J Clin Periodontol 2000; 27: 325–332 Copyright C Munksgaard 2000Printed in Denmark . All rights reserved

ISSN 0303-6979

Sofia Karapataki 1,2,Anders Hugoson 1,2 andHealing following GTR treatmentCarl F. Kugelberg 3

1Department of Periodontology, The Institutefor Postgraduate Dental Education,of bone defects distal to Jonkoping, Sweden; 2Department ofPeriodontology, Faculty of Odontology,Goteborg University, Goteborg, Sweden;3Department of Oral Surgery, Countymandibular 2nd molars afterHospital, Kalmar, Sweden

surgical removal of impacted 3rdmolarsKarapataki S, Hugoson A, Kugelberg CF: Healing following GTR treatment ofbone defects distal to mandibular 2nd molars after surgical removal of impacted 3rdmolars. J Clin Periodontol 2000; 27: 325–332. C Munksgaard, 2000.

AbstractAim: The purpose of this study was to study the healing, following guided tissueregeneration (GTR) treatment, of bone defects distal to mandibular 2nd molars(M2s) after surgical removal of impacted mesioangularly or horizontally inclinedthird molars (M3s) in patients >25 years.Method: 20 patients with bilateral soft tissue impacted M3s were included in thesplit-mouth study. The 2 sites to be treated in each patient were randomisedbefore the 1st operation as to which would undergo the test procedure and whichwould be the control site. After surgical removal of M3 at test sites, a resorbablepolylactic acid (PLA) barrier was attached to M2 to cover the post-surgical bonedefect. The flap was then replaced and sutured to cover the barrier. Control sitesunderwent the same procedure, as did the test sites, with the exception that nobarrier was placed. The clinical examinations performed were oral hygiene pre-and 12 months postoperatively and probing pocket depth 12 months postopera-tively. The alveolar bone level (ABL) at the distal surface of the M2, as deter-mined from radiographs taken at suture removal and 12 months postsurgery, waschosen to be the primary response variable.Results. Most bone defects showed healing up to 10%–20% of the tooth lengthat both test and control sites. 2 test and 2 control sites showed no improvement in

Key words: guided tissue regeneration;the bone level. The mean values of bone healing registered in mm from the ce-resorbable barrier; surgical removal; impactedmento-enamel junction (CEJ) were 2.6∫2.19 SD and 3.0∫2.20 SD for test and lower 3rd molars; split-mouth

control sites, respectively. Different factors affecting the healing result are dis-cussed. Accepted for publication 28 June 1999

The least traumatic way to remove acompletely or partially impacted toothis to reflect a flap, remove the bonecovering the tooth, and divide thetooth into sections which are removedseparately. Clinical investigations, how-ever, have shown that surgical removalof impacted mandibular 3rd molars(M3s) may result in intrabony defects(IBD) at the distal aspect of the 2ndmolar (M2) (Ash et al. 1962, Szmyd &

Hester 1963, Grondahl & Lekholm1973, Chin Quee et al. 1985, Marmaryet al. 1985, Kugelberg et al. 1985, Ku-gelberg 1990). In a retrospective studycomprising 215 patients, Kugelberg etal. (1985) found that 2 years aftersurgery, 43.3% of the cases exhibitedprobing pocket depths exceeding 7 mmand 32.1% showed intrabony defectsexceeding 4 mm.

Kugelberg (1990) compared the peri-

odontal healing 2 and 4 years after im-pacted lower M3 surgery in 51 cases. 2years postoperatively, 16.7% of thecases aged <25 years had IBD exceed-ing 4 mm compared to 40.7% in the agegroup >25 years. At the 4-year re-ex-amination, the corresponding figureswere 4.2% and 44.4% respectively.

It can be concluded that the use ofconventional impacted M3 surgery inpatients >25 years may result in IBD at

326 Karapataki et al.

the distal aspect of the adjacent M2 thatdo not improve over the years. These de-fects may require surgical treatment at alater time. Studies on periodontal woundhealing have resulted in the developmentof the treatment modality Guided TissueRegeneration (GTR), (Karring et al.1985, Nyman et al. 1980, 1982a, Gottlowet al. 1984, 1986). This treatment in-volves the placement of a barrier to coverthe periodontal defect in such a way thatthe gingival tissues (epithelium and con-nective tissue) are prevented from con-tacting the root surface during healing.At the same time, a space is formed be-tween the barrier and the root allowingperiodontal ligament (PDL) cells to re-populate the denuded portion of the rootand produce a new connective tissueattachment. New cementum with insert-ing connective tissue fibres as well as newbone will eventually be formed.

