12.13 Echocardiographic Guidance During Neonatal and Pediatric ECMO Cannulation is Not Necessary in All Patients

Posted on January 18, 2018

P. A. Salazar1, D. Blitzer2, S. C. Dolejs1, J. J. Parent3, B. W. Gray1  1Indiana University School Of Medicine,Division Of Pediatric Surgery, Department Of Surgery,Indianapolis, IN, USA 2Indiana University School Of Medicine,Division Of Cardiothoracic Surgery, Department Of Surgery,Indianapolis, IN, USA 3Indiana University School Of Medicine,Section Of Pediatric Cardiology, Department Of Pediatrics,Indianapolis, IN, USA

Introduction: Internal jugular cannula position is traditionally confirmed via plain film at the conclusion of the ECMO cannulation procedure. However, it may be difficult to estimate the location of the right atrium on plain films. A misplaced cannula can result in need for repositioning and increased morbidity. Echocardiography (ECHO) may be used during cannulation as a more accurate means of guiding cannula position.  The aim of this study is to study the effect of a protocol encouraging the routine use of ECHO at the time of cannulation.

Methods: We performed a retrospective review of patients at Riley Hospital for Children who received ECMO support using jugular venous cannulation from January 2013 through October 2016. We compared those who underwent ECHO (ECHO+) at the time of cannulation with those who did not (ECHO-). Our primary outcome was need for cannula repositioning after initial cannulation, with secondary outcome being cannula-related morbidity.  For categorical variables, Fisher's exact or Chi-square tests were used to assess for significance. For continuous variables, the median with interquartile ranges (IQR) are presented, and the Wilcoxon rank sum test was used to assess for significance. All statistical analysis was performed on SAS version 9.4 (Cary, NC).

Results: 89 patients met inclusion criteria: 26 ECHO+ (29%), 63 ECHO- (71%). Most of ECHO+ patients underwent dual-lumen VV cannulation (n=17, 65%), while 32% of ECHO- patients required VV support (p<0.003). Seven (27%) ECHO+ patients and 18 (28%) ECHO- patients had a history of cardiac surgery prior to ECMO (p=0.88). All patients had CXR to verify cannula position, and fluoroscopy was used in 4 ECHO+ patients but no ECHO- patients. There was a major mechanical complication in each group: atrial perforation from a guidewire during cannulation in ECHO+ and late atrial perforation from a loose cannula in ECHO-. Subsequent to cannulation, there were 0.58 ECHO studies per patient to verify cannula position in the ECHO+ group compared to 0.22 ECHO per patient in the ECHO- group (p=0.02).  Two (8%) ECHO+ patients required a cannula repositioning procedure for misplacement during the ECMO run, while 6 (10%) ECHO- patients required repositioning procedures (p=0.78).  In the VV ECMO subgroup, ECHO+ patients required no respositioning, while 4 (20%) ECHO- VV patients required repositioning (p=0.1).  Repositioning procedures resulted in no additional complications.  Survival to discharge was similar in both groups: 54% ECHO+ and 62% ECHO- (p=0.51).

Conclusion: Implementation of a protocol to perform ECHO during jugular cannulations for neonatal and pediatric peripheral ECMO did not result in significantly less repositioning procedures, complications, or ECHO studies performed per patient.  ECHO should be used to guide VV cannulation, as need for repositioning approached significance, and it may be a useful adjunct for surgeons during difficult cannulations.

 

12.10 ECMO Duration Predicts Survival in Congenital Diaphragmatic Hernia

Posted on January 18, 2018

S. M. Deeney1, D. D. Bensard1, T. M. Crombleholme1  1Children’s Hospital Colorado,Department Of Pediatric Surgery,Aurora, CO, USA

Introduction:
Physicians caring for patients supported on extracorporeal membrane oxygenation (ECMO) with congenital diaphragmatic hernia (CDH) may wish to know the chance of survival based on time on ECMO. There are limited data reporting the predicted survival outcomes of these patients as a function of ECMO support duration. We aim to describe survival rates in patients with CDH repaired on ECMO in relation to their duration of ECMO support.

Methods:
Retrospective patient data of all patients who underwent repair of CDH while on ECMO from 2008 through 2015 was collected at our institution. Statistical analysis was by logistic regression analysis and chi square, p<0.05.

Results:
There were 22 patients with 10 surviving to discharge. The total number of time spent on ECMO predicted survival to discharge in CDH patients (p=0.006). For every additional day on ECMO, the odds of survival changed by a factor of 0.86 (0.75-0.99). The odds of survival was 50% after 12 days on ECMO, and 25% after 21 days. There was no survival in our patients after 26 days on ECMO.

Conclusion:
Duration of ECMO support predicts survival in patients with CDH repaired while on ECMO, with low survival after 3 weeks and no survivors after 4 weeks. This information may be useful in guiding goals of care conversations.
 

12.11 Using Surgeon-Reported Categorization of Pediatric Appendicitis Severity to Predict Patient Outcomes

Posted on January 18, 2018

K. J. Baxter1, H. L. Short1, C. D. Travers2, K. F. Heiss1, M. V. Raval1  1Emory University School Of Medicine, Children’s Healthcare Of Atlanta,Division Of Pediatric Surgery, Department Of General Surgery,Atlanta, GA, USA 2Emory University School Of Medicine,Department Of Pediatrics,Atlanta, GA, USA

Introduction:  The purpose of this study was to evaluate a novel surgeon-reported categorization (SRC) schema for pediatric appendicitis severity.  We hypothesized that the SRC assigned prospectively at the time of surgery would be superior to standard surgical wound classification in predicting outcomes.

Methods:  We conducted an IRB-approved retrospective review of all appendectomies in children 1-18 years performed at two children's hospitals within a single institution from January to December 2016.  Interval appendectomies were excluded.  The SRC is defined as: simple acute (1), complicated gangrenous or adherent (2A), complicated with perforation and localized abscess (2B), or complicated with perforation and gross contamination (2C). Logistic regression was used to model surgical site infections (SSI) and returns to the system.  Cox proportional hazards survival analyses were used to model length of stay (LOS).  All models were adjusted for patient age, sex, and race.

