Alveolar Bone Preservation in Extraction Sockets Using Non ...

J Periodontol • August 2008

Alveolar Bone Preservation in Extraction Sockets Using Non-Resorbable dPTFE Membranes: A Retrospective Non-Randomized Study Oliver Hoffmann,*† Barry K. Bartee,‡ Christian Beaumont,† Adrian Kasaj,§ Giorgio Deli,i and Gregor-Georg Zafiropoulos†§

Background: The aim of this study was to investigate the clinical regeneration of extraction sockets using high-density polytetrafluoroethylene (dPTFE) membranes without the use of a graft material. Methods: A total of 276 extraction sockets were evaluated in 276 subjects (151 males and 125 females; mean age, 50.2 years; age range: 24 to 73 years). After extraction, flaps were elevated and a dPTFE membrane was placed over the extraction site. The flaps were repositioned and sutured into place. Primary closure was not obtained over the membranes. The cemento-enamel junctions of the adjacent teeth were used as reference points. Measurements were taken postextraction and 12 months after surgery in the same areas with the help of a stent and were defined as the distance from the reference points to the bone level. Hard tissue biopsies were taken from 10 representative cases during implant placement 12 months after socket preservation. The bone core samples were submitted for histologic evaluation. A stringent plaquecontrol regimen was enforced in all subjects during the 12month observation period. Results: A significant regeneration of the volume of sockets could be noted by histologic evaluation, indicating that the newly formed tissue in extraction sites was mainly bone. No influence of gender, smoking, age, or clinical bone level before treatment was found on the percentage of bone gain. Conclusion: The use of dPTFE membranes predictably led to the preservation of soft and hard tissue in extraction sites. J Periodontol 2008;79:1355-1369. KEY WORDS Periodontitis; regeneration.

* † ‡ §

Department of Periodontics, Loma Linda University, Loma Linda, CA. Blaues Haus, Dental Center, Du ¨ sseldorf, Germany. Department of Surgery, Texas Tech University Health Science Center, Lubbock, TX. Department of Operative Dentistry and Periodontology, University of Mainz, Mainz, Germany. i Division of Periodontology, Catholic University of the Sacred Heart, Rome, Italy.

A

lveolar ridge resorpion following tooth extraction is a frequently observed phenomenon that may decrease the possibility of placing dental implants or impair the esthetic results after prosthodontic treatment. Although the degree of bone loss (BL) varies among individual subjects and between anatomic sites, it is well accepted that as much as 40% of the alveolar height and 60% of the alveolar width may be lost in the first 6 months following extraction.1 Even subtle postextraction BL may have significant clinical effects, particularly in the esthetic zone. With the current emphasis on esthetic single-tooth replacement with dental implants, techniques to preserve natural bone and soft tissue contours are of great interest to clinicians. Studies showed that BL occurs in the vertical and horizontal planes, with the degree of horizontal BL typically exceeding the degree of vertical BL.1,2 In the horizontal plane, BL occurs largely at the expense of the facial cortical plate, increasing the risk for facial soft tissue recession, especially in the presence of a thin periodontal biotype.2 Interdental BL may lead to the loss of the interdental papilla.3,4 Although osseointegration of the dental implant may be successful, an anterior implant restoration is often judged an esthetic success or failure on the basis of soft tissue appearance.5-7 doi: 10.1902/jop.2008.070502

