Heparin-bonded versus standard polytetrafluoroethylene arteriovenous grafts: A Bayesian perspective on a randomized controlled trial for comparative effectiveness.

BACKGROUND: Heparin-bonded polytetrafluoroethylene grafts were marketed to improve hemodialysis access outcomes but are twice the cost of standard polytetrafluoroethylene. We launched a randomized trial of heparin-bonded polytetrafluoroethylene versus standard polytetrafluoroethylene for hemodialysis access to compare patency. Since the trial began, additional studies were published with heterogeneous findings. We performed an interim analysis by Bayesian methods using prior probability from meta-analysis of existing literature. METHODS: NCT01601873 is a randomized, blinded trial of heparin-bonded polytetrafluoroethylene versus standard polytetrafluoroethylene for dialysis access at 5 sites. Planned sample size was 200 with 1-year primary patency as the primary endpoint. At interim analysis (50% of sample size at 1 year), we also performed a meta-analysis for 1-year primary patency with a random effects model to compute summary rate ratio and standard-error estimates. Meta-analysis estimates formed a prior probability for a Bayesian Cox regression model, and trial data were reanalyzed to develop posterior probability of heparin-bonded polytetrafluoroethylene effectiveness at our hypothesized effect size. Futility analysis was conducted using posterior probability estimates. RESULTS: One hundred and five patients were enrolled at the time of interim analysis. One-year primary patency was 34.9% in the heparin-bonded-polytetrafluoroethylene group vs 32.7% in the standard-polytetrafluoroethylene group (P = .884). Summary rate ratio from the meta-analysis (1,209 patients) was 0.87 favoring heparin-bonded polytetrafluoroethylene (P = .33). Posterior hazard ratio from Cox regression was 0.90 (credible interval 0.70-1.13) favoring heparin-bonded polytetrafluoroethylene, which was not significant. Bayesian posterior probability of the a priori hypothesized 20% better patency with heparin-bonded polytetrafluoroethylene was 24%. Sample size to detect superiority with the small observed effect size would require about 3,800 subjects. CONCLUSION: Current evidence does not demonstrate sufficiently large benefit of heparin-bonded polytetrafluoroethylene over standard polytetrafluoroethylene for dialysis access to justify higher cost. Given similar 1-year patency rates, a conclusive finding of superiority was judged to be infeasible, and the trial was stopped for futility.

Impact of redo sternotomy on proximal aortic repair: Does previous aortic repair affect outcomes?

PURPOSE: Proximal aortic repair (AoR) in the setting of previous sternotomy may be associated with greater risk than primary repair. Our objective was to determine whether redo sternotomy increases the risk of adverse outcomes following proximal aortic surgery. METHODS: We reviewed all proximal AoRs from 1991 to 2014. Outcomes were compared between first-time AoR (non-redo = 1305) and redo AoRs, which were further classified into 3 categories: (1) previous acute type A aortic dissection (AAD) repair (redo-AAD = 146, 8.3%); (2) previous proximal aneurysm repair (redo-aneurysm = 165, 9.4%); and (3) previous cardiac (non-aortic) sternotomy (redo-cardiac = 145, 8.2%). Data were analyzed by contingency tables and multiple regression. RESULTS: In total, 456 of 1761 (25.9%) proximal AoRs had redo sternotomy. Aortic redos (redo-AAD and redo-cardiac) had significantly more connective tissue disorders (P < .001). On presentation, AAD was least common in aortic redos followed by cardiac redos (redo-cardiac) versus non-redos (5% vs 28% vs 31%, P < .001). At reoperation, 190 underwent ascending + hemiarch (21% redo-AAD, 50% redo-aneurysm, 53% redo-cardiac), 140 total arch (64% redo-AAD, 15% redo-aneurysm, 15% redo-cardiac), 110 elephant trunk (52% redo-AAD, 12% redo-aneurysm, 11% redo-cardiac), 159 AVR (36% redo-AAD, 42% redo-aneurysm, 25% redo-cardiac), and 100 aortic root (34% redo-AAD, 22% redo-aneurysm, 10% redo-cardiac). Except for pulmonary, redo sternotomy did not increase risk of postoperative complications. Thirty-day mortality after redo sternotomy was 14%-the greatest among cardiac redos. Over a median follow-up of 13 years, non-redos had significantly greater long-term survival (P < .001). Coronary artery disease was a significant predictor of mortality (P < .001). After adjustment for coronary artery disease, cardiac redos had the greatest long-term mortality risk (hazard ratio, 1.43, P < .005). Previous AoR did not significantly add risk above redo sternotomy alone (P = .734). CONCLUSIONS: Redo sternotomy is associated with increased risk for short- and long-term mortality after proximal aortic repair. Despite need for extensive repair, previous proximal aortic (for aneurysm or AAD) repair did not add further risk above that attributable to redo sternotomy.

