Differential presentation in acuity and outcomes based on socioeconomic status in patients who undergo thoracoabdominal aortic aneurysm repair.

OBJECTIVES: Socioeconomic differences can lead to differences in how patients present with surgical conditions. We attempted to determine whether socioeconomic status (SES) affects survival outcomes after thoracoabdominal aortic aneurysm (TAAA) repair. METHODS: We retrospectively reviewed prospectively collected data from 981 TAAA repairs performed on domestic (noninternational) patients between 2006 and 2016. We excluded patients <18 years old (n = 3), those with no available US home address (n = 114), those not within the race and ethnicity categories assessed (n = 30), and those lost to follow-up (n = 6), leaving 832 repairs for analysis. We derived patient SES by using US Census Bureau data to estimate median household income according to patient home address. Patients were grouped into 3 SES groups: high (n = 283), middle (n = 274), and low (n = 275). Multivariable logistic regression modeling was used to identify predictors of operative mortality. Kaplan-Meier curves and Cox proportional hazards regression were used to analyze the association between SES and survival. RESULTS: Operative mortality occurred in 9% (n = 76) of patients. Patients of low SES had greater rates of acute symptoms, dissection, and urgent or emergency TAAA repair. However, lower SES was not an independent predictor of operative death. Kaplan-Meier analysis and Cox proportional hazards modeling did not show a significant difference in mid-term survival by SES. CONCLUSIONS: In our TAAA series from a single, high-volume practice, SES differences did not appear to influence operative mortality rates. In addition, SES was not associated with a difference in mid-term survival. Efforts to understand and ameliorate the greater acuity of presentation in patients of low SES appear worthwhile.

Propensity score analysis in patients with and without previous isolated coronary artery bypass grafting who require proximal aortic and arch surgery.

OBJECTIVE: The risk posed by previous isolated coronary artery bypass grafting (CABG) in patients who require proximal aortic or aortic arch surgery is unclear. We compared outcomes of ascending aortic and arch procedures in patients with and without previous CABG. METHODS: Using propensity scores, we created 2 matched groups of patients who underwent proximal aortic surgery, including total arch repairs, at our institution: 126 patients who underwent isolated CABG before the index operation and 126 without previous CABG. Forty-four percent of aortic operations were emergency procedures. Eighty-six patients had a patent previous left internal mammary graft. We compared outcomes between the 2 groups and calculated Kaplan-Meier survival curves. RESULTS: The following outcomes were recorded for the patients with previous isolated CABG versus no CABG: operative mortality, 15.9% versus 11.1% (P = .3); 30-day mortality, 13.5% versus 7.1% (P = .1); persistent stroke, 6.3% versus 4.8% (P = .6); and renal failure necessitating hemodialysis at discharge, 7.9% versus 4.0% (P = .2). Previous CABG did not independently predict any adverse outcome, even though patients who underwent previous CABG more frequently needed intra-aortic balloon support (P < .01). The P value for the overall intergroup difference in long-term survival was .06. CONCLUSIONS: This is one of the largest studies yet reported to examine the impact of previous isolated CABG on proximal aortic or arch surgery outcomes. Although these results may be specific to aortic centers of excellence, in this complicated patient cohort, previous isolated CABG did not independently predict any adverse outcome. These results could serve as a benchmark for assessing future endovascular therapies.

A 23-year experience with the reversed elephant trunk technique for staged repair of extensive thoracic aortic aneurysm.

OBJECTIVE: The reversed elephant trunk technique permits staged repair of extensive thoracic aortic aneurysm in patients whose distal (ie, descending thoracic and thoracoabdominal) aorta is symptomatic or disproportionately large compared with their proximal aorta (ie, ascending aorta and transverse aortic arch). We present our 23-year experience with the reversed elephant trunk approach. METHODS: Between 1994 and 2017, 94 patients (median age 62 [46-69] years) underwent stage 1 reversed elephant trunk repair of the distal aorta. Fifty-three patients (56%) had aortic dissection, and 31 patients (33%) had heritable thoracic aortic disease. Eighty-eight operations (94%) were Crawford extent I or II thoracoabdominal aortic repairs. Twenty-seven patients (29%) underwent subsequent stage 2 repair of the proximal aorta; 14 patients (52%) required redo median sternotomy. The median time between the stage 1 and 2 operations was 18.8 (4.8-69.3) months. RESULTS: The operative mortality was 10% (9/94) for stage 1 repairs and 4% (1/27) for stage 2 repairs; 1 patient with heritable thoracic aortic disease died after stage 1 repair (1/31, 3%), and 1 patient died after stage 2 repair (1/13, 8%). Two patients (2%) had ruptures after stage 1 repair; 1 resulted in death, and 1 precipitated emergency stage 2 repair. In total, 36 patients (38%) who survived stage 1 repair died before stage 2 reversed elephant trunk completion repair could be performed. CONCLUSIONS: Managing extensive aortic aneurysm with the 2-stage reversed elephant trunk technique yields acceptable short-term outcomes. This technique is useful for the reversed elephant trunk in patients who require distal aortic repair before proximal repair and is particularly effective in patients with heritable thoracic aortic disease. The low number of patients returning for completion repair is concerning. Rigorous surveillance is needed.

