LAUSR

As suggested by Moris and coworkers,1 although there is accumulating evidence that platelets stimulate liver regeneration in rodent models and in humans, there is a paucity of highquality data to indicate by which mechanism(s) platelet exert their stimulating effect. Several mechanisms have been postulated including direct stimulatory effects of plateletderived molecules on hepatocyte proliferation, functional transfer of platelet RNA to the hepatocyte, and plateletmediated stimulation of influx of inflammatory cells, as we have reviewed recently.2 In addition, it has been suggested that plateletendothelial cell interactions drive liver regeneration.3 Unfortunately, unequivocal evidence to support or refute any of these mechanisms in plateletmediated liver regeneration in rodents and humans is lacking.2,4 Moris and coworkers also point to data suggesting a role of platelets in improving liver fibrosis. However, platelets have been shown to have both stimulating and inhibitory roles in fibrosis, depending on the context and experimental models used.5 In addition, we studied liver regeneration in mice with otherwise healthy livers. We (and others) have not assessed proliferation of nonparenchymal cells in experimental models in which platelets were depleted, but this would be of definite interest. Although I agree that liver function tests would nicely complement studies on plateletmediated liver regeneration, I do note that the combination of liverbody weight ratio with ki67 immunostainings is compatible with functional liver regeneration in otherwise healthy livers. To my knowledge, it is unknown whether platelet deposition in the liver remnant decreases portal flow by increasing vascular resistance or whether the number of platelets deposited within the liver remnant is too limited to cause a decrease in flow. The study by Starlinger6 does not address this specific question. We showed that deficiency of VWF not only virtually abolishes platelet influx, but also substantially delays liver regeneration, suggesting that VWF is a key player in plateletmediated stimulation of liver regeneration. This finding might be clinically relevant as VWF release can be stimulated by 1desamino8Darginine vasopressin (DDAVP). I agree it would be of interest to assess which platelet receptors (e.g. the glycoprotein Ib/IX/V complex or αIIbβ3) facilitate VWFmediated platelet influx in the liver remnant, but this may be less relevant in a therapeutic context. In aggregate, I concur with Moris and coworkers that crucial details on the mechanisms by which platelets stimulate liver regeneration are as yet unknown. Future work should combine development of strategies to stimulate plateletmediated liver regeneration in humans with in depth mechanistic studies.