Scientific Abstract Presentations: Abstract Topics Session (Not for CME) Thursday, October 9 Thu, Oct 9 1:30-3:00 p.m. CDT

PURPOSE: Autologous fat grafting is widely used for soft tissue augmentation for both reconstructive and aesthetic applications. However, unpredictable volume retention and requisite donor site morbidity remain a concern. Decellularized adipose extracellular matrix (ECM) sourced from allograft donors has emerged as a promising alternative. Ideally, antigenic cellular components are removed while bioactive and biomechanical properties of the matrix are retained to recruit host cells. While conventional decellularization methods employ abrasive chemicals agents that can denature proteins and leave toxic residues, we have engineered an injectable, human-derived adipose ECM that is decellularized via supercritical carbon dioxide (scCO2). This emerging technology promises a gentler, eco-friendly isolation of matrix components, promising an enhanced post-implantation tissue regeneration.

METHODS: Adipose tissue removed from abdominoplasty procedures was mechanically homogenized, delipidated with isopropanol at 37°C, and underwent a non-wetting scCO2 cleansing step. The tissue then underwent two cycles of scCO2 decellularization, cryomilling, and scCO2 terminal sterilization. Processed adipose ECM underwent DNA quantification to verify removal of cellular debris. Adipose ECM was subcutaneously injected into dorsum and cranial subcutaneous pockets of immunocompetent C57BL/6 mice through a 20-gauge catheter (0.4mL of 0.5g sample/mL PBS). Commercially available (cross linked) hyaluronic acid (HA) was injected (0.4mL, dorsal flank) as a control. Samples were explanted at 1-, 3- and 6-months for volumetric and immunohistochemical analysis.

RESULTS: Delipidated, decellularized adipose ECM demonstrated a mean DNA content of 47ng double-stranded DNA/mg of sample, below the accepted threshold of 50ng/mg for "complete" decellularization. After six months in vivo, mean adipose flank volume was 307mm3 (77% of initial volume), adipose cranial was 273mm3 (68%), and HA volume was 1393mm3 (348%), due to its hygroscopic nature. Cranial-site adipose grafts appeared more vascularized than flank-site, consistent with greater regional blood flow. No cases of graft extrusion, infection, or seroma/hematoma formation were observed. DAPI quantification revealed significantly greater cellular density in adipose samples over HA samples at one and three months (p<0.0001); cellular density declined over time in all groups as inflammation subsided. M1/M2 macrophage expression ratio at three months was 0.28 and 0.36 in adipose flank and cranial grafts, respectively, indicating a predominance of the pro-regenerative M2 phenotype over pro-inflammatory M1. Perilipin expression revealed the appearance of mature adipocytes/adipose tissue developing within injected adipose ECM after three months; few cells and no adipocytes were noted in HA explants. Similarly, CD31 expression demonstrated evidence of neo-vascularization after three months in adipose ECM samples only. Picrosirius red staining under polarized light demonstrated deposition of new fibrillar collagen networks in 1-month ECM (disorganized fibers, yellow/green birefringence), which became more aligned with maturation by three months (red birefringence). No new collagen deposition was seen in HA explants. Six-month histological/biomechanical analysis is ongoing.

CONCLUSION: We present an innovative scCO2-based decellularization process to fabricate an adipose ECM scaffold which demonstrated 73% volume retention after six months in vivo, with robust architectural remodeling including adipogenesis, angiogenesis, and collagen deposition. scCO2-decellularization holds the potential for an "off-the-shelf" biologically-derived adipose ECM that circumvents harsh chemical protocols and promises superior regenerative capacity with long-term volume restoration.

Introduction A new concept for facial contouring is proposed, utilizing light to in-situ harden an injectable regenerative implant. This implant is designed for smooth facial injections and changes its rheological properties in place, upon skin illumination. The photocuring technology enhances the filler's structure, stability, and cohesiveness, transforming it from its liquid form to a silicon-like facial implant. The light wavelength and intensity used for this transformation is safe and does not heat the patient's skin. This innovative injectable implant is composed of recombinant type I human collagen (rhCollagen) and hyaluronic acid. CollPlant's rhCollagen, known for its exceptional purity, uniformity and biofunctionality, promotes cells proliferation and tissue regeneration. Over time, the newly formed tissue replaces the injected material, offering a long-lasting, natural tissue rejuvenation.

