88.11 Analysis of Short-term Outcomes After Endovascular and Open Infrapopliteal Revascularization

Posted on January 26, 2017

H. J. Leraas1, S. S. Adkar1, Z. Sun1, B. F. Gilmore1, U. P. Nag1, C. A. Long1, W. S. Jones1, C. K. Shortell1, R. S. Turley1  1Duke University Medical Center,Durham, NC, USA

Introduction:  Consensus guidelines regarding optimal treatment of infrapopliteal arterial disease for critical limb ischemia (CLI) lack specificity with respect to surgical and endovascular interventions. Evidence that endovascular approaches to infrapopliteal disease are associated with lower 30-day morbidity than open surgical bypass consists predominantly of small case series without matched surgical comparisons. The objective of this study is to compare the 30-day outcomes of patients treated with endovascular or open interventions for infrapopliteal disease.

Methods: The 2011-2013 NSQIP Vascular database was queried for CLI patients undergoing femoral distal bypass, popliteal distal bypass, or tibial angioplasty. Surgical patients were propensity matched 2:1 to tibial angioplasty using the nearest neighbor method. Variables for matching were age, race, BMI, elective vs. emergency surgery, ASA class, rest pain vs. ischemic tissue loss, diabetes, renal failure, dialysis, wound classification, and smoking. Primary endpoints were 30-day major adverse events (death, stroke, MI), or major amputation. Secondary endpoints were post-operative wound complications, length of stay, and readmission.

Results: 317 endovascular patients were matched to 634 surgery patients. Types of surgery were femoral distal bypass with prosthetic (28%), femoral distal bypass with saphenous vein (48%), and popliteal distal bypass with saphenous vein (24%). All endovascular patients underwent tibial angioplasty/stenting. Median age (95% CI) was 54 (45-63) and 53 (44-62) years and median BMI (95% CI) was 27 (24-31) and 27 (24-31) for surgery and endovascular, respectively. Tissure loss was present in 77% of surgery patients and 80% of endovascular patients, with all remaining having rest pain. The need for emergency intervention (3.9% vs. 3.8%) was similar in both groups. While there were no significant differences in major amputation or 30-day mortality between the treatment groups, surgery had a higher incidence of combined stroke/MI (4.6% vs. 1.3%, p=0.009), post-operative wound infection (12.8% vs 7.3%, p =.01), and longer median hospital stay (8 days vs 3 days, p <.001). Surgery patients were also more likely to be discharged to a skilled care facility (25.9% vs 12.6%, p<.001) or be readmitted (4.4% vs 3.2%, p=.024). (Table 1) 

Conclusion: Endovascular treatment of infrapopliteal disease with CLI has similar 30-day major amputation and mortality with fewer stroke/MIs, wound infections, hospital days and readmissions compared to surgery. Long term comparisons are needed to provide objective data on which to form consensus guidelines for the optimal treatment for critical limb ischemia due to infrapopliteal arterial disease.

 

88.10 Open & Endovascular Ruptured AAA Repair Have Equivalent Outcome When Performed Weekdays vs Weekends

Posted on January 26, 2017

G. Gilot1, E. Abotsi1, G. Ortega1, C. Zogg2, D. Taghipour1, D. Tran1, E. Cornwell1, K. Hughes1  1Howard University College Of Medicine,Department Of Surgery,Washington, DC, USA 2Yale University School Of Medicine,New Haven, CT, USA

Introduction:

Studies have demonstrated that there may be an increased risk of postoperative complications for certain surgical procedures when performed during weekends.  This “weekend effect”, however, has not been studied for vascular surgical procedures such as repair of an abdominal aortic aneurysm (AAA). We undertook this study to determine if repair of ruptured AAA (rAAA) performed on weekends differed in outcomes as compared to rAAA operations performed on weekdays.    

Methods:

We conducted a retrospective review of the Nationwide Inpatient Sample (NIS) database to identify all patients who underwent an open or endovascular rAAA repair from 2007 to 2012. Data analyzed included patient characteristics including demographics and comorbidities, as well as postoperative morbidity and mortality. Operations were dichotomized into Weekday (Mon-Fri) and Weekend (Sat and Sun); and open versus endovascular. Multivariate analysis was performed adjusting for patient characteristics and comorbidities to evaluate postoperative outcomes by groups. 

Results:

A total of 15,418 patients had a rAAA repair. The majority were non-Hispanic White (87%), males (72%), with a mean age of 75 (SD±10) years. Overall mortality was 47.8%.  Postoperative complications occurred in 30.5% of patients.

 

In those undergoing open repair, (n=6,623), the mean age was 73 (SD±9) years with an overall mortality of 39.5% and overall complication rate of 43.1%. Most open rAAA repairs were performed on a weekday (72.4%). Comparing weekday versus weekend, open rAAA repair, there were similar rates of mortality (40.6% weekday vs. 39.1% weekend; p=0.264) and morbidity (43.5% weekday vs. 42.9% weekend; P=0.693).

 

In those undergoing endovascular repair, (n=2,170), the mean age was 74 (SD±10) years with an overall mortality of 26.5% and morbidity of 31.6%.  The mortality rate was 26.5%. The in-hospital complication rate was 31.6%. A majority of operations were performed on a weekday (75.2%). Comparing weekday versus weekend endovascular rAAA repair, there were similar rates of mortality (24.5% weekday vs. 32.7% weekend; OR [95%CI]: 1.30 [0.87-1.96]) and morbidity (31.9% weekday vs. 31.5% weekend; p=0.843).

Conclusion:

rAAA repair is associated with equivalent outcomes when performed on the weekend vs weekday.  Endovascular repair of rAAA is associated with superior outcomes. 

 

88.09 Race, Socioeconomic Factors and Leg Amputations among Patients with P.A.D. in Texas

Posted on January 26, 2017

N. R. Barshes1, K. D. Smith3, H. Serag3, B. J. Carter2, S. O. Rogers2  1Baylor College Of Medicine,Division Of Vascular And Endovascular Surgery, DeBakey Department Of Surgery,Houston, TX, USA 2University Of Texas Medical Branch,Galveston, TX, USA 3University Of Texas Medical Branch,Center To Eliminate Health Disparities,Galveston, TX, USA

Introduction:  Previous analyses of national data have suggested racial disparities in leg amputation rates. We sought to determine whether race- or insurance-based disparities in leg amputations occur among people in Texas with peripheral artery disease (PAD).

Methods:  Deidentified hospital admission data from the Texas Inpatient Public Use Data File was used to identify admissions associated with the diagnosis of PAD as well as either revascularization (endovascular or surgical procedures) or leg (i.e. above-ankle) amputation from 2004 to 2010. Multivariate regression models were used to identify factors independently associated with ER admission and leg amputation. All analyses were performed using Intercooled State v8.0 (College Station, TX), with p<0.05 considered significant. 

