From the over 400 di�erent bacterial speciesthat may colonise the subgingival microbiota,very few have demonstrated a clear etiologicalrole in the pathogenesis of periodontitis (1).Evidence for aetiology is based on the ful®lment

of several criteria de®ned by SOCRANSKY et al.(2). Using these criteria, only a clear evidencehas been demonstrated with Actinobacillus actino-mycetemcomitans, Porphyromonas gingivalis andBacteroides forsythus. However, other putative

Differences in thecomposition of thesubgingival microbiota oftwo periodontitis populationsof different geographicalorigin. A comparisonbetween Spain andThe NetherlandsSanzM, vanWinkelho� AJ, Herrera D, Dellemijn-Kippuw N, SimoÂn R, Winkel EG.Di�erences in the composition of the subgingival microbiota of two periodontitispopulations of di�erent geographical origin. A comparison between Spain andThe Netherlands. Eur J Oral Sci 2000; 108: 383±392. # Eur J Oral Sci, 2000

The purpose of this study was to compare the subgingival microbiota of twogeographically distinct patient populations using identical clinical and bacterio-logical methods. Adult patients with a diagnosis of periodontitis were consecutivelyselected according to pre-de®ned clinical criteria. Microbiological samples weretaken from the deepest four sites with bleeding. The samples were plated on bloodagar plates, for the determination of the total anaerobic counts and identi®cationof speci®c bacterial pathogens, and on TSBV and McConkey for isolation ofActinobacillus actinomycetemcomitans and enteric rods, respectively. Thirty-onepatients in Spain and 30 patients in The Netherlands were selected. Both patientgroups showed similar clinical characteristics, both in terms of age, gender andperiodontal clinical variables. A. actinomycetemcomitans was signi®cantly moreprevalent (23.3% vs. 3.2%) in the Dutch group, while Porphyromonas gingivaliswas signi®cantly more prevalent (64.5% vs. 36.7%) in the Spanish group.Bacteroides forsythus and most commensal periodontal pathogens showed similarprevalences, except Peptostreptococcus micros that was signi®cantly more frequentin the Dutch group (96.7% vs. 74.2%). In summary, the subgingival microbiotafrom the Spanish group was characterised by a high prevalence of P. gingivalisand low of A. actinomycetemcomitans, while the ¯ora from the Dutch group wascharacterised by a high prevalence of A. actinomycetemcomitans and P. micros.

Mariano Sanz1,2, Arie Jan vanWinkelhoff3, David Herrera1,2, NancyDellemijn-Kippuw3, Rosa Simo n2,Edwin Winkel4

1Section of Graduate Periodontology, Facultyof Odontology, Universidad Complutense deMadrid, Madrid, Spain, 2Laboratory ofMicrobiology, Faculty of Odontology,Universidad Complutense de Madrid, Madrid,Spain, 3Department of Oral Biology, SectionClinical Oral Microbiology, Academic Centrefor Dentistry Amsterdam, Amsterdam, TheNetherlands, 4Clinic of PeriodontologyAmsterdam, Amsterdam, The Netherlands

Mariano Sanz, Faculty of Odontology ± CiudadUniversitaria, Avda. RamoÂn y Cajal, s/n, 28040Madrid, Spain

Telefax: z34±91±3941910E-mail: marsan@eucmax.sim.ucm.es

Key words: adult periodontitis; periodontalpathogens; anaerobic bacteria; Spain;The Netherlands

Accepted for publication May 2000

Eur J Oral Sci 2000; 108: 383±392Printed in UK. All rights reserved

periodontal pathogens such as Campylobacterrectus, Eubacterium nodatum, Fusobacteriumnucleatum, Prevotella intermedia/nigrescens, Pepto-streptococcus micros, Streptococcus intermedius,Eikenella corrodens, enteric rods, Pseudomonassp., Selenomonas sp., Staphylococcus sp. andTreponema denticola only show moderate evidence,without demonstrating a clear ful®lment of allthe criteria (3). When the evidence for aetiology iscon®rmed, the detection of these periodontalpathogens would aid in the selection of a moretargeted therapy for the treatment of this disease.This will be particularly useful in patients forwhom the outcome of the standard mechanicaltherapy is not predictable (early onset perio-dontitis), or in whom this conventional periodontalmode of therapy fails (refractory periodontitis)(4±6).

When microbial diagnosis is used for targetingthe therapy and for evaluation of treatment out-comes, clinicians must have a clear idea of theprevalence of these putative pathogens in bothhealth and disease. For more than 20 yr, multiplestudies, mainly using analysis by means ofbacterial culturing, have tried to associate thepresence of di�erent putative pathogens withdisease status, showing a wide variety of results indi�erent populations and in di�erent environments(1, 7±10). From these studies one can concludethat the prevalence of speci®c putative peri-odontal pathogens varies between individualsfrom the same environment and between di�erentcountries.

Very few studies have compared directly theprevalence of periodontal pathogens in two distinctgeographical populations, using the same micro-biological methods and conditions. In one study,comparing Sudanese with Norwegian periodontitispatients, di�erences in transportation time couldaccount for most of the di�erences reported (11).When comparing Chinese patients living in theUSA with matched American patients (12), nodi�erences were found in their subgingival micro-¯ora, with the exception of a higher prevalenceof A. actinomycetemcomitans in the Americangroup. However, in that study the limited numberof patients and pathogens evaluated (black-pigmented Bacteroides, Fusobacterium sp. andA. actinomycetemcomitans) limit the value of theresults.

