Scientific Abstract Presentations: Hand 5 Sunday, October 29 Sun, Oct 29 8:00-9:30 a.m. CDT

Introduction: The role of surgery remains unclear in management of Radiation-induced brachial plexopathy (RIBP), with the predominant approach being conservative therapy.

Methods: A literature search was performed using the main online databases to find all related articles. Systematic review was performed including 29 studies (n=580) that described the clinical features of RIBP patients and outcomes after surgery.

Results: The most commonly reported symptom was sensory loss (n=295,59.8%), followed by motor deficits (n=279,56.6%), and neuropathic pain (n=267,54.1%). Sixty-five (56.0%) patients had panplexal involvement, and 51 (44.0%) patients had partial plexus involvement. The most common surgical interventions were neurolysis with omental or other flaps (n=108,45.6%), followed by neurolysis alone (n=71,29.9%). Overall, out of 237 patients that underwent surgery, 125 (52.7%) reported improved neuropathic pain. Motor and sensory deficits were improved in 46 (19.4%) and 39 (16.4%) patients, respectively. In patients that underwent surgical neurolysis with omental or other flaps, 57 (52.8%) patients had improvement in pain, followed by improvement in sensory and motor deficits in 17 (15.7%) and 13 (12.0%) patients, respectively. In patients that underwent surgical neurolysis alone, pain (n= 55, 77.5%) was most commonly improved, followed by motor (n= 17, 23.9%) and sensory deficits (n= 15, 21.1%), respectively.

Conclusion: Surgery is most effective in in alleviating pain, but has less satisfactory outcomes for motor and sensory improvement. For motor deficits, ulnar nerve fascicular transfer to the biceps branch and cable grafting of the musculocutaneous nerve have shown encouraging results.

Introduction: Perilunate dislocations are rare, yet serious, high-energy injuries with many limiting functional sequelae, including residual pain, stiffness, and arthritis of the wrist. In this study, we review clinical outcomes and their tools for measurement following operative fixation of perilunate fracture and dislocation.

Methods: A systematic review of studies pertaining to clinical outcomes following operative management of perilunate dislocations was conducted by querying Web of Science, PubMed/MEDLINE, Cochrane, and Embase databases. PRSIMA guidelines were followed. After removal of duplicates, 246 articles were screened by two independent reviewers.

Results: Forty-four studies met inclusion criteria for the study. Most were retrospective, single center studies, encompassing 885 patients (896 wrists). Most (99.4%) were male, with a mean age of 32.3. Mean follow-up was 50.3 months (minimum average 22.4 months), with 93.3% of wrists having undergone open reduction and internal fixation (ORIF). Most (86.3%) studies reported grip strength (mean: 36.0 kg; 76.4% of contralateral side). The most used clinical scoring assessment was the Mayo Wrist Score (MWS) (N=24 studies; mean: 75.6), followed by the Disabilities of Arm, Shoulder, and Hand (DASH) score (N=15 studies; mean= 21.51) and Patient-Related Wrist Evaluation (PRWE) score (N= 12 studies; mean= 26.7). The most reported complications postoperatively were posttraumatic arthritis and residual wrist pain. Nonunion rate did not differ significantly across surgical techniques (dorsal vs. volar vs. combined vs. arthroscopic).

Conclusion: Our systematic review suggests that measurement of clinical outcomes following surgical intervention for perilunate dislocations is not standardized, with a range of clinical scoring assessments being used. Furthermore, even with operative fixation, clinical outcomes for patients with perilunate dislocation remain overall varied, with some patients reporting residual pain and chronic osteoarthrosis.

