Quantitative detection of circulating endothelial cells in vasculitis: comparison of flow cytometry and immunomagnetic bead extraction

Clarke, Lindsey, Shah, Vanita, Arrigoni, Francesca, Eleftheriou, D., Hong, Ying, Halcox, Julian, Klein, Nigel and Brogan, Paul (2008) Quantitative detection of circulating endothelial cells in vasculitis: comparison of flow cytometry and immunomagnetic bead extraction. Journal Of Thrombosis And Haemostasis, 6(6), pp. 1025-1032. ISSN (print) 1538-7933

Abstract

BACKGROUND: Circulating endothelial cells (CECs) are biomarkers for endothelial cell (EC) injury and are quantified using immunomagnetic bead extraction (IBE), or flow cytometry (FC). Reports suggest that there is good agreement between these methods for CEC quantification. OBJECTIVES: We examined levels of agreement between these techniques in children with systemic vasculitis. METHODS: We added HUVEC or human pulmonary artery EC to whole blood to optimize FC gating strategies for EC. EC-optimized FC was then compared with IBE for CEC enumeration in 25 children with vasculitis and 20 healthy controls. RESULTS: Using Bland-Altman analysis, agreement between IBE and EC-optimized FC was poor in children with vasculitis (n = 25) and healthy controls (n = 20): IBE consistently detected higher values than the EC-optimized FC method: the mean difference between the two techniques was 60 CECs mL(-1), 95% CI +/-374 CECs mL(-1) (paired analyses of 45 individuals). Agreement was poorest for vasculitis patients: mean difference (IBE - EC-optimized FC) 120 CECs mL(-1), 95% CI +/-460 CECs mL(-1) (P = 0.018). We identified three reasons for this discrepancy: (i) sub-optimal FC gating parameters previously used for detecting CECs; (ii) inherent lack of sensitivity of FC compared with IBE for CEC rare event detection; and (iii) use of lysis buffers required for FC causing CEC lysis. CONCLUSIONS: There was poor agreement between EC-optimized FC and IBE for the quantification of CECs from children with active vasculitis and controls. We emphasize that in this clinical setting the two techniques are not directly comparable when comparing results obtained using these different methodologies.

Actions (Repository Editors)

Item Control Page Item Control Page