systemic anti-infective periodontal therapy. a systematic ... · † department of periodontics,...

Ann Periodontol

Systemic Anti-Infective Periodontal Therapy.A Systematic Review

Anne D. Haffajee,* Sigmund S. Socransky,* and John C. Gunsolley†

* Department of Periodontology, The Forsyth Institute, Boston, Massachusetts.† Department of Periodontics, College of Dental Surgery, University of Maryland, Baltimore, Maryland.

115

Background: Periodontal diseases are infections and thus systemically administered antibiotics are oftenemployed as adjuncts for their control. There are conflicting reports as to whether these agents provide a ther-apeutic benefit.

Rationale: The purpose of this systematic review is to determine whether systemically administeredantibiotics improve a primary clinical outcome measure, periodontal attachment level change.

Focused Question: In patients with periodontitis, what is the effect of systemically administered antibi-otics as compared to controls on clinical measures of attachment level?

Search Protocol: The PubMed database was searched from 1966 to May 2002. Searches were limitedto human studies published in English. Hand searches were performed on the Journal of Clinical Peri-odontology, Journal of Periodontology, and Journal of Periodontal Research. References in relevant papersand review articles were also examined.

Selection CriteriaInclusion criteria: Trials were selected if they met the following criteria: randomized controlled clinical

trials, quasi-experimental studies, and cohort studies of >1 month duration with a comparison group; sub-jects with aggressive, chronic, or recurrent periodontitis and periodontal abscess; use of a single or a com-bination of systemically administered antibiotics(s) versus non-antibiotic therapy; and a primary outcomeof mean attachment level change (AL).

Exclusion criteria: Studies involving the use of low-dose doxyclycline, combinations of locally plus sys-temic antibiotics, or where the control group included a systemically administered antibiotic were excluded.

Data Collection and Analysis: A mean difference in AL between groups was available for all papers used inthe meta-analysis. A standard deviation (SD) for the difference was used if provided or calculated from the SDor standard error of the mean (SEM) when provided for single measurements. Data were subset by antibioticemployed, type of adjunctive therapy, and disease type. Results were assessed with both fixed-effects andrandom-effects models.

Main Results1. Twenty-nine studies, 26 RCTs and 3 quasi-experimental (36 comparisons), met the entry criteria.

Total study population, both control and test groups, was estimated at over 1,200.2. Twenty-two studies (27 comparisons) were used in the meta-analysis, evaluating if the antibiotics pro-

vided a consistent benefit in mean AL change for different patient populations, for different therapies, and fordifferent antibiotics.

3. For the majority of the comparisons, systemically administered antibiotics exhibited a more positiveattachment level change than the control group in the study. The combined results were statistically significant(P

Systemic Anti-Infective Therapy Volume 8 • Number 1 • December 2003

BACKGROUNDPeriodontal diseases constitute a series of infectionscaused by the microorganisms that colonize at or belowthe gingival margin. These infections commonly leadto periodontal inflammation and often result in thedestruction of the supporting tissues of the teeth. Sincemost periodontal pathology is caused by bacteria, it isnot surprising that systemically administered antibi-otics have often been employed as adjuncts to aid intheir control. The organisms that cause these diseasesreside in unique structures termed biofilms that offerpartial protection to the colonizing organisms from thedefense mechanisms of the host as well as from theantibiotics used for their treatment.

The literature is replete with studies that havedescribed the effects of systemically administeredantibiotics on periodontal diseases. These reports rangefrom case reports to randomized clinical trials (RCTs).Most reports suggest benefits resulting from antibioticuse, although often these benefits were not “statisticallysignificantly better” than comparison groups. Unfortu-nately, the reports in the literature are quite heteroge-neous in terms of antibiotics employed, dosage, dura-tion, nature of adjunctive therapy, time of follow up,subject population, and the choice of outcome vari-ables.

RATIONALEThe purpose of this systematic review was to examinethe added effects, if any, of systemically administeredantibiotics on one commonly accepted primary outcomevariable, clinically determined gain in periodontal attach-ment level (AL).

FOCUSED QUESTIONThis systematic review attempted to answer the focusedquestion: “in patients with periodontitis, what is the effectof systemically administered antibiotics as comparedto controls on clinical measures of attachment level?”

SEARCH PROTOCOLData Sources and Search StrategiesPubMed was the primary database searched from 1966to May 2002.

The following search terms were used:For clinical conditions: periodontitis; periodontal dis-

eases; periodontal attachment loss; periodontitis juvenile;recurrent periodontitis; refractory periodontitis; alveolarbone loss; periodontal abscess; periodontal pocket; peri-odontics; tooth loss; and tooth mobility.

For periodontal treatment: antibiotic; antibiotics; amox-icillin; amoxycillin; amoxicillin-potassium clavulanatecombination; amoxicillin + clavulanic acid; penicillin;ampicillin; doxycline; tetracycline; minocycline; metron-idazole; flagyl; clindamycin; spiramycin; rovamycin;azithromycin; erythromycin; kanamycin; rifampin; rifam-picin; chloramphenicol; and ciprofloxacin.

Searches were limited to human studies in the Eng-lish language. Available abstracts were reviewed andsuitable papers requested. In addition, potentially suit-able papers without abstracts and publications thatcould not be adequately evaluated from the abstractwere also requested. Hand searches of Journal of Clin-ical Periodontology, Journal of Periodontology, andJournal of Periodontal Research and examination ofreferences in relevant papers and review articles wereused to supplement the search.

Selection Criteria: InclusionStudy types: The studies were limited to randomizedcontrolled clinical trials, quasi-experimental studies, andcohort studies of >1 month duration. A comparison groupwas mandatory.

Study populations: Participants included subjectswith aggressive periodontitis, chronic periodontitis,recurrent periodontitis, and periodontal abscess.

Intervention: Studies were limited to those using asingle systemically administered antibiotic versus non-antibiotic therapy or therapy using combined system-ically administered antibiotics versus non-antibiotictherapy. Antibiotic therapies could be stand-alone orcombined with mechanical therapies such as scalingand root planing (SRP) or periodontal surgery.

Selection Criteria: ExclusionStudies involving the use of low-dose doxycycline,local + systemic antibiotics, and where the control groupincluded a systemically administered antibiotic wereexcluded, as were animal and in vitro studies.

OutcomesThe primary outcome measured in the studies had tobe clinically determined attachment level (AL) change

116

Reviewers’ Conclusions1. The use of systemically administered adjunc-

tive antibiotics with and without SRP and/or surgeryappeared to provide a greater clinical improvementin AL than therapies not employing these agents.

2. The data supported similar effect sizes forthe majority of the antibiotics; therefore, the selec-tion for an individual patient has to be made basedon other factors.

3. Due to a lack of sufficient sample size formany of the antibiotics tested, it is difficult to pro-vide guidance as to the more effective ones.Ann Periodontol 2003;8:115-181.

KEY WORDSClinical trials, controlled; antibiotics/therapeutic use; periodontal diseases/drugtherapy; periodontal attachment level/drugtherapy; periodontal diseases/therapy;periodontal attachment level/therapy; reviewliterature; meta-analysis.

8006.qxd 12/30/03 3:20 PM Page 116

Ann Periodontol Haffajee, Socransky, Gunsolley

(e.g., mean AL change, % of sites exhibiting AL changeover a given threshold). Studies without this outcomewere excluded.

Methods of ReviewEach reviewer (ADH, SSS) independently examined thetitles and abstracts of the search output for the follow-ing criteria: 1) Was it a human study? 2) Of periodon-titis? 3) Involving a systemically administered antibiotic?4) Was it a randomized clinical trial, quasi-experimental,or cohort study? 5) Was there a comparison group?6) Was there a clinically determined attachment levelmeasure outcome? A “No” to any of these questionswould exclude the study.

The agreement between the 2 reviewers for eachpaper was tabulated in a 2 × 2 table. A paper was dis-carded at this stage if both reviewers agreed. Thesenumbers were recorded. If either reviewer “accepted”the paper, it was entered into a full-text review.

All possibly relevant full-text papers were screened byboth reviewers using the screening criteria outlinedabove. Reasons for exclusion were recorded. If there wasdisagreement, the reasons were discussed and resolved.

Data Collection and AnalysisData were independently abstracted from all eligiblefull-text papers using a review form by the reviewers.

Study Quality AssessmentThe following measures were used to insure quality ofdata collection. There was duplicate independentscreening and data abstraction; agreement scores werechecked using kappa statistics; a log of all studiesscreened and accepted or rejected was kept; the num-bers achieved from the initial search were recordedas well as the number of publications accepted andrejected. Reasons for rejection of studies at the full-textreview stage were recorded.

The following questions were asked of each study.Was assignment to treatment groups random? Wasthe treatment allocation concealed? Were the groupssimilar at baseline in terms of level of disease? Wereinclusion and exclusion criteria specified? Were out-come assessors masked to the treatment allocation?Was the care provider masked? Was the patientmasked? Were the point estimates and measure of vari-ability presented for the primary outcome measure?Did the analyses include an intention-to-treat analy-sis? Were drop-out rates presented and explained? Wereadverse events reported? Was compliance measured?

Data AnalysisA mean difference in AL between groups was availablefor all the studies employed in the meta-analyses. Astandard deviation (SD) for the difference was calculatedusing the following strategy. All standard errors (SE)were converted to SD. If there was a standard error ordeviation for the change, that was used. If only a stan-

dard error or deviation of an individual measure wasavailable, then it was assumed that the covariance was0 and the variance of the change was computed as thesum of the variance of the 2 measures. The same for-mula was applied in order to obtain the variance of thedifferences in changes between groups.

The outcome nearest to the 6-month time point wasused for the analysis, although this was not possible forall studies. Studies that examined only 1 or 2 sites werenot included in the analysis. It was assumed that theadjunctive effect to different mechanical therapies wouldresult in similar differences between groups. This wasexamined in one of the analyses. In different analyses,the data were subset according to systemic antibioticemployed, type of adjunctive therapy, and disease type.

