Possible effect of medically administered antibiotics on ...

Oral Microbioi Itnmunol 1989: 4: 77-81

Possible effect of medicallyadministered antibiotics on themutans streptococci: implicationsfor reduction in decay

Walter J. Loesche\ StephenA. Eklund^ David F. i\/lehlisch^Brian Burt^'Department of Biologic and MaterialsSciences, ^Program in Dental Public Health,University of iVlichigan, Ann Arbor,'Department of Community Dentistry andHuman Behavior, Southern Illinois University,U.S.A.

Loesehe WJ, Eklund SA, Mehlisch DE, Burt B. Po.ssible effeet of medically adtninis-tered antibiotics on the mutans streptococci: implications for reduction in decay.Oral Microbioi Immunol 1989: 4: 77-81.

The decline in dental caries in children in North America, Scandinavia, Britain,Ireland and tnany Commonwealth nations is well docutnented. The tnultiple usesof fluoride can account for tnost, but not all, of this reduction. In this investigation,data are provided which suggest a relationship between antibiotic usage formedical purposes and a declitie in both mutans streptococci (MS) and caties.Children attending Grades 1 and 2 in the Coldwater, Michigan school systemand who reportedly never received antibiotics had significantly higher pro-portions of MS in the fissure plaques of first molars than subjects who receivedantibiotics. The level of decay in the pritnary dentition was invetsely related tothe reported usage of antibiotics. The frequent usage of antibiotics could reducethe incidence of dental caries by delaying the colonization of the teeth by the MS.This was evaluated by a prospective study in infants to detennine what effectreported antibiotic usage would have on the colonization of newly erupting primaryteeth. Otily 2 of 10 infants cultured at 2 to 3 week intervals for periods up to 1year after tooth eruptioti became colonized by the MS. One had never receivedantibiotics and the second had been on antibiotics for a single 5-day period.Seven of the 8 non-colonized infants had received antibiotic therapy for periodsranging frotn 10 to 181 days. Both the Coldwater study and the prospective studyof infants suggested a relation between frequency of antibiotic usage for medicalpurposes and the MS levels on the teeth.

Key words: antibiotics: mutans streptococci.

W. J. Loesehe University of Michigan Schoolof Dentistry, Ann Arbor, MI48109, USA.

Accepted for publication September 8, 1988

The decline in dental caries in childrenresiding in North Atnerica, Scandina-via, Britain, Ireland and the variousCommonwealth nations is well docu-mented (10). Evidence points to waterfluoridation and fluoridated dentifricesas the tnajor factors tesponsible for thisreduction, but lluoride may not accountfor all the decline (3). The reduction hasoccurred in countries like Sweden whetethere is no water fiuoridation (18), andin non-fluoridated areas in the U.S. (6).Indeed, a reduction of decay in the pri-mary dentition is being observed in .la-pan (Matsukudo, pers. cotntn.) whetethere is neither water fluoridation norwidespread usage of fiuoride dentifrices.What then besides fluorides could becontributing to this decrease in dentalearies? There is no reason to suspectthat oral hygiene in the absence of flu-oridated dentifrices has played a role.

as even the best otal hygiene habits havelittle effect on the incidetice of decay(2, 13, 15), There is no evidence thata decreased consumptioti of sweetenedfoods or a decreased frequency ofsnacking has contributed to this decline(7) and a genetic change in the host inuntenable.

Since Keyes (16) demonstrated thetransmissible nature of dental decay inanimal tnodels, numerous investigatorshave ptovided evidence that the mutansstreptococci (MS), of which Streptococ-cus mutans is the predominant humantype, ate statistically associated with hu-man dental decay (21). If the MS arehutnan cariogens, then the decline indecay could reflect a decrease in theirlevels, or a delay in the titning of an MSinfection in childten (22). The salivatylevels of the MS and lactobacilli (alsoimplicated in hutnan decay) had de-

creased significantly in 9 to 12 year oldSwedish school children between 1972atid 1984, and this decrease was coittci-dent with a teduction in the tnean de-cayed and filled surface scores ftom 12.8to 3.0 (17). Likewise, the salivary levelsof U.S. children attenditig a pediatricdental clinic had decreased by about90% between 1973-74 and 1981-82(22).

