Stored L-PRF membranes: safety, morphology, tensile properties, and cell viability—an in vitro pilot study Authors: Jize Yu, Ana Castro Sarda, Nicolas Peredo, Merve Kübra Aktan, Bart Van Meerbeek, Marc Quirynen Published: 2026 ABSTRACT AbstractThis in vitro study aimed to verify the qualitative changes of leukocyte- and platelet-rich fibrin (L-PRF) membranes during their storage. L-PRF membranes from 10 volunteers were stored in their exudate (obtained after compression) at 4 °C for 1, 2, and 4 weeks. The following parameters were compared to the fresh L-PRF membranes: bacterial contamination, macroscopic dimensions, diameter of fibrin fibers and fibrin network porosity (SEM), tensile properties (Young’s modulus, tensile stress and strain), and leukocyte viability (live/dead staining & 3D analyses). No bacterial contamination was observed during the storage. After 4 weeks of storage, L-PRF membranes shrank 5.1% in area, and their fibrin fibers decreased by 48.4% in diameter. The Young’s modulus of membranes and the stress/strain at maximum load remained at similar levels. The viability of leukocytes decreased from 99.0% at baseline, to 69.1 / 28.5 / 24.6% after 1 / 2 / 4 weeks respectively, starting from the surface towards the interior side. L-PRF membranes could be stored for up to 4 weeks without bacterial contamination and with largely preserved tensile properties. However, leukocyte viability declined substantially over time, which may compromise the biological activity of stored membranes and should be carefully considered in potential applications. Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-54761-0.Keywords: Autogenous grafts, Platelet concentrates, Leukocyte- and platelet-rich fibrin, Cell- and tissue-based therapy, Tissue engineeringSubject terms: Materials science, Microbiology Introduction Wound healing is a vital process for the survival of all higher organisms, consisting of spatially and temporally overlapping stages including blood clotting, inflammation, cellular proliferation and extracellular matrix (ECM) remodeling1. At the initial stage of wound healing, the formation of blood clots not only provides a provisional shield protecting injury sites from pathogens, but also serves as a reservoir of cytokines and growth factors for platelet aggregation and cell recruitment, playing an indispensable role in the repair process2. Leukocyte- and platelet-rich fibrin (L-PRF) generally refers to an autologous fibrin matrix that contains most platelets and leukocytes present in peripheral blood. Since these components are concentrated in a dense fibrin scaffold by centrifugation, L-PRF possesses a number of characteristics that are essential in wound healing, including a barrier function, an antibacterial capacity3,4, analgesic activity, and the release of growth factors that enhance tissue regeneration5,6. A growing body of evidence supports L-PRF’s advantages in achieving improved clinical outcomes in not only periodontal regenerative therapies7–12 but also the treatments of extra-oral wounds13–15. Not all dentists or periodontists are authorized or feel confident performing a venipuncture. For these practitioners, it would be advantageous if L-PRF membranes could be prepared in advance and stored for a certain period. Additionally, when multiple repetitive interventions are planned, the ability to produce a larger number of membranes from a single blood draw would be highly beneficial. This approach would also support home healthcare providers treating chronic wounds (e.g., diabetic foot), who often apply L-PRF membranes on a weekly basis14,16,17, but may lack the equipment or time to prepare them chairside, unless they carry a centrifuge and allocate more than 15 min only for each preparation. Therefore, an easy strategy to store the L-PRF membranes and to preserve their regenerative capacity is demanded. Human whole blood can be stored at 4 °C for up to 35 days with additional anticoagulants18. However, current research in blood products’ storage mainly focuses on alleviating the storage lesions of red blood cells (RBC) and platelets like hemolysis19–21 and oxidative stress21–23, whereas the “buffy coat” has not been fully assessed and is even recommended to be removed before storage to reduce the risk of transfusion reactions24–27. However, the “buffy coat” contains most leukocytes and platelets and is considered indispensable in preserving L-PRF’s anti-microbial and regenerative properties. To date, only limited studies have investigated preservation strategies for L-PRF or its precursors. Isobe et al.28 demonstrated that L-PRF can be prepared from stored whole-blood samples; however, the impact of pre-storage on the structural and biological properties of the resulting membranes remains insufficiently characterized. More recently, Chai et al.29 reported that lyophilized horizontal PRF retains bioactivity and can promote wound healing through modulation of macrophage polarization and fibroblast migration. While promising, such approaches involve additional processing steps and may not reflect a simple, clinic-friendly solution for ready-to-use L-PRF membranes. In parallel, in vitro studies have examined L-PRF degradation by incubating L-PRF membranes in enzymatic or buffer-based medium30–32 (e.g., trypsin, plasmin, and phosphate buffered saline (PBS)) to simulate physiological breakdown in human body environment, but this concept is not applicable for their storage in vitro for later use, and there remains a lack of evidence regarding simple, clinically applicable storage protocols that maintain both structural integrity and biological functionality. Therefore, the aim of this in vitro study was to evaluate a patient- and clinic-friendly storage strategy for L-PRF membranes. We hypothesized that L-PRF membranes stored under a simple way could maintain (i) microbial safety, (ii) structural and (iii) morphological characteristics, (iv) tensile properties, and (v) leukocyte viability for a period of time, thereby retaining key features relevant for their potential clinical use. Methods Volunteer selection 10 systemically healthy adult volunteers were enrolled in this study with the approval of the ethical committee of KU Leuven (No. S58789, third amendment). The exclusion criteria included: anticoagulant medication or antibiotics 3 months before the study, pregnancy or lactation, history of any active systemic infection or psychological problems. All procedures were performed according to the Helsinki Declaration and the regulations of the University Hospital. An informed consent was obtained from all subjects after having explained the purpose of the study. L-PRF preparation & storage Venous blood from each volunteer was collected in silica-coated plastic vacuum tubes without anticoagulant and immediately centrifuged at 408 g for 12 min (IntraSpin, Boca Raton, FL, USA). The L-PRF clots were retrieved from the tubes and after gentle removal of the red blood cells, they were compressed into membranes using the Xpression Box® (Intra-Lock, Boca Raton, FL, USA). All procedures preparing L-PRF membranes were under aseptic conditions using sterile instruments to minimize contamination. All L-PRF membranes obtained from a single volunteer were stored together in a dedicated sterile tube at 4 °C until analysis. At each predefined time point (7, 14, and 28 days), one membrane was retrieved for analysis under open handling conditions using sterile instruments, thereby reflecting the intended clinical simplicity. These storage time points were selected to represent clinically relevant long-term storage scenarios, for example, in case a patient needs weekly applications to treat a chronic wound. Bacterial contamination At Day 0 (= a fresh membrane, < 20 min after preparation), and after 7, 14, and 28 days of storage, 50 µL of L-PRF exudate was sampled, serially diluted, and cultured on blood agar plates supplemented with hemin (5.0 mg/L), menadione (1.0 mg/L), and 5% sterile horse blood. Plates were incubated for 5 days under either aerobic or anaerobic conditions before colony-forming units (CFU) were quantified. To assess bacteria adherent to the membranes, stored L-PRF membranes were vortexed in 1 mL PBS, and 50 µL of the resulting solution was cultured using the same protocol. All samples were analyzed in duplicate to control for potential environmental or medium-related contamination. Macroscopic analysis Standardized pictures of a separately stored L-PRF membrane were captured at Day 0 (= a fresh membrane, less than 20 min after preparation)/7/14/28 with a camera (DSC-HX400V, SONY, Japan). The surface area was calibrated