Is Elevated Preoperative Glycosylated Hemoglobin Associated with Postoperative Infections?

S72

Scientific Forum Abstracts

However, it is unclear if the type of diabetes treatment (insulin or oral agents) has any effect on outcomes after EAS. METHODS: This was a matched retrospective cohort study using the ACS NSQIP database. Patients with DM undergoing EAS were divided into insulin and oral agent treatment groups. A 1:1 cohort matching of insulin-treated and oral agent-treated patients was performed. Matching criteria were sex, age, American Society of Anesthesiologists score, BMI category, operative procedure, corticosteroid use, and preoperative acute renal failure, pneumonia, systemic inflammatory response syndrome, sepsis, and septic shock. Outcomes of matched insulin- and oral agent-treated patients were compared with univariable and multivariable regression analysis. RESULTS: There were 7,401 patients with DM who underwent EAS, 3,182 (43%) of whom were insulin-treated and 4,219 (57%) treated with oral agents. Matching resulted in 2,280 matched cases, which formed the basis of this analysis. Insulintreated patients had significantly more overall complications (35.2% vs 30.7%, p¼0.001) and a longer hospital length of stay (LOS, median 9.0 vs 8.0 days, p¼0.003). In regression analysis, insulin-treated patients were more likely to die (adjusted odds ratio [AOR] 1.311, p¼0.011), to develop complications (AOR 1.285, p<0.001), and to have a longer total (AOR 1.128, p¼0.030) and postoperative LOS (AOR 1.094, p¼0.016). CONCLUSIONS: Insulin-treated patients with DM undergoing EAS have worse outcomes than oral agent-treated patients. This finding supports experimental work showing a possible immunosuppressive effect of insulin. Insulin-treated patients should be monitored and treated more intensively in anticipation of potential complications after EAS. International, Prospective Comparison of Open Inguinal Hernia Repair Techniques: Two-Year Quality of Life and Recurrence Outcomes in More than 1,300 Patients Tiffany C Cox, MD, Laurel J Blair, MD, Ciara Huntington, MD, Erwin VanGeffen, MD, PhD, Tanushree Prasad, Amy E Lincourt, PhD, MBA, Brent D Matthews, MD, FACS, Vedra A Augenstein, MD, FACS, B Todd Heniford, MD, FACS Carolinas Medical Center, Charlotte, NC; Jeroen Bosch Ziekenhuis, Eindhoven Area, Netherlands INTRODUCTION: Inguinal hernia repair (IHR) is one of the most common operations in the world. Recurrence and quality of life (QOL) outcomes are the predominant measures of success. Herein, we compare outcomes between IHR operative approaches. METHODS: Prospective, international, multi-institutional data were collected from 2007-2012. The Lichtenstein hernia repair (LichHR) was compared with a 3-dimensional hernia system (Gilbert) and plug&patch (PP). The QOL was quantified by the Carolinas Comfort Scale (CCS) preoperiatively, 1, 6, 12, and 24 months. RESULTS: A total of 1,419 IHRs in 1,341 patients were performed: 348 PP, 178 LichHR, and 893 Gilbert. Demographics included average age 57.415.5 years, 93.5% male, BMI 25.93.7kg/m2,

