729

In Vitro Antifungal Properties ofMouthrinses Containing AntimicrobialAgentsGiovanna Giuliana,* Giuseppe Pizzo,* Maria E. Milici,' Giuliana C. Musotto,' andRosalia Giangrecó'

The purpose of this study was to investigate the in vitro antifungal properties ofseven commercial mouthrinses containing antimicrobial agents. These included cetyl-pyridinium chloride (CPC), Chlorhexidine digluconate (CHX), hexetidine (HEX), san-

guinarine (SNG), and triclosan (TRN). The minimum fungicidal concentration (MFC)against six species of yeasts was determined by a broth macrodilution method. Thekill-time of mouthrinses at half the concentration of the commercial formulations wasalso determined. MFCs were achieved with each mouthrinse, except the SNG-con-taining mouthrinse; against all the organisms being tested. However, the CPC-con-taining mouthrinse appeared more active than the other products (P < 0.001). Therewere no significant differences in MFC values among CHX mouthrinse products, once

adjusted for initial concentration differences (P = 0.1). Kill-times of mouthrinsescontaining either CHX or CPC were less than or equal to 180 seconds with all thespecies of yeasts, and no significant differences were found among these products (P= 0.18). On the other hand, mouthrinses containing either TRN or HEX did not showa lethal effect on Candida albicans, Candida parapsilosis, or Candida guilliermondii.No kill-times were achieved with the SNG-containing mouthrinse. These results sug-gest that mouthrinses containing antimicrobial agents might represent an appropriatealternative to conventional antifungal drugs in the management of oral candidiasis.However, the effectiveness of antimicrobial mouthrinses as antifungal agents needs tobe evaluated in further clinical trials. 7 Periodontol 1997;68:729-733.

Key Words: Mouthrinses/therapeutic use; cetylpyridinium chloride/therapeutic use;chlorhexidine/therapeutic use; hexetidine/therapeutic use; sanguinarine/therapeutic use;triclosan/therapeutic use; fungal diseases/prevention and control.

Mouthrinses containing antimicrobial agents exert bene-ficial clinical effects when used as adjuncts in the treat-ment of periodontal disease. Antibacterial activity and ef-fectiveness in reducing or retarding plaque formationhave been comprehensively studied.1 The effects on theecological balance of the oral microflora have also beendocumented, but few data are currently available on theactivity against yeasts.12

These microorganisms constitute a part of the normal hu-man microbiota and are considered harmless commensals ofthe oral cavity. The dorsum of the tongue represents the pri-mary oral reservoir for these organisms, but mucosal surfacesand dental plaque can also harbor them.3 In systemically

*Department of Periodontology, University of Palermo, School of Den-tistry, Palermo, Italy.'Department of Hygiene and Microbiology, University of Palermo,School of Medicine.

healthy individuals, subgingival colonization by Candida spp.has been demonstrated in deep periodontal pockets,4-6 refrac-tory Periodontitis lesions,7-9 and failing osseointegrated dentalimplants.10 The emergence of these organisms in the subgin-gival flora may also occur as a result of prolonged systemicantibiotic therapy."12 It is uncertain whether yeasts representsecondary colonizers of established lesions or putative patho-gens contributing to disease progression. However, findingsin several investigations have suggested that Candida spp.may play a role in destructive periodontal diseases.4-71314

When changes occur in host defense mechanisms and oralenvironment, yeasts may induce mucosal and periodontal op-portunistic infections. In the last decade, the incidence ofthese infections has increased in parallel with the increase innumbers of inmtunocompromised patients, including HTV-in-fected individuals, transplant recipients receiving immunosup-pressive agents, and patients undergoing cytotoxic chemo-

730 ANTIFUNGAL PROPERTIES OF MOUTHRINSESJ PeriodontolAugust 1997

therapy and radiotherapy for cancer and bone marrow trans-

plantation.15-20 In this population, invasive oropharyngealyeast infections are always preceded by colonization of mu-

cosal surfaces.21 Therefore, an antifungal mouthrinse wouldbe useful in preventing oral yeast colonization and subsequentinvasive infection.

