Conclusions: Whole gene sequencing of the PTEN gene from apooled tumor sample has shown promising preliminary results.Known PTEN variants in pooled samples of HCC patients werereadily detectable using both the 454 GS and Solexa sequencingplatforms.

QS446. A COMPREHENSIVE ELECTRONIC RESEARCHRECORD SYSTEM FOR PRE-CLINICAL ASSESS-MENT OF CARDIOVASCULAR AND SURGICAL DE-VICES. Karen R. Wasiluk1, Andrew L. Rivard2, Heather P.Gammon1, Shoeb Mohammad3, Richard W. Bianco1; 1Uni-versity of Minnesota, Minneapolis, MN; 2University of Flor-ida, Gainesville, FL; 3Saba University School of Medicine,Saba, Netherlands Antilles

Introduction: Although electronic medical records have been im-plemented for various aspects of patient care, such as medicalrecords, operative records, progress notes, physician orders, patientinstructions, and discharge summaries, electronic medical recordsare not widely used for pre-clinical studies. The main objective ofpre-clinical testing is to provide a safety assessment for use inhuman patients. Under the current regulatory guidelines, thesestudies must be performed using Good Laboratory Practices (GLP),which involve extensive documentation of study-related data andarchival control appropriate for independent auditing. Methods: Webuilt a relational database using Microsoft Access 2003 based uponexisting GLP-compliant paper records. We established data integrityat the domain, entity, and referential level structuring and data-design independence. For data consistency and accuracy, we imple-mented interface design, validation and relationship rules withpassword-protection at both the user and administrator levels. Toalign our pre-clinical studies with subsequent clinical studies weincorporated use of the American Medical Association Current Pro-cedural Terminology (CPT) codes. Results: We created 100 formsand sub forms based upon the paper forms currently used to docu-ment our pre-clinical studies. Examples of these forms include pre-surgical histories, surgical procedure data, operative notes, labora-tory tests, and pathology. We linked tables by a unique animalidentification number, allowing the reuse of data in other associatedforms and structured queries. We obtained proof of concept by ret-rospective analyzing data entered from 5 studies encompassing 105implanted mechanical or tissue valves. In addition, we developedsecure access protocols to validate user identity and to create anaudit trail of the data entered, complete with error codes, to complywith GLP standards for electronic records established by the Foodand Drug Administration (FDA) and International Standards Orga-nization (ISO) 5840. Conclusions: We are the first to develop anelectronic medical record designed solely for pre-clinical researchpurposes, using a commercially available program. This databaseincorporates features facilitating compliance with regulatory guide-lines and allows for thorough and expedited quality assurance au-diting of GLP compliance. The electronic format allows for rapidanalysis of the data and preparation of reports or manuscripts. Thisdatabase facilitates comparison of data obtained from testing newdevices with historical data, allowing for a risk-based data analysis.This database is modifiable, both to meet changing needs as well asregulatory requirements. Although designed for pre-clinical use, thisdatabase could easily be adapted for clinical applications.

QS447. TWO-PORT LAPAROSCOPIC APPENDECTOMY: MIN-IMIZING THE MINIMALLY INVASIVE APPROACH.Lucian Panait1, Robert L. Bell2, Andrew J. Duffy2, Kurt E.Roberts2; 1Saint Mary’s Hospital, Waterbury, CT; 2Yale Uni-versity School of Medicine, New Haven, CT

Introduction: Three laparoscopic ports are traditionally required tocomplete a laparoscopic appendectomy. The purpose of this report isto describe a novel, innovative approach in which appendectomy can

be safely performed through a two-port technique. Methods: Eightconsecutive patients were prospectively assigned to undergo two-port laparoscopic appendectomies for presumed appendicitis at ourinstitution in June of 2007. Average age was 35 years. The techniqueinvolves the placement of a 12-mm infraumbilical port for the work-ing instruments and a 5-mm left lower quadrant port for the camera.A stay suture is tied in a loop on the posterior fascia of the anteriorabdominal wall, in the right upper quadrant. A pretied loop sutureplaced on the appendix is passed through the stay suture and thenthrough the port to the outside of the abdomen. The loop on theinside of the abdominal wall provides a fulcrum effect for retraction.This technique allows exposure of the base of the appendix andcompensates for the lack of the third port usually required for theretraction of the appendix. The mesoappendix and appendix arestapled and removed from the abdomen in an extrication bag. Re-sults: All eight laparoscopic procedures were completed withoutdifficulty. The mean operative time was 64 minutes. Length of hos-pitalization varied between 0 and 1 day. No immediate or late post-operative complications were encountered. Conclusion: In an agewhen surgeons are focusing on endo- and transluminal approachesfor access to the abdominal cavity, we favor laparoscopy for theenhanced exposure, instrument diversity and overall patient safety.Two-port laparoscopic appendectomy is a safe, novel approach andtechnique, which minimizes minimally invasive surgery to a newlevel with decreased invasiveness and better cosmesis.

