Skip Navigation



NDT Advance Access published online on November 11, 2008
Nephrology Dialysis Transplantation, doi:10.1093/ndt/gfn613
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
24/5/1506    most recent
gfn613v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Covic, A.
Right arrow Articles by Goldsmith, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Covic, A.
Right arrow Articles by Goldsmith, D.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author [2008]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org


Systematic review of the evidence underlying the association between mineral metabolism disturbances and risk of all-cause mortality, cardiovascular mortality and cardiovascular events in chronic kidney disease

Adrian Covic1, Prajesh Kothawala2, Myriam Bernal2, Sean Robbins3, Arpi Chalian2 and David Goldsmith4

1 ‘C.I. Parhon’ University Hospital and University of Medicine Gr T Popa Iasi, Romania 2 Cerner LifeSciences, CA, USA 3 Amgen (Europe) GmbH, Zug, Switzerland 4 Guy's Hospital, London, UK

Correspondence and offprint requests to: Adrian Covic, Dialysis and Renal Transplantation Center, ‘C.I. Parhon’ University Hospital, Blvd. Carol I Nr. 50, Iasi, 700503, Romania. Tel: +40-721-280246; Fax: +40-232-212237; E-mail: acovic{at}xnet.ro



   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 Conclusions
 References
 
Background. Chronic kidney disease (CKD) is a powerful risk factor for all-cause mortality and its most common aetiology, cardiovascular (CV) mortality. Mineral metabolism disturbances occur very early during the course of CKD but their control has been poor. A number of studies have assessed the relationship between all-cause mortality, CV mortality and events with mineral disturbances in CKD patients, but with considerable discrepancy and heterogeneity in results. Thus, a systematic review was conducted to assess methodological and clinical heterogeneity by comparing designs, analytical approaches and results of studies.

Methods. Medline, EMBASE and Cochrane databases were systematically searched for articles published between January 1980 and December 2007.

Results. Thirty-five studies were included in the review. All-cause mortality was the most commonly assessed outcome (n = 29). Data on CV mortality risk (n = 11) and CV events (congestive heart failure, stroke, myocardial infarction) (n = 4) are limited. The studies varied in populations scrutinized, exposure assessments, covariates adjusted and reference mineral levels used in risk estimation. A significant risk of mortality (all-cause, CV) and of CV events was observed with mineral disturbances. The data supported a greater mortality risk with phosphorus, followed by calcium and parathyroid hormone (PTH). The threshold associated with a significant all-cause mortality risk varied from 3.5–3.9 mg/dL (reference: 2.5–2.9) to 6.6–7.8 mg/dL (reference: 4.4–5.5) for high phosphorus, <3 mg/dL (reference: 5–7) to <5 mg/dL (reference: 5–6) for low phosphorus, 9.7–10.2 mg/dL (reference: ≤8.7) to >10.5 mg/dL (reference: 9–9.5) for high calcium, ≤8.8 mg/dL (reference: >8.8) to <9 mg/dL (reference: 9–9.5) for low calcium and >300 pg/mL (reference: 200–300) to >480 pg/mL (reference: ≤37) for PTH. Thresholds at which the CV mortality risk significantly increased were >5.5 (reference: 3.5–5.5) and >6.5 mg/dL (reference: <6.5) for phosphorus and >476.1 pg/mL (reference: <476.1) for PTH.

Conclusions. Serious limitations were observed in the quality and methodology across studies. In spite of enormous heterogeneity across studies, a significant mortality risk was observed with mineral disturbances in dialysis patients. Data on risk in pre-dialysis patients were less conclusive due to even more limited (numerically) evidence.

Keywords: chronic kidney disease; dialysis; mineral metabolism disturbances; risk of cardiovascular mortality; risk of mortality



   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 Conclusions
 References
 
Chronic kidney disease (CKD) is a worldwide public health problem. Approximately 16% of the population in Europe and 13% in the United States (US) have CKD [1,2]. The prevalence and incidence of end-stage renal disease in 2005 among European adults were 808 and 124 per million people (p.m.p), and in the US were 1569 and 347.1 p.m.p, respectively [3,4]. The 5-year survival rates among dialysis patients are low (Europe: 40.5%, US: 55.2%) and have remained constant over the past decade [3,4]. Even early CKD patients have poor survival; approximately 25% of these patients die while relatively few reach dialysis [5]. The treatment costs of CKD are high; they have increased 4-fold over the past 10 years to reach 42 billion US dollars, or 12.7% of the total Medicare budget [3].

CKD is a powerful risk factor and a risk multiplier for poor survival [6]. The mortality rates are considerably higher among CKD patients than among the general population (17.7 versus 5.5/100 patient-years) [7]. A recent analysis identified a 6- to 9-fold higher mortality risk among dialysis patients compared to the non-CKD cohort [8]. Cardiovascular disease (CVD) is the single largest cause of mortality, responsible for 40–50% of deaths among CKD patients [3]. CKD patients are more likely to die of CVD than to develop kidney failure, with mortality rates estimated to be 10–30 times higher in dialysis patients as compared to the healthy population [9].

The mortality epidemic in CKD patients may be attributed to a high prevalence and longer exposure [10,11] to traditional CV risk factors (such as diabetes, hypertension, obesity and hypercholesterolaemia) and involvement of new non-traditional risk factors (such as hyperhomocysteinaemia, abnormal lipoprotein levels, chronic inflammation and oxidant stress [12]). Some CKD patients are at an increased risk of mortality even with low blood pressure and reduced cholesterol, which has been referred to as ‘reverse epidemiology’ [13].

Deterioration of kidney function is also accompanied by mineral metabolism disturbances, which include decreased vitamin D and increased phosphorus and parathyroid hormone (PTH) levels. These mineral disturbances begin to develop early with only slight impairment in renal function [14]. However, control of mineral parameters in CKD patients has been poor [15]. Achievement of mineral control has been difficult with traditional treatments such as vitamin D and calcium binders [16]. Newer therapies that target all mineral parameters are expensive, emphasizing the need to understand the relationship between mineral disturbances and survival to make appropriate treatment recommendations.

This study is a systematic review that examines the relationship of mineral disturbances with all-cause mortality and its most common aetiology, CV mortality and CV events in CKD patients. Since there is evidence of discrepancy in results across studies and between observational data and prospective trials, the special focus of this analysis is on addressing methodological and clinical heterogeneity in designs, populations and analytical approaches among studies; this focus differentiates our study from previous reviews [17].



   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 Conclusions
 References
 
Literature review
The MEDLINE, EMBASE and Cochrane databases were systematically searched for English-language articles published from January 1980 to December 2007. The search strategy was comprised of the following medical subject headings (MeSH) and keywords from four categories: population [renal dialysis (MeSH), kidney failure chronic (MeSH), dialysis (MeSH), kidney failure (MeSH), end-stage renal disease (ESRD), chronic kidney disease (CKD), haemodialysis, peritoneal dialysis], interventions [calcitriol (MeSH), vitamin D (MeSH), calcium compounds (MeSH), cinacalcet hydrochloride, alfacalcidol, vitamin D analogues, paricalcitol, doxercalciferol, hectorol, maxacalcitol, phosphate binders, calcium carbonate, calcium acetate, sevelamer hydrochloride, lanthanum carbonate, ferric citrate], mineral metabolism parameters [phosphorus (MeSH), calcium (MeSH), parathyroid hormone (MeSH), P, Ca, PTH] and outcomes [mortality (MeSH), fatal outcome (MeSH), myocardial infarction (MeSH), cerebrovascular accident (MeSH), overall mortality, all-cause mortality, death, CV mortality, CV events, stroke, congestive heart failure, CHF, acute coronary syndrome, ACS, transient ischaemic attack, TIA].

The search strategy consisted of four components. The first part of the search combined keywords from ‘population’, ‘mineral parameters’ and ‘outcomes’ categories to capture articles that assessed the relationship between mineral metabolism abnormalities and risk of clinical outcomes in CKD patients. The second component combined keywords from ‘population’, ‘intervention’ and ‘outcomes’ categories to aid in identifying articles that assessed the effect of SHPT interventions in ameliorating the risk of clinical outcomes. The third component was the combination of all articles identified by the first two searches and the exclusion of all duplicate articles. The last component consisted of limiting the search to English-language articles published between January 1980 and December 2007. Table 1 describes the search strategy used to identify the published articles of interest.


