ABSTRACT
Background
Calcium homeostasis is critical for cardiovascular health in chronic kidney disease (CKD). Ionized calcium is biologically active, but clinical monitoring typically relies on total or albumin-corrected calcium, which may misclassify calcium status and mask abnormalities such as hidden hypocalcemia. We examined the associations of ionized calcium levels and hidden hypocalcemia with cardiovascular outcomes in dialysis patients.
Methods
We analyzed 637 maintenance dialysis patients from the prospective, multicenter ORCHESTRA cohort with ionized calcium measurements at baseline, 6, and 12 months. To minimize reverse causality, patients experiencing events within the first 12 months were excluded from time-to-event analyses. Total, albumin-corrected, and ionized calcium were categorized as low, normal, or high, and patients were further classified into concordant or discordant groups (both normal, hidden hypocalcemia, true hypocalcemia, hidden hypercalcemia, true hypercalcemia). Cox models assessed associations of baseline and longitudinal calcium categories with major adverse cardiovascular events (MACE; cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke) and all-cause mortality.
Results
During a median follow-up of 1.1 years after the 12-month assessment, 61 MACE and 52 deaths occurred. When evaluated using 12-month averaged values, total, corrected, and ionized calcium categories were not significantly associated with MACE or mortality. However, patients with ≥2 ionized hypocalcemic measurements had a higher risk of MACE (adjusted hazard ratio (HR) 3.62; 95% confidence intervals 1.25–10.46). Hidden hypocalcemia (normal total or corrected calcium with low ionized calcium) was associated with increased MACE risk both at baseline (total × ionized HR 2.94; corrected × ionized HR 2.81) and in longitudinal analyses (corrected × ionized ≥ 2 episodes HR 4.58). No calcium category was associated with mortality.
Conclusions
Total and albumin-corrected calcium were not associated with cardiovascular outcomes. By contrast, recurrent hidden hypocalcemia was associated with excess MACE risk. These findings suggest that evaluating discordance between total and ionized calcium may provide clinically relevant information beyond conventional calcium measures in dialysis patients.
Keywords: albumin-corrected calcium, cardiovascular events, chronic kidney disease, hidden hypocalcemia, ionized calcium
KEY LEARNING POINTS.
What was known:
Current KDIGO CKD-MBD guidelines do not specify whether to monitor total, corrected, or ionized calcium, leading to possible misclassification.
This study adds:
In this multicenter prospective dialysis cohort (n = 662), low ionized calcium—but not total or corrected calcium—was associated with increased cardiovascular risk.
Potential impact:
Repeated hidden hypocalcemia (normal total/corrected but low ionized calcium) conferred the greatest risk of major cardiovascular events, supporting ionized calcium measurement for risk stratification.
INTRODUCTION
Calcium is a fundamental regulator of cardiovascular, neuromuscular, and skeletal physiology. Roughly half of circulating calcium exists in the ionized form, which is biologically active and directly involved in myocardial contractility, vascular tone, and neuromuscular excitability. By contrast, the remainder is bound to albumin or complexed with anions and does not contribute to physiologic signaling. For this reason, ionized calcium is widely regarded as the gold standard for assessing calcium status.
In patients with chronic kidney disease (CKD), disordered calcium and phosphate metabolism is central to the pathogenesis of mineral bone disease and its associated cardiovascular complications. Despite the physiologic relevance of ionized calcium, clinical practice and international guidelines on CKD–mineral and bone disorder (CKD–MBD) continue to recommend monitoring serum total calcium, often adjusted for albumin concentration [1]. This approach reflects pragmatic considerations: measurement of ionized calcium requires rapid, anaerobic sampling, specialized equipment, and incurs higher costs, whereas total and albumin-corrected calcium are more readily available in routine laboratories.
However, reliance on total or albumin-corrected calcium is increasingly questioned. Several studies have demonstrated poor concordance between total or albumin-corrected calcium and ionized calcium in CKD, with albumin-based correction formulas prone to misclassification, particularly in the presence of hypoalbuminemia, metabolic acidosis, or hyperphosphatemia [2, 3]. Such discrepancies can lead to unrecognized states such as “hidden hypocalcemia,” in which total or corrected calcium appears normal but ionized calcium is reduced, a condition linked to increased cardiovascular risk in dialysis populations [4].
Against this backdrop, it remains unclear whether the combined use of total and albumin-corrected calcium, as currently recommended for routine monitoring, adequately captures the prognostic impact of calcium dysregulation in CKD. To address this knowledge gap, we analyzed serial measurements of total, albumin-corrected, and ionized calcium in a large multicenter CKD cohort, focusing on concordance and discordance patterns, and their associations with mortality and cardiovascular events.
MATERIALS AND METHODS
Ethical considerations
The ORCHESTRA study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Institutional Review Boards of all participating centers (principal IRB: Korea University Guro Hospital, approval no. 2019GR0060). Written informed consent was obtained from all participants at the time of enrollment. The present study is a secondary analysis of the ORCHESTRA cohort data, and the requirement for additional approval was waived by the Institutional Review Board of Korea University Guro Hospital.
