Search

Background Appropriate collection of quality of life (QOL) measurements for left ventricular assist device (LVAD) patients is challenging. Patient-Reported Outcomes Measurement Information System (PROMIS) is a popular tool that has been validated across multiple disciplines, but its applicability to the LVAD population remains unknown. Methods This single-center, retrospective review included LVAD patients that completed a PROMIS assessment and Kansas City Cardiomyopathy Questionnaire (KCCQ-12) survey at clinical encounters postoperatively. Patients completed computer adaptive PROMIS assessments for physical function, pain interference and depression. All PROMIS domains are designed to follow a normal distribution (mean T-score 50, standard deviation 10) in the general population. Assessments were aggregated over time and correlation between the KCCQ-12 summary score and each PROMIS domain was assessed individually. Results A total of 178 LVAD patients were included in the study. The median time between LVAD implantation and PRO collection was 16.5 [interquartile range, 7.9-37.8] months. Patients typically had worse physical function (T-score 38.8 [33.6-44.2]) but comparable pain (51.1 [38.7-59.2]) and depression (49.9 [41.7-57.5]) as the general population. The KCCQ-12 was more strongly correlated to PROMIS physical function (Spearman’s ρ = 0.746) than pain (ρ = -0.539) or depression (ρ = -0.591). Conclusions PROMIS provides a robust QOL data collection system that can be implemented in a clinical setting without imposing a significant burden. Using this more holistic system may allow for better patient-centered care in order to address QOL limitations imposed by LVAD support that are not directly related to heart failure symptoms.

Background Venoarterial (VA) extracorporeal membrane oxygenation (ECMO) undoubtedly saves many lives, but it is associated with a high degree of patient morbidity, mortality, and resource use. This study aimed to develop a machine learning algorithm to augment clinical decision making related to VA-ECMO. Methods Patients supported by VA-ECMO at a single institution from May 2011 to October 2018 were retrospectively reviewed. Laboratory values from only the initial 48 hours of VA-ECMO support were used. Data were split into 70% for training, 15% for validation, and 15% withheld for testing. Feature importance was estimated, and dimensionality reduction techniques were used. A deep neural network was trained to predict survival to discharge, and the final model was assessed using the independent testing cohort. Model performance was compared with that of the SAVE (Survival After Veno-arterial ECMO) score by using a receiver operator characteristic curve. Results Of the 282 eligible adult patients who were undergoing VA-ECMO, 117 (41%) survived to discharge. A total of 1.96 million laboratory values were extracted from the electronic medical record, from which 270 different summary variables were derived for each patient. The most important variables in predicting the primary outcome included lactate, age, total bilirubin, and creatinine. For the testing cohort, the final model achieved 82% overall accuracy and a greater area under the curve than the SAVE score (0.92 vs 0.65; P = .01) in predicting survival to discharge. Conclusions This proof of concept study demonstrates the potential for machine learning models to augment clinical decision making for patients undergoing VA-ECMO. Further development with multi-institutional data is warranted.

Background Despite limited evidence, systemic anticoagulation is the current standard of care for patients supported on venoarterial-extracorporeal membrane oxygenation (VA-ECMO). We hypothesized that not anticoagulating patients on VA-ECMO would decrease complications. Methods We retrospectively reviewed adult patients supported on VA-ECMO at our institution. Patients were stratified based on anticoagulation strategy. The primary outcome was a composite of hemorrhagic and thrombotic complications. Secondary outcomes included blood product use, incidence of heparin-induced thrombocytopenia (HIT), hospital length of stay, and in-hospital mortality. Results From May 2011 through January 2018, there were 203 eligible patients supported on VA-ECMO, 35% (75 patients) were not anticoagulated. Overall complication rates were significantly lower for the no anticoagulation group (57% versus 76%; P = .007) including a trend toward fewer hemorrhagic complications (53% versus 63%; P = .178) without increased risk of thrombosis (13% versus 21%; P = .147). The anticoagulated group required more transfusions of packed red blood cells (12.8 versus 1.09; P = .002) and platelets (3.0 versus 1.3; P = .009) and showed a higher incidence of HIT (8% versus 0%; P = .015). No difference was found in overall mortality (72% versus 62%; P = .165). Conclusions The absence of routine systemic anticoagulation for patients supported on VA-ECMO is not associated with higher mortality, pump failure, or thrombotic complications. Patients had a lower requirement for blood product transfusions, and there was no incidence of HIT. Patients supported on VA-ECMO without other indications for anticoagulation can be treated without systemic anticoagulation during their VA-ECMO course.

