Faculty Bibliography

BACKGROUND: The HeartWare left ventricular assist device (HVAD, HeartWare Inc, Framingham, MA) is the first implantable centrifugal continuous-flow pump approved for use as a bridge to transplantation. An infrequent but serious adverse event of LVAD support is thrombus ingestion or formation in the pump. In this study, we analyze the incidence of pump thrombus, evaluate the comparative effectiveness of various treatment strategies, and examine factors pre-disposing to the development of pump thrombus. METHODS: The analysis included 382 patients who underwent implantation of the HVAD as part of the HeartWare Bridge to Transplant (BTT) and subsequent Continued Access Protocol (CAP) trial. Descriptive statistics and group comparisons were generated to analyze baseline characteristics, incidence of pump thrombus, and treatment outcomes. A multivariate analysis was performed to assess significant risk factors for developing pump thrombus. RESULTS: There were 34 pump thrombus events observed in 31 patients (8.1% of the cohort) for a rate of 0.08 events per patient-year. The incidence of pump thrombus did not differ between BTT and CAP. Medical management of pump thrombus was attempted in 30 cases, and was successful in 15 (50%). A total of 16 patients underwent pump exchange, and 2 underwent urgent transplantation. Five patients with a pump thrombus died after medical therapy failed, 4 of whom also underwent a pump exchange. Survival at 1 year in patients with and without a pump thrombus was 69.4% and 85.5%, respectively (p = 0.21). A multivariable analysis revealed that significant risk factors for pump thrombus included a mean arterial pressure > 90 mm Hg, aspirin dose /= 3 at implant. CONCLUSIONS: Pump thrombus is a clinically important adverse event in patients receiving an HVAD, occurring at a rate of 0.08 events per patient-year. Significant risk factors for pump thrombosis include elevated blood pressure and sub-optimal anti-coagulation and anti-platelet therapies. This suggests that pump thrombus event rates could be reduced through careful adherence to patient management guidelines.

BACKGROUND: The incidence of percutaneous lead failure among patients supported with a HeartMate II left ventricular assist device is unknown. METHODS: All HeartMate II left ventricular assist device driveline dysfunctions reported to Thoratec Corporation were retrospectively reviewed. The location and severity of driveline failures and their association with adverse clinical outcomes were examined. Also, the effect of design modifications was evaluated. RESULTS: Between 2004 and October 2012, 12,969 HeartMate II pumps were implanted worldwide. The incidence of percutaneous lead dysfunction was 1,418 events occurring in 1,198 pumps (9.2% of pumps) over a cumulative support period of 13,932 patient-years (maximum, 8 years). Lead failure was mostly in the externalized part of the cable (87.2%). Lead dysfunction was managed by clamshell reinforcement of the external connector strain relief or by tape or silicone cable reinforcement in 76% of cases. Mortality or significant morbidity, including pump exchange or urgent transplant, or more complex cable repair occurred in 2.3% of all implanted pumps. The cumulative incidence of lead failures leading to major adverse clinical events has decreased with two lead design revisions: at 18 months postimplantation, the incidence was 6.2%+/-1.2% for the original design versus 2.2%+/-0.5% for the latest design change introduced in 2010 (log-rank p<0.001). CONCLUSIONS: Lead failures remain an important factor in the durability of left ventricular assist devices during long-term support. Most lead failures in the HeartMate II occurred in the externalized portion of the driveline, suggesting lead fatigue. The incidence of both internal and external lead failures has diminished since 2004 with improvements in lead design.

Heart failure remains one of the most prevalent diseases worldwide and in recent decades, left ventricular assist devices (LVADs) have become an important treatment option. With increasing device experience, there is particular interest in the use of LVADs as a bridge to recovery that allows the patient's heart to undergo reverse remodeling, whereby the device can be explanted and the heart can function at an improved state. There are many considerations that play a role in this process, including the ability of the device to unload the heart, the innate physiology of the heart to recover and the use of concomitant therapies. This review provides an overview of the most current literature as it pertains to these processes and gives a view into the future directions of LVADs as a tool for achieving myocardial recovery.

OBJECTIVES: We designed a device that applies motion-activated energy (vibration) to prevent chest-tube clogging and maintain tube patency. We evaluated the efficacy of this device in vitro and in vivo. METHODS: The motion-activated system (MAS) device assembly comprises a direct current motor with an eccentric mass (3.2 g, centroid radius of 4.53 mm) affixed to its motor shaft. The device was tested in vitro using a model of an obstructed chest tube, with clots of bovine blood and human thrombin. The in vivo study (in nine healthy pigs, 46.0 +/- 3.3 kg) involved a bilateral minithoracotomy and placement of 32-Fr chest tubes (with and without the device). Whole autologous blood (120 ml) was injected every 15 min into the right and left chest each over 120 min total. RESULTS: Chest-tube drainage over these 2 h using the MAS was significantly higher than that without the device (369 +/- 113 ml vs 209 +/- 115 ml; P = 0.027). CONCLUSIONS: Our results suggest that the motion-activation of the chest tubes may be an effective tool to maintain chest tubes patent. Further optimization of this technology is required to obtain more consistent prevention of clot deposition within or outside the chest tubes.