The data were analyzed using a standardized differ-ence as described by Fleiss.1 The results were checkedwith both a fixed-effects model and a random-effectsmodel and the results were consistent. To test for het-erogeneity both Cohen’s d (unadjusted)2 and Hedges’sg (adjusted)3 were used. Both tests had to be non-significant to support the lack of heterogeneity.

Details of Included StudiesThe search strategy outlined above identified 526studies; 446 of which met the established inclusioncriteria. Of these, the reviewers agreed on 68 for full-text review. There was disagreement on 12 studiesthat were also acquired in full-text format. Thus, therewas a 97.7% agreement between reviewers on theselection of the studies. After full-text review, 29 werefound to meet all inclusion and exclusion criteria,7 of the 29 were excluded from the meta-analysis forreasons described below. Thus, a total of 22 studiesproviding 27 comparisons, since some studies eval-uated more than a single agent, were analyzed in themeta-analysis.

Rationale for excluding the other 51 studies whichmet the defined criteria is explained immediately beforethe Results section.

The design of the 29 (36 comparisons) includedstudies is presented in Table 1 (page 118). The stud-ies are subset according to the systemically adminis-tered antibiotic(s). There was a single study employ-ing amoxicillin (AMOX),4 5 comparisons4-8 evaluatedAMOX plus metronidazole (MET), 4 comparisons9-12

evaluated AMOX + clavulanic acid (AMOX + CA), 2comparisons10,13 evaluated clindamycin (CLIN), 3comparisons13-15 evaluated doxycycline (DOXY) attherapeutic levels, 12 comparisons4,13,16-25 evaluatedMET, 1 comparison26 evaluated penicillin (PEN), 2comparisons27,28 evaluated spiramycin (SPIR), and 6comparisons12,27,29-32 tetracycline (TET). All studiesincluded either placebo (PLAC), SRP, or surgery (SURG)in either the control or test groups.

117

(text continued on page 121)

8006.qxd 12/30/03 3:20 PM Page 117


118

Table 1.

Design of Included Studies

TreatmentStudyDesign/ N

Reference Duration Groups Test Control Setting Funding Ranking*

AMOXICILLIN

Rooney et al.4 2002 RCT 6 months 4 AMOX + MET + SRP; 2 PLAC + SRP University Not stated IMET + PLAC + SRP; UKAMOX + PLAC + SRP:

AMOXICILLIN + METRONIDAZOLE

Lopez et al.5 2000 RCT 12 months 2 MET & AMOX PLAC University Government IChile

Berglundh et al.6 RCT 24 months 4 MET & AMOX + PLAC + SRP; University Government/ I1998 SRP; MET & AMOX PLAC Sweden industry

Winkel et al.7 2001 RCT 3 months 2 MET + AMOX + PLAC + SRP University Not stated ISRP The Netherlands

Rooney et al.4 2002 RCT 6 months 4 AMOX + MET + SRP; 2 PLAC + SRP University Not stated IMET + PLAC + SRP: UKAMOX + PLAC + SRP:

Tinoco et al.8 1998 RCT 12 months 2 AMOX + MET + SRP PLAC + SRP Multi-center Government IBrazil

AMOXICILLIN + CLAVULANIC ACID

Winkel et al.9 1999 RCT 12 months 2 AMOX + CA + SRP PLAC + SRP University Industry IThe Netherlands

Magnusson et al.10 Quasi 3 AMOX + CA + SRP; PLAC + SRP University Government II-11994 24 months CLIND + SRP USA

Purucker et al.11 RCT 54 weeks 2 AMOX + CA + SRP Local TET University Industry II-12001 fibers + SRP Germany

Haffajee et al.12 RCT 10 months 3 TET + SRP + MWF or PLAC + SRP + University Government I1995 AMOX + CA + MWF USA

SRP + MWF†

CLINDAMYCIN

Sigusch et al.13 2001 RCT 24 months 4 DOXY + SRP; MET + SRP University Not stated II-1SRP; CLIND + SRP Germany

Magnusson et al.10 Quasi 3 AMOX + CA + SRP; PLAC + SRP University Government II-11994 24 months CLIND + SRP USA

DOXYCYCLINE

Ng and Bissada14 1998 RCT 24 weeks 4 DOXY + SRP; DOXY PLAC + SRP; PLAC Unclear Not stated IUSA

Feres et al.15 1999 RCT 90 days 2 DOXY + SRP SRP University Government II-1USA

Sigusch et al.13 2001 RCT 24 months 4 DOXY + SRP; MET + SRP University Not stated II-1SRP; CLIND + SRP Germany

METRONIDAZOLE

Yilmaz et al.16 1996 RCT 42 days 4 MET; MET + SRP None; SRP University Industry II-1Turkey

Noyan et al.17 1997 RCT 42 days 4 MET; MET + SRP None; SRP Unclear Industry II-1Turkey

8006.qxd 12/30/03 3:20 PM Page 118


119

Table 1. (continued)



Reference Duration Groups Test Control Setting Funding Ranking

Lindhe et al.18 1983 RCT 50 weeks 4 MET + SRP; MET SRP; none University Government II-1Sweden

Walsh et al.19 1986 Quasi 3 months 3 MET SRP or none University Government/ II-1USA industry

Sigusch et al.13 RCT 24 months 4 DOXY + SRP; MET + SRP University Not stated II-12001 SRP; CLIND + SRP Germany

Loesche et al.20 RCT 4-6 weeks 2 MET + SRP PLAC + SRP University Government I1992 after completion USA

of debridement

Loesche et al.21 RCT 4-6 weeks 2 MET + SRP PLAC + SRP University Government I1991 after completion USA

of debridement

Rooney et al.4 RCT 6 months 4 AMOX + MET + SRP; 2 PLACS + SRP University Not stated I2002 MET + PLAC + SRP; UK

AMOX + PLAC + SRP

Clark et al.22 1983 RCT 6 months 2 MET + prophy PLAC + prophy Unclear Not stated II-1Canada

Soder et al.23 RCT 5 years 8 MET + SRP smokers; PLAC + SRP smokers Unclear Not stated II-11999 MET + SRP non- PLAC + SRP non- Sweden

smokers; MET + smokers; PLAC +SRP + SURG smokers; SRP + SURGMET + SRP + SURG smokers; PLAC +non-smokers SRP + SURG

non-smokers

Palmer et al.24 RCT 24 weeks 3 MET + SRP SRP; SRP + MET gel University Government II-11998 UK

Palmer et al.25 RCT 6 months 3 MET + SRP SRP; SRP + MET gel University Government II-11999 UK

PENICILLIN

Kunihira et al.26 RCT 62 weeks; 98 2 PEN + SRP + SURG SRP + SURG + PLAC Unclear Not stated I1985 weeks for subset USA

SPIRAMYCIN

Al-Joburi et al.27 RCT 24 weeks 3 SPIRA + SRP;TET + SRP PLAC + SRP Unclear Industry I1989 Canada

Bain et al.28 1994 RCT 24 weeks 2 SPIRA + SRP PLAC + SRP Multi-center Industry ICanada

TETRACYCLINE

Hellden et al.29 RCT 25 weeks 4 TET + SRP;TET SRP; none University Not stated II-11979 Sweden

Lindhe et al.30 RCT 50 weeks 4 TET + SRP;TET PLAC + SRP; PLAC University Government I1983 Sweden

Al-Joburi et al.27 RCT 24 weeks 3 SPIRA + SRP;TET + SRP PLAC + SRP Unclear Industry I1989 Canada (continued)

8006.qxd 12/30/03 3:20 PM Page 119


120




Reference Duration Groups Test Control Setting Funding Ranking

Ramberg et al.31 Quasi 13 years 2 TET + SRP SRP University Government II-12001 (numeric data for Sweden

1 year only)

Haffajee et al.12 RCT 10 months 3 TET + SRP + MWF PLAC + SRP + University Government I1995 or AMOX + CA + MWF USA

SRP + MWF

Palmer et al.32 RCT 12 months 2 TET + SRP followed PLAC + SRP followed University Not stated I1996 by SURG by SURG UK

* I: RCT; II-1: controlled trial without randomization.† Modified Widman flap.

Table 2.

Quality Assessment

Randomization Masking

Reference Randomization Method Allocation Concealment Patient Therapist Examiner

AMOXICILLIN

4 According to subject number Coded packages dispensed by hospital pharmacy Yes Yes Yes


5 Unclear Coded labels Yes Yes Yes

6 Unclear Unclear Yes Unclear Unclear

7 Unclear Unclear Yes Unclear Yes




9 Unclear Coded labels Yes Unclear Yes

10 Unclear Unclear Unclear Unclear Unclear

11 Unclear Unclear No No Unclear

12 Random number table Unclear Yes Unclear Yes

CLINDAMYCIN

13 Unclear Unclear No Unclear Yes


DOXYCYCLINE


15 Random number table Unclear No No Yes


8006.qxd 12/30/03 3:20 PM Page 120


Twenty-six comparisons were RCTs and 3 werequasi-experimental. Twenty-seven of the compari-sons took place in a university setting, 2 were multi-center studies, and the location of 7 was unclear. Fif-teen comparisons were funded by government, 7 byindustry, 2 by government and industry, and the fund-ing source was unknown for the remaining studies.The duration of the studies ranged from 6 weeks to5 years. The studies were carried out in 9 countries;10 in the United States, 6 each in United Kingdom andSweden, 4 each in Canada and Germany, 2 each inTurkey and The Netherlands, and 1 each in Brazil andChile.

Quality AssessmentQuality assessment of the studies is summarized inTable 2 (page 120). While randomization was employedin 26 studies, only 11 of the comparisons described the

method of randomization and just 6 defined allocationand concealment procedures. Masking of the subjectstook place in 20 of the comparisons, of the therapist in9, and of the examiner in 26. Masking was not done orwas unclear for the remainder of the studies. One exam-iner was employed in 8 studies and examiner calibra-tion was described in 9 of the comparisons. Inclusioncriteria were defined for all but 1 comparison and exclu-sion criteria were described for all studies. The numberof subjects completing the studies was documented inall studies but the reason for dropouts was usually notdescribed. In addition, the handling of dropouts in thedata analysis of many studies was not explicitly stated,although many studies appeared to employ only thesubjects that were retained to the end of the study inthe analyses.

