Soft-Tissue Changes After Using Collagen Matrix Seal or Collagen Sponge With Allograft in Ridge Preservation: A Randomized Controlled Volumetric Study

Horizontal and vertical morphological and dimensional alterations in both hard and soft tissues typically occur following tooth extraction. Alveolar bone resorption varies; at 6–7 months after extraction, horizontal loss is expected to occur at rates of 29%–63% (mean width reduction of 3.79 mm), with vertical loss expected of 11%–22% (mean height loss of 1.24 mm).^1–3  Several systematic reviews that have investigated the effect of alveolar ridge preservation (ARP) using various scaffolds or membranes on hard-tissue changes have pointed to a significantly greater reduction in vertical and horizontal bone resorption, although not complete elimination, as compared with unassisted socket healing.^1,3–9  A recent network meta-analysis of randomized controlled clinical trials¹⁰  ranked several ridge-preservation procedures using allografts, xenografts, or extraction only without any bone grafts, concluding that extraction-only procedures were the least effective treatments and that all other treatment options were associated with minimal width and height changes. In addition, when ARP procedures were used in combination with different types of membranes, no statistically significant differences existed between allografts (mineralized freeze-dried bone allograft [FDBA] or demineralized FDBAs) and xenografts for bone height and width.

The goal of ARP is to maintain the available volume of bone for dental implant placement and to minimize ridge dimensional changes. Studies of this process have largely focused on hard-tissue changes; few studies have been conducted on soft-tissue or volumetric changes after ARP procedures. In general, limited data are available on the effects of ARP techniques on soft-tissue volume alterations after tooth extraction. Loss of the “bundle bone” of intact sockets and the partial or total loss of the buccal plate have been shown to result in marked bucco-oral alterations, with higher resorption amounts at the buccal aspect and a collapse of the coronal buccal soft tissues. Insufficient buccal side volume may have negative effects on the esthetic outcome and cause food accumulation and patient discomfort.¹¹ 

Schropp et al¹²  conducted one of the first landmark volumetric cast analyses of soft tissue. They observed a 5- to 7-mm horizontal loss of volume, which corresponded with a 50% loss of alveolar bone width. Recently, noninvasive assessment methods using digital optical scanning have been applied to assess volume changes in oral tissues.^13–15  Thalmair et al¹⁶  and Schneider et al¹⁷  conducted volumetric analyses of various ARP techniques with and without xenografts, in conjunction with a free gingival graft or a collagen matrix seal (CMS). They observed a horizontal volume change in the contour of buccal soft tissue in all groups and a reduced volume loss compared with spontaneous healing. Today, it is still not known which ARP procedure best preserves soft-tissue profiles. Thus, any materials able to predictably maintain or increase soft-tissue volume are beneficial.

Collagen matrix seal, a non–cross-linked extracellular xenogeneic barrier (Mucograft, Geistlich Pharma AG, Wolhusen, Switzerland) with a diameter of 8 mm and designed to cover sockets after bone grafting, was recently introduced into clinical practice for ARP procedures. A recent clinical trial indicated that the use of a CMS, in combination with FDBA, minimizes bone resorption in sockets with a buccal vertical loss of <2 mm and is also potentially beneficial for soft-tissue preservation.¹⁸  The aim of the current study is to volumetrically assess buccal soft-tissue contour alterations following 4 months of healing after ARP using either a CMS or a collagen sponge (CS) with FDBA. This study is based on a recently published randomized clinical trial that assessed clinical and radiographic changes¹⁸  and observed a similar reduction in both groups in vertical buccal bone loss (0.30 mm in CMS and 0.79 mm in CS) and the coronal part of the ridge width (1.21 mm, 14.91% in CMS and 1.47 mm, 20.40% in CS) as well as a slightly better buccal gingival thickness at the coronal width in both groups (0.9 mm in CMS and 0.5 mm in CS).

This study investigates the volumetric analysis of a recently published randomized, controlled, 2 parallel-arms, prospective, single-center clinical trial examining the effect of CMS or CS with FDBA in ARP, for which the clinical and radiographic results have already been reported.¹⁸  This study followed the CONSORT statement (http://www.consort-statement.org/), was conducted in accordance with the Helsinki Declaration of 1975 as revised in 2003, was registered at ClinicalTrials.gov (NCT02697890), and was approved by the Institutional Review Board of Tufts University School of Dental Medicine (IRB 11360).

