A Single-center Experience of CardioMEMS in Patients With Left Ventricular Assist Devices

Brief Report Cardiomems (CMEMS) have been demonstrated to reduce hospitalizations in symptomatic HF. The purpose of our study is to describe our experience of CMEMS in our 27 left ventricular assist device (LVAD) patients. We found that there was a 56% reduction in HF hospitalizations at 6 mo and up to 29% reduction in HF hospitalizations at 12 mo pre- and post-CMEMS placement. Further investigation is warranted in a large cohort with a particular focus on the effects on the hemocompatibility LVAD complications. Brief Communication Left ventricular assist devices are the standard of care for the treatment of patients with advanced heart failure (HF). Despite this therapy, nearly 20% of patients have residual HF symptoms resulting in increased morbidity, mortality, and decrease in quality of life. The CardioMEMS (CMEMS) pulmonary artery (PA) sensor allows for frequent collection of hemodynamic information and has been proven to reduce HF hospitalizations in chronic HF patients.1,2 Data regarding efficacy in the LVAD population is limited to small case series. We describe our single-center experience using the CMEMS system in this patient population. We performed a retrospective study of 27 LVAD patients who underwent placement of the CMEMS device between August 2018 and January 2020 at Ascension St. Vincent Data was extracted through the electronic medical record systems. Patients were selected based on high-risk features such as frequent hospitalizations pre CMEMS implant. Baseline characteristics and outcome analysis including HF hospitalizations occurring in the 1 y pre- and 6 mo and 1 y post-CMEMS implantation in these patients were collected. All implants were performed after LVAD placement without interruption of anticoagulation, with the majority (89%) performed by the right internal jugular approach without interruption of anticoagulation. Technique was similar among implanting cardiologists including gaining access with an 11 French Terumo sheath and either a 260 cm Cardiomems guidewire or 300 cm Steelcore to selectively engage the appropriately sized vessel in the left PA. A hand injection of a mixture of ~10 mL of iodinated contrast and normal saline was then used under fluoroscopy for vessel identification, followed by re-engagement of the vessel with the wire and the implant then performed as described by the IFU. At the time of implant, goal PADP was set for the patient, and subsequent PA pressure data was then used to adjust diuretics, guideline-directed medical therapy, or speed adjustments in a stepwise fashion at the discretion of the patient’s primary cardiologist or CMEMS implanter. Baseline demographics were collected and included a mean age of 56 with 71% male with all with end-stage HF with LVAD support with HM3 (55%), HM2 (9%), and HVAD (36%) with 43% as a bridge to transplant. There was no significant difference in pulmonary vascular resistance between LVAD recipients with and without CMEMS. As seen in Figure 1, there was a 58% reduction in HF hospitalizations at 6 mo and up to 29% reduction in HF hospitalizations at 12 mo pre- and post-CMEMS placement. When HF hospitalization was further delineated by primary or secondary diagnosis, this reduction was even more marked (64% at 6 mo and 47% at 1 y, respectively). The procedure was remarkably safe in this anticoagulated, continuous flow population, with only 3% of patients developing scant hemoptysis that resolved without any intervention.Figure 1.: Primary endpoint of HF hospitalizations.Major limitations to these observations exist including the limited number of patients examined in a retrospective and single-center nature. Additionally, patients were selected for PA sensor implant at the discretion of their primary cardiologist and influenced by factors such as adherence to treatment recommendations and monitoring, distance traveled to LVAD center and other clinical characteristics that may introduce significant bias. Hospitalizations remain a large burden for LVAD patients, with some reports of >70% of patients needing hospitalization within the first-year postimplant.3,4 Elevated filling pressures may not only lead to HF hospitalizations but also lead to hemocompatibility related adverse events that may further complications associated with LVADs contributing to morbidity5,6 further hospitalizations. If the signal of reduced hospitalization for HF as both a primary and secondary diagnosis can be replicated in a larger population, PA sensor monitoring may present a safe and unique opportunity to decrease morbidity in a population that often struggles with the adverse events associated with LVAD therapy. Further investigation is required to examine the generalizability of these findings and whether other adverse events such as right ventricular failure, hemocompatibility related adverse events, or infection are truly impacted by more intensive hemodynamics monitoring.