Thomas Jefferson University Hospital Research Articles

Hemorrhage Sites and Mitigation Strategies After Pancreaticoduodenectomy.

Postpancreatectomy hemorrhage is an uncommon but highly morbid complication of pancreaticoduodenectomy. Clinical evidence often draws suspicion to the gastroduodenal artery stump, even without a clear source. To determine the frequency of gastroduodenal artery bleeding compared to other sites and the results of mitigation strategies. This cohort study involved a retrospective analysis of data for consecutive patients who had pancreaticoduodenectomy from 2011 to 2021 at Memorial Sloan Kettering Cancer Center (MSK) and Thomas Jefferson University Hospital (TJUH). Demographic, perioperative, and disease-related variables. The incidence, location, treatment, and outcomes of primary (initial) and secondary (recurrent) hemorrhage requiring invasive intervention were analyzed. Imaging studies were re-reviewed by interventional radiologists to confirm sites. Inclusion criteria were met by 3040 patients (n = 1761 MSK, n = 1279 TJUH). Patients from both institutions were similar in age (median [IQR] age at MSK, 67 [59-74] years, and at TJUH, 68 [60-75] years) and sex (at MSK, 814 female [46.5%] and 947 male [53.8%], and at TJUH, 623 [48.7%] and 623 male [51.3%]). Primary hemorrhage occurred in 90 patients (3.0%), of which the gastroduodenal artery was the source in 15 (16.7%), unidentified sites in 24 (26.7%), and non-gastroduodenal artery sites in 51 (56.7%). Secondary hemorrhage occurred in 23 patients; in 4 (17.4%), the gastroduodenal artery was the source. Of all hemorrhage events (n = 117), the gastroduodenal artery was the source in 19 (16.2%, 0.63% incidence in all pancreaticoduodenectomies). Gastroduodenal artery hemorrhage was more often associated with soft gland texture (14 [93.3%] vs 41 [62.1%]; P = .02) and later presentation (median [IQR], 21 [15-26] vs 10 days [5-18]; P = .002). Twenty-three patients underwent empirical gastroduodenal artery embolization or stent placement, 7 (30.4%) of whom subsequently experienced secondary hemorrhage. Twenty percent of all gastroduodenal artery embolizations/stents (8/40 patients), including 13% (3/13 patients) of empirical treatments, were associated with significant morbidity (7 hepatic infarction, 4 biliary stricture), with a 90-day mortality rate of 38.5% (n = 5) for patients with these complications vs 7.8% without (n = 6; P = .008). Ninety-day mortality was 12.2% (n = 11) for patients with hemorrhage (3 patients [20%] with primary gastroduodenal vs 8 [10.7%] for all others; P = .38) compared with 2% (n = 59) for patients without hemorrhage. In this study, postpancreatectomy hemorrhage was uncommon and the spectrum was broad, with the gastroduodenal artery responsible for a minority of bleeding events. Empirical gastroduodenal artery embolization/stent without obvious sequelae of recent hemorrhage was associated with significant morbidity and rebleeding and should not be routine practice. Successful treatment of postpancreatectomy hemorrhage requires careful assessment of all potential sources, even after gastroduodenal artery mitigation.

Outcome of Multiple Myeloma Patients Undergoing Allogeneic Stem Cell Transplant with Post-Transplant Cyclophosphamide - a Single Center Experience

Objective: The survival of multiple myeloma (MM) patients has improved significantly in the past decade with the development of new effective treatments including immunomodulators, monoclonal antibodies, and chimeric antigen receptor T cell (CAR-T) therapy. However, to date, none of the available treatments is curative for MM. In high-risk and relapsed patients, allogeneic stem cell transplant (allo-SCT) achieved long term survival in carefully selected MM patients. Here, we report our experience in MM patients who received allo-SCT at Thomas Jefferson University Hospital (TJUH). We aim to define the role of allo-SCT in the new era of MM treatment. Method: We retrospectively analyzed data of 57 MM patients who received allo-SCT at TJUH from 1996-2023. We used EZR software to analyze post-transplant outcomes including overall survival (OS) and disease-free survival (DFS) with Kaplan Meier method. Cumulative incidence (CI) calculated by EZR was used to determine relapse rate (RR) and non-relapse mortality (NRM). Median time to neutrophil and platelet engraftment was analyzed by Excel 2022. Results: The median age was 53 years old (range from 34-68 years old); 29 females and 28 males. Mean follow up time was 57.9 months (range from 0 to 253.6 months). Thirty-two patients received reduced intensity conditioning, 11 patients received myeloablative conditioning regimen, and 5 patients received non-myeloablative conditioning regimens. Thirty-one patients received matched related donor (MRD), 3 patients received matched unrelated donor (MUD), 21 patients received haploidentical (HI), and 2 patients received mismatched unrelated donor stem cell grafts. Time to neutrophil and platelet engraftment data was available for 33 and 29 patients, respectively. The median time to neutrophil engraftment was 12 days (range from 3-23 days). and the median time to platelet engraftment was 16 days (range from 6-51days). The 3-year and 5-year probabilities of OS were 53.2% (95% confidence interval is 38.7% to 65.7%) and 48.3% (95% confidence interval is 33.9% to 61.3%) respectively. The 3-year and 5-year probability of DFS were 37% (95% confidence interval is 21.6% to 52.5%) and 26.2% (95% confidence interval is 12.6% to 42%) respectively. Since 2009, most of the patients (33/36) in this study received post-transplant cyclophosphamide. Thus, we compared the OS for patients who received allo-SCT before 2009 and after 2009. The 5-years OS was 28.6% (95% confidence interval is 7.8% to 42.5%) and 62.7% (95% confidence interval is 43% to 77.2%) for patients who received allo-SCT before 2009 and after 2009 respectively (P=0.029). Among 43 patients with available relapse date data, the CI of relapse at 3 years was 36.9% (95% confidence interval 21.5%-52.4%) and the CI of NRM at 3 years was 26.1% (95% confidence interval 14%-39.9%). Thirty-five patients were evaluable for acute graft versus host disease (GVHD). Nine patients developed grade I-IV acute GVHD (25.7%) and 2 patients experienced grade III-IV acute GVHD (5.7%). Four out of 35 evaluable patients experienced chronic GVHD (11.4%): 2 patients had mild to moderate scleroderma, one patient had severe chronic skin GVHD and one had severe lung chronic GVHD. For 36 patients received allo-SCT after 2009, 3 died from relapse, 3 patients died from infection, 2 died from GVHD, 2 from unknown cause, one from brain bleeding, one from non-infectious pulmonary failure and one from acute heart failure. Conclusion: In MM patients with high-risk feature or relapsed/refractory disease, allo-SCT can be curative for approximately a third of the patients. Post-transplant cyclophosphamide coupled with contemporaneous MM treatments significantly increased the survival of allo-SCT MM patients in the modern era.