Chlorhexidine (CHX)-containing mouthrinses have beenused for prophylaxis against oral infections related to che-motherapy and radiotherapy. However, conflicting resultshave been reported with regard to the effect on oral yeastcolonization.22-27 On the other hand, there is little up-to-dateinformation concerning the antifungal properties of mouthrin-ses containing antimicrobials such as cetylpyridinium chloride(CPC), hexetidine (HEX), sanguinarine (SNG), and triclosan(TRN).1

The purpose of this study was to investigate the in vitroantifungal properties of seven commercial mouthrinses con-

taining antimicrobial agents. Experiments were performed to:

1) assess the antifungal activity of different mouthrinses bymeans of the minimum fungicidal concentration (MFC); and2) determine the kill-time at half the concentration of thecommercial formulations.

MATERIALS AND METHODS

Test OrganismsSix different species of yeasts were used: Candida albicans,Candida parapsilosis, Candida krusei, Candida guilliermon-dii, Torulopsis glabrata, and Saccharomyces cerevisiae.

Candida and Torulopsis strains were from the Departmentof Hygiene and Microbiology (University of Palermo, Italy).These strains were recent oral isolates from organ transplantpatients with oral candidosis. All isolates were identified bygerm tube formation in serum, chlamydospore production on

com meal agar, and carbon compound assimilation and fer-mentation by using two commercially available micromethodsystems.28*5 S. cerevisiae strain was supplied by the Depart-ment of Genetics (University of Palermo, Italy).

MouthrinsesThe mouthrinse products included in this study and their ac-

tive ingredients are shown in Table 1. Sodium fluoride (NaF)and sodium monofluorophosphate (MFP), incorporated intoPlax 45 and Dentosan S, respectively, have not been consid-ered active ingredients because fluorides were reported to beineffective as antifungal agents.29 Furthermore, our prelimi-nary experiments have demonstrated that NaF and MFP, atthe concentrations at which they are incorporated into Plax45 and Dentosan S (0.025 and 0.18 g/100 ml, respectively),have no antifungal activity against any of the fungal strains

API 20C, bioMérieux, Marcy-1 'Etoile, France.5BBL Mycotube, Becton Dickinson Microbiology Systems, Cockeys-ville, MD.

Table 1. Mouthrinse Products and Their Active IngredientsCorsodyl*Dentosan*Dentosan S1NeoCepacol1Oraseptic*Plak out5Plax 45«

Chlorhexidine 0.2%Chlorhexidine 0.2%Sanguinarine 3%Cetylpyridinium chloride 0.05%Hexetidine 0.1%Chlorhexidine 0.12%Triclosan 0.045%

*SmithKline Beecham Farmaceutici S.p.A., Baranzate di Bollate, Italy.'Warner Wellcome, Consumer Health Products S.Com.p.A., Pomezia,Italy.*Gruppo Lepetit S.p.A., Milano, Italy.5Byk Gulden Italia S.p.A., Cormano, Italy."Colgate Palmolive S.p.A., Roma, Italy.

tested in this study. All mouthrinses, available on the Italianmarket, were obtained commercially.

Susceptibility Testing ProcedureThe susceptibility of yeast isolates to antimicrobial mouthrin-ses was determined by a broth macrodilution method.30 Sterileglass tubes (13 X 100 mm) and Sabouraud dextrose broth(SDB)11 were used for the tests.

Yeast isolates, grown for 24 hours on Sabouraud dextroseagar (SDA)I! supplemented with chloramphenicol (1 mg/ml),were subcultured in SDB medium for 24 hours at 37°C. Theturbidity of each yeast suspension was adjusted to the 0.5McFarland standard by the addition of SDB medium to givea final inoculum of 1 X 106 to 2 X 106 colony forming unitsper milliliter (CFU/ml). Confirmation of the final inoculumsize was determined for all organisms by subcultures on SDAplates.