View this table:
[in this window]
[in a new window]

  		Table 1 Details of the search strategy applied in identifying published articles of interest

 
For inclusion in the review, the published articles had to meet the following criteria: the subjects were adults (≥18 years of age) with CKD or on dialysis; the clinical outcomes were the risk of all-cause mortality, CV mortality and CV events [myocardial infarction (MI), congestive heart failure (CHF), stroke, transient ischaemic attack (TIA), acute coronary syndrome (ACS)] with mineral disturbances or its treatment with SHPT interventions and the study design was observational or a clinical trial. We focused on both all-cause mortality and CV mortality, as inclusion of only CV mortality as an outcome may be misleading because labelling of CV deaths across the identified studies are usually based upon death certificates that carry a significant risk of misclassification. At the same time, the majority of deaths in the CKD population (40–50%) are attributed to CV causes [3], which emphasizes the need to focus on both all-cause and CV mortality. Articles were excluded if study designs were case reports, editorials or reviews and if the language of publication was non-English. Articles were also rejected if they assessed indirect surrogate CV outcomes such as vascular calcification and left ventricular hypertrophy (LVH). Vascular calcification and LVH correlate significantly with clinical endpoints and mineral parameters and thus are intermediate and surrogate endpoints for mortality. However, a finding of a significant relationship between mineral disturbances and vascular calcification does not necessarily translate into a significant association between mineral abnormalities and mortality. Thus, assessing clinical fatal endpoints (mortality, CV events) along with their surrogates (vascular calcification, LVH) may contribute to significant methodological heterogeneity.

Using these criteria, two reviewers independently reviewed an identical 10% random sample of the abstracts that were identified with the search strategy. Inter-rater agreement was assessed using the kappa statistic, and the remaining abstracts were split evenly between the reviewers once a sufficient level of agreement was achieved ({kappa} > 0.7). The full-text publication was obtained for each accepted abstract, and the review process was repeated for all full-text articles.

Data synthesis
A qualitative analysis was performed to summarize the results of the identified studies. The studies were categorized by type of outcome analysed. The number of studies reporting a significant risk (P < 0.05) of each outcome with abnormalities in every mineral parameter was calculated. The threshold mineral level above which the risk significantly increased was ascertained. Also, the risk of every outcome with mineral levels beyond the K/DOQI-recommended targets (phosphorus: 3.5–5.5 mg/dL; calcium: 8.4–9.5 mg/dL and PTH: 150–300 pg/mL) was measured. Clinical and methodological homogeneity across studies was explored with respect to populations studied [CKD stage, type of dialysis (incident versus prevalent), mode of dialysis (HD versus PD)], exposure assessments (continuous, categorical or dichotomous variable), risk measurements [risk (RR), hazard (HR), or odds ratio (OR)], and the reference mineral levels used in the risk measurement. Table 2 describes the baseline characteristics and level of clinical and methodological heterogeneity across studies. The types of covariates adjusted across the identified studies were also assessed to gain additional information on the consistency of analytical techniques used across studies and the ability to compare results. We captured the type of covariates that were controlled by each identified study in a table (Table 3) and then compared the adjusted covariates across studies to gain insight into the extent of methodological heterogeneity. We measured the number of studies that adjusted for each reported covariate and identified covariates, if any, that were controlled consistently across studies. Furthermore, the effects of treatments on the risk of outcomes were assessed.


View this table:
[in this window]
[in a new window]

  		Table 2 Baseline characteristics of observational studies assessing risk of mortality (all-cause, CV) and CV events

 

View this table:
[in this window]
[in a new window]

  		Table 3 Variations in covariates adjusted in studies assessing the risk of all-cause mortality with mineral metabolism disturbances

 
The methodology and reporting of the included studies were assessed based on the STROBE statement (The Strengthening the Reporting of Observational Studies in Epidemiology) [18]. The STROBE criteria [18] consist of a checklist of 22 items, which assesses the quality of the following components of an observational study: (1) title and abstract; (2) introduction: background/rationale, objectives; (3) methods: study design, setting, participants, variables, data sources/measurement, bias, study size, quantitative variables, statistical methods; (4) results: participants, descriptive data, outcomes data, main results and (5) discussion: key results, limitation and interpretation. A detailed description of each item along with the assessment of the included studies is reported in Table 4. In addition, we created a rating score and classified the quality of each publication as good [attained the criteria listed by at least 17 items (75%)], moderate [11–16 items (50%)] or poor (≤10 items).


View this table:
[in this window]
[in a new window]

  		Table 4 Assessment of the quality of observational studies assessing the risk of clinical outcomes with mineral abnormalities based upon the STROBE statement (22 items)

 


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 Conclusions
 References
 
Literature review
The search strategy identified 2534 references, of which 532 abstracts were selected for full-text review. Thirty-five of these met the inclusion criteria. The majority of rejected full-text articles failed to assess the outcomes of interest. Figure 1 describes the literature review process.


Figure 1
View larger version (29K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 A flow-chart diagram on the results of the literature review.

 
Twenty-seven studies assessed the relationship between mineral metabolism disturbances and CV outcomes and eight evaluated the effect of treating SHPT on outcomes of interest. All-cause mortality (n = 29) was the most commonly assessed outcome, followed by CV mortality (n = 11) and CV events (n = 4). Study designs of 35 studies were observational (n = 32), RCT (n = 1), pooled analysis of RCTs (n = 1) and post hoc analysis of RCT (n = 1). Table 2 outlines the baseline characteristics of identified studies.

All-cause mortality
Twenty-two studies [19–40] assessed the mortality risk with mineral metabolism disturbances. However, the studies were heterogeneous with respect to type of populations enrolled, used varying methods of mineral parameter assessment and controlled for different covariates. Of 22 studies, seven [22,24,30,32,34,38,39] enrolled incident and the rest, prevalent dialysis/pre-dialysis patients; five studies focused on HD and PD [22,24,30,32,37]; 14 studies [19–21,23,25,27–29,33,35–37,40,41], HD only and three studies [26,31,38], only pre-dialysis patients. Seven studies [19,20,25,29,30,32,36] assessed mineral parameters categorically; six [21,22,28,33,35,39], dichotomously; seven [23,24,26,27,31,38,40], continuously; one [37], both categorically and continuously and one [34] used all three techniques. The mineral levels used as a reference in measuring the risk lacked uniformity. The risk was expressed as RR (n = 14) [19,20,22,23,25,27–29,33–35,37,38,40] or HR (n = 8) [21,24,26,30–32,36,39]. Furthermore, evidence on the association between K/DOQI targets and mortality risk is limited (n = 4 [20,21,32,39]).

Only nine [19,20,22,23,25,26,32,37,39] of the 22 studies were rated as good quality based upon the STROBE criteria. The quality of the methodology and reporting of data was moderate (n = 8 [21,24,27,30,31,34,36,38]) or poor (n = 5 [28,29,33,35,40]) in a majority of studies. The major areas of weakness were the lack of assessment of mortality risk with time-dependent changes in mineral markers (n = 19), failure to report the reasons for choosing the adjusted covariates (n = 17), lack of information on the external validity of the results (n = 15) and how the study sample size was determined from the initial patient enrolment (n = 14).

Of a set of 20 covariates considered, only age was adjusted for consistently across studies. Although it is important to control for dialysis vintage due to its positive association with mortality, only nine studies [20,23,25, 27,28,33,35,36,40] adjusted for it. Very few studies controlled for differences in CKD severity by adjusting glomerular filtration rate (GFR) (n = 2 [31,38]), creatinine (n = 3 [20,25,27]) or CKD aetiologies (n = 2 [23,31]). Although six studies [22,24,30,30,32,42] enrolled HD and PD patients, only two [30,37] controlled for dialysis modality. Variations in adjustments for risk factors or its measures were noted for diabetes (n = 18 [19–25,27,28,30,31,33–38,40]), CV comorbidities (n = 13 [21,22,24,26,27,30–33,35,36,38,40]), smoking (n = 7 [19,22–24,30,31,36]), blood pressure (n = 3 [22,26,31]) and cholesterol (n = 3 [20,22,31]). Recently, a U-shaped relationship between bicarbonate and mortality was observed in CKD patients [43] but only two studies controlled for this [20,25]. Also, the influence of co-medications (vitamin D/phosphate binders) on mortality risk was considered in only three studies [25,26,30]. Tables 3 and 5 outline the variations in adjusted covariates and the results of identified studies, respectively.


