Calcification of the coronary artery is pathognomonic of coronary atherosclerosis and has been confirmed by intravascular ultrasound study and histopathology [1].

Therefore, coronary calcification is a well-established marker for coronary artery disease and predictor for cardiovascular morbidity and cardiovascular death [2]. Several studies demonstrated an association between future cardiovascular events and the amount of coronary calcium (CAC) [3, 4, 5, 6]. CAC screening can therefore be used for risk stratification for future cardiovascular events [4, 5]. Quantification of CAC can be done noninvasively by cardiac computed tomography (CT) [2, 4, 5]. The individual amount of CAC provides an individual risk prediction that is superior to conventional risk factors scores like the ATP III Score or PROCAM score [5]. A CAC of zero, is associated with a high negative predictive value [6]. In, higher scores, e.g., above the 75th percentile or above 400 are associated with an elevated cardiovascular risk independent of underlying risk factors [4].

Coronary calcifications can reflect the individual extent of coronary atherosclerosis. Thus, extensive calcification can be associated with coronary stenosis. The specificity of CAC in the diagnosis of relevant coronary stenosis is limited. The absence of coronary calcification, a score of zero, has a high negative predictive value and can be used for the exclusion of coronary artery disease.

Patients with severe chronic kidney disease (CKD) are at high risk for cardiovascular disease independent from concomitant risk factors [7, 8]. In addition, classic cardiovascular risk factors such as hypertension are common in patients with severe CKD. In addition to classic atherosclerotic lesions which are located in the arterial intima and are the consequences of multifocal, smoldering, immunoinflammatory processes [9] in patients with severe CKD media sclerosis is frequent. In media sclerosis, also known as Monckeberg’s sclerosis, muscular arteries undergo non-inflammatory medial calcification [10], which is asymptomatic. Progression of media sclerosis results in the loss of the cushioning function in blood vessels, thus inducing pseudo hypertension, left ventricular hypertrophy and altered coronary perfusion [10, 11, 12].

Because of this additional mechanism of vascular calcification in severe CKD patients, the predictive value of CAC could be altered. Therefore we investigated the predictive value of CAC in a population of severe CKD patients and compared it to a population with regular to moderate reduced renal function.

1094 individuals (724 males and 370 females, aged 62 $\pm$ 9.3 years) were included in the study. 393 individuals had a creatinine clearance $\le$30 mL/min/1.73 cm² (group 1) and 701 patients had a creatinine clearance $\ge$30 mL/min/1.73 cm² (group 2). The mean observation time was 6.2 $\pm$ 1.3 years. Patient characteristics for both groups are listed in Table 1. The mean number of risk factors per person in each group was 1.9 $\pm$ 1.0. There was no significant difference in age and risk factor distribution between the two study groups.

The mean CAC score was 152 $\pm$ 127 in all patients. The average calcium score was significantly higher in patients with severe kidney dysfunction than in patients with normal to moderate reduced renal function (207 $\pm$ 190 vs. 121 $\pm$ 169, p 0.001). Coronary calcifications could be excluded in 180 individuals (44 in group 1 and 136 in group 2).

The mean score in females was significantly lower compared to males (101 $\pm$ 78 vs. 178 $\pm$ 141, p 0.01). We found this difference in patients with normal to moderate reduced renal function (88.6 $\pm$ 71.2 vs. 138 $\pm$ 102, p 0.01) and in patients with severe CKD (174 $\pm$ 153 vs. 222 $\pm$ 171, p 0.01).

185 cardiovascular events (MI, myocardial revascularization, cardiac death) were observed. There was a significantly higher event rate of all cardiovascular events in patients with severe reduced renal function, p = 0.001. 19 patients in group 1 and 21 patients in group 2 died from cardiac death (p = 0.03). 107 patients suffered from MI, 46 in group 1 and 61 in group 2. 72 patients in group 1 and 94 patients in group 2 underwent myocardial revascularization, p 0.01. The distribution of cardiovascular events and event rates are given in Table 2.

In Tables 3,4, the risk ratio of patients for a calcium score above 400 is given in dependence of different cardiovascular risk factors. In a multivariate analysis, we could identify a significant correlation between a calcium score above 400 and age, male sex, hyperlipidemia, diabetes, smoking, and hypertension. In addition, severe renal insufficiency could be identified as an independent risk factor for an elevated calcium score.

Table 5 shows the hazard ratio for cardiovascular events in different score groups in group 1 and group 2. The hazard ratio was adjusted for sex, age, and coronary risk factors including hypertension, hypercholesterolemia, diabetes and smoking. In both groups we found a strong correlation between an increasing CAC score and an increasing hazard ratio for cardiovascular events. No cardiovascular events were observed in all patients without coronary calcification. The hazard ratio for cardiovascular events increased constantly with the CAC score in both groups, up to 7.3 (95% CI 4.0–10.8) in group 1 and 9.3 (95% CI 3.6–12.4) in group 2 for patients with a score above 1000. In corresponding score groups, the hazard ratio in patients with regular to moderately reduced renal function, group 2, was significantly higher compared to patients with severely impaired renal function.

CAC is a well-established risk marker for future cardiovascular events. In previous studies, CAC demonstrated a more accurate and individual risk stratification for future cardiovascular events compared to score systems using cardiovascular risk factors such as the ATP III risk score or PROCAM score [3, 4, 5, 14].

The aim of this study was to evaluate whether CAC is a reliable marker in patients with severe CKD. Our study population consisted of asymptomatic patients who had no signs of prior coronary artery disease.

