89.06 Strategies for Minimizing Complications and Length of Stay after Minimally Invasive Esophagectomy

Posted on January 31, 2019

R. E. Merritt1, P. Kneuertz2, L. Luo1, K. A. Perry1  1Ohio State University,Thoracic Surgery,Columbus, OH, USA, 2Ohio State Wexner Medical Center,Surgery,Columbus, OH, USA

Introduction:
 Minimally invasive esophagectomy is associated with significant morbidity which can significantly prolong the length of stay for patients. Minimizing major postoperative complications can facilitate protocoled care and allow patients to recover more rapidly, which can decrease the length of stay. A streamlined preoperative assessment, gastric ischemic preconditioning, and consistent collaboration between a specialized general surgeon and general thoracic surgeon are key elements that can improve the outcomes for minimally invasive esophagectomy.

Methods:
This is a retrospective study of 130 patients who underwent a total laparoscopic and thoracoscopic Ivor Lewis esophagectomy for esophageal carcinoma between August 2014 and June 2018. A total of 112 patients (86 %) underwent neoadjuvant chemoradiation. All of the 130 patients in the study underwent a laparoscopic gastric devascularization procedure a mean of 18.1 days ±  13.9 days prior to the esophagectomy. A total laparoscopic and right thoracoscopic esophagectomy was performed by a single team consisting of the same general surgeon and general thoracic surgeon. The intrathoracic esophagogastric anastomosis was performed with a circular EEA stapler in all 130 patients. A barium esophagram was obtained routinely on postoperative day number 6.  The target discharge date was postoperative day number 8.

Results:
A total of 130 patients underwent a total laparoscopic and thoracoscopic Ivor Lewis esophagectomy. A total of 121 patients (86.2%) had Adenocarcinoma and 8 patients (6.2 %) had squamous cell carcinoma. A total of 30 patients (23.08%) had postoperative complications. A total of 20 patients (15.4%) develped atrial fibrillation which was the most frequent complication. A total of 4 patients (3.1%) developed an anastomotic leak and 1 patient (0.77%) developed an esophago-bronchial fistula. There was 1 postoperative death (0.77%) in the cohort of patients. The median length of stay was 8 days. The mean length of stay for patient without complications was 8 days  ± 1.2 days and 12.4 days  ± 7.1 days for patients with one or more complications (p=0.002).

Conclusion:
The development of postoperative complications after minimally invasive Ivor Lewis esophagectomy significantly increases the length of stay. Performing the operation with a specialized tandem surgical team and including preoperative ischemic preconditioning of the stomach minimizes complications and facilitates on time discharges. This consistent minimally invasive approach for esophagectomy also demonstrated a low rate of anstomotic complications, which may also contribute signifinantly to a decreased length of stay.
 

89.05 Lung Resection Outcomes After Implementation of an Enhanced Recovery After Thoracic Surgery Program

Posted on January 31, 2019

G. J. Haro1, D. M. Jablons1, J. R. Kratz1  1University Of California – San Francisco,San Francisco, CA, USA

Introduction:

Enhanced recovery after thoracic surgery (ERATS) programs have not been widely utilized but have the potential to improve perioperative outcomes.  We developed an evidence-based, multidisciplinary ERATS program for our lung resection patients in an academic, quaternary-care center.  In this study, we evaluate our outcomes following ERATS implementation.

Methods:

The ERATS program was implemented in October 2017 and included preoperative education, increased utilization of minimally invasive surgery, and standardization of anesthesia techniques, hospital acuity, ancillary consultations, line management, and multimodality analgesia.  Patients were included in the study if they underwent elective lobectomy or sublobar resection by any technique for primary or secondary lung cancer from October 2015 to June 2018, which included two years before and nine months after the program.  Propensity scores based upon age, sex, race, comorbidity, diagnosis, technique (minimally invasive versus thoracotomy), and procedure were used to match patients and evaluate outcomes.

Results:

156 pre-ERATS and 65 post-ERATS patients were consecutively identified.  There were no substantial differences in age, sex, race, comorbidity, smoking history, or diagnosis, but there were slightly more lobectomies performed post-ERATS (Pre 44.9%-Post 53.9%).  Regarding adherence to the program, the proportion of minimally invasive cases increased by 21.0%, and the number of patients admitted to the ICU was reduced by 47.3%.  Foley catheters and chest tubes were removed 56.9% and 60.0% within goal, respectively, and the number of patients that ambulated at least three times on POD1 remained the same (Pre 66.0%-Post 63.1%).  In propensity score matched analysis that accounted for minimally invasive surgery (Table), the program was estimated to reduce length of stay by 1.1 days (95% CI 0.3-2.0; P=0.01), morbidity by 14.5% (95% CI 2.9%-30.3%; P=0.02), and the direct costs of surgery and hospitalization by $5,000 (95% CI $2,000-7,900; P<0.01).  Despite expedited hospitalizations, readmission remained minimal (Pre 5.1%-Post 1.5%; P=0.39).

Conclusion:

Our early experience displays that an ERATS program for lung resection reduces length of stay, morbidity, and direct costs without an increase in readmission.  Application of an ERATS program to other thoracic resections, such as esophagectomy or mediastinal mass resection, may show similar benefit

89.04 Outcomes with Percutaneous Right Ventricular Assist Device following Left Ventricular Assist Device

Posted on January 31, 2019

S. K. Sundararajan1, M. Cain3, B. Badu3, L. Durham3, L. Joyce3, N. Gaglianello2, D. Ishizawar2, M. Saltzberg2, D. Joyce3, A. Mohammed2  1Medical College Of Wisconsin,General Internal Medicine/Medicine/Medical College Of Wisconsin,Milwaukee, WI, USA 2Medical College Of Wisconsin,Cardiology/Medicine/Medical College Of Wisconsin,Milwaukee, WI, USA 3Medical College Of Wisconsin,Cardiothoracic Surgery/Surgery/Medical College Of Wisconsin,Milwaukee, WI, USA

Introduction:
Right ventricular (RV) failure is a frequent complication of left ventricular assist device (LVAD) implantation in patients with end-stage heart failure, impacting 20% of patients. RV failure post LVAD is associated with 40% 1-year mortality. Percutaneous support can be used in this scenario, however data supporting this approach is lacking. 

Methods:
A retrospective review was conducted of all patients who developed severe RV failure requiring use of a right internal jugular placed percutaneous right ventricular assist device (RVAD) following durable LVAD implant at our academic institution between 01/01/2017 and 07/31/2018. Patient demographics were reviewed in addition to key hemodynamic and laboratory values were collected pre-RVAD implantation (immediately prior to cannulation) and at the time of explant (post-RVAD). Pre-RVAD and post-RVAD data was compared using paired T tests. Survival at 30 days or at date of hospital discharge (whichever was longer) was considered as the primary outcome. Patient hemodynamic measures were evaluated as secondary outcomes.