Introduction: Breast cancer-related lymphedema is a prevalent cause of upper extremity lymphedema, affecting 17% of breast cancer survivors. Several risk factors, such as age, obesity, radiation, chemotherapy, axillary node dissection, hypertension, and taxane therapy, have been identified through prior studies. However, further research is required to explore other possible risk factors, such as the use of Beta-blockers (BB). Clinical decisions could be made to improve overall patient outcomes by identifying those at the highest risk. We aim to examine the association between BB usage and upper extremity lymphedema and cellulitis in patients who have undergone a mastectomy. Methods: Anonymized patient data from 75 healthcare organizations (HCOs) was obtained through the TriNetX platform. The data analysis was performed at healthcare organizations, and the aggregated results returned to the platform. To build the cohorts, we used ICD-10, CPT, and TNX-curated codes. The query and analysis were conducted in February 2023 using the analytic tools from the TriNetX platform. We compared two cohorts of mastectomy patients: one who received BB therapy within one year of surgery (cohort 1) and another who did not receive BB therapy (cohort 2). The cohorts were matched on age at Index, sex, race, ethnicity, body mass index, diabetes mellitus, hypertension, heart failure, chronic kidney disease, cancer, cellulitis and acute lymphangitis, lymphadenectomy, type of mastectomy, type of breast reconstruction, radiotherapy, chemotherapy, and use of calcium channel blockers. The outcomes of interest were the development of cellulitis and lymphedema in the first three years following mastectomy. After matching, the Risk Ratio (RR) with a 95% CI was calculated to evaluate cohort differences. Results: Between 2003 and 2017, a total of 68,568 subjects met the eligibility criteria. Following the propensity score matching, each cohort had a patient count of 16,128. After matching, the mean age at Index for the cohort 1 and cohort 2 was 61.1 (SD 13.4) and 61.3 (SD 13.1), respectively. Moreover, 97.8% were females, 77.1% were white, 1525 patients developed lymphedema, and 3657 patients developed cellulitis. The cohorts had significant differences in demographic and clinical characteristics, including age at Index (p=0.006), ethnicity (p=0.01), hypertension (p=0.002), heart failure (p=0.011), diuretic use (p<0.001), and BMI (p<0.001). There was a positive association between the use of BB within 1 year of mastectomy and cellulitis (RR = 1.179, 95% CI [1.109, 1.254], [p<0.001]). Additionally, the use of BB within 1 year of mastectomy increased the risk of developing lymphedema (RR = 1.121, 95% CI [1.016, 1.237], [p=0.022]). Conclusion: Our results propose that using BB within a year of mastectomy increases the risk of developing cellulitis and lymphedema. Further research addressing the unbalanced variables within the cohorts is required to exclude any potential confounders.

Background: Interventions for type B postaxial polydactyly include suture ligation and surgical excision. To date, there is a paucity of literature comparing the long-term outcomes of these procedures. Thus, this study sought to analyze and compare the patient-reported short and long-term outcomes of these procedures.

Methods: Following institutional review board (IRB) approval, the authors performed a retrospective review of patients who underwent primary suture ligation or surgical excision for type B postaxial polydactyly at our tertiary care institution between 2010 and 2016. Baseline demographic characteristics, age at the time of surgery, type of initial treatment (ligation vs. surgical excision), complications within 30 days after the initial procedure, and additional procedures were recorded. To specifically evaluate the long-term complications, a six-question survey was distributed from January, 2021 to March, 2022. The patients were queried about the incidence of sensitivity or pain, presence of hypertrophic scars, and/or persistent presence of bump ("nubbin") at the site of the excised supernumerary digit.

Results: A total of 158 responses accounting for 258 digits were attained for a 53% response rate. Overall, 67.4% were initially managed surgically (n=174) and 32.6% underwent suture ligation (n=84). Median age at the time of procedure across both cohorts was 49 [IQR 21, 97] days. Patients treated surgically were significantly older at the time of excision (median age 63.0 [33.3, 113.0] vs. 13.0 [0.0, 113.0] days, p<0.05). The short-term complication rate was 1.6%, accounting for four cases (ligation 1.5% vs. surgical excision 1.5%, p=0.964). Regarding long-term complications, the median age at survey was 8 [IQR 5.4, 10.2] years. Overall, the rate of long-term complications was 39.5% (ligation 51.5% vs. surgical excision 35.4%, p<0.05). The likelihood of postoperative sensitivity (ligation 12.1% vs. surgical excision 11.5%, p=0.88) and presence of hypertrophic scars (ligation 10.6% vs. surgical excision 15.1%, p=0.36) was comparable in both groups. However, the odds of nubbin in the excision group were 57% lower than the ligation group (OR: 0.44; 95% CI: 0.24, 0.78; P=0.006). These findings remained significant in the adjusted analysis.