1355

Socket Preservation Using dPTFE Membranes

Several clinical techniques and a variety of biomaterials have been introduced over the years in an effort to prevent or reduce postextraction BL. Initially, the concept of vital root retention was proposed based on the observation that bone resorption did not occur around teeth retained as abutments for partial dentures or overdentures, yet edentulous areas in the same jaw exhibited severe resorption. This technique was later abandoned because of soft tissue complications associated with the exposed roots.8,9 Using a similar concept, techniques for ridge preservation were introduced in the 1980s using hydroxyapatite in the form of root-shaped cones.10,11 Although the technique was successful in terms of ridge preservation, problems with soft tissue encapsulation and the resultant exfoliation of the cones led to the abandonment of this technique in favor of particulate hydroxyapatite materials. Clinical evaluation of these materials again indicated successful outcomes in terms of bone preservation; however, problems such as particle migration and loss prevented the widespread acceptance of the procedure. The advent of guided tissue regeneration membranes provided clinicians with a new method to potentially modify extraction socket healing. An early study12 using direct measurements between the alveolar ridge crest and fixed reference points indicated a statistically significant reduction in BL when using membranes alone or in combination with particulate materials. Membranes evaluated in that study included non-resorbable and bioabsorbable materials with and without particulate augmentation materials. Current methods to prevent ridge resorption include the use of particulate autografts, allografts, alloplasts, xenografts, and membranes manufactured from various materials, including those that are bioabsorbable or non-resorbable, naturally derived or synthetic.13-18 Each of these biomaterials provides certain advantages and disadvantages. Historically, the bioabsorbable and porous synthetic materials have required primary closure over the socket, a requirement that increases surgical complexity, reduces the amount of keratinized gingiva, and disrupts the natural architecture of soft tissues in the area. A membrane made of high-density polytetrafluoroethylene (dPTFE), designed specifically for use in socket grafting, which does not require primary closure was described in case reports.19 The successful use of this material was demonstrated in animal and clinical investigations.20-22 The aim of the present study was to determine the effectiveness of dPTFE membranes in preserving the dimensions of extraction sockets during an 8-month healing period. 1356

Volume 79 • Number 8

MATERIALS AND METHODS Study Population A total of 276 sockets of 361 extracted teeth in 276 subjects (151 males and 125 females; mean age, 50.2 years; age range: 24 to 73 years), who were scheduled for socket preservation, were evaluated in this retrospective, non-randomized study (Table 1). The sockets were assigned to one of two groups: single sockets, in cases where only one tooth in an area was extracted, or side-by-side sockets, in cases where two neighboring teeth were extracted (Tables 2 and 3). One socket form (single or side-by-side) was selected per subject. One hundred twenty – four (124) subjects smoking 1 to 10 cigarettes per day were categorized as smokers (44.9%; 45 females and 79 males; mean age, 49 years; age range: 24 to 73 years). Subjects who smoked >10 cigarettes per day were excluded from the study (Table 1). All subjects were examined and subsequently treated between January 1999 and December 2005 in the private periodontal practice of one of the authors (GGZ) in Du ¨ sseldorf, Germany. Cases were referred for treatment. The restoration of the augmented areas with fixed partial dentures or dental implants was planned in collaboration and consultation with the referring general dentist and the subject at a later time point during maintenance. Demographic data, medical and dental health history, and smoking status were determined by questionnaire; the periodontal status was determined by a comprehensive periodontal examination. All subjects had a history of chronic periodontitis with a minimum of four sites with clinical attachment levels (CAL) >4 mm, radiographic evidence of alveolar BL, and bleeding on probing (BOP) in at least four areas.23 All subjects were treated by scaling/root planing 4 to 6 months before beginning the study and demonstrated good oral hygiene and compliance (probing depth [PD]: 5.1 – 0.9 mm; CAL: 6.2 – 1.2 mm; BOP: 12%; and plaque index [PI]: 10%24). All subjects were informed about the treatment procedures and had ‡1 week after the information was given before the informed consent form was signed. The study was performed according to the Helsinki Declaration of 1975, as revised in 1983. Teeth scheduled for extraction and socket preservation at the initial examination were not extracted during the initial periodontal therapy. Instead, they were reduced in height to the gingival level and scaled and root planed together with the remaining teeth.The dentitionwas divided into sextants: a) teeth #1 through #5, b) teeth #6 through #11, c) teeth #12 through #16, d) teeth #17 through #21, e) teeth #22 through #27, and f) teeth #28 through #32. Subjects for the present retrospective non-randomized study were selected according to the following criteria: 1) no contraindications for treatment, such

Hoffmann, Bartee, Beaumont, Kasaj, Deli, Zafiropoulos


Table 1.

