52.08 Practical Adherence to the Step-Up Approach for Pancreatic Necrosis: An Institutional Review

Posted on January 26, 2017

V. Tam1, C. Umapathy4, M. Zenati3, S. Downs-Canner3, B. A. Boone1, J. Steve1, A. Zureikat1, K. K. Lee1, H. Zeh1, D. Yadav2, M. E. Hogg1  1University Of Pittsburgh,Surgical Oncology,Pittsburgh, PA, USA 2University Of Pittsburgh,Gastroenterology, Hepatology And Nutrition,Pittsburgh, PA, USA 3University Of Pittsburgh,Department Of Surgery,Pittsburgh, PA, USA 4University Of Pittsburgh,General Internal Medicine,Pittsburgh, PA, USA

Introduction:
Infected pancreatic necrosis is a highly morbid disease that was traditionally approached with an open necrosectomy. This approach was associated with rates of morbidity and mortality up to 95% and 39%, respectively. The multicenter randomized controlled PANTER trial published in 2010 in the New England Journal of Medicine proposed a “Step-Up” approach which demonstrated fewer major complications than conventional treatment, with comparable rates of mortality, and spared a major operation in one-third of patients. We sought to evaluate the practical adherence to the Step-Up approach at a single tertiary care institution, its temporal adoption into clinical practice, and impact on outcomes.

Methods:
This is a retrospective review of all patients treated at a tertiary care center with infected pancreatic necrosis between 2006 and 2014. Diagnosis was based on positive culture on pancreatic fine needle aspiration, or presence of an air filled necroma on computed tomography. “Modified Step-Up” (MSU) was defined as percutaneous or endoscopic drainage followed by additional percutaneous or endoscopic drainage, followed by any surgical intervention, including video-assisted retroperitoneal debridement and open necrosectomy. Patients were stratified into the “early” pre-PANTER (2006-2010) or ”late” post-PANTER (2010-2014) period. Rates of adherence to the MSU approach were compared as well as clinical outcomes. 

Results:
There were 130 patients with infected necrotizing pancreatitis in the overall cohort; 75(58%) and 55(42%) were treated in the early and late period. At baseline, patients admitted in the late period were more likely to have higher ASA scores (3-5 vs 1-2, 92% vs 39%, p<0.001). In the late period, adherence to MSU was 46%(n=25) vs. 27%(n=27) in the early period (p<0.05). Late period patients had a greater likelihood of percutaneous drainage (65% vs. 43%, p=0.012) and greater number of total median interventions (3 vs. 2, p<0.001), however had comparable rates of surgery (73% vs. 79%, p=0.432), including 34(85%) open necrosectomies in the late period vs. 55(93%) in the early period. There were no differences in length of hospital stay, rates of in-hospital mortality, long-term complications, or survival at 2-years following discharge. Patients in the late period were less likely to have a pancreatitis-related readmission (47% vs. 71%, p=0.007) or multiple readmissions (31% vs. 51%, p=0.024). 

Conclusion:
Overall, adherence to the MSU approach was 46% between 2010 and 2014. Patients treated during this period had lower rates of pancreatitis-related readmission and total readmissions, with similar rates of long-term complications and mortality compared to patients between 2006-2010. This study demonstrates that adoption of clinical guidelines can result in improved clinical outcomes. Barriers to implementation of the Step-Up approach should be identified to improve adherence rates. 

52.07 The role of simultaneous cystgastrostomy and necrosectomy for walled off pancreatic necrosis

Posted on January 26, 2017

M. R. Driedger1, F. R. Sutherland1, E. Dixon1, S. Gregg1, N. Zyromski2, C. G. Ball1  1University Of Calgary,General Surgery,Calgary, AB, Canada 2Indiana University School Of Medicine,Indianapolis, IN, USA

Introduction:  

Severe acute pancreatitis (SAP) occurs in 15% of patients with generalized pancreatitis. Walled off pancreatic necrosis (WOPN) is the most common end result of SAP.  When symptomatic, WOPN requires intervention. The aim of this study was to evaluate the role of simultaneous cystgastostomy and necrosectomy (CG/N) for WOPN.

Methods:

A retrospective review of patients with WOPN undergoing surgical management on a high volume pancreatic service over 11 years (2005-2016) was performed.  Outcomes included mortality, morbidity, intervention timing and symptom resolution. Statistics were descriptive.

Results:

Seventy-seven patients were analyzed (mean WOPN diameter=14.5 cm, 67.5% male, mean age=47 years). The majority were acutely ill, with an average preoperative length of stay of 28.7 days and 26% requiring preoperative support in the Intensive Care Unit (ICU). Preoperative complications were prevalent (42.9%), and included mesenteric vein thrombosis (37%), gastric outlet obstruction (19.2%), respiratory complications (19.2%), bacteremia (13.7%) and acute kidney injury (9.6%). Nearly all (93.5%) patients underwent an open trans-gastric CG/N while 6.5% received a cystjejunostomy. The median duration of time between the onset of SAP and operative intervention was 45.1 days. Forty-two percent of the cultured necrosum was infected with bacteria. Postoperative morbidity included infection (10.4%), bleeding (5.2%), fistula (5.2%) and re-operation (3.9%). Postoperative median hospital length of stay was 10 days (range 4-228) with 96.1% of patients discharged home. Mortality was 2.6% with 10.4% requiring postoperative ICU care. The mean length of follow up was 12 months with 87.7% of patients having complete clinical resolution of symptoms at an average of 7.3 weeks. Recurrent WOPN occurred in only 5.5% of patients at an average of 19 months after the index operation.

Conclusion:

Despite acutely ill and comorbid patients with large WOPN volumes, simultaneous CG/N offers a definitive single-stage solution in the vast majority of patients with minimal postoperative morbidity and rapid return to an asymptomatic state. Upon consideration of the minimal laparotomy required, this procedure represents the preferred approach for WOPN.

 

52.06 Prophylactic Antibiotic Use in Outpatient Anorectal Surgery for HIV Patients

Posted on January 26, 2017

S. A. Patel1, R. J. Kucejko1, J. L. Poggio1  1Drexel University College Of Medicine,Philadelphia, Pa, USA

Introduction:  Human immunodeficiency virus (HIV) has been a vexing challenge to healthcare providers since its discovery. As medical therapy increases survival among people with an HIV infection, surgeons have been seeing an increasing number of HIV-infected patients with anorectal pathologies, including anal dysplasia and cancer. There is, however, no data on the incidence and risk factors for postoperative complications, such as surgical site infections (SSIs), in HIV patients undergoing elective anorectal procedures, nor is there data to suggest the administration of prophylactic antibiotics. Regular antibiotics are costly and have side effects that urge for accurate targeting of use; the purpose of this study was to review a HIV-positive patient population that has undergone anorectal surgery for dysplasia and assess the need for antibiotic prophylaxis. 