Mouthrinses were serially diluted twofold in SDB mediumto obtain test broth. All dilutions, ranging from 1:2 to 1:1,024,were performed immediately prior to testing. The range offinal concentrations in test broth for each active ingredientwas as follows: CPC: 250 to 0.48 µg/ml; CHX: 600 to1.17 µg/ml for Plak out, and 1,000 to 1.95 µg/ml forCorsodyl and Dentosan, respectively; HEX: 500 to 0.97µg/ml; SNG: 15 to 0.029 mg/ml; TRN: 225 to 0.43µg/ml. To adjust CHX-containing mouthrinses for initialconcentration differences, Corsodyl, Dentosan, and Plakout were also appropriately diluted in SDB medium to

yield a 280 µg/ml start solution which was further dilutedas described above. Final concentrations of CHX rangedfrom 280 to 0.27 µg/ml.

Each test tube was inoculated with the test broth (4 ml)and the yeast inoculum (0.1 ml). Control tubes were in-oculated with either SDB medium alone or yeast inocu-lum in mouthrinse-free medium and used as contamina-tion and growth controls, respectively. Test and controltubes were incubated aerobically at 37°C for 24 hours.

Because of the turbidity that occurred in test brothwhen some mouthrinses were diluted in SDB, it was not

Difco Laboratories, Detroit, MI.

Volume 68Number 8 GIULIANA, PIZZO, MILICI, MUSOTTO, GIANGRECO 731

Table 2.Agents*

Minimum Fungicidal Concentrations ^g/ml) of Mouthrinses Containing Antimicrobial

MouthrinsesActive Ingredients1Yeasts

Corsodyl Dentosan Dentosan S NeoCepacol* Oraseptic Plak out Plax 45CHX CHX SNG CPC HEX CHX TRN2000 2000 30000 500 WOO 1200 450

Candida albicans 31.25 31.25 15000 7.81 31.25 37.5 14.06Candida parapsilosis 62.25 250 >15000 1.95 15.62 150 14.06Candida krusei 7.81 7.81 15000 1.95 7.81 9.37 14.06Candida guilliermondii 125 125 >15000 1.95 7.81 150 14.06Torulopsis glabrata 7.81 7.81 15000 7.81 31.25 9.37 28.12Saccharomyces cerevisiae 3.9 3.9 7500 1.95 7.81 4.68 14.06

*MFC values are the medians of triplicate determinations. A greater-than symbol (>) indicates that an MFCwas not achieved..'Italicized values indicate the concentration ^g/ml) of active ingredients.*The MFC values of NeoCepacol were significantly different from those achieved with CHX-containingmouthrinses, Oraseptic, and Plax 45 (P < 0.01). Dentosan S, which had incomplete data, was omitted fromthe statistical analysis.

Table 3. Minimum Fungicidal Concentrations ^g/ml) of CHX-Containing Mouthrinses*

Mouthrinses Corsodyl* Dentosan' Plak out'CHX Concentration ^g/ml) 2000 2000 1200

Yeasts

Candida albicans 35 35 35Candida parapsilosis 70 280 140Candida krusei 8.75 8.75 8.75Candida guilliermondii 140 140 140Torulopsis glabrata 8.75 8.75 8.75Saccharomyces cerevisiae 4.37 4.37 4.37

*Final concentrations of CHX ranged from 280 to 0.27 µg/ml. MFCvalues are the medians of triplicate determinations.'No significant differences in MFC values were found among the threeCHX-containing products (P = 0.1).