121



Quality Assessment

Subject Selection Defined Completion

Calibration Inclusion Criteria Exclusion Criteria Patients Acounted at End Dropouts in Analyses?

AMOXICILLIN

1 examiner Yes Yes Yes No


Yes Yes Yes Yes No

Unclear No Yes Yes N/A*

Unclear Yes Yes Yes N/A





Unclear Yes Yes Yes No

1 examiner Yes Yes Yes Unclear

Yes Yes Yes Yes N/A

CLINDAMYCIN

Yes Yes Yes Yes N/A


DOXYCYCLINE


Yes Yes Yes Yes N/A

Yes Yes Yes Yes N/A

(continued)

8006.qxd 12/30/03 3:20 PM Page 121


122

Table 2. (continued) Table 2. (continued)

Quality Assessment

Randomization Masking

Reference Randomization Method Allocation Concealment Patient Therapist Examiner

METRONIDAZOLE



18 Not stated Not stated Yes Unclear Yes

19 None N/A No No Yes


20 Random number table Unclear Yes No Yes



22 Not stated Sealed packages with coded numbers Yes Unclear Yes

23 Computer generated list in Sealed envelopes Yes Yes Yesblocks of 10

24 Random number table Unclear No No Yesafter stratification

25 Unclear Unclear No No Yes

PENICILLIN

26 Matched pairs and Not stated Yes Yes Yespharmacologistrandomly assigned drugs

SPIRAMYCIN

27 Unclear Unclear Yes Yes Yes

28 Coded envelopes from drug Coded envelopes from drug manufacturer Yes Yes Yesmanufacturer

TETRACYCLINE

29 Unclear Unclear No Unclear Unclear

30 Not stated Not stated Yes Unclear Yes

27 Unclear Unclear Yes Yes Yes

31 None N/A No No No

12 Random number table Unclear Yes Unclear Yes


* Not applicable or not available.

8006.qxd 12/30/03 3:20 PM Page 122


123

Quality Assessment

Subject Selection Defined Completion

Calibration Inclusion Criteria Exclusion Criteria Patients Acounted at End Dropouts in Analyses?

METRONIDAZOLE




Unclear No Yes Yes N/A

Yes Yes Yes Yes N/A




2 examiners measured everyone Yes Yes Yes N/A



Unclear Yes Yes Yes? No

PENICILLIN

1 examiner Yes Yes Yes N/A

SPIRAMYCIN

Not stated Yes Yes Yes No


TETRACYCLINE

Yes Yes Yes Yes N/A


Not stated Yes Yes Yes No

Yes Yes Yes Yes Unclear

Yes Yes Yes Yes N/A


8006.qxd 12/30/03 3:20 PM Page 123


124

Table 3.

Subject Populations

N Test Group N Control Group

Reference Baseline Completion Baseline Completion Disease % Males

AMOXICILLIN

4 66 overall 4 lost AMOX + MET = 15 66 overall PLAC = 15 Advanced Not statedMET = 16 4 lost chronicAMOX = 16


5 23 20 23 19 Active 15.3

6 8 8 8 8 Advanced 38

7 23 23 26 26 Chronic 43

4 66 overall 4 lost AMOX + MET = 15 66 overall PLAC = 15 Advanced Not statedMET = 16 4 lost chronicAMOX = 16

8 25 total in 10 25 total in 10 Aggressive Not statedboth groups both groups (LJP)


9 10 10 11 11 Chronic 29

10 AMOX + CA = 13 AMOX + CA = 12 4 4 Refractory Not statedCLIND = 4 CLIND = 4

11 15 13 15 15 Generalized Unclearaggressive

12 TET = 13 TET = 13 11 11 Active 55 PLAC, 62 TET;AMOX + CA = 10 AMOX + CA = 10 40 AMOX + CA

CLINDAMYCIN

13 DOXY = 12 DOXY = 12 10 10 Generalized 42MET = 15 MET = 15 aggressiveCLIND = 11 CLIND = 11

10 AMOX + CA = 12 AMOX + CA = 12 4 4 Refractory Not statedCLIND = 4 CLIND = 4

DOXYCYCLINE

14 8 (split mouth) 8 8 (split-mouth) 8 Chronic 56

15 10 10 10 10 Chronic 60% in both groups


8006.qxd 12/30/03 3:20 PM Page 124


125


Subject Populations

Mean AgeRacial/

% SmokersSystemically

Test Control All Age Range Ethnic Group Test Control Healthy

AMOXICILLIN

Not stated Not stated Not stated 20-45 Not stated Not stated Not stated Yes


46.7 44.1 Not stated 36-68 No 40 42.1 Yes

Not stated Not stated Not stated 35-58 No Not stated Not stated Yes

42 28-63 No 61 69 Yes


Not stated Not stated Not stated 12-19 Unclear Not stated Not stated Yes


49 39 44 28-66 No 50 45 Yes

Not stated Not stated 46 35-62 Not stated 16/21 (76%) 16/21 (76%) Yesoverall overall

Not stated Not stated 32 20-40 Caucasian Not stated Not stated Yes

44 TET; 48 Not stated 14-71 No Not stated Not stated Yes53 AMOX + CA

CLINDAMYCIN

Not stated Not stated 32.4 Not stated Not stated No smokers No smokers Yes

Not stated Not stated 46 35-62 Not stated 16/21 (76%) 16/21 (76%) Yesoverall overall

DOXYCYCLINE

Not stated Not stated Not stated 32-72 Not stated Unclear Unclear Yes

49 54 Not stated ≥20 years Not stated Not stated Not stated Yes


(continued)

8006.qxd 12/30/03 3:20 PM Page 125


126


Subject Populations



METRONIDAZOLE

16 6 6 6 6 Aggressive 33(EOP)

17 5 5 5 5 Chronic 30

18 8 8 8 8 Chronic 44advanced

19 6 6 6, 6 6, 6 Chronic? 0


20 46 both groups 15 46 both groups 18 Advanced 52chronic (high spiros and/or BANApositive)

21 50 both groups 17 50 both groups 19 Chronic 51

4 66 overall 4 lost AMOX + MET = 15 66 overall 4 lost PLAC = 15 Advanced Not statedMET = 16 ` chronicAMOX = 16

22 12 12 11 11 Chronic Not stated

23 98 in all groups MET + SRP smokers = 3; 98 in all groups PLAC + SRP Chronic 50at start MET + SRP non-smokers = 9; at start smokers = 3;

MET + SURG smokers = 13; PLAC + SRP non-smokers = 1;MET + SURG non-smokers = 7 PLAC + SURG

smokers = 18;PLAC + SURG

non-smokers = 10

24 Unclear 31 Unclear 27 SRP 26 MET gel Chronic 48

25 90 in all groups 10 smokers; 21 non-smokers 90 in all groups SRP: 9 smokers + Moderate to 4818 non-smokers; advanced

SRP + MET gel: chronic9 smokers +18 non-smokers

PENICILLIN

26 8 8 8 8 Aggressive Not stated

8006.qxd 12/30/03 3:20 PM Page 126


127


Subject Populations

Mean AgeRacial/



METRONIDAZOLE

Not stated Not stated 25.7 19-35 Not stated Not stated Not stated Yes



37.3 37.2, 37.0 37.2 23-59 No Not stated Not stated No; trichomonasinfection intest group


Not stated Not stated Not stated Not stated Not stated Not stated Not stated Yes

Not stated Not stated Not stated Not stated Not stated Not stated Not stated Yes


Adolescents Adolescents Not stated Not stated Not stated Not stated Not stated Mentallyretardedadolescents

36.5 36.1 Not stated Not stated Not stated MET = 50% PLAC = 66% Yes

44.7 50.5 SRP; Not stated 35-65 No 32 33 SRP; Yes48.1 SRP + 35 SRP + MET gel MET gel

42.1 smokers; SRP 50 smokers, Not stated 35-65 No Stratifed by Stratifed by Yes46 non-smokers 50.7 non-smokers; smoking smoking

SRP + MET gel 48.3 smokers,48 non-smokers

PENICILLIN

Not stated Not stated Not stated


Subject PopulationOver 600 subjects were recruited in each of the test andcontrol groups (Table 3; page 124). The uncertainty isdue to the fact that in some studies the total number ofsubjects was reported, but it was not clear how manysubjects were assigned to the test and control groups.The designation of the disease type by the investigatorswas not entirely clear reflecting the change in the clas-sification of periodontal diseases over time. The major-ity of studies appeared to include subjects who would beclassified as chronic periodontitis today, although 8 stud-ies included subjects with “aggressive disease” and 2studies involved “refractory” periodontitis. The male-female ratio was described in 22 of the comparisons andaveraged about 41% males across studies. All but 3 ofthe comparisons provided either an overall mean age,mean age for subjects in each group, or an age range.One study described the participants as “adolescents.”In general, the racial/ethnic characteristics of the subjectpopulation were not described, while the smoking char-acteristics of the subjects were described for 10 com-parisons. Subjects in the majority of studies were sys-temically healthy, although test subjects in one studyhad a trichomonas infection and one study examinedmentally retarded subjects.

Clinical MeasurementsThe nature of the clinical measurements performed inthe studies varied (Table 4; page 130). Twenty studiesreported on measurements on all teeth in the oral cav-ity (which may or may not have included third molars),while 2 comparisons reported on measurements from8 teeth, 5 on measurements from 4 teeth, 4 compar-isons on measurements from 1 tooth and in 3 com-parisons the numbers of teeth reported in the analyseswere unclear. When subsets of teeth were employed,selection was frequently based on baseline probingdepth (11 comparisons), with or without concomitantdemonstration of bone loss (3 comparisons). The num-ber of sites examined per tooth differed, ranging from1 to 6. Three comparisons employed the Florida probe,4 a constant force probe, and the rest a manual probe.Stents were employed in 13 comparisons, while dupli-cate measures of probing depth and attachment levelat each site were employed in 8 comparisons.