Materials and methods The injectable implant's properties were characterized using a rheometer and a compression tester. A long-term pivotal implantation study in rats was conducted to evaluate safety, tissue regeneration and lifting capacity. Histopathological analysis, including Hematoxylin and Eosin (H&E) and Masson's Trichrome staining, was performed at 3, 6 and 12 months to assess the local tissue response, regeneration and integration. Weekley external measurements were taken to monitor the product's projection and lifting capacity. A commercial mid-face hyaluronic acid filler served as control. In situ photocuring was further evaluated in human cadavers: the product was injected into multiple areas, primarily on the periosteum, including the jawline, chin, nasal dorsum and tip, zygoma, and temples. Following photocuring these regions were dissected to assess the material's properties after the hardening process. Findings from animal study, along with product characterization, and human cadaver trials, are presented.

Results
The injectable regenerative implant has a storage modulus several folds higher than that of a standard commercial HA filler, while still enabling smooth injection through a 27G needle. This is achieved thanks to the photocuring technology. When compared to commercial filler, a higher G' is achieved with the same injection force. The long term in-vivo study demonstrated superior lifting capability compared to commercially available midface HA dermal filler with a consistent improvement in tissue ingrowth and integration, at 3, 6 and 12 months of follow up. No adverse events were reported. Moreover product safety was evaluated in-vitro and in-vivo per FDA guidance document on the use of ISO 10993-1 confirming its biological safety and biocompatibility. Cadaver assessments further demonstrated that the photocuring process can reach the periosteum, transforming the injectable implant into a stable, gummy bear-like implant.

Conclusion The advantages of this innovative photocurable and injectable regenerative implant have been demonstrated through in-vitro, in-vivo and in-cadaver trials. CollPlant's minimal invasively injectable implant introduces a novel approach to facial and body aesthetics, with the potential to become a game changer in clinical practice.

Purpose: Soft augmentation plays a critical role in reconstructive and aesthetic procedures. Among the most popular filler materials are autologous fat and hyaluronic acid, but the former provides inconsistent graft survival and causes donor site morbidity. Although hyaluronic acid (HA) is available off-the-shelf, it requires reapplication every 1-2 years since it largely gets absorbed without forming new tissue. Pre-clinical porcine studies using composite collagen hydrogels (CCH), which consist of collagen microspheres within a lower-density collagen matrix, demonstrated significantly enhanced cellular infiltration and vascularization allowing for simultaneous skin grafting and neodermis formation in one surgical step. Herein we aim to investigate a flowable/injectable form of the CCH to assess its capacity for tissue integration and volume retention following subcutaneous implantation. Methods: Sterile 1% bovine atelocollagen microspheres (80-300um in diameter) were mixed with 0.3% telocollagen bulk at varying microsphere-to-bulk ratios (70:30, 50:50, 30:70) to create a homogeneous, flowable hydrogel. 500ul of each mixed collagen suspension was injected subcutaneously bilaterally on the dorsa of wildtype mice, where it underwent in situ gelation. The injected materials were evaluated over a 12-month period with volumetric analysis and histological assessment to characterize local tissue responses and long-term maintenance. Results: The injectable CCH (iCCH) exhibited excellent handling properties, allowing for smooth administration via an 18G needle without leakage. Over one year, the shape and projection of the injected collagen hydrogel was maintained across all groups, with the highest volume retention in the 70:30 composite collagen group (98.3% of initial volume). Histological analysis demonstrated a uniform distribution of collagen microspheres throughout the injected hydrogel, with significant cell infiltration and vascularization observed within the bulk collagen as early as 1 month. The collagen microspheres remained present at 12 months but decreased in size, maintaining 80% of their initial dimensions due to gradual degradation. The lower density bulk collagen was progressively replaced and remodeled by host collagen, in addition to the retained collagen microspheres, contributing to sustained volume retention. Mature vascular structures were noticed as early as 3 months and remained stable out to one year. Macrophage analysis showed a predominance of M2 macrophages over the course of the study, indicating favorable biocompatibility of the iCCH. Biomechanical studies are underway. Conclusions: Unlike traditional fillers that resorb over time without induction of new tissue formation, this simple but unique composite collagen hydrogel promotes neovascularization and host collagen deposition, resulting in nearly complete volume retention over 12 months, without any signs of chronic inflammation. This novel biologic composite collagen injectable holds great promise for both reconstructive and aesthetic applications.