Informed consent was acquired from each patient before the start of the study (by both Z.S.N. and Y.N.J.). The patients were enrolled and treated between August 2015 and May 2016. The main inclusion criterion was patients who had a treatment plan for delayed implant placement with the need for extraction and ARP. The included tooth had to be single rooted (excluding the lower incisors). Other necessary inclusion criteria were that patients had to be at least 18 years old, have no acute infection or active periodontal disease, smoke no more than 9 cigarettes per day, and have a buccal plate based on a cone-beam computerized tomography examination. Patients were excluded if they were pregnant or lactating or had any personal reasons to oppose periodontal surgery. Twenty-eight patients met all the inclusion criteria and were included in the study.

Convenience sampling with random assignment to groups using a computer-generated list was used, and each participant was assigned to a group on the day of surgery. One investigator (Y.N.J.) conducted block randomization and random permutation of treatment allocation within each block. Only 1 tooth was included per patient; investigators were not informed of the assigned treatment, and the group allocation for each participant and surgeon was concealed in a locked folder until the day of the surgery. Each record was assigned a number. The patients, biostatistician, and investigator (Z.S.N.) who performed all the volumetric measurements were blinded.

A soft polyether silicone impression (Impregum, 3M EPSE, Seefeld, Germany) was first taken of the relevant jaw. Then, before tooth extraction and ARP, cast models were fabricated from a type IV dental stone (Suprastone, Kerr Co, Orange, Calif). The treatment protocol has been described in detail previously.¹⁸  In brief, an atraumatic extraction using a flapless technique was conducted with periotomes and extraction forceps. FDBA (Mineross, Biohorizons IPH, Birmingham, Ariz) was used as a scaffold, and a CMS or a CS (HeliPLUG, type I bovine collagen 1 cm × 2 cm, Integra Life Sciences, Plainsboro Township, NJ) was adapted to seal the extraction socket and sutured with nonresorbable monofilament horizontal mattress and interrupted sutures.

Postoperative instructions included the prescription of 500 mg of amoxicillin (3 times a day for 8 days) and chlorhexidine gluconate mouth rinse 0.12% (twice a day for 3 weeks). Patients were seen again at 1, 2, and 4 weeks postoperatively. Sutures were removed at 2 weeks. Four months after extraction, a second impression was taken using the same impression material, and cast models were fabricated.

Casts of each patient at baseline and at 4 months were optically scanned and digitized with a 3-dimensional scanning system (Activity 2.8Ink, Smart Optics Corporation, Olso, Norway) to create Standard Tessellation Language files. These files were imported into a specific type of software (Exocad Dental CAD, Exocad GmbH, Darmstadt, Germany) and superimposed using a best-fit algorithm of 4 reference points (the cusp tips and buccal surfaces of adjacent teeth). The tooth to be extracted was removed virtually in the baseline cast with a third type of software (Meshmixer, Autodesk Research, San Francisco, Calif). Then, the surface area and volume in the buccal alveolar ridge area were measured, both before tooth extraction and at the 4-month follow-up (Figures 1 and 2).

The area of measurement was defined both mesially and distally by a line parallel to the mesial or distal aspects of the adjacent teeth, coronally by the top of the ridge surface, and apically by an apical line determined by the apical buccal horizontal hole of a stent used for clinical and radiographic measurements that corresponded to 10-mm apical to the cementoenamel junction of the tooth to be extracted, as described previously.¹⁸  Because of anatomic variation in each participant, the measured area was diverse between participants. However, it was the same in one participant site at baseline and at 4 months. Because the size of the focal area varied from site to site, in terms of differences in the extracted tooth size and adjacent teeth (premolars or incisors), the mean volume alteration per area was calculated as a distance-adjusted change and compared between baseline and 4 months using a distance adjustment in the buccal direction, to allow for direct comparison between sites.^16,17,19  To avoid exposure of any patient's identity or treatment assigned in the final analysis, each cast was assigned a number that linked it with all other files. All measurements were calculated twice by a calibrated investigator (Z.S.N.).

The surface area (mm²) and volume (mm³) were recorded, and changes were calculated both before extraction and 4 months after ARP. The distance adjusted (Δd [mm]) was calculated using the following equation Δd = Δ volume (mm³)/area (mm²).

The sample size calculation was described in the previous study.¹⁸  Fourteen participants per group (28 total) were needed to achieve 80% power to detect a 10% difference in bone width, with a 20% potential dropout. Based on the current outcome, however, using the G*power program (version 3.1.9, Faul F, Erdfelder E, Lang A-G, and Buchner A, Germany) and adjusting for distance, 20 participants per group (40 total) were needed to achieve the same power. Therefore, based on the means and standard deviations observed in both groups, the current study used a post hoc power analysis and 64% power to detect a true mean difference of 0.92 mm between the 2 groups.