TherapyThe therapeutic procedures employed in the studies wereremarkably heterogeneous (Table 5; page 132). Nineantibiotics or antibiotic combinations were employed,although MET alone (12 comparisons) or in combina-

128


Subject Populations



SPIRAMYCIN

27 96 in all groups TET = 27; SPIRA = 28 96 in all groups 24 Chronic Unclearadvanced

28 96 93 97 96 Chronic Not stated

TETRACYCLINE

29 6 6 6 6 Aggressive 58

30 7 7 7 7 Chronic 43

27 Unclear TET = 27; SPIRA = 28 Unclear 24 Chronic Unclearadvanced

31 42 35 99 80 Advanced 47

12 TET = 13 TET = 13 11 11 Active 55 PLAC, 62 TET;AMOX + AMOX + CA = 10 40 AMOX + CA

CA = 10

32 19 19 completed TET + 19 19 completed PLAC + Aggressive (LJP + GJP) 37SRP phase; SRP phase;

13 completed 13 completed SURG phase SURG phase

8006.qxd 12/30/03 3:20 PM Page 128


tion with AMOX (5 comparisons) was the most frequentlytested agent. The dosage and duration varied even withinindividual antibiotic regimens. MET was prescribed atdoses of 200 mg tid, 250 mg tid, 200 mg qid, 2 gmsonce, 200 mg per day, and 400 mg tid for durations thatincluded, 1, 7, 8, and 14 days. AMOX was administeredat dosages of 250 mg tid; 375 mg bid; 375 mg tid; and500 mg tid for 7, 8, or 14 days. AMOX + AC was givenat the dosage of 250 mg tid; 375 mg tid; 500 mg tid; or625 mg tid for 10, 14, or 30 days. CLIN was adminis-tered at the dosage of 150 mg qid for 8 days. DOXYwas given at the dosage of 100 mg per day for 14 or42 days (with a loading dose of 200 mg on Day 1), or200 mg a day for 8 days. PEN was administered at thedosage of 250 mg qid for 10 days. SPIR was given at thedosage of 1,500,000 IU bid for 14 days in 2 studies. TETwas prescribed at the dosage of 250 mg tid or 250 mgqid for 14, 21, or 30 days. One study employed 250 mgqid for 14 days followed by 250 mg once a day for 48weeks. In 3 studies, the antibiotic regimen was repeatedfrom 1 to 3 times. Mechanical debridement took placeafter antibiotic administration in 18 comparisons and wasgiven at the same time as SRP in 14 comparisons. Inthe remaining comparisons, there was either no mechan-ical debridement or the timing was unclear. Seven stud-

ies utilized a split-mouth design. The concomitantmechanical debridement also differed among studies.Eight comparisons involved no mechanical debridement(including 5 of the split-mouth studies). Six comparisonsinvolved surgical procedures and the remaining utilizedSRP as the concomitant and control therapy.

Compliance and Adverse EventsCompliance and adverse events were often not reportedin the cited studies (Table 6; page 136). Compliancewas measured in 11 comparisons using techniques thatincluded asking the patient, keeping a diary, pill counts,and supervised usage. Adverse events were recordedin 17 of the comparisons. They ranged from none in 6comparisons to 39% of subjects in the test groupexhibiting diarrhea when provided with AMOX plusMET. The most frequently reported adverse eventappeared to be diarrhea.

Excluded StudiesAs noted above, 51 of the full-text studies reviewed werenot included in the review. Studies were excluded for thefollowing reasons: 29 because they contained no attach-

129


Subject Populations

Mean AgeRacial/



SPIRAMYCIN

Not stated Not stated 46 ≥35 No Not stated Not stated Yes

48.5 47.3 Not stated ≥35 No Not stated Not stated Yes

TETRACYCLINE

Not stated Not stated 34 27-42 No Not stated Not stated Yes

Not stated Not stated Not stated 37-52 Not stated N/A N/A Yes

Not stated Not stated 46 ≥35 No Not stated Not stated Yes

41.2 42.1 41.7 24-60 No 60 56 Yes

44 TET; 48 Not stated 14-71 No Not stated Not stated Yes53 AMOX + CA

20.5 18.5 19.5 12-25 Not stated Not stated Not stated Yes


8006.qxd 12/30/03 3:20 PM Page 129


130

Table 4.

Clinical Assessment of Probing Depth and Attachment Levels

Number Teeth Selection Number Stent DuplicateReference Examined Criteria of Sites/ Tooth Type of Probe Used? Measures

AMOXICILLIN

4 All* N/A† 4 Manual No No


5 All N/A 6 Florida No Yes

6 All N/A 4 Manual Yes No

7 All N/A 6 Constant force No No

4 All N/A 4 Manual No No

8 All (only those See previous 4 Manual No Nowith PD ≥5 or CAL >2 in analyses)


9 All N/A 6 Constant force Yes No

10 1 active site Active = 1 active site Florida probe Yes Yesand 1 matched AL >3 SDs‡ and 1 matchednon-active site for the subject non-active site

11 4 Tooth with 6 Controlled force Unclear Yesdeepest PD electronicin each quadrant

12 All N/A 6 Manual No Yes

CLINDAMYCIN


10 1 active site Active = 1 active site Florida Yes Yesand 1 matched AL >3 SDs and 1 matchednon-active site for the subject non-active site

DOXYCYCLINE

14 All (or 10)? Unclear 6 Manual Yes No



METRONIDAZOLE

16 4 1 site with 6 Manual Yes Nobaseline (BPD) ≥5 mm

17 4 1 site with 6 Manual Yes NoBPD ≥5 mm

18 8 BPD ≥6 mm; 1 Manual No No≥50% bone loss

8006.qxd 12/30/03 3:20 PM Page 130


131


Clinical Assessment of Probing Depth and Attachment Levels

Number Teeth Selection Number Stent DuplicateReference Examined Criteria of Sites/ Tooth Type of Probe Used? Measures

19 2 BPD ≥5 mm 1 Manual No No





22 2 1 site on each 1 Manual Yes Yestooth withdeepest BPD of ≥5


24 All N/A 6 Florida No No

25 All, but only deep BPD ≥4.6 Unclear Florida No NoPD reported (Florida) or

≥5 mm manual

PENICILLIN


SPIRAMYCIN

27 4 1 pocket ≥7 mm 2 Manual (Williams) Yes No

28 2 BPD ≥7 mm 1 Manual Yes No

TETRACYCLINE


30 8 BPD ≥6 mm; 1 Manual No No≥40% bone loss; premolars and/or incissors

27 4 1 pocket ≥7 mm 2 Manual (Williams) Yes No



32 Affected Affected teeth 6 Constant force No Noteeth (PD ≥4 mm,

AL ≥2 mm;bone loss ≥4 mm)

* At least 28 teeth if present.† Not applicable or not available.‡ Standard deviations.

8006.qxd 12/30/03 3:20 PM Page 131


132

Table 5.

Antibiotic and Mechanical Debridement Procedures

Total dosage Reference Antibiotic 1 Dosage Duration (gms) Repeated? And/or Antibiotic 2 Dosage

AMOXICILLIN

4 AMOX 250 mg tid* 7 days 5.25 No Or MET (see 200 mg tidbelow)


5 MET 250 mg tid 7 days 5.25 (×3) = 15.75 Yes And AMOX 500 mg tid

6 MET 250 tid 14 days? 10.5 No And AMOX 375 mg bid

7 MET 250 mg tid 7 days 5.25 No And AMOX 375 mg tid

4 MET 200 mg tid 7 days 4.2 No And AMOX 250 mg tid

8 MET 750 mg/day 8 days 6.0 No And AMOX 1500 mg/day


9 AMOX + CA 625 mg tid 10 days 18.75 No – N/A N/A

10 AMOX + CA 250 mg AMOX + 14 days 15.75 No Or CLIND qid;‡ No dosage given125 mg CA tid

11 AMOX + CA 500 mg tid 14 days 21.0 No – N/A§ N/A

12 AMOX + CA 375 mg tid 30 days 33.75 No Or TET 250 mg tid (see below)

CLINDAMYCIN

13 CLIND 150 mg qid 8 days 4.8 No – N/A N/A

10 CLIND qid; no dosage 14 days – No Or AMOX + CA 250 mg AMOX + given 125 mg clavulanic

acid tid

DOXYCYCLINE

14 DOXY 200 mg day 1 6 weeks 4.3 No – N/A N/Athen 100 mgper day

15 DOXY 200 mg loading 14 days 1.5 No – N/A N/Adose, 100 mgper day

13 DOXY 200 mg/day 8 days 1.6 No Or MET 500 mg bid

8006.qxd 12/30/03 3:20 PM Page 132


133



Total dosage When Versus Concomitant Control Duration (gms) Repeated? Debridement Split Mouth? Therapy Therapy Maintenance?