Conclusion: This study suggests that suture ligation can be used in select cases without increasing the prevalence of long-term pain or sensitivity. Time to treatment remains a key variable in deciding either treatment.

Background Health care expenditure has been continuously increasing with widespread variation of spending across the United States. In recent years, increased attention has been drawn around a value-based health care system with emphasis to push toward cost-conscious system and improve health outcomes relative to cost. One of the strategies to decrease the costs is uncovering variations in spending. This study aims to investigate nationwide cost variation for digital replantation.

Methods The data was retrieved from the Health Cost and Utilization Project National Inpatient Sample (NIS) database from 2016 to 2019. All patients age 18 years or older who had any single digit amputation and underwent replantation were included in this study. Patients were identified by the use of the International Classification of Disease, Tenth Revision (ICD-10) diagnosis and procedure codes. The primary variable of interest was hospital cost. Sociodemographic variables and hospital-level characteristics were analysed. Furthermore, patients were separated into four groups based on the hospital region, defined by NIS as Northeast, Midwest, South and West. Multivariable linear regression was implemented to evaluate predictors for cost variation for digital replantation.

Results Over the study period from 2016 to 2019, a total of 414 patients underwent digital replantation. The average age was 41 years, with majority of patients being white male (n=250, 85%). Majority of patients had private insurance (n=143, 35%) or other insurance (n=131, 32%). Hospitals performing replantation were more likely to be teaching hospital (n=385, 93%) and have large bedsize (n=293, 71%). The median length of stay was 5 days and median cost of stay was $91,805. Revision amputation was performed in 137 (33%) cases. When comparing regions, there was a statistically significant difference in cost with West being the most expensive and South least (median $114,792 and $78,295, respectively). Using multivariable analysis, increased length of stay (mean difference (95% CI), $10209 ($8458-$11961), p<0.001) and revision amputation (mean difference (95% CI), $30670 ($13049-$48292), p<0.001) were predictive of higher costs, while small bedsize hospital was predictive of lower costs (mean difference (95% CI), -$51473 (-$77563-$25384), p<0.001).

Conclusion In this study, there was significant nationwide cost variation for digital replantation with hospitals in the West region having highest expenditure. The results of this study highlight the constant need to improve excess utilization costs while maintaining optimal patient outcomes.

Introduction: Breast cancer-related lymphedema (BCRL) is a chronic, disabling condition that progresses over time, frequently develops following surgical breast cancer treatment, and affects between 2 and 3 million people in the United States. Carpal tunnel syndrome (CTS) is the most common median nerve neuropathy, particularly in women, which accounts for 90% of all neuropathies, with almost 500,000 hand surgery procedures performed annually to correct it. While lymphedema can potentially increase the susceptibility of median nerve compression, the relationship between BCRL and CTS hasn't been well established yet. This study aims to explore any association between BCRL and CTS development in order to advance clinical understanding and treatment of both disorders. Methods: This study was conducted with anonymized data accessed via the TriNetX platform, which is being increasingly utilized to perform real-world data studies that are helpful in clinical practice. Data from approximately 111,962 patients who underwent mastectomy from 50 HCOs were obtained via TriNetX.
This analysis was run on the Research Network, which has around 108 million patients from 75 HCOs in four countries. The analysis is performed at HCO, with only aggregated results returned to the platform. We utilized ICD-10, CPT, and TNX-curated codes to build our cohorts. After creating the cohorts, the analysis was conducted on February 2023 using the analytic tools built into the TriNetX platform. We compared mastectomy patients' incidence of CTS based on the BCRL status (cohort 1 patients with lymphedema, cohort 2 patients without lymphedema). In addition, the cohorts were matched on age, sex, race, ethnicity, BMI, hypertension, diabetes mellitus, congestive heart failure, chronic kidney disease, type of mastectomy procedure, and the presence of different kinds of rheumatoid arthritis with and without rheumatoid factor. Patients also got matched based on using Anastrozole and Glucocorticoids, and patients with a history of CTS were excluded from the study. The outcome of interest was the development of CTS in the first 5 years after the incidence of lymphedema. Results: There were 111,962 mastectomy encounters before matching in the TriNetX database and 9290 after matching during the 5-year period. The mean age at the index for the cohort 1 was 57.9 years (SD 12.5) and for cohort 2 was 58 years (SD 12.7), and 99.4% of patients were females. The mean BMI for the cohort 1 was 30.4 (SD 6.8) and for cohort 2 was 30.2 (SD 6.9). The percentage of patients with comorbidities was 19.7%, 49.7%, and 6.6% for DM, HTN, and at least one type of rheumatoid arthritis disorder, respectively. After matching, the number of patients with CTS was 443 (4%), and there was no significant difference in the risk of CTS between postmastectomy patients with and without lymphedema (RR 1.091, 95% CI (0.910, 1.308), P= 0.348). Conclusion: There is no meaningful relationship between BCRL and the risk of CTS development. Therefore, focusing on other possible CTS risk factors in future studies can help to describe and manage this phenomenon.