Purpose This study describes the clinical outcomes associated with the use of Ovine-Reinforced Hybrid Mesh (OviTex) in alloplastic breast reconstruction performed by a single surgeon after mastectomy. To date, published studies regarding OviTex use in breast reconstruction are lacking.

Methods and Materials A retrospective chart review was conducted on patients treated by a single surgeon between 2019 and 2023, who underwent immediate or delayed breast reconstruction using OviTex. Outcomes reviewed include reconstructive failure, implant exposure, infection, seroma, and hematoma. Descriptive statistics were performed using chi-squared and Fisher's exact test for small sample sizes. All statistical analyses were performed using IBM SPSS v29 (Armonk, NY).

Results Seventy-nine patients (134 breasts) were included. The average age at the operation was 51.26 (SD=13.27) years and the average BMI at the procedure was 26.57 (SD=4.75). The majority of reconstructions involved direct-to-implant procedures (65% vs. 35% tissue expander), with a predominance of immediate reconstructions (97% vs. 3% delayed) and pre-pectoral placement (72% vs. 28% dual-plane). There was an average implant size of 426 (SD=125) mL and an average intraoperative expander fill of 235 (SD=88) mL. Reconstructive failure occurred in 29 breasts (22%), all in immediate reconstructions, primarily due to exposure (20%), infection (10%), or a combination of both (8%). Four breasts (3%) developed a seroma, two of which also required surgical intervention for exposure. No hematomas were observed. Among breasts requiring reoperation, 21% had prior radiation, 90% involved the cancer side, and 34% were current or former smokers. Reconstructive failure was significantly higher following direct-to-implant compared to tissue expanders (p<0.001) and on the cancer side compared to prophylactic reconstructions (p<0.001). Of the 25 patients who experienced an RTOR, there was a significant difference between the 23 (92%) patients who experienced an exposure with an average RTOR time of 2.20 (SD=2.29) and the two (8%) patients who only experienced an infection with an average RTOR time of 0.88 (SD=0.40) (p=0.038). No significant differences in reconstructive failure were noted regarding placement location (pre-pectoral vs. dual-plane) or radiation treatment.

Conclusion This study presents the clinical outcomes of a single surgeon's use of OviTex in breast reconstruction. Continued adherence to evidence-based practices is essential for optimizing surgical strategies in this field.

Background and Purpose

Tissue-engineered constructs offer an alternative to traditional reconstruction methods for microtia. Auricular cartilage-derived chondrocytes are frequently used as a cell source for construct generation. However, their application is limited due to low cellular yields, slow proliferation, and phenotypic drift during in vitro expansion. Our previous research demonstrated that costal cartilage-derived chondrocytes (CCs) are a viable alternative to auricular chondrocytes. Nonetheless, identifying optimal conditions for maximal cell harvest and functionality to generate constructs remains an active area of research. Thus, we aimed to further characterize CCs and determine whether the location of cell harvest along the costal cartilage segment exhibited distinct cellular properties that could affect chondrocyte function.