A resorbable polylactic acid (PLA)barrier (GUIDOR matrix barrierA Gu-idor AB, Huddinge, Sweden) with amulti-layer matrix design into whichgingival connective tissue can grow hasbeen introduced for use in GTR ther-apy. The function of this barrier allowsfor the body’s own regeneration of peri-odontal ligament, cementum, and bone.The safety and efficacy of the barrierhave been documented in GTR studiesin monkeys (Gottlow et al. 1992) as wellas in humans (Laurell et al. 1992,Gottlow et al. 1992).

The aim of this investigation was tostudy the healing following GTR treat-ment of bone defects distal to man-dibular second molars after surgicalremoval of impacted mesioangularly orhorizontally inclined 3rd molars in pa-

Fig. 1. An example of a case. 2 impacted lower 3rd molars (M3) at approximately the samelevel as that of the M2 cemento – enamel junction (CEJ). The alveolar crest is located Ω4mm apical to CEJ.

tientsΩ25 years, using a resorbable(PLA) barrier.

Material and Methods

The study was designed as an open label,split-mouth randomised study. Twentypatients, male and female, Ω25 years ofage participated (mean age 35.2, range25–43 years). Each had 2 impacted M3s,which were at approximately the samelevel as that of the M2 cemento-enameljunction (CEJ). The alveolar crest (D)was located Ω4 mm apical to the CEJ(Fig. 1). All patients were healthy from aperiodontal point of view.

Inclusion criteria

O Willingness to participate and >25years of age.

O Good general health as evidenced bythe preoperative history.

O 2 impacted M3s with mesioangular(26æ–75æ) or horizontal (±75æ) incli-nation.

Exclusion criteria

O Patients with a history of atopy that,according to the investigator, wouldpreclude periodontal surgery and theuse of antibiotics.

O Patients who after the pre-experi-mental treatment fail to maintaingood oral hygiene.

O Patients with a systemic conditionthat, according to the investigator,would preclude surgery.

O Patients with acute infectious lesionsin areas intended for surgery.

O Participation in another clinical trial.

O Alcohol or drug abuse or any con-ditions associated with poor compli-ance.

Withdrawal criteria

O Failure to return for post-operativetreatment and evaluation.

O Patient request to be released from thestudy. No reasons needed to be given.

O The discretion of the investigator.Patients could be withdrawn at anytime by the investigator.Patients were informed that their

continued co-operation in attending allappointments for evaluation wasnecessary. Patients who failed to appearfor one or more evaluation visits wererescheduled as early as possible and no-tations made in the patient records. Alldata from such patients were reportedfor specific evaluation.

Replacement criteria

Patients who were withdrawn from ordropped out from the study during thepre-experimental treatment were re-placed by additional patients. All pa-tients were numbered consecutively.Drop-outs were analysed for the safetyevaluation as judged from data avail-able.

Pre-experimental treatment

Following patient selection and in-formed consent, each patient was givencareful instruction in proper oral hy-giene techniques. When needed, supra-gingival scaling was carried out. Ad-ditional instruction and reinforcementof oral hygiene was provided accordingto individual needs. Not more than 30%of the tooth surfaces covered by visibleplaque was accepted in order to fullfillthe criteria to take part in the study.

Presurgical procedures

Prior to M3 removal, the patients wereexamined and a general medical historywas recorded, including smoking habitsand use of oral contraceptives. Anamn-estic information concerning previousinconveniences or infections in the M3region was also recorded. The clinicalexamination was carried out at the buc-cal, distal, and lingual aspects of theM2 adjacent to the extraction site.After patients’ selection, third molarsurgery was performed. Before the op-eration, one of the two sites to be

Healing following GTR treatment 327

treated in each patient was randomlychosen to be the test site; the other be-came the control site. The time lapse be-tween the two surgeries was approxi-mately one month.

3rd molar surgery

The M1 and the M2 adjacent to the ex-traction site were polished with a rub-ber cup and an abrasive paste. The pa-tients then rinsed their mouths with asolution of 0.2% clorhexidine digluco-nate (HibitaneA Dental 0.2%, ICI).

All patients were treated under asep-tic conditions. Local anaesthesia wasgiven (XylocaineA adrenaline 20 mg/ml,ASTRA; adrenaline 12.5 mg/ml). Astandardised surgical technique of asingle incision to reflect a buccal and alingual flap was used. The anterior endof the flaps extended to the mesial as-pect of the M1. Osteotomy and sec-tioning were performed with a low-speed rotary instrument under constantirrigation with sterile saline. After thetooth was removed, the extractionsocket was carefully cleansed; all fol-licular remnants and granulation tissuewere removed and the area was cleanedwith a saline lavage. In cases of partiallyimpacted molars, epithelium under theflap was removed. Exposed root sur-faces were subjected to meticulous scal-ing and root planing. Extra care wastaken not to damage the remainingperiodontal ligament at the distal as-pect of the M2. Saline lavage was usedto clean the operated area.