Methods:  A retrospective chart review was performed of all HIV positive patients seen as an outpatient in the Colorectal Surgery Division from 2007 to 2014. Basic demographics and clinical data such as date(s) of surgery, follow-up visit(s), and antibiotic prophylaxis, as well as preoperative CD4 count and HIV viral load, were recorded for 229 patients and 362 procedures. Postoperative exam notes were reviewed to determine the presence of an SSI. To analyze the data, patients were stratified according to CD4 count and viral load. The proportion of patients who received prophylactic antibiotics was assessed and the SSI rate was calculated. A chi-squared analysis was performed to assess whether SSI risk was elevated in this population.

Results: SSIs occurred in two of 193 (1.04%) cases for which antibiotic prophylaxis was not administered and in none of the 36 cases with antibiotic prophylaxis and available post-operative reports. One SSI occurred in a 51-year-old male with a preoperative CD4 count of 612/μL and viral load of zero. The other SSI occurred in 57-year-old female with an unknown CD4 count and viral load. A chi squared analysis showed the incidence of SSIs in the groups with and without antibiotic prophylaxis was not significantly different (p=0.540). 

Conclusion: Our study found an SSI incidence of 1.04% in HIV-infected patients with CD4 counts above 50/μL, which does not suggest an elevated risk compared to the general population reported incidence of 3-11% for clean-contaminated wounds. Antibiotics are costly and have many side effects; based on our study, prophylactic antibiotics are not indicated for outpatient anorectal procedures in HIV patients with CD4 counts above 50/μL. 
 

52.05 Age is not associated with Readmission in Patients Undergoing Pancreatectomy or Colectomy Procedures

Posted on January 26, 2017

M. N. Mustian1, L. E. Goss1, D. Chu1, L. Theiss1, J. Christein1, C. Balentine1, M. S. Morris1  1University Of Alabama At Birmingham,Surgery,Birmingham, AL, USA

Introduction: More than one third of operations in the United States are performed on individuals aged 65 years or older. These patients are at high risk for unplanned readmission, prolonged hospitalization, and postoperative complications. This study analyzes the relationship between aging and perioperative outcomes for patients undergoing pancreatectomy and colectomy at a major academic medical center. We hypothesize that elderly patients would have higher rates of readmission and worse postoperative outcomes.

Methods: We used National Surgical Quality Improvement Project data from our institution to evaluate 1964 patients having colectomy or pancreatectomy between 2010 and 2016. We stratified by age: < 50, 50-59, 60-69, 70-79, and ≥80 years old. The primary endpoint was 30 day unplanned postoperative readmission.  Secondary endpoints included length of stay, mortality, postoperative readmission rates and NSQIP complications. Univariate and multivariate analysis were performed. We used stepwise backwards logistic regression analyses to identify risk factors for readmission.

Results: In our cohort of 1964 patients, 692 (35%) patients had pancreatectomies and 1272 (65%) colectomies.  The age distribution included:  23% age <50, 26% age 50-59, 28% age 60-69, 17% age 70-79 and 6% age 80 and older.  Median postoperative length of stay was 6 days and was similar across all age groups.  Median readmission rate was 13.6% and similar across age groups.  Major complications were observed in 216 patients and resulted in different rates of readmission based on age (<65 years 40% vs ≥65 years 27% p=0.03).  The risk of mortality increased steadily with age from <1% in the youngest group to 4.2% in the oldest group.  On multivariable analysis, there was no significant relationship between age and risk of readmission, but both minor (OR 1.76 CI 1.15-2.68) and major (OR 4.6 CI 3.2-6.5) complications were associated with increased odds of readmission.

Conclusion: Our data did not reveal an association between age and risk of readmission.  Postoperative complications are associated with readmission.  However, younger patients experiencing major complications were more likely to be readmitted when compared to older patients.  Reasons for readmission in elderly patients may differ from those of younger patients.  Further work is needed to understand the relationship between age and readmission to design programs to address the unique discharge needs of elderly patients.

52.04 Laparoscopic Partial Colectomy Reduces Length of Stay and Mortality in Patients with Ascites

Posted on January 26, 2017

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

Introduction:
Ascites increases perioperative complications and risk of death, but is not an absolute contra-indication for subtotal colectomy. It remains unclear whether this risk can be minimized using a laparoscopic versus open approach. We hypothesize that laparoscopic surgery reduces risk of complications and death in patients with ascites.

Methods:
Data was retrospectively analyzed from 2,419 patients with ascites who underwent laparoscopic or open colectomy from 2005 to 2013 using the American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) database. Post-operative outcomes were compared using two-sample tests of proportions, or using two-sample T-tests after checking for variance equality with Welch’s approximation for degrees of freedom. Odds ratios (OR) and incidence rate ratios (IRR) for complications and death were calculated using univariable logistic regression or linear regression respectively. P values < 0.05 two-tailed were considered statistically significant. 

Results:
218 patients (9%) with ascites underwent laparoscopic colectomy. Laparoscopic surgery was associated with lower 30-day mortality (15.6% versus 24.3%, OR 0.58, 95% CI 0.39 – 0.84, P = 0.004) and shorter hospital length of stay (9 days versus 15 days, IRR 0.0046, 95% CI 6E-4 – 0.034, P < 0.001). There was no significant difference in operative time (145 minutes versus 146 minutes, P = 0.69) or superficial surgical site infections (7.3% versus 7.7%) between laparoscopic versus open surgery.

Conclusion:
Laparoscopic colectomy reduces 30-day mortality and hospital length of stay in patients with ascites. This surgical modality may help mitigate the additional risk associated with ascites in patients undergoing subtotal colectomy.
 

52.03 Trends in Utilizing Laparoscopic Colorectal Surgery Over Time in 2005-2014 Using the NSQIP Database

Posted on January 26, 2017

C. M. Hambleton Davis1,2, B. A. Shirkey3, L. W. Moore1, H. R. Bailey1, X. L. Du2, M. V. Cusick1  1Houston Methodist Hospital,Department Of Surgery,Houston, TX, USA 2The University Of Texas School Of Public Health,Department Of Epidemiology,Houston, TX, USA 3Oxford University,Oxford Clinical Trials Research Unit / Centre For Statistics In Medicine, NDORMS,Oxford, , United Kingdom

Introduction:  Laparoscopy, originally pioneered by gynecologists, was first adopted by general surgeons in the late 1980’s. Since then, laparoscopy has been adopted in the surgical specialties and colorectal surgery for treatment of benign and malignant disease. Formal laparoscopic training became a required component of surgery residency programs as validated by the Fundamentals of Laparoscopic Surgery (FLS) curriculum; however, some surgeons may be more apprehensive of widespread adoption of minimally invasive techniques. Although an overall increase in the use of laparoscopic techniques is anticipated over a ten-year period, it is unknown if a similar increase will be seen in higher risk or more acutely ill patients.