Clinical Characteristics and Long-Term Outcomes of Midaortic Syndrome.

BACKGROUND: Midaortic syndrome (MAS) is a rare congenital or acquired condition marked by segmental or diffuse stenosis of the distal thoracic and/or abdominal aorta and its branches. The optimal approach to medical or interventional management of MAS and long-term outcomes in adults are not well defined. We reviewed MAS cases to characterize the natural history of aortic disease, identify prognostic factors, and evaluate the durability of invasive interventions. METHODS: We conducted a retrospective review of patients with MAS who presented to Memorial Hermann Hospital and Baylor College of Medicine between 1997 and 2018. We categorized cases according to demographic and clinical manifestations, etiologies, the extent of aortic involvement, interventions, and vascular outcomes. RESULTS: We identified a cohort of 13 patients with MAS. The etiology of MAS was identified in 6 cases, including genetic syndromes (neurofibromatosis type 1 (2/13), Williams syndrome (1/13), fibromuscular dysplasia (2/13), and Takayasu arteritis (1/13)). Mean age at first documented clinical event was 25.2 (2-67) years, but cases with genetic etiologies presented significantly younger (18.2 years). The most common primary anatomic site was the suprarenal and infrarenal aorta (zones 5-8). Extra-aortic locations involved the renal (4/13), celiac (3/13), and superior mesenteric (3/13) arteries. Clinical manifestations included hypertension (13/13), claudication (9/13), and postprandial abdominal pain (5/13). All patients with available follow-up data underwent at least one surgical or endovascular intervention (range: 1-8). Postoperative complications included renal failure requiring postdischarge hemodialysis and respiratory failure. There were no deaths in long-term follow-up. CONCLUSIONS: MAS is a complex vasculopathy with substantial variability in clinical presentation and anatomic distribution. Extensive disease frequently requires multiple invasive interventions and results in refractory hypertension, which may predict subsequent clinical events. A multidisciplinary approach with long-term monitoring is essential for preservation of end-organ function and quality of life in this debilitating disease.

Spinal cord deficit after 1114 extent II open thoracoabdominal aortic aneurysm repairs.

OBJECTIVE: Crawford extent II repairs are the most extensive thoracoabdominal aortic aneurysm operations and pose the greatest risk of postoperative spinal cord deficit. We sought to examine spinal cord deficit after open extent II thoracoabdominal aortic aneurysm repair to identify predictors of the most serious type: persistent paraplegia or paraparesis. METHODS: We included 1114 extent II thoracoabdominal aortic aneurysm repairs performed from 1991 to 2017. Intercostal/lumbar artery reattachment (n = 959, 86.1%) and cerebrospinal fluid drainage (n = 698, 62.7%) were used to mitigate the risk of postoperative spinal cord deficit. We used univariate and multivariable analyses to examine spinal cord deficit and identify predictors of persistent paraplegia or paraparesis, defined as paraplegia or paraparesis present at the time of early death or hospital discharge. RESULTS: Spinal cord deficit developed after 151 (13.6%) repairs: 86 (7.7%) cases of persistent paraplegia or paraparesis (51 paraplegia; 35 paraparesis) and 65 (6.1%) cases of transient paraplegia or paraparesis. Patients with spinal cord deficit were older (median 68 vs 65 years, P < .001) and had more rupture (6.6% vs 2.2%, P = .002) and urgent/emergency repair (25.2% vs 16.9%, P = .01) than those without. Persistent paraplegia or paraparesis developed immediately in 47 patients (4.2%) and was delayed in 39 patients (3.5%). Urgent/emergency repair (relative risk ratio, 2.31; P = .002), coronary artery disease (relative risk ratio, 1.80, P = .01), and chronic symptoms (relative risk ratio, 1.76, P = .02) independently predicted persistent paraplegia or paraparesis. Reattaching intercostal/lumbar arteries (relative risk ratio, 0.38, P < .001) and heritable disease (relative risk ratio, 0.36, P = .01) were protective. Early and late survival were poorer in those with persistent paraplegia or paraparesis than in those without. CONCLUSIONS: Spinal cord deficit after extent II thoracoabdominal aortic aneurysm repairs remains concerning; survival is worse in patients with persistent paraplegia or paraparesis. The complexity of spinal cord deficit and persistent paraplegia or paraparesis warrant further study.