At test sites, a resorbable polylacticacid (PLA) barrier (GUIDORA biore-sorbable matrix barrier) was used. Thebarrier was attached to the M2 to coverthe defect and secured to the M2 by thesling ligature preplaced in the device.The flap was replaced and sutured tocompletely cover the barrier. At controlsites no barrier were used. Thus theonly difference in the procedures car-ried out at the test and the control sitewas whether a barrier was used.

Postsurgical procedures

All patients received systemic antibiotictherapy (V-penicillin, KåvepeninA 2gr¿2 per os for 5 days). Analgesics wereprescribed according to individualneeds. The flap sutures were removedafter 14 days. The patients were advisedto rinse their mouth with 0.2% solutionof chlorhexidine digluconate (HibitanA

Dental, ICI, Sweden) twice daily for 6–

8 weeks following surgery and refrainfrom toothbrushing in the treated area.

Assessments

The clinical examinations were per-formed by two surgeons which werecarefully calibrated to each other.

Clinical examination

Sites registered at the M2 were disto-buccal, distal, and distolingual surfaces.

Oral hygiene. The plaque index (PLI)(Silness & Loe 1964) was registered pre-operatively and 12 months postopera-tively.

Probing pocket depth (PPD). ThePPD was measured with a 1-mm gradedmanual probe 12 months postopera-tively.

Radiographic examination

Radiographic examinations were madepreoperatively, at suture removal, and 6and 12 months postoperatively. To as-sess the anatomy and position of theM3s, the pre-operative radiographic ex-amination included a panoramic radio-graph, a posterioanterior radiographwith the mouth open, and at least twointra-oral X-ray films. Details of theradiographic technique have been pre-viously described by Kugelberg et al.(1986).

In addition, standardised radio-graphs were taken of the experimentalteeth using commercially availablefilmholder devices (Eggen 1969) de-signed to hold a vertically-oriented X-ray film. Each filmholder device wasadapted to the occlusal surface by theaid of individual impressions (Presi-dentA). The intra-oral radiograph takenafter suture removal was used as a base-line, because of the difficulty of deter-mining the bottom of the defect withthe M3 in place.

All radiographic measurements weremade 2¿ at the end of the study by oneof the authors. The radiographs wereplaced on an illuminator with diffusewhite light, which was masked to thearea of the mounted radiographs. Theimages were analysed with the aid ofobservation binoculars according toMattsson (1953).

The following variables were regis-tered at suture removal and 12 monthspostoperatively:

(a) Bone level (BL). A transparentplastic ruler with 10 equidistant divi-

sions was placed on the radiograph toestimate the bone level in tenths of thetotal length of the tooth. The distancefrom the CEJ to the apex in per cent oftooth length was recorded as well as thedistance (%) from the apex to the bot-tom of the defect (BL) at the distal sur-face of the M2 root adjacent to the de-fect.

(b) Alveolar bone level (ABL). Thedistance between the CEJ and the bot-tom of the bone defect was measuredwith the aid of a transparent plasticruler graduated in mm. The measure-ments were made on a line parallel tothe distal surface of the M2 root ad-jacent to the defect.

(c) Root resorption. Assessments werealso made regarding the prevalence ofradiolucent areas, e.g. root resorption-caries at the M2 distal surface, through-out the study. They were classified as ac-tive and inactive ones according to theirradiographic appearance, without orwith respectively a clear demarcation ofthe dentine defect as well as their ap-pearance throughout the study.

Adverse Events (AEs)

Possible adverse reactions to the re-sorbable barrier were evaluated by clin-ical observations and open questioningthroughout the study. A distinction wasdrawn between AEs normally experi-enced in conjunction with M3 surgeryand reactions normally not encoun-tered. AEs were recorded prior tosurgery (baseline); at 14 days; and 1, 2,6, and 12 months postoperatively.

All adverse events were recorded withinformation about degree of severity(i.e. mild, moderate, severe), date of on-set, duration, and action taken regard-ing the study device.

The degree of severity of the adverseevents was scored as follows:

1. mild no significant inter-ference with normal ac-tivities, acceptable;

2. moderate significant interferencewith normal daily activi-ties, but still acceptableto the patient;

3. severe unacceptable interferencewith normal daily activi-ties. Treatment changedor stopped.

A separate registration was madewhen a flap on either a test or a controlsite was not completely sealed after su-ture removal.