Demographic Characteristics of the Study Population (N = 276) Age (years) Gender

Smoker

n

%

Mean

Minimum

Maximum

Female

No Yes Total

80 45 125

29.0 16.3 45.3

50.3 46.3 48.9

31 24 24

67 66 67

Male

No Yes Total

72 79 151

26.1 28.6 54.7

52.0 50.6 51.2

26 26 26

71 73 73

Total

No Yes Total

152 124 276

55.1 44.9 100.0

51.1 49.0 50.2

26 24 24

71 73 73

Table 2.

Distribution of the Treated Sockets in the Study Population Gender (n [%])

Smokers (n [%])

Age (years)

Female

Male

No

Yes

Mean (range)

Total (n [%])

Maxilla

59 (43.7)

76 (56.3)

79 (58.5)

56 (41.5)

49.98 (26 to 71)

135 (100.0)

Mandible

66 (48.8)

75 (53.2)

73 (51.8)

68 (48.2)

50.38 (24 to 73)

141 (100.0)

Maxilla: anterior

12 (63.2)

7 (36.8)

11 (57.9)

8 (42.1)

48.95 (27 to 71)

19 (100.0)

Mandible: anterior

8 (36.4)

14 (63.6)

13 (59.1)

9 (40.9)

48.73 (29 to 63)

22 (100.0)

Maxilla: posterior

47 (40.5)

69 (59.5)

68 (58.6)

48 (41.4)

50.15 (26 to 69)

116 (100.0)

Mandible: posterior

58 (48.7)

61 (51.3)

60 (50.4)

59 (49.6)

50.69 (24 to 73)

119 (100.0)

Single sockets

88 (46.1)

103 (53.9)

105 (55.0)

86 (45.0)

50.18 (24 to 73)

191 (100.0)

Side-by-side sockets

37 (43.5)

48 (56.5)

47 (55.3)

38 (44.7)

50.20 (26 to 71)

85* (100.0)

125 (45.3)

151 (54.7)

152 (55.1)

124 (44.9)

50.18 (24 to 73)

276 (100.0)

Total

* In the side-by-side socket group, two teeth can be found per evaluated unit, i.e., the 85 sites consist of 170 extraction sockets.

as systemic diseases (e.g., diabetes), pregnancy, the use of prescription medications or consumption of recreational drugs; 2) the use of one membrane only per surgical site; 3) no more than two neighboring extraction sockets; in cases of two neighboring sockets, both had to be in the same sextant; 4) the buccal and lingual/palatal plate of the extraction socket had to be present with a maximum BL of 50% of the vertical dimension; however, sockets with BL in the buccal furcation area were included if one root was covered by bone and BL did not extend into the apical third of the other root; sockets with BL in the lingual furcation area were excluded from the study; 5) reasons for extraction were severely increased mobility, caries, fracture, Class III furcation involvement, endodontically non-

salvageable complications, or rejection of endodontic treatment by the subject; 6) teeth adjacent to the extraction sockets had to be free of overhanging or insufficient restoration margins; the restorations were replaced if insufficient margins were present; 7) third molars were excluded; 8) subjects smoking £10 cigarettes a day were considered smokers. Those smoking more were excluded from the study and were not considered eligible for the treatment described; 9) subjects who used prescription medication on a regular basis were excluded from the study; and 10) subjects who began using a medication during the study period, including the maintenance phase, were excluded from the final analysis because medications may affect the healing process, and the present study 1357



Table 3.

