51.04 Factors Associated with Complications in Medical and Surgical Management of Diverticulitis

Posted on January 31, 2019

S. Wei1, K. M. Mueck1, A. A. Radwan1, C. Wan1, C. E. Wade1, T. C. Ko1, S. G. Millas1, L. S. Kao1  1McGovern Medical School at UTHealth,Acute Care Surgery,Houston, TX, USA

Introduction:  Complications during hospitalization for diverticulitis are difficult to classify using traditional tools such as the Clavien-Dindo system, since only 10% of patients require surgical intervention during the same admission. The Adapted Clavien-Dindo in Trauma (ACDiT) grading system is advantageous over the traditional Clavien-Dindo score because it is applicable to emergent surgeries and to patients managed non-operatively.  We have shown that ACDiT is applicable to acute diverticulitis patients, and we aimed to identify factors associated with ACDiT ≥ grade 2 complications in acute diverticulitis patients managed medically or surgically. ACDiT score of 2 means the complication required pharmacologic treatment (not including antiemetics, antipyretics, analgesics, diuretics, and electrolytes) or unexpected blood transfusions, but did not require unplanned procedures or intensive care unit admission.

Methods:  We performed a retrospective cohort study of patients hospitalized for acute diverticulitis admitted to surgery between 2011 – 2016 at a safety-net hospital. Baseline demographics and hospitalization data were collected. ACDiT scores were assigned; scores range from 0 to 5b, with 0 indicating no deviation from treatment plan and 5b indicating death despite active treatment. Univariate analysis was performed. Inverse probability weighted (IPW) propensity scores were assigned for surgical management, and IPW regression analysis was used to determine factors associated with ACDiT ≥ grade 2.

Results: Of 260 patients, 177 (68%) were managed medically. There were no differences in age, sex, race, Charleston Co-morbidity Index (CCI), or intraabdominal drain placement based on management strategy. On multivariable analysis, percutaneous drainage was associated with higher odds of ACDiT ≥ grade 2 with medical and surgical management. Higher CCI increased the odds of ACDiT ≥ grade 2 with medical management, while open surgery increased the odds of ACDiT ≥ grade 2 with surgical management. On IPW propensity score analysis, Hinchey 3, percutaneous drainage, and surgical management had 11-, 9-, and 3-times higher odds of having a complication of ACDiT ≥ grade 2 (Table).

Conclusion: The ACDiT score can be used to grade complications in acute diverticulitis patients managed medically or surgically, and to identify factors contributing to worse outcomes regardless of management strategy. Factors associated with ACDiT ≥ grade 2 include Hinchey 3, percutaneous drainage, and surgical management. ACDiT should be considered as a tool that can be used to benchmark outcomes for acute diverticulitis and to compare the effectiveness of strategies addressing risk factors for complications.

51.03 Comparison of Robotic Versus Laparoscopic and Open Repair for Inguinal Hernias

Posted on January 31, 2019

D. T. Lammers1, J. P. Kuckelman1, J. Bingham1  1Madigan Army Medical Center,Department Of General Surgery,Tacoma, WA, USA

Introduction:
Inguinal hernia repair is one of the most common surgical procedures performed worldwide. The development of the robotic assisted inguinal hernia repair theoretically allows for optical and technical advantages over laparoscopic surgery with improved post-operative pain and recovery over open methods. These theories are yet to be clearly described with a paucity of data comparing robotic inguinal hernia repairs to other commonly performed methods. We characterize our experience with robotic assisted inguinal hernia repair compared to open and laparoscopic approaches.

Methods:
Retrospective review using a prospectively collected data base of all inguinal hernia repairs over 7 years at a single institution. Data was grouped to compare between robotic, laparoscopic and open cases. Comparisons were made for all perioperative data, including patient demographics, intra and post-operative outcomes. Statistical significance was set at a p value of 0.05 comparing mean using ANOVA and Chi-square analysis.

Results:
A total of 277 matched cases met inclusion criteria and were compared. There were no statistical differences in age, gender, or preoperative comorbidities between groups. BMI was significantly higher in the robotic group when compared to laparoscopic and open inguinal hernia repairs at 31 vs 26 and 27, respectively (p=0.001). Operative times were found to be significantly longer with robotic cases at an average of 146 minutes vs 75 minutes in the open group and 86 minutes with laparoscopic cases (p=<0.001). Greater than 30 day follow up was accomplished in 95% of patients. Readmission within 30 days occurred more frequently with the open group (2.4%) when compared to both laparoscopic (1.2%) and robotic (0%) groups (p=0.03). There were no differences seen between groups with regards to post-operative complications including surgical site infections, return to the operating room, length of stay, thromboembolic events, and death (Figure 1).

Conclusion:
Robotic inguinal hernia repair was preferentially performed in larger patients with significantly better or equivalent outcomes when compared to laparoscopic or open inguinal hernia repairs, although associated with longer operative times. Robotic repairs are a viable and safe option for inguinal hernias.
 

51.02 Analysis of the Pediatric Appendicitis Score as a Clinical Adjunct

Posted on January 31, 2019

J. Stevens4, N. Vaughan3, L. Burkhalter2, G. Wools2, A. Alder1,2  1University Of Texas Southwestern Medical Center,Department Of Surgery,Dallas, TX, USA 2Children’s Medical Center,Department Of Pediatric Surgery,Dallas, Tx, USA 3Baylor Scott and White Medical Center,Department Of Surgery,Dallas, TEXAS, USA 4University Of Texas Southwestern Medical Center,Medical School,Dallas, TX, USA

Introduction:  Acute appendicitis is the most common cause for urgent surgical intervention in children. Accurate and timely diagnosis of pediatric appendicitis is thought to minimize complications. The pediatric appendicitis score (PAS) was developed by Samuel in 2002 and has been refined to help guide decision-making for diagnosing appendicitis with a goal to limit unnecessary imaging or procedures and to lower hospital costs. PAS is the core of a practice guideline that was implemented at our institution in September 2012 with intent to minimize unnecessary imaging and lower negative appendectomy rates. The purpose of this study was to evaluate the integration of the PAS into our appendicitis pathway to determine appropriateness of utilization.

Methods:  This is a retrospective review of all patients at an urban, referral children’s hospital whose evaluation for appendicitis included a PAS from July 2017 to December 2017. Data analyzed included imaging rates, appendectomy rates and pathology reports.

