Cardiovascular disease (CVD) has consistently remained the leading cause of morbidity and mortality in end-stage renal disease (ESRD) patients [1–3]. Although many traditional and non-traditional risk factors have been identified [2], CVD prevalence still continues to be 20 times higher in ESRD patients than in the general population [1].

Atherogenic dyslipidemia and chronic inflammation have been considered major risk factors for cardiovascular (CV) morbidity and mortality in ESRD patients [1,4,5]. In turn, changes in the balance of oxalate have been reported to be associated with dyslipidemia [6], systemic inflammation [7–9], and thus a high risk of CVD independent of its cause [10,11]. It has been demonstrated that plasma oxalic acid (POx) concentration increases according to the progression of chronic kidney disease (CKD) and reaches its highest level in ESRD patients [12,13].

Moreover, recently, it has been speculated that oxalate is a uremic toxin originating from the colonic microbial metabolism, which indirectly suggests its contribution to systemic inflammation associated with CKD [14]. In addition, in a series of early in vitro studies, Levin et al. [15] and Recht et al. [16] have provided the first evidence that oxalate may be an atherogenic toxin in uremic conditions. Although hyperoxalemia is a well-known occurrence in ESRD patients, the clinical data on the association between POx and CV outcomes in the dialysis population has never been analysed. Considering the above, we therefore hypothesized that oxalate might be strongly involved in atherogenesis and the inflammatory pathway and could result in an increased risk of CVD in ESRD patients.

To address this issue, this prospective, observational, single-centre pilot study was primarily designed to characterize the lipid profile and the pattern of pro-inflammatory cytokines according to POx concentration and to evaluate the potential role of elevated POx in the development of CVD risk. We also aimed to test the feasibility of the study for future large-scale research.

Oxalate is an ionic form of a potentially toxic oxalic acid that is primarily eliminated by the kidneys [8,13]. Loss of kidney function leads to a decrease in renal oxalate clearance and hyperoxalemia in ESRD patients [11]. The accumulation of oxalate is associated with inflammation [8,9,13], high risk factors for CVD [20,21] and an increased mortality rate [8,10] in kidney stone disease patients. However, oxalate has never been investigated as a potential contributor to atherosclerosis, systemic inflammation and CVD risk in ESRD patients. To our knowledge, this is the first study to prospectively evaluate the association between POx concentration and CVD events in the dialysis cohort during a 2-year follow-up period using multivariable Cox regression analysis and a set of sensitivity analyses. Nevertheless, atherogenic dyslipidemia and systemic inflammation are proven risk factors for CVD incidence in ESRD patients and can be important independent confounders. Therefore, in the present study we separately characterized the association between both the lipid profile and the pro-inflammatory markers and POx concentration, which has never been reported before.

Our study resulted in a number of new findings. First, an increasing trend in the atherogenic lipoprotein fractions and the pro-inflammatory markers and a linear decrease in HDL-C was significantly associated with elevated POx concentration. Second, POx concentration was directly associated with blood TG, AIP, IL-6, and MCP-1 levels. Third, POx concentration ≥ 62.9 μmol/L was significantly associated with experienced CVD events independent of other examined CVD risk factors and can be considered a useful marker for predicting CVD events in dialysis patients.

Numerous studies have demonstrated increased CVD morbidity in kidneys stone formers [10,20,21]. Recently, several hypotheses have been proposed to explain the association between kidney stones and CVD: (1) urinary stones and CVD have similar traditional atherosclerotic risk factors, including diabetes, hypertension, metabolic syndrome and obesity; (2) a common origin of both diseases is based on the response-to-injury mechanism, including oxidative stress and inflammation; and (3) kidney stone disease leads to renal function decline, which in turn could increase CV morbidity and mortality [6,8,11,20,22]. In our opinion, all hypotheses mentioned above could doubtless be extrapolated to indicate an association between elevated POx and CVD in ESRD patients.

A wide variety of scientific data indicate that there are common risk factors for calcium oxalate (CaOx) stone diseases and CVD. It has been demonstrated that both diseases are significantly associated with older age, diabetic status, smoking habits, hypertension, hypercholesterolemia, hyperlipidemia, and obesity in these patients compared with controls, and this association remains significant after adjusting for multiple potential confounders [10,23,24].

Apart from the traditional factors mentioned above, there are a number of other uremia-related CVD risk factors in ESRD patients, such as anemia, mineral and bone disorders, malnutrition, overhydration and electrolyte imbalance, that in turn aggravate the prognosis [1,25,26]. We did not focus on well-known risk factors of CVD in our study. However, except for POx concentration, roughly similar confounders for the risk of CVD events (age, diabetes, Hb, TG, BMI, iPTH, and IL-6) were found among baseline parameters in our sample (Supplementary Table 1).

In the present study, we focused on POx concentration and its potential role in dyslipidemia and systemic inflammation associated with CVD risk in ESRD patients. It is somewhat surprising that notwithstanding the fact that the comparative analysis clearly identified a gradually increasing trend in the atherogenic lipoprotein fractions associated with elevated POx, the partial correlation analysis indicated that POx was significantly associated only with TG. It should be emphasized that it was difficult to compare the results of our study with the results of previous findings, as most studies in this field investigated only the association between dyslipidemia and the presence of nephrolithiasis [24,27]. It is unfortunate that the authors did not take into account POx concentration. To the best of our knowledge, in contrast to our findings, a negative association between POx concentration and LDL-C in HD patients was reported in only one study [28].