Results: 29,128 revascularization procedures and 6,482 leg amputations were performed in Texas from 2004-2009 for PAD-related diagnoses. The unadjusted incidence rates of leg amputation were 5.0 per 100,000 total population per year (per 100K/yr) for non-Hispanic white persons versus 7.2 for black persons, 3.1 for Hispanic persons, and 0.7 for Asian persons. Leg amputation rates also ranged from 2.7 per 100K/yr in zip codes in lowest quartile of poverty prevalence to 5.0 per 100K/yr in the middle two quartiles and 6.8 per 100K/yr in the highest quartile of poverty prevalence. Hospital admission through the emergency room was much more common among those without insurance (odds ratio [OR] 2.2, p<0.001) or only Medicaid coverage (OR 1.1, p=0.002) and was much less common among those with Medicare, HMO/PPO, or private insurance coverage (odds ratio [OR] 0.68-0.76, all p<0.0001). After adjustment for clinical factors (incl. foot infection, comorbidities), demographic features (incl. age, gender), and geography (viz. Texas public health region), leg amputations without antecedent revascularization attempts occurred much more frequently in patients that were categorized as black (odds ratio [OR] 2.1, p<0.001) or Hispanic (OR 1.6, p<0.001), those with Medicaid coverage (OR 2.1, p<0.001), and those that were uninsured (OR 2.0, p<0.001; Table 1). Overall model R2 was 0.16. Race/ethnicity, Medicaid coverage, or uninsured status was not associated with an increased rate of leg amputation in patients that had undergone revascularization.

Conclusion: Leg amputations among people with PAD in Texas vary widely, with higher risk-adjusted rates occuring in people who are uninsured, insured only by Medicaid, or are categorized as black or Hispanic. State-wide efforts should focus on addressing these existing health disparities. 

88.07 Mortality following Endovascular versus Open Repair of Abdominal Aortic Aneurysm in the Elderly.

Posted on January 26, 2017

S. Locham1, R. Lee1, B. Nejim1, H. Aridi1, M. Faateh1, H. Alshaikh1, M. Rizwan1, J. Dhaliwal1, M. Malas1  1Johns Hopkins University School Of Medicine,Surgery,Baltimore, MD, USA

Introduction:
Prior RCTs have reported better perioperative outcomes following endovascular aneurysm repair (EVAR) as compared to open aneurysm repair (OAR). EVAR-1 and DREAM trial reported significantly higher mortality for OAR as compared to EVAR. However most of these studies excluded the elderly.  Age is a well-known risk factor for postoperative death and the efficacy of these approaches remains controversial in the elderly population. The aim of the study is to provide recent real world outcomes using the NSQIP database (2010-2014) exclusively looking at the predictors of mortality in a large cohort of elderly population in the United States. 

Methods:

Using the NSQIP targeted vascular database (2010-2014), we identified all patients over 70 years of age who underwent OAR and EVAR for non-ruptured AAA. Explanatory analyses using Pearson’s Chi-square and Student’s t-tests were performed. Univariate and multivariable logistic regression analyses were implemented to examine postoperative morbidities and mortality adjusting patient demographics and characteristics.

Results:
A total of 5,332 non-ruptured AAA repairs were performed [OAR: 809 (15%) vs. EVAR: 4,523 (85%)]. The majority of patients were male (77%) and white (81%) with mean age of 78 ± 6 years. Diabetes mellitus and obesity were more prevalent in the EVAR group (15% vs. 12%, p=0.01) and (30% vs. 25%, p=0.002), respectively. Whereas, history of chronic obstructive pulmonary disease (COPD) (22% vs. 19%, p=0.02) and smoking status (35% vs 23%, p<0.001) were more likely to be seen in patients undergoing an OAR. On average the operative time in minutes (250 vs. 151) and mean length of stay in days (11 vs. 3) was also longer for patients undergoing OAR versus EVAR (p<0.001). The mortality was higher following OAR versus EVAR (8% vs 3%, p<0.001). Compared to EVAR, OAR was associated with higher rates of cardiac (7% vs. 2%), renal (7% vs. 1%), pulmonary (20% vs. 3%) and any wound complications (4% vs. 2%) (all p<0.05). After adjusting for patients’ characteristics and comorbidities, OAR was associated with 3 times higher mortality than EVAR [OR(95%CI): 3.04(2.01-4.57), p<0.001]. The predictors of mortality in our elderly cohort were age, female gender, smoking status, functional dependency, history of COPD, steroid use, bleeding disorders, progressive renal failure, transfusion, aneurysm diameter and Type IV TAAA. 

Conclusion:

Our study reflects contemporary real world outcomes following repair of non-ruptured AAA in the elderly. Endovascular approach was associated with significant reduction in the risk of postoperative cardiac, pulmonary and renal complications the elderly. Open repair was associated with 3 fold increase in mortality compared to EVAR and should be avoided in the elderly. Further prospective studies involving geriatric population is required to better understand the predictors of mortality following AAA repair. 

88.05 Lower Extremity Bypass: Are We Meeting Guidelines?

Posted on January 26, 2017

S. S. Adkar1, R. S. Turley1, L. Youngwirth1, E. Benrashid1, C. K. Shortell1, L. Mureebe1  1Duke University Medical Center,Department Of Surgery,Durham, NC, USA

Introduction:

The American Heart Association recommends use of an anti-platelet agent (APA) and a statin in patients with symptomatic peripheral arterial disease. The extent of guideline adherence by clinicians and patients is unknown. Given the inherent morbidity of lower extremity bypass (LEB), including limb loss, loss of patency, cardiac risk and bleeding complications, we sought to assess current usage of APA and statins and their effects on LEB outcomes.

Methods:
Data were obtained from the 2011-2013 ACS-NSQIP participant user files with LEB targeted data. Patients with LEB performed for aneurysmal disease and trauma were excluded from this analysis.  Patient, procedural characteristics, and 30-day postoperative outcomes were compared using Pearson Χ2 tests for categorical variables and Wilcoxon rank-sum tests for continuous variables. The primary outcome measures were graft patency rates, perioperative bleeding, and mean hospital length of stay.

