The Circulating Inactive Form of Matrix Gla Protein Is a... : Clinical Journal of the American Society of Nephrology (original) (raw)
Cardiovascular diseases account for 50% of all deaths in a chronic kidney disease (CKD) setting (1). Vascular calcification (VC) in the media-intimal arterial layers contributes significantly to the greater mortality in this population (2,3). In fact, it has been repeatedly demonstrated that patients suffering from advanced CKD present VC to a greater extent than individuals with normal renal function (4–7). Although many factors may influence the occurrence and progression of this condition, the fact that 20% to 40% of the patients in most CKD cohorts (8–10) do not develop detectable VC (despite exposure to well-known environmental triggers, such as uremia, diabetes, and hyperphosphatemia) suggests that naturally occurring VC inhibitors have an important role in preventing this disease process.
Matrix Gla protein (MGP) is a 10-kD protein secreted by chondrocytes and vascular smooth muscle cells in the arterial media (11). It is the first protein known to act as a calcification inhibitor in vivo (12–14), probably by directly inhibiting calcium precipitation and crystallization in the vessel wall (15) and antagonizing bone morphogenetic protein-2 (which regulates osteoblast differentiation, and thus bone formation (16)). In particular, MGP only exerts its anticalcification activity after posttranslational γ-glutamyl carboxylation of five glutamate residues—a crucial activation step that depends on the availability of vitamin K (17,18). Moreover, there is evidence that to be adequately secreted and become fully active, MGP must undergo further phosphorylation (on three serine residues) (19).
Although recent studies have demonstrated an inverse association between total uncarboxylated MGP (ucMGP) and VC in hemodialysis patients (20–22), data for investigating an association with earlier stages of CKD are still lacking. Moreover, this measured total ucMGP level contains both phosphorylated and dephosphorylated fractions or fragmented MGP, which might exert distinct biologic functions. It is thus important to use an assay that enables the measurement of the circulating, inactive form; i.e., dephosphorylated ucMGP (dp-ucMGP).
Hence, the objective of the present study was to specifically determine (for the first time) plasma levels of dp-ucMGP in a cohort of patients at different stages of CKD and to evaluate the association between plasma dp-ucMGP on the one hand and aortic calcification and mortality on the other.
Materials and Methods
Patient Selection
Over an 18-month period (from January 2006 to June 2007), a total of 150 Caucasian, prevalent CKD patients were recruited from the Nephrology Department's outpatient clinic at Amiens University Hospital. All patients gave their informed, written consent. The study was approved by the local independent ethics committee and was performed in accordance with the ethical principles of the Declaration of Helsinki.
Included patients had to be older than 40 years, with a confirmed diagnosis of CKD (defined as ongoing hemodialysis or having two previous estimated Cockcroft and Gault (23) creatinine clearance values <90 ml/min per 1.73 m2 3 to 6 months apart). Stage 5D CKD patients had been receiving thrice-weekly hemodialysis for at least 3 months. The noninclusion criteria included the presence of chronic inflammatory disease, atrial fibrillation, complete heart block, abdominal aorta aneurysm, aortic and/or femoral artery prosthesis, primary hyperparathyroidism, kidney transplantation, or any other acute cardiovascular event in the 3 months before screening. The 107 patients who met all inclusion criteria and had available dp-ucMGP measurements were included in the present analysis. The 43 patients who were not included in the analyses did not differ from the analyzed patients in terms of age, sex, diabetes and cardiovascular disease status, body mass index, or CKD stage distribution.
Study Protocol
All patients were hospitalized for the day to perform the laboratory blood tests, BP measurements, a lateral lumbar x-ray, and multislice spiral computed tomography (CT) scan. Hence, for a given patient, all examinations were performed between 9 a.m. and 2 p.m. on the same day. Hemodialysis patients were preferentially seen on a dialysis-free day or, if this was not possible, the morning before the dialysis session. A patient interview focused on comorbidities and the personal disease history (especially any previous vascular events). The patients' medical records were reviewed to identify and describe any concomitant medications. Six patients were taking coumarins: three for thrombosis/stenosis of the internal carotid artery, two for deep venous thrombosis, and one after a cardiac valve replacement. For descriptive purposes, patients who reported the current or past use of insulin and/or oral hypoglycemic agents were considered to be diabetic. Previous cardiovascular disease was defined as a history of any of the following events: myocardial infarction, stroke, heart failure, angina pectoris, or surgical procedures for angina or coronary/peripheral artery disease (including percutaneous transluminal angioplasty).