Superficial Femoral Artery Occlusion Reduces Aortofemoral Bypass Graft Patency.

OBJECTIVES: The effect of superficial femoral artery (SFA) occlusion on long-term patency of aortofemoral bypasses (AFBs) for aortoiliac occlusive disease (AIOD) was examined. METHODS: The AIOD database was reviewed to identify risk factors for AFB failure. The status of the SFA at AFB procedure was categorised as patent; diseased treated (DT), if the SFA occlusion was intervened on concomitant to AFB; or diseased untreated (DU), if the SFA was occluded but not revascularised. Censoring hierarchies for primary patency and patent graft survival time were constructed. Data were analysed by contingency table, Kaplan-Meier, and Cox regression analysis. RESULTS: Between 2004 and 2015, 122 AFB (9 unifemoral, 113 bifemoral) for AIOD were performed. Seventy-five (61%) were female and the mean age was 60 ± 10 years. At the time of AFB, 50 (41%) had occluded SFAs (DT/DU). Of these, 15 had concomitant SFA revascularisation (i.e., DT) at the time of AFB. Patients with occluded SFAs had greater history of prior aortoiliac/infrainguinal procedures (aortoiliac 54% vs. 22%, infrainguinal 58% vs. 25%, both p < 0.001), Trans-Atlantic Inter-Society Consensus II classification of femoropopliteal type D lesions (78 vs. 10%, p < 0.001), Rutherford 4-6 categories (80% vs. 57%, p = 0.011), and longer hospital stay (median 11 vs. 7 days, p < 0.004). SFA status did not affect 30 day mortality (overall 9%); however, sub-analysis showed DT had significantly higher mortality than DU (p < 0.03). Over a median follow up of 7.7 (IQR 4.3-11.4) years, primary patency at one and five years was 98.3% and 91.2% in patients with patent SFAs, 87.9% and 82.7% in DU, and 72.7% and 43.6% in DT (p < 0.001), respectively. On multivariable analysis, low baseline glomerular filtration rate (HR 1.01, p = 0.022), DT (HR 3.7, p = 0.020), Rutherford 4-6 (HR 9.1, p = 0.048), and occluded SFA (HR 3.9, p = 0.009) adversely affected primary patency of AFBs. Long-term mortality was not different between the SFA status groups (p = 0.279). CONCLUSION: Baseline SFA occlusion predicted a fourfold increased hazard of primary AFB failure. Concomitant SFA revascularisation did not improve AFB durability and was associated with increased in hospital mortality.

Risk of Mortality After Resolution of Spinal Malperfusion in Acute Dissection.