Results: 1741 patients were evaluated with 503 undergoing appendectomy. 423(24.3%) patients had a complete PAS with the remaining missing portions of the PAS, most commonly lab results. 1501(92%) patients had an ultrasound and 339(20.8%) had a CT with 66(4%) having imaging done before the PAS was filled out. 109 patients had conclusive imaging from an outside hospital and were excluded from these results. Overall compliance with the PAS protocol was 11.3% with 96.6% of patients with a completed PAS >7 having imaging. 

Conclusion: The PAS has not become a valuable tool as part of our appendicitis pathway to reduce over-imaging of children and lower negative appendectomy rates. Compliance with the guideline (PAS >7) would have resulted in a reduction of ultrasound and CT utilization of 243(16.2%) and 78(23%), respectively. In contrast, compliance would have doubled the negative appendectomy rate from 4 to 8%. Often, imaging is ordered prior to completion of the PAS. The default approach to any patient with possible appendicitis appears to be an ultrasound first and possibly a CT if it is still inconclusive. The PAS has not proven to be an important component of an acute appendicitis practice guideline at a busy tertiary children’s facility with a high volume of patients with appendicitis. A practice guideline that reflects our current practice potentially would save time, money and prevent patients from unnecessary radiation exposure from CT scans.

 

51.01 The Feasibility of Extracorporeal Membrane Oxygenation (ECMO) in Burn and Inhalation Injury Patients

Posted on January 31, 2019

T. D. Reid1, Y. Mikhaylov-Schrank1, C. Gaber1, P. D. Strassle1, R. Maine1, S. M. Higginson1, A. G. Charles1, C. Beckman1, B. A. Cairns1, L. Raff1  1University Of North Carolina At Chapel Hill,Chapel Hill, NC, USA

Introduction:
Burn inhalation patients are at risk for Acute Respiratory Distress Syndrome (ARDS) given pulmonary damage, systemic cytokine release, and large volume fluid resuscitation. As many as 86% of mechanically ventilated burn patients suffer from ARDS. Extracorporeal Membrane Oxygenation (ECMO) is a useful adjunct in patients with severe ARDS after failure of maximal ventilatory therapy. However, few studies have looked at the utility of ECMO following burn inhalation injury. We hypothesized that the use of ECMO in burn and inhalation injury patients is both safe and effective.

Methods:
This is a retrospective review of prospectively collected ECMO program data at the University of North Carolina. Patients included in the study were All adult and pediatric patients with burns and/or inhalation injury with ARDS that underwent Veno-venous (VV) or Veno-arterial (VA) ECMO cannulation between November 2008 and October 2017. Baseline characteristic information was collected. Primary outcomes included mortality on ECMO and 30-day mortality. Secondary outcomes included critical care and ECMO related complications. Frequencies and percentages were presented for categorical data and medians and interquartile ranges were presented for continuous data.

Results:
Of the 21 patients in this study, 16 (76%) were male. Six (29%) patients had burns only, 3 (14%) had inhalation injury only, and 12 (57%) had both burns and inhalation injury. Median percent burn was 28% of total body surface area. Patients had a median age of 48 years (IQR 26-55) with a range of 2 to 72 years. Median hours on ECMO was 116 hours and 90% percent of cannulations were VV. Substance abuse was common in this population at 33%. Eight (38%) patients required hemodialysis, which was performed via the ECMO circuit, and 12 (57%) patients were placed on a lasix infusion. Tracheostomy was performed in 18 (86%) patients. One (5%) patient died while on ECMO from cardiac causes. Total 30 day mortality was 19% (n=4) and 90-day mortality was 24% (n=5). These additional deaths were sepsis-related. Eight (38%) patients had ECMO-related complication; 3 (14%) had minor bleeding, 3 (14%) had bleeding requiring transfusion of more than 2 units, 1 (5%) had a deep venous thrombosis at the cannula site, 1(5%) had a malpositioned cannula, and 1(5%) had an arrhythmia. Only two of the patients who died had a complication related to ECMO. Both patients had bleeding requiring transfusion, however both patients died of sepsis unrelated to the bleeding. 

Conclusion:
In this study, 30-day survival was 81%, and 90-day survival was 76%. While 38% of patients had complications, the majority were minor and did not lead to morbidity or mortality. These numbers are comparable to the current literature on ECMO unrelated to burns, that demonstrate a survival of approximately 60-75%. ECMO in burn and inhalation injury patients appears to be safe and effective. Larger trials are needed to examine the use of ECMO in this population.
 

50.20 The Prognostic Value of CT Angiography in Endoscopic Intervention of Acute Lower GI Bleeds

Posted on January 31, 2019

A. Zhong1, C. Divino1  1Mount Sinai School Of Medicine,General Surgery,New York, NY, USA

Introduction:
Diagnostic modalities for lower gastrointestinal bleeds (LGIB) include endoscopy, mesenteric angiography, capsule endoscopy, nuclear RBC scans, and most recently CT angiography (CTA). The advantages of CTA include a sensitivity and specificity of 98.4% and 93.3%, visualization of the entire abdomen, and expediency. There are no clear guidelines to help providers decide on which diagnostic or interventional modality is optimal for their patient with a LGIB, often leading to confusion and unnecessary invasive workup. We propose that CTA’s can safely be used as an initial diagnostic modality in guiding intervention, specifically endoscopy, in acute LGIB’s.

Methods:
A single-institution retrospective chart review was performed of a cohort of patients who had procedure codes for endoscopy, abdominal CT angiography, and an ICD code for lower GI bleed over a period of 18 years (2000-2018).

Results:
185 patients were identified into the cohort. 51 of those patients had a CTA to diagnose an acute LGIB. A total of 69 CTA’s were performed in those 51 patients. 27/69 CTA’s had a subsequent endoscopic intervention. 22/27 CTA’s were negative for intraluminal contrast extravasation on arterial or venous phase, and 5 were positive. 17/22 (77.3%) negative CTA’s had subsequent diagnostically negative endoscopic procedures. 18/22 (81.8%) negative CTA’s had subsequent endoscopic procedures that resulted in unsuccessful intervention. Out of the 69 CTA’s, only 2 resulted in an AKI in dialysis dependent ESRD patients. No patients required surgical intervention. There were no mortalities.