Study population
We conducted this study within the framework of the ORCHESTRA (Korean Dialysis Cohort for Mineral, Vascular Calcification, and Fracture) study, a nationwide, multicenter, prospective cohort that enrolled ∼900 consecutive patients receiving maintenance dialysis at 16 university-affiliated hospitals and one Veterans’ Health Service Medical Center in Korea between May 2019 and January 2021 [5]. Eligible participants were adults (≥19 years) undergoing hemodialysis or peritoneal dialysis for at least 3 months. Exclusion criteria included hospitalization at the time of enrollment, history of cancer within 5 years, cardiovascular or cerebrovascular events within 6 months, immunosuppressive therapy within 6 months, or inability to comply with study procedures. For the present analysis, we included only 662 patients with available baseline measurements of both total corrected calcium and ionized calcium.
Data collection
At enrollment, detailed demographic and clinical information, comorbidities, dialysis-related data, and medication use were collected using standardized case report forms. Laboratory parameters included hemoglobin, serum albumin, phosphate, total calcium, corrected calcium (calculated using serum albumin), and ionized calcium, as well as intact parathyroid hormone (PTH) and other biochemical variables. Total and ionized calcium values were measured in certified hospital laboratories using routine automated assays. Because participating centers reported calcium values using different units (mg/dl, mmol/l, or mEq/l), all calcium values were harmonized using standardized conversion formulas prior to analysis to ensure inter-center comparability [6]. Corrected calcium was calculated using the Payne formula: corrected calcium (mg/dl) = total calcium (mg/dl) + 0.8 × [4.0 − albumin (g/dl)] [7]. In the ORCHESTRA cohort, serial laboratory tests were scheduled every 6 months up to 24 months; for the present analysis, however, we utilized repeated measurements of corrected calcium and ionized calcium at baseline, 6 months, and 12 months only. Corrected calcium at each time point (baseline, 6 months, and 12 months) was calculated using the contemporaneously measured serum albumin concentration at the corresponding visit.
Exposure definitions
Total and corrected calcium was categorized as low (<8.5 mg/dl), normal (8.5–10.2 mg/dl), or high (>10.2 mg/dl), and ionized calcium was categorized as low (<1.12 mmol/l), normal (1.12–1.32 mmol/l), or high (>1.32 mmol/l). At baseline, patients were further stratified into concordant and discordant groups: “both normal” when total/corrected and ionized calcium were within the normal range, “true hypocalcemia” when both values were below the normal range, “true hypercalcemia” when both were above the normal range, “hidden hypocalcemia” when corrected calcium was normal but ionized calcium was low, and “hidden hypercalcemia” when corrected calcium was normal but ionized calcium was high. For longitudinal analyses, exposure was defined by repeated calcium measurements at 0, 6, and 12 months, using either the mean value across time points or the occurrence of at least one abnormal value during follow-up, and only patients with at least two calcium measurements during this period were included.
Outcomes
The primary outcomes were major adverse cardiovascular events (MACE), defined as cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke and all-cause mortality, in line with the ORCHESTRA protocol. Outcomes were ascertained prospectively through chart review and patient follow-up visits.
For time-to-event analyses, calcium exposure phenotypes were defined using measurements obtained at baseline, 6 months, and 12 months. To ensure temporal ordering between exposure and outcome, only events occurring after the 12-month assessment were included.
Statistical analysis
Baseline characteristics were summarized as mean ± standard deviation (SD) or median [interquartile range] for continuous variables and as proportions for categorical variables. Comparisons across calcium categories were performed using Student’s t-test, χ² test, or nonparametric tests, as appropriate.
Calcium exposure phenotypes were defined using measurements obtained at baseline, 6 months, and 12 months. For time-to-event analyses, follow-up was calculated from the 12-month time point to the occurrence of the outcome or censoring, thereby ensuring temporal separation between exposure assessment and outcome events.
Time-to-event outcomes were analyzed using Cox proportional hazards regression models. Hazard ratios (HRs) with 95% confidence intervals (CIs) were estimated after adjustment for potential confounders, including age, sex, body mass index, dialysis modality and duration, diabetes, hypertension, cardiovascular history, calcium and non-calcium based phosphate binder use, active and inactive vitamin D analog use (alfacalcidol, calcitriol, paricalcitol, and cholecalciferol), cinacalcet use, antiplatelet therapy, statin therapy, hemoglobin, albumin, total CO₂, intact PTH, and serum phosphate at baseline. To minimize overadjustment, covariate adjustment differed according to the primary exposure. When total or corrected calcium categories were analyzed as exposures, ionized calcium was included as an adjustment variable to account for physiologically active calcium levels. Conversely, when ionized calcium was the primary exposure, total calcium was included as a covariate. In analyses examining discordant calcium phenotypes, both total and ionized calcium were included in multivariable models to isolate the independent effect of discordant calcium status beyond absolute calcium concentrations.
To evaluate whether the association between hidden hypocalcemia and MACE differed according to key clinical factors, interaction terms were introduced between hidden hypocalcemia and selected variables (intact PTH, antiplatelet therapy, statin therapy, and use of active vitamin D analogs or cinacalcet). Significance of interaction was assessed from the coefficient of the cross-product term in the multivariable Cox model. In cases with sparse events where standard Cox regression failed to provide finite estimates, Firth’s penalized Cox regression was applied.