Methods

Costal Cartilage-derived chondrocytes were harvested from the distal (sternal) and proximal (osteochondral junction) regions of costal cartilage in three paired pediatric donors undergoing autologous microtia reconstruction. The chondrogenic properties of CCs were evaluated in native tissue histology, 2D cell cultures, and pellet cultures. Cells were isolated by digesting the tissue in type II collagenase. The chondrogenic properties of the CCs were assessed via immunofluorescence and immunohistochemistry to detect type II collagen expression. Cell proliferation assays were also performed using the Cell Counting Kit-8 (Abcam CCK-8). Additionally, native tissue and pellet cultures were evaluated histologically via hematoxylin and eosin, Alcian blue for glycosaminoglycans and Masson's Trichrome for overall collagen. Images were quantified using ImageJ. Student t-tests were performed to evaluate statistical comparisons in RStudio.

Results

Distal costal cartilage tissue samples yielded significantly more chondrocytes than proximal cartilage samples. The histological analysis demonstrated that distal cartilage contained more densely packed cells and exhibited more type II collagen and glycosaminoglycan expression. Enzymatic digestion of the distal costal cartilage yielded 1.75-fold the number of cells per gram isolated from proximal cartilage. Native tissue histology demonstrated a higher type II collagen expression in distal CCs (p=0.074). Pellet cultures from distal CCs exhibited a significantly higher type II collagen expression (p<0.05) compared to proximal CC cultures. These results were supported by increased type II collagen expression in the distal CCs on immunofluorescence.

Conclusion

CCs within the distal and proximal regions of costal cartilage possess distinct properties. These findings suggest that CCs from distal tissue segments closer to the sternal junction may serve as a more viable cell source for cartilage scaffold seeding and construct generation. Further investigation is necessary to assess the function of distal and proximal CCs on decellularized auricular scaffolds to evaluate their chondrogenic potential.

Purpose Gender-affirming mastectomy (GAM) may leave behind varying amounts of breast tissue conferring an unknown, nonzero risk of developing breast cancer. Current breast cancer screening guidelines provide minimal evidence-based direction tailored to this patient population. This study aimed to assess the current practices by plastic surgeons when recommending breast cancer screening following GAM.

Methods A 12-item electronic survey was distributed to 5802 members of the American Society of Plastic Surgeons (ASPS) in 2024. Survey items focused on familiarity with current screening guidelines, individual screening practices, and confidence in recommendations. Descriptive analyses were performed.

Results A total of 582 responses were received and among respondents, 25.4% perform GAM. Of the currently available breast screening recommendations, 31.1% of respondents follow the WPATH Standards of Care, 20.9% guidelines for cisgender women, 12.8% ACR Appropriateness Criteria, 9.5% Fenway Health guidelines, 5.4% UCSF guidelines, and 20.3% none. Only 1 in 4 plastic surgeons express confidence in providing the appropriate recommendations. The overwhelming majority (93.9%) of plastic surgeons support the development of more specific, evidence-based, and transgender-inclusive screening guidelines.

Conclusion This is the first study to provide insight on plastic surgeon's perspectives on the current breast cancer screening guidelines for transgender individuals. Our findings reveal significant variability in practice among ASPS members regarding breast cancer screening after GAM. There is a critical need to develop improved guidelines in order to provide equitable cancer screening surveillance. Longitudinal research aimed at understanding breast cancer risk following GAM is needed to aid in the development of evidence-based screening protocols.

Introduction: Dupuytren's Disease (DD) is a progressive fibrotic disorder affecting the palmar fascia of the hand. A key feature of DD pathogenesis is disruption of normal collagen fiber anisotropy, a highly ordered pattern that provides mechanical flexibility. This organization is lost in DD, as myofibroblasts produce disordered collagen fibers that contribute to tissue stiffness and contracture (1).

Image analysis algorithms, such as TWOMBLI (The Workflow Of Matrix BioLogy Informatics), have been developed to enable the quantitative assessment of collagen fiber anisotropy and changes to the extracellular matrix (ECM). TWOMBLI is an ImageJ plugin designed to analyze ECM patterns by calculating parameters related to fiber alignment (2). It has been used in various disease models to provide insights into the structural remodeling associated with fibrosis (3).

