93.14 Aortoiliac Thromboembolic Events in Younger Women: A Retrospective Case Review

Posted on January 18, 2018

M. P. Kochuba1, E. Choi1, F. Schmieder1, R. Dhanisetty1, P. Van Bemmelen1, S. Golarz1  1Temple University Hospital,Department Of Surgery,Philadelphia, PA, USA

Introduction: Aortoiliac occlusive disease is well-defined in the literature and usually affects older individuals with a history of smoking, HTN, HLD, and DM.  The disease is often progressive and can be asymptomatic or can present with claudication, rest pain or tissue loss (when combined with disease at other levels).  Acute presentation tends to be extremely morbid.  Acute aortic occlusion has mortality up to 50% and acute iliac occlusion and/or embolization can result in threat to life and limb.  Again, these are typically seen in much older patients.   We have noted a population of much younger female patients presenting with acute aorto-iliac pathology with devastating outcomes.  We have identified a cohort of these individuals and will discuss their presentation, demographics, individual risk factors, treatment algorithms, and outcomes.

Methods: We conducted a retrospective case review of all young females meeting an age criteria of 18-55 years, who presented to the vascular surgery service at our institution between the years 2014-2016 with aortoiliac thrombosis or embolization. Co-mobordities (smoking status, BMI, HLD, hypercoagulable state), vessel characteristics (size, atherosclerosis) and race were collected for evaluation. Data analysis was performed with SPSS data software. Significance was assumed for p<0.05.

Results: Nine patients were identified that met our selected criteria, the mean age was 44 with a BMI of 34, 67% white 33% AA. All received operations. The most common comorbidity was smoking which was observed in 100% of the patients identified. Other comorbidities that were noted include PVD/presence of plaque on CT in 89%, elevated TG, total cholesterol and LDL in 33%, and elevated HDL in 11%.  No distinct hyper-coagulable conditions were identified. A statistically significant difference in mean infrarenal aorta diameter was noted, 14.86mm (SD2.72mm) with a literature average of 19.3mm (SD2.5mm), p 0.001. A trend toward smaller mid common iliac diameter was also noted, with right/left diameters 8.8mm(SD1.75mm)/9.3mm(SD2.37mm) and an average literature diameter of 10.4mm(SD1.9mm) with p values of 0.087/0.230.  Of the 9 patients in this review, 5 avoided amputations with the other 4 having amputations ranging from digital to BKA. No mortalities occured. 

Conclusion: Aortoiliac thromboembolic events in younger women may be a distinct variant of aortoiliac occlusive disease.  This variant appears to have a much more aggressive presentation with high morbidity.  Aggressive surgical management is warranted but outcomes tend to be poor.  After completion of surgical interventions we recommend strict lipid control and strict abstinence from tobacco.  All patients should remain on anticoagulation indefinitely, if possible, due to the devastating consequence of reoccurrence.  Although sample size is small, diminutive vessel size may contribute to these events.  Further study is needed. 

 

93.15 Characterization of the Strength of the Genetic Component Associated with Abdominal Aortic Aneurysms

Posted on January 18, 2018

R. Vissepo1, B. Shue2, J. Hinson3, D. Melzer4, J. E. Indes2  1University Of Connecticut School Of Medicine,Farmington, CT, USA 2University Of Connecticut Health Center,Department Of Surgery,Farmington, CT, USA 3University Of Connecticut Health Center,Department Of Cardiology,Farmington, CT, USA 4University Of Connecticut Health Center,Center On Aging,Farmington, CT, USA

Introduction:  Abdominal aortic aneurysm (AAA) is a common medical condition that contains a genetic and inheritable component. A number of different genes, including genes of the extracellular matrix (ECM) and the immune system, have been identified and found to be associated with the development or progression of AAAs. The aim of this study was to create a database and characterize the influence of candidate genes to help describe the inherited component of AAA. 

Methods: A systematic search of the PUBMED database using the search terms “AAA” and “Gene” was performed to identify candidate genes contributing to the etiology of abdominal aortic aneurysms. Inclusion criteria included single studies and review articles in which the focus was the association between genes and AAA. Exclusion criteria included duplicates, editorials, isolated animal studies and studies whose main focus was not the genetic component of AAA. The results from the studies were collected and analyzed in order to determine their quality. Genes were considered “strong” if the majority of the reviewed studies determined a genetic association with AAA. The genes in which the majority of the studies determined that no genetic association existed were considered “weak” genes. Genes placed in the “moderate” category only had one study or review article which showed a genetic association with AAA. The genes were gathered from genome-wide association studies and from genetic association studies, using sequence variants, on patients with AAA and controls.

Results: A total of 9 review articles and 12 studies were obtained from the systematic search. The search identified 96 gene candidates, including sequence variants. After assessment studies, the number of genes associated with abdominal aortic aneurysm amounted to 55 genes, while 41 genes were found to have no association with AAA. Of the genes that were considered to have an association with AAA, 1 was considered to have a strong association, cyclin-dependent kinase inhibitor 2B antisense RNA (CDKN2BAS), 22 genes were considered to have a moderate association, while 32 of the genes were considered to have a weak association. After an analysis of the different population sets associated with the strong candidate gene, CDKN2BAS, the relative risk was determined to be 1.143 (95% CI, 1.102-1.184). 

Conclusion: The characterization of the genes that have known associations with AAA may be useful in future investigations in order to identify new genes and mechanisms in which abdominal aortic aneurysms occur. The identification of strong candidate genes, such as CDKN2BAS, may help guide studies on the inherited and genetic component of abdominal aortic aneurysms.