AMOXICILLIN

7 days – No After No PLAC + SRP PLAC + UnclearPLAC + SRP


7 days 10.5 (×3) Yes (0, 4, N/A No None PLAC No= 31.5 8 months)

14 days 10.5 No During Yes SRP/None PLAC + SRP/None Unclear

7 days 7.9 No After (6 weeks) No SRP PLAC No

7 days 5.25 No After No SRP PLAC + UnclearPLAC + SRP

8 days 12.0 No After No SRP + Widman PLAC + SRP + OHI† + supra flap at deep Widman flap at and subgingival sites deep sites scaling at 1,

3 months andthen every3 months


N/A – N/A After (6 weeks) No SRP PLAC + SRP Yes 3 monthsintervals

10 days – No After No SRP PLAC + SRP Not stated

N/A – N/A After (2 months) No SRP SRP + local TET fibers Yes 3-4‡ weekintervals

30 days – No During active No SRP + MWF SRP + MWF Yes every treatment 3 months

CLINDAMYCIN

N/A – N/A After enhanced No SRP followed SRP followed by YesSRP by enhanced enhanced SRP

SRP

10 days No After No SRP PLAC + SRP Not stated

DOXYCYCLINE

N/A – N/A After Yes None; SRP PLAC; SRP Unclear

N/A – N/A During SRP No SRP SRP No

8 days – No After enhanced No SRP followed by SRP followed by YesSRP enhanced SRP enhanced SRP (continued)

8006.qxd 12/30/03 3:20 PM Page 133


134




METRONIDAZOLE

16 MET 250 mg tid 7 days 5.25 No – N/A N/A


18 MET 200 mg qid 14 days 11.2 (×3) Yes 0, 10, – N/A N/A= 33.6 20 weeks

19 MET 2 gm once Single dose 2.0 No – N/A N/A

13 MET 500 mg bid� 8 days 8.0 No Or DOXY 200 mg/day



4 MET 200 mg tid 7 days 4.2 No Or AMOX (see 250 mg tidabove)





PENICILLIN

26 PEN (phenoxymethyl) 250 mg qid 10 days 10.0 (×5) Yes at – N/A N/A(week 13) = 50 weeks 26,

38, 50and 62

SPIRAMYCIN

27 SPIRA 1,500,000 IU¶ 14 days 42,000,000 IU No Or TET 250 mg qidbid

28 SPIRA 1,500,000 IU 14 days 42,000,000 IU No – N/A N/Abid

TETRACYCLINE

29 TET 250 mg qid 14 days 14.0 (× 2) Yes; 5 weeks – N/A N/A= 28 later

30 TET 250 mg qid for 14 days then 98.0 No – N/A N/A14 days then 48 weeks250 mg oncea day for48 weeks

8006.qxd 12/30/03 3:20 PM Page 134


135




METRONIDAZOLE

N/A – N/A During SRP Yes None; SRP None; SRP OHI every2 weeks

N/A – N/A During SRP Yes None; SRP None; SRP OHI every2 weeks

N/A – N/A During Yes SRP; none SRP; none OHI at each monitoring visit

N/A – N/A No SRP No None SRP or none No

8 days – No After enhanced No SRP followed by SRP followed by YesSRP enhanced SRP enhanced SRP

N/A – N/A After SRP No SRP PLAC + SRP No

N/A – N/A During SRP No SRP PLAC + SRP No

7 days – No After No PLAC + SRP PLAC + PLAC + SRP Unclear

N/A – N/A After prophylaxis No Prophylaxis PLAC + prophylaxis Not stated

N/A – N/A After No SRP; SRP + SURG PLAC + SRP; PLAC + OHI + SRP everySRP + SURG 6 months

N/A – N/A Started on the No SRP SRP; SRP + MET gel Yes; prophylaxis +2nd of 2 SRP OHI at visits 4 weeks only

N/A – N/A After SRP No SRP SRP; SRP + MET gel Not stated(twice)

PENICILLIN

N/A – N/A During flap No SRP + flap surgery SRP + flap surgery + Yes every surgery and PLAC 3 monthsat recall visits

SPIRAMYCIN

14 days – No During SRP No SRP SRP + PLAC Not stated

N/A – N/A During SRP No SRP SRP + PLAC Not stated

TETRACYCLINE

N/A – N/A During first Yes SRP or none SRP or none Yesadministration

N/A – N/A During Yes SRP; none PLAC + SRP; OHI at each PLAC + none monitoring visit

(continued)

8006.qxd 12/30/03 3:20 PM Page 135


136




27 TET 250 mg qid 14 days 14.0 No Or SPIRA 1,500,000 IU bid

31 TET 250 mg qid 21 days 21.0 No – N/A N/A

12 TET 250 mg tid 30 days 22.5 No Or AMOX + CA 375 mg tid

32 TET 250 mg qid 14 days 14.0 No – N/A N/A

* 3 times a day.† Oral hygiene index.‡ 4 times a day.§ Not applicable or not available.� 2 times a day.¶ International units.# Modified Widman flap.

Table 6.

Description of Compliance and Adverse Events

Compliance Reference Measured How? Adverse Events (AE) AE % Affected Test AE % Affected Control

AMOXICILLIN

4 Yes Asked patient Reported None None


5 Yes Pill count Reported None None

6 No N/A* Not stated N/A N/A

7 Yes Pill count Diarrhea 39 None

7 Rash 9 4

7 Nausea after alcohol 4 None

7 Softened stools 0 8


8 Yes Daily supervision Reported None None


9 Yes Diary, returned Diarrhea 20 18.2meds, phone for compliance,AEs

10 No N/A No N/A N/A

11 No N/A Yes “Few” had “mild “Few” reported “pressure”GI symptoms” from fibers

12 No N/A Not stated N/A N/A

8006.qxd 12/30/03 3:21 PM Page 136


137




14 days – No During SRP No SRP SRP + PLAC Not stated

N/A – N/A During No SRP SRP Yes

30 days – No During active No SRP + MWF SRP + MWF# Yes every treatment 3 months

N/A – N/A After No SRP + SURG at PLAC + SRP + SURG Not statedsites with PD at sites with PD ≥5 mm + BOP ≥5 mm + BOP




CLINDAMYCIN

13 No N/A Not reported N/A N/A


DOXYCYCLINE

14 Yes? Reinforced verbally Reported None None



METRONIDAZOLE




19 Unclear Unclear Not stated N/A N/A


20 Yes Oral report by subject Yes Occasional metallic taste Unclear

21 Yes Oral report by subject Yes Legs cramps 6% None


22 Yes Supervised usage Not reported N/A N/A




PENICILLIN

26 Unclear N/A Not stated N/A N/A (continued)

8006.qxd 12/30/03 3:21 PM Page 137


138




SPIRAMYCIN

27 Yes Pill count Reported 3.7 TET Not stated

28 Yes Pill count Reported 1.1 periodontal symtoms; 1.0 periodontalGI upset/diarrhea 2.2 symptoms

TETRACYCLINE



27 Yes Pill count Reported 3.7 TET Not stated

31 No N/A None reported N/A N/A



* Not available.

Figure 1.Meta-analysis of mean attachment level changes by patient periodontal disease diagnosis (aggressive or chronic).

8006.qxd 12/30/03 3:21 PM Page 138


139

ment level measurements or included uninterpretabledata;33-61 11 because there was no comparison groupor that group included another antibiotic;62-72 5 becauseguided tissue regeneration barriers or other materialswere inserted into the defects;73-77 3 because another ver-sion of the same study and data were selected;78-80 and3 because they were reviews or meta-analyses.81-83

MAIN RESULTSTwenty-nine studies involving 36 comparisons werefound to be eligible for inclusion after review (Table 1;page 118).4-32 Tables 7 through 12 present the attach-ment level and probing depth findings from these stud-ies.

Meta-Analysis Results Using Mean AttachmentLevel ChangeOf the 36 comparisons described in the tables, only27 were selected for meta-analyses. One study4 involv-ing 3 comparisons and a second study8 involving onecomparison were omitted because the attachmentlevel change was provided only as percent of sites. Twostudies were excluded because no SD or SEM were pro-vided.17,29 Two studies that evaluated 1 site per subjectwere excluded.10,19 Finally, one study that provided data

only at 5 years after treatment was excluded.23 Themeta-analyses emphasized full-mouth mean attachmentlevel change rather than change at selected subsets ofsites that may have been provided by the authors. Threemeta-analyses were carried out to evaluate the data.The analyses evaluated if antibiotics were consistentlyadjunctive for different patient populations, adjunctiveto different therapies, and different antibiotics.

Figure 1 (page 138) is the forest plot of the analy-sis divided by patient population. For all of the stud-ies and for the 2 sets of studies divided by the popu-lation studied, the antibiotic groups showed statisticallysignificantly better results than the control groups. Alltests of heterogeneity were not significant, indicatingthat the outcomes of the studies were consistent. Addi-tionally, the results were significantly different betweenpatient populations, indicating that the aggressive peri-odontitis groups had a larger adjunctive benefit.

Figure 2 separates the adjunctive effect by themechanical therapy employed. In the case of therapiesadjunctive to SRP, there was a significant benefit for theuse of antibiotics and the studies provided a consis-tent finding as indicated by a non-significant test forheterogeneity. The analysis was less clear for the

Figure 2.Meta-analysis of mean attachment level change grouped by type of mechanical therapy.

8006.qxd 12/30/03 3:21 PM Page 139


140

Figure 3.Meta-analysis of mean attachment level changes grouped by type of antibiotic used.

Table 7.

Mean Attachment Level and/or Mean Attachment Level Change at Different Post-Treatment Time

Control

Mean AL Mean AL Reference Time Site Type Treatment Baseline SD/SEM at End SD/SEM Change*


5 2 months All PLAC −0.13 ± 0.18 sd




8006.qxd 12/30/03 3:21 PM Page 140


141

adjunctive effect of antibiotics to either no other mechan-ical therapy or adjunctive to SRP plus surgery. In the 4studies with no mechanical therapy,5,7,18,22 there was asignificant test for heterogeneity and a significant dif-ference between groups for the random-effects model of0.093. While one may question combining these 4 stud-ies due to the significant test for heterogeneity, all of thestudies favored the antibiotic groups and the overall testfor significance was borderline. The significance levelfor studies adjunctive to SRP and surgical therapy was0.055. While a borderline significance level, the studieswere consistent by a lack of a significant test for het-erogeneity. Thus the data did support a case for anadjunctive effect in these 4 studies.