Results:
Of 41 patients who received LVAD, 10 (24%) developed severe RV failure postoperatively. Baseline characteristics include age 51±14.5 years, with 30% females. Incidence of ischemic cardiomyopathy was 20%. Most patients received LVAD as destination therapy (80%). RVAD support via percutaneous cannulation was maintained for a median duration of 15.5 days (Range – 7 to 23 days). Survival at 30 days or to hospital discharge was 100%. RVAD support was successfully weaned in 9 patients (90%), while the remaining 1 patient underwent cardiac transplantation. We observed a reduction in central venous pressure from 17.2±4.5 mm(Hg) to 11±3.1 mm(Hg), p=0.004 and an increase in LVAD flow from 4.5±0.8 l/min to 5.1±0.4 l/min, p=0.018. There was a trend towards increase in mixed venous oxygen saturation from 61.5±11.1% to 67.0±7.0%, p=0.224. Other biochemical parameters collected at the time of percutaneous RVAD support and 24-48 hours after RVAD explant are described in the table below.

Conclusion:
Use of a right internal jugular placed percutaneous RVAD for treatment of severe RV failure after LVAD implantation provides superior early postoperative survival than traditional management techniques. These findings support broader implementation of this technique to further characterize its benefit.

89.03 Trends in Mortality and Readmissions following Endovascular and Open Infra-inguinal Revascularization

Posted on January 31, 2019

H. Khoury1, Y. Sanaiha1, S. Rudasill1, H. Xing1, A. Mardock1, R. Jaman2, H. Gelabert2, P. Benharash1  1David Geffen School Of Medicine, University Of California At Los Angeles,Cardiothoracic Surgery,Los Angeles, CA, USA 2David Geffen School Of Medicine, University Of California At Los Angeles,Los Angeles, CA, USA

Introduction:  While short-term outcomes of endovascular and open surgical revascularization in patients with peripheral artery disease have been previously reported, 30-day readmission and resource use following these procedures remain unknown. 

Methods:  We used the 2010-2015 Nationwide Readmissions Database and the International Classification of Diseases, Ninth Edition to identify patients with peripheral artery disease undergoing either endovascular or open infra-inguinal bypass revascularization. Student’s t-test and chi-squared test were used to compare baseline continuous and categorical variables, respectively. Multivariable logistic regression was used to evaluate the independent association between the two procedures and readmission, index complications, and mortality. 

Results: Of an estimated 782,593 patients diagnosed with peripheral artery disease (PAD), 352,587 and 304,928 underwent endovascular and open surgery revascularization, respectively. Compared to patients who underwent open revascularization, endovascular patients were more commonly female (44.8 vs. 36.7%, P<0.001), and older (69.4 vs. 67.2 years, P<0.001). Moreover, they had a higher Elixhauser Comorbidity Index (4.3 vs. 3.8, P<0.001) and cost of index hospitalization ($25,288 vs. 21,949, P<0.001), in addition to experiencing higher rates of 30-day readmission (16.0 vs. 13.7%, P<0.001), in-hospital complications (21.1 vs. 19.0%, P<0.001), and in-hospital mortality (2.1 vs. 1.7%, P<0.001). 30-day readmission for endovascular and open revascularization increased the overall hospitalization cost by an estimate of $16,841 and $16,235, respectively. In contrast, the risk-adjusted multivariable analysis found open revascularization to be independently associated with increased odds of 30-day readmission (OR, 1.11; 95% CI 1.08 – 1.14), index complications (OR, 1.16; 95% CI 1.13 – 1.19) and mortality (OR, 1.26; 95% CI 1.16 – 1.35), compared to those who underwent endovascular revascularization. Trend analysis revealed a decreasing annual incidence of PAD, endovascular and open revascularization procedures (Figure).

Conclusion: Despite lower rates of adverse events compared to endovascular, open infra-inguinal surgical revascularization is independently associated with increased risk of short-term readmission, complications and mortality. Aside from anatomic factors, these findings should be considered in the selection of appropriate surgical therapy for arterial occlusive disease of the lower extremities.

 

89.02 Algorithm-Based Troubleshooting for Bleeding during Thoracoscopic Anatomic Pulmonary Resections

Posted on January 31, 2019

H. Igai1, R. Yoshikawa1, F. Osawa1, T. Yazawa1, M. Kamiyoshihara1  1Japanese Red Cross Maebashi Hospital,Department Of General Thoracic Surgery,Maebashi, GUNMA, Japan

Introduction: Few studies have reported on the effects of intraoperative complications such as vessel injury during thoracoscopic anatomic pulmonary resections. We evaluated intraoperative vessel injury and assessed troubleshooting methods during thoracoscopic anatomic pulmonary resections.

Methods: Between April 2012 and March 2018, 378 patients underwent thoracoscopic anatomic pulmonary resection, 40 of whom were identified as having intraoperative vessel injury. Significant vessel injury was defined as bleeding that needed compression times of more than 30 seconds for hemostasis. In our department, we treat significant bleeding based on the algorithm shown in Figure 1. We analyzed the injured vessel and the hemostatic procedure employed, then compared the perioperative outcomes in patients with (n=40) or without (n=338) vessel injury. Additionally, we examined the data on a year-by-year basis from April 2012, and perioperative results were compared in each year.

Results: The surgical procedures examined included 32 lobectomies (80%) and 8 segmentectomies (20%). The vessel injured was a branch of the pulmonary artery in 22 cases (55%), a branch of the pulmonary vein in 13 (32.5%), and a different vessel in 5 (12.5%). The procedure was converted into a thoracotomy in five cases (12.5%) to achieve hemostasis. Hemostasis was achieved by applying a thrombostatic sealant in 26 cases (65%), compression with a cotton stick or adjacent lung parenchyma in 10 (25%), and some other way in 4 (10%).
Although patients without vessel injury had a shorter operation time (207 vs 240 min., p=0.0005), less intraoperative blood loss (53 vs 286ml., p<0.0001), and shorter duration of chest tube drainage (3 vs 3.9days, p=0.02), there were no significant differences in the length of postoperative hospitalization (8 vs 13days, p=0.14) or morbidity (15 vs 23%, p=0.25).
The occurrence rate of intraoperative significant bleeding in the last year measured was similar to that in the first year measured (6.8 vs 10.5%, p=0.23), although other perioperative results had significantly improved in that time frame.

Conclusion: Thoracoscopic anatomic pulmonary resection is feasible and safe if the surgeon performs appropriate hemostasis although vascular hazards might be inherent during thoracoscopic anatomic pulmonary resection, regardless of surgeon’s experience. Application of sealants and compression techniques using a cotton stick or adjacent lung parenchyma are important techniques to achieve hemostasis in the event of significant vessel injury during thoracoscopic anatomic pulmonary resection.