possible to determine the minimum inhibitory concentra-tion. Therefore, the antifungal activity of mouthrinses was

assessed by means of the minimum fungicidal concentra-tion (MFC). MFC was determined by subculturing 20 µ aliquots from each tube in the broth macrodilution seriesonto SDA plates. The plates were incubated for a further48 hours under the same conditions. MFC was defined as

the lowest concentration of each mouthrinse that com-

pletely inhibited the growth of the strain under test. Twen-ty µ aliquots from control tubes were plated onto SDAplates to check the medium sterility and viability of thetest organisms. All antifungal susceptibility experimentswere performed in triplicate.Kill-Time DeterminationYeast inoculum (0.1 ml) was added to a series of 12 testtubes containing 2.5 ml of each mouthrinse diluted in 2.5ml of sterile saline. Control tubes were inoculated withyeast inoculum (0.1 ml) and saline (5 ml) only. Each testtube was vortexed at the medium setting and 20 µ ali-quots were removed at 15-second intervals for up to 180seconds, and immediately subcultured onto SDA plates.Twenty µ aliquots from control tubes vortexed for 180

seconds were subcultured as above to exclude a possiblekilling effect of vortexing. Test and control plates were

incubated aerobically at 37°C for 48 hours. Kill-time wasdefined as the shortest interval that yielded no growth or

one discrete colony. All experiments were performed intriplicate.

Statistical AnalysisThe differences in MFC and kill-time data were tested forsignificance using Friedman's two-way analysis of vari-ance. A multiple comparisons test was used to distinguishwhich mouthrinse differed significantly.31 The signifi-cance level was set at < 0.01. Mouthrinses which hadincomplete data for either MFC or kill-time were omittedfrom the statistical analysis.

RESULTSMFC and kill-time results are expressed as medians oftriplicate determinations. MFCs of the mouthrinses againstthe six species of yeasts are shown in Table 2. The resultsof the MFC determination showed that mouthrinses wereeffective against test organisms. The concentration of ac-

tive ingredients in these products, in fact, was in excess

of the corresponding MFC. However, an MFC was notachieved with Dentosan S against C. parapsilosis and C.guilliermondii. Friedman's analysis of variance, followedby multiple comparisons, indicated that MFC values ofNeoCepacol were significantly lower than those achievedwith all the other mouthrinses (P < 0.001). Furthermore,no significant differences were found in MFC values ofCHX mouthrinse products, once adjusted for initial con-

centration differences of CHX (Table 3).Table 4 shows the kill-times achieved with different

mouthrinses when used at half the concentration of thecommercial formulations. All mouthrinses, except Den-tosan S, showed rapid kill-times with S. cerevisiae (15 to30 seconds), but, in general, longer contact periods were

required for the other organisms tested. CHX-containing

732 ANTIFUNGAL PROPERTIES OF MOUTHRINSESJ PeriodontolAugust 1997

Table 4. Kill-Times (sec) Achieved With Mouthrinses Containing Antimicrobial Agents*Yeasts Corsodyl* Dentosan1 Dentosan S NeoCepacol* Oraseptic Plak out* Plax 45

Candida albicans 60 30 >180 .105 >180 60 >180Candida parapsilosis 105 60 >180 150 >180 150 >180Candida krusei 105 75 >180 30 165 120 45Candida guilliermondii 120 120 >180 120 >180 120 >180Torulopsis glabrata 135 180 >180 60 75 135 45Saccharomyces cerevisiae 15 15 > 180 30 15 15 15

*Kill-time values are the medians of triplicate determinations at half the concentration of commercial formulations. A greater-than symbol (>) indicatesthat a kill-time was not achieved.'No significant differences were found in kill-time values among the CHX- and CPC-containing products (P = 0.18). Mouthrinses which had incompletedata were omitted from the statistical analysis.

mouthrinses and NeoCepacol showed lethal effect on allthe species of yeasts within 180 seconds. On the otherhand, Plax 45 and Oraseptic did not show fungicidal ac-

tivity against C. albicans, C. parapsilosis, and C. guil-liermondii. No kill-times were achieved with Dentosan S.Differences in kill-time values among mouthrinses in-cluded in statistical analysis were not significant (P =

0.18). Although there was a trend towards a more rapidfungicidal activity of Dentosan, no significant kill-timedifferences were found among the CHX-containingmouthrinses (P = 0.04).