Conclusion:
CTA should be the universal initial diagnostic modality in a patient with an acute LGIB when they present to an inpatient setting. It is fast, safe, and effective in identifying bleeds and potential sources. CTA’s can result within hours of presentation, have a better adverse event profile than other modalities, and have the highest sensitivity and specificity of all modalities. CTA’s have been shown to predict mesenteric angiographical diagnosis and intervention. A negative CTA is predictive of a negative endoscopic intervention, signifying that not all LIGB’s require additional attempts at diagnosis or intervention. A negative CTA can predict that a patient with a LGIB can be safely observed with transfusions as necessary. The results of the CTA should be used to guide clinical decision making in order avoid unnecessary work up, waste of healthcare resources, and potential risk from additional procedures.
 

50.19 Comparative Analysis of Long-term Outcomes of Open, Laparoscopic, and Robotic Inguinal Hernia Repair

Posted on January 31, 2019

C. Timmerman1, H. Zhu1, T. Pham1, S. Kukreja1, S. Huerta1  1University Of Texas Southwestern Medical Center,Dallas, TX, USA

Introduction:  Many techniques are currently available for the repair of inguinal hernias. As inguinal hernias are one of the most common operations performed by general surgeons, any aspect associated with outcomes and cost should be analyzed. We hypothesized that open inguinal hernia repair (OHR) is associated with superior outcomes and less operative time compared to laparoscopic (LHR) and robotic (RHR) repair.

Methods:  This is a single institution retrospective review of patients undergoing open (n=1100), laparoscopic (n=128) and robotic (n=71) inguinal hernia repair at the VA North Texas Health Care system between 7/05 and 6/17. Univariate analysis was performed using Fisher’s Exact Test for categorical and Student’s T-Test for continuous variables. We excluded 61 patients with both bilateral and recurrent hernia. For the remaining 1238 patients (964 unilateral, 165 bilateral, and 109 recurrent) variables with a univariable p≤0.15 were entered in a backward selection algorithm to yield the parsimonious multivariable regression model. Multivariable logistic regression analyses (MVA) were used to assess the association between treatment and overall complication rate, adjusting for hernia type (unilateral, bilateral, and recurrent). Data are expressed as means ± SD and significance was established at a p≤0.05 (two-sided).

Results: All patients were men and slightly overweight. Compared to the OHR, the LHR patients were three years older. Complex hernias were substantially more common in the LHR and RHR compared to the OHR cohort. All patients had similar comorbidities except for a history of cardiac disease which was more common in patients with OHR compared to both LHR and RHR (all p’s <0.05). Univariate analysis showed that: OR time [65.5±26.1 vs. 78.4±27.1 vs. 117.5±61.8 (both p’s<0.001)]; inguinodynia [1.5% vs. 26.6% vs. 28.2% (both p’s < 0.001)]; and overall complications [11.2% vs. 34.4% vs. 38% (both p’s < 0.001)] were fewer for OHR compared to both LHR and RHR. Recurrence for OHR was similar to LHR, but less than RHR [1.7% vs. 3.9 vs. 5.6% (p=0.1 OHR vs. LHR; p=0.04 OHR vs. LHR)]. Adjusting for hernia type, RHR was significantly associated with a longer OR time compared with OHR (p<0.001), while LHR is significantly associated with a shorter OR time compared with OHR (p<0.001). MVA also showed that LHR or RHR surgery has a significantly higher overall complication risk compared to OHR.

Conclusion: For unilateral hernia repair, the open approach remains the gold standard operation. LHR and RHR are associated with a higher rate of overall complications. Randomized controlled trials are needed to more conclusively demonstrate the best approach to IHR.

 

50.18 Increasing Use of Thyroidectomy as Definitive Treatment for Hyperthyroidism

Posted on January 31, 2019

A. Asban1, A. Anue1, H. Chen1  1University Of Alabama at Birmingham,Department Of Surgery,Birmingham, Alabama, USA

Introduction:
Thyroidectomy is a definitive treatment for hyperthyroidism. The initial discussion to undergo thyroidectomy is usually carried out between a patient and a primary care physician or an endocrinologist. At our institution, it is unknown how often patients with hyperthyroidism are referred for thyroidectomy and what are the common reasons for referral. Therefore, the purpose of this study is to examine the trend of thyroidectomy over a 6-year period and to identify reasons for referral. 

Methods:
We identified 237 patients with hyperthyroidism underwent thyroidectomy from January 2016 to December 2016. To examine the trend of thyroidectomy over the study period, patients were divided into six groups according to the year of thyroidectomy, group 1 to group 6, from 2011 to 2016, respectively. For each group, patients’ charts were reviewed for the reasons why patient was referred for thyroidectomy (primary outcome), time from diagnosis and/or start of antithyroid drugs (ATDs) to thyroidectomy, as well as the trend and total number of thyroidectomies each surgeon did during the study period (secondary outcomes). Differences in primary and secondary outcomes between groups were examined.

Results:
The mean age was 44 ± 15 years, 73% were women, and 54% were Caucasian. Majority of patients presented with palpitation 65%, weight loss 50% and heat intolerance 44%. The median preoperative TSH was 0.008 mU/L. A significant increase in the rate of thyroidectomy over the study period was observed where 31 patients underwent thyroidectomy in 2011 compared to 61 patients in 2016. Among the total cohort, the most common reasons patients were referred for thyroidectomy were resistance or intolerance to ATDs followed by patient’s preference and presentation with obstructive symptoms with no statistically significant difference between groups. The median time from diagnosis to surgery was 8 months (0 to 204 months) and 7 months from initiation of ATDs to thyroidectomy with no significant difference between groups.  A total of 13 surgeons operated in during the study period with no noticeable change in volume over time.  

Conclusion:
 An increase of thyroidectomy rate was observed at our institution over the last 6 years. Patients mostly referred due to resistance or intolerance to antithyroid medications, patients’ preference of surgery and presentation with obstructive symptoms. ?
 