Results:
During the study period, 5500 total LEB were identified. Demographics and major complications are displayed in Table 1. We stratified patients treated with both statin and APA (59.4%), statin only (9.3%), APA only (20.3%), and those receiving neither medication (11%). Smoking was more prevalent in the group receiving neither medication and diabetics were more often treated with either an APA or statin (p<.01). Graft patency rates were higher in patients receiving APA compared to other groups (51.3% vs 44.4%, p < .001). The incidence of bleeding requiring transfusion or a second procedure was higher in patients receiving a statin but not APA alone (13.2% vs 19%, p<.01). Mean hospital length of stay was significantly lower in patients receiving APA alone (8 days vs 9 days, p<.001). No significant difference in post-operative myocardial infarction or stroke was observed between treatment groups (p=.762). No differences in wound infection or mortality was observed between groups.

Conclusion:

Both the American Heart Association and the Society for Vascular Surgery have given their strongest recommendations that patients with atherosclerotic lower extremity arterial disease be treated with APA and statins (literature grade: Level 1A). We fail to accomplish this in 4 out of every 10 patients undergoing LEB nationally. Furthermore, APA, and not statin, treatment is associated with a shorter post-operative length of stay, but statin use (not APA) is associated with an increased risk of bleeding after surgery. This dichotomy between guidelines and our observed data suggests an ambiguous interplay between statins and APA that will require concentrated exploration.

88.04 Experience with Tibial/Peroneal Polytetrafluoroethylene Bypasses for Critical Limb Ischemia (CLI)?

Posted on January 26, 2017

N. J. Gargiulo4, E. C. Lipsitz2, N. Cayne1, G. Landis3, F. J. Veith1  1New York University School Of Medicine,New York, NY, USA 2Albert Einstein College Of Medicine,Bronx, NY, USA 3North Shore University And Long Island Jewish Medical Center,Manhasset, NY, USA 4Clinch Valley Medical Center,Vascular Surgery,Richlands, VIRGINIA, USA

Introduction:  Polytetrafluoroethylene (PTFE) tibial and peroneal bypasses without vein cuffs, patches or arteriovenous fistulas have been advocated for critical limb ischemia in circumstances when autologous saphenous vein is not available. This reviews a 30-year experience.

Methods:  A retrospective analysis was performed on a group of 377 patients with critical limb ischemia requiring revascularization between July 1977 and June 2011. These 377 patients underwent 411 PTFE bypasses to a tibial or peroneal artery without any adjunctive procedure.  Cumulative life table primary and secondary patency and limb salvage rates were calculated for those bypasses performed between July 1977 through June 1987 (Group I). These were compared to those performed between July 1987 through June 1997 (Group II) and July 1997 through June 2011 (Group III).  Ethnic background, TASC distribution, hemoglobin A1C levels, and inflammatory mediators (CRP, IL-6, and IL-10) were measured in a subset of Group III patients.  Multivariate logistic regression was used to calculate intergroup differences with significance determined as P<0.02.

Results: Five- and 10-year primary patency and five- and 10-year limb salvage for Group I patients was 28.6%, 9.0%, 55.0%, and 27% respectively, for Group II patients was 27.3%, 9.9%, 50.0%, and 26%,  respectively, and for Group III  patients was 34.0%, 11%, 73.3%, and 33% respectively.  Patency and limb salvage for Group III patients exceeded that observed in Group I and II patients which correlated with the implementation of several perioperative strategies.  Group III patients manifested a statistically better outcome compared to Group I and II patients.  Interestingly, Group III patients manifested a greater distribution of TASC II D atherosclerotic disease and a greater percentage of Latino patients compared to Group I and II patients, and in this subset of patients had a trend towards a statistically worse five- and 10-year primary patency and five- and 10-year limb salvage rate (P=0.12).  There was no direct correlation between hemoglobin A1C level, PTFE patency or successful limb salvage rates in any of the Groups (P>0.02).  There was a trend towards a statistical correlation to PTFE graft patency and inflammatory mediators (CRP, IL-6 and IL-10, P=0.17).

Conclusion:  PTFE bypasses without adjunctive procedures to infrapopliteal arteries is an acceptable alternative option for those patients without autologous vein facing imminent amputation in this small cohort of patients.  

88.03 Natural History of Dialysis Interventions Based Upon Initial Dialysis Access Type

Posted on January 26, 2017

E. S. Lee1,2, K. C. Chun1, T. Yenumula1, A. S. Schmidt1, K. M. Samadzadeh1, A. Rona1, A. Gonzalves1, M. D. Wilson3, R. E. Noll1,2, E. S. Lee1,2  1VA Northern California Health Care System,Surgery,Mather, CA, USA 2University Of California, Davis,Surgery,Sacramento, CA, USA 3University Of California, Davis,Department Of Public Health Sciences, Division Of Biostatistics,Sacramento, CA, USA

Introduction: End-stage renal disease (ESRD) patients face an unknown number of dialysis access interventions (DAI) that could occur throughout their lifetime on dialysis. The purpose of this study is to determine the frequency of DAI per year while on dialysis to better inform prospective dialysis patients.

Methods: A retrospective chart review of ESRD patients on dialysis for a minimum of 1 year was conducted at a large regional medical center. The frequency of DAI per year was the primary outcome measure. An intervention is defined as any procedure (fistulogram, catheter placement, or new access) done to regain dialysis access for the patient. Average days between interventions is determined by time between first dialysis access until date of death or end of study analysis (Dec. 31, 2015), divided by total number of interventions. The primary outcome was then evaluated with other patient factors such as initial access type (catheter, fistula, or graft), time between DAI, time on dialysis, age, hypertension, diabetes, smoking, cholesterol, triglycerides, statin use, blood thinner use, body mass index, coronary artery disease, peripheral vascular disease, chronic obstructive pulmonary disease, and stroke. A general linear model (GLM) was fit to test for associations between the measured variables and average days between interventions. The Tukey correction for multiple comparisons was used to compare across groups in variables significant from GLM.

Results: A total of 166 patients (mean ± standard deviation; 67.2 ± 10.3 years) were analyzed from 1991 to 2015 in this study. The patients are comprised of 54.2% (n=90) white, 37.4% (n=62) African American, and 8.4% other. Patients averaged 1.4 ± 1.4 DAI per year (7.2 ± 8.5 DAI total, range: 1 to 37 DAI) within an average follow up length of 5.6 ± 3.8 years while on dialysis. The average days between interventions based upon initial access type were: catheter (n=86) 475 ± 359.8 days, graft (n=16) 516.1 ± 448.5 days, and fistula (n=64) 546.8 ± 595.5 days (Table).

Conclusion: ESRD patients can expect 1 to 2 interventions per year to maintain dialysis access, regardless of initial dialysis access type. Although not statistically different, longer days between interventions can be expected with the fistula versus other access types and that the average life expectancy after starting dialysis is 5 to 6 years.