Therefore, studies with simultaneous comparisonof two distinct populations using identical meth-odology are needed in order to ®nd out if there aredi�erences in the subgingival micro¯ora composi-tion between subjects from di�erent geographicallocations and environments. One possible con-sequence of true geographical di�erences in the

prevalence of periodontal pathogens wouldbe a di�erent therapeutic approach, especially inrelation to antibiotic therapy.

The purpose of this investigation was tostudy the subgingival micro¯ora of two distinctpatient populations, both demonstrating moderateto severe periodontitis, one in The Netherlandsand one in Spain, using identical clinical andbacteriological methods.

Material and methods

Patients

Sixty-one patients with a diagnosis of adult perio-dontitis were consecutively selected to participatein this study from those attending the graduateperiodontal clinics in the Faculty of Odontology,University Complutense of Madrid (Spanish popu-lation) and in the Clinic of Periodontology ±Amsterdam (Dutch population). Both patientpopulations met with the following entrancecriteria: i) age 425 yr; ii) 3 or more teeth in eachquadrant of the dentition with at least one site perquadrant with a probing pocket depth §5 mm andwith bleeding on probing; and iii) radiographicevidence of alveolar bone loss in each quadrant ofthe dentition. Patients were excluded from thestudy if they: i) had a history of previous perio-dontal treatment; ii) had used systemic or topicalantimicrobial therapy 4 wk prior to the study; andiii) were pregnant.

At baseline, information on current systemicdisorders, current medications, history of use ofantibiotics, and smoking habits were recorded.The habit of tobacco smoking was recorded as:never smoked (0); former smoker 41 yr ago (1);former smoker 51 yr ago (2); current smoker 1±10cigarettes/d (3); and current smoker 410 cigar-ettes/d (4).

Microbiological sampling

Based on the patient's clinical and radiographicperiodontal records, the deepest bleeding pocket(mm) with the maximum amount of clinical attach-ment loss (mm) in each quadrant of the dentition(total 4 sites) were selected. At these selected sites,also supragingival plaque accumulation (PI0, PI1),bleeding on probing (BI0, BI1), and suppuration(SI0, SI1) were determined.

After careful removal of supragingival plaquedeposits and isolation of the sampling sites usingcotton rolls and gentle air-drying, two sterilepaper points (Fine, West Palm Beach, CA, USA)were inserted consecutively into the depth of thepocket and left in place for 10 s. Paper points from

384 Sanz et al.

all 4 selected periodontal sites were pooled in 2.0 mlof reduced transport ¯uid (RTF) (13). Sampleswere stored at 4³C and processed within 2 h aftersampling.

Microbiological methodology

Paper point samples were vortexed for 30 s and10-fold serially diluted in phosphate-bu�ered saline(PBS). A 100 mL of each dilution was plated onnon-selective blood agar medium (Oxoid no 2;Oxoid, Basingstoke, UK), supplemented with5% horse blood, haemin (5 mg/L) and menadione(1 mg/L) for the determination of the total anaer-obic counts and identi®cation of speci®c bacterialpathogens. Samples were also plated on trypticasesoy serum-bacitracin-vancomycin plates (TSBV)(14) for the selective isolation and counting ofA. actinomycetemcomitans, and on McConkey agarplates for the isolation of enteric rods andpseudomonads.Blood agar plates were examined after 7 and

14 d of anaerobic incubation (80% N2, 10% H2,10% CO2 at 37³C); TSBV plates after 3±5 d ofincubation at 37³C in air with 5% CO2; andMcConkey plates after 2±3 d of aerobic incubationat 37³C.Total anaerobic microbial counts were evaluated

on blood agar plates. Presence and numbers ofthe putative periodontal pathogens P. gingivalis,P. intermedia, B. forsythus, P. micros, C. rectus, andF. nucleatum were determined on the anaerobicnon-selective blood agar plates. Identi®cation ofthe selected bacterial species was based on Gramstain and cell morphology, aerotolerance, produc-tion of catalase and then con®rmed using di�erentstandard biochemical tests (Rapid ID 32A;BioMerieux, Marcy-l'EÂ toile, France) (15, 16).Colonies were counted and the percentage of thetotal ¯ora for each pathogen was calculated. Totalcounts of A. actinomycetemcomitans were per-formed on TSBV plates. Identi®cation of thisbacterial species was based on its typical colonymorphology (star-like inner structure), a positivecatalase reaction and a set of speci®c enzymes(APIZYM; BioMerieux) (17). Colonies fromMcConkey plates were identi®ed based on thereaction pro®le of the API 20 E (BioMerieux).

