Sundareswaran, Kartik S.

Introduction: Advanced age is a risk factor for poor outcomes following left ventricular assist device (LVAD) implant. A means of identifying elderly patients with acceptable mortality risk following LVAD surgery is not currently available. The HeartMate II (HMII) Risk Score (HMRS) was recently developed and validated, providing good risk discrimination for predicting survival in patients undergoing HMII implant. Independent survival predictors in the model included age and markers of renal function, coagulopathy and nutrition. We hypothesize that the HMRS will provide utility for risk stratification of the elderly LVAD candidate. Methods: Elderly patients (age >65 years) enrolled into the HMII Destination Therapy (DT) and Bridge to Transplant trials were selected for the study (n=444). Complete data for HMRS calculation was available in 98% (n=435). Preoperative patient HMRS was calculated according to the formula: HMRS = ln(1.29)*decade of age + ln(0.71)*Albumin + ln(1.37)*Cr + ln(1.86)*INR + ln(0.52)*LVAD Center Volume >=15 + ln(0.67)*Implant after 2007. Patients were trichotimized into high (HMRS >1.0), medium (HMRS 0.50-1.0), and low (HMRS <=0.50) risk categories (as previously defined in model development) based on HMRS. The primary endpoint was survival at 1 year following LVAD implant, estimated using Kaplan-Meier methods. Results: The mean cohort age was 71±4 years and 84% (n=372) had devices implanted for DT. The table shows the HMRS along with mean±stddev values for the score's components. All between-group differences were statistically significant (Kruskal-Wallis p<0.05), although differences in age were not clinically significant. Survival based on HMRS strata is shown in the figure. Patients in the low and medium risk groups had survivals >80% at 1 year following LVAD implant. Conclusions: The HMRS can be used to identify elderly patients with survivals comparable to that of the general LVAD population.

Introduction and hypothesis: Recent data have indicated an increase in the rate of early, confirmed pump thrombosis of the HeartMate II (HMII) left ventricular assist device. Factors related to this increase were not known, but one of the hypotheses was change in clinical management over time. The PREVENT ( PREVEN tion of Heartmate II Pump T hrombosis through Clinical Management) study was designed to evaluate the impact of clinical practices developed to reduce the risk of HMII pump thrombosis. Methods: PREVENT is a prospective, multi-center, single-arm, non-randomized study of up to 300 patients at up to 25 sites. Patient and pump management, including implant technique, anticoagulation, and pump settings are guided by protocol (Table 1). Primary endpoint is incidence of confirmed pump thrombosis at 3 months, with a total follow-up of 6 months. Presented here are the preliminary results from the first 100 subjects enrolled. Results: Median age was 61 years [25-79], 87% were male, and 79% were destination therapy. At implant, 9% were in INTERMACS Profile1, 68% Profiles 2-3, and 22% Profiles 4-7. Preliminary analysis of 90-day outcomes demonstrated an incidence of 1% for confirmed and 2% for suspected thrombosis with a 38% incidence of INTERMACS-defined bleeding. Adherence to the protocol recommendations was as follows: 91% of the pumps were anchored, 97% of the patients were bridged with heparin, median pump speed at 1 month was 9190 [range: 8400-10000], with 93% of the patients having a pump speed > 8600, and 78% of the patients having pump speeds > 9000. Median INR per patient was 2.2 [IQR: 2.0-2.4]. Overall Kaplan-Meier survival was 93%. Conclusions: Preliminary results are encouraging and suggest good adherence to the PREVENT protocol and low rates of pump thrombosis without a resultant increase in bleeding complications. Overall survival after HM II implantation remains excellent.

