The present study was focused to discern the contribution of uremia and the RRT to the development of endothelial activation and damage. The ex vivo and in vitro approaches applied revealed that uremia per se causes a proinflammatory state on the endothelium, as derived from results in pre-dialysis patients. The hemodialysis technique with the current advances (generalized use of biocompatible membranes and ultrapure water, among others) did not exhibit a damaging effect on the endothelium additionally to that observed in patients with advanced CKD managed conservatively. Interestingly, peritoneal dialysis was the most proinflammatory condition, as demonstrated by a higher presence of soluble markers of endothelial activation and damage in plasma, and a more intense activation of both p38 MAPK and NFκB signaling pathways in cultured endothelial cells. These effects could be attributed, at least in part, to the glucose and its degradation products present in the PD dialysis fluids.
Endothelial activation and damage could be the earliest indicator of subclinical cardiovascular disease. In the laboratory, measurement of plasma levels of different molecules and/or other elements, either discharged or up-regulated in an activated endothelium, could be useful to assess endothelial damage and activation 
. Considering that to date there is no universal marker of endothelial damage, we have evaluated different indicators. Soluble adhesion receptors were evaluated by a multiplex system and showed significant activation of the endothelium in all uremic patients, being more notable in the PreD and PD groups. In relation to VWF plasma levels, our present data is consistent with previous observations in hemodialyzed patients 
being significantly higher in all the uremic patients than in controls. However, the short range in which values are included might explain the lack of differences found between the studied groups. Levels of circulating endothelial cells (CEC) have been described to correlate well with VWF plasma levels 
. There is previous evidence generated in HD patients of increased CEC and their association with future cardiovascular events 
. In our present study, blood CEC counts were higher in all the uremic patients when compared to the control group, especially in the group of peritoneal dialysis.
The analysis of plasma markers of endothelial damage indicates that the toxic environment of uremia causes inflammation on the endothelium, as observed in the non dialyzed CKD patients. This inflammatory effect is also seen in patients under both types of RRT. From the results obtained, it can be predicted that PD exhibits the most deleterious effect on the endothelium and, interestingly, current HD techniques have diminished the harmful effect previously ascribed to this treatment.
Previous studies by our group and others were mostly performed in hemodialyzed patients. Results have demonstrated the existence of endothelial activation and damage in CKD both in vivo
and in vitro
. Plasma levels of endothelium-derived proteins 
, some of them vasoactive factors, are increased in these patients 
. Endothelium-dependent vasodilation, the gold-standard method to assess endothelial function in vivo
, is also decreased in these patients 
. Exposure of cultured EC to sera from hemodialyzed patients accelerates cell cycle and proliferation, with a more prothrombotic extracellular matrix and a proinflammatory phenotype with a higher expression of cell adhesion molecules 
, which represents one of the earliest pathological changes in immune and inflammatory diseases such as atherosclerosis 
. Moreover, a proteomic characterization of these changes demonstrated a differential expression of proteins associated with inflammation and oxidative stress in cells exposed to the uremic condition, related to the NFκB signaling pathway 
. In fact, exposure of endothelial cells in culture to sera from hemodialyzed patients induces activation of p38 MAPK 
and of NFκB 
. In hemodialyzed patients there is presence of the uremic toxics but also of those components released by blood cells that become activated by the procedure itself. However, results from the present study indicate that there is not an additional deleterious effect of hemodialysis over that observed in the predialysis condition. Probably, the generalized use of more biocompatible dialysis membranes and ultrapure water, among other advances, has reduced blood cell stress with less contribution of proinflammatory cytokines. Interestingly, PreD patients showed a more significant inflammatory state than HD patients. These results are not fully in agreement with those by Merino et al. 
. In their study, PD seems to exhibit a lower damaging condition on the endothelium than HD and predialysis. Differences between both studies may be due to the different experimental approaches applied but especially to the fact that in our study patients with CKD did not have evidence of previous cardiovascular disease and other known cardiovascular risk factors.
The present study provides the first evidence that sera from PD patients have a greater activating effect on p38 MAPK and NFκB, two intracellular key markers of inflammation and cell damage. The p38 MAPK protein kinases affect a variety of intracellular responses, with well-recognized roles in inflammation, cell-cycle regulation, cell death, development, differentiation, senescence and tumorigenesis 
. NFκB seems to act by regulating the expression of several genes involved in tumorigenesis, including anti-apoptotic proteins, cyclooxygenase-2, matrix metalloproteinase-9, genes encoding adhesion molecules, chemokines, and inflammatory cytokines; and cell cycle regulatory genes 
. Therefore, both p38 MAPK and NFκB participate in the proinflammatory responses and exhibit a clear role in the development of inflammatory and immunological diseases.
According to the USRDS 2011 Annual Data Report, mortality rates of dialyzed patients have declined in the last years probably reflecting changes in catheter utilization, improved cardiovascular disease care, and changes in infectious complications. However, cardiovascular disease is still the major cause of the limited life expectancy in patients on substitutive therapies 
Patients under peritoneal dialysis are often hypertensive and sometimes volume overloaded 
, which is related to endothelial damage in PD patients 
. Glucose is absorbed to a large extent from the dialysate, and conventional PD results in an almost unique metabolic situation involving continuous 24-hour absorption of glucose. One common and important side effect of this treatment is weight gain and accumulation of body fat stores 
. These patients develop hyperlipidemia, with high levels of low-density lipoprotein and triglycerides. Advanced glycation end-products (AGEs), which are believed to promote atherosclerosis through interaction with endothelial receptors 
, are commonly accumulated in CKD patients due to decreased renal clearance, could be overproduced in PD patients due to the high exposure to glucose and glucose degradation products (GDP). Bioincompatibility of conventional glucose-based peritoneal dialysis fluids (PDF) has been partially attributed to the presence of GDP generated during heat sterilization of PDF 
, which are thought to contribute to cellular dysfunction and membrane damage during peritoneal dialysis 
. In accordance with these previous results, our present study shows that the glucose load and the presence of GDP could play a role in the development of endothelial damage among these patients. Icodextrin is a colloid osmotic agent, derived from maltodextrin, used as aqueous solution for peritoneal dialysis. The osmotic activity of icodextrin keeps the solution inside the peritoneum for 10 to 16 hours without being significantly metabolized. Due to its chemical characteristics, Icodextrin reduces the burden of glucose overexposure. From this perspective, the activating effect of Icodextrin on the transcription factor (and to a lesser extent on the inflammation-dependent protein p38 MAPK) is difficult to explain. It can however not be excluded that the pH of the Icodextrin solution could exert a damaging effect on the endothelium in our in vitro studies.
From our present results it can be concluded that there is endothelial activation and damage associated with CKD, as demonstrated by the increased presence of plasma markers and by the in vitro studies in cultured endothelial cells. The uremic state seems to be a major cause of endothelial damage, probably through the activation of transcription factors, such as NFκB, which are related to inflammation. However, while improved hemodialysis procedures do not seem to have an additional deleterious effect, our different experimental approaches applied indicate that peritoneal dialysis seems to exert a more intense proinflamatory action on the endothelium that could be due, at least in part, to the increased glucose load. Studies to elucidate the potential molecular mechanisms involved should be the aim of future investigation.