Fluoridation does not sectn to lowerMS levels (4, 30). However, antibioticsdo temporarily reduce the salivaty MSlevels (25), and during the titne spanthat decay has been declining the tnedi-cal usage of antibiotics, has iticreasedsubstantially (3). This taises the possi-bility that some part of the decline inboth MS and in decay could be second-ary to the medical usage of antibiotics(3, 22). The possiblity that antibioticusage could reduce decay has been sug-

78 Loesehe et al.

gested by studies of caries prevalence inchildren with rheumatic fever (14, 20)and by clinical observations (5).

There has been a marked decline inmortality due to rheumatic fever in theU.S. (12, 26). A retrospective study hasshown that while the decline in occur-rence of rheumatic fever antedated theavailability of antibiotics, the rates ofdecline in national mortality due torheumatic cardilis accelerated fourfoldwith the advent of antibiotics (26). Asthe Group A streptocci that are impli-cated in rheumatic fever and the MS aresensitive to the same antibiotics, thisdata on rheumatic fever is supportive ofthe hypothesis that frequent antibioticusage for medical purposes may alsohave coincidently, contributed to the de-cline of the MS and indirectly to dentaldecay. This possibility led us to re-exam-ine our data files on S. mutatis (22-24),to find situations in which informationon antibiotic usage had been obtainedin order to determine whether a re-lationship between antibiotic usage anda decline in MS and caries could bedemonstrated.

Material and methods

Coldwater. All children attendingGrades 1 and 2 in the Coldwater Michi-gan school system (age 6 to 7 years)were given information concerning aprospective clinical study seeking toshow a relationship between MS colon-ization and dental decay (8). A ques-tionnaire on antibiotic history was dis-tributed to the parents of the 411children enrolled in the study andanswers were received from 390 sub-jects, a 94% response rate. The anti-biotic questions related to the parentsrecall as to whether their children usedantibiotics on an average 1) 3 or moretimes per year; 2) once or twice per year;3) less than once per year; or 4) never.The responses from these 390 subjectswere tabulated and became the indepen-dent variable upon which a) the initial

caries score in the primary teeth and b)the proportions of selected bacteria inplaque samples taken from the occlusalfissures were stratified.

All permanent and primary teethwere examined for dental decay by thesame examiner. Caries was recorded bythe decayed-filled surfaces (dfs) index inprimary teeth. Caries was diagnosed onall coronal tooth surfaces only whensoftness or a definite break in the conti-nuity of the enamel surface could bedetected with an explorer. No bite-wingradiographs were taken.

Plaque samples for microbiologicalassay were taken from the occlusal sur-faces of the lower first molars (2 samplesper patient) with a No. 26-gauge sterileneedle held with a hcmostat. Each need-le which contained from 10-* to 107 co-lony-forming-units (CFU) of bacteriawas dropped into a reduced transportfluid (RTF) (23) and transported backto the laboratory. After dispersion, ser-ial dilution in RTF and plating, theplaque samples were incubated anaero-bically under an atmosphere of 85%N2, 10% Hj and 5% CO^ within 3 to 5h after their collection. The media usedand the procedures for counting thecolonies of MS, Streptococcus sanguis,lactobacilli, VeiUonella and Actinomyceshave been described previously (24). Aratio of the MS to S. sanguis was calcu-lated for each plaque sample (23). Inplaques where S. .sanguis outnumberedthe MS, the reciprocal value was calcu-lated and reported as a negative num-ber. The lower first molars weresampled as the colonization of their fis-sures within 6 months after their erup-tion was thought to reflect the "load"of MS that was present on the primaryteeth.

Infants. The study population consistedof ten infants who were edentulous atthe start of the observation period andwho lived in households where at leasttwo of the family members had levels ofMS at or above 1 x lO'' CFU per ml of

saliva. Bacterial samples of the edentul-ous infants were taken at approximately14-day intervals by swabbing the oralmucosa overlying the mandibular andmaxillary ridges with a sterile cotton-tipped applicator. This applicator wasrubbed across the sterile surface of anMM 10-10% sucrose-bacitracin (3 //g/ml)-agar (MM lO-SB agar) (23). As sooiias 2 or more teeth had fully erupted ineach infant the swabbing was discon-tinued in favor of sampling each toothsurface with unwaxed dental fioss se-cured in a sterile floss holder. That por-tion of the fioss containing the plaquewas removed from the holder by cuttingthe floss with sterile scissors and placingit in the RTF.

