Comparative mechanical analysis of collagen membranes derived from different tissues in an in vitro model simulating membrane-stabilized horizontal GBR Authors: Xueying Bai, Yufei Xue, Jiao Li, Yi Liu, Zhou Ye, Huiyi Yan, Yi Zhou Published: 2026 ABSTRACT AbstractBackgroundThis study evaluates and compares the mechanical behavior of two collagen membranes in an in vitro model simulating membrane-stabilized horizontal guided bone regeneration (GBR).Materials and methodsMegreen® (bovine pericardium, BP) and Bio-Gide® (porcine dermis, PD) were tested under wet conditions. Two models were developed: (1) a balloon dilation test to assess the membrane’s capability to accommodate bone graft material and (2) a bone graft compression model to evaluate tear resistance during graft compaction.ResultsIn the tensile testing, BP demonstrated a significantly superior modulus of elasticity (146.71 ± 45.08 vs. 6.08 ± 2.66 MPa), maximum load tolerance (30.71 ± 8.42 vs. 4.49 ± 1.36 N), and tensile stress at maximum load (14.62 ± 4.01 vs. 1.61 ± 0.49 MPa; p < 0.001). Conversely, PD achieved high tensile extension at maximum load (6.32 ± 1.93 vs. 1.64 ± 0.32 mm) and percent tensile strain at maximum load (52.69 ± 16.03 vs. 13.63 ± 2.65%; p < 0.001). In the balloon dilation test, PD presented a slightly greater expansion diameter (4.43 ± 0.17 vs. 4.34 ± 0.05 mm; p = 0.047). In the compression model, BP significantly outperformed PD in maximum load (40.86 ± 4.54 vs. 7.90 (7.72–10.92) N) and displacement at maximum load (5.61 ± 0.60 vs. 3.65 ± 0.82 mm; p < 0.001).ConclusionsBP exhibited higher tensile strength and compression resistance, whereas PD showed greater deformability. These findings indicate distinct mechanical differences between the two collagen membranes under the tested conditions and may provide a mechanical basis for membrane selection in membrane-stabilized horizontal GBR. Background Guided bone regeneration (GBR) has been successfully applied to horizontal alveolar bone reconstruction with implant success rates comparable to those achieved in native bone, while being less invasive than block bone grafting [1–3]. Among the available approaches for horizontal bone augmentation, the Sausage Technique, a specific membrane-stabilized horizontal GBR method proposed by Dr. Istvan Urban, has been successfully used in clinical practice for several years and has demonstrated favorable bone gain with a relatively low complication rate in previous studies [4–8]. Natural collagen membranes are commonly used in this technique because of their biocompatibility, low incidence of soft tissue complications, and favorable handling characteristics [4, 5]. Compared with cross-linked collagen membranes and synthetic membranes, they are generally more deformable and better able to adapt to surface contours [9–11]. However, this deformability may also limit their ability to maintain space, which remains a major challenge in GBR [2, 5, 12, 13]. While greater stiffness and strength enhance space-maintaining capacity, these properties are generally achieved through higher crosslink density, which can hinder cellular infiltration and vascularization, thereby delaying tissue integration and prolonging the foreign body response [14–16]. Membranes with lower stiffness, by contrast, favor biological integration but are prone to premature barrier failure [15, 16]. Balancing these competing demands highlights the clinical relevance of systematically characterizing the mechanical behavior of natural collagen membranes under procedure-specific loading conditions. In membrane-stabilized horizontal GBR, collagen membranes are fixed with titanium pins to stabilize the grafted area and maintain the regenerative space more effectively than non-fixed membranes [6]. However, pin fixation may also create localized stress concentration, increasing the risk of membrane tearing. Accordingly, the membrane must combine sufficient extensibility to accommodate the graft material with adequate tear resistance during stretching and graft compaction [11, 17]. Because these stresses are localized and procedure-dependent, conventional mechanical tests alone may not fully reflect membrane behavior under clinically relevant conditions [5, 12, 18–21]. Although natural collagen membranes play a central role in the membrane-stabilized horizontal GBR, their mechanical behavior under these clinically relevant loading conditions remains insufficiently characterized [5, 18]. In particular, comparative data on their ability to accommodate graft volume and resist tearing during pin fixation