View this table:
[in this window]
[in a new window]

  		Table 5 Summary of results of studies assessing the risk of all-cause mortality with mineral metabolism disturbances

 
Phosphorus
Seventeen studies [19,20,23–25,27–30,32–37,39,40] assessed the mortality risk with high phosphorus in dialysis patients; all but one study demonstrated a significant relationship. Of six studies assessing phosphorus continuously [23,24,27,34,37,40], three reported a significant rise in relative risk by 4% [40], 5% [23] and 26% [34], respectively, with every 1 mg/dL increase and two, by 56% [37] and 341% [24], respectively, for every 1 mmol/L increase in phosphorus. The lack of a significant relationship in the remaining study [27] may be attributed to the small number of mortality events and failure to adjust for covariates such as hypertension, smoking and cholesterol. Eight studies used categorical models [19,20,25,29,30,32,34,36]; all reported significant increases in mortality risk at higher levels of phosphorus as compared to the reference range. The mortality risk increased significantly at phosphorus levels from 5–5.5 mg/dL (RR: 1.1, reference: 4–5) [20]; 5.5–6 mg/dL (RR: 1.25, reference: 4–5); 5–6.5 mg/dL (RR: 1.94, reference: 3–5) [34]; >5.5 mg/dL [HR: 1.4 (HD) and 1.6 (PD), reference: 3.5–5.5] [32]; >6 mg/dL (HR: 1.57, reference: 4.3–5.1 [30] and 5–6 [25]); 6.4–7.5 mg/dL (HR: 1.1, reference: ≤4.4) [36]; >6.5 mg/dL (RR: 2.02, reference: 3–5); 6.6–7.8 mg/dL (RR: 1.18, reference: 4.4–5.5) [19]; >7.5 mg/dL (HR: 1.19, reference: ≤4.4); 7.9– 16.9 mg/dL (RR: 1.39; reference: 4.4–5.5) and ≥11 mg/dL (RR: 2.47; reference: 4–5). A similar trend was also observed in dichotomous models (n = 4 [28,33–35]), which reported a significant increase in the relative risk of mortality by 111% at >5 mg/dL (reference: <5) [34], 11% at >5.5 mg/dL (reference: ≤5.5) [33] and 17% at >7.5 mg/dL (reference: ≤7.5) [28,35].

The relationship between high phosphorus and mortality risk remained significant even in pre-dialysis patients [26,31,38]. In continuous models, every 1 mg/dL increase in phosphorus was associated with a significant increase in risk by 10% [31], 33% [26] and 62% [38] in pre-dialysis patients. One study [26], using a categorical model, reported a significant increase in hazard of all-cause mortality in pre-dialysis patients at phosphorus levels between 3.5– 3.9 mg/dL (HR: 1.32), 4–4.4 mg/dL (HR: 1.34), 4.5–4.9 mg/dL (HR: 1.83) and ≥5 mg/dL (HR: 1.9) as compared to the reference level of 2.5–2.9 mg/dL.

Two studies assessed the clinical significance of controlling phosphorus within the K/DOQI targets. A significant mortality risk was observed in HD (HR: 1.4) and PD patients (HR: 1.6) with levels >5.5 mg/dL (reference: 3.5–5.5) [32]. Another study observed a significant amelioration in risk, by 33% in patients with phosphorus within the K/DOQI target [39].

Three studies [20,25,29] in dialysis patients reported a significant risk of mortality with low phosphorus at a cut-off level of <5 mg/dL (reference: 5–6 mg/dL), <3 mg/dL (reference: 5–7 mg/dL) and <4 mg/dL (reference: 4–5).

Calcium
Nine studies [19,25,27,30,32,36,37,39,40] assessed the mortality risk with high calcium in dialysis patients; six [25,27,30,36,39,40] reported a significant relationship. In studies using continuous models, every 1 mg/dL increase was associated with a significant rise in the relative mortality risk, by 12% [40] and 22% [27]. However, the risk was not significantly higher with a 1 mmol/L increase in calcium [37]. Of five studies analysing calcium categorically [19,25,30,32,36], three observed significant increases in the mortality risk with higher calcium levels as compared to the reference range. The mortality risk significantly increased by 11% at calcium levels from 9.7– 10.2 mg/dL (reference: ≤8.7) [36], 52% at >9.73 mg/dL (reference: 8.97–9.33) [30], 14% at >10.2 mg/dL (reference: ≤8.7) and >10.5 mg/dL (risk estimate: not reported, reference: 9–9.5) [25]. The absence of a significant relationship in the remaining two studies [19,32] may be attributed to different reference mineral levels and failure to adjust for dialysis vintage and comorbidities.

Two studies assessed the impact of achieving K/DOQI calcium targets (8.4–9.5 mg/dL) on mortality in dialysis patients but reported conflicting results [32,39]. One study used a dichotomous model and reported that patients with levels within the K/DOQI target had a significantly lower risk (20%) when compared to levels beyond it [39]. On the other hand, categorizing calcium with the K/DOQI target as the reference did not show any significant change in the mortality risk with levels >9.5 mg/dL in either HD or PD patients (HR: 1 and 0.9) [32]. Evidence of the mortality risk with high calcium in pre-dialysis patients is absent.

Two [22,25] of the five studies [19,22,25,30,32] demonstrated a significant relationship between low calcium and mortality risk. One study identified an increased mortality risk at calcium levels <9 mg/dL upon assessing calcium categorically (reference: 9–9.5) [25]. Another dichotomized calcium and observed a 2.31-fold greater relative risk with levels ≤8.8 mg/dL (reference: >8.8) [22].

Parathyroid hormone
Eleven studies [19–21,25,27,30, 32,36,37,39,40] analysed the relationship between high PTH and risk of all-cause mortality in dialysis patients; only seven [20,25,27,30,36,39,40] showed any significant association. Data on mortality risk with PTH changes in pre-dialysis patients are lacking. Two studies assessed mortality with every 100 pg/mL increase in PTH and reported significant increases in the relative risk, by 1% [40] and 4% [27], respectively. However, the risk did not increase significantly with a 1 pmol/L increase in PTH levels [37]. Categorical models revealed a significant increase in the mortality risk in only four [20,25,30,36] of the seven studies [19–21,25,30,32,36]. The categories of PTH levels at which the mortality risk increased significantly were >300 pg/mL (risk estimate: not reported, reference: 200–300) [25]; >308 pg/mL (HR, 1.68, reference: 160–308) [30]; >480 pg/mL (HR: 1.17, reference: ≤37) [36]; 600–900 pg/mL (RR: 1.08, reference: 150–300) [20]; 900–1200 pg/mL (RR: 1.18, reference: 150–300) and >1200 pg/mL (RR: 1.24, reference: 150–300).

Four studies [20,21,32,39] assessed the association between mortality risk and K/DOQI targets but showed conflicting results depending upon the categorization choices. Categorical analysis identified a significant risk at PTH levels between 600 and 900 pg/mL [20] but not at >300 pg/ mL [21,32] relative to the reference of 150–300 pg/mL. Analysing PTH dichotomously, one study reported a significant decrease in risk by 11% in patients with levels within the target compared with those beyond it [39]. No significant relationship was observed between low PTH levels and risk of mortality.

Efficacy of SHPT interventions in reducing the risk of mortality
Seven studies [30,44–49] assessed the effect of interventions on mortality risk. Teng et al. [44] observed a significant reduction in risk, by 16% in paricalcitol as compared to calcitriol users (HR: 0.84). The outcome difference may be explained by the non-random assignment of therapies and lack of control of CV co-morbidities, co-medications and risk factors such as hypertension and cholesterol. In contrast, a recent study by Tentori et al. [45] did not identify any significant differences in the mortality rates between doxercalciferol or paricalcitol and calcitriol groups (HR: 0.95). Two studies [30,46] reported a significant risk reduction among vitamin D as compared to non-users (HR: 0.8 and 0.74) but failed to control for CV risk factors, dialysis vintage and co-medications. A pooled analysis of RCTs [47] reported no significant reduction in mortality among cinacalcet users as compared to standard care (RR = 0.81). This outcome may be explained by the lack of statistical power in the pooled RCTs to identify between-group differences in mortality rates. Two studies [48,49] compared the efficacy of sevelamer and calcium on mortality risk but showed conflicting results. Suki et al. [48] showed similar rates in sevelamer and calcium groups (15 versus 16.1/100 patient-years). In contrast, Block et al. [49] reported significantly lower rates among sevelamer as compared to the calcium group (5.3 versus 10.6/100 patient-years). However, this finding should be interpreted with caution as switching of medications was allowed and the results were derived from post hoc analysis of an RCT that considered all-cause mortality as a secondary endpoint.