328 Karapataki et al.

Table 1. Plaque scores (PLI) at the M2s distobuccal (DB), distal (D), and distolingual (DL) surfaces at baseline and 12 months postoperativelyfor the test and the control groups

No. sites

DB surface D surface DL surfacePLIscore test control test control test control

Baseline 0 16 18 11 14 7 9

1 4 2 7 4 11 9(3,8,15,17) (5,7) (3,4,7,8,15,16,18) (5,7,8,20) (3,4,6,7,8,10,11, (2,3,4,5,7,8,10,11,13)

14,15,16,17)

2 2 2 2 2(17,19) (17,19) (18,19) (17,19)

12 months 0 14 12 12 8 8 7post- 1 4 7 5 10 9 9operatively (3,8,13,19) (3,5,8,13, (3,5,9,13,15) (3,4,5,8,9,11, (3,4,5,6,7,8, (3,4,5,6,7,9,

15,17,18) 13,15,17,19) 12,14,15) 11,13,20)

2 2 1 3 2 3 4(15,17) (14) (16,17,19) (14,16) (16,17,19) (14,16,17,19)

Figures outside the parentheses display the number of patients with PLI 0, 1 or 2. Figures within parentheses refer to the actual patient.

Statistical analysis

The analysis addressed the followingquestion: Does the resorbable barrier

Fig. 2.

Fig. 3.

Figs. 2, 3. Radiographs taken presurgically (a), at suture removal (b), and 12 months post-operative (c) in 2 patients. Barriers were used to cover the defect in both cases.

improve the healing of bone in a popu-lation of patients >25 years of age whohave had mesioangularly or horizon-tally inclined M3s extracted? As this in-

vestigation was a split-mouth studywith a randomised allocation of devicewithin the patient, statistical methodsfor related samples were used. Continu-ous variables were analysed using theStudent t-test. The alveolar bone levelmeasured in radiographs (ABL) waschosen to be the primary response vari-able to judge the effectiveness of thetreatment. A difference of 1 mm (ABL)was the minimum height that was con-sidered to be clinically important. Allstatistical tests were performed two-tailed and at the 5% level of signifi-cance.

Ethical Requirements

Declaration of HelsinkiThis study was performed according tothe Declaration of Helsinki. All pa-tients were informed about the studydesign and their right to withdraw fromthe study at any time and that such adecision did not negatively affect theirfuture treatment. The patients gavetheir informed consent to participate inthe study.

Results

In Table 1, the PLIs recorded at theM2’s distobuccal, distal, and distolingu-al surfaces 12 months postoperativelyare presented. The number of patientsas well as which ones had scores of 1 or2 can be seen. Most of the patients hadlow plaque scores. No score 3 was seenduring the study. No differences be-

Healing following GTR treatment 329

tween the test and the control sites werenoticed.

Patients who experienced adverseevents at either the test or the controlsite were recorded. AEs were rare anddid not differ from those usually experi-enced in third molar surgery. Two pa-tients reported moderate pain, swelling,and suppuration 1 month postopera-tively at their test sites; three patientsreported the same at their control sites.Two patients experienced temporaryparaesthesia. Those who received extraantibiotics underwent their second op-eration 2 months after the first one.

The flaps at the test sites in patientsnos. 11, 12, 13, 16, and 18 were notfound to be completely sealed at sutureremoval and the barriers in these caseswere exposed. The flaps at the controlsites in patients nos. 12, 14, and 19 werenot found to be completely sealed at su-ture removal either. The M3s in thesecases were semi-impacted.

Figs. 2, 3 present radiographs takenpresurgically, at suture removal, and 12months postoperatively for two pa-tients. Barriers were used to cover thedefect in both cases.

In Table 2, the PPDs 12 months post-operatively are shown. The values of allthe patients are presented. Mean values(SD) for the test sites at the M2’s disto-buccal, distal, and distolingual surfaceswere 2.9∫1.8 mm, 3.2∫1.8 mm, and3.0∫1.7 mm respectively. For controlsites, the mean values were 3.4∫2.0mm, 3.9∫2.1 mm and 3.1∫1.7 mm re-spectively. There were no statistical dif-ferences between test and control sites.

Fig. 4. BL at the distal surface of the M2 immediately after suture Fig. 5. BL at the distal surface of the M2 immediately after sutureremoval and 12 months postoperative for test sites. The first bar for removal and 12 months postoperative for control sites. The 1st bareach patient site represents the distance from the CEJ to the apex in for each patient site represents the distance from the CEJ to the apexper cent of tooth length. HΩCalculated ‘‘normal’’ BL (3 mm apical in per cent of tooth length. HΩCalculated ‘‘normal’’ BL (3 mm apicalto the CEJ). The 2nd and 3rd bars represent the distance from the to the CEJ). The second and third bars represent the distance frombottom of the defect to the apex at suture removal and 12 months the bottom of the defect to the apex at suture removal and 12 monthspostoperative, respectively. postoperative, respectively.