Methods:  Using the American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) database from 2005 to 2014, colorectal procedures were identified by CPT codes and categorized to open or laparoscopic surgery. The proportion of colorectal surgeries performed laparoscopically was calculated for each year. Separate descriptive statistics were collected and categorized by age and BMI. ASA classification and emergency case status variables were added to the project to help assess complexity of cases.

Results: During the ten-year study period, the number of colorectal cases increased from 3,114 in 2005 to 51,611 in 2014 as more hospitals joined NSQIP. A total of 277,376 colorectal cases were identified, 114,359 (41.2%) of which were performed laparoscopically. The use of laparoscopy gradually increased each year, from 22.7% in 2005 to 41.2% in 2014. Laparoscopic procedures were most commonly performed in the youngest age group (18-49 years), overweight and obese patients (BMI 25-34.9), and in ASA Class 1-2 patients. Over the ten-year time period, there was a noted increase in the use of laparoscopy in every age, BMI, and ASA category, except ASA 5. The percent of emergency cases receiving laparoscopic surgery also doubled from 5.5% in 2005 to 11.5% in 2014.

Conclusion: Over a ten-year period, there was a gradual increase in the use of laparoscopy in colorectal surgery. Further, there was consistent increase of laparoscopic surgery in all age groups, including the elderly, in all BMI classes, including the obese and morbidly obese, and in most ASA classes, including ASA 3-4, as well as emergency surgeries. These trends suggest that minimally invasive colorectal surgery appears to be widely adopted and performed on more complex or higher risk patients. 

 

52.02 Incisional Hernia Risk after Prophylactic Mesh Augmentation: A Systematic Review and Meta-Analysis

Posted on January 26, 2017

S. Shakir1, Z. Borab2, M. A. Lanni1, M. G. Tecce1, J. MacDonald3, W. W. Hope4, J. M. Weissler1, M. J. Carney1, J. P. Fischer1  1University Of Pennsylvania,Plastic Surgery,Philadelphia, PA, USA 2Drexel University College Of Medicine,Philadelphia, Pa, USA 3University Of Western Ontario,London, ONTARIO, Canada 4New Hanover Regional Medical Center,Surgery,Wilmington, NC, USA

Introduction:  Incisional hernia (IH) is a morbid and challenging complication with an extraordinary financial burden to the healthcare system. With nearly 350,000 repairs and expenditures in excess of $3 billion annually, there is a clear need for reparative strategies to diminish hernia recurrence. As greater emphasis is being placed on disease prevention, and as surgical technique and mesh technology evolves, the surgical paradigm must transition from a reparative approach to preventative action. Restoration of tensile strength is paramount in reducing risk of IH, yet the optimal intervention is not known. This systematic review and meta-analysis reviews incidence of IH and complications after elective laparotomy using either PMA or conventional fascial repair.

Methods:  A systematic review of the literature was conducted in accordance with PRISMA guidelines to identify studies comparing PMA to primary suture closure (PSC) repair in elective, midline laparotomies during index operation. The primary outcome of interest was development of IH. Inclusion criteria included patients between the ages of 18-75 years and at least 1 IH risk factor (BMI ≥ 25 kg/m2, prior abdominal surgery, history of AAA, ≥45 years of age, and smoking or history of COPD). Exclusion criteria included prior hernia or existing abdominal mesh, non-elective cases, history of metastatic cancer, immunosuppression, active infection, life expectancy less than 24 months, and pre-existing pregnancy.

Results: Overall, 14 studies were included. Of the total 2114 patients, 1152 participants underwent PMA. PMA significantly decreased the risk of IH compared to PSC (RR=0.15, 95% CI 0.07-0.30). PMA reduced the risk of IH regardless of mesh location (i.e. onlay, retrorectus or preperitoneal), however, PMA use was associated with an increased rate of seroma (RR 1.95, 95% CI 1.31 to 2.91; 10 studies). PMA patients are at increased risk for chronic wound pain compared to PSC [RR=1.70, 95% CI 1.04 to 2.78).

Conclusion: The existing literature provides high level evidence demonstrating that prophylactic mesh lowers the rate of IH after elective, open intra-abdominal surgery. Risk stratification models in conjunction with an increased emphasis on preventative medicine, strategies such as PMA offer safe and efficacious risk reduction in high-risk laparotomy patients. Given the evidence gap, however, further study is undoubtedly warranted.  

 

52.01 Morbidity and mortality for patients with sinistral portal hypertension undergoing splenectomy

Posted on January 26, 2017

M. Neuwirth1, A. J. Sinnamon1, R. R. Kelz1, G. C. Karakousis1, M. K. Lee1  1Hospital Of The University Of Pennsylvania,Department Of Surgery,Philadelphia, PA, USA

Introduction: Patients with sinistral portal hypertension frequently develop gastric varices and are therefore at risk of life-threatening hemorrhage.  Splenectomy is the treatment of choice as it largely eliminates the bleeding risk, but these procedures likely carry increased risk relative to splenectomy performed for other reasons.   We sought to compare the specific morbidity and mortality of splenectomy performed for sinis-tral portal hypertension versus other indications using a national cohort.

Methods: The American College of Surgeons National Surgical Quality Improvement Program (ASC NSQIP) was accessed from the years 2005-2014 for patients undergoing open or laparoscopic splenectomy.  Patients with an operative diagnosis of gastric vari-ces were selected for inclusion in the case cohort.  These patients were compared to two separate control groups: hypersplenism or splenomegaly, and all other indi-cations (OI) for splenectomy with 25 or more cases (excluding trauma).  Pearson’s chi-squared or Wilcoxon rank-sum tests were applied to categorical variable com-parisons as indicated.

Results: Of the 7,522 splenectomies performed during the study period, we identified 28 performed for gastric varices, 1,186 performed for hypersplenism or splenomeg-aly, and 4,128 in the OI group.  Notably, patients in the gastric varices group had lower rates of thrombocytopenia and were similar to the comparison groups with respect to their total bilirubin and INR.  Compared to OI, patients with gastric vari-ces more often had dependent functional status (21.4 vs 5.9%, p < .001), ASA class > III (92.8 vs. 62.5%, p < .001), and albumin < 3 (57.1% vs 11.4%, p < 0.001).  There was no difference in mortality for the varices group as compared with the splenomegaly group (3.6% vs 2.4%; p = 0.678) or OI (3.6% vs 2.4%; p = 0.671).  As depicted in figure 1, serious morbidity was significantly increased in the varices group versus the control group (53.6% vs 19.7%; p < 0.001) and versus the sple-nomegaly group (53.6% vs 23.2%; p < 0.001).  Most major post-operative compli-cations in the varices group were due to respiratory (39.2%), bleeding (42.9%) and sepsis (25.0%) related events.