BACKGROUND: The DuraHeart is a continuous centrifugal-flow left ventricular assist device that uses active magnetic levitation for impeller positioning designed for improved hemocompatibility and durability. This study reviews the results of the US trial with specific attention to hemolysis, thrombotic complications, and pump failure. METHODS: The US SUSTAIN trial was a multicenter, prospective, single-arm observational study in advanced heart failure patients listed for transplantation. Follow-up was complete in 100% of the patients at 6 months. RESULTS: Sixty-three patients were enrolled at 23 centers. Forty-six patients (73%) reached the primary end points of survival to transplantation, alive on the original device at 180 days and listed for transplantation, or explant for recovery. Median duration of support was 267 days (range, 10 to 952 days) with a total support time of 46 patient-years. There was no clinical hemolysis reported during the study. Mean lactate dehydrogenase values peaked at day 4 and significantly decreased during support (435+/-236 U/L and 297+/-142 U/L on day 3 and day 180, respectively). There were no cases of pump thrombosis reported, and 3 cases of pump thrombus "in transit" (0.06 events/patient-year) were observed. There were 6 (10%) cases of magnetic levitation system failure, all secondary to cable wire fractures (0.12 events/patient-year). All patients were hemodynamically stable with the backup hydrodynamic mode. Major adverse events included gastrointestinal bleeding (0.52 events/patient-year), ischemic and hemorrhagic strokes (0.17 events/patient-year and 0.09 events/patient-year, respectively), and driveline infections (0.67 events/patient-year). CONCLUSIONS: The DuraHeart demonstrated good hemocompatibility; however, the reliability of full magnetic levitation systems should be a high priority in future pump designs.

Right ventricular failure is a difficult problem to manage and typically carries a dismal prognosis. In the setting of post-left ventricular assist device implantation (LVAD), right ventricular dysfunction both in the early and late stages is of particularly high incidence and concern. There are currently no agreed upon preoperative algorithms to predict patients at risk for this problem, thus adding another level of complexity to treatment. Furthermore, there is no current technology available for chronic right ventricular support and the devices currently in use are LVADS modified to adapt to the right circulatory system. This review provides an overview of right ventricular failure, particularly after LVAD implantation, and describes the survival outcomes and continued challenges in this area.

BACKGROUND: Drive-line infections (DLIs) frequently complicate ventricular assist device (VAD) support. We sought to describe the detailed effects of DLIs over time in patients with continuous-flow VADs, including the onset, risk factors, organisms involved, association with invasive infections, and outcomes. METHODS: We reviewed data for patients with HeartMate II VADs (HMII) who were implanted at the Cleveland Clinic from October 2004 to September 2011 and followed through December 2011. DLIs were defined according to published criteria. RESULTS: DLIs developed in 45 of 194 HMII VADs over a median period of 232 days (range 22 to 883 days). Hazard for DLI was 2.0%/month, but transiently peaked at 11%/month at 7.5 months after implant. Pseudomonas aeruginosa accounted for 31%, 42% and 55% of initial, final and deep DLIs, respectively. Of the 40 superficial DLIs, 13 (32.5%) became deep. DLI-associated bacteremia and hospitalization occurred in 14 of 45 (31%) and 30 of 45 (67%), respectively. All patients received antibiotics (median 171 days), but only 3 of 44 (6.8%) developed an antibiotic complication. DLIs increased the risk for death while on VAD support (HR 2.20, 95% CI 1.20 to 4.05; p = 0.01). Six and 12 months after DLI, mortality was 9.8% and 31%, but the competing event of transplantation occurred successfully in 20% and 28%, respectively. CONCLUSIONS: Most DLIs begin superficially with peak hazard at 7.5 months after implant. Depth of infection and infecting organism may evolve over months on support, with Pseudomonas becoming more prominent. Although effectively managed for prolonged periods, DLIs are associated with reduced survival on VAD support. Earlier transplantation is the most successful approach to treatment.

Patients on peripheral extracorporeal membrane oxygenation (ECMO) are at risk for lower extremity ischemia. Effective monitoring is needed to identify complications quickly and allow timely correction. Near-infrared spectroscopy has been used extensively in cerebral monitoring during cardiac surgery. We present its use in monitoring lower extremity perfusion in patients on ECMO. Five patients on ECMO had near-infrared spectroscopy monitors placed on the calf of both lower extremities. Continuous real-time tissue oxygen saturation data (stO2) was displayed and recorded. Two patients had lower extremity complications in the leg with the arterial cannula. The patients with complications had lower stO2 in the cannulated leg at the time of ECMO insertion, larger differences in stO2 between the legs at the time of insertion, lower nadir stO2s, and larger peak differences in stO2 between the legs than patients without limb complications. The use of near-infrared spectroscopy for continuous monitoring of tissue oxygenation in the lower extremities in patients on ECMO may allow early identification of patients with lower extremity complications.

Number of left ventricular assist device (LVAD) implantations increases every year, particularly LVADs for destination therapy (DT). Right ventricular failure (RVF) has been recognized as a serious complication of LVAD implantation. Reported incidence of RVF after LVAD ranges from 6% to 44%, varying mostly due to differences in RVF definition, different types of LVADs, and differences in patient populations included in studies. RVF complicating LVAD implantation is associated with worse postoperative mortality and morbidity including worse end-organ function, longer hospital length of stay, and lower success of bridge to transplant (BTT) therapy. Importance of RVF and its predictors in a setting of LVAD implantation has been recognized early, as evidenced by abundant number of attempts to identify independent risk factors and develop RVF predictor scores with a common purpose to improve patient selection and outcomes by recognizing potential need for biventricular assist device (BiVAD) at the time of LVAD implantation. The aim of this article is to review and summarize current body of knowledge on risk factors and prediction scores of RVF after LVAD implantation. Despite abundance of studies and proposed risk scores for RVF following LVAD, certain common limitations make their implementation and clinical usefulness questionable. Regardless, value of these studies lies in providing information on potential key predictors for RVF that can be taken into account in clinical decision making. Further investigation of current predictors and existing scores as well as new studies involving larger patient populations and more sophisticated statistical prediction models are necessary. Additionally, a short description of our empirical institutional approach to management of RVF following LVAD implantation is provided.