Comparison of Dehydrated Human Amnion-Chorion and Type 1 Bovine Collagen Membranes in Alveolar Ridge Preservation: A Clinical and Histological Study

Alveolar ridge collapse is a consequence of tooth extraction that often complicates future implant placement. The preservation of alveolar ridge dimensions is essential to optimizing implant outcomes from both a functional and esthetic perspective. A systematic review conducted by Hammerle et al²  reported postextraction average ridge width and height reduction of 3.8 mm and 1.24, respectively. Schropp et al¹  reported a 50% reduction in the original ridge width over a 1-year healing period. Two-thirds of this reduction occurred within the first 3 months after extraction. Therefore, alveolar ridge preservation (ARP) procedure to avoid ridge dimensional changes is recommended.¹ 

Alveolar ridge preservation techniques utilize a variety of grafting materials and membranes with varying degrees of resorption reported.^3–6  Fickl et al⁶  conducted a randomized clinical study to evaluate the use of different ARP techniques. The study demonstrated that significantly better ridge width preservation was achieved using a combination of a barrier membrane or soft tissue punch and xenograft than socket grafting alone.⁶  In a systematic review and meta-analysis conducted by Troiano et al,⁷  the combination of bone grafting and a resorbable membrane provided better outcomes of ARP compared with spontaneous healing. The meta-analysis reported a mean width and height reduction of 2.19 and 1.72 mm, respectively.

The use of a resorbable collagen membrane (RCM) as an exposed barrier in ARP is well documented in the dental literature. This is attributed to the favorable biological and physical properties of type 1 collagen that include biocompatibility,⁸  affinity for periodontal ligaments and gingival fibroblasts, facilitation of hemostasis, clot stabilization,^9,10  handling simplicity,¹¹  lack of immunogenic reactivity,¹²  and space maintenance capability.¹³  Clinical studies performed by Iasella et al,¹⁴  Kutkut et al,¹⁵  and Cook and Mealey¹⁶  used collagen membranes in open-barrier technique for ARP. The percentages of vital bone (VB) and the ridge dimensions in these studies were considered favorable for implant placement.

The human amnion-chorion membrane (ACM) is a resorbable, lightweight thin barrier (300 μm)^17,27 ; derived from the human ACM. Anatomically, the human ACM is composed of 2 layers, the amnion and the chorion, which forms the inner and outer layers of the amniotic sac, respectively. Both layers originate from the trophoblast layer, a cell layer that have been reported to produce no foreign-body reaction due to the lack of human leukocyte antigens.¹⁹  The pluripotency of the ACM cells allows them to differentiate into numerous cell types including osteoblasts.^20–22  Histologically, the amnion consists of a single cell layer resting on a basement membrane. The basement membrane is composed of structural proteins, such as collagen types III, IV, and V, and an abundance of laminin 5. Laminin 5 is an extracellular matrix that plays a major role in the binding and adherence of gingival epithelial cells.²³  In addition, the amnion layer contains functional cytokines that are important for cell signaling, such as fibroblast growth factor (FGF), platelet-derived growth factor (PDGF), keratinocyte growth factor (KGF), basic fibroblast growth factor (bFGF), and transforming growth factor β (TGFβ).²⁴  Furthermore, tissue inhibitor of metalloproteinase-1 (TIMP-1), which is a glycoprotein that inhibits extracellular matrix degradation and improves cellular proliferation, is also a component of the amnion layer.²⁴  The chorion layer is composed of collagen types I, III, IV, V, and VI and laminins.^24–26 

Application of ACM in ARP is plausible because of the biological properties. Two case series studies evaluated the use of ACM in ARP with either primary¹⁷  or secondary wound closure (open barrier technique).¹⁸  In these reports, the outcome of ARP was encouraging. Only 1 randomized split-mouth clinical trial was published comparing the outcome of ARP using ACM to that achieved by dense polytetrafluoroethylene (d-PTFE) membrane. The authors found that ACM yields comparable results in preservation of ridge dimensions as d-PTFE when the membranes were left exposed. However, the ACM group had significantly higher bone volume density and more osteoid formation compared with the d-PTFE group at 3-month histomorphometric analysis.