Figure 3 shows the results by the type of antibiotic.While it would have been desirable to determine whichantibiotic was the most effective, it is difficult becauseof the number of antibiotics (8) for the 27 comparisonsin the analysis. The combination of all of the studies wassignificant and the studies were consistent as noted pre-viously. Furthermore, the effect for studies grouped bythe type of antibiotic was similar ranging from 0.188 to0.555, excluding the 2 studies that used SPIR that hadan effect of 0.06.27,28 It appears that statistical powerdetermined which antibiotics demonstrated a significantadjunctive effect. There were significant adjunctive effectsfor TET (5 studies; 135 subjects)12,27,30-32 and MET (9studies; 125 subjects)13,16,18,20-22,24,25 with non-signifi-cant tests of heterogeneity. The remaining antibiotics inwhich the number of studies ranged from 1 to 3 and thenumber of subjects ranged from 8 to 120 lacked signif-icance, with a borderline effect for AMOX plus MET.Only the 2 studies of SPIR had sufficient subjects(i.e., 120) to suggest that the adjunctive effect of theantibiotic was ineffective.27,28

Mean Attachment Level ChangeThe data describing mean attachment level change atall sites and subsets of sites are presented in Table 7

(page 140). The meta-analyses describing the overallchanges in the subjects were presented above. Severalstudies reported mean AL changes at sites subsetaccording to baseline probing depth. In some instances,the data were provided at multiple time points. Basedon data closest to the 6-month time point and usingany of the antibiotics, it was found that at sites withshallow baseline probing depths (


142



Control




6 2 months All SRP + PLAC 0.6 ± 0.2 sd





6 2 months BPD 4-5 mm SRP + PLAC 0.6 ± 0.2 sd



6 2 months BPD >5 mm SRP + PLAC 1.0 ± 0.2 sd



7 3 months All PLAC 4.0 1.3 sd 3.6 1.1 sd 0.4

7 3 months BPD 0-3 mm PLAC −0.31 ± 0.29 sd

7 3 months BPD 4-6 mm PLAC 0.68 ± 0.44 sd

7 3 months BPD >6 mm PLAC 1.46 ± 0.70 sd


9 3 months All SRP 7.3 0.3 sd 6.8 0.3 sd 0.5



9 12 months BPD 0-3 mm SRP 0.1

9 12 months BPD 4-6 mm SRP 0.6

9 12 months BPD >6 mm SRP 1.7

10 2 years 1 active PLAC + SRP 0.84 ± 0.43 sd

10 2 years 1 control PLAC + SRP −0.35 ± 0.19 sd

11 15 weeks 4 teeth SRP + Local TET 12.0 1.8 sd 11.6 1.8 sd 0.4




12 10 months All SRP + MWF‡ + PLAC 0.02 ± 0.11 sem

12 10 months BPD


143



Test

Mean AL Mean ALTreatment Baseline SD/SEM at End SD/SEM Change* Difference†

AMOX + MET 0.45 ± 0.35 sd 0.83

AMOX + MET 0.46 ± 0.38 sd 0.89

MET + AMOX + SRP 0.6 ± 0.4 sd 0

MET + AMOX + SRP 0.8 ± 0.4 sd 0.1

MET + AMOX + SRP 1.1 ± 0.3 sd 0.3

MET + AMOX 0.2 ± 0.2 sd 0.3

MET + AMOX 0.2 ± 0.2 sd 0.5

MET + AMOX + SRP 0.4 ± 0.3 sd −0.2

MET + AMOX + SRP 0.6 ± 0.3 sd −0.1

MET + AMOX + SRP 0.9 ± 0.3 sd 0.2

MET + AMOX + SRP 1.3 ± 0.6 sd 0.3

MET + AMOX + SRP 1.8 ± 0.5 sd 0.5

MET + AMOX + SRP 2.1 ± 0.4 sd 0.6

AMOX + MET 3.9 1.1 sd 3.2 1.0 sd 0.7 0.3

AMOX + MET −0.14 ± 0.32 sd 0.17

AMOX + MET 0.88 ± 0.38 sd 0.20

AMOX + MET 1.97 ± 0.79 sd 0.51


AMOX + CA + SRP 7.4 0.9 sd 6.9 0.8 sd 0.5 0

AMOX + CA + SRP 7.4 0.9 sd 6.8 0.8 sd 0.6 0

AMOX + CA + SRP 7.4 0.9 sd 6.9 0.9 sd 0.5 0

AMOX + CA + SRP 0.1 0

AMOX + CA + SRP 0.7 0.1

AMOX + CA + SRP 1.1 −0.6

AMOX + CA + SRP 1.0 ± 1.42 sd 0.16

AMOX + CA + SRP −0.13 ± 1.29 sd 0.22

SRP + AMOX + CA 12.3 1.5 sd 12.1 1.5 sd 0.2 −0.2

SRP + AMOX + CA 12.3 1.5 sd 11.4 1.1 sd 0.9 0.4

SRP + AMOX + CA 12.3 1.5 sd 11.4 1.2 sd 0.9 −0.4

SRP + AMOX + CA 12.3 1.5 sd 11.2 1.2 sd 1.1 0.4

AMOX + CA + SRP + MWF‡ 0.68 ± 0.23 sem 0.66

AMOX + CA + SRP + MWF 0.16 ± 0.20 sem 0.41

(continued)

8006.qxd 12/30/03 3:21 PM Page 143


144



Control


12 10 months BPD 4-6 mm SRP + MWF + PLAC 0.27 ± 0.20 sem

12 10 months BPD >6 mm SRP + MWF + PLAC 1.73 ± 0.44 sem

CLINDAMYCIN



13 6 months BPD ≥6 mm SRP 9.6 0.86 sd 8.2 0.77 sd 1.4


10 2 years 1 active PLAC + SRP 0.84 ± 0.43 sd

10 2 years 1 control PLAC + SRP −0.35 ± 0.19 sd

DOXYCYCLINE

14 3 weeks All PLAC + SRP 9.0 1.9 sd 10.3 2.4 sd −1.3




14 3 weeks All PLAC 9.5 2.3 sd 10.0 1.1 sd −0.5

14 6 weeks All PLAC 9.5 2.3 sd 9.2 0.4 sd 0.3



15 90 days All SRP 3.15 0.93 sd 3.19 1.17 sd −0.04





METRONIDAZOLE

16 42 days 24 sites None 9.0 1.2 sd 9.1 1.5 sd −0.1 ± 0.3 sdmeasured

16 42 days 24 sites SRP 9.1 2.0 sd 8.4 2.1 sd 0.8 ± 0.3 sdmeasured

17 42 days 24 sites None 8.73 – 8.72 – 0.01measured

17 42 days 24 sites SRP 9.56 – 8.97 – 0.59measured

18 2 weeks BPD ≥6 mm SRP 0.2 ± 0.2 sem

8006.qxd 12/30/03 3:21 PM Page 144


145



Test




CLINDAMYCIN

CLIND + SRP 6.1 0.96 sd 4.4 1.00 sd 1.7 1.1

CLIND + SRP 6.1 0.96 sd 4.2 0.87 sd 1.9 1.5

CLIND + SRP 10.4 1.04 sd 7.4 1.00 sd 3.0 1.6

CLIND + SRP 10.4 1.04 sd 7.5 1.17 sd 2.9 2.4

CLIND + SRP −0.15 ± 0.14 sd −0.99

CLIND + SRP −0.48 ± 0.61 sd −0.13

DOXYCYCLINE

DOXY + SRP 7.8 1.2 sd 7.6 0.4 sd 0.2 1.5

DOXY + SRP 7.8 1.2 sd 7.5 0.6 sd 0.3 0.9

DOXY + SRP 7.8 1.2 sd 7.5 0.5 sd 0.3 0.8

DOXY + SRP 7.8 1.2 sd 7.4 0.5 sd 0.4 1.3

DOXY 8.0 1.1 sd 8.1 0.7 sd −0.1 0.4

DOXY 8.0 1.1 sd 8.0 0.4 sd 0 −0.3

DOXY 8.0 1.1 sd 7.9 0.4 sd 0.1 0.6

DOXY 8.0 1.1 sd 7.9 0.8 sd 0.1 1.0

DOXY + SRP 3.13 0.89 sd 3.00 0.84 sd 0.13 0.17

DOXY + SRP 6.0 1.05 sd 4.8 0.83 sd 1.2 0.6

DOXY + SRP 6.0 1.05 sd 5.1 0.9 sd 0.9 0.5

DOXY + SRP 10.1 1.21 sd 8.1 0.83 sd 2.0 0.6

DOXY + SRP 10.1 1.21 sd 8.7 1.02 sd 1.4 0.9

METRONIDAZOLE

MET 8.9 1.1 sd 8.6 1.2 sd 0.3 ± 0.3 sd 0.4

MET + SRP 8.7 1.1 sd 7.9 1.4 sd 0.8 ± 0.4 sd 0

MET 8.68 – 8.31 – 0.37 0.36

MET + SRP 9.61 – 8.61 – 1.0 0.41

MET + SRP −0.1 ± 0.1 sem −0.3

(continued)