Introduction: Lymphovenous bypass (LVB) is a microsurgical procedure used to restore lymphatic drainage in patients with lymphedema. We sought to determine if the change in limb volume after LVA differed between BMI groups in patients with upper extremity lymphedema. Methods: Adult patients that underwent upper extremity LVB were included. The mean percentage change in volume difference was calculated as follows: % Change = [(postoperative difference) – (preoperative difference)/(preoperative difference)] * 100, where postoperative difference is the absolute difference between the affected and unaffected limbs' volume at a specific endpoint and the preoperative difference is the absolute difference between the affected and unaffected limbs' volume preoperatively. The endpoints were 2 weeks , 2 months, and 6 months after the surgery. Subjects were grouped by BMI into 'Normal Weight' (18.5 - 24.9 kg/m2), 'Overweight' (25 - 29.9 kg/m2), and 'Obesity' (≥ 30 kg/m2).Shapiro-Wilk tests were used to assess normality and differences per endpoint among BMI groups were evaluated using Kruskal-Wallis tests. The limb volume change is expressed as a percentage. Data is presented as mean ± standard deviation. All statistical analyses were performed using R version 4.2.2 using the RStudio IDE version 2022.12.0.353. Results: Sixty-eight patients were identified ('Normal Weight', n=7; 'Overweight', n=24; 'Obesity', n=37). The Kruskal-Wallis test was used to assess for significant differences in endpoint values among the three BMI categories. The test, which compares median values, was not statistically significant at any endpoint (χ2 = 1.61, df = 2, p = 0.45 for week 2; χ2 = 3.94, df = 2, p = 0.14 for month 2; χ2 = 4.24, df = 2, p = 0.12 for month 6; χ2 = 0.14, df = 2, p = 0.93 for year 1). Despite this, large differences in the mean volume change were observed between the BMI categories. The mean percentage change in volume difference among BMI categories at 2 weeks was as follows: 'Normal Weight', -156% ± 169; 'Overweight', -59.88% ± 116; 'Obesity', -45.6% ± 79.6). At 2 months, the mean percentage change in volume difference among BMI categories was the following: 'Normal Weight', 12.3% ± 81.4; 'Overweight', -16.8% ± 68.6; 'Obesity', -38.2% ± 64.3). The mean percentage change in volume difference among BMI categories at 6 months was: 'Normal Weight', -32.5% ± 30.8; 'Overweight', -20.8% ± 26.3; 'Obesity', -55% ± 26.4). At 1 year, the mean percentage change in volume difference among BMI categories was the following: 'Normal Weight', -54.5% ± 102; 'Overweight', -36.6% ± 51.3; 'Obesity', -46% ± 18.4). Conclusion: In conclusion, no significant differences were observed in endpoint values among the three BMI categories using the Kruskal-Wallis test. However, there were notable differences in the mean percentage change in volume difference between BMI categories at different endpoints. These findings suggest that BMI may have an impact on volume change following LVB. However, the relatively small sample size might have affected both the statistical significance and the normality of the endpoints.