 

88.02 Bundling Of Reimbursement For Inferior Vena Cava Filter Placement Decreased Procedural Utilization

Posted on January 26, 2017

M. J. TerBush1, E. L. Hill1, J. Guido1, A. Doyle1, J. Ellis1, G. R. Morrow1, M. Stoner1, K. Raman1, R. J. Glocker1  1University Of Rochester,Surgery,Rochester, NY, USA

Introduction: On January 1, 2012, reimbursement for inferior vena cava filters (IVCF) became bundled by the Centers for Medicare and Medicaid Services (CMS). This resulted in a 70% decrease in RVUs associated with ICVF placement from 15.6 RVUs to 4.71 RVUs.  Our hypothesis was that procedural utilization would decrease following this change. We previously performed an analysis which revealed no significant changes in utilization. As new data have become available, we have revised our analysis in an effort to identify practice pattern changes.
 

Methods: We analyzed data from 2010-2014 using 5% inpatient, outpatient, and carrier files of Medicare limited data sets, analyzing IVCF utilization, controlling for total diagnosis of deep vein thrombosis (DVT) and pulmonary embolism (PE) (ICD – 9 codes 453.xx and 415.xx, respectively).

 

Results: In 2010 and 2011, the rates per 10,000 DVT/PE diagnoses were 918 and 1052, respectively (average 985). In 2012, 2013, and 2014, rates were 987, 877, and 605, respectively (average 823). The included figure demonstrates these trends graphically across different specialties. Comparing each year individually, there is a significant difference (p<0.0001) with 2012, 2013, and 2014 having lower rates of ICVF utilization. Comparing averages between the 2010-2011 and 2012-2014 groups, there is also a significant decrease in utilization after bundling (p<0.0001).

Conclusion: These data demonstrate that adjusted IVCF deployment rates dropped after the introduction of a bundled code with a reduced RVU and professional fee reimbursement value. This correlation may be evidence of a supply-sensitive medical service, and a successful realignment based on procedural valuation. More data from 2015 to present will be needed to show if this decrease in utilization continues to persist today.

 

88.01 Human Adiponectin Correlates with Severe Carotid Plaque Calcification

Posted on January 26, 2017

V. T. Hurst1,2, S. E. Deery1, G. Sharma2, M. D. Coll1,2, M. Tao2, K. Trocha2, A. Longchamp2, C. K. Ozaki2, R. J. Guzman1  1Beth Israel Deaconess Medical Center,Vascular And Endovascular Surgery,Boston, MA, USA 2Brigham And Women’s Hospital,Vascular And Endovascular Surgery,Boston, MA, USA

Introduction:  The mechanisms underlying calcification of atherosclerotic plaques remain obscure. Since carotid plaque calcification is associated with stability, and the adipose associated hormone adiponectin underlies multiple vasculo-protective pathways, we hypothesized that human adiponectin would positively correlate with severe carotid plaque calcification in a compartment specific (e.g. perivascular adipose) manner.

Methods:  Fifty-five patients who received either a carotid endarterectomy (n=38) or open lower extremity revascularization (n=17) and who had a carotid duplex ultrasound were studied. After informed consent, blood, perivascular and subcutaneous adipose samples, and medical history were obtained. Carotid bulb plaque was assessed via duplex ultrasonography, and patients were stratified into two groups: none/mild or severe calcification. Potentially relevant biomarkers were measured by multiplex bead immunoassay after tissue protein isolation, and data were normalized to initial adipose tissue mass. Wilcoxon Rank Sum testing was used to compare adiponectin levels in none/mild versus severe calcification patients. Categorical variables were presented as counts (percentages); continuous variables were presented as mean (standard deviation) or median (interquartile range), based on the normality of distribution. Differences between those with none/mild calcification and those with severe calcification were assessed using the Fisher’s exact test for categorical variables and either the Student T test or the Mann Whitney U test for continuous variables, where appropriate. All tests were 2-sided, and a P-value of less than 0.05 was considered significant. Statistical analysis was conducted using STATA 14.1.

Results: Of the clinical/biologic factors evaluated, carotid plaque calcification most strongly associated with adiponectin. In all compartments assayed, adiponectin levels positively linked to severe carotid plaque calcification (note log scale).

Conclusion: Human plasma, subcutaneous, and perivascular adiponectin levels positively correlate with carotid plaque calcification. These findings suggest relationships between adipose associated biomediators and vascular calcification. Furthermore, adiponectin based interventions may serve as novel strategies toward vascular plaque stabilization.

 

87.20 Case- Based Learning in Critical Care: An Appraisal of Current Literature

Posted on January 26, 2017

S. F. McLean1, A. H. Tyroch1, C. Ricci1, A. H. Tyroch1  1Texas Tech University Health Sciences Center At El Paso,Surger,El Paso, TX, USA

Introduction: Case—based learning (CBL)is an inquiry-based learning paradigm requiring some type of inquiry by the student, based on  actual or simulated cases, discrete learning objectives, and  with mentorship to  achieve the objectives.  A literature search was completed in order to assess its use in critical care education.

Methods: A literature search was completed using OVID and Pubmed.gov. Key word searches used “Critical Care” and separately Emergency Medicine, Medicine and Surgical Critical Care and “Case-Based” and “learning”.   Abstract total was 595. Abstracts were discarded due to not critical care topics, no data, not CBL, no English or Spanish translation. Method of teaching was assessed and categorized based on main teaching method.  Chi-square testing was used for categorical variables.

Results:  595 Abstracts were retrieved, out of these 39 articles were kept. Key  disciplines were Surgical Specialties (8, 20%), Medicine (26, 67%), and Anesthesia (2, 6%). 5 continents were represented, with North America having 30 papers (77%).  Methods of delivery were live (27, 69%) and live plus online (6, 15%), live plus written (4, 10%), online only (2, 5%).  Method of teaching was of 9 categories, the most common was simulation plus didactics (10, 26%), then simulation cases only (9, 23%), case-based non-simulation only (5, 12.8%), written materials followed by CBL (4, 10%).  19 (49%) of studies listed CBL as part of the course.  Learners were students to post-graduates.   Student numbers were 3-413, 14 papers (36%) reporting on practitioners, 25 (64%) were in-training or students.
Primary assessments were categorized into 5 types:  written tests, 13 (33%),observed skills clinical exam OSCE), 10 (26%), survey of learners, 6 (15%), review of practice behaviors, 4(10%), review of patient outcomes, 3 (7.7%).   Post course learner’s survey was used in 24 (62%), this was the sole course evaluation in 6 (15%).  
 Learners who were still in training for their specialty had more written tests (13 studies vs. 0 in practitioner level courses)  as primary course evaluations, and fewer review of patient outcomes ( 0 in trainees vs. 3 in practitioners), (p=.008). In training studies used OSCE (6) more often than practitioner studies (4).