QS448. FROM MOLECULE TO MAN: MULTI-SCALE TRANS-LATIONAL AGENT BASED MODELING OF INFLAM-MATION. Gary An; Northwestern University, Chicago, IL

Inflammation is a central, modulating process in many surgicaldiseases, such as sepsis, trauma, transplant immunology, cancer,obesity, and atherosclerosis. However, inflammation is not an inher-ently detrimental process: inflammation is required for successfulimmune response and wound healing necessary in surgical patients.Thus, inflammation is a prototypical example of an internal regula-tory system that exhibits a paradoxical set of behaviors, at oncenecessary and potentially detrimental, that characterize a “complex”system. There is an increasing recognition of the challenge of ma-nipulating biocomplex internal regulatory systems in the face ofdisease, particularly when they are as ubiquitous as inflammation. Itis virtually impossible to effectively account for the “unintendedconsequences” of such an intervention. This is due to a great dealfrom the multi-scalar organization of biological systems (gene )protein ) cell ) tissue ) organ ) organism), another hallmark ofbiocomplexity. Research groups are oriented towards working withinthese levels of organization, but are challenged in applying discov-ered information across their boundaries. This difficulty is related tononlinearities between the generative mechanisms at one level andthe observed phenomenon at the next. Furthermore, research groupsoften work on parallel tracks with little cross-communication, lead-ing to greater challenges in the integration and concatenation ofbasic research information. A critical manifestation of this challengeis the difficulty in the translation of information from the lab to thebedside in the development of medical therapeutic agents. There hasbeen a growing recognition of the need to apply mathematical mod-eling and simulation Methods to address issues of biocomplexity andtranslational knowledge integration. Agent Based Modeling (ABM)is a computational modeling technique that focuses on the behaviorsof the components of a system. ABM is well suited for creatingmodular translational multi-scale models: 1) ABM focuses on classi-fying the behavior of constitutive components in a structure inherentto biology; 2) ABM uses this classification structure to group agentsby similar behaviors and function; 3) Component behavior can oftenbe expressed in conditional statements (“if-then”), therefore the re-sults of wet lab experiments can be translated to the behavioral rulesof agents intuitively; and 4) Heterogeneous individual agent behav-

445ASSOCIATION FOR ACADEMIC SURGERY AND SOCIETY OF UNIVERSITY SURGEONS—ABSTRACTS

ior is aggregated into population behavior that mirrors the behaviorof the higher-hierarchical system as a whole. Herein are presented aseries of ABMs of inflammation developed at multiple levels of res-olution leading up to simulated organ function and organ-organinteractions. These levels include: 1) ABMs of intracellular signalingin response to inflammatory stimuli, 2) ABMs matching in-vitro cellculture response to inflammatory cytokines, 3) ABMs of different celltypes aggregated to simulate tissue/whole organ behavior, 4) ABMslinked to simulate an organ-organ axis of pathophysiology (the gut-lung axis of injury) and finally 5) an overall modeling architecturethat incorporates organ-organ crosstalk and multiple tissues andorgans to create a “virtual patient.” These virtual patients will formthe basis of simulated in-silico clinical trials that will be the vitalcomponent in the future of translational drug design and the devel-opment of personalized medicine.

QS449. HAND-HELD PET PROBES AS A TOOL FOR INTRA-OPERATIVE LOCALIZATION OF CANCER. Vivian E.Strong, Charles Galanis, Christopher Riedl, Valerie Longo,John Humm, Steven Larson, Yuman Fong; Memorial Sloan-Kettering Cancer Center, New York, NY

Introduction: Positron Emission Tomography (PET) detects re-lease of gamma emissions to aid diagnosis and evaluation of cancer.However, correlating preoperative PET images with intraoperativefindings remains a challenge and the smallest deposits detectableare 1cm, making detection of metastatic deposits challenging.Positron emitters used for PET not only release long-range gammaemissions, but also short-range beta emissions that can be detectedwith a novel hand-held probe. We investigated whether beta emis-sions detected by hand-held portable PET probes would correlatewith conventionally detected gamma emission and offer a novel wayto better localize tumor intraoperatively. Methods: Beta and gammaemissions were correlated in multiple murine tumor models (breast,gastric, squamous cell, and pancreatic cancer). Mice (n�10) wereinjected with 4 microCu 18-Fluorodeoxyglucose and underwentmicro-PET imaging. Portable probe measurements of gamma andbeta emission from tumor and normal tissue were made in vitro, invivo, and ex vivo. Tissues were excised, weighed, and cut for auto-radiography and hematoxylin and eosin staining. Results: The por-table PET probe detected beta and gamma emission from all tumors.Areas of detection corresponded to regions of intensity on PET im-aging and to areas positive for tumor by H�E staining. The smallestdeposits found in vivo were 4mm in size. Additionally, gamma andbeta emission exhibited a strong positive correlation (R�0.8).

