Prognosis and risk factors of cardiac valve calcication in Chinese end-stage renal disease patients

Cardiac valve calcication (CVC) was an important risk factor for cardiovascular complication. But the prevalence, clinical features and risk factors for CVC in ESRD patients were not fully clear at present. In this study, we explored the possible risk factors and clinical characteristics of CVC happened in Chinese ESRD patients. We conducted a retrospective case-control study on 433 cases of ESRD patients who received maintenance dialysis (MHD) for at least 3 months in the First Aliated Hospital of Chongqing Medical University from October 2014 to December 2015. 93 patients were conrmed to happen cardiac valve calcication (CVC) by echocardiography, and 200 patients without CVC in the same period as control, matched with age and gender. The demographic data, clinical characteristics, and laboratory parameters of the two groups were analyzed.


Introduction
Epidemiological and clinical studies demonstrated that calci cation in vascular or tissue was common, and cardiovascular complication was the leading cause of mortality in patients with end-stage renal disease (ESRD). Braun et al [1] reported that over one half adult hemodialysis patients had evidence of cardiac valve calci cation (CVC), which happened more early and accelerated more dramatically [2].
Growing studies showed CVC was tightly associated with cardiovascular complication in ESRD patients.
Moreover, cardiovascular complication signi cantly shorten the life span of the patients with ESRD [3], which was an important predictor for all-cause mortality and cardiovascular mortality [4].
The pathophysiology of CVC, though not entirely determined, is de nitely multifactorial. Inaddition to conventionalriskfactors including age, gender, smoking, and primary disease such as hypertension or diabetes mellitus, ESRD patients have some unique risk factors for accelerated soft tissue calci cation. In ESRD patients, metabolic disorders characterized by abnormal calcium and phosphorus levels may play a key role [4], CVC and vascular calci cation are associated with clinical syndromes because of chronic kidney disease-mineral bone metabolic disorders (CKD-MBD). However, other factors which conduce to CVC formation are not fully determined.
The aim of the study was to investigate the epidemiological characteristics of CVC, the risk factors of CVC, the relationship between CVC and prognosis in Chinese ESRD patients.

Study Subjects
We conducted a retrospective cohort study and cross-sectional study that included 433 ESRD patients who had received more than 3 months of MHD in the dialysis center of the First A liated Hospital of Chongqing Medical University between October 2014 and December 2015. Exclusion criteria for the study included ESRD patients with hemodialysis vintage less than 3 month, aging more than 75 year-old, combination with congenital heart disease, infective endocarditis, rheumatic heart disease and these without complete data. 293 patients were rolled in this study in the end, including 160 male and 133 female, aged from 43 to 74, the average age was 64.0±7.2 year-old. 93 ESRD patients con rmed CVC by echocardiography as CVC group. 200 ESRD patients without CVC, matching age and gender, were choosed as control group in the same period ( Figure 1). The etiology of ESRD were shown in Figure 2, including 99 cases of primary gomerulonephritis, 96 cases of diabetes mellitus, 56 cases of primary hypertension, 13 cases of polycystic kidney, 12 cases of obstructive nephropathy, and 11 cases of interstitial nephritis ( Figure 2). All patients underwent hemodialysis twice or three times a week.
The study was approved by the Ethics Committee of the First A liated Hospital of Chongqing Medical University and written informed consent was obtained from all the patients.

Clinical Data Collection
General information including age, gender, cigarette smoking, duration of hemodialysis, diabetes, hypertension and clinical manifastations including cardiovascular complication and cerebrovascular diseases (CVD) were collected. Systolic, diastolic and pulse pressures were measured before hemodialysis at 2-wk intervals were averaged for 3 months.
Pre-dialysis laboratory variables including hemoglobin, albumin, calcium, phosphorus, intact parathyroid hormone (iPTH), serum lipid and high sensitivity C-reactive protein (hs-CRP) were collected when the patients were rolled in the study. Calculating corrected serum calcium according to the serum calcium level. Calcium was corrected for serum albumin levels<40g/L as follows: Corrected calcium mmol/l = calcium mmol/l +0.2 × [4 -serum albumin(g/dl)] [5].

Echocardiography
All subjects underwent echocardiography with Philips IE33 echocardiograph equipped with a S5-1 PureWave array probe (frequency 1.7/3.3MHz) to veri ed the presence of CVC. CVC was de ned as bright echoes of more than 1 mm on one or more cusps of the aortic valve or mitral valve or mitral annulus [6].

Computer tomography scan
To determine vascular calci cation, all ESRD patients underwent a chest and abdominal computer tomography (CT) scan on a SOMATOM Perspeative scanner, few patients did the coronary artery CT. A calci ed plaque in coronary artery and aorta was considered present if CT value was more than 130 scores [7].