Is incidental splenectomy during thoracoabdominal aortic aneurysm repair associated with reduced survival?

OBJECTIVE: The effect of incidental splenectomy during thoracoabdominal aortic aneurysm repair is unknown. We hypothesized incidental splenectomy was associated with decreased late survival. METHODS: We studied 1056 thoracoabdominal aortic aneurysm repairs from 2006 to 2016. Exclusion criteria were age less than 18 years (n = 9), prior splenectomy (n = 2), and intraoperative death (n = 3). This left 1042 thoracoabdominal aortic aneurysm repairs for analysis (median age, 65 years; interquartile range, 56-72), including 221 (21%) that were reoperations. Multivariable modeling identified predictors of operative mortality in the total cohort. Moreover, to adjust for baseline differences, propensity score matching was performed to examine the frequency of these outcomes in the total cohort (n = 132 pairs) and the early survivors (n = 110 pairs). Late survival was estimated by the Kaplan-Meier method, and risk of late mortality was assessed by Cox proportional hazards regression. RESULTS: Incidental splenectomy was performed in 135 patients (13%), 36% of whom underwent reoperation. Operative mortality rates of the incidental splenectomy and nonincidental splenectomy groups were 16% versus 8% in both the overall study (P = .005) and the propensity score-matched (P = .07) cohorts. In multivariable analysis, incidental splenectomy independently predicted operative mortality (odds ratio, 2.2; 95% confidence interval, 1.21-3.94; P = .008). For early survivors, incidental splenectomy did not increase the risk of late mortality. Survival estimates of matched early survivors did not differ between the incidental splenectomy and nonincidental splenectomy groups (P = .29). CONCLUSIONS: Incidental splenectomy during thoracoabdominal aortic aneurysm repair was associated with increased operative mortality but not reduced late survival. Splenic preservation is encouraged when feasible.

Neurologic complications after the frozen elephant trunk procedure: A meta-analysis of more than 3000 patients.

OBJECTIVE: The frozen elephant trunk technique's safety regarding spinal cord ischemia has been questioned. We used a meta-analysis to determine the rates of adverse neurologic events and mortality. METHODS: We searched PubMed/Medline, Embase, Scopus, and Cochrane databases (inception to April 2018) to identify studies of neurologic events after the frozen elephant trunk procedure. Separate meta-analyses were conducted with random-effects models to assess frozen elephant trunk associations with spinal cord ischemia, stroke, operative mortality, and all adverse events combined. Subgroup analyses compared outcomes in patients with acute versus nonacute type A dissection and aneurysm and with different extents of coverage. RESULTS: Thirty-five studies (total N = 3154) met inclusion criteria. The pooled rates of the outcomes of interest were 4.7% (95% confidence interval, 3.5-6.2) for spinal cord ischemia, 7.6% (95% confidence interval, 5.0-11.5) for stroke, and 8.8% (95% confidence interval, 7.0-10.9) for operative mortality. The spinal cord ischemia event rate was higher with stent length 15 cm or greater or coverage to T8 or beyond than with stent length of 10 cm (11.6% vs 2.5%, P < .001). Adverse event rates in patients with acute type A aortic dissection versus nonacute dissection or aneurysm were as follows: mortality 9.2% versus 7.6% (P = .46), stroke 9.3% versus 6.6% (P = .51), and overall adverse events 22.0% versus 16.5% (P = .41). CONCLUSIONS: As the frozen elephant trunk procedure becomes more popular, accurate data regarding outcomes are vital. We associated the frozen elephant trunk technique with (nonsignificantly) more adverse events overall in acute type A dissection cases. Stent length of 10 cm was associated with significantly less risk of spinal cord ischemia. Using a stent 15 cm or greater or coverage extending to T8 or farther should be avoided.

Acute type I aortic dissection with or without antegrade stent delivery: Mid-term outcomes.