50.17 Totally Extraperitoneal Approach for Complex Abdominal Wall Reconstruction

Posted on January 31, 2019

S. J. Kumar1, Z. D. Warriner1, Y. W. Chang1, M. A. Plymale1, D. L. Davenport1, A. Wade1, R. W. Edmunds1, J. S. Roth1  1University Of Kentucky,General Surgery,Lexington, KY, USA

Introduction: We describe five years’ experience of totally extraperitoneal approach (TEP) with component separation for complex ventral hernia repair (VHR). Complex ventral and incisional hernia management with abdominal wall reconstruction (AWR) has typically involved obligatory peritoneal entry for adhesiolysis, with subsequent risk of enterotomy. We have previously demonstrated that totally extraperitoneal (TEP) approach to AWR is feasible and results in shorter operative times with similar complication rates.  Our objective is to review continued experience with TEP hernia repair at our institution, specifically addressing technique, decision-making, and outcomes.

Methods: A retrospective review of TEP cases performed over five years.  TEP involves hernia sac identification and preservation. Hernia sac is dissected circumferentially until edges of intact anterior fascia identified. Posterior component separation performed as required for fascial closure. Hernia sac is then imbricated within the preperitoneal space or posterior rectus sheath in the midline. Mesh is placed as retromuscular sublay and linea alba restored ventral to mesh.

Results: Between January 2012 and December 2016, we used this technique for 166 cases. Four cases required intraperitoneal entry to explant densely adhered mesh. 86.1% of cases had ≤ 1 prior repair and 89.2% ASA wound class 1. Median defect size 135cm2 and mostly Rives-Stoppa or transversus abdominis release performed for component separation. Median operative time was 175 minutes, blood loss 100ml, and incidence of enterotomy was 0%. Median length of stay (4 days) and time to return of bowel function (4 days) were favorable.  Overall wound complication rate was 27.1%, specifically 9% required seroma drainage and 3% (five patients) required re-operation for various wound or mesh complications. As of April 2018, 4 (four) patients returned back to our institution for SBO, all of which resolved with conservative management.

Conclusions: Totally extraperitoneal hernia repair can be performed safely, with a low risk of enterotomy and post operative small bowel obstruction in selected patients. The TEP approach allows for hernia repair with avoidance of both peritoneal entry and adhesiolysis.  Future studies are required to validate these results.

 

50.16 The Microbiome of Gastrointestinal Perforations: Does it Matter?

Posted on January 31, 2019

V. T. Daniel1, D. V. Ward2, C. I. Kiefe3, B. A. McCormick2, H. P. Santry4
 1University Of Massachusetts Medical School,Department Of Surgery,,Worcester, MA, USA 2University of Massachusetts Medical School,Center For Microbiome Research,Worcester, MA, USA 3University Of Massachusetts Medical School,Department of Quantitative Health Sciences,,Worcester, MA, USA 4The Ohio State University Wexner Medical Center,Department Of Surgery,Columbus, OHIO, USA

Introduction: Although many pro-inflammatory conditions have been shown to have decreased microbiome diversity (alpha) as well as dysbiosis, little is known about the gut microbiome of patients with gastrointestinal perforations with inflammatory etiologies. In addition, it is unclear whether gut dysbiosis plays a role in the development of poor outcomes of surgical patients. Therefore, the objective of our study was to prospectively evaluate outcomes of patients with gastrointestinal perforations who underwent surgical intervention, to characterize the microbiome of these patients, and to assess the alpha diversity of these patients who develop poor outcomes.

Methods: Patients with stomach, small intestine, and large intestine perforations who underwent surgical intervention at a single institution were included in this prospective, translational study. 16srRNA gene sequences extracted from swabs at the perforation site were analyzed and then the reads were clustered and classified to microbial genome using QIIME. Specific taxonomic abundances were assessed. Analysis of composition of microbiomes (ANCOM) was used to assess differences of alpha diversity within groups. Outcomes assessed were 30-day mortality, 30-day postoperative sepsis, and all-cause 30-day readmission rate.

Results: Interim analysis demonstrated overall 28 subjects with stomach (18%), small intestine (36%), and large intestine (46%) perforations underwent surgical intervention. The majority were males (68%) with a mean age of 66 years (SD 16 years) who were not smokers (71%) nor used steroids (86%).  The 30-day mortality rate was 14% and postoperative sepsis rate was 50%. 30-day readmission rate was 11%. Bacteroidetes dominated the gut microbiome of patients with gastrointestinal perforations. Although not significant, microbiome alpha diversity was lower for the following groups: those who died within 30 days postoperatively compared to those who did not (p=0.59), those with postoperative sepsis compared to those without (p=0.76), and those who were readmitted within 30 days compared to those who were not (p=-.67).

Conclusion: In our small sample size, we did not find any significant differences in microbiome alpha diversity among patients with gastrointestinal perforations who had poor postoperative outcomes; however these preliminary data demonstrate high mortality among those with gastrointestinal perforations and suggest possible similarities between gut microbiome of patients with gastrointestinal perforations. Further research is needed to better characterize the microbiome of a larger sample of patients with gastrointestinal perforations compared to controls, and furthermore, those with gastrointestinal perforations who develop poor outcomes.

 

50.15 Outcomes of Ventral Hernia Repair in the Obese and Morbidly Obese: a Single Institution NSQIP Review

Posted on January 31, 2019

F. Gleason1, K. Feng1, S. Baker1, P. Washburn1, C. Perkins1, J. Richman1, M. Morris1, A. Parmar1  1University Of Alabama at Birmingham,Birmingham, Alabama, USA

Introduction:

Post-operative outcomes of ventral hernia repairs (VHR) in obese and morbidly obese patients are poorly defined. To identify the association between obesity and postoperative outcomes, we reviewed our experience in this patient population. We hypothesized that postoperative morbidity and readmission would increase with increasing body mass index (BMI). 

Methods:
We identified all patients undergoing elective VHR at our institution from 2012 to 2017 who were included in the American College of Surgeons National Surgical Quality Improvement Program (NSQIP). Hernia specific characteristics were abstracted through chart review and patients with missing information were excluded. Hernia width was classified using European Hernia Society width classification:  W1 (<4cm), W2 (4-10cm), and W3 (>10cm).  Patients were stratified by BMI category based on the World Health Organization definitions of obesity: preobese (25-29.9), obese class I (30-34.9), obese class II (35-39.9), and obese class III (≥40). Surgical site occurrence (SSO) was defined as any surgical site infection or wound disruption. Descriptive statistics were calculated, and factors associated with SSO and readmissions in a bivariate analysis were included in a logistic regression model.

