58.11 Preoperative Frailty Correlates with Postoperative Outcomes in Major Abdominal and Thoracic Surgery

Posted on January 31, 2019

M. M. Mrdutt1, B. Robinson1, E. Bird1, H. Papaconstantinou1, C. Isbell1  1Baylor Scott & White Medical Center-Temple,Department Of Surgery,Temple, TEXAS, USA

Introduction:   Frailty is a measure of physiologic reserve and an emerging metric for risk stratification.  Ideally frailty assessment would be objective, easily administered  during a preoperative assessment, and offer timely information with regard to postoperative complication risk.  We examined the correlation of frailty with postoperative outcomes following major abdominal and thoracic surgery to determine its potential as a risk-assessment tool.

Methods:   Frailty was prospectively measured at a single institution in all elective surgery patients using the Modified Hopkins Frailty Score.  Frailty classification—low (0), intermediate (1-2), or high (≥3)—was calculated based on shrinking (unintentional recent weight loss 10 pounds or greater in the last year), handgrip strength, hemoglobin, and ASA classification.  Demographics and surgical outcomes were obtained from institutional procedure-targeted National Surgical Quality Improvement Program (NSQIP) data for major abdominal (esophagectomy, pancreatectomy, hepatectomy, colectomy, proctectomy, nephrectomy, cystectomy) and thoracic (pulmonary resection) procedures January 2016-June 2017.  Outcomes included any NSQIP complication, readmission, unplanned reoperation or mortality within 30 days, along with discharge location and prolonged length of stay (> 75th percentile of 2015 NSQIP national average).  Logistic regression was performed accounting for demographics, comorbidities and frailty for each complication.   

Results:  Of the 548 patients identified, 426 (77.8%) had a complete frailty evaluation.  Cases included pulmonary resections (17.3%) and abdominal procedures (82.7%); the majority of patients were classified as intermediate (76.5%) or high (10.3%) frailty.  Incidence of any NSQIP complication, readmission and discharge to a facility correlated with frailty classification in univariate analysis (Figure, p<0.05).  Logistic regression demonstrated an increased odds ratio (OR) with increasing frailty for any NSQIP complication (intermediate OR 3.6, high OR 8.9), readmission (intermediate OR 7.5, high OR 15.3) and discharge to facility (intermediate OR 3.3, high OR 10).  There was no significant association with unplanned reoperation, prolonged LOS or mortality. 

Conclusion:  Higher frailty correlates with increased postoperative complications, readmission and discharge location.  These findings provide external validation for the Modified Hopkins Score in major non-cardiac elective cases and provide a link between postoperative morbidity and preoperative frailty.  Preoperative frailty assessment should contribute to perioperative patient optimization and care strategies, specifically in high risk procedures such as those captured by NSQIP procedure targeted cases. 

58.10 Social Determinants of Falls: Are falls a disease of poor white females?

Posted on January 31, 2019

K. Sairafian2, C. Towe2,5, L. Brown1,2, L. A. Kreiner1,2, M. Crandall4, E. R. Haut3, V. P. Ho1,2  3Johns Hopkins University School Of Medicine,Baltimore, MD, USA 4University of Florida College of Medicine Jacksonville,Jacksonville, FLORIDA, USA 5University Hospitals,Thoracic And Esophageal Surgery,Cleveland, OHIO, USA 1MetroHealth Medical Center,Cleveland, OH, USA 2Case Western Reserve University School Of Medicine,Cleveland, OH, USA

Introduction:  

Falls are a leading cause of morbidity and mortality in the elderly.  It is unknown if there are racial or other disparities associated with falling. We sought to determine which social and/or demographic variables are associated with falls in the outpatient Medicare population.

Methods:

We examined data from the 2013 Medicare Current Beneficiary Survey Public Use File, a representatively sampled cross-sectional survey.  Fall was defined as at least one self-reported fall in the preceding year.  We performed a logistic regression, adjusted for survey data characteristics, to determine social/demographic factors (age, sex, race, ethnicity, income, education level, and marital status) associated with fall. In these data, patients <65 have more chronic disease than the general population to meet Medicare eligibility. Presence of physical or cognitive limitations were included in the analysis as possible confounders.  Data are presented as percent (± standard error). Adjusted odds ratios are presented with 95% confidence intervals.

Results:

13,924 Medicare beneficiaries, representing 47 million people, were included.  26.6% (±0.4) reported falling. Females, patients <65 or >74, and patients with physical/cognitive limitations were more likely to report a fall (Table).  Minority patients and males had significantly fewer self-reported falls than white patients and females (See Table, p<0.001 for each).  Low income patients (OR 1.17 [1.04-1.33]) were also significantly more likely to report a fall.

Conclusions:

Black and Hispanic Medicare patients are significantly less likely to have reported a fall than white patients. This finding differs from other health-related disparities in which minorities most commonly experience higher risk or more severe diseases.  These data may also represent differences in self-reporting, indicating disparities in self-reported data in these cohorts. Further studies on social factors related to falling are needed in this population.

58.09 Age-Related Microbiome Differences in Surgical Site Infections

Posted on January 31, 2019

R. Khatri2, T. L. Hedrick1, K. A. Popovsky1, R. G. Sawyer1,2  1University Of Virginia,Charlottesville, VA, USA 2Western Michigan University School of Medicine,Kalamazoo, MICHIGAN, USA

Introduction: Surgical site infections (SSI) contribute to overall morbidity and mortality of the surgical patient. There are numerous risk factors for the development of SSI. We propose that there are variations in the type of pathogen in SSI between different age groups.

Methods: We evaluated patients who underwent surgery at a university hospital from 1997 – 2017 that developed SSI. Patients were divided into age-related cohorts, age ≤ 45 years and age ≥65 years. We assessed immune response, disease severity, types of pathogens within wounds, mortality, and antibiotic therapy. Proportions were compared through use of chi square test and continuous variables were compared using Student’s t test.

Results: As compared to patients age ≥65 years who developed SSI (n=575), patients in age group ≤45 years (n=598) had increased leukocytosis (14.2K vs 13.4K, p=0.06), presence of fever (30.8% vs 20.9%, p<0.0001), and Tmax (37.9oC vs 37.6oC, p<0.0001) at time of diagnosis. Of those infections with positive growth on culture, patients age ≥65 years had higher percentage of Enterococcus spp (27.4% vs 17.5%; p=0.006) and P.aeruginosa (15.8% vs 9.6%; p=0.03) as primary wound pathogen as compared to those patients age ≤45 years. In addition, those patients age ≥65 had greater crude mortality rates (10.6% vs 2.7%; p<0.001). Those in age group ≤45 years had longer overall antibiotic treatment length (12.4 days vs 10.3 days, p<0.001) but were more likely to be prescribed oral antibiotics on discharge (51.3% vs 34.4%, p<0.0001).