Table 2. Probing pocket depth (PPD) at the M2s distobuccal (DB), distal (D), and distolingual(DL) surfaces 12 months postoperatively for all patients; values (mm) at the test and controlsites

Test ControlPatientno. DB D DL DB D DL

1 0 0 0 0 0 02 3 3 3 5 4 43 0 3 3 4 4 44 3 3 3 3 5 55 4 3 3 3 4 46 3 0 0 0 0 07 0 4 4 5 7 58 3 3 4 4 5 39 3 3 3 4 3 3

10 0 0 0 0 0 011 4 3 3 4 3 312 3 5 4 4 5 313 3 5 6 7 5 514 7 6 6 3 4 615 4 3 5 4 5 316 4 7 2 7 8 417 4 4 2 4 6 318 4 3 2 1 3 219 3 3 3 4 5 320 2 3 4 2 2 2

Figs. 4, 5 present the BL for all pa-tients at test as well as control sites. Inradiographs the distance from the CEJto the apex is given for each site in percent of tooth length. An average valueof the distance between the CEJ and‘‘normal’’ bone level (3 mm apical tothe CEJ) is marked for every patient/site. The distance (%) from the apex tothe bottom of the defect (BL) is pre-sented at suture removal and 12 monthspostoperatively.

Most defects showed healing of the

bone defect up to 10%–20% of thetooth length at both the test and con-trol sites. 2 test and 2 control sitesshowed no improvement in the heightof the bone level.

Table 3 shows the difference in heightof alveolar bone level (ABL) in mm asmeasured in the radiographs at theM2’s distal surfaces between suture re-moval and 12 months for all patientsand for both test and control sites. Themean value (SD) for the test sites was2.6∫2.19 mm and for the control sites

330 Karapataki et al.

Table 3. Difference (mm) in height of al-veolar bone level (ABL) in radiographs at theM2’s distal surfaces between suture removaland 12 months postoperatively for all pa-tients; values at the test and control sites

Difference (mm)Patientno. test control

1 π4 π12 π2 π33 π7 π54 0 π55 ª1 π16 π0.5 π0.57 π5 π38 π2.5 π29 π4 π2

10 π2 π2.511 π1 012 π1.5 ª113 π7 π7.514 π3 π215 π3 π316 π4 π517 π2 π618 π1 π519 π4 π320 0 π5

3.0∫2.20 mm. No statistically signifi-cant differences existed between testand control sites.

In the test group, 4 radiolucent areasat the M2 distal surface were judged tobe inactive and 1 active at 12 months.The corresponding figures for the con-trol sites were 5 and 3 respectively(Table 4).

Discussion

According to Kugelberg (1990), there isa risk that a bone defect at the distalaspect of the second mandibular molarwill not heal after surgical removal ofsome types of impacted third molars inpatients above the age of 24. The aimof this investigation was to study thehealing following GTR treatment witha PLA barrier as well as to investigatewhether the risk of developing such de-fects in the bone could be minimisedusing the GTR technique. This wasevaluated with help of radiographic BLand ABL. According to the results noclear benefit of using a PLA barrier wasfound. A resolution of the original de-fect occurred at both the test and con-trol sites in most cases. The individualvariation was high. Thus, some of thedefects healed completely with an up to7-mm gain in bone tissue while a fewothers deteriorated.

In this study, the split-mouth design

Table 4. Prevalences of root resorption at baseline and 12 months postoperatively; the figuresrefer to the actual patients

Root resorption

Inactive Active

baseline 12 months baseline 12 months

test 2,4,18,20 2,4,18,20 16control 9,16,17,20 9,14,16,17,20 12 3,11,12

was chosen as it was supposed to ex-clude patient specific characteristics.The healing result of a single site, how-ever, is dependent on a number of fac-tors at baseline that may differ from siteto site (Kugelberg 1990). In each pa-tient, the locations of the M3s in re-lation to the M2s, as determined fromthe radiographs, were chosen to besimilar. The actual contact between theM3 and the M2 as well as the destruc-tion produced by the surgical procedureitself, though, could be very different.Differences during M3 removal such asamount of bone removal, the actualsurgical trauma of the area, and thelength of the operation could be factorsthat are important for the healing pro-cess and that could not be discountedwith the use of a split-mouth design.

An attempt was made to record PPDdepth as well as PAL before surgical re-moval of the M3. This proved to bevery difficult due to the location of theM3 in relation to the M2. Often, theprobe met the occlusal surface of theM3 and the depth of the defect wasunderestimated; especially in semi-im-pacted teeth, the CEJ of the M2 couldbe located beneath the M3. PPD resultsare therefore only given 12 monthspostoperatively, and the radiographicalveolar bone level (ABL) was chosento be the primary response variable tojudge the effectiveness of the treatment.

PPD values did not always corre-spond well with ABL values. This couldbe due to thick soft tissue or even toerror at ABL measurements due to abone defect not diagnosed behind thickbuccal and lingual bone walls. EvenABL values did not always correspondto BL measurement. This could in turnbe attributed to small alterations in ver-tical position of the X-ray conus whenthe radiograph was taken. Thus accord-ing to ABL results 6 test and 5 controlsites revealed no clinically importantimprovement between baseline and 12months postoperatively.