Limited well controlled clinical studies are published on the use of ACM as a barrier membrane in ARP. Therefore, the primary objective of this study was to compare the use of a human ACM to a collagen membrane in an exposed-barrier ridge preservation technique. The second aim was to determine if intentional membrane exposure compromises ridge dimensions and bone vitality.

The current clinical study was approved by the Institutional Review Board of the University of Kentucky and was registered with clinicaltrials.gov (ID# NCT03290638). All patients were recruited from the dental clinics of the University of Kentucky College of Dentistry between October 2013 and October 2015 and signed a written informed consent. Entry criteria included patients with partial edentulism who had at least 1 nonsalvageable posterior tooth that was planned for extraction and delayed implant placement. Exclusion criteria included patients who had received radiation therapy, were taking antiresorptive medication, had uncontrolled systemic diseases, were heavy smokers (more than 10 cigarettes daily), were pregnant or breastfeeding. Teeth with acute odontogenic infections, or extraction sockets that demonstrated facial bony dehiscence of 4 mm or greater were excluded.

The enrolled patients were randomly assigned into either control group (received RCM membrane) or test group (received ACM membrane) by a computer-generated randomization table. Extraction sockets of both groups were grafted with cortical demineralized freeze-dried bone allograft (DFDBA) particulate (particle size, 0.125–0.850 mm, 0.5 cc, Maxxeus). Extraction sockets of the test and control groups were covered with ACM membrane (BioXclude, Snoasis Medical) and RCM membrane (Mem-Lok, BioHorizons), respectively. Membranes were sutured utilizing an open barrier technique for both groups.

A comprehensive oral and periodontal examination was performed during the screening visit. Any subject who fulfilled the inclusion criteria were enrolled in the study after obtaining research and clinical consent forms. An alginate impression was taken for each subject and a diagnostic cast was generated; a baseline stent was used fabricating by trimming the crown of the unsalvageable tooth (Triad, DENTSPLY). Four reference holes were drilled using a carbide straight fissure bur through the stent at the mid-buccal, mid-palatal, mid-mesial, and mid-lingual positions, corresponding to the 4 points of ridge height measurements determined on the cast.

All procedures described above were the standard of care except for bone core biopsy, which was the research component.

All subjects were given prophylactic antibiotics, either 2 g of Amoxicillin or 600 mg of clindamycin (in case of a penicillin allergy). Subjects were also premedicated with 600 mg of ibuprofen for initiating pain control and an anti-inflammatory response. A mouth rinse consisting of 0.12% chlorhexidine (Chlorhexidine Gluconate Oral Rinse, 0.12%: PerioGard Colgate) for 30 seconds was performed before the surgical procedure. After administration of a local anesthetic, a sulcular incision was made and extended to the midline of the adjacent teeth with no releasing incisions. A full-thickness buccal and lingual flaps were elevated to adequately expose the buccal and lingual plates of the socket for direct clinical measurements. Atraumatic tooth extraction was carried out. Cautionary procedures performed to avoid facial wall fractures included the use of PIEZOSURGERY (PIEZOSURGERY touch, Mectron), periotomes (Salvin Dental), root sectioning burs (Brasseler), or elevators (Hu-Friedy).

After extraction, clinical baseline ridge height and width measurements were obtained. Socket debridement and copious saline irrigation was performed. The extraction sockets were grafted with a mixture of cortical DFDBA particulate (particle size, 0.125–0.850 mm, 0.5 cc, Maxxeus) that had been hydrated in sterile water for 30–45 minutes.

In the test group, an ACM membrane was placed on the grafted socket, whereas in the control group, an RCM membrane was placed. Both membranes were extended at least 3 mm below the socket crest on the buccal and lingual sides and secured with 3-0 chromic gut horizontal mattress sutures (Ethicon Suture).

