195

234

ASSOCIATION FOR ACADEMIC SURGERY AND SOCIETY OF UNIVERSITY SURGEONS—ABSTRACTS

tained via endoscopic biopsy. Liver and kidney allografts utilize liver enzymes and creatinine, respectively as predictive markers of allograft dysfunction. No such marker is available for intestinal allograft dysfunction. A pilot study of stool calprotectin, a marker for active inflammatory bowel disease, suggested that this cytosolic protein found in neutrophils and macrophages may correlate with rejection episodes (optimal cutoff ⬍ 92 mg/L). Hypothesis: Calprotectin levels can be used instead of protocol intestinal allograft biopsies as a non-invasive method to detect rejection in intestinal transplant recipients. Methods: With IRB approval and informed patient consent (n⫽16), stool samples were collected for measurement of calprotectin levels after intestinal transplantation and correlated with allograft histology from protocol biopsies. This prospective study will enroll a total of 40 patients in order to have adequate power to determine if protocol biopsies can safely be eliminated. Samples were prospectively collected, stored, and then analyzed using the CALPREST® ELISA kit (NovaTec Immundiagnostica, Germany). Results: In the 16 patients enrolled, 4 patients were treated for acute allograft rejection. Median calprotectin levels obtained within 48 hours prior to or at the time of biopsy were markedly elevated 502 mg/L (range 60-1574). Two of the 4 patients had severe rejection resulting in explantation of the allograft or patient death. Calprotectin levels remained elevated during and after treatment in both these patients. In the other 2 patients calprotectin levels returned quickly to normal after treatment and resolution of the rejection episodes. In 2 additional patients, allograft histology was consistent with rejection, but the patients were not treated due to lack of symptoms. Calprotectin levels were ⬍ 20 mg/L in both patients and repeat biopsies revealed no rejection. Out of 119 additional samples with normal histology, viral enteritis, non-specific enteritits (NSE) or other pathology only 2 (1.7%) patient samples demonstrated elevated calprotectin levels (92 and 99 mg/L, respectively). Neither of these 2 patients developed rejection or had further elevations in their calprotectin levels. Conclusions: At the time of this interim analysis, calprotectin appears to have excellent sensitivity and specificity for intestinal allograft rejection and will likely be a good non-invasive test for allograft monitoring. In the future, endoscopic biopsies will likely be reserved for confirmation of rejection episodes. 100

Calprotectin mg/L (x10)

90 80 70 60 50

shortage of pediatric whole-organ liver allografts, two strategies have been developed to increase the donor liver pool: cadaveric segmental liver allografts (which include both reduced partial-liver allografts and split-liver allografts), and living donor liver allografts. Previous studies have suggested that living donor liver transplantation (LDLT) results in superior patient survival for recipients less than 2 years of age when compared with cadaveric segmental liver transplantation (CSLT) and whole-organ liver transplantation (WLT). However, these studies used data obtained prior to 2001. Analysis of outcomes in patients transplanted after 2001 is warranted as the field of liver transplantation has incorporated improved surgical techniques and practices which are associated with better outcomes. Objective: To determine which allograft type is associated with superior patient and allograft survival rates in pediatric recipients transplanted in the current era. Methods: Data on all patients less than 12 years of age who underwent liver transplantation in the United States between 2/2002 and 12/2004 was obtained from the United Network of Organ Sharing (UNOS) database. After adjusting for multiple donor and recipient characteristics, the impact of allograft type on 1-year post-transplant patient and allograft survival was assessed with a multivariate Cox proportional hazards model. A single subgroup analysis was performed for recipients less than 2 years of age. Results: Of the 1,260 patients included in the study, 659 (52.3%) patients underwent WLT, 412 (32.6%) CSLT, and 189 (15%) LDLT. After risk-adjustment there was no significant difference in 1-year patient survival rates for WLT (94.2%), LDLT (92.8%), or CSLT (90.7%) recipients (p-values for the pairwise comparisons: 0.53 for WLT vs. LDLT, 0.41 for LDLT vs. CSLT, and 0.07 for WLT vs. CSLT). There was also no difference in 1-year allograft survival rates for WLT (84%), LDLT (86.4%), or CSLT (80.1%) recipients (p-values for the pairwise comparisons: 0.44 for WLT vs. LDLT, 0.07 for LDLT vs. CSLT, and 0.16 for WLT vs. CSLT). Likewise, for patients less than 2 years of age (n⫽755), the adjusted 1-year patient survival rates: WLT (95.5%), LDLT (92.1%), and CSLT (91.3%)did not differ significantly (p-values for the pairwise comparisons: 0.19 for WLT vs. LDLT, 0.79 for LDLT vs. CSLT. and 0.06 for WLT vs. CSLT). In this subgroup, allograft survival rates: WLT (83.2%), LDLT (86.3%), CSLT (80%) also did not differ significantly (p-values for the pairwise comparisons: 0.41 for WLT vs. LDLT, 0.12 for LDLT vs. CSLT, and 0.35 for WLT vs. CSLT). Conclusion: Whole organ, cadaveric segmental, and living donor liver allografts achieve similar outcomes in pediatric patients, including those less than 2 years of age. In the current era of liver transplantation, this most recent study justifies the use of all available liver allograft types to minimize waitlist mortality for pediatric patients with end-stage liver disease. 195. LIVING KIDNEY DONOR RELATIONSHIP IN CAUCASIAN AND AFRICAN-AMERICAN POPULATIONS AND IMPLICATIONS FOR TARGETED DONOR EDUCATION PROGRAMS. M. Cooper, M. Phelan, S. Jacobs, J. Nogueira, R. Haririan, E. Schweitzer, L. Campos, B. Philosophe, S. Bartlett; University of Maryland Medical System, Baltimore, MD