The purpose of this study was to use TWOMBLI as a novel mechanism to empirically evaluate collagen alignment and anisotropy in Dupuytren's Disease.

Methods: Samples of nodule and cord tissue were obtained from two Dupuytren's patients undergoing palmar fasciectomy, processed, and mounted on glass slides. Slides were stained with Picrosirius Red (PSR) and imaged at 40x magnification under both brightfield and polarized light. Four regions of interest were captured from each sample.

Images were processed through TWOMBLI to generate line masks corresponding to collagen fibers. Parameters associated with collagen anisotropy were then calculated and outputs were compared between donors and tissue type.

Results: Compared to cord tissue, nodules exhibited a higher number of total matrix endpoints and lower number of hyphal growth units (HGU) – calculated as the average number of branch points per fiber length. Neither of these were statistically significant. Decreased HGU and increased endpoints have both been previously described as markers of more advanced fibrosis, signifying that Dupuytren's nodules display greater disruption of normal collagen alignment than cords.

Results also varied greatly between the individual donors, demonstrating potential patient-specific differences.

Conclusions: These results are limited by small sample size and lack of statistical significance. However, these findings provide insight into Dupuytren's pathogenesis and highlight the utility of TWOMBLI as a quantitative approach in characterizing fibrotic changes to the ECM. Further efforts are needed to evaluate these differences. Future studies should aim to incorporate a larger sample size and non-fibrotic samples as a negative control.

This study offers a novel approach to assessing fibrotic changes in collagen alignment in Dupuytren's Contracture. Though image analysis algorithms have been used previously to evaluate fibrosis, to our knowledge, this is the first attempt to quantify differences in collagen alignment in Dupuytren's Contracture or other fibrotic hand diseases.

References:

1. Walthall J, Anand P, Rehman UH. Dupuytren Contracture. In: StatPearls. Treasure Island (FL): StatPearls Publishing; February 26, 2023.
2. Wershof E, Park D, Barry DJ, et al. A FIJI macro for quantifying pattern in extracellular matrix. Life Sci Alliance. 2021;4(3):e202000880. Published 2021 Jan 27. doi:10.26508/lsa.202000880
3. Wyetzner RH, Segal EX, Jussila AR, Atit RP. Topographical changes in extracellular matrix during skin fibrosis and recovery can be evaluated using automated image analysis algorithms. FEBS Lett. 2024;598(24):2995-3004. doi:10.1002/1873-3468.14987

Introduction: Dupuytren's disease is a common fibrotic condition of the palm and digits. Central to the pathogenesis of the disease is the activation of Dupuytren's myofibroblasts (DMFs), which produce disorganized collagen fibers and alter the extracellular matrix (1).

Current standard treatment options involve excising or breaking down the fibrotic tissue – either via palmar fasciectomy or injection of collagenase clostridium histolyticum. However, these current modalities are associated with high rates of recurrence, usually within two years (2). We previously demonstrated that platelet rich plasma (PRP) and ASCs together induce transformation of DMFs into adipocyte-like lipid-laden cells in vitro with reduced markers of fibrosis in a rodent model (3). However, this treatment is inherently limited by person-to-person variability in PRP composition.

The purpose of this study was to determine if any individual component of PRP could act as a functional replacement in the presence of ASCs. Following a literature review and preliminary testing, insulin-like growth factor I (IGF-I) was selected as a potential candidate.

Methods: DMFs were isolated from palmar fascia taken from Dupuytren's patients. PRP was pooled from ten healthy donors and added to DMFs co-cultured with ASCs in vitro to serve as a positive control. Experimental groups consisted of DMFs with IGF-I alone and DMFs co-cultured with ASCs with IGF-I.

DMFs were also treated with insulin-like growth factor I receptor (IGF-IR) inhibitors as a negative control. Additionally, IGF-I depleted PRP was employed as another negative control. IGF-I was depleted from the pooled PRP using an antibody-based method and verified with ELISA.