Microbiological control procedures

Laboratories in Spain and The Netherlandsused identical microbiological protocols, such aspreparing the transport ¯uid and isolation media.A calibration test prior to the study was performed.Four pooled samples from 4 deep periodontalpockets from 4 untreated adult periodontitis

patients were taken in the Department of Perio-dontology, Faculty of Odontology of Madrid,Madrid, Spain. The samples, taken with fourpaper points per site, were split in two RTF vials,and the cooled samples (510³C) were sent toThe Netherlands within 24 h. During this time, thesamples in Madrid were kept at 4³C. Upon arrivalin The Netherlands, the samples were processedin both countries as described above. The percen-tages of microbiota for each tested pathogenobtained in each laboratory were compared, repre-senting a total of 32 pairs. The analysis of thesecomparisons resulted in no statistical di�erences,both in regards to the frequency of detection(Chi-square, P-value=0.80) and to the proportionsof total micro¯ora (paired t-test, P-value=0.10).The similarities between both sets of data can beseen in Table 1.

Data analyses

Clinical variables: Clinical variables of the samplingsites values were averaged (probing pocket depthand clinical attachment level) or expressed aspercentage of positive sites (plaque, bleeding onprobing, suppuration) within each patient. Con-tinuous data from both patient groups were com-pared using Student's t-test for two independentsamples.

Microbiological variables: The total anaerobiccounts were calculated on blood-agar plates(expressed in total colony-forming units/sample).Counts for each speci®c bacterial species, aswell as its percentage of total ¯ora, were alsocalculated for each patient. Data between bothpatient populations was compared using theStudent's t-test for two independent samples. TheChi-square test was used to assess if di�erencesin the prevalence of the various periodontalpathogens between both patient groups weresigni®cant.

The level of con®dence was established at95%, and P50.05 was considered statisticallysigni®cant.

Table 1

Microbiological control procedures: results (frequency ofdetection) of 4 samples (32 pairs of studied bacterial species)

processed in both laboratories

The Netherlands

Positive Negative Total

Spain

Positive 12 4 16Negative 3 13 16Total 15 17 32

Composition of the subgingival microbiota 385

Results

Thirty-one patients in Spain (Spa-group) and30 patients in The Netherlands (Ned-group) wereinvolved. Both patient groups showed similarcharacteristics with comparable mean age (43.1 yr(range 26±62) versus 43 yr (range 29±63)) andgender (55% of females versus 57%). The presenceof a concomitant systemic disease was reportedby a comparable number of patients in both groups(21.4% in Spain and 16.7% in The Netherlands).The intake of daily medications was more frequentin the Spanish patient group (32.1% versus 3.2%),however the composition of these drugs wereconsidered irrelevant to the results of this study.Smoking habits were di�erent between the twopatient samples (P50.05). The proportion ofheavy smokers (410 cig/d) was close to 50% inboth groups. The percentage of individualssmoking 1±10 cig/d was higher in the Ned-group(26.7% versus 3.2%), and the percentage non-smoking patients was higher in the Spa-group(22.6% versus 6.7%).

The periodontal clinical variables showed similarcharacteristics between both groups (Table 2). The

mean probing pocket depth was 7.0 mm in theNed-group versus 6.2 mm in the Spa-group. Themean clinical attachment level was 8.0 mm versus7.0 mm, the percentage of sites bleeding on probingwas 98.3% versus 98.4%, and the percentage ofsites with suppuration was 27.4% versus 25.8%.The Ned-group demonstrated a better level of oralhygiene when compared with the Spa-group(45.8% of sites free of plaque versus 21.8%).

With this similar clinical conditions and identicalbacteriological methods, the prevalence of speci®cperiodontal pathogens demonstrated signi®cantdi�erences between the two population samples.Table 3 shows the percentage of samples positivefor the di�erent speci®c micro-organisms. In thegroup of pathogens with a presumed etiologicalrole, A. actinomycetemcomitans was signi®cantlymore prevalent (23.3% vs. 3.2%) in the Ned-group(P~0.002), while P. gingivalis was signi®cantlymore prevalent (64.5% vs. 36.7%) in the Spa-group(P~0.029). B. forsythus showed similar percentagesin both patient groups. In the group of commensalperiodontal pathogens, most of the species showedsimilar prevalences, except P. micros that wassigni®cantly more frequent in the Ned-group

Table 2

Clinical data of the Spanish and Dutch patient groups

Spain The Netherlands

Plaque index % sites PI0 21.8 45.8% sites PI1 78.2 54.2

Bleeding index % sites BI0 1.6 1.7% sites BI1 98.4 98.3

Probing pocket depth Mean 6.2 7.0SD 0.8 1.2Range 4.5±8.2 5.8±9.8

Clinical attachment level Mean 7.0 8.0SD 1.4 2.5Range 5.0±10.5 5.0±11.3

Suppuration % patients 51.6 51.7Mean number of sites per patient 1.03 1.11

Table 3

Prevalence of putative periodontal pathogens in the subgingival microbiota of adult patients with periodontitis, in Spain (31 patients) andThe Netherlands (30 patients)

Spain The Netherlands Chi-square

(+) patients prevalence (+) patients prevalence P-value

A. actinomycetemcomitans 1 3.2% 7 23.3% 0.020P. gingivalis 20 64.5% 11 36.7% 0.030P. intermedia 23 74.2% 27 90.0% 0.108B. forsythus 20 64.5% 22 73.3% 0.457P. micros 18 58.1% 29 96.7% 0.000F. nucleatum 31 100.0% 30 100.0% 1.000C. rectus 5 16.1% 11 36.7% 0.068P. melaninogenica 6 19.4% 5 16.7% 0.785

(+) patients, number of patients positive for each bacterial species.

