Discharge Glomerular Filtration Rate (GFR) as a Predictor of Outcomes Post Left Ventricular Assist Device Placement

Abstract

Methods: We identified HM II implants from 9/2009 to 10/2013 at our single center. Patients who required RVAD or ECMO post-operatively or had LVAD exchange were excluded. We created two cohorts: patients with a power spike (defined as pump power O10 watts) during the first 21 post-operative days (n 5 49) vs. patients without a power spike (n 5 150). Over 70,000 pump powers were recorded (median time between measurements of 60 6 83 minutes). Suspected pump thrombosis was defined as direct observation of obstructive thrombus in the pump or conduit following pump exchange or severe hemolysis (defined as lactate dehydrogenase level O1,000 mg/dL in the absence of a kinked inflow or outflow cannula). Results: Over a mean follow-up of 13.2 6 10.8 months, suspected pump thrombosis occurred in 32 of 196 patients. Kaplan-Meier estimates of freedom from pump thrombosis (Figure 1) were significantly higher in patients who did not have power spikes compared with patients who had power spikes (Wilcoxon, P 5 0.026). In Cox proportional hazard models adjusting for age, BMI, INTERMACS profile the HR for hemolysis was 1.91 (95% CI 0.90 4.04, p50.09). The median time between power spike and suspected pump thrombosis was 41 6 231 days, with nine of the eleven events occurring within 103 days of the power spike. Of the 32 patients with suspected pump thrombosis, 20 underwent pump exchange. Fourteen of the 20 had confirmed pump thrombosis. One patient had pannus on his outflow graft, and 5 patients had pumps removed with no thrombus seen in the OR. Conclusion: We conclude that there appears to be an association between HM II LVAD patients with unexplained early power spikes during the first 21 days of HM II support and subsequent suspected pump thrombosis. However, after adjusting for known factors associated with pump thrombosis, this association loses statistical significance. Thus, this association may represent an early event in the natural history of pump thrombosis, an inherent difference in the pump itself, or a marker of patient risk due to intrinsic patient factors.