Results:  The cohort included 397 children and the combined morbidity (SSI and revisits) rate was 9.8%.  Over the course of a 15 month implementation, surgeon compliance with SRC documentation increased from 33.5% to 85.9% and chart review revealed 100% concordance of SRC with reported operative findings.  When modeling combined morbidity, SRC displayed improved model calibration (Hosmer-Lemeshow statistic from 0.418 to 0.829), but had similar outcome discrimination to wound class (C-statistic 0.661 vs. 0.657).  SRC showed an advantage over wound class in predicting SSI alone (C-statistic 0.740 vs. 0.684).  SRC better predicted LOS compared to wound class (Figure).

Conclusion:  Despite an overall low morbidity rate in this cohort, SRC improved prediction of SSI and LOS when compared to wound class.  SRC implementation is feasible and provides a more granular assessment of appendicitis severity which may guide future quality improvement efforts through development of grade-specific care pathways.

 

12.12 Single-Visit Surgery Offers Added Convenience and Excellent Family Satisfaction

Posted on January 18, 2018

C. A. Justus1, A. Milewicz1, M. Wortley1, F. Denner1, R. Bogle1, K. Ceyanes1, S. Shah1  1Texas Children’s Hospital,Pediatric General Surgery,Houston, TEXAS, USA

Introduction: The traditional model for elective ambulatory surgical care includes three separate visits to the surgeon: an initial consultation, a second for outpatient surgery, and a third for postoperative follow-up.  Single-Visit Surgery (SVS) is an alternative model of ambulatory surgical care that increases convenience to patients and their families by decreasing the burden of multiple visits.  SVS consolidates care into a single appointment where patients with straightforward surgical problems are evaluated in the morning and undergo a surgical procedure later that same afternoon.  In April 2016 SVS was introduced at a tertiary-care freestanding children’s hospital for the following conditions: umbilical hernia (over 3 years), inguinal hernia (over 12 months), hydrocele (over 12 months), and epigastric hernia.  Our objective for this study was to evaluate our early experience and conduct a survey of our patient’s caregivers to evaluate their satisfaction with SVS.

Methods: We retrospectively reviewed the medical records of patients that were seen as part of SVS from April 2016 through December 2016.  Data collected included demographics, diagnoses, procedures performed, and distance traveled to the hospital.  Additionally, adult caregivers of SVS patients were contacted and asked to participate in a telephone survey.  The telephone survey evaluated their satisfaction with SVS using a 5-point Likert scale.

Results: There were 43 patients seen through SVS during the study period, and 63% were male.  The median age was 7-years-old (IQR, 4.5 – 10).  The median roundtrip patient commute to the hospital was 30 miles (IQR, 23 – 64).  Of the 43 patients evaluated through SVS, 40 (93%) of them underwent surgery.  The most common procedure performed was inguinal hernia repair (n=20), followed by umbilical hernia repair (n=17), and epigastric hernia repair (n=1).  Of the 40 patients that had surgery, 27 (68%) of the families were contacted and participated in the telephone survey. Of those responding, 93% were strongly satisfied, and 7% were satisfied with the care through SVS. All families said they would recommend the SVS program to a friend.

Conclusion: Single-Visit Surgery is an alternative model of ambulatory surgical care that adds convenience to the patient experience and results in excellent family satisfaction. 

 

12.08 Rigorous Failure Criteria Results in High Yield Post-Operative Imaging for Pediatric Appendicitis

Posted on January 18, 2018

K. T. Anderson1,2, M. A. Bartz-Kurycki1,2, K. Tsao1,2  1McGovern Medical School, The University Of Texas Health Sciences Center At Houston,Houston, TX, USA 2Children’s Memorial Hermann Hospital,Houston, TX, USA

Introduction:  Use of computed tomography (CT) for the diagnosis of pediatric appendicitis is declining due to increased awareness of the harms of radiation exposure. However, CT is still frequently employed following appendectomy to evaluate for superficial surgical site infection, intra-abdominal abscess, and small bowel obstruction. The purpose of this study was to examine the efficacy of a post-operative failure criteria for post-operative imaging use in pediatric appendicitis patients and its impact on change in patient management.

Methods:  Pediatric patients who underwent appendectomy for appendicitis between July 2009 and May 2017 were included. Complicated appendicitis was defined as gangrenous or perforated based on surgical diagnosis. Patients who did not meet discharge criteria, defined by fever (temperature >38°C), leukocytosis (white blood cell (WBC) count >12,000/mm3), lack of diet tolerance and/or continued abdominal pain by post-operative day (POD) 5-7 were deemed non-responders. Imaging was recommended by departmental guidelines at POD 5-7 in non-responders. Imaging modality, ultrasound only (US), US and CT (US+CT) or CT only was at provider discretion. The primary outcome was any intervention, a composite of reoperation, drainage/aspiration procedures, opening a wound, or changing antibiotics.

Results: 3,214 pediatric appendectomy patients were identified, with 12% not responding to initial management. Most non-responders (78%) underwent post-operative imaging; CT only was the predominant modality (69%), followed by US+CT (19%), and US only (12%). Non-responders required a change in management only 32% of the time. On multivariate regression, complicated disease (OR 2.6, 95% CI 1.5-4.6) and presence of fever (OR 2.6, 95% CI 1.5-4.4) were associated with need for intervention but not leukocytosis (p=0.18), POD of diet (p=0.24), or continued abdominal pain (p=0.59). A patient with complicated disease and fever at POD 5, had an 87% probability of requiring an intervention. Given the positive likelihood ratios (+LR) of different imaging modalities (US only +LR= 0.8; US + CT +LR 2.1; CT only +LR=1.5), the post-test probability of requiring an intervention after imaging was 84% (US), 93% (US+CT), and 91% (CT).

Conclusion: In pediatric patients who did not respond to initial management, meeting strict failure criteria resulted in a high yield of those requiring intervention after appendectomy. However, type of post-operative imaging did not sufficiently discriminate between those who required a change in management, which will require further delineation. Reserving ionizing radiation to those who necessitate it for the intervention, such as a drainage procedure, may reduce unnecessary post-operative CT use.