Results:
A total of 334 patients underwent elective VHR and had complete data on hernia characteristics.  The mean age was 58.1 (+ 13.0), 57% (n=190) were female, 80.2% (n=268) were ASA class III, and 14.7% (n=49) were active smokers.  Average BMI was 31.4 (range 17.8-55.5) and the majority of patients had a BMI<30 (47.9%), followed by 27.3% with a BMI 30-35, 13.8% BMI 35-40, and 11.1% with a BMI >40. Average hernia length was 8.6cm (0.4-45cm) and hernia widths were W1 (n=115, 34.4%), W2 (n=148, 44.3%), and W3 (n=71, 21.3%), and were similarly distributed across BMI categories. Operative approach included open (70.0%), laparoscopic (24.6%) and robotic (5.4%). Mesh was used in 62% of cases (71.4% intraperitoneal underlay, 5.5% preperitoneal underlay, 15.9% onlay, 0.3% unknown). The remaining repairs were component separation (27.0%) and primary suture repair (10.8%). Postoperative SSO occurred in 5.7% (n=19) of the cases, and the 30 day readmission rate was 7.2%. There was a trend towards increasing SSO by BMI groups and unplanned readmission, although these were not statistically significant. In a logistic regression model adjusting for hernia width, duration of operation, and patient smoking history, increasing BMI was only weakly associated with increased SSO (OR 1.081, CI 1.02-1.15) and unplanned readmission (OR 1.06, CI 1.01-1.12).

Conclusion:
We demonstrated that acceptable immediate post-surgical outcomes for elective ventral hernia repair are possible in a select population of older, obese patients with large hernia defects. While increasing BMI was associated with increased SSO and readmissions, these effects were relatively small.

50.14 The Utilitarian Gastrojejunostomy: Evaluation of Indications and Techniques on Surgical Outcomes

Posted on January 31, 2019

B. A. Potz1, C. C. Ciambella1, V. L. Garcia1, K. P. Charpentier1, W. G. Cioffi1, T. J. Miner1  1Rhode Island Hospital,Surgical Oncology,Providence, RI, USA

Introduction:  A gastrojejunostomy (GJ) is commonly created to manage a broad range of general surgical problems.  Understanding the differences in groups of patients undergoing this procedure may allow improved patient selection and perioperative management.

Methods:  10 year retrospective review of a prospective surgical database. 386 consecutive adult patients were evaluated for operative indications, gastric outlet obstruction (GOO) scores, technical details, perioperative management, surgical outcomes, NGT tube utilization.  Surgical intent to identify palliative operations was determined by previously established criteria. 

Results: 295(76%) of the 386 GJ were performed as part of a cancer operation.    301 (77%) of the GJ were associated with partial gastrectomy (PG) included with the surgical procedure.  Antecolic reconstruction was performed in 282 (73%).  Cancer operations were associated with severe (GOO=0, 46%, p<0.001) and moderate preoperative symptoms (GOO1-3, 26%, p<0.001).   GJ with concurrent PG (235/386, 61%) was more frequently performed in asymptomatic patients (205/287 (87%), p<0.001).  Palliative intent was documented in 22% (64/295) and associated with cancer operation without PG (60/60 100%, p<0.001).  Following operation, documented symptom improvement was noted in 85% (84/99) of patients who initially presented with obstructive symptoms.  There was no difference in length of stay (mean 7 days) or major complication rates (14%) between groups(specify which “groups” here).  30-day mortality was associated with operations performed with palliative intent (11/64 (17%), p<0.001).  GJ without PG was associated with postoperative NGT placement more frequently than operations with PG (81/85 (95%), p<0.001).  GJ without PG were also associate with longer duration of tube placement (mean 1 vs 5 days, p<0.001) and more frequent NGT issues requiring replacement (16/85 (19%), p=0.004).  There was no significant difference in rates of delayed gastric emptying (51/386 (13%)) between groups. 

Conclusion: There are distinct differences amongst groups of patients undergoing an operation which includes GJ.  Active symptom management is more frequently required for cancer patients.  30-day mortality is associated with palliative operations.  Regardless of indication, prolonged NGT management is associated with GJ without PG probably due to worse gastric drainage.  Appreciation of such factors not only improves patient selection and counseling, but also will allow more precise analysis of administrative data in the future.

 

50.13 Pre-Hospital Caloric Deficit in Surgical Patients

Posted on January 31, 2019

J. Sadeghi1, K. Duh1,2, R. Barerra1  1North Shore University And Long Island Jewish Medical Center,Manhasset, NY, USA, 2Donald and Barbara Zucker School of Medicine at Hofstra/Northwell,Hempstead, NY, USA

Introduction: The timing, route, source and amount of nutrition for surgical patients with substantial caloric deficits remain active areas of study. There is benefit to starting nutrition early in surgical patients with large caloric deficits, but the data is based on in-hospital nutrition deficits and does not take into account patients’ pre-hospital courses. There has been minimal research and no guidelines set forth regarding pre-hospital nutrition deficits in surgical patients. Large pre-hospital caloric deficits with inadequate or delayed nutritional supplementation may lead to poorer outcome measures including length of stay, functional status, and 30-day readmission rates.

Methods:  We performed a retrospective review of 50 surgical patients over one year admitted to two metropolitan surgical centers with a primary admitting diagnosis of small bowel obstruction, acute pancreatitis, or diverticulitis and assessed their pre-hospital and inpatient caloric deficit. Pre-hospital deficits were estimated using patient-reported days with significant nausea, emesis, and absent oral intake. Inpatient deficits were estimated using total days kept NPO. Patients were classified as either mildly malnourished (2,500-5,000 Kcal) or moderate-severely malnourish (>5,000 Kcal) and compared to patients with no pre-hospital deficit for length of stay, status on discharge as measured by ambulatory status and disposition, and 30-day readmission rate.

Results:The average lengths of stay for the no deficit group, the mild deficit group, and the moderate-severe deficit groups were 7.79, 8.14, and 14 days respectively. The rates of independent functional ambulatory status upon hospital discharge were 72.41%, 69.23%, and 62.5%. The rates of discharge home were 89.65%, 100%, and 85.71%. The 30-day readmission rates were 20.69%, 21.43%, and 42.86%.