DISCUSSIONYeasts may induce oral opportunistic infections in im-munocompromised patients. Candida spp., particularlyCandida albicans, are the most frequently reported etio-logical agents.19-32 Oral candidosis has increased duringthe last decade and constitutes a major clinical problembecause oral mucosa and periodontal tissue represent an

important port of entry for serious systemic fungal infec-tions.16-33-34 Furthermore, it has been suggested that sub-gingival occurrence of yeasts may contribute to periodon-tal disease progression.4-71314

This study was the first to our knowledge to evaluatea number of antimicrobial mouthrinses for in vitro anti-fungal properties. MFC and kill-time were determinedagainst six species of yeasts chosen as representative ofthe oral fungal flora in health and disease.-3·5·12·32 Themouthrinses evaluated, with the exception of Dentosan S,were effective against all the species under test at muchlower concentrations than those present in commercialformulations. However, NeoCepacol has shown to be themost effective against yeasts. Kill-time determinationsshowed that only CHX and CPC mouthrinses were fun-gicidal against all of the test organisms within 180 sec-onds.

Among antimicrobials incorporated into mouthrinses,only CHX has received considerable study as an antifun-gal agent. Little is known about the mechanism of actionagainst yeasts. It has been suggested that the fungicidalactivity of CHX is due to its penetration of the region ofthe glycerol moieties, with subsequent breakdown of thepermeability barrier and leaking of cytoplasmatic con-

tents.35 In vitro studies have provided evidence of a re-markable activity against Candida spp.,26-36 but resultsfrom clinical studies have been equivocal. CHX rinsinghas been reported to significantly reduce the incidence oforal candidosis in patients undergoing chemoradiothera-py 22,27 However, a number of clinical trials did not sup-port its effectiveness in eradicating oral yeasts and pre-venting systemic fungal infections of oral origin.23"26 Thediscrepancy between good in vitro and poor in vivo ac-

tivity of CHX may be explained by several factors: in-activation by food compounds or saliva, interaction withdrugs topically administered, short rinsing time, dilutionby oral fluids, or emergence of resistant strains. Salivaand nystatin have been reported to significantly inhibit thein vitro antifungal activity of CHX.37-38 An antagonisticinteraction may also occur with excipients and additivesincorporated into mouthrinse formulations, but we foundthat the different CHX-containing mouthrinse productsexhibited similar antifungal properties. This suggests thatthe different ingredients incorporated into Corsodyl, Den-tosan, and Plak out did not affect antifungal activity ofCHX. The emergence of fungal strains resistant to CHXhave not been observed in clinical studies.26 However,according to the criteria of Slots et al.,39 some of the testedstrains could be considered resistant, because MFCs were

higher than 70 µg/ml. Therefore, CHX-containing mouth-rinses would be used after susceptibility testing of yeastisolates.

In conclusion, our results indicate that antimicrobialmouthrinses included in this study possess antifungalproperties. NeoCepacol was more effective than the otherproducts, while Dentosan S showed poor antifungal ac-

tivity. NeoCepacol and CHX-containing mouthrinses,when used at half concentration, showed lethal effectwithin 180 seconds. These findings suggest that mouth-rinses containing antimicrobial agents might represent an

appropriate alternative to conventional antifungal drugs inthe management of oral candidosis. Furthermore, a localdelivery system containing CPC could be useful in thetreatment of periodontal diseases associated with subgin-gival occurrence of yeasts. However, the effectiveness ofantimicrobial mouthrinses as antifungal agents needs tobe evaluated in clinical trials.

Volume 68Number 8 GIULIANA, PIZZO, MILICI, MUSOTTO, GIANGRECO 733

AcknowledgmentsThe authors would like to thank Prof. Matteo D'Angelo(Department of Periodontology, School of Dentistry, Uni-versity of Palermo, Italy) for his valuable critical advicein preparation of the manuscript.

REFERENCES1. Ciancio SG. Chemical agents: Plaque control, calculus reduction and

treatment of dentinal hypersensitivity. Periodontal 2000 1995;8:75-86.