Conclusion: Our data suggest that those in the older age group have less robust immune response and are colonized with more aggressive pathogens, which may require inpatient treatment with parenteral antibiotics. These findings may help tailor perioperative antibiotic prophylaxis according to age group and most common pathogens, and help prevent or improve overall morbidity from SSI.

 

 

58.08 Management of Blunt Kidney Injuries in Geriatric versus Non-Geriatric Trauma Patients

Posted on January 31, 2019

J. Saluck1,2, M. Crawford2, A. A. Fokin2, A. Tymchak1,2,3, J. Wycech2,3, M. Gomez3, I. Puente1,2,3,4  1Florida Atlantic University,College Of Medicine,Boca Raton, FL, USA 2Delray Medical Center,Trauma Services,Delray Beach, FL, USA 3Broward Health Medical Center,Trauma Services,Fort Lauderdale, FL, USA 4Florida International University,College Of Medicine,Miami, FL, USA

Introduction:
Nonoperative management (NOM) is the standard of care in majority of blunt kidney injuries (BKI). However, little is known about the effect of age on outcomes of NOM. The aim of this study was to assess the efficacy of NOM in BKI in geriatric vs non-geriatric patients.

Methods:
This IRB approved retrospective cohort study included 86 adult patients, admitted to a level 1 trauma center between 2012 and 2017 with computed tomography (CT) confirmed kidney injuries. 49 patients had isolated BKI, 25 BKI with other solid organ injuries, 9 BKI plus hollow viscus injury and 3 BKI plus abdominal vascular injuries. Patients were separated into 2 groups: Group I ≥65 years old (n=21) and Group II <65 years old (n=65). Age, Injury Severity Score (ISS), kidney organ injury scale (OIS) grade, comorbidities, packed red blood cells transfused within 24 hours of admission (PRBC24), rates of hemoperitoneum, angiography, embolization, Intensive Care Unit length of stay (ICULOS), hospital length of stay (HLOS) and mortality were compared.

Results:

Mean age in Group I was 82.2 vs 36.8 years in Group II (p<0.001). The two groups had similar mean ISS (19.0 vs 22.4; p=0.4) and the same mean OIS grade of 2.3 (p=0.8). Group I had statistically more comorbidities than Group II (90.5% vs 60.0%; p=0.01). Main comorbidities in Group I included hypertension, anticoagulation therapy prior to trauma (p<0.001), and cardiovascular disease. Main comorbidities in Group II included hypertension, obesity, and substance abuse.

NOM was attempted in 100% of Group I patients and in 76.9% of Group II patients (p=0.02). The frequency of attempted NOM was lower in Group II due to higher prevalence of other abdominal injuries. For attempted NOM patients in Groups I and II, mean ISS (19.0 vs 18.1, p=0.9) and OIS grade (2.3 vs 2.1, p=0.7) were similar, which adds to the comparability of the groups. Attempted NOM was successful in 100% of Group I patients and in 93.8% of Group II (p=0.3). Of the 3 patients that failed NOM in Group II, 2 patients failed due to a liver injury and 1 patient due to a spleen injury.

For both groups rates of PRBC24 (33.3% vs 41.5%), hemoperitoneum (52.4% vs 61.5%), angiography (9.5% vs 18.5%), embolization (4.8% vs 4.6%), ICULOS (5.9 vs 7.4 days) and HLOS (8.1 vs 10.5 days) were similar (p>0.4). Group I tended to have higher mortality than Group II (19.0% vs 10.8%; p=0.3) but it did not reach statistical significance. Concomitant traumatic brain injury was the leading cause of mortalities in Groups I and II, at 50.0% and 57.1% respectively. None of the mortality in either group was due to the kidney injury.

Conclusion:
Even with more comorbidities, the advanced age was not a contraindication for NOM and did not affect the success of NOM in geriatric patients with BKI. The severity of kidney injury in both age groups was similar and did not affect the frequency of attempted NOM.

58.07 What are the Clinical Outcomes for Damage Control Laparotomy in Patients of Advanced Age?

Posted on January 31, 2019

A. A. Smith1, C. Guidry1, P. McGrew1, J. Friedman1, R. Schroll1, C. McGinness1, J. Duchesne1  1Tulane University,Surgery,New Orleans, LA, USA

Introduction:  Damage Control Laparotomy (DCL) is an integral component in the immediate management of critically ill trauma patients to control hemorrhage and intra-abdominal contamination. Patients of advanced age have less physiologic reserve and an altered response to traumatic injuries when compared to younger patients. As the population in the United States continues to age, the number of DCLs in patients of advanced age will ultimately increase.  There is a paucity of literature on outcomes for older patients managed with DCL. The objective of this study was to provide evidence for outcomes in older population who received DCL for trauma.

Methods:  A retrospective chart review of consecutive adult patients with DCL for abdominal trauma at a Level I trauma center was conducted from 2012-2017. The patients were stratified into two groups, advanced age (AA) for patients 40 years and older and younger age (YA) for patients less than 40 years of age.

Results: A total of 149 patients with DCLs were identified with an average age of 34.0 (range, 19-81 years). In regards to patient demographics, there was no difference in ISS (p=0.16), mechanism (p=0.44), and initial INR (p=1.0). The AA group did, however, have significantly lower ED SBP (p=0.01) and significantly higher initial fibrinogen (p<0.0001). When analyzing outcomes and interventions, AA patients received MTP more frequently (p=0.03). There was a trend toward increased mortality in the AA group (23% vs 11%) when compared to YA group, though this did not reach significance (p=0.08). Of significance, the AA group had an overall shorter time to mortality (4.5+0.4 vs 8.9+1.2 days, p=0.02).

Conclusion: With an aging population, it is likely that the number of DCLs in older patients will increase. AA patients managed with DCL had decreased initial ED SBP with more utilization of MTP resources and overall shorter time to mortality. Future research should emphasize strategies that will develop optimal management and resource utilization of older trauma patients.