All analyses were performed using R version 4.4.3 (R Foundation for Statistical Computing, Vienna, Austria). A two-sided P < .05 was considered statistically significant.
Sensitivity analyses
We performed sensitivity analyses to examine whether the association between discordant calcium status and outcomes was modified by key clinical factors. Stratified analyses were conducted according to intact PTH levels, the use of active vitamin D analogs (alfacalcidol, calcitriol, or paricalcitol), and cinacalcet therapy. Interaction terms between discordant calcium categories and each factor were included in Cox models, and interaction P values were obtained from likelihood ratio tests.
In addition, competing risk analyses were performed using Fine–Gray subdistribution hazard models, treating death as a competing event for MACE.
RESULTS
Baseline characteristics of the study population
After excluding patients who experienced clinical events within the first 12 months, a total of 637 patients were enrolled in the study (Table 1). The mean age was 59.0 ± 11.9 years, and 54.6% were male. Diabetes mellitus was present in 49.0% of the cohort. At baseline, the mean total calcium level was 8.85 ± 0.78 mg/dl, and the mean ionized calcium level was 1.13 ± 0.10 mmol/l. The mean intact PTH level was 347.0 ± 681.0 pg/ml.
Table 1:
Baseline characteristics of the study population.
| Variables | N = 637a |
|---|---|
| Age, years | 59.0 ± 11.9 |
| Male sex, % | 348 (54.6%) |
| BMI, kg/m2 | 23.3 ± 3.66 |
| RRT status, % | |
| HD | 527 (82.7%) |
| PD | 110 (17.3%) |
| ESKD cause, % | |
| DM | 249 (39.1%) |
| HTN | 115 (18.1%) |
| Chronic glomerulonephritis | 139 (21.8%) |
| Polycystic kidney disease | 21 (3.3%) |
| Chronic tubulo-interstitial disease | 25 (3.9%) |
| Others | 78 (12.2%) |
| NA | 10 (1.6%) |
| DM, % | 312 (49.0%) |
| HTN, % | 572 (89.8%) |
| HBV, % | 41 (6.4%) |
| HCV, % | 17 (2.7%) |
| History of heart disease, % | 177 (27.8%) |
| History of cerebral disease, % | 53 (8.3%) |
| Medication, % | |
| Calcium based P binder | 180 (28.3%) |
| Non-calcium based P binder | 373 (58.6%) |
| Statin | 315 (49.5%) |
| Antiplatelet | 347 (54.5%) |
| Alfacalcidol | 44 (6.9%) |
| Cholecalciferol | 88 (13.8%) |
| Calcitriol | 181 (28.4%) |
| Paricalcitol | 110 (17.3%) |
| Cinacalcet | 110 (17.3%) |
| Baseline laboratory findings | |
| Hemoglobin, g/dl | 10.8 ± 1.22 |
| Serum albumin, g/dl | 3.96 ± 0.34 |
| Total CO2, mmol/l | 23.8 ± 3.16 |
| Serum phosphate, mg/dl | 5.04 ± 1.37 |
| Total calcium, mg/dl | 8.85 ± 0.78 |
| Albumin-corrected calcium, mg/dl | 8.88 ± 0.73 |
| Ionized calcium, mmol/l | 1.13 ± 0.10 |
| Intact PTH, pg/ml | 347.0 ± 681.0 |
aValues are presented as mean ± SD or n (%).
Abbreviations: RRT, renal replacement therapy; HD, hemodialysis; PD, peritoneal dialysis; ESKD, end-stage kidney disease; DM, diabetes mellitus; HTN, hypertension; HBV, hepatitis B virus; HCV, hepatitis C virus.
When classified by averaged calcium levels over 0–12 months, the distribution differed substantially between total/corrected calcium and ionized calcium (Table 2).
Table 2:
Associations of averaged corrected and ionized calcium levels with MACE and mortality.
| Total calcium | Corrected calcium | Ionized calcium | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Outcome | Group | Total N | Event N | aHRa (95% CI) | P value | Total N | Event N | aHRa (95% CI) | P value | Total N | Event N | aHRa (95% CI) | P value |
| MACE | hypocalcemia | 147 | 17 | 2.22 (0.94–5.23) | .070 | 129 | 10 | 1.35 (0.55–3.31) | .512 | 339 | 46 | 2.23 (0.94–5.31) | .070 |
| normocalcemia | 471 | 42 | 1 (ref) | – | 491 | 48 | 1 (ref) | – | 290 | 15 | 1 (ref) | – | |
| hypercalcemia | 19 | 2 | 0.14 (0.001–1.29) | .093 | 17 | 3 | 1.02 (0.19–5.55) | .984 | 8 | 0 | – (no events) | – | |
| Mortality | hypocalcemia | 147 | 19 | 1.17 (0.52–2.62) | .703 | 129 | 11 | 0.74 (0.29–1.87) | .523 | 339 | 33 | 2.39 (1.00–5.69) | .050 |
| normocalcemia | 471 | 33 | 1 (ref) | – | 491 | 39 | 1 (ref) | – | 290 | 18 | 1 (ref) | – | |
| hypercalcemia | 19 | 0 | – (no events) | – | 17 | 2 | 1.29 (0.25–6.58) | .757 | 8 | 1 | 3.36 (0.30–37.99) | .328 | |
aAdjusted HRs (aHRs) were derived from Cox proportional hazards regression models adjusted for age, sex, body mass index at baseline (M0), RRT status and duration, diabetes mellitus, hypertension, history of heart disease, history of cerebrovascular disease, use of calcium-based and non-calcium phosphate binders, antiplatelet agents, and statins, as well as baseline hemoglobin, serum albumin, total calcium, total CO₂, intact PTH, serum phosphate, and use of active vitamin D analogs (alfacalcitol, cholecalciferol, calcitriol, paricalcitol) and cinacalcet.