2. Walker CB. Microbiological effects of mouthrinses containing an-

timicrobials. J Clin Periodontol 1988;15:499-505.3. Arendorf TM, Walker DM. The prevalence and intra-oral distribu-

tion of Candida albicans in man. Arch Oral Biol 1980;25:1-10.4. Slots J, Rams TE, Listgarten MA. Yeasts, e.nteric rods and pseudo-

monads in the subgingival flora of severe adult Periodontitis. OralMicrobiol Immunol 1988;3:47-52.

5. Rams TE, Slots J. Candida biotypes in human adult Periodontitis.Oral Microbiol Immunol 1991;6:191-192.

6. Danién G, Wikström M. Occurrence of enteric rods, staphylococciand Candida in subgingival samples. Oral Microbiol Immunol 1995;10:42-46.

7. Slots J, Feik D, Rams TE. Age and sex relationships of superin-fecting microorganisms in Periodontitis patients. Oral Microbiol Im-munol 1990;5:305-308.

8. Listgarten M, Lai C, Young V. Microbial composition and patternof antibiotic resistance in subgingival microbial samples from pa-tients with refractory Periodontitis. J Periodontol 1993;64:155-161.

9. Olsvik B, Hansen BF, Tenover FC, Olsen I. Tetracycline-resistantmicroorganisms recovered from patients with refractory periodontaldisease. J Clin Periodontol 1995;22:391-396.

10. Alcoforado GAP, Rams TE, Feik D, Slots J. Microbial aspects offailing osseointegrated dental implants in humans. J Parodontol1991;10:11-18.

11. Rams TE, Babalola OO, Slots J. Subgingival occurrence of entericrods, yeasts and staphylococci after systemic doxycycline therapy.Oral Microbiol Immunol 1990;5:166-168.

12. Helovuo H, Hakkarainen K, Paunio K. Changes in the prevalenceof subgingival enteric rods, staphylococci and yeasts after treatmentwith penicillin and erythromycin. Oral Microbiol Immunol 1993;8:75-79.

13. DeWitt G, Cobb C, Killoy W. The acute periodontal abscess: Mi-crobial penetration of the soft tissue wall. Int J Periodontics Re-storative Dent 1985;5:39-51.

14. Gonzales S, Lobos I, Guajardo A, et al. Yeasts in juvenile Perio-dontitis. Preliminary observations by scanning electron microscopy.J Periodontol 1987;58:119-124.

15. Peterson DE, Minan GE, Overholser CD, et al. Microbiology ofacute periodontal infection in myelosuppressed cancer patients. JClin Oncol 1987;5:1461-1468.

16. Odden K, Schenck , Koppang HS, Hurlen . Candidai infectionof the gingiva in HIV-infected persons. J Oral Pathol Med 1994;23:178-183.

17. Grbic JT, Mitchell-Lewis DA, Fine JB, et al. The relationship ofcandidiasis to linear gingival erythema in HIV-infected homosexualmen and parenteral drug users. J Periodontol 1995;66:30-37.

18. Tomar SL, Swango PA, Kleinman DV, Burt BA. Loss of periodontalattachment in HIV-seropositive military personnel. J Periodontol1995;66:421-428.

19. Nikoskelainen J. Oral infections related to radiation and immuno-suppressive therapy. J Clin Periodontol 1990;17:504-507.

20. King GN, Healy CM, Glover MT, et al. Prevalence and risk factorsassociated with leukoplakia, hairy leukoplakia, erythematous can-

didiasis, and gingival hyperplasia in renal transplant recipients. OralSurg Oral Med Oral Pathol 1994;78:18-26.

21. Henderson DK. Defining risks for nosocomial candidemia. Arch In-tern Med 1989;149:2172-2173.

22. Ferretti GA, Ash RC, Brown AT, Parr MD, Romond EH, Lillich TT.Control of oral mucositis and candidiasis in marrow transplantation:A perspective, double blind trial of Chlorhexidine digluconate oralrinse. Bone Marrow Transplant 1988;3:483-493.