Cardiovascular risk factors were similarly distributed in both groups. There was a significantly higher number of patients on statin therapy in the group with severe CKD with a similar distribution of hyperlipidemia. This may be due to more frequent blood checks in the group of patients with severe CKD, which then lead to tighter control of lipid levels and more frequent administration of statins.

Severe CKD was associated with increased CAC compared to patients with normal to moderately reduced renal function. In patients with severe CKD, we could demonstrate that conventional cardiovascular risk factors led to higher amounts of CAC (Table 3). In addition, elevated CAC was found with increasing age and in male patients. These findings are consistent with previous studies in patients without severe CKD [4, 15]. Severe CKD itself could be identified as an independent risk factor for coronary calcification and as an independent risk factor for cardiovascular events.

Due to the protective effects of gestagen/progesteron on atherosclerosis, the CAC of females is lower compared to males in the same age group [16]. We found a higher calcium score in males, both in the normal to moderately reduced renal function group and in the severe CKD group. Despite a higher overall score, the difference was smaller in patients with severe CKD. This shows that the protective effect of the female sex is less relevant in this group. Due to the additional calcification induced by renal insufficiency, which occur regardless of gender, the scores could be slowly adjusted.

As demonstrated in prior studies [15, 17], patients with a CAC score of zero had a very low risk for future cardiovascular events (Table 5). Patients with a calcium score of zero on two consecutive examinations showed the best coronary disease prognosis [17]. CT angiographic studies have shown that patients with a score of 0 can have non-calcified plaques, but usually without relevant stenosis. Overall, however, the calcium score correlates with the extent of total atherosclerosis, so that a score of 0 indicates a low risk of atherosclerosis.

Over a follow up of 6.2 years in both groups, no patient with a CAC score of zero suffered from a cardiac event in our study. This demonstrates the excellent negative predictive value of a CAC score of zero even in high-risk populations. All 44 patients with a CAC score of 0 in the severe CKD group had no cardiovascular events during follow up regardless of existing risk factors.

Calcification is an almost ubiquitous pathological process in patients with end stage renal disease [9]. Therefore, the higher amount of CAC in the severe CKD group is not surprising. In severe CKD patients, the cardiovascular risk increased with an increasing CAC score (Table 5). Still, the hazard ratio of cardiovascular events in corresponding groups was lower than in patients with normal to moderately reduced renal function. In these patients, the equivalent CAC score was associated with a higher risk compared to patients with severe impaired renal function (Table 5). This could be derived from severe CKD specific calcifications in the media (Mönckenberg’ sclerosis), that might not contribute to additional cardiovascular risk at a similar level of conventional atherosclerosis.

It is known that the incidence of major cardiovascular events increases with higher CAC scores [18]. Therefore, in patients with severe CKD, the CAC score slightly overestimates the future cardiovascular risk compared to patients with normal to moderately reduced renal function because specific calcifications in the media are included in the CAC. In these patients with severe CKD, the CAC score is a valuable tool for cardiovascular risk assessment and the risk stratification for future events, especially with its high negative predictive value in patients with a score of 0. As described in other populations, in patients with severe CKD, the risk for cardiovascular events increased consistently with CAC. Therefore, it can be used to guide risk reduction therapy, for example lipid lowering with statin therapy. It is reasonable to assume that the number needed to treat to prevent cardiovascular event decreases with higher CAC scores, as shown in a previous study [19].

Firstly, this is a single center study from Germany consisting of patients sent to our institution for a cardiological examination. Therefore, this cannot be considered an unselected population. Still, it can be regarded as a homogenous population without signs of CAD on study entry, as ECG, stress ECG, and echocardiography showed normal findings.

Secondly, due to the concentration of patients with CKD, the number of patients in this study is relatively small compared to other studies on coronary artery calcifications and cardiovascular events. However, the distribution of conventional risk factors is consistent with a typical population with cardiovascular risks in Europe. All patients received treatment according to current guidelines.

Thirdly, cardiac death was confirmed by autopsy in only 27% of patients, nevertheless in 66%, a prior MI was proven by ECG and laboratory testing. In 7%, clinical symptoms indicated a fatal infarction. So, it is reasonable to assume a cardiac death occurred in these patients.

In our asymptomatic high-risk population, we could demonstrate the importance of values for future cardiovascular events. A CAC of 0 in this high-risk group is associated with an excellent prognosis for event free progression. We could also demonstrate that risk prediction with CAC is still possible and independent in patients with severe CKD even if the CAC in patients with severe renal dysfunctions overestimated the risk for future cardiovascular events in comparison to patients with normal to moderately reduced renal function.

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

MG: Writing original draft, Data curation, Formal analysis. KL: Writing original draft, Data curation, Formal analysis. MW: Data curation. FZ: Data curation. AB: Statistics, Supervision, Conceptualization. MW, FZ, AB have been involved in drafting the manuscript and revised it critically for important intellectual content; All authors gave final approval of the version to be published. All authors have participated sufficiently in the work to take public responsibility for appropriate portions of the content; All authors agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

The research protocol was authorized by the local Clinical Institutional Review Board (Ethics commitee of the LMU Munich Germany, ethics approval number is 237-09) and is in accordance with the declaration of Helsinki. All patients or their families/legal guardians gave their written informed consent before they participated in the study.

Not applicable.

This research received no external funding.

The authors declare no conflict of interest.