40 30 20 10 0

NSE

Rejection

PTLD

Viral Enteritis

Normal Indeterminate

194. IS THERE A PREFERRED ALLOGRAFT TYPE FOR PEDIATRIC LIVER TRANSPLANT RECIPIENTS? N. S. Becker 1, N. R. Barshes 1, T. S. Nguyen 2, J. Rojo 2, J. A. Rodriguez 1, T. A. Aloia 1, C. A. O’Mahony 1, J. A. Goss 1; 1Michael E. Debakey Department of Surgery, Baylor College of Medicine, Houston, TX, 2Department of Statistics, Rice University, Houston, TX Background: Liver transplantation is the only curative option for pediatric patients with end-stage liver disease. In response to the

Introduction: Kidney transplantation is presently recognized as the treatment of choice for patients with renal failure. The number of individuals awaiting transplantation continues to grow exponentially while cadaveric organ donation has remained stagnant. Living kidney donation (LKD) has expanded in an attempt to meet that need. The introduction of the laparoscopic nephrectomy technique brought more donors for evaluation to academic transplant centers. The opportunities for LKD vary greatly among socioeconomic and racial groups. We reviewed our series of living donors to evaluate donor relationships in various ethnic groups in order to improve donor education programs with the expectation of increasing opportunities for dialysis patients via LKD. Methods: Donor and recipient records for 1000 laparoscopic donor nephrectomy patients were reviewed (1996-2005). An additional

ASSOCIATION FOR ACADEMIC SURGERY AND SOCIETY OF UNIVERSITY SURGEONS—ABSTRACTS 857 records of potential recipients presenting with an initial donor were analyzed to assess donor relationship (2000-2005). We compared the relationship of the recipient to both the initial donor identified and the indvidual who underwent nephrectomy in both the Caucasian (CC) and African-American (AA) populations. Results: There were 693 CC and 263 AA live kidney donors. During the same time period 64% of CRT recipients were AA, p⫽.001. In the CC population 57% of the donors were female while in the AA population females accounted for 62% of donors. In the AA population donors were comprised of first-degree relatives 71%, seconddegree relatives 10%, and unrelated donors 19%. In the CC population donors were comprised of first-degree relatives 57%, second-degree relatives 6%, and unrelated donors 37% (p⬍.0001 for unrelated CC v. AA). Spousal donation is more common in CC donors (13%) than in AA donors (6%), p⫽.001. Donation from child to parent is more common in AA (33%) than in CC donors (15%), p⬍.0001. Donation from parent to child is similar in both CC (9%) and AA cohorts (8%). Donation among siblings is similar in CC (30%) and AA donors (32%) as well, p⫽NS. Potential AA adult kidney transplant recipients predominantly identified a child as a first potential donor in 63% of cases and 48% were successfully cleared for donation. In contrast, 69% of potential CC recipients identified a spouse as a first potential donor yet only 23% became live kidney donors, p⬍.001 and p⬍.001. Conclusion: Living donor transplantation occurred in greater proportion in the Caucasian population. In contrast, CRT was performed in AA in greater numbers. The majority of live renal donors in both the CC and AA population were female and a first degree relative. Perception of potential donors varies significantly between the two populations as does the success of ensuring medical clearance for that individual to undergo laparascopic donation. Although spousal donation was more common in the CC population, the majority of spouses initially identified were not suitable donors. Children predominate in AA which are those predominantly identified at initial screening. This knowledge has directed our center toward improved efforts in education for potential kidney recipients in the ability to identify a live donor for AA patients and in the recognition of the most suitable/medically stable donor for CC. This, in turn, has led to a dramatic increase in the number of LKD at our center. 196. LIMITED FEEDBACK AND VIDEO TUTORIALS OPTIMIZE LEARNING AND RESOURCE UTILIZATION DURING LAPAROSCOPIC SIMULATOR TRAINING. D. Stefanidis 1, J. R. Korndorffer, Jr. 2, B. T. Heniford 1, D. J. Scott 3; 1 Carolinas Medical Center, Charlotte, NC, 2Tulane University School of Medicine, New Orleans, LA, 3UT Southwestern Medical Center, Dallas, TX Introduction: Performance feedback is essential for learning motor skills but optimal methods have not been defined for simulator-based training. Video tutorials may facilitate skill acquisition. The purpose of this study was to determine the impact of instructor feedback and video tutorials on skill acquisition during proficiency-based laparoscopic suturing training. Methods: Performance data from a prospectively maintained database were reviewed for three groups of novices (n⫽34 medical students) who completed proficiency-based laparoscopic suturing training on a FLS-type videotrainer model as part of two separate IRBapproved randomized controlled trials. All trainees viewed a video tutorial, completed baseline testing and practiced during one-hour sessions to a previously validated proficiency level (score ⫽ 512) with standardized feedback provided by a single expert instructor. Group I (n⫽9) received intense feedback during the early stages of skill acquisition and had no further video exposure, Group II (n⫽13) received limited feedback (⬍10 min per session) and no further video exposure, and Group III (n⫽12) had limited feedback (⬍10 min. per session) and watched the video tutorial delib-