 

12.09 Atresia and Gastroschisis: Results from a Multicenter Study

Posted on January 18, 2018

S. Raymond1, S. D. St. Peter1, C. Downard1, F. Qureshi1, E. Renaud1, P. D. Danielson1, S. Islam1  1University Of Florida,Pediatric Surgery,Gainesville, FL, USA

Purpose:  Atresia is one of the most common associated anomalies in gastroschisis (GS), and renders the anomaly a complex one. A number of single center studies suggest that atresia has minimal effect on morbidity and mortality in GS, while others suggest otherwise. The purpose of this study was to report outcomes in the largest series of atresias in GS.

 

Methods: A retrospective cohort of GS from 8 institutions from 2006-2013 was created and clinical data collected for pre-, peri- and post-natal variables. Patients with a diagnosis of atresia were selected from this cohort for analysis, excluding those with bowel resection due to perforation or ischemia unrelated to atresia. Comparison was performed with other complex GS patients. Student’s t test and Fisher’s exact test were used for statistical evaluation and a p value of <0.05 was considered significant.  

 

Results: Out of 566 cases of GS, 123 (21.7%) were complex and 51 had an atresia. Atresias were primarily in the jejunum and ileum (67%), and 23.5% were multiple. Despite 84% of GS found prenatally, there was no prenatal atresia diagnosis. Atresia patients had a 9.8% mortality rate in the NICU and a 15.8% mortality at follow up. Atresia’s LOS was over 3 months and 70% required repeat laparotomy. At discharge, 43% of atresia patients required TPN. A comparison of atresia with other complex GS shows atresia to be associated with higher rate of intestinal failure and a trend to higher LOS, rate of sepsis, and readmissions (Table).

 

Conclusions: GS associated atresia carries high morbidity and mortality rates, and has significantly higher rates of intestinal failure than other complex GS. A majority required multiple operations, readmissions, and long-term central lines. Identification of an atresia with GS should prompt a transfer to a facility with higher capability. 

 

12.07 Predictive value of Ultrasonography in Diagnosis of Appendicitis in Children

Posted on January 18, 2018

J. A. McKean1, S. Ayub1, D. Rajderkar1, M. M. Mustafa1, J. A. Taylor1, S. D. Larson1, S. Islam1  1University Of Florida,Pediatric Surgery,Gainesville, FL, USA

Introduction:

Abdominal pain in children is one of the most frequent causes of an emergency department (ED) visit. Ultrasound evaluation for appendicitis in that setting has become almost standard of care imaging to reduce radiation exposure from computed tomography. However, this has led to a practice that encourages indiscriminate use of ultrasound (US) and potentially reduces its effectiveness as a diagnostic tool. The purpose of this study was to better understand the utilization and predictive value of US for diagnosis of appendicitis in children.

 

Methods:

All children who underwent an ultrasound evaluation for abdominal pain in the ED over a 5-year period were included in the study. Patients having US for trauma, or for gallbladder disease were excluded. Data regarding clinical presentation, laboratory evaluation, imaging results, and outcomes were collected. The entire cohort was divided based on age, gender, appendicitis score (PAS), and US results and comparative statistics performed. Students t test, Fischer’s exact test, and the Mann-Whitney tests were performed where appropriate and a p value <0.05 was considered significant.  

 

Results: 1650 patients were identified with US evaluation in the ED. A total of 746 children had evaluation for appendicitis. Overall mean age was 11 years, 50.8% were female, and the mean WBC count was 11.88/mm3. Seventy percent of the cohort had a moderate risk PAS score (4-7), while 21% were low risk and 9% were high risk. US results were 63.8% non-visualization (NV), 12.2% positive, and 23.9% negative for appendicitis; a definitive result on US was more likely during the daytime(P=0.002). Further analysis of the NV subset revealed no difference in age or gender, 74% had a moderate PAS score, and 22% underwent a subsequent CT scan. Table 1 shows the difference between NV US patients who were admitted for observation only, discharged from the ED, and who were admitted and underwent an appendectomy. Low risk PAS patients had a 60% NV US and 3% false positive rate. Patients with a high PAS had positive US diagnosis in 43% cases. 

Conclusion:

US utilization for children with abdominal pain is high, even in the setting of a low PAS, where it was not useful. There was a likely reduction in CT scans for patients with a high PAS. Patients with a NV appendix were very common and the use of laboratory and clinical criteria as well as the C-reactive protein were helpful in management. These data will be used to help refine the US use in children in the ED.

 

12.05 Engaging Families Through Shared Knowledge: RCT of Open Access to a Rapid Learning Healthcare System

Posted on January 18, 2018

D. O. Gonzalez1, Y. Sebastiao1, J. N. Cooper1, M. Levitt1, R. J. Wood1, K. J. Deans1  1Nationwide Children’s Hospital,Columbus, OH, USA

Introduction: A rapid learning healthcare system (RLHS) can deliver near-real time data to physicians and families about a disease and its outcomes based on specific patient characteristics. Giving families access to a RLHS may increase patient engagement, improve their knowledge, and lead to better outcomes. This study investigated the impact of allowing families of pediatric patients with complex colorectal diseases access to a RLHS on patient-centered outcomes. 

Methods:  We created a RLHS that integrates pre-specified data elements and validated surveys within the clinical workflow into the electronic health record. The RLHS is an interactive dashboard which contains information on a number of data points, including demographics, quality of life (QOL), surgical outcomes, complications, and continence. We performed a randomized trial of caregivers of children <18 years of age with an anorectal malformation, Hirschsprung disease, or functional constipation visiting our colorectal center. Prior to their initial office visit, participants were randomized to either standard surgical consultation or open access to the RLHS in addition to a standard consultation. To determine the effect of open access to the RLHS on patient-centered outcomes, we assessed healthcare satisfaction, quality of life (QOL), parent activation, health literacy, and caregiver knowledge about their child’s diagnosis. Outcomes between groups were compared at the end of the initial office visit and at 30 days. For participants randomized to the RLHS group, system usability was assessed and an exit interview conducted.

Results: Of 126 participants, 62 were randomized to the RLHS group and 64 were randomized to standard consultation. There were no differences in age, gender, or diagnosis between the patients and no differences in demographics between the caretakers in both groups. At the end of the initial clinic visit, there were no differences in healthcare satisfaction, QOL, parent activation, health literacy, and knowledge of disease between groups. After 30-day follow-up, there were no differences in healthcare satisfaction, QOL, and knowledge of disease. The usability and learnability of the RLHS were ranked 73.0 and 80.4, respectively, on a 100-point scale. Most participants reported that the RLHS included a lot of useful information and was easy to use. Approximately 25.0% of participants reported feeling overwhelmed when looking at the information on the RLHS, but only 3.6% reported that it increased their stress level. The majority of participants (83.9%) would recommend that we provide RLHS access to all families.