BACKGROUND: Spinal cord ischemia (SCI) may develop in patients presenting with acute aortic dissection. We sought to determine how SCI and its recovery affect outcomes. METHODS: We reviewed patients with SCI in acute type A aortic dissection (ATAAD) and acute type B aortic dissection (ATBAD) from September 1999 to May 2014. SCI was defined as paraplegia or paraparesis present on admission. Monoparesis/plegia, paraesthesia, or numbness was defined as ischemic neuropathy. All ATBAD patients were managed with antiimpulse therapy, with selective intervention for rupture, rapid aortic expansion, malperfusion, or intractable pain. ATAAD patients were managed with urgent proximal aortic replacement. RESULTS: Neurologic symptoms were present in 178 (18.2%) of 978 acute dissections (482 ATAAD and 496 ATBAD). Of these 178 patients, SCI presented in 52 patients (29.2%; 80.1% male; mean age, 57 years). On admission paraplegia was present in 24 (46.2%), paraparesis in 10 (19.2%), paresthesia/numbness in 27 (51.9%), and leg ischemia in 25 (48.1%). Aortic operations were performed in 27 SCI patients (51.9%). Symptom resolution was seen in 30 (57.7%). The 30-day mortality was 19.2% and was significantly less in those with resolution of SCI (6.7% vs 36.4%, p = 0.012). When surgical intervention was required in ATBAD with SCI, mortality was 50% (p = 0.039). SCI and symptom resolution significantly affected overall survival. SCI is associated with significantly increased risk of overall mortality (hazard ratio, 2.9; p < 0.001), and SCI resolution completely offsets this risk (hazard ratio, 0.28; p = 0.003). These effects were consistent between ATAAD and ATBAD (p = 0.554). CONCLUSIONS: SCI in acute aortic dissection portends a poor prognosis. However, reversal of deficits is associated with a long-term survival outcome comparable to patients unaffected with SCI.

Determinants and outcomes of nonoperative management for blunt traumatic aortic injuries.

OBJECTIVE: The natural history and parameters for successful nonoperative management of blunt traumatic aortic injuries (BTAIs) involving the descending aorta are poorly understood. We examined our experience with nonoperative BTAI treatment (anti-impulse, blood pressure) and evaluated for determinants of successful outcomes. METHODS: We performed a review of our institutional prospective trauma registry database for all BTAI patients from 1999 to 2015. Computed tomography angiography was used to classify aortic injuries on the basis of severity: grade I, intimal tear; grade II, intramural hematoma; grade III, aortic pseudoaneurysm; and grade IV, free rupture. Grade IV injuries were excluded from nonoperative management. Baseline characteristics, clinical outcomes, and follow-up lesion resolution were compared within the medically managed cohort and between surgical and nonoperative groups using univariate and multivariable analysis. RESULTS: Among 338 BTAI patients admitted between 1999 and 2015, 67 BTAI patients were managed nonoperatively; 26 (54%) had grade I BTAI, 22 (46%) had grade II, and 2 (4%) had grade III. Both grade III injuries required a late thoracic endovascular aortic repair after initial medical management and were excluded from analysis. In all, 48 were managed with initial medical therapy, and the remaining 19 died on admission or before definitive treatment. Among the 48 medically managed, the median age was 34 years, and 14 (29%) were female. Six of the 48 (12%) were transferred from other facilities. There was no significant difference in baseline characteristics or early outcomes between BTAI grades. Median injury resolution time was 39 days for grade I and 62 days for grade II (P = .03). Compared with a surgical cohort, BTAI grade and Abbreviated Injury Scale score for the chest were the only significant determinants of propensity to operate. CONCLUSIONS: Based on these limited data, it appears that patients with minimal aortic injuries (grades I and II) may be managed medically, with the majority resolving within 8 weeks. Minimal aortic injury is associated with low mortality and excellent intermediate-term outcomes. Further prospective studies are required to validate these findings.

Metabolic Syndrome but not Obesity Adversely Affects Outcomes after Open Aortoiliac Bypass Surgery.

