Dialysate copeptin and peritoneal transport in incident peritoneal dialysis patients

Chronic kidney disease (CKD) affects millions of people worldwide and end-stage renal disease (ESRD) patients have a high risk of premature cardiovascular death. Apart from traditional risk factors such as high age, smoking, hypertension and/or diabetes, the presence of persistent low-grade systemic inflammation with increased concentration of pro-inflammatory cytokines has been acknowledged as an important underlying mechanism of cardiovascular disease in ESRD [1‐4]. The causes of inflammation include comorbidities other than ESRD, decreased renal function with retention of uremic toxins, imbalance between increased production and reduced clearance of pro-inflammatory cytokines, increased oxidative stress, and the dialysis procedure itself.

Peritoneal dialysis (PD), a method of renal replacement therapy in which the patient’s peritoneum is used as a membrane through which water and solutes are exchanged between blood and dialysate, may contribute to intraperitoneal inflammation due to the bioincompatibility of dialysis fluids, episodes of PD-related peritonitis, and catheter infections [5, 6]. Local peritoneal inflammation has been linked to higher peritoneal solute transport rate (PSTR) as assessed by dialysate/plasma (D/P) ratio of creatinine in peritoneal equilibration test (PET) [7]. There is an ongoing search for biomarkers associated with peritoneal transport rate; so far, the dialysate concentration of interleukin-6 (IL-6) has been found to be strongly related to D/P ratio of creatinine [2, 8].

Copeptin is a 39-amino-acid glycosylated peptide of molecular mass approximately 5 kDa, synthesized in the hypothalamus as a C-terminal part of pre-pro-hormone of arginine vasopressin (AVP) [9, 10]. It is difficult to measure the concentration of AVP in the blood due to its instability and short half-life. As copeptin and AVP are released in equimolar amounts and copeptin is stable for more than 24 h after blood withdrawal, copeptin is increasingly used as an easily measured surrogate marker of vasopressin [11]. However, the biologic role of copeptin in circulation is still unknown. High circulating concentrations of copeptin have been linked to a decline in glomerular filtration rate (GFR) and greater risk of new-onset CKD [12]. Moreover, copeptin has been reported to be associated with increased risk of cardiovascular complications such as hypertension, myocardial infarction, and heart failure but also with inflammatory response in sepsis [13, 14]. However, it is not known to what extent copeptin and AVP may influence peritoneal transport.

The aim of this study was to assess if copeptin concentrations in plasma and dialysate are related to peritoneal dialysis transport parameters and residual renal function (RRF) in incident PD patients.

Mean plasma and dialysate copeptin concentrations are shown in Table 1. There was a strong positive correlation between the plasma and dialysate concentrations of copeptin (Rs = 0.52, p = 0.001), see Fig. 1.

Our study is the first to report that dialysate copeptin, a surrogate marker for AVP, is associated with peritoneal transport assessed by PET in PD patients who were investigated 4–6 weeks following initiation of PD therapy. A main target for active AVP is in the kidneys, where it increases aquaporin-mediated water re-absorption from the filtrate, increases permeability to urea by influencing urea transporters, and increases sodium absorption. So far, there is no known biological role for copeptin, and the role of AVP, if any, in peritoneal transport is also not known.

We report that dialysate copeptin was inversely correlated with D/P creatinine, a marker of the transport properties of the capillary wall reflecting the combined impact of vascular surface area (determined by the number of perfused capillaries) and permeability (determined by the number of inter-endothelial small pores). The inverse correlation between dialysate copeptin and D/P creatinine suggest that AVP/copeptin could act as a vasoconstrictive and permeability active agent, or in other ways could influence the transport properties of the peritoneal vasculature. As plasma copeptin did not correlate with D/P creatinine, this appears to be mainly a local effect.

Whereas, so far, studies on copeptin in PD patients are lacking, our results agree with previous observations demonstrating elevated circulating concentrations of copeptin in patients with CKD. Thus, the mean concentration of plasma copeptin in our study, 80.7 ± 37.3 pg/ml (corresponding to 20.07 ± 9.28 pmol/L), was significantly higher than the mean concentration of 4.2 pmol/L in 359 healthy individuals reported by Morgenthaler et al. [10]. While Bhandari et al. in their study found significantly higher plasma copeptin concentration in healthy volunteer males compared to females (4.3 vs. 3.2 pmol/L) [15], we did not observe a difference between the sexes.

The previous studies show that the higher the copeptin concentration, the higher the risk of decline of renal function and progression of CKD [16, 17]. Thus, Zittema et al. in their study in IgA nephropathy patients found that a higher level of copeptin was associated with the severity and progression of the disease [18]. Similar results were obtained in studies in patients with autosomal dominant polycystic kidney disease (ADPKD) [19, 20].

