2022 Abstracts

Introduction: Heart Rate Variability (HRV) is an established measure of stress with higher levels of HRV correlating with lower stress. However, it is unknown whether attending surgeons’ leadership skill impact residents’ stress level. The aim of this study was to assess the relationship between attending surgeons’ leadership skills and residents’ stress levels assessed by HRV.

Methods: This prospective observational study was conducted at a single Midwestern United States academic medical center. Trauma attending surgeons and residents rotating on the Trauma and Critical Care services were enrolled. HRV was recorded with a WHOOP strap worn continuously either on the wrist or the bicep. The highest HRV ever recorded for each subject during the study period was considered the baseline HRV. We used the ratio between the daily HRV value and the maximum HRV ever recorded for each participant to determine the participant’s level of stress. Participants completed a weekly survey that assessed the attending surgeons’ leadership skills. We used multivariable repeated measures gamma regression to assess the relationship between attending leadership skills and resident stress.

Results: Three attending surgeons and sixteen residents were enrolled. There was no significant relationship between attending surgeons’ leadership skills and residents' stress. Results from the repeated measures gamma regression model show a 3.4% increase in HRV (less resident stress) per hour increase in sleep and a 11.6% decrease in HRV (more resident stress) for those working in the ICU compared to those working on Trauma service.

Conclusion: This study revealed that residents Attending surgeons’ leadership skills were not associated with residents’ stress. More sleeping time was associated with less residents’ stress.

Introduction: This paper discusses two cases of ovarian tumor (primary and secondary) in pediatric patients undergoing ovarian tissue cryopreservation (OTC). In general, there has been limited research on how to offer OTC in the setting of primary ovarian tumor or ovarian metastasis, and cryopreserved ovarian tissue may pose some risk of reseeding malignancy. The following case series describes an approach to fertility preservation in ovaries found to contain metastatic disease and/or known primary ovarian cancer at the time of OTC, with a focus on transparency to patients and families regarding the balance between fertility preservation and risk of future malignancy. These are some of the first documented cases of such findings in the history of the Fertility & Hormone Preservation & Restoration Program at Ann & Robert H. Lurie Children’s Hospital of Chicago. The importance of appropriate fertility preservation counseling, OTC, routine pathology, and implications for future fertility of these patients will be discussed.

Methods: A 13-year-old pubertal girl with metastatic ovarian rhabdomyosarcoma and a 4-year-old prepubertal girl with primary ovarian Sertoli-Leydig cell tumor and DICER1 variant were counseled for and underwent OTC. Laparoscopic right oophorectomy was performed in each patient. For the patient with known primary ovarian tumor, the ovary was bisected longitudinally and half was submitted for cryopreservation per institutional protocol. The ovarian tissue was examined by pathology and processed for cryopreservation in our gonadal tissue processing suite.

Results: Germ cell-containing follicles and previously unknown metastatic rhabdomyosarcoma were identified by histological examination of a 4mm punch biopsy taken from the unilateral oophorectomy specimen in the 13-year-old patient. Examination of the bisected ovary specimen revealed no residual tumor nor significant pathological changes reported in the 4-year-old patient. After multidisciplinary discussion of pathology results, both patients opted to store cryopreserved tissue for future use and have not yet requested autologous transplantation.

Conclusion: Our findings highlight the importance of OTC being incorporated alongside standard care for all patients with fertility-threatening diagnoses, including primary ovarian tumor. Furthermore, routine pathology performed alongside oophorectomy may identify ovarian metastasis which will inform future use of ovarian tissue. Multidisciplinary discussion determined sufficient tissue for pathology evaluation and appropriate medical decision making. Our findings support the use of OTC and the flexibility of a multidisciplinary team as part of standard oncologic care for prepubertal and pubertal patients and underscore the importance of histological evaluation of the ovarian tissue prior to final decisions for long-term storage.

Introduction: Poor operative ergonomics may lead to muscle fatigue, musculoskeletal injury, and burnout. Taking microbreaks has shown to decrease perceived discomfort and fatigue in the operating room (OR). We hypothesized that strategically placing posters reminding surgeons to adjust table height and take microbreaks would increase the use of these strategies and decrease physical pain.

Methods: A poster, developed by a physiatrist expert on surgeon work-related injuries, summarizing strategies to improve posture while operating and to encourage microbreaks was placed at the entrance of every OR in a single academic center. An anonymous and voluntary pre- and post-poster survey was administered to attending surgeons inquiring frequency of microbreaks, ergonomic discussions, ergonomic adjustments, and physical pain. The post-survey was administered six months after poster placement.

Results: We surveyed 141 surgeons and received 66 responses for the pre-poster survey and 32 for the post-poster survey. There was no significant difference in the proportion of surgeons who adjusted table height (96.8% vs 96.9%; p-value = 0.9), took microbreaks (35.9% vs 37.6%; p-value = 0.8) or discussed ergonomics more in the OR (62.1% vs 56.3%, p-value = 0.6) before vs after posters were displayed. There was no improvement in physical discomfort before vs after posters (50% vs 45.2%, p-value = 0.7). While 41% agreed that the posters might help, most responders were neutral or unaware the posters had been posted (47%).

Conclusions: The use of posters outside the OR does not appear to be effective in encouraging use of microbreaks and proper posture in the OR. While most surgeons who responded to the survey discussed ergonomics during an operation, real-time feedback, additional training, and incorporation of an ergonomics checklist as part of the “timeout” are potential dynamic interventions that could improve operative ergonomics, fatigue, and burnout among surgeons.

Introduction: One of the primary treatments for microtia is two-stage autogenous auricular reconstruction. The first stage involves a costal cartilage framework inset, and the second stage elevates the framework using fascial flaps, wedge cartilage grafts, and skin grafts. The goal of this study was to modify the conceptual classifications and surgical approach to auricular reconstruction, with a focus on small concha-type microtia.

Methods: A narrative literature review was conducted to identify existing understandings and approaches to microtia and autogenous auricular reconstruction. Findings from the senior author's 20 years of independent practice were then conceptualized to produce updated surgical pearls.

Results: 1) Small concha-type microtia is often overlooked or thought to be a simple indentation in the conchal bowl region. The small concha is often a cavity, instead of an indentation, and located anteriorly to the normal conchal bowl. 2) The proper location of the ear can be determined by constructing an “auricular rectangle,” which utilizes multiple topographic references of the unaffected side in unilateral microtia: top of upper helix, caudal end of lobule, Frankfurt Horizontal line, shape of hairline, and face mask. 3) Surgical technique has evolved from a V- to W-shaped incision along the posterior surface of the auricle to maximize the skin surface area and create the deep concha. The W-shape also creates a superior advancement of the vestige. 4) The hallmark of small concha-type microtia is that small concha is not located in the correct anatomic location, so the skin flap elevated off the indent is too anteriorly positioned to be used for the tragus as much of the literature suggests. 5) Although conventional teachings do not include skin trimming, the senior author finds that horizontal wound closure has been successful in providing uninterrupted blood supply while preventing ischemia.

Conclusions: These modifications advance our understanding of microtia classifications and how the surgical approach can be tailored to best utilize each component of the vestige. The functional outcomes of auricular reconstruction are especially important in light of the coronavirus disease 2019 (COVID-19) pandemic where the ability to wear masks relies on external ear function.

Introduction: While neoadjuvant chemotherapy (NACT) has become standard in localized pancreatic cancer, NACT-induced weight loss is a potential area of concern for patients who undergo surgery. The goal of this study is to determine the effects of weight loss on post-pancreatectomy complications for patients who received NACT.

