Normal Platelet Function, but Increased Platelet Activation In Vivo in Diabetic Patients

Increased platelet function and thromboxane (Tx) A2 generation in diabetes mellitus have been suggested by some investigators (1), but not by others (2, 3). Such discrepancies may be due to limitations of the test systems used to investigate platelet function ex vivo. More recently urinary excretion of products of arachidonic acid metabolism was measured in' diabetic patients as an index of platelet activation in vivo. In patients with diabetes without evidence of macrovascular complications, platelet function ex vivo and urinary excretion of 2,3 dinor-TxB2 did not differ from non-diabetic controls (4), while urinary excretion of 11dehydro-TxB2 was significantly higher in diabetic patients with clinical evidence of macrovascular disease than in controls (5). Increased excretion of TxA2 metabolites in urine from diabetic patients with severe atherosclerosis involving large blood vessels had also been reported (6). On the basis of these findings, we measured both platelet response to arachidonic acid and capacity to synthesize TxA2 ex vivo and TxA2 biosynthesis in vivo in seven insulin-treated patients (3 non-insulin dependent and 4 insulin-dependent) and in 5 healthy volunteers. Mean duration of diabetes was 19.8 years (range 10-33) and mean HbA1 value was 10.7 ± 1.0%. All patients had retinopathy (3 proliferative, 4 non proliferative), but no signs of clinical nephropathy. Two of them had clinical signs of macrovascular complications. Serum TxB2 measurements and aggregation studies were performed as described (7). A 6-h urine collection was performed in the morning for the analysis of 11-dehydro-TxB2 excretion, according to a method previously validated (8). Platelet capacity to generate TxA2 during spontaneous blood clotting in vitro as well as platelet aggregation response to sodium arachidonate were similar in diabetic patients and healthy controls. In contrast, the urinary excretion of 11-dehydro-TxB2 was significantly higher in patients than in controls (Table 1). Administration of indobufen (200 mg x 2), a cyclooxygenase inhibitor, to patients resulted in an almost complete suppression of serum TxB2 generation (>95% inhibition) and of platelet aggregation and in a very marked reduction of 11-dehydro-TxB2 urinary excretion (19.6 ± 12.4 vs 113.9 ± 31.2 ng/h). The discrepancy between in vitro and in vivo platelet function in diabetes could be the consequence of platelets essentially normal in a test tube being exposed in vivo to atherosclerotic or otherwise damaged vascular walls. The abnormal platelet-vessel