Background Left ventricular assist device (LVAD) therapy has been the standard of care for selected patients with advance heart failure. Even though considerable strides have been achieved with the introduction of the newest centrifugal pump, therapy is still burdened with significant perioperative complications. Smaller devices, along with improved techniques and instruments, have encouraged the adoption of minimally invasive cardiac surgery (MICS) techniques for LVAD implantation to improve perioperative outcomes. Methods We describe a technique for complete sternal-sparing (CSS) HeartMate 3 (Abbott Laboratories, Abbott Park, IL) left ventricular assist device (LVAD) implantation using bilateral thoracotomies and discuss early clinical outcomes of the first ten consecutive patients who underwent CSS implantation of the HeartMate 3 LVAD at our institution. Results The median length of stay in the intensive care unit was 3.5 days. No patients required reexploration for postoperative bleeding. There was no incidence of right ventricle failure, stroke, renal failure, hepatic failure, or myocardial infarction. The median length of hospitalization after LVAD implantation was 14.5 days (interquartile range, 12 to 17 days). Conclusions Our early outcomes suggest that the CSS approach may reduce the incidence of right ventricular failure, bleeding, intubation time, and intensive care unit length of stay. Further studies are needed to fully elucidate the advantages of CSS LVAD implantation compared with median sternotomy.

Background The Cox-Maze III remains the gold standard for the surgical treatment of atrial fibrillation. However, the "cut-and-sew" technique is time consuming and technically challenging. The pulmonary veins are the source of ectopy in the majority of patients with atrial fibrillation. The safety and efficacy of bipolar radiofrequency to electrically isolate the pulmonary veins was evaluated in a prospective multi-center trial. Methods Beginning in January 2002, 30 patients at three medical centers underwent pulmonary vein isolation using bipolar radiofrequency and were followed for 6 months. Twenty-four of the patients also underwent a modified Cox-Maze III. Electrical isolation of the pulmonary veins was confirmed with intraoperative pacing. Pulmonary vein patency was assessed by magnetic resonance imaging or three-dimensional computed tomography in 15 patients at 1 month. Results Mean age was 60.9 ± 11.7 years. Nineteen patients had paroxysmal atrial fibrillation. All pulmonary veins were isolated in every patient. The left pulmonary veins underwent 3.0 ± 1.4 applications for a total of 26.4 ± 10.5 seconds. The right pulmonary veins underwent 2.8 ± 1.1 applications for a total of 26.3 ± 12.6 seconds. There was no operative mortality. At 1 month, imaging revealed no evidence of pulmonary vein stenosis. At 6 months, 96% of patients were in normal sinus rhythm. Conclusions The use of bipolar radiofrequency for electrical isolation of pulmonary veins and to replace other Cox-Maze III incisions is safe and effective at controlling atrial fibrillation. This emerging technology may shorten and simplify the surgical management of atrial fibrillation.

Background . This multicenter prospective trial was designed to assess the safety and efficacy of using a robotically-assisted microsurgical system to create endoscopic coronary anastomoses. Methods . Thirty-two patients scheduled for elective primary coronary surgery underwent endoscopic anastomosis of the left internal thoracic artery (LITA) to the left anterior descending (LAD) artery. Three thoracic ports (two for instruments and one for a camera) were placed, and a robotic system was used to perform the LITA–LAD graft. Conventional techniques were used to perform the other grafts. Thirty-one patients underwent median sternotomy and 1 patient underwent a limited anterior thoracotomy. Results . Graft flow was measured in the operating room and averaged 37 ± 19 mL/min. Mean anastomosis time was 24 ± 9 minutes. There were three intraoperative revisions (9%). Two were for inadequate flow and one for an inadvertent injury. Each of these grafts was successfully revised by hand. There were no technical failures of the robotic system. Average postoperative length of stay was 5.5 ± 2.7 days. There were three reoperations for bleeding, but none of these were related to the LAD anastomosis. Two months following the operation, selective angiography revealed a graft patency of 93%. The patients have been followed for 16 ± 4 months. Conclusions . This initial prospective multicenter trial documents the feasibility of robotically-assisted coronary bypass grafting. Further trials are warranted to establish the safety and efficacy of this new technology.

