Objectives Although frailty has recently been examined in various populations as a predictor of morbidity and mortality, its effect on thoracic aortic surgery outcomes has not been studied. The objective of the present study was to evaluate the role of frailty in predicting postoperative morbidity and mortality in patients undergoing proximal aortic replacement surgery. Methods A retrospective analysis of a prospectively maintained database was performed for all patients undergoing elective and nonelective proximal aortic operations (root, ascending aorta, and/or arch) at a single-referral institution from June 2005 to December 2012. A total of 581 patients underwent proximal aortic surgery, of whom 574 (98.8%) were included in the present analysis; 7 were excluded because of incomplete data. Frailty was evaluated using an index consisting of age > 70 years, body mass index < 18.5 kg/m2, anemia, history of stroke, hypoalbuminemia, and total psoas volume in the bottom quartile of the population. One point was given for each criterion met to determine a frailty score of 0 to 6. Frailty was defined as a score of ≥ 2. Risk models for length of stay > 14 days, discharge to other than home, 30-day composite major morbidity, 30-day composite major morbidity/mortality, and 30-day and 1-year mortality were calculated using multivariate regression modeling. Results Of the 574 patients, 148 (25.7%) were defined as frail (frailty score ≥ 2). The unadjusted 30-day/in-hospital and long-term outcomes were significantly worse for the frail versus nonfrail patients in all but 1 of the outcomes analyzed; no difference was found in the 30-day readmission rates between the 2 groups. In the multivariate model, a frailty score of ≥ 2 was associated with discharge to other than home and 30-day and 1-year mortality. Conclusions Frailty, as defined using a 6-component frailty index, can serve as an independent predictor of discharge disposition and early and late mortality risk in patients undergoing proximal aortic surgery. These frailty markers, all of which are easily assessed preoperatively, could provide valuable information for patient counseling and risk stratification before proximal aortic replacement.

Malposition, embolization, fracture, and migration of endovascular devices are unfortunate consequences of endovascular intervention and will be encountered at some point by nearly every practitioner. The existing literature on foreign body retrieval consists of large single-institution series and case reports. We provide an overview of this recent literature, clarifying what devices are being lost, what symptoms occur as a result, and how retrieval is being performed. We have identified all case series and case reports since the year 2000, summarized the results, and made some general observations and recommendations that may be useful to the practitioner faced with the prospect of retrieving a fractured medical device, malpositioned coil, or migrated inferior vena cava filter.

Background Extracorporeal membrane oxygenation (ECMO) is being increasingly used as a bridge to lung transplantation. Small, single-institution series have described increased success using ECMO in spontaneously breathing patients compared with patients on ECMO with mechanical ventilation, but this strategy has not been evaluated on a large scale. Methods Using the United Network for Organ Sharing database, all adult patients undergoing isolated lung transplantation from May 2005 through September 2013 were identified. Patients were categorized by their type of pretransplant support: no support, ECMO only, invasive mechanical ventilation (iMV) only, and ECMO + iMV. Kaplan-Meier survival analysis with log-rank testing was performed to compare survival based on type of preoperative support. A Cox regression model was used to determine whether type of preoperative support was independently associated with survival, using previously established predictors of survival as covariates. Results Approximately 12,403 primary adult pulmonary transplantations were included in this analysis. Sixty-five patients (0.52%) were on ECMO only, 612 (4.93%) required only iMV, 119 (0.96%) were on ECMO + iMV, and the remaining 11,607 (94.6%) required no invasive support before transplantation. One-year survival was decreased in all patients requiring support, regardless of type. However, mid-term survival was similar between patients on ECMO alone and those not on support but significantly worse with patients requiring iMV only or ECMO + iMV. In multivariable analysis, ECMO + iMV and iMV alone were independently associated with decreased survival compared with nonsupport patients, whereas ECMO alone was not significant. Conclusions In patients with worsening pulmonary disease awaiting lung transplantation, those supported via ECMO with spontaneous breathing demonstrated improved survival compared with other bridging strategies.

Native aortic valve regurgitation developed and/or progressed after left ventricular assist device implantation, with this effect being more pronounced in continuous flow left ventricular assist device recipients. However, the preoperative aortic valve regurgitation grade failed to correlate with the development of substantial aortic valve regurgitation after left ventricular assist device implantation. After left ventricular assist device implantation, aortic valve regurgitation had a small, but discernible, clinical effect, with some patients developing severe aortic valve regurgitation and requiring aortic valve procedures. These data have implications for the long-term management of left ventricular assist device recipients, in particular as the durability of implantable continuous flow left ventricular assist device therapy improves.

Objective The choice of cerebral perfusion strategy for aortic arch surgery has been debated, and the superiority of antegrade (ACP) or retrograde (RCP) cerebral perfusion has not been shown. We examined the early and late outcomes for ACP versus RCP in proximal (hemi-) arch replacement using deep hypothermic circulatory arrest (DHCA). Methods A retrospective analysis of a prospectively maintained database was performed for all patients undergoing elective and nonelective hemiarch replacement at a single referral institution from June 2005 to February 2013. Total arch cases were excluded to limit the analysis to shorter DHCA times and a more uniform patient population for whom clinical equipoise regarding ACP versus RCP exists. A total of 440 procedures were identified, with 360 (82%) using ACP and 80 (18%) using RCP. The endpoints included 30-day/in-hospital and late outcomes. A propensity score with 1:1 matching of 40 pre- and intraoperative variables was used to adjust for differences between the 2 groups. Results All 80 RCP patients were propensity matched to a cohort of 80 similar ACP patients. The pre- and intra-operative characteristics were not significantly different between the 2 groups after matching. No differences were found in 30-day/in-hospital mortality or morbidity outcomes. The only significant difference between the 2 groups was a shorter mean operative time in the RCP cohort (P = .01). No significant differences were noted in late survival (P = .90). Conclusions In proximal arch operations using DHCA, equivalent early and late outcomes can be achieved with RCP and ACP, although the mean operative time is significantly less with RCP, likely owing to avoidance of axillary cannulation. Questions remain regarding comparative outcomes with straight DHCA and lesser degrees of hypothermia.