Cardiovascular mortality
Eight studies [23,27,31,40,50–53] evaluated the CV mortality risk with mineral disturbances. Of eight studies, six [23,27,40,51–53] included HD patients only; one [50], HD and PD patients and one [31], a CKD stage 3 or 4 population. Only one study [50] evaluated the CV mortality risk with mineral concentrations outside the K/DOQI targets. Only one study [23] was rated to be of ‘good’ quality as it met the criteria on 18 of the 22 STROBE items. The quality of the remaining studies varied from moderate (n = 4 [27,31,50,52]) to poor (n = 3 [40,51,54]). Only one study [50] clearly defined the confounders and the reasons for its adjustment. Similarly, only one study [50] assessed the relationship between time-dependent changes in mineral markers and risk of cardiovascular mortality.

Mineral parameters were assessed continuously [27,31, 40,51,53], dichotomously [50,52] or both [23]. Of 18 covariates considered, only age was adjusted uniformly. Enormous inconsistency was observed in control of covariates such as dialysis vintage (n = 5) [23,27,40,51,53], GFR (n = 1) [31] or creatinine (n = 1) [23,51]. Traditional CV risk factors were not adjusted uniformly. One study [50] assessed a population of HD and PD patients but did not adjust for the dialysis modality. Methodologies of risk measurement were either RR [23,27,40,51–53] or HR [31,50]. Table 6 describes the results of the studies on CV mortality.


View this table:
[in this window]
[in a new window]

  		Table 6 Summary of results of studies analysing the risk of cardiovascular mortality with changes in mineral metabolism parameters

 
Phosphorus
Six studies [23,27,40,50,52,54] assessed phosphorus in dialysis patients; all reported a significant risk of CV mortality with high phosphorus. Four studies, assessing phosphorus continuously, reported a significant increase in the relative risk of mortality, by 10% [40], 9% [53], 13% [27] or 9% [23] with every increase of 1 mg/dL. Using a dichotomous model, one study [50] assessed the risk with phosphorus levels above the K/DOQI target; a significant risk was noted in HD (HR = 1.5) and PD patients (HR = 2.4) with levels >5.5 mg/dL (reference: 3.5–5.5). Two studies [23,52] assessed the risk with levels >6.5 mg/dL; the relative risk estimates were significant but differed with reference levels (2.4–6.5 mg/dL: 1.36; <6.5 mg/dL: 2.5). The mortality risk in CKD patients varied depending upon the covariates controlled. In continuous models, a significant association with hyperphosphataemia was observed upon adjustments of demographics and CV risk factors (HR = 1.43). However, the risk was no longer significant after adjustment for the CKD severity [31].

Calcium
Four studies [27,40,50,53] analysed the CV mortality risk with high calcium in dialysis patients. In continuous models, every 1 mg/dL increase was associated with a significantly higher (13% [40], 14% [53] and 28% [27]) relative risk of mortality. However, a dichotomous analysis of the effect of calcium control within K/DOQI targets on CV mortality yielded no significant increase in risk in HD (HR: 1) and PD patients (HR: 1) with levels >9.5 mg/dL (reference: 8.4–9.5) [50]. Only one study assessed the risk in patients with low calcium. The risk with calcium levels <8.4 mg/dL as compared to the reference range of 8.4–9.5 mg/dL was high in PD (HR: 2.8) and HD patients (HR: 1.5) (reference: 8.4–9.5) but did not reach statistical significance [50].

Parathyroid hormone
Six studies [27,40,50–53] analysed PTH in dialysis patients; five found a significant CV mortality risk with high PTH. As a continuous variable, a 100 pg/mL increase resulted in a significant increase in the relative risk of CV mortality by 2% [40,53], 7% [51] and 8% [27]. When analysed dichotomously, a significant increase in risk (RR = 3.9) was found in patients with PTH levels >476.1 pg/mL [52] (reference: <476.1). However, it is doubtful whether controlling PTH within the K/DOQI targets decreases the mortality risk, as is evident by a non-significant risk in HD (HR = 0.8) and PD patients (HR = 1.3) at levels >300 pg/mL (reference: 150–300) [50]. The relationship of CV mortality with PTH is modest, unlike those of phosphorus and calcium. The CV mortality risk in patients with PTH levels below the lower limit set by K/DOQI or <150 pg/mL (reference: 150–300) was also non-significant in HD (HR = 0.9) and PD patients (HR = 1.2) [50].

Effect of SHPT interventions in ameliorating risk of cardiovascular mortality
Three studies [46,48,55] assessed the impact of treatments on the CV mortality risk in dialysis patients. Shoji et al. [55] conducted a multivariate analysis and reported a significantly reduced risk among alfacalcidol users as compared to non-users (HR = 0.37). The outcome difference may be explained by unadjusted factors such as co-medications, CV traditional risk factors and severity of CKD. Teng et al. [46] reported a significantly lower incidence of CV mortality with injectable vitamin D as compared to non-users (7.6/100 versus 14.6/100 person-years). However, the study did not control for confounders such as oral vitamin D, phosphate binders, CV medications, CV risk factors and dialysis vintage. An RCT by Suki et al. [48] did not report any significant differences in CV mortality rates between sevelamer or calcium arms (8/100 versus 8.6/100 patient-years; HR = 0.93). However, CV mortality was a secondary endpoint, indicating a lack of statistical power to detect any between-group differences.

Cardiovascular events
Only four studies [22,26,36,56] assessed the risk of CV events with mineral disturbances. Two studies enrolled HD and PD patients [22,56]; one, HD [36] and one [26], pre-dialysis patients. Two [22,56] enrolled incident dialysis patients, and the rest [36], chronic. Outcomes assessed included CHF (n = 2) [22,56], MI (n = 1) [26] and a composite outcome (MI, CHF, TIA or stroke) (n = 1) [36]. CV events were assessed as primary endpoints in two [36,56] studies. None of the studies independently assessed stroke, TIA or ACS. Data on the effect of treatments on CV events are lacking. The quality of observational assessing CV events was good in two studies [22,26] and moderate in the remainder [36,56].

One study [26] measured mineral parameters as a continuous, two [22,56] dichotomous and one [36] categorical variable. Among 20 covariates considered, only age, diabetes and hypertension were adjusted for consistently. Nutritional deficiencies, a risk factor for cardiac decompensation, were considered in three [22,26,56] of the four studies. CV event traditional risk factors were controlled for inconsistently. Only three studies [26,36,56] considered the confounding effect of renal function impairment; they adjusted dialysis vintage or creatinine. Only one study [26] adjusted for the potential treatment effect of CV medications on the risk of CV events. Risk estimates were calculated as HR (n = 2) [26,36], RR (n = 1) [22] and OR (n = 1) [56]. Data on the CV event risk with mineral levels beyond the K/DOQI targets were not found. Table 7 reports the results of identified studies.


View this table:
[in this window]
[in a new window]

  		Table 7 Summary of results of studies analysing the risk of CV morbidity with changes in mineral metabolism parameters

 
Phosphorus
Three studies [26,36,56] assessed the CV event risk with high phosphorus; all reported a significant relationship. One study [36] assessed ‘any CV event’ categorically in HD patients; it reported a significant risk at phosphorus levels of 4.5–5.3 mg/dL (HR: 1.06), 5.4– 6.3 mg/dL (HR: 1.16), 6.4–7.5 mg/dL (HR: 1.14) and >7.5 mg/dL (HR: 1.25) as compared to the reference range of ≤4.4 mg/dL. Using a dichotomous model identified a significantly higher risk of CHF in HD and PD patients at concentrations ≥6.8 mg/dL (OR = 1.34) as compared to the reference level of <6.8 mg/dL [56]. For every 1 mg/dL increase, the risk of MI in CKD patients significantly increased (HR = 1.35) [26].

Calcium
Two studies [36,56] analysed the CV event risk with high calcium in HD patients; both reported a significant relationship. One study assessed calcium categorically and reported a significant risk of any CV event only with severe increases in calcium, >10.2 mg/dL (reference: ≤ 8.7) (HR = 1.08) [36]. However, a dichotomous analysis identified a significantly higher risk of CHF at levels ≥8 mg/dL (reference: <8) (OR = 1.41) [56]. Only one study [22] investigated CHF with low calcium and reported a significantly greater risk of de novo (RR: 2.43) and recurrent CHF (RR: 2.66) with calcium ≤8.8 mg/dL as compared to those with levels >8.8 mg/dL.

Parathyroid hormone
Only one study [36] assessed the risk of ‘any CV event’ with high PTH in HD patients. This study reported a significant increase in risk with PTH >480 pg/mL (reference: ≤37) (HR = 1.12).



   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
Conclusions
 References
 
This study is a systematic review that compares and contrasts designs, analytical approaches and results of studies on the risk of all-cause mortality, CV mortality and CV events with mineral disturbances among CKD patients. Serious limitations were noted in the number of studies and the methodological quality of evidence. Very few studies (n = 12) focused on the relation between mineral parameters and CV outcomes. Evidence in pre-dialysis patients is scarce (n = 3), which is in contrast to the high prevalence of incipient CKD and its recognition as a risk factor for accelerated CVD. The number of interventional RCTs is highly limited (n = 2).