with a ruler besides and the size was calculated with Software ImageJ (v2.14.0/1.54f). Scanning electron microscopy (SEM) A Nova NanoSEM 450 (FEI, Oregon, USA) was used to take representative images of the surface of the face part of the membrane (the area that was previously in contact with the red blood cells, harboring the highest concentration of platelets and white blood cells), and of the middle part of the membrane (called the body part) to visualize the fibrin network in high-resolution using a magnetic immersion lens for nano-scale spatial resolution under back-scattered electron detector with a 5 keV acceleration voltage and a 5 mm working distance. The diameter of the fibrin fibers and the porosity of the fibrin network were analyzed with Software ImageJ. Tensile tests The tensile properties of the L-PRF membranes were determined on a universal testing machine equipped with a 10 N load cell (Instron 5800, Norwood, USA). L-PRF membranes were first trimmed into a uniform “dog-bone” shape (the middle part being 4 mm in width) before being mounted between two clips at 10 mm without any tension. The membranes were then slowly stretched at a fixed rate of 1 mm/min until their rupture. Applied force [N] and tensile extension [mm] were recorded in real time using an Instron Bluehill 3 software. Stress was calculated from the applied force and the cross-sectional area of the membranes, while strain was calculated from the crosshead displacement and normalized to the initial length (10 mm). A stress-strain curve was subsequently plotted, by which the following parameters were determined: Young’s modulus, tensile stress at maximum load and tensile strain at maximum load. To calculate the Young’s modulus of L-PRF membranes accurately, only the initial linear region of the stress-strain curve was used. Live/dead staining Eight L-PRF membranes from each storage time for live/dead staining were washed with saline before their incubation with 3.5 µM SYTO 9 (Thermo Fisher Scientific, USA) and 4 µM PI. For image acquisition, a Nikon NiE upright microscope equipped with a Yokogawa CSU-X spinning-disk module with Prime BSI camera (Teledyne Photometrics) was used in combination with a 20x Plan Fluor objective. The setup was controlled by NIS-Elements (Nikon Instruments Europe B.V.), by which large-scanning images were acquired. Three regions of interest (ROIs) were selected in the face part of membranes using standardized settings for 3D structure reconstruction at high magnification and semi-quantitative analysis of leukocyte viability. For post processing/image analysis, NIS-Elements and Software ImageJ were used. Statistical analysis Data were presented as means ± standard deviations using GraphPad Prism 10.0 software. The Shapiro-Wilk test was applied to assess the normality of the data. Subsequently, one-way analysis of variance (ANOVA) and Tukey’s post hoc test were utilized to compare significant differences in the mean values of each storage time with a confidence interval of 95% (p < 0.05). Results Baselines of volunteers Ten subjects (5 females and 5 males) participated in this study (Table 1). The mean age was 50.4 ± 17.8 years (range 22–69 years). No complications during blood collection were reported. Table 1Baselines of volunteers.VolunteerBaselinesGenderAge (year)ComorbiditiesSmoking historyMedication1Female58NoNoNo2Male69NoNoNo3Female59NoNoNo4Female22NoNoNo5Male43NoNoNo6Male28NoNoNo7Male31NoNoNo8Male67NoNoAnti-cholesterol9Female65AsthmaNoNo10Female62OsteoporoseNoAnti-cholesterol Baselines of volunteers. Bacterial contamination The number of colony-forming units (CFU) following 5 days of aerobic and anaerobic incubation is summarized in Supplementary Tables 1 and 2. Among the 320 aerobic and 320 anaerobic culture plates, bacterial growth was observed in 20 and 7 plates, respectively. Most positive plates (18/27) exhibited a single CFU, while a small number showed 2–5 CFUs (n = 4) or more than 5 CFUs (n = 5). Importantly, bacterial growth was never reproduced in duplicate samples, and only a single colony morphology was identified in each case. No consistent trend in CFU counts was observed across storage time points. Macroscopic analysis The outlines of each membrane over the storage period were nearly identical (Fig. 1a). After the first 7-day storage, L-PRF