 

85.06 A National Survey of Sexual Harassment Among Surgeons

Posted on January 31, 2019

A. Nayyar1, S. Scarlet1, P. D. Strassle1, D. W. Ollila1, L. M. Erdahl3, K. P. McGuire2, K. K. Gallagher1  1University Of North Carolina At Chapel Hill,Department Of Surgery,Chapel Hill, NC, USA 2Virginia Commonwealth University,Department Of Surgery,Richmond, VA, USA 3University Of Iowa,Department Of Surgery,Iowa City, IA, USA

Introduction:

Emerging data suggests that experiencing sexual harassment in the workplace is a common occurrence for women across all professions, which can be harmful personally and professionally.  While recent studies have reported the incidence of sexual harassment in medicine, no study has examined the nature or scope of sexual harassment experienced by surgeons.

 

Methods:

An anonymous, electronic survey was distributed via a web-based platform to members of American College of Surgeons (ACS), Association of Women Surgeons (AWS), and through targeted social media platforms from April-July 2018. Questions pertained to workplace experiences and frequency of sexual harassment in the past 12 months. Sexual harassment was defined according to the U.S. Equal Employment Opportunity Commission. Fisher’s exact and Mantel-Haenszel tests were used to assess differences in respondent characteristics.

Results:

1,005 individuals completed the survey. 74% (n=744) identified as women, a response rate of 18% among US women surgeons. 25% (n=249) of respondents identified as men which represents 1% of male surgeons in the US – these responses were analyzed separately. Respondents reported employment by an academic institution (51%), community medical center (13%), private practice (15%), or other (19%). For both genders, the most common specialties were general surgery (34%), trauma surgery (10%), and surgical oncology (8%). Overall, 58% (n=432) of women surgeons experienced sexual harassment within the 12-month period preceding the survey, compared to 25% (n=61) of men, p<0.0001. Among women, the most common forms of harassment reported were “verbal or physical conduct (e.g. body language)” (53%), “unwanted sexual advances or physical contact” (23%), and “comments about sexual orientation” (10%). Women trainees were more than twice as likely to experience harassment as compared to attending surgeons (OR 2.52, 95% CI 1.78, 3.56, p<0.0001). The majority (84%) of incidents of harassment reported by women as part of the survey were not reported to any institutional authority. The most common reasons for not reporting was “fear of a negative impact on my career” (43%), “fear of retribution” (32%), and “fear of being dismissed and/or inaction towards perpetrator” (31%) (Figure).  

Conclusion:

Our study indicates that there is an alarming prevalence of unreported sexual harassment experienced by women surgeons in the US.   Combined with the documented sexual harassment of women physicians in other medical specialties, we believe this finding demonstrates an urgent need to improve the safety of the healthcare workplace, not just for women surgeons, but for all.
 

85.05 A National Evaluation of Motor Vehicle Crashes among General Surgery Residents

Posted on January 31, 2019

C. R. Schlick1, D. B. Hewitt1, C. M. Quinn1, R. J. Ellis1,2, K. E. Shapiro1, K. Y. Bilimoria1,2, A. D. Yang1  1Feinberg School Of Medicine – Northwestern University,Surgical Outcomes And Quality Improvement Center, Department Of Surgery,Chicago, IL, USA 2American College Of Surgeons,Division Of Research And Optimal Patient Care,Chicago, IL, USA

Introduction:  Motor vehicle crashes (MVCs) are a leading cause of resident mortality. However, the overall rate of MVCs in general surgery residents is unknown, as are the effects of duty hour regulations on MVCs. Our objectives were to (1) examine the frequency of “nodding off” while driving, near miss MVCs, and MVCs in a national cohort of general surgery residents and (2) evaluate whether assignment to the Flexible Policy arm of the FIRST trial impacted MVC outcomes.

Methods:  Residents from all US ACGME-accredited general surgery training programs were surveyed following the 2017 American Board of Surgery In-Training Examination (ABSITE). Outcomes of interest were nodding off while driving, near-miss MVCs, and MVCs after being on call. Exposure variables included frequency of duty hour violations, poor psychiatric well-being (as measured by the GHQ-12 assessment), and FIRST trial study arm. Group-adjusted cluster chi-square and hierarchical regression models with program-level intercepts were used to measure associations between resident/program factors and outcomes.

Results: Among 7,391 general surgery residents (response rate 99.3%) from 260 programs, 34.7% reported nodding off while driving, 26.6% a near-miss MVC, and 5.0% an MVC over the preceding 6 months. Nodding off while driving was associated with more frequent 80-hour rule violations (59.8% with 5+ months with violations vs. 27.2% with 0, OR 3.70 [95% CI 2.90-4.72]) and poor psychiatric well-being (41.2% in worst quartile vs. 23.4% in best, OR 1.80 [95% CI 1.55-2.09]). Near-miss MVCs were similarly associated with more frequent 80-hour rule violations (53.6% with 5+ months with violations vs. 19.2% with 0, OR 4.09 [95% CI 3.20-5.22]) and poor psychiatric well-being (36.7% in worst quartile vs. 12.9% in best, OR 3.07 [95% CI 2.60-3.64]). Residents were more likely to report MVCs with more frequent duty hour violations (14.0% for 5+ months with violations vs. 3.5% for 0, OR 3.01 [95% CI 2.08-4.35]) or poor psychiatric well-being (9.8% in worst quartile vs. 1.8% in best, OR 4.57 [95% CI 3.17-6.58]). FIRST trial study arm was not significantly associated with nodding off (34.0% in flexible policy vs. 36.3% in standard, OR 0.83 [95% CI 0.65-1.05]), near-miss MVCs (25.9% in flexible policy vs. 27.5% in standard, OR 0.85 [95% CI 0.66-1.08]), or MVCs (4.8% in flexible policy vs. 4.7% in standard, OR 0.97 [95% CI 0.67-1.41]).

Conclusion: Nodding off while driving, near-miss MVCs, and MVCs were more likely in residents with frequent duty hour violations and poor psychiatric well-being, but were not associated with FIRST trial study arm. Thus, although extended work hours may contribute to resident MVCs, flexible duty hour policies in surgical training are not associated with increased risk. Future interventions focused on minimizing violations of the 80-hour maximum weekly duty hour limit and improving psychiatric well-being may improve surgical resident driving safety.

 

85.04 Implementation of a comprehensive initiative to reduce over-prescribing of postoperative opioids

Posted on January 31, 2019

A. S. Chiu1, M. Freedman-Weiss1, P. Yoo1  1Yale University School Of Medicine,Surgery,New Haven, CT, USA

Introduction: Surgeons almost universally prescribe opioids postoperatively for analgesia, and often do so in far excess of what is needed—leaving unused pills susceptible to diversion and abuse. However, surgeons also have the opportunity to help offset the opioid epidemic by leading with practice changes. We sought to decrease the amount of opioid prescribed postoperatively through a multifaceted program.

Methods: Starting in the academic year 2016, an initiative to reduce over-prescribing of postoperative opioids was introduced at a single academic center. Based on the results of a needs assessment survey, this multi-pronged program included: resident education on postoperative analgesia and opioid prescribing, a grand rounds on the opioid epidemic, a change of the default number of opioid pills in an electronic medical record, and the distribution of a postoperative analgesic guideline with recommended doses.

