[Hemolysis index: qualification and evaluation of usefulness in patients on ECMO]
Julie Fettig1, Véronique Wurtz2, André Gillibert3
1Biochimie Générale, CHU de Rouen, 1 rue de Germont, 76000 Rouen, France.
To assess the validity of the hemolysis index (HI) as a method for plasma hemoglobin measurement and its relevance in the clinical-biological monitoring of patients on extracorporeal membrane oxygenation (ECMO). HI results from a Roche Cobas® analyzer (HIc) were compared with those obtained using the reference Drabkin and Sysmex methods. A full method validation was performed in accordance with SH GTA 04 (scope B). A retrospective study was then conducted in patients on ECMO to evaluate whether the HI could predict hemolysis events within 24 hours of ECMO initiation, according to ECMO type, and its potential association with mortality. HIc values were strongly correlated with reference measurements (p < 0.0001), and validation criteria were fully met. The study included 106 patients on ECMO (mortality rate: 55.7%). Survival was higher in patients with HIc < 10 mg/dL compared to those with HIc > 50 mg/dL (severe hemolysis). Early hemolysis was not associated with increased mortality. Severe hemolysis occurred significantly more frequently with veno-arterial ECMO circuits than with veno-venous circuits (p = 0.003). HIc offers analytical advantages (speed, automation, low cost) and appears to be a reliable surrogate for plasma hemoglobin measurement. IHc could be used to predict hemolysis events in patients on ECMO. Further studies, in particular kinetic studies, are needed to confirm the relevance of HI in routine hemolysis monitoring and to help differentiate in-vivo hemolysis from pre-analytical artifacts.
To assess the validity of the hemolysis index (HI) as a method for plasma hemoglobin measurement and its relevance in the clinical-biological monitoring of patients on extracorporeal membrane oxygenation (ECMO). HI results from a Roche Cobas® analyzer (HIc) were compared with those obtained using the reference Drabkin and Sysmex methods. A full method validation was performed in accordance with SH GTA 04 (scope B). A retrospective study was then conducted in patients on ECMO to evaluate whether the HI could predict hemolysis events within 24 hours of ECMO initiation, according to ECMO type, and its potential association with mortality. HIc values were strongly correlated with reference measurements (p < 0.0001), and validation criteria were fully met. The study included 106 patients on ECMO (mortality rate: 55.7%). Survival was higher in patients with HIc < 10 mg/dL compared to those with HIc > 50 mg/dL (severe hemolysis). Early hemolysis was not associated with increased mortality. Severe hemolysis occurred significantly more frequently with veno-arterial ECMO circuits than with veno-venous circuits (p = 0.003). HIc offers analytical advantages (speed, automation, low cost) and appears to be a reliable surrogate for plasma hemoglobin measurement. IHc could be used to predict hemolysis events in patients on ECMO. Further studies, in particular kinetic studies, are needed to confirm the relevance of HI in routine hemolysis monitoring and to help differentiate in-vivo hemolysis from pre-analytical artifacts.