Conclusion: Although providing access to the RLHS did not affect patient-centered outcomes, the majority of patients recommended providing families access to this type of system. Although access to the information in the RLHS overwhelmed some of the patients, it did not affect the stress level of the vast a majority of patients.

12.06 Post-operative Complications in Children with Down Syndrome Correlate with Other Comorbidities

Posted on January 18, 2018

M. A. Bartz-Kurycki1,2, K. T. Anderson1,2, M. T. Austin1,2, L. S. Kao1, K. Tsao1,2, K. P. Lally1,2, A. L. Kawaguchi1,2  1McGovern Medical School, University Of Texas Health Sciences Center At Houston,Pediatric Surgery,Houston, TX, USA 2Children’s Memorial Hermann Hospital,Houston, TX, USA

Introduction:  Down syndrome (DS) is associated with multiple co-morbidities, which may increase the risk of post-operative complications. Physicians perceive that DS patients have a higher risk of post-operative complications; however, the literature remains unclear. This study evaluated the risk of post-surgical complications for abdominal and thoracic procedures in children with and without DS.

Methods:  The National Surgical Quality Improvement Program Pediatric (NSQIP-P) database was queried for patients under 18 years of age who underwent non-cardiac abdominal and thoracic operations (by CPT codes) between 2012 and 2016. Analysis compared patients based on the presence or absence of DS. The primary outcome was a composite of all post-operative complications as defined by NSQIP-P. Analysis utilized chi square, student’s t-test, and univariate and multivariate logistic regression. Potential pre-operative risk factors, including DS and those found in the NSQIP-P risk calculator, were evaluated for an association with post-operative complications. Variables were tested for interactions.

Results: 91,478 surgical patients were included with a mean age of 7.2 years ± 6.1 years. Of those patients, 57.8% were male and 1,476 (1.6%) had a diagnosis of DS. Baseline covariates demonstrated significant differences; patients with DS had higher rates of pre-operative nutritional support (38.8% vs. 15.0%), developmental delay (61.9% vs. 10.4%), and cardiac risk factors (76.5% vs 13.8%). The overall rate of post-operative complications was 10.9% and patients with DS demonstrated a higher proportion than controls (16.2% vs 10.8%, p<0.001). On univariate analysis, DS was associated with increased odds of post-operative complications (OR 1.6 95%CI 1.4-1.9) compared to the non-DS group; however, this risk was reversed when adjusting for all other covariates (aOR 0.86 95%CI 0.7-1.1; Table). Univariate analysis of individual surgical complications suggested an increased risk of sepsis (2.6 95%CI 1.8-3.6) and surgical site infection associated with DS (OR 1.6 95%CI 1.3-2.0) but were not statistically significant on multivariate analysis.

Conclusion: Although a greater proportion of post-operative complications were observed in patients with DS, when adjusting for other risk factors, DS was not an independent risk factor. The increased rate of complications is likely related to the presence of certain risk factors that are more common in DS, such as hematologic disorders, cardiac risk factors, nutritional supplementation, and ventilator dependence. Pre-operative counseling and optimization for patients with DS should be tailored to the individual based on their co-morbidities.

12.04 Nonoperative Anesthesia Time for Common Surgical Procedures in Young Children

Posted on January 18, 2018

K. Williams1, B. Nwomeh2, T. A. Oyetunji3  1Howard University College Of Medicine,General Surgery,Washington, DC, USA 2Nationwide Children’s Hospital,Columbus, OH, USA 3Children’s Mercy Hospital- University Of Missouri Kansas City,Kansas City, MO, USA

Introduction: Increasing attention is being paid to the neurotoxicity and potential long-term cognitive effects of general anesthesia (GA) in children younger than 3 years old. This study aims to describe the most common pediatric surgical procedures requiring GA in children less than 3 years, and the potential discordance in nonoperative anesthesia time  and operative anesthesia time.

Methods: The National Surgical Quality Improvement Program Pediatric (NSQIP-P) database for 2012-2013 was queried for all children 3 years old or less who underwent GA for an inpatient pediatric general surgical procedure. Demographic data, procedure by Current Procedural Terminology (CPT) code, case type, operating time, anesthesia time, and American Society of Anesthesiologists (ASA) class were descriptively analyzed. Those who underwent additional procedures during the same anesthesia period were excluded. For each procedure, the difference between total anesthesia time and total operating time was calculated, yielding the nonoperative anesthesia time.

Results: A total of 5143 patients were identified. Of these, 63% were male and 68% were White. The median age at admission was 64 days (IQR 28-294) and 78% were infants. Most cases were elective (59%), and the most common ASA class was 2 (38%). The most common procedures performed in children younger than 3 years were pyloromyotomy (21%), laparoscopic gastrostomy (8.6%), and unlisted procedures on the stomach (6.6%). The median nonoperative anesthesia times were 45 mins (IQR 37,55), 48 mins (IQR 37, 64), and 48 mins (IQR 38, 62) respectively. (Table 1)

Conclusion: For the most common procedures requiring GA in children less than 3 years, the median nonoperative anesthesia time exceeds operative anesthesia time by more than 45 minutes, suggesting that this is a potentially modifiable target for decreasing anesthetic exposure. The provider and systems level factors that contribute to this time discordance need to be explored.

 

12.03 Earlier Feeding after Congenital Diaphragmatic Hernia Repair Associated with Shorter Hospitalization

Posted on January 18, 2018

S. M. Deeney1, D. D. Bensard1, T. M. Crombleholme1  1Children’s Hospital Colorado,Department Of Pediatric Surgery,Aurora, CO, USA

Introduction:
The benefits of early enteral feeding have been demonstrated in expedited recovery after surgery protocols for various procedures such as adult colorectal surgery. Little is known regarding the effect of earlier enteral feeding in patients after congenital diaphragmatic hernia (CDH) repair.

Methods:
Retrospective patient data of all patients who underwent CDH repair from 2008 through 2015 was collected at our institution, excluding patients who died prior to initiation of enteral feeding (n=64). Statistical analysis was by Student’s t test, chi square, and Fisher’s exact test, p<0.05.