Methods: The National Surgical Improvement Quality Program database from 2014 to 2019 was utilized to identify patients who were diagnosed with PDAC, received NACT, and were identified to have weight loss >10% of their body weight in the six months prior to operation. Univariate analysis was conducted using Fisher’s Exact Test, Pearson’s Chi-squared Test was used to compare clinical variables, and multivariate analysis was performed using logistic regression.

Results: Of the 5,590 PDAC patients who received NAC, 913 (16%) experienced post-NAC weight loss. From 2014 to 2019, incidence of weight loss among NAC patients remained stable, but incidence of complications among weight loss patients slightly decreased. Those who experienced weight loss were found to have a greater incidence of unplanned intubation (3.8% vs. 2.2%, p = 0.004), ventilator use greater than 48 hours (3.7% vs. 1.8%, p<0.001), and septic shock (3.9% vs. 1.8%, p<0.001). Weight loss did not lead to differences in delayed gastric emptying, pancreatic fistula, and unplanned readmission. In a multivariate logistic regression model, NAC-induced weight loss, history of COPD, and hypertensive medication use independently predicted a postoperative complication.

Conclusions: Weight loss caused by NAC is a significant independent predictor of post-operative complications. Nutritional measures to stabilize weight during the chemotherapeutic regimen should be considered to decrease complications and improve post-operative quality of life.

Introduction: Scars can have significant morbidity and negatively impacts psychological, functional, and cosmetic outcomes as well as the overall quality of life, especially among ethnic minorities. The objective of this study is to evaluate African American and White patients' perception of their scar’s impact on symptoms, appearance, psychosocial health, career, and sexual well-being using validated assessment tools.

Method: A total of 675 abdominoplasty and breast surgery patients from four providers completed the SCAR-Q, and Career/Sexual Well-Being (CS) scales via phone or email. A higher score on both assessments indicates a more positive patient perception.

Results: Of 675 respondents, 77.0% were White, and 23.0% were African American. White patients scored significantly higher on the SCAR-Q (232 ± 79 vs. 203 ± 116), appearance (66 ± 26 vs. 55 ± 29), and CS (16 ± 2 vs. 15 ± 5) scales than African American patients (P<.001, P<.001, P<.001 respectively). There was no significant correlation between duration after surgery and symptoms or appearance scores for African American patients (P=.11, P=.37). There was no significant correlation between patient age and SCAR-Q score or time after surgery and psychosocial scores.

Conclusions: African American patients are more likely to have lower perceptions of their scarring appearance, symptoms, psychosocial impact, career impact, and sexual well-being impact than White patients. Scar appearance and symptoms are less likely to improve over time for African American patients. This study highlights the need to address patient ethnicity when considering further follow-up, counseling, or other measures to enhance scar perception.

Introduction: Critical limb-threatening ischemia (CLTI), a severe form of peripheral artery disease, carries increased risks of limb loss and mortality. This study aims to identify factors associated with higher level reamputation and 1-year mortality after lower extremity amputation for CLTI.

Methods: This is a single-center retrospective review of patients who underwent amputation for CLTI between 2014 and 2017. Patients with acute limb ischemia or ankle disarticulations were excluded. Patient- and limb-level data were collected from the electronic medical record. Outcomes were unplanned higher-level reamputation (UHRA) and 1-year mortality after index amputation. We report unadjusted bivariate associations and adjusted odds ratios (AOR) from logistic regression models of the association between each preamputation risk factor and each outcome controlling for age, race, sex, and amputation type.

Results: During the study period, 183 patients underwent 203 amputations (median age 65 years [IQR 57, 75]; 70.9% males). Amputations included 118 toe (58.1%), 20 transmetatarsal (TMA; 9.9%), 37 below-knee (BKA; 18.2%), and 29 (14.3%) amputations at or above the knee. Indications for amputation included gangrene (n=82, 40.4%), infection (n=88; 43.4%), ulceration (n=25; 12.3%), and rest pain (n=8; 3.9%). Median follow-up was 302 days [IQR 62, 1348]. Thirty-six (17.7%) limbs had an UHRA, most occurring after toe amputation (n=26; 22.0%) or TMA (n=6; 30.0%). Revascularization prior to or concomitant with amputation was associated with increased UHRA (25.3% versus 12.5%, p=0.03). Risk factors associated with increased odds of UHRA included non-ambulatory status (AOR 6.74, CI 1.74-26.18; p<0.01), toe pressure < 30 mm Hg (AOR 4.90, CI 1.52-15.78; p<0.01), and a monophasic or absent ankle waveform (AOR 3.12, CI 1.30-7.46; p=0.01). All-cause 1-year mortality was 17.2% (n=32 patients). Risk factors for 1-year mortality included CAD (AOR 3.93, CI 1.56 – 9.87; p<0.01), CHF (AOR 4.90, CI 1.96 – 12.29; p=0.001), ESRD (AOR 7.54, CI 3.10 – 18.34; p<0.001), and non-independent ambulatory status (AOR 4.31, CI 1.20 – 15.49; p=0.03). Males had lower odds of 1-year mortality compared to females (AOR 0.37, CI 0.15 – 0.89; p=0.03). UHRA was not associated with 1-year mortality.

Conclusion: Determining the appropriate level of amputation for patients with CLTI can be difficult. Rates of UHRA after toe amputations and TMA are high. Patients with CLTI requiring amputation, regardless of subsequent UHRA, are at high risk of 1-year mortality. Larger multi-center datasets and hierarchical modeling of surgeon, patient, and limb-level factors may reveal further insights.

Introduction: Assigned sex at birth, sexual orientation, and gender identity (ASAB/SOGI) have been routinely excluded from demographic data collection tools, including in electronic medical record (EMR) systems. We assess the ability of adding structured ASAB/SOGI data capture to improve identification of transgender and nonbinary (TGNB) patients compared to using only International Classification of Diseases (ICD) codes and text mining and comment on the ethics of these cohort formation methods. 

Methods: We conducted a retrospective chart review to classify patient gender at a single institution using ICD-10 codes, structured ASAB/SOGI data, and text mining for patients presenting for care between March 2019 and February 2021. We report each method’s overall and segmental positive predictive value (PPV).

Results: We queried 1,530,154 EMR records from our institution. Overall, 154,712 contained relevant ICD-10 diagnosis codes, ASAB/SOGI data fields, or text mining terms; 2,964 were manually reviewed. This multi-pronged approach identified a 1,685 TGNB patient cohort. The initial PPV was 56.8%, with ICD-10 codes, ASAB/SOGI data, and text mining having PPV of 99.2%, 47.9%, and 62.2%, respectively. 

Conclusions: This is one of the first studies to use a combination of structured data capture with keyword terms and ICD codes to identify TGNB patients. Our approach revealed that though structured ASAB/SOGI documentation was less than 10% in our health system, 1,343/1,685 (79.7%) of TGNB patients were identified using this method. We recommend that health systems promote patient EMR documentation of ASAB/SOGI to improve health and wellness among TGNB populations, while centering patient privacy.

Introduction: We aimed to improve accuracy of administrative data for perforated appendicitis classification. Summary Background Data: Reliance on International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) diagnosis codes may misclassify perforated appendicitis with resultant research, fiscal, and public health implications.