386 Sanz et al.

(96.7% vs. 58.1%). No superinfecting organisms,including enterics, pseudomonae and Candida weredetected. Table 4 shows the proportion of thedi�erent pathogens in the subgingival micro¯ora inthe two samples. No signi®cant di�erences werefound for A. Actinomycetemcomitans, representingless than 5% of the ¯ora. The mean percentageof P. gingivalis ranged between 20±30%, andB. forsythus amounted to 7%. Regarding the totalnumber of bacteria per sample, both populationsshowed similar number of bacteria (Table 4).

Discussion

Table 5 shows the results from di�erent studiesassessing the prevalence of periodontal pathogensin the subgingival microbiota from di�erent perio-dontitis populations and geographical locations.Table 6 describes the di�erences in the populationsstudied, their disease status, and the samplingmethodology. From these studies one may con-clude that the prevalence of speci®c putativeperiodontal pathogens varies between individuals

from the same environment and between di�erentgeographical locations. Most of these studies werecross-sectional, and di�erent populations anddi�erent bacteriological methods were used.Therefore, direct associations and comparisonsare di�cult, and it is not clear if geographical dif-ferences condition di�erences in the microbiota, orif the high variability reported is the consequenceof using di�erent sample populations and di�erentbacteriological methods and culture conditions.

The results from this study demonstrated thattwo sampled populations from di�erent geograph-ical locations with similar periodontal clinicalcharacteristics harboured a micro¯ora with signi-®cantly qualitative di�erent proportion of bacteria,especially those that are supposedly etiologicallyrelevant to the disease process. A. actinomycetem-comitans was found in 23.3% of the samples fromDutch patients, which is well within the range ofprevalence (between 20 and 40%) reported in theliterature (11, 18±28), although both higher andlower prevalence has been reported (29±33).However, Spanish patients very seldom harbouredthis bacterial species (3.2% of the samples).

Table 4

Comparative total anaerobic counts and proportions of total microbiota of different periodontal pathogens in Spain and in TheNetherlands

Spain The Netherlands

A. actinomycetemcomitans Mean%* 0.5% 3.0%SD 0.1% 2.4%Range* 0.5% 0.2±11.2%

P. gingivalis Mean% 21.6% 28.6%SD 16.1% 18.5%Range 2.6±58.0% 1.6±71.3%

P. intermedia Mean% 7.4% 6.3%SD 6.9% 6.9%Range 0.5±31.8% 0.1±25.8%

B. forsythus Mean% 7.3% 7.7%SD 7.1% 6.8%Range 0.1±32.2% 0.9±31%

P. micros Mean% 4.0% 7.2%SD 3.2% 7.1%Range 0.2±12.4% 0.4±26.8%

F. nucleatum Mean% 6.9% 9.3%SD 9.7% 12.0%Range 0.1±43.0% 0.5±56.6%

C. rectus Mean% 2.2% 4.1%SD 1.5% 2.7%Range 0.1±8.1% 0.8±8.7%

P. melaninogenica Mean% 4.0% 6.7%SD 2.0% 3.2%Range 0.9±7.9% 1.6±13.6%

Total counts Mean 1.576107 1.346107

Maximum 1.096108 4.606107

Minimum 5.736105 1.106106

SD 2.476107 1.156107

Median 7.606106 9.306106

*Mean percentage of total microbiota and range, in positive samples. SD, standard deviation.

Composition of the subgingival microbiota 387

Table 5

Results on the prevalence of periodontal pathogens reported in studies from different countries and environments

Prevalence (%)

Country Authors Year Aa Pg Pi Bf Pm Fn Cr Pmela Ec

Cameroon Ali et al. (18) 1997 28 57 38 19 28Kenya Dahle n et al. (21) 1989 40 70 100Sudan Ali et al. (11) 1994 28 36 40 68China Dahle n et al. (19) 1995 20 27 93 87 33China Dahle n et al. (19) 1995 13 40 100 8 40Japan Hagiwara et al. (32) 1998 14 38 43 38 9Greece Kamma et al. (20) 1994 20 100 100 100 100 100 100Italy Piccolomini et al. (28) 1997 27Norway Ali et al. (11) 1994 33 87 78 100Rumania Ali et al. (22) 1996 42 76 18 53Sweden Dahle n & WikstroÈ m (23) 1995 38 31 76Sweden Slots et al. (24) 1986 40±47 51±41 59Sweden Slots et al. (24) 1986 6 0 60Sweden Papapanou et al. (25) 1993 25 14 58 80 81 66Switzerland** McNabb et al. (30) 1992 63 67 97 60 97* 70* 80 40The Netherlands van Dalen et al. (37) 1998 91The Netherlands van der Weijen et al. (26) 1994 38 43 54The Netherlands Rodenburg et al. (29) 1990 54 48 63USA Rams et al. (36) 1993 81USA Kornman et al. (33) 1991 10 33 48 15 10USA Slots et al. (42) 1990 32 45USA Chen et al. (43) 1989 100USA Lotufo et al. (35) 1994 67USA Rams et al. (44) 1992 63Dominican Rep. Slots et al. (34) 1991 37 75 87 100 33

*Prevalence of the genus. **Included only immigrants from di�erent countries.Aa ± A. actinomycetemcomitans, Pg ± P. gingivalis, Pi ± P. intermedia, Bf ± B. forsythus, Pm ± P. micros, Fn ± F. nucleatum, Cr ± C. rectus, Ec ± Eikenella corrodens, Pmela ± Prevotellamelaninogenica.