Purpose: Reconstruction of upper extremity soft tissue wounds with exposed bone and tendon remains a challenge. Skin substitutes and dermal matrices have been utilized to assist in creating well vascularized wound bed to facilitate skin grafting. Recent literature has demonstrated the use of Novosorb Bio-degradable Temporizing Matrix (BTM) in management of complex wounds. We hypothesize that BTM is safe and effective for reconstruction of complex upper extremity wounds.

Methods: A retrospective, IRB-approved chart review was performed for all patients who underwent reconstruction of complex upper extremity soft tissue defects with BTM between January 2017 and May 2022. Demographic data, comorbidities, wound etiology, wound size, secondary surgery and complications were recorded.

Results: 51 patients were identified using a CPT query of the electronic medical record. Patient population included 39 males and 12 females with an average age of 44.3 years. Nineteen patients (37.3%) were active smokers and 7 patients (13.7%) had diabetes. Wound etiology included trauma (n=30, 58.8%), burns (n=12, 23.5%), infection (n=8, 15.7%), and vasopressor-related injury (n=1, 2.0%). Twenty-four patients (47.1%) had wounds with exposed bone and 27 patients (52.9%) had exposed tendon/muscle.

The average size of BTM template used was 162.5 (range: 1.5-1000) cm2. Average time from BTM application to complete wound closure was 90.1 (range: 11-207) days. Twenty patients (39.2%) re-epithelialized spontaneously after removal of the sealing layer and did not require skin grafting. Average wound size for these patients was 58.5 (range: 2-200) cm2. Time to wound closure by secondary intent was 123.7 (range: 35-190) days. Twenty-seven patients (52.9%) underwent skin grafting and the average wound size in this cohort was 248.6 (range: 15–1000) cm2. Time to skin grafting from BTM application was 51.7 (range: 24-126) days. When comparing size of wounds, those who did not require skin grafting had significantly smaller wounds compared to those who required skin grafting (58.5 cm2 vs 248.6cm2; p = 0.002).

Overall, 49 patients of 51 (96.1%) achieved successful wound closure. The two patients who failed reconstruction required revision finger amputation and secondary flap reconstruction. Complications occurred in 14 patients and included template infection (n=10), template fluid collection (n=5) and template dehiscence (n=3). Five patients required revision surgery for infection and three of these patients required repeat application of BTM.

Conclusion: Here, we demonstrate that Novosorb BTM is safe and effective for management of complex upper extremity wounds with exposed bone and tendon. We have also found that secondary skin grafting may not be necessary in patients with smaller wounds.

Introduction: Lymphovenous Bypass (LVB) is many hospitals' first-line treatment for lymphedema. Many aspects of its effects, however, remain unknown. This study aimed to analyze the relationship between LVB anastomosis and the outcomes of surgery for lymphedema in the upper and lower extremities.

Methods: Adult patients that underwent upper extremity LVB were included. The mean percentage change in volume difference was calculated as follows: % Change = [(postoperative difference) – (preoperative difference)/(preoperative difference)] * 100, where postoperative difference is the absolute difference between the affected and unaffected limbs' volume at a specific endpoint and the preoperative difference is the absolute difference between the affected and unaffected limbs' volume preoperatively. The endpoints were 2 weeks, 2 months, and 6 months after the surgery. Patients were categorized based on the number of anasotmosis ('1 to 4 anastomosis', '5+ anastomosis') and type of anastomosis (combined,' 'end-to-end,' 'end-to-side'). After evaluating the endpoints for normality using Shapiro-Wilk tests, differences per endpoint among the number and the type of anastomosis were evaluated using Kruskal-Wallis tests. The limb volume change is expressed as a percentage. Data is presented as mean ± standard deviation.

Results: 70 patients were identified for this study ('1 to 4 anastomosis' = 57, '5+ anastomosis' = 13 for the number of anastomosis and 'combined' = 27, 'end-to-end' = 28, 'end-to-side' = 15 for the type of anastomosis). Kruskal-Wallis's test assessed for significant differences in endpoint values among the number and type of anastomosis categories.

