Comparison of nine platelet function tests used to determine responses to different aspirin dosages in people with...

The antiplatelet efficacy of aspirin (ASA) is reduced in type 2 diabetes (T2D). As the best ex vivo method of measuring ASA efficacy remains uncertain, we compared nine platelet function tests to assess responsiveness to three ASA dosing regimens in 24 T2D patients randomized in a three-treatment crossover design to ASA 100 mg/day, 200 mg/day, or 100 mg twice daily for 2-week treatment periods. Platelet function tests compared were as follows: light transmission aggregometry (LTA)-0.5 mg/mL of arachidonic acid (AA) and 10 µM adenosine diphosphate (ADP); multiplate whole blood aggregometry (WBA)-0.5 mM AA and 6.5 µM ADP; platelet function analyzer (PFA)-100™-collagen and ADP (CADP) and collagen and epinephrine (CEPI); VerifyNow™-ASA; and urinary 11-dehydro-thromboxane B2 (TxB₂) and serum TxB₂. All cyclo-oxygenase (COX-1)-dependent tests and some COX-1-independent tests (PFA-CEPI, LTA-ADP) demonstrated significant reductions in platelet reactivity with all ASA doses. Two COX-1-independent tests (WBA-ADP and PFA-CADP) showed no overall reduction in platelet reactivity. Overall classifications for detecting all ASA doses, compared to baseline, were as follows: very good-LTA-AA (k = 0.95) and VerifyNow™-ASA (k = 0.85); good-serum TxB₂ (k = 0.79); moderate-LTA-ADP (k = 0.59), PFA-100™-CEPI (k = 0.56), urinary TxB₂ (k = 0.55), WBA-AA (k = 0.47); and poor-PFA-100™-CADP (k = -0.02) and WBA-ADP (k = -0.07). No significant kappa statistic differences were seen for each test for each ASA dose. Correlations for each test with serum TxB₂ measurements were as follows: very good-VerifyNow™-ASA (k = 0.81, R² = 0.56) and LTA-AA (k = 0.85, R² = 0.65); good-PFA-100^TM-CEPI (k = 0.62, R² = 0.30); moderate-urinary TxB₂ (k = 0.57, R² = 0.51) and LTA-ADP (k = 0.47, R² = 0.56); fair-WBA-AA (k = 0.31, R² = 0.31); and poor-PFA-100™-CADP (k = 0.04, R² = 0.003) and WBA-ADP (k = -0.04, R² = 0.0005). The platelet function tests we assessed were not equally effective in measuring the antiplatelet effect of ASA and correlated poorly amongst themselves, but COX-1-dependent tests performed better than non-COX-1-dependent tests.