The observational studies were clinically and methodologically diverse in populations studied and analytical strategies used. Studies assessed prevalent or incident, HD, PD or pre-dialysis patients. Methodologically, the studies were heterogeneous in technique of mineral parameter assessment, method of risk calculation, adjusted covariates and reference mineral levels. Mineral parameters were evaluated categorically, dichotomously or continuously. Categorical analysis is a rigorous technique that captures the threshold range of mineral levels associated with significant risk. In contrast, a continuous model does not identify the threshold level at which the risk begins to increase and tends to force a linear relationship between mineral parameters and outcome risk. Dichotomous analysis prevents the comparability of results across studies. However, fewer than half of the observational studies evaluated mineral parameters categorically (n = 9). A large sample used continuous only (n = 10), dichotomous only (n = 7) or both (n = 1). The mineral levels used as a reference in calculating risk lacked consistency, thus further limiting the ability to compare results across studies. Table 8 describes the variations in method of mineral parameter assessment and risk calculation across studies.


View this table:
[in this window]
[in a new window]

  		Table 8 Variations in the method of mineral parameter assessment and risk measurements across the identified studies

 
In spite of this heterogeneity, the review identified a significant relationship between mineral parameters and all-cause mortality, CV mortality and CV events in dialysis patients. However, data on risk of outcomes in pre-dialysis patients was less conclusive due to limited evidence. Data supports a greater mortality risk with phosphorus followed by calcium and PTH. All but one study assessing phosphorus showed a significant all-cause mortality risk. However, the strong association may be due to ‘negative publication bias’. In contrast, the risk of mortality with calcium and PTH was significant in only six of nine studies and seven of eleven studies, respectively. No significant relationship was observed between low PTH levels and risk of all-cause mortality. However, a recent cross-sectional study by Nakai et al. [57] observed a significantly lower risk of all-cause mortality with low PTH levels at a threshold level of 60–119 pg/mL as compared to the reference level of 180– 359 pg/mL (HR = 0.891; P = 0.0173). This finding may be attributed to utilization of baseline PTH levels instead of time-dependent PTH levels in the measurement of mortality risk and misclassification of mineral parameter levels due to the cross-sectional nature of the database. The finding needs to be confirmed by assessing the risk of mortality with time-dependent concentration of low PTH levels.

Continuous assessment of mineral parameters in dialysis patients resulted in significant all-cause mortality risk estimates that varied from 1.04–1.26 (every 1 mg/dL increase) and 1.56–3.41 (1 mmol/L increase) for phosphorus, 1.12–1.22 for calcium (1 mg/dL increase) and 1.01–1.04 for PTH (100 pg/mL increase). In dichotomous models, the mortality risk estimates varied from 1.11 (>5.5 versus ≤ 5.5 mg/dL) to 2.11 (>5 versus <5 mg/dL) for high phosphorus. The risk estimates in categorical models ranged from 1.1 (5–5.5 versus 4–5 and 6.4–7.5 versus ≤4.4 mg/dL) to 2.47 (≥11 versus 4–5 mg/dL) for high phosphorus, 1.07 (8.8–9.2 versus ≤8.7 mg/dL) to 1.52 (>9.73 versus 8.97–9.33 mg/dL) for high calcium and 1.08 (600–900 versus 150–300 pg/mL) to 1.68 (>308 versus 160–308 pg/mL) for high PTH.

With respect to CV mortality, continuous assessment of mineral parameters resulted in risk estimates that varied from 1.09–1.13 for phosphorus (every 1 mg/dL increase), 1.13–1.28 for calcium (every 1 mg/dL increase) and 1.02–1.08 for PTH (every 100 pg/mL increase). Dichotomous analysis resulted in risk estimates that varied from 1.36 (>6.5 versus 2.4–6.5 mg/dL) to 2.5 (>6.5 versus <6.5 mg/ dL) for high phosphorus, 1 for high calcium (>9.5 versus 8.4–9.5 mg/dL) and 0.8 (>300 versus 150–300 pg/mL) to 3.9 (>476.1 versus <476.1 pg/mL) for high PTH levels. Described differences in adjusted confounders, reference mineral levels and method of mineral parameter assessment are likely to be the major factors responsible for the variability in observed risk estimates.

This finding is consistent with previous reviews. Kestenbaum et al. [17] reviewed only selected studies and found a significant association of phosphorus, calcium and in some cases PTH with mortality among CKD patients. Our special aim of evaluating the quality and addressing the heterogeneity across all studies differentiates this analysis from the review by Kestenbaum et al. [17]. Nevertheless, the relationship between serum phosphorus and CVD has been further solidified by studies conducted in patients without CKD and/or CVD. A post hoc analysis [58] identified a graded independent relation between high phosphorus and CV risk in prior MI patients; most of them did not have CKD-related overt hyperphosphataemia. A prospective evaluation of Framingham offspring study participants showed a similar relationship in patients without CVD or CKD, thus indicating a direct link between phosphorus and vascular risk [59]. A review by Raggi et al. [60] highlights the contribution of SHPT and vitamin D deficiency—two early CKD complications—to the enormous CVD burden, which may be responsible for the increased mortality risk among CKD patients. Finally, a recent prospective cohort study by Tentori et al. [61] observed a significant risk of mortality with high phosphorus (>7 mg/dL), high calcium (>10 mg/dL) and severe increases in PTH levels (>600 pg/mL).

There are several mechanisms that may explain the association of mineral parameters with greater mortality risk; development of CVD and vascular calcification are the most likely links, followed by infections. High phosphate and calcium may promote calcification by causing osteogenic differentiation of the vascular smooth cells and mineralization of the collagenous extracellular matrix secreted by the differentiated cells [62,63]. However, there are cases in which elevated concentrations of calcium and phosphate may not induce vascular calcification because of inhibition by pyrophosphate. However, the presence of chronic renal failure may create a deficiency of pyrophosphate activity by increasing levels of tissue non-specific alkaline phosphatase and thus promoting vascular calcification [64,65]. This calcification is known to be associated with arterial stiffening, increased pulse pressure, decreased coronary perfusion and LVH [66,67], all of which might directly contribute to CV mortality [68]. Furthermore, arterial stiffening can create abnormalities in microcirculation and thus cause poor wound healing and infection-related deaths [23]. Excess phosphate may also influence CV risk by increasing PTH or decreasing 1,25-dihydroxyvitamin D levels. PTH excess has been implicated in cardiac fibrosis [69], impaired cardiac contractility [70], impaired endothelial vasodilatory function [71] and LVH [72]. Lower levels of vitamin D may decrease cardiac contractility [73] and increase coronary calcification [74]. Lower levels of vitamin D may also lead to upregulation of the renin-angiotensin axis, resulting in the development of hypertension and LVH [75,76].

The association between hypophosphataemia and mortality may be due to poor nutritional intake [77], as well as other non-nutritional factors such as reduced cardiac contractility. These can result in respiratory insufficiency, cardiac arrhythmias and significant bone diseases such as osteomalacia, which may increase the fracture rates [78]. Several studies also demonstrated a significant risk of mortality with low calcium levels. Mechanisms involved include the stimulation of parathyroid glands by low calcium, resulting in excess PTH levels, an important risk factor for CVD and mortality. Also, hypocalcaemia can cause cardiac failure, given the role of calcium in maintaining cardiac contractility [79].

The significant association of mineral metabolism parameters with CV morbidity may have significant long-term consequences on quality of life. The published literature has shown a significant impact of CV events such as stroke and CHF on various domains of health-related quality of life [80,81]. Thus, it may be of considerable interest to understand the indirect impact of mineral metabolism disturbances on health-related quality of life. Currently, published data on this topic are lacking and are highly warranted.