Conclusion: Large caloric deficits lead to poorer surgical outcomes, but pre-hospital caloric deficits are not routinely studied. Based on our preliminary results, we found that >5,000 Kcal pre-hospital deficits increase hospital length of stay and 30-day readmission rates. We suggest that pre-hospital caloric deficit should be routinely considered both in determining timing of supplemental nutrition and in future study protocols examining supplemental nutrition.

 

50.11 Laparoscopic Versus Open Common Bile Duct Exploration: Trends And Outcomes in Choledocholithiasis

Posted on January 31, 2019

M. L. Warren1, T. Wyatt1, R. Dev1, B. K. Patel1, J. Luo2, Y. Zhang2,3, K. Y. Pei1  1Texas Tech University Health Sciences Center,Surgery,Lubbock, TX, USA 2Yale School of Public Health,Environmental Health Sciences,New Haven, CT, USA 3Yale School of Medicine,Section Of Surgical Outcomes And Epidemiology,New Haven, CT, USA

Introduction:

 

There is renewed interest in performing primary laparoscopic common bile duct exploration for choledocholithiasis, but endoscopic retrograde cholangiopancreatography has largely replaced common bile duct exploration while surgical volume and experience are likely low.  Despite increasing experience and familiarity with advanced laparoscopic skills, it is unknown whether US surgeons are increasingly adopting laparoscopic common bile duct exploration for common bile duct stones.

 

Methods:

 

The ACS NSQIP database was queried for patients undergoing laparoscopic (CPT code 47564) or open common bile duct exploration (CPT code 47610) for diagnosis of choledocholithiasis (identified by ICD 9 and ICD 10 codes) from 2005 to 2016.  Trends information was evaluated as percentages of total procedures performed from NSQIP participating hospitals.  Standard descriptive statistics was analyzed and multivariable logistic regression were utilized to compare outcomes of interest including complications, mortality, reoperation, and length of stay.

 

Results:

 

A total of 1073 procedures were included for analysis.  Among NSQIP participating hospitals, the majority of explorations were performed laparoscopically but the percentage of laparoscopic common bile duct exploration remains largely unchanged (Figure 1).  After adjusting for patient characteristics, laparoscopic common bile duct exploration was associated with decreased overall complications [OR 0.25 95% CI (0.15-0.40)] and length of stay [OR 0.10 95% CI (0.06-0.16)].  There were no differences in 30-day mortality [OR 0.87 95% CI (0.15-5.00)]or reoperation [OR 0.19 95% CI (0.02-2.23). 

Conclusion:

 

Most NSQIP participating hospitals perform laparoscopic common bile duct exploration but overall experience with common bile duct explorations were low in general.  Laparoscopic exploration was associated with decreased overall complication and length stay.

50.10 Laparoscopic Cholecystectomy Value Calculation – Giving "Value" A Numeric Figure

Posted on January 31, 2019

N. Shahzad1, U. F. Bhatti1, F. Mannan1, N. A. Pasha1, H. Zafar1  1Aga Khan University Medical College,Surgery,Karachi, Sindh, Pakistan

Introduction:

Porter et al. defined "value in healthcare" as health outcomes achieved per dollar spent and the concept revolves around the patient and not the provider. It is challenging to measure the value however as giving numeric value to health outcomes is novel concept. Objective of our study was to calculate and compare "value" delivered by individual surgeons for laparoscopic cholecystectomy.

Methods:

Data was collected for laparoscopic cholecystectomies performed over two years from Jan 01, 2016 till Dec 31, 2017. Only elective cholecystectomies performed for non-inflammed gall bladder were included in the analysis.  Patients who had to be converted to open from laparoscopy were excluded as that increased the cost of care.

We devised formula for value calculation. Some operational definitions are as follows,

Minor post-operative issue: Stay more than 24 hours but less than 5 days, more than 1 clinic follow up visits within 30 days of operation.

Major post-operative Issues: Stay more than 5 days, emergency room visit or re-admission within 30 days of operation due to issues related to cholecystectomy

Numerator: Value of numerator was 100 if no major or minor issues encountered, 50 if some minor issue happened and 0 if any major issue happened.

Denominator: It was ratio of cost of individual cholecystectomy to the average cost of cholecystectomy performed in the study period

Value: Numerator / Denominator

Average value along with 95% and 99% confidence limits were calculated for the duration of study along with value provided by individual surgeons. Data has been plotted in funnel chart.

Results:

A total number of 1840 cholecystectomies were performed by nine surgeons in the study period out of which 1402 met the selection criteria. 817 went home within 24 hours of operation, while 16 patients stayed more than 5 days due to various reasons. 131 patients came to follow up clinic more than once after discharge from hospital. 62 patients had either emergency room visit or were re-admitted within 30 days of admission due to complications related to cholecystectomy. Mean +/- Standard Deviation of value provided was 83.95 +/- 43.07. Fig 1 shows funnel plot with 95% and 99% confidence limits along with values provided by individual surgeons. Value provided by surgeons “F” and “D” was greater than 99% upper confidence limit while value provided by surgeon “C” was below 99% lower confidence limit.

Conclusion:

Value provided by surgeons varies remarkably from one surgeon to another. Underlying factors need to be further explored to improve value.

Limitations:

Formula for value calculation needs to be validated.

Strengths:

First study to give numeric figure to concept of value

50.09 Effectiveness of a Central Line Associated Blood Stream Infection Protocol in a Pediatric Population

Posted on January 31, 2019

J. S. Graham1, M. Mathis2, L. Wilkinson2, S. Anderson2, C. Hutto3, K. Monroe4, A. Jones4, R. Dimmitt5, D. Galloway5, C. Martin2  1University of Alabama School of Medicine,Birmingham, AL, USA 2University of Alabama at Birmingham,Surgery/Pediatric Surgery,Birmingham, Alabama, USA 3University of Alabama at Birmingham,Infectious Disease/Pediatrics,Birmingham, Alabama, USA 4University of Alabama at Birmingham,Emergency Medicine/Pediatrics,Birmingham, Alabama, USA 5University of Alabama at Birmingham,Gastroenterology;Hepatology And Nutrition Services/Pediatrics,Birmingham, Alabama, USA

Introduction: Long-term parenteral nutrition administered by a central venous catheter (CVC) is often needed for pediatric patients with intestinal failure. Central line-associated bloodstream infections (CLABSIs) are a common cause of life-threatening bacteremia and sepsis in this patient population, secondary to long-term CVC use. When a parenteral nutrition (PN)-dependent patient presents with fever and other infection-concerning-symptoms, prompt recognition and care are needed.  Expedited antimicrobial treatment has been shown to decrease mortality and morbidity in patients with sepsis.

