Association between Arterial Stiffness, Carbamylation, and Mortality in Patients Undergoing Coronary Angiography with No or Mild Chronic Kidney Disease

Authors: Clara Daschner, Marcus E. Kleber, Ksenija Stach, Goekhan Yuecel, Faeq Husain-Syed, Niklas Ayasse, Anders H. Berg, Winfried März, Bernhard K. Krämer, Babak Yazdani

Abstract

Introduction

The processes of atherosclerosis, inflammation, and carbamylation are closely linked in cardiovascular (CV) disease, but the potential of carbamylation burden as a CV mortality predictor is unclear, especially in patients with no or mild chronic kidney disease (CKD). This study aimed to investigate whether elevated carbamylated albumin (C-Alb), as a surrogate marker for carbamylation burden, is associated with mortality and arterial stiffness/atherosclerotic burden in patients with no or mild CKD, using pulse pressure (PP) as a marker for arterial stiffness.

Methods

We measured C-Alb in 3,193 participants of the Ludwigshafen Risk and Cardiovascular Health study who had been referred for coronary angiography and followed up for 10 years.

Results

The mean age was 62.7 years, and 30.4% were female. Mean blood pressure was 141/81 mm Hg, and mean C-Alb was 5.54 mmol/mol. Increase in C-Alb levels was associated with older age; female sex; increased PP, high-sensitivity C-reactive protein, and interleukin-6 levels; and increased incidence of coronary artery disease (CAD), peripheral artery disease (PAD), and carotid stenosis. In contrast, BMI, diastolic blood pressure (DBP), albumin, and the proportion of active smokers decreased with increasing C-Alb levels. In particular, C-Alb showed a highly significant correlation with CAD Friesinger (Pearson correlation coefficient [r] = 0.082, p < 0.001) and Gensini score (r = 0.066, p < 0.001). The area under the curve (AUC) for all-cause mortality prediction by the European Society of Cardiology Heart Score (ESC-HS) significantly improved from 0.719 to 0.735, and the AUC for CV mortality prediction based on C-Alb increased from 0.726 to 0.750 in patients without previously known CV disease. C-Alb correlated directly and significantly with PP (r = 0.062, p < 0.001), which was consistently the strongest predictor of mortality across all C-Alb tertiles. The hazard ratios (HRs) for all-cause mortality per 10 mm Hg increase (or 1,000 mm Hg/min increase for double product [DP]) in the 1st tertile of C-Alb were 1.18, 1.13, 1.11, and 1.11 for PP, mean arterial pressure (MAP), systolic blood pressure (SBP), and DP, respectively, but the HR for DBP did not reach significance. In the 3rd tertile of C-Alb, the HRs were 1.13, 1.05, and 1.09, for PP, SBP, and DP, respectively, but the HR for MAP did not reach significance.

Conclusion

C-Alb may be a valuable biomarker for assessing CV risk and improving mortality prediction even in patients with no or mild CKD. The findings support the notion of a crosslink between carbamylation, inflammation, atherosclerosis, and mortality. While these results are promising, further research is needed to fully elucidate the role of C-Alb in CV disease progression and risk stratification.

Introduction

Inflammation is known to play a pathogenic role in all stages of atherosclerosis – from plaque initiation to rupture [1–6]. In addition, inflammation is closely linked to arterial stiffness, an independent predictor of cardiovascular (CV) events [7, 8]. Inflammation significantly contributes to the stiffening of large arteries through processes such as atherosclerosis, endothelial dysfunction, oxidative stress, vascular calcification, and extracellular matrix degradation [9]. Moreover, inflammation can lead to structural changes in the vascular wall, including fragmentation and stiffening of elastic fibers, thus contributing to increased arterial stiffness [1, 3, 6, 10]. Recent research has shown that vascular elastic fibers are prone to a nonenzymatic posttranslational protein modification process called carbamylation that is associated with an increase in their stiffness at the molecular level and may have deep consequences on the mechanical properties of the vascular wall. Thus, atherosclerosis, inflammation, arterial stiffness, and carbamylation may function as interconnected processes that contribute to CV disease. Moreover, carbamylation, together with other posttranslational modifications and traditional CV risk factors, may act as a crucial factor in the etiology of CV diseases [10].