 

93.04 Variability of Pre-operative Toe Pressures in Patients with Chronic Limb-threatening Ischemia (CLI)

Posted on January 18, 2018

A. M. Flores1, M. Fakhari1, J. Darling1, M. L. Schermerhorn1,2, R. J. Guzman1,2  1Beth Israel Deaconess Medical Center,Vascular Surgery,Boston, MA, USA 2Harvard School Of Medicine,Boston, MA, USA

Introduction: In patients with chronic limb-threatening ischemia (CLI), assessment of pedal perfusion is often complicated by non-compressible ankle arteries resulting in false elevation of the ankle-brachial index (ABI). When the ABI cannot be calculated, toe pressures are often used as a surrogate measure. However, the utility of toe pressure measurements has been questioned. We thus sought to investigate the variability of pre-operative toe pressures, whether they correlated with preoperative cardiovascular risk factors, and if they predicted long-term outcomes in patients with CLI.

Methods: Consecutive patients presenting with signs and symptoms of CLI between 2010 and 2015 were identified by review of the institutional electronic medical record. All patients who underwent pre-operative vascular lab assessment by arterial Doppler with toe pressure measurements were then used for the present study. Demographic and cardiovascular risk factors were recorded as well as results from pre-operative, non-invasive arterial studies performed within 1 month of a procedure on the index limb. Extent of occlusive disease was assessed by determining TASC classifications for the femoral-popliteal and tibial levels on pre-operative imaging studies. A composite TASC score of the index limb was derived by adding the two values. The index limb, interventions, and outcomes including amputation and survival were identified by review of the medical record. Toe pressures from the index limb were used for all statistical analyses.

Results:  We identified 124 patients that met inclusion criteria. Of these, 83% had diabetes, 54% had a history of tobacco use, and 21% had a history of renal failure requiring dialysis. Symptoms of ischemic rest pain were present in 15%, ulcers in 68%, and gangrene in 17% of patients. The mean (standard deviation) for toe pressure measurements was 32 (29.5). The median (range) was 29 mm Hg (0-119). Of the 124 index limb toe pressures, 21 (17%) were over 60, 10 (8%) were over 80 mm Hg, and one patient had non-compressible toe vessels. There was no correlation identified between toe pressures and age or cardiovascular risk factors; however, toe pressures were lower in patients on hemodialysis (p=0.023), and in those with higher TASC scores (p=0.003). In outcomes analyses, pre-operative toe pressures did not predict major amputation or 3-year survival.

Conclusion:  In a small cohort of patients with CLI, pre-operative toe pressures demonstrated significant variability and did not correlate with demographic or cardiovascular risk factors, but they were lower in patients with renal failure and those with more extensive occlusive disease. They did not predict amputation or long-term survival.

93.05 Ultrasound Guided Foam Sclerotherapy: Does Size Matter?

Posted on January 18, 2018

J. T. Toniolo1,2, D. Munteanu1, A. Russell1, N. Chiang1, H. Hao1, J. Chuen1,2  1The Austin Hospital,Department Of Vascular Surgery,Melbourne, VICTORIA, Australia 2The University Of Melbourne,Department Of Surgery,Melbourne, VICTORIA, Australia

Introduction:
Ultrasound-guided foam sclerotherapy (UGFS) is a low-cost and minimally invasive treatment option for varicose veins. There is a relative paucity of outcome reports.

Methods:
UGFS at a tertiary hospital between 2010 and 2017 were studied. Treatments were either the greater saphenous vein (GSV) or short saphenous vein (SSV). Pre-treatment, in-treatment, 6-week and 1-year post-treatment ultrasound reports were analysed. Vein diameters were measured at 3cm below the respective junctions. Primary outcome was to determine if vein diameter would predict recanalisation.

Results:

456 treatments were completed in 290 patients. 372 (81%) targeted the GSV. Mean vein diameters of the GSV were smaller than SSV [5.5 +/- 1.9mm (GSV), 6.2 +/- 2.6mm (SSV); p=0.024]. 109 (24%) had CEAP of at least 4.

 

At six weeks, complete obliteration, partial recanalisation and complete recanalisation rates for GSV were 55%, 29% and 13% respectively. Of the 251 veins with complete obliteration at six-weeks, partial and complete recanalisation at one-year were 21% and 26%, respectively. Regarding the SSV system, complete obliteration, partial recanalisation and complete recanalisation at 6-weeks were 51%, 30% and 13%, respectively. Similarly, partial and complete recanalisation at one-year were 26% and 18%, respectively. Using Student-t test, increasing vein diameter was associated with partial or complete recanalisation at 6 weeks [5.4 +/- 2.0mm (obliteration) cf. 5.9 +/- 2.2mm (recanalisation); p=0.036] and 1 year [4.7 +/- 1.7mm (obliteration) cf. 5.7 +/- 2.0mm (recanalisation); p=0.036). Vein diameter of greater than 6mm was not associated with recanalisation (p=0.07). 15 (3%) new onset DVTs, and 11 (2%) injection-related ulcers reported on follow-up.

Conclusion:
More than 50% achieved complete obliteration at 6 weeks. Significant sonographic recanalisation developed at one-year. Increased vein diameter was associated with recanalisation. Their impact on clinical recurrence is unknown.

92.19 Comparative analysis of pancreatic cancer clinical pathways

Posted on January 18, 2018

S. Willis5, A. Ardeljan1,3, D. Maurente1,5, A. Johns1, H. Abdul1, M. Bustos1,3, S. Susie2, M. Ghali1, O. M. Rashid1  1Holy Cross Hospital,Michael and Dianne Bienes Comprehensive Cancer Center,Fort Lauderdale, FL, USA 2Massachusetts General Hospital,Cancer Center,Boston, MA, USA 3University Of Miami Miller School Of Medicine,Miami, FL, USA 4Charles E. Schmidt College Of Medicine At Florida Atlantic University,Boca Raton, FL, USA 5Nova Southeastern University College Of Osteopathic Medicine,Fort Lauderdale, FL, USA

Introduction:  There is no consensus combination therapy clinical pathway (CP) based on randomized prospective studies for the management of borderline resectable pancreatic adenocarcinoma (BPA) and locally advanced pancreatic cancer (LAPC). Therefore, BPA and LAPC CP's vary by institution. We recently reviewed outcomes of different CP‘s and now seek to compare results to further evaluate optimal therapy. 