and graft compression are limited. Accordingly, the present study employed in vitro models simulating key steps of membrane-stabilized horizontal GBR to compare two commercially available collagen membranes derived from different tissues. The study aimed to determine whether the two membranes exhibit different mechanical behaviors under simulated clinical loading conditions, particularly in terms of tensile properties, space-accommodating capacity, and tear resistance. Methods Two commercially available natural collagen membranes, clinically used for GBR, were selected for this study. Megreen® (Shaanxi Reshine Biotech Company, Xi’an, China) is a bovine pericardium-derived collagen membrane that retains the natural collagen fiber structure of the bovine pericardium, with type I collagen as the main component. Bio-Gide® (Geistlich Biomaterials, Baden-Baden, Germany) is a porcine dermis-derived collagen membrane composed primarily of type I and type III collagen. BP and PD represent Megreen® and Bio-Gide®, respectively. No human or animal subjects were involved in this study; therefore, ethics approval was not required. To simulate practical clinical applications, the collagen membranes were wetted in all test groups. The collagen membranes were soaked in isotonic sodium chloride solution for durations determined according to the manufacturer’s instructions (BP for 5 min and PD for 30 min). For thickness measurements, digital calipers (Shenzhen Yuanhengtong Technology Co., Ltd., Shenzhen, China; resolution: 0.001 mm; error range: ±0.005 mm) were used. Ten layers of collagen membranes were superimposed for thickness measurement, and the average thickness per membrane was calculated and used as the standardized input value. All specimens were randomly assigned to the experimental groups, and the testing order was randomized. The definitions of failure endpoints for the three mechanical test models are summarized in Table 1. Membrane orientation was standardized at the macroscopic level in all test models, whereas the intrinsic microscopic fiber orientation of individual membrane specimens could not be precisely controlled. Sample size calculation was performed based on a pilot study using G*Power 3.1, with an alpha level of 0.05 and a statistical power of 90%, to determine the appropriate number of specimens for each testing procedure. Table 1Definition of failure endpoints across three mechanical test models.ModelDimensionDefinitionUniaxial tensile test Failure event Membrane rupture, identified by a sudden and irreversible drop in the load-displacement curve. Detection method Automatically detected by the instrument's built-in load-cell sensor system, which continuously monitored tensile force throughout the test. Outcome recorded The following parameters were recorded: elastic modulus (MPa), maximum load (N), maximum tensile stress (MPa), maximum extension (mm), and maximum tensile strain (%).Balloon dilation model Failure event First visible macroscopic tearing or perforation of the membrane at the titanium pin fixation site (pin-membrane interface). Rationale for site selection The fixation site represents a stress concentration zone: under progressive loading, direct contact between the hydrated collagen membrane and the rigid titanium pin promotes localized deformation and eventual structural failure at this interface. Test termination Testing terminated immediately upon visual confirmation of membrane damage at the fixation site.Bone graft compression model Failure event First visible macroscopic tearing or perforation of the membrane at the titanium pin fixation site (identical definition to balloon dilation model). Rationale for site selection Under compressive loading, the pin-membrane contact zone acts as a stress concentrator; the hydrated membrane undergoes localized tearing at this interface, consistent with the mechanism described above. Outcome recorded Peak load (N) and displacement (mm) recorded at the exact moment failure is confirmed. Definition of failure endpoints across three mechanical test models. Uniaxial tensile test The membranes were cut into standard rectangular dimensions of 7 × 20 mm using a custom-made guillotine (Hengxin Instrument Equipment Co., Ltd., Hebei, China). Equal numbers of samples from each group (n = 15) under wet conditions were securely fastened to a material-testing machine (Instron ElectroPuls E1000, Buckinghamshire, England). Subsequently, the membranes were subjected to a longitudinal tensile force at a consistent rate of 1 mm/min until they reached their tear points (Fig. 