membranes showed a slight decrease in area from (3.2 ± 0.5 cm2) to (3.1 ± 0.6 cm2, p < 0.05), which remained stable with an area of (3.1 ± 0.6 cm2) and (3.1 ± 0.5 cm2) after a 14- and a 28-day storage, respectively. The proportional dimensions, related to the area of day 0, are 94.9 ± 3.8% at day 7 (p < 0.01), 94.6 ± 5.2% and 95.0 ± 4.5% after a 14 and 28-day storage, respectively (Fig. 1b). Macroscopic analysis of L-PRF membranes during their storage. (a) Representative superimposed outlines of one L-PRF membrane over time. Scale bar: 2 cm. (b) Quantitative analysis of the area of the same L-PRF membrane over time. Histogram chart for comparison in mean area between storage times, line chart for percentage change compared to the initial size. (n = 10). One-way ANOVA with Tukey’s test. *p < 0.05, **p < 0.01. Scanning electron microscopy (SEM) SEM analysis revealed the presence of intact red blood cells and leukocytes up to 14 days of storage, whereas such structures were rarely observed after 28 days (Fig. 2a), suggesting progressive cellular degradation (Fig. 2a). The mean porosity maintained at a similar level during the storage (17.5 ± 8.0% for fresh membranes, and 22.8 ± 9.1%, 15.4 ± 3.1% and 20.1 ± 6.6% after a 7/14/28-day storage, Fig. 2b). The mean diameter of single fibers gradually decreased during the storage, from (135.6 ± 21.7 nm) in fresh membranes to (102.0 ± 12.4 nm) after 7 days (p < 0.01), reaching (99.0 ± 16.3 nm) and (70.0 ± 13.4 nm) after 14 (p < 0.001) and 28 days (p < 0.0001), respectively (Fig. 2c). SEM analysis of L-PRF membranes during their storage. (a) Representative SEM images of L-PRF membranes during its storage. Scale bar: 10 μm for the face and body part and 2 μm for ROIs in the body part. (b) Semi-quantitative analysis of the change in the porosity of L-PRF membranes (n = 8). (c) Semi-quantitative analysis of the change in the diameter of single fibers (n = 8). One-way ANOVA with Tukey’s test. **p < 0.01, ***p < 0.001, ****p < 0.0001. SEM scanning electron microscope, ROIs regions of interest. Tensile tests Most membranes showed a linear change in extension with an increasing loading (Fig. 3a). The Young’s modulus was 0.09 ± 0.03 MPa for fresh membranes, and 0.09 ± 0.02 MPa, 0.11 ± 0.02 MPa and 0.09 ± 0.02 MPa after a 7/14/28-day storage (Fig. 3b). The tensile stress at maximum load was also similar (0.11 ± 0.04 MPa for fresh membranes, and 0.09 ± 0.02 MPa, 0.12 ± 0.03 MPa, and 0.10 ± 0.03 MPa after a 7/14/28-day storage, respectively, Fig. 3c). For the tensile strain at maximum load, the fresh membranes showed better flexibility (1.26 ± 0.25 times of initial length) than the stored membranes (1.17 ± 0.31 times, 1.09 ± 0.29 times, 1.16 ± 0.15 times after a 7/14/28-day storage). However, no statistically significant differences were recorded between any two storage times in Young’s modulus or stress/strain at maximum load (Fig. 3d). Tensile tests on L-PRF membranes (fresh or after a 7/14/28-day storage). (a) Representative real-time loading-extension curve recorded by Instron Bluehill 3 software. Quantitative analysis of (b) Young’s modulus, (c) the tensile stress at maximum load, and (d) the tensile strain at maximum load of L-PRF membranes (n = 10). One-way ANOVA with Tukey’s test. No statistically significant differences between storage times were observed. Live/dead staining The leukocytes showed uneven distribution in the membranes due to the centrifuge procedure (Fig. 4a). Most leukocytes were located in the “face” portion of L-PRF membranes, near to the interface with the red blood cells, while rare fluorescence signal could be found in the “body” portion of the membranes. The 3D reconstructions in the “face” portion showed that most leukocytes were concentrated close to the membrane surface within a depth of approximately 48–63 μm (Fig. 4b, c). The proportion of live leukocytes significantly decreased from 99.0% in fresh membranes, to 69.1/28.5/24.6% after a 7/14/28-day storage (p < 0.05, Fig. 4d). Live/Dead staining of leukocytes in L-PRF membranes (fresh or after a 7/14/28-day storage). (a) Representative images of entire L-PRF membranes (red dashed circle represents the face portion of the membrane; white dashed circle represents the whole membrane). Scale bar: 5 mm. (b) Representative 3D reconstructions of ROIs. Scale bar: 100 