The cells were assessed at four weeks using Oil Red-O or Bodipy to reveal lipid droplets.

Results: In the presence of ASCs, IGF-I successfully transformed DMFs into adipocyte-like cells in vitro, demonstrating lipogenic effects similar to PRP. DMFs treated with ASCs+PRP and ASCs+IGF-I showed significant induction of lipid-filled cells.

When DMFs were co-cultured with IGF-I-depleted PRP and ASCs, lipid-laden cells were reduced by 64%. Addition of an IGF-I receptor inhibitor to DMFs+ASCs+PRP and DMFs+ASCs+IGF-I reduced the percentage of lipid-laden cells from 84% to 13% and 80% to 11%, respectively.

Conclusions: This study offers mechanisms for discovering new therapeutic targets for treating skin fibrosis, using Dupuytren disease as a model with possible implications in other fibrotic conditions. Identifying IGF-I as a functional replacement for PRP in inducing transformation of DMFs to adipocyte-like cells in vitro is both a novel and important finding given the translational limitations of PRP as a treatment modality.

However, the use of ASCs has similar limitations. Further research efforts are needed to elucidate which paracrine mediators secreted by ASCs are responsible for the synergistic effects with IGF-I.

References: 1. Khaliq F, Orji C. Dupuytren's Contracture: A Review of the Literature. Cureus. 2024;16(12):e74945. Published 2024 Dec 2. doi:10.7759/cureus.74945 2. Nayar SK, Pfisterer D, Ingari JV. Collagenase Clostridium Histolyticum Injection for Dupuytren Contracture: 2-Year Follow-up. Clin Orthop Surg. 2019;11(3):332-336. doi:10.4055/cios.2019.11.3.332 3. Ziegler ME, Staben A, Lem M, et al. Targeting Myofibroblasts as a Treatment Modality for Dupuytren Disease. J Hand Surg Am. 2023;48(9):914-922. doi:10.1016/j.jhsa.2023.06.007

Purpose Patients with neuropathic pain of the head and neck often experience concurrent headache disorders (HDs), particularly migraines and headaches, which may complicate the diagnosis and treatment planning. The overall impact of nerve decompression on improvement of these individual conditions is unknown. We aimed to describe the clinical features of nerve pain, as well as concomitant HDs, and report on postoperative outcomes of these three conditions following nerve decompression surgery.

Methods From January 2023 to January 2025, 236 patients with a diagnosis of headache disorder were prospectively enrolled at two institutions. Screening for nerve decompression surgery was performed by the senior authors based on the following criteria: (a) failure of conservative treatment, (b) symptoms and physical exam findings suggestive of nerve compression, (c) pain drawing suggesting nerve pain, and (d) positive response to nerve blocks. Demographics, pain characteristics, and accompanying symptoms were collected through patient surveys. The primary treatment outcome was evaluated by symptom reduction in terms of pain frequency (days/month), duration (hours/day), and intensity (0-10).

Results In total, 178 (75.4%) patients reported nerve pain, 149 (63.1%) migraines, and 121 (51.3%) headaches. Of those with nerve pain, 152 (85.4%) were diagnosed with at least 1 concomitant HD. The average age of nerve pain onset was 35.8 +/-18.0 years, which was significantly later compared to both migraine (24.3 +/-13.3, p<0.001) and headache (26.8+/-18.2, p=0.002). When comparing these three pain conditions, nerve pain was more frequently described as starting at the back of the head compared to other anatomic locations (n=143 [80.3%], p<0.001). Headache pain quality was more commonly classified as dull (n= 47 [38.9%], p=0.03) whereas nerve pain quality was typically described as shock-like (n= 77 [43.3%], p<0.001) or stabbing (n= 112 [62.9%], p<0.001). At baseline, nerve pain occurred at significantly greater mean frequency in days per month (23.4 +/-18.4 days/month, p<0.001) and duration (17.8 +/-19.5 hours/day, p=0.007) than migraines or headaches. Nerve decompression surgery was performed in 45 (19.1%) patients. Of these, 36 (80.0%) also experienced migraines, and 26 (57.8%) had headaches. Postoperatively, significant decreases in median pain frequency, duration, and intensity were observed for nerve pain, migraine, and headaches (p<0.05). The mean postoperative follow-up was 8 months.

