The true incidence and risk factors of hemolysis in patients with Impella CP placement

Abstract Background The Impella, a micro axial-flow pump, is effective in treating cardiogenic shock. However, compared to other mechanical circulatory support devices, its high pump speeds can result in complications like hemolysis and subsequent renal failure. Previous reports have defined hemolysis based on the elevation of LDH levels or the presence of hemoglobinuria, but this approach may underestimate its occurrence. Hemolysis is commonly assessed using the hemolysis index, which is calculated through biochemical auto-analysis based on the absorbance measure of the red color change. However, since the hemolysis index is not highly sensitive, it is recommended to measure plasma free hemoglobin (PFHb) for accurate diagnosis of hemolysis. Purpose The aim of this single-center retrospective cohort study is to investigate the frequency of hemolysis in Impella CP using PFHb measurements and its association with the Impella support level. Methods Between December 2019 and June 2023, we included 116 patients who had the Impella CP device implanted at our hospital. For these patients, PFHb was measured twice daily during the Impella placement period (up to 13 days). A PFHb value of 0.06 or higher was considered indicative of hemolysis, at which point a haptoglobin preparation of 4000 units was administered. Separately, the hemolysis index was defined as hemolysis at a level of 41 mg/dL. We investigated the relationship between daily Impella settings, PFHb, and hemolysis index for each patient. Results The average age was 70 ± 13 years. 59% underwent ECPELLA, and 15% were introduced to renal replacement therapy (RTT) due to severe renal failure. During Impella placement, hemolysis was diagnosed in 33% of patients based on PFHb, while only 18% were diagnosed with hemolysis according to the hemolysis index. Upon separate analysis for Impella and ECPELLA during the initial three days, the support levels of Impella were significantly higher in the Impella group on Days 1 and 2, with no significant difference noted on Day 3. Moreover, PFHb levels were significantly higher in the Impella group on Day 1, while on Day 3, the levels were higher in the ECPELLA group. Patients on RRT had significantly higher max PFHb levels than those not on RRT (0.19 vs. 0.06 g/dL; p < 0.001). Next, we investigated the relationship between PFHb levels and Impella support levels in patients with either Impella or ECPELLA over 711 data points. Dividing Impella support into low (P1-3), middle (P4-6), and high (P7-9) categories, PFHb values were significantly higher at high P levels compared to low and middle P levels. Similar results were observed for both Impella and ECPELLA. Conclusion There is a possibility of underestimating hemolysis associated with Impella if PFHb measurement is not performed. Hemolysis showed a significant correlation with higher P levels (P7 and above) for both Impella and ECPELLA.