Methods:  We performed a literature review of all BPA and LAPC CP reports, BPA diagnostic criteria, pancreatectomy rate, margin status, pathologic response (grade IIa-IV) rate (PR), disease free (DFS), disease specific (DSS), and overall survival (OS) were reviewed and compared. Standard statistical methods were used for statistical comparison. 

Results: There were 9 studies which met inclusion criteria. 2,745 entered 9 CP’s. 1,750 met NCCN/AHPBA diagnostic criteria, 144 MDACC and 257 Surgeon Judgement. Of these patients, 21.2% – 93% completed neoadjuvant therapy. The neoadjuvant regimens included: 5-Fluorouracil, Paclitaxel, Cabecitabine, Gemcitabine, Cisplain, Folfirinox, XRT, EBT, IORT, GT, GTX, SBRT, and/or CHXRT. The resection rate was 38%-90%, R0 resection rate was 35-97%, pathological response for GII, GIII, GIV was 14.8%-57%, 14%-82% and 0%-17%, respectively. Median follow up was 11-27 months. Median overall survival after resection was from 8-40 months, median disease free survival was 17-24 months. Without resection median overall survival was 11-24 months, and 8-15 months after metastasis. 

Conclusion: Comparative analysis demonstrates that many BPA and LAPC patients benefit from the reported CP’s. However, the differences in reported resection and pathologic response to treatment rates between institutional CP’s warrant further investigation to determine optimal therapy.
 

93.02 Time to Complications after Endovascular Repair of Abdominal Aortic Aneurysm: A 3 Years Analysis

Posted on January 18, 2018

V. Pandit1, M. N. Khan1, M. Zeeshan1, M. Hamidi1, F. S. Jehan1  1University Of Arizona,Tucson, AZ, USA

Introduction:
The rates of complications following endovascular repair (EVAR) of abdominal aortic aneurysm (AAA) are well known; however, the data regarding timing are opaque. Accordingly, we sought to assess the median time-to-event for principal postoperative complications within 30 days following surgery.

Methods:
Patients undergoing EVAR of AAA were identified within the American College of Surgeons National Surgical Quality Improvement Program database (2013–2015). Primary endpoint was time-to-complication. Secondary endpoints included length-of-stay (LOS), re-admission and 30-day mortality. Multivariable regression models assessed the predictors for pre-/post-discharge complications and the effect of time-to-complication on secondary outcomes.

Results:
Overall, 12,279 patients underwent EVAR of AAA. The overall complication rate was 27%. 23.1 % patients had major complications while 3.9% patients suffered from minor complications. The median LOS was 2[1-3] days and patients were discharged on post-op day 1[1-3]. The majority of major complications (88.1 %), including bleeding / transfusion, renal, septic, deep venous thrombosis or pulmonary embolism, pulmonary, cardiac and neurologic, occurred prior to discharge. Conversely, the relatively minor complications, including wound and urinary tract infections, occurred predominantly post-discharge (70.7 %). The median time to major complications was 3.7 versus 14 days for minor complications. In multivariable analyses, age (p < 0.01), American Society of Anesthesiologists score ≥ 3 (p < 0.01), while male gender (p < 0.01), hypertension (p < 0.01) and diabetes (p < 0.01) were predictors of post-discharge complications. For a given complication, time-to-complication did not affect the odds for mortality (p = 0.45) or re-admission (p = 0.79).

Conclusion:
Approximately one in four patients suffers a complication following endovascular repair of abdominal aortic aneurysm; major complications tend to occur early with the majority occurring pre-discharge. Knowledge regarding the timing and risk factors for complications may facilitate improved patient–physician communication, both at admission and at discharge.
 

92.18 Pancreaticodudenectomy for Periampullary Carcinoid tumors: Outcome analysis using NSQIP.

Posted on January 18, 2018

K. Memeh1, H. Aziz1, T. S. Riall1, T. Jie1  1University Of Arizona,Surgery,Tucson, AZ, USA

Introduction:
Very few data exist on the outcome of patients with periampullary carcinoid. There are suggestions in the literature that these tumors have a different outcome from other small bowel carcinoid after surgical resection, however most of these data are from institution-based case series.The aim of this study is to describe the outcomes of patients who underwent pancreaticodudenectomy for periampullary carcinoid tumors using a national database.

Methods:

We queried the NSQIP database for years 2008 – 2012 inclusive and identified all patients who underwent pancreaticodudenectomy for periampullary carcinoid tumor. Patient demographics,peri-operative data, post-operative morbidity and 30-day mortality were obtained and analysed. 

Results:

14,977 patients who underwent pancreaticodudenectomy were identified from the NSQIP for years 2008-2012. Of these, 150 patients underwent pancreaticodudenectomy for a carcinoid tumor. The mean age was 57.12 +/-12 years, and 54% of the patients were of female gender.75% were performed in an open fashion. Overall morbidity was 29% and 30-day mortality was 2%. Pancreatic fistula rate was 15%.

Conclusion:

Our study highlights outcomes of pancreaticodudenotomy for periampullary carcinoid tumor using a national database which has not been described before in literature.Within the limitations of the study, morbidity and mortality indices are described in the hope that it may add to an update to existing literature as well as serve as a guide to patient education regarding potential outcome following surgical resection of ampullary carcinoid.
 

92.15 Satistfaction with Sedation in Colonoscopy: A Systematic Review and Meta-Analysis

Posted on January 18, 2018

F. Dossa1, B. Medeiros2, S. A. Acuna1, N. N. Baxter1  1University Of Toronto,Division Of General Surgery,Toronto, ON, Canada 2Western University,Department Of Biology,London, ON, Canada

Introduction:
The use of propofol for sedation in colonoscopy adds considerable expense to the procedure. Proponents of propofol defend its use through claims of increased patient and endoscopist satisfaction and greater procedural efficiency; however, these outcomes have not been formally compared between propofol and the commonly used alternative combination of midazolam/fentanyl. We aimed to compare satisfaction and efficiency outcomes between these agents.

