Complication profile, failure to rescue, and mortality following elective endovascular aortic aneurysm repair.

INTRODUCTION: Understanding the relationship between patient risk factors, postoperative complications, and morbidity and mortality is important when considering elective endovascular aortic aneurysm repair (E-EVAR) performed to prevent aneurysm rupture mortality. We aimed to stratify complications in E-EVAR and explore their relationship with postoperative death. METHODS: E-EVAR cases from 2012 NSQIP were identified. 30-day complications were categorized as major (MAJCX) or minor (MINCX) using the Clavien-Dindo classification. Failure to rescue (FTR) was defined as death following a complication. Univariate and multivariate analyses were performed to identify associations between patient risk factors, complications, and mortality. Significance set at P < 0.05. RESULTS: 3344 E-EVAR's were analyzed, with 155 (4.6%) MINCX, 106 (3.2%) MAJCX, and 39 (1.2%) mortality. Significant univariate risk factors differed between MINCX (preoperative dyspnea 27% vs 19%, COPD 32% vs19%, HTN 87% vs 79%, functional dependence 9% vs 3%) and MAJCX (female sex 33% vs 18%, preoperative diabetes 30% vs 17%, dyspnea 40% vs 19%, COPD 46% vs 20%, anticoagulant use 20% vs 11%, and functional dependence 13% vs 3%). 24 of 39 (62%) of deaths were preceded by a complication. FTR was more frequent following MAJCX than MINCX (16% vs 4.5%, P = 0.002), and occurred most commonly after renal failure with dialysis (33% mortality with complication), cardiac arrest (33%), septic shock (22%), and reintubation (22%). Independent predictors of MAJCX included female sex (OR 2, P = 0.001), COPD (OR 2, P = 0.009), and anticoagulant use (OR 2, P = 0.001). Mortality was independently predicted by MAJCX (OR 29, P < 0.001), MINCX (OR 8, P < 0.001), and preoperative renal failure (OR 11.6, P < 0.001). CONCLUSION: The majority of deaths within 30 days following E-EVAR are preceded by a complication; both MAJCX and MINCX predict mortality. FTR is more common after MAJCX; prevention efforts should take this into account. Identified risk factors should be taken into consideration when considering E-EVAR.

A quality improvement project to improve inferior vena cava filter retrieval.

OBJECTIVE: Poor retrieval rates for retrievable inferior vena cava filters (R-IVCFs) have been reported throughout the literature, with poor follow-up a common cause. In 2009, we reported a retrieval rate of 18% despite an initial follow-up rate of 85%. Use of a registry has been shown to improve retrieval rates. As a quality improvement project, in May 2012, the vascular surgery fellowship implemented a reiterative registry to track R-IVCFs placed at Walter Reed National Military Medical Center to improve retrieval rates. We report the results in 125 patients after 38 months. METHODS: Patients receiving an R-IVCF were entered into a registry. All patients were reviewed monthly using an electronic health record. When there was no longer an indication for the R-IVCF, the patient was scheduled for an outpatient appointment with a vascular surgeon followed by retrieval. Rates of retrieval, technical success, dwell time, indication, complications, and demographics were collected. RESULTS: There were 125 R-IVCFs placed between May 2012 and June 2015; 52 filters were placed for therapeutic and 73 for prophylactic indications. Our follow-up rate improved to 94%. A total of 79 filters were retrieved (63% absolute retrieval rate). Excluding patients who died before retrieval and patients with a permanent indication, 77% of filters were retrieved. The average dwell time was 101.5 days (7-460 days), and 63% of successful R-IVCF retrievals were within 3 months of placement. Technical success for retrieval was 92%. There were two major complications from retrievals (1.5% of retrievals). CONCLUSIONS: The creation of an R-IVCF registry promoted ongoing follow-up with patients. In our earlier experience, retrieval rates were poor despite a high follow-up rate. The use of a reiterative registry improved our retrieval rate by 45% and increased our follow-up rate to 94%. These results emphasize the importance of repetitive follow-up for R-IVCFs. Despite a follow-up rate >90%, around a third of R-IVCFs were not retrieved.

Permacol interposition graft as an alternative to vein in contaminated wounds using a rabbit model.

BACKGROUND: Vascular injuries are common in trauma and often involve massive soft tissue injury and segmental arterial loss. Current practice uses either autogenous vein or polytetrafluorethylene (PTFE) for interposition grafting in arterial injuries. Decision making between autogenous or synthetic conduit pivots around the physiological state of the trauma patient. Vein is known to increase operative times in an already physiologically depleted patient, whereas synthetic graft can be simply pulled from the shelf. However, when used in contaminated wounds, PTFE is prone to chronic infection and subsequent graft failure. An alternative synthetic conduit resistant to infection would be ideal for such situations. Permacol (Tissue Science Laboratories, Inc, Andover, MA), a biosynthetic material, has demonstrated resistance to bacterial contamination in contaminated hernia repairs. When fashioned into a tubular structure, this material may be useful as an alternative vascular conduit in contaminated trauma wounds. METHODS: New Zealand white rabbits were randomized to one of 4 groups: Permacol graft (P) without bacterial contamination (n = 9), Permacol graft with bacterial contamination (CP; n = 9), autogenous vein graft without bacterial contamination (V; n = 9), or autogenous vein with bacterial contamination (CV; n = 9). All groups then underwent interposition grafting of the right common carotid artery. Grafts were contaminated by applying Staphylococcus aureus (1 × 10(5) colonies/0.1 mL) directly to the exposed surface of the graft on completion of the arterial repair. Each graft was then excised at day 42, and segments were collected for histologic evaluation, bacterial counts, and real-time polymerase chain reaction. RESULTS: Of the 36 rabbits used in this study, 3 animals in the CV group died within 72 hr of surgery. There was no difference in early mortality between P and V (0% vs. 0%; P = 1.0); however, early mortality was higher in the CV compared with the CP group (33% vs. 0%; P = 0.023). At 42 days, histologic evaluation of graft patency demonstrated no difference between P and V (67% vs. 33%; P = 0.157); however, patency was higher in CP than CV (56% vs. 12%; P = 0.040). In addition, no difference was found between the 2 contaminated groups in regard to the number of bacteria present on each graft material. CONCLUSIONS: Permacol as an interposition graft is a feasible alternative to vein in a contaminated setting and shows resistance to infection in a rabbit model. Future studies are needed to evaluate this material in larger animal models.