 

Conclusion: CBL is used throughout the world to teach critical Care topics to students in training and practitioners.  Simulation is the most common way to deliver CBL. Evaluations differ significantly depending on learner education level. .

 

87.19 Social Indices as Potential Measures of Patient’s Health-Related Quality of Life

Posted on January 26, 2017

S. Yi1,3, S. Mukhopadhyay2,3  1George Washington University,Washington, DC, USA 2University Of Connecticut,General Surgery,Storrs, CT, USA 3Harvard School Of Medicine,Program In Global Surgery And Social Change,Brookline, MA, USA

Introduction:  Traditional surgical outcomes have been measured in clinical terms such as mortality, margin of resection, or infection rates. More holistic measures such as health-related quality of life (HRQoL) have become increasingly important to capture less tangible outcomes of quality health care.

At a broader level, social indices attempt to evaluate quality of life across multiple domains. These comprehensive summary measures may better reflect effects of health outcomes (e.g. economic productivity or increased happiness). Despite the importance of these indices in social analyses, their use is poorly understood in health system prioritization. We examine these indices and assess their feasibility of use to evaluate growing surgical health systems.

 

Methods:  We compiled leading indices from the Social Progress Imperative (Social Progress Index, SPI), UN Sustainable Development Solutions Network (World Happiness Index, WHI), UN Development Programme (Human Development Index, HDI), Organisation for Economic Co-operation and Development (Better Life Index, BLI), World Economic Forum (Global Competitiveness Index, GCI), Oxford Poverty & Human Development Initiative (Multidimensional Poverty Index, MPI), Legatum Institute (Prosperity Index, PI), and The Economist Intelligence Unit (Where-to-be-born Index, WBI). Indicators only directly related to health and medical care were extracted (Table 1). Other related indicators such as access to clean water or sanitation were not included.

 

Results: All eight indices included at least one measure of health. All indices but the MPI used life expectancy as a key health indicator, but differed in when it was measured (SPI at 60 years of age, the BLI at 80, 65, 60, 40, and at birth; all others measured at birth only). Three (WHI, HDI, WBI) used life expectancy at birth as the only health indicator. Five indices used infant mortality rates, while only two used maternal mortality rates, as well. Three cited undernourishment as a health indicator.

 

Conclusions: All examined social indices include an indicator for measuring health. Life expectancy is the most common indicator, with only the MPI excluding it. The MPI instead uses child mortality rate and undernourishment. The index with the most numerous health indicators is the PI. Interestingly, other common measures of quality health systems, such as access to timely care, workforce density, or complication rates, have not been included in these social indices. As surgery becomes a more integral component of developing health systems, it is important to utilize effective health indicators that reflect the quality of care and health that individuals experience.

87.18 Upper Gastrointestinal Endoscopy: Indications and Findings in Kumasi, Ghana

Posted on January 26, 2017

C. Dally1, J. Valenzuela2, A. Merchant2, O. Gunter3  1Komfo Anokye Teaching Hospital,Surgery,Kumasi, ASHANTI, Ghana 2University Of Cincinnati,Surgery,Cincinnati, OHIO, USA 3Vanderbilt University Medical Center,Surgery,Nashville, TN, USA

Introduction:
Abdominal pain remains a large proportion of presenting complaints when seeking medical assistance. Comprehensive patient information is scarce in Africa to follow these patients and characterize epidemiology, pathology, and treatment. Endoscopy remains vital to the diagnosis and treatment of upper gastrointestinal diseases including peptic ulcer disease (PUD), gastrointestinal reflux disease (GERD), gastrointestinal tumors and causes of bleeding. Direct visualization along with photography and biopsy can lead to definitive diagnosis and collaborative efforts for treatment. This study looks at presenting complaints of patients who underwent endoscopy in Kumasi, Ghana and correlates them with endoscopic findings, CLO test, and need for biopsy.

Methods:
Between 2012 and 2015, 1077 endoscopies were performed by a single-operator at three institutions in Kumasi, Ghana.  No image-capturing ability was available. A retrospective review was completed that evaluated age, sex, presenting complaint, endoscopic findings, CLO test for H. Pylori, and pathology reports when available. 

Results:
The average patient age for endoscopy was 44.5 years.  Sixty percent of endoscopies were performed on woman and 40% on men.  Forty percent of the clinical complaints leading to endoscopy were abdominal/epigastric pain.  Another 37% of patients were thought to have gastritis, peptic ulcer disease, or gastrointestinal reflux disease.  Eighty-nine percent of patients underwent CLO test; 56% of those patients had a positive CLO test indicating H. Pylori infection.  The remaining 11% could not undergo CLO testing due to lack of testing strips. Thirty-eight percent of endoscopic findings included gastritis.  Of 1065 patient records, 12.4% had lesions or ulcers biopsied. 

Conclusion:
The overwhelming findings show gastritis from endoscopy. The incidence of H. Pylori remains high.  Current patient data does not indicate whether patients received a proton-pump inhibitor trial prior to endoscopy and complete biopsy results are not available, sometimes due to affordability. Improved documentation and a database can further characterize epidemiology of abdominal pain in developing countries. One major limitation in endoscopy reports is the lack of pictures to accompany the descriptive report.  Image-capturing ability can confirm endoscopic findings, assist the referring physician with treatment plan and patient education, as well as lead to multi-disciplinary efforts in treatment.
 

87.17 Neonatal Surgical Outcomes In A Tertiary Care Center In India

Posted on January 26, 2017

A. Ranjit2, R. Shrestha1, A. Prasad3, K. P. Devkota1, R. Kulshrestha3  1Nepal Medical College And Teaching Hospital,Surgery,Kathmandu, BAGMATI, Nepal 2Brigham And Women’s Hospital,Center For Surgery And Public Health,Boston, MA, USA 3Sir Ganga Ram Hospital,Pediatric Surgery,Delhi, NEW DELHI, India

Introduction:

Low and Middle-Income countries (LMIC) bear the majority of the global pediatric surgical burden. Despite increasing volumes of pediatric surgery being performed in LMIC, outcomes of these surgeries in LMIC remains unknown due to lack of robust data. To that end, the objective of our study was to collect data on and evaluate neonatal surgical outcomes at a tertiary level center in India.

Methods:

The surgical outcomes data of all neonates undergoing major surgical procedure between February 1, 2015 and December 31, 2015, at Sir Ganga Ram Hospital, a tertiary level center in New Delhi, India was collected prospectively. Patient demographics, preoperative clinical characteristics and postoperative outcomes including surgical site infection (SSI), sepsis, length of stay (LOS), number of ventilator days and in-hospital mortality were recorded. Descriptive statistics were used to determine the rates of various postoperative outcomes.

