Arterial and venous blood sampling is equally applicable for coagulation and fibrinolysis analyses.

OBJECTIVES: No consensus exists upon whether arterial and venous blood samples are equivalent when it comes to coagulation analyses. We therefore conducted a comparative cohort study to clarify if arteriovenous differences affect analyses of primary and secondary hemostasis as well as fibrinolysis. METHODS: Simultaneous paired blood samplings were obtained from a cannula in the radial artery and an antecubital venipuncture in 100 patients immediately before or one day after thoracic surgery. Analyses of platelet count and aggregation, International Normalized Ratio (INR), activated partial thromboplastin time (APTT), antithrombin, thrombin time, fibrinogen, D-dimer, rotational thromboelastometry (ROTEM), thrombin generation, prothrombin fragment 1 + 2, and an in-house dynamic fibrin clot formation and lysis assay were performed. RESULTS: No differences were found between arterial and venous samples for the far majority of parameters. The only differences were found in INR, median (IQR): venous, 1.1 (0.2) vs. arterial, 1.1 (0.2) (p<0.002) and in prothrombin fragment 1 + 2: venous, 289 (209) pmol/L vs. arterial, 279 (191) pmol/L (p<0.002). CONCLUSIONS: The sampling site does not affect the majority of coagulation analyses. Small differences were found for two parameters. Due to numerically very discrete differences, they are of no clinical relevance. In conclusion, the present data suggest that both samples obtained from arterial and venous blood may be applied for analyses of coagulation and fibrinolysis.

Short-term biological variation of serum glial fibrillary acidic protein.

OBJECTIVES: Serum glial fibrillary acidic protein (GFAP) is an emerging biomarker for intracerebral diseases and is approved for clinical use in traumatic brain injury. GFAP is also being investigated for several other applications, where the GFAP changes are not always outstanding. It is thus essential for the interpretation of GFAP to distinguish clinical relevant changes from natural occurring biological variation. This study aimed at estimating the biological variation of serum GFAP. METHODS: Apparently healthy subjects (n=33) had blood sampled for three consecutive days. On the second day, blood was also drawn every third hour from 9 AM to 9 PM. Serum GFAP was measured by Single Molecule Array (Simoa™). Components of biological variation were estimated in a linear mixed-effects model. RESULTS: The overall median GFAP value was 92.5 pg/mL (range 34.4-260.3 pg/mL). The overall within- (CVI) and between-subject variations (CVG) were 9.7 and 39.5%. The reference change value was 36.9% for an increase. No day-to-day variation was observed, however semidiurnal variation was observed with increasing GFAP values between 9 AM and 12 PM (p<0.00001) and decreasing from 12 to 9 PM (p<0.001). CONCLUSIONS: Serum GFAP exhibits a relatively low CVI but a considerable CVG and a marked semidiurnal variation. This implies caution on the timing of blood sampling and when interpreting GFAP in relation to reference intervals, especially in conditions where only small GFAP differences are observed.