The comparison between the median values was not significant at any endpoint for the number of anastomosis analyses (χ2 = 3.3826, df = 1, p-value = 0.06589 for week 2; χ2 = 0.12267, df = 1, p-value = 0.7262 for month 2; χ2 = 3.0823, df = 1, p-value = 0.07915 for month 6; χ2 = 3.0925, df = 1, p-value = 0.07865 for year 1). Despite this, there were large differences in the mean volume changes when looking at the number of anastomosis categories. The mean percentage change in volume difference between the number of anastomosis categories at 2 weeks was '1 to 4 anastomosis', -51.5% ± 72.5 and '5+ anastomosis, -92.6% ± 132. At 2 months, the mean percentage change was '1 to 4 anastomosis', -36.3% ± 49 and '5+ anastomosis, -6.73% ± 99.3. At 6 months, the mean percentage change was '1 to 4 anastomosis', -31.8% ± 26.1 and '5+ anastomosis, -60% ± 31.1. For 1 year, the mean percentage change for the number of anastomosis categories was '1 to 4 anastomosis', -31.2% ± 43.7 and '5+ anastomosis, -69.1% ± 33. Additionally, the test did not show statistically significant results at any endpoint for the type of anastomosis (χ2 = 0.95797, df = 2, p-value = 0.6194 for week 2; χ2 = 0.0048961, df = 2, p-value = 0.9976 for month 2; χ2 = 2.1007, df = 2, p-value = 0.3498 for month 6; χ2 = 0.73709, df = 2, p-value = 0.6917 for year 1). However, the mean percentage change in volume difference among type of anastomosis categories showed large differences at each endpoint. At 2 weeks, the mean percentage change was 'combined,' -156% ± 169; 'end-to-end,' -59.88% ± 116; 'end-to-side,' -45.6% ± 79.6. At 2 months, the mean percentage change was 'combined,' -156% ± 169; 'end-to-end,' -59.88% ± 116; 'end-to-side,' -45.6% ± 79.6. At 6 months, the mean percentage change was 'combined,' -156% ± 169; 'end-to-end,' -59.88% ± 116; 'end-to-side,' -45.6% ± 79.6. Lastly, at 1 year, the mean percentage change for type of anastomosis categories was 'combined,' -156% ± 169; 'end-to-end,' -59.88% ± 116; 'end-to-side,' -45.6% ± 79.6.

Conclusion: In the statistical analysis, our data indicates no relationship between the number or type of anastomosis and lymphedema measurements at each endpoint. However, there are significant differences in the mean percentage change in volume difference between the number and type of anastomosis groups at different endpoints. These findings imply that age may have an effect on volume change following LVB. However, the relatively small sample size may have influenced the statistical significance and normality of the endpoints.

Background: Nerve injuries from gunshot wounds to the upper arm can cause significant morbidity and loss of function. However, indications for surgical exploration and nerve reconstruction remain unclear as both low- and high-grade injuries can present with an abnormal neurological exam.

Methods: Adult patients presenting with a history of isolated gunshot wound to the upper arm between 2010-2019 at a single urban level 1 trauma center were screened for inclusion in this retrospective study. Patient demographics, neurological exam findings, concurrent injuries, and intra-operative findings were gathered. Bivariate analysis was performed to characterize factors associated with nerve injuries.

Results: There were 139 adult patients with isolated brachial gunshot wounds, and 49 patients (35%) presented with an abnormal neurological exam and significantly associated with concurrent humerus fractures (39% vs 21%, p=0.026) and brachial artery injuries (31% vs 2%, p<0.001). Thirty of these 49 patients were operatively explored. Fifteen patients were found to have observed nerve injuries during operative exploration including 8 patients with nerve transections. The radial nerve was the most commonly transected nerve (6), and among the 16 contused nerves, the median (8) was most common.

Conclusions: Nerve injury from upper arm gunshot wounds is relatively common with directly traumatized nerves in at least 39% and nerve transection in at least 16% of patients with an abnormal neurological exam. Timely referral to a hand and/or peripheral nerve surgeon for close clinical follow-up and functional reconstruction with nerve grafts, tendon transfers, and nerve transfers is recommended.