Conclusion:Patients that undergo splenectomy for sinistral portal hypertension are at substan-tially increased risk for morbidity but not mortality as compared with splenectomy for splenomegaly or other indications.  Most significant are risks of respiratory, renal, bleeding or sepsis-related events.  Careful patient selection must be employed in this cohort. 

 

34.10 Increasing Number and Age of Blood Increases Mortality in Massively Transfused Trauma Patients

Posted on January 25, 2017

R. Uhlich1, R. Patel1, J. Pittet1, P. Bosarge1, M. Marques1, H. Wang1, J. Kerby1  1UAB,Acute Care Surgery,Birmingham, AL, USA

Introduction:
Numerous studies have examined the association between blood age and mortality among trauma patients. This study was designed to account for the time-varying nature of exposure to stored blood and its effect on mortality.

Methods:
Patients receiving at least one unit of red blood cells between 2011 and 2014 were included. Blood bank data was queried for the age of blood and time of transfusion. Demographics, injury, and clinical characteristics were obtained from the trauma registry. The time at which a patient received at least one unit of blood stored > 7 days, > 14 days, or > 21 days was calculated. A Cox proportional hazards model adjusted for age, Injury Severity Score, and injury mechanism estimated hazard ratios (HRs) for the association between death in 24 hours and receiving at least one unit of stored blood. A separate model was created for each blood age category, and exposure to at least one unit of stored blood was entered as a time-varying covariate. Models were stratified by whether the patient received > 10 units of blood in 24 hours.

Results:
1,654 patients received at least one unit of PRBC (187 massively transfused). Increased hazard of 24-hour mortality was observed for those with exposure to at least one unit of blood > 7 days (HR 1.83, 95% CI 1.31-2.56), > 14 days (HR 1.49, 95% CI 1.05-2.10), or > 21 days (HR 1.75, 95% CI 1.17-2.62). The hazard increased when patients received at least 4 units of blood > 14 days (HR 3.54, 95% CI 2.20-5.71) or 21 days (HR 5.21, 95% CI 2.53-10.76). Patterns of association were observed among massively transfused patients.

Conclusion:
Exposure to increasing number and age of stored blood increases the hazard of 24-hour mortality for patients requiring large blood volume replacement. Clinical trials evaluating the effect of blood storage age on clinical outcomes in trauma are warranted.
 

34.09 Risk of Death in Elderly Blunt Trauma Patients: Complications Count

Posted on January 25, 2017

V. Gahlawat1, J. A. Vosswinkel1, A. J. Singer1, M. J. Shapiro1, J. E. McCormack1, E. Huang1, R. S. Jawa1  1Stony Brook University Medical Center,Stony Brook, NY, USA

Introduction:  A variety of factors are thought to influence outcomes following geriatric trauma. However, there is little detailed information regarding the effects of complications on outcomes following admissions for blunt trauma in the elderly.

Methods:  We performed a retrospective review of the trauma registry at a suburban regional trauma center from 2010 to 2015 for all elderly (>65 y) blunt trauma admissions with hospital length of stay (LOS) ≥ 3 days. Deaths in Emergency Department, burns and cardiac arrest were excluded. Patients were divided into three groups- No Complications [NC], minor complications only [MC], and major complications with or without minor complications [SC]. Major and Minor complications were defined as per TQIP. Univariate and multivariate analyses were performed.

Results: There were 2,469 admissions meeting inclusion criteria, NC (n=1,984), MC (n=211) and SC (n=274).  There was no significant difference amongst groups in terms of age (median years; NC- 81, MC- 82, SC- 81, p=0.60) and frequency of low fall as the mechanism of injury (NC- 74.7%, MC- 73.5%, SC- 72.3%, p=0.64). Pre-admission Do Not Resuscitate directive (DNR, NC- 6.6%, MC- 10.4%, SC- 9.9%, p=0.025) and Injury severity score (ISS, median [IQR]; NC- 9[5.5-13] , MC- 10[9-16], SC- 12[9-17], p<0.001) were lower in no complications group. With regards to outcomes, patients who suffered major complications after blunt trauma had increased intensive care unit (ICU) LOS, mechanical ventilation, hospital LOS, in-hospital mortality and fewer discharge to home (Table 1). In-hospital mortality increased with increasing number of major complications (0-2.6%, 1-12%, 2-29%, 3-20.0%, 4-50%). On multivariate stepwise forward logistic regression analysis, factors significantly associated with in-hospital mortality (Odds Ratios with 95%CI) were mechanical ventilation (9.56 [5.08-18.00]), pre-admission DNR (5.01 [2.64-9.49]), ICU stay (2.12 [1.06-4.23]), low fall mechanism of injury (2.07 [1.12-3.83]), major complications (1.56 [1.17-2.06]), ISS (1.08 [1.05-1.12]), and age (1.04 [1.01-1.07]).

Conclusion: Low fall was the most frequent mechanism of injury. There was a near linear relationship between the number of major complications and in-hospital mortality in elderly patients admitted with blunt trauma. The multivariate model had excellent discriminative characteristics for mortality, with an AUCROC of 0.911. The probability of death during hospitalization increased by a factor of 1.56 with each major complication (i.e. OR 6.24 with 4 complications). Once even a single major complication has occurred, great vigilance is warranted to prevent in-hospital mortality. Further study in a larger cohort is warranted.

34.08 Trauma-Induced Coagulopathy is associated with an Early, Discoordinated Inflammatory Response

Posted on January 25, 2017

R. A. Namas1, X. Zhu1, D. Liu1, O. Abdul-Malak1, J. Sperry1, Y. Vodovotz1, T. Billiar1  1University Of Pittsburgh,General Surgery,Pittsburgh, PA, USA

Introduction: Trauma-induced coagulopathy (TIC) is often associated with a broad systemic inflammatory response that can predispose patients to follow a complicated clinical course. Despite significant recent advancements associating post-traumatic inflammation with TIC, a better understanding of this complex interaction is needed. To characterize the systemic inflammatory response accompanying TIC, we analyzed an extensive time course of circulating inflammatory mediators coupled with data-driven modeling.