M0, baseline; M6, 6 months; M12, 12 months; RRT, renal replacement therapy.
Longitudinal calcium measurements and risks of MACE and mortality
During a median follow-up of 1.1 years [IQR, 1.0–2.2], 61 MACE and 52 deaths were recorded. Based on averaged total calcium, 23.1% of patients were hypocalcemic and 3.0% were hypercalcemic; neither abnormality was significantly associated with MACE or mortality. For corrected calcium, 20.3% were hypocalcemic and 2.7% hypercalcemic, with no significant association with either outcome. By contrast, when classified according to averaged ionized calcium levels, hypocalcemia showed a numerically higher risk of MACE (adjusted HR 2.23, 95% CI 0.94–5.31) and mortality (adjusted HR 2.39, 95% CI 1.00–5.69). However, these associations did not reach conventional statistical significance (Table 2).
Frequency of abnormal calcium and risks of clinical outcomes
When the frequency of abnormal calcium measurements across baseline, 6 months, and 12 months was considered, repeated abnormalities of total calcium showed no significant association with MACE, but patients with a single episode of total hypocalcemia had a significantly lower mortality risk (adjusted HR 0.21, 95% CI 0.06–0.75). Repeated corrected calcium abnormalities were not associated with either MACE or mortality. By contrast, for ionized calcium, patients with two or more hypocalcemic measurements had a markedly higher risk of MACE (adjusted HR 3.62, 95% CI 1.25–10.46), whereas neither single nor repeated hypercalcemia was significantly associated with MACE or mortality (Table 3).
Table 3:
Associations of the frequency of abnormal total, corrected, and ionized calcium measurements (M0, M6, and M12) with MACE and mortality.
| Total calcium | Corrected calcium | Ionized calcium | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Outcome | Group | Total N | Event N | aHRa (95% CI) | P value | Total N | Event N | aHRa (95% CI) | P value | Total N | Event N | aHRa (95% CI) | P value |
| MACE | ≥2 hypocalcemia | 124 | 16 | 1.88 (0.75–4.73) | .181 | 126 | 10 | 1.88 (0.68–5.20) | .224 | 299 | 48 | 3.62 (1.25–10.46) | .017 |
| 1 hypocalcemia | 121 | 10 | 0.98 (0.40–2.43) | .969 | 126 | 15 | 2.15 (0.97–4.75) | .058 | 93 | 5 | 1.17 (0.29–4.62) | .826 | |
| all normal | 291 | 25 | 1 (reference) | – | 280 | 25 | 1 (reference) | – | 156 | 5 | 1 (reference) | – | |
| 1 hypercalcemia | 67 | 9 | 1.47 (0.59–3.64) | .403 | 69 | 8 | 1.75 (0.67–4.58) | .254 | 29 | 1 | 0.60 (0.06–5.93) | .665 | |
| ≥2 hypercalcemia | 10 | 1 | 0.14 (0.001–1.34) | .098 | 11 | 3 | 3.90 (0.65–23.42) | .137 | 7 | 0 | – (no events) | – | |
| Mortality | ≥2 hypocalcemia | 124 | 18 | 0.88 (0.34–2.23) | .780 | 126 | 15 | 1.22 (0.46–3.19) | .692 | 299 | 27 | 1.44 (0.54–3.87) | .468 |
| 1 hypocalcemia | 121 | 3 | 0.21 (0.06–0.75) | .016 | 126 | 6 | 0.63 (0.23–1.72) | .369 | 93 | 8 | 1.58 (0.54–4.57) | .395 | |
| all normal | 291 | 27 | 1 (reference) | – | 280 | 22 | 1 (reference) | – | 156 | 9 | 1 (reference) | – | |
| 1 hypercalcemia | 67 | 3 | 0.55 (0.15–1.98) | .361 | 69 | 6 | 1.65 (0.55–4.98) | .373 | 29 | 3 | 1.77 (0.36–8.76) | .485 | |
| ≥2 hypercalcemia | 10 | 0 | – (no events) | – | 11 | 2 | 2.59 (0.41–16.50) | .314 | 7 | 1 | 16.82 (1.71–165.55) | .016 | |
aAdjusted HRs (aHRs) were adjusted for age, sex, BMI at baseline (M0), RRT status and duration, diabetes, hypertension, history of cardiovascular disease, phosphate binder use, antiplatelet therapy, statin therapy, hemoglobin, albumin, total calcium, total CO₂, intact PTH, serum phosphate, and use of active vitamin D analogs and cinacalcet.
M0, baseline; M6, 6 months; M12, 12 months; RRT, renal replacement therapy.
Discordance between corrected and ionized calcium and clinical outcomes
Given the discrepant associations of total/corrected and ionized calcium with outcomes, we further classified patients into concordant and discordant categories and assessed their relationships with clinical outcomes using baseline data and repeated longitudinal measurements.