Outcomes of patients placed on comfort care at a large, urban, quaternary care medical center.

229 Background: End-of-life care in the typical hospital setting has several drawbacks including lack of interdisciplinary hospice teams, less comfortable environments, and lack of provider experience. In addition, patients at the end of their life are often severely symptomatic and imminently dying and thus have barriers to appropriate disposition. Few studies have explored the outcomes of patients placed on comfort care with respect to hospice disposition. The objective of this study was to perform a retrospective analysis of patients who transitioned to comfort care at Thomas Jefferson University Hospital (TJUH) to better understand the ultimate disposition of patients placed on comfort care. Methods: We conducted a retrospective study of patients placed on the comfort care order set between July 1st, 2021, until June 30th, 2022. A report of all patients placed on our institutions “comfort care order set” was run using Qlik analytics and data platform. Each individual patient chart was then analyzed to collect multiple clinical variables, including diagnoses, palliative care involvement, time on comfort care, hospital length of stay (in total and after comfort care order set placed), healthcare utilization prior to hospitalization, and IV medications used for symptom management. Participants were limited to those 18 years or older. IRB approval was obtained as per institutional guidelines. Results: 541 patients were included in the analysis. 424 (78.37%) patients died while still admitted to the hospital. This amounts to 1.5 patients a day requiring comfort care support. 60% of patients on comfort care were in the intensive care unit (ICU). The ICU population spent on average 11.04 (+/- 24) hours on comfort care. 19% of these patients were on comfort care for longer than 12 hours. Subspecialty and hospital medicine patients were on comfort care for 3.3 (+/- 3.3) days and 5 (+/- 9.5) days respectively. Within oncology patients, the average time was 3 (+/- 2.4) days. 94% of the patients placed on comfort care in the ICU died in the hospital as compared to 53% of subspecialty and 59% of hospital medicine patients. 40% of non-ICU patients were discharged with hospice services. The average length of stay for the study population was 12.5 (+/- 13.0) days. Conclusions: The majority of patients placed on comfort care died during their hospitalization demonstrating a real need for comprehensive end of life care and immediate hospice services for our patient population. For those discharged with hospice services, they spent an excessive amount of time in the hospital waiting for services to be arranged. Establishing a hospital based inpatient hospice unit has the potential to improve the patient and family experience, provide better end of life care, and reduce length of stay. Previous studies have also demonstrated decreased costs. Ongoing work is underway at our institution to provide optimal end of life care for this patient population.

Assessment of Concurrent Personal and Family History of Hematologic Malignancies in Patients Referred for Genetic Counseling at Thomas Jefferson University Hospital (TJUH), Sidney Kimmel Cancer Center (SKCC) from 2014-2021

Introduction: Studies have suggested that 5% of adult patients with MDS and AML have inherited mutant alleles that predispose to disease development. More than 50 hereditary cancer syndromes have been defined-many specifically associated with myeloid and lymphoid malignancies. Yet hereditary hematologic malignancies (HHMs) continue to go unrecognized. The implications for patients, especially regarding family donor selection for allogeneic stem cell transplant, necessitate greater awareness. Referrals for genetic counseling at SKCC have been triggered primarily by a personal or family history of solid tumors. The aim of this study was to assess the frequency of hematologic malignancies in the setting of concurrent solid tumor diagnosis or significant family history of cancer. Methods: In order to gain preliminary insight into the unmet need for genetic counseling for HHMs at our institution, we queried our Cancer Genetics REDCap database (IRB-approved #16D.743) for all patients who had a cancer diagnosis, a family history of cancer, and a hematologic malignancy (HM). Cancer diagnoses and family history were collected at time of genetic counseling by a genetic counselor and were primarily patient reported. After identifying patients with self-reported HMs, HM diagnoses were confirmed by medical history and pathology report, if available, in our electronic medical record system. Results: The database included 5073 patients who were evaluated at SKCC's Cancer Genetics Program between 2014 and 2021. 3150 patients had a known cancer diagnosis including: breast (1812), prostate (349), ovarian (219), colon/rectal (218), pancreatic (157), melanoma (123), kidney (83), lung (18), and other (708). 4734 patients had a family history of cancer. 2913 patients had both a personal cancer diagnosis and a family history of cancer. Overall, 58 (2%) of these patients had a hematologic malignancy. Among patients with a HM, average age at HM diagnosis was 47 years (median 50, range 5-80) and average age at time of genetics counseling was 60 years (median 63.5, range 35-83). 48 of the patients were female. 51 were White/Caucasian, 6 were Black/African American, and 1 was Asian. 1 patient was Hispanic and 8 patients were Ashkenazi Jewish. There was a wide range of HM diagnoses (Table 1). 39 (67%) patients had a concurrent solid malignancy. The 58 patients with a HM and family history of cancer had on average 5.2 family members with a malignancy (median 4, range 1-14). 53 (91.4%) of the patients had at least 1 first degree relative with a malignancy (Table 2). 17 (29%) patients had a family member with a known HM, including 11 (19%) with a first degree relative. HMs identified among first and second degree relatives included: CLL (N=5) Non-Hodgkin lymphoma (N=3), Hodgkin lymphoma (N=3), Waldenstrom macroglobulinemia (N=2), lymphoma not otherwise specified (N=3), AML (N=1), leukemia NOS (N=3), multiple myeloma (N=2), and polycythemia transformed to myelofibrosis (N=1). Of the 40 HM patients who had germline genetic testing performed, alterations were identified in 22. 10 of the patients had 11 pathogenic or likely pathogenic variants in CHEK2, ATM, FANCC, NBN, MSH2, MSH6, BRCA2, and/or FH. 6 (60%) of these patients had alterations in genes known to be associated with HHMs. 21 variants of unknown significance were identified in 14 patients, including in 2 patients with pathogenic variants. Conclusions: Approximately 2% of patients referred for genetic counseling for a personal malignancy and family history of cancer had a concurrent diagnosis of a HM. This likely under-represents the number of patients with HHMs, particularly since 6 of the 10 patients identified to have pathogenic or likely pathogenic variants had alterations in genes associated with HHMs. The findings that 67% of the HM patients had a second malignancy and that HM patients had on average 5 family members with cancer highlight the importance of identifying high risk patients to guide surveillance for earlier detection. Earlier recognition is crucial, as the average age of genetics counseling was 60. Despite striking family cancer histories, germline variants were not identified in several patients, suggesting that there may be germline variants that have yet to be identified. Efforts to expand genetic screening that focus on earlier identification, standardized germline testing for high risk patients, and appropriate surveillance will be important. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

