Low iPTH Can Predict Vascular and Coronary Calcifications in Patients Undergoing Peritoneal DialysisKim S.C. · Kim H.W. · Oh S.W. · Yang H.N. · Kim M.-G. · Jo S.-K. · Cho W.Y. · Kim H.K.
Division of Nephrology, Department of Internal Medicine, Korea University Anam Hospital, College of Medicine, Seoul, Korea Corresponding Author
Background: There is substantial evidence that low levels of serum intact parathyroid hormone (iPTH) are associated with vascular calcium deposition and subsequent increased cardiovascular risk in chronic kidney disease patients. The purpose of this study was to determine the relationship between the serum iPTH level, and vascular and coronary artery calcifications (VCs, CACs) and arterial stiffness in peritoneal dialysis (PD) patients. Methods: In this cross-sectional study, 93 PD patients were included. VCs, CACs and arterial stiffness were measured by simple X-rays of the hands and pelvis, multi-slice coronary CT and brachial-ankle pulse wave velocity (BaPWV). Results: Patients were divided into 3 groups according to iPTH levels. The prevalence of severe VCs (VC score ≧3) was highest in the low iPTH group (<150 pg/ml). In multivariate analysis, the presence of diabetes mellitus and a low iPTH were shown to be significant risk factors for severe VCs. In addition, a simple VC score of ≧1 was a significant variable for predicting severe CACs (CAC score ≧400). Conclusion: Low iPTH and the presence of diabetes mellitus are thought to be independent risk factors for predicting VCs. VCs determined by simple X-ray can further predict the coexistence of CACs that ultimately might contribute to increased cardiovascular risk in PD patients.
Copyright © 2010 S. Karger AG, Basel
In patients with chronic kidney disease (CKD), >50% of mortality is attributed to cardiovascular diseases. When compared with age-matched populations, the mortality rate from cardiovascular events is 10–100 times higher in CKD patients than healthy patients [1,2]. In addition to traditional cardiovascular risk factors, uremia, chronic inflammation, vascular calcifications (VCs; representing CKD-mineral bone disorder [MBD]), and arterial stiffness have recently been suggested to contribute to increased cardiovascular morbidity and mortality .
Although clinical and experimental studies have demonstrated that VCs are increased in patients with high- or low-turnover bone disease, the precise mechanisms of the underlying bone turnover state or the intact parathyroid hormone (iPTH) level leading to VCs is largely unknown . In the past, VCs were considered to be the result of passive accumulation of calcium due to an increase in phosphate or calcium phosphate products, or degenerative changes. However, recent studies have shown that VCs are the result of an active cell-regulated process with the possibility of transformation of vascular smooth muscle cells into osteocytes or chondrocytes [5,6].
Arterial stiffness has also been reported to be increased in patients with CKD, and a recent study from the Framingham Heart Study showing a correlation between arterial stiffness and albuminuria suggested that arterial stiffness in CKD patients might also play an important role in increased cardiovascular morbidity and mortality .
The aim of this study was to examine the association between the iPTH level, VCs, coronary artery calcifications (CACs), and arterial stiffness, and also to identify factors that independently predict the development of VCs in peritoneal dialysis (PD) patients.
Patients and Methods
In this cross-sectional study we included 93 stable PD patients being followed in the Division of Nephrology of the Korea University Anam Hospital with >6 months of PD. Patients with a history of parathyroidectomy, malignancy, or advanced liver disease were excluded. Patient baseline characteristics, including gender, age, body weight, height, duration of PD, and history of hypertension and diabetes were recorded. The mean values of serum albumin, total cholesterol, alkaline phosphatase, calcium, phosphate, HbA1c, and hemoglobin levels that had been measured every month for 18 months were used.