When examining the mean PAL gainfound 6 months postoperatively in the

study by Oxford et al. (1997), it wasfound to be only 0.13 mm for test(GTR) sites and 0.25 mm for controlsites. These results could be due to theabove-mentioned difficulties of probingPAL at baseline in the actual region.Even Pecora et al. (1993) presentedmean values of recordings made atthree sites adjacent to the defects.

Some evidence has been presentedthat use of GTR techniques facilitatesnew attachment formation and thatcells responsible for this regenerationappear to originate from the peri-odontal ligament (Nyman et al. 1982a).At the present study an attempt wasmade to visualise what was left of theperiodontal ligament at the distal sur-face of the M2 and record it by photo-graphic methods or by simply drawinga picture of the situation. Duringsurgery, however, the shape of the defectwas not easy to see due to bleeding aswell as for anatomic reasons. The actualtype of intrabony defect (e.g., 2 or 3-wall) may be of importance. Bone re-moval buccally was limited as much aspossible, but if the M3 was located dis-tobuccally, a combination of defectswas already created. In some cases thebone crest was so thin that it by necess-ity would be resorbed during healing;the odds for healing or regeneration upto the original level of the bone crestdecreased.

In any case, the appearance of theperiodontal ligament covering the distalsurface of the M2, depending on howthe M3 was positioned in relation to it,differed widely from site to site. Thearea not covered by periodontal liga-ment could start from the CEJ and benarrow or wide, short or long, and runin an apical direction but the peri-odontal ligament could also be com-pletely absent from the area. One hypo-thesis for the variation of results couldbe that short or long narrow areas de-prived of periodontal ligament are moreeasily repopulated by ligament cells andtherefore also have a higher degree ofhealing than do wide areas with no

Healing following GTR treatment 331

periodontal ligament simply because ofa more abundant sources of cells for theactual area to be covered.

As the sizes of the postsurgically-created defects differed, the proper sizeof barrier was not always available.Since the area close to the distal part ofthe M2 was always covered, the barriermight be unable to cover the distal partof the defect fully. PLA barrier lackstiffness, and the size of the defects aswell as the additional pressure of thethick gingival tissue of the area mayalso have displaced the barrier apically.These clinical difficulties could be ofimportance for the healing process ascreation and maintenance of space be-neath the barrier is crucial for the out-come of a regeneration technique(Tonetti et al. 1996).

Flap opening was noticed in somecases due to a shortage of gingivaltissue and an increase in tension of theflaps after suturing. No obvious effecton the healing result, however, could betraced to flap relapse.

The PLI values 12 months post-operatively show that the distal portionof the M2 is difficult to clean by oralhygiene measures. This fact, however,does not seem to affect the results. Pa-tients who lost bone (radiographically)such as patients no. 5 (test site) and no.12 (control site) barely had plaque.Others like patients no. 16 (test site) andno. 17 (control site) with PLI scores of2, gained significant amounts of bone.

The effect of non-erupted M3s onroot resorption has been studied byNemcovsky et al. (1996) among others.In this study, radiolucent areas judgedto have active root resorption proceed-ing apically may have affected the re-sults, in patients nos. 11 and 12 (controlsites). Patient no. 3, however, gained asignificant amount of bone despiteradiographic signs of root resorption.Radiographic interpretation has itslimitations (Chapnich & Endo 1989),however, and the number of sites withroot resorption in this study was toofew to allow any conclusion to be made.

Adverse events as pain, swelling, andsuppuration, though moderate, did notseem to affect the results.

It can be concluded from this studythat much research is still needed to cla-rify the healing process at the distal sur-face of the second molar after surgicalremoval of the M3. The results did notreveal any clear-cut benefit of using aresorbable GTR barrier in conjunctionwith surgical removal of an impacted

M3. The reason may be that in severalof the treated sites there was no or onlya very limited periodontal engagementof the distal root surface of the M2, e.g.the root surface was not involved. Thusthe healing could proceed normally, asthat following any routine tooth extrac-tion with more or less complete bonefill. An intrabony defect distal to theM2 following impacted M3 surgerywould thus depend on the existingamount of periodontal ligament of theM2 and whether this was affected bydisease before surgery.

Acknowledgements

The authors gratefully acknowledge Dr.Rolf Karlsson for statistical advice andthe Jonkoping County Council and Gu-idor AB, Goteborg, Sweden for theirfinancial support.