Both test and control groups received postoperative antibiotic therapy for 10 days as follows: 500 mg of amoxicillin 3 times daily or 150 mg of clindamycin 4 times daily. For pain control, all patients were prescribed 800-mg ibuprofen tablets. The control group was instructed to use 0.12% chlorhexidine mouthwash twice a day for 2 weeks. As per the ACM manufacture instructions, ACM membrane contains bioactive charged proteins that may bind to the chlorhexidine cation and reduce the rate of cellular migration across the membrane. Therefore, the test group was prescribed 1.5% hydrogen peroxide wash twice a day for 2 weeks. Wound healing was evaluated 2 weeks after extraction using clinical macro lens photography.

Full-thickness buccal and lingual flaps were elevated to adequately expose the buccal and lingual plates of the socket for direct clinical measurements. The clinical measurements were obtained in 2 timepoints as follows:

• Baseline visit:Immediately after extraction, clinical baseline ridge height measurements were obtained using the custom-made stent as a reference and a UNC-15 periodontal probe at 4 sites: mid-mesial, mid-distal, mid-buccal, and mid-lingual. The ridge width measurements were obtained using a metal bone caliper at the height of contour (Hartzell and Son). The clinical measurements were rounded to the nearest half of a millimeter.

• Implant placement visit: Before harvesting the bone core biopsy, ridge height measurements were repeated using the same baseline stent, and ridge width measurements were repeated as well.

At the time of implant placement (average healing period of 19.5 weeks), a mid-crestal incision was performed and full-thickness buccal and lingual flaps were elevated. The ridge height and width measurements were repeated. A trephine drill (internal diameter, 2 mm; outer diameter, 2.75 mm; Straumann) was used at a length of 8 mm to collect the bone core from the center of the grafted socket. Implant site preparation and implant placement (Straumann) was performed according to the clinical protocol (Figure 2a through t). If the thread of the body of the implant was exposed, bone grafting buccal to the implant was performed. Implants were successfully installed during implant placement appointments for all subjects without the need for lateral ridge augmentation and staged implant placement protocol. The implants were assessed radiographically after placement and at the second-stage surgery for bone level implants or at the time of abutment installation for tissue level implants based on the clinical protocol.

The bone specimens were placed and stored in 10% neutral buffered formalin at 4°C for 2 days. In the laboratory, the samples were dehydrated in graded alcohol, then embedded in paraffin wax. The bone specimens were sectioned transversely at the peripheries and the middle of the bone core to sample representative regions of interest. At least 3 bone sections (∼5 μm) were obtained from each specimen using Leica microtome (Leica RM2255 Biosystems). The sections were mounted on slides, stained with hematoxylin and eosin, and examined under light microscopy at 100× (Semi-Motorized Microscope, Olympus BX53).

Histomorphometric analysis of each section was performed by a blinded investigator using specialized software (BIOQUANT). In each section, 5 areas were examined and quantified as follow:

• Vital bone was defined as bone with viable nucleated osteocytes within the lacunae. Vital bone was generally associated with regions of vasculature and connective tissues (CTs) or graft particles (GP) that were integrated and connected to the VB region (Figure 3a).

• Graft particles were identified as either lamellar bone with empty lacunae that were devoid of nuclei or bone chips resulting from trephination or sectioning of the bone core. Graft particles were associated with the margins of the core or were laying or floating on the CTs and were not connected to or integrated with VB (Figure 3b).

• Newly mineralizing bone (NMB) was identified as well-organized collagen fibrils that were darker in color than the surrounding CTs, had no lamella or lacunae, were within a dense active area of CTs, and appeared to be bridging the various components of the specimen (Figure 3a).

• Connective tissue was identified as loose fibrous tissue with fibroblasts or adipocytes, vasculature, and irregularly organized collagen fibers (Figure 3a and b).

• Voids were identified as empty tissue spaces between various components of the section.

For each section, the mean percentages of VB, NMB, GP, CT, and voids were measured and calculated per area.

A sample size of 12 in each group provides 80% power. Sample size and power calculation were based on previous randomized controlled clinical trials. The data are presented as the means and standard deviations. All outcomes of interest were continuous variables. The group outcomes were compared using 2-sample t tests. The data were analyzed using JMP Software (JMP, Version 9.4., SAS Institute, Inc). The level of significance was set at P < .05.

