Endothelial dysfunction in uremic patients on continuous ambulatory peritoneal dialysis (CAPD)

Endothelial dysfunction is associated with diabetic microand macroangiopathy as well as with the decline in creatinine clearance. It has been suggested that endothelial dysfunction presents in patients (pts) on continuous ambulatory peritoneal dialysis (CAPD). Th e objective of this study was to examine the plasma biomarkers of endothelial dysfunction and their association with IMT of carotid arteries in diabetic and nondiabetic patients on CAPD. Th is study included  CAPD pts ( with type II diabetes and  non-diabetic pts) mean age . years ± .. Plasma von Willebrand factor (vWF) activity, serum albumin, glucose, total cholesterol, triglycerides and lipoprotein (a) levels, as well as serum level of homocysteine, parathyroid hormone (PTH) in plasma and microalbuminuria was determined. Ultrasound examination of carotid arteries was performed in all patients by measured bilateral intima-media thickness of carotid artery (CIMT). Mean IMT value was signifi cantly higher in type  DM patients (. ± . mm) compared to non-diabetic patients (. ± . mm) on peritoneal dialysis (p<.). Th ere was also a signifi cant diff erence in lipids /triglycerides and Lp (a)/, procoagulation (fi brinogen, von Wilebrand factor, factor VIII) and infl ammatory markers (CRP) level between type  DM and non-diabetic CAPD patients. A stepwise multiple regression analysis revealed that log triglycerides and factor VIII were independent factors for the IMT. Th e results of this research impose that diabetic type  CAPD patients have developed systemic alteration of endothelial function and higher risk of cardiovascular complications compared to non-diabetic CAPD patients. ©  Association of Basic Medical Sciences of FBIH. All rights reserved


INTRODUCTION
Due to the significant increase in cardiovascular morbidity and mortality in patients with terminal renal failure, yet more than two decades ago, it was hypothesized about "accelerated atherosclerosis" in such patient population.Noninvasive method of ultrasound examination has found wide application in the general population to identify high-risk patients for cardiovascular incidents and asymptomatic atherosclerosis [], and ultrasound of the carotid artery has served as a useful predictor of systemic atherosclerosis [].In recent years, many studies have confi rmed the high prevalence of atherosclerotic lesions in blood vessels of uremic patients [, ] with the impossibility of a full explanation of their occurrence by the traditional risk factors, often present in these patients.Factors of atherosclerotic disease in chronic renal disease, primarily in its terminal stage, significantly coexist with other cardiovascular risk factors in the development and progression of vascular complications.Chronic renal disease with creatinine clearance below  ml/min for a period longer than three months, regardless of the cause by which it is induced, has been recognized as an important factor that affects the integrity of the endothelium and the occurrence of endothelial dysfunction.Impaired complex endothelial function is manifested by changes in concentration and activity of some vasoactive substances, such as von Willebrand factor (vWF), factor VIII, fibrinogen, homocysteine.Changes in intima-media thickness (IMT) of carotid arteries are a reflection of the morphological changes in the endothelium of blood vessels.Recent studies suggest that measuring the concentration of specifi c vasoactive substances, or biomarkers in plasma, such as von Willebrand factor, factor VIII, fibrinogen, homocysteine, along with traditional and uremia specifi c risk factors, can predict the existence of endothelial dysfunction and development of micro and macrovascular changes [, ].Vascular endothelial dysfunction is a signifi cant factor in the pathogenesis of diabetic micro-and macroangiopathy.On the other hand, the decline in creatinine clearance is associated with endothelial dysfunction and development of coronary and generalized atherosclerosis.In the study, Yilmaz et al. [], which included  patients at diff erent stages of chronic kidney disease, endothelial dysfunction was confi rmed in all stages of chronic kidney disease, with a strict correlation with the degree of reduction in glomerular fi ltration rate.It remains unclear how present is endothelial dysfunction in uremic patients on continuous ambulatory peritoneal dialysis (CAPD) and whether the prevalence of endothelial dysfunction in this category of patients depends on the presence of diabetes.Th e aim of this study was to examine the concentration of fi brinogen, von Willebrand factor, factor VIII and homocysteine in plasma as biomarkers of endothelial dysfunction and their association with IMT of carotid arteries in diabetic and nondiabetic patients on CAPD, as well as the correlation of IMT with uremia associated and other traditional risk factors inherent in the presence of vascular endothelial dysfunction, which, with its endocrine and paracrine function, represent an important organ in maintaining cardiovascular homeostasis.