 

The amount of opioid prescribed postoperatively between July 2016 and July 2018 was collected for the five most common short stay general surgery procedures (laparoscopic appendectomy, laparoscopic cholecystectomy, partial mastectomy, and laparoscopic and open inguinal hernia repair). Patients discharged after 48 hours were excluded to avoid confounding of complex hospital stays. To compare different opioids, prescriptions were converted to Morphine Milligram Equivalents (MME). Linear regression was used to evaluate prescribing over time and controlled for patient age, gender, race, insurance, and operation.

Results: There were a total of 6,109 qualifying operations over this 24-month period, including 2,071 laparoscopic cholecystectomies, 1,541 partial mastectomies, 1,404 inguinal hernia repairs (614 open and 790 laparoscopic), and 1,093 laparoscopic appendectomies. The population was predominately female (60.5%), Caucasian (71.0%), and privately insured (54.4%).

The mean opioid dose prescribed during this time was 168.5 MME (SD 102.5). The monthly average steadily declined over time (Figure 1); during the first 3 months, the average postoperative opioid prescription was 196.3 MME compared to 106.3 MME (p<0.01) in the final 3 months. This difference over a 2-year period is the equivalent of 12 pills of 5mg of oxycodone per prescription. Multivariable regression demonstrated a 4.4 MME (95%CI -5.0 to -3.6) decrease in average opioid prescription per month.

Conclusion: A dedicated and comprehensive program aimed at eliminating the over-prescription of opioids by surgeons reduced the amount of postoperative opioid prescribed by nearly half. Surgeons must acknowledge both their role in the societal problem of opioid abuse and their ability to deliver sustained and reproducible improvements.

85.03 Impact Of Traumatically Brain-Injured Donors On Outcomes After Heart Transplantation

Posted on January 31, 2019

A. Suarez-Pierre1, T. C. Crawford1, X. Zhou1, C. Lui1, C. D. Fraser1, E. Etchill1, K. Sharma2, R. Higgins1, G. J. Whitman1, A. Kilic1, C. W. Choi1  1Johns Hopkins University School Of Medicine,Cardiac Surgery,Baltimore, MD, USA 2Johns Hopkins University School of Medicine,Cardiology,Baltimore, MARYLAND, USA

Introduction: Heart transplant recipients of traumatically brain-injured (TBI) donors have been reported to have inferior survival and increased rates of coronary artery vasculopathy in single-center studies with limited sample sizes. This study sought to examine the impact of TBI donors on outcomes after heart transplantation. 

Methods:  We identified all adult heart transplants performed between January 2007 and December 2016 (inclusive) in the OPTN database. Patients undergoing repeat or heterotopic heart transplantation were excluded from the study. Recipients were dichotomized based on donor cause of death (TBI versus non-TBI), propensity-scored across 22 variables with known associations with mortality, and matched 1:1 without replacement. The primary endpoint was all-cause mortality at 1-, 3-, and 5-years after transplantation. Secondary endpoints were 3- and 5-year survival conditional on 1-year survival and rates of coronary artery vasculopathy at 1-, 3-, and 5- years after transplant. 

Results: In the study period, 20,244 patients underwent heart transplantation. TBI was the primary cause of death of all donors (53.4%; 10,816/20,244) and among TBI donors, blunt injury (59.6%; 6,443/10,816) and gunshot wound (35%; 3,781/10,816) were the most common mechanisms of traumatic brain injury. Propensity matching generated 6,919 pairs with adequate covariate balance (all standardized mean differences < 0.07). Risk-adjusted survival was similar between recipients of TBI donors and non-TBI donors at 1-year (90.5% vs 90.0%, log-rank p=0.32), 3-years (84.1% vs 83.3%, log-rank p=0.25), and 5-years (78.1% vs 77.5%, log-rank p=0.34). Risk-adjusted survival conditional on 1-year survival, was also similar at 3-years (92.8% vs 92.6%, log-rank p=0.57) and 5-years (86.2% vs 86.1%, log-rank p=0.74). The risk-adjusted rates of coronary artery vasculopathy did not differ either at 1-year (8.0% vs 7.7%, log-rank p=0.60), 3-years (20.6% vs 20.4%, log-rank p=0.77), or 5-years (30.6% vs 30.4%; long-rank p=0.78).

Conclusion: In the largest analysis of TBI donors in heart transplantation, we found similar survival and rates of coronary artery vasculopathy to those who received hearts from non-TBI donors out to 5 years. These findings should allay concerns over continued transplantation with this unique donor population.

85.02 Learning from England’s Best Practice Tariff: Process Measure Pay-for-Performance Can Improve Outcomes

Posted on January 31, 2019

C. K. Zogg1,2,3, D. Metcalfe3, A. Judge4, D. C. Perry3, M. L. Costa3, B. J. Gabbe5, A. H. Haider2  1Yale University School Of Medicine,New Haven, CT, USA 2Brigham And Women’s Hospital,Center For Surgery And Public Health,Boston, MA, USA 3University of Oxford,Oxford, United Kingdom 4University of Bristol,Bristol, United Kingdom 5Monash University,Melbourne, Australia

Introduction: Since passage of the Patient Protection and Affordable Care Act in 2010, Medicare has renewed efforts to improve the quality of older adult health through the introduction of an expanding set of outcome-based readmission and mortality pay-for-performance (P4P) measures. Among trauma patients, potential P4P has met with mixed success given concerns about the heterogeneous nature of patients that trauma providers treat and resultant variations in outcome measures. A novel approach taken by the National Health Service in England could offer a viable alternative plan. The objective of this study was to assess the effectiveness of the 2007-2010 English provider consensus-driven, process measure-based P4P Hip Fracture Best Practice Tariff (BPT) on improving trauma outcomes.

Methods: Quasi-experimental interrupted time-series and difference-in-difference analysis of 2000-2014 death certificate-linked data from England (Hospital Episode Statistics), Scotland (Scottish Morbidity Records), and the United States (100% Medicare all-payer claims). The study compared before-and-after differences in English temporal trends relative to those of Scotland and the US. Outcomes included: 30/90/365-day mortality, readmission, index hospital length of stay, and time to surgery. The study also assessed projections for the number of lives saved and readmissions averted were the BPT to be implemented in Scotland and the US.

Results: A total of 878,860 English, 97,487 Scottish, and 2,994,748 US index fractures were included among adults ≥65y. Following BPT introduction in England, 30-day mortality decreased instantaneously by an absolute value of -2.6 (95%CI -3.5, -1.7) percentage-points and continued to drop by an average of -0.2 (-0.4, -0.1) percentage-points per year (DID-Scotland: -1.6; DID-US: -2.2). 90-day mortality decreased more precipitously, dropping by an absolute value of -5.6 (-7.1, -4.2) percentage-points and an annual average thereafter of -0.2 (-0.5, 0.0) percentage-points per year (DID-Scotland: -1.9; DID-US: -2.9). Similar improvements were observed in readmission (e.g. 30-day ITSA: -1.4 [-2.3, -0.5]), time to surgery, and length of stay. Projections suggest that were the BPT to be implemented in Scotland and the US (Figure), by 2030, as many as 1,377 Scottish and 11,434 US lives could be saved.