At baseline, when classified by total calcium × ionized calcium, 315 patients had both calcium parameters within the normal range, six had true hypercalcemia, 143 had true hypocalcemia, and 161 were classified as hidden hypocalcemia (normal total but low ionized calcium). When classified by corrected calcium × ionized calcium, 315 were both normal, eight had true hypercalcemia, 133 had true hypocalcemia, and 172 were classified as hidden hypocalcemia.
In multivariable analyses of discordance groups defined at baseline (Table 4), hidden hypocalcemia was consistently associated with a higher risk of MACE (adjusted HR 2.94, 95% CI 1.16–7.42 for total × ionized; adjusted HR 2.81, 95% CI 1.11–7.16 for corrected × ionized). True hypocalcemia demonstrated differing patterns according to the classification method (total calcium × ionized versus corrected calcium × ionized), and was not consistently associated with MACE across models. For all-cause mortality, no clear association was observed between calcium discordance categories and risk. Groups with hypercalcemia had very small numbers and showed no significant associations with either outcome.
Table 4:
Associations of baseline total, corrected, and ionized calcium concordance/discordance groups with clinical outcomes.
| Total calcium x ionized calcium discordance | Corrected calcium x ionized calcium discordance | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Outcome | Group | Total N | Event N | aHRa (95% CI) | P value | Total N | Event N | aHRa (95% CI) | P value |
| MACE | true hypocalcemia | 143 | 13 | 1.58 (0.44–5.73) | .486 | 133 | 13 | 5.55 (1.40–21.92) | .014 |
| hidden hypocalcemia | 161 | 26 | 2.94 (1.16–7.42) | .022 | 172 | 26 | 2.81 (1.11–7.16) | .030 | |
| both normal | 315 | 21 | 1 (reference) | – | 315 | 21 | 1 (reference) | – | |
| hidden hypercalcemia | 10 | 1 | 2.62 (0.26–26.68) | .417 | 8 | 1 | 2.23 (0.22–22.64) | .498 | |
| true hypercalcemia | 6 | 0 | – (no events) | – | 8 | 0 | – (no events) | – | |
| Mortality | true hypocalcemia | 143 | 17 | 0.66 (0.21–2.04) | .468 | 133 | 13 | 0.43 (0.12–1.52) | .193 |
| hidden hypocalcemia | 161 | 7 | 0.39 (0.13–1.18) | .095 | 172 | 11 | 0.49 (0.17–1.36) | .172 | |
| both normal | 315 | 27 | 1 (reference) | – | 315 | 27 | 1 (reference) | – | |
| hidden hypercalcemia | 10 | 1 | 1.96 (0.20–19.08) | .561 | 8 | 0 | – (no events) | – | |
| true hypercalcemia | 6 | 0 | – (no events) | – | 8 | 1 | 5.59 (0.48–65.56) | .171 | |
aAdjusted HRs (aHRs) were adjusted for age, sex, BMI at baseline (M0), RRT status and duration, diabetes, hypertension, history of cardiovascular disease, phosphate binder use (calcium- and non-calcium based), antiplatelet therapy, statin therapy, hemoglobin, albumin, total calcium, total CO₂, intact PTH, serum phosphate, and use of active vitamin D analogs (alfacalcidol, cholecalciferol, calcitriol, paricalcitol) and cinacalcet.
RRT, renal replacement therapy.
When longitudinal measurements were considered (Table 5), patients with hidden hypocalcemia at two or more time points had a significantly higher risk of MACE when classified by corrected calcium × ionized calcium (adjusted HR 4.58, 95% CI 1.25–16.77), and a similar but non-significant trend when classified by total calcium × ionized calcium (adjusted HR 2.69, 95% CI 0.79–9.19). No significant association was observed between longitudinal calcium discordance categories and all-cause mortality. Analyses of true or hidden hypercalcemia were not performed in Table 5 because patient numbers were too small for meaningful statistical evaluation (true hypercalcemia: 16 with one occurrence and only two with two or more; hidden hypercalcemia: 16 with one and only four with two or more).
Table 5:
Association of the frequency of calcium concordance/discordance (M0–M12) with clinical outcomes.