Patients
Th is study included  stable uremic patients on CAPD,  with type II diabetes and  non-diabetic patients ( M,  F, mean age . years ± .) from Outpatient Department of Clinic for Nephrology, CCU Sarajevo.In the study have not been included patients who were on CAPD shorter than  months, patients with congestive heart disease and patients with manifest peripheral blood vessels disease of inflammatory or idiopathic nature, as well as those who have not signed informed consent.The study was performed in period from  st of January to the  th of June .Th e average duration of dialysis treatment in these patients was . ± . months of dialysis.Patients were evaluated by standard physical examination, laboratory tests and ultrasound examination of carotid arteries.All patients included in the study signed upon informed consent with careful explanation of the study procedures.Approval for the study was obtained by the local Ethics Committee (Ethical committee approval N° -; ..).All procedures on human subjects were performed in accordance with the latest version of Helsinki Declaration.

Design of study and measurements
Th e study was designed as a cross-sectional comparison between CAPD patients with type II DM and nondiabetic CAPD patients.Blood pressure was measured and venous blood samples were taken in the morning on an empty stomach.Laboratory procedures vWF activity was assessed by standard ristocetin-dependent platelet aggregation procedure, using reagents such as BC von Willebrand reagent (Dade Behring, USA).Serum albumin, glucose, total cholesterol, triglycerides and lipoprotein (a) /Lp(a)/ levels were measured by quantitative kinetic spectrophotometric method using a computerized autoana-lyzer AXION (Menheim Boehringer, Germany) and reagents from Boehring diagnostics.Level of homocysteine in serum was determined by enzyme immunoassay technique on the appliance Abbott AxSYM, Germany.Th e level of parathyroid hormone (PTH) in plasma was measured by RIA method.Microalbuminuria was determined by nephelometry.

Ultrasound of carotid artery
In all CAPD patients high-resolution B-mode ultrasound examination of common carotid artery an ATL  ultrasound machine equipped with - MHz transducer was applied.Intima-media thickness of carotid artery (CIMT) was measured in all patients  cm above the carotid bifurcation.IMT diameters were calculated as the average by bilateral IMT measured diameter.Carotid atherosclerosis was defined if a CIMT is greater than . mm and / or if there is presence of carotid plaque that protrudes more than . mm in the vascular lumen [].

Statistical analysis
Each value was expressed as the mean ±SE.Diff erences between two groups were analyzed by the Mann-Whitney test.Univariate correlation coeffi cients were determined by Pearson analysis.A stepwise multiple regression analysis was applied to examine the relationship between IMT and a set of clinical parameters including age, glucose, duration of PD treatment, blood pressure (MBP), hemoglobin, cholesterol, log-triglycerides, log-Lp(a), Von wilebrand factor, factor VIII, log-fibrinogen, homocysteine, albumin, microalbuminuria, intact parathyroid hormone (PTH), calcium, phosphate, calcium phosphate, CRP.Since triglycerides, fi brinogen and Lp(a) were non-normally distributed in this study, we normalized these data by log transformation before entering a stepwise multiple regression.The significant independent variables were ordered according to their standardized eff ect, defi ned as regression coeffi cient/standard error of the regression (β).P values less than . were considered statistically signifi cant.All statistical calculations were performed with the SPSS  software (version ., SPSS Inc, Chicago, Illinois, USA).

RESULTS
Baseline characteristics of the patients included in the study are shown in Table .Mean values of intima media thickness, lipids, procoagulant and inflammatory markers in serum of patients on peritoneal dialysis are shown in Table .Type  DM was observed in ., while hypertension was observed in . of patients on PD.Mean IMT value was significantly higher in type  DM patients (.±.mm) compared to non-diabetic patients (. ± . mm) on peritoneal dialysis (p<.)(Figure ).