Carbamylation occurs when free amino groups on proteins spontaneously react with cyanate, a urea dissociation product [11]. Urea builds up in the patient’s blood as kidney function declines, and along with it, the carbamylation burden also increases. In a manner similar to that of hemoglobin A1c levels and time-averaged glucose concentrations, the proportion of carbamylated albumin (C-Alb), a marker of protein carbamylation burden in general, increases with the time-averaged blood urea concentration [12]. Accordingly, increased levels of carbamylated proteins were not only found in the plasma but also in the tissues of 75% nephrectomized mice compared to control mice. Further, the carbamylated protein tissue concentrations in the heart, kidney, and aorta were found to be 3.9-, 3-, and 2.3-fold higher than normal at 5 weeks following nephrectomy [13]. Accordingly, higher C-Alb levels were linked to a higher possibility of developing end-stage renal disease (ESRD) in a cohort of 3,111 participants with chronic kidney disease (CKD) stages 2–4 [14]. CKD is a condition in which carbamylation is promoted due to hyperuremia. Carbamylated lipoproteins are more atherogenic, and uremic, pro-carbamylating conditions stimulate calcification processes that contribute to arterial stiffening [15]. Interestingly, we have recently shown in about 3,200 patients with no to mild CKD that C-Alb acts as a significant predictor of all-cause and CV mortality, even after multivariate adjustment for estimated glomerular filtration rate (eGFR), among other factors (Yazdani, Drechsler 2024 unpublished data). Based on these findings and the possible links between inflammation, carbamylation, and atherosclerosis, it would be interesting to explore the mortality prediction potential of C-Alb in patients with CV disease who have no or mild CKD. Accordingly, the aim of this study was to investigate whether the C-Alb level, as a surrogate marker for overall carbamylation load, is associated with blood pressure parameters (as markers of CV disease) and CV and all-cause mortality in patients with no or mild CKD.

In this study, we focused on pulse pressure (PP) as a clinically well-known surrogate parameter for vascular stiffness that can be easily calculated by subtracting diastolic blood pressure (DBP) from systolic blood pressure (SBP). In fact, we have recently shown that increase in PP is associated with age, diabetes, obesity, atherosclerotic burden (i.e., the number of atherosclerotic vascular beds), impaired renal function, and abnormal catecholamine levels, but it is not affected by renin [16–18]. Additionally, we have demonstrated in a population with a medium to high CV risk with widespread use of antihypertensive medication (>85%), particularly among patients with coronary artery disease (CAD) and HF, that PP and double product (DP, defined as the product of SBP and heart rate) are powerful predictors of CV and all-cause mortality [16]. In line with our previous findings, the Framingham Heart Study was able to demonstrate in a cohort of 1,924 patients, who were not receiving antihypertensive medication at baseline, that an increase of 10 mm Hg in PP, SBP, and DBP was significantly associated with an increase in the risk of developing CAD by 23%, 16%, and 14%, respectively, after a mean follow-up of 14.3 years [19]. Moreover, the RENAAL study on 1,513 patients with nephropathy, arterial hypertension, and type 2 diabetes showed that a baseline PP ≥70 mm Hg was associated with an increased risk of serum creatinine doubling, ESRD, or death [20]. This significant predictive value of high PP for the risk of ESRD was also confirmed in a study from Singapore involving 30,636 participants [21]. Thus, the reported links between PP and ESRD and between C-Alb and ESRD point to a possible link between PP and carbamylation burden and further support the possibility of using C-Alb as a marker of CV risk and mortality.