Conclusion Nerve pain, migraine, and headache often coexist, and patients may experience overlap in these distinct conditions. Although nerve decompression surgery is primarily aimed at alleviating nerve pain, this study also demonstrates improvement in migraine and headache symptoms. These findings suggests that nerve compression/irritation may trigger and exacerbate concomitant HDs.

Purpose The greater occipital nerve (GON) is the largest sensory nerve in the body. Compression or irritation of the GON can lead to debilitating neuropathic pain known as occipital neuralgia (ON). This study analyzes 14-years of experience to develop an algorithm for surgical management of occipital neuralgia (ON).

Methods
A prospective cohort study was conducted of 1405 patients who were screened for nerve decompression surgery at two institutions from January 2011 to January 2025. Surgical candidacy was assessed by the senior authors based on the following criteria: (a) failure of conservative treatment, (b) symptoms and physical exam findings suggestive of nerve compression, (c) pain drawing suggesting nerve pain, and (d) positive response to nerve blocks. Demographics, pain characteristics, and accompanying symptoms were collected through patient surveys. The primary treatment outcome was evaluated by symptom reduction in terms of pain frequency (days/month), duration (hours/day), and intensity (0-10).

Results A total of 284 patients with occipital neuralgia underwent 533 primary GON decompressions. There were 53 cases of GON reoperation with neurectomy performed in 37 (13%) patients on 43 (8.1%) nerves. At an average follow-up time of 15 months, patients that underwent GON reoperation achieved similar reduction in pain frequency (p=0.449), duration (p=0.849), and intensity (p=0.622) to those only requiring primary decompression. GON neurectomy techniques consisted of the proximal clamp, segmental cautery, protected plane approach (n= 18 [40.5%]), muscle burial (n= 14 [26.4%]), reset neurectomy with autograft (n= 8 [15.1%]), RPNI (n= 6 [11.3%]), nerve cap (n= 5 [9.4%]), and TMR with RPNI (n=2 [3.8%]). All patients achieved at least 80% improvement in pain in the cohort that underwent the proximal clamp, segmental cautery, and protected plane approach compared to only 28.6% of patients in the muscle burial group (p=0.001). A total of 48 patients (17.0%) underwent surgery to address a secondary pain site, most commonly involving the supraorbital/supratrochlear nerves (n= 25 [52.1%]). Patients that underwent secondary site surgery achieved comparable improvement in pain frequency (p=0.381), duration (p=0.683), and intensity (p=0.682) to those only requiring primary decompression.

Conclusion Following primary decompression of the GON for treatment of ON, secondary surgeries may be required if pain reoccurs in the same location or is present at a different pain site. With the proposed algorithm, outcomes following secondary surgery were similar to those of primary decompression.

Purpose

The velopharyngeal port is a sphincter that isolates the soft palate and lateral/posterior walls of the nasopharyngeal cavity from that of the oral cavity. In velopharyngeal insufficiency (VPI), as seen in patients with clefts, there is inadequate closure of this port during speech, resulting in hypernasality. This study aimed to automate the process of identifying velopharyngeal port closure by applying a Convolutional Neural Network (CNN) to nasopharyngoscopy studies in both patients with clefts and without clefts.

Methods

A custom CNN (termed NasoNet) was developed to analyze nasopharyngoscopy videos of patients with and without VPI. Video frames were extracted at a rate of 25 frames per second, resulting in 30,620 frames, and the CNN was trained on a patient-level analysis. The CNN's architecture consisted of four convolutional layers designed to identify the status of the velopharyngeal sphincter in each frame as either "open" or "closed". Accuracy and area under the curve (AUC) were obtained to measure performance specifically.