 

58.06 Clinical Frailty Scores Predict Re-admission for Fall Following Trauma

Posted on January 31, 2019

V. H. Hatcher1, D. Skeete1, K. S. Romanowski2  1University Of Iowa,Surgery,Iowa City, IA, USA 2University Of California – Davis,Surgery,Sacramento, CA, USA

Introduction: Falls are a significant cause of morbidity and mortality in the elderly. Frailty scales have been developed, but most cannot be utilized in retrospective studies. The Canadian Study of Health and Aging clinical frailty scale (CSHA CFS) is a 7-point clinical opinion scale validated to predict mortality and institutionalization in elderly internal medicine patients. We hypothesize that patients with higher admission frailty will be admitted more frequently with falls post-index admission.

Methods:  Charts of patients ≥50 years of age admitted for traumatic injuries from 2010 to 2015 were reviewed. Demographics, admission data, and injury severity score (ISS) were collected. Frailty scores were calculated using the CSHA CFS. Statistical analyses were performed using R.

Results: Data were collected from 804 patients (70.3 ± 13.4 years), including 380 men (47.2%). Thirty patients (3.73%) died of their injuries. Frailty scores were similar between survivors and those who died. Mean ISS was 9.83 ± 7.92 and not different between frail (CSHA CFS 5-7) and non-frail patients (CSHA CFS 1-4). One hundred and sixteen (14.4%) patients previously presented with falls. Frailty scores of patients with a history of falls were higher than those of their counterparts (4.64 ± 1.0 vs. 3.85 ± 1.13; p < 0.001). Frailty scores of patients who were readmitted with a fall after traumatic injury were higher than those of their counterparts (4.49 ± 0.97 vs. 3.89 ± 1.15; p < 0.001). The number of falls in the year post-admission of frail patients was higher than that of non-frail patients (1.31 ± 0.74 vs. 0.92 ± 0.77; p < 0.001). On multivariate linear regression analysis, CSHA CFS predicted the number of falls in the year post-admission, while controlling for age and ISS (p < 0.001).

Conclusion: Frailty predicts fall readmission post-trauma and number of falls in the year following trauma admission, but does not influence mortality. 

 

58.05 Pre-Injury Cerebral Atrophy in Traumatic Brain Injury Patients Correlates with Decreased Mortality

Posted on January 31, 2019

S. Mansour1,2, A. A. Fokin2, A. Zuviv2, J. Wycech2,3, M. Crawford2, A. Tymchak1,2,3, I. Puente1,2,3,4  1Florida Atlantic University,College Of Medicine,Boca Raton, FL, USA 2Delray Medical Center,Trauma Services,Delray Beach, FL, USA 3Broward Health Medical Center,Trauma Services,Fort Lauderdale, FL, USA 4Florida Atlantic University,College Of Medicine,Boca Raton, FL, USA

Introduction:
Geriatric patients are at increased risk for falling and sustaining traumatic brain injuries (TBI) than their non-geriatric counterparts. It has been shown that elderly patients may have cerebral atrophy that may in turn affect outcomes of intracranial hemorrhage; however this claim has not been studied adequately in TBI patients. We hypothesized that patients with pre-injury cerebral atrophy would have lower morbidity associated with TBI due to increased volume available for hematoma expansion.

Methods:
This IRB approved retrospective cohort study included 346 TBI patients on pre-injury Aspirin, Clopidogrel, or both, between the ages of 17 and 101, who were delivered to a level 1 trauma center between 1/1/2015 and 3/30/2018. Patients were divided into 2 groups: Group A did not have cerebral atrophy (n=148) and Group B had cerebral atrophy documented on computed tomography (CT) reports (n=198). Patients were excluded if they were also on anti-coagulants. Age, Injury Severity score (ISS), Revised Trauma score (RTS), Glasgow Coma score (GCS), Rotterdam CT score (RCT), Marshall CT score (MCT), incidence of intracranial hemorrhage (ICH), midline shift, platelet function and status, platelet transfusion, need for neurosurgical intervention, Intensive Care Unit length of stay (ICULOS), hospital LOS (HLOS), rate of readmission and mortality were compared.

Results:

Between Groups A and B mean values for ISS (12.9 vs 12.3), RTS (7.6 vs 7.7), GCS (14.2 vs 14.3), incidence of ICH (85.5% vs 86.9%), platelet count (215.0 vs 209.8), Platelet Function Assay (PFA)-100 epinephrine (194.7 vs 188.0), PFA-100 adenosine diphosphate (ADP) (127.0 vs 160.0), Thromboelastography Platelet Mapping (TEG-PM) % inhibition ADP (39.4% vs 39.8%), TEG-PM % inhibition of arachidonic acid (60.1% vs 56.6%), Prothrombin Time (11.5 vs 11.0 seconds), Partial Thromboplastin Time (25.7 vs 26.5 seconds), platelet transfusion (37.2% vs 44.1%), neurosurgical intervention (4.7% vs 5.1%), ICULOS (3.5 vs 2.9 days), HLOS (3.8 vs 3.8 days), and readmission rate (5.5% vs 5.1%) were similar (all p>0.07).

Group B compared to Group A had significantly lower RCT (2.7 vs 2.5; p=0.002), MCT (1.2 vs 1.0; p=0.002), midline shift (12.8% vs 3.5%; p=0.001) and mortality (14.2% vs 7.6%; p=0.04) (Fig. 1).

Conclusion:
Cerebral atrophy was associated with less severe damage and lower mortality in head trauma, despite similar injury severity, when compared to patients with no atrophy. Our findings suggest that increased intracranial volume may allow TBI patients to accommodate hematoma expansion, alleviating clinical presentation and reducing the likelihood of adverse outcomes.

58.04 Sarcopenia and Frailty: Similar Yet Distinct Measurements in Geriatric Trauma and Emergency Surgery

Posted on January 31, 2019

H. K. Weiss1, M. Errea2, B. W. Stocker1, N. Weingarten1, K. E. Engelhardt3, B. Cook2, J. A. Posluszny2  1Feinberg School Of Medicine – Northwestern University,Chicago, IL, USA 2Northwestern University,Department Of Surgery,Chicago, IL, USA 3The Medical University of South Carolina,Department Of Surgery,Charleston, SC, USA

Introduction:   The diagnoses of sarcopenia and frailty are often intertwined—both identify less healthy patients with higher complication rates and worse outcomes. However, sarcopenia is a function of muscle density alone, while frailty measurements take into account a multidimensional understanding of frailty, including patient comorbidities, emotional health, nutrition, cognitive and physical function. We hypothesize that sarcopenia and frailty measure different forms of disability. The objective of our study is determine the concordance of sarcopenia and frailty in geriatric trauma and emergency general surgery (TEGS) patients.

