Our practice is a university-based referral center with most referrals coming from
                                            within the university, a local Veterans Affairs medical center and private practice
                                            offices locally and as far away as four neighboring states. Information accompanying
                                            each referral can be quite varied and often included one or a combination of the
                                            following: referral letter, photo, pathology report and clinical notes.
                                        
                                            A system to triage cases into one of two categories, large or small, was developed
                                            based on clinical data, location, size and/ or histology characteristics (Table
                                            1). The system was all-or-none,
                                        

                                            such that if any criteria met the definition of a "large" case, the case was categorized
                                            as such. Data was gathered for cases from July 1, 2008 to June 30, 2009 including:
                                            the triage category, number of stages to tumor clearance and type of repair.
                                        
                                            Data were analyzed to determine the average number of stages to tumor clearance
                                            for the "large" and "small" categories as well as the distribution of types of reconstructions
                                            for each category (Tables 2 and 3).
                                        
                                            Statistical analysis, using the Pearson's chi-square, was utilized in testing the
                                            null hypothesis that there is no relationship between cancer size and the reconstruction
                                            type; in other words, reconstruction type is independent of the cancer size, showed
                                            a P value for the test to be <0.005 (Table 4). This confirms there is a statistically
                                            significant relationship between the cancer size and reconstruction type.
                                        

 
						