1). The distance between clamps was maintained at 12 mm. Load–displacement data were automatically and continuously recorded throughout the tensile test. The mechanical behavior of the two membrane types was characterized using five parameters: elastic modulus, maximum load, maximum tensile stress, maximum extension, and maximum tensile strain. a-c. Tensile test procedure Balloon dilation model Equal membrane samples (n = 10) from each group under wet conditions were cut into standard rectangular dimensions of 7 × 10 mm using a custom-made guillotine. Four titanium pins (Trausim, China) were used to secure the collagen membrane onto the horizontal surface of a Class I bone block model, positioned 1.5 mm away from the membrane edge. The PTCA balloon dilation catheter is specifically designed to withstand high pressure and is commonly used in cardiovascular surgery for balloon angioplasty in narrowed sections of the coronary arteries or bypass grafts, enhancing myocardial perfusion [22]. A balloon with a diameter of 4.5 mm and a length of 15 mm (CONQUEROR™ SC PTCA Balloon Catheter, APT Medical, China) was inserted beneath the membrane to simulate its capacity to accommodate the underlying bone graft material (Fig. 2). The opposing end of the balloon was connected to a pressurization device that displayed the pressure value. To reduce potential confounding, each balloon was used for a single application, and the water pressure was increased at a constant rate of 0.2 atm/s across all groups. Pretest results consistently showed that membrane tearing occurred at the pin fixation site. Therefore, in this experimental model, failure was operationally defined as the onset of membrane tearing at the pin fixation site, as detailed in Table 1. The maximum compression load and displacement at the onset of membrane tearing were recorded automatically and continuously to compare the space-accommodating capacity and tear resistance of the two membrane types along the pin axis. a, b. Balloon dilation test procedure in the simulated membrane-stabilized horizontal GBR in vitro Bone graft compression model Equal membrane samples (n = 10) from each group under wet conditions were cut into standard square dimensions of 7 × 20 mm using a custom-made guillotine. Custom-cut polyamide (PA) resin blocks measuring 15 mm in length, 15 mm in width, and 10 mm in height were prepared. A central space of 7 × 7 × 10 mm was reserved within each block to simulate and construct a bone defect model in vitro. To simulate the membrane-stabilized horizontal GBR in vitro, eight titanium pins were employed to secure the membrane to the block. A custom-made tool was used to guide pin fixation, ensuring that each pin was positioned at an equal distance of 1.5 mm from the edges of the block. In addition, the tool ensured that the pins were inserted perpendicular to the surface of the block. All pins were fixed by the same experimenter across all groups. The reserved space within the block was filled with 0.2 g of bone graft particles in a wet state. The graft material used in this study was THE Graft™ (Purgo Biologics Inc., Seongnam-si, Gyeonggi-do, Korea), a porcine-derived xenograft composed of a porous anorganic cancellous bone mineral matrix (mainly hydroxyapatite) with a particle size of 0.25–1.0 mm. Utilizing a material-testing machine, pressure was applied to a rigid round rod (R: 3.4 mm, H: 25 mm) at a constant rate of 1 mm/min, mimicking the compression process of bone graft particles in the membrane-stabilized horizontal GBR. Pretest results showed that membrane tearing consistently occurred at the pin fixation site. Accordingly, failure in this experimental model was defined as tearing at the pin fixation site, as detailed in Table 1. The maximum compression load and displacement at the onset of membrane tearing were recorded automatically and continuously. These two parameters were used to compare the tear resistance of the two membrane types under different axial loading directions relative to the pins in this simulated horizontal GBR compression model (Fig. 3). Bone graft compression test procedure in the simulated membrane-stabilized horizontal GBR in vitro Statistical analysis Statistical analyses were performed using SPSS version 20.0. Data distribution was assessed using the Shapiro–Wilk test. Normally distributed variables were analyzed using the independent-samples t-test and are presented as mean ± standard deviation (SD), while non-normally distributed variables were analyzed using the Mann–Whitney U test and are presented as median (range). In