BACKGROUND: Although the incident risk of peripheral artery disease increases in patients with metabolic syndrome, several authors report favorable outcomes in obese patients after arterial bypass surgery. We examine the effect of the so-called "obesity paradox" and metabolic syndrome on outcomes after open aortoiliac bypass surgery. METHODS: We identified patients between 2004 and 2015 who had open surgical bypass for aortoiliac occlusive disease. We excluded patients with endovascular repair and those treated primarily for aneurysmal disease. Variables that were analyzed included preoperative medical history, Trans-Atlantic Inter-Society Consensus Document on Management of Peripheral Arterial Disease II classification, Rutherford classification, intra-operative, and postoperative outcomes. Metabolic syndrome was defined by World Health Organization criteria: diabetes and 2 or more of dyslipidemia, hypertension, and obesity (body mass index > 30 kg/m2). Data were analyzed by stratified Kaplan-Meier and multiple Cox regression for outcomes including long-term mortality and reintervention rate. RESULTS: There were 154 open bypass surgery patients during the study period with a median age of 60 years (interquartile range [IQR] 53-68), median glomerular filtration rate 76.1 mL/min (IQR 54-102), and 58% female prevalence. In all, 53 patients had metabolic syndrome (4%), and 14 patients (9%) were obese but did not have metabolic syndrome. Primary bypass graft patency was 89.0 ± 2.7% at 1 year and 77.4 ± 4.1% at 5 years and was not significantly different between metabolic syndrome, obese, and nonmetabolic syndrome patients. Reintervention rate for the entire cohort was 25.3 ± 3.7% at 1 year and 40.6 ± 4.7% at 5 years. In those with and without metabolic syndrome, reintervention rate at 1 and 5 years was 33.0 ± 6.8% vs. 21.1 ± 4.2% and 56.1 ± 7.9% vs. 30.7 ± 5.4%, respectively (log-rank P = 0.003). In multivariable analyses, metabolic syndrome (hazard ratio [HR] 1.8, P = 0.036) and critical limb ischemia (CLI) (HR: 3.2, P = 0.001) were the only independent predictors of reintervention. Neither obesity nor the individual components comprising metabolic syndrome was a risk for reintervention. Multivariate analysis demonstrated age, female gender, CLI, and nonobesity as the independent risk factors for long-term mortality. CONCLUSIONS: Our study supports the "obesity paradox" that obesity by itself is not a risk factor for reintervention and was a protective factor for mortality after open aortoiliac bypass surgery. Bypass graft patency and major amputation rates were not affected. Although the individual components do not predispose to worse outcome, metabolic syndrome is a constellation of factors that, together, are associated with adverse events.

Injury grade is a predictor of aortic-related death among patients with blunt thoracic aortic injury.