Methods:   We reviewed our QI database of geriatric (≥65 year old) TEGS patients.  As part of this project, all geriatrics patients are screened for frailty using the validated TEGS frailty screening tool.  Patients in whom a CT of the abdomen and pelvis was obtained were identified and total psoas index (TPI) (area of right and left psoas muscle at L3/2) was calculated, with lower TPIs signifying sarcopenia.  Patients were then compared for frailty and sarcopenia.

Results:  117 geriatric TEGS patients were screened for frailty.  Of all patients, 78 (67%) had a CT of the abdomen and pelvis.  Of these 78 patients, 22 (28%) were frail.  Mean TPI for all patients was 2.34±0.77.  Mean TPI for non-frail patients was 2.34±0.81 and was 2.33±0.65 for frail patients (p=0.97).  Since 28% of our patients were frail, we then compared patients with the 28% (n=23) lowest TPI for frailty.  Only five (23%) of these patients were frail (p=0.59).  Using the lowest quintile of TPI (15 patients) to define sarcopenia, as is done in other studies, only 3 (20%) of these sarcopenic patients were frail (p=0.53).

Conclusion: Although both sarcopenia and frailty identify patients at higher risk for complications and worse surgical outcomes, sarcopenia does not specifically reflect frailty.  As such, frailty and sarcopenia are not concordant measures of illness. Further study will help elucidate the true relationship between frailty and sarcopenia and the clinical implications of sarcopenia for geriatric TEGS patients.

 

58.02 Frailty Severity Predicts Poor Outcome After First-time Lower Extremity Revascularization

Posted on January 31, 2019

L. Gonzalez1,2, M. Kassem1,2, A. Owora3, S. Cardounell1, M. Monita1, S. Brangman1, V. Gahtan1,2  1State University Of New York Upstate Medical University,Vascular And Endovascular Surgery,Syracuse, NY, USA 2Syracuse VA Medical Center,Surgery,Syracuse, NY, USA 3Syracuse University,Falk College School Of Public Health,Syracuse, NY, USA

Introduction:  Frailty severity is a predictor of poor outcome after vascular surgery. The modified frailty index (mFI) has been validated as a prognostic assessment tool in large scale databases of patients with peripheral arterial disease. Our objectives were to determine the predictive utility of the mFI after first-time lower extremity revascularization and to identify biomarkers of frailty in patients with peripheral arterial disease. Hypotheses: (1) frailty severity is associated with adverse outcome after revascularization and (2) select preoperative data may serve as biomarkers of frailty.

Methods:  A retrospective cohort study was performed of all first-time revascularizations [open surgery (OS) and endovascular surgery (ES)] in male veterans at a single institution (2003-2016). Multivariable logistic and Cox proportional hazard regression models were used to examine the relationship between the mFI and post-operative short-term (30-day morbidity, readmission, and re-intervention) and long-term (up to 2-year incidence of re-intervention, amputation, or mortality) outcomes, respectively. 

Results: 431 patients met inclusion criteria (OS n=188; ES n=243), with a mean age of 66±9 years and median follow up of 16 months. Treatment groups were similar in baseline characteristics, pre-operative lab values, and polypharmacy tallies. Mean mFI was 0.39±0.16 for the OS group and0.38±0.15 for the ES group (p=0.43). 30-day complications (aOR 4.89; 95%CI: 1.67-14.33) and early readmissions (aHR 3.32; 95%CI: 1.16-9.55) were increased in the OS group compared to the ES group. Frailty severity did not predict risk of re-intervention in either group.  Kaplan Meier analysis showed an increased risk of amputation, death, and the composite outcome of amputation and/or death in both treatment groups with increasing frailty when stratified by frailty severity (p<0.005 for all).  Multivariate analysis confirms that frailty independently predicts major amputation (aHR 2.16; 1.06-4.39), mortality (aHR 2.62; 95%CI: 1.17-5.88), and the composite outcome (aHR 1.97; 95%CI: 1.06-3.68) in the cohort as a whole. Hypoalbuminemia is correlated with increased mFI in the ES group (p<0.01), but only showed a trend with mFI in the OS group (p>0.05).  Independent of treatment assignment and preoperative mFI, higher albumin concentration was associated with lower risk of amputation (aHR: 0.58; 95% CI: 0.36 -0.94) and mortality (aHR: 0.45; 95% CI: 0.25-0.83). Higher hemoglobin concentration was also independently predictive of limb salvage (aHR 0.72 95%CI: 0.62-0.84).

Conclusion: Frailty severity is predictive of short- and long-term outcomes after lower extremity revascularization. Hypoalbuminemia and anemia are associated with higher mFI and independently predicted poor outcome after revascularization, suggesting albumin and hemoglobin concentration may serve as true biomarkers of frailty in this population. 
 

58.01 Older Age Increases Mortality And Stroke Risk at One Year After Carotid Revascularization

Posted on January 31, 2019

S. J. Aitken1,2, S. J. Aitken1,2  1Concord Repatriation General Hospital,Institute Academic Surgery (Vascular),Sydney, NEW SOUTH WALES (NSW), Australia 2University Of Sydney,Concord Clinical School,Sydney, NSW, Australia

Introduction:
Cardiac and neurological complications following carotid revascularization have been associated with an increased risk of mortality, especially in older patients. This study reports mortality and stroke following carotid revascularization in Australia, comparing outcomes up to 1 year for those with and without complications within 30days and in younger and older patients.

Methods:
Routinely collected hospital data on all patients in New South Wales (NSW), Australia, were linked to state-wide mortality records. All patients who underwent carotid revascularization (endarterectomy or stenting) between 2010-2012 were selected. Primary outcomes of mortality or stroke were measured at 30days, 90days and 1 year. Secondary outcomes were complications within 30days, length of stay and hospital readmission within 90days. Differences in outcomes between younger (aged less than 75 years old) and older (aged 75 years and older) patients were evaluated. Complications were assessed at 30days, including stroke and major adverse cardiac events (MACE). Outcomes were assessed with multivariable Cox regression and Kaplan Meier survival analysis.