Results:
Patients who started enteral feeding 5 days or earlier following CDH repair had a statistically shorter length of hospitalization than those fed 6 days or later postoperatively (51±42 vs 85±125 days, p=0.03). There was no significant difference between groups in survival to discharge (95% vs 85%, p=0.3), ventilator time (24±44 vs 95±167 days, p=0.07), postoperative total infection rates (38% vs 45%, p=0.16) and wound complication rates (2% vs 15%, p=0.09).

Conclusion:
There is an association between earlier enteral feeding and decreased length of hospitalization in patients who have undergone CDH repair. All other factors studied did not reach statistical significance, possibly due to underpowering. These results need to be validated in a prospective randomized, controlled study.
 

12.01 Maintaining Oncologic Integrity with Minimally Invasive Resection of Pediatric Embryonal Tumors

Posted on January 18, 2018

H. M. Phelps1, G. D. Ayers2, J. M. Ndolo3, H. L. Dietrich4, K. D. Watson5, M. A. Hilmes3, H. N. Lovvorn6  1Vanderbilt University Medical Center,School Of Medicine,Nashville, TN, USA 2Vanderbilt University Medical Center,Division Of Cancer Biostatistics,Nashville, TN, USA 3Vanderbilt University Medical Center,Pediatric Radiology,Nashville, TN, USA 4Vanderbilt University Medical Center,School Of Nursing,Nashville, TN, USA 5Vanderbilt University Medical Center,Pediatric Hematology/Oncology,Nashville, TN, USA 6Vanderbilt University Medical Center,Pediatric Surgery,Nashville, TN, USA

Introduction:  Embryonal tumors arise typically in infants and young children and are often massive at presentation. Treatment is multimodal, and while complete resection is a critical element, surgery can interrupt therapy. When appropriate, minimally invasive surgery (MIS) offers a potential means to minimize treatment delays. However, the use of MIS to resect embryonal tumors remains controversial regarding the oncologic integrity of this approach.

Methods:  A retrospective review of embryonal tumors treated at a single institution over a 15-year period was conducted to: 1) assess candidacy of embryonal tumors for MIS, and 2) evaluate outcomes for patients undergoing MIS versus open resection. Query of the institution’s cancer registry identified pediatric patients treated for intracavitary embryonal tumors from 2002 to 2017. To assess amenability for MIS, tumor volume (TV) and image-defined risk factors (IDRF, neuroblastic tumors only) were measured radiographically at time of diagnosis and immediately before resection. Stage, Children’s Oncology Group risk stratification, procedure-related details, delay to next dose of chemotherapy, relapse-free survival (RFS), and overall survival (OS) were evaluated. Wilcoxon, Pearson chi-square, and log-rank tests were performed.

Results: A total of 201 patients were treated for neuroblastic tumors (NBL, n=101), Wilms tumor (WT, n=66), hepatoblastoma (n=23), rhabdomyosarcoma (RMS, n=10), and pancreatoblastoma (n=1). Among these patients, 175 tumors were resected either open (n=151, 86%) or by MIS (n=24, 14%; 20 NBL, 3 WT, 1 RMS). Of the 174 with complete data at time of analysis, the median TV at resection was 84.8 ml [IQR 20.4, 372.5]. For NBL cases, a significantly greater proportion of MIS resections (n=17, 94%) had no IDRF when compared to open resections (n=31, 48%; p<0.001). For the entire cohort, RFS at 5 years was 0.78 [CI 0.71–0.85] for open resection versus 0.90 [CI 0.78–1.00] for MIS (p=0.463). OS at 5 years was 0.87 [CI 0.81–0.93] for open resection versus 1.00 [CI 1.00–1.00] for MIS (p=0.294). The largest TV resected via MIS was 93.4 ml, so subgroup comparisons were adjusted for TV<100 ml. No significant difference in margin status between open resection (n=68) and MIS (n=23) was observed, and MIS was associated with significantly less blood loss, shorter hospital stays, shorter operating time, and quicker return to the next chemotherapy cycle (Table 1).

Conclusion: For appropriately selected patients, MIS resection of pediatric embryonal tumors, particularly NBL, maintains an acceptable oncologic integrity while minimizing treatment delays, but large tumor volume, vascular encasement, and small patient size limit its broader applicability.

12.02 Does Muscle Biopsy change the treatment of Pediatric muscular disease?

Posted on January 18, 2018

N. Le1, J. Sujka1, J. Sobrino1, L. A. Benedict1, R. Rentea1, H. Alemayehu1, T. Oyetunji1, S. St. Peter1  1Children’s Mercy Hospital- University Of Missouri Kansas City,Kansas City, MO, USA

Introduction:
Muscle disease presentation is highly variable. Tissue biopsy is performed to confirm diagnosis and to guide therapy however it is unclear if this changes management. The purpose of our study was to determine if muscle biopsy changed patient diagnosis or treatment, which patients were most likely to benefit from muscle biopsy, and the complications resulting from muscle biopsy.

Methods:

With IRB approval a retrospective chart review of all patients less than 18 years old undergoing muscle biopsy between Jan 2010 and Aug 2016 was performed.  Demographics, patient presentation, change in treatment and diagnosis, hospital course, and follow up were evaluated. T-test and descriptive statistical analysis was performed; all means reported with standard deviation.

Results:

A total of 90 patients underwent a muscle biopsy at our institution during the study period. Mean age at time of biopsy was 6.5 years (±5) with the most common site of biopsy was the vastus lateralis. Of these patients only 37% (n=34) had a definitive diagnosis by muscle biopsy, in the remaining patients 27% (n=25) were normal and 35% (n=31) were non diagnostic. Of all patients biopsied, 39% (n=35) had a change in their diagnosis with only 37% (n=34) having a change in their treatment course from the pathology result.

Among the 34 patients who had a change in their treatment, the most common diagnosis was Inflammatory muscle disease at 44% (n=15) followed by those with muscular dystrophy, 23% (n=8). In the 56 patients who did not have a change in treatment, the most common diagnosis was hypotonia at 30% (n=17) followed by those patients whose diagnosis remained undetermined at 25% (n=14). However, the third most common diagnosis that did not change treatment was inflammatory with 21% (n=12). Two of these patients had definitive diagnosis from their biopsy but the remaining 10 were previously suspected of, and being treated for, myositis.