Beta emission showed a stronger correlation than gamma emissionwith overall tissue radioactivity. Conclusion(s): This study demon-strates that gamma emission detected by conventional PET imagingcorrelates directly with beta emission. Further, this study suggeststhat compared to detection of gamma emission, beta emission detec-tion can offer superior real-time localization of tumor by detecting

smaller deposits than possible with gamma emissions. Intraopera-tive portable PET probe may become a useful way to exploit tumorbiology and PET technology to guide surgical therapy.

QS450. IMPROVED ASSESSMENT OF MELANOMA MARGINUSING ULTRAVIOLET LIGHT. Christopher R. Rouse,Summer R. Youker, Marie Y. Hurley, Cherise M. Cortese,Scott W. Fosko, Eddy C. Hsueh, Frank E. Johnson; SaintLouis University, St. Louis, MO

Introduction: Judging the border of a primary melanoma affects theplan of excision. Currently this relies on examination using visible light.Ultraviolet light (UV) can alter the apparent border, but the outcomesof UV-aided margin assessment have not been reported. Methods: Insitu and invasive primary melanoma lesions were evaluated with bothvisible light and UV, using a hand-held Wood’s lamp. Shallow incisionswere made at the visible border of the lesion using each type of light.The visible-light margin and the UV margin were marked with differ-ent ink colors if they were not identical. The lesions were excised usingthe visible-light margin as a reference. Microscopic margins were eval-uated via standard Methods. The visible-light margin and the UVmargin were compared. The distances between the microscopic marginand both visible macroscopic margins were measured using an ocularmicrometer. Results: 27 lesions were examined; 17 enhanced beyondthe visible-light border with UV. 12 slides from10 lesions were evalu-able, due to early learning-curve difficulties. The maximum distancebetween the UV and visible-light margins was 4.6 � 2.9 mm (mean �SD). In 8 of the 12 slides, the UV margin was closer than the visible-light margin to the microscopic margin; in 4, the visible-light marginwas closer than the UV margin. The distance from the visible-lightmargin to the microscopic margin was 3.6 � 4.3 mm (mean � SD). Thedistance from the UV margin to the microscopic margin was 0.2 � 4.5mm. Conclusions: The UV margin extended past the visible-lightmargin in 17/27 (63%) of lesions. The microscopic margin was moreclosely approximated by the UV margin in 8/12 (67%) and by thevisible-light margin in 4/12 (33%). UV methodology is simple to use,quick, reproducible, and inexpensive. Improving the surgeon’s ability tojudge the microscopic margin by UV might permit a decrease in theallowable margin, possibly improving cosmesis without compromisingcure rates.

SUS NEW MEMBER POSTERS A

P1. BLOOD PRODUCT TRANSFUSIONS IN PEDIATRICTRAUMA PATIENTS ARE NOT ASSOCIATED WITH AN IN-CREASE IN MORBIDITY AND MORTALITY. SUS NEW MEM-BER POSTER SESSION: THU 2/14 4:30 PM. Cynthia Leaphart,Megan Sippey, Kathy Gismondi, Barbara A. Gaines; Children’s Hos-pital of Pittsburgh, Pittsburgh, PA

Abstract not available

P2. THE BUSINESS PRACTICES IN AN ACADEMIC DE-PARTMENT OF SURGERY: CAN WE SURVIVE? C. DanielSmith; Mayo Medical School, Jacksonvile, FL

Abstract not available

P3. OPEN VERSUS LAPAROSCOPIC PYLOROMYOTOMYFOR PYLORIC STENOSIS: A PROSPECTIVE RANDOMIZEDCONTROLLED TRIAL. Shawn D. St Peter, Sr., George W. Hol-comb III, Casey M. Calkins, Sr., John P. Murphy, Sr., Walter S.Andrews, Sr., Ronald J. Sharp, Sr., Charles L. Snyder, Sr., Daniel J.Ostlie; Children’s Mercy Hospitals and Clinics, Kansas City, MO

Abstract not available

446 ASSOCIATION FOR ACADEMIC SURGERY AND SOCIETY OF UNIVERSITY SURGEONS—ABSTRACTS