Methods:
We systematically searched MEDLINE, Embase, and the Cochrane Library (to March 28, 2017) to identify randomized controlled trials comparing outcomes between propofol +/- short-acting opioids (fentanyl, remifentanil, alfentanil) and midazolam/fentanyl used for colonoscopy. Outcomes included post-procedure and delayed recall of patient pain and satisfaction, endoscopist satisfaction, procedural time, and recovery time. Standardized mean differences (SMD) were meta-analyzed using random effects models.

Results:
We identified 5 studies reporting at least one satisfaction or efficiency outcome of interest. There were no statistically significant differences in post-procedure (SMD 0.44; 95% CI: -0.03, 0.90) or delayed assessment (SMD 0.39; 95% CI: -0.34, 1.11) of patient satisfaction between propofol and midazolam/fentanyl. Similarly, there were no significant differences in post-procedure (SMD -0.31; 95%CI -0.88, 0.26) or delayed recall (SMD -0.12; 95%CI -0.33, 0.08) of patient pain between midazolam/fentanyl and propofol. We were unable to meta-analyze endoscopist satisfaction due to a lack of studies reporting this outcome. Procedural time was shorter with propofol use (SMD -0.22; 95%CI: -0.40, -0.04); however, the absolute differences in procedure time within studies ranged from only 0.5 to 1.7 minutes. There were no significant differences in recovery time between sedative combinations. Notably, for all outcomes assessed, we found greater differences between than within studies, suggesting a greater effect of endoscopic technique or depth of sedation used than sedative administered.

Conclusion:
Despite the beliefs of greater patient satisfaction and procedural efficiency with propofol versus midazolam/fentanyl in colonoscopy, we did not find evidence to support these claims. Given the large between study differences observed in this study, satisfaction and efficiency may depend more greatly on endoscopist technique and depth of sedation used than on sedative selection. Given the increased cost associated with propofol use, these findings should be considered when selecting sedative agents for lower gastrointestinal endoscopy. 

 

92.12 Effectiveness of Liposomal Bupivacaine in Ostomy Reversal: A Retrospective Review

Posted on January 18, 2018

D. G. McKeown1, C. Sokas1, A. Nevler1, S. Goldstein1, G. Isenberg1, B. Phillips1  1Thomas Jefferson University Hospital,Colorectal Surgery,Philadelphia, PA, USA

Introduction:

Our objective was to assess the clinical efficacy of Liposomal Bupivacaine (LB) in patients undergoing ileostomy and colostomy reversal and its effect on average length of stay

Methods:

We conducted a single institution retrospective review of consecutive patients undergoing elective reversal and closure of either an ileostomy or colostomy from January 2012 to December 2016. Liposomal bupivacaine was approved for use at our institution between May 2013 to September 2016. The primary outcomes were postoperative hospital length of stay (LOS) and cumulative opioid usage calculated as morphine equivalents (MEQ)

Results:

A total of 154 patients were evaluated. 87 patients received LP and 67 patients received a standard dose of local anesthetic. There was no significant difference between the two cohorts. The mean length of stay for the non-LB group was 4.27 days compared to 3.45 days in the LB group (p=0.009). We then defined early discharge as a discharge less or equal to three days and we found that patients who received LB were more likely to receive an early discharge with an odds ratio of 2.1 (CI 1.13 – 4.13) p=0.23. Cumulative opioid use 78.8 morphine equivalents (meq) in the non-LB group versus 75.7 in the LP group (p=0.66). 32.8% of the LP group received a PCA vs 25.3% of the non-LBA group and there was no statistical difference in LOS between these two groups (p=0.36) Secondary outcomes looked at the effect of non-opioid analgesia and anti-emetics on length of stay. When we examined anti-emetic usage, we noted that patients who did not require the administration of anti-emetics were more likely to be discharged early (p=0.05, OR = 0.37 (CI = 0.19 – 0.72))  

Conclusion:

The usage of LB for local wound anesthesia after colostomy and ileostomy reversal is associated with decreased LOS, however opioid usage between the non-LB and LB groups were similar. Despite this, LB appears to offer the benefit of decreasing costs associated with longer hospital stay. A multi-institutional prospective randomized control trial would help to elucidate further. 

 

 

92.14 Acute Diverticulitis: Evolving trends in operative versus non-operative management

Posted on January 18, 2018

B. Zangbar1, S. Imtiaz1, V. Roudnitsky1, L. Dresner1, H. Talus1, A. Schwartzman1, F. Serafini1, R. Gruessner1  1State University Of New York Downstate Medical Center,Surgery,Brooklyn, NY, USA

Introduction:  Diverticular disease is among the most prevalent diseases in United States and imposes a vast healthcare burden to population. Treatment of diverticulitis has evolved over time with reportedly increasing primary anastomosis and non-operative treatment. The aim of our study was to assess these changes in patterns of practice over an 8 year period using a national database.

Methods:  We used the 2004 to 2011 nationwide inpatient sample (NIS) database to analyze the care received by 412,163 patients admitted with acute diverticulitis, and 47,574 patients operated for diverticulitis. Patient’s with Left hemicolectomy, Colostomy, Ileostomy, and percutaneous abscess drainage were identified and Jonckheere-Terpstra trend analysis and logistic regression analysis was used.

Results: Overall there was an increase in rate of admission for acute diverticulitis from 2004 to 2011. There was a decrease in rate of operations for diverticulitis from 5,832 (12.5%) to 5,823 (10.2%) per year during the same time period. The rate of primary anastomosis is decreased from 2,306 (5.0%) in 2004 to 1,660 (2.9%) in 2011 while patients admitted with acute diverticular disease were relatively younger during the same time period (64±16 in 2004 vs 63±16 in 2011, p=0.001). Gender, Race, and comorbidities was unchanged. Mortality during hospitalization decreased from 1.8% in 2004 to 1.5% in 2011.

Conclusion: Our study shows that non-operative management is clearly increasing over time with one main exception. In those that did receive operations  there is a significantly decreased rate of primary anastomosis. The cause of this change, contrary to the prior trend, is unknown. Further research is needed to investigate the role of non-operative management and more selective operative management for primary anastomosis.