Methods: From a cohort of 472 blunt trauma survivors, 114 patients had TIC (defined by admission INR ≥ 1.3). After excluding patients with history of anticoagulant intake pre-trauma, 98 TIC patients (71 males [M] and 27 females [F], age: 39.7 ± 2, injury severity score [ISS]: 23.7 ± 1) were matched to 98 non-TIC patients (71/27 M/F, age: 39.8 ± 2, ISS: 23.4 ± 1) for age, gender ratio, and ISS using IBM SPSS®. Three samples within the first 24 h were obtained from all patients and then daily up to day 7 post-injury and assayed for twenty four inflammation biomarkers using Luminex™. Two-way analysis of variance was used to determine statistical significance (p<0.05) between the TIC and non-TIC sub-groups. Dynamic network analysis (DyNA) was used to infer dynamic connectivity and complexity among the inflammatory mediators.

Results: ICU length of stay (LOS), total LOS, and days on ventilation were statistically significantly prolonged in the TIC group when compared to non-TIC group. In addition, the TIC group had a greater requirement of operative intervention within the first 24 h post-admission. The TIC group had a higher degree of organ dysfunction from days 1 to 7 when compared to the non-TIC group. Importantly, circulating levels of IL-6, IL-10, MCP-1, MIG, IP-10, and IL-8 and were significantly elevated in the TIC group. DyNA suggested that the inflammatory response in the non-TIC group had a higher coordinated degree of interconnectivity while the response in TIC consisted of multiple sparse nodes with reduced interconnectivity within the initial 4 h post-injury.

Conclusion: These results suggest that post-traumatic coagulopathy, identified by elevated admission INR, is associated with a markedly differential inflammatory response when compared to patients that present without TIC despite similar injury patterns. Reduced dynamic network connectivity in the TIC patients suggests a discoordinated inflammatory response that might promote immune dysfunction and hence worse outcome.

 

34.07 Palliative Care Assessment for Injured Patients: Perspectives from Trauma and Palliative Care Teams

Posted on January 25, 2017

M. J. Keating1, N. D. Patel1, M. Nishtala1, C. W. Towe1, C. S. Koniaris2, J. George2, V. P. Ho1  1Case Western Reserve University School Of Medicine,Surgery,Cleveland, OH, USA 2Case Western Reserve University School Of Medicine,Medicine,Cleveland, OH, USA

Introduction:
Palliative care is a medical specialty focused on improving quality of life and alleviating symptoms for patients facing life-threatening illness. Barriers to the integration of Palliative Care team (PCT) services into the care of trauma patients have not been clearly defined. We sought to prospectively evaluate whether trauma team (TT) and PCT members differed regarding perceived benefit of PCT (PBP) consultation in trauma patient care.

Methods:
TT and PCT clinicians attended weekday trauma service sign out for 12 weeks. Based on verbal report, each member of the TT and PCT independently assessed whether patients might benefit from a PCT consultation.  TT included trauma surgeons (TS), advance practice providers (APPs), and residents. Sign out typically involved 4-6 TT members. PCT included physicians and APPs; sign out typically involved 1-2 PCT members. Data were prospectively collected regarding assessments, demographics, injury severity, and outcomes. Patients who received a PCT consult were excluded from subsequent assessments. PBP between the TT and PCT clinicians was compared. Secondarily, we sought to identify clinical outcomes associated with PBP. Chi-square and Student’s t-test were used to compare groups (p<0.05 considered significant).

Results:
186 patients (median age 47.6, SD 23.9) had 2013 assessments performed by TT members and 522 assessments by PCT members. Mean injury severity score was 11.9 (SD 9.9). Mean length of stay was 5 days (SD 10.9). There were 5 deaths. TT members identified 76 patients (41%) as having PBP at least once during the hospital stay, compared with 59 patients (32%) identified by PCT (p<0.001), with 78% concordance. TS identified fewer patients with PBP than PCT (22% vs 32%, p<0.001), with 49% concordance with PCT. Eight (5%) patients received a formal PCT consult.  Patients identified as having PBP by any clinician were significantly more likely to change code status to “Do Not Resuscitate,” and were more likely to be discharged to a destination other than home.

Conclusion:
TT and PCT providers identify a high proportion of trauma patients who might benefit from PCT evaluation. Despite this, consultations are rarely requested. Further exploration should be performed to determine barriers to PCT consultation.

34.06 Outcomes Of Patients with Traumatic Brain Injury in Skilled Nursing Facilities

Posted on January 25, 2017

S. N. Lueckel1,2, D. S. Heffernan1,3, T. Kheirbek1,3, M. D. Connolly1,3, S. F. Monaghan1,3, C. A. Adams1,3, W. G. Cioffi1,3, K. Thomas2  3Brown University,School Of Medicine,Providence, RI`, USA 1Rhode Island Hospital,Department Of Surgery Division Of Trauma Surgery And Critical Care,Providence, RI, USA 2Brown Universtiy,School Of Public Health,Providence, RI, USA

Introduction:
Traumatic brain injury (TBI) is a leading cause of death and disability in the United States. In 2010, 2.5 million people suffered TBI at a cost of $76.5 billion. Within the survivors, TBI remains a leading contributor to long term disability. It is estimated that 5.3 million people are living with physical, emotional, cognitive and behavioral disabilities attributable to TBI, many of whom require placement in long term skilled nursing (SNFs). Despite this very large population of TBI patients, very little is known about the long term outcomes of TBI survivors, including rates of discharge to home or risk of death in long term nursing facilities.

Methods:
This is a retrospective review of the prospectively maintained Federal Minimum Data Set (MDS) combined with the CMS Vital Status database from 2012-2013. Records were reviewed for clinical characteristics upon admission to the SNF including cognitive function (CFS), ability to communicate, and motor function. Activities of daily living were reassessed at 30 days post SNF admission to calculate Functional Improvement at 30 and 60 days and 1 year (FI). Records were also reviewed for discharge to home, readmission rates and death at 30 and 60 days and 1 year. For analysis we used robust Poisson regression to estimate relative risk (RR).

Results:
Overall, 65,099 individuals were admitted to SNFs with a TBI diagnosis in the US. The mean age was 71.6 years with 26% over the age of 85yrs. Overall, poor cognitive or functional status upon presentation to a SNF was associated with increased risk for poorer outcomes. Patients who were unable to communicate upon presentation to the SNF had a 42% lower risk of showing any FI at 30 days compared to those who could communicate. Patients with motor dysfunction had a 17% lower risk of showing any FI. Patients with communication impairment and patients with motor impairment had lower risk of being discharged to home, 86% and 85% respectively. Overall, older patients (> 65yrs) with TBI had a 3.6 times higher risk of death at 30 days in SNF compared to younger patients with TBI (RR=3.6, 95%CI=2.9,4.3). The risk of death was higher in patients with poor cognitive function was (RR=8.9, 95%CI=7.4, 10.6) significant motor impairment (RR=5 95%CI=4.5, 5.6) and in patients with impairment in communication (RR=5.5, 95%CI=5.0, 6.0) compared to those without the respective deficits.