addition to P values, between-group mean differences with 95% confidence intervals and Hedges’ g were calculated for normally distributed variables, whereas rank-biserial correlation was reported as the effect size for non-normally distributed variables. Statistical significance was set at p < 0.05. Results Uniaxial tensile test The mean thickness was 0.307 mm for the PD group and 0.404 mm for the BP group. Table 2 shows the descriptive statistics of the uniaxial tensile test results. Detailed between-group effect sizes and 95% confidence intervals are presented in Table 2. Figure 4a-e depict the measured tensile mechanical parameters for both membrane groups during the tests. Table 2Descriptive data of in vitro experimentsPropertiesMembraneSample size (n)Descriptive dataP-valueEffect size95% CIElastic modulus [MPa]BP15146.71 ± 45.08P < 0.0014.29[115.64, 165.62]PD156.08 ± 2.66Max. load [N]BP1530.71 ± 8.42P < 0.0014.23[21.51, 30.91]PD154.49 ± 1.36Max. tensile stress [MPa]BP1514.62 ± 4.01P < 0.0014.44[10.79, 15.25]PD151.60 ± 0.49Max. extension (mm)BP151.64 ± 0.32P < 0.0013.31[3.61, 5.76]PD156.32 ± 1.92Max. tensile strain [%]BP1513.63 ± 2.65P < 0.0013.31[30.10, 48.01]PD1552.69 ± 16.03Max. diameter (mm)BP104.34 ± 0.05P = 0.0470.95[0.00, 0.18]PD104.43 ± 0.12Max. compression force [N]BP1040.53 (33.90–51.75)P < 0.001r = 1.00—PD107.90 (6.94–11.94) *Max. compression displacement (mm)BP105.61 ± 0.60P < 0.0012.62[1.28, 2.64]PD103.65 ± 0.82Data are presented as mean ± SD for normally distributed variables and as median (range) for non-normally distributed variables. Hedges' g is reported for normally distributed outcomes. For the maximum-load outcome in the bone graft compression model, rank-biserial correlation is reported as the effect size, and no 95% confidence interval is shown Descriptive data of in vitro experiments Data are presented as mean ± SD for normally distributed variables and as median (range) for non-normally distributed variables. Hedges' g is reported for normally distributed outcomes. For the maximum-load outcome in the bone graft compression model, rank-biserial correlation is reported as the effect size, and no 95% confidence interval is shown Results from tensile tests (a-e) and balloon dilation tests (f) The BP group exhibited significantly higher values for elastic modulus (146.71 ± 45.08 MPa vs. 6.08 ± 2.66 MPa), maximum tensile load (30.71 ± 8.42 N vs. 4.49 ± 1.36 N), and maximum tensile stress (14.62 ± 4.01 MPa vs. 1.61 ± 0.49 MPa) compared with those of the PD group, with all differences being statistically significant (p < 0.001). Conversely, the maximum tensile displacement (6.32 ± 1.93 mm vs. 1.64 ± 0.32 mm) and maximum tensile strain (52.69 ± 16.03% vs. 13.63 ± 2.65%) were significantly higher in the PD group than in the BP group, both with statistically significant differences (p < 0.001). Figure 5 shows the engineering stress-strain curves obtained from the tensile tests. In both groups, stress increased with strain until a peak was reached, followed by a decrease. After the peak, the BP group showed a more pronounced decrease in the curve, whereas the PD group showed greater deformation before failure. Characteristic tensile stress-strain curve Balloon dilation test Figure 4f presents the balloon dilation test results for both membrane groups. The mean maximum diameter of the BP group (4.34 ± 0.05 mm) was slightly lower than that of the PD group (4.43 ± 0.17 mm), with a statistically significant difference (p = 0.047). Collagen membrane failure was consistently observed only at the sites of membrane fixation in all groups. Bone graft compression Collagen membrane failure consistently occurred at the pin fixation site across all groups. As shown in Fig. 6, the BP group exhibited significantly higher maximum load values than the PD group (40.86 ± 4.54 N vs. 7.90 [7.72–10.92] N; p < 0.001). Likewise, the displacement at maximum load was significantly greater in the BP group than in the PD group (5.61 ± 0.60 mm vs. 3.65 ± 0.82 mm; p < 0.001). Results from bone graft compression tests Figure 7 presents the load–displacement curves obtained from the bone graft compression test. Overlap between the curves of the two membrane groups was observed; however, the BP group exhibited higher maximum load and greater displacement at failure than the PD group. Characteristic load-displacement curve Discussion Membrane-stabilized horizontal GBR imposes three specific mechanical requirements on natural collagen membranes. First, the membrane should allow easy manipulation with forceps during placement of the bone