Derivation and validation of an equation for adjustment of neuron-specific enolase concentrations in hemolyzed serum

Abstract

Objective: To derive and validate a formula to allow for adjustment of serum neuron-specific enolase (NSE) concentrations based on the amount of hemolysis in the sample. To compare the accuracy of qualitative and quantitative assessment of hemolysis.

Design: Derivation and validation of a formula.

Subjects: Eighty-three children <11 years of age were used for the derivation. Twenty-three children <11 years without brain injury and 30 with brain injury were used for the validation.

Interventions: We evaluated the amount of hemolysis quantitatively using a hemocue and qualitatively by visualization. We measured NSE concentrations by enzyme-linked immunosorbent assay. Using a subset of subjects, we used linear regression analysis to derive an equation to adjust serum NSE concentrations based on the degree of hemolysis. We prospectively validated the formula in a different group. NSE < or =11.5 ng/mL was considered normal.

Results: The formula derived was as follow: Adjusted NSE = Unadjusted NSE - (1.659) (square root of the hemocue). In the validation, 96% of children without brain injury had a normal adjusted NSE; 93% of children with brain injury had an abnormal adjusted NSE. Qualitative analysis of hemolysis was inconsistent and both overassessed and under-assessed hemolysis as quantified by hemocue.

Conclusions: We retrospectively derived and prospectively validated an equation for adjusting serum NSE concentrations based on the amount of hemolysis in the sample. Use of this formula will allow for accurate measurement of NSE even in hemolyzed sample and may improve its usefulness as a marker of brain injury in children. Qualitative assessment of the degree of hemolysis is not accurate and should not be used.