Conclusion:
Our results further suggest that among a population with TBI admitted to SNFs, the likelihood of adverse outcomes varies significantly by key clinical and demographic characteristics. Understanding this can help set expectations to patients, families, as well as providers.  Moreover, this data might help guide future therapies and calls for dedicated TBI rehabilitation facilities. 
 

34.05 The “Lethal Triad” in Blunt Traumatic Hemorrhagic Shock: What is Contributing to “Lethal?”

Posted on January 25, 2017

J. O. Hwabejire1, C. E. Nembhard1, T. A. Oyetunji3, W. R. Greene2, M. Williams1, E. E. Cornwell III1, S. M. Siram1  1Howard University College Of Medicine,Surgery,Washington, DC, USA 2Emory University Hospital,Surgery,Atlanta, GEORGIA, USA 3Northwestern University Feinberg School Of Medicine,Surgery,Chicago, ILLINOIS, USA

Introduction:  The combination of acidosis, coagulopathy and hypothermia in a trauma patient is a harbinger of death. Resuscitation and control of bleeding are two key tenets of trauma care designed to halt this vicious cycle. We examine clinical variables that contribute to mortality in blunt traumatic shock patients presenting with this triad.   

Methods:  The Inflammation and the Host Response to Injury database was analyzed. Patients who, on presentation to the emergency room (ER), had the triple combination of severe hyperlactatemia (serum lactate >4 mg/dL), coagulopathy (INR >1.5) and hypothermia (body temperature ≤ 36 °C) were included. Univariate analyses were used to compare survivors and non-survivors while multivariable analysis was used to determine predictors of mortality.

Results: A total of 172 patients met all three criteria. The mean age of the cohort was 39 years, 70% were male, and 90% were White. Their overall mortality was 30.8%. There was no difference in pre-injury comorbidities, body mass index, Injury Severity Score, multiple organ dysfunction score, ER systolic BP, ER heart rate, ER body temperature, crystalloid volume administered within 12 hours, WBC count, and platelet count between survivors and non-survivors. Compared to survivors, non-survivors were older (46±22 vs. 37±18 years, p=0.005), more coagulopathic (ER INR 2.6±1.5 vs. 2.1±1.2, p=0.021), had higher ER lactate (7.8±3.2 vs. 6.5±2.2 mg/dL, p=0.002), higher APACHE II score (37±6 vs. 31±5, p<0.001), larger volume of transfused blood within 12 hours (6848±5574 vs. 3232±2779 mL, p<0.001) and were more likely to have a cardiac arrest (50.9% vs. 2.5%, p<0.001) or myocardial infarction (7.5% vs. 0.8%, p=0.032).  Non-survivors were more likely to undergo angiographic embolization (37.7% vs. 14.3%, p=0.001) or an operative thoracic procedure (thoracotomy, sternotomy, or VATS, 26.4% vs. 7.6%, p=0.01), although they had similar laparotomy rates (52.8 % vs. 49.6%, p=0.694). Independent predictors of mortality in this cohort include APACHE II score (OR: 1.15, CI: 1.04-1.28, p=0.005), cardiac arrest (OR: 21.21, CI: 5.06-88.87, p<0.001), and angioembolization (OR: 4.31, CI: 1.45-12.83, p=0.009). For patients who underwent angiographic embolization, mortality was 54.1%, and for those who suffered a cardiac arrest, mortality was 90%. 

Conclusion: In blunt trauma patients with hemorrhagic shock who met criteria for the lethal triad on presentation to the ER, angiographic embolization, APACHE II score, and cardiac arrest independently predict mortality. The exact role of angiographic embolization, which should be a life-saving procedure, deserves further study.

 

34.04 Complications in the Morbidly Obese After ORIF of Isolated Open Tibia Fractures

Posted on January 25, 2017

N. N. Branch1, R. Wilson1  1Howard University College Of Medicine,Washington, DC, USA

Introduction: Obesity is known to be associated with postoperative complications. Few articles investigate the association between obesity and isolated open tibia fractures {IOTF), thus understanding how this condition will affect patient outcomes is critical. We aim to determine perioperative complications after open reduction and internal fixation (ORIF) of IOTF in morbidly obese (MO) patients using a large national sample.

Methods: Using ICD-9 codes we reviewed the National Trauma Data Bank (NTDB) from 2007- 2010. Patients 18 years and older with open IOTF who underwent ORIF at level I or II trauma centers were identified. Multivariate logistic regression and univariate analyses were used to investigate postoperative complications. The NTDB defines obesity as a body mass index (BMI) of 40 or greater, which for the purposes of this study is reference as morbid obesity.

Results: 7,201 cases met inclusion criteria. The majority were white males ages 25-44 with private insurance injured in motor vehicle collisions. 248 (3.44%) of those patients were MO. On multivariate analysis morbidly obese patients had a 40% increased odds (OR: 1.41 CI: 1.07-1.84 p=0.014) of undergoing ORIF after hospital day 2 and were two times more likely to develop acute respiratory distress syndrome (OR: 2.0 Cl: 1.08-3.71 p=0.028). MO patients were more likely to develop superficial (OR: 3.19 Cl: 1.17-8.74 p=0.024), organ/space (OR: 1.75 Cl: 1.08-2.85 p=0.024), or any surgical site infection (OR: 1.93 Cl: 1.23-3.03 p=0.004). MO patients were two times more likely to have at least one complication (OR: 2.01 Cl: 1.35-2.99 p=0.001), and more than four times more likely to develop cardiac arrest (OR: 4.28 Cl: 1.31-13.71 p=0.014). Mortality and length of stay were not associated with being MO (Table 1).

Conclusion: Morbid obesity was found to be associated with increased perioperative complications in IOTF. These patients are at greater risk of infectious complications, most notably superficial surgical site infections. Delays in time to surgical fixation may be secondary to concurrent trauma related injuries while optimizing the patient for surgery. Despite having a significantly higher risk of cardiopulmonary complications, morbidly obese patients did not have an increased mortality.

