Joyce, David L.

Introduction: We compared outcome of repeat CABG (RCABG) and PCI in patients with previous CABG. Hypothesis: RCABG may improve survival versus PCI. Methods: From 1993 through 2009, 3,058 patients underwent revascularization after previous CABG. RCABG patients (n=530) were grouped by the use of arterial grafts (n=396), or saphenous vein grafts (SVG; n=134). At previous CABG, 37% of RCABG patients received IMA graft. PCI patients (n=2,528) were grouped into balloon angioplasty (BA; n=491), bare metal stent (BMS; n=1,267), and drug-eluting stent (DES; n=769). PCI patients were further stratified by the treated target into native coronary artery (n=1,637), bypass grafts (n=674), or both (n=216). Propensity score analysis matched 287 patients. Results: RCABG had increased 30-day mortality compared with PCI (HR=3.92; 95% CI, 2.25-6.84; P <.001) in multivariable analysis, while in propensity matched analysis it did not ( P =.09). Overall survival, however, was greater for RCABG compared with PCI in multivariable analysis (HR=0.70; 95% CI, 0.59-0.83; P <.001), as well as in propensity matched analysis (10-year survival rates were 52% vs 37%, respectively [ P =.005]). Compared with RCABG, PCI involving bypass grafts had increased mortality (HR, 1.39; 95% CI, 1.09-1.78; P =.008), whereas PCI targeting only the native coronaries did not (HR, 1.12; 95% CI, 0.85-1.47; P =.43); BMS patients had increased mortality (HR, 1.34; 95% CI, 1.06-1.68; P =.013), but DES did not (HR, 1.28; 95% CI, 0.77-2.12; P =.35). At RCABG, 318 patients (60%) received an IMA graft and late survival was similar to those receiving SVG; The use of IMA in previous CABG was associated with improved survival (HR=0.82; 95% CI, 0.72-0.92; P <.001]. Conclusions: Long-term survival after previous CABG is improved with RCABG compared to PCI, especially when PCI involves bypass grafts. There is no difference in long-term survival between RCABG using arterial grafts vs SVG only if IMA is used in either previous or RCABG.

Introduction: Frailty has recently been shown to be associated with increased risk of death after continuous-flow left ventricular assist device (CF-LVAD) implantation as destination therapy (DT), but no data exists specifically analyzing its association with specific early and late adverse events. Methods: Between February 2007 and May 2014, 251 patients (205 male) underwent primary implantation of CF-LVAD at our institution; 157 (63%) were implanted as DT and preoperative frailty assessment was available in 96 pt. Frailty was identified in 58/96 (60%) patients according to the deficit index (31 impairments, disabilities and comorbidities) and defined as > 25% deficits. Age (67 vs. 64 years, p=0.12), male sex (79% vs. 87%, 0.63), chronic kidney disease (64% vs. 63%, p=0.74), creatinine (1.7 vs. 1.5, p=0.1), hemoglobin (11.6 vs. 11.5, p=0.98), and albumin levels (3.7 vs. 3.8, p=0.53) were similar between frail and non-frail pt, while diabetes (59% vs. 13%, p<0.001), ischemic etiology (64% vs. 45%, p=0.04), and preop balloon pump (50% vs. 26%, p=0.01) were significantly different. Results: All patients underwent CF-LVAD implantation and cardiopulmonary bypass time was similar between groups (102 vs. 111 min., p=0.3). There were no significant differences in early adverse events between frail and non-frail pt, including inotrope duration (120 vs. 96 hr., p=0.11), right ventricular failure (19% vs. 13%, p=0.44), respiratory failure (26% vs. 18%, p=0.34), acute renal failure (16% vs. 13%, p=0.69), and hepatic dysfunction (16% vs. 13%, p=0.69). Median length of stay (20 vs. 16 days, p=0.25) and early mortality (12% vs. 8%, p=0.48) were also similar. During a median follow up of 2 years with 224 patient/years of support, any GI bleed (43% vs. 42%, p=0.79), any thromboembolic event (34% vs. 39%, p=0.74), and driveline infection (10% vs. 26%, p=0.052). Conclusions: In this select cohort of patients, frailty did not portend higher risk of adverse events after CF-LVAD implantation for DT despite prior observation that late survival is reduced in these patients. These findings are extremely important in patient selection and counseling of patients prior to LVAD implantation.