Distribution of Sockets by Area Region (n [%]) Anterior

Posterior

Total (n [%])

Single sockets Maxilla Mandible Total

16 (16.2) 8 (8.7) 24 (12.6)

83 (83.8) 84 (91.3) 167 (87.4)

99 (100.0) 92 (100.0) 191 (100.0)

Side-by-side sockets Maxilla Mandible Total

3 (8.3) 14 (28.6) 17 (20.0)

33 (91.7) 35 (71.4) 68 (80.0)

36 (100.0) 49 (100.0) 85 (100.0)

Total sockets Maxilla Mandible Total

19 (14.1) 22 (15.6) 41 (14.9)

116 (85.9) 119 (84.4) 235 (85.1)

135 (100.0) 141 (100.0) 276 (100.0)

was not designed to standardize or control for a medication variable. Examiner’s Calibration All measurements were taken by one periodontist who was not the surgeon who performed the procedures. After initial training of the technique on casts, the examiner was calibrated by repeated measurements of 40 single and 40 side-by-side extraction sockets on the day of extraction as well as 2 and 4 days thereafter, achieving an intraexaminer reliability of 90% (data not shown). The examiner was recalibrated once a year by measuring 10 extraction sockets of each form (single/side-by-side) following the initial protocol. Clinical Examination Prior to the surgical phase, a comprehensive dental and periodontal examination was completed on all subjects. Periodontal evaluation included measuring PD, CAL, BOP, tooth mobility, furcation involvement, and plaque scores (data not shown). Diagnostic casts were made and were mounted on a semiadjustable articulator¶ using a face-bow and a bite registration. Occlusal analysis was performed, diagnostic wax-ups were prepared on the articulated casts, and restorative treatment needs were determined. Caries lesions on the neighboring teeth were restored with composite resin as part of the initial therapy. Surgical Procedure A modification of the technique described by Bartee17 was used. An intrasulcular incision extending to the adjacent teeth was made, and a full-thickness flap 1358

was elevated. No vertical releasing incisions were made. All surgeries were performed by the same surgeon. Extractions were done using atraumatic technique, and the socket was curetted carefully and irrigated with sterile saline solution (Figs. 1A, 1B, and 2A). The socket preservation was performed using a non-resorbable dPTFE membrane# alone without the use of any soft or hard tissue grafts (Figs. 1C and 2B). No further steps were taken to secure the membrane in place. The flap was repositioned and sutured in place with interrupted sutures** (Figs. 1D and 2C). The membrane was left partially exposed during the healing period (Figs. 1D and 2C). Medication and Postoperative Care Subjects scheduled for surgery were prescribed analgesics†† (100 mg, once daily for 4 days) and a systemic antibiotic‡‡ (600 mg, once daily for 6 days). Subjects were instructed to start the medication 1 day before surgery. All subjects rinsed twice daily with 0.1% chlorhexidine digluconate solution§§ starting 1 day before surgery until 1 week after membrane removal. Sutures were left for 10 days. The membrane was removed 4 weeks after surgery. Measurements The cemento-enamel junction (CEJ) of the adjacent teeth were used as reference points. The study reports only measurements at the same area of the selected defect. Measurements were rounded up to the next millimeter. The examiner who performed the measurements was not the surgeon who provided the surgical treatment. The clinical parameters described below were assessed using a periodontal probe.ii A stent was fabricated before the extraction using a cast (Figs. 3A through 3C). The tooth to be extracted was removed from the cast, and two clasps for the adjacent teeth were fabricated. The two clasps were connected by a resin plate that was located approximately at the height of the CEJ of the adjacent teeth. In cases in which the tooth to be extracted was the most distal tooth in the sextant, a denture tooth was attached in the distal position of the stent to guarantee sufficient stability (Figs. 3B and 3C). Markings/holes were made in the following positions of the resin plate: mesial in the center of the stent, distal in the center of the stent, in the direct center, buccal-mesial, buccal-direct, buccal-distal, palatal-mesial, palatal-direct, and palatal-distal; the ¶ # ** †† ‡‡ §§ ii