8006.qxd 12/30/03 3:21 PM Page 145


146



Control






18 2 weeks BPD ≥6 mm None 0.2 ± 0.4 sem

18 10 weeks BPD ≥6 mm None 0.1 ± 0.1 sem

18 20 weeks BPD ≥6 mm None −0.1 ± 0.3 sem



19 1 month All (2 sites) SRP 5.5 2.3 sd 4.5 1.7 sd 1.0

19 3 months All (2 sites) SRP 5.5 2.3 sd 4.3 1.5 sd 1.2

19 3 months All (2 sites) None 6.0 1.2 sd 5.8 1.3 sd 0.2



13 6 months BPD ≥6 SRP 9.6 0.86 sd 8.2 0.77 sd 1.4

13 24 months BPD ≥6 SRP 9.6 0.86 sd 9.1 1.23 sd 0.5

20 4-6 weeks BPD ≤3 mm SRP −0.33 ± 0.19 sem

20 4-6 weeks BPD 4-6 mm SRP 0.32 ± 0.16 sem

20 4-6 weeks BPD ≥7 mm SRP 1.03 ± 0.33 sem

21 4-6 weeks BPD ≤3 mm SRP −0.15 ± 0.09 sem

21 4-6 weeks BPD 4-6 mm SRP 0.27 ± 0.16 sem

21 4-6 weeks BPD ≥7 mm SRP 0.54 ± 0.26 sem

22 1 month 2 sites PLAC + prophylaxis 7.81 1.29 sd 7.21 1.17 sd 0.6

22 3 months 2 sites PLAC + prophylaxis 7.81 1.29 sd 7.53 1.37 sd 0.28

22 6 months 2 sites PLAC + prophylaxis 7.81 1.29 sd 7.47 1.63 sd 0.34

23 5 years All smokers PLAC + SRP 4.2 0.7 sd 3.3 0.4 sd 0.9

23 5 years All Non-smokers PLAC + SRP 6.3 Only 1 3.0 3.3subject

23 5 years All smokers PLAC + SRP + SURG 4.1 0.8 sd 4.2 0.9 sd −0.1

23 5 years All Non-smokers PLAC + SRP + SURG 4.6 0.7 sd 3.6 0.6 sd 1.0

24 8 weeks BPD ≥5 mm SRP 0.36 ± 0.35 sd

24 8 weeks BPD ≥5 mm SRP + MET gel 0.39 ± 0.52 sd

8006.qxd 12/30/03 3:21 PM Page 146


147



Test


MET + SRP 0.4 ± 0.1 sem −0.3

MET + SRP 0.9 ± 0.2 sem −0.2

MET + SRP 1.5 ± 0.2 sem 0.2

MET + SRP 1.8 ± 0.2 sem 0.2

MET 0.1 ± 0.1 sem −0.1

MET 0.2 ± 0.1 sem 0.1

MET 0.4 ± 0.2 sem 0.5

MET 0.5 ± 0.1 sem 0.8

MET 0.8 ± 0.2 sem 1.1

MET 5.9 1.0 sd 5.8 1.3 sd 0.1 −0.9

MET 5.9 1.0 sd 5.4 1.0 sd 0.5 −0.7

MET 5.9 1.0 sd 5.4 1.0 sd 0.5 0.3

MET + SRP 6.2 1.00 sd 4.3 0.74 sd 1.9 1.3

MET + SRP 6.2 1.00 sd 4.0 1.07 sd 2.2 1.8

MET + SRP 9.8 1.62 sd 6.7 1.22 sd 3.1 1.7

MET + SRP 9.8 1.62 sd 6.4 1.12 sd 3.4 2.9

MET + SRP 0.04 ± 0.19 sem 0.37

MET + SRP 0.79 ± 0.16 sem 0.47

MET + SRP 1.69 ± 0.33 sem 0.66

MET + SRP −0.05 ± 0.09 sem 0.10

MET + SRP 0.40 ± 0.16 sem 0.13

MET + SRP 0.86 ± 0.26 sem 0.32

MET + Prophylaxis 7.21 0.85 sd 6.48 0.82 sd 0.73 0.13



MET + SRP smokers 4.2 0.6 sd 3.2 0.7 sd 1.0 0.1

MET + SRP non-smokers 3.8 0.4 sd 3.1 0.4 sd 0.7 −2.6

MET + SRP + SURG smokers 4.4 0.5 sd 4.4 0.8 sd 0 0.1

MET + SRP + SURG 4.0 0.4 sd 3.8 0.7 sd 0.2 −0.8

MET + SRP 0.59 ± 0.51 sd 0.23

MET + SRP 0.59 ± 0.51 sd 0.20

(continued)

8006.qxd 12/30/03 3:21 PM Page 147


148



Control


24 24 weeks BPD ≥5 mm SRP 0.51 ± 0.43 sd

24 24 weeks BPD ≥5 mm SRP + MET gel 0.47 ± 0.65 sd

25 2 months BPD ≥5 Smokers SRP 0.26 ± 0.47 sd

25 6 months BPD ≥5 Smokers SRP 0.47 ± 0.46 sd

25 2 months BPD ≥5 Smokers SRP + MET gel 0.50 ± 0.26 sd

25 6 months BPD ≥5 Smokers SRP + MET gel 0.46 ± 0.25 sd

25 2 months BPD ≥5 Non- SRP 0.41 ± 0.28 sdsmokers

25 6 months BPD ≥5 Non- SRP 0.53 ± 0.43 sdsmokers

25 2 months BPD ≥5 Non- SRP + MET gel 0.34 ± 0.60 sdsmokers

25 6 months BPD ≥5 Non- SRP + MET gel 0.48 ± 0.80 sdsmokers

PENICILLIN

26 26 weeks All SRP + SURG 0.93 0.19 sem 0.83 0.24 sem 0.10




26 26 weeks BAL ≥2 mm SRP + SURG 3.68 0.34 sem 2.49 0.39 sem 1.19




26 62 weeks BPD ≥6 mm SRP + SURG 2.52 ± 0.32 sem

26 62 weeks BPD ≤5 mm SRP + SURG 1.08 ± 0.17 sem

SPIRAMYCIN

27 2 weeks BPD 1-3 mm PLAC + SRP 7.50 0.84 sem 7.90 0.93 sem −0.40


27 12 weeks BPD 1-3 mm PLAC + SRP 7.50 0.84 sem 7.36 0.90 sem 0.14





8006.qxd 12/30/03 3:21 PM Page 148


149



Test


MET + SRP 0.67 ± 0.67 sd 0.16

MET + SRP 0.67 ± 0.67 sd 0.20

MET + SRP 0.32 ± 0.26 sd 0.06

MET + SRP 0.43 ± 0.57 sd −0.04

MET + SRP 0.32 ± 0.26 sd −0.18

MET + SRP 0.43 ± 0.57 sd −0.03

MET + SRP 0.72 ± 0.55 sd 0.31

MET + SRP 0.79 ± 0.70 sd 0.26

MET + SRP 0.72 ± 0.55 sd 0.38

MET + SRP 0.79 ± 0.70 sd 0.31

PENICILLIN

PEN + SRP + SURG 1.39 0.21 sem 1.05 0.16 sem 0.34 0.24



PEN + SRP + SURG 1.39 0.21 sem 1.00 0.20 sem 0.39 −0.07





PEN + SRP + SURG 2.50 ± 0.25 sem −0.02

PEN + SRP + SURG 0.73 ± 0.14 sem −0.35

SPIRAMYCIN

SPIR + SRP 7.29 0.57 sem 7.08 0.50 sem 0.21 0.61

SPIR + SRP 7.29 0.57 sem 6.94 0.49 sem 0.35 0.55

SPIR + SRP 7.29 0.57 sem 6.75 0.49 sem 0.54 0.40

SPIR + SRP 7.29 0.57 sem 6.67 0.67 sem 0.62 0.92

SPIR + SRP 9.11 0.27 sem 8.80 0.26 sem 0.31 0.00

SPIR + SRP 9.11 0.27 sem 8.09 0.30 sem 1.02 0.39

SPIR + SRP 9.11 0.27 sem 8.32 0.30 sem 0.79 −0.14

(continued)

8006.qxd 12/30/03 3:21 PM Page 149


150



Control



27 2 weeks BPD ≥7 mm PLAC + SRP 10.75 0.29 sem 9.81 0.29 sem 0.94








TETRACYCLINE

29 8 weeks All None 0.02

29 8 weeks All SRP 0.31

29 25 weeks All None 0.05

29 25 weeks All SRP 0.30

30 10 weeks BPD ≥6 mm PLAC + SRP 0.6 ± 0.2 sem




30 10 weeks BPD ≥6 mm PLAC 0.1 ± 0.2 sem

30 20 weeks BPD ≥6 mm PLAC 0.0 ± 0.3 sem

30 30 weeks BPD ≥6 mm PLAC −0.1 ± 0.3 sem

30 50 weeks BPD ≥6 mm PLAC −0.4 ± 0.4 sem



27 12 weeks BPD 1-3 PLAC + SRP 7.50 0.84 sem 7.36 0.90 sem 0.14

27 24 weeks BPD 1-3 PLAC + SRP 7.50 0.84 sem 7.80 0.82 sem −0.30





27 2 weeks BPD ≥7 PLAC + SRP 10.75 0.29 sem 9.81 0.29 sem 0.94


8006.qxd 12/30/03 3:21 PM Page 150


151



Test


SPIR + SRP 9.11 0.27 sem 8.31 0.30 sem 0.80 −0.22

SPIR + SRP 10.60 0.27 sem 9.75 0.30 sem 0.85 −0.09

SPIR + SRP 10.60 0.27 sem 9.30 0.29 sem 1.30 0.18

SPIR + SRP 10.60 0.27 sem 9.14 0.28 sem 1.46 0.02

SPIR + SRP 10.60 0.27 sem 9.14 0.26 sem 1.46 −0.08

SPIRA + SRP 10.72 0.15 sem 9.74 0.16 sem 0.98 0.21

SPIRA + SRP 10.72 0.15 sem 9.13 0.18 sem 1.59 0.32

SPIRA + SRP 10.72 0.15 sem 8.87 0.20 sem 1.85 0.46

SPIRA + SRP 10.72 0.15 sem 8.85 0.16 sem 1.87 0.29

TETRACYCLINE

TET 0.23 0.21

TET + SRP 0.25 −0.06

TET 0.32 0.27

TET + SRP 0.49 0.19

TET + SRP 1.0 ± 0.5 sem 0.4

TET + SRP 1.3 ± 0.4 sem 0.2

TET + SRP 1.4 ± 0.6 sem 0.0

TET + SRP 1.7 ± 0.3 sem 0.3

TET 0.2 ± 0.3 sem 0.1

TET 0.4 ± 0.3 sem 0.4

TET 0.6 ± 0.4 sem 0.7

TET 0.7 ± 0.3 sem 1.1

TET + SRP 7.25 0.46 sem 7.81 0.46 sem −0.56 −0.16

TET + SRP 7.25 0.46 sem 7.14 0.45 sem 0.11 0.31

TET + SRP 7.25 0.46 sem 7.31 0.45 sem −0.06 −0.20

TET + SRP 7.25 0.46 sem 7.06 0.55 sem 0.19 0.49

TET + SRP 9.06 0.26 sem 8.76 0.31 sem 0.30 −0.01

TET + SRP 9.06 0.26 sem 8.24 0.27 sem 0.82 0.19

TET + SRP 9.06 0.26 sem 8.19 0.29 sem 0.87 −0.06

TET + SRP 9.06 0.26 sem 8.00 0.27 sem 1.06 0.04

TET + SRP 10.58 0.25 sem 9.82 0.28 sem 0.76 −0.18

TET + SRP 10.58 0.25 sem 8.90 0.28 sem 1.68 0.56

(continued)

8006.qxd 12/30/03 3:21 PM Page 151


152



Control




31 12 months All SRP 0.16 ± 0.5 sd

12 10 months All SRP + MWF + PLAC 0.02 ± 0.11 sem

12 10 months BPD 6 SRP + MWF + PLAC 1.73 ± 0.44 sem

32 3 months All affected teeth PLAC + SRP 2.66 1.18 sd 2.31 1.27 sd 0.35

32 3 months All affected teeth PLAC + SRP + SURG 2.62 0.90 sd 2.31 1.07 sd 0.31



32 3 months Proximal sites of PLAC + SRP 3.22 1.25 sd 2.74 1.36 sd 0.48affected teeth

32 3 months Proximal sites of PLAC + SRP + SURG 3.25 0.99 sd 2.77 1.13 sd 0.48affected teeth



* A negative value in the Change column indicates improvement.† A positive value in the Difference column favors test over control.‡ Modified Widman Flap.