Laboratory Procedures
Blood samples were collected on the same morning and before the other investigations were performed. Selected variables were measured after the samples had been frozen and stored at −80°C. Serum calcium, phosphate, cholesterol, hemoglobin, and creatinine (Scr) levels were assayed in an on-site biochemistry laboratory using standard autoanalyzer techniques (the Modular IIP® system from Roche Diagnostics, Basel, Switzerland). Serum-intact parathyroid hormone (iPTH 1 to 84) was determined in a chemiluminometric immunoassay (Liaison N-tact PTH CLIA®; Diasorin, Stillwater, MN). Plasma levels of dp-ucMGP were determined using a recently developed sandwich ELISA involving two anti-MGP monoclonal antibodies. An antibody directed against the nonphosphorylated sequence 3 to 15 in human MGP is used as capture antibody, whereas the antibody directed against the noncarboxylated sequence 35 to 49 is used as a detection antibody (VitaK BV, Maastricht, the Netherlands). The synthetic peptide dpMGP3-15-(AADO)-ucMGP35-54 was used as a standard. The intraassay and interassay variability values were 5.6% and 9.9%, respectively. Reference plasma dp-ucMGP values were determined in 77 healthy, age-matched controls (mean ± SD: 466 ± 257 pM; median: 430 pM; interquartile range: 305 to 563 pM). C-reactive protein (normal value: 2.87 mg/L, ranging from nondetectable to 3.0 mg/L), albumin, and cystatin C (CysC) levels were determined by laser nephelometry (BNProSpec; Siemens Healthcare, Dade Behring, Marburg, Germany). To describe the true GFR as closely as possible, the estimated GFR combining Scr and CysC measurements was calculated for all nondialyzed patients according to the recently published (24) equation: 177.6 × Scr−0.65× CysC−0.57× age−0.20 (×0.82 if female). Patients were then classified into CKD stages, according to the National Kidney Foundation's Kidney Disease Outcomes Quality Initiative (K/DOQI) guidelines (25). Because there were low numbers of patients in CKD stages 2 and 5, the study population was stratified into CKD stages 2 to 3, 4 to 5, and 5D for descriptive and analytical purposes.
Abdominal Aorta Imaging with Plain Radiography
A technique similar to that described by Kauppila et al. (26) was used to obtain images of the lower abdominal aorta and generate an aortic calcification score. All x-rays were reviewed by two investigators, and a consensus on the interpretation was reached in all cases (17).
Multislice Spiral CT
To quantify the presence and extent of aortic calcifications, each patient underwent a multislice spiral CT scan. All examinations were performed with a 64-detector scanner (Lightspeed VCT®; GE Healthcare, Milwaukee, WI). Detailed technical information on the procedure is provided elsewhere (27).
Survival
Death records were compiled prospectively by considering all patients included at least 20 months before the study end date (March 1, 2009). Each set of medical records was reviewed, and the cause of death was assigned by a physician on the basis of the available clinical information. For out-of-hospital deaths, the patient's general practitioner was interviewed to gain pertinent information on the cause.
Statistical Analyses
Data were expressed as the mean ± SD, median and range, or percentage, as appropriate. For analytical purposes, patients were divided into two groups according to the median [dp-ucMGP] (i.e., above and equal to or below 921 pM). The CKD stages were entered as a continuous variable (recoded from 1 to 3, with 3 meaning stage 5D patients). Intergroup comparisons were performed using a χ2 test for categorical variables and either t test or the Mann-Whitney test for continuous variables. For parameters presenting non-Gaussian distribution, logarithmic normalized values were considered in tests that assume normally distributed variables. Univariate and then multivariate linear regression analyses were used to select factors that were independently associated with [dp-ucMGP]. Thereafter, to see whether higher dp-ucMGP levels could have contributed to a higher aortic calcification score in the study cohort, we performed a multiple linear regression analysis with the aortic calcification score as the dependent variable, and [dp-ucMGP] and selected clinical and biochemical characteristics (CKD stage, age, previous cardiovascular disease, and systolic arterial pressure) as covariates. Because there was a good correlation between the aortic calcification score on x-ray and the aortic calcification score on CT (Spearman correlation coefficient = 0.873, P < 0.0001), the latter was used as the reference measurement in all further analyses. The Kaplan-Meier actuarial method was used to estimate overall survival for the median dp-ucMGP level, and the log-rank test was used to compare survival curves. Univariate and multivariate analyses of mortality were performed by using a Cox proportional hazards model of death as a function of the plasma dp-ucMGP level (either categorized by the median or as a continuous variable). Variables that were significantly associated with death in the univariate analyses were fed into predefined, noncumulative models in the multivariate analysis. In view of the small size of the present cohort, additional Cox regression analyses included a propensity score adjustment; this takes account of each individual's probability of exposure to measured, confounding variables (i.e., CKD stage, hemoglobin, and the aortic calcification score), as detailed elsewhere (28). A P value ≤0.05 was considered to be statistically significant. All statistical analyses were performed using SPSS software (SPSS Inc., Chicago, IL), version 13.0 for Windows (Microsoft Corp., Redmond, WA).