Plasma copeptin levels are increased also in hemodialyzed patients and in kidney transplant recipients [21, 22]. In hemodialyzed subjects, the concentration of copeptin positively correlates with pre-dialysis body fluid volume [22]. In the study by Artunc et al., plasma copeptin showed a positive correlation with time on dialysis and a negative correlation with residual diuresis [23]. This is in consistence with our study which showed an inverse correlation between plasma copeptin and residual diuresis. Moreover, Ettema et al. showed that in hemodialyzed patients, plasma copeptin levels rise during hemodialysis and that clearance of vasopressin by HD was higher than that of copeptin [24, 25]. On the other hand, Rasche et al. showed that plasma copeptin levels, which were approximately 11-fold higher in hemodialyzed patients compared to normal population, decrease during the hemodialysis (HD) session. The relative reduction of copeptin during HD correlated inversely with the Kt/V ratio, indicating that copeptin might be eliminated by HD [26].

In our study, plasma copeptin correlated with renal Kt/V but not with peritoneal Kt/V, suggesting that kidneys play a major role in the elimination of copeptin. Similar to Rasche et al. [26], we did not find a significant association of copeptin with age or gender.

Our results confirmed the important role of dialysate interleukin-6 concentration, reflecting the intensity of intraperitoneal inflammation, in predicting PSTR, particularly in the early phase of PD treatment [8, 27]. This is in consistence with the results of our study which showed that intraperitoneal IL-6 was associated with PSTR both in univariate and multivariate analyses. It has been found that IL-6 is present in the drained dialysate in higher concentrations that can be explained by diffusion from plasma, implying its local production [28]. In the present study, the concentration of dialysate IL-6 was almost fivefold higher that of plasma IL-6.

In contrast, the higher concentration of plasma as compared to dialysate copeptin and the correlation between dialysate and plasma copeptin (Fig. 1) suggest that dialysate copeptin is mainly stemming from transport of circulating copeptin into the dialysate. Dialysate copeptin showed a negative correlation with PSTR (Fig. 2) and a positive correlation with peritoneal ultrafiltration—which is inversely related to PSTR. We speculate that the mechanisms of this phenomenon might involve increased transport of copeptin from blood to dialysate due to convective transport through the pores of the endothelium with the higher ultrafiltration flow and, in addition, the impact of vasoconstriction, leading on the one hand to a reduction of the number of perfused microcapillaries, but, on the other hand, to increased hydrostatic pressure—a driving force for increased solute transport—in the capillaries of the peritoneal tissue. Although dialysate copeptin was associated with D/D0 glucose in univariate analysis, these different mechanisms, including not only solute transport through small pores, but also convective transport and ultrafiltration might be the reason that dialysate copeptin did not turn out to be the significant determinant of D/D0 glucose in multivariate analysis. Vasopressin contributes to vasoconstriction by different mechanisms: stimulation of the renin–angiotensin–aldosteron system (RAAS), direct effect on smooth muscle cells, or via secretion of endothelin-1 (ET-1) from endothelial cells, which aggravates endothelial dysfunction [13, 29, 30]. Kocyigit et al. in a study in ADPKD patients found that copeptin might be considered as a marker of endothelial dysfunction [31]. The influence of vasopressin administered intravenously on peritoneal membrane was examined in dogs; AVP infusion led to the decrease in dialyzed urea, which was interpreted as a decrease in the total membrane area available for diffusion caused by the decrease in splanchnic blood flow [32]. Vasoactive changes in peritoneal membrane vessels lead to changes in PSTR [33]. The reason for the finding in the present study that dialysate copeptin—and not plasma copeptin—was associated with peritoneal transport parameters is unclear.

The results of the present study should be considered given some important limitations. First, the examined group of patients is relatively small and investigations were carried out during the initial weeks of patients starting PD therapy. Second, given the cross-sectional character of the study, the cause–effect relation cannot be determined. Thirdly, it is not clear if or to what extent copeptin as such may act on the peritoneal vasculature, and AVP was not measured. Further studies are needed to elucidate the role of copeptin in the function of peritoneal membrane in PD patients.

In summary, we report that an increased concentration of dialysate copeptin, a surrogate marker of AVP, associates with two markers of PSTR, i.e., inversely with D/P creatinine and positively with D/D0 glucose, and with increased peritoneal ultrafiltration during PET, while no such associations were observed for plasma copeptin. Whether these findings reflect changes associated with local effects of AVP or if dialysate copeptin per se may influence PSTR is unclear and should be explored in future studies involving larger number of patients.