23. Weisdorf DJ, Bostrom , Raether D, et al. Oropharyngeal mucositiscomplicating bone marrow transplantation: Prognostic factors andthe effect of Chlorhexidine mouth rinse. Bone Marrow Transplant1989;4:89-95.

24. Spijkervet FKL, van Saene HKF, Panders AK, Vermey A, van SaeneJJM, Fidler V. Effect of Chlorhexidine rinsing on the oropharyngealecology in patients with head and neck cancer who have irradiationmucositis. Oral Surg Oral Med Oral Pathol 1989;67:154-161.

25. Wahlin YB. Effects of Chlorhexidine mouthrinse on oral health inpatients with acute leukemia. Oral Surg Oral Med Oral Pathol1989;68:279-287.

26. Thurmond JM, Brown AT, Sims RE, et al. Oral Candida albicansin bone marrow transplant patients given Chlorhexidine rinses: Oc-currence and susceptibilities to the agent. Oral Surg Oral Med OralPathol 1991;72:291-295.

27. Epstein JB, Vickars L, Spinelli J, Reece D. Efficacy of Chlorhexidineand nystatin rinses in prevention of oral complications in leukemiaand bone marrow transplantation. Oral Surg Oral Med Oral Pathol1992;73:682-689.

28. Baron EJ, Peterson LR, Finegold SM. Bailey & Scott's DiagnosticMicrobiology. St. Louis: Mosby-Year Book; 1994:44.

29. Laine R Meurman JH, Murtomaa H, et al. One-year trial of the effectof rinsing with an amine fluoride-stannous-fluoride-containingmouthwash on gingival index scores and salivary microbial countsin lymphoma patients receiving cytostatic drugs. / Clin Periodontol1993;20:628-634.

30. Baron EJ, Peterson LR, Finegold SM. Bailey & Scott's DiagnosticMicrobiology. St. Louis: Mosby-Year Book; 1994:14.

31. Siegel S, Castellan NJ. Statistica Non Parametrica. Milano, Italy:McGraw-Hill; 1992:232-243.

32. Barchiesi F, Morbiducci V, Ancarani F, Scalise G. Emergence oforopharyngeal candidiasis caused by non-albicans species of Can-dida in HIV-infected patients. Eur J Epidemiol 1993;9:455-456.

33. Heimdahl A, Mattsson T, Dahllöf G, Lönnquist , Ringden . Theoral cavity as a port of entry for local and systemic early infectionsin patients treated with bone marrow transplantation. Oral Surg OralMed Oral Pathol 1989;68:711-716.

34. Working Party of the British Society for Antimicrobial Chemother-apy. Chemoprophylaxis for candidosis and aspergillosis in neutro-

penia and transplantation: A review and recommendations. J Anti-microb Chemother 1993;32:5-21.

35. Elferink JGR, Booij HL. Interaction of Chlorhexidine with yeastcells. Biochem Pharmac 1974;23:1413-1419.

36. Hiom SJ, Furr JR, Russell AD, Dickinson JR. Effects of Chlorhex-idine diacetate on Candida albicans, C. glabrata and Saccharomy-ces cerevisiae. J Appi Bacteriol 1992;72:35-40.

37. Spijkervet FKL, van Saene JJM, van Saene HKF, Panders AK, Ver-mey A, Fidler V. Chlorhexidine inactivation by saliva. Oral SurgOral Med Oral Pathol 1990;69:444-449.

38. Barkvoll , Attramadal A. Effect of nystatin and Chlorhexidine dig-luconate on Candida albicans. Oral Surg Oral Med Oral Pathol1989;67:279-281.

39. Slots J, Rams TE, Schonfeld SE. In vitro activity of Chlorhexidineagainst enteric rods, pseudomonads and acinetobacter from humanPeriodontitis. Oral Microbiol Immunol 1991;6:62-64.

Send reprint requests to: Dr. Giovanna Giuliana, Via Vodige 6, 90144Palermo, Italy.

Accepted for publication January 6, 1997.