The primary outcome variable for this exploratory analysis was the distance being adjusted 4 months after tooth extraction. Statistical analysis was performed following the per protocol analysis using a statistical software program (SAS, version 9.3, SAS Institute Inc, Cary, NC; SPSS 20, IBM Corporation, Armonk, NY). For each treatment group, descriptive statistics were used to determine the mean, standard deviation, median, minimum, and maximum values. A Kolmogorov–Smirnov test, used to check the normality assumption, revealed that the volume changes were not normally distributed; however, all baseline values were normally distributed. The reliability of all measurements was calculated by intraclass correlation. Baseline measurements of demographic characteristics, surface area, and volume were compared using a chi-square test, Fischer exact test, or independent-sample t test. Differences in volume and surface area within groups were tested using Wilcoxon rank-sum tests. The difference in the distance adjusted between groups was assessed using a Mann-Whitney U test. The number needed to treat (NNT) depended on detecting a difference of 1 mm between the 2 treatment groups and was evaluated based on differences between the group means. This calculated an average number of participants needing a treatment option to benefit a 1-mm lower volume change over the other. The level of significance was set at P < .05. all statistical analysis was performed by Z.S.N., Dental Public Health, School of Dentistry, King Abdulaziz University, Jeddah, Saudi Arabia.

All patients completed the study without any withdrawals (28 patients total; 14 in the CMS group and 14 in the CS group) or adverse events. Seventeen participants were male, with a mean age of 55.4 ± 15.0 years. Four participants were smokers. Most of the extracted teeth were maxillary (23 teeth), whereas 17 were premolars (Table 1). There were no statistically significant differences between the 2 groups regarding age, gender, smoking, or tooth location.

All volumetric measurements were calculated twice, and the intraclass correlations for all of them were greater than 0.90. Table 2 shows the changes in surface area and volume after 4 months of healing. No baseline measurements were statistically significantly different; no statistically significant differences in surface area were found in any group. Surface area loss was (mean ± SD) 71.44 ± 1189.09 mm² in the CMS group and 139.56 ± 557.92 mm² in the CS group. A significant decrease in volume was observed within both groups (P < .001). The CMS group had a 239.58 ± 231.89 mm³ loss, and the CS group had a 337.23 ± 310.18 mm³ volume loss.

Table 3 shows the distance-adjusted alterations from baseline to 4 months. Mean buccal soft-tissue loss and minimum/maximum distance loss were less in the CMS group (0.88 ± 0.52 mm and 0.2–2.15 mm, respectively) than in the CS group (1.63 ± 1.03 mm and 0.3–3.68 mm, respectively). However, there was no statistically significant difference between the groups in distance-adjusted loss (P = .2742).

Table 4 displays the number of sites with a loss of between <0.5 mm and >2 mm in terms of adjusted distance and the NNT to reach a difference in mean distance of at least 1 mm. An NNT of 2.8 was found when comparing the CMS and the CS groups. In other words, almost every third patient who received the CMS showed less than 1 mm of contour reduction as compared with those who received the CS.

After tooth extraction, resorption at the buccal aspect of the alveolar ridge and a collapse of the coronal buccal soft tissues are typically observed. In addition to hard-tissue alterations, substantial soft-tissue remodeling is usually also present. Insufficient buccal soft-tissue contour may have an undesirable effect on the esthetic shape of implant prostheses and pontic area. Thus, ARP procedures must preserve the original anatomical soft-tissue contours and maintain a sufficient volume of bone to allow for implant placement.

This study was conducted to assess the effect of ARP using CMS or CS with FDBA on soft-tissue contour level. After 4 months, reduction of the buccal ridge contour was found to be lower in the CMS group (mean ± SD, 0.88 ± 0.52 mm) than in the CS group (mean ± SD, 1.63 ± 1.03 mm). These findings are comparable with recent volumetric studies that showed that ARP techniques can minimize but not entirely compensate for alveolar ridge reduction.^14,16,17 

After using a xenogenic bone substitute, a free gingival graft and a combination of both, Thalmair et al¹⁶  reported a mean horizontal buccal ridge contour reduction of 0.8 ± 0.5 mm to 1.45 ± 0.7 mm.¹⁶  They discovered that different ARP techniques resulted in less reduction than unassisted socket healing, observing a loss of 2.3 ± 1.1 mm. Schneider et al¹⁷  documented a similar finding, using both demineralized bovine bone mineral with 10% collagen, sealed with a punch modification of collagen matrix (very similar to CMS) and an autogenous soft-tissue punch graft with a loss of 1.2 ± 0.5 mm, which also compared favorably with unassisted socket healing, showing a loss of 1.8 ± 0.8 mm. Barone et al¹⁴  compared the volumetric changes after ARP with porcine bone in 2 different forms (collagenated corticocancellous and cortical), in combination with a membrane, and concluded that neither technique was able to preserve a stable pristine alveolar crest, with a volume reduction ranging between 244 and 349 mm³.

