ASSOCIATION FOR ACADEMIC SURGERY AND SOCIETY OF UNIVERSITY SURGEONS—ABSTRACTS
209
were significantly lower in GSNO-treated bioreactors (310⫾38 IU/L; n⫽3) compared to controls(919⫾188 IU/L;n⫽4; p⬍.05), but synthetic function remained unaltered as measured by albumin levels in the media (115⫾19 ug/day/cell inoculum GSNO-treated bioreactors vs. 110⫾13 ug/day/cell inoculum in controls; p⫽0.851) at 24 hrs. Conclusion: Addition of the NO donor GSNO reduces rat liver cell apoptosis within a 3-D perfusion bioreactor system and promotes liver cell aggregation and tissue reformation at 24 hours. GSNO treated bioreactors remain metabolically active and show significantly lower levels of cellular injury with intact biosynthetic function. Further studies will be required to evaluate the impact of GSNO treatment of liver bioreactors in clinical applications.
127. MULTIVARIATE ANALYSIS OF RISK FACTORS FOR ALLOGRAFT LOSS IN BLACK AND WHITE RECIPIENTS OF RENAL ALLOGRAFTS. C. E. Simpkins, D. Chang, R. A. Montgomery, E. Kraus, A. A. Zachary, R. Ugarte, C. Handley, D. Segev, D. Warren, J. K. Melancon; The Johns Hopkins University, School of Medicine, Baltimore, MD. Introduction: Ethnicity-based inequities in access to medical care, delivery of medical interventions, and outcomes following interactions with the health care system have been described across a broad range of medical fields. The goal of this study was to assess for organ allocation and patient level differences that may contribute to the explanation of disparity in allograft outcomes following renal transplantation for black vs. white recipients. Methods: Retrospective cohort study of prospectively collected registry information of a single university hospital experience from 1/1/1994 to 12/31/2004. Adult white (W) and black (B) recipients of a renal allograft from a living (LD) or deceased donor (DD) were identified. Outcome measures included renal allograft survival by Kaplan-Meier product limit estimate and multivariable analysis of independent risk factors for renal allograft loss by Cox proportional hazards methodology. Analyses were performed between ethnic groups by donor type. Results: 454 B recipients and 735 W recipients were identified for analysis. Live donor renal transplantation was more commonly performed in W vs. B recipients (63.8% vs. 19.3% of all LD transplant procedures for W vs. B, respectively). Median duration on the UNOS deceased donor waiting list was longer for B recipients of both LD and DD organs (LD: 235 days vs. 160 days, p⫽0.03; DD: 824 days vs. 526 days, p⬍0.001). Duration of cold ischemic preservation time (CIT) was longer for B vs. W recipients of DD organs (30 hours vs. 24 hours, p⬍0.001). By log-rank test, W recipients had an improved death-censored allograft survival experience relative to B recipients in both the live donor and deceased donor groups (p⫽0.007 & p⫽0.009, respectively). Kaplan-Meier estimates of death-censored allograft survival are demonstrated in Fig. 1. The absolute difference in deathcensored allograft survival between B and W recipients of LD organs was minimal at five years (90.9% vs. 93.9%, respectively). On multivariable analysis, median adjusted household income (MAHI) was found to be a statistically significant predictor of allograft loss for B recipients (HR:2.38 for MAHI$35k , 95% CI: 1.17 - 4.82). A trend in increased risk for allograft loss from ⬎0 HLA-DR mismatches was observed for B recipients (HR: 1.94, 95% CI: 0.96-3.93). For W recipients, the interaction between cadaveric donation and CIT (HR: 1.03, 95% CI: 1.02 - 1.05) as well as ⬎0 HLA-DR mismatches (HR: 2.05, 95% CI: 1.08-3.87) were found to be statistically significant predictors of allograft loss. Conclusion: Prominent differences exist in renal allograft outcomes between white and black recipients in the United States. Complex, multifactorial contributors to these outcomes, including physiologic responses to ischemia-reperfusion injury and socioeconomic status as suggested by this study, appear to play a role in the differences in allograft survival that have been observed between these two groups of recipients.
