Background: Improved post-operative outcomes have been demonstrated in gastrointestinal procedures where a narcotic sparing strategy has been utilized. Data for pancreaticoduodenectomy (PD) patients is limited. This study reviews an institutional database for outcomes based on initial analgesic strategy.Methods: 1004 consecutive patients who underwent PD at Emory University between 2010 and 2017, were included in the analysis. Patients were divided into groups based on primary analgesic strategy employed: epidural alone (EPI), patient controlled opiate analgesia (PCA), dual (dual-PCA/EPI) and other (non-PCA/EPI). Postoperative outcomes for each group were analyzed utilizing univariate and multivariate linear regression.Results: 448 (44.6%) patients were treated with EPI, 300 (29.9%) were given a PCA, 78 (7.8%) had dual-PCA/EPI and 178 (17.7%) had non-PCA/EPI analgesia. On univariate analysis, increased BMI (p = 0.030), PCA use (p < 0.001), venous thromboembolism (VTE) (p < 0.001), post-operative pancreatic fistula (POPF) (p < 0.001) and Ileus/delayed gastric emptying (DGE) (p < 0.001) were all correlated with increased LOS. On multivariate linear regression, VTE (b-coefficient 9.07, p = 0.004) POPF (8.846, p = 0.001), Ileus/DGE (4.464, p = 0.004) and PCA use (1.75, p = 0.003) were associated with significantly increased LOS. Conclusion: A primary narcotic sparing strategy is associated with a significantly reduced LOS and lower rates of Ileus/DGE. Mean opiate usage was significantly lower in the EPI and non-EPI/PCA groups.

BACKGROUND:Morbidity after pancreaticoduodenectomy (PD) has been reported to be about 30-53%. These complications can double hospital costs. We sought to explore the financial implications of complications after PD in a large institutional database. METHODS:A retrospective analysis of patients undergoing PD from 2010-2017 was performed. Costs for index hospitalization were divided into categories: operating room, postoperative ward, radiology and interventional radiology. Complications were categorized according to the Clavien-Dindo classification. Univariable and mutivariable analysis were performed. RESULTS:Median cost of index admission for 997 patients who underwent PD was $23,704 (range $10,988-$528,531). Patients with major complications incurred significantly greater median costs compared to those without ($40,005 vs $21,306, p < 0.001). Patients with postoperative pancreatic fistula (POPF) grade A, B and C had progressively increasing costs ($32,164, $50,264 and $102,013, p < 0.001). On multivariable analysis ileus/delayed gastric emptying, respiratory failure, clinically significant POPF, thromboembolic complications, reoperation, duration of surgery >240 minutes and male sex were associated with significantly increased costs. CONCLUSION:Complications after PD significantly increase hospital costs. This study identifies the major contributors towards increased cost post-PD. Initiatives that focus on prevention of complications could reduce associated costs and ease financial burden on patients and healthcare organizations.

BACKGROUND:The use of neoadjuvant chemotherapy (NAC) for pancreatic ductal adenocarcinoma (PDAC) has increased in recent years. Limited data exists on the impact of NAC on biliary microbiome.METHODS:Patients who underwent pancreaticoduodenectomy (PD) for PDAC between 2014 and 2017 were reviewed. Patients were stratified into two groups based on their NAC status for comparison.RESULTS:Of 168 patients included, 63 (37.5%) received NAC. Patients who received NAC exhibited significantly increased growth of Gram-negative anaerobic bacteria (p = 0.043). Patients in the non-NAC group were more likely to grow pathogens resistant to ampicillin-sulbactam (47% vs 21%, p = 0.007), cefazolin (49% vs 28%, p = 0.040), cefoxitin (42% vs 11%, p = 0.009) and cefuroxime (26% vs 4%, p = 0.019). NAC status did not impact infectious postoperative outcomes, including SSIs.CONCLUSION:Patients who did not receive NAC were more likely to grow pathogens resistant to cephalosporins. Perioperative antibiotic prophylaxis should be tailored to cover Gram-negative organisms and enterococci.