OBJECTIVE: The current Society for Vascular Surgery Clinical Practice Guidelines suggest urgent (<24 hours) thoracic endovascular aortic repair for grade (G) II to G IV blunt thoracic aortic injuries (BTAIs). The purpose of this study was to determine whether some patients may require more emergency treatment. METHODS: We reviewed imaging variables of prospectively collected BTAI patients between 1999 and 2014. We used computed tomographic angiography to classify BTAIs into four categories: G I, intimal tear; G II, intramural hematoma; G III, aortic pseudoaneurysm; and G IV, free rupture. Specific examination of G III injuries was undertaken in an effort to predict aortic-related mortality (ARM) before repair. For this subset, we examined pseudoaneurysm size, lesion/normal aortic diameter ratio, and mediastinal hematoma location and size. RESULTS: Among 331 patients with BTAIs, 86 died before imaging. Admission computed tomographic angiography was available for 205 patients (71.2% male; mean age, 39.3 years) with BTAIs (24 G I, 49 G II, 124 G III, 8 G IV). The mean Injury Severity Score was 35.6, and 22.4% had hypotension (<90 mm Hg). Overall mortality was 11.2% (G I/G II, 4.1%; G III/G IV, 15.3%; P = .02). ARM was 2.4% (G I/G II, 0%; G III/G IV, 3.8%; P = .09). ARM was significantly greater in G IV (3 of 8 [37.5%]) than G III (2 of 124 [1.6%]) vs G I/II (0 of 73 [0%]) injuries (P < .0001). Medical management alone was used in 53 (20 G I, 18 G II, 13 G III, and 2 G IV). Open repair was performed in 51 (3 G I, 9 G II, 36 G III, and 3 G IV) at a mean time to repair (TTR) of 10.6 hours. Thoracic endovascular aortic repair was conducted for 101 patients (1 G I, 22 G II, 75 G III, and 3 G IV) at a mean TTR of 9.4 hours. Median TTR for the overall population of BTAI patients was 24.0 hours from admission. (G I, 64.5 hours; G II, 24.0 hours; G III, 19.7 hours; and G IV, 3.5 hours). ARM occurred in four of five patients before planned repair (2 G III and 2 G IV), 7.0 ± 3.6 hours from admission. No G I/II ARM occurred. Among eight G IV injuries, there were three ARMs. Focus on G III injuries through regression analysis demonstrated that early clinical/imaging variables (eg, mediastinal hematoma dimensions and lesion/normal aortic diameter ratio) were not significant predictors of ARM. CONCLUSIONS: Injury grade is a predictor of ARM among patients with BTAIs. Aggressive use of the current Society for Vascular Surgery Clinical Practice Guidelines at a busy level I trauma center resulted in low rates of ARM. In this setting, identification of additional physiologic and radiographic criteria indicating the need for emergency (vs urgent) repair of aortic pseudoaneurysms remains elusive.

Intentional left subclavian artery coverage during thoracic endovascular aortic repair for traumatic aortic injury.

BACKGROUND: Thoracic endovascular aortic repair (TEVAR) is widely used for treatment of traumatic aortic injury (TAI). Stent graft coverage of the left subclavian artery (LSA) may be required in up to 40% of patients. We evaluated the long-term effects of intentional LSA coverage (LSAC) on symptoms and return to normal activity in TAI patients compared with a similarly treated group whose LSA was uncovered (LSAU). METHODS: Patients were identified from a prospective institutional trauma registry between September 2005 and July 2012. TAI was confirmed using computed tomography angiography. The electronic medical records, angiograms, and computed tomography angiograms were reviewed in a retrospective fashion. In-person or telephone interviews were conducted using the SF-12v2 (Quality Metrics, Lincoln, RI) to assess quality of life. An additional questionnaire was used to assess specific LSA symptoms and the ability to return to normal activities. Data were analyzed by Spearman rank correlation and multiple linear and logistic regression analysis with appropriate transformations using SAS software (SAS Institute, Cary, NC). RESULTS: During the study period, 82 patients (57 men; mean age 40.5 ± 20 years, mean Injury Severity Score, 34 ± 10.0) underwent TEVAR for treatment of TAI. Among them, LSAC was used in 32 (39.5%) and LSAU in 50. A group of the LSAU patients (n = 22) served as matched controls in the analysis. We found no statistically significant difference in SF-12v2 physical health scores (ρ = -0.08; P = .62) between LSAC and LSAU patients. LSAC patients had slightly better mental health scores (ρ = 0.62; P = .037) than LSAU patients. LSAC patients did not have an increased likelihood of experiencing pain (ρ = -0.0056; P = .97), numbness (ρ = -0.12; P = .45), paresthesia (ρ = -0.11; P = .48), fatigue (ρ = -0.066; P = .69), or cramping (ρ = -0.12; P = .45). We found no difference between groups in the ability to return to activities. The mean follow-up time was 3.35 years. Six LSAC patients (19%) died during the follow-up period of unrelated causes. CONCLUSIONS: Intentional LSAC during TEVAR for TAI appears safe, without compromising mental or physical health outcomes. Furthermore, LSAC does not increase the long-term risk of upper extremity symptoms or impairment of normal activities.