To examine VCs, we used the method described by Adragao et al. . Plain radiographic films of both hands and the pelvis were obtained when the patients were enrolled. The pelvic radiographic film was arbitrarily divided into 4 sections by 2 imaginary lines (a horizontal line over the upper limit of both femoral heads and a median vertical line over the vertebral column). For each hand, the film was divided by a horizontal line over the upper limit of the metacarpal bones. Only the linear calcifications in the iliac, femoral, radial, and digital arteries were counted. The presence of calcifications in each section was counted as a score of 1 and the absence of calcifications was counted as a score of 0; the simple VC score (SVCS) was the sum of all the sections, ranging from 0 to 8. Adragao et al.  found that a VC score of ≧3 was an independent predictor of cardiovascular mortality, cardiovascular hospitalizations, and fatal or non-fatal cardiovascular events. According to this result, we defined severe VCs as a SVCS of ≧3. Analysis of VCs was performed by one experienced radiologist blinded to the patient information.
We took multi-slice coronary CT (MSCT) only in patients who did not have residual renal function and who agreed. The CAC score is expressed as modified Agatston units . Rumberger et al.  reported that a CAC score of ≧400 indicates severe and extensive atherosclerotic disease. Patients with calcification scores in this range are very likely to have obstructive coronary artery disease, with a high risk of developing symptomatic myocardial ischemia . According to this result, we defined severe CACs as a CAC score of ≧400.
Brachial-ankle pulse-wave velocity (BaPWV) was measured using a Colin non-invasive vascular screening device (Colin, Co., Ltd., Courbevoie, France), which simultaneously records the bilateral arm and ankle blood pressures, the pulse volume of the brachial and posterior tibial arteries, the heart sounds, and an electrocardiogram.
All analyses and calculations were performed using SPSS, version 12.0 (SPSS, Inc., Chicago, Ill., USA). Data are presented as the means ± standard deviation (SD) or median (interquartile range), according to the distribution. Comparisons of 3 PTH groups were performed by one-way ANOVA using the Tukey test or Kruskal-Wallis tests, according to the distribution. Pearson χ2 tests were used to analyze nominal data. Comparisons of 2 PTH groups were performed by independent-samples t test or Mann-Whitney tests, according to the distribution. In multivariate analysis, a binary logistic regression model was used. Pearson correlation, Spearman rank correlation, and linear regression analysis were used to analyze the factors assumed to be correlated with BaPWV. A p value of <0.05 was considered statistically significant.
The mean age of the patients was 54.5 ± 12.7 years, and the mean duration of PD was 50.4 ± 41.2 months. Thirty-nine (41.9%) patients had a history of diabetes, and 81 patients (87.1%) had hypertension. According to the mean iPTH level, the patients were divided into 3 groups: 28 patients were in the low group (iPTH <150 pg/ml); 35 patients were in the intermediate group (iPTH 150–400 pg/ml), and 30 patients were in the high group (iPTH >400 pg/ml). Measurement of the BaPWV was determined in 82 patients, and a MSCT was performed on 30 patients. The baseline characteristics according to the iPTH level are shown in table 1.
|Table 1. Baseline characteristics according to the iPTH level|
Patients who belonged to the highest tertile of iPTH had a significantly longer duration of PD, higher serum alkaline phosphatase, calcium, and phosphate levels, and an increased left ventricular mass index. However, the proportion of patients with severe calcifications (SVCS ≧3) was highest in the low iPTH group (39.3%), followed by the high iPTH group (7 patients, 23.3%) and the intermediate iPTH group (4 patients, 11.4%); this difference was statistically significant between the 3 groups (p = 0.035). However, neither the CAC score measured by MSCT nor the BaPWV (cm/sec) was significantly different according to the iPTH level. We also compared the net fluid removal rate according to different iPTH tertiles and found that there was no significant difference (table 1).
Patients with severe VCs had a significantly higher prevalence of diabetes and higher HbA1c levels. They also had significantly lower serum albumin levels, a higher BaPWV, and the proportion of patients with a low iPTH was significantly higher (table 2). In addition, the CAC score in patients with severe VCs was significantly increased. To identify the risk factors that independently predict severe VCs, a binary logistic regression model was applied for multivariable analysis. A SVCS of ≧3, age, cholesterol, HbA1c, albumin, the presence of diabetes, and the iPTH status were included as variables in logistic regression analysis (forward method). A low iPTH level (OR 5.75, p = 0.023) and the presence of diabetes (OR 31.3, p < 0.001) were independent risk factors for severe VCs (SVCS ≧3) in PD patients (table 3).