Presenter: Syed Omair Nadeem MD | Emory University Background: Increased utilization of neoadjuvant therapy (NT) for pancreatic adenocarcinoma (PDAC) has allowed for potential downsizing of the primary tumor, improved patient selection for curative resection and has demonstrated a safe and efficacious effect on survival of these patients. However, there exists limited data on patient factors that influence pathologic response in patients presenting with PDAC. We sought to explore the effect of diabetes and smoking on the pathologic response of tumors post NT. Methods: Patients with biopsy confirmed PDAC who received NT and underwent pancreaticoduodenectomy between January 2010 and December 2019 at a tertiary care center were reviewed. Surgical pathology reports were reviewed to determine treatment effect. Pathologic tumor response was assessed based on the College of American Pathologists (CAP) grading system. Grades 0-2 were defined as showing a treatment response while grade 3 was defined as no treatment response. Results: Overall 180 patients underwent pancreaticoduodenectomy at our institution. FOLFIRINOX was the most commonly employed neoadjuvant regimen (n=105, 58.3%) followed by Gemcitabine combination chemotherapy (n=68, 37.8%), while the rest got other treatment regimens. SBRT was utilized in 38 (21.1%) patients, while CRT was utilized in 17 (9.4%) patients. 20 patients (11.0%) had a complete or near complete pathologic response (CAP grade 0, 1 respectively), 53 (29.4%) patients had a partial response (CAP grade 2) and 107 (59.4%) patients had poor or no response to NT (CAP grade 3). Median overall survival was 13.6 months. On univariate analysis for factors predicting pathologic response, a diagnosis of Diabetes was associated with a significantly increased chance of poor or no pathologic response (p=0.033). On multivariate regression analysis, Diabetes again was culpable for predicting poor pathologic response (OR=0.262; 0.109-0.633, p=0.003). No other patient factors including age, gender, BMI, ethnicity, tobacco-use or other comorbidities and neoadjuvant treatment options (FOLFIRINOX vs Gemcitabine based therapy) were found to be significant for predicting pathologic tumor response. Despite Diabetes being associated with poor pathologic response, there were no differences in overall survival between diabetics and non-diabetics using Kaplan-Meier survival estimates (Figure1). Conclusion: Our results demonstrate that a diagnosis of Diabetes was associated with a significantly increased chance of predicting poor pathologic response to neoadjuvant therapy. This warrants the need for further discussion and randomized controlled trials to study patient factors which influence pathologic response to neoadjuvant therapy, which can help improve patient selection for neoadjuvant treatment.

Minimally invasive approaches to major liver resection have been limited by presumed difficulty of the operation. While some concerns arise from mastering the techniques, factors such as tumor size and liver parenchymal features have anecdotally been described as surrogates for operative difficulty. These factors have not been systematically studied for minimally invasive right hepatectomy (MIRH). Seventy-five patients who underwent MIRH during 2007–2016 by the senior author were evaluated; these were compared to control group of open right hepatectomy. Demographics, operative, and post-operative variables were collected. Operative times and estimated blood loss, two objective parameters of operative difficulty were correlated to volume of hepatic resection, parenchymal transection diameter and liver parenchymal features using regression analysis. Thirty-eight (50.6%) resections were performed for malignant indications. Average tumor size was 5.7 cm (±3.6), mean operative time was 196 min (±74), and mean EBL was 220 mL (±170). Average transection diameter was 10.1 cm (±1.7). There was no correlation between operative difficulty with parenchymal transection diameter or presence of steatosis. Blood loss was higher with increased right hepatic lobe volume and body mass index. This analysis of a very defined anatomical resection suggests that the often quoted radiographic and pathologic features indicative of a challenging procedure were not significant in determining operative difficulty.

Background Optimal diagnostic and surgical approaches for patients with bile duct injuries (BDI) remain debated. This study reviews results from a standardized approach to management of high-grade BDIs at a North American center. Design Patients undergoing surgical repair for BDIs over a 15-year period were included. Post-operative outcomes and biliary patency rates were calculated using imaging, laboratory values, and patient interviews. Results A total of 107 consecutive patients underwent repair for BDIs. Bismuth grade I/II injuries were identified in 46 patients (41%), grade III/IV in 41 (38%), grade V in 11 patients (10%), and 9 (10%) were unclassified. BDI anatomy was commonly identified using magnetic resonance imaging (MRI) (75%). Concomitant arterial injuries were identified in 30 (28 with formal angiography). Fifteen had early repairs (within 4 days) and remainder interval repairs (median: 65 days). Hepp-Couinaud repair was method of choice (83%). Estimated primary biliary patency was 100% at 30 days and 87% at 5 years. Conclusion With appropriate referral to a specialist, surgical reconstruction of BDIs can have excellent outcomes, even with accompanying arterial injuries. Based on our experience, MR as first imaging modality and supplemental angiography served as the optimal diagnostic strategy. Delayed repair, using Hepp-Couinaud technique, with selective liver resection results in high long-term patency rates.