All saliva (household members) andplaque samples were processed withinone hour after their collection. Theywere sonified for 1000 watt/s, seriallydiluted in RTF and plated on theMMiO-SB agar and MSB agar plates(11), using a spiral plating technique(23). The sampling of the children wasrepeated at 2 to 3 week intervals over a5 to 17-month period of observation.Each infant was monitored for usageof antibiotics for medical purposes overthis time interval. This information wasobtained from the mother when con-tacted by telephone to schedule a sam-pling appointment or when questionedabout the infants health at the time ofsampling. These antibiotics were mainlyoral penicillins prescribed for ear infec-tions, "strep" throat, or bronchial infec-tions.

Results

The survey of self-reported antibioticusage in the Coldwater children showedthat only 7% of the subjects had neverreceived antibiotics; 49% received themless than one occasion per year; 32%received them one to two times per yearand I2"/D received them on an average3 or more times per year. It was not

Tabte I. Mean percentage of scleeted plaque species in oeelusal fissure plaque as a function of reported usage of antibiotics - Coldwater

ReportedUsage>3 times/yr1-2 times/yr< 1 time/yrNever

No. ofTeeth

8221832546

MS"

12.9'' + 21.413.8+18.513.1 + 18.419.7 + 24.8''

5". sanguis

10.4+17.48.1 +8.49.0 + 8.48.4 + 7.8

Percentage of Cultivable Count + Standard DeviationLactobacilli

1.4±2.61.3±4.11.7±7.20.9 + 2.1

VeiUonella

5.9 + 8.05.0 + 9.25.5 + 8.25.2 ±6.4

Aetinoniyees sp.

8.6+13.06.4 + 9.16.8+10.07.6±I2.9

MS/5', sangui.s'

-7.5 + 29.1-2.4 + 27.1-5.7 + 29.1

2.2''±29.3''a) mutans streptococci.b) value is significantly different; p = 0.03 Scheffe contrast of never vs all other categories.c) negative values are the reciprocal of MS/S. .satiguis ratio, when S. satiguis is greater than the MS.

Antibiotics and tnutans streptococci 79

possible to identify the types of anti-biotics used.

The teeth in the Coldwater subjectswho never received antibiotics had sig-nificantly higher proportions of MS inthe fissure plaques of newly-eruptingfirst molars than did the teeth in thesubjects who received antibiotics (p =0.03, Scheffe contrast (Table 1). Withinthe antibiotic users there was no obvi-ous effect of frequency of usage uponthe percentage of MS. The reduction inMS was exceptional among the plaqueflora that were monitored, as no anti-biotic related changes in the proportionsof S. sanguis, lactobacilli, VeiUonella, oran unspeciated Actinomyces species wasfound (Table I). When the MS to 5.sanguis ratio was calculated (23), theMS outnumbered 5. sanguis only in thechildren who had never received anti-biotics (Table I). In the other groupsthe MS to S. sanguis ratio was negative,indicating the relative dominance of S.sanguis in the fissure plaques.

The level of decay in the primary den-tition was inversely related to the re-ported usage of antibiotics. Childrenwho had never received antibiotics hadsignificantly more decay than childrenwho had received antibiotics 3 or moretimes per year (Table 2). Children whohad received antibiotics 2 or less timesper year had levels of decay that wereintermediate to the "never" and "3 ormore" group. All antibiotic groups andthe never group had comparable num-bers of primary teeth (13.5 to 14.2),(Table 2), so that missing teeth did notseem to account for the observed differ-ences.