Results
As shown in Figure 1, plasma dp-ucMGP levels increased with CKD stage. In comparison with age-matched controls, this increase became statistically significant from CKD stage 4 onward. When considering only predialysis patients at CKD stages 2 to 5 (n = 67), we still observed an inverse, exponential association between plasma dp-ucMGP levels and the estimated GFR (Figure 2).
Plasma dp-ucMGP levels by CKD stage. *P < 0.001 versus CKD stages 2 and 3. The dotted lines indicate the upper and lower boundaries of the plasma dp-ucMGP range in age-matched controls.
Exponential relationship between plasma dp-ucMGP levels and the estimated GFR, for patients at CKD stages 2 to 5 (n = 67), r 2 = 0.268, P < 0.0001.
Table 1 presents the main clinical and biochemical characteristics and vascular measurements for the entire study cohort and as a function of the median dp-ucMGP level (921 pM). Patients with higher plasma dp-ucMGP levels were significantly older, more likely to be taking coumarins, and more frequently in late-stage CKD (stages 4, 5, and 5D). They also had lower hemoglobin levels, higher C-reactive protein levels, and a higher aortic calcification score. Univariate linear regression analyses with plasma dp-ucMGP as a dependent variable yielded essentially the same results (Table 2). Furthermore, a multivariate linear regression analysis including all variables significantly associated with plasma dp-ucMGP levels in a univariate analysis (Table 2) identified the CKD stage (P < 0.0001), hemoglobin levels (P < 0.0001), age (P = 0.002), and current coumarin use (P = 0.006) as independent factors (_r_2 = 0.432). This conclusion was not modified by excluding the six patients on coumarins from our analyses.
Main clinical characteristics and baseline data as a function of plasma dp-ucMGP levels
Univariate linear regression analysis: variables associated with plasma dp-ucMGP levels
With respect to VC, there was a positive, statistically significant association between the aortic calcium score and plasma dp-ucMGP levels (Figure 3). Other variables significantly associated with the aortic calcium score are listed in Table 3. Furthermore, in a multivariate linear regression analysis (including all clinical and biochemical characteristics significantly associated with the aortic calcium score in the univariate analyses), only age and plasma dp-ucMGP levels were independently associated with the aortic calcium score (Table 4). In an analysis restricted to predialysis patients, the independent association between the aortic calcification score and plasma dp-ucMGP levels remained significant after adjustment for the estimated GFR (n = 67; _r_2 = 0.125, P = 0.018).
Linear relationship between aortic calcification score and plasma dp-ucMGP levels (n = 101), r 2 = 0.143, P < 0.0001.
Univariate linear regression analysis: variables associated with the aortic calcification score on CT (logarithmic normalized)
Multivariate linear regression analysis: variables independently associated with the aortic calcification score on CT (logarithmic normalized)
During the follow-up period (mean duration: 802 ± 311 days; median: 846 days; range: 10 to 1129 days), 34 patients died (20 from cardiovascular events, 7 from infectious disease, and 7 from other causes). In a crude analysis, plasma [dp-ucMGP] > 921 pM predicted overall mortality (P = 0.006 in a log-rank test comparing the curves) (Figure 4). Further univariate and multivariate, noncumulative Cox regression analyses (Table 5) including other variables that had been shown to significantly influence mortality risk in this population (age, per 1-year increment: risk radio [RR] = 1.04; P = 0.013; hemoglobin, per 1-g/dL increment: RR = 0.73, P = 0.003; aortic calcification score, per 1-unit increment: RR = 1.23, P < 0.0001; CKD stage: RR = 1.84, P = 0.007) confirmed that the plasma dp-ucMGP level was an independent predictor of mortality, with attenuation after adjustment for the aortic calcification score (model 4). It is noteworthy that in cumulative models that considered either all these covariates together or included the calculated propensity score (to better adjust for confounders, as detailed in Materials and Methods), the association between plasma dp-ucMGP level and mortality lost its significance (model 5).