Zusammenfassung

Die Heilung nach GTR-Behandlung von kno-chernen Defekten distal von 2. unteren Mola-ren nach chirurgischer Entfernung von impak-tierten 3. MolarenDer Zweck dieser Studie war es, die Heilungnach gesteuerter Geweberegeneration (GTR)von Knochendefekten distal der unteren 2.Molaren (M2s) nach chirurgischer Entfer-nung von impaktierten mesial gekippten oderhorizontal verlagerten 3. Molaren (M3s) beiPatienten >25 Jahren zu studieren. 20 Pati-enten mit bilateral im Weichgewebe impak-tierten M3s wurden in diese split-mouth-Stu-die einbezogen. Die zwei zu behandelndenSeiten jedes Patienten wurden vor der 1. Ope-ration randomisiert und Test- oder Kontroll-seiten zugeordnet. Nach der chirurgischenEntfernung des 3. Molaren auf der Testseitewurde eine resorbierbare Polylactidmembran(PLA) am 2. Molaren befestigt, um den post-chirurgischen Knochendefekt zu bedecken.Die Lappen wurden dann reponiert und ver-naht, um die Membran zu bedecken. DieKontrollseiten unterlagen der gleichen Proze-dur wie die Testseiten, allerdings ohne Mem-braneinbringung. Die klinischen Prufungenbestanden in der pra- und 12-monatigenpostoperativen oralen Hygiene und der Son-dierungstiefe 12 Monate postoperativ. Dasalveolare Knochenniveau (ABL) an der di-stalen Flache des M2, das von Rontgenbil-dern zum Zeitpunkt der Nahtentfernung und12 Monate postoperativ bestimmt wurde,war die erste Ergebnisvariable. Die meistenKnochendefekte zeigten eine Heilung bis zu10–20% der Zahnlange sowohl auf den Test-als auch Kontrollseiten. 2 Test- und 2 Kon-trollseiten zeigten keine Verbesserung derKnochenhohe. Der mittlere Wert der Kno-chenheilung, der von der Schmelz-Zement-Grenze (CEJ) in mm registriert wurde, war2.6∫2.19 SD fur die Testseiten und 3.0∫2.20

SD fur die Kontrollseiten. Verschiedene Fak-toren, die die Heilung betreffen, werden dis-kutiert.

Resume

Guerison suivant le traitement GTR de lesionsosseuses en distal des 2emes molaires mandi-bulaires apres enlevement chirurgical des dentsde sagesse inclusesLe but de cette etude a ete d’etudier la gueri-son faisant suite au traitement par regenera-tion tissulaire guidee (RTG) de lesions osseu-ses situees en distal des deuxiemes molairesmandibulaires apres extraction chirurgicalede dents de sagesse impactees avec une angu-lation mesiale ou horizontale chez des pa-tients >25 ans. 20 patients ayant des dentsde sagesse bilaterales incluses uniquement ausein des tissus mous ont ete inclus dans cetteetude par bouche divisee. Les 2 sites devantetre traites chez chaque patient ont ete repar-tis au hasard avant l’operation pour deciderdu processus qui serait applique. Apres l’ex-traction chirurgicale des dents de sagesse auniveau du site operatoire test, une membraneen acide polylactique resorbable (PLA) a eteattachee a la deuxieme molaire pour recou-vrir la lesion osseuse. Le lambeau a ete repla-ce et suture pour recouvrir cette derniere. Lessites controles ont subi le meme processus ala difference qu’aucune membrane barrieren’etait placee. Les examens cliniques investi-gues etaient: le releve de l’hygiene buccaleavant et 12 mois apres l’operation, et le son-dage a douze mois. Le niveau osseux alveo-laire a la surface distale de la 2eme molaire,determine sur des radiographies lors del’enlevement des sutures et douze mois apres,a ete choisi comme variable principale. Laplupart des lesions osseuses montraient uneguerison jusque 10% a 20% de la longueur dela dent tant dans les sites tests que controles.2 sites tests et 2 controles n’ont montre aucu-ne amelioration en ce qui concerne le niveauosseux. Les valeurs moyennes de guerison os-seuse enregistrees en mm depuis la jonctionemail-cement etaient de 2.6∫2.19 mm et3.0∫2.20 mm pour respectivement les sitestests et controles. Differents facteurs affec-tant la guerison sont discutes.

References

Ash, M., Costich, E. R. & Hayward, J. R.(1962) A study of periodontal hazards ofthird molars. The Journal of Periodonto-logy 33, 209–219.

Chapnick, L. & Endo, D. (1989) Externalroot resorption: An experimental radio-graphic evaluation. Oral Surgery OralMedicine Oral Patholgy 67, 578–582

Chin Quee, T., Gosselin, D., Millar, E. &Stamm, J. (1985) Surgical removal of fullyimpacted mandibular third molar. The in-fluence of flap design and alveolar boneheight on the periodontal status of the sec-ond molar. Journal of Periodontology 56,625–630.

332 Karapataki et al.

Eggen, S. (1969) Standardiserad intra-oralrontgenteknik. Sveriges Tandlakarfor-bunds Tidning 6, 867.