 

92.10 Predictors of Conversion from Laparoscopic to Open Lysis of Adhesion for Small Bowel Obstruction

Posted on January 18, 2018

D. Asuzu1, G. Chao1, K. Y. Pei1  1Yale University School Of Medicine,General Surgery,New Haven, CT, USA

Introduction:
Laparoscopic lysis of adhesion (LLOA) for small bowel obstruction (SBO) is associated with shorter operative times and lower complication rates compared to open lysis of adhesion (OLOA). However, intra-operative conversion from LLOA to OLOA still occurs even at experienced centers. Factors associated with this conversion remain poorly understood. 

Methods:

We retrospectively analyzed data from 9,920 patients undergoing OLOA (CPT 44005) for SBO (ICD10 560.81 and 560.9) in the prospectively collected American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) database from 2005 to 2013. 92 cases were converted from LLOA to OLOA. Factors independently associated with conversion were identified using multivariable logistic regression adjusting for age, gender, weight and pre-operative functional status.

Results:
Age and weight were significant risk factors for conversion from LLOA to OLOA, with adjusted odds ratio (OR) per decade of 0.86, 95% confidence interval (CI) 0.76 – 0.97, P = 0.017; and adjusted OR per 50 lbs of 1.24, 95% CI 1.01 – 1.52, P = 0.048 respectively. Other factors significantly associated with conversion from LLOA to OLOA included pre-operative albumin, (adjusted OR 1.60, 95% CI 1.16 – 2.22, P = 0.005), aspartate aminotransferase (AST, adjusted OR 1, 95% CI 1 – 1.01, P = 0.038) and white blood cell (WBC) count (adjusted OR 0.93, 95% CI 0.88 – 0.99, P = 0.018). 

Conclusions:
Patients who are younger but weigh more are more likely to be converted from LLOA to OLOA. Patients with higher synthetic liver function and patients with high transaminases and lower WBC counts are also more likely to be converted from LLOA to OLOA. Our results warrant verification in large independent prospective datasets. 

92.11 Differing Rates of C. Difficile Infection in Patients with Ostomy Reversal versus Colon Resection

Posted on January 18, 2018

C. L. Charlton1, D. Chen1, V. Pandit1, A. Cruz1, D. Sessinou1, P. Vij1, V. Nfonsam1  1University Of Arizona,Department Of Surgery,Tucson, AZ, USA

Introduction:

Colon resections and ostomies are commonly performed surgical procedures.  Clostridium difficile (C. diff) is a known disease process associated with worse outcomes but variations exists among patients developing C. diff. The aim of the study was to assess the prevalence of C. diff among patient with colon resection with primary anastomosis (CR), ileostomy (IS), and colostomy (CS).

Methods:

We queried the National Impatient Sample (NIS) for the year 2011 selecting patients with age ≥ 18. Patients were stratified into: large-large colon resection with primary anastomosis (CR), small-large colon resection with primary anastomosis (IR), ileostomy (IS), and colostomy (CS). The primary outcome measure was development of C. diff infection. Statistical analysis was performed.

Results:

A total of 6,013,105 patients were assessed of which 20,312 patients were included. 49.6% had CR, 21.9% had IS and 24.7% had CS. Patients with CS (OR: 1.4, p=1.43×10^-6) were more likely to develop C. diff compared to patients with CR, while patients with IS (OR: 0.74, p=0.52) were less likely to develop C. diff compared to patients with IR. On comparing CS to IS, patients with CS were more likely to develop C. diff (OR: 1.8, p=3.62×10^-8). There was no difference in developing C. diff between ostomy reversal and colon resection. 

Conclusion:

Patients with colostomy or large colon resection were more likely to develop C. diff than patients with ileostomies or small intestine resection. However, there did not appear to be any difference between the rates of C. diff infection when comparing whether a patient has ostomy reversal or a colon resection.

 

 

92.08 Sarcoma Resection: Outcomes Related to Methods of Reconstruction and Tumor Locations

Posted on January 18, 2018

M. H. Nguyen1, J. Carr2, L. A. Damitz1, H. J. Kim2, C. Hultman1  1University Of North Carolina At Chapel Hill,Plastic Surgery Deparment,Chapel Hill, NC, USA 2University Of North Carolina At Chapel Hill,Department Of Surgical Oncology,Chapel Hill, NC, USA

Introduction: Neoadjuvant radiotherapy prior to surgical resection improves survival in patients with high-grade sarcomas, but less is known about the impact of tumor location, tumor type, and method of reconstruction on outcomes such as morbidity, recurrence, and mortality.

Methods:  We performed a retrospective cohort analysis of all patients from our institutional sarcoma registry who underwent neoadjuvant radiotherapy, prior to surgical resection, from 1998-2015 (n=136).  Both Surgical Oncology and Plastic Surgery were involved in pre-operative planning for wound closure.  Patients were separated into type of reconstruction (flap vs complex repair) and location of tumor (extremity vs torso).  Outcome measures were complications, length of stay, recurrence, and survival.   Groups were compared by t-test and chi-square analysis, with significance assigned to p values <0.05. 

Results: When compared to complex closure, the use of myocutaneous or fasciocutaneous flaps did not appear to confer any benefit, in terms of overall or wound complication rates, infection rates, recurrence, or survival (NS).   Regarding location, tumors of the extremities, compared to torso, were associated with improved outcomes, in term of length of stay (3.8 vs 9.5 days, p<0.001), local recurrence (23.8% vs. 41.1%, p< 0.05) and disease-free survival (65% vs. 44.6%, p<0.05). The majority of soft tissue sarcomas in the extremities was liposarcoma (28.75%), while leiomyosarcoma was the most common tumor in the torso (40.35%). 80.3% of the non-extremity tumors were retroperitoneal, 36% of which had positive margins.