128. IN VIVO INDUCTION OF REGULATORY T-CELLS BY A NOVEL BONE-MARROW DERIVED CELL POPULATION IN THE CONTEXT OF SUCCESSFUL ALLOTRANSPLANTATION. V. R. Shinde Patil, K. N. Taylor, Y. L. Colson; Brigham and Women’s Hospital, Boston, MA. The field of transplantation biology has seen dramatic improvements in short term graft survival rates, however, poor long-term survival secondary to infection and rejection continues to be a major impediment to successful transplantation. As a means to improve graft outcomes and minimize complications associated with continual immunosuppressive therapy, current research efforts have focused on strategies to induce effective immunocompetent transplantation tolerance. The potential of regulatory T-cell tolerance has prompted the development of regimens to elicit functional regulatory T-cells under in vitro or in vivo conditions. In this report, we characterize a novel approach for the in vivo induction of fully allogeneic regulatory T-cells that results in donor-specific transplantation tolerance without graft-versus-host disease (GVHD). Lethally conditioned B10.BR (H-2k) allogeneic murine recipients were cotransplanted with either purified hematopoietic stem cells (SC) and a rare CD8⫹T-cell receptor (TCR)- bone marrow (BM) derived facilitating cell (FC) population (SC⫹FC), or SC along with BM-derived CD8⫹TCR⫹ T-cells (SC⫹TBM) from C57BL/6 (B6: H-2b) donors. Recipients were assessed for survival, SC alloengraftment, histologic evidence of acute GVHD and tolerance to donor and third-party skin allografts, 4-6 weeks following BM transplantation. Flow cytometric analysis was employed to determine donor chimerism, characterize reconstituted cellular subsets and quantify CD4⫹CD25⫹ regulatory T-cell induction. Specific immunomodulatory cytokines involved in tolerance and GVHD were assessed using ELISA. Flow cytometric analysis of the spleens and BM of SC⫹FC and SC⫹TBM recipients revealed nearly twice the numbers of CD4⫹CD25⫹ regulatory T-cells in SC⫹FC recipients compared with SC⫹TBM recipients. The induction of donor regulatory T-cells in SC⫹FC recipients was associated with long-term survival, a high degree of donor chimerism (96 ⫾ 1.3 %) with complete reconstitution of all cell lineages, and transplantation tolerance for donor, but not, third-party skin grafts. Tolerance, alloengraftment and the clinical and histologic absence of GVHD in these recipients was associated with a downregulation in pro-inflammatory cytokines involved in GVHD, namely IFN␥, TNF␣, IL1 and IL6. In contrast, SC⫹TBM recipients engrafted poorly (30 ⫾ 5.13%) and succumbed to low-grade GVHD associated with donor T-cell infiltration in the gut and increased systemic levels of these same pro-inflammatory cytokines. In summary, our data demonstrate that transplantation of FC into fully allogeneic recipients promotes robust SC engraftment, inhibits GVHD and induces donor-
210
ASSOCIATION FOR ACADEMIC SURGERY AND SOCIETY OF UNIVERSITY SURGEONS—ABSTRACTS
specific transplantation tolerance through the downregulation of pro-inflammatory cytokines and the in vivo induction of donor CD4⫹CD25⫹ regulatory T-cells. This novel strategy for eliciting regulatory T-cells in the context of successful allotransplantation has tremendous potential for the induction of donor-specific tolerance and improved long-term survival in transplant patients. 129. THE TRANSCRIPTION FACTOR INTERFERON REGULATORY FACTOR-1 (IRF-1) MEDIATES LIVER DAMAGE FOLLOWING ISCHEMIA-REPERFUSION INJURY. A. Tsung, M. Stang, G. Jeyabalan, A. Nakao, N. Ritchlow, M. H. Chan, D. A. Geller; Unviersity of Pittsburgh Medical Center, Pittsburgh, PA. Introduction: Hepatic ischemia occurs in the settings of liver transplantation, hepatic resection, and trauma. The intracellular signaling events contributing to local hepatocellular damage are only partially understood. Interferon regulatory factor-1 (IRF-1) is a nuclear transcription factor that regulates the expression of genes involved in both innate and acquired immunity; however its function in liver injury is unknown. Therefore, we sought to determine the role of IRF-1 in hepatic ischemia-reperfusion (I/R) injury. Methods: Partial warm ischemia was produced in the left and median lobes of WT and IRF-1 KO mice. Adenoviral gene delivery of IRF-1 (AdIRF-1) was also performed in C57BL6 mice. Serum and tissue samples were obtained after reperfusion to measure liver enzymes, histology, and inflammatory mediators. Results: In mice undergoing 60 minutes of ischemia, IRF-1 protein expression increased as early as 30 minutes after reperfusion (fig. A). IRF-1 KO mice were significantly protected from hepatic I/Rinduced damage (Table), in association with markedly lower hepatic TNF, IL-6, ICAM-1, and iNOS mRNA levels compared to WT controls (data not shown). Further, we found that hepatic JNK, but not p38, MAPK activation was lower in IRF-1 KO mice (fig. B). Total JNK was not affected. Transfection of normal mice with Ad-IRF-1 (3⫻109 pfu) resulted in increased liver damage even without an ischemic insult (Table). This injury was associated with increased phospho-JNK activity (fig. C) and hepatic iNOS expression (not shown). Since IRF-1 was found to have an important role during hepatic I/R injury, we also examined for signals that regulated IRF-1 gene expression in cultured hepatocytes. IFNg and IFNb were strong inducers of IRF-1 mRNA (⬎15-fold) in a time- and dose-dependent manner, while TNFa and IL-1b also induced IRF-1 mRNA to a lesser extent (2-3 fold). IL-6 or LPS had no effect on IRF-1 expression. Resting hepatocytes exhibited minimal basal IRF-1 mRNA, similar to normal liver in vivo. Conclusion: This study demonstrates that IRF-1 exerts a harmful role in hepatic I/R injury by modulating the expression of multiple inflammatory mediators. We further show that IRF-1 mediated liver damage involves the activation of JNK, and that hepatocellular IRF-1 expression itself is regulated by specific cytokines. Table
ALT (IU/L) Histology
I/R:WT
I/R:IRF1 KO
Ad-null
Ad-IRF-1
1,697 ⫾ 184 ⫹⫹⫹⫹
641 ⫾ 72* ⫹ (necrosis)
62 ⫾ 28 0
1,530 ⫾ 230* ⫹⫹⫹
Data are mean ⫾ SEM n ⫽ 6 per group *indicates p ⬍ 0.05 vs. WT or Ad-null)
CLINICAL TRIALS/OUTCOMES 130. ELECTIVE MAJOR SURGERY ON FRIDAY LEADS TO HIGHER MORTALITY AS COMPARED WITH MONDAY
THROUGH WEDNESDAY. S. Zare 1, K. M. Itani 2, T. Schifftner 3, W. G. Henderson 3, S. F. Khuri 2; 1Boston VA Healthcare System and Boston University, West Roxbury, MA, 2 Boston VA Healthcare System and Harvard University, West Roxbury, MA, 3The National Surgical Quality Improvement Project (NSQIP), University of Colorado Health Outcomes Program, Denver, CO. Introduction: Weekend hospital admissions for emergency conditions are associated with higher mortality when compared to weekday admissions. The effect of weekend on elective surgical admissions is undocumented. Methods: Using the VA National Surgical Quality Improvement Program (NSQIP) database, we compared the 30-day mortality and morbidity of elective major surgeries performed on Fridays with similar surgeries performed on Mondays through Wednesdays during fiscal years 2000-2004. The patients were divided into three groups: the FLOOR group (n⫽20,665) included elective surgeries that usually require post-operative admission to a regular floor: colectomy, enterectomy, ventral hernia repair, radical prostatectomy and nephrectomy; the ICU group (n⫽4,591) included elective surgeries that usually require post-operative admission to an intensive care unit: Abdomino-aortic aneurysm repair, pneumonectomy including lobectomy, craniotomy, craniectomy, and hepatectomy; the OUTPATIENT group (n⫽24,322) included outpatient surgeries: laparoscopic cholecystectomy, umbilical hernia repair and arthroscopy. A univariate analysis was performed comparing the 3 groups on 33 co-morbid conditions, 14 pre-operative laboratory tests and 4 operative and intra-operative variables that were prospectively collected. A logistic regression analysis was then used to test the effect of Friday surgery on mortality and morbidity in the presence of characteristics that were significant (p⬍0.20) in the univariate analysis. Results: Surgeries performed on Fridays in the FLOOR group were associated with significantly higher 30-day mortality when compared with surgeries performed on Mondays through Wednesdays (OR, 1.364; 95% CI, 1.043-1.782) (see Table). 30-day morbidity (1 or more complications) was similarly elevated (OR, 1.157; 95% CI, 1.048-1.278). Within the ICU and OUTPATIENT groups, no significant differences were observed in outcomes between surgeries performed on Fridays versus Mondays through Wednesdays. Conclusions: Patients who undergo major elective operations on Fridays and are admitted to a regular floor have a significantly higher rate of death and complications than those whose procedures were performed on Mondays through Wednesdays. 30-day Mortality
Floor group: Friday vs. MondayWednesday ICU group: Friday vs. MondayWednesday Outpatient group: Friday vs. MondayWednesday
30-day Morbidity
p-value
Odds Ratio
95% CI of Odds Ratio
p-value
Odds Ratio
95% CI of Odds Ratio
0.0231
1.364
(1.043, 1.782)
0.0039
1.157
(1.048,1.278)
0.2597
1.214
(0.867, 1.699)
0.7897
1.024
(0.858,1.223)
0.4858
0.788
(0.404, 1.539)
0.5361
0.945
(0.791,1.130)
131. ANALYSIS OF SURGICAL ERRORS IN CLOSED MALPRACTICE CLAIMS AT FOUR LIABILITY INSURERS. Gawande AA; Brigham and Womens Hospital Background: The relative importance of the different factors that can cause surgical error is unknown. Analysis of malpractice claims files may identify the leading factors. Methods: From 4 liability insurers, we randomly selected 444 closed claims alleging surgical error. Surgeon-reviewers examined the medical and insurance files for each claim to systematically determine whether an injury due to surgical error had occurred and, if so, what cognitive-, system-, and patient-related factors contributed. Detailed descriptive information