Results:

3008 carotid revascularization procedures were performed between 2010 and 2013; 20% for symptomatic carotid disease (n=598). Carotid endarterectomy was the most common procedure (n=2280, 76%), with 728 patients (24%) having carotid stenting. The median age was 72 years (SD 10), with more males than females having carotid revascularization (M:F ratio 69%:31%).  Mortality at 30days was 0.8% (n=26), 90days 1.4% (n=43) and 1 year 3.9% (n=112). Postoperative stroke occurred in 14 patients at 30days (0.5%), 90days 1.1% (n=32) and 1 year 1.7% (n=52).  17.3% of patients had a major complication within 30days (n=522). Median length of stay was 3 days (IQR 7). 25% of patients (n=754) had a readmission for any cause at 90days. After adjusting for age, gender and procedure type, patients aged 75 years or older were at higher mortality risk than younger patients (HR 2.7, 95%CI 2.2-3.3, P<.0001) at 1 year. After adjusting for age, gender and procedure type, older patients had a higher risk of stroke at 1 year (HR 2.4, 95%CI 1.9-2.8, P<.0001) than younger patients. Stroke risk was also associated with carotid stenting and major complications.   MACE occurring within 30days predicted 1 year stroke (HR 2.1, 95%CI 1.6-2.9, P<.0001) and death (HR 2.0, 95%CI 1.5-2.7, P<.0001). Older patients had a higher incidence of MACE (IRR 1.9, 95%CI 1.4-2.6, P<.0001) and complications (IRR 1.2, 95%CI 1.1-1.4, P.007) within 30days than younger patients.

Conclusion:

Older age increased risk for all adverse outcomes including mortality, stroke, complications, increased length of stay and readmission. Postoperative complications also increased the risk of mortality and stroke at 1 year. Targeted strategies to improve perioperative care in older patients are required to reduce complications associated with postoperative mortality.

57.20 The Utility of ECMO After Liver Transplant: Experience at a Single High-Volume United States Center

Posted on January 31, 2019

M. E. Pulcrano1, H. J. Braun1, D. J. Weber1, B. Padilla1, N. L. Ascher1  1University Of California – San Francisco,Department Of Surgery,San Francisco, CA, USA

Introduction: Extracorporeal membrane oxygenation (ECMO) is a method to artificially support respiratory and/or cardiac function when conventional techniques fail. Liver transplant recipients constitute a special patient group at high risk for developing pathologies such as acute respiratory distress syndrome and pneumonia in the settings of resuscitation and immunosuppression, respectively. Additionally, conditions such as portopulmonary hypertension and hepatopulmonary syndrome are directly related to cirrhosis and can contribute to respiratory distress both before and after transplantation. Over the past two decades, ECMO has been described as a treatment modality for acute pulmonary and/or cardiac disease following orthotopic liver transplantation (OLT) in the adult and pediatric populations. Here we present a series of OLT adult recipients placed on ECMO after transplantation for both respiratory and cardiac indications, which constitutes the largest described review in the United States.

 

Methods: The Extracorporeal Life Support Organization database of all patients at our academic institution who had undergone ECMO cannulation was cross-referenced with the institution’s liver transplant patient database, between 2002 and 2018. The patient and disease characteristics were identified, and the types of cannulation and outcomes were described

Results: Eight patients were identified, five men and three women aged 28-68. One patient was cannulated intraoperatively and the initiation of ECMO otherwise ranged from 3 days to 6 months after transplant in the remainder of the cohort. The cannulation time ranged from 1 day to 1.5 months. Four patients were placed on venovenous ECMO, for which the indications were hypoxemic respiratory failure (n=3) and inferior vena cava (IVC) obstruction (n=1). Four patients were placed on venoarterial ECMO for right heart failure and massive pulmonary embolism (PE). Three of eight patients survived to discharge and remain alive today. Of the deceased patients (n=5), three required ECMO as complications of pulmonary disease (portopulmonary hypertension (n=1) and hepatopulmonary syndrome (n=2)), one for technical complication of IVC obstruction, and one for a massive PE. Of the three patients who survived to discharge, one was cannulated intraoperatively for a massive PE, one was cannulated for hypoxemic respiratory failure in the setting of massive resuscitation for bleeding and coagulopathy, and one was cannulated for right heart failure. These patients remain alive from 5-13 months after transplantation.

Conclusion: ECMO is a useful modality to consider in liver transplant patients with severe cardiopulmonary failure. This adjunctive support is particularly effective if the etiology of cardiopulmonary failure is recognized promptly and is thought to be transient. This is the largest series in the United States and demonstrates a 38% survival rate, which is comparable to other reports in the literature from Asia.

57.19 Post Transplant Nutrition Status: Resolution of Hypoalbuminemia

Posted on January 31, 2019

S. Atoa1, M. A. Maluccio1, R. S. Mangus1  1Indiana University School Of Medicine,Surgery / Transplant,Indianapolis, IN, USA

Introduction:
Poor nutrition status is associated with worse post-operative outcomes including complications, prolonged hospitalization and increased costs. Malnutrition is associated with weakness and frailty. There are many ways to measure malnutrition and frailty including laboratory markers to assess protein stores, functional tests to assess strength, and body imaging to assess muscle mass. This study uses serum albumin levels, measured before and after transplant, as a marker for nutrition status. We assess the utility of various methods of pre-transplant testing to determine post-transplant nutrition status.

Methods:
Adults patients undergoing liver, kidney or pancreas transplant were included in this study. Nutrition status was assessed using weight and body mass index (BMI), 5 meter walk test and computed tomography assessment of muscle and fat stores. Each of these tests was then correlated with post transplant serum albumin levels. Measures of nutrition status included a scaled scoring of core muscle mass, and visceral and subcutaneous fat stores, protein and albumin serum levels, functional status, body weight and BMI.

Results:
There were 184 patients analyzed. All transplant organ groups had an acute decrease in their serum albumin level in the post-transplant period, with an improvement to the normal range between 20 and 50 days. Liver patients lagged behind the kidney and pancreas groups in resolution of hypoalbuminemia at all time points up to 40 days (p<0.01). Among liver transplant patients, the groups that experienced a better return to normal serum albumin levels included younger patients (

Conclusion:
This study identifies an acute drop in serum albumin that occurs post transplant. This decrease returns to the normal range with kidney sooner than pancreas, which is sooner than for liver patients. Malnutrition is clearly worse for the liver failure patients. Those that resolve their hypoalbuminemia the fastest are those who are female and younger and have good physiologic strength. However, the group with the worst sarcopenia returned to normal faster than those with lesser muscle loss. These findings suggest a physiologic mechanism which stimulates normalization of serum albumin levels, which is a marker for improved nutrition status. This mechanism may be linked to baseline body muscle mass.
 