Conclusion: In this cohort of patients who had neoadjuvant radiotherapy, prior to sarcoma resection, extremity tumors were associated with better prognosis, in term of decreased length of stay, postoperative morbidity, recurrence, and improved disease-free survival, compared to tumors of the torso.  Liposarcoma was more common in extremity, while leimoyosarcoma was more common in the torso.  Interestingly, complex closure of the defects, despite neoadjuvant radiotherapy, achieved similar results as flap reconstruction, indicating that patients can be preoperatively sorted into the most appropriate reconstructive option.   Complex repair should be considered, as a collaborative effort between Surgical Oncology and Plastic Surgery, in carefully selected patients, to avoid donor site morbidity and achieve equivalent outcomes.

 

92.07 Laparoscopic Cholecystectomy in Patients Supported with a Left Ventricular Assist Device (LVAD)

Posted on January 18, 2018

V. Suresh1, M. Bishawi5, B. Bryner5, M. Manning4, C. Patel3, J. Rogers3, C. Milano5, J. Schroder5, M. Daneshmand5, C. A. Sommer2  1Duke University Medical Center,School Of Medicine,Durham, NC, USA 2Duke University Medical Center,Division Of Acute Care Surgery, Department Of Surgery,Durham, NC, USA 3Duke University Medical Center,Division Of Cardiology, Department Of Medicine,Durham, NC, USA 4Duke University Medical Center,Division Of Cardiac Anesthesia, Department Of Anesthesia,Durham, NC, USA 5Duke University Medical Center,Division Of Cardiothoracic Surgery, Department Of Surgery,Durham, NC, USA

Introduction: An increasing number of end-stage heart failure patients are supported with Left Ventricular Assist Device (LVAD) implantation and must be maintained on a consistent anticoagulation regimen.  These patients are experiencing prolonged survival, and in some, the development of new biliary disease. Thus, the objective of this study was to describe the outcomes and management of LVAD patients undergoing laparoscopic cholecystectomy.

Methods: This study was a retrospective single center review. Adult patients supported on an implanted, continuous flow LVAD from January 1, 2007 to December 31, 2016 were included. Baseline characteristics were collected via retrospective chart review and the institutional LVAD registry. All laparoscopic cholecystectomies were performed in the operating room, utilizing cardiac anesthesia with endotracheal intubation, while LVAD settings were monitored by a trained perfusionist. Physical palpation and visual inspection were used to determine the course of the driveline and avoid driveline injury. Preoperative and postoperative hematology lab values, such as hemoglobin, international normalized ratio (INR), were collected via chart review. Continuous variables were presented as mean ± standard deviations, and compared using the Student t-test. Categorical variables were presented as proportions and percentages, and compared using chi-squared test or Fisher Exact test as appropriate. Statistical significance was established at a p < 0.05.

Results: After screening 798 patients, 5 (0.63%) were found to have undergone laparoscopic cholecystectomies after LVAD implantation. For 4 patients (80%), the indication was symptomatic cholelithiasis, and one patient (20%) had symptomatic acalculous cholecystitis. The average time from LVAD implantation to laparoscopic cholecystectomy was 254.4 ± 158.0 days.  Average pre-operative INR was 1.34 ± 0.30.  Average preoperative hemoglobin was 11.28 ± 2.41. Pre-operative FFP was administered to one patient. Preoperative Vitamin K was given to one patient.  All patients were on Coumadin pre-operatively and admitted prior to their operations for bridging with a heparin drip. Average post-operative change in hemoglobin was -1.16 ± 1.97. The only major post-operative complication in this cohort was the development of an abdominal wall hematoma in one patient, which was treated with drainage in the operating room and post-operative blood transfusion. The average length of stay (LOS) was 13.2 ± 4.6 days. 3 patients (60%) took an average of 12 days to reach therapeutic INR as the main driver for prolonged hospitalization.

Conclusions: Laparoscopic cholecystectomies can be performed safely in LVAD patients.  Prolonged hospital stay is mainly due to time to re-achieve INR goals. 

92.03 Retrospective Review of TAP Block with Liposomal Bupivicaine During Laparoscopic Enterocolectomy

Posted on January 18, 2018

L. M. Deppe2, C. Kasal2, K. Mathis1, D. W. Larson1  1Mayo Clinic Rochester,Colon And Rectal Surgery,Rochester, MN, USA 2Mayo Clinic Health System,Red Wing, MN, USA

Introduction:
Data has suggested value of TAP (transversus abdominus plane) neural blockade in
procedure related pain management as part of an enhanced recovery surgical program. Our goal was to
retrospectively analyze pain management using combined bupivacaine and liposomal bupivacaine as
part of a rural community laparoscopic colorectal and surgical program.

Methods:
Records of consecutive laparoscopic segmental colectomy and enterectomy patients
managed with an enhanced recovery program whose procedures were performed by four community
surgeons over an eighteen month period were reviewed under IRB approval. Patients who returned to
the operating room within thirty days were excluded from analysis. In place of an intrathecal block, an
ultrasound (US) guided bilateral TAP block was performed by the operating surgeon using 266 mg of
liposomal bupivacaine and 50 mg of 0.25% bupivacaine with epinephrine using the technique previously
described. Comparison was performed to previously published results of an enhanced recovery program
(which we also adopted) utilized by our Rochester, Minnesota based colorectal surgical system
colleagues.

Results:
Twenty four patients were included in analysis. Two patients were excluded; one requiring reexploration
for hemorrhage after anticoagulation and another with an anastomotic leak (who eventually
succumbed to respiratory failure). Three enterectomies were included (for malignancy and
inflammatory bowel disease) as well as twelve right hemicolectomies and nine sigmoid/left
hemicolectomies (eighteen for neoplasia and the remainder for diverticulitis). We found compliance
with enhanced recovery elements varied greatly and was generally surgeon dependent. No TAP block
site related reactions were noted. Median length of stay was three days, consistent with previously
published academic institutional results. Our ASA 1-2 population was 58%, comparing favorably with
published experience by our colorectal team which was 81%. Mean body mass index and median oral
morphine equivalents were 30.6 kg/m2 versus 26.9 kg/m2 and 4.5 mg versus 37.5 mg respectively.
Forty-six percent of patients never took an opiate versus a rate of 26% by our academic colleagues. There
were no readmissions within thirty days, although one patient did develop a deep incisional surgical site
infection. Though not formally timed, performing the above block was felt to add approximately five minutes to case length.