Upper gastrointestinal swallow study following bariatric surgery: institutional review and review of the literature.

BACKGROUND: The aim of this study was to determine the efficacy of routine upper gastrointestinal imaging following the three forms of laparoscopic bariatric surgery completed at our institution (laparoscopic Roux en Y gastric bypass (LRYGB), laparoscopic sleeve gastrectomy (LS), and laparoscopic adjustable gastric banding (LAGB)). METHODS: Radiograph reports were reviewed from the period of January 2005 to July 2010. During that time, 129 patients underwent LRYGB, 209 underwent LS, and 12 patients underwent LAGB. Of those patients, 120 LRYGB patients, 188 LS patients, and 11 LAGB patients underwent upper gastrointestinal studies on postoperative days (POD) 1 or 2. RESULTS: Of the 319 total patients who underwent UGI, no contrast leaks were found. One LRYGB patient was found to have stenosis of the jejunojejunal anastomosis and was taken to the operating room for revision. A total of ten patients went on to develop leaks: four LRYGB patients, six LS patients, and zero LAGB patients. CONCLUSIONS: The results of our study show that a positive UGI study for stricture has a specificity of 100 %. In terms of leak, which offers a much higher risk of significant morbidity and mortality, UGI was unable to find any on postoperative days 1 or 2. Based on the results of this study, our institution has stopped completing routine UGI on POD 1 following bariatric surgery.

Comparison of staple-line leakage and hemorrhage in patients undergoing laparoscopic sleeve gastrectomy with or without Seamguard.

Laparoscopic sleeve gastrectomy (LSG) has been recognized as a primary procedure for the surgical management of morbid obesity. Staple-line leaks and hemorrhage are two associated complications. Staple-line buttressing materials have been suggested to decrease these complications. When used during LSG, few published papers exist that compare the incidence of leak or hemorrhage to that of nonreinforced staple-lines. The purpose of this study was to compare the incidence of leak and hemorrhage in patients who did and did not receive reinforcement with Seamguard (W.L. Gore & Associates, Flagstaff, AZ). This is a retrospective analysis of patients undergoing LSG. All patients met National Institutes of Health criteria and each had an extensive preoperative evaluation. Data was collected from inpatient and outpatient medical records. Fifty-nine patients received reinforcement and 80 patients did not. There was no significant difference in mean body mass index, age, or gender make-up between the two groups. The overall incidence of leak was 3.60 per cent. The incidence was 3.39 per cent in patients who received reinforcement and 3.75 per cent in those who did not. This was not statistically significant. There was no incidence of staple-line hemorrhage in either group. There is no conclusive evidence that Seamguard reduces staple-line leakage or hemorrhage. Studies involving a larger number of patients are necessary before recommending staple-line reinforcement.

Laparoscopic sleeve gastrectomy: outcomes at a military training center.

Laparoscopic sleeve gastrectomy (LSG) has gained support as a single-staged and stand-alone bariatric procedure. Reports of excess weight loss of 35 to 83 per cent, reduction in comorbidities, and decreased operative morbidity have garnered support for LSG. This study represents an initial outcome analysis of LSG performed solely at a military treatment center. This study is a retrospective analysis of all patients receiving LSG at Dwight D. Eisenhower Army Medical Center from September 2007 to December 2009. The patients were planned for a stand-alone procedure. One hundred and fifteen patients received LSG over this time period with a mean body mass index of 45.5 +/- 6.2 (range 35.1-58.3). The average age was 47.4 +/- 12.5 years. Diabetes mellitus was seen in 47 per cent and 68 per cent of patients had hypertension. The mean and median length of operation was 124 +/- 48 and 115.5 minutes. The mean percentage of excess weight loss was 16.6 +/- 6.40 per cent at 1 month, 31.5 +/- 7.6 per cent at 3 months, 41.2 +/- 13.9 per cent at 6 months, and 53.7 +/- 12.5 per cent at 1 year from surgery. One or more of patient's preoperative diabetic or hypertensive medications were improved postoperatively in 18.7 per cent and 16.3 per cent, respectively. Incidence of major complications occurred in 4.35 per cent of patients in this study to include four leaks (3.4%), one death (0.87%), and 10 readmissions. Midterm analysis of outcomes related to LSG as a single-stage bariatric procedure is promising as long-term outcome data is collected; the efficacy of this procedure as a sole bariatric procedure will continue to be borne out.