Statistical analysis
Statistical analysis were performed using SPSS 21.0 statistical software. Data distribution was tested for normality using Kolmogorov-Smirnoff test. All numeric data with normal distribution was shown as means ± SD, analyzed by Student's t-test. Numeric variables with skew distribution were expressed as median (interquartile range), compared by Mann-Whitney U test. Categorical variables were expressed as percentages, analyzed by chi-square test. The relation between CVC and its associated complication was evaluated by using Spearman's correlation analysis. The risk factors associated with CVC were identi ed by using Binary logistic regression analysis. P < 0.05 was considered statistically signi cant.

Prevalence of cardiac valve calci cation in ESRD patients
In our single-center study, of all 433 ESRD patients, there were 30.3% ESRD patients with cardiac valve calci cation (CVC). Of 93 ESRD patients with CVC, the most common was aortic valve calci cation with prevalence of 60.2%, followed by mitral valve calci cation with prevalence of 26.9%, and 12 (12.9%) ESRD patients occurred to both mitral and aortic valve calci cation.

Basic clinical characteristics in ESRD patients with CVC
Of 293 ESRD patients enrolled in this study, 93 ESRD patients with CVC included 55 male and 38 female, aged from 44 to 74, the average age was 64.0 ± 7.0 year-old. 200 ESRD patients without CVC included 105 male and 95 female, aged from 43 to 74, the average age was 64.0 ± 7.0 year-old. Age and gender matched between the two groups.
We explored the risk factors for CVC in ESRD patients, and found no difference of smoking history between the two groups (Table 1). However, we found that the prevalence of diabetes mellitus and hypertension in ESRD patients were signi cantly higher in CVC patients than that of non-CVC (nCVC) patients, and the same as systolic pressure, diastolic pressure, pulse pressure and dialysis vintage (P < 0.05) ( Table 1).

Nutrition and in ammation related factors in ESRD patients with CVC
At the same time, we analyzed the nutrition and in ammation related biochemical indicators in ESRD patients with CVC, the results showed that the hemoglobin, albumin, and HDL were signi cantly lower in ESRD patients with CVC (P < 0.05). However, no statistical differences of TC, LDL, TG, and hs-CRP were found between the two groups ( Table 2). Mineral-bone metabolism associated parameters in ESRD patients with CVC As alterations of mineral-bone metabolism were associated with increased risk of CVC. In this study, we analyzed the changes of serum mineral-bone metabolism associated parameters in ESRD patients with CVC, and found that corrected serum calcium, and calcium-phosphorus product in CVC were higher than that in nCVC patients(P < 0.05). No statistical differences in serum phosphorus and iPTH were found between the two groups (Table 3). All factors with P < 0.10 at univariate analysis were included in a binary logistic regression model with backward selection. As shown in Table 4, longer dialysis vintage (P = 0.006, OR = 2.25), higher corrected serum calcium (P = 0.046, OR = 2.04), the prevalence of diabetes (P = 0.037, OR = 1.81), and higher pulse pressure (P < 0.001, OR = 3.22) were independent risk factors for CVC in ESRD patients, while higher serum albumin (P = 0.047, OR = 0.54) was protective factor.  In this article, we found that there were a great number of ESRD patients with different degree calci cation in aortic, and coronary artery (Fig. 3). Moreover, in CVC patients, the incidence of aortic artery calci cation, coronary artery calci cation, and both aortic and coronary artery calci cation was higher than that in nCVC patients (P < 0.001) ( Table 5). The incidence of aortic calci cation in ESRD patients with CVC was nearly 1.44 times as much as nCVC, and the prevalence of coronary artery calci cation was 1.74 times, similarly, both aortic and coronary artery calci cation was 1.87 times. Cardial and cerebrovascular complications in ESRD patients with CVC We further analyzed the relationship between cardiovascular complications and CVC. We found that the prevalence of cardial-cerebrovascular complications increased signi cantly in ESRD patients with CVC. The incidence of arrhythmia, heart failure, and coronary heart disease (CHD) in CVC patients was 1.8, 2.5, and 1.4 times compared with nCVC patients, respectively (P < 0.05), and atrial arrhyttmia was the most common arrhythmia. However, no difference of cerebrovascular complications(CVD) was found between the two groups. The risk of all-cause mortality in CVC group was 3.7 times higher than that of nCVC group (P < 0.05). Of 16 died patients, 11 died of cardial-cerebrovascular complications, 5 died of severe infection during the followed-up from 3 months to 10 years (Table 6). We further used Spearman's correlation analysis to determine the relation between CVC and cardiovascular complications in ESRD patients. Our results showed that CVC in ESRD patients were obviously associated with aortic calci cation, coronary artery calci cation, heart failure, and all-cause mortality (P < 0.05). However, no relationship between CVC and arrhythmia or CHD was found (Table 7).