Conclusion:

Incorporating a surgeon performed US guided TAP block as part of an enhanced recovery
program resulted in similar length of stay, favorable oral morphine equivalent utilization and avoidance
of narcotic compared to published data utilizing an intrathecal block. Further study is indicated to
directly compare surgeon directed TAP block versus intrathecal injection for post-operative pain
management.

92.04 The perioperative outcome of pelvic exenteration for gastrointestinal and gynecologic malignancies.

Posted on January 18, 2018

N. M. Hinkle1, P. Dickson1, R. B. Interiano1, D. Shibata1, G. Munene4, J. Monroe1, L. Smiley2, M. Reed2, J. Deneve1  1University Of Tennessee Health Science Center,Surgery,Memphis, TN, USA 2University Of Tennessee Health Science Center,Obstetrics And Gynecology,Memphis, TN, USA 3University Of Tennessee Health Science Center,Medical Oncology,Memphis, MS, USA 4West Michigan Cancer Center,Surgery,Kalamazoo, MI, USA

Introduction: Locally advanced gastrointestinal and gynecologic malignancies can be managed with curative intent surgery in appropriately selected patients.  We review the outcomes of patients undergoing pelvic exenteration (PEx) for these neoplasms. 

Methods:  An IRB-approved retrospective review was performed of patients with primary and recurrent disease undergoing PEx with clinicopathologic variables and perioperative outcomes described. 

Results: Forty-two patients underwent PEx: median age 55.1 years, female (71%), Caucasian (67%), prior chemotherapy (79%). Pathology included the rectum (36%), colon (17%), ovary (12%), and other (35%).  Type of PEx performed included: anterior (10%), posterior (55%) and total exenteration (36%).  Median operative time was 499 minutes (167-893). Urinary diversion was performed in 43%.  Median hospital stay was 11.5 days (6-27). Overall complication rate was 57% with major morbidity associated with 33% of the cases and multiple complications occurring in 26%. There was also one 30-day mortality (2.4%).  Adjuvant chemotherapy was administered to 64%.  Eighteen (43%) suffered from local recurrence and 26% developed distant metastases.  At a median follow up of 18 months (0-84), 45% were alive without disease, 24% alive with disease, 24% dead of disease and 7% dead of other causes.  Median recurrence free survival (RFS) was 25 months and overall survival (OS) was not reached. RFS at 1- and 5-years was 84% and 31% with 1- and 5-year OS being 80% and 58%, respectively.  

Conclusion: Aggressive resection with PEx is possible with acceptable morbidity and outcome.  A multidisciplinary approach with PEx is a sound treatment option for appropriately selected patients with gastrointestinal and gynecologic malignancies.

 

89.20 Implementation of Enhanced Recovery to Improve Perioperative Outcomes at a Community Hospital

Posted on January 18, 2018

D. N. Maurente1, A. Ardeljan1, A. Johns1, S. Willis1, H. Abdul1, M. Bustos1, S. Sennhauser1, M. Ghali1, A. Rashid1, M. Perez1, O. M. Rashid1  1Holy Cross Hospital,Michael And Dianne Bienes Comprehensive Cancer Center,Fort Lauderdale, FL, USA 2Massachusetts General Hospital,Boston, MA, USA 3University Of Miami Miller School Of Medicine,Miami, FL, USA 4Charles E. Schmidt College Of Medicine At Florida Atlantic University,Boca Raton, FL, USA

Introduction:
To improve postoperative length of stay (LOS) and readmissions (RA) in bowel surgery, enhanced recovery protocols (ERP) were developed promoting a multidisciplinary approach with minimal anesthesia, early ambulation and enteral alimentation, and multimodality analgesia. This study evaluated ERP at a community Hospital (HCH) for improved patient outcomes.

Methods:
Patient charts for DRG 329, 330, and 330 were retrospectively reviewed at HCH in 2017 to compare outcomes in ERP versus non-ERP cases. The Medicare claims database (CMS) was also retrospectively reviewed by the same DRG codes to determine LOS and RA. Standard statistical comparisons were used to determine significance.

Results:

At HCH: DRG 329 mean LOS for non ERP was 13.0833 days (n = 12) versus 3.375 (n = 8), p = < 0.001; DRG 330 mean LOS for non ERP was 10.861 days (n = 36) versus 4.583 (n = 24), p = < 0.001; and DRG 331 mean LOS for non ERP was 7.272 days (n = 11) versus 3.348 (n = 23), p = 0.004. 

Based on comparable CMS data, HCH DRG 329 LOS improved from 90th to top 10th percentile (n = 238,907); DRG 330 from 90th to 28th percentile for DRG 330 (n = 285,423); and DRG 331 from 90th to 46th percentile for (n = 126,941), p < 0.001. HCH RA in ERP and non-ERP cases was 3% at 30 and 90 days. CMS RA for DRG 329 was 25.1% at 90 days and 9.9% at 30 days; DRG 330 RA was 18.3% at 90 days and 6.6% at 30 days; and DRG 331 RA was 11% at 90 days and 3.9% at 30 days. 

Conclusion:
Implementation of ERP for bowel surgery at HCH improved outcomes, and future ERP research in non-bowel surgery is recommended.

89.19 Liver Resection for Metastatic Colon Cancer: During or Separate from Colon Surgery??

Posted on January 18, 2018

L. G. Leijssen1,2, A. M. Dinaux1,2, C. R. Ferrone1,2, H. Kunitake1,2, L. G. Bordeianou1,2, D. L. Berger1,2  1Massachusetts General Hospital,General And Gastrointestinal Surgery,Boston, MA, USA 2Harvard School Of Medicine,Brookline, MA, USA

Introduction:  ?Concomitant hepatectomy in patients with primary colon cancer and synchronous liver metastasis is feasible. The aim of this study is to assess the impact of combined hepatectomy and colectomy compared to separate procedures on both short- and long-term outcomes.?