Conclusion: In contrast to outcome-based P4P, process measure P4P such as that implemented through the English Hip Fracture BPT could result in significant improvements in outcomes for US patients while remaining more applicable to heterogeneous trauma populations and acceptable to trauma providers. As efforts to improve older adult health continue to increase, there are important lessons to be learned from initiatives like the BPT

85.01 National Evaluation of Gender Discrimination and Sexual Harassment in U.S. General Surgery Residency Programs

Posted on January 31, 2019

Y. Hu1,3, R. Ellis1, D. B. Hewitt1,6, A. D. Yang1, J. Buyske7, D. B. Hoyt8, T. Nasca9, K. Y. Bilimoria1  1Feinberg School Of Medicine – Northwestern University,Chicago, IL, USA 3Ann & Robert H. Lurie Children’s Hospital,Pediatric Surgery,Chicago, IL, USA 6Thomas Jefferson University,Philadelphia, PA, USA 7American Board Of Surgery Inc,Philadelphia, PA, USA 8American College Of Surgeons,Chicago, IL, USA 9Accreditation Council for Graduate Medical Education,Chicago, IL, USA

Introduction: Discrimination and harassment create a hostile work environment that adversely affects morale, productivity, and patient care. However, national prevalence estimates of discrimination and harassment, sources of mistreatment, and associated factors are unknown.

Methods: Residents training in all Accreditation Council for Graduate Medical Education (ACGME)-accredited general surgery programs were administered a survey immediately following the 2018 American Board of Surgery In-Training Examination (ABSITE). The survey queried experiences with gender discrimination, pregnancy/childcare discrimination, and sexual harassment. Program-level proportions of female residents and faculty were obtained from the Association of American Medical Colleges. Gender-stratified multivariable regression models, adjusted for resident clustering within programs, were developed to examine associations with resident and program characteristics.

Results: Among 7,409 respondents (99.3% response rate), 31.9% reported gender discrimination, 7.2% pregnancy/childcare discrimination, and 10.3% sexual harassment. Gender discrimination was reported by 65.1% of female residents, most commonly from patients/families (49.2%), nurses/staff (23.6%), and attendings (17.6%). Gender discrimination was more likely to be reported by senior residents, women training in the Northeast, and men in programs with the highest proportions of female residents (all p<0.05). Pregnancy/childcare discrimination was experienced by 13.1% of women, mostly from other surgeons (attendings 43.9%, co-residents 23.5%). Pregnancy/childcare discrimination was more likely in female senior residents, married residents, divorcées/widows, women training in the largest programs, women in non-military programs, and women in the Northeast and Midwest (all p<0.05).  Sexual harassment was reported by 19.9% of female residents, most frequently from patients/families (31.2%), attendings (30.9%), and co-residents (15.4%).  Sexual harassment was more likely in senior residents, unmarried residents, women in non-military programs, women training in the Northeast, women in programs with lower proportions of female residents, and women in programs with the highest proportion of female faculty (all p<0.05). There was wide variation in program-level rates of (proportion of residents reporting) gender discrimination (0% to 66.7%) and sexual harassment (0% to 37.5%).

Conclusion: Gender discrimination, pregnancy/childcare discrimination, and sexual harassment are frequent experiences in general surgery residency, particularly for women. Programmatic variation indicates that low mistreatment rates are feasible. The sources (patient/families vs. attendings/co-residents) vary and must be considered when developing potential mitigation strategies.  These results begin to provide insight on how to build safer, more equitable, and more effective educational environments.

81.10 Renal Function Changes Following Fenestrated Endovascular Abdominal Aortic Aneurysm Repair (fEVAR)

Posted on January 31, 2019

M. M. Oberdoerster1, M. M. Wynn1, P. D. DiMusto1  1University Of Wisconsin,Madison, WI, USA

Introduction:  Fenestrated endovascular abdominal aortic aneurysm repair (fEVAR) has been approved for clinical use since 2012.  One possible complication of this repair is impairment in renal function. We sought to assess the clinical outcomes related to renal function over time in patients undergoing fEVAR at our tertiary referral center.

Methods:  A retrospective review was conducted of prospectively collected data on all patients undergoing fEVAR at our institution between 2012 and 2017.  Patient characteristics, procedural variables, laboratory values, and imaging characteristics were collected. Serum creatinine was measured preoperatively, and at 1 month, 6 months, 1 year and yearly thereafter. Estimated glomerular filtration rate was calculated using the Cockcroft Gault equation.

Results: A total of 58 patients were included: 42 men and 16 women with an average age of 75 years. The average follow up time was 469 days; 6 patients were lost to follow up.  A total of 111 out of 116 main renal arteries were successfully revascularized. Only one main renal artery that was planned for revascularization was not due to the inability to place a stent intraoperatively. The other four were above the graft and did not require revascularization.  There were 11 accessory renal arteries that were covered with the aortic graft. 

Eighteen patients (31%) had advanced chronic kidney disease (CKD) prior to the repair; 17 stage III, 1 stage IV.  Nine patients (15.5%), including 3 of the 18 who had CKD prior to the repair, had an increase of at least 30% from baseline creatinine over two or more follow up visits.  All nine had evidence of post-operative renal insult including infarct, renal artery stenosis, or occlusion. Four patients (7%) with no prior history of CKD progressed to stage II, 13 (22%) patients progressed from stage II to III, and 3 patients (5%) progressed from stage III to stage IV over the follow up period. Only the patient with stage IV CKD at the time of the repair went on to require dialysis 8 months after his procedure. 

Eight patients (13.7%) developed stenosis in one of the renal artery stents.  Five of these patients had worsening of their renal function with progression by one CKD stage.  Two patients had an intervention due to the stenosis, but only one was successful.  Four additional patients developed a renal artery occlusion, three of which had a progression of CKD by one stage.  None of these patients went on to dialysis. A total of 24 patients (41%) developed renal infarction on imaging over the follow up period. Nine of these patients had progression of their CKD stage, however none went on to dialysis.  

Conclusion: Our review demonstrates that while kidney dysfunction can occur over the long term following fEVAR, rates of worsening renal function are relatively low.  Additionally, in patients with pre-operative CKD, fEVAR remains safe and effective with low rates of progression to dialysis.  
 