34.03 Can We Truly Impact The Incidence Of Post-Traumatic Seizures Using Anti-Epileptic Drug Prophylaxis?

Posted on January 25, 2017

M. B. Singer1, B. Zangbar1, K. Williams1, B. Joseph1, A. Tang1, N. Kulvatunyou1, P. Rhee1, T. O’Keeffe1  1University Of Arizona,Division Of Trauma, Critical Care, Burns, And Emergency Surgery,Tucson, AZ, USA

Introduction:  Patients who sustain traumatic brain injury (TBI) are at risk for post-traumatic seizures (PTS). The reported incidence of early PTS varies widely from 2% to 14% but there is disagreement over which patients are at highest risk. We hypothesize that the PTS rate is lower than previously reported and that specific types of brain injury are not predictive of PTS.

Methods:  We conducted a retrospective cohort analysis of all TBI patients admitted to our level one academic trauma center over a nine year period (January 1, 2006 to December 31, 2015). Demographic and injury data including ISS, head AIS, TBI type, history of seizure disorder, time of seizure and prophylactic AED use were collected for all patients who experienced PTS. Seizures were defined by clinical criteria due to the fact that electroencephalography is not routinely performed at our institution. Multivariate logistic regression was used to identify independent predictors of PTS.

Results: 10,001 TBI patients were evaluated at our institution during the nine-year study period, 180 (1.8%) of whom experienced PTS. 63 (34.8%) of these patients seized in the field or in the trauma bay and 23 (12.7%) had a prior history of seizure disorder. Of the 118 patients who seized after hospital admission, 28 (23.7%) were receiving AED prophylaxis at the time of PTS.  Head abbreviated injury scale (AIS) ?  3, injury severity score, and subdural hematoma were independent predictors of PTS (Table 1). PTS did not independently predict mortality.

Conclusion: The rate of PTS at our institution is significantly lower than other published reports. This may reflect a change in the natural history of PTS or changes in intensive care unit management. In light of the high rates of pre-admission and breakthrough PTS (approximately 50%), we question the effectiveness of routine AED prophylaxis in reducing early PTS. Specific risk factors for PTS remain elusive.
 

34.02 Concussion Among The Elderly: A Silent Epidemic

Posted on January 25, 2017

A. Caiado1, S. Armen1, R. Staszak1, J. Chandler1, K. Fitzgerald1, S. Allen1  1Penn State Hershey Medical Center,Hershey, PA, USA

Introduction: A significant amount of attention has been given to the identification and consequences related to mild traumatic brain injury (mTBI) or concussion among young athletes and combat veterans. The interest in concussion is the result of the insidious yet devastating long-term sequelae of these seemingly minor head injuries. The effects of concussion among the elderly are largely unstudied; furthermore, the incidence and identification of concussions in this age group is not well delineated. The goal of this study was to investigate the incidence of the diagnosis of concussion among the elderly population as compared to an injury matched younger cohort. We hypothesized that elderly patients who meet criteria for the diagnosis of a concussion are under-diagnosed compared to a younger, injury matched cohort.

Methods: The trauma registry of an academic Level 1 trauma center was retrospectively queried over a 2 year study period. Adult patients (>18 years of age) with an abbreviated injury score (AIS) head of 1-2 from a blunt mechanism of injury and who met criteria for mTBI as set forth by the American Association of Neurosurgeons (AANS) were evaluated. Demographic information, as well as Glasgow coma score (GCS), the GCS motor score (MS), injury severity score (ISS), blood alcohol concentration (BAC), ICU length of stay, and hospital length of stay were studied. The charts of all patients were queried for the specific diagnoses of mTBI or concussion (n=618). An older cohort (>65 years of age) (n=231) was then compared to a younger cohort (18-64 years) (n=387) matched on MS, GCS and ISS. The difference in the incidence of the diagnosis of concussion was calculated. Chi-squared tests as well as student’s t-tests were used for statistical analysis as appropriate.

Results:There was a steady and statistically significant decrease in the incidence of the diagnosis of concussion across the older decades of life. Nearly 90% of patients 18-24 years of age were diagnosed with concussion as compared to just over 50% of those over the age of 75 years and is inversely related to the incidence pre-existing dementia. There were no differences in ISS, MS and GCS among the two study groups. Concussion was formally diagnosed significantly less in older patients as compared to the injury matched younger group.

Conclusion:Mild traumatic brain injuries are common among patients who sustain blunt injuries. However, while younger patients are frequently diagnosed with concussion, only 50% of elderly adults who meet criteria for mTBI are formally diagnosed with concussion. The failure to recognize this ubiquitous injury among the elderly may lead to less interventions and significantly poorer long-term outcomes, especially in those with a prior history of dementia. Future studies should focus on early and aggressive interventions for concussion in the elderly in an effort to mitigate the negative sequelae of these injuries and improve quality of life.

 

34.01 Bile Leak After Operative AAST Grade III-V Liver Injuries, Risk Factors and Management

Posted on January 25, 2017

R. J. Miskimins1, A. A. Greenbaum1, P. Kilen2, R. Preda1, S. W. Lu1, T. R. Howdieshell1, S. D. West1  1University Of New Mexico HSC,Department Of Surgery,Albuquerque, NM, USA 2University Of New Mexico HSC,School Of Medicine,Albuquerque, NM, USA

Introduction: Bile leak from the intrahepatic biliary tree is a major cause of morbidity after high grade liver injury. The rate of bile leak after hepatic trauma ranges from 0.5-21%. The risk of a bile leak increases with higher grade injury, however other risk factors have not been characterized. The aim of our study was to clarify the incidence, risk factors and management of intrahepatic bile leak following laparotomy for high grade liver injury.

Methods:  A retrospective review of patients with complex liver injuries, defined as AAST grade III-V, who underwent laparotomy from Jan 2008 to July 2015 at an ACS-verified Level I trauma center was performed. Patients who died within 72 hours or under the age of 14 were excluded. Bile leak was defined as bilious output lasting more than 14 days from a surgically or interventional radiology percutaneous drain (IRPD). The grade of liver injury, number of laparotomies, operative techniques, use of hepatic angioembolization (HAE), placement of surgical drains, reason for laparotomy, number of readmissions, and interventions for management of bile leak were recorded. The institutional trauma database was used to obtain demographics, initial vital signs, ISS, length of stay (LOS), ICU LOS, and mechanism of injury (MOI). Statistical analyses were performed using Chi-squared and Fisher exact tests for categorical data, and the Mann–Whitney U-test for continuous variables. P-values < 0.05 were significant.