μm. Semi-quantitative analysis of the average values of 3 ROIs per sample in: (c) the leukocyte layer thickness, (d) the leukocyte survival rate, and (e) the differences in leukocyte survival rate between surface part and interior part (n = 8). One-way ANOVA with Tukey’s test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. ROIs regions of interest. Furthermore, spatial analysis revealed reduced viability in the outermost 30% of the leukocyte layer compared to the inner regions, with significant differences observed at 7 days (p < 0.01) and 28 days (p < 0.05; Fig. 4e). (Check Supplementary video 3 for 3D reconstruction animation). Discussion In this study, L-PRF membranes were stored together in one tube and all handling afterwards was performed in an open environment to simulate a condition where a clinician would like to work with membranes that are prepared in advance. L-PRF exudate provides a native biochemical environment that closely mimics in vivo conditions, which eliminates the risk of immune reactions and xenogeneic contamination. Therefore, only the exudate obtained during L-PRF preparation was selected as the storage medium. The absence of a consistent increase of bacterial contamination over time, together with the non-reproducible nature of positive cultures, suggests that the observed minimal bacterial growth most likely resulted from the preparation of the blood agar plates and/or from the environment during plating. As such the proposed storage procedure seems to be safe. However, within the limitations of methodology, more sensitive detection methods (e.g., PCR-based techniques) might be required to strengthen the conclusion. The degradation of L-PRF membranes during storage could partially reduce their benefits through the release of cellular and biological components before the final application. In this study, however, L-PRF membranes showed morphological stability throughout their 4-week storage, with only a 5% shrinkage in size after the first week, which was probably caused by the consistent clot contraction activated by platelets after L-PRF preparation33. The long-term stability of fibrin was also observed in tensile tests, in which, during the 4 weeks of storage, no statistically significant differences were observed in Young’s modulus or the tensile stress/strain at maximum load. In this study, notable intragroup variability was observed in the tensile test results, indicating that a larger sample size would be required to draw more robust conclusions. Furthermore, the dimensional and mechanical stability during storage may vary depending on preparation protocols (e.g., L-PRF, A-PRF, A-PRF+)34–36 and centrifugation systems37–39. Therefore, future studies with larger cohorts are warranted to validate and extend these findings. The mechanical behavior of L-PRF membranes involves an interplay between the mechanical characteristics of fibrin itself (internal structure of fibrin fibers) and the structure of the networks40,41. Under a scanning electron microscope, the diameter of a single fiber in L-PRF membranes significantly reduced only after a 28-day storage, which might explain the well-maintained physical properties in tensile tests. L-PRF membranes possess a few advantages essential in anti-fibrinolysis, partially due to the concentrated platelets that mediate clot contraction, expelling more unbound lytic proteins (e.g., tPA, plasminogen)42–44 and partially due to the dense fibrin network with small pores that decrease the penetration of lytic enzymes45 and the activation/access to plasmin cleavage sites46. Furthermore, fibrin promotes the activation of FXIIIa, which further cross-links fibrinolysis inhibitors such as thrombin-activatable fibrinolysis inhibitor (TAFI), α2-antiplasmin, and plasminogen activator inhibitor (PAI)−247,48, protecting L-PRF membranes from internal fibrinolysis. Hence, storing L-PRF in the form of membranes instead of clots should be given priority to preserve their fibrin structure against degradation. Due to the difference in density, the cell components in peripheral blood can be separated via centrifugation and trapped inside the L-PRF clot after fibrin crosslinking. Miron et al.49 and Andrade et al.38 observed, histologically, a massive accumulation of leukocytes immediately under the red blood cell layer, while