388

Sanzet

al.

Table 6

Methodology used in studies describing the prevalence of periodontal pathogens in different countries

Material Disease Status Sampling Media

Authors Country Year n Age Dx Prev. treat. Sites Method Transp Within General TSBV Other

Ali et al. (18) Cameroon 1997 21 20±62 AP No 2 2 pp VGMAIII 36±40 h BA yes yesDahle n et al. (21) Kenya 1989 20 30±65 General ? 1 1 pp/curet VGMAIII 4/36±48 h BBA yes noAli et al. (11) Sudan 1994 25 22±70 P. No 1 3 pp VGMAIII 40±48 h BA yes yesDahle n et al. (19) China 1995 15 55±69 General No (worst group) 1 1 pp VGMAIII 10 h BBA yes yesDahle n et al. (19) China 1995 15 55±69 General No (best group) 1 1 pp VGMAIII 10 h BBA yes yesHagiwara (32) Japan 1998 21 43±75 AP Untreated 3? 1 pp PRAS immed CDC yes yesKamma et al. (20) Greece 1994 10 25±35 RPP No 46 m 1 3 pp RTF 10 min ETSA yes noPiccolomini (28) Italy 1997 30 na AP Untreated ? curet RTF 30 min ETSA yes noAli et al. (11) Norway 1994 18 30±61 P. No 1 3 pp VGMAIII 24 h BA yes yesAli et al. (22) Rumania 1996 36 30±68 AP No 1 3 pp VGMAIII 36±40 h BA yes yesDahle n & WikstroÈ m (23) Sweden 1995 535 na P. No; various 1 3 pp VGMAIII ? BA yes yesSlots et al. (24) Sweden 1986 61 19±79 P. No (active) 1 3 pp VGMAIII ? BBA yes noSlots et al. (24) Sweden 1986 20 30±75 P. No (non-active) 1 3 pp VGMAIII ? BBA yes noPapapanou et al. (25) Sweden 1993 192 30±65 General ? 6 1 pp VGMAIII 24 h BBA yes yesMcNabb et al. (30) Switzerland 1992 30 35±44 General No 1 3 pp RTF 15 min TS-BA yes novan Dalen et al. (37) The Netherlands 1998 123 24±68 AP No; 43 m 4 2 pp RTF 24 h BA no novan der Weijen et al. (26) The Netherlands 1994 91 19±54 P. 50 no; 41 yes 4 2 pp RTF 48 h BA yes noRodenburg et al. (29) The Netherlands 1990 138 14±70 P.severe No 3±4 2 pp RTF 45 min BA yes noSlots et al. (34) Dominican Rep. 1991 24 18±60 P. No 3 1 pp VGMAIII 24±48 h BBA yes yesRams et al. (36) USA 1993 1447 36±89 AP-Ref. Various 3 1 pp VGMAIII 48 h BA no yesKornman et al. (33) USA 1991 21 na AP No 1±2 curet VGMAIII 24 h TS-BA yes yesChen et al. (43) USA 1989 11 35±61 AP No 46 m 1 3 pp PRAS 0 TS-BA no yesSlots et al. (42) USA 1990 3075 12±93 Ref. Yes 42 m 3 1 pp VGMAIII 16±40 h BA yes yesSlots et al. (27) USA 1990 1624 15±89 P. Yes, various 3 1 pp VGMAIII 4/24±48 h BBA yes noLotufo et al. (35) USA 1994 80 na P. ? 3 1 pp VGMAIII 0±144 h BBA no noRams et al. (44) USA 1992 545 36±82 AP Most treated 3 1 pp VGMAIII 4±48 h BBA no yesSlots et al. (45) USA 1988 500 na AP Most treated 3 1 pp VGMAIII 24±48 h EBBA yes yesRams et al. (38) USA 1992 907 36±89 AP Most treated 3 1 pp VGMAIII 24±48 h EBBA no noRams et al. (46) USA 1990 506 36±89 AP Most treated 3 1 pp VGMAIII 24±48 h EBBA no yes

AP., adult periodontitis. General, General population. Ref., refractory periodontitis. P., periodontitis. pp, paper points. Prev. treat., previous periodontal treatment. BA, blood agar.CDC, commercial agar. TSA, tripticase soy agar. BBA, Brucella blood agar. TS-BA, tripticase soy blood agar. TSBV, tripticase-serum-bacitracin-vancomicin. ETSA, enriched tripticasesoy agar.

Compositio

nofthesubgingiva

lmicro

biota

389

The reasons for this discrepancy can not beexplained by the use of di�erent methodologies,since a controlled standardised procedure wascarried out, and therefore the di�erences must beecological. However, when present, the proportionsof A. actinomycetemcomitans in respect to the total¯ora were similar between both patient populations(55% in both groups), which also agrees withpreviously published results (11, 21, 22, 24, 26, 27).Those studies showed that the presence ofA. actinomycetemcomitans in periodontitis patientsdoes not follow a geographical pattern, althoughAsian populations seemed to demonstrate a lowerprevalence.

