Background Myeloperoxidase has shown potential as a marker for prognosis of coronary artery disease, but presently little is known about preanalytical handling of samples for quantifying myeloperoxidase. The present study was conducted to evaluate the effect of collection tube and freeze-thaw cycles on myeloperoxidase concentrations. Methods Acute coronary syndrome patients ( n = 88) were enrolled after obtaining written informed consent from coronary care unit of a tertiary care hospital (January 2012-June 2014). About 5 mL venous blood was collected from patients and divided into serum, lithium heparin, ethylenediaminetetraacetic acid and sodium citrate tubes. Except serum, all tubes were kept on ice immediately after collection. Samples were centrifuged at -4℃, separated immediately after centrifugation and stored at -40℃ until analysis. Myeloperoxidase was quantified by in-house and commercial assays and re-quantified after five freeze-thaw cycles. Results Myeloperoxidase concentrations, (serum samples) determined by commercial and in-house assays correlated well (r = 0.946) ( P < 0.001) and were higher in serum samples. Within plasma, myeloperoxidase concentrations were slightly higher in ethylenediaminetetraacetic acid (307.7 ± 52.1) and lower in lithium heparin (290.3 ± 49.2) and sodium citrate (221.4 ± 40.3) but not statistically significant. Correlation between myeloperoxidase concentrations (in-house enzyme-linked immunosorbent assay) after first cycle and fifth freeze-thaw cycle dropped to r = 0.448 ( P < 0.001). Conclusion Myeloperoxidase concentrations are comparable in three types of plasma tubes when samples are placed on ice immediately, centrifuged at low temperatures and separated immediately after centrifugation. Multiple freeze-thaw cycles have an effect on myeloperoxidase and should be avoided for quantifying myeloperoxidase.