Background Hand injuries represent one of the most common pediatric traumas and account for around 1.7% of pediatric emergency room visits in the United States each year. While most hand fractures are managed conservatively, an estimated 10% will ultimately require surgical intervention. As such, transfer to specialized centers is common for pediatric hand trauma patients. These transfers have the potential to significantly reduce patient morbidity; however, when specialized care is not indicated, they can place undue burden on families to travel outside their communities while diverting resources from urgent cases. Our project aims to identify factors associated with patient transfer in pediatric hand trauma.

Methods A retrospective review was performed of patients under 18 years of age who were evaluated for hand trauma at one pediatric Level I trauma center between 2010 and 2020. Variables studied included patient demographics, etiology of trauma, medical history, and associated injuries. Patients were categorized based on transfer status, and factors associated with increased likelihood of transfer were identified. Finally, choice of management and outcomes were recorded for each hand fracture.

Results A total of 1151 patients met inclusion criteria. Of these, 308 (26.8%) were transferred from an outside institution. Certain injury types were associated with a significantly higher likelihood of transfer; specifically, scaphoid fractures (RR 7.63, CI 1.80-72.58), index finger injuries (RR 1.57, CI 1.01-2.43), and fingertip injuries (RR 1.62, 1.08-2.44) were more likely to be transferred (p<0.04), as opposed to phalangeal or metacarpal fractures. Mechanisms of injury, such as motorized vehicle accidents (MVA), and animal bites were also associated with increased risk of transfer (RR 6.06, CI 1.90-19.35, p<0.001; RR 13.47, CI 1.59-114.25, p=0.002, respectively). Finally, rural geography was associated with 2.89 times greater risk of transfer compared to patients living in urban or suburban areas (RR 2.89, CI 1.67-5.02, p<0.001).

Conclusion Pediatric hand trauma is one of the most common causes of emergency room visits in the United States each year. Understanding factors that influence the likelihood of transfer to specialized institutions is critical to optimizing patient care in the management of these injuries.

Introduction: Traumatic amputations of the finger and thumb comprise some of the most common injuries in the United States each year. However, there is variability in the reported success rates for replantation of digits, which have ranged from 48% to 97%.(1,2) Various factors including sociodemographic, surgical technique, and medical comorbidities have been associated with outcomes.(1,2) The purpose of our study is to assess the outcomes and success rate of digit replantation and identify predictors of digit survival.

Methods: We conducted a retrospective review to evaluate all patients with operative traumatic digit amputations at our institution from 2012-2022. Data on patient demographics, mechanism of injury, operative details, postoperative complications and follow-up were recorded. Pearson chi-square and independent t-tests were performed, with statistical significance set at p < 0.05.

Results: A total of 60 traumatic digit amputations were taken to the operating room for attempted replant. A total of 29 digit replants (63%) were performed, and 35 revision amputations were performed in the operative room after initial debridement (86.2% males, 68.9 % Hispanic). For digits that were replanted, crush/avulsion was the most common mechanism of injury (72.4%). Average age of patients with attempted replants was 32.5 years (SD 21.2). Average time from arrival at ER to operating room was 4.7 hours (SD 4). Majority of the attempted replants were at the MCPJ or through the metacarpal. The survival rate was 58.6%. Replants with and without vein grafts had similar success rates (58.3% vs 60%, p= 0.95). Digit survival rates were higher in patients with sharp injuries as compared to crush/avulsion injuries (87.5% vs 42.9%, p=0.05). Risk of failure of replant was higher in patients with diabetes and/or hypertension. Mean follow-up was 9.19 months. Of all patients with successful replants, 56.3% were compliant with hand therapy postoperatively.

Conclusion: Digit replantation after a traumatic injury may be considered in selected individuals. Careful preoperative assessment of mechanism of injury, comorbidities and demographics play a significant role in short term and long term success of digit replantation.

