The study was in rats but it has been replicated in humans as well, even though the replication study in humans used much higher doses of aspirin (human dose 90mg/kg). This study used smaller doses of aspirin equivalent to 650mg taken every 6 hours. If this works, it could be a much safer dosing regimen especially for people prone to bleeding.
http://www.jci.org/articles/view/11559
"...Insulin-stimulated whole-body glucose uptake was decreased by 37% with the lipid infusion (Figure 1c). Salicylate pretreatment prevented decreases in insulin-stimulated whole-body glucose uptake caused by lipid infusion. Decreases in insulin-stimulated whole-body glucose uptake were due to a 41% decrease in insulin-stimulated glucose uptake in skeletal muscle (soleus) with lipid infusion (Figure 1d). Salicylate pretreatment also prevented the lipid-induced decrease in insulin-stimulated skeletal muscle glucose uptake. Furthermore, insulin-stimulated whole-body glycolysis and glycogen/lipid synthesis were significantly decreased by 27% and 43%, respectively, with the lipid infusion (Figure 2, a and b). Salicylate pretreatment prevented these decreases in insulin-stimulated whole-body glucose metabolic flux caused by lipid infusion. Similar to the decreases in insulin-stimulated whole-body glucose metabolic flux with lipid infusion, insulin-stimulated skeletal muscle glycolysis was decreased by 36% with lipid infusion (Figure 2c), but the most profound change was a 66% decrease in insulin-stimulated skeletal muscle glycogen synthesis with lipid infusion (Figure 2d). Salicylate pretreatment also prevented the lipid-induced decreases in insulin-stimulated skeletal muscle glucose metabolism."
http://www.jci.org/articles/view/11559
"...Insulin-stimulated whole-body glucose uptake was decreased by 37% with the lipid infusion (Figure 1c). Salicylate pretreatment prevented decreases in insulin-stimulated whole-body glucose uptake caused by lipid infusion. Decreases in insulin-stimulated whole-body glucose uptake were due to a 41% decrease in insulin-stimulated glucose uptake in skeletal muscle (soleus) with lipid infusion (Figure 1d). Salicylate pretreatment also prevented the lipid-induced decrease in insulin-stimulated skeletal muscle glucose uptake. Furthermore, insulin-stimulated whole-body glycolysis and glycogen/lipid synthesis were significantly decreased by 27% and 43%, respectively, with the lipid infusion (Figure 2, a and b). Salicylate pretreatment prevented these decreases in insulin-stimulated whole-body glucose metabolic flux caused by lipid infusion. Similar to the decreases in insulin-stimulated whole-body glucose metabolic flux with lipid infusion, insulin-stimulated skeletal muscle glycolysis was decreased by 36% with lipid infusion (Figure 2c), but the most profound change was a 66% decrease in insulin-stimulated skeletal muscle glycogen synthesis with lipid infusion (Figure 2d). Salicylate pretreatment also prevented the lipid-induced decreases in insulin-stimulated skeletal muscle glucose metabolism."