Methods:  All patients who underwent a hepatectomy for metastatic colon cancer at our tertiary center from 2004 to 2014 were included. We compared patients who underwent hepatectomy during (DC) with before or after colectomy (BAC) and partial (PH) against extended hepatectomy (EH). ?

Results: We included 18 DC and 44 BAC patients (6.3% before and 63.5% after colectomy). DC patients had more transverse tumors (P=0.009), less sigmoid cancer (P=0.047), and a higher readmission rate (P<0.001). EH was performed in a comparable numbers of patients (DC 61.1% vs. BAC 61.4%). Mean survival duration was non-significantly shorter for DC patients (36 vs. 48 months; P0.075). R0 resection was achieved in 91.7% of PH and 97.4% of EH patients (P=0.308). Mean survival duration was non-significantly shorter for EH patients (49 vs. 42 months; P0.298). In adjusted analyses, there was no difference in overall survival between DC and BAC patients (HR 1.126 [95% CI: 0.46-2.76] P=0.795), or PH and EH patients (HR 1.042 [95% CI: 0.46-2.36] P=0.921).?

Conclusion: This study underscores the oncologic safety of simultaneous hepatectomy in patients with primary colon cancer and synchronous liver metastasis.?

 

89.14 The impact of thrombocytopenia on pure laparoscopic hepatectomy for hepatocellular carcinoma

Posted on January 18, 2018

T. Cheung1, C. Lo1  1The University Of Hong Kong,Hong Kong, HONG KONG, Hong Kong

Introduction:
 Bleeding in the presence of thrombocytopenia has been a major issue in liver resection even in open procedures. Although laparoscopic liver resection has been reported to be a safe and effective treatment even in patients with liver cirrhosis, there is very little evidence in patients with thrombocytopenia.

Methods:

Between October 2002 and February 2016, 132 patients had undergone pure laparoscopic liver resection for hepatocellular carcinoma (HCC). Thrombocytopenia was defined as platelet counts <100x 109/L. Amongst 132 patients received laparoscopic liver resection, 17 patients who had thrombocytopenia (Group 1) and 111 patients who did not have thrombocytopenia (Group 2) with the same solitary tumor and clear resection margin characteristics were chosen for comparison. 

The immediate operation outcome and survival including operation morbidity were compared. The disease free survival and overall survival were also compared.

Results:

Comparing Group1 to Group 2, the median operation time was 184 minutes vs 175 minutes p=0.817, the median blood loss was 200ml vs 100ml (P=0.347). Hospital stay was 4 days vs 4 days (P<0.888), postoperative complication was 3(17.6%) vs 8(7.2%) (p=0.334), median disease free survival was 20 months vs 71 months (P<0.001) and the median overall survival was >120 months vs 136 months (P=0.298). The median disease free survival for stage II HCC was 12 months vs 71 months (p=0.001).

 

Conclusion:

Thrombocytopenia does not have an adverse effect immediately on laparoscopic liver resection. Long term outcome has to be evaluated further in future. 

89.12 Outcomes of Extended Hepatectomy for Hepatobiliary Tumors. Who is More Important Than Where

Posted on January 18, 2018

A. M. Attili1,2, I. Sucandy1, N. Patel1, J. Spence1, K. Luberice1, T. Bourdeau1, S. Ross1, A. Rosemurgy1  1Florida Hospital Tampa,General Surgery,Tampa`, FL, USA 2University Of Central Florida,General Surgery,Orlanda, FL, USA

Introduction:  Hepatectomy is the gold standard curative treatment for hepatic neoplasms in patients with preserved liver function. Many large tumors involving central liver segments require extended hepatectomy (EH) to gain complete resection with negative margins,however some patients with impaired liver function are offered non surgical options due to high morbidity and mortality following this major operation. Outcome data is relatively limited with the majority of extended hepatectomies offered only at major hepatobiliary centers. We aim to describe outcomes of EH at our hepatobiliary center.

Methods:  With institutional review board approval, all patients undergoing hepatectomy between 2012-2017 were prospectively followed. Patient demographic data, perioperative outcomes, and short/long-term survival data were collected and analyzed. Data are presented as median (mean ± SD).

Results: A total of 91 patients underwent hepatectomy (open and robotic approach) within the study period with 10 patients undergoing EH. The majority of patients who underwent EH were women (70%), age of 63 (60.3 ± 16.5) years, body mass index of 24 (24.7 ± 3.8) kg/m2, and MELD score of 11 (10.9 ± 1.6). Six patients underwent an extended right hepatectomy (resecting segment IV-VIII), while 4 patients underwent extended left hepatectomy (resecting segment II-V and VIII). Indications were Klatskin tumor (30%), hepatocellular carcinoma (30%), intrahepatic cholangiocarcinoma (20%), and metastatic neuroendocrine tumor (20%). Operative time was 224 (253.1 ± 111.7) minutes with estimated blood loss of 500 (845 ± 1002.9) ml. No intraoperative complications were seen. One patient had blood transfusion. Negative resection margins (R0) were achieved in 9 patients. One patient with R1 resection margin had hepatocellular carcinoma involving a deep intrahepatic portion of the right hepatic duct, where bile duct resection followed by hepaticojejunostomy was unsafe to perform. Two patients experienced postoperative complications (pleural effusion requiring thoracentesis in one patient and respiratory failure leading to multisystem organ failure and death in another). Length of intensive care unit stay was 2(2.1 ± 1.5) days, and hospital stay was 5(5.6 ± 2.6) days. Readmissions to the hospital were seen in 3 patients (failure to thrive in 2 patients and development of postoperative ascites with pulmonary embolism in 1 patient). 80% of the patients are currently alive with median follow up of 41.2 months. 

Conclusion: Despite major concerns of postoperative complications and liver failure for patients with a component of liver dysfunction who has otherwise resectable tumors, EH can be a feasible curative option for these patients. Clinical success mainly depends on preoperative planning, experience of hepatobiliary team, and optimum postoperative care.

Next Page »