Methods: We conducted a retrospective study of randomly sampled patients aged ≤ 18 years diagnosed with acute appendicitis from 8 children’s hospitals. Patients were identified using the Pediatric Health Information System (PHIS), and true perforation status was determined by medical record review. We used PHIS data elements to develop two algorithms using machine learning (ML) approaches through a 5-fold cross-validation. ML algorithm performance was compared against algorithms that exclusively relied on ICD-10-CM codes using area under the curve (AUC) and other measures.

Results: Of 1037 included clinically validated encounters, 377 (36.3%) patients were identified to have perforated appendicitis. The two algorithms developed using ML approaches primarily leveraged ICD-10-CM codes and length of stay as predictors. ML developed algorithms had a significantly higher accuracy than algorithms relying exclusively on ICD-10-CM (p-value<0.01): the AUC were 0.92 and 0.91 for both ML-developed algorithms, while the AUC were 0.80 and 0.78 for both ICD-10-CM algorithms.

Conclusions: This study provides an algorithm that can improve the accuracy of perforated appendicitis classification using commonly available elements in administrative data. We recommend that this algorithm is used in future appendicitis classification to ensure valid reporting, hospital-level benchmarking, and fiscal or public health assessments.

Introduction: Cirrhosis of the liver affects over 2 million adult Americans and is implicated in the death of more than 50,000 patients each year; however, accurate identification of cirrhosis in electronic health records (EHR), currently done through International Classification of Disease (ICD) codes, has limited success. The objective of this study was to leverage machine-learning algorithms with natural-language processing (NLP) to identify patients with cirrhosis in radiology report text.

Methods: We trained a neural network model (word2vec) to learn word associations from radiology reports from Northwestern Medicine’s Enterprise Data Warehouse were used to train the model. From the word2vec algorithm, synonyms for key organ-findings were generated based upon the cosine-similarity of each word-vector and validated by physicians (‘cirrhotic’, ‘hepatocellular’, ‘ascites’, ‘portal hypertension’, ‘hepatomegaly’ etc.). Comparisons of the word-similarity based upon of the word-embedding vectors was evaluated using the t-distributed stochastic neighbor embedding statistic. Leveraging validated word2vec embeddings (computer-generated word meanings), deep-learning (xgboost) was utilized for detection of cirrhosis within radiology report text. Data was split into a train and test set (70:30). Highly specific cirrhosis ICD-codes were used as positive control (> 98% specificity). Non-cirrhosis radiology reports were used as the negative outcome. Chart-review of 1,100 total radiology reports is being performed by clinicians for the final validation set.

Results: Utilizing t-distributed stochastic neighbor embedding, the cluster for cirrhosis-related terms (“cirrhotic”, “hepatitis” …) was significantly different from terms pertaining to other organs (p<0.001). A balanced dataset of 6,332 radiology reports (3,166 cirrhosis-positive, 50%) were used for the deep learning model. After learning on the training-set, the deep learning model had a Sensitivity = 81%, Specificity = 91%, with the area under the receiver operating characteristic curve (ROC) = 0.891 on the 30% test set. Cirrhosis Classification using only ICD Codes had a Sensitivity = 97.9%, Specificity = 43.0%, ROC = 0.71. When comparing the ROC curves, the Deep learning model significantly outperformed the ICD-classification with an 0.18 greater area under the ROC curve (P<0.001). The positive predictive value (PPV) with deep learning was 96.4%, compared to 78% using only ICD codes. The difference between PPV (deep learning) and (ICD) was 18.4%.

Conclusions: A deep learning model leveraging radiology report text better detects cirrhosis patients than cirrhosis ICD-codes in an EHR dataset. This is impactful for identifying patients in need for guideline-adherent care and risk prediction.

Introduction: Minimally invasive surgery, including robotic-assisted (RATS) and video-assisted thoracoscopic surgery (VATS), has become increasingly common for treatment of non-small-cell lung cancer (NSCLC). However, it is unknown how geographic disparities impact outcomes. The objective of this study is to evaluate how trends in geographic disparities influence access and perioperative outcomes for patients undergoing MIS for NSCLC.

Methods: Patients diagnosed between 2010-2018 with stage 0-IIIA (N0-N1) resected NSCLC were identified using the National Cancer Database (NCDB). Patients were included if they had a single malignancy, intact geographic data, and underwent surgery at the reporting facility. Rural-urban trends in use of MIS and differences in perioperative outcomes were evaluated. Odds of surgical approach were determined while adjusting for patient, clinical, hospital, and geographic characteristics.

Results: Overall, 127,865 patients met inclusion criteria with 11.9% classified as rural (n=15,214). Patients were categorized into those who underwent RATS (n=16,790, 8.5% rural), VATS (n=42,645, 10.3% rural), and thoracotomy (OT) (n=68,428, 13.8% rural). For rural residents, the median travel distance to treatment with MIS was 50.2 (IQR 31.6-77.2) miles while nonrural residents traveled a median of 10.4 (IQR 4.9-21.7) miles (p<0.001). Use of MIS increased from 19.3% (2010) to 58.0% (2018) for rural residents and 26.5% (2010) to 66.7% (2018) for nonrural residents (all P<0.001). OT remains a more common operative approach for rural residents (42.0%; 2018) compared with nonrural residents (33.3%; 2018) as a proportion of all surgical approaches (all P<0.001). Rural residents had decreased odds of receiving RATS (aOR 0.66, 95% CI 0.54-0.82 ). Overall, MIS had decreased 30-day (1.2% RATS vs 1.5% VATS vs 2.4% OT) and 90-day mortality (2.1% RATS vs 2.6% VATS vs 4.3% OT) (Table 1; all P<0.001). The number of hospitals providing MIS to rural lung cancer populations increased from 256 (2010) to 370 (2018) (p<0.01). The number of hospitals providing MIS to nonrural populations increased from 735 (2010) to 828 (2018) (p<0.01).

Conclusions: There are persistent geographic disparities in access to MIS for treatment of NSCLC. Rural residence is associated with decreased odds of MIS, and MIS is associated with improved perioperative outcomes.

Introduction: We describe a reliable approach for alternative double nerve transfer of the medial triceps nerve branch and thoracodorsal nerve to the axillary nerve to increase axonal input. We present a review of these patients compared to other axillary nerve transfers performed, as well as outcomes for end-to-end compared to reverse end-to-side nerve transfer.

Methods: After Institutional Review Board approval, a retrospective review of patients who underwent nerve transfer surgery for improvement of shoulder abduction at Harborview Medical Center and Northwestern Memorial Hospital between 2012 and 2021 was conducted. Subsequently, patients were prospectively contacted to fill out a DASH survey, with an option to upload a video of their active range of motion.

Results: Twenty-seven patients were included in the final analysis. Eighteen patients completed the prospective arm of the study (66% response rate). Seventy-nine percent of patients (15/19) who underwent the alternative double nerve transfer achieved an MRC of 4 or greater, and mean measured AROM was 140.25° and 136.91° for shoulder abduction and flexion, respectively. Comparing end-to-end vs. reverse end-to-side neurorrhaphy, outcomes including follow-up, mean post-operative MRC, mean change in MRC, DASH, abduction AROM, and flexion AROM were not statistically different.

Conclusions: We showed improvements in shoulder abduction by utilizing the thoracodorsal nerve, in addition to the medial triceps branch, to increase the axonal donation to power the axillary nerve, without the sacrifice of the spinal accessory nerve. Further, we demonstrated substantial improvements with reverse end-to-side coaptation when intra-operative stimulation of the axillary nerve revealed some residual function.

