36.5. Variation in Mesh Utilization for Incisional Hernia Repair: Effect on Recurrence. M. T. Hawn, C. Snyder, S. H. Gray, L. Graham, C. C. Vick; Univeristy of Alabama at Birmingham, Birmingham, AL Background: Mesh placement during ventral incisional hernia re- pair has been shown to result in superior outcomes, however, there is signiﬁcant variation in adoption of this technique. We performed a multi-institutional study to understand how variation in surgical technique inﬂuences outcomes. Methods: This is a retrospective co- hort study of all patients undoing elective open incisional hernia re- pair at 16 VA hospitals between 1998 and 2002. Patients who had a concomitant procedure, laparoscopic repair, pre-existing infection, placement of absorbable mesh or more than one hernia repair during the study period were excluded. Operative notes and post operative course was physician abstracted from the medical record. Hospital rates for type of hernia repair, mesh placement and recurrence were calculated. Bivariate and multivariable linear regression analyses were performed to assess factors associated with hospital recurrence rates. Results: There were 968 procedures, 803 were primary her- nias, and 165 were recurrent hernias. Overall, 586 (60.5%) underwent mesh repair and 382 (39.5%) suture repairs. The rate of mesh place- ment by hospital ranged from 27.6-88%. At a median follow up of 73 months, 217 (22.4%) hernias recurred and ranged from 12% to 38.9% by hospital. Rates of mesh placement and recurrence were both higher for recurrent hernias than primary hernias (see table). On bivariate regression analysis, hospital rate of overall mesh place- ment (R2 ¼ 0.45, p ¼ 0.004), rate of primary repair mesh placement (R2 ¼ 0.37, p ¼ 0.01), and rate of recurrent mesh placement (R2 ¼ 0.66, p < 0.001) were strongly correlated with hospital recur- rence rate. Hospital volume (range 25 to 128 repairs) (R2 ¼ 0.005, p ¼ 0.80) and hospital proportion of recurrent repairs (range 12- 24%) (R2 ¼ 0.02, p ¼ 0.61) were not associated with hospital recur- rence rate. Multivariable linear regression adjusting for volume and case mix found hospital rate of mesh placement to be signiﬁcantly as- sociated with hospital recurrence rate and for every 10% increase in the rate of mesh use, was associated with a 2.6% decrease in recur- rence rate (p < 0.01). Conclusions: Hospitals that adopted a higher rate of mesh repair for elective open incisional hernia repair had lower recurrence rates. These data support the efﬁcacy of mesh repair pre- viously proven in clinical trials is highly translatable to effective prac- tice in the ﬁeld. Further studies on attribute risk of complications to mesh placement are ongoing. 36.6. Acute Care Surgery Performed by Sleep Deprived Residents: Are Outcomes Affected? A. Yaghoubian, 1 A. H. Kaji, 1 B. Ishaque, 1 J. Park, 1 D. K. Rosing, 1 S. L. Lee, 2 B. E. Stabile, 1 C. de Virgilio 1 ; 1 Harbor-UCLA Medical Center, Torrance, CA; 2 Kaiser Permanente, Los Angeles Medical Center, Los Angeles, CA Background: The Institute of Medicine recently recommended fur- ther reductions in resident duty hours, including a 5-hour rest time for on call residents after 16 hours of work, as a way of providing bet- ter protection for patients against fatigue-related errors. Yet no data is available regarding outcomes of operations performed by surgical trainees working beyond 16 hours in the current 80-hour workweek era. Methods: A retrospective review of all laparoscopic cholecystec- tomies (LC) and appendectomies performed by surgery residents at a public teaching hospital from July 2003 through March 2009. Oper- ations after 10 PM were performed by residents who began their shift at 6 AM and had thus been working 16 hours. A comparison was performed with operations performed between 6 AM and 10 PM (Day- time) and 10 PM and 6 AM (Nighttime). Outcome measures were total complications, bile duct injury, conversion to open operation, length of surgery, and mortality. Results: Over the 7-year study period, 2908 LC and 1726 appendectomies were performed. Appendectomies were performed laparoscopically in 73% in patients in both time pe- riods. On multivariable analysis, there was no difference in outcomes between the two groups. Conclusion: The two most commonly per- formed operations performed at night by sleep deprived residents have similar favorable outcomes compared to those performed during the day. Instituting a 5-hour rest period at night is unlikely to improve outcomes of these commonly performed operations. 36.7. Expensive ‘‘Olives’’ - Does Hospital Type Affect Pyloro- myotomy Outcomes? Analysis of the Kids’ Inpatient Database. M. V. Raval, 1 M. E. Cohen, 2 K. A. Barsness, 3 D. J. Bentrem, 4 J. D. Phillips, 5 M. Reynolds 3 ; 1 Division of Research and Optimal Patient Care, American College of Surgeons, Division of Pediatric Surgery, Children’s Memorial Hospital; Department of Surgery, Northwestern University, Chicago, IL; 2 Division of Research and Optimal Patient Care, American College of Surgeons, Chicago, IL; 3 Division of Pediatric Surgery, Children’s Memorial Hospital; Department of Surgery, Northwestern University, Chicago, IL; 4 Department of Surgery, Northwestern University, Chicago, IL; 5 Division of Pediatric Surgery, University of North Carolina at Chapel Hill, Chapel Hill, NC Introduction: Pyloromyotomy for hypertrophic pyloric stenosis is the most common surgical procedure performed on infants. Referrals to specialized pediatric hospitals have been advocated to improve outcomes. The purpose of this study was to determine if hospital type affects lengths of stay (LOS), charges, and morbidity. Methods: Patients undergoing pyloromyotomy for hypertrophic pyloric steno- sis were identiﬁed in the Kids’ Inpatients Database from 2000, 2003, and 2006. Freestanding children’s hospitals (CH) were com- pared to children’s units within general hospitals (CHUG) and gen- eral/non-children’s hospitals (GH). Results: Of the 10,969 patients, 24.7% received care at 30 CH, 34.6% received care at 94 CUGH, and 40.7% received care at 662 GH. CH averaged 30 pyloromyotomies per hospital per year of study as compared to 13 at CUGH and 2 at GH (medians: CH 37, CUGH 18, and GH 5). Overall mean LOS was 2.62 (median 2), mean charges were $10,984 (median $8,756), and complication rate was 1.8%. Least squares means adjusted LOS were 2.41 days for CH, 2.75 days for CUGH, and 2.82 days for GH (P < 0.01, Figure A). Least squares adjusted mean charges were $11,160 for CH, $12,284 for CUGH, and $10,197 for GH (P ¼ 0.01, Figure B). CH had the lowest unadjusted complication rate at 1.2% compared to 1.6% at CUGH and 2.2% at GH (overall P < 0.01). GH were more likely to have patients with prolonged LOS (4 days) compared to CH after adjusting for patient and hos- pital factors (OR 1.7, 95%CI 1.2-2.5). After accounting for LOS, CUGH had a higher likelihood of high charges ($11,057) compared to CH (OR 3.4, 95%CI 1.03-11.18). Adjusted mean charges rose from ASSOCIATION FOR ACADEMIC SURGERY AND SOCIETY OF UNIVERSITY SURGEONS—ABSTRACTS 311