Using the T-test we compared those who did or did not have a change in treatment based on their pathology. There was no statistically significant difference in the patient’s weight, age, operative duration, or length of follow up post operatively. One patient was found to have a complication from surgery, malignant hypothermia. No patients returned to the operating room secondary to operative complications. Mean length of follow up for all patients was 2.75 years (±2).

Conclusion:

Muscle biopsy could be considered to diagnose patients with symptoms consistent with inflammatory or dystrophic muscular disease though the likelihood of this altering the patient’s treatment course is less than 50%.

11.18 Pancreatic Neuroendocrine Tumor (PNET) Imaging Features are Predictive of Biology

Posted on January 18, 2018

A. Fang1, E. Tashakori1, C. Farinas1, M. Mederos1, A. McElhany1, S. Mohammed1, N. Villifane1, W. E. Fisher1, G. Van Buren1  1Baylor College Of Medicine,Michael E. DeBakey Department Of Surgery, The Elkins Pancreas Center,Houston, TX, USA

Introduction:
Pancreatic neuroendocrine tumors (PNETs) are a heterogeneous group of tumors with variable malignant potential. While most PNETs appear as solid, well-circumscribed, and enhancing masses on computed tomography (CT), their heterogeneous nature can lead to differences in their stereotypical appearance. Therefore, the objective of this study was to test the hypothesis that PNETs with atypical features on CT were associated with more aggressive pathologic features.

Methods:
Through our prospectively maintained Pancreas Surgery Registry, we retrospectively analyzed the radiologic and pathologic features of resected PNETs between January 2005 and December 2015. After independent verification, the CT characteristics such as size, morphology, and enhancement pattern were compared to their histopathologic findings using Chi square and Fisher’s exact tests. Multivariate logistic regression models were generated with backward selection method and a removal p value of 0.1. 

Results:
We analyzed 52 patients who underwent preoperative CT imaging and PNET resection. 29 lesions demonstrated abnormal imaging characteristics such as hypoenhancement (4), isoenhancement (1) calcifications (7), and cystic appearance (19). On univariate analysis, cystic appearance was significantly associated with lower grade and less lymphovascular invasion (p = 0.02, and 0.04, respectively). Bile duct dilation and lymphadenopathy were significantly associated with higher grade and stage, (p = 0.02 and 0.01, respectively). Tumor size significantly correlated with higher stage, positive margin, and lymphovascular invasion (p = 0.02, 0.03, and 0.02, respectively). On multivariate analysis, when controlling for lymphadenopathy, bile duct dilation, and image size, cystic lesions were a significant predictor of lower staging (p = 0.003, 0.01, and 0.01, respectively) and lower rates of lymphovascular invasion (p = 0.04). Hypoenhancment, isoenhancment, and calcifications did not correlate with aggressive pathologic findings.

Conclusion:

This cohort study demonstrated that PNETs with cystic appearance were less aggressive. Conversely, characteristics such as lymphadenopathy, bile duct dilation, and larger lesion size were predictors of aggressive pathologic characteristics. However, enhancement pattern and the presence of calcifications on CT were not associated with more aggressive features. 

11.19 Smoking and Next Generation Sequencing Mutation Signature in Melanoma

Posted on January 18, 2018

K. Loo1, I. Soliman1, M. Renzetti1, T. Li1, H. Wu1, B. Luo1, A. Olszanski1, S. Movva1, M. Lango1, N. Goel1, S. Reddy1, J. Farma1  1Fox Chase Cancer Center,Philadelphia, PA, USA

Introduction: The use of molecular profiling to characterize tumors is becoming increasingly utilized to guide and tailor therapies for personalized treatment in the setting of malignant melanoma. Furthermore, smoking has been identified as a largely preventable cause of cancer mortality. Yet it remains to be seen whether smoking has a causative or protective effect in the setting of malignant melanoma. Using Next Generation Sequencing (NGS), we investigated a panel of 50 targetable cancer-related gene mutations in melanoma tumors. The principle aim of this study was to investigate the correlations between previous history of smoking with genetic mutations among individual genes, as well as total mutation burden in patients with malignant melanoma.

Methods: A retrospective study was conducted to include both primary and recurrent malignant melanoma tumor samples. Utilizing a prospective database, we identified a cohort of patients whose tumor tissue samples underwent NGS sequencing analysis for somatic mutations of 50 cancer-related genes. Within this cohort, clinical and pathological data were also collected. A univariate analysis was conducted using Fisher’s exact and Wilcoxon tests to compare patients with previous history of smoking to never smokers to investigate differences in each cohort’s molecular profile.  

Results: A total of 173 patients with malignant melanoma whose tumor tissue specimens underwent NGS sequencing were analyzed in this study cohort. Median age at diagnosis was 65 (range 21-94) and 64% were male (n=111). The smoking cohort was divided into never smokers (n=72) versus current or former smokers (n=101). Of the 168 patients with staging data, 9% of patients were Stage I melanoma (n=15), 30% Stage II (n=50), 49% with Stage III (n=83), and 12%with Stage IV (n=20).

In the total cohort, 277 mutations were identified affecting 34 unique genes. No mutations were found in 12% of patients (n=20), while 47% of patients (n=82) had 1 mutation, 24% (n=41) had 2 mutations, 9% (n=16) had 3 mutations, and 8% (n=14) had 4 or more mutations. The most common mutations among patients with a history of smoking were BRAF v600E (27.7%, vs. 19.4% in never smokers) and CDKN2A (12.87%, vs. 9.72% in never smokers) genes. Conversely, the most common genes among never smokers were NRAS (34.72%, vs. 31.7% in smokers) and TP53 (22.22%, vs. 21.8% in smokers). The overall mutation burden in the never smoker cohort was 1.59 versus 1.60 in the current and former smoker cohort (p=0.94).

Conclusion: This study demonstrated no significant association of overall mutational burden or increased incidence of individual gene mutations to smoking status, additional studies are needed to identify the effect of smoking on melanoma tumor characteristics with a larger sample size. Further studies with additional tumor biomarkers are additionally warranted to discern the impact of smoking on malignant melanoma tumors. 

