Novel Use of EnSite NavX System for Real-Time Cardiac Hemodynamic Measurements

S50 Journal of Cardiac Failure Vol. 14 No. 6S Suppl. 2008 these devices poses a challenge because patients often do not have palpable pulse and rarely have an ausculatory blood pressure. Convenient and reliable methods of assessing blood pressure are needed. The goal of the study is to develop a better understanding of non-invasive techniques to measure blood pressure in these patients and determine the most reliable method or measuring systolic blood pressure. Methods: In the immediate post operative period when patients had an arterial line, we measured blood pressure by automatic blood pressure cuff, auscultation, doppler ultrasound, and palpation at various time intervals. Results: Blood pressure was measured in 7 LVAD recipients at a total of 37 separate time points. We were able to obtain a Doppler pressure 33 times (a success rate of 89.2%). By comparison, we were able to obtain a pressure 21 times by automatic cuff (56.8%), 6 by auscultation (16.2%) and 2 by palpation (5.4%). There was a strong correlation between Doppler measurements and arterial line systolic pressure (r2 5 0.794, p # 0.0001). We also examined this relationship in the presence or absence of vasopressors (epinephrine or vasopressin). Of 20 measurements when patients were on pressors, we were able to obtain a Doppler pressure 16 times and the correlation coefficient was 0.559, (p 5 .0122). All 17 measurements attempted with patients not on vasopressors yielded a doppler pressure with a correlation coefficient of 0.926. (p ! 0.0001)

Conclusions: In patients with continuous flow VADs, a blood pressure can occasionally be obtained by auscultation or automatic blood pressure machines. Doppler measurements are the most reliable method and closely correlates to the systolic pressure. Furthermore, in the setting of vasopressors, an arterial line is the only accurate way to assess blood pressure.

158 A Propensity Adjusted Comparison of Mitral Valve Repair Versus Replacement in Patients with Severe LV Dysfunction Eric S. Weiss1, Nishant D. Patel1, Duke E. Cameron1, John V. Conte1, Ashish S. Shah1; 1Division of Cardiac Surgery, Dept of Surgery, The Johns Hopkins University School of Medicine, Baltimore, MD

157 Novel Use of EnSite NavX System for Real-Time Cardiac Hemodynamic Measurements Kyungmoo Ryu1, Stuart P. Rosenberg1, Allen Keel1, Kjell Noren1, Wenbo Hou1, Steve Koh1, Jeff Schweitzer1, Curt Deno1, John Hauck1, Thao Thu Nguyen1, Michael Yang1; 1St. Jude Medical, Inc, Sylmar, CA Up to 30% of cardiac resynchronization therapy (CRT) patients do not realize benefit from CRT. Thus, any tool that improves CRT patient outcomes would be of value. We present here a feasibility study on an EnSite NavX based cardiac performance measurements (CPM) method, hypothesizing that it provides real-time, operator-independent assessment of myocardial wall motion and cardiac performance during CRT implants. Methods: We studied 3 HF canine models induced by 4e6 weeks of rapid right ventricular apical (RVA) pacing and left bundle branch block induced by RF ablation. Leads were implanted in right atrium (RA), RVA, and accessible coronary sinus (CS) branch (dog 1: great cardiac vein [GCV]; dogs 2 and 3: lateral branch [LB]). At the terminal study a multielectrode catheter was placed in another accessible CS branch (dog 1: LB; dogs 2 and 3: GCV). A pressure catheter was placed by retrograde access in the left ventricle (LV). Real-time positions of all intracardiac electrodes were recorded by the EnSite NavX mapping system during 4 pacing interventions: RA, RA + RV, RA + LV, and RA + BIV. Wall to wall dyssynchrony, temporal dispersion of peak motion, and magnitude of the RV-to-LV vector were computed from recorded 3-D positioning of all electrodes over 10 consecutive beats in each configuration. dP/dtMax for each intervention was also similarly computed. Results: A representative example of computed RV-to-LV dyssynchrony, temporal dispersion of peak motion, and RV-to-LV vector is shown in the figure. The trend of dP/dtMax (line graph) across the different interventions resembles that of the NavX CPM-derived LV motion parameters (bar graph). The magnitude of the RV-to-LV vector demonstrated the improved systolic and diastolic function of the LV when BIV was paced. Conclusions: NavX CPM demonstrated to be a promising approach for trending cardiac performance, guiding CRT implantation, and device setting. The results warrant a human feasibility study.

Objective: The optimal mitral valve (MV) procedure for patients undergoing CABG with functional mitral regurgitation (MR) is unknown. We examined our experience with MV repair vs. replacement in patients (pts) with functional MR and ejection fraction (EF) !35%. Methods: We reviewed pts with EF ! 35% who underwent CABG with mitral ring annuloplasty repair vs. chordal sparing replacement from 1995e2006. We assessed short and mid-term outcomes including postoperative inotrope requirement, cardiac performance, mortality and recurrence of MR in a propensity adjusted multivariable logistic regression model. Results: Of 93 pts meeting inclusion criteria, 63 had MV repair and 30 had replacement. Repair and replacement patients were matched for age (63y vs. 65y, p 5 0.6) and preoperative EF (25% vs 28%, p 5 0.8). Propensity scores predicted repair accurately and refect underlying differences in patient characteristics (0.87 vs 0.28, repair vs replacement, p ! 0.001, ROC 0.94) Replacement patients had longer postoperative length of stay (25 vs. 18 days, p 5 0.05) Inotrope requirements (first 48 hours) were not different between groups nor was 24 hour postoperative cardiac index (CI) (2.7 vs 2.5, p 5 0.5).Thirty-day mortality was 11% (7/63) and 10% (3/30) for the repair and replacement groups, respectively (p 5 1.0) Actuarial 1 and 5 year survival was 86% and 63% for the repair group and 73% and 50% for the replacement group (p 5 0.23).