Methods: A 36 month, IRB approved retrospective chart review was conducted on TPN-dependent patients with intestinal failure who present with a fever to Children’s of Alabama’s ED and were admitted to the hospital. Outcomes of interest were adherence to protocol, unplanned transfers, and length of stay.

Results: 44 patients were included in the study, 28 were in the first 18-month period and 26 were in the second, with 10 patients in both populations. Post-protocol implementation, mean time from ED admission to antibiotic ordered and ED admission to antibiotic administered were lower (2:17±1.34 vs. 0:46±0.46, p<0.001, and 2:46±1:42 vs. 1:19±0:49, p<0.001), mean time between antibiotic administration and admission to the floor was greater (2:37±1:02 vs. 1:56±1:25, p=0.025), number of infectious disease consultation was greater (23.5% vs. 46.7%, p=0.006), floor to ICU transfers were lower (28.6% vs. 6.5%, p=0.009), readmission within 30 days was greater (5.9% vs. 22.8%, p=0.009) and mean length of stay was similar (7.67±4.82 vs. 6.93±3.25, p=0.283).

Conclusion: Here we show the value and importance of expedited antimicrobial treatment and a multidisciplinary approach to the treatment of each patient. A prospective analysis of the patients being readmitted within 30 days is recommended to determine the source of increased infection incidences and readmission rates.?

 

50.08 Greater rates of postoperative abscess in open appendectomy for complicated appendicitis

Posted on January 31, 2019

M. I. Orloff1, J. Lu1, N. Matolo1, S. Kolakowski1, D. Vyas1, A. Dayama1  1San Joaquin General Hospital,Surgery,French Camp, CA, USA

Introduction:  Laparoscopic appendectomy is the standard of care for non-perforated appendicitis, however its role in complicated appendicitis remains unclear. In this study we compared perioperative outcomes of open appendectomy (OA), laparoscopic appendectomy (LA), and laparoscopic converted to open appendectomy (LCOA) in the U.S. Adult population.

Methods:  We reviewed the ACS-NSQIP targeted appendectomy data sets from 2016 to identify patients with complicated appendicitis, who underwent an appendectomy. Complicated appendicitis was defined as perforated appendicitis with and without abscess. The primary outcomes of our study were intraabdominal abscess, surgical site infection, length of stay, reoperation and 30-day mortality. Multivariate logistic regression was performed to determine the association of surgical approach and intraabdominal abscess formation.

Results: A total of 2826 patients met our inclusion criteria – 2505 underwent LA, 185 underwent LCOA and 149 underwent OA. The rate of surgical site infection was lower in the LA cohort (1.2%) compared to the LCOA (9.4%) and OA (8.8%) cohorts, p < 0.01. The rate of postoperative intraabdominal abscess was 9.1% in the LA, 10.5% in LCOA and 18.2% in OA cohort, p < 0.01. Length of stay was lowest in patients who underwent LA 3.2 days (d), compared to 5.8 d in LCOA and 6.7 d OA cohorts, p < 0.01. The rate of reoperation was 2.2% in the LA cohort compared to 5.5% in the LCOA and 6.8% in OA cohorts, p < 0.01. No difference was seen in mortality between the three cohorts. The multivariate analysis revealed a statistically significant association between intraabdominal abscess formation and the OA cohort (OR 1.98, CI 1.25 – 3.14). However, no statistical difference was observed between the LOCA and LA cohorts with intraabdominal abscess formation (OR 0.97, CI 0.58 – 1.63).

Conclusion: Analysis of a contemporary national dataset showed laparoscopic appendectomy is associated with shorter LOS, less wound complications, and lower rates of postoperative abscess formation when compared to open appendectomy.  This goes against historical comparisons of the two surgical techniques.  The heterogeneity of studies demonstrates the need for randomized controlled trials to better elucidate the optimal management of complicated appendicitis. 

 

50.07 Defining Clinically Relevant Opioid Sparing Effects of Ketamine in the Peri-Operative Period

Posted on January 31, 2019

C. M. Trevino1, K. Gibbons1, C. Mitchell1, W. Peppard1  1Medical College Of Wisconsin,Surgery,Milwaukee, WI, USA

Introduction: Opioids are often prescribed for acute treatment of severe pain in the post-operative setting. Opioids have the propensity to cause addiction and many patients who utilize them become dependent, even patients undergoing surgery. It has been shown that 10-20% of surgical patients who are opioid naïve pre-operatively became dependent on opioids after surgery. Multimodal analgesia regimens have been implemented to limit opioid consumption perioperatively. Ketamine has been studied as an opioid-sparing agent, reducing acute adverse effects, and long-term dependence. The primary objective of this study was to determine the clinical correlation between the opioid sparing effects and incidence of nausea and vomiting in perioperative ketamine use at 24 hours post-op.

Methods: A systematic review was conducted utilizing studies that evaluated perioperative ketamine use compared to an opioid therapy control group. Included studies must have reported total opioid use and incidence of nausea and vomiting at 24 hours post-op in both control and ketamine groups. Studies that were excluded failed to report this data or did not report the data at time interval of interest.

Results: Overall, 24 studies where eligible and completed between 1993-2013, with 1456 patients available for analysis. A total of 753 patients received ketamine perioperatively for multimodal pain management (ketamine group) and 673 patients received traditional opioid pain regimens (control group). At 24 hours, the ketamine group experienced 35% relative reduction of total opioid use compared to the control group. Patients in the ketamine group also experienced less nausea and vomiting compared to the control group (19% vs 30%, p<0.001), leading to a 37% relative risk reduction of nausea and vomiting when ketamine was utilized perioperatively. Despite these findings, there was no significant correlation (correlation coefficient r = -0.31) between post-operative nausea and vomiting with the reduction of opioid use at 24-hours.

Conclusion: The addition of ketamine to perioperative, multimodal analgesia regimens leads to significant reductions in total opioid use and nausea and vomiting at 24 hours postoperatively. While the reduction in opioid use was not significantly correlated with the reduction in nausea and vomiting, these findings remain clinically significant for surgical patients.