11.20 Postoperative Opioid Consumption in Cancer Patients after Curative-Intent Surgery

Posted on January 18, 2018

J. A. Balch1, J. S. Lee1,2, V. Parashar1,2, J. B. Miller3, S. M. Bremmer1,2, J. V. Vu1,2, L. A. Dossett1,2  1University Of Michigan,Department Of Surgery,Ann Arbor, MI, USA 2University Of Michigan,Center For Health Outcomes And Policy,Ann Arbor, MI, USA 3University Of Michigan,Center For Bioethics & Social Sciences In Medicine,Ann Arbor, MI, USA

Introduction:  

Multiple studies demonstrate that opioid prescriptions far exceed patient consumption after surgery. These excess opioids can be inappropriately used or diverted for nonmedical use, thereby contributing to the national epidemic of opioid-related morbidity and mortality. Cancer patients are at particular risk for chronic pain and opioid use, and have higher rates of persistent opioid use after curative-intent surgery. To more accurately assess the amount of opioid needed postoperatively in cancer patients, we systematically reviewed the literature to evaluate opioid consumption in cancer patients after curative-intent surgery.

Methods:

MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Trials were searched to identify studies describing opioid consumption in cancer patients after curative-intent surgery. The primary outcome was the amount of opioid consumed within 24 hours after the operation concluded. Reported opioid amounts were converted to oral morphine equivalents (OME) for comparison. Three reviewers independently screened studies for inclusion, extracted data, and assessed study quality.

Results:

Of the 31 studies reviewed in full, sixteen eligible studies reported opioid consumption for five types of curative-intent procedures, including breast, gynecologic, colorectal, renal, and lung cancer. We used 24 hours as a time point as this number represents peak postoperative opioid consumption. The amount of opioid consumed in this time point varied widely across surgical disciplines, ranging from 4 OME for breast cancer to 208 OME for gynecologic cancer (Figure1). Within each discipline, the amount of opioids consumed also varied widely depending on the use of non-opioid analgesics, peripheral nerve blocks, and epidural analgesia. For example, for breast cancer, the study group with the lowest reported opioid consumption (4 OME; Albi-Feldzer 2013) used scheduled acetaminophen and ketoprofen. In contrast, the breast cancer study group with the highest opioid consumption (72 OME; Terkawi 2014) had no standardized protocol for non-opioid analgesics.

Conclusion:

In cancer patients undergoing curative-intent surgery, opioid consumption varies widely depending on the use of non-opioid analgesics, peripheral nerve blocks, and epidural analgesia. Standardized pain management protocols may reduce variation in opioid consumption and minimize excessive opioid prescribing.
 

11.16 Perioperative Chemoradiation Does Not Influence Incisional Hernia Formation After Cancer Resection

Posted on January 18, 2018

S. Kavalukas1, R. Baucom2, L. Huang1, S. Phillips1, C. Bailey1, R. Pierce1, M. Holzman1, K. Sharp1, B. Poulose1  1Vanderbilt University Medical Center,Department Of Surgery,Nashville, TN, USA 2Baylor University Medical Center,Dallas, TX, USA

Introduction:  The effect of adjuvant therapy on subsequent ventral incisional hernia (VIH) formation after cancer resection is unknown.  This study assessed the impact of adjuvant therapy on long term VIH formation after tumor resection for abdominal malignancy.

Methods:  Patients undergoing resection of abdominal malignancy were retrospectively identified and followed up to 2 years by computed tomography (CT) scan for VIH formation.  Those who received either chemotherapy and/or radiation 6 months before or after resection (adjuvant therapy, AT) were compared to patients without adjuvant therapy (no adjuvant therapy, NAT).  Cox proportional hazards (CPH) regression was used identify factors associated with VIH formation over time while adjusting for multiple confounding factors.

Results: 485 patients underwent abdominal cancer resections (AT n=105; NAT n=380). The median age was 58 yrs for the AT group and 61 for the NAT group. 40% of the AT group and 41% of the NAT group were female. The clinical cancer stage breakdown was statistically significant between the 2 groups (AT had more stage 4 and NAT had more stage 1 cancers). The proportion of patients free of VIH at 24 months was 46% in the AT group and 39% in the NAT group (Figure, p=0.62).  AT was not found to be associated with increased VIH compared to NAT in the CPH model (HR=0.817, 95% CI: (0.567, 1,177).

Conclusion: The rate of incisional hernia formation after resection for abdominal malignancy does not appear to be influenced by adjuvant chemo- or radiotherapy. Continued evaluation of risk factors and the role of hernia prevention is important to maintain quality of life for cancer survivors. 

 

11.17 The Emerging Role of Surgery in Melanoma Patients Treated with Immune Checkpoint Inhibitors (ICI)

Posted on January 18, 2018

C. Puza1, P. Mosca1, A. K. Salama2, H. Howard3, D. Agnese3, A. Terando3, D. G. Blazer1, R. Scheri1, G. Beasley1  1Duke University,Department Of Surgery,Durham, NC, USA 2Duke University,Division Of Medical Oncology,Durham, NC, USA 3Ohio State University,Division Of Surgical Oncology,Columbus, OH, USA

Introduction: The emergence of novel ICI has resulted in dramatic improvements in survival for patients with metastatic melanoma. Relative to traditional chemotherapy, the types of disease response patterns to ICI therapy can be more complex, including mixed responses and pseudoprogression.  Specifically, some lesions may regress while new lesions appear, or tumors may remain stable in size for long periods of time. As the role of surgery in these scenarios is continuing to evolve, the purpose of this study was to explore outcomes associated with surgery following ICI therapy. 

Methods:  A retrospective study was conducted at two centers and included patients with melanoma who underwent surgery following treatment with monotherapy or combination therapy with anti-PD-1 and/or anti CTLA-4 checkpoint blockade.  Data collected included: treatment regimen, toxicities, operative reports, pathology, and clinical plus radiographic follow-up.