OBJECTIVE: We determined the effect of antegrade stent delivery in the descending thoracic aorta on short- and mid-term clinical and imaging outcomes for patients who underwent repair of acute DeBakey type I aortic dissection. METHODS: Outcomes were evaluated for 178 patients who underwent acute type I aortic dissection between 2005 and 2016 (standard repair, n = 115 [64.6%]; antegrade stent delivery, n = 63 [35.4%]). Propensity score match and multivariable analyses were performed to assess outcomes. RESULTS: The stent and standard repair groups had similar rates of operative mortality (30-day or in-hospital) (12.7% vs 17.4%, P = .41), persistent stroke (6.3% vs 5.3%, P = .75), and persistent paraplegia/paraparesis (1.6% vs 0.9%, P = 1.0). Propensity score match analysis indicated that the operative mortality rate was higher in the standard repair group (P = .059), which the multivariable analysis confirmed. The persistent stroke rate was nonsignificantly higher in the stent group (P = .66). Persistent paraplegia/paraparesis rates were similar in both groups (P = 1.0), and the overall rates of spinal cord ischemia were nonsignificantly higher in the stent group (P = .18). During follow-up (mean duration, 4.6 ± 3.6 y), computed tomography showed that stented patients more often had remodeling of the descending thoracic aorta (P = .0002) and somewhat more often had remodeling of the thoracoabdominal aorta (P = .13). Stented patients also had fewer subsequent procedures (P = .25). The 3- and 5-year survivals were 73.3% ± 6.9% and 49.9% ± 7.6% in the matched stented group and 66.3% ± 9.4% and 41.6% ± 7.7% in the matched standard group, respectively (P = .015 for overall survival). CONCLUSIONS: In the short term, antegrade stent delivery was associated with less operative mortality. In the mid-term, promising remodeling of the false lumen was seen in stented patients, as were (nonsignificantly) lower rates of subsequent procedures in the thoracoabdominal aorta. Mid-term survival was also greater in the stented patients.

Tracheostomy After Thoracoabdominal Aortic Aneurysm Repair: Risk Factors and Outcomes.

BACKGROUND: Respiratory failure, the most frequent complication after thoracoabdominal aortic aneurysm (TAAA) repair, necessitates tracheostomy in severe cases. We examined risk factors for and outcomes of tracheostomy after TAAA repair. METHODS: We reviewed the records of 1267 consecutive patients who underwent TAAA repair. Patients with a preexisting tracheostomy were excluded. Extensive repairs (Crawford extent I or II) were performed in 716 patients (56.6%). Stepwise logistic regression analysis was used to identify risk factors for postrepair tracheostomy. RESULTS: Tracheostomy was necessary in 140 patients (11.1%). Operative mortality was significantly higher in patients with tracheostomy (27.9%) than in those without (5.8%; p < 0.001). As expected, tracheostomy patients had longer intensive care unit stays (24 vs 4 days, p < 0.001) and hospital stays (57 vs 10 days, p < 0.001) than nontracheostomy patients. Patients with tracheostomy were frequently transferred for additional long-term acute care or hospitalization (107, 76.4%), and many died after transfer (24/107, 22.4%). Kaplan-Meier curves showed markedly poorer late survival in patients with tracheostomy than in those without (47.9% ± 4.3% vs 87.3% ± 1.0% at 1 year; 27.8% ± 4.8% vs 68.6% ± 1.6% at 5 years). Independent predictors of post-TAAA repair tracheostomy included acute aortic dissection, chronic renal insufficiency, underweight body mass index, hypertension, history of stroke, extent II repair, diabetes, age at least 70 years, and greater platelet transfusion volume. CONCLUSIONS: Patients who undergo tracheostomy after TAAA repair have a high risk of early and late mortality as well as prolonged hospitalization. Strategies for improving survival outcomes in tracheostomy patients warrant investigation.

In the endovascular era, is elective open aortic arch surgery in elderly patients still justified?

OBJECTIVE: The data supporting performing elective aortic arch surgery in patients aged 75 years or older are equivocal. We evaluated short- and long-term outcomes after elective arch surgery in patients aged ≥75 years to determine whether complex arch operations are justified in such patients. METHODS: Over a 10-year period, 805 patients aged 50 to 89 years underwent elective proximal or total arch surgery. Composite adverse outcome was defined as operative mortality, persistent (ie, present at discharge) neurologic event, or persistent hemodialysis. Multivariable logistic regression was performed in the entire group. RESULTS: Multivariable analysis showed that age at admission independently predicted composite adverse outcome, operative mortality, and prolonged (>48 hours) ventilator support (P < .0001 for all), but not stroke. The same results were shown in a subgroup analysis in which older age (80-89 years) was associated with composite adverse outcome, operative mortality, and prolonged ventilator support. In a Cox proportional hazards regression model adjusted for antegrade cerebral perfusion time and prior history of renal disease, patients aged 50 to 74 years had significantly better overall survival than patients aged ≥75 years (P < .001). CONCLUSIONS: As endovascular technology evolves, having benchmark data from likely endovascular-therapy candidates is critical. This study, among the few to focus on elective aortic arch surgery in elderly patients, suggests that surgical intervention carries risk and that novel endovascular therapies are needed.

