A randomized, single-blinded, single-center clinical trial was conducted to evaluate the clinical feasibility and bone regeneration capacity around peri-implant dehiscence defects of two types of membrane. The study was approved by the Institutional Review Board (IRB) for Clinical Research at Dental Hospital of Yonsei University (approval no. 2-2013-0021). All patients provided written fully informed consent in accordance with IRB guidelines for enrollment, and the study was conducted in accordance with the Declaration of Helsinki and the Guidelines on Good Clinical Practice [19].

A total of 43 patients who needed single-tooth implant treatment from August 2013 to October 2014 were included in this clinical study. The following inclusion criteria were applied: (1) male or female aged ≥20 years, (2) healthy systemic condition (including well-controlled medical illnesses), (3) a vertical dehiscence defect (only on the buccal side) of ≥3 mm present immediately after implant placement, (4) secure primary stability of the implant, and (5) no allergic reaction to collagen. The following exclusion criteria were applied: (1) severe or uncontrolled systemic disease, (2) advanced or untreated periodontitis, (3) pregnancy or breastfeeding, (4) history of radiation therapy to the head or neck, (5) hormones or bisphosphonate therapy affecting bone or connective tissue metabolism, and (6) heavy smoking (>20 cigarettes/day).

The required sample size was determined using the two-sided t-test at an alpha level of 0.05 and a statistical power of 80%. The threshold for differences in bone regeneration capacity between the NCM and DCM groups was set to 1.0 mm, and the standard deviation was assumed to be the same for both groups according to the results of a previous study [2021]. These parameters resulted in a required sample size of 28 patients, and so 30 patients were enrolled (15 patients in each group) to account for a potential dropout rate of 10%. The statistical power was calculated using G* Power 3.1 (University of Duesseldorf, Germany) [22].

Randomization took place after implant placement using online databases for clinical trials (Sealed Envelope™, sealedenvelope.com). The 30 enrolled patients were assigned to either the NCM group (n=15) or the DCM group (n=15) according to computer-generated random numbers. None of the patients knew whether they received the control or experimental membrane until after the end of the study.

All steps in the surgical procedures and all evaluation parameters were calibrated in training and calibration sessions. All patients received antibiotics (amoxicillin 500 mg or roxithromycin 150mg daily), a single dose of analgesic (ibuprofen 200mg), and mouthwash (GUM Activital, Sunstar, Osaka, Japan) after implant surgery for 7 days. Full-thickness flaps were elevated, with vertical incisions made when necessary. A surgical stent was prepared for the optimal implant position, and a sandblasted, large-grit,acid-etched (SLA) surface internal-type implant fixture was placed in accordance with the manufacturer’s recommended protocol. A sealed randomization envelope was opened to allocate augmentation of the defect, with either porcine dermis-derived non-cross-linked type I and III collagen (BioGide® Geistlich Biomaterials, Wolhusen, Switzerland) or porcine pericardium-derived type I collagen membrane (OssGuide®, Bioland, Cheongju, Korea) and xenograft bone substitutes (BioOss®, Geistlich Biomaterials, Wolhusen, Switzerland, and CollaOss®, Bioland, Cheongju, Korea). After implant placement in the ideal prosthetic position, the horizontal and vertical dehiscence defect was augmented. A collagen membrane was trimmed so that it extended 2–3 mm from the defect margin. Flaps were sutured with 6-0 absorbable sutures (Monosyn 6–0, B. Braun Aesculap, Tuttlingen, Germany), with a horizontal periosteal releasing incision used where necessary to attain primary and tension-free closure of the flap. Follow-up was performed three times during 8 weeks and additional care was arranged according to the needs of individual patients.

All horizontal and vertical defects were measured using a 15-mm UNC periodontal probe (CP 15 UNC, Hu-freidy, Chicago, IL, USA) at the time of implant installation and re-entry surgery. The same trained and calibrated examiners carried out all measurements. The following parameters were measured: (1) defect width (DW), measured as the linear distance between the widest points on the medial and distal sides of the buccal aspect; (2) defect height (DH), measured as the linear distance from the top of the implant platform to the initial bone-to-implant contact at the buccal aspect; (3) a change in the defect width (ΔDW), calculated as DW (re-entry surgery) – DW (baseline); and (4) a change in the defect height (ΔDH), calculated as DH (re-entry surgery) – DH (baseline).

The ease of manipulating and maintaining NCM and DCM in implant surgery was clinically assessed using scores on a visual analog scale (VAS) that ranged from 0 (very good) to 10 (very poor). The following parameters were measured: (1) manipulation of the membrane, degree of hydrophilicity and handling during the GBR procedure; and (2) maintenance of augmented bone substitutes, and whether covering the membrane after augmentation with bone materials improved the stability.

Cone-beam computed tomography (CBCT; Alphard Vega, Asahi Roentgen, Kyoto, Japan) was performed to assess the horizontal thickness (HT) immediately after augmentation and at the time of re-entry surgery. The following parameters were measured (Figure 1): (1) horizontal thickness of horizontal augmented bone located 1, 2, and 3 mm below the top of the implant platform (HT1, HT2, and HT3); (2) changes in horizontal thickness at each level, calculated as HT (re-entry surgery) – HT (baseline) (ΔHT1, ΔHT2, and ΔHT3); (3) density of the newly formed bone located 1 mm below the top of the implant platform, assessed using the Hounsfield unit (HU) scale with computed-tomography image-processing software (OnDemand3D®, CyberMed, Seoul, Korea) (bone density of HT1).