Introduction: Resident burnout is increasingly recognized as a symptom of an unsupportive workplace. The objectives of this study were to describe resident perceptions of program responsiveness and to identify associated factors.

Methods: We used a convergent mixed-methods design. A cross-sectional survey was administered to all U.S. general surgery residents following the 2020 ABSITE, querying perceptions of their learning environment, burnout, thoughts of attrition and suicide, and career satisfaction. Multivariable logistic regression models adjusting for program/resident characteristics assessed associations of program responsiveness with aspects of the learning environment and resident wellness. 366 interviews and 27 focus groups with residents and faculty were conducted during in-person visits to 15 residency programs. Transcripts were analyzed thematically using inductive and deductive logics until thematic saturation was achieved.

Results: Of the 7233 clinically active residents from 323 programs who completed the survey (85.5% response rate), 5256 had data for all outcomes of interest. 72.1% (n=3791) reported satisfaction with program responsiveness. These residents were significantly less likely to report 80-hour workweek violations (OR 0.22, 95%CI 0.18-0.26), burnout (OR 0.47, 95%CI 0.41-0.53), thoughts of attrition (OR 0.32, 95%CI 0.27-0.38), and suicidality (OR 0.52, 95%CI 0.41-0.65). They were significantly more likely to report satisfaction with career choice, personal life, and work-life balance (all p<0.001). Factors associated with improved perception of program responsiveness included larger program size (50+vs<23 residents; OR 1.48, 95%CI 1.01-2.19), having faculty mentorship (OR 2.64, 95%CI 2.22-3.14), having meaningful input into schedules (OR 3.31, 95%CI 2.74-4.00), and feeling comfortable speaking up (OR 4.20, 95%CI 3.47-5.09). Qualitative analysis identified the following components of program responsiveness: (1)core values reflecting a shared understanding of the importance of resident voice in shaping the training experience and the program, including mutual trust and respect between residents and faculty, transparency and communication, resident unity, and resident participation in and ownership of program improvement; (2)structural constructs that reflect and support responsiveness; (3)mechanisms for supporting resident agency, including resources and leadership support, faculty advocacy, and resident leadership opportunities.

Conclusion: Program responsiveness in surgical residency is associated with improved resident wellness. Programs should develop formal channels to elicit and concretely address resident concerns, provide opportunities for resident representation, and entrust residents with the flexibility and autonomy to make decisions that support their education and wellness.

Introduction: Regionalization of surgery for non-small-cell lung cancer (NSCLC) to high-volume centers (HVCs) improves perioperative outcomes but increases travel distance for patients who receive care at these centers. Increased distance may decrease rates of adjuvant chemotherapy (AC). However, the relationship of distance, volume, and receipt of AC with outcomes is unknown. The purpose of this study is to evaluate the association of distance, volume, and receipt of AC with overall survival among patients with NSCLC.

Methods: Patients with stage II-IIIA (N0-N1) NSCLC were identified between 2004-2018 using the National Cancer Database. Patient travel distance to their surgical facility was categorized into quartiles (<6.4, 6.4 to <14.7, 14.7 to <35.1, and ≥35.1 miles), and HVCs were defined in accordance with LeapFrog criteria as those performing ≥40 annual resections. Patient characteristics and odds of receiving AC at any center were determined. Survival analysis was performed using Kaplan-Meier curves and adjusted Cox Proportional Hazards models.

Results: Overall, 48,226 patients with surgically resected stage II-IIIA (N0-N1) NSCLC met criteria for inclusion. Of the cohort, 51.7% received AC, 16.7% traveled <6.4mi to LVCs, and 15.2% traveled ≥35.1mi to HVCs (p<0.001). Among stage II-IIIA patients who traveled ≥35.1mi to HVCs, 46.0% received AC vs 54.1% who traveled <6.4mi to LVCs (aOR 0.66, 95% CI 0.58-0.74; p<0.001; reference). Patients with Stage II-IIIA NSCLC who traveled ≥35.1mi to HVCs and did not receive AC had higher mortality than those who traveled <6.4mi to LVCs and received AC (aHR for mortality 1.31, 95% CI 1.21-1.42).

Conclusions: Longer travel distance is associated with decreased odds of receiving adjuvant chemotherapy. Furthermore, patients with stage II-IIIA (N0-N1) NSCLC who traveled ≥35.1mi to high-volume centers for surgery and did not receive adjuvant chemotherapy had lower overall survival compared to patients who traveled <6.4mi to low-volume centers for surgery but received adjuvant chemotherapy. Understanding the reason for lack of receipt of adjuvant chemotherapy is necessary to improve delivery and maximize the benefit of travel to high-volume centers for surgery.

Introduction: While much is known about risk factors for physician suicide, it is unknown which, if any, are generalizable across specialties. Specialty specific culture may influence a physician’s likelihood of distress. This study aims to examine individual and program level factors that are associated with suicidality in general surgical residents.

Methods: A voluntary cross-sectional survey was administered after the 2019 ABSITE. Residents were asked about their perceptions of the learning environment (including mistreatment), burnout symptoms (via a modified abbreviated Maslach Burnout Inventory), and whether they had considered suicide in the past year. Associations between suicidality and resident and/or program characteristics were assessed using multivariable logistic regression.

Results: Overall, 6,956 residents (85.6%) from 301 surgical programs responded to the survey. Of the 6,567 residents who completed the suicidality question, 289 (4.1%) answered that they had considered suicide in the previous year. Significantly increased odds of suicidality were found in residents who were single (OR 1.57, 95% CI 1.14-2.16) and/or identified as LGBTQ (OR 1.68, 95% CI 1.02-2.78). Suicidality was significantly associated with negative responses regarding workload and job demands (e.g, more 80 hour violations; OR 1.48, 95% CI 1.07-2.06), organizational culture (e.g., greater programmatic emphasis on blame after adverse events; OR 1.65, 95% CI 1.16-2.36), meaning in work (e.g., less operative autonomy; OR 1.58, 95% CI 1.10-2.27), and exposure to sexual harassment (OR 2.18, 95% CI 1.59-2.98) and bullying (OR 2.34, 95% CI 1.49-3.68). Multiple factors were not associated with suicidality including post-graduate year, gender, and race/ethnicity.

Conclusion: Although suicidality is often considered and intervened upon as an individual-level issue (e.g., mental health), it is associated with multiple modifiable aspects in the learning environment. Continued work on acknowledging and augmenting the role of educators in creating safe training spaces that support residents in learning and recovering from error, as well as in gaining autonomy, may be important to reduce resident suicidality and prevent loss of life.

Introduction: Femoral artery access (FAA) is a fundamental clinical skill across multiple specialties. Simulation-based mastery learning (SBML) is a rigorous form of competency-based education that improves procedural skills and clinical outcomes. The objective of our study was to assess the impact of a FAA SBML curriculum on procedural competency among cardiology fellows (CFs) and vascular surgery trainees (VSTs).

Methods: First-year CFs (Postgraduate year 4) and all VSTs (postgraduate years 1-7) at a single institution were eligible for the curriculum from June 2020-December 2021. The SBML curriculum consisted of a pretest, video didactics, deliberate practice with feedback, and a post-test. FAA was performed on a novel simulator that allows for ultrasound guidance and simulated fluoroscopy. Learners were evaluated by trained faculty using a 17-item skills checklist and a minimum passing score (MPS) of 96%. The checklist was created and MPS was set by a multidisciplinary panel of experts. Learners who did not achieve the MPS at post-test received additional training and were retested until the MPS was met (final post-test). Each element of the checklist was assigned a potential complication of introduction of air or vascular injury if done incorrectly. Checklist scores and number of potential complications were compared from pretest to final post-test among trainees.