Incidence and Treatment Outcomes of Vocal Fold Movement Impairment After Total Arch Replacement.

BACKGROUND: Vocal fold movement impairment (VFMI) secondary to neuronal injury is a known risk after aortic surgery. Total arch replacement is technically challenging, and the incidence of vocal fold movement impairment secondary to neuronal injury after this surgery is unknown. This study examined the incidence of VFMI after total arch replacement and medialization treatment outcomes. STUDY DESIGN: Retrospective cohort study. METHODS: All patients who underwent total arch replacement at a tertiary care center over 11 years (2006-2017) were identified through an institutional database. End points included evidence of VFMI on flexible laryngoscopy, time to diagnosis, time to treatment, need for reintubation, and intensive care unit (ICU) and hospital length of stay. RESULTS: Of the 358 patients who underwent total arch replacement, 63 (20%) were diagnosed with VFMI during their initial inpatient stay. Fifty patients (79%) VFMIs were left-sided, nine (14%) were right-sided, and four (6%) were bilateral. Thirty-nine patients (62%) underwent inpatient vocal fold medialization: 28 (72%) by injection laryngoplasty and 11 (28%) by type 1 thyroplasty. Those with unilateral VFMI had longer ICU (8.9 days) and hospital (19.4 days) than those with no VFMI (5.7 and 16.1 days). Among patients with unilateral VFMI, those who underwent inpatient vocal fold medialization trended toward shorter ICU (6.2 vs. 14.4 days, P = .03) and hospital stays (20.1 vs. 23.3 days, P = .4) than patients who did not have a medialization procedure. CONCLUSION: The overall incidence of VFMI after total arch replacement in our series was 20%. Both the right and left vocal folds are potentially at risk from a total arch replacement; consequently, the distribution of injury in our cohort was more heterogeneous than in other series. LEVEL OF EVIDENCE: 3 Laryngoscope, 129:699-703, 2019.

Early-Stage Acute Kidney Injury Adversely Affects Thoracoabdominal Aortic Aneurysm Repair Outcomes.

BACKGROUND: Acute kidney injury (AKI) necessitating renal replacement therapy adversely affects outcomes after thoracoabdominal aortic aneurysm (TAAA) repair. The effects of earlier stages of AKI are less known. We hypothesized that earlier stages of AKI would reduce early survival after TAAA repair. METHODS: We analyzed prospectively collected data from 1,056 consecutive TAAA repairs from our institution (2006 to 2016). We excluded patients less than 18 years of age, those with preexisting renal disease, and three patients who died intraoperatively, resulting in 873 patients. The Kidney Disease Improving Global Outcomes criteria grouped patients into three AKI stages; stage 3 necessitated initiation of renal replacement therapy. Multivariable modeling identified operative mortality predictors. Kaplan-Meier analysis assessed 1-year survival. RESULTS: Of 873 patients, 642 (73.5%) had no AKI and 231 (26.5%) had postoperative AKI (mild/stage 1, n = 92 [10.5%]; moderate/stage 2, n = 44 [5%]; severe/stage 3, n = 95 [10.9%]). Operative death occurred in 65 patients (7.4%): 14 (2.2%) with no AKI, 5 (5.4%) with mild AKI (p = 0.07 versus no AKI), 8 (18.2%) with moderate AKI (p = 0.02 versus mild), and 38 (40%) with severe AKI (p = 0.01 versus moderate). In multivariable analysis, moderate AKI independently predicted death (relative risk ratio: 9.4, 95% confidence interval: 3.4 to 25.9). Kaplan-Meier 1-year survival was 91.1% ± 1.2% for no AKI, 84.6% ± 3.9% for mild AKI (p = 0.07 versus no AKI), 67.4% ± 7.6% for moderate AKI (p = 0.01 versus mild), and 46.6% ± 5.3% for severe AKI (p = 0.02 versus moderate; p < 0.0001 across all groups). CONCLUSIONS: Moderate/stage 2 AKI reduced early survival after TAAA repair. Prevention, earlier detection, and optimal medical management of AKI may improve survival.

The impact of preoperative chronic kidney disease on outcomes after Crawford extent II thoracoabdominal aortic aneurysm repairs.