Warm cardioplegic induction improves the ischemically "stressed" adult heart. However, it is rarely used in infants, despite the fact that many newborn hearts are stressed by other factors such as hypoxia. The need for amino acids as well as their mechanism of action has also not been studied.We first assessed the role of cardioplegic induction temperature in 10 nonhypoxic neonatal piglets undergoing 70 minutes of multidose blood cardioplegic arrest. Five piglets (group 1) received a cold (4 degrees C) induction, and 5 (group 2) a warm (37 degrees C) induction. Twenty-six other piglets underwent ventilator hypoxia (fraction of inspired oxygen, 8% to 10%) for 60 minutes before cardiopulmonary bypass (stress). Six piglets (group 3) then underwent 70 minutes of cardiopulmonary bypass without ischemia (hypoxia controls), and 20 underwent 70 minutes of cardioplegic arrest. Five of these (group 4) received cold cardioplegic induction, and 15 received warm induction; in 5 of these (group 5), the warm cardioplegic solution contained amino acids, in 5 others (group 6), it was unsupplemented, and in the remaining 5 (group 7), nitroglycerin was added to determine the role of vasodilation. Myocardial function was assessed by pressure-volume loops (expressed as a percent of control), and coronary vascular resistance was measured with cardioplegic infusions.In nonhypoxic (normal) piglets, cold (group 1) and warm (group 2) induction completely preserved systolic function (end-systolic elastance, 100% versus 104%) and preload recruitable stroke work (100% versus 102%), with minimal increase in diastolic compliance (162% versus 156%). Hypoxia-reoxygenation alone (group 3) depressed systolic function (end-systolic elastance, 51%+/-2%) and preload recruitable stroke work (54%+/-3%), and raised diastolic stiffness (260%+/-15%). The detrimental effects of reoxygenation persisted (unchanged from reoxygenation alone) with cold induction (group 4) or warm induction without amino acids (groups 6 and 7). In contrast, warm induction with amino acids (group 5) restored systolic function (end-systolic elastance, 105%+/-3%; p0.001 versus groups 3, 4, 6, and 7) and preload recruitable stroke work (103%+/-2%; p0.001 versus groups 3, 4, 6, and 7), and decreased diastolic stiffness (154%+/-7%; p0.001 versus groups 3, 4, 6, and 7). However, there was no difference in myocardial oxygen consumption in hypoxic hearts receiving a warm induction (6.9 versus 6.5 versus 7.3 mL/g per 5 minutes) (groups 5, 6, 7), and coronary vascular resistance was lowest with nitroglycerin (group 7).Cardioplegic induction can be given either warm or cold in nonhypoxic neonatal hearts. In contrast, only warm induction with amino acids repairs the hypoxic injury, but the primary mechanism of action is not related to increased metabolic activity or vasodilation.

Background Complex use of arterial conduits has resurrected concerns about the adequacy of conduit flow. The T-graft is the extreme example of this trend. Our purpose was to identify the limitation of single source inflow and to compare flow capacity with completion coronary flow. Methods Between February 1999 and November 2001, 372 patients underwent total arterial revascularization with the T-graft alone. Intraoperative flows were recorded for each limb of the T-graft before and after distal anastomoses in 204 patients. Independent predictors of T-graft flow were identified by multivariate analysis. Results Free flow for the radial arterial (RA) limb was 161 ± 81 mL/min, the internal thoracic artery (ITA) limb 137 ± 57 mL/min (combined 298 ± 101 mL/min) versus simultaneous limb flow of 226 ± 84 mL/min giving a flow restriction of 24% ± 14%. Completion coronary flow was 88 ± 49 mL/min for the RA, 60 ± 45 mL/min for the ITA, and 140 ± 70 mL/min for both limbs simultaneously to give a flow reserve (vs simultaneous free flow) of 160% or 1.6. Independent predictors of completion RA limb flow are RA proximal diameter ( p = 0.005), number of anastomoses ( p = 0.018), and target stenosis ( p = 0.005). Conclusions A flow reserve of 1.6 compares favorably with an ITA flow reserve of 1.8 at 1-month postoperatively and 1.8 for both the ITA T-graft and the ITA/RA T-graft at 1-week postoperatively as reported by others. Proximal RA diameter and competitive coronary flow influence completion T-graft flow. These data quantitate the limitation of single source inflow of the T-graft configuration and support its continued use.

Background Pinacidil solutions have been shown to have significant cardioprotective effects. Pinacidil activates both sarcolemmal and mitochondrial potassium-adenosine triphosphate (K ATP ) channels. This study was undertaken to compare pinacidil solution with University of Wisconsin (UW) solution and to determine if the protective effect of pinacidil involved mitochondrial or sarcolemmal K ATP channels. Methods Thirty-two rabbit hearts received one of four preservation solutions in a Langendorff apparatus: (1) UW; (2) a solution containing 0.5 mmol/L pinacidil; (3) pinacidil with Hoechst-Marion-Roussel 1098 (HMR-1098), a sarcolemmal channel blocker; and (4) pinacidil with 5-hydroxydecanote, a mitochondrial channel blocker. Left ventricular pressure-volume curves were generated by an intraventricular balloon. All hearts were placed in cold storage for 8 hours, followed by 60 minutes of reperfusion. Results Postischemic developed pressure was better preserved by pinacidil than by UW. This cardioprotective effect was eliminated by 5-hydroxydecanote and diminished by HMR-1098. Diastolic compliance was better preserved by pinacidil when compared with UW. This protection was abolished by the addition of 5-hydroxydecanote and moderately decreased by HMR-1098. Conclusions Our results support the superiority of pinacidil over UW after 8 hours of storage. The cardioprotective role of pinacidil is mediated primarily by the mitochondrial K ATP channel.