J Am Coll Surg

and 8.6% recurrent hernias. Operative time for LichHR was 56.421.7 minutes, vs 49.221.1 minutes for Gilbert (p<0.001), and 49.218.9 minutes for PP (p<0.001); no difference for Gilbert vs PP (p¼0.26). Average follow-up was 25.611.5 months (75% 2year). Recurrence, seroma, and infection rates were equivalent for all groups (p>0.05). At 1 month, Gilbert patients had less pain (22.6% vs 32.3%, p¼0.017), mesh sensation (11.7% vs 21%, p¼0.004), and activity limitation (15.5% vs 27.2%, p¼0.001) compared with the LichHR patients. In activity limitation, PP was equivalent to LichHR, but Gilbert was superior to PP (15.5% vs 22.8%, p¼0.004). At 2 years, QOL for LichHR and PP patients were equivalent, yet Gilbert patients had significantly less pain (7.9% vs 15%, p¼0.02) and mesh sensation (10.1% vs 18.8%, p¼0.01) than LichHR. Gilbert was also superior to PP in pain (7.9% vs 14.1%, p¼0.01) and mesh sensation (10.1% vs 15.3%, p¼0.04). CONCLUSIONS: The repair techniques had equal recurrence rates, and the LichHR and PP also demonstrated equal short- and longterm QOL. The Gilbert repair, however, showed superior 1-month and 2-year QOL outcomes compared to LichHR and PP. The variance between the most common IHR techniques appears to be QOL. Is Elevated Preoperative Glycosylated Hemoglobin Associated with Postoperative Infections? Joseph M Blankush, Aron Y Soleiman, Trung L Tran, MD, I Michael Leitman, MD, FACS Icahn School of Medicine at Mount Sinai, New York, NY INTRODUCTION: Perioperative hyperglycemia is associated with poor outcomes and infectious complications. Elevated blood glucose negatively affects the immune response, and preoperative glycemic control has been shown to reduce postoperative infections. However, the correlation between elevated preoperative glycosylated hemoglobin (HbA1c) and postoperative infections remains unclear. METHODS: Adult patients undergoing non-emergent procedures across all surgical subspecialties, from January 2010 to July 2014, had HbA1c measured during their preoperative assessment. Patients were divided into those with a preoperative HbA1c<6.5% and those with HbA1c 6.5%. Data from the first 1,100 consecutive patients in each group were captured. Patient charts were reviewed for evidence of postoperative infection (surgical site infection [SSI], pneumonia [PNA], urinary tract infection [UTI], and/ or deep wound/surgical space abscess formation as defined by CDC criteria) within 30 days of surgery. RESULTS: To date, 1,100 and 530 records have been analyzed in the <6.5% and 6.5% HbA1c groups, respectively. Postoperative infections occurred in 4.9% of the 6.5% HbA1c group vs 3.8% in the <6.5% HbA1c group (odds ratio [OR] 1.30; 95% CI, 0.79-2.14; p¼0.30). Specific infectious etiologies, including surgical site infection (OR 1.29; 95% CI 0.64-2.60; p¼0.47), PNA (OR 2.61; 95% CI, 0.70-9.76; p¼0.15), UTI (OR 1.14; 95% CI, 0.54-2.41; p¼0.72), and surgical space abscess (OR 1.39; 95% CI, 0.23-8.32; p¼0.72) were more common in the elevated HbA1c population but did not represent a statistically significant difference. This was also true for

Vol. 221, No. 4S1, October 2015

Scientific Forum Abstracts

overall infections in patients with considerably elevated HbA1c (8.0% HbA1c; OR 1.37; 95% CI, 0.69-2.70; p¼0.37). CONCLUSIONS: Elevated preoperative HbA1c is not significantly associated with postoperative infections. Lightweight vs Midweight Polypropylene Mesh in 948 Open Ventral Hernia Repairs (OVHR) Laurel J Blair, MD, Ciara Huntington, MD, Tiffany C Cox, MD, Tanushree Prasad, Amy E Lincourt, PhD, MBA, Vedra A Augenstein, MD, FACS, B Todd Heniford, MD, FACS Carolinas Medical Center, Charlotte, NC INTRODUCTION: In 2005, lightweight mesh (LWM) was introduced with the promise of improved quality of life (QOL) and equal recurrence rates. Little information proving or denying either has been put forth. Our aim was to compare the long-term outcomes of LWM vs mid-weight mesh (MWM) in open ventral hernia repair (OVHR). METHODS: Prospective data were collected from 2000 to 2014 for all elective OVHR using synthetic LWM or MWM. Emergent, contaminated, and non-LWM or MWM operations were excluded. Carolinas Comfort Scale (QOL) was recorded at 1, 6, and 12 months. Statistical evaluation included Pearson’s chi-square, Fisher’s exact tests, and Wilcoxon 2-sample tests. RESULTS: There were 948 OVHRs, 400 LWM, and 548 MWM. The MWM patients tended to be more complex than LWM patients, with longer operative times, larger defects requiring larger meshes and more component separations, more frequent panniculectomy, and longer inpatient stay (Table, p<0.0001). With 33.6 months average follow-up, despite the case complexity, MWM had fewer recurrences (0.55% vs 5.0%, p<0.0001). In obese patients, LWM had greater recurrence than MWM (5.9% vs 0.9%, p¼0.0015). However, even in nonobese patients, LWM patients continued to have higher recurrence rates (3.1% vs 0.0%, p¼0.01). There was no difference in rates of mesh infection or other postoperative complications (p>0.05)(Table). QOL scores were no different at 1, 6, or 12 months (p>0.05). Table. Patient Characteristics and Operative Details in Elective Open Ventral Hernia Repair, Lightweight vs Midweight Mesh Operative characteristics