Results:
A total of 37 neonatal surgeries were performed during the study period. The mean age of the neonates on the day of surgery was 7 days (range: 1-30 days). Most of the neonates (72.9%, n=27) were males. About 40% (n=15) of the neonates were preterm and 15 (40.5%) of them were small for gestational age. The average LOS was 22 days (range: 2-80 days). In our study,10 neonates (28.6%) needed ventilation for 48 hours or less after surgery and 5 neonates (13.5%) were kept postoperative Nil per Oral (NPO) for more than 10 days. Out of 37 neonates, 4 (10.80%) developed a surgical site infection and 8 neonates (21.6%) had postoperative sepsis. The in-hospital mortality rate among neonates undergoing a surgical procedure during the study period was 8.1 deaths per 100 neonates.

Conclusion:

Our study demonstrates the feasibility of data collection to study neonatal surgical outcomes in low and middle-income countries and might help encourage other centers in LMIC to conduct pediatric surgical outcome research.

 

 

87.16 Effect of Bariatric Surgery on Cardiovascular Risk Factors: Single Institution Retrospective Study

Posted on January 26, 2017

M. A. Al Suhaibani1, A. Al Harbi1, A. Alshehri2, E. Alghamdi2, R. Aljohani2, S. Elmorsy3, J. Alshehri3, A. Almontashery3  1Qassim University,Medicine,Qassim,Saudi Arabia 2Umm-alqura University,Makkah,Saudi Arabia 3King Abdullah Medical City,Makkah,Saudi Arabia

Introduction:
Bariatric surgeries became a trend for obese individuals with or without comorbidities. It can result in remission and control of cardiovascular risk factors. We aim to describe bariatric surgeries effect on cardiovascular risk factors including glycemic control, blood pressure, lipid profile and body weight at King Abdullah Medical City (KAMC) at Makkah, Saudi Arabia. 

Methods:
Retrospective cohort study including all obese patients who underwent bariatric surgery at KAMC between 2013 to February 2016 . 

Results:
Total of 566 patients, 58.1% (n=329) were female with mean age 34.8±10.2 years and 26.9% (n=152) were smokers. Almost all of them underwent laparoscopic gastric sleeve (95.9%, n=543). Diabetes and hypertension were on the top of comorbidities (24.8%, n=118 and 18.9%, n=107 respectively). After 24 months of follow up there was significant reduction in the baseline mean of Body Mass Index (BMI) (47.6 kg/m2) and Glycosylated Hemoglobin (HbA1c) (6.8%) to 31.8 kg/m2 and 5.7% respectively (p<0.001). Baseline High Density Lipoprotein (HDL) (44.4mg/dl), non-HDL cholesterol (147.1 mg/dl) and Triglyceride (TAG) (123.6 mg/dl) also showed significant improvement to reach 52.3 mg/dl, 137.7 mg/dl and 78.5 mg/dl at 12 months of post-operative follow up (p<0.001).Out of 79 hypertensive patients, 57 patients had remission of hypertension in last follow up (p<0.001). Among 103 pre-operative diabetic patient 71 (68.9%) had complete remission, and 17 (16.6%) had partial remission in last follow up (p<0.001).The predicted 10-years risk of CVD decreased significantly from 7.4% to 5.5% (p<0.001).

Conclusion:
Bariatric surgeries were significantly effective for controlling obesity related cardiovascular risk factors and decrease its 10-years predicted risk within a maximum 2 years period of follow up.
 

87.15 Intraoperative Perfusion Assessment of High-Risk Amputation Stumps Predict Area of Necrotic Eschar

Posted on January 26, 2017

G. S. De Silva1, K. Saffaf1, L. A. Sanchez1, M. A. Zayed1,2  1Washington University In St. Louis,Department Of General Surgery/Division Of Vascular Surgery,St. Louis, MISSOURI, USA 2Veterans Affairs St. Louis Health Care System,St. Louis, MISSOURI, USA

Introduction:  More than 130,000 extremity amputations are performed in the U.S. per year. In 40% of patients, poor amputation site healing requires stump revision and/or re-amputation.  This contributes to added patient morbidity, disability, and healthcare costs.  We hypothesize that inadequate tissue perfusion is associated with poor amputation stump healing.  We evaluated this using non-invasive Laser-Assisted Fluorescent Angiography (LAFA; SPY Elite® system) in the peri-operative setting.

 

Materials and

Methods:  A pilot group of ‘higher-risk’ patients were evaluated prospectively at the time of major lower extremity amputation.  Immediately following stump creation, LAFA was intra-operatively performed. Rate of arterial inflow and peak perfusion were determined using densitometry analysis.  Post-operative stump healing was serially evaluated for 4-6 weeks using a modified Bates-Jensen Wound Assessment Tool. Non-parametric Spearman correlation analysis was performed to evaluate stump perfusion and healing variables.

 

Results:  In a cohort of 8 patients (100% smoking, 75% diabetic), the least globally well-perfused stumps had the highest necrotic eschar scores (p=0.04), as well as increased volume of eschar (p=0.05).  Similarly, amputation stumps with lower perfusion scores just along the surgical suture line were more likely to also develop a necrotic eschar (R2=0.834, p<0.05), and increased eschar volume (R2=0.842, p<0.05).  We observed no correlation between low stump perfusion scores and higher iliac and common femoral arterial runoff scores (major arterial occlusions).

 

Conclusions:  In ‘higher-risk’ patients, peri-operative perfusion assessments of amputation stumps using LAFA can help predict potential areas of necrotic eschar formation.  Intra-operative determination of areas of decreased amputation stump perfusion may encourage corrective intervention or anticipate subsequent wound care needs.

 

87.13 Sarcopenia Predicts Mortality of Trauma Patients Requiring Intensive Care

Posted on January 26, 2017

N. A. Lee1, A. Khetarpal3, L. Wolfe1, S. Demasi2, A. Stiles2, M. Aboutanos1, P. Ferrada1  1Virginia Commonwealth University,Department Of Surgery,Richmond, VA, USA 2Virginia Commonwealth University,School Of Medicine,Richmond, VA, USA 3Virginia Commonwealth University,Department Of Radiology,Richmond, VA, USA

Introduction:
As the population of the United States ages, there has been a disproportionately larger increase in the amount of elderly trauma patients. Elderly patients have worse outcomes when controlling for injury severity independent of age. Frailty, a syndrome characterized by increased vulnerability to stressors leading to functional impairment and adverse outcomes, has been found to negatively affect the outcomes of surgical patients and critically ill patients, but evaluating frailty in trauma patients has proven difficult. Sarcopenia, or decreased muscle mass and function, is a measurable factor associated with frailty. We hypothesize that sarcopenia can be used as a surrogate for frailty to independently predict risk of mortality.