Results

The final dataset for the patient-level analysis comprised 24,010 frames, including 13,167 frames classified as 'cleft' and 10,843 as 'non-cleft'. Within the cleft category, 5,926 out of the 13,167 cleft frames were manually labeled as 'insufficiently closed', while 2,628 out of the 10,843 non-cleft frames were labeled as 'closed'. Naso-Net achieved a frame-level accuracy of 95.41% and 94.79% for non-VPI and VPI frames with AUC values of 97.43% and 97.50%. For patient-level analysis, the model achieved an accuracy of 85% for non-VPI and 82% for VPI patients, with AUC values of 95.40% and 84.50%.

Conclusion

This study demonstrates that CNN may be developed for automated analysis in nasopharyngoscopy studies of patients with clefts. The reliable performance of Naso-Net demonstrates the potential of automated analysis to both improve VPI diagnosis and increase access to highly specialized care.

Purpose: Abnormal fibroproliferative scars are characterized by hypervascularity and excessive extracellular matrix accumulation. They are symptomatic, with a high potential for recurrence, and have a notable impact on patient quality of life. Research highlights the immune system's critical role in the development of these scars (1-2), but little is known about what drives their recurrence after surgical excision. Our study aims to explore gene activity in the peripheral tissue of abnormal scars to identify key mRNA molecules that may contribute to their regrowth.

Methods: Clinically defined keloid scars (n=4), hypertrophic scars (n=2), and normal skin (n=2) were collected with patient consent. All scars were removed to alleviate patient pain and discomfort. Histological evaluation identified 3 x 3 mm areas encompassing both the scar tissue and the adjacent surrounding tissue. These areas were excised and underwent spatial gene expression analysis using the Visium Cytassist platform. The marginal tissue of scars that regrew after excision (n=3) was compared to that of scars that did not regrow (n=3). Raw mRNA sequencing data were processed using 10X Cell Ranger Software to generate expression profiles for each spot and identify clusters of spots with similar profiles. The Seurat package in R (version 4.4.0) was employed to determine significant differences in gene expression between groups. Statistical significance was set at p ≤ 0.05.

Results: Four clinically defined keloid scars were collected from two African American (AA) females, one Asian female, and one White male, averaging 50.5 ± 17.2 years old. Three scars had a history of repeated corticosteroid injections, with one treated scar regrowing 84 days post-excision. Two clinically defined hypertrophic scars were collected from an AA female and a mixed-race female, averaging 47.5 ± 9.19 years old. One scar received laser treatment, yet both scars regrew post-excision at 100 and 267 days, respectively. Two normal skin samples were obtained from an AA and a White female, averaging 58.5 ± 2.12 years old. Compared to scars that did not regrow, those that did showed a significant decrease in wound healing pathways regulated by YAP1 (Z-score: 0.54, p= 6.29e-18 vs Z-score: 4.23, p= 6.51e-25) and TGF-β (Z-score: 2.58, p= 5.54e-62 vs Z-score: 7.58, p= 9.29e-65) in the peripheral tissue. Additionally, the peripheral tissue of scars that regrew exhibited a significant increase in the inflammatory immune response regulated by TNF (Z-score: 4.35, p= 8.84e-69 vs Z-score: 2.30, p= 3.02e-50) and STAT1 (Z-score: 6.21, p= 7.82e-20 vs Z-score: 4.28, p= 1.09e-8).

Conclusions: Our findings suggest that abnormal scar regrowth is associated with a decreased expression of wound healing genes and increased expression of inflammation-related genes in the peripheral scar tissue. This study uncovers potential targets for the development of novel therapies and highlights marginal tissue as an area of focus for future research and scar treatment applications.

References: 1. Aramaki-Hattori N, et al. Relationship between Keloid Formation and YAP/TAZ Signaling. Plast Reconstr Surg Glob Open. 2017;5(6):e1357. 2017 Jun 13. 2. Wang X, et al. Identification and characterization of four immune-related signatures in keloid. Front Immunol. 2022;13:942446. 2022 Jul 27.