Results: Of 17 identified patients, 7 patients had received anti-CTLA-4 therapy, 4 anti-PD-1 therapy, and 6 anti-CTLA-4 therapy plus anti-PD-1 therapy before surgery. Five patients were being treated in the adjuvant setting and developed new lesions while 12 patients were being treated for metastatic disease and underwent surgery for persistent disease on imaging.   Seventeen patients underwent 18 operations including: 4 small bowel resections, 2 splenectomies, 4 wide local excisions, 4 groin dissections, 1 craniectomy, 1 mesenteric mass resection, 1 axillary dissection, and 1 lung nodule resection.  There were no major reported complications from surgery.  Seventeen of 18 masses were confirmed to be persistent melanoma on surgical pathology while 1 was a desmoid tumor.  At median follow up of 10-months, 1 patient has died, 8 are alive with known disease, and 8 continue to have no further evidence of disease since the time of surgery. 

Conclusion: In this small group of patients receiving ICI therapy for melanoma, surgery was well tolerated. Surgery may benefit select patients with mixed responses to ICI therapy. Indications for surgery in this population warrant further exploration.  

 

11.14 RAS Mutation Confers Prognostic Significance in Patients Undergoing CRS-HIPEC for Colorectal Cancer

Posted on January 18, 2018

Z. Morgan1, A. Krepline1, M. Hembrook1, S. Tsai1, K. K. Christians1, H. Mogal1, T. C. Gamblin1, C. N. Clarke1  1Medical College Of Wisconsin,Division Of Surgical Oncology,Milwaukee, WI, USA

Introduction: Approximately 5% of patients with colorectal cancer (CRC) will present with peritoneal carcinomatosis (PC) with a mean overall survival (OS) of 6-months if left untreated. Cytoreductive surgery (CRS) and hyperthermic intraperitoneal chemotherapy (HIPEC) is an aggressive surgical approach to treat peritoneal carcinomatosis (PC).  The role of this procedure in CRC continues to evolve.  There remains a significant need to further characterize the natural history of CRC carcinomatosis and identify prognostic factors to facilitate better risk stratification for prospective CRS-HIPEC patients. We performed a single institution study of CRC patients undergoing CRS-HIPEC with curative intent to identify prognostic factors associated with recurrence or overall survival.

Methods: Patients with CRC evaluated for CRS-HIPEC at the Regional Therapies Program at the Medical College of Wisconsin from 2010 -2017 were identified. Patients with non-CRC pathology, progression of disease precluding surgical intervention and/or HIPEC for palliation were excluded.  Patients who had CRS only or did not receive at least 60 minutes of HIPEC were excluded. Clinicopathologic data including age, sex, PCI score, completeness of cytoreduction, lymphovascular invasion, neutrophil-lymphocyte ratio, histology, microsatellite stability, BRAF and RAS mutation status were collected and analyzed.

Results: 47 patients underwent CRS- HIPEC with curative intent. Median PCI score was 14 [IQR: 6-21]. 34 (72%) patients had complete (CC0) resection, 11(23%) had CC1 (≤0.25 cm residual tumor) resection, 2 (4%) had CC2 (0.25-2.5 cm) resection.  6 (13%) of CRC were MSI-high.  22 (47%) were RAS mutant, 4 (9%) BRAF mutant.  At median follow-up of 2 years, 23 (48%) died of disease with a median overall survival (OS) of 19 months [IQR: 10-27], 36 (77%) patients developed recurrence with a median disease free survival (DFS) of 7 months [IQR: 5-12].   No factors analyzed reached significance for OS. RAS mutation status and LVI were the only significant predictors of decreased DFS (p= 0.02 and 0.03 respectively) on univariate analysis. On multivariate analysis neither remained significant.  

Conclusion: CRS HIPEC can achieve improved survival in patient with PC from CRC, however better risk stratification is needed for patient selection.  RAS mutation status is an independent marker of poor prognosis and may provide enhanced prognostic information in these high risk patients.  Larger cohort studies are needed to validate these findings.

 

11.15 Impact of Marital Status on Presentation and Management of Early Stage Melanoma

Posted on January 18, 2018

C. E. Sharon1, A. J. Sinnamon1, M. E. Ming2, E. Y. Chu2, R. R. Kelz1, R. E. Roses1, D. L. Fraker1, G. C. Karakousis1  2Hospital Of The University Of Pennsylvania,Dermatology,Philadelphia, PA, USA 1Hospital Of The University Of Pennsylvania,Endocrine And Oncologic Surgery,Philadelphia, PA, USA

Introduction:
Early detection of melanoma is associated with improved patient outcomes. There is data to suggest that spouses or partners may facilitate detection of melanoma prior to the onset of regional and distant metastases. Less well known is the influence of marital status on the detection of early clinically localized melanoma. We sought to evaluate the impact of marital status on T stage at time of presentation for early stage disease and decision for SLN biopsy in appropriate patients.

Methods:
Patients at least 18 of age without evidence of regional or distant metastases were identified using Surveillance Epidemiology and End Results (SEER 2010-2014). The main independent variable of interest was marital status, categorized as married, never married, divorced, and widowed. Separate analysis dichotomizing patients as married or unmarried was also performed. Chi square test was used to evaluate for significant differences in distribution of T stage at presentation by marital status. Multivariable analysis using ordered logistic regression was performed to adjust for additional patient factors. 

Results:
A total of 56,718 patients were identified for study. Most patients were married (n=39,448, 70%). Others were never married (n=8,374, 15%), divorced (n=4,024, 7%), or widowed (n=4,872, 9%). Distribution of T stage at diagnosis was significantly different by marital status (p<0.001). Forty-three percent of married patients presented with T1a disease, compared to 40% of never married patients, 36% of divorced patients, and 30% of widowed patients (p<0.001). Conversely, 10% of widowed patients presented with T4b disease compared to only 4% of married patients (p<0.001). The association between marital status and higher T stage at presentation remained significant among all non-married groups after adjustment for income, age, gender, state of residence, urban versus rural, and high school education level. There was no observed effect modification between marital status and gender (p=0.17). Independent of T stage and other patient factors, never married and widowed patients were also less likely to undergo sentinel lymph node biopsy (SLNB) in lesions over 1mm in thickness, for which SLNB is routinely recommended (p<0.001).

Conclusion:
Married status is associated with earlier presentation of localized melanoma, which has important implications on prognosis and extent of surgery. Moreover, patients who are never married or widowed are less likely to undergo SLNB for lesions where it is routinely recommended. Marital status should be considered when counseling patients for melanoma procedures and when recommending frequency of screening and follow-up to optimize patient care. 
 

« Previous Page
Next Page »