Methods:
In this retrospective cohort study, trauma patients aged 45 years or greater requiring ICU care at our level-1 trauma center between April 2015 and January 2016 were assessed for sarcopenia by averaging the psoas muscle body cross-sectional area at the level of the L4 pedicles, and dividing by the cross-sectional area of the L4 body at the same level. Patients were excluded if there was traumatic injury in the L4 area, or if no CT was obtained. This psoas:lumbar vertebral index (PLVI) was then cross-referenced with data from the local trauma registry including outcomes, comorbidities, complications, and injury severity scores. Our primary outcome was in-hospital mortality. Statistical analyses including Wilcoxon rank sum test and stepwise logistic regression was performed with SAS 9.4 with a significance level of 0.05.

Results:
Over the study period, 715 patients were identified, of which 528 were eligible and assessed for sarcopenia by calculating the PLVI (median 0.95, range 0.34 – 1.91). There were 41 deaths in the study population. Patients who died had smaller PLVI (0.79, 0.40 – 1.28 vs 0.96, 0.34 – 1.91, p=0.0014), were older (72, 47 – 92 vs 62, 45 – 98, p < 0.0001), and had higher injury severity scores (26, 1 – 45 vs 14, 1 – 75, p<0.0001). Stepwise logistic regression was performed with comorbidities and complications which were approaching significance on univariate analysis. With these factors included, age was not a significant predictor of mortality; however, ISS and PLVI were included as significant contributors to mortality in addition to history of renal failure, malignancy, and prehospital DNR status, and new onset stroke, and renal failure. The odds ratio of 0.135 (0.029 – 0.623) demonstrates a strong association with lower PLVI and mortality.

Conclusion:
Sarcopenia as measured by PLVI is an independent predictor of mortality in trauma patients older than 45 requiring critical care. This may provide an opportunity to further risk-stratify this high-risk patient population on admission.

87.12 Survival After Massive Transfusion for Trauma: The Type of Injury Matters.

Posted on January 26, 2017

D. R. Fraser1, A. Snow1, C. F. McNicoll1, P. J. Chestovich1, A. J. Chapman1, J. J. Fildes1  1University Of Nevada,Acute Care Surgery,Las Vegas, NV, USA

Introduction:

Trauma patients often receive large volume blood transfusions for life-threatening bleeding. Despite routine use of Massive Transfusions (MT) (≥ 10 units of packed red blood cells [PRBC]) in devastating injuries, the threshold for transfusion futility has yet to be established. When a trauma patient receives ≥ 10 units of PRBC, the surgeon must shepherd the continued resuscitative use of PRBC judiciously. Injury patterns may predict survival in MT patients, and guide blood product stewardship.

 

Methods:

A retrospective review of our level 1 trauma center’s registry was conducted to identify all trauma patients that received at least one blood product between June 2010 and June 2015. All PRBC, fresh frozen plasma (FFP), cryoprecipitate, and platelets transfused during the first 24 hours of admission were tabulated. Primary outcome was 30-day survival, by Abbreviated Injury Scale (AIS) body region, for patients who received ≥ 10 units of PRBC. Secondary outcomes included 30-day survival of patients who received ≥ 15 units of PRBC, by AIS body region. Stepwise reverse logistic regression was performed in STATA version 11, with statistical significance of p<0.05.

 

Results:

We identified 435 patients that received at least one blood product during the study period, with 75.4% men, mean age of 42.8 years, and 72.4% blunt injuries. Of these, 116 (26.7%) patients received a MT, and 319 (73.3%) did not. The range of all blood products transfused was 1 to 203 units. The range for the 415 patients that received any PRBC was 1 to 80 units. The survivor with the largest PRBC transfusion received 57 units. Percent 30-day survival by PRBC quantity transfused was 65.9% (n=299), 44.0% (n=91), 40.0% (n=20) and 20.0% (n=5) for groups of 1-9, 10-25, 26-50, and >50 units of PRBC transfused, respectively. The odds ratio (OR) for 30-day survival in MT patients compared to non-MT patients was 0.92 (p=0.01), after accounting for age, FFP:PRBC ratio, platelets, cryoprecipitate, and AIS body regions. Extremity (OR 4.98, p=0.002) and abdominal (OR 4.63, p=0.008) injuries correlated with improved survival in MT patients (Table 1). Chest injuries were associated with worse 30-day survival in MT patients, though not significant (OR 0.35, p=0.07). These effects persisted for PRBC transfusions ≥ 15 units.

 

Conclusion:

Survival in trauma patients requiring PRBC transfusion worsens with increasing volume. In MT patients, abdominal and extremity injuries had improved 30-day survival, while no significant difference could be found for other body regions. These findings will inform the surgeon’s decision making process for trauma patients requiring MT, and potentially improve the utilization of blood bank resources.

 

87.11 Epidural Anaesthesia For Traumatic Rib Fractures Is Associated With Worse Outcome: A Matched Analysis

Posted on January 26, 2017

K. M. McKendy1, K. Boulva1, L. Lee1, A. N. Beckett1, D. L. Deckelbaum1, P. Fata1, K. A. Khwaja1, D. S. Mulder2, T. S. Razek1, J. R. Grushka1  1McGill University,General And Trauma Surgery,Montreal, QC, Canada 2McGill University,Cardiothoracic Surgery,Montreal, QC, Canada

Introduction: To determine the effect of epidural anaesthesia on the incidence of respiratory complications and in-hospital mortality in adult patients with rib fractures after blunt trauma.

 

Methods: All adult patients presenting at a university-affiliated level I trauma center from 2004 to 2013 with at least one rib fracture secondary to blunt trauma were queried from a prospectively entered database. Patients who had a combined blunt-penetrating mechanism of injury, simultaneous intracranial haemorrhage or traumatic brain injury, or underwent a laparotomy or thoracotomy were excluded from the analysis. Epidurals were placed within the initial 24 hours of presentation according to the treating physician’s preferences. Main outcome measures were respiratory complications (pneumonia, DVT/PE, and respiratory failure) and 30-day in-hospital mortality. Coarsened exact matching was used to account for differences in patient-level factors (age, sex, Injury Severity Score [ISS], number of rib fractures, flail segment, bilateral rib fractures, chest tube insertion, pulmonary contusion, and year of injury) between those who received epidural anaesthesia (EPI) and those who did not (NEPI) in a one-to-one fashion. Subgroup analyses were performed based on age (≥65 and < 65 years), number of rib fractures (≥3 and <3 fractures), burden of trauma (ISS ≥16 and ISS < 16), and bilaterality of fractures. Statistical significance was defined as p<0.05.