Healthy infants living in householdsin which two or more family membershad salivary MS levels higher than lC/ml were followed for periods rangingfrom 6 to 16 months after their teetherupted. Only 2 children became detec-tably colonized by MS. One had never

Table S. Relation between antibiotie therapy in infants and deteetion of mutans streptoeocei(MS) on primary teeth

Household

F 2F 12F 5F 4F 8F 11F 9F 7F 3F 10

MSDeteeted

yes (16)"yes (8)no (16)no (12)no (6)no (6)no (5)no (8)no (12)no (2)

Antibiotictherapy

noyesnoyesyesyesyesyesyesyes

Days onantibiotic

050

1310131984

18190

Monthdentate

119

1614676

10126

% of dentate limeon antibioties

0.01.70.03.04.95.09.4

29.148.864.1

a) number of teeth erupted at eotielusion of study.

received antibiotics, whereas, the sec-ond had been on antibiotics for a single5-day period (Table 3).MS were not de-tected on any teeth in the other 8children even though sampling con-tinued on a 2-3 week basis over a 6to 16 month interval. Seven of these 8children had received antibiotie therapymainly with penicillin type medicationfor periods ranging from 10 to 181 days.

The number of days the teeth wereexposed to antibiotics was divided bythe number of days the teeth were in thetnouth. The resultant values indicatedthat teeth not colonized by MS wereexposed to antibiotics for periods rang-ing from 3% to 65% of their existencein the tnouth (Table 3). The proportionof dentate titne that the infants weretepottedly on antibiotics was ranked ac-cording to whether or not their teethbecame infected with MS. This rankingjust tnissed statistical significance (Kru-skal-Wallis, p = 0.06).

Discussion

The results deseribed were obtainedretrospectively from studies designedfor other purposes, and are deficient inmany respects, i.e., the type of anti-biotics used, were the subjects, es-

Table 2. Mean number oT deeayedand filled primary teeth as a funetion of reported level ofantibiotie use at baseline - Coldwater

ReportedAntibiotieUse

Never< I/year1 to 2/yearS: 3/year

No. ofchildren

2519212251

Mean No. ofPrimary Teeth

13.5 + 3.4'14.0 + 2.814.2±3.014.2 + 3.1

dt"

1.00.80.70.6

ft

1.81.51.71.0

Mean

d + ft"

2.92.32.41.5

p value''

0.050.080.06

a) d = decayed primary teeth; 1 = filled teeth.b) p value obtained for comparison of 3 or more per year group with each of the other

groups, Scheffe contrast.e) mean + standard deviation.

pecially the infants, on antibiotics dur-ing satnpling? Thus the findings shouldbe interpreted with caution. Both thecross-sectional Coldwater study and theprospective study of infants suggest arelationship between frequency of anti-biotic usage and MS levels on the teeth.The Coldwater study further suggestedthat increased antibiotic usage was as-sociated with a teduced incidence of de-cay in the pritnary dentition.

The infortnation on antibiotic usagein both studies was obtained by inquiryof the parents which could be con-sidered as potentially unreliable, es-pecially in the Coldwater study wherelongtenn recall was involved. Onemight expect in the Coldwater studythat the identification of those childrenwho either tiever used antibiotics, orused them 3 or tnore times per year, tobe the tnost reliable data obtained frotnthe questionnaire, as the parents wouldbe likely to recall this infortnation accu-rately. It was the differences betweenthese "never" and "frequent" antibioticusers that were statistically significantboth in tenns of propot tions of the MSin fissure plaque (Table 1) and df scores(Table 2).

Maltz and Zickert (25) have shownthat a single 10-day episode of penicillinsignificantly redueed the salivary levelsof MS for several weeks, but had a mini-mal effect on 5. sanguis and lactobacilli.A similar reduction in plaque MS pro-portions without a reduction in S. .satt-guis and lactobacilli was found in theantibiotic users relative to the non-anti-biotic users in the Coldwater popula-tion. This bacteriological sitnilarity be-tween the 2 studies suggests that thequestionnaire used in the present studytnay have reliably identified the non-antibiotic users.

An association between frequentantibiotic usage and a decrease in dental

80 Loesehe et al.

caries has been noted previously in pa-tients with exceptional medical histor-ies. Two retrospective epidemiologicstudies showed that young rheumaticfever patients on a daily prophylacticpenicillin regimen had significant per-centage reductions in caries levels whencompared to untreated controls (14, 20).Children with immunodeficiencies whowere extensively, but not continually,treated with antibiotics had a meanDMF tooth score of 2.3, whereas immu-nodeficient children who averaged only5 days on atibiotics per year had a DM Ftooth score of 5.4 (28). A study of thecaries experience of children with tetra-cyeline discolored teeth showed a mod-est, but significantly lower DMFSscores for the complete dentition, de-ciduous molars and first permanent mo-lars in comparison with paired eontroldentitions in children without anydiscoloration (5).