|Table 2. Characteristics according to the simple vascular calcification score|
|Table 3. Binary logistic regression analysis of risk factors associated with vascular calcification score of ≧3 (forward method)|
Despite the small number of patients who underwent MSCT for evaluation of CACs (n = 30), we analyzed the patient characteristics. Patients who had MSCT had a longer PD duration (62 vs. 22 months, p = 0.001), lower hemoglobin levels (10.1 vs. 10.5 g/dl, p = 0.007), and higher phosphate levels (5.53 vs. 4.84 mg/dl, p = 0.015) than patients for whom MSCT was not performed. However, the other variables showed no statistically significant differences. Patients were divided according to a CAC score of 400 into 2 groups. Patients with a CAC score of ≧400 were older, had a significantly higher prevalence of diabetes, more VCs (SVCS ≧1 and ≧3), and higher BMI and HbA1 levels, while the diastolic blood pressure was lower, indicating increased pulse pressure (table 4). A binary logistic regression model was applied for multivariate analysis regarding the risk factors affecting severe CACs (CAC score ≧400). The presence of diabetes, a SVCS of ≧1, age, BMI, HbA1c, and diastolic blood pressure were included as variables in logistic regression analysis (forward method). A SVCS of ≧1 (OR 76.0, p = 0.001) was an independent risk factor for severe CACs (CAC score ≧400; table 5).
|Table 4. Characteristics according to a CAC score of ≧400|
|Table 5. Binary logistic regression analysis of factors associated with a CAC score of ≧400 (forward method)|
We identified 31 patients who had a urine volume of >100 ml/day and examined the correlation between urine volume and VCs, or iPTH levels. We found no significant correlation between them, suggesting residual renal function might have no effect on VCs (data not shown).
In this study, we observed the following: (i) the prevalence of VCs in PD patients was highest in the lowest tertile of the iPTH group, followed by the highest and intermediate tertiles; (ii) patients with severe peripheral VCs (SVCS ≧3) had lower plasma albumin levels, stiffer arteries, higher CACs, and the probability of falling into the lowest tertile of the iPTH group; (iii) diabetes mellitus and a low iPTH level were independent predictors of VCs, and (iv) CACs had a good correlation with peripheral VCs.
VCs and CACs frequently complicate patients undergoing PD and are associated with increased cardiovascular morbidity and mortality [8,11,12]. A recent report showed that cardiac valve calcifications are also associated with chronic inflammation, malnutrition, and uremia . In addition to traditional cardiovascular risk factors, bone turnover status, associated altered plasma iPTH levels, calcium, and phosphate levels might contribute to VCs and increase subsequent cardiovascular risks [14,15].
Among CKD-MBD, low-turnover bone disease, or adynamic bone disease, characterized by a low iPTH level is more likely to be associated with soft tissue or VCs, partly due to the decreased calcium phosphate-buffering capacity of bone. However, there have been many controversies about the association of the iPTH level and VCs. While London et al.  first reported that a high arterial calcification score was associated with bone histomorphometry suggestive of low bone activity and adynamic bone disease, Coen et al.  found no prominent association between low iPTH level and the severity of coronary calcium deposits in HD patients. Furthermore, they observed more frequent hypercalcemia and hyperphosphatemia in patients with very high iPTH levels, suggesting that high iPTH could be the strongest predictor of VCs. In our study, we also observed that plasma calcium and phosphate levels were significantly higher in patients in the highest tertile of iPTH. However, severe VCs were most prevalent in the lowest tertile of iPTH, followed by the highest and intermediate tertiles, showing a U-shape relationship between iPTH level and VCs. These results suggest that factors other than the plasma level of calcium phosphate or iPTH play an important role in the development of VCs. In addition to uremia, chronic inflammation, which is known to be involved in abnormal VCs, and variations in the plasma levels of calcification inhibitors, such as fetuin-A, osteopontin, bone-morphogenic protein, or fibroblast growth factor that potentially affect the phenotypic transformation of vascular smooth muscle cells into osteocytes or chondrocytes, are thought to contribute to VCs.