Results: Nineteen CFs and eleven VSTs completed the curriculum. Most trainees (76.9%) had clinical experience with FAA. All learners eventually achieved the MPS. The median checklist score improved from pretest (13, interquartile range [IQR], 5-16) to final post-test (17, IQR 16-17, p=0.002) for and for VSTs from 13 (IQR, 10-16) to 17 (IQR, 17-17; P = .01). The number of potential air and vascular complications significantly decreased from pretest to final post-test for CFs and VSTs. Trainees with no prior FAA clinical experience (N=6) had lower pre-test scores and higher possible complications compared to those with >5 prior FAA clinical experiences (N=8, p=0.002). There were no significant differences in post-test scores between those with no prior FAA experience versus >5 FAA clinical experiences.

Conclusions: Trainees from both disciplines displayed significantly improved FAA skills after completing SBML. There was a significant difference in test scores and complications compared to those with no prior FAA experience to those with the most FAA experience with significant improvement in test scores and complications after completing SBML. Further studies are needed to assess skill transfer to patient care and the impact of this training on procedural complications.

Introduction: Significant racial and ethnic disparities in mortality and access to liver transplant exist among patients with cirrhosis in the greater Chicago metropolitan area. Increased community vulnerability has been associated with poor health care outcomes, but it has not been studied in cirrhosis. We hypothesized that community social vulnerability independently predicts cirrhosis outcomes.

Methods: A cohort of adult (>18) patients with cirrhosis was procured from HealthLNK (2006-2012), a large multi-center population-based electronic health record cohort from six large health systems in the greater Chicago metropolitan area. The data were merged with the United Network for Organ Sharing data, and Illinois Department of Public Health death data. Cause of death was categorized by blinded clinician death certificate review. Demographics, ICD, CPT codes, and medication data were collected. Community social vulnerability was calculated for 5-digit ZIP-codes using the CDC’s Social Vulnerability Index (SVI), a composite index score for community social determinants of health. Adjusted competing risks of all-cause mortality, transplant, liver-related death, and non-liver related death were estimated by Fine-Gray sub-distribution hazard model. The net effect of Social Vulnerability (SVI) on each outcome was calculated by comparing a propensity-matched patient in the most vulnerable ZIP-code (high SVI) to an identical patient in the least vulnerable ZIP-code (low SVI).

Results: In this cohort of 19,906 patients with cirrhosis, mean age was 57.1 yrs, mean follow-up 2.6 yrs, and 8,475(43%) were Female. 9,038(45%) were White, 4,399(22%) Black, 3,258(16%) Hispanic, 520(3%) Asian, and 2,691(14%) Other. ZIP-codes with high social vulnerability (SVI) had more Black (46%) and Hispanic (29%) patients, more patients with HCV (48%) and alcoholic cirrhosis (44%) (univariate; p<0.01). ZIP-codes with low SVI had predominantly patients who were White (72%), had private insurance (34%), NASH cirrhosis, and HCC (p<0.01). Increased SVI was associated with a 30% increase in all-cause mortality (SHR=1.3, 95%CI=1.2-1.55, p<.001) and an 83% decreased chance to receive a liver transplant (SHR=0.17, 95%CI=0.13-0.21, p<0.001). Increased SVI was associated with a 39% increase in non-liver related death (SHR=1.39, 95%CI=1.14-1.7, p<.001) and was not significantly associated with liver-related death (SHR=0.96, 95%CI=0.85-1.11, p=0.52). Ongoing subgroup analysis suggests these effects vary significantly for specific patient populations, such as Black or Hispanic patients.

Conclusion: For patients with cirrhosis, high Social Vulnerability (SVI) is associated with higher all-cause mortality, non-liver related mortality, and less liver transplantation.

Introduction: Mass Casualty Incidents (MCI’s) have devastating consequences throughout the U.S., highlighting a failure of primary and secondary triage policy in urban healthcare systems. It is imperative to provide optimal care for trauma victims in the shortest amount of time while avoiding burden on Level 1 trauma centers. Reinforcement learning (RL) can leverage patient and hospital data to program a computational agent acting as incidence control. By providing the agent with rewards and goals in a simulated emergency environment, it is possible to generate optimal policies for various MCIs to assess and develop a standardized, nationally accepted trauma triage practice.

Methods: A discrete event simulation (DES) was developed to model an urban, inclusive trauma system environment complete with simulated trauma patients from a pre-defined MCI using fuzzy logic. The MCIs modeled are the 9/11 terrorist attack, Hurricane Katrina, the Pulse Nightclub shooting, the Station Nightclub fire, and the Continental Airlines Flight 1404 aviation accident. Patient parameters were simulated based on Monte Carlo methods for preliminary results using 2002-2016 data from NTDB; these were subject to feature engineering for patient labeling, survivability and salvageability scores. A baseline agent traverses the DES following a Markov Decision Process (MDP) and queueing theory. Pandas (python) is used to translate the MDP to Q-learning to track action-state values and subsequent optimal policy generation as the RL agent completes multiple simulated episodes of the DES. The complete triage algorithm for the RL agent is programmed based on considerations and recommendations from the 2019 ASPR TRACIE “Mass Casualty Trauma Triage Paradigms and Pitfalls” and the CDC Field Triage Guidelines.

Results: Simple Q-learning demonstrates that an agent can generate an optimal policy of triaging red labeled patients to high-level trauma centers first, then yellows to low level trauma centers while avoiding burden on high-level centers, and finally triaging green labeled patients. The agent reliably tracks fully treated patients and minimizes under- and over-triage rates to near zero in about 200 training episodes with an epsilon-greedy strategy.

Conclusions: The results of the simulation optimizations may provide evidence-based prioritization processes for U.S. trauma centers. The goal is to preserve as many lives as possible by optimizing for injury-specific allocation given the critical seconds that distinguish trauma case survival rates from others. As real-time case reporting develops, the agent could better balance large patient loads with real-time knowledge of bed capacity and available resources.

Introduction: While children’s hospitals provide a significant proportion of highly specialized pediatric surgical care in the United States, the value of children’s hospitals compared to non-children’s hospitals for routine surgical procedures is unknown. Our objective was to evaluate the value of children’s hospitals for routine surgical procedures by assessing clinical outcomes and payments data.

Methods: This retrospective cohort study examined outcomes and payments for 13 commonly performed pediatric surgical procedures from 2010 to 2015 using administrative data from the Health Care Cost Institute (HCCI). The primary exposure was tier of children’s hospital status, defined using self-reported pediatric services, affiliation with pediatric focused programs, and validated based on proportion of pediatric admissions.

Results: There were 368,220 pediatric patients who underwent one of the procedures of interest. 220,899 (60.0%) were male. 118,977 (32.2%) had their procedure at a freestanding children’s hospital (CH-A), 75,256 (20.4%) at a children’s hospital attached to an adult hospital (CH-B), and 173,987 (47.3%) at a non-children’s hospital (NCH). The average payment for all procedures at CH-A was $6,533.56, at CH-B $5,847.50, and at NCH $5,034.25. The overall complication rate was 0.004±0.06 at CH-A, 0.01±0.07 at CH-B, and 0.003±0.06 at NCH. Readmission rates at 30, 60, and 90 days were the same at all hospitals. After adjusting for complex chronic conditions, zip code, year, surgery, surgery setting, and observable patient, hospital, and county characteristics, we estimate that payments for inpatient common procedures were 39% higher at CH-A than NCH (p<0.001) and 2% lower at CH-B than NCH (p=0.93). Payments for outpatient common procedures were 34% higher at CH-A than at NCH (p<0.001) and 3% higher at CH-B than NCH (p=0.42).