Identification of the threshold at which the mortality risk starts to increase significantly is important for clinical practice and development. The threshold level at which the all-cause mortality risk significantly increased varied from 3.5–3.9 mg/dL (reference: 2.5–2.9) to 6.6–7.8 mg/dL (reference: 4.4–5.5) for high phosphorus, <3 mg/dL (reference: 5–7) to <5 mg/dL (reference: 5–6) for low phosphorus, 9.7–10.2 mg/dL (reference: ≤8.7) to >10.5 mg/dL (reference: 9–9.5) for high calcium, ≤8.8 mg/dL (reference: >8.8) to <9 mg/dL (reference: 9–9.5) for low calcium and >300 pg/mL (reference: 200–300) to >480 pg/mL (reference: ≤ 37) for PTH. The threshold level associated with significant CV mortality risk varied from >5.5 mg/dL (reference: 3.5–5.5) to >6.5 mg/dL (reference: <6.5) for phosphorus and >476.1 pg/mL (reference: <476.1) for PTH. The threshold level associated with significant CV event risk varied from ≥4.5 mg/dL (reference: ≤4.4) to ≥6.8 mg/ dL (reference: <6.8) for phosphorus, ≥8 mg/dL (reference: <8) to >10.2 mg/dL (reference: ≤8.7) for calcium and >480 pg/mL (reference: ≤37) for PTH. Only some of the threshold levels of phosphorus and calcium are within the recommended K/DOQI targets. Therefore, these guidelines may begin to look lenient and may require reappraisal [82].

Also, current guidelines for mineral metabolism control were developed based upon observational data, due to paucity of RCT data [82]. Interventional RCTs are long overdue and urgently needed to establish a causal relationship between mineral parameters and risk of patient-centred outcomes. The EVOLVE (Evaluation of Cinacalcet Therapy to Lower CV Events) RCT [83] will try to ascertain the impact of mineral metabolism control on mortality.

Our review has some limitations. The results could not be quantitatively synthesized in a meta-analysis due to the significant clinical and methodological heterogeneity observed across the identified studies. Enormous clinical diversity was observed with respect to enrolled study populations. The enrolled population varied from pre-dialysis, HD or PD patients or a combination of HD and PD. The studies were methodologically heterogeneous in methods of mineral parameter assessment, techniques of risk estimation, mineral levels used as a reference and covariates adjusted. Mineral metabolism parameters were evaluated as categorical, continuous or dichotomous variables. Risk estimates were reported in a diverse manner as a hazard, risk or odds ratio. The reference mineral levels used to measure the risk of clinical outcomes at abnormal mineral levels varied considerably and thus limited the comparability of results across studies. Also, no consistency was observed in the type of covariates adjusted across the identified studies. All of the above factors suggest that across inclusion criteria, populations and methodology across studies, there is a lack of homogeneity needed for pooling results or creating summary estimates across studies via a meta-analysis [84].

Our review only analysed published articles of interest. Thus, unpublished data as a result of ‘negative publication bias’ may have been missed by our review. Studies with significant, positive association between mineral metabolism parameters and risk of all-cause mortality, CV mortality or events are more likely to be published as compared to studies with negative, non-significant results resulting in ‘negative potential bias’ [85,86]. Thus, a number of studies that did not find any significant relationship between mineral disturbances and clinical outcomes may never get published and thus may have not been included in our review. The presence of publication bias may have resulted in over-representation of positive studies in our systematic review, which in turn may have biased our review towards a positive result. Publication bias is an inherent limitation of any systematic review and should be always taken into account when analysing or reading it.



   Conclusions
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Conclusions
References
 
Serious limitations were observed in the number, quality and methodology of studies that reported the association of mineral metabolism disturbances with risk of all-cause mortality, CV mortality and CV events. In spite of these drawbacks, the data showed a significant risk of assessed outcomes with mineral disturbances. Also, the data supported a greater mortality risk with phosphorus followed by calcium and PTH. Data on clinical risk in pre-dialysis patients were less conclusive due to limited (numerically) evidence. The threshold associated with significant all-cause mortality risk varied from 3.5–3.9 mg/dL to 6.6–7.8 mg/dL for high phosphorus, <3 mg/dL to <5 mg/dL for low phosphorus, 9.7–10.2 mg/dL to >10.5 mg/dL for calcium and >300 pg/mL to >480 pg/mL for PTH. Thresholds at which the CV mortality risk significantly increased were >5.5 and >6.5 mg/dL for phosphorus and >476.1 pg/mL for PTH.

Given these findings it is of the utmost clinical importance to learn whether these associations indicate a cause or are coincidental, to better understand the importance of mineral control in decreasing mortality by conducting additional interventional RCTs and thus to more confidently recommend interventions. Future efforts should focus on systematically reviewing the relationship between mineral metabolism parameters and other important clinical outcomes such as risk of fractures, considering the 4-fold increase in fracture risk identified by the USRDS (United States Renal Data System) in dialysis patients as compared to the general population [3].



   Acknowledgments
 
This study was funded by Amgen (Europe) GmbH.

Conflict of interest statement. Dr Covic has served as a consultant, speaker or member of an advisory board for Amgen, Roche, Affymax and Fresenius Medical Care. Dr Goldsmith has served as a speaker or member of an advisory board for Amgen, Genzyme and Shire Pharmaceuticals. Drs Kothawala, Bernal and Chalian are employees of Cerner LifeSciences, a consulting company that provides services to the pharmaceutical industry. Sean Robbins is an employee of Amgen (Europe) GmbH.