only few cells in the remaining part of the membrane. In this study, the full membrane scans with a confocal microscope confirmed this uneven distribution of leukocytes in L-PRF membranes. Despite the structural stability exhibited, a pronounced and progressive decline in leukocyte viability was observed, indicating that biological functionality is substantially compromised over time. Given the established role of leukocytes in antimicrobial defense, cytokine release, and immunomodulation, their loss of viability may reduce the regenerative potential of stored L-PRF membranes50–52. Regarding the pattern of leukocyte apoptosis during storage across the thickness of the membrane, the leukocytes close to the surface were found to be more susceptible to physical damage. At the surface the leukocytes are of course directly exposed to the byproducts of hemolysis and the pH change in the storage medium. The hemoglobin and iron from ruptured RBCs can promote reactive oxygen species, damaging cell structure and potentially impairing leukocyte viability and function53,54. Additionally, the elevated extracellular potassium caused by hemolysis can disrupt the ionic balance and signaling cascades of leukocytes, affecting cell activation and inflammatory response53,55. Therefore, a complete removal of the RBCs from the L-PRF membranes might be considered before storage to prolong the lifespan of L-PRF membranes. Moreover, PRF prepared from different g-force56 and rotation angulation57 vary in leukocyte dispersion within the fibrin scaffold, potentially altering the patterns of viability reduction of leukocytes. Several limitations of this study should be acknowledged. Firstly, this is a pilot study, with some variables requiring larger sample size. Therefore, variables with smaller effects (e.g., size change, strain at maximum load) should be interpreted cautiously. Secondly, the in vitro design does not fully replicate the complex in vivo environment, particularly with respect to enzymatic activity and immune interactions. Thirdly, even though all experiments were done by the same researcher, blinding was only presented for the SEM analysis and confocal microscope tests. Finally, functional assays assessing growth factor release or antimicrobial activity were not included in this study, limiting conclusions regarding biological performance. Conclusion Within the limitations of this in vitro study, L-PRF membranes demonstrated relative structural and mechanical stability during storage at 4 °C for up to four weeks, with no consistent evidence of bacterial contamination during storage. However, a pronounced and progressive decline in leukocyte viability was observed, indicating that biological functionality may be substantially compromised over time. As no direct functional assays were performed, the extent to which stored L-PRF membranes retain their biological efficacy remains unclear. Future studies incorporating functional assessments and larger sample sizes are required to determine optimal storage conditions and clinically relevant time frames. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 2 Supplementary Material 3 Author contributions J.Y. performed tests and data analysis for Table 1; Figs. 1, 2, 3 and 4; wrote the main manuscript text. A.C. provided supports in blood harvesting and sample preparation. N.P. provided technical support and performed data analysis for Fig. 4. M.K.A. performed scanning electron microscope tests. B.V.M. provided technical training and suggestions on tensile tests. M.Q. provided supports in blood harvesting and sample preparation and supports in manuscript writing. All authors reviewed the manuscript. Funding Jize Yu’s stay and research conducted as a PhD student in the Department of Oral Health Sciences of KU Leuven was funded by the China Scholarship Council (File No. 202206170027). Data availability All data supporting the findings of this study are available within the paper and its Supplementary materials. Declarations 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. References Associated Data Supplementary Materials Supplementary Material 1 Supplementary Material 2 Supplementary Material 3 Data Availability Statement All data supporting the findings of this study are available within the paper and its Supplementary materials.