In contrast, P. gingivalis showed a higherprevalence in Spain, as compared both with therange of prevalence reported in the literature andwith the data obtained from the Dutch patients.The presence of P. gingivalis in 36.7% of thesamples from the Dutch group is in accordancewith the range (27±51%) reported in the literature(11, 19, 23, 24, 26, 29, 31±34), but both higher (11,18, 20±22, 30) and lower prevalences (24, 25) havebeen reported. On the contrary, the 64.5% shownin the Spanish samples is higher than the prevalencereported by most other studies. Similar to theresults obtained with A. actinomycetemcomitans,the proportions of this pathogen in the total ¯ora inpositive sites were similar, showing in both samplespercentages over 20%, in accordance to what hasbeen reported in other studies (11, 20, 22, 24, 26,29). Again, those studies showed that the presenceof these bacteria demonstrates a great variability,but without showing a clear geographical pattern.

B. forsythus was found with similar frequency inboth populations, comparable with results fromother studies in the literature (30, 35), althoughboth higher (20) and lower (18) percentages havebeen reported. More studies are necessary tocon®rm the real prevalence of this bacterial speciesin adult periodontitis patients.

In regard to commensal periodontal patho-gens, both patient samples showed similarproportions, except for the percentage of P. microsthat was signi®cantly higher in the Dutch group.P. intermedia/nigrescens showed comparable resultsbetween both groups with percentages within therange (75±100%) of ``high occurrence'' studiesdescribed in the literature (11, 19±21, 23, 30, 31, 34).Also similar mean percentages of ¯ora were cal-culated in both populations, around 7%, whichagrees with the range (2.4±7.4%) found in earlierstudies (11, 21, 22, 24, 26, 27, 29). F. nucleatum wasdetected in all samples in both populations, whichagrees with the range of prevalence described inthe literature, 80±100% (11, 20, 25, 30, 34). Thepercentages of total ¯ora were higher than those in

earlier studies (0.4±3.7%) (11, 22), and closer to the13.6% reported in Greece in rapidly progressiveperiodontitis patients (20). C. rectus showed asimilar presence in both patient groups, howeverthis prevalence is much lower than the 70±100%reported in some studies (19, 20, 25, 30, 36).However, a comparable prevalence has beenreported in the USA (33) and in The DominicanRepublic (34). The reasons for these discrepanciesand the relative importance of this bacterial speciesin the aetiology of periodontitis are still unclear.

P. micros was signi®cantly more prevalent inThe Netherlands, con®rming data from previousstudies in this country (37), and also similar toprevalence reported in Greece (20) and TheDominican Republic (34). The percentage shownin the Spanish population is also similar to resultsreported in the USA (38). P. micros levels seem tobe associated with smoking and may explain inpart the higher prevalence of this bacterial speciesin the Dutch population.

Superinfecting micro-organisms were not foundin the studied populations. Enteric rods,Staphylococci and yeast are not usually found inthe subgingival micro¯ora or in very low numbers,except in speci®c populations, such as in studiesin The Dominican Republic (34), Sudan (11) andRumania (22) where their prevalence has beenfound to be very high (460%). These bacteriawill only grow in speci®c environmental condi-tions or patient populations (immunode®cient,undernourished, previous antibiotic medications,etc.)

The rationale for the use of pooled samples of thefour deepest sites in each quadrant was based ondi�erent earlier studies, showing that this samplingstrategy has the highest chances of detectingP. gingivalis (39, 40), P. intermedia (40) and otherperiodontal pathogens (41).

When comparing the data from the two perio-dontitis populations studied, we could thus identifytwo distinct patterns in their subgingival micro-biota. The microbiota from the Spanish patientgroup was characterised by a high prevalence ofP. gingivalis and a low prevalence of A. actino-mycetemcomitans, while the ¯ora from the Dutchgroup was characterised by a high prevalence ofA. actinomycetemcomitans and P. micros. Thesedi�erences may have therapeutic consequencesmainly in what regards to the prescription ofsystemic antibiotics. It further emphasises thatthis mode of therapy should be based on the resultsof local microbial testing for presence of targetmicroorganisms rather than on clinical criteriaonly. From this study, we can not elucidate if thesedi�erences have any pathogenic or therapeuticimplications, but a di�erent geographical pattern

390 Sanz et al.

has been demonstrated. Longitudinal studies withsequential microbiological monitoring are neededin order to corroborate these di�erences and to®nd out if they are clinically meaningful.

References

1. MOORE W. Microbiology of periodontal disease. J Perio-dont Res 1987; 22: 335±341.

2. SOCRANSKY SS, HAFFAJEE AD. E�ects of therapy onperiodontal infections. J Periodontol 1993; 64: 754±759.

3. CONSENSUS REPORT. Periodontal diseases: pathogenesis andmicrobial factors. Ann Periodontol 1996; 1: 926±932.

4. SHILOAH J, PATTERS MR, DEAN JW, BLAND P, TOLEDO G.The survival rate of Actinobacillus actinomycetemcomitans,Porphyromonas gingivalis, and Bacteroides forsythus follow-ing 4 randomized treatment modalities. J Periodontol 1997;68: 720±728.

