5) 26 (44.1)Dysplasia 113 (33.5) 33 (55.9)

3.2. Comparison of general leukoplakia and candidal leukoplakia

To define the differences in clinicopathologic parameters betweengeneral leukoplakia and candidal leukoplakia, a comparative analysiswas preformed (Table 2). The average age of the patients with generalleukoplakia was 54.6 years compared with that of 60.7 years of thepatients with candidal leukoplakia (Student t test, P = .002), with adifference in the age group (≤60, N60 years) (Fisher exact test, P =.003). Significant difference in lesion site (Fisher exact test, P = .011)was also observed, whereas differences in sex were not observedbetween the 2 groups.

To evaluate the candidal infection risk of patients with oralleukoplakia, clinicopathologic parameters were analyzed by using thelogistic regression model (Table 2). On univariate analysis, sex wasnot associated significantly with infection risk. The elderly patient(N60 years) was associated with 2.43-fold (95% CI, 1.39-4.27; P =.002) increased infection risk compared with the nonelderly patient.The lesion located on the tongue was associated with 2.11-fold (95%CI, 1.18-3.76; P = .012) increased infection risk compared with thatlocated on the other oral sites. The lesion had dysplasia was associatedwith 2.52-fold (95% CI, 1.44-4.41; P = .001) increased infection riskcompared with the lesion had hyperplasia.

To further assess the influence of each factor, we did multivariateanalysis. The 3 factors retained statistical significance. The infectionrisk of the elderly patient with oral leukoplakia (≥60 years) washigher than the nonelderly patient (adjusted OR, 2.28; 95% CI, 1.28-4.06; P = .005). The lesion located on the tongue was associated withincreased infection risk compared with that located on the other oralsites (adjusted OR, 1.89; 95% CI, 1.04-3.46; P = .038). The infectionrisk of the lesion that had dysplasia was higher than the lesion thatonly had hyperplasia (adjusted OR, 2.02; 95% CI, 1.13-3.63; P= .018).

4. Discussion

Previous studies have suggested that the presence of Candidainvasion was associated with the development of oral dysplasia andsquamous cell carcinoma [11-16]. The current study attempts toelucidate the epidemiological and clinicopathologic features ofcandidal infection in biopsies of a large series of patients withpremalignant oral leukoplakia (n = 396) from eastern China andassess the clinical implications of candidal infection in oral leukopla-kia. Multivariate regression analysis revealed that old age, tonguelesion, and presence of dysplasia were significant risk factors ofcandidal infection in oral leukoplakia.

We found that the frequency of candidal infection in biopsies of oralleukoplakiawas 15.9%, within the range reported in the literature [4,17].We observed the average age at diagnosis of candidal leukoplakia waslarger than that reported by Arendorf et al [18]. The sex ratio was equal

Fig. 2. Representative histopathology of HE and PAS staining of candidal leukoplakia. Candidal hyphae were invading the epithelium of hyperplasia (A and B) and dysplasia (C and D).Arrow shows candidal hyphae.

Table 2Comparison of general leukoplakia and candidal leukoplakia

Patients Generalleukoplakia

Candidalleukoplakia

P UnivariateLRA

P MultivariateLRA

n (%) 337 59 OR(95% CI)

OR(95% CI)

P

Age (y)Mean (SD) 54.6 (13.3) 60.7 (15.4) .002Range 7-87 11-90

Age group .003≤60 224 (68.7) 28 (47.5) 1.00 (ref) 1.00

(ref)N60 102 (31.3) 31 (52.5) 2.43

(1.39-4.27)

.002 2.28(1.28-4.06)

.005

Sex .888Female 170 (50.4) 29 (49.2) 1.00 (ref)Male 167 (49.6) 30 (50.8) 1.05

(0.61-1.83)

.855

Site .011Others 175 (51.9) 20 (33.9) 1.00 (ref) 1.00

(ref)Tongue 162 (48.1) 39 (66.1) 2.11

(1.18-3.76)

.012 1.89(1.04-3.46)

.038

Histology .001Hyperplasia 224 (66.5) 26 (44.1) 1.00 (ref) 1.00

(ref)Dysplasia 113 (33.5) 33 (55.9) 2.52

(1.44-4.41)

.001 2.02(1.13-3.63)

.018

Abbreviation: LRA, logistic regression analysis.

39L. Wu et al. / Annals of Diagnostic Pathology 17 (2013) 37–40

in our area, but a significant male predilection was found in London,United Kingdom [8]. Williams [16] reported that candidal leukoplakiaaffected the following oral sites in decreasing order of frequency: thebuccal mucosa, palate, and tongue, whereas we observed that thetongue was the most common site, followed by buccal mucosa, andlesions on the palate and other sites were uncommon. These wereprobably because of the differences in study populations and geograph-ical areas between the current cohort and previous reports.