Background: Post–cardiac surgery bleeding can have devastating consequences, and it is unknown if warfarin-induced rapid international normalized ratio (INR) rise during the immediate postoperative period increases bleed risk. Objective: To determine the impact of warfarin-induced rapid-rise INR on post–cardiac surgery bleeding. Methods: This was a single-center, retrospective chart review of post–cardiac surgery patients initiated on warfarin at Mayo Clinic Hospital, Rochester. Patients were grouped based on occurrence or absence of rapid-rise INR (increase ≥1.0 within 24 hours). The primary outcome compared bleed events between groups. Secondary outcomes assessed hospital length of stay (LOS) and identified risk factors associated with bleed events and rapid rise in INR. Results: During the study period, 2342 patients were included, and 56 bleed events were evaluated. Bleed events were similar between rapid-rise (n = 752) and non–rapid-rise (n = 1590) groups in both univariate (hazard ratio [HR] = 1.22; P = 0.594) and multivariable models (HR = 1.24; P = 0.561). Those with rapid-rise INR had longer LOS after warfarin administration (discharge HR = 0.84; P = 0.0002). The most common warfarin dose immediately prior to rapid rise was 5 mg. Risk factors for rapid-rise INR were low body mass index, female gender, and cross-clamp time. Conclusion and Relevance: This represents the first report to assess warfarin-related rapid-rise INR in post–cardiac surgery patients and found correlation to hospital LOS but not bleed events. Conservative warfarin dosing may be warranted until further research can be conducted.

Background: The FOXO3a (forkhead box O3a)-BNIP3 (B-cell lymphoma 2/adenovirus E1B 19kDa interacting protein 3) pathway modulates mitochondrial dynamics and function and contributes to myocardial remodeling in rodent models of heart failure. We sought to investigate the expression of this pathway along with the expression of mitochondrial biogenesis (PGC-1α [peroxisome proliferator-activated receptor-γ coactivator-1α]), dynamics (DRP-1 [dynamin-related protein 1], OPA-1 [optic atrophy 1], and MFN 2 [mitofusin 2]), and oxidative phosphorylation (citrate synthase and electron transport chain complexes) markers and COX IV (cytochrome C oxidase) activity in myocardium from patients with valvular or ischemic heart disease and heart failure with preserved ejection fraction (HFpEF) or heart failure with reduced ejection fraction (HFrEF). Methods and Results: Subepicardial left ventricular biopsies (10×1×1 mm 3 ) were obtained at aortic valve replacement (HFpEF AVR , n=5; and HFrEF AVR , n=4), coronary artery bypass grafting (HFpEF CABG , n=5; and HFrEF CABG , n=5), or left ventricular assist device implantation (HFrEF LVAD , n=4). Subepicardial biopsies from patients with normal left ventricular function (n=2) and from donor hearts (n=3) served as controls (normal). Relative to normal, mitochondrial fragmentation and cristae destruction were evident, and mitochondrial area was decreased in HFpEF; 1.00±0.09 versus 0.71±0.08; P =0.016. These mitochondrial morphological changes were more pronounced in HFrEF (0.54±0.06); P =0.002 HFpEF versus HFrEF. BNIP3 (monomer+dimer) expression was increased in HFpEF (3.99±2.44) and in HFrEF (5.19±1.70) relative to normal; P =0.004 and P <0.001, respectively. However, BNIP3 monomer was increased in HFrEF (4.32±1.43) compared with normal (0.99±0.06) and HFpEF (1.97±0.90); P =0.001 and 0.004, respectively. The HFrEF group uniquely showed increase in DRP-1 expression (1.94±0.38) and decreases in PGC-1α expression (0.61±0.07) and COX IV activity (0.70±0.10) relative to normal; P =0.013, P <0.001, and P <0.001, respectively, with no significant change in electron transport chain complexes expression. Conclusions: These findings in human myocardium confirm studies in rodents where contractile dysfunction is associated with activation of the FOXO3a-BNIP3 pathway and altered mitochondrial dynamics, biogenesis, and function.

Abstract Aim The lungs—and particularly the alveolar-capillary membrane—may be sensitive to continuous flow (CF) and pulmonary pressure alterations in heart failure (HF). We aimed to investigate long-term effects of CF pumps on respiratory function. Methods and results We conducted a retrospective study of patients with end-stage HF at our institution. We analysed pulmonary function tests [e.g. forced vital capacity (FVC) and forced expiratory volume in 1 s (FEV1)] and diffusing capacity of the lung for carbon monoxide (DLCO) from before and after left ventricular assist device (LVAD) implantation and compared them with invasive haemodynamic studies. Of the 274 patients screened, final study analysis involved 44 patients with end-stage HF who had CF LVAD implantation between 1 February 2007 and 31 December 2015 at our institution. These patients [mean (standard deviation, SD) age, 50 (9) years; male sex, n = 33, 75%] received either the HeartMate II (Thoratec Corp.) pump (77%) or the HeartWare (HeartWare International Inc.) pump. The mean (SD) left ventricular ejection fraction was 21% (13%). At a median of 237 days post-LVAD implantation, we observed significant DLCO decrease (−23%) since pre-implantation (P &lt; 0.001). ΔDLCO had an inverse relationship with changes in pulmonary capillary wedge pressure (PCWP) and right atrial pressure (RAP) from pre-LVAD to post-LVAD implantation: ΔDLCO to ΔPCWP (r = 0.50, P &lt; 0.01) and ΔDLCO to ΔRAP (r = 0.39, P &lt; 0.05). We observed other reductions in FEV1, FVC, and FEV1/FVC between pre-LVAD and post-LVAD implantation. In mean (SD) values, FEV1 changed from 2.3 (0.7) to 2.1 (0.7) (P = 0.005); FVC decreased from 3.2 (0.8) to 2.9 (0.9) (P = 0.01); and FEV1/FVC went from 0.72 (0.1) to 0.72 (0.1) (P = 0.50). Landmark survival analysis revealed that ΔDLCO from 6 months after LVAD implantation was predictive of death for HF patients [hazard ratio (95% confidence interval), 0.60 (0.28–0.98); P = 0.03]. Conclusions Pulmonary function did not improve after LVAD implantation. The degree of DLCO deterioration is related to haemodynamic status post-LVAD implantation. The ΔDLCO within 6 months post-operative was associated with survival.