The use of individualized 3D-printed models on trainee and patient education, and surgical planning for robotic partial nephrectomies.

3D printing is a growing tool in surgical education to visualize and teach complex procedures. Previous studies demonstrating the usefulness of 3D models as teaching tools for partial nephrectomy used highly detailed models costing between $250 and 1000. We aimed to create thorough, inexpensive 3D models to accelerate learning for trainees and increase health literacy in patients. Patient-specific, cost-effective ($30-50) 3D models of the affected urologic structures were created using pre-operative imaging of 40 patients undergoing partial nephrectomy at Thomas Jefferson University Hospital (TJUH) between July 2020 and May 2021. Patients undergoing surgery filled out a survey before and after seeing the model to assess patient understanding of their kidney, pathophysiology, surgical procedure, and risks of surgery. Three urological residents, one fellow, and six attendings filled out separate surveys to assess their surgical plan and confidence before and after seeing the model. In a third survey, they ranked how much the model helped their comprehension and confidence during surgery. Patient understanding of all four subjects significantly improved after seeing the 3D model (P < 0.001). The urology residents (P < 0.001) and fellow (P < 0.001) reported significantly increased self-confidence after interacting with the model. Attending surgeon confidence increased significantly after seeing the 3D model (P < 0.01) as well. Cost-effective 3D models are effective learning tools and assist with the evaluation of patients presenting with renal masses, and increase patient, resident, and fellow understanding in partial nephrectomies. Further research should continue to explore the utility of inexpensive models in other urologic procedures.

Efficient Targeting of CD38 in Mature T-Cell Neoplasms with Daratumumab and Allogeneic NK Cells

