Abstract 4341144: Prospective Use of Hemolysis Index as a Surrogate for Clinical Hemolysis in Patients Receiving Temporary Mechanical Circulatory Support: Pilot Study

Introduction: Rates of temporary mechanical circulatory support (tMCS) have increased by more than 300% since FDA approval in the 1990s. With increased utilization, it is imperative for critical care providers to diagnose and manage the most common adverse events associated with MCS, such as hemolysis. Hemolysis is a known independent risk factor for mortality and acute renal failure. Despite these clinical implications, not every hospital system has onsite access to plasma free hemoglobin (pfHb), the gold standard diagnostic test for hemolysis. The Hemolysis Index (HI) is a new spectrophotometric method initially developed to detect blood sample quality in chemistry analyzers and has become a promising avenue for reliable, inexpensive, and fully automated hemolysis detection. Previous studies have shown a strong correlation between the hemolysis index and plasma free hemoglobin (r > 0.90), however, no studies have validated HI as a surrogate marker for hemolysis in patients receiving tMCS. Hypothesis: HI is a feasible, rapid, and reliable alternative to pfHb with a correlation coefficient > 0.90 in patients undergoing tMCS. Methods: We conducted a single-site prospective cohort pilot study at Maine Medical Center in Portland, Maine. Eligible participants were adults admitted to the cardiac or cardiothoracic ICU for cardiogenic shock with or without cardiac arrest who underwent tMCS with Impella or ECMO. Daily HI and pfHb levels were obtained for each participant. Analysis was conducted using Spearman’s correlation coefficients to establish a relationship between pfHb and HI among tMCS patients. Results: A total of 116 blood samples were collected from 40 patients, with the majority receiving mechanical support with Impella (73%) versus ECMO (27%). Acute coronary syndrome (42%) was the most common cause for tMCS, followed by decompensated heart failure (25%). The average length of tMCS support was 7 days with a 30-day mortality rate of 40%. Analysis revealed a Spearman’s correlation coefficient (ρ) of 0.54 between pfHb and HI (p<0.001) and 0.58 between HI and LDH (p<0.001). The relationship was similar between patients who received Impella only support and those who received ECPELLA (ECMO plus Impella). Conclusion: Despite prior literature showing a very strong correlation between HI and pfHb, our prospective single-center pilot study did not demonstrate a strong enough relationship for HI to be used as a surrogate marker for pfHb in tMCS patients (ρ < 0.9).