CVC was common complication in ESRD patients
Vascular or tissue calci cation is a highly prevalent condition at all stages of chronic kidney disease (CKD), and increasingly recognized to be a common complication in ESRD patients, which represents an important part of the CKD-MBD complex. Documents [1,8]reported that two thirds of the adult hemodialysis patients had electron beam computed tomographic evidence of coronary artery calci cation (CAC) and that over half had cardiac valve calci cation. Consistent with the report of literature, in this study, the average annual incidence of CVC was 30.3% in our dialysis center, and aortic valve calci cation was the most frequent site, accounting for 60.2%. This was followed by mitral valve calci cation, with prevalcence of 26.9%, and 12.9% ESRD patients had both mitral and aortic valve calci cation.

Risk factors for CVC in ESRD patients
It has been thought that valve calci cation and atherosclerotic calci cation have certain common factors and a similar pathogenesis. The traditional risk factors for CVC included hypertension [11]and diabetes. In this study, rstly, we found that ESRD patients with CVC had a longer history of hypertension and a higher systolic pressure, diastolic pressure and pulse pressure, and regression analysis con rmed pulse pressure was the independent risk factor for CVC. Study [12]had found that the stability of hemodynamics was very important to maintain normal cardiac valve function. In ESRD patients, the way of hemodialysis, ultra ltration volume, and blood ow during hemodialysis could make blood pressure uctuation and hemodynamic abnormality, which may lead to dysfunction of endothelial cells and stromal cells of cardiac valve and in ammatory, remodeling and eventually developed CVC. Furthermore, increased pulse pressure caused by decreased artery diastolic and compliance after vascular calci cation, was an independent risk factor for CVC in turn.
Much study had reported that vascular calci cation was more serious in the ESRD patients with diabetes, and impaired fasting glucose and diabetes could be used as predictors of vascular calci cation in ESRD patients [13]. In this study, we also found a higher prevalence of CVC in ESRD patients with diabetes. High blood glucose and hyperinsulinemia, in ammatory state, glycosylation product accumulation, and endocrine disorders of diabetes, and so on, all could damage endothelial cells and stromal cells of cardiac valve, eventually led to CVC [14].
It was important to nd that even though there were certain common factors to the pathogenesis of cardiac valve calci cation and atherosclerotic calci cation, however, the pathogenesis in the ESRD patients was likely different from the calci cation observed in the general population. For example, the calci cation score in ESRD patients was not only substantially higher than age-matched and gendermatched patients with coronary artery disease but also progressed rapidly within short period. Furthermore, When CVC was initially present, the speed, and the degree of calci cation increased as age or dialysis vintage [15]. In this study, we also con rmed dialysis vintage was an independent risk factor in ESRD patients with CVC. Arjona Barrionuevo et al [16]had con rmed CVC and vascular calci cation increased as dialysis vintage in ESRD patients including diabetic nephropathy. Goldsmith et al [17]observed 38 MHD patients with 10 ~ 25 years, vascular calci cation prevalence increased from 39% at dialysis onset to 92% after the average dialysis vintage for 16 years, with a mean onset 9.7 years after initiating dialysis.
ESRD patients receiving MHD were shown to be a MICA syndrome including malnutrition, in ammation, atherosclerosis, and calci cation, and serum albumin re ected the nutritional status in ESRD patients [18]. However, the relationship between nutritional conditions, in ammation and CVC in ESRD patients was still unclear. In this study, we also found that ESRD patients has a lower level of hemoglobin and serum albumin, with increased vascular calci cation, which was consistent with the prior reports [18]. Interestingly, although serum lipid played an important role in the process of cardial-cerebrovascular disease development, our data showed no signi cant difference of serum TC, LDL, and TG was found between in ESRD patients with and without CVC. In fact, in patients with ESRD, peripheral serum lipid levels did not re ect lipid deposition in the tissue, and the severity of arteriosclerosis [19]. It had been reported that lipid adjusting drugs didn't delay the process of vascular calci cation effectively in ESRD patients, reduced fatality rate because of cardial-cerebrovascular diseases [19,20]. However, we found that much lower peripheral serum HDL in ESRD patients with CVC, indicating HDL may be a protect factor for CVC. As we all know, HDL may reduce cholesterol deposition, resist atherosclerosis, to protect artery [18,21]. This nding was similar to the report of Japan hemodialysis population, but in contrast to the western hemodialysis population [22], the mechanism may be associated with generous beckgroud but need to be explored. Otherwise, although a state of chronic in ammation exited in ESRD patients received MHD because of biocompatibility in dialyser and dialysis uid [22], no statistical difference in hs-CRP between the two groups, showing that In ammation may not be a independent factor but a synergy with malnutrition for CVC in ESRD patients [18]. Further detection of in ammatory medium may be need to con rm the role of in ammatory in CVC in ESRD patients.