Fifty patients were enrolled into the study. However, 7 patients with 8 sites (5 in the test group and 3 in the control group) did not return following ARP and thus were excluded from the study. A total of 43 patients (16 males and 27 females) with an age range from 25 to 77 years (mean age 54 years) completed the study (Figure 1). A total of 51 sites were examined. Twenty-six sites were included in the test group and 25 sites in the control group. Forty-three implants were placed as follows: 22 (12 nonmolar sites and 9 molar sites) in the control group and 21 (8 nonmolar sites and 11 molar sites) in the experimental group. Forty-three bone cores were collected.

Two subjects in the control group required simultaneous ridge augmentation during implant placement. Healing or epithelialization of the mucosa after bone grafting and intentional membrane exposure was clinically observed using clinical photography. Both groups exhibited clinical evidence of epithelialization at 2 weeks without signs of infection (Figure 4k through r).

Healing view after 2 weeks is as follows (Figure 4k through r): (k and m) the human amnion membrane was left exposed; (l and m) epithelialization at 2 weeks after extraction; (o and q) the collagen membrane was left exposed; (p and r) epithelialization at 2 weeks after extraction.

Five of the 43 bone core biopsies were excluded from the experimental group, and 3 samples were excluded from the control group as a result of inadequacy that impeded their processing. Analysis showed no statistically significant differences in the mean percentages of VB (P = .7567), GP (P = .7082), NMB (P = .8765), CT (P = .5428), or voids (P = .6606) between the groups.

Table 1 shows the percentage and standard deviations of VB, GP, NMB, CT, and voids for nonmolar teeth, molar teeth, and the total number of teeth for the test and control groups.

Three sites (2 in the experimental group and 1 in the control group) were excluded from the clinical measurements due to facial bony dehiscence of 4 mm or greater after extraction. These sites however were treated according to the surgical protocol and included in the histomorphometric analysis (Figure 4a through j).

The difference in ridge height was not statistically significant between the groups (P = .7648). Test and control groups exhibited a mean ridge height loss of 0.4 ± 0.11 mm and 0.4 ± 0.13 mm, respectively. Furthermore, no statistically significant difference was observed in ridge width reduction between the groups (P = .1622). Test and control groups exhibited an average ridge width loss of 0.2 ± 0.24 mm and 0.5 ± 0.4 mm, respectively.

Table 2 shows the means and standard deviations of ridge dimensional change in millimeter for nonmolar teeth, molar teeth, and the total number of teeth for test and control groups.

Alveolar ridge preservation outcomes seems to be influenced by the selection of regenerative biomaterial. The decision making on which biomaterial to use is dependent on multiple factors, of which is clinical judgment based on the available evidence that either support or discourage its application. Although there is strong evidence in favor of RCM and nonresorbable d-PTFE in open barrier ARP, evidence supporting ACM efficacy in ARP is lacking. In the current study, the objective was to compare the percentage of newly formed VB and alveolar ridge dimensional changes between ACM and RCM membranes in open barrier technique ARP.

Studies investigating the effect of ACM on ARP outcomes reported similar percentage of VB to the ACM membrane group in the current study. Cobb and Wallace¹⁷  published a case series study in which VB and ridge dimensional changes were assessed after an average healing time of 13 weeks following primary closure ARP with ACM membrane. The mean percentage of newly formed VB was 54%, which is consistent with the results of the current study. The percentage of VB in the current study compares favorably with that reported in Holtzclaw's ARP case series (mean percentage of VB is 29.28%),³³  in which a ACM membrane was used as an exposed barrier. The control group (RCM) in this study also has similar percentage of VB as published studies that used RCM as exposed barrier,^37,38  for nonmolar^16,15  or molar teeth.³²  In the current study, the percentage of VB was assessed in an average healing period of 19.5 weeks. Beck and Mealey³⁷  showed no significant difference in bone vitality between 14 and 27 weeks of healing using mineralized allografts, with a mean vitality of 45.8% and 45% after 27 weeks, respectively.