Kaplan-Meier estimates of overall mortality as a function of the median plasma dp-ucMGP level.
Multivariate Cox regression analysis of risk factors at baseline for all-cause mortality
Discussion
In the present study, we show for the first time that plasma dp-ucMGP levels augment progressively in a CKD setting. Moreover, renal dysfunction was the only independent determinant of higher circulating dp-ucMGP levels in our cohort. Importantly, these higher dp-ucMGP levels were independently associated with aortic calcification and showed a limited association with the overall mortality risk in the study cohort.
In contrast to our data, previous studies have reported significantly lower serum levels of ucMGP in dialyzed adult and pediatric patients than in healthy controls (20,29,30). Moreover, an inverse association between estimated GFR and ucMGP serum levels in patients with stable cardiovascular disease has been demonstrated (31). Recently, data from Cranenburg et al. (22) have evidenced that circulating levels of ucMGP were inversely correlated with the extent of coronary artery calcification in hemodialysis patients. It is important to note that these studies used an assay that did not differentiate between phosphorylated and dephosphorylated forms of ucMGP or fragments of MGP. In fact, it has been shown that ucMGP accumulates at arterial calcification sites (32), possibly by binding through its negatively charged phosphoserine residues (22). Consequently, up-regulation of MGP transcription in response to vascular stress (as well as absorptive forces in already calcified vessels) would influence ucMGP levels (31). Importantly, the present study featured a new assay that distinguishes the phosphorylation status of ucMGP, so that only the dephosphorylated uncarboxylated form of MGP was specifically measured; this may explain the apparently conflicting results with respect to previous assays. Indeed, it was recently demonstrated that the formation of phosphoserine residues actively contributes to MGP's calcification inhibition activity (19). Furthermore, because of the absence of phosphoserine residues, dp-ucMGP is thought to be released more easily into the circulation than phosphorylated MGP forms; the latter usually present extremely low circulating levels in the highly calcified CKD population, which may limit their use as a biomarker.
Importantly, our present findings indicate that elevation of serum levels of the inactive form of MGP parallels the progression of CKD and thus seems to be an important contributor to (or even a biomarker of) the severity of the VC frequently observed in the CKD setting. The high dp-ucMGP levels could reflect a low dietary intake of vitamin K, resulting in lower metabolic availability and thus less MGP γ-glutamyl carboxylation in patients with more advanced CKD. Additionally, it has been suggested that vitamin K status in hemodialysis patients is also influenced by the apolipoprotein E genotype. Lipoprotein-bound apolipoprotein E regulates the clearance of circulating vitamin K in the liver and in peripheral tissues (33). It is interesting to note that the vitamin K requirement in extrahepatic tissues (for γ-glutamyl carboxylation of MGP, for example) appears to be greater than that for the liver itself (for the modification of coagulation factors). Therefore, a subclinical vitamin K deficiency could impair the function of vitamin K-dependent calcification modulators (such as MGP and osteocalcin) without greatly affecting coagulation. This hypothesis is reinforced by the observation of low phylloquinone (vitamin K1) concentrations and undercarboxylation of osteocalcin (another vitamin K-dependent protein) in dialysis patients, relative to the general population (34). Moreover, coumarin vitamin K antagonists (which are used as coagulation inhibitors) have been identified as a risk factor for VC in the general population (35,36). In contrast, high vitamin K intake has been shown to regress warfarin-induced medial calcinosis in rats (37), and a high dietary intake of menaquinone (vitamin K2) has been associated with (1) a lower risk of coronary heart disease in the general population (38) and (2) reduced coronary calcification in postmenopausal women (39). In our study, patients on coumarins presented significantly higher levels of dp-ucMGP, which probably reflected an accumulation of this inactive form of MGP. In agreement with these observations, we recently reported (30) that plasma dp-ucMGP levels increase significantly with the use of vitamin K antagonists and, conversely, decrease after vitamin K supplementation. This finding indeed suggests that vitamin K has an important role in modulating dp-ucMGP levels. Additionally, a pilot study with vitamin K supplementation in CKD 5D patients clearly demonstrated a dose-dependent decrease in dp-ucMGP levels over time, with a return to baseline levels during the washout phase (Westenfeld et al., ASN Renal week 2008, TH-FC044). Alternatively, uremic toxins might impair the function of γ-glutamyl carboxylase. In such a case, a uremic state would affect MGP carboxylation even if vitamin K status were adequate. This latter hypothesis remains to be tested, however.