   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Conclusions
 References
 

  1. Coresh J, Selvin E, Stevens LA, et al. Prevalence of chronic kidney disease in the United States. JAMA (2007) 298:2038–2047.[Abstract/Free Full Text]
  2. Meguid EN, Bello AK. Chronic kidney disease: the global challenge. Lancet (2005) 365:331–340.[Web of Science][Medline]
  3. United States Renal Data Systems. USRDS 2007 Annual Data Report: Atlas of End-Stage Renal Disease in the United States (2007) Bethesda, MD: National Institute of Health, National Institute of Diabetes and Digestive and Kidney Diseases.
  4. European Renal Association–European Dialysis Transplant Association Annual Data Report 2005.
  5. Daly C. Is early chronic kidney disease an important risk factor for cardiovascular disease? A background paper prepared for the UK Consensus Conference on early chronic kidney disease. Nephrol Dial Transplant (2007) 22(Suppl_9):ix19–ix25.
  6. Go AS, Chertow GM, Fan D, et al. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med (2004) 351:1296–1305.[Abstract/Free Full Text]
  7. Foley RN, Murray AM, Li S, et al. Chronic kidney disease and the risk for cardiovascular disease, renal replacement, and death in the United States Medicare population, 1998 to 1999. J Am Soc Nephrol (2005) 16:489–495.[Abstract/Free Full Text]
  8. United States Renal Data Systems. USRDS 2005 Annual Data Report: Atlas of End-Stage Renal Disease in the United States (2005) Bethesda, MD: National Institute of Health, National Institute of Diabetes and Digestive and Kidney Diseases.
  9. Sarnak MJ, Levey AS, Schoolwerth AC, et al. Kidney disease as a risk factor for development of cardiovascular disease: a statement from the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention. Circulation (2003) 108:2154–2169.[Free Full Text]
  10. Hostetter TH. Chronic kidney disease predicts cardiovascular disease. N Engl J Med (2004) 351:1344–1346.[Free Full Text]
  11. Locatelli F, Pozzoni P, Tentori F, et al. Epidemiology of cardiovascular risk in patients with chronic kidney disease. Nephrol Dial Transplant (2003) 18(Suppl 7):vii2–vii9.
  12. Menon V, Gul A, Sarnak MJ. Cardiovascular risk factors in chronic kidney disease. Kidney Int (2005) 68:1413–1418.[CrossRef][Web of Science][Medline]
  13. Kalantar-Zadeh K, Block G, Humphreys MH, et al. Reverse epidemiology of cardiovascular risk factors in maintenance dialysis patients. Kidney Int (2003) 63:793–808.[CrossRef][Web of Science][Medline]
  14. Zehnder D, Landray MJ, Wheeler DC, et al. Cross-sectional analysis of abnormalities of mineral homeostasis, vitamin D and parathyroid hormone in a cohort of pre-dialysis patients. The chronic renal impairment in Birmingham (CRIB) study. Nephron Clin Pract (2007) 107:c109–c116.[CrossRef][Web of Science][Medline]
  15. Wei M, Taskapan H, Esbaei K, et al. K/DOQI guideline requirements for calcium, phosphate, calcium phosphate product, and parathyroid hormone control in dialysis patients: can we achieve them? Int Urol Nephrol (2006) 38:739–743.[CrossRef][Web of Science][Medline]
  16. Al Aly Z, Gonzalez EA, Martin KJ, et al. Achieving K/DOQI laboratory target values for bone and mineral metabolism: an uphill battle. Am J Nephrol (2004) 24:422–426.[CrossRef][Web of Science][Medline]
  17. Davenport A, Crabtree J, Androjna C, et al. Tumour necrosis factor does not increase during routine cuprophane haemodialysis in healthy well-nourished patients. Nephrol Dial Transplant (1991) 6:435–439.[Abstract/Free Full Text]
  18. Vandenbroucke JP, von Elm E, Altman DG, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. Ann Intern Med (2007) 147:W163–W194.[Medline]
  19. Block GA, Hulbert-Shearon TE, Levin NW, et al. Association of serum phosphorus and calcium x phosphate product with mortality risk in chronic hemodialysis patients: a national study. Am J Kidney Dis (1998) 31:607–617.[Web of Science][Medline]
  20. Block GA, Klassen PS, Lazarus JM, et al. Mineral metabolism, mortality, and morbidity in maintenance hemodialysis. J Am Soc Nephrol (2004) 15:2208–2218.[Abstract/Free Full Text]
  21. Chang JJ, Concato J, Wells CK, et al. Prognostic implications of clinical practice guidelines among hemodialysis patients. Hemodial Int (2006) 10:399–407.[CrossRef][Medline]
  22. Foley RN, Parfrey PS, Harnett JD, et al. Hypocalcemia, morbidity, and mortality in end-stage renal disease. Am J Nephrol (1996) 16:386–393.[Web of Science][Medline]
  23. Ganesh SK, Stack AG, Levin NW, et al. Association of elevated serum PO(4), Ca x PO(4) product, and parathyroid hormone with cardiac mortality risk in chronic hemodialysis patients. J Am Soc Nephrol (2001) 12:2131–2138.[Abstract/Free Full Text]
  24. Jassal SV, Douglas JF, Stout RW. Prognostic markers in older patients starting renal replacement therapy. Nephrol Dial Transplant (1996) 11:1052–1057.[Abstract/Free Full Text]
  25. Kalantar-Zadeh K, Kuwae N, Regidor DL, et al. Survival predictability of time-varying indicators of bone disease in maintenance hemodialysis patients. Kidney Int (2006) 70:771–780.[CrossRef][Web of Science][Medline]
  26. Kestenbaum B, Sampson JN, Rudser KD, et al. Serum phosphate levels and mortality risk among people with chronic kidney disease. J Am Soc Nephrol (2005) 16:520–528.[Abstract/Free Full Text]
  27. Kimata N, Albert JM, Akiba T, et al. Association of mineral metabolism factors with all-cause and cardiovascular mortality in hemodialysis patients: the Japan dialysis outcomes and practice patterns study. Hemodial Int (2007) 11:340–348.[CrossRef][Medline]
  28. Leggat JE Jr, Orzol SM, Hulbert-Shearon TE, et al. Noncompliance in hemodialysis: predictors and survival analysis. Am J Kidney Dis (1998) 32:139–145.[Web of Science][Medline]
  29. Lowrie EG, Lew NL. Death risk in hemodialysis patients: the predictive value of commonly measured variables and an evaluation of death rate differences between facilities. Am J Kidney Dis (1990) 15:458–482.[Web of Science][Medline]
  30. Melamed ML, Eustace JA, Plantinga L, et al. Changes in serum calcium, phosphate, and PTH and the risk of death in incident dialysis patients: a longitudinal study. Kidney Int (2006) 70:351–357.[CrossRef][Web of Science][Medline]
  31. Menon V, Greene T, Pereira AA, et al. Relationship of phosphorus and calcium-phosphorus product with mortality in CKD. Am J Kidney Dis (2005) 46:455–463.[CrossRef][Web of Science][Medline]
  32. Noordzij M, Korevaar JC, Boeschoten EW, et al. The Kidney Disease Outcomes Quality Initiative (K/DOQI) Guideline for Bone Metabolism and Disease in CKD: association with mortality in dialysis patients. Am J Kidney Dis (2005) 46:925–932.[CrossRef][Web of Science][Medline]
  33. Port FK, Pisoni RL, Bragg-Gresham JL, et al. DOPPS estimates of patient life years attributable to modifiable hemodialysis practices in the United States. Blood Purif (2004) 22:175–180.[CrossRef][Web of Science][Medline]
  34. Rodriguez-Benot A, Martin-Malo A, Alvarez-Lara MA, et al. Mild hyperphosphatemia and mortality in hemodialysis patients. Am J Kidney Dis (2005) 46:68–77.[CrossRef][Web of Science][Medline]
  35. Saran R, Bragg-Gresham JL, Rayner HC, et al. Nonadherence in hemodialysis: associations with mortality, hospitalization, and practice patterns in the DOPPS. Kidney Int (2003) 64:254–262.[CrossRef][Web of Science][Medline]
  36. Slinin Y, Foley RN, Collins AJ. Calcium, phosphorus, parathyroid hormone, and cardiovascular disease in hemodialysis patients: the USRDS waves 1, 3, and 4 study. J Am Soc Nephrol (2005) 16:1788–1793.[Abstract/Free Full Text]
  37. Stevens LA, Djurdjev O, Cardew S, et al. Calcium, phosphate, and parathyroid hormone levels in combination and as a function of dialysis duration predict mortality: evidence for the complexity of the association between mineral metabolism and outcomes. J Am Soc Nephrol (2004) 15:770–779.[Abstract/Free Full Text]
  38. Voormolen N, Noordzij M, Grootendorst DC, et al. High plasma phosphate as a risk factor for decline in renal function and mortality in pre-dialysis patients. Nephrol Dial Transplant (2007) 22:2909–2916.[Abstract/Free Full Text]
  39. Tentori F, Hunt WC, Rohrscheib M, et al. Which targets in clinical practice guidelines are associated with improved survival in a large dialysis organization? J Am Soc Nephrol (2007) 18:2377–2384.[Abstract/Free Full Text]
  40. Young EW, Akiba T, Albert JM, et al. Magnitude and impact of abnormal mineral metabolism in hemodialysis patients in the Dialysis Outcomes and Practice Patterns Study (DOPPS). Am J Kidney Dis (2004) 44(Suppl 2):34–38.[CrossRef][Medline]
  41. Young EW, Albert JM, Satayathum S, et al. Predictors and consequences of altered mineral metabolism: the Dialysis Outcomes and Practice Patterns Study. Kidney Int (2005) 67:1179–1187.[CrossRef][Web of Science][Medline]
  42. Frimat L, Loos-Ayav C, Panescu V, et al. Early referral to a nephrologist is associated with better outcomes in type 2 diabetes patients with end-stage renal disease. Diabetes Metab (2004) 30:67–74.[Web of Science][Medline]
  43. Bommer J, Locatelli F, Satayathum S, et al. Association of predialysis serum bicarbonate levels with risk of mortality and hospitalization in the Dialysis Outcomes and Practice Patterns Study (DOPPS). Am J Kidney Dis (2004) 44:661–671.[CrossRef][Web of Science][Medline]