Methods

Study Cohort

The Ludwigshafen Risk and Cardiovascular Health (LURIC) study is a monocentric hospital-based cohort study that enrolled 3,316 patients of German ancestry residing in the southwest of Germany around the cities of Ludwigshafen, Mannheim, and Heidelberg, who underwent coronary angiography between July 1997 and January 2000. The main indications for coronary angiography were acute chest pain and a positive noninvasive cardiac stress test. The exclusion criteria were the presence of other acute cardiac diseases such as decompensated HF, decompensated valvular disease, acute noncardiac diseases such as infection, endocrine disease or any type of surgery within the previous 3 months, chronic polymorbid disease where noncardiac disease was predominant (i.e., chronic renal failure and hemodialysis, severe rheumatic arthritis, and persistent incapacitation after accident/trauma), history of malignant disease within the last 5 years, and inability to understand the purpose of the study [22]. Participants who met the study criteria and were included, underwent a detailed physical examination, including assessment of vital signs, and a detailed blood test.

Using an automated oscillometric device (Omron MX4; Omron Healthcare GmbH, Hamburg, Germany), blood pressure was measured with the patient in the supine position for at least 10 min. At least three consecutive measurements of SBP and DBP were obtained at intervals of 30 s. If the SBP, DBP, or heart rate measurements varied by >10 mm Hg, >5 mm Hg, or >5 beats per minute (except during atrial fibrillation), respectively, they were regarded as invalid and then repeated. Both valid and invalid measurements were recorded. The invalid measurements were immediately identified as such. Omron devices are clinically validated [23, 24] and regularly tested according to the quality control standards of the Federal Medizinproduktgesetz (Federal Regulation for Medical Instruments). Only stable measurements that matched the reproducibility criteria were entered into the database. One part of the questionnaire investigated whether blood pressure had been measured according to protocol, that is, whether the measurements were made over a 10-min period with the patient at rest in the supine position.

Laboratory Assessments

Venous blood was drawn under standardized conditions after overnight fasting at the time of entry into the study. Within 30 min after venipuncture, the obtained blood was centrifuged at 3,000 g for 10 min, immediately aliquoted, and frozen at −80°C until further analysis. Serum C-Alb was measured and reported in SI units as the ratio of millimoles of C-Alb (defined by carbamylation of Lys-549) to moles of total albumin using high-performance liquid chromatography and tandem mass spectrometry. A complete description of the mass spectrometric assay for C-Alb and its analytic validation have previously been described [12], with the modification that instead of reporting C-Alb as percent carbamylated albumin (%C-Alb), C-Alb is herein reported in units of mmol/mol (mmoles carbamylated albumin per mole of uncarbamylated albumin). C-Alb values were available for 3,193 study participants, and these were used for all analyses presented in this manuscript.

High-sensitivity C-reactive protein (CRP) and cystatin C were measured by immunonephelometry (N-High-Sensitive CRP; N-Latex Cystatin C, Dade Behring, Marburg, Germany) using a Behring Nephelometer II. Creatinine level was measured using the CREA assay (Roche, Germany) on a Hitachi 717 analyzer, and the glomerular filtration rate was estimated by using the 2012 CKD-EPI eGFRcreat-cys equation [25]. Renin was measured by an immunoradiometric assay (Active renin/Berthold Multi-Crystal Counter LB2014), and aldosterone was measured with an immunoradiometric assay (Coat-a-count aldosterone/Berthold Multi-Crystal Counter LB2014). High-sensitive cardiac troponin was assessed using the immunoassay of MODULAR E170 series automated analyzers (Roche Diagnostics, Switzerland), and N-terminal pro-B-type natriuretic peptide (NT-proBNP) was measured by electrochemiluminescence on an Elecsys 2010 Analyzer (Roche Diagnostics, Switzerland). Total cholesterol, high-density lipoprotein cholesterol, and triglyceride concentrations were quantified with enzymatic reagents from WAKO (Neuss, Germany) on a WAKO 30R analyzer. LDL and very low-density lipoprotein cholesterol were separated by ultracentrifugation in a Beckman LM-8 ultracentrifuge in 100-μL volumes with a VT-51.2 rotor (Beckman Coulter). Cotinine was determined by radioimmunoassay using Nikotin-Metabolit RIA (DPC Biermann GmbH, Bad Nauheim, Germany). Cholinesterase was measured on a Hitachi 717 analyzer using the CHE assay (Roche, Mannheim, Germany). Glycated hemoglobin was measured by immunoassay (hemoglobin A1c UNIMATE 5; Hoffmann-La Roche, Grenzach-Wyhlen, Germany). The study protocol and the analytic methods used here have been reported previously [22].