Presenter: Syed Omair Nadeem MBBS | Emory University Background: Postoperative infectious complications constitute a major source of morbidity in pancreatic surgery. For patients with pancreatic ductal adenocarcinoma (PDAC), the use of neoadjuvant chemotherapy is becoming more commonly adopted. Yet data is limited regarding its impact on the biliary microbiome. It has been hypothesized that the currently recommended peri-operative antibiotic coverages may be inadequate prophylaxis for patients undergoing pancreatic resection. The aim of this study was to determine the effect of neoadjuvant therapy on the biliary microbiome in patients with and without preoperative stenting as a surrogate for overall patient microbiome status Methods: Demographic, operative, post-operative, and biliary pathogen morphology and antibiotic sensitivity data was reviewed from consecutive patients with pathologically confirmed PDAC from January 2014 to December 2017. Data was expressed as percentages. Chi square and ANOVA testing was performed for comparison between groups. Results: A total of 263 patients with PDAC were identified. 258 cases with complete stent data were included in the analysis. Of them 199 (77.1%) had received preoperative stenting, amongst whom 71 (35.7%) had received neoadjuvant chemotherapy. Ancef/Flagyl (n=167, 84%) was the most common antibiotic combination for perioperative prophylaxis followed by Clindamycin/Aztreonam (n=16, 8%). There were no significant differences in baseline characteristics between the people who did and did not receive neoadjuvant chemotherapy. The most common pathogens isolated from biliary cultures were Klebsiella Pneumoniae (n=31, 18.5%), Enterobacter Clocae (n=20, 11.9%) and E.Coli (n=16, 9.5%). Patients who did not receive neoadjuvant chemotherapy were more likely to grow, fungi (5.5% vs 0% p=0.039), E.Coli and Enterobacter (25% vs 12.7%, p=0.039). However, growth of Gram-negative anaerobes (32.4% vs 18.8%, p= 0.030) and unspecified bacterial species (31% compared to 14.8%, p=0.007) was higher in the patients who received neoadjuvant chemotherapy. Patients who did not receive neoadjuvant therapy were more likely to grow gram-negative pathogens resistant to Ampicillin-Sulbactam (47% vs 21%, p=0.007), Cefazolin (49% vs 28%, p=0.040), Cefoxitin (42% vs 11%, p=0.009) and Cefuroxime (26% vs 4%, p=0.019) than the patients who did receive neoadjuvant chemotherapy. No differences were seen in resistance to Ampicillin (p=0.376), Aztreonam (p=0.977), Cefepime (p=0.172), Ceftriaxone (p=0.183) and Piperacillin-Tazobactam (p=0.36) (figure 1). Stratified by stent status, there were no differences in resistance patterns among patients who did and did not receive neoadjuvant chemotherapy. Conclusion: Neoadjuvant chemotherapy significantly alters the biliary microbiome in patients with PDAC undergoing pancreaticoduodenectomy. Patients who did not receive neoadjuvant chemotherapy were more likely to grow pathogens resistant to cephalosporin antibiotics and have fungal colonization of their bile compared to patients who did receive neoadjuvant chemotherapy. Perioperative antibiotic coverage should be tailored to cover enterococci and gram negative organisms, especially for the subset of patients who had a biliary stent placed

Jajja, Raheel MD; Omair Nadeem, Syed MBBS; Lovasik, Brendan MD; Sharma, Jyoti MD, FACS; Sarmiento, Juan M. MD, FACS Author Information

Omair Nadeem, Syed MBBS; Jajja, Raheel MD; Scott Davis, S. MD, FACS; Sharma, Jyotirmay MD, FACS; Manuel Sarmiento, Juan MD, FACS Author Information