  44. Teng M, Wolf M, Lowrie E, et al. Survival of patients undergoing hemodialysis with paricalcitol or calcitriol therapy. N Engl J Med (2003) 349:446–456.[Abstract/Free Full Text]
  45. Tentori F, Hunt WC, Stidley CA, et al. Mortality risk among hemodialysis patients receiving different vitamin D analogs. Kidney Int (2006) 70:1858–1865.[CrossRef][Web of Science][Medline]
  46. Teng M, Wolf M, Ofsthun MN, et al. Activated injectable vitamin D and hemodialysis survival: a historical cohort study. J Am Soc Nephrol (2005) 16:1115–1125.[Abstract/Free Full Text]
  47. Cunningham J, Danese M, Olson K, et al. Effects of the calcimimetic cinacalcet HCl on cardiovascular disease, fracture, and health-related quality of life in secondary hyperparathyroidism. Kidney Int (2005) 68:1793–1800.[CrossRef][Web of Science][Medline]
  48. Suki WN, Zabaneh R, Cangiano JL, et al. Effects of sevelamer and calcium-based phosphate binders on mortality in hemodialysis patients. Kidney Int (2007) 72:1130–1137.[CrossRef][Web of Science][Medline]
  49. Block GA, Raggi P, Bellasi A, et al. Mortality effect of coronary calcification and phosphate binder choice in incident hemodialysis patients. Kidney Int (2007) 71:438–441.[CrossRef][Web of Science][Medline]
  50. Noordzij M, Korevaar JC, Bos WJ, et al. Mineral metabolism and cardiovascular morbidity and mortality risk: peritoneal dialysis patients compared with haemodialysis patients. Nephrol Dial Transplant (2006) 21:2513–2520.[Abstract/Free Full Text]
  51. Kato A, Takita T, Furuhashi M, et al. Association between seroprevalence of anti-chlamydial antibodies and long-term cardiovascular mortality in chronic hemodialysis patients. Atherosclerosis (2006) 188:120–125.[CrossRef][Web of Science][Medline]
  52. Marco MP, Craver L, Betriu A, et al. Higher impact of mineral metabolism on cardiovascular mortality in a European hemodialysis population. Kidney Int Suppl (2003) 85:S111–S114.[Medline]
  53. Young EW, Albert JM, Satayathum S, et al. Predictors and consequences of altered mineral metabolism: the Dialysis Outcomes and Practice Patterns Study. Kidney Int (2005) 67:1179–1187.[CrossRef][Web of Science][Medline]
  54. Pool PE, Seagren SC, Salel AF. Metabolic consequences of treating hypertension. Am J Hypertens (1991) 4(Pt 2):494S–502S.[Medline]
  55. Shoji T, Shinohara K, Kimoto E, et al. Lower risk for cardiovascular mortality in oral 1alpha-hydroxy vitamin D3 users in a haemodialysis population. Nephrol Dial Transplant (2004) 19:179–184.[Abstract/Free Full Text]
  56. Stack AG, Bloembergen WE. A cross-sectional study of the prevalence and clinical correlates of congestive heart failure among incident US dialysis patients. Am J Kidney Dis (2001) 38:992–1000.[Web of Science][Medline]
  57. Nakai S, Akiba T, Kazama J, et al. Effects of serum calcium, phosphorous, and intact parathyroid hormone levels on survival in chronic hemodialysis patients in Japan. Ther Apher Dial (2008) 12:49–54.[CrossRef][Web of Science][Medline]
  58. Tonelli M, Sacks F, Pfeffer M, et al. Relation between serum phosphate level and cardiovascular event rate in people with coronary disease. Circulation (2005) 112:2627–2633.[Abstract/Free Full Text]
  59. Dhingra R, Sullivan LM, Fox CS, et al. Relations of serum phosphorus and calcium levels to the incidence of cardiovascular disease in the community. Arch Intern Med (2007) 167:879–885.[Abstract/Free Full Text]
  60. Raggi P, Kleerekoper M. Contribution of bone and mineral abnormalities to cardiovascular disease in patients with chronic kidney disease. Clin J Am Soc Nephrol 2008; 3(3): 836–843.
  61. Tentori F, Blayney MJ, Albert JM et al. Mortality Risk for Dialysis Patients With Different Levels of Serum Calcium, Phosphorus, and PTH: The Dialysis Outcomes and Practice Patterns Study (DOPPS). Am J Kidney Dis 2008; 52(3): 519–530.
  62. Shanahan CM. Inflammation ushers in calcification: a cycle of damage and protection? Circulation (2007) 116:2782–2785.[Free Full Text]
  63. Ketteler M, Westenfeld R, Schlieper G, et al. Pathogenesis of vascular calcification in dialysis patients. Clin Exp Nephrol (2005) 9:265–270.[CrossRef][Medline]
  64. Lomashvili KA, Cobbs S, Hennigar RA, et al. Phosphate-induced vascular calcification: role of pyrophosphate and osteopontin. J Am Soc Nephrol (2004) 15:1392–1401.[Abstract/Free Full Text]
  65. Lomashvili KA, Garg P, Narisawa S, et al. Upregulation of alkaline phosphatase and pyrophosphate hydrolysis: potential mechanism for uremic vascular calcification. Kidney Int (2008) 73:1024–1030.[CrossRef][Web of Science][Medline]
  66. Gusbeth-Tatomir P, Covic A. Causes and consequences of increased arterial stiffness in chronic kidney disease patients. Kidney Blood Press Res (2007) 30:97–107.[CrossRef][Web of Science][Medline]
  67. London GM. Cardiovascular calcifications in uremic patients: clinical impact on cardiovascular function. J Am Soc Nephrol (2003) 14(Suppl 4):S305–S309.[Abstract/Free Full Text]
  68. Budoff MJ, Shaw LJ, Liu ST, et al. Long-term prognosis associated with coronary calcification: observations from a registry of 25,253 patients. J Am Coll Cardiol (2007) 49:1860–1870.[Abstract/Free Full Text]
  69. Amann K, Ritz E, Wiest G, et al. A role of parathyroid hormone for the activation of cardiac fibroblasts in uremia. J Am Soc Nephrol (1994) 4:1814–1819.[Abstract]
  70. Rostand SG, Drueke TB. Parathyroid hormone, vitamin D, and cardiovascular disease in chronic renal failure. Kidney Int (1999) 56:383–392.[CrossRef][Web of Science][Medline]
  71. Kosch M, Hausberg M, Vormbrock K, et al. Impaired flow-mediated vasodilation of the brachial artery in patients with primary hyperparathyroidism improves after parathyroidectomy. Cardiovasc Res (2000) 47:813–818.[Abstract/Free Full Text]
  72. Saleh FN, Schirmer H, Sundsfjord J, et al. Parathyroid hormone and left ventricular hypertrophy. Eur Heart J (2003) 24:2054–2060.[Abstract/Free Full Text]
  73. Zittermann A, Schleithoff SS, Tenderich G, et al. Low vitamin D status: a contributing factor in the pathogenesis of congestive heart failure? J Am Coll Cardiol (2003) 41:105–112.[Abstract/Free Full Text]
  74. Watson KE, Abrolat ML, Malone LL, et al. Active serum vitamin D levels are inversely correlated with coronary calcification. Circulation (1997) 96:1755–1760.[Abstract/Free Full Text]
  75. Li YC, Kong J, Wei M, et al. 1,25-Dihydroxyvitamin D(3) is a negative endocrine regulator of the renin-angiotensin system. J Clin Invest (2002) 110:229–238.[CrossRef][Web of Science][Medline]
  76. Xiang W, Kong J, Chen S, et al. Cardiac hypertrophy in vitamin D receptor knockout mice: role of the systemic and cardiac renin-angiotensin systems. Am J Physiol Endocrinol Metab (2005) 288:E125–E132.[Abstract/Free Full Text]
  77. Koch M, Lund R, Oldemeyer B, et al. Refeeding hypophosphatemia in a chronically hyperphosphatemic hemodialysis patient. Nephron (2000) 86:552.[CrossRef][Web of Science][Medline]
  78. de Francisco AL. Secondary hyperparathyroidism: review of the disease and its treatment. Clin Ther (2004) 26:1976–1993.[CrossRef][Web of Science][Medline]
  79. Henrich WL, Hunt JM, Nixon JV. Increased ionized calcium and left ventricular contractility during hemodialysis. N Engl J Med (1984) 310:19–23.[Abstract]
  80. Lee DT, Yu DS, Woo J, et al. Health-related quality of life in patients with congestive heart failure. Eur J Heart Fail (2005) 7:419–422.[Abstract/Free Full Text]
  81. Haacke C, Althaus A, Spottke A, et al. Long-term outcome after stroke: evaluating health-related quality of life using utility measurements. Stroke (2006) 37:193–198.[Abstract/Free Full Text]
  82. Gal-Moscovici A, Sprague SM, Hutchison AJ, et al. Which of the K/DOQI guidelines for bone disease in dialysis patients should be changed? Semin Dial (2007) 20:24–32.[CrossRef][Web of Science][Medline]
  83. Chertow GM, Pupim LB, Block GA, et al. Evaluation of Cinacalcet Therapy to Lower Cardiovascular Events (EVOLVE): rationale and design overview. Clin J Am Soc Nephrol (2007) 2:898–905.[Abstract/Free Full Text]
  84. Thompson SG. Why sources of heterogeneity in meta-analysis should be investigated. Br Med J (1994) 309:1351–1355.[Free Full Text]
  85. Easterbrook PJ, Berlin JA, Gopalan R, et al. Publication bias in clinical research. Lancet (1991) 337:867–872.[CrossRef][Web of Science][Medline]
  86. Chalmers I. Underreporting research is scientific misconduct. JAMA (1990) 263:1405–1408.[Abstract/Free Full Text]
Received for publication: 1. 7.08
Accepted in revised form: 7.10.08


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 HeartHome page
C J Ferro, C D Chue, R P Steeds, and J N Townend
Is lowering phosphate exposure the key to preventing arterial stiffening with age?
Heart, November 1, 2009; 95(21): 1770 - 1772.
[Abstract] [Full Text] [PDF]

Home page
 Nephrol Dial TransplantHome page
D. H. Smith, E. S. Johnson, M. L. Thorp, A. Petrik, X. Yang, and D. K. Blough
Outcomes predicted by phosphorous in chronic kidney disease: a retrospective CKD-inception cohort study
Nephrol. Dial. Transplant., August 11, 2009; (2009) gfp387v1.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
24/5/1506    most recent
gfn613v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Covic, A.
Right arrow Articles by Goldsmith, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Covic, A.
Right arrow Articles by Goldsmith, D.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?