References: 1. Mulders M A, Neuhaus V, Becker S J, Lee S G, Ring D C. Replantation and revascularization vs. amputation in injured digits. Hand (NY) 2013;8(03):267–273. 2. Cheng G L, Pan D D, Yang Z X, Qu Z Y. Replantation of digits amputated at or about the distal interphalangeal joint. Ann Plast Surg. 1985;15(06):465–473.

Purpose: Early repair of neurotmetic peripheral nerve injury in the upper extremity offers better outcomes than delayed repair. Conventional wisdom suggests that sensory nerves may be more tolerant of delay than motor nerves, but the differential impact of delay on motor and sensory nerves has not been well-characterized. Furthermore, It is unknown how much delay can be tolerated before a nerve repair is futile. This study aims to elucidate how sensory and motor outcomes differ after nerve repair in the upper extremity through a systematic review of the literature and meta-analysis of individual participant data.

Methods: PubMed was queried for original articles describing outcomes after repair of median, ulnar, or radial nerves published between 1970 and 2022. Articles were included if they described individual patient outcomes for sensory or motor function according to British Medical Research Council grading. Individuals who had isolated digital nerve injuries were excluded. The effects of clinical variables on sensory and motor outcomes were assessed by univariate and multivariate regressions using generalized linear mixed models. A subset analysis was performed on mixed motor/sensory nerves to further assess the differential effect of delay on sensory and motor outcomes.

Results: Of 3108 articles meeting search criteria, 22 articles reporting on 445 nerve repairs were ultimately included in the analysis. These included 165 (37%) median, 219 (49%) ulnar, and 61 (14%) radial nerve injuries. Five percent were sensory-only, seven percent were motor-only, and 87% were mixed. On univariate analysis, a 9-month delay to repair was associated with a modest decline in satisfactory outcomes for sensory nerves (96% for <9mo delay versus 83% for >9mo delay) but a precipitous decline for motor nerves (74% for <9mo delay versus 48% for >9mo delay). On multivariate regression, older age, the use of a graft, and greater delay were independently associated with worse outcomes in motor nerves (p<0.05 for each), while only greater delay was associated with worse outcomes in sensory nerves (p<0.05). Subset analysis of mixed nerves in which motor and sensory components were repaired concurrently demonstrated that motor recovery rarely ever outperformed sensory recovery (<1% of the time) while sensory recovery commonly outperformed motor recovery (20% of the time) (p<0.01).

Conclusion: This literature review of individual participant data for upper extremity nerve repair supports our hypothesis that sensory recovery is more tolerant of delay than motor recovery. Consequently, the absence of sensory recovery after repair of a mixed nerve is highly associated with poor motor recovery. Outcomes after delayed repair in this data set were surprisingly good for both motor and sensory nerves, but this may be due to publication bias. Additional studies are necessary to explain the mechanisms underlying these findings.

Introduction: Patient-reported carpal tunnel syndrome (CTS) severity following treatment is difficult to predict using standard analytical methods. Artificial intelligence (AI) and deep learning neural networks commonly outperform other prediction methods. The aim of this study was to design a decision aid predicting personalized patient-reported outcomes following treatment for CTS using AI.

Methods: We built a deep-learning neural network from a recent prospective comparative study. We used a comprehensive dataset that includes baseline data, medical comorbidities, sensory and motor recovery, and patient-reported outcomes. We held out a testing dataset for internally validating the model that was not used during the training procedure. The model compared outcomes following carpal tunnel release surgery to nighttime splinting. The outcome was defined as an improvement in CTS severity beyond the minimal clinically important difference for the validated Boston Carpal Tunnel Questionnaire.

Results: A total of 93 patients were included in the study. The mean age was 56 years (SD=12.6), and 63 of the participants (68%) were female. Fifty-seven patients (61%) underwent an open carpal tunnel release, and 36 patients (39%) used nighttime splinting. Following the validation, the model accurately predicted the outcomes in (15/19) 79% of the cases. The model was off by a maximum of 5% probability in all four mispredictions. The area under the curve for the model was 85%, indicating excellent performance.

Conclusion: AI models can predict personalized patient-reported outcomes in CTS with excellent performance and can be considered a decision aid for CTS patients. The benefits of personalized predictions will be explored qualitatively in future studies.