81.09 Assessment of the “Weekend Effect” in Lower Extremity Vascular Trauma

Posted on January 31, 2019

A. K. Jundoria1, B. Grant1, O. A. Olufajo1, E. De La Cruz1, D. Metcalfe2, M. Williams1, E. E. Cornwell1, K. Hughes1  1Howard University College Of Medicine,Washington, DC, USA 2University of Oxford,Nuffield Department Of Orthopaedics, Rheumatology And Musculoskeletal Sciences (NDORMS),Oxford, OX3 9BU, United Kingdom

Introduction:  Numerous studies have suggested that compared to the weekday, weekend admissions may be associated with worse patient outcomes across a range of patient diagnoses. Lower extremity vascular trauma is increasingly common and requires immediate/urgent surgical intervention. Although this weekend effect has been reported for several domains, it has not been elucidated in vascular trauma. The objective of this study was to determine if there is a weekend effect in patient outcomes of lower extremity vascular trauma (LEVT).

Methods:  Retrospective data was retrieved from the National Inpatient Sample database, a 20% stratified sample of the United States inpatient population, from 2005 – 2014. Patients ages 18 and above with International Classification of Diseases, 9th Edition codes indicating trauma to the lower extremity vessels were included. Patients and hospital characteristics were extracted including age, sex, race, insurance type, median household income, Injury Severity Score, Charlson comorbidity score, Abbreviated Injury Scale (AIS) for extremity body region, and location/teaching status of hospital. Outcomes (mortality, amputation, hospital length of stay, and discharge disposition) among patients admitted on weekdays versus patients admitted on weekends were measured. Independent factors associated with outcomes were identified using multivariable regression models. Supplementary analyses were performed using patients with only isolated LEVT, which was defined by AIS of zero in every body region except extremity.

Results: There were 9282 patient records with LEVT (2866 admitted on the weekend vs. 6416 admitted on the weekday). Compared to patients admitted on the weekday, patients admitted on the weekends were more likely to be younger than 45 years old (67.6% vs. 55.4%), males (80.5% vs. 74.6%), and uninsured (22.1% vs. 17.2%) [all p < 0.001]. Comparison of outcomes for patients on weekend vs. weekday showed mortality of 3.80% vs. 3.29% [p = 0.209], amputation rates of 7.85% vs. 7.19% [p = 0.258], hospital length of stay (LOS) of 15.5 days vs. 13.8 days [p = 0.009], and discharge home rates of 57.3% vs. 56.1% [p = 0.271]. The multivariable regressions showed the following outcomes for weekend vs. weekday admissions: mortality (Odds Ratio, OR [95% Confidence Interval, CI]); 1.06 [0.79-1.40], amputation (OR [95% CI]); 1.09 [0.89-1.30], discharge home (OR [95% CI]); 0.95 [0.85-1.06] and hospital LOS (predicted mean LOS [95% CI]); 0.33 [-0.34-1.00].

Conclusion: This study demonstrated there was no weekend effect identified in patients admitted with LEVT in the United States. This suggests that there is likely no difference in the level of care given to lower extremity vascular trauma patients, regardless of whether they present on a weekend or on a weekday.

 

81.08 Modified Metabolic Syndrome is Associated with Poor Access Outcomes Following EVR.

Posted on January 31, 2019

M. D. Balceniuk1, P. Zhao1, L. Cybulski1, M. C. Stoner1  1University Of Rochester,Division Of Vascular Surgery,Rochester, NY, USA

Introduction:
Metabolic syndrome (MetS) is a constellation of five clinical and laboratory findings associated with adverse cardiovascular outcomes as well as wound healing outcomes. MetS has also been reported to be associated with type 2 endoleak following endovascular aneurysm repair (EVR). We report impact of modified metabolic syndrome (mMetS) on groin access complications following EVR.

Methods:
The Vascular Quality Initiative (VQI) endovascular aortic repair module 2008-2016 was evaluated. MetS classification was modified based on the variables available within the VQI registry. Patients were considered to have mMetS (study group) if they had all three diagnosis of body mass index (BMI) >30, hypertension and diabetes. Patients missing one or more of these diagnoses were placed in the non-mMetS (control group) group. Access site complications were evaluated between patients with mMetS and those without.

Results:
Over 30,000 patients were included in the analysis, with over 3000 patients met the criteria for mMetS (Table). The non-metabolic group were significantly older and had higher rates of females and white race. The mMetS cohort had higher rates of coronary artery disease, chronic obstructive pulmonary disease and congestive heart failure. Groups were similar regarding access type for percutaneous (65% versus 65%, p=0.987).  Surgical site infections (SSI) were significantly higher in the mMetS group (24 (0.783%) vs 87 (0.317%), p?0.001), whereas no difference was noted for access site hematoma between groups.  Additionally, sub-group regression analysis of the mMetS cohort identified open access type to be a significant predictor of post-operative SSI (p=0.028).

Conclusion:
Our data demonstrates that there is a significantly higher rate of SSI in patients with mMetS following EVR. We have also shown that percutaneous access in patients with modified metabolic syndrome significantly reduced the incidence of post-operative SSI. These data suggest that percutaneous access should be attempted in EVR patients with metabolic syndrome to mitigate the risk of open femoral access SSI.
 

81.07 Transfemoral Carotid Artery Stenting Is Inferior To Carotid Endarterectomy In The Community

Posted on January 31, 2019

J. Nicklas1, J. Albright1, E. Jerzual1, A. Obi1, P. Henke1  1University Of Michigan,Vascular Surgery,Ann Arbor, MI, USA

Introduction: The Carotid Revascularization Endarterectomy versus Stenting Trial (CREST) suggested that there was no significant difference between carotid endarterectomy (CEA) and transfemoral carotid artery stenting (CAS) in the endpoint of major adverse cardiovascular events (MACE) including stroke, myocardial infarction, or death but a 2 fold increased risk of stroke with CAS. Limited reports exist in real world hospital outcomes for these procedures.

Methods: 11,290 patients (pts) who underwent CEA and 2,391 pts who underwent CAS in 34 hospitals between 1/3/2012 and 2/28/2018 were assessed for baseline demographics, anatomical variables, medications, and major outcomes, at 30 days. Statistical modeling was done with univariable and multivariable analysis for stroke and MACE.