It is particularly noteworthy that we evaluated the clinicopatho-logic parameters in candidal infection risk of patients with oralleukoplakia and identified a subpopulation that was more liable tocandidal infection. We found that the average age at diagnosis ofcandidal leukoplakia was larger than general leukoplakia, and theinfection risk in the elderly patient (N60 years) with oral leukoplakiawas higher than in the nonelderly patient, which maybe correlate toage-related alteration contributing to the oral microflora andmicroenvironment. The tongue was the predominant site of oralleukoplakia, and the infection risk of the tongue was still significantlyhigher than other oral sites. In addition, the infection risk of the lesionthat had dysplasia was higher than the lesion that only hadhyperplasia,which further confirmed thatCandida infection correlatedwith the presence of oral dysplasia [4,8,9]. Based on the clinicalimplications of the current study, antifungal therapy was furtherrecommended to be the routine treatment of patientswith oral tongueleukoplakia older than 60 years and who had epithelial dysplasia.

In summary, the current study was a clinicopathologic analysis ofcandidal infection in a relatively large cohort of patients with oralleukoplakia. The clinicopathologic features of candidal leukoplakia ineastern China were elucidated. A point to highlight was that we

40 L. Wu et al. / Annals of Diagnostic Pathology 17 (2013) 37–40

identified a subpopulation that was more liable to candidal infection.Elderly patients with oral tongue leukoplakia with epithelial dysplasiahad much higher risk of candidal infection. Antifungal therapy wasfurther recommended to be the routine treatmentof this subpopulation.

Acknowledgments

The authors thank Dr Jiang Li (Department of Oral Pathology, NinthPeople's Hospital, Shanghai Jiao Tong University School of Medicine,Shanghai, China) for support and help in getting the medical records.This study was supported by Shanghai Health Bureau (ZYSNXD-CC-ZDYJ004), Science and Technology Commission of Shanghai(11DZ1972600 and 10DZ1974200), and Shanghai Natural ScienceFoundation (11ZR1420400).

References

[1] Williams D, Lewis M. Pathogenesis and treatment of oral candidosis. J OralMicrobiol 2011;3:5571.

[2] Samaranayake LP. Host factors and oral candidosis. In: Samaranayake LP,MacFarlane TW, editors. Oral candidosis. London: Wright; 1990. p. 66–103.

[3] Cawson RA, Lehner T. Chronic hyperplastic candidosis-candidal leukoplakia. Br JDermatol 1968;80:9–16.

[4] Sitheeque MA, Samaranayake LP. Chronic hyperplastic candidosis/candidiasis(candidal leukoplakia). Crit Rev Oral Biol Med 2003;14:253–67.

[5] Liu W, Shi LJ, Wu L, et al. Oral cancer development in patients with leukoplakia—clinicopathological factors affecting outcome. PLoS One 2012;7:e34773.

[6] Liu W, Wang YF, Zhou HW, et al. Malignant transformation of oralleukoplakia: a retrospective cohort study of 218 Chinese patients. BMC Cancer2010;10:685.

[7] van der Waal I. Potentially malignant disorders of the oral and oropharyngealmucosa; terminology, classification and present concepts of management. OralOncol 2009;45:317–23.

[8] Barrett AW, Kingsmill VJ, Speight PM. The frequency of fungal infection in biopsiesof oral mucosal lesions. Oral Dis 1998;4:26–31.

[9] McCullough M, Jaber M, Barrett AW, et al. Oral yeast carriage correlates withpresence of oral epithelial dysplasia. Oral Oncol 2002;38:391–3.

[10] World Health Organization. In: Pindborg JJ, Smith CJ, van der Wall I, editors.Histological typing of cancer and precancer of the oral mucosa. 2nd ed. Berlin:Springer-Verlag; 1997. p. 21–31.

[11] Bastiaan RJ, Reade PC. The prevalence of Candida albicans in the mouths of tobaccosmokers with and without oral mucous membrane keratoses. Oral Surg Oral MedOral Pathol 1982;53:148–51.

[12] Krogh P, Hald B, Holmstrup P. Possible mycological etiology of oral mucosalcancer: catalytic potential of infecting Candida albicans and other yeasts inproduction of N-nitrosobenzyl- methylamine. Carcinogenesis 1987;8:1543–8.

[13] O'Grady JF, Reade PC. Candida albicans as a promoter of oral mucosal neoplasia.Carcinogenesis 1992;13:783–6.

[14] Nagy KN, Sonkodi I, Szoke I, et al. The microflora associated with human oralcarcinomas. Oral Oncol 1998;34:304–8.

[15] Dwivedi PP, Mallya S, Dongari-Bagtzoglou A. A novel immunocompetent murinemodel for Candida albicans–promoted oral epithelial dysplasia. Med Mycol2009;47:157–67.

[16] Williams DM. Chronic hyperplastic candidiasis and squamous carcinoma. Br JDermatol 1969;81:125–7.

[17] Arendorf TM. Factors affecting oral candidal colonisation in health and disease. BrDent J 1984;147:267–72.

[18] Arendorf TM, Walker DM, Kingdom RJ, et al. Tobacco smoking and denturewearing in oral candidal leukoplakia. Br Dent J 1983;155:340–3.