5. RENVERT SN, WIKSTROÈ M MB, DAHLEÂ N GG, SLOTS J,EGELBERG J. E�ect of root debridement on the eliminationof Actinobacillus actinomycetemcomitans and Bacteroidesgingivalis from periodontal pockets. J Clin Periodontol1990; 17: 345±350.

6. RENVERT SN, WIKSTROÈ M MB, DAHLEÂ N GG, SLOTS J,EGELBERG J. On the inability of root debridement andperiodontal surgery to eliminate Actinobacillus actino-mycetemcomitans from periodontal pockets. J Clin Perio-dontol 1990; 17: 351±355.

7. VAN WINKELHOFF AJ, RAMS TE, SLOTS J. Systemicantibiotic therapy in periodontics. Periodontol 2000 1996;10: 45±78.

8. VAN WINKELHOFF AJ, PAVICIC M, DE GRAAFF J. Antibioticsin periodontal therapy. In: LANG NP, KARRING T, eds.Proceedings of the 1st European Workshop in Perio-dontology. London: Quintessence Books, 1994; 258±273.

9. HAFFAJEE AD, SOCRANSKY SS. Microbial etiological agentsof destructive periodontal diseases. Periodontol 2000 1999;5: 78±111.

10. SOCRANSKY SS, HAFFAJEE AD. Microbiology of perio-dontal disease. In: LINDHE J, KARRING T, LANG NP, eds.Clinical periodontology and implant dentistry, 1st edition.Copenhagen: Munksgaard, 1997: 138±188.

11. ALI RW, BAKKEN V, NILSEN R, SKAUG N. Comparativedetection frequency of 6 putative periodontal pathogensin Sudanese and Norwegian adult periodontitis patients.J Periodontol 1994; 65: 1046±1052.

12. CAO CF, AEPPLI DM, LILJEMARK WF, BLOMQUIST C,BANDT CL, WOLFF LF. Comparison of plaque micro¯orabetween Chinese and Caucasian population groups. J ClinPeriodontol 1990; 17: 115±118.

13. SYED SA, LOESCHE WJ. Survival of human dental plaque¯ora in various transport media. Appl Microbiol 1972; 24:638±644.

14. SLOTS J. Selective medium for isolation of Actino-bacillus actinomycetemcomitans. J Clin Microbiol 1982; 15:606±609.

15. VAN WINKELHOFF AJ, CLEMENT M, DE GRAAFF J. Rapidcharacterization of oral and nonoral pigmented Bacteroidesspecies with the ATB Anaerobes ID system. J ClinMicrobiol 1988; 26: 1063±1065.

16. WINKEL EG, VAN WINKELHOFF AJ, TIMMERMAN MF,VANGSTED T, VAN DER VELDEN U. E�ects of metronidazolein patients with ``refractory'' periodontitis associatedwith Bacteroides forsythus. J Clin Periodontol 1997; 24:573±579.

17. SLOTS J. Enzymatic characterization of some oral andnonoral Gram-negative bacteria with the API-ZYMsystem. J Clin Microbiol 1981; 14: 288±294.

18. ALI RW, JOHANNESSEN AC, DAHLEÂ N GG, SOCRANSKY SS,SKAUG N. Comparison of the subgingival microbiota ofperiodontally healthy and diseased adults in NorthernCameroon. J Clin Periodontol 1997; 24: 830±835.

19. DAHLEÂ N GG, LUAN W-M, BAELUM V, FEJERSKOV O,CHEN X. Periodontopathogens in elderly Chinese withdi�erent periodontal disease experience. J Clin Periodontol1995; 22: 188±200.

20. KAMMA JJ, NAKOU M, MANTI FA. Microbiota of rapidlyprogressive periodontitis lesions in association with clinicalparameters. J Periodontol 1994; 65: 1073±1078.

21. DAHLEÂ N GG, MANJI F, BAELUM V, FEJERSKOV O. Black-pigmented Bacteroides species and Actinobacillus actino-mycetemcomitans in subgingival plaque of adult Kenyans.J Clin Periodontol 1989; 16: 305±310.

22. ALI RW, VELCESCU C, JIVANESCU M-C, LOFTHUS B,SKAUG N. Prevalence of 6 putative periodontal patho-gens in subgingival plaque samples from Romanianadult periodontitis patients. J Clin Periodontol 1996; 23:133±139.

23. DAHLEÂ N GG, WIKSTROÈ M MB. Occurrence of enteric rods,staphylococci and Candida in subgingival samples. OralMicrobiol Immunol 1995; 10: 42±46.

24. SLOTS J, BRAGD L, WIKSTROÈ M MB, DAHLEÂ N GG. Theoccurrence of Actinobacillus actinomycetemcomitans,Bacteroides gingivalis and Bacteroides intermedius indestructive periodontal disease in adults. J Clin Periodontol1986; 13: 570±577.

25. PAPAPANOU PN, SELLEÂ N A, WENNSTROÈ M JL, DAHLEÂ N G. Ananalysis of the subgingival micro¯ora in randomly selectedsubjects. Oral Microbiol Immunol 1993; 8: 24±29.