235

erately during each session. Feedback was quantified on a 0 (none) to 4 (extensive) Likert-scale per repetition by the instructor. Personnel and materials cost was calculated. Comparisons were by ANOVA; p⬍0.05 was considered significant. Results: Baseline performance was similar in all groups and all participants achieved the proficiency level on two consecutive attempts. As shown below, Group III required the shortest training time and resources to reach proficiency. Conclusions: Limited instructor feedback is superior to intense feedback during the early stages of proficiency-based laparoscopic simulator training. Coupled with video tutorials, this type of feedback may accelerate the learning curve and improve resource utilization by minimizing the need for instructor involvement.

Group Comparison Group I (n⫽9)

Group II (n⫽13)

Group III p-value (n⫽12) (ANOVA)

Instructor Feedback per Repetition † 1 ⫾ 0.3 0.23 ⫾ 0.1 0.27 ⫾ 0.1 Video Tutorial Viewings (#) 1⫾0 1⫾0 8⫾1 Training Time to Proficiency (min) 148 ⫾ 52 112 ⫾ 46 86 ⫾ 30 Repetitions to Proficiency (#) 50 ⫾ 17 38 ⫾ 13 31 ⫾ 9 Cost per trainee $156 ⫾ 50 $51 ⫾ 18 $43 ⫾ 13

⬍0.001* ⬍0.001 ⬍0.02* ⬍0.02* ⬍0.001*

† 0-4 Likert scale * Pair-wise comparisons not significant between groups II and III

197. ASSESSMENT OF COMMUNICATION SKILLS OF SURGICAL RESIDENTS USING THE SOCIAL SKILLS INVENTORY (SSI). I. B. Horwitz 1, S. K. Horwitz 2, M. L. Brandt 3, F. C. Brunicardi 3, B. G. Scott 3, S. S. Awad 3; 1University of Texas - School of Public Health, Houston, TX, 2Lamar University, Beaumont, TX, 3Baylor College of Medicine, Houston, TX Background: The ability of surgeons to effectively communicate has been found to decrease medical errors, reduce conflict among caregivers, and improve the overall quality of patient care. Additionally, consequences of poor communication have been linked to increased malpractice suits and patient non-compliance. Recently, the ACGME has mandated the inclusion of interpersonal and communication skills as part of the core competencies. However, attempting to assess this specific competency has proven difficult. The Social Skills Inventory (SSI), a widely utilized psychometric instrument, provides a multi-level analysis of verbal and non-verbal communication. Our objective was to use the SSI to assess the communication skills of surgical residents. Methods: Under an IRB approved protocol, surgical residents (n ⫽ 64) completed the SSI to assess their communication skills. The SSI is comprised of 90 statements that respondents rate on a 5-point Likert scale. SSI factors include two dimensions: Emotional Scores which measure non-verbal communication skills (Emotional Expressivity (EE), Emotional Sensitivity (ES) and Emotional Control (EC)), and Social Scores which measure verbal communication skills (Social Expressivity (SE), Social Sensitivity (SS) and Social Control (SC)). In addition, an Equilibrium Index (EI) is derived to assess if the social skills of respondents are appropriately balanced. Demographic data such as age, gender, post-graduate year (PGY), and previous managerial experience was also collected. Statistical analysis was performed using SPSS. Results: An internal consistency estimate of each SSI scale was computed and found to have overall item reliability coefficient of 0.89. It was found that the 6 SSI scales approximated a normal distribution. The distribution of the EI scores was asymmetrical and negatively skewed. Large vari-