OBJECTIVE: To determine whether preoperative chronic kidney disease (CKD) is predictive of poor outcomes in patients who undergo Crawford extent II thoracoabdominal aortic aneurysm (TAAA) repair. METHODS: Data were collected from patients with CKD (defined as a preoperative estimated glomerular filtration rate <60 mL/min/1.73 m2; n = 399) and without CKD (n = 604) who underwent extent II TAAA repair during 1991 to 2016. We used univariate, multivariable, and propensity score matching analyses to compare outcomes between these 2 groups. RESULTS: Compared with patients without CKD, patients who presented with CKD were older and had greater rates of comorbidities, including coronary artery disease, cerebrovascular disease, and peripheral vascular disease. Patients with CKD had higher rates of operative mortality and adverse events. After propensity analysis, patients with CKD had greater rates of adverse event and renal failure necessitating dialysis, but had comparable rates of operative death to patients without CKD. Multivariable modeling indicated that CKD independently predicted adverse event (relative risk ratio [RRR] = 1.61; P = .01) and renal failure (RRR = 1.86; P = .02) after repair. After adjustment for median age, patients with CKD had substantially worse mid-term survival than those without (23.9 ± 2.4% vs 48.5 ± 2.5% at 10 years; P < .001). CONCLUSIONS: In patients who present with CKD, extent II open TAAA repair carries considerable risks of operative death and adverse events. Further investigation is needed to improve renal protection during such repair.

Chronic Type I and Type III aortic dissections: a propensity analysis of outcomes after open distal repair.

OBJECTIVES: To compare short-term outcomes, long-term survival and reinterventions in patients requiring surgery after chronic Type I and chronic primary Type III aortic dissections. METHODS: Over an 11-year period, 466 patients underwent thoraco-abdominal aortic aneurysm repair for chronic Type III (n = 239) and Type I (n = 227) aortic dissections. Short-term outcomes and reinterventions were evaluated by multivariable regression analysis for the entire group; propensity matching produced 169 pairs. RESULTS: Mortality was 6% (n = 28) in the overall cohort and 6.2% (n = 14) and 5.9% (n = 14) in those with chronic Type I and Type III aortic dissections, respectively. Overall stroke and persistent spinal cord deficit rates were 4.0% and 2.6%, respectively, in the Type I group and 1.3% and 3.8% in the Type III group. In the propensity-matched patients, analysis showed no neurological differences between the 2 groups, but respiratory failure was significantly more frequent in the chronic Type I group (30.2% vs 15.4%; P = 0.001). Multivariable analysis identified chronic Type I dissection as an independent risk factor for postoperative pulmonary complications (odds ratio 1.612; 95% confidence interval 1.060-2.452; P = 0.026) and an association between chronic Type I dissection and stroke (odds ratio 4.013; 95% confidence interval 1.026-15.698; P = 0.046). Six-year survival was 74.4% ± 4.1% and 74.4% ± 4.6% in the chronic Type I and Type III groups, respectively (P = 0.87). CONCLUSIONS: Short- and long-term mortality and reintervention rates were comparable after open repair for chronic Type I and primary chronic Type III aortic dissections. Respiratory failure was more frequent in the chronic Type I aortic dissection group.

Postoperative Chylothorax After Thoracoabdominal Aortic Aneurysm Repair.

Chylothorax is a potentially deadly complication that can occur after thoracoabdominal aortic aneurysm (TAAA) repair. We describe our contemporary experience (2005-2014) with this complication, our efforts to identify perioperative variables associated with it, and our attempts to assess treatment outcomes. We reviewed the records of 1092 consecutive patients who underwent TAAA repair between 2005 and 2014. Standard bivariate analysis was used to test for between-group differences. Eleven patients (0.9%) developed postoperative chylothorax. Nonoperative management was used in 8 of these patients (73%); 1 patient died after a lengthy hospital stay (297 days). The other 3 patients required thoracotomy with direct ligation; 1 of these patients required a second operation. Patients who developed chylothorax appeared to be similar to other patients in age, sex, extent of aneurysm, and metabolic or cardiovascular comorbidities. Patients who developed postoperative chylothorax were more likely to require drainage of a pleural effusion (P = 0.005), tracheostomy (P = 0.02), and longer stays in the intensive care unit (median, 6 [2-24] days, P < 0.001) and the hospital (median, 35 [24-88] days, P = 0.001), and these patients were more likely to develop a graft infection (n = 2, P < 0.001). The extent of TAAA repair (Crawford I-IV), reoperation, and clamping proximal to the left subclavian artery were not significantly associated with postoperative chylothorax. Chylothorax after TAAA repair can often be managed nonoperatively. Development of postoperative chylothorax may lead to significant morbidity, longer hospitalization, and increased likelihood of graft infection.

