Stroke and death risk in ventricular assist device patients varies by ISHLT infection category: An INTERMACS analysis.

BACKGROUND: Ventricular assist device (VAD) patients often experience infections, which increase the risk of stroke and mortality. Using the definitions of the International Society for Heart and Lung Transplantation (ISHLT), we have characterized differences in clinical outcomes for categories of infection: VAD-specific (e.g., pump component related); VAD-related (e.g., bloodstream infection, BSI); and non-VAD infections (e.g., pneumonia). METHODS: Querying of the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) identified 16,597 continuous-flow VAD recipients. Categories of infection were tested in multivariate models to determine the risk of stroke and death. RESULTS: After implant, 7,046 patients (42%) developed an infection at a median of 69 (interquartile range 12 to 272) days. A majority were non-VAD infections (49%), followed by VAD-related (26%) and VAD-specific infections (25%). BSIs were the most common form of VAD-related infection (92%), and the majority (59%) had no associated infection, that is, idiopathic bacteremia. Internal pump component infections were rare (0.003 event per patient-year [EPPY]). Infected VAD patients had a higher prevalence of stroke compared to patients without an infection (18% vs 11%, p < 0.001). The lowest stroke rate occurred after a VAD-specific infection (0.11 EPPY) compared with VAD-related (0.17 EPPY) and non-VAD infections (0.15 EPPY, p < 0.001). Hemorrhagic strokes were more common than ischemic strokes in all infection groups and highest after a VAD-related infection (0.13 EPPY). One-year survival after an infection was 87% in VAD-specific infections, as compared with VAD-related (71%) and non-VAD infections (72%, p < 0.001). CONCLUSIONS: The ISHLT categorization of VAD infections unveils notable differences in associated risk of stroke and mortality. A re-assessment of transplant prioritization for eligible infected VAD patients may be useful to increase transplant-related survival benefit.

Multicenter experience with durable biventricular assist devices.

BACKGROUND: Severe right ventricular failure necessitating a right ventricular assist device (RVAD) complicates 6% to 11% of left ventricular assist device (LVAD) implants. Patient outcomes for those receiving durable continuous-flow VADs in a biventricular configuration (i.e., BiVAD) have been reported in limited case series. METHODS: Data from United States centers with ≥ 6 BiVAD implants were collected. Characteristics and outcomes of patients receiving contemporaneous (i.e., same surgery) vs staged implantation of the HVAD as a BiVAD were compared. RESULTS: From 2011 to 2017, 46 patients received durable BiVADs and had the following characteristics: median age, 46 years (interquartile range [IQR], 19-67 years), non-ischemic cardiomyopathy (80%), bridge to transplant (83%), Interagency Registry for Mechanically Assisted Circulatory Support Profile 1 or 2 (92%), use of temporary circulatory support (37%), right atrial pressure 19 mm Hg (IQR, 14-23 mm Hg), and cardiac index of 1.6 liters/min/m2 (IQR, 1.2-2.1 liters/min/m2). Operative mortality was 33%. Equal numbers of patients received a right atrial or right ventricular implant. Contemporaneous BiVAD implantation occurred in 31 patients (67%), and compared with 15 patients (33%) with staged implants, these patients had a shorter intensive care unit length of stay of 12 days (IQR, 7-23 days) vs 42 days (IQR, 28-48 days, p = 0.035) and were more likely to be discharged from the hospital on BiVAD support (61% vs 27%, p = 0.04). RVAD thrombosis developed in 17 patients (37%). Patients with contemporaneous BiVAD implants had a 1-year survival of 74% compared with 40% in staged BiVAD patients (p = 0.11). CONCLUSIONS: Patients receiving durable BiVADs represent a critically ill patient population with severe biventricular failure who have high operative mortality and RVAD thrombosis rates. The 1-year survival for patients receiving contemporaneous BiVADs in experienced centers mirrors other contemporary durable biventricular support strategies.

Unrecognized Left Heart Failure in LVAD Recipients: The Role of Routine Invasive Hemodynamic Testing.

The role of routine right heart catheterizations (RHCs) in left ventricular assist device (LVAD) patients is undefined. We analyzed 105 continuous-flow LVAD recipients who underwent an RHC approximately 3 months after implant. In 38 patients, LVAD speed was ramped with the goal of optimizing hemodynamics. Our cohort consisted of 71 (68%) HeartMate II (HMII) and 34 (32%) HeartWare (HVAD) patients. Thirty patients (29%) had either a reduced cardiac index (CI ≤ 2.2 L/min/m), elevated pulmonary capillary wedge pressure (PCWP > 18 mm Hg), or both. A subgroup of 38 patients (19 with abnormal hemodynamics) underwent LVAD ramping. With LVAD ramping, normalization of hemodynamics was achieved in 13 (68%) patients with abnormal hemodynamics. In ramped patients, the CI increased from 2.1 L/min/m (2.0-2.3) to 2.5 L/min/m (2.3-2.6; p = 0.004), and the PCWP dropped from 21 mm Hg (20-26) to 18 mm Hg (14-21, p < 0.001). The 6-minute walk distance improved from 338 m (253-394) to 353 m (320-442, p = 0.041). A 400 rpm change in HMII speed was like a 130 rpm change in HVAD speed and led to a change in cardiac output (CO) of 0.3 L/min. The correlation between device-reported flow and measured CO for both the HMII (Rs = 0.50, p < 0.001) and HVAD (Rs = 0.47, p < 0.001) was moderate. At 3 months after LVAD implant, most patients have normal hemodynamics. Of those patients with abnormal hemodynamics, LVAD ramping results in normalization of hemodynamics and improvement in 6-minute walk distance.