Coen et al.  reported that PTH serum levels of <150 pg/ml are associated with adynamic bone disease in at least half of the cases and also with mixed osteodystrophy or normal bone in another 50%. These results are in agreement with other publications [19,20,21], while according to the KDOQI guideline, PTH serum levels of >400 pg/ml are known to be usually associated with high-turnover bone disease. Therefore, we defined an iPTH level of <150 as the low PTH group and an iPTH level of >400 as the high PTH group.
In our study, severe VCs were observed in 23.7% of the PD patients, and patients with severe VCs had a significantly higher prevalence of diabetes (90.9 vs. 26.8%), lower plasma albumin levels, and higher BaPWV than patients without severe VCs. In addition, >50% of patients with severe VCs fell into the category of the lowest tertile of iPTH, further suggesting the strong association between lower iPTH levels and VCs. The method for quantifying VCs which we used in this study was first suggested in 2004 by Adragao et al.  who measured the iliac, femoral, radial, and digital arteries. These are all muscular arteries that are prone to developing medial calcifications, which are known to be more prevalent in patients with CKD [5,22].
Regardless of the causative factors, VCs and/or increased arterial stiffness are commonly observed in patients undergoing PD and are thought to be important in mediating increased cardiovascular risks. A recent prospective observational study by Adragao et al.  confirmed that a higher SVCS, pulse wave velocity, and pulse pressure might be used to predict mortality risk.
CACs in PD patients are associated with the coronary atherosclerotic plaque burden and increase the risk of plaque instability or myocardial infarction . The CAC score measured by MSCT in patients with severe VCs was also significantly higher than in patients without VCs. The number of patients with severe CACs (CAC score ≧400) was 6 (66%) among patients in the severe VC group (SVCS ≧3), but there were no patients with severe CAC in the non-severe VC group (SVCS <3). This difference was still evident if a lower VC score standard (SVCS ≧1) was used (88.9 vs. 9.5%). These findings suggest that peripheral VC can reliably predict CAC status and can represent the risk of further cardiovascular events, such as myocardial infarction. This is clinically important because the possible deleterious effects of contrast for coronary angiography or MSCT on loss of residual renal function in these patients sometimes outweigh the benefit of these diagnostic maneuvers, and therefore careful decision should be given before proceeding with MSCT or conventional coronary angiography in terms of preserving residual renal function.
Despite several meaningful findings, there are a number of limitations to our study. First this was a cross-sectional study enrolling a relatively limited number of patients in a single center. The cross-sectional nature of our study did not allow identification of other factors associated with VCs. Second, analysis of the exact dose of calcium-containing salts or active vitamin D treatment that might affect bone status, iPTH, or plasma calcium and phosphate levels was not performed. However, in our center, strict guidelines to prescribe calcium-containing salts or sevalamer are followed, according to the Kidney Disease Improving Global Outcome guidelines.
In conclusion, VCs are prevalent in PD patients and show a U-shaped relationship with the iPTH level, being highest in the lowest tertile of the iPTH level. The presence of diabetes and being in the lowest tertile of iPTH were also independent predictors of VCs. CACs were strongly associated with peripheral VCs. A low serum iPTH level and high SVCS can be used as a noninvasive tool for predicting CACs. More studies enrolling a larger number of patients are needed to better understand the pathogenesis of VCs and the subsequent development of treatment targets.
Prof. Won Yong Cho, MD, PhD
Division of Nephrology, Department of Internal Medicine
Korea University Anam Hospital
5Ka, Anam-Dong, Sungbuk-Ku, Seoul 136-705 (Korea)
Tel. +82 2 920 5599, Fax +82 2 927 5344, E-Mail email@example.com
S.C. Kim and H.W. Kim contributed equally to this work.
Received: December 22, 2009
Accepted: April 28, 2010
Published online: August 6, 2010
Number of Print Pages : 7
Number of Figures : 0, Number of Tables : 5, Number of References : 23
Nephron Clinical Practice
Vol. 117, No. 2, Year 2011 (Cover Date: January 2011)
Journal Editor: El Nahas M. (Sheffield)
ISSN: 1660-2110 (Print), eISSN: 1660-2110 (Online)
For additional information: http://www.karger.com/NEC