SAM-2, SAM Praezisionstechnik, Gauting, Germany. Cytoplast, Regentex GBR-200, Oraltronics, Bremen, Germany. Ethibond, Excel 3-0, Johnson & Johnson, St.-Stevens-Woluwe, Belgium. Voltaren, Novartis Pharma, Nuremberg, Germany. Clindamycin, Ratiopharm, Ulm/Donautal, Germany. Chlorhexamed Fluid, GlaxoSmithKline, Buehl, Germany. UNC-15, Hu-Friedy, Leimen, Germany.



Figure 1. Surgical procedure, case 1, maxilla. A) Initial radiograph. B) Surgical site after extraction and debridement of the extraction sockets. C) Membrane placed. D) Flaps repositioned and sutured in place. E) Surgical site immediately after membrane removal. F) Surgical site after membrane removal (4 weeks after surgery).

Figure 2. Surgical procedure, case 2, mandible. A) Surgical site after extraction and debridement of the extraction sockets. B) Membrane placed. C) Flaps repositioned and sutured in place. D) Surgical site immediately after membrane removal. E) Surgical site 8 weeks after membrane removal. F) Surgical site 8 weeks after membrane removal. After staining with iodide solution, the unchanged position of the mucogingival junction is visible; no loss of keratinized gingival is noted.

line from buccal-direct to palatal-direct was defined as the ‘‘buccal–palatal line.’’ Measurements were taken during the surgery after tooth extraction, elevation of the flap, and removal of any remaining soft tissue. One year after the initial surgery, measurements and orthopantomographs were repeated by placing

the stent into the same position and bone sounding the area with the same periodontal probe¶¶ after delivering a local anesthetic (Fig. 3D).

¶¶ UNC-15, Hu-Friedy.

1359



newly formed tissue. Additionally, panoramic radiographs were taken at the 12-month follow-up examination. Implant insertion and/or prosthodontic treatment was performed 14 months after membrane removal. Data Analysis BL measured at mesial, distal, buccal, palatal, and central sites as well as in the buccal–palatal direction was described in absolute values and percentages. The percentage of BL versus the baseline clinical bone level (CBL) was expressed as means and minimal and maximal losses. The variables describing BL were reported individually for all cases reviewed as well as in groups by factors of interest investigated, i.e., socket form, jaw, and region. Bone gain (BG) in vertical defects was expressed as means – SD and minimal, maximal, Figure 3. Fabrication of the stent. A) Cast with stent. B and C) Modified stent for cases with missing and median values. distal adjacent tooth. D) The stent is placed; taking the measurements. Interval-scaled variables were examined for their normal distribution with the Kolmogorov-Smirnov test. Those Histology that deviated significantly from the normal distribuHard tissue biopsies were taken using a trephine bur## tion were analyzed with non-parametric tests. For from 10 representative cases during implant placement all others, parametric tests were used. 12 months after socket preservation. At the time of the The independent-samples t test was used to detect biopsies, clinical conditions at augmented sites in all any effects of gender and smoking on BL in vertical subjects were similar, and the clinical appearance imdefects. The effects of these two variables on the plied that the tissue had healed. The bone core samples percentage of BL were established with the Mannwere submitted for histologic evaluation (Figs. 4A Whitney U test. The Pearson correlation was used to through 4C). Samples were decalcified, stained with investigate the influence of age on BG in case of verhematoxylin and eosin, and evaluated under a light mitical defects. croscope at ·200 and ·400 magnification (Figs. 4D BL was taken to reflect absolute changes without and 4E). regard to baseline CBL, whereas the percentage of BL was considered to show changes versus baseline Supportive Periodontal Therapy and CBLs. The effects of the baseline CBL on the mesial, Prosthodontic Treatment distal, buccal, and palatal BLs were evaluated with the During the first 8 weeks after surgery, subjects underx2 test, those on the direct center were evaluated with went maintenance therapy with a dental hygienist the Kruskal-Wallis test, and those on the buccal–palonce a week. Subjects were instructed to rinse with atal BL were evaluated with the Mann-Whitney U test. 0.1% chlorhexidine digluconate*** once per day. The effects of CBL on the percentage of BL were anAfter this time, subjects were enrolled in a supportalyzed with the Kruskal-Wallis test at the mesial, ive periodontal therapy schedule consisting of regular distal, buccal, and palatal sites and with the Spearman recall appointments every 2 months. During the recall correlation at the direct center and buccal–palatal appointments, oral hygiene instructions were given, sites. BOP and PI was measured, supra- and subgingival deClassification tree analysis was performed to shed bridement was performed, chlorhexidine stains were light on the effects of the socket form, the jaw, and the removed, and the teeth were polished. PD was mearegion as well as BOP, age, gender, and smoking on sured at 6-month intervals (data not shown); howBL and percentage of BL. This served to rank the ever, no PD measurements were recorded during the maintenance visits at the teeth neighboring ## Aesculap, Tuttlingen, Germany. the sockets to prevent any possible damage to the *** Chlorhexamed Fluid, GlaxoSmithKline. 1360