Conclusion: For children undergoing lower acuity, commonly performed surgical procedures, clinical outcomes are equivalent at children’s hospitals and non-children’s hospitals, but the procedures were associated with higher payments and, thus, overall lower value care. To ensure delivery of optimal value to patients and payers, further research is needed to evaluate mechanisms to ensure access, decrease costs, and improve value at both children’s and non-children’s hospitals.

Introduction: Patient education programs improve wellness and decrease healthcare resource utilization for patients undergoing intestinal stoma creation. We sought to characterize the impact of patient factors on a skills-based perioperative stoma education program and effects on postoperative ostomate wellbeing, stoma-related problems, and healthcare resource utilization.

Methods: Ostomates participating in a national stoma educational program between 2007-2017 retrospectively completed assessments exploring demographics, stoma care confidence, postoperative healthcare resource utilization and stoma-related problems in the early postoperative period. Composite confidence scores (CCS) were created to characterize confidence in performing various stoma care tasks. Composite problem scores (CPS) were created to describe the frequency of various stoma-related problems. Compiled responses were examined to determine associations with CCS, CPS, and postoperative healthcare resource utilization.

Results: 1178 patients responded (colostomy=558, ileostomy=399, urostomy=202, unknown=19) of which 74.8% were preoperatively sited for a stoma. Ostomates were least confident troubleshooting a leaking pouch (74%=”no confidence/experience”) and had the most problems applying a new pouch (75%=”always/often” problematic). Overall, 32% of patients were uncomfortable leaving home, 31% worried about self-care, and 24% felt sad/depressed. Postoperatively, 77% of ostomates utilized home nursing, 61% required surgeon clinic visits , and 18% required ED visits. High CCS patients were younger (63y vs. 69y, OR=0.97, 95%CI:0.95-0.98, p<0.001), had lower CPS (12.5 vs. 19, OR=0.92, 95%CI:0.89-0.95, p< 0.001), less frequently required nursing visits (70% vs. 83%, OR=0.87, 95%CI:0.8-0.95, p=0.002), and were less often female (44% vs. 56%, OR=0.6, 95%CI:0.39-0.96, p=0.034). High CPS patients were more likely to have lower CCS (9.8 vs. 13.8, OR=0.84, 95%CI:0.8-0.9, p<0.001), require home nursing visits (82% vs. 72%, OR=1.1, 95%CI:1.01-1.2, p=0.031) and require stoma nurse telephone calls (61% vs. 46%, OR=1.2, 95%CI:1.02-1.46, p=0.041). Patients requiring physician rescue visits had higher mean CPS (16.3 vs. 14.0, OR=1.03, 95%CI:1.00-1.05 p=0.027) and more frequently required nursing visits.

Conclusion: Despite educational efforts, stoma creation profoundly impacts patient-reported measures of wellbeing and healthcare resource utilization in this large sample of ostomates. High confidence with stoma care was associated with decreases in stoma-related problems and resource utilization. Efforts focusing on improving patients’ stoma care confidence may improve postoperative wellbeing, decrease problems and resource utilization. However, women and older patients may require focused stoma education strategies.

Introduction: Trauma systems coordinate timely triage and care from the moment of injury. However, fewer than half of U.S. states allocate stable funding towards trauma systems and care. This study examines the association between state trauma funding and mortality among critically injured patients.

Methods: Patient encounters with an injury diagnosis and Injury Severity Score (ISS) >15 were extracted from 2016 and 2017 Healthcare Cost and Utilization Project (HCUP) State Emergency Department Databases (SEDD) and State Inpatient Databases (SID) from five states (FL, MA, MD, NY, WI). These states capture a broad range of geographic and demographic variability and allow for patient linkage across encounters. These data were merged with the American Hospital Association (AHA) Survey and publicly available state trauma funding data obtained from each state’s Department of Public Health. Patients were linked across emergency and inpatient encounters to determine triage status, defined as appropriate triage (admitted to a Level I or II trauma center (TC)), under-triage (admitted to a Level III, IV, or non-TC), or re-triage (emergently transferred from an emergency department to a Level I or II TC). We tested the association between state trauma funding and mortality using a hierarchical logistic regression, controlling for triage status with a trauma funding by triage status interaction term and adjusting for age, sex, race, primary payer, Elixhauser comorbidity score, and ISS.

Results: A total of 242,299 patients with ISS >15 met inclusion criteria. Median age was 52 years (IQR=28-73). Median ISS was 17 (IQR=16-25). Two states (MA, NY) allocated $0.00 per capita trauma funding, and three states (WI, FL, MD) allocated between $0.09-$1.80 per capita trauma funding. Compared to patients in states with no trauma funding, patients in states with trauma funding experienced decreased adjusted odds of mortality (OR=0.75 [0.60-0.93]). Funding was associated with decreased adjusted odds of mortality among all triage statuses, with the lowest adjusted odds of mortality among re-triaged patients (OR=0.63 [0.46-0.87]).

Conclusions: State trauma funding is associated with lower adjusted mortality among critically injured patients, especially among those who are re-triaged. Allocating state funding towards trauma systems and care may lower adjusted mortality in many states.

Introduction: Upfront surgery is potentially curative treatment for extrahepatic cholangiocarcinoma (eCCA), though survival is poor. Evidence regarding the role of neoadjuvant therapy (NAT), including chemotherapy or chemoradiation (CRT) is limited. The objectives of this study were to (1) characterize treatment trends, (2) identify factors associated with use of NAT, and (3) evaluate the association between NAT and postoperative outcomes and overall survival.

Methods: This was a retrospective cohort study of patients who underwent surgery for eCCA identified in the National Cancer Database (2004-2017). The proportion of patients receiving NAT was assessed over the study period. Factors associated with the use of NAT were evaluated with cluster-adjusted multivariable logistic regression. The associations between NAT and postoperative outcomes, including margin-negative (R0) resection, 30-day readmission, 30-day and 90-day mortality, and overall survival (OS), were assessed with propensity-score adjusted multivariable logistic regression and cox proportional hazard models. Primary analysis compared surgery first to NAT, and stratified analysis assessed differences between surgery first, neoadjuvant chemotherapy, and neoadjuvant CRT groups.

Results: Between 2004-2017, 8040 patients (median [IQR] age 67 [59-75]; 2957 women [36.8%]) underwent surgery for eCCA. Of these, 417 (5.2%) received NAT (chemotherapy: 215 (51.6%), CRT: 202 (48.4%)). NAT increased during the study period 2.9% to 8.4% (p<0.001). Factors associated with the use of NAT included age <50 (vs >75, OR 4.32, 95% CI 2.55-7.29) and stage 3 disease (vs 1, OR 1.68, 95% CI 1.12-2.52). Compared with surgery first, patients who received NAT had higher odds of R0 resection (OR 1.49, 95% CI 1.10-2.02), lower 30-day mortality (OR 0.51, 95% CI 0.27-0.97) and 90-day mortality (OR 0.58, 95% CI 0.35-0.97), and improved median survival (35.1 vs 25.3 months, logrank <0.001). On stratified analysis, neoadjuvant chemotherapy was not associated with differences in any outcomes. However, neoadjuvant CRT was associated with improvement in R0 resection (OR 3.52, 95% CI 2.11-5.86), 90-day mortality (OR 0.33, 95% CI 0.13-0.81), and median survival (47.8 vs 25.3 months, logrank <0.001) compared to surgery first.