After a healing period of 16 weeks, when there was sufficient keratinized tissue around the top of the implant cover screw, a thin strip-shaped soft-tissue biopsy sample was obtained prior to performing re-entry surgery and the connection of the abutment. The biopsy samples were fixed in 10% neutral formalin for 10 days, and then trimmed and dehydrated in a graded series of alcohol solutions. All specimens were stained with hematoxylin-eosin and Masson’s trichrome stains. The slides were observed under a light microscope (BX50, Olympus, Tokyo, Japan).

During concurrent research that is currently still in progress, we determined that the use of different bone substitutes in the NCM and DCM groups interfered with evaluations of membrane efficiency. Therefore, after obtaining re-approval from the IRB, we used the same bone graft materials (BioOss®) in both groups, and altered the study to focus on the clinical feasibility of the membrane.

The mean±standard deviation values and 95% confidence intervals were estimated for each study group. Statistical analyses were performed with IBM SPSS Statistics (Version 21.0, IBM Corp, Armonk, NY, USA), using independent t-tests to compare the results between the NCM and DCM groups (P<0.05).

The 30 enrolled patients who fulfilled the inclusion and exclusion criteria comprised 16 males and 14 females with a mean age of 53.3 years (range, 31 to 75 years). Two patients dropped out during the follow-up: one patient in the DCM group showed signs of local infection such as gingival swelling, redness, and pus discharge at the 4-week checkup, and so the implant fixture was removed; the other patient was in the NCM group and showed early exposure of the cover screw at the 8-week checkup, and so an emergency surgical procedure involving healing around the abutment connection was performed (Figure 2). The remaining 28 patients (mean age, 53.4 years; range, 31 to 75 years) experienced no critical adverse events (Table 1).

The implants were distributed as follows: incisor, n=10 (35.7%); bicuspid, n=7 (25%); and molar, n=11 (39.3%). The 30 implants investigated in these patients comprised the following 4 models, all of which had an internal-connection design: Implantium and NR line® (Dentium), n=22 (78.6%); TS III® (Osstem), n=2 (7.1%); Bone Level® (Straumann), n=3 (10.7%); and Luna® (Shinhung), n=1 (3.6%). The diameter of the most commonly used implant was 4.8 mm (n=8, 28.6%), while the other implants had the following diameters: 3.8 mm (n=8, 28.6%), 4.3 mm (n=5, 17.9%), 3.3 mm (n=2, 7.1%), and 3.6, 4.1, 4.5, 5.0, and 6.0 mm (n=1 each, 17.9%). The implants had the following lengths: 10 mm (n=19, 67.9%), 8 mm (n=5, 17.9%), 12 mm (n=2, 7.1%), and 8.5 and 9 mm (n=1 for each, 7.1%, Table 2).

In the present study we attempted to determine the clinical feasibility of using DCM compared with NCM for treating human peri-implant dehiscence defects. We therefore conducted experiments that excluded bone graft materials from the evaluations, and assessed only the membranes themselves.

Five defect sites (two sites in the NCM group and three sites in the DCM group) showed soft-tissue dehiscence defects and membrane exposure during the early healing period. Despite the presence of plaque accumulation and signs of mild inflammation around the soft-tissue dehiscence defects, additional necrosis and secondary dehiscence defects did not occur. Spontaneous secondary closures were successfully completed, and no significant differences were observed between the two groups at the 8-week follow-up.

The DW value decreased from 3.8±1.3 to 0.4±0.9 mm in the NCM group and from 3.5±1.1 to 1.7±1.6 mm in the DCM group; the corresponding ΔDW values were 3.5±1.2 and 1.7±2.2 mm, respectively. Similarly, the DH value decreased from 5.1±2.4 to 0.2±0.6 mm in the NCM group and from 4.5±2.2 to 1.1±1.2 mm in the DCM group, with corresponding ΔDH values of 5.0±2.5 and 2.9±2.3 mm, respectively. The ΔDW and ΔDH values were statistically significant between groups (P=0.016 and P=0.031, respectively; Table 3).

The clinical assessment of the ease of manipulation and maintenance was based on the VAS scores in the two groups. The VAS score for membrane manipulation did not differ significantly between the NCM group (0.5±1.0; range, 0 to 3) and the DCM group (1.3±1.7; range, 0 to 5; P=0.119). In contrast, the VAS score for the maintenance of augmented bone substitutes did differ significantly between the two groups: 0.3±0.7 (range, 0 to 2) in the NCM group and 1.3±1.1 (range, 0 to 3) in the DCM group (P=0.012, Table 4).

The mean HT and corresponding ΔHT values were comparable in the two groups (Table 5). The HT values decreased and the corresponding ΔHT values increased by similar amounts in the two groups, and there were no statistically significant differences among any of the ΔHT values (Table 4).

The bone density of HT1 as assessed at the time of re-entry surgery did not differ significantly between the NCM group (769.5±218.2 HU; range, 421.1 to 1163.7 HU) and the DCM group (739.6±275.1 HU; range, 280.4 to 1179.4 HU; P=0.752).

There was insufficient keratinized gingiva around the implant in six patients, and so histological slides from the remaining 10 DCM and 12 NCM specimens were examined by light microscopy. Soft tissue biopsy was performed to confirm the presence of the remaining membrane fragment. Little Newly formed bone was observed around the grafted bone particles, which were dispersed above the bone bed in all groups. No membrane remnants were observed in any samples of the NCM group. In contrast, partially resorbed DCM leaflets exhibiting structural integrity were clearly identified between the gingival connective tissue and bone substitute materials in the DCM group at magnifications of 100- and 200-fold (Figure 3). Inflammatory processes were not considered to be present in either group, and so these histological results indicate that DCM exhibited sufficient biocompatibility.