Calcium-phosphate metabolic disorders were most common complication in ESRD patients, and also be thought as an unique nontraditional risk factor of cardial-cerebrovascular calci cation. Hypocalcemia and/or hyperphosphatemia could stimulate excessive PTH secretion, caused secondary hyperparathyroidism, metabolic bone disease, artery, and other tissue metastatic calci cation, eventually formed a vicious circle, increased the mortality in ESRD patients [23]. Studies had found that high phosphorus, and calcium-phosphorus product were the independent risk factors of deaths in ESRD patients [24]. But the relationship between calcium, phosphorus metabolism disorders and CVC in Chinese ESRD patients was still to be further study. In this study, we found that adjusted serum calcium and the calcium-phosphorus product in CVC patients were signi cantly elevated compared with nCVC group, while PTH signi cantly decreased, and further logistic regression analysis showed that high adjusted serum calcium was an independent risk factor for CVC. Long-term exposure to high calcium-phosphorus environment could lead to cell phenotypic change, osteogenesis cell form, osteogenesis transcription factors mediate osteogenesis, which was similar to the formation of bone [25,26], eventually lead to cardiac valve, artery and the whole body metastatic calci cation. In ESRD patients, low PTH secondary from hypercalcemia, caused decreased bone calci cation because of the reduced osteoblast activity and more serious metastatic calci cation resulting from excess serum calcium. Oddly, this study didn't nd CVC was directly related with serum phosphorus, which needed to expand the sample size to further con rmed.
The predicted value of CVC in cardial-cerebrovascular complications in ESRD patients Cardial-cerebrovascular events were the most important cause of death in ESRD patients, which was due to the early-onset and progressive vascular calci cation in ESRD patients. Research had found that any part of the artery wall calci cation increased 3 ~ 4 times risk of death, and cardiovascular event [9]. However, in ESRD patients, the relation between CVC and artery calci cation remained undetermined. In this study, ESRD patients with CVC had more overt vascular calci cation, reached 90.1%, and the prevalence of aortic or coronary artery calci cation in CVC group was 1.44 ~ 1.87 times more than that of nCVC group. Furthermore, our data showed that ESRD patients with CVC had a higher prevalence of arrhythmia, heart failure, CHD, i.e. 1.4-2.5 times compared with nCVC group, and the risk of all-cause mortality was 3.7 times higher than that of nCVC. Spearman's correlation analysis further demonstrated that CVC was relevant for the aortic calci cation, coronary artery calci cation, heart failure, and all-cause mortality. Therefore, CVC may be a marker of vascular calci cation and, cardiovascular events. Calci ed cardiac valve reduced itself compliance, led to sclerosis or stenosis, decreased ventricular out ow, nally contributed to ventricular hypertrophy, resulting in myocardial ischemia, arrhythmia, heart failure, and even sudden death, which was particularly signi cant in those with both aortic and mitral valve calci cation [10]. Similarly, some documents [3,4] has indicated that CVC itself was a superior predictor of clinical outcomes in ESRD patients and closely associated with an increased risk of cardiovascular events and all-cause mortality. Taken together, all of these ndings indicated that CVC may re ect a poor clinical prognosis in ESRD patients.
In summary, in this retrospective case-control study, it showed that CVC was a common complication in ESRD patients, and a danger signal for severe atherosclerosis and cardiovascular events in ESRD patients. Diabetes, increased pulse pressure, longer dialysis vintage, and higher adjusted serum calcium were independent risk factors for CVC in ESRD patients, while higher serum albumin was a protective factor.

Declarations
Ethics approval and consent to participate The study was approved by the Ethics Committee of the First A liated Hospital of Chongqing Medical University .Written informed consent was obtained from all the patients, and all patients were identi ed by numbers, not their real names.

Consent for publication
Not applicable.

Competing interests
The authors declare that there is no competing interests.

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

Authors' contributions
All authors reviewed the nal version and approved the content. JQX performed data collection and helped with the data interpretation and manuscript writing. XMC helped with the design of the study and manuscript writing. CTL helped with the collection of biological samples. XGD designed the experiment, performed the literature search, and was responsible for manuscript writing and data analysis and interpretation. Figure 1 Flow chart of study design