These findings suggest that very fre-quent usage of antibiotics in medicallycompromised children can reduee theincidence of dental decay. This can beenvisioned as due to the more or lesscontinued suppression of the MS in theplaque. However, in medically healthychildren such as our Coldwater subjects,in order for the medieal usage of anti-bioties to be coincidentally loweringdental decay, it is necessary to posit thatthe temporary suppression of MS in theplaques occurs during a stage or a per-iod in the tooth maturation, in whichthe tooth is most vulnerable to a cariog-enic attach by the MS. This would bethe period shortly after tooth eruptionand would be dependent upon the MSbeing among the initial colonizers of thefissure depths (22).

If during the period of tooth erup-tion, the child is reeeiving or has justreceived antibiotics that have activityagainst the MS, there is the possibilitythat the salivary levels of the MS willbe so low that the fissure will becomecolonized by bacteria other than the MS(29). There is data that shows if the MScolonization of the dentition is delayedor does not occur, then the caries experi-ence in the primary dentition is quitelow or absent, whereas if colonizationoccurs shortly after eruption, decay iscommon. 1 hus, when the MS were notdetected in 2-year-old children, thesechildren were caries free at age 4, where-as children with detectable MS at age 2had a df score of over 10 at age 4(1).In another investigation, only 1 of 36children without detectable MS coloniz-

ation in their first 4 years of life haddecay, whereas 75% of the children in-fected by age 15 months had decay byage 4 (19). If primary incisors becomecolonized with MS shortly after erup-tion, they soon become carious (27).

These findings, combined with the re-sults from our infant study, provide aconnection between antibiotic usageand reduced colonization of the MS inthe primary dentition. In 7 of these in-fants frequent usage of antibiotics andabsence of MS were apparently linkedevents. If the antibiotic usage in these10 infants is representative of its usageby other children, such as those livingin Coldwater, then the antibiotic associ-ated reduction in decay and MS pro-portions noticed in the Coldwaterchildren could reflect absent or delayedcolonization by the MS when the teetherupted and were most prone to a cari-ogenic challenge. This colonization-de-layed, infection-prevented model couldthen account for some of the decline indental decay that has been noted inmany countries. However, the actualdemonstration of this effect would re-quire a prospective study where typesand duration of antibiotic usage couldbe correlated with data on MS coloniz-ation and levels, and actual caries ex-perience of the children.

Acknowiedgements

This research was supported by grantsDE02731 and DE045()4.

References

1. Alaluusua S, Rcnkonen OV. Streptoeoe-eus tnutans establishment and dental car-ies experienee in ehildren from 2- to 4-years-old. Seand J Dent Res 1983: 91:453-457.

2. Axelsson P, Lindhe J. Effeet of oralhygiene instruction and professionaltoothcleaning on caries and gingivitis insehoolchildren. Community Dent OralEpidemiol 1981: 9: 251-255.

3. Bohannan HM. The impaet of deereas-ing caries prevalence: Implications fordental education. J Dent Res 1982: 61:1369-1377.

4. Bowden GHW, Odluin O, Nolettc N,Hamilton IR. Microbial populationsgrowing in the presence of fluoride atlow pH isolated from dental plaque ofehildren living in an area with fluori-dated water. Infect Immun 1982: 36:247-254.

5. Brearly LJ, Porteus JR. Charaeteristicsatid experience of tetraeyeline-affecteddentitions. J Dent Res 1973: 52: 508-516.

6. Brunelle JA, Carlos JP. Changes in theprevalenee of dental earies in U.S. sehoolehildren, 1961-1980. J Dent Res 1982:61 (spec, iss.): 1346-1351.

7. Burl BA. The future of the caries decline.J Public Health Dent 1985: 45: 261-269.

8. Burt BA, Loesehe WJ, Eklund SA. Sta-bility of selected plaque speeies and theirrelationship to earies in a child popula-tion over 2 years. Caries Res 1985: 19:193-200.

9. Fitzgerald RJ, Keyes PH. Demon-stration of the etiologic role of strepto-cocci in experimental caries in the ham-ster. J Am Dent Assoe 1960: 61: 9-19.