graft particles and pin fixation, while minimizing the risk of perforation [5, 7]. Second, it should provide sufficient mechanical support after pin fixation to maintain space stability and contain the bone graft material, while also exhibiting adequate deformability and extensibility to accommodate the graft [4, 5, 23]. Third, during graft compaction, it must have sufficient toughness to withstand shear forces from rigid titanium pins and resist damage from coarse bone graft particles. Therefore, a comprehensive understanding of the mechanical behavior of natural collagen membranes is essential for optimizing this procedure. Based on this rationale, the present study developed a series of in vitro models to evaluate membrane behavior under simulated clinically relevant conditions. Operability comparisons In the uniaxial tensile test, the elastic modulus of BP was approximately 25 times that of PD, and the maximum tensile stress under wet conditions was also significantly higher in BP, indicating greater resistance to deformation and superior tensile strength. In contrast, the tensile strain of PD was approximately 4 times that of BP, suggesting better intrinsic extensibility. In the present study, the maximum tensile strength, maximum load, and maximum extension of wet PD were 1.61 ± 0.49 MPa, 4.49 ± 1.36 N, and 6.32 ± 1.93 mm, respectively, which were generally consistent with previous reports. Zhang, et al. [24] reported a tensile strength of 1.16 MPa for wet PD. Similarly, Yamanaka, et al. [21] reported that the maximum elongation of Techgraft® (a collagen membrane derived from bovine pericardium under wet conditions) was 44.03 ± 7.13%. Another study by Raz, et al. [12] reported a tensile strength of 1.68 ± 0.54 MPa for wet PD, with a maximal tensile load of 3.7 ± 1.43 N and maximal tensile extension of 2.3 ± 0.25 mm. It should be noted that the uniaxial tensile test provides only a simplified evaluation of basic mechanical properties and cannot fully reproduce the complex mechanical stresses to which collagen membranes are subjected in clinical settings [5, 25]. Bone graft accommodation test Regarding bone graft accommodation, the balloon dilation test showed that, under the conditions of the present model, PD had a significantly greater space-accommodating capacity than BP, although the magnitude of the difference was small. As the balloon diameter increased, “failure” in both membranes initiated at the titanium pin fixation sites; however, the failure patterns differed. BP showed abrupt rupture, whereas PD failed more gradually, often undergoing a slow transition from a transparent state to final failure. It remains unclear whether, under clinical conditions, overstretching of PD to this pre-failure transparent state may compromise its barrier function. The stress–strain curves showed that PD had no clearly defined transition from elastic to plastic deformation and underwent marked plastic deformation before fracture, resulting in greater elongation. By contrast, BP experienced a noticeable stress reduction after reaching maximum stress during plastic deformation, which could be defined as failure. This difference may be attributed to the natural structure of bovine pericardium, which enhances its ability to resist tensile forces from the pins and effectively counter crack propagation. At present, direct comparative data on the tensile properties, space-accommodating capacity, and tear resistance of these two specific collagen membranes under titanium pin loading are lacking. To the best of our knowledge, the present study is the first to compare these two collagen membranes in a simulated membrane-stabilized horizontal GBR model, with particular emphasis on space-accommodating capacity and tear resistance. A comparable study by Bozkurt, et al. [26] exclusively reported the measured suture retention strength of PD. Although suture retention testing assesses unidirectional tear resistance, tear testing performed along the same loading axis in a simulated membrane-stabilized horizontal GBR model may allow a more comprehensive comparison between materials. This approach may also provide valuable insight into the energy or force required for crack propagation through the membrane [25]. Tear resistance Regarding tear resistance, since failure in both membranes during the balloon dilation test occurred predominantly by tearing at the titanium pin fixation sites, a bone graft compression test was further established to better simulate the clinical graft compaction process and to evaluate the membranes’ resistance to