The available data indicate that oxidative stress and inflammation are the major risk factors for accelerated atherosclerosis [6,29]. Reactive oxygen species oxidize LDL-C to minimally modified LDL-C (mm-LDL-C), which induces the secretion of various cytokines, including MCP-1 [6,30]. In turn, MCP-1 is directly involved in monocyte recruitment, resulting in transformation into foam cells due to modified lipoprotein. Therefore, MCP-1 is one of the earliest events in the pathogenesis of atherosclerosis [30]. In addition to MCP-1, oxidatively modified LDL-C induces the synthesis of pro-inflammatory cytokines, such as IL-6 and TNF-α [30,31]. On the other hand, these mediators are strongly involved in oxalate-induced inflammation [32–34]. A crucial role of MCP-1, IL-6, and TNF-α in the oxalate-induced inflammatory response of the kidneys has been demonstrated in early in vitro and experimental studies [33–35]. In a recent in vitro study, Dominguez-Gutierrez et al. [7] found that CaOx could activate human monocytes, which promoted M1 macrophage development and produced the inflammatory response. Monocytes or neutrophils exposed to oxalate crystals produce TNF-α, IL-6, and MCP-1. As a result, these identical inflammatory pathways in atherosclerosis and hyperoxalemia became the main reason for the initial selection of this cytokine panel in our study. It is unfortunate that clinical studies on the association between POx and cytokine concentrations in ESRD patients have never been performed. Therefore, the results obtained from our study could not be directly compared with the results of previous reports. Nevertheless, in support of the previously mentioned studies, our findings have provided preliminary clinical evidence that elevated POx concentration is associated with increased levels of IL-6, TNF-α, and MCP-1 in ESRD patients. However, it should be noted that the correlation analysis did not confirm the association between POx and serum TNF-α, while the univariate Cox regression analysis substantiated the association between POx and the only pro-inflammatory marker, IL-6. Moreover, we did not observe an association between POx and CRP. The obtained results contradict the existing data, including that for patients with nephrolithiasis [8], and require further clarification. Possible explanations for the contradictory results may be the clinically stable condition of the patients during the entire 2-year follow-up period and the small sample size.

Several studies have found that kidney stone disease is strongly associated with CKD progression and ESRD and that both diseases can lead to CVD [36,37]. POx has been defined as a significant predictor of kidney function decline not only in nephrolithiasis but also in primary hyperoxaluria and enteric hyperoxaluria [12,13]. Moreover, Mulay et al. [38] have identified numerous clinically important CKD complications, including hyperphosphatemia, hyperparathyroidism, hypertension and cardiac fibrosis, in the proposed experimental oxalate diet-induced model of CKD. Thus, both POx and uremia-related complications could lead to CVD and accelerate the process in ESRD patients. A statistically significant association between POx and CVD events during the 2-year follow-up period was demonstrated in the present study. It is important to note that this association remained significant after adjusting the model for the effects of the potential confounding factors previously described.

According to our results, POx concentration ≥ 62.9 μmol/L can predict CVD risk independent of CKD-related factors in dialysis patients. Our data are consistent with the findings of the following two studies. In a very early study, Tomson et al. [39] demonstrated a significant association between POx and vascular calcification in HD patients. More recently, Gulhan et al. [28] found that aortic pulse wave velocity was independently predicted by POx, glucose and TG. This finding undoubtedly confirms a significant role of POx in the pathogenesis of adverse CV outcomes in dialysis patients. However, notwithstanding the statistical significance and sufficient power of our findings in the Cox regression analysis, the sample size was relatively small, and a causal association between POx and CVD events could not be established, although the post hoc analysis showed that a minimum of 41 CVD cases would be required for the analysis to achieve power of 0.90 and alpha of 0.05 in further investigations.

There are several important limitations that should be acknowledged. First, this was an observational study, and the examined parameters (POx concentration and cytokines) were measured at only one time point. Therefore, causality could not be established. Second, the study sample size was relatively small, and the study cohort represented a select group of patients with a clinically stable condition with the target level of Kt/V. This selection criterion could lead to greatly exaggerated findings in our research. In addition, due to the sample size it was not possible to perform subgroup analysis, and we could not rule out the effects of other potential confounding variables (age, sex, dialysis modality, etc.) on the results. Third, we did not evaluate the effects of other factors that could be associated with elevated POx concentration (changes in Ca homeostasis or the administration of phosphate binders and other medications).

Notwithstanding these limitations, our study is the first to report a direct association between elevated POx concentration, atherogenesis, inflammation and CVD risk in ESRD patients. The results of our study have brought to light many questions that require further investigation. More research with a larger cohort is needed to confirm this preliminary evidence and to validate POx as a proposed biomarker for CVD risk in clinical practice. However, we believe our findings can be useful in determining sample size considerations and can serve as the basis for future research projects.