Introduction: Mature T-cell neoplasms (MTCN) are a heterogeneous and aggressive group of lymphoid neoplasms with very limited options for precision immunotherapy. Targeted immunotherapy with antibodies directed against surface markers on tumor cells has emerged as an effective treatment for B-cell neoplasms, but the development of immunotherapy strategies for MTCN has been much slower. CD38 is expressed at low levels in a subset of normal resting T-cells. Data on the frequency and level of CD38 expression is MTCN are lacking. Our goal was to study the expression and stability of CD38 on a spectrum of neoplastic T-cell populations and assess the potential anti-tumor effect of anti-CD38 monoclonal antibodies in combination with allogeneic natural killer (NK) cells in MTCN.Methods and results: We searched the Thomas Jefferson University Hospital (TJUH) pathology records for all cases of MTCN for which immune-phenotypical characterization of the neoplastic T-cell population by multi-color flow cytometry (FC) was available. CD38 expression was evaluated in cases where an abnormal T-cell population (defined as loss of one on more pan T-cell markers and/or a skewed CD4 to CD8 ratio). A total of 103 unique patients with MTCN were identified. Of these, 51 had at least one biopsy whereby tumor cells had an abnormal immunophenotype which could then be assessed for CD38 expression. CD38 was expressed to some extent in all but 2 cases with expression levels on peripheral T-cell lymphoma-not other specified (PTCL-NOS)(% Mean±SEM = 80.84±10.26, N=11), angioimmunoblastic T-cell lymphoma (AITL) (% Mean±SEM = 80.56±7.34, N=6), nodal PTCL with T follicular-helper (T FH) phenotype (% Mean±SEM = 55.00±12.72, N=7), anaplastic large cell lymphoma (ALCL) (% Mean±SEM = 77.38±10.75, N=3), large granular lymphocytic leukemia (LGLL) (% Mean±SEM = 80.27±7.49, N=4), T-cell prolymphocytic leukemia (T-PLL) (% Mean±SEM = 88.26±4.20, N=7), cutaneous T-cell lymphoma (CTCL) (% Mean±SEM = 49.52±14.77, N=7), adult T‐cell leukemia/lymphoma (ATLL) (% Mean±SEM = 76.68±10.45, N=3), hepatosplenic T-cell lymphoma/monomorphic epitheliotropic intestinal T-cell lymphoma (HSTCL/MEITL) (% Mean±SEM = 75.00±11.37, N=3). The medians and ranges of the MFI of CD38 on CD38+ tumor cells were the following: PTCL-NOS 29.46 (3.1-115.62), AITL 13.64 (2.67-29.41), other PTCL-T FH 5.87 (0-48.67), ALCL 11.13 (3.6-11.28), LGL (13.52 (9.72-18.25), T-PLL 6.49 (3.09-18), CTCL 8.36 (0-116.75), ATLL 27.17 (3.54-60.5), HSTCL/MEITL 15.96 (6.96-167.85). We also measured surface expression of CD38 on the patient-derived MTCN cell lines HuT-78, HuT-102, Jurkat, H9, HH, and MOTN1, all of which expressed CD38, to determine which of these cell lines could be used for in vitro experiments. We next evaluated if the CD38 molecule is an effective target for antibody-mediated therapy in MTCN, by testing the ability of daratumumab (dara) to enhance antibody-dependent cellular cytotoxicity (ADCC) elicited by NK cells. For this, we purified normal NK-cells from TJUH Blood Bank leukoreduction filters and cultured with recombinant IL-15 for 48 hours prior to all experiment. T-cell lines and primary MTCN cells were treated with increasing concentrations (0.1 µg/mL - 2 µg/mL) of dara or isotype control. NK-cells were added at Effector:Target ratio of 5:1 and incubated for 4 hours at 37⁰ Celsius. Cytotoxicity was measured by LDH release assay. Dara induced significant cell lysis starting at doses as low as 0.1μg/mL in both T-cell lines and primary MTCN cells, reaching maximum cytotoxicity at 0.5-2μg/mL (mean±SEM cytotoxicity in isotype vs dara treated cells= 50.0±5.05% vs 97.5±2.5%, N=4, p-value=0.0002). The degree of ADCC induction also correlated with interferon-gamma (IFN-g) release by NK cells in vitro for both T-cell lines and primary MTCN cells.Conclusions: The majority of MTCN analyzed (N=49, 96%) showed any degree of CD38 expression by FC with a wide variation of intensity, including within the same subtype. Allogeneic NK cells efficiently elicited dara-mediated ADCC of tumor cells from all MTCN subtypes and produced abundant IFN-g. These data highlight the potential of targeting CD38 in MTCN with anti-CD38 antibodies and allogeneic NK cells. The strong CD38 expression observed in most tumor cells from ultra-rare and very aggressive subtypes of MTCL opens the door to much needed new treatment strategies. DisclosuresBrammer: Celgene: Research Funding; Kymera Therapeutics: Consultancy; Seattle Genetics: Speakers Bureau. Chakravarti: Kiadis Pharma: Patents & Royalties. Porcu: Viracta: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Innate Pharma: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; BeiGene: Membership on an entity's Board of Directors or advisory committees, Research Funding; Incyte: Research Funding; Daiichi: Honoraria, Research Funding; Kiowa: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Spectrum: Consultancy; DrenBio: Consultancy.

Associations between age and stage at presentation among black and white colorectal cancer patients.

e15542 Background: Blacks have a 25% higher incidence of colorectal cancer compared to their white societal counterparts. Additionally, the overall mortality rate among black colorectal cancer patients is 50% higher than that of whites. It is suggested that the etiology of this disparity may be inadequate screening for colorectal cancer among racial communities. However, little is known about the correlations between age and stage at presentation among black and white colorectal cancer patients at Thomas Jefferson University Hospital (TJUH). Objective: The objective of this study is to explore diagnostic trends that may unveil differences in age and stage at presentation between black and white colorectal cancer patients. Methods: De-identified patient data was obtained from The Oncology Data Services Department (Cancer Registry) of TJUH. The population cohort (n= 529) included newly diagnosed colorectal cancer patients treated at TJUH from 2015-2019 and included information regarding race, sex, age at presentation, stage at presentation, histological code, tumor markers: KRAS, NRAS, BRAF, MS1, treatment received, surgical findings: tumor size, lymph node involvement, presence of distant metastases at first surgery, response to chemotherapy &amp; disease-free survival. The cohort was divided by age (&lt;50, 50-65, &gt;65). Patient age was compared against AJCC stage at presentation. Results: Findings reveal 12.38% of blacks &amp; 14.29% of whites presented before age 50. 36.19% of blacks and 30.19% of whites presented between ages 50-65. 51.43% of blacks &amp; 55.42% of whites presented after age 65. Additionally, the average age at presentation among blacks was 62.71 years vs. 65.71 years among whites. Lastly, the average stage at presentation among blacks was between 2C and 3A while the average stage at diagnosis among whites was between least 2B and 2C. Conclusions: Within this population, similarities exist among age at presentation between black and white patients, however, differences in stage at presentation exist between blacks and whites. Further research is needed explore how these findings inform the poor clinical outcomes seen among black colorectal cancer patients. Future analysis of this cohort will explore trends regarding time to treatment and treatment received after diagnosis, as similar studies have found differences among racial populations.

The use of 3D printed models on trainee and patient experience for partial nephrectomies.

363 Background: 3D printing is a growing tool in surgical education due to the ability to visualize organs, tissue, and masses from multiple angles before operating on a patient. Previous studies using highly detailed and expensive 3D models costing between $1,000-250 per model have been shown to enhance patient and trainee comprehension of tumor characteristics, goals of surgery, and planned surgical procedure for partial nephrectomies. In our study we aim to use simpler and less expensive models in a greater range of patients receiving partial nephrectomies to determine the use of 3D models in patient, resident, and fellow education. Methods: 3D models of the effected kidney, mass, renal artery, and renal vein were created using preoperative imaging of undergoing partial nephrectomies at Thomas Jefferson University Hospital (TJUH) costing $35 per model. Residents and fellows filled out 3 surveys assessing their surgical plan and their confidence in the chosen plan at 3 time points: 1) Before seeing the model, 2) After seeing the model before surgery, and 3) After surgery. Ten patients filled out 2 surveys about their understanding of the kidney, their disease, the surgery they will undergo, and the risks involved with surgery before and after seeing the model. Results: Based on surveys to assess for surgical plan and confidence given to resident and fellow surgeons before and after seeing the 3D model, confidence significantly increased. Surveys given after surgery assessing anatomic and surgical comprehension found that resident and fellow surgeons rated the helpfulness of the models on their anatomical comprehension 7.6 out of 10 and the help of the models on their surgical confidence 7 out of 10. Patient understanding of their kidney, disease, and surgery significantly increased after seeing the 3D model, but the risks associated with surgery did not significantly increase. The extent that the model helped the patients learn about the kidney, their disease, the surgery, and the risks related to surgery were rated an average of 8.33, 9.67, 9.5, and 8.83 out of 10, respectively. Conclusions: Patient-specific 3D models for partial nephrectomies increase resident and fellow confidence in surgical approach and helped patients learn about their disease and feel comfortable going into surgery. Thus, it is important to continue to explore 3D models as an educational tool for both trainees and patients and potentially include 3D models as part of the standard of care. Further research could continue to explore the utility of 3D models as a pre-operative educational tool for both patients and trainees in other surgical fields.