| Total calcium x ionized calcium discordance | Corrected calcium x ionized calcium discordance | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Outcome | Group | Total N | Event N | aHRa (95% CI) | P value | Total N | Event N | aHRa (95% CI) | P value |
| MACE | all normal | 135 | 5 | 1 (reference) | – | 134 | 4 | 1 (reference) | – |
| 1 hidden hypocalcemia | 146 | 17 | 2.16 (0.68–6.82) | .191 | 136 | 17 | 3.21 (0.92–11.22) | .068 | |
| ≥2 hidden hypocalcemia | 159 | 28 | 2.69 (0.79–9.19) | .115 | 163 | 31 | 4.58 (1.25–16.77) | .022 | |
| Mortality | all normal | 135 | 7 | 1 (reference) | – | 134 | 7 | 1 (reference) | – |
| 1 hidden hypocalcemia | 146 | 14 | 1.43 (0.47–4.32) | .525 | 136 | 11 | 1.08 (0.32–3.65) | .898 | |
| ≥2 hidden hypocalcemia | 159 | 12 | 1.35 (0.36–5.12) | .659 | 163 | 15 | 1.35 (0.35–5.12) | .663 | |
| MACE | all normal | 135 | 5 | 1 (reference) | – | 134 | 4 | 1 (reference) | – |
| 1 true hypocalcemia | 100 | 12 | 4.05 (0.91–18.05) | .066 | 108 | 15 | 8.85 (1.64–47.85) | .011 | |
| ≥2 true hypocalcemia | 88 | 11 | 5.04 (0.93–27.27) | .061 | 86 | 7 | 6.62 (0.95–46.41) | .057 | |
| Mortality | all normal | 135 | 7 | 1 (reference) | – | 134 | 7 | 1 (reference) | – |
| 1 true hypocalcemia | 100 | 5 | 0.73 (0.14–3.97) | .718 | 108 | 7 | 1.13 (0.26–4.83) | .868 | |
| ≥2 true hypocalcemia | 88 | 11 | 1.33 (0.19–9.39) | .772 | 86 | 10 | 2.81 (0.45–17.66) | .272 | |
aAdjusted HRs (aHRs) were adjusted for age, sex, BMI at baseline (M0), RRT status and duration, diabetes, hypertension, history of cardiovascular disease, phosphate binder use (calcium- and non-calcium based), antiplatelet therapy, statin therapy, hemoglobin, albumin, total calcium, total CO₂, intact PTH, serum phosphate, and use of active vitamin D analogs (alfacalcidol, cholecalciferol, calcitriol, paricalcitol) and cinacalcet.
RRT, renal replacement therapy.
Kaplan–Meier analyses demonstrated that, whether classified by total or corrected calcium, the cumulative incidence of MACE increased in proportion to the number of hidden hypocalcemia episodes, whereas mortality curves showed no significant separation (Fig. 1).
Figure 1:
Kaplan–Meier curves for cumulative incidence of MACE and all-cause mortality according to the frequency of hidden hypocalcemia. Patients were stratified by the number of hidden hypocalcemia episodes (0, 1, or ≥2) based on either total calcium (a and b) or corrected calcium (c and d). (a) MACE with total calcium, (b) mortality with total calcium, (c) MACE with corrected calcium, and (d) mortality with corrected calcium.
Baseline characteristics according to the frequency of hidden hypocalcemia
To explore baseline factors that might contribute to the increased risk of MACE observed in patients with hidden hypocalcemia, baseline characteristics were compared according to the number of hidden hypocalcemia episodes (Table 6). Patients with more frequent hidden hypocalcemia exhibited progressively higher body mass index and a greater prevalence of diabetes mellitus. Male sex proportion decreased across categories. The use of statins and antiplatelet agents was likewise more common in groups with a higher number of hidden hypocalcemia episodes, consistent with a greater underlying cardiovascular risk profile. Conversely, serum albumin showed a stepwise decrease across categories, whereas corrected calcium and total CO₂ were progressively higher. Regarding vitamin D analogs, alfacalcitol use decreased, whereas cholecalciferol and calcitriol use increased significantly across categories, while paricalcitol and cinacalcet use did not show significant trends. Other demographic factors, including age, sex, dialysis modality, and blood pressure, showed no significant linear trend.
Table 6:
Baseline characteristics according to the number of hidden hypocalcemia episodes.
| Variables | 0Hidden hypocalcemiaa | 1Hidden hypocalcemiaa | ≥2Hidden hypocalcemiaa | P for trend |
|---|---|---|---|---|
| n = 252 | n = 136 | n = 163 | ||
| Age, years | 58.9 ± 11.42 | 58.95 ± 11.54 | 57.63 ± 13.11 | .316 |
| Male sex, % | 131 (52.0) | 59 (43.4) | 69 (42.3) | .044 |
| BMI, kg/m2 | 22.92 ± 3.20 | 23.24 ± 3.88 | 23.89 ± 3.90 | .009 |
| HD among RRT, % | 205 (81.3) | 115 (84.6) | 126 (77.3) | .376 |
| DM, % | 110 (43.7) | 68 (50.0) | 92 (56.4) | .011 |
| HTN, % | 220 (87.3) | 119 (87.5) | 152 (93.3) | .072 |
| HBV, % | 15 (6.0) | 10 (7.4) | 12 (7.4) | .553 |
| HCV, % | 6 (2.4) | 6 (4.4) | 4 (2.5) | .854 |
| History of heart disease, % | 63 (25.0) | 42 (30.9) | 46 (28.2) | .408 |
| History of cerebral disease, % | 16 (6.3) | 15 (11.2) | 12 (7.4) | .581 |
| Medication, % | ||||
| Calcium based P binder | 84 (33.3) | 37 (27.4) | 38 (23.8) | .034 |
| Non-calcium based P binder | 156 (62.4) | 79 (59.0) | 100 (62.5) | .952 |
| Statin | 106 (42.2) | 66 (49.3) | 102 (64.2) | <.001 |
| Antiplatelet | 112 (44.4) | 82 (61.2) | 100 (62.5) | <.001 |
| Alfacalcitol | 30 (12.4) | 5 (3.8) | 3 (1.9) | <.001 |
| Cholecalciferol | 4 (1.7) | 26 (19.5) | 50 (31.2) | <.001 |
| Calcitriol | 46 (19.1) | 46 (34.8) | 71 (44.4) | <.001 |
| Paricalcitol | 46 (18.8) | 21 (15.8) | 27 (16.9) | .581 |
| Cinacalcet | 40 (16.4) | 28 (21.2) | 33 (20.6) | .254 |
| Laboratory finding | ||||
| Hemoglobin, g/dl | 10.70 ± 1.25 | 10.88 ± 1.21 | 10.88 ± 1.15 | .101 |
| Serum albumin, g/dl | 4.02 ± 0.34 | 3.95 ± 0.32 | 3.91 ± 0.33 | .001 |
| Total CO2, mmol/l | 23.61 ± 3.35 | 23.53 ± 3.15 | 24.49 ± 2.84 | .009 |
| Serum phosphate, mg/dl | 5.09 ± 1.47 | 5.07 ± 1.28 | 4.98 ± 1.29 | .461 |
| Total calcium, mg/dl | 8.80 ± 0.86 | 8.86 ± 0.82 | 8.92 ± 0.56 | .133 |
| Albumin-corrected calcium, mg/dl | 8.79 ± 0.81 | 8.91 ± 0.75 | 8.99 ± 0.47 | .004 |
| Intact PTH, pg/ml | 327.69 ± 237.17 | 507.71 ± 1393.05 | 296.85 ± 247.29 | .894 |
aValues are presented as mean ± SD or n (%). P for trend was calculated across the three categories.