 

Results: A total of 1360 (EPI 329, NEPI 1031) patients met the inclusion criteria (mean age 54.2 years (SD19.7), 68% male). Overall, the mean number of rib fractures was 4.8 (SD3.3) ribs (21% bilateral) with a high total burden of injury (mean ISS 19.9 (SD8.9)). The incidence of respiratory complications was 13% (180/1360) and mortality 4% (53/1360) in the unmatched cohort. After matching, 204 EPI patients were compared to 204 NEPI patients with no differences in demographics and traumatic characteristics (. In matched analysis, EPI patients experienced more respiratory complications (19% vs. 10%, p=0.009) but no differences in 30-day mortality (5% vs. 2%, p=0.159) compared to NEPI patients. There were also no differences in duration of mechanical ventilation (EPI 148 h (SD167) vs. NEPI 117 (SD187), p=0.434) and intensive care unit length of stay (EPI 6.5 d (SD7.6) vs. NEPI 5.8 d (SD9.1), p=0.626). Total length of stay was higher in the EPI group (16.6 d (SD19.6) vs. 12.7 d (SD15.2), p=0.026). This relationship remained unchanged in all subgroup analyses.

 

Conclusion: Epidural anaesthesia is associated with increased respiratory complications and mortality after traumatic rib fractures. Alternate analgesic strategies and rib fracture fixation should be further investigated to treat these severely injured patients.

 

 

 

 

 

87.10 IS CT ALONE SUFFICIENT TO EXCLUDE CLINICALLY RELEVANT CERVICAL SPINE INJURY?

Posted on January 26, 2017

J. S. Hanna2, V. Sim2, C. King1, A. Rayner3, R. Gupta2  1Rutgers Robert Wood Johnson Medical School,Department Of Radiology,New Brunswick, NJ, USA 2Rutgers Robert Wood Johnson Medical School,Division Of Acute Care Surgery,New Brunswick, NJ, USA 3Rutgers Robert Wood Johnson Medical School,Department Of Surgery,New Brusnwick, NJ, USA

Introduction:

Missed cervical spine injuries may result in devastating morbidity and mortality. Debate continues over the optimal method for identifying clinically relevant cervical spine injuries. The adequacy of computed tomography (CT) alone and the role of magnetic resonance imaging (MRI) in assessing ligamentous injury remains controversial. We hypothesize that CT scan alone is neither 100% sensitive nor specific because of equipment and provider heterogeneity.

 

Methods:

A prospectively maintained database at a level one trauma center was queried for all trauma patients with a negative cervical spine CT scan who required an MRI for persistent midline cervical tenderness (MCT) or altered mental status (AMS) between 2011 and 2016. 838 patients were identified for which a retrospective chart review was performed to identify admission characteristics, imaging findings, and clinical management.

 

Results:

The identified cohort was composed of 649 patients with persistent MCT and 189 patients with AMS. MRI identified clinically relevant injuries not seen on CT in 5% of patients with MCT and AMS each. In the MCT group 4% were managed with a cervical collar and 1% required surgery. In the AMS group 4.8% were managed with a cervical collar and 0.2% with surgery. In the absence of a cervical spine fracture, the sensitivity and specificity of CT alone in detecting ligamentous injury in this cohort is 2.3% and 100% respectively.

 

Conclusion:

These data suggest that a negative CT alone is insufficient to exclude injuries resulting in cervical spine instability. Although recent studies have suggested CT scan is 100% sensitive and specific, we believe that equipment and interpreter heterogeneity substantially decrease the sensitivity. We recommend that each institution perform an internal validation prior to developing protocols which rely solely on CT imaging to exclude clinically relevant cervical spine injuries.

87.09 Burn Injury in Diabetics Leads to Significant Increases in Hospital-Acquired Infection and Mortality

Posted on January 26, 2017

F. N. Williams1, P. Strassle2, S. Jones1, B. Cairns1  1UNC,Surgery/Burns,Chapel Hill, NC, USA 2UNC,Surgery,Chapel Hill, NC, USA

Introduction: Outcomes in burn patients, including mortality, are affected by age of the patient and the extent of the burn injury. This is reflected in the revised Baux score, which is calculated by adding age plus percent total body surface area burned plus 17 if there is an inhalational injury. The result corresponds to a predicted mortality. Tight glucose control following burn injury improves complication and mortality rates. It remains unclear whether a pre-existing diagnosis of diabetes in burn patients influences key outcomes such as infectious complications and mortality.

Methods: The Burn Center registry, Hospital Epidemiology database, electronic medical records, and billing data were linked. Adult patients (≥18 years old) admitted between January 1, 2004 and December 31, 2013 with a burn injury were included. Only the first hospitalization within this time frame was included. Diabetes mellitus was identified using both comorbidities listed on the burn registry and d diagnostic codes attached to the inpatient hospitalization (ICD-9-CM 250). Multivariable Cox proportional hazard models were used to estimate the increased risk of diabetes on 60-day mortality and hospital-acquired infections, after adjusting for patient and burn characteristics. Only patients hospitalized for 2 or more days (i.e. at risk for infection as per CDC definitions) were included in HAI analyses.

Results: 5,539 patients met the inclusion criteria. 665 (11.8%) had a diabetes mellitus (DM) diagnosis. Diabetic patients were significantly more likely to be female (34.1% vs. 26.6%, p<0.0001), African American (36.9% vs. 26.0%, p<0.0001), and older (median age 56.7 years old vs. 39.9 years old, p<0.0001). Diabetic patients were more likely to have contact burns (8.9% vs. 4.7%, p=<0.0001) and inhalational injury (11.0% vs. 8.1%, p=0.01). No differences were seen in median burn size (4.0% vs. 4.0%, p=0.44). The median revised Baux scores was also higher among diabetics (64.0 vs. 47.6), p<0.0001. Patients with DM were more likely to be admitted to the ICU (14.3% vs. 10.6%, p<0.0001) and were hospitalized for a median 11 days (interquartile range [IQR] 4 – 26), compared to a median 7 days (IQR 2 – 13) for patients without diabetes, p<0.0001. Only 242 patients (4.4%) were hospitalized longer than 60 days and administratively censored prior to discharge or death. Overall, 243 (4.4%) died during their inpatient hospitalization. After 30 days, diabetic patients had a higher mortality risk (RD 0.03, 95% CI 0.00, 0.05) compared to non-diabetic patients, and after 60 days the risk was higher (RD 0.07, 95% CI 0.01, 0.12). Patients with DM were significantly more likely to have an HAI after 60 days.

Conclusion: Comorbid conditions can lead to worse outcomes. Diagnosed diabetics fare worse after burn injury than matched non-diabetics.

 

« Previous Page
Next Page »