The radial artery (RA) can be used as either an aortocoronary (RA-Ao) or composite graft (T graft). Optimum use for the RA has yet to be established. We compared RA patency with these two techniques.Between October 1993 and June 2001, 1505 patients underwent coronary artery bypass grafting using the RA as either a composite (n = 1022) or RA-Ao graft (n = 483). Angiograms performed on 203 (13.5%) patients with signs or symptoms of ischemia at an average of 26.1 +/- 18.5 months postoperatively were reviewed.Patients with RA-Ao grafts had a greater incidence of postoperative angiography versus patients with composite grafts (19% versus 11%; p0.01). Patients receiving T grafts had a greater number of anastomoses per patient (4.1 +/- 0.6 versus 3.0 +/- 1.0; p0.01) and a higher incidence of total arterial revascularization (100% versus 41%; p0.01). Regardless of grafting strategy, patency was significantly worse for targets of the right coronary artery (58% T graft; 67% RA-Ao; p0.01 for both) and for targets with less than or equal to 70% stenosis (59% T graft; 57% RA-Ao; p0.01 for both). The site of proximal anastomosis failed to effect RA patency (relative risk, 1.2; 95% confidence interval, 0.7 to 1.8; p = 0.50).The site of the proximal anastomosis does not appear to influence patency. Both RA-Ao and composite conduits are sensitive to target location and stenosis. Advantages of composite grafting include greater conduit length and minimizing aortic manipulation at the expense of increased complexity and the potential for hypoperfusion. These factors should be considered when choosing an RA grafting strategy.

Background Alternative energy sources have been proposed for the transvenous and surgical treatment of atrial fibrillation. This study examined the physiologic consequences of a novel energy source, bipolar radiofrequency energy, in a chronic animal model in order to determine its ability to produce transmural lesions on the beating heart. Methods Five dogs underwent baseline pacing from the following target areas: right and left atrial appendage, superior and inferior vena cavae, and right and left pulmonary veins. A cuff of atrial myocardium, proximal to the target tissue was clamped and ablated between the arms of the bipolar radiofrequency energy device. Tissue conductance was used as a transmural indicator. After ablation, the pacing protocol was repeated. Baseline and postablation pulmonary vein flows were measured. Animals were survived for 30 days, and permanent electrical isolation was evaluated by pacing, epicardial mapping, and histology. Results Mean ablation time was 5.0 ± 1.8 seconds and mean peak tissue temperature was 46.7°C ± 2.8°C. All lesions (30/30) acutely and permanently isolated atrial tissue. There was no change in pulmonary vein flow. Mapping studies with pacing of atrial tissue on both sides of the lesion confirmed isolation. Histology demonstrated that all lesions were linear, continuous, and transmural with no thrombus formation or stenosis. Conclusions Bipolar radiofrequency energy rapidly produced permanent transmural linear lesions on the beating heart. Measurement of tissue conductance reliably predicted transmural lesions. This new technology may enable the development of a less invasive, surgical approach to atrial fibrillation.

Background Paraplegia or paraparesis after otherwise successful thoracic or thoracoabdominal aortic reconstruction is a devastating complication for patient and physician. Interventions for its prevention have focused primarily on the intraoperative period. We have recently noted a significant incidence of delayed-onset neurologic deficit. Methods We reviewed our most recent 5-year experience with thoracic and thoracoabdominal reconstruction to examine the incidence of and potential contributors to delayed paraplegia or paraparesis. Results Between June 1996 and June 2001, 60 patients (29 men, 31 women) underwent repair of isolated thoracic (n = 26) or thoracoabdominal aortic aneurysm (Crawford I, n=7; Crawford II, n=14; Crawford III, n=12; Crawford IV, n=1) by the cardiac and vascular surgical services collaboratively. Repair was performed endovascularly in 6, and open with either circulatory arrest in 12, partial left heart bypass in 37, or partial femorofemoral bypass in 5. Operative mortality was 9.3% (5 of 54 patients) for open repair and 0% for endovascular repair. Paraplegia or paraparesis occurred in 6 (10%) patients of which 83.3% (5 of 6) were delayed in onset. All patients with delayed paraplegia or paraparesis had degenerative aneurysms of Crawford extent II (n = 3) or III (n = 2), had intraoperative left heart bypass, and had perioperative spinal drainage. Delayed paraplegia or paraparesis occurred up to 27 days postoperatively, and was associated with a documented episode of hypotension in 60% (3 of 5) of patients. Conclusions Improvements in intraoperative management may have reduced immediate paraplegia or paraparesis among vulnerable patients only to leave them at risk of delayed-onset deficit. Postoperative care, including assiduous attention to avoidance of even transient hypotension, must be tailored to this patient population.