Our findings also demonstrate an association between plasma dp-ucMGP levels and overall mortality. However, this association was rather weak and was lost after cumulative adjustment for age and the calculated propensity score. This inconsistency may result (at least in part) from the small size of the study cohort and the complexity of the etiology of cardiovascular disease in a CKD setting. It seems also possible that plasma dp-ucMGP is simply a biomarker that reflects the VC burden. In theory, higher levels of the inactive form of MGP would initially favor VC. This high calcification burden would lead to a burst in MGP production, further exhausting vitamin K stores, limiting MGP activation, and resulting in additional VC. Additional research is needed to confirm and unravel the possible role of dp-ucMGP as a biomarker in CKD and to establish whether interventions that decrease plasma dp-ucMGP levels would result in better outcomes in a CKD setting.
Limitations of the present study include the small sample size and the fact that vitamin K intake and serum levels were not assayed. Moreover, plasma dp-ucMGP levels were measured at a single time point; at present, nothing is known about the time course of release of this biomarker and its persistence in the circulation. In contrast, one of the study's main strengths relates to the fact that this was the first specific analysis of circulating, inactive dp-ucMGP in a cohort of patients at different stages of CKD.
In conclusion, plasma dp-ucMGP levels augment progressively in a CKD setting and were directly correlated with the severity of aortic calcification in our study population. Plasma dp-ucMGP could thus be regarded as a surrogate marker for VC in CKD. Further experimental work on the potential relationship between vitamin K status and dp-ucMGP levels and interventional studies evaluating the influence of vitamin K repletion on dp-ucMGP levels, VC, and outcomes in the CKD setting are warranted.
Disclosures
None.
This study was funded by a grant from Amiens University Hospital (Projet Hospitalier de Recherche Clinique 2006/0100 (27/03/2006)). D.V.B. and F.C.B. received postdoctoral grants from the Picardy Regional Council/University of Picardy Jules Verne and postdoctoral scholarships from the National Council for Scientific and Technological Development, Brazil.
Published online ahead of print. Publication date available at www.cjasn.org.
References
1. Parfrey PS, Foley RN: The clinical epidemiology of cardiac disease in chronic renal failure. J Am Soc Nephrol 10: 1606–1615, 1999
2. London GM, Guerin AP, Marchais SJ, Metivier F, Pannier B, Adda H: Arterial media calcification in end-stage renal disease: Impact on all-cause and cardiovascular mortality. Nephrol Dial Transplant 18: 1731–1740, 2003
3. Lehto S, Niskanen L, Suhonen M, Ronnemaa T, Laakso M: Medial artery calcification. A neglected harbinger of cardiovascular complications in non-insulin-dependent diabetes mellitus. Arterioscler Thromb Vasc Biol 16: 978–983, 1996
4. Schwarz U, Buzello M, Ritz E, Stein G, Raabe G, Wiest G, Mall G, Amann K: Morphology of coronary atherosclerotic lesions in patients with end-stage renal failure. Nephrol Dial Transplant 15: 218–223, 2000
5. Ibels LS, Alfrey AC, Huffer WE, Craswell PW, Anderson JT, Weil R 3rd. Arterial calcification and pathology in uremic patients undergoing dialysis. Am J Med 66: 790–796, 1979
6. Braun J, Oldendorf M, Moshage W, Heidler R, Zeitler E, Luft FC: Electron beam computed tomography in the evaluation of cardiac calcification in chronic dialysis patients. Am J Kidney Dis 27: 394–401, 1996