High Incidence of Burst Suppression during Propofol Sedation for Outpatient Colonoscopy: Lessons Learned from Neuromonitoring.

Background Although anesthesia providers may plan for moderate sedation, the depth of sedation is rarely quantified. Using processed electroencephalography (EEG) to assess the depth of sedation, this study investigates the incidence of general anesthesia with variable burst suppression in patients receiving propofol for outpatient colonoscopy. The lessons learned from neuromonitoring can then be used to guide institutional best sedation practice. Methods This was a prospective observational study of 119 outpatients undergoing colonoscopy at Thomas Jefferson University Hospital (TJUH). Propofol was administered by CRNAs under anesthesiologists' supervision. The Patient State Index (PSi™) generated by the Masimo SedLine® Brain Root Function monitor (Masimo Corp., Irvine, CA) was used to assess the depth of sedation. PSi data correlating to general anesthesia with variable burst suppression were confirmed by neuroelectrophysiologists' interpretation of unprocessed EEG. Results PSi values of <50 consistent with general anesthesia were attained in 118/119 (99.1%) patients. Of these patients, 33 (27.7%) attained PSi values <25 consistent with variable burst suppression. The 118 patients that reached PSi <50 spent a significantly greater percentage (53.1% vs. 42%) of their case at PSi levels <50 compared to PSi levels >50 (p=0.001). Mean total propofol dose was significantly correlated to patient PSi during periods of PSi <25 (R=0.406, p=0.021). Conclusion Although providers planned for moderate to deep sedation, processed EEG showed patients were under general anesthesia, often with burst suppression. Anesthesiologists and endoscopists may utilize processed EEG to recognize their institutional practice patterns of procedural sedation with propofol and improve upon it.

Unplanned hospital admissions are a marker of decreased overall survival in patients with solid tumor malignancies.

e19118 Background: For patients being treated for solid tumor malignancies, an unplanned hospitalization is a significant event that may help predict outcomes. Prior studies have demonstrated a 1 year mortality rate as high as 73% for oncology patients following an unplanned admission. As our therapies continue to improve, better understanding of the significance of acute hospitalizations is needed to help influence management decisions for oncology providers. Methods: We conducted a retrospective review of all patients admitted to a solid oncology inpatient service at Thomas Jefferson University Hospital (TJUH) who were discharged between January 1st and March 31st 2019. We excluded all patients who had not established care with a provider at TJUH prior to the index admission. We collected data on age, sex, length of stay (LOS), diagnosis, treatment activity, ECOG performance status (PS), 30 day and total re-admissions, survival and whether hospice was discussed during index admission. We censored data at 180 days after date of index discharge. We evaluated overall survival (OS) using the Kaplan-Meier method and compared survival to 30-day readmission and ECOG status using a log-rank test. Results: During our investigational time period there were 263 discharges, which yielded a study population of 182 patients. Median LOS was 5 days and 55% of patients were female. Gastrointestinal and genitourinary cancers were the most prevalent, observed in 32% and 28% of patients respectively. 75% of patients were on active treatment at time of admission. Survival data was available in 174 patients and at 6 months, OS was found to be 47%. 35% of patients were readmitted within 30 days of their index discharge. Patients readmitted within 30 days had a significantly lower OS than patients without a readmission (31% vs. 56%, P = 0.002). 70% of patients had a PS of 0 or 1 prior to index admission and OS in this group was significantly better in comparison to patients with a higher ECOG score (57% vs. 33%, P = 0.001). Hospice was discussed with 31% patients during index admission. Conclusions: We observed a 6 month overall survival of 47% following an unplanned admission and this was lower for patients that were re-admitted within 30 days. Higher PS scores predicted worse outcomes; however, the majority of patients had a favorable PS prior to index admission. This study suggests that following an unplanned admission, oncology patients are at elevated risk for poor outcomes and that providers should reassess goals of care at follow up clinic visits.

Outcomes of out-of-hospital extracorporeal membrane oxygenation transfers: significance of initiation site and personnel.