Results: Demographics in CREST and our population were similar in age (69 vs 70 years), gender (65% vs 60% male), and race (93% vs 93% white). However, in our population, the 30-day incidence of MACE among pts undergoing CEA vs CAS was 2.8% (n=321) vs 5.9%, (n=141), p<0.001 and the incidence of stroke was 2.8% in CEA (n=309) vs 4.4% in CAS (n=105), p<0.001. After multivariable adjustment, CAS was associated with increased stroke risk – (OR = 1.323; 95% confidence interval = 1.018, 1.718, P = .036) and MACE- (1.261; 1.004, 1.584, P = .046) as compared with CEA.
Preoperative risk factors for stroke after CEA included: female gender – (1.33; 1.06-1.67), non-ambulatory status – (1.54; 1.17-2.02), contralateral carotid artery occlusion – (2.19; 1.51-3.17), restenosis of a prior CEA (2.22; 1.21-4.07), prior stroke (2.33; 1.56-3.17), and restenosis of a prior CAS (3.85; 1.31-11.34). Significant perioperative risks of stroke include: non-patch closure – (1.44; 1.03-2.02) and eversion CEA technique – (1.50; 1.02-2.22). 30-day MACE analysis showed that the use of a shunt – (1.28; 1.03-1.59) and renal failure requiring dialysis – (2.44; 1.25-4.74) increased the risk.
Preoperative risk factors for stroke after CAS included: low BMI – (1.06; 1.03-1.10), lack of anemia – (1.56; 1.05-2.57), diabetes – (2.18; 1.43-3.32), and non-white race – (2.25; 1.23-4.10). Significant perioperative risks include: increased operative time – (1.55; 1.23-1.95), embolic protection failure – (1.85; 1.01-3.39), and tracheostomy – (10.35; 1.72-62.33). 30-day MACE factors also included: advanced age – (1.024; 1.01-1.04), lack of Ace/ARB medications – (1.51; 1.05-2.15), >=2 coronary arteries >70% obstructed – (2.60; 1.28-5.27), and left main stenosis >50% – OR 2.84 (1.16-6.91).

Conclusion: Our large multi-hospital real world practice registry suggests that CEA in community practice is associated with better outcomes than transfemoral CAS, and diverges from the CREST results. Differential factors were associated with CEA and CAS risks for stroke and MACE.

 

81.06 Does Kidney Transplant Increase the Risk of Ipsilateral Lower Extremity Deep Venous Thrombosis?

Posted on January 31, 2019

S. Ahmed1, D. Kim2, Z. Al Adas1, M. Weaver1, J. Lin2, T. Nypaver1, A. Shepard1, L. Malinzak2, L. S. Kabbani1  1Henry Ford Health System,Department Of Vascular Surgery,Detroit, MI, USA 2Henry Ford Health System,Department Of Transplant Surgery,Detroit, MI, USA

Introduction:

There is limited information on the development and laterality of symptomatic deep vein thrombosis (DVT) following kidney transplantation. In this study, we want define the incidence of DVT in this population and determine if the side of the DVT corresponds to the side of the transplanted kidney.

Methods:

We performed a retrospective review of all kidney transplant recipients from January, 2004 to August, 2014 at our institution and who subsequently developed symptomatic DVT. Kidney transplant recipients and confirmed DVT patients were obtained as two separate data files and were matched to obtain our cohort. Patients with concomitant pancreatic transplants, repeat, and bilateral kidney transplants were excluded. We used Cohen’s kappa statistic to test the agreement between the surgical incision site of the kidney transplant to the side at which the DVT occurred.

Results:

A total of 1827 kidney transplant recipients were performed between January 2004 to August 2014. A total of 877 kidney transplant recipients met the inclusion criteria as our total cohort. From our total cohort, 217 recipients underwent ultrasounds to rule out DVT. A total of 41 kidney transplant patients received a positive duplex ultrasound. The incidence of DVT in our kidney transplant cohort was 4.7%. The most common period of DVT diagnosis was in the perioperative period within the first 4 weeks. A Cohen kappa statistic of -0.02 occurred between the surgical incision site of the kidney transplant and the side of DVT occurrence. Large positive kappa statistic values indicate agreement, whereas large negative values indicate disagreement. Approximately 64.6% of transplant patients with a positive duplex ultrasound had a 1:1 correlation to the side of DVT, although this did not reach statistical significance. There was no statistically significant difference in patient sex, race, or age between the two groups.

Conclusion:

The incidence of symptomatic DVT in this cohort was 4.7%, which is lower than that reported in the literature. DVT was highest during the first four weeks postoperatively. There was an increased rate of ipsilateral DVTs to the kidney transplant, although this did not reach statistical significance. ?

 

81.05 New 5-Factor Frailty Index Can Predict Outcomes in Patients Undergoing Endovascular Aneurysm Repair

Posted on January 31, 2019

V. Pandit1, M. Zeeshan1, S. Jhajj1, A. Lee1, K. R. Goshima1, C. Weinkauf1, W. Zhou1, T. Tan1  1University Of Arizona,Department Of Surgery,Tucson, AZ, USA

Introduction:
The modified frailty index (mFI-11) is a NSQIP-based 11-factor index that has been proven to adequately reflect frailty and predict mortality and morbidity. In past years, certain NSQIP variables have been removed from the database; as of 2015, only 5 of the original 11 factors remained. The predictive power and usefulness of these 5 factors in an index (mFI-5) have not been proven in patients undergoing endovascular repair (EVAR) of abdominal aortic aneurysm (AAA). The goal of our study was to compare the mFI-5 to the mFI-11 in terms of value and predictive ability for mortality, postoperative infection, and unplanned 30-day readmission.

Methods:
The mFI was calculated by dividing the number of factors present for a patient by the number of available factors for which there were no missing data. Spearman's rho test was used to assess correlation between the mFI-5 and mFI-11. Predictive models, using both unadjusted and adjusted logistic regressions, were created for each outcome for EVAR using 2005-2012 NSQIP data, the last year all mFI-11 variables existed.

Results:
A total of 14,387 patients were included. Mean age was 71.6±6.5. Overall, 22.1% patients were frail. A total of 24.1% had in-hospital complications, while 4.9% patients died. Correlation between the mFI-5 and mFI-11 was above 0.9 across all outcomes for patients undergoing EVAR. Adjusted and unadjusted models showed similar c-statistics for mFI-5 and mFI-11, and strong predictive ability for mortality, postoperative complications and 30-days readmission (Table 1).

Conclusion:
The mFI-5 and the mFI-11 are equally effective predictors of postoperative outcomes in patients undergoing endovascular aneurysm repair of abdominal aortic aneurysm and the mFI-5 is a strong predictor of postoperative complications, mortality and 30-d readmission. It has credibility for future use to study frailty within the NSQIP database. It also has potential in other databases and for clinical use.
 

81.04 Is Non-ruptured AAA Repair Still a Worthwhile Solution for Nonagenarians?

Posted on January 31, 2019

L. A. Huntress3, J. Kalenik2, V. Dombrovskiy3, S. G. Huang3, R. Shafritz3, S. Rahimi3  2University of Georgia,Athens, GA, USA 3Rutgers RWJMS,Division Of Vascular Surgery,New Brunswick, NJ, USA

Introduction:  The practicality of non-ruptured abdominal aortic aneurysm (AAA) repair in patients aged 90 and above (90+) remains in question.  This study assessed the short-term outcomes of elective endovascular aortic repair (EVAR), and compared them between patients aged 90+ and patients aged 80-89.

Methods: The National Inpatient Sample (NIS) was queried from 2012-2015, and patients aged 80 and above who had undergone elective EVAR were identified. Postoperative complications, hospital mortality, hospital length of stay, and total hospital cost in patients aged 90+ were compared to those aged 80-89 using Chi square test; multivariable logistic regression analysis controlled for age, gender, race, comorbidities, and hospital characteristics; and Wilcoxon rank sum test. Total hospital cost was adjusted to 2015 U.S. dollars.