Abbreviations: RRT, renal replacement therapy; HD, hemodialysis; DM, diabetes mellitus; HTN, hypertension; HBV, hepatitis B virus; HCV, hepatitis C virus.
Effect modification by PTH and medications
We next examined whether the associations of discordant calcium groups with cardiovascular events were modified by intact PTH levels or medication use, including antiplatelet agents, statins, active vitamin D analogs, and cinacalcet. No significant interaction was observed with high PTH (adjusted HR 1.30, 95% CI 0.35–4.84; P = .701), statin use (adjusted HR 0.79, 95% CI 0.21–2.96; P = .732), active vitamin D use (adjusted HR 4.21, 95% CI 0.96–18.43; P = .057) or cinacalcet use (adjusted HR 1.49, 95% CI 0.27–8.29; P = .649).
Competing risk sensitivity analyses
Fine–Gray competing risk analyses were performed to account for death as a competing event. Overall, these analyses yielded results consistent in direction with the primary Cox proportional hazards models. However, estimates were numerically unstable in hypercalcemic subgroups because of sparse cardiovascular events. Accordingly, Fine–Gray results are presented as sensitivity analyses in Supplementary Tables S1–S4.
DISCUSSION
In this longitudinal study of patients with CKD, we compared the prognostic implications of repeated measurements of total, albumin-corrected, and ionized calcium. When averaged calcium values over 0–12 months were analyzed, neither total, corrected, nor ionized calcium was significantly associated with MACE or all-cause mortality in multivariable models. However, when calcium concordance was examined, hidden hypocalcemia (normal total or corrected but low ionized calcium) was consistently associated with a higher MACE risk in both baseline and longitudinal analyses. Patients with recurrent hidden hypocalcemia had a higher cardiovascular risk profile, but no significant effect modification was observed according to PTH level, statin use, vitamin D analog use, or cinacalcet therapy.
Our results extend prior observations on the limitations of both total and albumin-corrected calcium in CKD. Albumin-corrected calcium has long been used in clinical practice as a surrogate for ionized calcium, but several studies have demonstrated its poor diagnostic performance, particularly in CKD where hypoalbuminemia, hyperphosphatemia, and metabolic acidosis are common [2, 3]. Payne’s original correction formula specifically excluded patients with renal disease, yet it remains widely applied in nephrology [7]. Subsequent investigations consistently show that albumin-corrected calcium frequently misclassifies true calcium status, underestimating hypocalcemia in the setting of low bicarbonate and overestimating hypercalcemia when albumin is reduced [2, 8]. In our study, corrected calcium was paradoxically higher in patients with repeated hidden hypocalcemia despite low ionized calcium. Recent data in dialysis and advanced CKD populations confirm that albumin-corrected calcium is inferior even to uncorrected total calcium in approximating ionized calcium [9]. Furthermore, a large US dialysis cohort showed that adjustment for serum albumin markedly changed the relationship between total calcium and mortality, indicating substantial confounding and highlighting the limited prognostic value of both total and albumin-corrected calcium [10]. Consistent with these observations, neither total nor corrected calcium showed significant associations with cardiovascular outcomes in our analysis.
The clinical relevance of hidden calcium abnormalities—defined as discordance between corrected and ionized calcium—has been increasingly recognized. A Japanese single-center cohort of incident hemodialysis patients reported that hidden hypocalcemia was strongly associated with all-cause mortality and cardiovascular hospitalization, whereas apparent hypocalcemia was not [4]. By contrast, a US dialysis cohort demonstrated that hidden hypercalcemia, rather than corrected calcium–defined hypercalcemia, was associated with higher mortality risk [11]. Our study extends these observations by demonstrating that recurrent hidden hypocalcemia, rather than a single averaged ionized calcium measurement, was associated with a significantly increased risk of MACE. This underscores the importance of longitudinal assessment of calcium status rather than reliance on corrected calcium alone.