Clinical Definitions

As per the criteria of the American Heart Association, CAD was defined as visible luminal narrowing (20% stenosis or more) in at least one of 15 coronary segments [22]. Coronary 1-vessel disease, coronary 2-vessel disease, and coronary 3-vessel disease were defined as stenosis of 50% or higher in one, two, or three of the major coronary arteries, namely, the left anterior descending artery, ramus circumflexus, and right coronary artery, respectively [22]. The echocardiographic classification of HF has been published in detail, along with a summary of other clinical definitions [22]. According to the definition published by Paulus et al. [26], HF with preserved ejection fraction was defined as preserved left ventricular function with an ejection fraction >45% (echocardiographic or invasive) and the presence of diastolic HF. Diastolic dysfunction was diagnosed in 388 patients (85%) based on mean pulmonary capillary wedge pressure >12 mm Hg or left ventricular end-diastolic pressure >16 mm Hg. In the remaining 71 patients (15%), diastolic dysfunction was diagnosed based on an elevated NT-proBNP concentration (>220 pg/mL) and electrocardiographic evidence of atrial fibrillation. Smoking status was assessed both by history and by plasma levels of cotinine, with the cutoff being >15 μg/L.

Follow-Up Assessments

Patients were followed up for a median of 9.9 years. Data on vital signs were obtained from local registries. Two experienced clinicians who were blinded to the patient characteristics reviewed death certificates, medical records of local hospitals, and autopsy data independently. They classified the causes of death. In cases involving disagreement or uncertainty concerning the coding of a specific cause of death, the decision was made by the principal investigator (W.M.). During the follow-up, 967 (30.2%) participants died, with 604 (19.0%) cases of CV mortality that included sudden cardiac death (n = 253), fatal myocardial infarction (n = 103), death due to congestive HF (n = 146), death after intervention to treat CAD (n = 26), fatal stroke (n = 54), and other causes of death due to CAD (n = 19).

Statistical Analysis

Continuous data are presented as the mean and standard deviation (SD) when normally distributed or as the median and 25th and 75th percentile for variables with non-normal distribution. Categorical data are presented as percentages. Statistical differences between groups and continuous variables were determined using analysis of variance (ANOVA). Non-normally distributed variables were log-transformed before they were entered into the analysis. The chi-square test was used for analysis of categorical variables. Missing values for covariates were imputed using the R package “Hmisc” (v5.1–2). The correlation of log C-Alb values with markers of arterial stiffness and CAD was assessed using Pearson correlation analysis. The plots showing the association of log C-Alb values with the number of vascular beds affected by atherosclerosis and the number of coronary vessels with stenosis were generated using the R package “ggstatsplot” (v0.12.3).

Cox proportional hazard models were built to assess the effect of PP, mean arterial pressure (MAP), SBP, DBP, and DP on all-cause and CV mortality within tertiles of C-Alb. The proportional hazard assumption was checked by examination of scaled Schoenfeld residuals. All tests were two-sided, and a p value <0.05 was considered to indicate statistical significance. All analyses were carried out using R v4.4.0 (http://www.r-project.org). Harrell’s C was calculated using the R package “hmisc” (v4.2-0). Receiver operating characteristics (ROC) curves were calculated and compared using the method of Delong as implemented in the R package “pROC” (v1.18.5), and the NRI for censored survival data was calculated using the R package “nricens” (v1.2).

