The fate of endoleaks after endovascular aneurysm repair and the impact of oral anticoagulation on their persistence.

BACKGROUND: The impact of anticoagulation on late endoleaks after endovascular aneurysm repair (EVAR) is unclear despite multiple investigators studying the relationship. The purpose of this study was to determine if long-term anticoagulation impacted the development of late endoleaks and if specific anticoagulants were more likely to exacerbate the development of endoleaks. METHODS: Using the Society for Vascular Surgery Vascular Quality Initiative database, patients undergoing EVAR between 2003 and 2019 for abdominal aortic aneurysms were evaluated. Patients were divided into two groups: those without a late endoleak and those with a late endoleak. Bivariate analysis was performed to assess preoperative, intraoperative, postoperative, and long-term follow-up variables. A multivariable analysis was done to determine associations of independent variables with late endoleaks. Patients were further subcategorized based on anticoagulation status before and after EVAR, specific type of anticoagulation, and the presence of an index endoleaks (diagnosed at the time of EVAR) to determine the subsequent frequency of late endoleaks. RESULTS: A total of 29,783 patients were analyzed with 2169 (7.3%) having a late endoleak identified. Several risk factors were related to late endoleaks, including anticoagulation before and after EVAR (odds ratio [OR], 4.23; 95% confidence interval [CI], 2.57-6.96; P < .001), anticoagulation after EVAR (OR, 1.88; 95% CI, 1.43-2.49; P < .001), and index endoleak (OR, 1.45; 95% CI, 1.26-1.66; P < .001). The frequency of late endoleaks in patients anticoagulated before and after EVAR and after EVAR as compared with those never anticoagulated was 16.89% and 14.40% vs 6.95%, respectively (both P > .001). No difference in late endoleaks were noted for patients treated with warfarin and novel oral anticoagulants. The most common type of index and late endoleak identified was type II, but patients with type I, type II, and type IV index endoleaks were more commonly found to have type I, type II, and type IV late endoleaks, respectively. The frequency of late endoleaks in patients with both an index endoleak and anticoagulation after EVAR was 20.42% as compared with patients with only anticoagulation after EVAR (14.63%; P = .0015) and with patients with index endoleaks not anticoagulated (10.06%; P < .00001). CONCLUSIONS: Late endoleaks were more common in patients treated with anticoagulation after EVAR. No difference in late endoleak frequency was detected between anticoagulation with warfarin and novel oral anticoagulants. Patients on anticoagulation and those with an index endoleak were at a higher risk of having a late endoleak.

Factors affecting patients' functional status and their impact on outcomes of ruptured abdominal aortic aneurysms.

OBJECTIVE: Despite advances in surgical techniques, ruptured abdominal aortic aneurysm (rAAA) remains associated with extremely high mortality. Several preoperative risk factors have been shown to predict poor prognosis after rAAA repair. Notably, a preoperative dependent functional status has previously been shown to be associated with poor outcomes after several vascular surgery procedures. The purpose of this study was to examine the effect of preoperative functional status on postoperative outcomes for patients undergoing repair of rAAA. METHODS: Patients with rAAA were identified in the American College of Surgeons National Surgical Quality Improvement Program database for the year 2013. The patients' demographics, procedural data, and postoperative outcomes were analyzed. Patients were then divided into two groups: dependent functional status and independent functional status. Preoperative variables and outcomes were compared between these two groups. A multivariate logistic regression analysis was then conducted to assess independent risk factors that predispose to dependent functional status. RESULTS: A total of 1239 patients underwent repair of rAAAs. Of these, 34 patients did not have a recorded functional status. The total number of patients analyzed was therefore 1205 (male, 78%; female; 22%; group I, dependent functional status, n = 62 [5%]; group II, independent functional status, n = 1143 [95%]). Bivariate analysis identified the following variables as having a significant association with dependent functional status: age >80 years (odds ratio [OR], 8.70; confidence interval [CI] 1.18-64.43; P = .002), female sex (OR, 2.89; CI, 1.71-4.87; P < .001), dyspnea (OR, 3.77; CI, 2.0-7.13; P < .001), dialysis (OR, 7.55; CI, 3.21-17.73; P < .001), insulin-dependent diabetes mellitus (vs nondiabetic: OR, 3.76; CI, 1.39-10.21; P = .033), current smoker (OR, 0.41; CI, 0.22-0.77; P = .005), and hypertension (OR, 2.86; CI, 1.4-5.87; P = .004). Preoperative functional status had no effect on the following postoperative outcomes: surgical site infection, cardiac arrest, unplanned intubation, readmission, return to the operating room, and death. Dependent functional status was associated with increased length of hospital stay (group I, median of 10 days; group II, median of 7 days). CONCLUSIONS: Dependent functional status is considered to be strongly associated with poor outcomes after surgical operations. Our study shows that functional status has little or no bearing on the outcomes of operations for rAAA and that preoperative dialysis, female sex, advanced age, and dyspnea are strong predictors of dependent functional status. Dependent functional status should not be used to exclude patients with rAAA from being offered surgical treatment.

Outcomes with and without axillary node dissection for node-positive lumpectomy and mastectomy patients.

BACKGROUND: American College of Surgeons Oncology Group Z0011 trial of select node-positive breast cancer patients demonstrated no survival or recurrence differences between SLN/axillary lymph node dissection (ALND) vs SLN. Our comparable node-positive lumpectomy and mastectomy populations should have similar outcomes. METHODS: An Institutional Review Board approved, retrospective review of pathologic SLN (N1) cases was performed. Treatment, recurrence, and survival were collected. Statistics was analyzed via exact chi-square test with Monte Carlo estimation, Kaplan-Meier curves, and log-rank tests. RESULTS: Of 528 node-positive patients, 318 patients met criteria: 28 (21.7%) lumpectomy, 32 (16.9%) mastectomy had SLN; 101 (78.2%) lumpectomy, 157 (83.0%) mastectomy had SLN + ALND. Median age was 57.5 years for SLN and 53 years for SLN + ALND (P = .003). Mean positive nodes were 1.1 for SLN and 1.47 for SLN + ALND (P = .0018). Chemotherapy use differed (SLN = 73.5%, SLN + ALND = 89.7%, P = .0032). Stage and recurrence were higher for SLN + ALND (P = .0001, P = .007). No difference in comorbidities, nodes retrieved, extracapsular extension, radiation, hormone therapy, or overall survival was observed. CONCLUSION: In clinically node-negative breast cancer patients, ALND for N1 disease has no impact on short-term recurrence or survival.