57.18 Obesity Is Not a Predictor of Poor Outcomes in Simultaneous Heart-Kidney Transplant

Posted on January 31, 2019

M. Gunder1, C. Biefeld1, V. Dioguardi1, K. N. Lau1, A. Di Carlo1, S. Karhadkar1  1Temple University,Surgery,Philadelphia, PA, USA

Introduction

Prior studies have demonstrated a relationship between obesity and increased risk for morbidity and mortality after kidney transplantation. Therefore, patients with a high body mass index (BMI) can be denied listing status, or have longer waitlist times. The relationship between obesity and simultaneous heart-kidney transplant (SHKT) outcomes has not been studied.

Methods

A retrospective analysis of all patients listed in the United Network for Organ Sharing database undergoing SHKT between 2006-2018 was performed. Patients were grouped into categories based on BMI, including underweight (BMI 16.00-18.49), normal (BMI 18.50-24.99), overweight (BMI 25.00-29.99), or obese (BMI 30.00-60.00). Data was collected on patient outcomes including length of stay, incidence of delayed graft failure, heart or kidney allograft failure, resumption of maintenance dialysis, and overall patient survival. Fisher exact test or ANOVA analysis were performed to assess correlation between BMI and SHKT outcomes. Kaplan Meier analysis was used to project patient and allograft survival across BMI groups.

Results

A total of 1054 simultaneous heart-kidney transplants were performed during the time period studied. 33 recipients were underweight, 375 were normal weight, 400 were overweight, and 246 were obese. The overall mean survival for all patients was 3238 days. Overall rate of kidney graft failure was 8.8% and of heart graft failure was 5.2%. Interestingly, a higher BMI was not associated with an increased length of stay (p=0.712), incidence of DGF (p=0.786), incidence of kidney (p=0.581) or heart (p=0.734) graft failure, or decreased overall survival (p=0.160).

Conclusion

Obesity does not correlate with adverse outcomes after simultaneous heart-kidney transplantation. A similar rate of kidney allograft failure, heart allograft failure, and patient mortality was seen in all groups. A higher BMI should not preclude potential recipients from being listed for SHKT. 

57.17 Identifying Challenges in Steroid Minimization Protocols for Kidney Transplant Immunosuppression

Posted on January 31, 2019

N. Taylor1, A. Fagenson1, K. Lau1, A. Di Carlo1, A. Diamond1, S. Karhadkar1  1Temple University,SURGERY / ABDOMINAL ORGAN TRANSPLANTATION,Philadelpha, PA, USA

Introduction:
Corticosteroids are a pharmacological staple for prevention of graft rejection in renal transplantation. Deleterious effects of long term steroid use have prompted attempts to implement maintenance immunosuppression protocols that minimize steroids with little success due to higher rejection rates. Furthermore, other factors such as African American race are postulated to increased rates of rejection. Conventionally, this has resulted in reluctance to use steroid-free or steroid minimization protocols in African American patients. We sought to determine whether steroid-free or steroid minimization protocols resulted in increased negative outcomes in African American patients when compared to their non-African American counterparts.

Methods:
A retrospective review of all consecutive renal transplantation patients at an Urban University Hospital from 2013 through 2018 was performed. Patients were stratified by race. Continuous variables were compared using student independent t-tests and categorical variables were compared using Chi-squared and Fischer exact tests. Patients were further grouped based on steroid minimization maintenance protocol (no more than 5mg of prednisone per day) and steroid-free maintenance protocol.

Results:
A total of 227 kidney allograft recipients were identified; 124 (54.6%) African Americans and 103 (45.4%) non-African Americans. There were significantly more non-African American patients that received steroid minimization maintenance immunosuppression protocol (n=86, 30.8% AA vs 51.6% non-AA, p = 0.002). Additionally, 34 patients received a steroid-free maintenance immunosuppression protocol. There were no racial disparities in this group. Initiation of steroid minimization and steroid free protocols was based on identification of risk factors for acute rejection such as PRA, prior renal transplant and presence of donor specific antibodies.

Overall, the African American group had more episodes of acute rejection (22.1% AA vs 10.2% non-AA; p=0.019) for all comers, irrespective of steroid utilization. Of the patients that underwent a steroid minimization protocol there was no difference in acute rejection between the African American and non-African American groups. Of the patients that underwent a steroid-free protocol there was no difference in acute rejection or graft failure between the African American and non-African groups.

Conclusion:

Contrary to the notion that African American patients should remain on corticosteroids during the maintenance period, our data suggests that there is no significant increase in risk of rejection when African American patients undergo steroid minimization or steroid-free maintenance immunosuppression protocols following renal transplantation. Other factors such as drug compliance and pharmacodynamics may play a role in immune modulation in this group of patients.

57.16 Bortezomib Based Induction Therapy Decreases Delayed Graft Function in High Risk Kidney Transplantion

Posted on January 31, 2019

J. Lambdin1, N. Koizumi2, K. Mahendran1, J. K. Melancon1  1George Washington University School Of Medicine And Health Sciences,Transplant Institute And Division Of Transplant Surgery,Washington, DC, USA 2George Mason University,School Of Policy, Government And International Affairs,Arlington, VIRGINIA, USA

Introduction:  Over 100 000 candidates await a kidney transplant on the national United Network for Organ Sharing (UNOS) waitlist. Among these candidates, one-third have preformed antibodies against human leukocyte antigens (HLA). These patients fall under the category of high risk kidney transplantation compared to non-sensitized patients. Those patients: 1) undergoing ABO incompatible kidney transplant; 2) with donor specific antibodies; 3) with a low level of nonspecific antibodies; or 4) with a prior transplant/s also fall into this category. These high risk transplant recipients demonstrate a higher rate of delay graft function and graft failure.

Our objective was to demonstrate that our unique Bortezomib based desensitization protocol could be an effective induction therapy among high risk kidney transplant candidates to reduce delayed graft function and acute antibody mediated rejections. Our hypothesis is that humoral responses has been an underappreciated culprit in both delayed graft function and as an interface with cell medicated rejection responses. Bortezomib's anti humoral mechanism we belive to be an important adjunct in the immunospressive armamentraium. 