multidirectional tearing. The results showed that BP had a maximum load-bearing capacity approximately 5 times greater than that of PD and a displacement at failure approximately 1.5 times higher. The load–displacement curves further confirmed the superior toughness of BP. The distinct stop-feedback observed during bone graft filling and compaction with BP was consistent with the failure pattern identified in the balloon dilation test, suggesting that BP is able to withstand greater forces during graft compaction. In contrast, PD exhibited a less distinct transition from plastic deformation to failure, indicating that more careful handling may be required clinically to minimize the risk of membrane perforation or damage [21]. Previous studies have shown that bovine pericardium-derived collagen membranes retain their native collagen architecture, characterized by dense, wavy, and multidirectionally oriented fibers, which contributes to their excellent multidirectional tear resistance [27, 28]. In contrast, collagen fibers in dermal tissues form a relatively loose network [29]. Consistent with these structural differences, BP demonstrated greater toughness and resistance to multidirectional tearing. In the present model, BP also resisted multidirectional tearing induced by titanium pin fixation, supporting its mechanical suitability for clinical application [30]. Within the mechanical limits of collagen membranes, precise control of compression force is essential to achieve an optimal compaction threshold that promotes vascular ingrowth, blood perfusion, and graft revascularization [31]. Further studies are needed to refine material-specific compression strategies for the treatment of bone defects [31]. Histological characteristics The mechanical differences between BP and PD can be attributed to differences in their underlying tissue microarchitecture. In pericardial tissue, the fibrous layer accounts for approximately 74.8% of the total thickness and constitutes the primary load-bearing component. Collagen fibers interspersed with elastic fibers form a heterogeneous and anisotropic architecture, which provides resistance to tearing in multiple directions and explains the mechanical behavior observed in BP [12, 32]. PD, by contrast, is a natural collagen membrane derived from porcine dermal type I and III collagen. Owing to its non-crosslinked collagen matrix, PD exhibits greater deformability than BP [24]. Scanning electron microscopy shows a relatively dense collagen organization in PD, characterized by interwoven fibers forming larger chain-like structures and increased surface area density [12, 33]. During tear resistance testing, applied shear forces are partially accommodated by material flow, which limits crack propagation [20, 32, 33]. Unlike tensile loading, fiber deformation during tearing is accompanied by internal fiber rearrangement, whereby applied energy is dissipated through changes in fiber orientation rather than direct structural failure. This mechanism accounts for the greater extensibility observed in PD compared with BP [32, 33]. Consistent with the intrinsic properties of porcine dermal collagen fibers, PD therefore exhibits lower tensile strength and tear resistance [34]. Clinical application characteristics This study showed that, compared with PD, BP had higher toughness, greater resistance to handling-related damage, and better overall handling stability, whereas PD exhibited greater ductility, suggesting a potential advantage in accommodating the bone graft. During the graft compaction, BP withstood significantly higher loads, indicating potential relevance in situations involving larger defects that require greater compaction forces. The greater deformability of PD could allow better adaptation to irregular bone surfaces; however, its lower tensile strength necessitates careful clinical handling to avoid membrane damage. The inherently low toughness of natural collagen membranes may limit their widespread clinical use, whereas improved mechanical strength may enhance clinical effectiveness [29, 35]. However, collagen membranes derived from natural materials often exhibit a trade-off between tensile strength and deformability; achieving an optimal balance between these properties may improve clinical handling and broaden their indications [12]. Understanding the mechanical behavior of collagen membranes may assist clinicians to select appropriate membrane types and sizes based on graft-site characteristics, thereby optimizing surgical procedures and minimizing material waste. Nevertheless, membrane