Results

Patient Characteristics at Baseline

Baseline C-Alb data were available for 3,193 of the 3,316 participants in the LURIC study. The cohort had a mean age of 62.7 years, and 30.4% of the participants were female. The mean SBP was 141 mm Hg; DBP was 81 mm Hg, and PP was 60.2 mm Hg. In order to analyze the association of C-Alb with anthropometric data and biomarkers, we divided the cohort into tertiles of C-Alb. Patients in the 1st, 2nd, or 3rd C-Alb tertiles had average C-Alb values of 4.4 mmol/mol, 5.54 mmol/mol, and 7.45 mmol/mol, respectively. Increase in C-Alb was associated with older age; female sex; higher PP, urea, cystatin C, creatinine, NT-proBNP, high-sensitivity C-reactive protein (hsCRP), and IL-6 levels; greater prevalence of CAD, peripheral artery disease, and carotid stenosis; and greater prevalence of prior myocardial infarction, prior stroke, and prior bypass. In contrast, increase in C-Alb was associated with a decrease in BMI, DBP, triglyceride levels, albumin levels, and the proportion of active smokers. The proportion of patients with only one atherosclerotic vessel remained unchanged with increasing C-Alb, although the proportion of patients with two and three atherosclerotic vascular beds increased significantly from 10.1% to 13.5% and 1.41% to 2.54%, respectively. On the other hand, there was no change in fasting glucose, HbA1c, SBP, MAP, DP, or the proportion of former smokers as C-Alb increased (Table 1).

Association of C-Alb with Blood Pressure Parameters

C-Alb values at baseline correlated differently with the blood pressure parameters examined. C-Alb showed a highly significant direct correlation with PP (Pearson correlation coefficient [r] = 0.062, p < 0.001, Fig. 1a) and an indirect correlation with MAP (Pearson correlation [r] = −0.050, p = 0.005, Fig. 1b) and DBP (Pearson correlation [r] = −0.070, p < 0.001, Fig. 1d). On the other hand, C-Alb did not show a significant correlation with SBP (p = 0.320; Fig. 1c) or DP (p = 0.654; Fig. 1e).

Correlation of C-Alb with Severity of CAD and Atherosclerotic Altered Vascular Beds

C-Alb showed a highly significant correlation with the Friesinger score, a quantitative indicator of the severity of CAD based on coronary angiography data (Pearson correlation [r] = 0.082, p < 0.001, Fig. 2a). In addition, C-Alb also showed a highly significant correlation with the Gensini score, which is used to quantify the severity of CAD based on coronary angiography findings by taking into account the degree of stenosis and also the specific location of the stenosis within the coronary artery tree (Pearson correlation [r] = 0.066, p < 0.001, Fig. 2b).

Additionally, we have illustrated the distribution of log C-Alb values for each CAD category by constructing violin plots (Fig. 3). As shown in the figure, the log C-Alb values increased significantly from no CAD to 3-vessel CAD: 1.72, 1.72, 1.75, and 1.78 for no CAD, 1-vessel CAD, 2-vessel CAD, and 3-vessel CAD, respectively (Welch’s ANOVA, p < 0.001). Pairwise comparisons using the Games-Howell test proved that the differences were significant.

Each violin plot in Figure 4 displays the distribution of log C-Alb values within each category of 0, 1, 2, or 3 atherosclerotic vascular beds. The mean log C-Alb values increased significantly with increase in the number of atherosclerotic vascular 1.71, 1.74, 1.80, and 1.83 for 0, 1, 2, and 3 atherosclerotic vascular beds, respectively (Welch’s ANOVA, p < 0.001). Pairwise comparisons using the Games-Howell test, adjusted for multiple comparisons, confirmed that the differences were significant (p = 0.002).

Performance in Risk Prediction Models

We investigated whether the addition of C-Alb to established risk prediction algorithms for individuals with or without a clinical history of CV disease would improve risk prediction. We selected a subgroup of patients with measured C-Alb values who had not experienced a previous myocardial infarction or stroke at the study baseline (n = 1,393) and calculated the European Society of Cardiology Heart Score (ESC-HS) [27] with and without C-Alb (Table 2). The area under the curve (AUC) for all-cause mortality prediction by ESC-HS significantly improved from 0.719 to 0.735 with the addition of C-Alb; for CV mortality, it improved significantly from 0.726 to 0.750 with the addition of C-Alb; each in patients without previously known CV disease.