Methods:  A total of 126 patients underwent deceased-donor kidney transplants at our center between 01/01/2015 and 04/23/2018. Of these, 65 patients received Bortezomib based induction therapy as they were high risk candidates. All patients were given maintenance immunosuppression with Tacrolimus, Mycophenolate Mofetil and Prednisone as per standard of care, regardless of their risk stratification.   The outcomes were assessed based on: 1) occurrence of delayed graft function; and 2) presence of protocol biopsy proven rejection; and 3) graft survival. Using the UNOS data, we performed a propensity score analysis to extract high risk transplant recipients from the SRTR national transplant database who are similar to the recipients who received the Bortezomib based induction therapy at our center. Logistic regression and survival analyses were performed to compare the outcomes between the intervention recipients and the matched sample. 

Results: The Bortezomib based desensitization therapy was effective (p=0.01) in the matched sample analysis. Adjusting for the common covariates and the propensity score, those recipients who received the therapy was about 36% less likely to experience a delayed graft function. 

Conclusion: We conclude that our Bortezomib based desensitization/induction therapy decreases delayed graft function in high risk kidney transplant recipients. We belive that this will impact very favorably on the long term graft and patient survial. 

57.15 Lung Transplant Listing Practices: Survival in Candidates Who Improve Clinically to Delisting

Posted on January 31, 2019

S. E. Rudasill1, Y. Sanaiha1, H. Xing1, A. Mardock1, H. Khoury1, P. Benharash1  1David Geffen School Of Medicine, University Of California At Los Angeles,Cardiac Surgery,Los Angeles, CA, USA

Introduction:
Listing criteria for lung transplantation influence demand for transplant. Occasionally, listed candidates clinically improve and are delisted, but their outcomes have not been elucidated. This study examined a national database for the characteristics and survival of primary and re-transplant lung transplant candidates who improved to delisting.

Methods:
This was a retrospective study of patients listed for lung transplantation between 1987-2015 in the United Network for Organ Sharing (UNOS) database. The last permanent waiting list status change was classified into transplanted, improved to delisting, or deteriorated to delisting, with those dying before or refusing transplant excluded from analysis. Survival time was calculated using the linked Social Security Administration date of death, and analysis was performed via the Kaplan-Meier method. Adjusted Cox hazards models predicted five-year mortality, while multivariable logistic regression identified the patient characteristics predicting improvement to delisting. Transplant centers were organized into quartiles by volume and analyzed for differences in the proportion of candidates improving or deteriorating.

Results:
Of 13,688 candidates, 12,188 (89.0%) were transplanted, 1,046 (7.6%) deteriorated to delisting, and 454 (3.3%) improved to delisting. Patients who improved were younger (48.5 vs. 50.9 years, p<0.001), less likely male (39.1 vs. 54.6%, p<0.001), and more likely to use life support measures at listing (7.3 vs. 2.9%, p<0.001) relative to those transplanted. Cox regressions showed increased five-year mortality for improved (HR=1.21 [1.07-1.38], p=0.002) and deteriorated (HR=3.36 [3.11-3.64], p<0.001) candidates relative to those transplanted, but two-year survival was highest in improved candidates (Figure 1). The short-term survival benefit for those improving was observed in primary (log rank p<0.001) but not re-transplant (log rank p=0.767) candidates. Those most likely to improve to delisting were older than 60 years (OR=1.96 [1.26-3.05], p=0.003), listed for primary pulmonary hypertension (OR=2.27 [1.13-4.56], p=0.022), and used life support measures at listing (OR=3.63 [2.15-6.13], p<0.001) relative to those transplanted. The percentage of candidates improving to delisting increased with increasing hospital volume (p<0.001).

Conclusion:
Primary lung transplant candidates improving to delisting faced lower short-term but higher five-year mortality relative to transplanted candidates. This effect did not persist for re-transplantation. Institutional volume may influence variations in listing practices. Understanding the characteristics and outcomes of this delisted population can better inform future candidate selection criteria.

57.14 Changing independent predictors of length of stay in pediatric kidney transplantation

Posted on January 31, 2019

E. L. Godfrey1, E. T. Pan1, D. Yoeli3, A. Rana2  1Baylor College Of Medicine,Houston, TX, USA 2Baylor College Of Medicine,Division Of Abdominal Transplantation,Houston, TX, USA 3University Of Colorado Denver,Department Of Surgery,Aurora, CO, USA

Introduction:

Factors influencing graft and patient survival in pediatric kidney transplant recipients have been studied, but the outcome of length of stay has rarely been examined, despite its usefulness as a marker of perioperative morbidity. Examining changing predictors of length of stay provides both tools for clinical decision making as well as an assessment of the impact of the current kidney allocation system (KAS); the KAS was introduced in 2014 with a specific focus on reducing ethnic and regional disparities associated with the previous Share 35 system.

Methods:

The United Network for Organ Sharing provided de-identified data on 5,111 patients under the age of 18 listed for kidney transplant from 2005 to 2017, under both the Share 35 and Kidney Allocation System. Age, race, insurance status, a diagnosis of FSGS, KDPI, listing status (active, urgent, or critical), dialysis use and glomerular filtration rate (GFR) at transplant, donation after cardiac death, BMI, regional or national share status, whether the donor organ was on a pump, and cold ischemia time thresholds were all assessed against a length of stay greater than 7 days, 10 days, and 14 days. Univariate regression was performed to identify potential predictors, and multivariate regression was performed to determine which factors were independent predictors of prolonged stay.

Results:
Predictors of prolonged stay have changed between the two allocation systems. Younger age at time of transplant is a predictor of longer stays across all groups. Under the Share 35 system, Hispanic race is a predictor of shorter stays, but being non-Hispanic and non-white is a predictor of longer stays. Being on dialysis before transplant and having a lower GFR are also associated with stays longer than 10 days. In the KAS, increased BMI is also associated with shorter stays, excluding overweight or obese status, which did not have a significant relationship with length of stay. Lower GFR remained a predictor of prolonged stay in general, while either public or private insurance coverage and dialysis before transplant were predictive of stays over 10 days. 

Conclusion:

Characteristics indicating poorer native kidney function prior to transplant have been, and remain, predictors of prolonged post-transplant hospitalization. Older age and higher BMI predict shorter hospital course; studies have suggested long-term graft and patient survival are superior in younger recipients, but this effect may only start to be observed longer after transplant: further investigation is needed to reconcile these apparently contradictory findings. The change in race as a predictive factor between Share 35 and KAS corroborates recent studies reporting that transplant rates have increased for non-white recipients, and suggests that this increased access may reduce racial differences in perioperative outcomes. Further study of how the KAS accomplishes this should be utilized to create more equitable future allocation systems.