Ridge dimensional changes in this study correspond to those reported in the literature for ACM membranes,³³  and RCM membranes whether primary closure was attempted^30,31  or not.^14–16  In the current study, the implemented ARP protocol provided consistent dimensional changes outcome irrespective of the location of the grafted socket in the mouth, a finding that is consistent with studies assessing molar³²  and nonmolar sites.^14–16 

Although the manufacturer's instruction for ACM use in socket grafting (if all 4 walls are intact) recommended only tucking the membrane under the gingival collar prior to suturing, the study protocol necessitates elevation of full thickness flap for direct ridge dimensional changes measurements. Interestingly enough, the amount of alveolar bone height loss in the current study is close to the amount of crestal bone loss around teeth after full thickness flap elevation.^28,29  Barone et al³⁴  concluded that grafted sites allowed placement of larger implants and required fewer augmentation procedures during implant placement compared with naturally healed sites. In the current study, the shortest implant placed was 8 mm, and the narrowest implant diameter was 4.1 mm. Fewer implants needed simultaneous grafting at the time of implant placement.

The fundamental goal that has been the main focus of ARP is preservation of alveolar bone dimensions. Although soft tissue changes translated by loss of keratinized tissue width is well established especially if primary closure is attempted, little to no attention was given to soft tissue dimensional preservation as a goal of ARP procedures. Open barrier ARP has an advantage over traditional primary closure ARP in preserving vestibular depth and increasing the amount of keratinized tissue while achieving similar ridge preservation outcomes³⁶; thus, allowing for optimal hard and soft tissue preservation rather than future site development. Barone et al³⁵  reported a gain of 1.8 ± 0.8 mm of keratinized gingiva using open barrier ARP approach and a loss of 1.7 ± 0.6 mm of keratinized gingiva using the primary closure ARP approach. The flap approach included 2 vertical releasing incisions to achieve primary closure. In the current study, no vertical releasing incisions were made for flap elevation; instead, sulcular incisions were extended to the middle of the adjacent teeth on the buccal and lingual aspects. No primary closure was achieved. Ridge epithelization over both ACM and RCM membranes was observed 2 weeks after extraction with no evident signs of infection during the healing process in either group.

One limitation in the biopsy technique in the current study is the inability in ensure exclusion of sampling the septum area of molar teeth. However, comparing percentage of VB and dimensional changes to nonmolar sites revealed no significant difference between the subgroups. Another limitation that may have affected histomorphometric outcomes is the variation in the timeline of bone harvesting (approximately 19.5 weeks). Future studies assessing molar sites could implement strict sampling technique to ensure exclusion of septal bone and use strict sampling timeline to provide better overall analysis of samples harvested within the same healing period.

In the current study, ACM and RCM membranes were covered by gingiva 2 weeks after extraction. No soft tissue measurements were compared between the groups. Future studies may consider evaluation of soft tissue dimension improvement with ACM membrane.

Future studies could also focus on the thickness of the membrane and the residual concentration of the following constitutions after epithelial closure: collagen fibrils (types I, III, IV, V, and VI), laminin 5, FGF, PDGF, TGFβ, KGF, bFGF, KGF, and TIMP-1.

Histologic and clinical outcomes after open barrier ARP using ACM membrane is comparable with sites treated with RCM membranes. Both membranes yield uneventful healing and successful ARP outcomes that are appropriate for implant placement.

Abbreviations

ACM:

amnion-chorion membrane

ARP:

alveolar ridge preservation

bFGF:

basic fibroblast growth factor

CT:

connective tissue

DFDBA:

demineralized freeze-dried bone allograft

d-PTFE:

dense polytetrafluoroethylene

FGF:

fibroblast growth factor

GP:

graft particles

KGF:

keratinocyte growth factor

NMB:

newly mineralizing bone

PDGF:

platelet-derived growth factor

RCM:

resorbable collagen membrane

TGFβ:

transforming growth factor β

TIMP-1:

tissue inhibitor of metalloproteinase-1

VB:

vital bone

This study was partially funded by Graduate Periodontics, College of Dentistry University of Kentucky and Snoasis Medical.

All authors declare that they have no conflicts of interest to report.