Results: 117 patients met inclusion criteria, 29 (25%) developed a bile leak (BLG) and 88 (75%) had no leak (NLG).  There was no significant difference between the groups in age, sex, MOI, initial vitals, ISS, ICU LOS or reason for laparotomy. The BLG had higher grades of injury (Grade 5: 45% vs 10%, Grade 4: 41% vs 31%, Grade 3: 14% vs 60%,  P <0.01), longer hospital LOS (29 vs 21 days, p <0.01) and were more likely to be readmitted (41% vs 15%, P<0.01). No significant difference in the rates of perihepatic packing, argon beam hepatorrhaphy, gelfoam packing, or suture hepatorrhaphy was observed. The BLG required more laparotomies (3.5 vs 2.2, p<0.01), were more likely to have excisional debridement (38% vs 9%, p<0.01), and HAE (38% vs 6%, P=0.03). Ninety-seven percent (n=28) in the BLG had perihepatic drains placed prior to abdominal closure, and 65% (n=19) of bile leaks were managed entirely with these drains. Seventeen percent (n=5) required IRPD, 14% (n=4) underwent ERCP and insertion of biliary stent in addition to the perihepatic surgical drains, and 3% (n=1) underwent both IRPD and ERCP.

Conclusion: In patients with AAST Grade III-V liver injury who require laparotomy, the grade of injury, use of hepatic embolization and excisional debridement are assoicated with development of bile leak.Those who develop a bile leak have longer hospital LOS and are more likely to be readmitted. The majority of bile leaks can be managed conservatively with perihepatic drain placement prior to definitive abdominal closure.

13.20 The Bifid Recurrent Laryngeal Nerve – Anatomical Details & Operative Implications

Posted on January 25, 2017

J. C. Lee1,2, A. Kiu1, P. Chang1, J. Serpell1,2  1The Alfred Hospital,Department Of General Surgery,Melbourne, VICTORIA, Australia 2Monash University,Endocrine Surgery Unit,Melbourne, VICTORIA, Australia

Introduction:  The identification and preservation of the recurrent laryngeal nerve (RLN) is paramount during thyroid surgery. Due to the slenderness of the branches, a RLN with an extralaryngeal bifurcation is at higher risk of intraoperative injury. When bifid, the motor fibres of a bifid RLN are located mainly in the anterior branch, and the sensory fibres in the posterior branch. However, it has not been documented whether the motor or sensory branch is likely to be thinner and therefore more prone to injury. This study aimed to measure the widths of the bifid RLN trunk and its branches, and to determine their possible associations with demographic factors. 

Methods:  This is a prospective observational study over 18 months at The Alfred Hospital, Melbourne, Australia, in patients undergoing thyroid surgery. The widths of the RLN trunk and branches were measured with Vernier calipers to the nearest 0.1 mm. Demographic data including age, gender, height, weight, and body mass index (BMI) were collected. Nerve widths were compared using Student’s t-test, and RLN widths and demographic data were correlated with Spearman correlation co-efficient (Stata 13).

Results: A total of 150 RLNs were eligible for inclusion during the 12-month study period. Of those, 34 bifid RLNs were identified in 32 patients, and therefore included in the analysis. The main RLN trunk had a mean width of 2.37 (range 1.7 – 4.0) mm. Whereas the mean widths for the anterior and posterior branches were 1.55 (0.8 – 2.5) mm and 1.33 (0.5 – 2.9) mm respectively. Both the anterior and posterior branches were significantly smaller than the main trunk (both p < 0.01). However, the branches were not statistically different from each other in their widths. Body weight and BMI positively correlated to the widths of both the anterior branch (p = 0.003 & p = 0.01 respectively) and posterior branch  (p = 0.02 & p = 0.04 respectively). There was no correlation between age, height and either the main trunk or branches of the RLN.

Conclusion: As expected, the width of the RLN trunk is significantly greater than either of the branches of a bifid RLN. The knowledge of this may help alert the thyroid surgeon to the possibility of a bifid RLN during the process of dissecting along the RLN. More importantly, the similarity in the widths of the branches suggests that it is not possible to determine if a fine nerve branch is likely to be the anterior (motor) or posterior (sensory) branch. Low body weights or BMI may be a clue to possible delicate RLN branches.

 

 

13.19 Postoperative Complications in Patients with Inflammatory Bowel Disease

Posted on January 25, 2017

S. Stringfield1, S. Ramamoorthy1, L. Parry1, S. Eisenstein1  1University Of California,Surgery,San Diego, CA, USA

Introduction:  Patients with Inflammatory Bowel Disease (IBD) are at high risk for postoperative complications. Many patients will receive anti-TNF medications or other biologic medications prior to surgery. There is still controversy as to whether anti-TNFs are associated with complications. Many new biologic medications have not been studied in surgical patients. The purpose of this study is to identify rates and types of postoperative complications in patients with IBD who have undergone abdominal surgery, and identify predictors of these complications. 

Methods:  Retrospective review of patients with IBD who underwent abdominal surgery at our institution June 2014-June 2016. Preoperative, perioperative, and postoperative data was collected. Categorical variables were analyzed using Fisher’s exact test or Chi-square test and continuous variables were analyzed using two sided t-test for independent means. Univariate and multivariate analyses were performed using binary logistic regression. 

Results: We identified 155 abdominal operations performed on IBD patients. Overall complication rate was 40%, with infectious complications the most common with rate of 27% overall. Univariate analysis showed predictors of complications to be age (p=0.028, OR 0.98), BMI (p=0.02, OR 0.93), recent weight loss (p=0.029, OR 2.12), and intraoperative blood loss (p=0.006, OR 0.996). Current use of any biologic medication was not a significant predictor (p=0.144), however vedolizumab use was a predictor (p=0.041, OR 2.46). On multivariate analysis, age (p=0.014, OR 1.03), BMI (p=0.027, OR 1.09), weight loss (p=0.041, OR 2.14), emergent case (p=0.018, OR 2.74), and vedolizumab use (p=0.016, OR 3.27) remained significant predictors of complications. Forty-one percent of patients were on a biologic medication at time of surgery. These patients were more likely to have Crohn’s Disease (59% v 26%, p<0.001), lower preoperative hemoglobin (10.9 v 12.0, p=0.0004) and albumin (3.6 v 3.9, p=0.027), to be on thiopurines (31% v 11%, p=0.003) or steroids (55% v 14%, p<0.001) at the time of surgery, and undergo emergent surgery (36% v 16%, p=0.008). Patients on biologics had a 47% overall and 28% infectious complication rate. Patients not on biologic medications had a 35% overall and 25% infectious complication rate. Complication rates did not vary significantly, except risk of bleeding requiring a transfusion was higher in patients on biologic medications (23% v 11%, p=0.047). 

Conclusion: Patients with IBD have a high rate of postoperative complications. Predictors of complications include age, BMI, weight loss, intraoperative blood loss, and vedolizumab use. Only rates of hematologic complications varied significantly between patients on biologic medications and those not on biologics. 

 

« Previous Page
Next Page »