Figure 4. Histologic sampling. A) Site before taking of the histologic sample, flap elevated. B) Trephine bur with sample. C) Retrieved sample. D and E) Histologic sections of newly formed bone. (Hematoxylin and eosin; original magnification: D, ·200; E, ·400.) F) Radiograph of the sampling area.

considered influencing variables in terms of the intensity of their effect and to detect interactions between them. A 5% significance level was defined for all statistical procedures. All tests were two-tailed. Statistical software††† was used throughout.

RESULTS A total of 276 sockets (191 single and 85 side-byside) in 276 subjects were treated (Tables 1 and 2). A total of 135 sockets (49%) were in the maxilla, and 141 sockets (51%) were in the mandible (Table 3); 44.9% of the subjects treated were smokers (Table 1). None of the subjects enrolled in this study reported any unusual pain or discomfort, abscess, swelling, or allergic reactions during the course of treatment. Membranes were left partially exposed after surgery. No sign of acute inflammation, exudate, or pain was detected. Plaque accumulation was observed on exposed surfaces of the membranes. After membrane retrieval, non-epithelialized soft tissue was found in the areas previously covered by the membrane (Figs. 1E and 2D). This tissue completely reepithelialized clinically within 4 weeks after membrane removal (Figs. 1F, 2E, and 2F). Nevertheless, a slight but clearly distinguishable difference in color compared to the adjacent mucosa persisted (Figs. 1F, 2E, and 2F). Clinically, the whole keratinized gingiva was preserved. Bone tissue with a regular trabecular structure was found histologically. Osteocytes as well as osteoblasts signaling active bone formation were present. Local-

ized areas of bone marrow with lymphocytes and, rarely, granulocytes were seen (Figs. 4D and 4E). Subjects who had undergone previous periodontal therapy were examined at each supportive periodontal therapy visit during the maintenance phase; none of the subjects had BOP >8% (range 5% to 8%). CBL ranged from 3 to 7 mm at mesial, distal, buccal, and palatal sites (Table 4). The baseline CBL in the direct center was 4 to 6.5 mm (mean, 5.1 mm). The buccal–palatal CBL was 5 to 14 mm (mean, 9.8 mm) at baseline (Table 4). One year after surgery, BL was estimated at 0.5 to 2 mm (Table 5). No correlation between baseline CBL and BL was found mesially, distally, buccally, or palatally (x2 test, P >0.100) in the direct center of the sockets (Kruskal-Wallis test, P >0.500) or at the buccal– palatal site (U test, P >0.500). However, the percentage of BL correlated significantly with the baseline bone level (CBL) at all sites (Kruskal-Wallis test, P