In a subgroup of patients with stable CAD (n = 1,215), we tested the incremental predictive value of C-Alb using the Vienna and Ludwigshafen Coronary Artery Disease (VILCAD) risk score [28] (Table 2). The AUC for all-cause and CV mortality prediction by the VILCAD risk score for patients with previously known CV disease increased with the addition of C-Alb, but the improvement in prediction ability was not significant (Table 2).

The addition of hsCRP to the ESC-HS resulted in an AUC of 0.742 (95% CI: 0.712–0.773) for all-cause mortality and 0.759 (95% CI: 0.722–0.795) for CV mortality. Similarly, the addition of hsCRP to the VILCAD score yielded an AUC of 0.743 (95% CI: 0.714–0.773) for all-cause mortality and 0.742 (95% CI: 0.708–0.775) for CV mortality (Table 2). Thus, both hsCRP and C-Alb demonstrate comparable improvements in the predictive accuracy of these risk scores.

Mortality Prediction Ability of Blood Pressure Parameters in C-Alb Tertiles

The hazard ratios (HRs) for all-cause mortality for PP, MAP, SBP, and DP were 1.18, 1.13, 1.11, and 1.11, respectively, in the 1st tertile of C-Alb. In contrast, the HR for DBP did not reach significance. In the 2nd C-Alb tertile, the HRs for all-cause mortality prediction based on PP, MAP, SBP, DBP, and DP were 1.19, 1.18, 1.14, 1.12, and 1.1, respectively. In the 3rd tertile of C-Alb, the HRs for PP, SBP, and DP were significant at 1.13, 1.05, and 1.09, respectively, while the HR for MAP was not significant (Fig. 5).

With regard to CV mortality, PP showed robust HRs of 1.23, 1.14, and 1.11 in the 1st, 2nd, and 3rd C-Alb tertiles, respectively, while DP showed HRs of 1.11, 1.09, and 1.09 in the 1st, 2nd, and 3rd C-Alb tertiles, respectively. SBP and MAP had HRs of 1.14 and 1.16, respectively, in the 1st tertile of C-Alb, and 1.12 and 1.19, respectively, in the 2nd tertile of C-Alb, but neither of them reached significance in the 3rd C-Alb tertile. The HR for DBP did not reach significance in the 1st and 3rd C-Alb tertiles, but in the 2nd C-Alb tertile, the HR for DBP was significant at 1.17 (Fig. 6). In both Figures 5 and 6, the HR is shown for every 10 mm Hg increase in the case of PP, MAP, DBP and SBP, and for every 1,000 mm Hg/min increase in the case of DP.

Discussion

In the present study, we have demonstrated the impact of carbamylation on blood pressure parameters and arterial stiffness in a cohort of patients with no or mild CKD. PP, one of the most commonly used indicators of arterial stiffness showed a significant direct correlation with the extent of albumin carbamylation, which serves as a surrogate marker for overall carbamylation. Interestingly, DBP seemed to have a bigger influence on this association than SBP. A possible explanation could be that carbamylation affects not only albumin and other plasma proteins but also tissues and blood vessels. Consequently, the central large vessels, in particular, lose their elasticity; this, in turn, impairs their Windkessel function and leads to reduced DBP and high PP. This finding is in alignment with the study conducted by Doué et al. [10] in which protein carbamylation in mice was induced by feeding with cyanate-supplemented water and found to cause carbamylation of elastic fibers in the vascular wall [20]. They used atomic force microscopy and demonstrated an increase in the stiffness of elastic fibers and also demonstrated a heightened aortic pulse wave velocity. Based on these findings, it would be interesting to explore the mechanisms via which carbamylation affects vascular elasticity in the future.