Extracorporeal membrane oxygenation is an accepted therapy option for refractory cardiac or respiratory failure. The outcomes of cases initiated at non-extracorporeal membrane oxygenation centers and subsequently transported for management to an extracorporeal membrane oxygenation center require further investigation. Retrospective institutional review board-approved database research and chart reviews were performed on referrals for extracorporeal membrane oxygenation initially admitted to an outside non-extracorporeal membrane oxygenation center hospital (OSH) then transferred to our extracorporeal membrane oxygenation center (Thomas Jefferson University Hospital (TJUH)). Unstable patients were placed on extracorporeal membrane oxygenation at OSH (Group A) before transport, while others were initiated at our certified extracorporeal membrane oxygenation center (Group B) upon arrival. Group A was further subdivided into patients cannulated by OSH personnel (Group AOSH) or TJUH transport team (Group ATJUH). Outcomes and complications were compared between the different initiation sites and personnel. A total of 108 patients were transferred from August 2010 to June 2018. The technical complication rate for all Group A patients was 33/49 (67%), while that of Group B was 24/59 (41%); p = 0.006. Within Group A, Group AOSH had a greater technical complication rate with 29/33 (88%) than Group ATJUH with 4/16 (25%); p < 0.001. extracorporeal membrane oxygenation survival rate was 34/49 (69%) in Group A and 43/59 (73%) in Group B; p = 0.690. The extracorporeal membrane oxygenation survival rate for Group AOSH and Group ATJUH was 21/33 (64%) and 13/16 (81%), respectively; p = 0.210. Promising extracorporeal membrane oxygenation survival rates were observed in transferred patients. The complication rates related to cannulation technique were significantly higher when patients were initiated at non-extracorporeal membrane oxygenation centers, especially when placed by personnel from non-extracorporeal membrane oxygenation centers.

Pericardial Effusion with Tamponade Physiology in a Patient with Multiple Myeloma

A 78-yeaer old African American female with a past medical history of IgA Kappa Multiple Myseloma was transfered to the Cardiovascular Intensive Care Unit (CVICU) at Thomas Jefferson University Hospital (TJUH) after being diagnosed with a pericardial effusion with tamponade physiology at an outside hospital.

1719 Esophageal Adenocarcinoma Arising From &lt;1 cm of Intestinal Metaplasia at the Gastroesophageal Junction: Is &lt;1 cm Actually Low Risk?

INTRODUCTION: Barrett’s esophagus (BE) is defined by length, which has been shown to correlate with the risk for esophageal adenocarcinoma (EAC). Although long-segment (&gt;3 cm) and short-segment (between 1 and 3 cm) BE are well-established definitions, current ACG practice guidelines recommend against defining columnar goblet-cell metaplasia in the distal esophagus &lt;1 cm as BE. This specific variation of metaplasia disease has been officially defined as “specialized IM of the esophagogastric junction” (SIM-EGJ) over alternatives such as ultra-short segment BE. Given the relatively low incidence of EAC in this group and the high degree of inter-observer variability in defining metaplasia of this length, current guidelines recommend against biopsy in this setting. Individual cases in practice however have demonstrated reasons to question this paradigm. CASE DESCRIPTION/METHODS: A 42 year old white man with obesity and 10 years of reflux symptoms was diagnosed with a GE junction nodule on index endoscopy. Cold forceps biopsy of the nodule revealed high grade dysplasia, prompting referral to the BE Treatment Center at Thomas Jefferson University Hospital (TJUH). His repeat EGD revealed salmon-colored mucosa suggestive of intestinal metaplasia in an area less than 1 cm in length (Z-line at 39 cm). An 8 mm nodule was identified in the area of abnormal mucosa and removed by endoscopic mucosal resection. Pathology of the nodule revealed a T1a moderately differentiated intramucosal adenocarcinoma arising in BE. All margins were negative for invasive carcinoma, dysplasia, and intestinal metaplasia. The patient patients underwent endoscopic eradication therapy (EET) with radiofrequency ablation at the GE junction and in subsequent endoscopy achieved complete eradication of intestinal metaplasia (CE-IM) and dysplasia (CE-D). DISCUSSION: Although current guidelines advise against biopsying &lt;1 cm of variability in Z line given its status as a relatively low risk lesion, data used to generate many of these risk predictions for EAC exclude visible lesions in their studies. Therefore in practice there may be under-evaluation of SIM-EGJ, potentially missing advanced dysplastic or cancerous disease. This case illustrates that the consideration BE risk factors and all endoscopic findings, such as nodules, need to be taken together when assessing abnormal mucosa at the GE junction, even with &lt;1 cm variability of the Z line.

New Graduates are Changing Trends in Operating Room Traffic

The Jefferson College of Nursing (JCN) is affiliated with Thomas Jefferson University Hospital (TJUH), which is one of the top-ranked hospitals in the nation. TJUH is recognized by the American Nurses Credentialing Center as a Magnet® hospital for quality patient care, nursing excellence, and innovations in professional nursing practice. Since JCN is an integral part of Jefferson’s premier academic health center, it is imperative that the two entities maintain a partnership in clinical care education. JCN offers a concept based curriculum which promotes practice excellence, innovation, population health, and inter-professional collaboration. The innovative curriculum offers senior learners a selective immersion in which students can now request the perioperative department as a rotational site. Having this unique opportunity for seven weeks challenges students to collaborate with leadership and staff in the perioperative environment and promotes professional interactions in an otherwise unfamiliar department. Additionally, entering the perioperative environment is the perfect setting to observe quality, safety, communication, and stimulate clinical reasoning amongst a very vulnerable patient population. For those students whose interests were peaked or career paths confirmed, joining the perioperative team as a new graduate begins with the innovative program at JCN then continues with the hospital sponsored perioperative 101 class and nurse residency program. Exposure to procedural areas is minimal at best for most nursing programs and having the perioperative department as part of JCN’s selective immersion will not only benefit our graduates, it will also help support nursing stability for an otherwise hard to staff specialty division.

