► Click here to download the original poster (PDF)
P. SUCHON1,2, E. MACCARY3 and P.E. MORANGE1,2
1Aix-Marseille Univ, INSERM, INRAE, C2VN, Marseille, France; 2Laboratory of Haematology, La Timone Hospital, Marseille, France; 3HORIBA Medical International, Montpellier, France
Diagnosis of antithrombin (AT), protein C (PC) and protein S (PS) deficiencies is based on plasmatic assays. Functional assays are usually performed in first-line as they can detect both qualitative and quantitative deficiencies. The Yumizen G1550 is a recently launched fully automated hemostasis analyser. HORIBA has developed reagents for the thrombophilia screening.
The aim of the study was to compare the performance of the couple HORIBA/Yumizen G1550 with the local configuration (Stagoor Hyphen BioMed/STA-R Max) for the measurement of antithrombin, protein C and protein S.
All 3 HORIBA reagents were functional assays measuring AT, PC and PS activities.
The local instrument was the StagoSTA-R Max. The following reagents were used for measuring AT and PC activities: STA StachromAT III (Stago), HEMOCLOT PC (Hyphen BioMed).
For all 3 parameters, repeatability was assessed using 2 levels of controls (30 replicates). Reproducibility was estimated after 25 runs of 2 levels of controls. Values of repeatability and reproducibility were previously calculated for the local configuration (repeatability: 10 replicates of plasma sample; reproducibility: 30 runs of control samples).
Passing Bablokand Bland Altman methods were used for the comparison. Of note, the local configuration did not allow a comparison for PS measurement (antigenic free PS measured with STA LiatestFree PS –Stago). Agreement between the configurations was assessed using a 80% cut-off for AT deficiency and 70% for PC deficiency.
Repeatability and reproducibility
As compared with local configuration the HORIBA/Yumizen configuration displayed good performances for AT and PC activities. Of note, PC activity CVs were lower with the Yumizen configuration for both repeatability and reproducibility. As expected, CVs were higher for PS activity (HORIBA) than antigenic free PS (Stago).
22 samples were tested for the comparison of AT (5 deficiencies according to the Stago configuration): bias = 11%; r = 0.96. 21 samples/22 were correctly classified by the HORIBA configuration (kappa = 0.87). Comparison of PC configurations was performed using 67 samples, including 16 deficiencies according to the Stago configuration: bias = 8%; r = 0.72. Conflicting classifications were observed for 10 samples (kappa = 0.63).
Correlation
Horiba configuration displayed excellent performances for AT deficiency screening. Only 1 sample was misclassified: 1 AT deficiency by the Horiba configuration in a patient treated by unfractionated heparin. Horiba configuration showed good performances for classifying PC samples. Although, the Horiba configuration overestimated PC deficiencies (90% of conflicting samples).
Evaluation of PS assay needs to be performed using chronometric local assay.
Horiba reagents showed good repeatability and reproducibility performances as compared with the local configuration.
Comparison between the 2 configurations was excellent for AT deficiency screening. Only 22 samples have been tested, including 6 deficiencies. More samples should be tested for fully validating the performances.
The Horiba configuration seems to overestimate PC deficiencies. Although 14/15 (93%) PC deficiencies were correctly classified.
Do you have any questions or requests? Use this form to contact our specialists.