Results: A total of 26,115 patients were estimated: 24,210 (92.7%) aged 80-89 and 1,905 (7.3%) aged 90+. Compared to octogenarians, patients aged 90+ were more likely to develop postoperative cardiac complications (OR [odds ratio]=1.68; 95%CI [confidence interval] 1.29-2.18), stroke (OR=3.19; 95%CI 1.59-6.39), urinary tract infection (OR=1.39; 95%CI 1.06-1.82), and bleeding (OR=1.41; 95%CI 1.23-1.62), the last of which required more blood transfusions (OR=1.31; 95%CI 1.14-1.52). However, hospital mortality did not differ among both age groups, and no differences in mortality among various races or genders were observed. At the same time, mortality overall in urban teaching and non-teaching hospitals was lower than in rural hospitals (OR=0.37; 95%CI 0.22-0.62 and OR=0.43; 95%CI 0.25-0.74, respectively). Hospital length of stay (median= 2 days, IQR [interquartile range] 1-3 days in both groups) and total hospital cost (median= $27,950; IQR $21,193-36,079 in octogenarians vs median= $27,203; IQR $20,436-36,332 in nonagenarians; P=0.47) did not differ between the two age groups

Conclusion: Although patients aged 90+ are at an increased risk of some postoperative complications following elective EVAR, hospital mortality and hospital resource utilization in this group of patients are not greater than in octogenarians. Elective EVAR should be considered in nonagenarians with non-ruptured abdominal aortic aneurysm.

 

81.03 Hepatoportal Venous Trauma: Analysis of Incidence, Morbidity and Mortality

Posted on January 31, 2019

S. Maithel1, A. Grigorian1, N. Kabutey1, B. Sheehan1, S. Gambhir1, J. Nahmias1  1University Of California – Irvine,Surgery,Orange, CA, USA

Introduction:

Traumatic injuries to the superior mesenteric, portal and hepatic veins are rare with an incidence of roughly 0.1%. However, the mortality rates are high ranging from 45-52.7% for superior mesenteric vein (SMV), 50-70% for portal vein (PV), and 50-100% for hepatic vein (HV) injuries in small previous single center reports. We hypothesize that SMV injury is associated with lower risk of mortality compared to HV and PV injury in adult trauma patients.

Methods:

The Trauma Quality Improvement Program database (2010-2016) was queried for patients with injury to either the SMV, PV, or HV. A multivariable logistic regression model was used for analysis.

Results:

From 1,403,466 patients, 509 patients had SMV injury, 357 patients had PV injury and 255 patients had HV injury. Compared to patients with PV and HV injuries, patients with SMV injuries were older (39 years vs 29 years, p<0.01), had lower injury severity score (25 vs 26, p<0.01), and a lower percentage of severe (grade >3) abbreviated injury scale for abdomen (57.6% vs 72.3%, p<0.01). A higher percentage of SMV injuries were from blunt mechanism compared to portal and hepatic vein injury (60.3% vs 48.1%, p<0.01). Patients with a SMV injury had a longer length of stay (9 days vs 6 days, p=0.01), higher rates of concurrent bowel resection (38.1% vs 9.9%, p<0.01), and lower mortality (36% vs 47.9%, p<0.01) compared to patients with PV and HV injuries. However, after controlling for covariates, traumatic SMV injury increased risk of mortality (OR 2.37, CI=1.55-3.62, p<0.001) in adult trauma patients as did PV injury (OR 3.74, CI=2.29-6.12, p<0.001) and HV injury (OR 3.44, CI=1.95-6.07, p<0.001).

Conclusion:

Traumatic SMV injury is associated with a lower rate of mortality compared to injuries of the HV and PV. SMV injury greater than doubles the risk of mortality in adult trauma patients, whereas HV injury more than triples the risk and portal vein injury nearly quadruples the risk of mortality.

 

81.02 Image Based 3D CT Decreases Radiation Exposure During Fenestrated Endovascular Aortic Aneurysm Repair

Posted on January 31, 2019

H. Weissler1, K. Southerland1, S. Nag1, C. Long1, B. Gilmore1, M. Turner1, L. Olivere1, M. Cox1, C. Shortell1  1Duke University Medical Center,Vascular Surgery,Durham, NC, USA

Introduction: Fenestrated endovascular aortic aneurysm repair (FEVAR) has expanded the benefits of endovascular aortic aneurysm repair (EVAR) to a population of patients who would have otherwise been anatomically unfit for endovascular repair.  However, FEVAR is associated with high radiation doses and contrast loads due to its increased complexity.  Three-dimensional (3D) fusion computed tomography (CT) merges the preoperative CT with intraoperative imaging to create a vascular mask and has been shown to decrease radiation and contrast use during FEVAR.  Currently available 3D fusion systems use hardware-based (i.e. operating table) tracking to position the overlay on the fluoroscopic image.  This is labor intensive and often leads to inaccurate overlays.  Our institution recently implemented a novel, cloud-based 3D fusion system which uses the patient’s vertebral anatomy rather than the operating table to register and create the overlay.  This system has been shown in prior studies to be highly accurate and decrease radiation dose required for EVAR. The purpose of this study was to determine if radiation dose reduction during FEVAR would occur with this new 3D fusion strategy.

Methods:
Our institutional database was reviewed to identify patients who underwent elective FEVAR.  Patients treated using our cloud-based 3D fusion software CT were compared to patients treated in the immediate 6 months prior to the implementation of 3D fusion CT.  Primary end points included patient radiation exposure (mGy), contrast use (mL), and fluoroscopy and procedure times (minutes).

Results:
Thirty-three patients underwent FEVAR from October 2016 through June 2018, twenty prior to implementation of 3D fusion CT and thirteen after. There was no difference between these groups regarding demographics, BMI  or comorbidities. Radiation dose was significantly decreased following 3D fusion CT implementation (5735 ± 2651 mGy versus 3503 ± 2422 mGy, p=0.019).  In addition, there was a significant decrease in the number of FEVARs requiring a high radiation dose (> 2Gy) with 3D fusion CT (9 vs 19, p=0.044).  There was no difference in fluoroscopy time (72.7 ± 16.9 minutes versus 62.9 ± 15.1 minutes, p=0.061), procedure time (257.6 ± 100.1 minutes versus 213.3 ± 41.1 minutes, p=0.118) and contrast volume (94.5 ± 34.7 mL versus 72.8  ± 37.8 mL, p=0.168) between the two groups.

Conclusions:
These results demonstrate that the use of an intraoperative image-based 3D fusion CT strategy based on the patient’s vertebral anatomy rather than hardware can significantly decrease radiation exposure during FEVAR.  Endovascular solutions to aortic pathology will undoubtedly continue to expand; therefore radiation safety will be paramount.  Image based 3D fusion CT has the potential to improve clinical and safety outcomes for both patients and providers.

 

« Previous Page
Next Page »