In elective arch surgery with circulatory arrest, does the arterial cannulation site really matter? A propensity score analysis of right axillary and innominate artery cannulation.

OBJECTIVE: The preferred arterial cannulation site for elective proximal aortic procedures requiring circulatory arrest varies, and different sites have been tried. We evaluated the relationships between arterial cannulation site and adverse outcomes, including stroke, in patients undergoing elective aortic arch surgery. METHODS: We reviewed the records of 938 patients who underwent elective hemiarch or total arch surgery with circulatory arrest between 2006 and 2016. Five cannulation sites were used: the right axillary (n = 515; 54.9%), innominate (n = 376; 40.1%), and right common carotid arteries (n = 15; 1.6%), each with a side graft; the ascending aorta (n = 19; 2.0%); and the femoral artery (n = 13; 1.4%). Multivariable logistic regression analysis was used to model the effects of cannulation site on adverse outcomes for the entire cohort and for a subcohort of 891 patients who underwent innominate or axillary artery cannulation. Propensity-matching yielded 564 patients (282 pairs) from the right axillary and innominate artery groups. RESULTS: For the entire cohort, mortality, stroke, and composite adverse outcome (operative death or persistent stroke or renal failure at hospital discharge) rates were 7.0%, 4.1%, and 9.8%. In the multivariable analysis of the axillary/innominate subcohort, cannulation site did not independently predict operative mortality, persistent stroke, or composite adverse event. These results were confirmed with the propensity-matched analysis, where both axillary and innominate artery cannulation provided equivalent composite adverse event rates, operative death rates, and overall stroke rates. CONCLUSIONS: During elective arch surgery, right axillary artery cannulation and innominate artery cannulation (both via a side graft) produce excellent results and can be used interchangeably.

Unplanned Readmissions After Open Thoracoabdominal Aortic Aneurysm Repair.

BACKGROUND: Although reducing the incidence of unplanned readmission after thoracoabdominal aortic aneurysm (TAAA) repair represents an important opportunity to improve outcomes, predictors of readmissions are not known. We sought to characterize and identify factors associated with unplanned readmission after discharge in survivors of open TAAA repair. METHODS: Through prospective phone contact and retrospective record review, we determined the frequency and characteristics of unplanned readmissions within 30 days of discharge in 363 patients who were discharged after open TAAA repair. We used univariate and multivariable analyses to identify factors associated with readmission. RESULTS: There were 44 unplanned readmissions in 40 patients (11%). After readmission, 11 patients underwent operations, and 17 underwent nonsurgical procedures, the most common of which was thoracentesis (n = 9). Readmitted patients tended to have lower preoperative estimated glomerular filtration rates (p = 0.045), higher frequencies of preoperative sleep apnea (p = 0.009) and postoperative pulmonary (p = 0.04) and infection (p = 0.02) complications, and longer hospital stays (p = 0.01) than patients without readmissions. Patient age, urgency of operation, and extent of TAAA repair were similar in patients with and without readmissions. Multivariable analysis identified sleep apnea (relative risk ratio [RRR] 3.21, 95% confidence interval [CI]: 1.51 to 6.82, p = 0.002), postoperative infection (RRR 4.34, 95% CI: 1.32 to 14.25, p = 0.02), renal failure necessitating dialysis (RRR 3.14, 95% CI: 1.04 to 9.46, p = 0.04), and visceral artery stenting (RRR 2.43, 95% CI: 1.09 to 5.44, p = 0.03) as significant predictors of readmission. CONCLUSIONS: Patients with renal dysfunction, sleep apnea, or postoperative infection were particularly likely to be readmitted; optimizing the management of these factors may reduce early readmission after TAAA repair.

Early versus late inpatient awake transcervical injection laryngoplasty after thoracic aortic repair.

