Unexpected Contribution of Pulsatility to Pump Thrombosis

Purpose Diminished pulsatility during continuous flow left ventricular assist device (CF-LVAD) support has been hypothesized to be a causative factor for adverse events, such as neurological events (NE), pump thrombus (PT) and gastrointestinal bleeding (GIB). We assessed outcomes between CF-LVAD patients according to measured pulsatility. Methods Log-files from 61 HVAD (HeartWare) patients with pulsatility data were analyzed. Mean, peak, trough flow and pulsatility ( peak - trough flow ) were extracted. Flow pulsatility index (PI), defined as pulsatility/mean flow , was calculated for each patient at 1, 3, 6, 9 and 12 months. Patients were categorized into low, intermediate and high PI tertiles ( table 1) . Baseline demographic factors and outcomes (NE, GIB, PT) were compared. NE included both cerebrovascular accidents and transient ischemic attacks. One way ANOVAs and chi-square analyses were performed to analyze continuous and categorical variables respectively. Results There were no significant differences in demographics between the 3 groups according to pulsatility. Surprisingly, there was a significantly higher rate of PT in the high PI group compared to the low and intermediate groups (p=0.013). Factors that contribute to the PI demonstrated that low trough flow was an independent predictor of PT according to logistic regression (OR 0.205, p=0.028). Marginally higher rates of NE were found in the high PI group compared to the low and intermediate groups, however this was not statistically significant (p=0.30). There were no differences in rates of GIB between groups . Conclusion High PI, particularly driven by low trough flow, was associated with an increased risk of PT highlighting an important therapeutic target for potential prevention or intervention for PT.