Conclusions: The use of NAT remains uncommon. Compared to surgery first, NAT, particularly neoadjuvant CRT, was associated with improved postoperative outcomes and OS. These observational data suggest expanding the use of neoadjuvant CRT for eCCA.

Introduction: Same-day discharge (SDD) following appendectomy for acute, uncomplicated appendicitis in children improves healthcare delivery efficiency and has been shown to be safe without increasing complications or readmissions. As SDD following appendectomy increases, utilization among diverse populations has not been explored to evaluate equitable care delivery. Our objective was to determine whether race and ethnicity are associated with rates of SDD and post-discharge healthcare utilization. We hypothesized that racial and ethnic minority groups would have lower rates of SDD.

Methods: This retrospective cohort study used data from the 2015-2019 American College of Surgeons National Surgical Quality Improvement Program-Pediatric (NSQIP-Pediatric) clinical registry and included children who underwent appendectomy. Patients with complicated appendicitis were excluded. Primary exposure was race (American Indian/Alaskan Native, Asian, Black/African American, Native Hawaiian/Pacific Islander, White) and ethnicity (Hispanic, non-Hispanic). The primary outcome was SDD, and secondary outcomes included post-discharge emergency department (ED) visits and hospital readmissions.

esults: Of 37,579 simple appendicitis patients, SDD after appendectomy occurred in 10,012 (26.6%). White patients had a higher rate of SDD than other racial groups (n=7,670; 27.1%; p<.0001). Hispanic patients had a higher rate of SDD (n=2,851; 29.4%) compared to non-Hispanics (n=6,627, 26.5%; p<.0001). On multivariable analysis, Black or African American (Black/AA) race was associated with lower likelihood of SDD (adjusted odds ratio [aOR]: 0.85; 95% confidence interval [95%CI]:0.79-0.92; p<.0001). Hispanic ethnicity was associated with higher likelihood of SDD (aOR: 1.19; 95%CI:1.12-1.25; p<.0001). Likelihood of postoperative ED visits was higher in Black/AA patients (aOR: 1.36; 95% CI:1.14-1.62; p<.001) and Hispanic patients (aOR: 1.37; 95% CI:1.12-1.58; p<.0001). Hospital readmission rates were similar across groups. When race and ethnicity were combined, Hispanic White patients had the highest overall rate of SDD (n=2,073; 30.5%; p<.0001).

Conclusions: Rates of SDD following appendectomy vary among racial and ethnic groups. Postoperative ED visits are higher among racial and ethnic minorities, but hospital readmission rates are similar. Interventions to achieve equitable healthcare delivery for care including SDD after appendectomy are needed.

Introduction: Morton’s interdigital neuroma is a common cause of severe foot pain that is detrimental to a patient’s quality of life. Operative interventions for this condition include neurolysis, decompression, or neurectomy, which may result in persistent post-operative pain and true neuroma formation, with failure rates reported between 15 to 49%. Targeted muscle reinnervation (TMR) is a surgical technique that has demonstrated efficacy in the treatment of neuroma pain. The objective of this study was to evaluate the efficacy of TMR for the treatment of Morton’s neuromas that have failed previous surgeries.

Methods: A retrospective chart review was performed between March 2018 and April 2021. Eight patients underwent TMR of the medial plantar nerve (MPN), lateral plantar nerve (LPN), and/or sensory plantar common digital nerves at Northwestern Memorial Hospital. All patients were non-amputees who had previously undergone surgery for Morton’s neuroma and now had recurrent symptoms. Operative and post-operative outpatient notes were reviewed. The anatomy of the relevant nerves in the foot, as well as the operative techniques employed for TMR of neuromas in the foot, were also described.

Results: Of the 8 patients reviewed receiving TMR for neuroma pain in the foot, all of whom had previously undergone excisions of digital nerve neuromas: 7 patients had complete or near complete resolution of their pain and symptoms; 1 patient had no improvement upon follow-up. The mean follow-up period in this study was 395 days.

Discussion: TMR has emerged as a successful technique for the management of neuroma pain. In this study, we have outlined the relevant neural anatomy and described the surgical technique for the use of TMR for the management of recurrent Morton’s neuromas in patients where past surgical interventions have failed. The use of TMR in treating recurrent Morton’s neuromas revealed good clinical outcomes and no procedure-specific complications.

Introduction: Aggregated data show that Black patients receive disproportionately lower rates of cosmetic surgery than their Caucasian counterparts. Similarly, our lab has shown that social media representation is lower among Black patients for breast reconstruction surgery, and we expect that this could be the case in cosmetic surgery as well. In this study, we explore the social media representation of Black patients and physicians in the five most common cosmetic surgery procedures: Rhinoplasty, Blepharoplasty, Abdominoplasty, Breast Augmentation, and Liposuction.

Methods: We collected data from RealSelf, the most popular social media site for sharing cosmetic surgery outcomes. Our lab assessed the skin tone of 1000 pictures of patients in each of the top five cosmetic surgeries according to the Fitzpatrick scale, a commonly utilized skin tone range. Additionally, we gathered Fitzpatrick scores of 72 providers who posted the photos within each surgical category.

Results: We found that Black patients and providers are underrepresented in Rhinoplasty, Blepharoplasty, Breast Augmentation, and Liposuction compared to the general population (defined by the US Census Bureau) while being proportionately represented in Abdominoplasty. Additionally, we found that patients most often matched Fitzpatrick scores when both had scores of two, while patients with a score of five and six rarely matched their provider’s score.

Conclusions: We suspect that the underrepresentation of Black patients and providers in social media for cosmetic surgery discourages patients from pursuing cosmetic surgeries. Therefore, it is essential to properly represent patients to encourage patients interested in considering cosmetic surgery.

Introduction: Pancreatectomy for adenocarcinoma carries significant risk for adverse short-term outcomes. However, little is known about intermediate-term outcomes – between 3-6 months – that may better guide informed consent and shared decision-making. Our objectives were to (1) identify factors associated with intermediate-term mortality and (2) characterize hospital variation in intermediate-term mortality following pancreatectomy.

Methods: Patients undergoing curative-intent pancreatectomy for adenocarcinoma were identified in the National Cancer Database (2006-2018). Logistic regression examined predictors of intermediate-term mortality. Hospital-level rates of intermediate-term mortality were assessed, and factors associated with high mortality hospitals were evaluated.

Results: Of 37,619 patients at 582 hospitals, 4,474 (11.9%) died within 6 months of surgery, most in the intermediate term (53.1%). Patients were less likely to die in the intermediate term with neoadjuvant therapy (OR: 0.85, 95% CI: 0.74-0.97) and adjuvant therapy (OR: 0.27, 95% CI: 0.24-0.30). Oncologic factors most predictive of intermediate-term mortality were stage (III vs. I, OR: 2.8, 95% CI: 2.1-3.7), grade (high vs. low, OR: 2.5, 95% CI: 2.0-3.1), and positive surgical margins (OR: 2.0, 95% CI: 1.8-2.3). Hospital-level intermediate-term mortality varied significantly, from 0.5% (n=118) to 14.2% (n=112) between lowest and highest quintile hospitals (p<0.001). Patients at high intermediate-term mortality hospitals were less likely to have private insurance (29.8% vs. 40.4%, p<0.001), receive neoadjuvant therapy (13.0% vs. 21.7%, p<0.001) or adjuvant therapy (60.1% vs. 71.1%, p<0.001), and were more likely to have positive margins (24.0% vs. 21.5%, p=0.03).