In the present study, we have shown that the level of C-Alb (as a marker of protein carbamylation) increases significantly as the severity of CAD increases (no CAD <1-vessel CAD <2-vessel CAD <3-vessel CAD). This is also corroborated by the positive correlation between C-Alb levels and both the Friesinger and Gensini scores, which are used to quantify CAD severity based on coronary angiography data. Additionally, our findings demonstrate that C-Alb levels increase significantly as the degree of involvement of atherosclerotic vascular beds increases, that is, across CAD, carotid stenosis, and peripheral artery disease. This is indicative of a positive relationship between C-Alb levels and the extent of atherosclerotic involvement of vessels, regardless of their central or peripheral location. A possible explanation for this observed link is that LDL is also carbamylated and its carbamylation has negative effects on vascular health. For example, it has been reported that endothelial dysfunction can be caused by carbamylated LDL through activation of the lectin-like oxidized LDL receptor-1 and increase in production of reactive oxygen species production, which leads to the uncoupling of endothelial nitric oxide synthase [29]. Additionally, carbamylated LDL exhibits atherogenic properties, such as binding to macrophage receptors, leading to cholesterol accumulation and foam cell formation, and stimulating vascular smooth muscle cell proliferation [30]. All these processes are known to be associated with vascular stiffness and further support the potential of C-Alb as a biomarker or risk factor for CV disease severity, even in patients with no or mild CKD.

Another important finding that emerged from this study is that the addition of C-Alb significantly improved all-cause and CV mortality prediction by ESC-HS in patients without previously known CV disease. Similarly, the VILCAD score showed a numerical improvement in mortality prediction with the same trend, although it did not reach statistical significance. These improvements are noteworthy, as even small absolute increases in AUC are meaningful in risk prediction models, especially for high-stakes outcomes like all-cause and CV mortality. Furthermore, it is important to highlight that this effect was demonstrated in a patient cohort with no or only mild CKD. Therefore, we suggest including C-Alb in risk scores to maximize their predictive accuracy. However, it is currently not clinically available everywhere. Currently, C-Alb can only be measured using liquid chromatography and tandem mass spectrometry. Following investments in the necessary equipment, our laboratory’s cost per sample is approximately USD 1.72 (excluding labor). Sample processing requires 2 h, and the instrument analysis itself takes 7 min. In line with this observation, in our previous study in participants with no or mild CKD, C-Alb showed the strongest predictive association with death from congestive HF, with a hazard ratio of 7.19 per unit increase in log C-Alb (Yazdani, Drechsler 2024 unpublished data). These findings further underline the impact of protein carbamylation on CV health. In addition, PP showed the highest mortality association per 10 mm Hg increase in almost all C-Alb tertiles compared to the other blood pressure parameters SBP, DBP, MP, and DP. This finding is in line with the negative correlation observed between C-Alb and vascular elasticity. Notably, this was noted for both all-cause and CV mortality rates and is consistent with a recently published review by Noels et al. that highlighted the role of posttranslationally modified proteins in promoting oxidative stress, inflammation and fibrosis, as well as inducing vascular damage and prothrombotic effects, thus contributing to CKD progression and/or increased CV risk in patients with CKD [31].

IL-6, an inflammatory cytokine, has been identified as a significant prognostic marker for CV and all-cause mortality in dialysis patients. A meta-analysis of 28 studies has shown that elevated IL-6 levels are associated with a heightened risk of mortality [32]. Furthermore, Chertow et al. [33] showed in a phase 2b trial that inhibiting IL-6 with the monoclonal antibody clazakizumab could reduce inflammatory biomarkers, such as high-sensitivity CRP, in patients undergoing dialysis. This is consistent with our results, as increase in C-Alb was accompanied by a significant increase in both hsCRP and IL-6. These findings further support the previously reported links between carbamylation and inflammation. As elevated CRP and C-Alb levels occur simultaneously in patients with CKD, it would be interesting to further explore the association between C-Alb, CRP, and IL-6 levels to more accurately estimate the risk of CV mortality, morbidity, progression to ESRD, and possible treatment options. In terms of therapeutic options, Perl et al. demonstrated in 2016 that extended hemodialysis sessions reduce C-Alb levels more effectively than conventional dialysis [34], with a similar effect observed on hsCRP levels [35]. Further, extended hemodialysis has been shown to have a better survival outcome [36, 37]. However, extended hemodialysis may be difficult to implement due to practical and logistical challenges. As it requires longer treatment sessions, it can burden both the patients and the healthcare resources. The infrastructure of dialysis centers might not be sufficient to accommodate extended dialysis on a larger scale. Thus, it is important to explore other options to reduce CRP and C-Alb levels to improve survival outcomes in patients who are not accessible for extended hemodialysis.