One-step versus two-step diagnostic testing for gestational diabetes: a randomized controlled trial

Objective: To evaluate the incidence of gestational diabetes mellitus (GDM) using the one-step as compared with the two-step approach.Study design: This was a parallel group nonblinded randomized trial conducted at Thomas Jefferson University Hospital (TJUH) in Philadelphia, Pennsylvania from June 2016 to December 2016. The primary outcome was GDM incidence in the one-step compared to the two-approach. Pregnant women without a history of pregestational diabetes were offered screening for GDM at gestational age 24–28 weeks. Obese women, defined as having a BMI ≥30 kg/m2, as well as those with a history of a pregnancy complicated by GDM, a history of a macrosomic baby (>4000 g), or with polycystic ovarian syndrome (PCOS), were offered early screening at their initial prenatal visit, and screening was repeated at 24–28 weeks if initially normal. Women were excluded if they had pre-existing diabetes or had a history of bariatric surgery. Women who were eligible were randomized in a 1:1 ratio to either the one-step or two-step approaches. A sample size of 142 women was planned per group. Women randomized to the one-step approach, after an overnight fast, were given a 2-h glucose tolerance test, which consisted of a 75-g glucose load. Blood glucose levels were measured fasting, at 1 h and 2 h after the glucose load. Diagnostic cutoffs for GDM diagnosis were one value of either fasting ≥92 mg/dL, 1 h ≥180 mg/dL, or 2 h ≥153 mg/dL, respectively. Women randomized to the two-step approach were given a nonfasting 50-g glucose load, and the blood glucose level was measured an hour after the glucose load. If that value was ≥135 mg/dL, the patient had a 3-h glucose tolerance test consisting of a 100-g glucose load. Diagnostic cutoffs for GDM diagnosis for this 3-h test were ≥2 abnormal values of fasting ≥95 mg/dL, 1 h ≥180 mg/dL, 2 h ≥155 mg/dL and 3 h ≥140 mg/dL, respectively. All analyses were done using an intention-to-treat approach, evaluating women according to the treatment group to which they were randomly allocated.Results: Two hundred eighty-four women agreed to take part in the study and underwent randomization from June 2015 to December 2015. Of them, 249 completed the screening and were followed up for the primary endpoint. Out of the 249 women who completed the screening, 123 were assigned to the one-step group and 126 to the two-step group. GDM occurred in 10 women (8.1%) in the one-step group, and 7 women (5.6%) in the two-step group (p = .42). Preeclampsia, preterm birth (PTB), induction of labor, mode of delivery and incidence of gestational age (OASIS) were not significantly different. Perinatal outcomes were similar as well.Conclusions: Screening for GDM with one-step, compared with the two-step approach, resulted in a similar incidence of GDM.

Database Quality and Access Issues Relevant to Research Using Anesthesia Information Management System Data.

For this special article, we reviewed the computer code, used to extract the data, and the text of all 47 studies published between January 2006 and August 2017 using anesthesia information management system (AIMS) data from Thomas Jefferson University Hospital (TJUH). Data from this institution were used in the largest number (P = .0007) of papers describing the use of AIMS published in this time frame. The AIMS was replaced in April 2017, making this finite sample finite. The objective of the current article was to identify factors that made TJUH successful in publishing anesthesia informatics studies. We examined the structured query language used for each study to examine the extent to which databases outside of the AIMS were used. We examined data quality from the perspectives of completeness, correctness, concordance, plausibility, and currency. Our results were that most could not have been completed without external database sources (36/47, 76.6%; P = .0003 compared with 50%). The operating room management system was linked to the AIMS and was used significantly more frequently (26/36, 72%) than other external sources. Access to these external data sources was provided, allowing exploration of data quality. The TJUH AIMS used high-resolution timestamps (to the nearest 3 milliseconds) and created audit tables to track changes to clinical documentation. Automatic data were recorded at 1-minute intervals and were not editable; data cleaning occurred during analysis. Few paired events with an expected order were out of sequence. Although most data elements were of high quality, there were notable exceptions, such as frequent missing values for estimated blood loss, height, and weight. Some values were duplicated with different units, and others were stored in varying locations. Our conclusions are that linking the TJUH AIMS to the operating room management system was a critical step in enabling publication of multiple studies using AIMS data. Access to this and other external databases by analysts with a high degree of anesthesia domain knowledge was necessary to be able to assess the quality of the AIMS data and ensure that the data pulled for studies were appropriate. For anesthesia departments seeking to increase their academic productivity using their AIMS as a data source, our experiences may provide helpful guidance.

White Paper: Improving Handoff Culture in Intensive Care Unit to Floor Handoffs

The frequency of handoffs between providers has increased since the 2011 Accreditation Council for Graduate Medical Education (ACGME) work hour restrictions, generating concerns over the quality of these handoffs and their impact on patient safety. At Thomas Jefferson University Hospital (TJUH), the 2016 Safety Culture Survey revealed that across all specialties, many residents felt that “things fall through the cracks” when transferring patients from one unit to another. The interdepartmental Housestaff Quality and Safety Leadership Council (HQSLC) at TJUH sought to improve handoffs at our institution and identified two areas of focus: (1) standardizing the language of handoffs with a commonly accepted handoff technique (IPASS), and (2) standardizing the process of handoffs from the ICU to the floor. Qualitatively, resident comfort with handoffs improved with no adverse impact on time to patient movement between units. This project demonstrated the difficulty of changing the handoff culture at an institution, establishing lasting change via a new EMR system, and training housestaff of a new handoff method. Future directions include monitoring compliance with the new standardized handof f curriculum, and determining whether these efforts and interventions translate to improved patient safety at our institution.

Does the Fetus With Klinefelter Syndrome Have a Disproportionately Long Femur Length at the Anatomy Ultrasound? [9P

INTRODUCTION: To determine if children with a known diagnosis of Klinefelter Syndrome (KS) had a fetal anatomy ultrasound demonstrating long femur length (FL). METHODS: Case series of males with Klinefelter Syndrome (KS) receiving medical care in the pediatric endocrinology division at Thomas Jefferson University Hospital (TJUH) in 2014. We invited the mothers of these male children to participate in our study. Women who agreed to participate signed a medical records release form, which was sent to the delivering obstetrician requesting prenatal and delivery records. Records were reviewed for demographics, biometric measurements from the 18-22 week anatomy ultrasound, and delivery outcomes. Primary outcome was FL (calculated percentile based on Hadlock) from the anatomy ultrasound. RESULTS: In 2014, there were 100 males with KS being cared for at TJUH. 52 women agreed to participate in the study. We received medical records on 40 of these women (77%); of these, 30 records had the anatomy ultrasound report and delivery information. Mean gestational age at the time of the anatomy ultrasound 19.71 weeks (SD ± 1.29 weeks). 41% of women had an amniocentesis and a known prenatal diagnosis of KS. Femur length percentile was computed based on gestational age at the time of the ultrasound; mean FL percentile was 51.6% (SD ± 14.7%, range 8.7%-77.6%). CONCLUSION: The phenotype of males with Klinefelter Syndrome commonly results in tall stature, with a disproportionately long femur from early childhood. However, fetal femur length does not appear to predict an increased risk for Klinefelter Syndrome at the time of anatomy ultrasound.