AbstractIntroductionTo evaluate the outcomes of video‐assisted thoracoscopic surgery (VATS) during transvenous lead extractions (TLEs).Methods and resultsNinety‐one high‐risk patients who underwent TLE in the operating room from January 1, 2015, to March 31, 2017, were included in the study. Of these, 9 patients underwent VATS during TLE. Their clinical characteristics, indications for lead extraction, and complications associated with TLE in the 9 patients who had VATS were compared with those for the 82 patients who did not have VATS. The mean (SD) age of the study patients was 61 (17) years (64.8% were male). The lead dwell time, number of leads extracted, and clinical comorbidities were similar between the 2 groups. Superior vena cava (SVC) tear occurred in 2 of the 9 patients in VATS group and in 1 of the 82 in the non‐VATS group (22.2% vs. 1.2%, P = 0.03). Of the 2 patients in the VATS group who had SVC tears, in 1 the tear was visualized immediately and there was no hemodynamic compromise. In the other patient, the SVC tear was within the pericardium; the blood pressure recovered quickly after sternotomy and repair. Both patients had complete lead extraction and survived hospitalization. The patient in the non‐VATS group who had an SVC tear had a successful repair but died of postoperative complications.ConclusionsUtilization of VATS to facilitate TLE is beneficial for early recognition of SVC tear and timely surgical repair in select high‐risk patients.

Background Diabetes mellitus (DM) is a risk factor for morbidity and mortality in patients with heart failure. The effect of DM on post–left ventricular assist device (LVAD) implantation outcomes is unclear. This study sought to investigate whether patients with DM had worse outcomes than patients without DM after LVAD implantation and whether LVAD support resulted in a better control of DM. Methods and Results We retrospectively reviewed 341 consecutive adults who underwent implantation of LVAD from 2007 to 2016. Patient characteristics and adverse events were studied and compared between patients with and without DM. One hundred thirty-one patients (38%) had DM. Compared with patients without DM, those with DM had higher rates of ischemic cardiomyopathy, LVAD implantation as destination therapy, and increased baseline body mass index. In a proportional hazards (Cox) model with adjustment for relevant covariates and median follow-up of 16.1 months, DM was associated with increased risk of all-cause mortality (hazard ratio, 1.73; 95% confidence interval: 1.18–2.53; P =0.005) and increased risk of nonfatal LVAD-related complications, including a composite of stroke, pump thrombosis, and device infection (hazard ratio, 2.1; 95% confidence interval: 1.35–3.18; P =0.001). Preoperative hemoglobin A1c was not significantly associated with mortality or adverse events among patients with DM. LVAD implantation resulted in a remarkable decrease in hemoglobin A1c levels (7.4±1.9 pre-LVAD versus 6.0±1.5 and 6.3±1.4 after 3 and 12 months post-LVAD, respectively; P <0.0001) and a significant reduction in requirements of DM medications. Conclusions DM is associated with increased rates of all-cause mortality and major adverse events despite favorable glycemic control after LVAD implantation.

Wait times have increased for patients approved for heart transplants. We reviewed United Network for Organ Sharing (UNOS) data for 14,242 patients listed for isolated heart transplant (2009–2013) to develop a risk score model for timing left ventricular assist device (LVAD) implantation in bridge-to-transplant patients. We used a multivariable Cox proportional hazards regression model with subsequent bootstrap resampling for internal validation to develop a scoring system that combined risk factors, weighted by the corresponding regression coefficients, to define an individual’s risk score. Four risk factors were identified (body mass index, blood type, region, and urgency status) to be significantly and independently associated with wait time (p < 0.001), showing adequate model discrimination (C = 0.704) and calibration. Higher risk scores correlated with shorter wait times. Our model corresponded closely with observed transplant rates, predicting longer wait times for lower status, larger size, certain blood groups, and some UNOS regions. This tool has the potential to more accurately describe the wait-time duration for an individual patient, which may influence care decisions. The wait-time discrepancies (blood types/regions) reinforce the need to reevaluate the geographic-allocation policy. The proposed review of the UNOS heart allocation policy may make this model especially relevant.