The PREVENtion of HeartMate II pump Thrombosis through clinical management (PREVENT) study was a multicenter, prospective investigation to evaluate the rate of pump thrombosis (PT) with adoption of a uniform set of surgical and medical practices for left ventricular assist device implantation. We sought to quantify pump position at baseline and retrospectively define a pump position associated with poor clinical outcomes. Chest x-rays at baseline were prospectively obtained per protocol. Pump pocket depth, inflow cannula (IC) angle relative to the pump, and IC angle relative to the vertical were measured. Pumps falling in the tail-ends of the IC angle and pump pocket depth distributions were categorized as having an extreme pump position within the PREVENT study. Patients with extreme pump position had a significantly higher risk of confirmed and suspected PT, hemolysis, and elevated lactate dehydrogenase. In a multivariable analysis of survival free of confirmed PT, extreme pump position was an independent risk factor (hazard ratio = 3.6; 95% confidence interval = 1.5–8.9; p = 0.006) when adjusting for differences in pump speed and anticoagulation level. Our analysis shows that HeartMate II pump position at implant can significantly impact event-free survival and the incidence of adverse events at 6 months.

Background— The HeartMate II (HMII) destination therapy (DT) trial demonstrated significant improvements in outcomes in continuous-flow left ventricular assist devices compared with patients implanted with the pulsatile-flow HeartMate XVE. The primary hypothesis of the current study is that trial patients enrolled after the initial data cohort would have better clinical outcomes. Methods and Results— Two hundred eighty-one patients who underwent HMII for DT from May 2007 to March 2009 (Mid Trial [MT] group) were compared with the initial 133 HMII patients from March 2005 to May 2007 (Early Trial [ET] group). Patient entry criteria were the same during the 2 time periods. Survival, adverse events, and quality of life were compared between the 2 groups. Baseline characteristics were similar between the groups. Compared with the ET group, patients in the MT group had reduced adverse event rates for bleeding requiring transfusions (1.66 versus 1.13 events per patient-year, P <0.001), sepsis (0.38 versus 0.27, P =0.025), device-related infections (0.47 versus 0.27, P <0.001), and hemorrhagic stroke (0.07 versus 0.03, P =0.01). Other event rates were similar between groups including ischemic stroke (0.06 versus 0.05 events per patient-year, P =0.57). Survival at 1 year in the MT group was 73% versus 68% in the ET group ( P =0.21). Additionally, there was a significant reduction in deaths caused by hemorrhagic stroke ( P =0.01). Quality of life improvements were significant in both the groups ( P <0.001). Conclusions— The benefit of DT therapy with the HMII is confirmed in subsequent trial patients, with improved adverse event rates and a strong trend for improvements in survival. Clinical Trial Registration— URL: http://www.clinicaltrials.gov . Unique identifier: NCT00121485.

Although the Fontan procedure has been enormously successful in palliation of single-ventricle patients, many seem to experience progressive failure of the Fontan circulation over time. Ventricular assist devices (VADs) have developed into stable platforms for long-term support of adult patients with heart failure. Given the success of axial flow devices, it was hypothesized that the technology could provide clinical benefit to failing Fontan patients. The aim of this study was to use a computer model to evaluate VAD support in failing Fontan physiology. A computer model of Fontan circulation with heart failure was developed and the HeartMate II (HM II) (Thoratec Corp) axial flow ventricular assist device was connected to this model in a systemic configuration to examine its impact. Cardiac catheterization data from 7 patients (8 catheterization studies) with failing Fontan physiology were applied to this model to evaluate the impact of using the HM II in this manner. When the HM II was used in a systemic configuration at 8000 rpm, there was a 35% decrease in the systemic venous pressure in the Fontan circuit and a 63% decrease in pulmonary capillary wedge pressure with a resultant 41% increase in CI. The model also predicted patient-specific parameters where the VAD may not benefit the patient, such as fixed elevated pulmonary vascular resistance, low systemic ventricular end-diastolic pressure, and high unresponsive systemic vascular resistance. These data suggest a potential benefit from application of axial flow VAD technology in the management of failing Fontan physiology. Clinical correlation will allow for refinement of this model as a predictive tool in discerning which patients may benefit from placement of a VAD and what issues must be addressed prior to implanting the device.