7. Goodman WG, Goldin J, Kuizon BD, Yoon C, Gales B, Sider D, Wang Y, Chung J, Emerick A, Greaser L, Elashoff RM, Salusky IB: Coronary-artery calcification in young adults with end-stage renal disease who are undergoing dialysis. N Engl J Med 342: 1478–1483, 2000
8. Moe SM, O'Neill KD, Reslerova M, Fineberg N, Persohn S, Meyer CA: Natural history of vascular calcification in dialysis and transplant patients. Nephrol Dial Transplant 19: 2387–2393, 2004
9. Yuen D, Pierratos A, Richardson RM, Chan CT: The natural history of coronary calcification progression in a cohort of nocturnal haemodialysis patients. Nephrol Dial Transplant 21: 1407–1412, 2006
10. Barreto DV, Barreto Fde C, de Carvalho AB, Cuppari L, Draibe SA, Dalboni MA, Moyses RM, Neves KR, Jorgetti V, Miname M, Santos RD, Canziani ME: Phosphate binder impact on bone remodeling and coronary calcification–results from the BRiC study. Nephron Clin Pract 110: 273–283, 2008
11. Price PA, Urist MR, Otawara Y: Matrix Gla protein, a new gamma-carboxyglutamic acid-containing protein which is associated with the organic matrix of bone. Biochem Biophys Res Commun 117: 765–771, 1983
12. Munroe PB, Olgunturk RO, Fryns JP, Van Maldergem L, Ziereisen F, Yuksel B, Gardiner RM, Chung E: Mutations in the gene encoding the human matrix Gla protein cause Keutel syndrome. Nat Genet 21: 142–144, 1999
13. Teebi AS, Lambert DM, Kaye GM, Al-Fifi S, Tewfik TL, Azouz EM: Keutel syndrome: Further characterization and review. Am J Med Genet 78: 182–187, 1998
14. Luo G, Ducy P, McKee MD, Pinero GJ, Loyer E, Behringer RR, Karsenty G: Spontaneous calcification of arteries and cartilage in mice lacking matrix GLA protein. Nature 386: 78–81, 1997
15. Shanahan CM: Mechanisms of vascular calcification in renal disease. Clin Nephrol 63: 146–157, 2005
16. Zebboudj AF, Imura M, Bostrom K: Matrix GLA protein, a regulatory protein for bone morphogenetic protein-2. J Biol Chem 277: 4388–4394, 2002
17. Murshed M, Schinke T, McKee MD, Karsenty G: Extracellular matrix mineralization is regulated locally; different roles of two gla-containing proteins. J Cell Biol 165: 625–630, 2004
18. Shearer MJ: Vitamin K. Lancet 345: 229–234, 1995
19. Schurgers LJ, Spronk HM, Skepper JN, Hackeng TM, Shanahan CM, Vermeer C, Weissberg PL, Proudfoot D: Post-translational modifications regulate matrix Gla protein function: Importance for inhibition of vascular smooth muscle cell calcification. J Thromb Haemost 5: 2503–2511, 2007
20. Hermans MM, Vermeer C, Kooman JP, Brandenburg V, Ketteler M, Gladziwa U, Rensma PL, Leunissen KM, Schurgers LJ: Undercarboxylated matrix GLA protein levels are decreased in dialysis patients and related to parameters of calcium-phosphate metabolism and aortic augmentation index. Blood Purif 25: 395–401, 2007
21. Cranenburg EC, Vermeer C, Koos R, Boumans ML, Hackeng TM, Bouwman FG, Kwaijtaal M, Brandenburg VM, Ketteler M, Schurgers LJ: The circulating inactive form of matrix Gla Protein (ucMGP) as a biomarker for cardiovascular calcification. J Vasc Res 45: 427–436, 2008
22. Cranenburg EC, Brandenburg VM, Vermeer C, Stenger M, Muhlenbruch G, Mahnken AH, Gladziwa U, Ketteler M, Schurgers LJ: Uncarboxylated matrix Gla protein (ucMGP) is associated with coronary artery calcification in haemodialysis patients. Thromb Haemost 101: 359–366, 2009
23. Cockcroft DW, Gault MH: Prediction of creatinine clearance from serum creatinine. Nephron 16: 31–41, 1976