Gottlow, J., Lundgren, D., Nyman, S.,Laurell, L. & Rylander, H. (1992) Newattachment formation in the monkey,using GuidorA a bioabsorbable GTR-de-vice. International Dental Research 71,special issue.

Gottlow, J., Nyman, S., Karring, T. &Lindhe, J. (1984) New attachment forma-tion as the result of controlled tissue re-generation. Journal of Clinical Periodonto-logy 11, 494–503.

Gottlow, J., Nyman, S., Laurell, L., Falk, H.,Fornell, J. & Johard, G. (1992) Clinical re-sults of GTR-therapy using a bioabsorb-able device, GuidorA. International DentalResearch 71, special issue.

Gottlow, J., Nyman, S., Lindhe, J., Karring,T. & Wennstrom, J. (1986) New attach-ment formation in the human periodon-tium by guided tissue regeneration. Casereports. Journal of Clinical Periodontology13, 604–616.

Grondahl, H. G. & Lekholm, U. (1973) In-fluence of mandibular third molars on re-lated supporting tissues. InternationalJournal of Oral Surgery 2, 137–142.

Karring, T., Isidor, F., Nyman, S. & Lindhe,J. (1985) New attachment formation onteeth with a reduced but healthy peri-odontal ligament. Journal of Clinical Peri-odontology 12, 51–60.

Kugelberg, C. F., Ahlstrom, U., Ericson, S. &Hugoson, A. (1986). Periodontal healingafter impacted lower third molar surgery.Precision and accuracy of radiographic as-sessment of intrabony defects. Interna-

tional Journal of Oral and MaxillofacialSurgery 15, 675–686.

Kugelberg. C. F., Ahlstrom, U., Ericson, S.,Hugoson, A. (1985) Periodontal healingafter impacted lower third molar surgery.A retrospective study. International Journalof Oral Surgery 14, 29–40.

Kugelberg, C. (1990) Periodontal healing 2and 4 years after impacted lower third mo-lar surgery. A comparative retrospectivestudy. International Journal of Oral andMaxillofacial Surgery 19, 341–345.

Laurell, L., Gottlow, J., Nyman, S., Falk, H.,Fornell, J. & Johard, G. (1992) Gingivalresponse to GuidorA a bioabsorbable de-vice in GTR-therapy. International DentalResearch 71, special issue.

Marmary, Y., Brayer, L., Tzukert, A. & Fell-er, L. (1985) Alveolar bone repair follow-ing extraction of impacted mandibularthird molars. Oral Surgery, Oral Medicineand Oral Pathology 60, 324–326.

Newcovsky, C. E., Libfeld, H., Zubery, Y.(1996) Effect of non-erupted 3rd molarson distal roots and supporting structuresof approximal teeth. A radiographic sur-vey of 202 cases. Journal of Clinical Peri-odontology 23, 810–815.

Nyman, S., Karring, T., Lindhe, J. & Plant-en, S. (1980) Healing following implan-tation of periodontitis affected roots intogingival connective tissue. Journal of Clin-ical Periodontology 7, 394–401.

Nyman, S., Gottlow, J., Karring, T. &Lindhe, J. (1982a) The regenerative poten-tial of the periodontal ligament. An ex-perimental study in the monkey. Journal ofClinical Periodontology 9, 257–265.

Silness, J. & Loe, H. (1964) Periodontal dis-

ease in pregnancy. Correlation betweenoral hygiene and periodontal condions.Acta Odontologica Scandinavica 22, 121–135.

Szmyd, L. & Hester, W. R. (1963) Creviculardepth of the second molar in impactedthird molar surgery. Journal of OralSurgery, Anaesthesiology Hospital DentalService 21, 185–189.

Matsson, O. (1953) A magnifying viewer forphotofluorographic films. Acta Radiolo-gica 39, 412.

Oxford, G. E., Quintero, G., Stuller Ch. B.,Stuller B. & Gher M. E. (1997) Treatmentof 3rd molar-induced periodontal defectswith guided tissue regeneration. Journal ofClinical Periodontology 24, 464–469.

Pecora, G., Celletti R., Davarpanah, M., Co-vani, U. & Etienne, D. (1993) The effectsof guided tissue regeneration on healingafter impacted mandibular 3rd-molarsurgery: 1-year results. The InternationalJournal of Periodontics & Restorative Den-tistry 13, 397–407.

Tonetti, M., Pini Prato, G. & Cortellini, P.(1996) Factors affecting the healing re-sponse of intrabony defects followingguided tissue regeneration and access flapsurgery. Journal of Clinical Periodontology23, 548–556.

Address:

Anders HugosonPublic Dental Service AdministrationP.O. Box 1024S-551 11 JonkopingSweden