Table 8.

Percent of Sites Gaining or Losing Attachment Over Various Thresholds at Different Times

Control Test

Reference Time Site Type Threshold Treatment % Sites SD/SEM Treatment % Sites SD/SEM


5 Baseline All AL loss ≥2 mm PLAC 3.14 1.45 sd AMOX + MET 3.3 1.63 sd

5 2 months All AL loss ≥2 mm PLAC 3.06 2.09 sd AMOX + MET 1.17 1.24 sd





8006.qxd 12/30/03 3:21 PM Page 152


therapy (Table 9; page 156). Only one study with 3comparisons presented data in this format.4 The dataindicated that AMOX alone, MET alone, or the 2 antimi-crobials used in combination reduced the percent ofsites with attachment level >6 mm and increased thepercent of sites with attachment level


154



Control Test



5 Baseline All AL gain ≥2 mm PLAC 0.53 0.78 sd AMOX + MET 0.57 1.03 sd

5 2 months All AL gain ≥2 mm PLAC 0.60 1.35 sd AMOX + MET 1.48 1.18 sd






6 2 months BPD 4-5 mm AL gain ≥2 mm SRP + 11 7 sd MET + AMOX + 7 7 sdPLAC SRP



6 2 months BPD >5 mm AL gain ≥2 mm SRP + 21 7 sd MET + AMOX + 40 18 sdPLAC SRP



8 12 months All “diseased AL gain >2 mm PLAC + 15.82 7.53 sd AMOX + MET + 34.70 11.67 sdsites” MWF + SRP + MWF†

SRP

8 12 months All “diseased AL loss >2 mm PLAC + 24.32 19.98 sd AMOX + MET + 10.04 11.59 sdsites” SRP + SRP + MWF

MWF


10 24 months All AL loss ≥3 SDs + PLAC + 11.3 (N 5.7 sd AMOX + CA + 4.1 (N 5.7 sd0.5 mm SRP not %) SRP not %)increase in PD

10 24 months All AL gain ≥3 SD + PLAC + 0.3 (N 6.0 sd AMOX + CA + 7.4 (N 6.0 sd0.5 mm SRP not %) SRP not %)decrease in PD

12 10 months All AL loss ≥2 mm SRP + 7.74 2.40 sem AMOX + CA + 3.83 1.70 semMWF + SRP + MWFPLAC

12 10 months All AL gain ≥2 mm SRP + 7.27 1.47 sem AMOX + CA + 23.44 6.61 semMWF + SRP + MWFPLAC

8006.qxd 12/30/03 3:21 PM Page 154


155


Percent of Sites Gaining or Losing AttachmentOver Various Thresholds At Different Times

Difference* Clincally Favors Test/Control

2.41 Test

0.04 Test

0.88 Test

1.44 Test

0.75 Test

1.04 Test

1.67 Test

1.88 Test

−4 Control

0 Same

3 Test

19 Test

18 Test

21 Test

18.88 Test

14.28 Test


7.2 Test

7.1 Test

3.91 Test

16.17 Test

(continued)

Change in Percent of Sites Increasing or Decreasingin Probing DepthTwo studies8,12 (with 3 comparisons) described dataas the percentage of sites that showed probing depthincreasing or decreasing over certain millimeter thresh-olds (Table 11; page 168). All comparisons presentedfavored the test over the control.

Percent of Sites in Different Probing DepthCategoriesTable 12 (page 170) presents the percentage of sitesin different probing depth categories at baseline and atdifferent time-points post-therapy. Eight studies with 10comparisons provided data in this for-mat.4,7,9,18,24,30,31,32 When data were limited to obser-vations closest to 6 months, the test was favored overthe control in 9 of 9 comparisons examining an increasein percentage of shallow pockets and in 8 of 12instances reporting a decrease in percentage of pock-ets ≥5 mm.

DISCUSSIONOutcome MeasureThe primary outcome measure employed in this reviewwas clinically determined attachment level. This variableappears to be the most commonly employed for studiesof the usefulness of adjunctive antimicrobial therapyand has gained universal acceptance by the profes-sion. A secondary outcome of probing depth was alsoincluded although no formal meta-analyses were car-ried out using this parameter. Other secondary clinicalmeasurements, such as plaque, gingival redness, sup-puration, or bleeding on probing, were not includedsince they were less consistently reported and may beof less importance. Another widely accepted primaryoutcome variable would be change in alveolar boneheight or density. However, since this variable was infre-quently assessed in clinical trials involving systemicadministration of antibiotics, it was not evaluated. Aninitial attempt was made to evaluate microbiologicalchanges that accompanied the use of systemicallyadministered antimicrobial agents. However, the datawere sparse in that limited numbers of studies, sam-ples, and species were reported. The interested readermay seek a recent comprehensive review by Slots andTing.84

Experimental DesignOne distressing feature of performing this systematicreview was the heterogeneity of the experimentaldesigns employed to assess the efficacy of systemicallyadministered antimicrobial agents. The designs variedenormously in terms of the antibiotics or antibioticcombinations employed; the dosage and duration ofadministration; whether the antibiotic administrationwas repeated or sustained over a long period of time;

8006.qxd 12/30/03 3:21 PM Page 155


156



Control Test


CLINDAMYCIN

10 24 months All AL loss ≥3 SD + PLAC + SRP 11.3 (N 5.7 sd CLIND + SRP 7.0 (N 5.7 sd0.5 mm not %) not %)increase in PD

10 24 months All AL gain ≥3 SD + PLAC + SRP 0.3 (N 6.0 sd CLIND + SRP 5.5 (N 6.0 sd0.5 mm not %) not %)decrease in PD

PENICILLIN

26 26 weeks BAL ≥2 mm AL gain ≥2 mm SRP + SURG 44.8 9.4 sem PEN + SRP + SURG 50.0 7.8 sem




26 26 weeks BAL ≥2 mm AL loss ≥2 mm SRP + SURG 6.4 2.3 sem PEN + SRP + SURG 9.2 2.2 sem




TETRACYCLINE

12 10 months All AL loss ≥2 mm SRP + MWF + 7.74 2.40 sem TET + SRP + MWF 4.08 1.33 semPLAC

12 10 months All AL gain ≥2 mm SRP + MWF + 7.27 1.47 sem TET + SRP + MWF 15.73 4.92 semPLAC

* A positive value in the Difference column favors test over control.† Modified Widman flap.

Table 9.

Percent Sites in Different Attachment Level Categories at Different Times

Control

Reference Time point Site Type Treatment % at Baseline SD/SEM % at End SD/SEM Change*

AMOXICILLIN

4 1 month AL ≤3 mm PLAC + PLAC + SRP 48.0 20.7 sd 55.1 22.2 sd 7.1

4 3 months AL ≤3 mm PLAC + PLAC + SRP 48.0 20.7 sd 51.6 21.8 sd 3.6

4 6 months AL ≤3 mm PLAC + PLAC + SRP 48.0 20.7 sd 56.3 20.7 sd 8.3

4 1 month AL ≥6 mm PLAC + PLAC + SRP 24.3 13.8 sd 18.4 12.4 sd 5.9

8006.qxd 12/30/03 3:21 PM Page 156


the nature of concomitant mechanical debridement;the timing of mechanical therapy in relation to antimi-crobial usage; the duration and timing of post-therapymonitoring; the number of subjects and the number ofsites evaluated per subject and the nature of sitesselected; the nature of the periodontal diseases beingtreated; severity of disease; the subject population eval-uated; the nature of the probing techniques; the use ofsplit-mouth designs; and the degree to which mask-ing of subject, therapist, and examiner was carried out.This formidable list of differences among studies com-plicated data analysis and led to data aggregationsthat may not have been entirely optimal. Thus, analy-ses performed examined all of the antibiotics togetheras well as subsets of antibiotics. Another problem wasthat the duration of monitoring differed and somestudies provided multiple time points of post-therapymonitoring. A time of 6 months was chosen for themeta-analysis since this was close to the median valuefor the time-points provided. Unfortunately, not all stud-ies provided 6-month data and thus, the closest datato that time point were employed.

Number of SubjectsThe number of subjects in the studies examined in thisreview ranged widely from 4 to 96 subjects per groupwith a median value of 13 subjects per group. Manyof the studies concluded that there was no “statisti-cally significant” difference between test and controlgroups although there were frequently better outcomesfor the test group, as summarized in this review. Thelack of statistical significance for the studies may havebeen due to too few subjects; i.e., the studies wereunder-powered. This led to the meta-analyses per-formed in the present review. The low number of sub-

157


Percent of Sites Gaining or Losing AttachmentOver Various Thresholds At Different Times

Difference* Clincally Favors Test/Control

CLINDAMYCIN

4.3 Test

5.2 Test

PENICILLIN

5.2 Test

4.8 Test

1.2 Test

−6.2 Control

−2.8 Control

−3.0 Control

−1.9 Control

−5.4 Control

TETRACYCLINE

3.66 Test

8.46 Test


Percent Sites in Different Attachment Level Categories at Different Times

Test

Treatment % at Baseline SD/SEM % at End SD/SEM Change* Difference†

AMOXICILLIN

AMOX + PLAC + SRP 55.8 14.8 sd 63.4 19.6 sd 7.6 0.5




(continued)


8006.qxd 12/30/03 3:21 PM Page 157


Percent Sites

systemic anti-infective periodontal therapy. a systematic ... · † department of periodontics,...

Documents