Lightweight mesh

Midweight mesh

p Value

Operative time, min Defect size, cm2 Mesh size, cm2 Component separation, % Panniculectomy, % Length of stay days Seroma, % Hematoma, % Wound infection, % Mesh infection, %

154  92 91  135 584  476 13.0 19.6 4.8  6.6 5.0 0.75 4.0 0.3

182  84 201  228 772  438 35.8 29.4 6.0  4.6 7.7 0.36 6.4 1.5

<0.0001* <0.0001* <0.0001* <0.0001* 0.0005* <0.0001* 0.10 0.41 0.11 0.06

*p<0.05 considered significant.

S73

CONCLUSIONS: Lightweight mesh, commonly favored in OVHR based on anecdotal reports of improved postoperative QOL, has equivalent QOL in long-term follow-up and an increased long-term failure rate compared with OVHR with MWM. Despite increased complexity in the MWM group, LWM demonstrates no clinical or QOL advantages and therefore should not be used for OVHR. Liposomal Bupivicaine Reduces Narcotic Use and Time to Flatus in Patients Undergoing Laparotomy Atuhani S Burnett, MD, PhD, Brian Faley, PharmD., Donald A McCain, MD, PhD, FACS, Themba L Nyirenda, PhD, Zubin M Bamboat, MD Hackensack University Medical Center, Hackensack, NJ INTRODUCTION: Narcotic use after laparotomy can prolong ileus, length of stay, and increase complications and hospital costs. Development of a sustained release formulation of bupivacaine (Exparel, Pacira), is a novel option for pain control. Addition of Exparel to multimodal therapy (MMT) to optimize postoperative pain has never been studied. METHODS: Data were collected prospectively on 61 consecutive patients undergoing laparotomy for a variety of malignancies. All patients were treated with MMT consisting of IV acetominophen, Toradol (Roche), and morphine or Dilaudid (Purdue Pharma) patient-controlled analgesia (PCA). Thirty-one patients were additionally treated with 40 mL of Exparel infiltrated during fascial closure (Table). Endpoints were narcotic use, time to flatus, length of stay, and complications. RESULTS: Both groups had similar demographics, administration of acetominophen, Toradol, and complication rates. Average PCA use for the first 72 hours was 78 mg of morphine for the Exparel group and 112 mg in the control group (p¼0.04). When divided into 24-hour increments, immediate postoperative (0-24 hours) PCA use was similar between groups. However, PCA use for the Exparel group was decreased by 46% during the 24-48 hour period (p¼0.038), and again by 55% during the 48-72 hour period (p¼0.019). Time to flatus was decreased to 3.0 days in the Exparel group compared with 4.0 days in control group (p¼0.005). There was a trend toward shorter stay in the Exparel group (p¼0.83). Table. Average Postoperative Morphine Use Time period, h

0-24 24-48 48-72 0-72

MMT, mg (n¼30)

MMT + Exparel, mg (n¼31)

p Value

49.4 41.6 22.7 112

44.9 22.5 10.2 78

0.32 0.038 0.019 0.04

MMT, multimodal therapy.

CONCLUSIONS: Exparel use decreases postoperative narcotic use, time to flatus, and may reduce length of stay. Prospective randomized trials are needed to verify whether Exparel should be considered a standard component of MMT to minimize pain after laparotomy.