BACKGROUND: Vocal fold movement impairment (VFMI), a known complication following thoracic aortic surgery, has been associated with poorer surgical outcomes, including higher pulmonary complications and longer length of stay (LOS). Awake transcervical injection laryngoplasty in the inpatient setting serves to augment pulmonary toilet function for affected patients. This study investigates clinical outcomes of patients who underwent early versus late injection laryngoplasty following aortic surgery. METHODS: A 5-year review (2011-2016) of 35 patients who underwent inpatient injection laryngoplasty for VFMI following aortic repair was conducted. Data included demographics, surgery parameters, laryngologic evaluation, pulmonary complications, LOS, and mortality. Early injection was defined as within 5 days from extubation. Statistical analyses were performed using SPSS, version 17.0 (IBM Corp., Armonk, NY). RESULTS: All 35 patients (mean age 56, 77% male) underwent left vocal-fold injection laryngoplasty without complication. There were 15 (43%) early procedures and 20 (57%) late procedures. Mean LOS for early and late injection groups (13 vs. 20 days, respectively) significantly differed (P = 0.04, 95% confidence interval (CI) 0.3 to 14.4). Early laryngoplasty cohorts experienced less pulmonary complications (20%) than those who had late medialization (50%), but this did not reach significance (P = 0.06, 95% CI -0.3 to 8.1). The most common complication type in both groups was therapeutic bronchoscopy. CONCLUSION: Early awake injection laryngoplasty within 5 days from extubation is feasible and may improve clinical outcomes for patients with VFMI following aortic repair. Further prospective studies on this approach for VFMI after thoracic surgery are needed. LEVEL OF EVIDENCE: 4. Laryngoscope, 128:144-147, 2018.

Reoperative surgery on the thoracoabdominal aorta.

OBJECTIVE: Since the advent of endovascular repair for aortic aneurysms, many centers have justified the use of endovascular approaches in patients with previous open distal aortic repair by deeming these patients "high risk" because of their previous operation. We sought to determine whether patients who undergo reoperative repair for thoracoabdominal aortic aneurysm (TAAA) have worse outcomes than patients who undergo non-reoperative repair. METHODS: We reviewed our data on 3379 TAAA repairs from 1986 to 2016. We compared patients' preoperative characteristics, surgical variables, and outcomes among reoperative (n = 726) and non-reoperative (n = 2653) cases. Furthermore, we examined reoperative indications to identify repairs performed because of repair failure (n = 93) and reoperations performed as an adjacent extension of repair (n = 633). A multivariable analysis was conducted to identify predictors of adverse events by using relevant preoperative and intraoperative factors. RESULTS: The operative mortality rate did not significantly differ between groups (8.1% for reoperative vs 7.3% for non-reoperative; P = .5); in addition, reoperative repair was not associated with an increased risk of adverse event. However, Kaplan-Meier survival analysis showed that over the first 10 years, the reoperative groups fared significantly worse than the non-reoperative group (P < .001) (survival estimates at 10 years: 23.9% ± 4.9% for patients with repair failure, 28.4% ± 2.0% for those with extension of repair, and 40.1% ± 1.1% for non-reoperative repairs). CONCLUSIONS: We were unable to detect noteworthy differences in early outcomes between reoperative and non-reoperative TAAA repair. However, mid-term results indicate worse survival for patients who undergo reoperative surgery.

Open descending thoracic or thoracoabdominal aortic approaches for complications of endovascular aortic procedures: 19-year experience.

OBJECTIVES: Endovascular aortic repair is increasingly being used to treat aneurysms, dissections, and traumatic injuries, despite its unknown long-term durability. We describe our 19-year experience with open descending thoracic and thoracoabdominal aortic repair after endovascular aortic repair. METHODS: Between 1996 and 2015, 67 patients were treated with open distal arch, descending thoracic, or thoracoabdominal aortic repair, or extra-anatomic bypass repair with aortic extirpation for complications after endovascular repair of the thoracic (n = 45, 67%) or abdominal (n = 22, 33%) aorta. The median interval between procedures was 18.0 months (interquartile range, 3.9-44.9). Indications for open repair included expanding aneurysm (n = 56), infection (n = 11), fistula (n = 8), aneurysm rupture (n = 5), pseudoaneurysm (n = 2), and restenosis (n = 1). Open repair involved partial (n = 9, 13%) or complete (n = 56, 84%) device removal or device salvage (n = 2, 3%) through a thoracoabdominal (n = 58, 87%) or thoracotomy (n = 9, 13%) incision. Eight patients (12%) underwent emergency procedures. RESULTS: There were 3 early (operative) deaths (2 with preoperative device infection) and 19 late deaths during a median follow-up of 35.8 months (interquartile range, 16.8-52.8 months). Overall 1- and 5-year survivals were 85% ± 4% and 60% ± 8%, respectively. Four patients had open repair failures necessitating reoperation; 2 patients had preoperative infection, and both died (1 early and 1 late). CONCLUSIONS: Open repair for complications after endovascular procedures is not uncommon. Experienced centers can yield acceptable outcomes, especially in patients without infection. Close surveillance is mandatory after endovascular aortic repair.