Regional Market Competition is Associated with Aneurysm Diameter at the Time of EVAR.

BACKGROUND: Local market competition has been previously associated with more aggressive surgical decision-making. For example, more local competition for organs is associated with acceptance of lower quality kidney offers in transplant surgery. We hypothesized that market competition would be associated with the size of an abdominal aortic aneurysm (AAA) at the time of elective endovascular aneurysm repair (EVAR). METHODS: We included all elective EVARs reported in the Vascular Quality Initiative database (2012-2018). Small AAAs were defined as a maximum diameter <5.5 cm in men or <5.0 cm in women. We calculated the Herfindahl-Hirschman Index (HHI), a measure of physician market concentration (higher HHI = less market competition), for each US census region. Multilevel regression was used to examine the association between the size of AAA at EVAR and HHI, clustering by region. RESULTS: Of 37,914 EVARs performed, 15,379 (40.6%) were for small AAAs. There was significant variation in proportion of EVARs performed for small AAAs across regions (P < 0.001). The South had both the highest proportion of EVARs for small AAAs (44.2%) as well as the highest market competition (HHI 50), whereas the West had the lowest proportion of EVARs for small AAAs (35.0%) and the lowest market competition (HHI 107). Adjusting for patient characteristics, each 10 unit increase in HHI was associated with a 0.1 mm larger maximum AAA diameter at the time of EVAR (95% CI 0.04-0.24 mm, P = 0.005). CONCLUSIONS: Physician market concentration is independently associated with AAA diameter at time of elective EVAR. These data suggest that physician decision-making for EVAR is impacted by market competition.

Practice patterns in arteriovenous fistula ligation among kidney transplant recipients in the United States Renal Data Systems.

BACKGROUND: Arteriovenous fistulas (AVF) and grafts (AVG) have been associated with significant cardiac morbidity that often improves after ligation. However, AV access ligation after kidney transplant (KT) is controversial due to concern for potential long-term allograft failure. We investigated US trends in AV access ligation after KT and the association between ligation and allograft failure. METHODS: All adult Medicare patients on pretransplant hemodialysis with a functioning AVF or AVG who underwent first-time KT were studied using the United States Renal Data Systems (January 2011 to December 2013). Post-transplant AV access ligation was determined using current procedural terminology codes. The incidence of post-transplant AV access ligation was described, and characteristics for patients undergoing ligation vs no ligation were compared. Kaplan-Meier curves and Cox proportional hazard models were then used to determine the association of AV access ligation with long-term allograft failure and all-cause mortality after accounting for patient characteristics, donor characteristics, and variation in transplant center practices. RESULTS: A total of 16,845 patients with functioning AVF/AVG received a KT during the study period. Of these, 779 (4.6%) underwent post-transplant AV access ligation. The proportion of patients who underwent ligation varied substantially between transplant centers, ranging from 0% (43.0% of centers) to >10% (11.0% of centers). Transplant recipients who underwent access ligation were more likely to be female (40.4% vs 36.6%), had lower median body mass index (27.6 vs 28.4 kg/m2), spent longer on dialysis pretransplant (4.2 vs 4.0 years), and were less likely to have renal failure secondary to diabetes compared with other etiologies (25.0% vs 34.9%) (all, P ≤ .03). Patients who underwent ligation were also more likely to have steal syndrome (77.2% vs 4.1%) and AV access infectious or aneurysmal complications (2.7% vs 0.7%) (both, P < .001). After adjusting for donor and recipient characteristics, increasing age (adjusted hazards ratio [aHR], 1.01; 95% confidence interval [CI], 1.00-1.01), increasing years on dialysis (aHR, 1.06; 95% CI, 1.00-1.13), zero human leukocyte antigen mismatch (aHR, 1.82; [95% CI, 1.09-3.05), and steal syndrome (aHR, 41.00; 95% CI, 34.56-48.64) were associated with post-transplant AV access ligation. Black race (aHR, 0.82; 95% CI, 0.69-0.98) and congestive heart failure (aHR, 0.66; 95% CI, 0.54-0.82) were negatively associated with ligation. Three-year allograft failure occurred in 4.9% ± 1.3% transplant recipients who underwent access ligation vs 9.5% ± 0.5% transplant recipients with functioning access (log-rank, P = .30), and was not significantly different between groups after risk adjustment (aHR, 0.81; 95% CI, 0.47-1.40). There was also no significant association between AV access and all-cause mortality after risk adjustment (aHR, 0.84; 95% CI, 0.46-1.54). CONCLUSIONS: Post-transplant AV access ligation is uncommon and generally reserved for patients with steal syndrome. Importantly, ligation is not associated with post-transplant allograft failure, which occurs in less than 10% of patients at 3 years. There also appears to be no reduction in all-cause mortality with AV access ligation. These data suggest that AV access ligation after KT can likely be reserved for access-related complications because the systemic benefits appear to be minimal.

Venous duplex scanning for suspected deep vein thrombosis: results before and after elimination of after-hours studies.

OBJECTIVE: A 24-hour venous duplex ultrasound (VDU) for suspected deep vein thrombosis (DVT) imposes significant resource burdens on a hospital. We hypothesize that termination of after-hours services increases empiric therapy without affecting clinical outcomes. METHODS: A retrospective review of patients evaluated by the emergency department (ED) for suspected DVT in 2005 and 2007. Demographics, empiric treatment, and complications were compared using propensity scores and multivariate regression models. RESULTS: In 2005 and 2007, 318 and 365 patients, respectively, had VDU after referral by the ED. In all, 49 (16%) tests during 2005 were after-hours, with 31 and 25 acute DVTs in 2005 and 2007, respectively. More patients received empiric treatment in 2007: 51 (14%) vs 26 ([8%]; P = .019) and tended to be more likely to have acute DVT: 7 (28%) vs 3 ([10%]; P = .08). We detected no complications from empiric anticoagulation and no difference in outcome. Estimated annual savings were $11 864. CONCLUSIONS: Elimination of around-the-clock VDU can render substantial savings to hospitals without adverse consequence in the management of acute DVT.