24. Stevens LA, Coresh J, Schmid CH, Feldman HI, Froissart M, Kusek J, Rossert J, Van Lente F, Bruce RD 3rd, Zhang YL, Greene T, Levey AS: Estimating GFR using serum cystatin C alone and in combination with serum creatinine: a pooled analysis of 3,418 individuals with CKD. Am J Kidney Dis 51: 395–406, 2008
25. National Kidney Foundation: K/DOQI clinical practice guidelines for chronic kidney disease: Evaluation, classification, and stratification. Am J Kidney Dis 39: S1–S266, 2002
26. Kauppila LI, Polak JF, Cupples LA, Hannan MT, Kiel DP, Wilson PW: New indices to classify location, severity and progression of calcific lesions in the abdominal aorta: A 25-year follow-up study. Atherosclerosis 132: 245–250, 1997
27. Barreto DV, Barreto FC, Liabeuf S, Temmar M, Boitte F, Choukroun G, Fournier A, Massy ZA: Vitamin D affects survival independently of vascular calcification in chronic kidney disease. Clin J Am Soc Nephrol 4: 1128–1135, 2009
28. Fitzmaurice G: Confounding: Propensity score adjustment. Nutrition 22: 1214–1216, 2006
29. Shroff RC, Shah V, Hiorns MP, Schoppet M, Hofbauer LC, Hawa G, Schurgers LJ, Singhal A, Merryweather I, Brogan P, Shanahan C, Deanfield J, Rees L: The circulating calcification inhibitors, fetuin-A and osteoprotegerin, but not matrix Gla protein, are associated with vascular stiffness and calcification in children on dialysis. Nephrol Dial Transplant 23: 3263–3271, 2008
30. Schurgers LJ, Cranenburg EC, Vermeer C: Matrix Gla-protein: The calcification inhibitor in need of vitamin K. Thromb Haemost 100: 593–603, 2008
31. Parker BD, Ix JH, Cranenburg EC, Vermeer C, Whooley MA, Schurgers LJ: Association of kidney function and uncarboxylated matrix Gla protein: Data from the Heart and Soul Study. Nephrol Dial Transplant 24: 2095–2101, 2009
32. Schurgers LJ, Teunissen KJ, Knapen MH, Kwaijtaal M, van Diest R, Appels A, Reutelingsperger CP, Cleutjens JP, Vermeer C: Novel conformation-specific antibodies against matrix gamma-carboxyglutamic acid (Gla) protein: Undercarboxylated matrix Gla protein as marker for vascular calcification. Arterioscler Thromb Vasc Biol 25: 1629–1633, 2005
33. Kohlmeier M, Saupe J, Drossel HJ, Shearer MJ: Variation of phylloquinone (vitamin K1) concentrations in hemodialysis patients. Thromb Haemost 74: 1252–1254, 1995
34. Pilkey RM, Morton AR, Boffa MB, Noordhof C, Day AG, Su Y, Miller LM, Koschinsky ML, Booth SL: Subclinical vitamin K deficiency in hemodialysis patients. Am J Kidney Dis 49: 432–439, 2007
35. Schurgers LJ, Aebert H, Vermeer C, Bultmann B, Janzen J: Oral anticoagulant treatment: Friend or foe in cardiovascular disease? Blood 104: 3231–3232, 2004
36. Koos R, Mahnken AH, Muhlenbruch G, Brandenburg V, Pflueger B, Wildberger JE, Kuhl HP: Relation of oral anticoagulation to cardiac valvular and coronary calcium assessed by multislice spiral computed tomography. Am J Cardiol 96: 747–749, 2005
37. Schurgers LJ, Spronk HM, Soute BA, Schiffers PM, DeMey JG, Vermeer C: Regression of warfarin-induced medial elastocalcinosis by high intake of vitamin K in rats. Blood 109: 2823–2831, 2007
38. Geleijnse JM, Vermeer C, Grobbee DE, Schurgers LJ, Knapen MH, van der Meer IM, Hofman A, Witteman JC: Dietary intake of menaquinone is associated with a reduced risk of coronary heart disease: The Rotterdam Study. J Nutr 134: 3100–3105, 2004
39. Beulens JW, Bots ML, Atsma F, Bartelink ML, Prokop M, Geleijnse JM, Witteman JC, Grobbee DE, van der Schouw YT: High dietary menaquinone intake is associated with reduced coronary calcification. Atherosclerosis 203: 489–493, 2009
Copyright © 2010 The Authors. Published by Wolters Kluwer Health, Inc. All rights reserved.