 

50.06 The Association of Body Mass Index with Postoperative Outcomes after Elective Hernia Repairs

Posted on January 31, 2019

J. R. Giacolone1, S. Torres-Landa2, J. Cohen3,4, G. Hoeltzel2, R. Swendiman2, D. Dempsey2, N. Williams2, K. Dumon2  1Perelman School of Medicine at the University of Pennsylvania,Philadelphia, PA, USA 2Hospital Of The University Of Pennsylvania,Department Of Surgery,Philadelphia, PA, USA 3Hospital Of The University Of Pennsylvania,Department Of Medicine, Renal-Electrolyte And Hypertension Division,Philadelphia, PA, USA 4University Of Pennsylvania,Perelman School Of Medicine, Center For Clinical Epidemiology And Biostatistics,Philadelphia, PA, USA

Introduction:  Patients operated for a PEH are often either underweight or obese; however, the association between all BMI classes and 30-day outcomes in patients who underwent laparoscopic hernia repairs has not been studied. The aim of this study was to evaluate the association between body mass index (BMI) and post-operative outcomes in elective hernia repairs. 

Methods:  We performed a retrospective study of patients who underwent an elective laparoscopic inguinal, femoral, umbilical, ventral, para-esophageal, epigastric and Spigelian hernia repair in the ACS NSQIP database (2005-2015). Patients were divided into BMI groups (< 18.5, 18.5 – 24.9, 25.0 – 29.9,30.0 – 34.9, 35 – 39.9,and ≥ 40.0 kg/m2). A multivariable logistic regression model was developed to characterize the association between BMI class and outcomes (mortality, readmission, reoperation, and overall complications).

Results: The median (IQR) age of the 9,641 patients who met inclusion criteria was 57 (46-67) and 36.8% were women. Across each BMI group, there were significant differences in age, race, gender, smoking status, frailty index, and ASA class (p < 0.05). Underweight BMI (<18.5) was associated with increased risk of readmission (OR = 1.61, p < 0.05). Patients with a BMI 25.0-29.9 (OR = 0.86), 30.0-34.9(OR = 0.86), 35-39.9 (OR = 0.81), and ≥ 40 (OR = 0.74) (p < 0.05) were associated with decreased readmission rates.  

Conclusion: Underweight patients had an increased risk for readmissions but not for mortality after elective laparoscopic hernia repairs. Higher BMI was associated with a diminished risk for readmission, but not for mortality or reoperations. Overweight patients had a decreased risk for overall complications.  
 

50.05 Anterior Component Separation and Phasix Mesh Placement with or without Panniculectomy: 175 Patients

Posted on January 31, 2019

M. P. Lundgren1, C. Kustera1, D. G. McKeown1, E. G. Rosato1, F. Palazzo1, K. A. Chojnacki1, M. Jenkins1, P. J. Greaney1  1Thomas Jefferson University Hospital,General Surgery,Philadelphia, PA, USA

Introduction:  Panniculectomy at the time of abdominal wall reconstructionis controversial, with current reviews reporting higher surgical site occurrences (SSO) when included. At our institution we prefer anterior component separation (ACS) and Phasix mesh onlay for AWR, and panniculectomy for patients with pannus. Herein, we compare perioperative, SSO and recurrence rates at our institution after our preferred AWR method with and without panniculectomy.

Methods:  Data was gathered retrospectively. Statistical analysis was performed using Fisher’s exact test to determine significance of difference with or without panniculectomy for SSO and perioperative outcomes, and readmission. Student’s t-test was used to evaluate  differences between operative time (OT) and length of stay (LOS).

Results: 175 patients who underwent AWR between September 2014-June 2017 were included. 59 patients underwent AWR with panniculectomy. The mean OT for the panniculectomy group was 4.1 hours versus 3.4 without (p=0.07). The mean LOS for the panniculectomy group was 4.4 days versus 3.4 without (p= 0.02). There were no significant differences in SSO or readmission (See table below). The recurrence rate in the panniculectomy group was 12% (7/59), with a mean time to recurrence of 1.4 years. This was not significantly different from the 17% recurrence rate without panniculectomy (p=0.38), with a mean time to recurrence of 1.5 years.   

Conclusion: Patients who undergo AWR with ACS and Phasix onlay mesh placement, the choice to perform panniculectomy should be based on discussion between the surgeon and patient, as there are no significant differences between SSO or recurrence rates. 

 

50.04 Surgeon-Dependent Factors Influence Rate of Ventral Incisional Hernia

Posted on January 31, 2019

N. S. Patel1, L. A. Israelsson2, J. S. Thompson1, D. J. Zhou1, S. Aravind1, M. A. Carlson1  1University Of Nebraska College Of Medicine,General Surgery,Omaha, NE, USA 2Sundsvall Sjukhus,Surgery,Sundsvalls, Sweden

Introduction:   

Prevailing opinion has de-emphasized the surgeon as a risk factor for ventral-incisional hernia (VIH); however,

recent controlled data has suggested that VIH risk is surgeon-dependent. Our objective was to determine the

relationship of surgeon-dependent factors with VIH incidence in published data.

Methods:

A systematic review was performed on comparative studies (1960–2015) which determined the effect of incision

choice (vertical midline=VMI; transverse=TI, lateral paramedian=LPI), suture use (nonabsorbable=NA, rapidly

absorbable=RA, slowly absorbable=SA), closure technique (mass vs. layered), or stitch length (VMI only; short

vs. long) on the incidence of primary VIH after uncomplicated laparotomy in adults. Exclusion criteria were

follow-up <12 months, <40 incisions/treatment group, and/or presenceof AAA, immunosuppression, mini-

laparotomy, and/or mesh implantation.

Results:

In 42 comparative studies (median incisions/treatment group=115; range=45–1,111) including 32 controlled trials,

1,383 primary VIHs developed from 15,305 incisions (raw VIH rate=9.0%;median rate=7.7%; range=0–29.2%).

Selection of TI or LPI instead of VMI in the incision category or use of short as opposed to long stitch length (for

VMI) produced an ~80% or ~60% decrease in median VIH incidence, respectively (Fig.1).

Conclusion:

Large differences in VIH rate in studies comparing incisional choice or stitch length suggested that these two

surgeon-dependent factors influence VIH incidence. To minimize VIH formation, published data support

selection of TI or LPI for incision, or short stitch length if VMI is selected.

« Previous Page
Next Page »