57.13 Effect of Transarterial Chemoembolization on Hepatic Artery Quality and Post-Transplant Complications

Posted on January 31, 2019

M. A. Selim1, M. Zimmerman1, J. Kim1, E. Hohenwalter2, W. Rilling2, J. C. Hong1  2Medical College Of Wisconsin,Department Of Radiology. Division Of Vascular And Intervention Radiology.,Milwaukee, WI, USA 1Medical College Of Wisconsin,Department Of Surgery. Division Of Transplant Surgery,Milwaukee, WI, USA

Introduction: Transarterial chemoembolization (TACE) is commonly utilized to downstage irresectable hepatic tumors prior to liver transplant (LT). Vascular complications of the hepatic artery (HA) after TACE may impact its suitability for vascular anastomosis during LT. We sought to examine HA quality in patients with pre-LT TACE.

Methods: We conducted a single center analysis from our prospective database of all adult LT performed in our center between October 2012 and May 2018. Median follow up was 26 months. Outcomes were compared between patients who received pre-LT TACE and those who did not.

Results: 162 transplants were included. Group I did not receive pre-LT TACE (n=110), Group II did (n=52). . Patients with TACE tended to be older males with viral hepatitis. Other characteristics are outlined in Table 1.
Our standard target for arterial anastomosis is the native common HA. For patients with TACE (Group II) only two cases (4%) required an aortic conduit to be created for arterialization because of inadequate CHA (dissection or poor vessel quality), compared to 8% in Group I, p=0.4.
The incidence of post-LT vascular complications was similar, Group I=6.3% and Group II=5.7% (p=0.9). Most complications were late (>30 days). Only one case in the TACE group was due to native HA dissection that necessitated revision and creating an aortic conduit.
Cox regression analysis showed that TACE wasn’t associated with increased risk of vascular complications. There was no difference in patient or graft survival between the two groups (p=0.1 and 0.2 respectively)

Conclusion: We found no association between TACE and higher rates of arterial vascular complications or other post-LT adverse outcomes. Careful post-LT follow up is important for early detection and management of any vascular complication.

 

57.12 Use of the Montreal Cognitive Assessment (MoCA) in Pre-transplant Evaluation

Posted on January 31, 2019

F. Olumba1, C. S. Hwang1, S. Levea1, B. Tanriover1, Y. Liang1, P. Vagefi1, M. MacConmara1, C. Hwang1  1UT Southwestern Medical Center,Dallas, TX, USA

Introduction:
Decreased cognitive function is associated with higher mortality in end stage renal disease (ESRD) and kidney transplant recipients. The Montreal Cognitive Assessment (MoCA) is a quick test often used to screen for cognitive impairment. In addition, the Stanford Integrated Psychosocial Assessment for Transplantation (SIPAT) is a scoring system increasingly used to quantify psychosocial risk. We questioned if MOCA and SIPAT scores would predict patients listed for kidney transplantation.

Methods:
A single center retrospective analysis was performed including all adult patients (age ≥18 years) with either CKD 5 or ESRD  who were evaluated for kidney transplantation between January 1, 2016 and January 1, 2018. Non-parametric comparison tests were executed to assess differences in MoCA and SIPAT scores between subjects successfully listed for transplant and those not listed during the study period. Regression analysis was done to determine the relationship of MoCA and SIPAT scores to listing outcome. 

Results:
880 patients were included for analysis. The average age at the time of referral or evaluation was 54.1± 13.8 years. The most common cause of ESRD was from diabetes and hypertension (26.8%), with the majority of subjects having ESRD (58.1%). During the study period, 255 (29.0%) of patients were placed on the waitlist. The median score for the non-listed group vs. listed groups was 24 vs. 25, however a  score ≥ 23 oresulted in an OR of 1.88 (95% CI 1.18 to 2.99, p= 0.0078) for being placed on the transplant list. Furthermore, of those patients not listed with MOCA ≥ 23 no statistically significant difference was found between those listed with respect to age, sex, race, etiology of ESRD, dialysis use or six-minute walk distance. Interestingly, SIPAT scores of  those of ≥ 23 non-listed patients were significantly higher (9 v 12) indicating increased psychosocial (p= 0.003).

Conclusion:
Combining MOCA and SIPAT scores may have useful predictive value to determine suitable candidates for kidney transplantation.

57.10 Outcomes in Patients Transplanted for NASH – Bigger is Better

Posted on January 31, 2019

Y. Liang1, D. Desai1, P. Vagefi1, C. Hwang1, M. MacConmara1  2UT Southwestern Medical Center,Medicine,Dallas, TX, USA 1UT Southwestern Medical Center,Surgery,Dallas, TX, USA

Introduction:
Non-alcoholic steatohepatitis (NASH) will likely become the most common indication for liver transplantation during the next decade. We sought to determine the impact of recipient and donor selection practices on outcomes in NASH recipients. 

Methods:
The United Network for Organ Sharing database was queried to identify all liver transplant recipients from 2000 to 2017. Pediatric recipients (< 18 years) were excluded from the analysis. Donor and recipient demographic, clinical and biochemical data were extracted and risk factors for outcome after transplant were compared between NASH and all other cause recipients. Patient and graft survival rates were calculated for both groups. Categorical differences were compared using the unpaired Student's t-test and nominal variables using either the Chi Square test. A p-value of <0.01 was considered significant. 

Results:
88,602 adult liver transplants were completed of which NASH recipients accounted for 9,689 transplants. NASH recipients were older (58.7 vs. 53.2 years, p<0.001) with higher MELD scores (23.5 vs. 22.1, p<0.001). Donor age >60 years (17.1% vs. 12.9% p<0.001), donor macrosteatosis over 30% (3.2% vs. 2.1%, p<0.001) and donor donation after cardiac death (DCD) status (5.8% vs. 4.5% p<0.001) were all significantly higher in NASH recipients. Cold ischemia time (6.5 vs. 6.9 hours, p<0.001) was shorter for NASH recipients. Interestingly, graft survival was significantly better for NASH recipients (Figure 1, p<0.001). 

Conclusion:
NASH recipients have superior outcomes following liver transplantation despite utilization of higher risk donor livers and should be considered for offers of marginal organs. 

« Previous Page
Next Page »