26. VAN DER WEIJDEN GA, TIMMERMAN MF, REIJERSE E,WOLFFE GN, VAN WINKELHOFF AJ, VAN DER VELDEN U.The prevalence of A. actinomycetemcomitans, P. gingivalisand P. intermedia in selected subjects with periodontitis.J Clin Periodontol 1994; 21: 583±588.

27. SLOTS J, FEIK D, RAMS TE. Actinobacillus actinomycetem-comitans and Bacteroides intermedius in human perio-dontitis: age relationship and mutual association. J ClinPeriodontol 1990; 17: 659±662.

28. PICCOLOMINI R, DI BONAVENTURA G, CATAMO G,PICCIANI C, PAOLANTONIO M. In vitro antimicrobialsusceptibility of periodontopathic Actinobacillus actino-mycetemcomitans to roxithromycin and erythromycin.Oral Microbiol Immunol 1997; 12: 366±371.

29. RODENBURG JP, VAN WINKELHOFF AJ, WINKEL EG, GOENEÂ

RJ, ABBAS F, DE GRAAFF J. Occurrence of Bacteroides gin-givalis, Bacteroides intermedius, and Actinobacillus actino-mycetemcomitans in severe periodontitis in relation to ageand treatment history. J Clin Periodontol 1990; 17: 392±399.

30. MCNABB H, MOMBELLI A, GMUÈ R R, MATHEY-DINCË S, LANG

NP. Periodontal pathogens in the shallow pockets ofimmigrants from developing countries. Oral MicrobiolImmunol 1992; 7: 267±272.

31. PREUS HR, ANERUD A, BOYSEN H, DUNFORD RG, ZAMBON

JJ, LOÈ E H. The natural history of periodontal disease. Thecorrelation of selected microbiological parameters withdisease severity in Sri Lankan tea workers. J Clin Perio-dontol 1995; 22: 674±678.

32. HAGIWARA S, TAKAMATSU N, TOMINAGA Y, UMEDA M.Subgingival distribution of periodontopathic bacteria inadult periodontitis and their susceptibility to minocycline-HCl. J Periodontol 1998; 69: 92±99.

33. KORNMAN KS, NEWMAN MG, ALVARADO R, FLEMMIG TF,NACHNANI S, TUMBUSCH J. Clinical and microbiological

Composition of the subgingival microbiota 391

patterns of adults with periodontitis. J Periodontol 1991; 62:634±642.

34. SLOTS J, RAMS TE, FEIK D, TAVERAS HD, GILLESPIE GM.Subgingival micro¯ora of advanced periodontitis in theDominican Republic. J Periodontol 1991; 62: 543±547.

35. LOTUFO R, FLYNN J, CHEN C, SLOTS J. Molecular detectionof Bacteroides forsythus in human periodontitis. OralMicrobiol Immunol 1994; 9: 154±160.

36. RAMS TE, FEIK D, SLOTS J. Campylobacter rectus in humanperiodontitis. Oral Microbiol Immunol 1993; 8: 230±235.

37. VAN DALEN PJ, VAN WINKELHOFF AJ, VAN STEENBERGEN T.Prevalence of Peptostreptococcus micros morphotypes inpatients with adult periodontitis. Oral Microbiol Immunol1998; 13: 62±64.

38. RAMS TE, FEIK D, LISTGARTEN MA, SLOTS J. Peptostrepto-coccus micros in human periodontitis. Oral MicrobiolImmunol 1992; 7: 1±6.

39. MOMBELLI A, MCNABB H, LANG NP. Black-pigmentingGram-negative bacteria in periodontal disease. II. Screen-ing strategies for detection of P. gingivalis. J Periodont Res1991; 26: 308±313.

40. MOMBELLI A, MCNABB H, LANG NP. Black-pigmentinggram-negative bacteria in periodontal disease. I.

Topographic distribution in the human dentition.J Periodont Res 1991; 26: 301±307.

41. WIKSTROÈ M MB, RENVERT SN, DAHLEÂ N GG, JOHNSON T.Variance in recovery of periodontitis-associated bacteriacaused by sampling technique and laboratory processing.Oral Microbiol Immunol 1991; 6: 102±106.

42. SLOTS J, FEIK D, RAMS TE. Age and sex relationships ofsuperinfecting microorganisms in periodontitis patients.Oral Microbiol Immunol 1999; 5: 305±308.

43. CASEY CHEN C-K, DUNFORD RG, REYNOLDS HS, ZAMBON

JJ. Eikenella corrodens in the human oral cavity. J Perio-dontol 1989; 60: 611±616.

44. RAMS TE, FEIK D, YOUNG V, HAMMOND BF, SLOTS J.Enterococci in human periodontitis. Oral Microbiol Immu-nol 1992; 7: 249±252.

45. SLOTS J, FEIK D, RAMS TE. Prevalence and antimicrobialsusceptibility of Enterobacteriaceae, Pseudomonadaceae andAcinetobacter in human periodontitis. Oral MicrobiolImmunol 1990; 5: 149±154.

46. RAMS TE, FEIK D, SLOTS J. Staphylococci in humanperiodontal diseases. Oral Microbiol Immunol 1990; 5:29±32.

392 Sanz et al.