Conclusion: Nearly 12% of patients died within 6 months of surgery, over half occurring in the intermediate term. Treatment and oncologic-related factors were most predictive of intermediate-term mortality, and wide hospital variation was identified. These data should be considered when discussing the role of curative-intent pancreatectomy for cancer.

Introduction: The American Board of Plastic Surgery (ABPS) has been collecting practice data on operative repair of flexor tendon lacerations since 2006 as part of its Continuous Certification program.

Methods: Data on operative repair of flexor tendon lacerations from 2006-2014 were reviewed and compared to those from 2015-2020. National practice trends observed in these data were evaluated and reviewed alongside published literature and evidence-based medicine.

Results: 780 patients with flexor tendon laceration injuries were included. Mean patient age was 38 years; mean time between tendon injury and first evaluation was four days, and the mean time from injury to operative repair was 12 days. Four-strand sutures remain the most common technique of tendon repair (57%). In the recent cohort, there were significant decreases in tourniquet use (94% vs. 89%), general anesthesia (88% vs. 74%), and monofilament sutures (44% vs. 35%), with a significant increase reported in preserving the A1 pulley (20% vs. 29%). Postoperative movement was described as “almost full range of motion” or “good” in 70% of cases, and 74% of patients were satisfied with their results. Postoperative adverse events were reported in 26% of cases, with the most common complications being tendon adhesions (14%) and rupture (3%).

Conclusion: Review of ABPS tracer data for operative repair of flexor tendon lacerations establishes a framework by which surgeons can evaluate how their current practice aligns with that of their peers, and whether their practice patterns remain current relative to recommendations from evidence-based medicine.

Introduction: Breast cancer is the leading cause of cancer death in women worldwide. In sub-Saharan Africa, breast cancer has a lower incidence than high-resource settings but a higher mortality-to-incidence ratio. Limited literature regarding the clinical characteristics of breast cancer is available in many regions of sub-Saharan Africa. We aim to characterize the clinical attributes and outcomes of women with breast cancer in Botswana.

Methods: The prospective Thabatse cohort began enrolling participants (>18, biopsy-confirmed cancer) in 2010 from the four main oncology referral centers in Botswana. Baseline clinical data was obtained, and participants were followed quarterly for 5 years. Certification of death was obtained from families, providers, and death certificates.

Results: 902 women with breast cancer in Botswana, enrolled between 2010 and January 2021, were included. Median age at enrollment was 51.6 (IQR 43.0-62.9). 273 (30.3%) were HIV+, 606 (67.2%) HIV-, and 23 (2.5%) had unknown HIV status. Immunohistochemistry was reported for 686 (76.1%) participants, with 442 (64.4%) tumors being ER/PR+, 74 (10.8%) HER2+, and 170 (24.8%) triple-negative. At diagnosis, stage was known for 798 (88.5%), with 28 (3.5%) having stage I disease, 255 (32.0%) stage II, 416 (52.1%) stage III, and 99 (12.4%) stage IV. 597 (66.2%) participants had surgery; 62 (10.4%) had a lumpectomy, and 535 (89.6%) had a mastectomy. 501 (55.5%) participants received chemotherapy; 356 (39.5%) received radiotherapy. 5-year survival was 46.9% (95%CI 43.3-50.8%). Cox proportional hazard models explored the effect of HIV status and receptor status on survival. In unadjusted analyses, HIV+ participants had a hazard ratio (HR) of 1.37 (95%CI 1.11-1.67); receptor status, using ER/PR+ as baseline, resulted in a HR of 1.32 (95%CI 0.93-1.89) for HER2+ tumors, and 1.44 (95%CI 1.12-1.85) for triple-negative tumors. Adjusted analyses resulted in HIV+ HR of 1.17 (95%CI 0.93-1.47), HER2+ HR of 1.53 (95%CI 1.04-2.24), and triple-negative HR of 1.60 (95%CI 1.22-2.10).

Conclusion: This study demonstrates the characteristics of breast cancer in Botswana utilizing an ongoing prospective study. Participants on average presented at late stage, received invasive surgery, and had limited access to chemotherapy and radiotherapy. Additionally, HIV positivity and triple-negative immunohistochemistry were correlated with worse outcomes; however, HIV positivity was not significant when adjusted for other covariates.

Introduction: Social media use by plastic surgeons may contribute to the overall increase in breast reconstruction in the United States. However, recent data show a concerning decrease in breast reconstruction in African American women. The purpose of this study was to analyze the inclusion of African American women in social media posts for breast reconstruction, with the premise that this may be a possible contributing factor to decreasing rates of breast reconstruction in this population.

Methods: Data from several social media platforms were obtained manually on December 1, 2019. Each image was analyzed using the Fitzpatrick scale as a guide.

Results: A total of 2580 photographs were included that met the authors' criteria. Only 172 photographs (6.7 percent) were nonwhite. This study surveyed 543 surgeons, 5 percent of whom were nonwhite. The analysis of the results from the random sample of the top plastic surgery social media influencers showed that only 22 (5 percent) of the photographs uploaded were nonwhite patients. Furthermore, 30 percent of surgeons did not have any photographs of nonwhite patients uploaded.

Conclusions: Numerous factors can contribute to the disparity between the growing trend of white patients seeking reconstructive surgery compared to the decreasing trend of African American patients, one of which may be the disparity in their representation in social media, particularly among common platforms and social media influencers. This study highlights the evolving factors that may impair African American breast cancer patients' access to safe, effective breast reconstruction, which must be identified and resolved.

Introduction: Although the use of minimally invasive surgery (MIS) has become common, there has been little research into disparities in the receipt of MIS for treatment of endometrial cancer. This study aims to assess racial-ethnic and insurance status disparities in the proportion of patients with stage I-III endometrial cancer who receive MIS according to the Commission on Cancer guidelines.

Methods: Patients who were diagnosed with stage I-III endometrial cancer from 2010-2018 were identified from the National Cancer Database. Differences in patient characteristics and perioperative outcomes for patients who received guideline concordant MIS versus those who did not were examined. Odds of receiving MIS were determined while controlling for sociodemographic, clinical, and hospital characteristics.

Results: 217,760 surgically treated patients with stage I-III endometrial cancer were included. Overall, 9.7% (n=21,040) were non-Hispanic Black, 6.3% (n=13,726) were Hispanic, and 9.4% (n=20,520) had Medicaid or no insurance. On multivariable analysis, non-Hispanic Black race (OR 0.40, 95% CI 0.33-0.49) and Hispanic ethnicity (OR 0.60, 95% CI 0.48-0.76) were associated with lower odds of receiving guideline concordant MIS. Additionally, those with either Medicaid or no medical insurance (OR 0.50, 95% CI 0.43-0.58) had lower odds of receiving MIS. Nonadherence to MIS guidelines was associated with worse 30-day (0.88% vs 0.22% MIS) survival, worse 90-day (2.0% vs 0.53% MIS) survival, and worse 30-day unplanned readmissions (3.92% vs 1.92% MIS; all P<0.001).

Conclusion: There are racial-ethnic and insurance status disparities in Commission on Cancer MIS guideline concordance for endometrial cancer. MIS is associated with improved short-term outcomes.