10. Glass RL. The first international confer-enee on the deelining prevalenee of den-tal earies. J Dent Res 1982: 61:130I-I383.

11. Gold OC, Jordan HV, Van Houte J. Aseleetive medium for Streptoeoeeus mu-tans. Areh Oral Biol 1973: 18:1356-1364.

12. Gordis L. The virtual disappearance ofrheumatie fever in the United States:lesions in the rise and fall of disease.Circulation. 1985: 72: 1155-1162.

13. Granath LE, Martinsson T, Matsson L,Nilsson G, Sehroder U, Soderholm B.Intra-individual effect of daily super-vised llossing on caries in schoolchildren. Conim Dent Oral Epidemiol1979: 7: 147-150.

14. llandelman SL, Hawes RR. The effeetof long-term antibiotie therapy on anti-biotic resistance of the salivary flora. JOral Ther Pharm 1964: 1: 23-44.

15. Horowilz AM, Suomi JD, Pelerson JK,Lyman AB. Effeets of supervised dailydental plaque removal by children. II. 24months' results. J Public Health Dent1977: 37: 180-188.

16. Keyes PH. The infectious and transmiss-ible nature of experimental dental earies.Findings and implieations. Areh OralBiol 1960: 1: 304-320.

17. Kloek B, Krasse B. Caries status andmierobial conditions in children in 1973and 1984. Seand J Dent Res 1987: 95:13-17.

18. Koch G. Evidence for deelining cariesprevalenee in Sweden. J Dent Res 1982:61: 1340-1345.

19. Kohler B, Andreen I, Jonsson B. Theeffeet of caries-preventive measures inmothers on dental caries and the oralpresenee of the baeteria Streptoeoeeusmutans and lactobacilli in their children.Arch Oral Biol 1984: 29: 879-883.

20. Littleton NW, White CF. Dental findingsfrom a preliminary study of children re-eeiving antibiotie therapy. J Amer DentAssoc 1964: 68: 520-525.

21. Loesche WJ. Role of Stre/ttoeoeeus tnu-tans in human dental deeay. MicrobiolRev. 1986a: 50: 353-380.

22. Loesche WJ. Decline in Streptoeoeeusw«(fl/).s-associated caries secondary tomedieal usage of antibiotics. In: Molecu-lar Microbiology and immunobiology of

Antibiotics and mutans streptoeoeei 81

Streptoeoeeus tnutatis. Hamada et al.,eds. Amsterdam: Elsevier 1986b.pp.371-380.

23. Loesche WJ, Straffon LH. Longitudinal 26.investigation of the role of Streptoeoeeusmutans in human fissure decay. InfectImmun 1979: 26: 498-507.

24. Loesche WJ, Eklund S, Earnest R, BurtB. Longitudinal investigation of bacteri-ology orhumiUi fissure decay: cpidemiol- 27.ogical studies in triolars shortly aftereruption. Infect Immun 1984: 46:765-772.

25. Maltz M, Zickert 1. Effect of penicillinon Streptoeoeeus tnutatis, Streptoeoeeus 28.

sanguis, and lactobacilli in hamsters andin man. Scand J Dent Res 1982: 90:193-199.Massell BF, Chute CG, Walker AM,Kurland GS. Penicillin and the tnarkeddecrease in morbidity and mortalityfrom rheumatic fever in the UnitedStates. N Engl J Med 1988: 318:280-286.Masuda N, Tsutsumi N, Sobue S, Ha-mada S. Longitudinal survey of the dis-tribution of various serotypes of Strepto-eoeeus tnutans in infants. J Clin Micro-biol 1979: 10: 497-502.Robertson PB, Wright TE III, Mackler

BF, Lenertz DM, Levy BM. Periodontalstatus of patients with abnormalities ofthe itnmune system. J Periodont Res1978: 13: 37-45.

29. Svanberg M, Loesche WJ. Salivary con-centration of Streptoeoeeus tnutans andStreptoeoeeus satiguis and the colotiiz-ation of artificial fissures in humans bythese organisms. Arch Oral Biol 1977:22: 441^47.

30. Van Houte J, Aasenden R, Peebles TC.Oral colonizatioti of Streptoeoeeus mu-tatis in human subjects with low caries ex-perience given fluoride suppletiients frombirth. Arch Oral Biol 1978: 23: 361-366.