Objectives The aim of this study was to establish multidisciplinary care for patients with transfusion-dependent thalassaemia (TDT) by creating a TDT quality improvement (QI) collaborative in a resource-constrained setting. This study presents our initial experience of creating this collaborative, the baseline characteristics of the participants, the proposed QI interventions and the outcome metrics of the collaborative. Design and setting TDT QI collaborative is a database comprising patients with TDT from four centres in Karachi, Pakistan. Study variables included symptoms of cardiac or endocrine dysfunction, physical examination including anthropometry and Tanner staging, chelation therapy, results of echocardiography, T2* cardiac MRI (CMR) and serum ferritin. The main outcome of this collaborative was improvement in TDT-related morbidity and mortality. Interventions addressing the key drivers of outcome were designed and implemented. Results At the time of reporting, the total number of patients in this database was 295. Most patients reported cardiac symptoms corresponding to New York Heart Association class 2. Approximately half (52%, n=153) of the patients demonstrated severe myocardial iron overload (T2* <10 ms). Majority of the patients (58%, n=175) were not on adequate chelation therapy. There was no difference in echocardiographic measures of systolic and diastolic left ventricle among the different spectrums of iron overloaded myocardium. Conclusion Using T2* CMR and endocrine testing, we have identified significant burden of iron siderosis in our patients with TDT. Lack of adequate iron load assessment and standardised management was observed. Interventions designed to target these key drivers of outcome are the unique part of this QI-based TDT registry.

1Medical College, AgaKhanUniversity, Karachi, Pakistan 2Department of Pediatrics andChildHealth, AgaKhanUniversity, Karachi, Pakistan 3AfzaalMemorial Thalassemia Foundation, Karachi, Pakistan 4Laboratory andClinical Department, Fatimid Foundation, Karachi, Pakistan 5Kashif Iqbal ThalassemiaCareCentre, Karachi, Pakistan 6PaediatricDepartment, SalisburyDistrict Hospital, Salisbury, UnitedKingdom 7Department of Pediatrics andChildHealth, AgaKhanUniversity, Karachi, Pakistan Correspondence SalmanKirmani,DepartmentofPediatrics and ChildHealth,AgaKhanUniversity, Stadium Road,Karachi 74800,Pakistan. Email: salman.kirmani@aku.edu Abstract Background: Endocrinopathy due to iron overload is the most common morbidity whereas myocardial siderosis causing toxic cardiomyopathy is the leading cause of mortality among patients with transfusion dependent thalassemia major (TDTM). If detected early, this can be treated with aggressive chelation. T2* cardiac magnetic resonance imaging (CMR) guided chelation protocols are now the gold standard but have limited availability in low and middle-income countries.Wehypothesized thatmarkers of endocrine dysfunctionwould correlatewith T2*CMR and can be used to predict the severity of myocardial siderosis and guide chelation therapy.

BackgroundEndocrinopathy due to iron overload is the most common morbidity whereas myocardial siderosis causing toxic cardiomyopathy is the leading cause of mortality among patients with transfusion dependent thalassemia major (TDTM). If detected early, this can be treated with aggressive chelation. T2* cardiac magnetic resonance imaging (CMR) guided chelation protocols are now the gold standard but have limited availability in low and middle-income countries. We hypothesized that markers of endocrine dysfunction would correlate with T2* CMR and can be used to predict the severity of myocardial siderosis and guide chelation therapy. MethodologyWe undertook a multicenter retrospective study of 280 patients with TDTM to assess the prevalence of endocrinopathies and the predictive value of a number of individual and composite markers of endocrinopathy with T2* CMR. ResultsThe prevalence of hypogonadism, stunting, hypoparathyroidism, and hypothyroidism was 82%, 69%, 40%, and 30%, respectively. The sensitivity of hypogonadism and stunting predicting severe myocardial siderosis was 90% and 80%, respectively. ConclusionWe conclude that clinical markers of endocrine dysfunction, especially hypogonadism (positive likelihood ratio [LR+]=1.4, 95% confidence interval [CI]=1.0-1.9; positive predictive value [PPV]=77%, 95% CI=70-82; negative predictive value [NPV]=57%, 95% CI=34-77] and stunting (LR+=1.3, 95% CI=1.1-1.6; PPV=64%, 95% CI=60-69; NPV=55%, 95% CI=45-64) in TDTM can predict severe myocardial siderosis and can potentially guide chelation therapy, especially where access to T2* CMR is limited.