damage caused by localized stress concentrations or structural weaknesses during or after placement cannot be completely excluded and should be carefully considered to avoid compromised bone augmentation outcomes [18]. It should be emphasized that these findings were obtained under in vitro conditions. Although they may provide useful insight into membrane handling characteristics and graft compaction behavior, direct extrapolation to the clinical setting should be made with caution. In vivo membrane performance is likely to be influenced by multiple additional factors, including defect morphology, tissue support, hydration status, surgical manipulation, and healing-related biological conditions. Limitations The present study has several limitations. With regard to the experimental model, only single-walled defects and Class I bone conditions were included. In clinical practice, however, more complex defect morphologies, larger grafting volumes, and different bone qualities are common, and these factors may impose different mechanical demands on collagen membranes as well as affect pin stability, vascularization, and regenerative outcomes. Accordingly, the generalizability of the present findings is limited. At the statistical level, no adjustment for multiple comparisons was performed, which may have increased the risk of type I error. In addition, as an in vitro study, this work could not reproduce the dynamic biological processes involved in clinical healing, such as blood supply, cellular activity, and tissue remodeling, all of which may substantially influence the in vivo performance of collagen membranes. Different hydration times may have affected the mechanical behavior of the collagen membranes and should be standardized or systematically controlled in future studies to enable a more direct comparison of intrinsic material-related mechanical differences. Additionally, no local mechanical or microstructural assessment was performed at the fixation sites. Despite these limitations, the present findings provide supplementary mechanical and functional data on existing collagen membranes. They may serve as a reference for predicting the performance of membrane-stabilized horizontal GBR in specific clinical scenarios and may also support the rational design and use of collagen membranes, rather than advocating the preferential selection of any particular product. Future studies incorporating digital image correlation or microstructural evaluation at the fixation sites may help further clarify the contribution of local mechanical and structural factors to the failure mechanism. Conclusions This study provides mechanical property data for two collagen membranes under simulated in vitro mechanical conditions, showing that BP had higher tensile strength, whereas PD exhibited greater extensibility. Based on these experimental findings, it may be hypothesized that the higher stiffness of BP could potentially limit bone augmentation under certain clinical conditions, while the greater deformability of PD may render it more susceptible to deformation when subjected to the stresses generated by titanium pin fixation in the membrane-stabilized horizontal GBR. Taken together, these findings suggest that the clinical selection of collagen membranes may involve a trade-off between mechanical strength and deformability. However, the specific clinical relevance of these observations remains to be established in future clinical studies. Authors' contributions X.Y. Bai: Conceptualization, Methodology, Software, Data curation, Investigation, Validation, Formal analysis, Writing original draft, Review & editing. Y.F. Xue: Methodology, Data curation, Investigation, Validation. J. Li: Methodology, Data curation, Investigation, Validation. Y. Liu: Funding acquisition, Writing original draft, Visualization, Supervision. Z. Ye: Writing original draft, Methodology, Supervision, Project administration. H.Y. Yan: Conceptualization, Methodology, Writing original draft, Supervision, Visualization, Project administration.Y. Zhou: Conceptualization, Methodology, Writing original draft, Supervision, Project administration. Funding The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the research grant 2022CFD005 from the Innovation and Development Fund of Hubei Province Natural Science Foundation. Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Huiyi Yan, Email: huiyiyan@whu.edu.cn. Yi Zhou, Email: dryizhou@whu.edu.cn. References Associated Data Data Availability Statement No datasets were generated or analysed during the current study.