DISCUSSION
Epidemic increase in the incidence of chronic kidney disease in the world is the result of the global pandemic of type  diabetes, associated with a high prevalence of metabolic syndrome, and diabetic kidney disease.In diabetic patients endothelial dysfunction is the earliest event that precedes the alteration of endothelial integrity and the appearance of micro-and macrovascular complications [, ].
Previous studies have shown that the arterial vascular disease and hypertrophy of the left ventricle are the main causes of high cardiovascular mortality in dialysis patients [].Atherosclerosis is the most common pathological process leading to cardiovascular disease.It is a disease of intima of blood vessels, for which has been proven that it starts with endothelial dysfunction, progression of plaque formation and accumulation of macrophages, the formation of fi brous caps and the presence of necrotic core.Recent studies indicate a link between endothelial dysfunction and hypertension, on the one hand [, ] and loss of renal function, on the other hand [].
In patients on peritoneal dialysis arterial diseases are characterized by a high level of calcifi cation in the arterial wall, in response to hemodynamic humoral abnormality in chronic uremia, which leads to the increased stiff ness of large arteries like the aorta and common carotid arteries (CCA), especially loosing elasticity of arteries [].Results of our study have shown that mean IMT value was signifi cantly higher in type  DM patients compared to non-diabetic patients on peritoneal dialysis, with signifi cantly higher levels of microalbuminuria, calcium, Ca x P product, as well as a signifi cant diff erence in lipids / triglyceride, Lp (a)/, procoagulation markers (von Willebrand factor, factor VIII) and infl ammatory markers (CRP) in type  DM patients compared to non-diabetic patients on peritoneal dialysis, which supports the concept of malnutrition-infl ammation-atherosclerosis syndrome [].
The development of endothelial dysfunction in type  diabetes mellitus involved hyperglycemia, insulin resistance, impaired lipid metabolism and metabolism of lipopro-teins, and oxidative stress [].Hyperglycaemic recurrent episodes, with the accumulation of advanced glycation end products (AGEs) and the induction of plasminogen activator inhibitor type-, lead to a chronic endothelial dysfunction [].Elevated glucose level stimulates the synthesis of growth factors and vasoactive substances, contributing to increased permeability of endothelial cells and indirect disruption of their functions.It has been shown that endothelial function in diabetics is being disturbed by components of the metabolic syndrome [], particularly pronounced in patients on CAPD.Th erefore, the search for biomarkers of endothelial dysfunction is an important step in improving the diagnostic and therapeutic approach in these patients.
Numerous studies indicate that the von Willebrand factor is a biomarker of endothelial dysfunction.Von Willebrand factor is primarily synthesized in endothelial cells and released into circulation when endothelial integrity is compromised by harmful same agents.Th is factor, inducted from endothelial cells, rapidly binds to collagen fi bres at the site of vascular damage.Action of shear stress is a subject of conformational change, by which it becomes a substrate for adhesion and aggregation of platelets.Bongers et al. [], were examining a number of biological markers, and have found that vWF is also an indicator of risk for coronary heart disease and stroke.Chronic renal failure is recognized as a clinical situation which imposes the greatest risk for endothelial integrity.Systemic endothelial dysfunction appears in the early stages of chronic kidney disease, increasing the risk of cardiovascular morbidity and mortality in patients with chronic kidney disease [].In patients with chronic renal insuffi ciency elevated levels of vWF were found [].Our results indicate that the von Willebrand factor is increased in CAPD patients with type  DM, as well as intima-media thickness.Triglycerides and factor VIII were independent predictors of carotid intima-media thickness, presenting a strong surrogate of endothelial dysfunction and atherosclerosis in these patients.Dyslipidemia with hyperglycaemia contributes to endothelial dysfunction, the earliest stage of atherosclerosis, as well as the infl ammatory process in the atherosclerosis, by presenting higher concentration of fi brinogen and CRP value in CAPD patients with type  DM in our study.We found a statistically signifi cant correlation between microalbuminuria and intima-media thickness which indicates the close relationship between microalbuminuria and endothelial dysfunction.Previous studies suggest that endothelial dysfunction may be a link between microalbuminuria and cardiovascular disease [].Many peritoneal dialysis patients already have significant vascular lesions before initiating dialysis in response to humoral abnormality in chronic ureamia.This study has confirmed that carotid ultrasonography is one of the useful non-invasive imaging techniques for the assessment of intima-media thickness, which is in strong relationship with values of biomarkers of endothelial dysfunction.Endothelial dysfunction in CAPD patients is especially expressed in patients with type  diabetes.Existent endothelial dysfunction in diabetic kidney disease may have additive eff ects on cardiovascular risk and therefore the endothelial function tests must be accepted as a valid parameter of cardiovascular pathology assessment in these patients.

CONCLUSION
In conclusion, the results of this research impose that factor VIII and triglycerides are independent predictors of intima-media thickness.IMT is signifi cantly higher in DM type  patients on CAPD in comparison to non-diabetic patients on CAPD.Significantly higher values of von Willebrand factor and triglycerides in CAPD patients with DM type  compared to the same in non-diabetic patients impose that diabetic patients have developed systemic alteration of endothelial function and higher risk of cardiovascular complications compared to non-diabetic CAPD patients.

DECLARATION OF INTEREST
Th e authors declare no confl ict of interest for this study.

FIGURE 1 .
FIGURE 1. Intima media thickness values in patients on peritoneal dialysis (PD) with and without type 2 diabetes mellitus (DM 2).The bars represent mean and error bars represent SE.

FIGURE 2 .
FIGURE 2. Relationship between intima media thickness and logarithmically transformed triglycerides (log-triglycerides) (A), factor VIII (B) and Von Wilebrand factor (C) in patients on peritoneal dialysis

TABLE 1 .
Baseline characteristics of the patients on peritoneal dialysis.

TABLE 2 .
Mean values of intima media thickness, lipids, procoagulant and infl ammatory markers in serum of patients on peritoneal dialysis.

TABLE 4 .
Stepwise multiple regression analysis of the carotid intima-media thickness for the determinants of independent factors Dependant variable-IMT; R 2 -0.396