Physician Recommendations for Celiac Disease Screening in First-Degree Relatives

Introduction: Screening first-degree relatives of patients with celiac disease (CD) offers an opportunity to diagnose CD in a high-risk population. The purpose of this study was to determine if patients with CD are receiving physician recommendations for first-degree relative screening by creating, validating, and distributing a survey to members of the National Foundation for Celiac Awareness (NFCA). Methods: A 12-question online survey was created and sent to consecutive consenting patients who were seen in the Division of Gastroenterology at Thomas Jefferson University Hospital (TJUH) for a visit billed for CD. Two weeks after completion of the survey, patients were sent a second survey containing 4 key questions for validation. These questions focused on CD diagnosis, receiving physician recommendation for relative screening, discussing the recommendation with relatives, and discussion leading to relative screening. Kappa (K) coefficients were calculated for each of the validation questions to measure answer agreement. After completion of survey validation, the 12-question online survey was distributed via email and social media to members of the NFCA, and results were collected over a 3-month period. Results: The survey was piloted with 87 patients at TJUH, and test-retest reliability of the 4 key questions was generally high (test-retest agreement >90% and K >0.80). The main analyses were based on data from 677 NFCA and 82 TJUH respondents with self-reported CD. Respondents were predominantly female (NFCA: 91%, TJUH: 74%, p< 0.001) and were aged 18 to 86 (mean = 45). Compared to NFCA respondents, TJUH respondents were more likely to report a diagnosis of CD by a GI physician (78% vs. 59%, p = 0.001) and through a small bowel biopsy (82% vs. 68%, p = 0.011). The proportion reporting receiving a physician recommendation for relative screening was substantially higher for TJUH than for NFCA (78% vs. 44%, p <0.001). However, among those receiving such a recommendation, almost all reported discussing the recommendation with family members (98% for both surveys). TJUH and NFCA respondents also reported similar rates of relative screening after discussion (TJUH: 71%, NFCA: 79%, p = 0.183) and ultimate CD diagnosis among screened relatives (TJUH: 18%, NFCA: 27%, p = 0.069). Conclusion: The results from this validated survey suggest that a significant percentage of patients with CD are not receiving the recommendation from their physicians for first-degree relative screening. This is a missed opportunity to raise awareness of and diagnose CD in a high-risk population, especially considering that relative screening is a high-impact recommendation resulting in very high rates of family member discussion, screening, and diagnosis.

1072

Introduction: In 2011 we observed that the 28-day mortality rate for sepsis at Thomas Jefferson University Hospital (TJUH) was significantly higher than the rate of the University Health System Consortium (UHC) (34% vs. 21%). We determined that changes would need to be made system-wide and across multiple departments and disciplines to significantly improve outcomes. Methods: An interdisciplinary committee was established which included physicians, nurses, pharmacists, administrators, and support staff from the Emergency Department (ED), the Medical Intensive Care Unit (MICU), Infectious Diseases (ID), and Information Services (IS). The committee examined each phase of sepsis recognition and treatment from first contact in the ED to eventual MICU transfer. Targeted education programs were developed specifically for physicians, nurses, and pharmacy staff. An electronic protocol was created to streamline the recognition and treatment of sepsis. The protocol created new standards for all providers in the care team. Innovations included automatic electronic alerts based on vital signs, First Line Orders (FLOs) for triage nurses, easy-to-use treatment order sets, stocking ED Pyxis® machines with all medications needed in the protocol, and intranet based resources allowing all staff to easily access the protocol. Continual process improvement is imperative to the initiative and the committee meets monthly to review performance. Results: All patients with a diagnosis-related group (DRG) code of sepsis were analyzed for 28-day mortality. Mortality in the six month time period before implementation of the sepsis protocol was 24.9% (n=170) compared with 16.23% (n=289) (p=0.049) in the six month time period post intervention. Reduction in the mortality-index (observed deaths TJUH/expected deaths UHC) from sepsis over was also reduced, from 1.31 in the pre-implentation period to 0.94 (p>0.05). Data on process measures associated with of optimal sepsis care were collected over a shorter two-month time period and compared to a sample of historical controls. Blood cultures were drawn before the administration of antibiotics 57.9% of the time prior to protocol initiation and 96.9% of the time after protocol initiation. The rate of early antibiotic administration increased from 57.9% pre-intervention to 84.4% after protocol initiation. Adequate fluid resuscitation in hypotensive patients increased from 62.5% to 93.8%. Initiation of vasopressors for persistent hypotension increased, from 23.1% to 77.8%. Data collection and analysis is ongoing. Conclusions: We report the successful development and implementation of a protocolized system for improved recognition and management of sepsis. With implementation of a targeted educational program and a protocolized system we were able to increase compliance with key interventions known to decrease mortality and morbidity, particularly aggressive fluid resuscitation and timely antibiotic therapy and ultimately significantly improve our outcomes. We believe that key elements to our success were: 1) Interdisciplinary and interdepartmental involvement of all the stakeholders in sepsis care 2) A targeted educational program emphasizing the importance of early recognition of sepsis and the value of protocolized care. 3) Close collaboration with IS to find new and innovative uses for our current EMRs, including the development of automated sepsis alerts to facilitate early recognition of patients at risk as well as bundled order sets 5) A process of continuous review of performance with regular feedback to practitioners.