An adequate amount of soft-tissue volume is of great significance to achieving a positive esthetic outcome in the anterior zone. In this study, the CMS group had more sites with a soft-tissue loss of <1 mm than the CS group (8 vs 3, respectively). Five CMS group and five CS group sites had a loss of 1–2 mm. Only 1 site in the CMS group lost >2 mm, as compared with 6 sites in the CS group. Every third socket that received a CMS showed less than 1 mm of contour reduction compared with those that received a CS. A 1-mm reduction of the soft-tissue profile loss is clinically relevant to reduce subsequent soft-tissue augmentation procedures, simplify implant placement, and optimize functional and esthetic outcomes. In fact, long-term soft-tissue volume gain after a connective tissue graft during implant placement or at stage 2 surgery is on average 1 mm.²⁰ 

In the previous study, hard- and soft-tissue changes in radiographic bone height and width and buccal tissue thickness were observed in both groups.¹⁸  Although no significant differences in vertical or horizontal bone changes were found between the 2 groups, the CMS was slightly better in increasing the mean crestal buccal soft-tissue thickness than the CS (0.9 mm vs 0.5 mm). In the current study, a significant decrease in soft-tissue contour was noticed within both groups; however, there was no significant difference between groups in distance-adjusted loss. Surface area, volume, and mean buccal soft-tissue loss were lower in the CMS group than in the CS group. The NNT analysis also showed potential benefits of the CMS; almost every third patient treated with the CMS displayed less than 1 mm of contour reduction compared with the CS.

In the previous study, the mean loss of coronal radiographic alveolar ridge width was 1.21 ± 1.22 mm and 1.47 ± 1.29 mm in the CMS and CS groups, respectively, with a mean increase in crestal buccal soft-tissue thickness of 0.9 mm in the CMS and 0.5 mm in CS groups. In the current study, the mean buccal soft-tissue loss was 0.88 ± 0.52 mm and 1.63 ± 1.03 mm in the CMS and CS groups, respectively. It seems that soft-tissue– and alveolar bone–level changes after ARP do not completely follow each other, confirming an observation previously reported by Schneider et al.¹⁷  It is not clear whether the horizontal volume loss occurred due to loss of soft tissue or underlying hard bone. Chappuis et al²¹  observed a spontaneous soft-tissue thickening, especially in thin bone phenotypes, with a 7-fold increase in thickness after single extraction in the esthetic zone. Iasella et al²²  reported an increase in soft-tissue thickness of 0.4 mm in single-extraction sites, as compared with a decrease of 0.1 mm in sites treated with FBDA and a collagen membrane. Another possible explanation for these observations is that a topographical difference exists between the bone location and soft-tissue measurements.

The current methodology used, involving optical scanning and superimposition, offers several advantages over prior approaches: a noninvasive character, lack of radiation, and ease of application. Admittedly, a limitation of this study is that the accuracy of the impressions and casts may affect that of the method. Two additional methodological limitations should be noted: the tooth to be extracted at the baseline cast was extracted only virtually, and the use of the distance-adjusted (Δd [mm]) measurement representing the volume alterations was averaged over the measured area and not at a specific point.¹⁹ 

The effects of the soft-tissue thickness and the site location in the alveolar arch (eg, bicuspids vs incisors) were not investigated in this study. It is generally assumed that a thin biotype and anterior sites are more susceptible to resorption.⁸  Although CBCT analysis (hard and soft tissue) of pre- and postprocedure would provide better data, this study used only model analysis, and there is a plan to include more data resources in the future.^23–28  Although the previous volumetric studies showed highly precise and reproducible results with a measurement error less than 10 μm,^13–17  there is a possibility of inaccuracy resulting from any of the steps, such as the impression/casting technique or software superimposition. Moreover, the results of this study cannot be generalized to various surgical procedures. A further limitation is the lack of power of the current study because of the small sample size, which is another reason that no significant differences between groups were detected. Further studies with a larger sample size are needed to verify whether the use of CMS provides a protecting effect in soft-tissue contour reduction after ARP.

After tooth extraction, both ARP barriers were found to be unable to completely prevent soft-tissue contour alterations. Application of a CMS in combination with FDBA was slightly better in minimizing the amount of volume loss compared with the use of a CS. Every third socket treated with a CMS showed less than 1 mm of contour reduction as compared with a CS.

Abbreviations

ARP

alveolar ridge preservation

CMS

collagen matrix seal

CS

collagen matrix

FDBA

freeze-dried bone allograft

NNT

number needed to treat

The authors acknowledge our independent statistician, Dr Matthew Finkelman, Department of Public Health, School of Dental Medicine, Tufts University, Boston, Massachusetts, for his effort in reviewing the methodology and the article.

The authors declare no conflicts of interest related to this study.