A primary limitation of our study is that we included only patients of German origin, as this restricts the generalizability of our findings. In addition, both numerator and denominator are important in this mathematical approach. Serum albumin levels decrease slightly from 4.4 g/dL in the first and second C-Alb tertiles to 4.26 g/dL in the third C-Alb tertile, possibly affecting the C-Alb value in the 3rd tertile, which is the ratio of carbamylated albumin to total albumin. Furthermore, C-Alb was only measured at the study entry time point, and longitudinal measurements were not analyzed. Further randomized controlled trials on long-term changes in C-Alb and the effect of possible interventions are needed to understand further the causal links between carbamylation and vascular function.

In conclusion, our study demonstrates the association of carbamylation with PP, a surrogate marker of arterial stiffness, and atherosclerotic burden in a population with no or only mild CKD. Based on the findings, C-Alb could be considered as a component in risk models for CV and all-cause mortality to improve their predictive accuracy. Overall, the findings suggest a complex interplay between inflammation, arterial stiffness, and carbamylation in the development and progression of atherosclerosis, particularly in the context of early CKD. Understanding these mechanisms may lead to the development of novel therapeutic strategies targeting carbamylation to reduce the burden of CV disease.

Statement of Ethics

The Landesärztekammer Rheinland-Pfalz provided ethical approval for this study (#837.255.97[1394]). The study protocol was in accordance with the Declaration of Helsinki, and all study participants provided their written informed consent to participate in the study.

Conflict of Interest Statement

Prof. Dr. med. Bernhard K. Krämer reports lecture fees and/or advisory board memberships and/or study participation from Astellas, Bayer, Boehringer Ingelheim, Chiesi, RiePharm, Pfizer, Sanofi, Servier, AstraZeneca, and Vifor Pharma, all outside the submitted work. He is the past president of the German Hypertension Society DHL. Dr. rer. nat. Marcus E. Kleber is employed by SYNLAB Holding Deutschland GmbH. Prof. Dr. med. Winfried März reports grants and personal fees from Aegerion Pharmaceuticals, AMGEN, Sanofi, Alexion Pharmaceuticals, BASF, Abbott Diagnostics, Numares AG, Berlin-Chemie, Akcea Therapeutics; grants from Siemens Healthineers, AstraZeneca, Bayer Vital GmbH, bestbion dx GmbH, Boehringer Ingelheim Pharma GmbH Co KG, Immundiagnostik GmbH, Merck Chemicals GmbH, MSD Sharp and Dohme GmbH, Novartis Pharma GmbH, Olink Proteomics; and other from SYNLAB Holding Deutschland GmbH, all outside the submitted work. The remaining authors have no conflicts of interest to declare.

Funding Sources

None to specify.

Author Contributions

Babak Yazdani and Clara Daschner designed the research, wrote the first draft of the manuscript, and did the literature research. In addition, Babak Yazdani interpreted the results as a senior author and discussed them with the current literature in a leading role. Marcus Kleber made statistical analyses. Bernhard K. Krämer, Faeq Husain-Syed, and Niklas Ayasse helped discuss the results and thereby made contributions to the manuscript. Anders Berg determined C-Alb in the laboratory, proofread the manuscript, and made important intellectual contributions with Gökhan Yücel and Ksenija Stach. Winfried März is the principal investigator of the LURIC study and was instrumental in performing of the study and providing the data for analysis.