The mechanisms linking hidden hypocalcemia to cardiovascular events are likely multifactorial. Ionized calcium plays a central role in myocardial excitation–contraction coupling and electrical stability, and its reduction predisposes to QT prolongation, arrhythmogenesis, and impaired diastolic function [12–15]. CKD-related acidosis further reduces ionized calcium by shifting the equilibrium between bound and free calcium, thereby exacerbating the discrepancy with corrected calcium. Consistent with these pathophysiological effects, our cohort showed that patients with repeated hidden hypocalcemia had a heavier burden of cardiovascular risk factors at baseline—including higher body mass index, more diabetes, lower albumin, and greater use of statins and antiplatelet agents—providing a clinical milieu that could amplify the adverse consequences of low ionized calcium. Importantly, hidden hypocalcemia is easily overlooked in practice because corrected calcium is often interpreted as “normal,” leading to under-recognition and delayed or less targeted clinical management [4]. Thus, both biological effects of low ionized calcium and clinical inertia due to misclassification may contribute to the observed risk.
Earlier CKD-MBD guidelines (2009 KDIGO) recommended measuring ionized calcium whenever feasible, recognizing its physiological relevance and diagnostic value [16]. The 2017 KDIGO update, however, referred only to “serum calcium” for routine monitoring and stated that ‘mild and asymptomatic hypocalcemia (including that associated with calcimimetic treatment) can be tolerated to avoid inappropriate calcium loading’ [1], without specifying whether ionized or total calcium should be used. This more general wording leaves some uncertainty regarding the optimal biochemical target and the preferred method of assessment. In our analysis, the prognostic signal was primarily observed in discordant calcium states rather than in absolute calcium concentrations alone. These observations suggest that a monitoring strategy relying mainly on total or corrected calcium, as often occurs in clinical practice, may not fully capture clinically relevant hypocalcemia. It may therefore be reasonable to consider more frequent ionized calcium measurement or improved surrogate markers that better reflect biologically active calcium, particularly in patients with hypoalbuminemia, acidosis, or unexplained cardiovascular risk.
This prospective, multicenter cohort included a large number of dialysis patients, enhancing statistical robustness and generalizability. Repeated measurements of corrected and ionized calcium enabled dynamic assessment beyond baseline values. Most prior studies in CKD relied on total or albumin-corrected calcium [17–19], which showed associations between hypercalcemia and adverse outcomes but are prone to misclassification. By contrast, outcome-based analyses of ionized calcium remain rare, particularly in dialysis populations [4]. Our study addresses this gap by providing longitudinal prognostic evidence focused on ionized calcium.
Nonetheless, several limitations should be noted. The small numbers in discordant subgroups reduced statistical power, and hypercalcemia was uncommon, reflecting current dialysis practice that avoids overt hypercalcemia. Ionized calcium measurement is technically demanding and susceptible to pre-analytical variability (e.g. sample handling, pH), which may attenuate associations. In addition, although dialysis-related treatment parameters—such as vascular access type, dialysis modality, dialysate calcium concentration, and dialysis dose—may influence cardiovascular outcomes, these variables were not included in the final multivariable models because of substantial missingness (∼14% to over 50%), which would have considerably reduced the analyzable sample size and increased the risk of selection bias. Finally, despite multivariable adjustment and sensitivity analyses (including stratification by PTH, vitamin D analogs, and cinacalcet), residual confounding from factors such as nutritional or inflammatory status cannot be excluded.
In conclusion, total or corrected calcium showed no significant association with mortality or cardiovascular events across multiple analytic strategies. Although ionized calcium alone did not demonstrate a statistically significant independent association, recurrent hidden hypocalcemia was consistently associated with an increased risk of cardiovascular events. These findings highlight the clinical importance of recognizing patients with hidden hypocalcemia who may otherwise be misclassified as normocalcemic when relying solely on total or corrected calcium.
Supplementary Material
ACKNOWLEDGEMENTS
We thank the ORCHESTRA investigator group, clinical centers, Korea University Guro Hospital, Young Joo Kwon, MD(PI), Shin Young Ahn, MD (SubPI), Gang Jee Ko, MD, Ji Eun Kim, MD, Eun Jung Cho, MD, Donghyuk Kang, MD, So Yeon Bae, and Ji Yeon Baek, RN, Gachon University Gil Medical Center, Ji Yong Jung, MD, Yeonsoo Lee, RN, the Catholic University of Korea Daejeon St. Mary’s Hospital, Yu Ah Hong, MD, Leeji Bae, RN, the Catholic University of Korea Eunpyeong St. Mary’s Hospital, Bum Soon Choi, MD, Tae Hyun Ban, MD, Yujung Choi, RN, the Catholic University of Korea Uijeongbu St. Mary’s Hospital, Sun Ae Yoon, MD, Eunhwa Ju, RN, Gangnam Severance Hospital, Jong Hyun Jhee, MD, Hyeong Cheon Park, MD, Suin Kang, RN, Kyung Hee University East-West Neo Medical Center, Yang Gyun Kim, MD, Ju-Young Moon, MD, Sang Ho Lee, MD, Yunmi Go, RN, Kyung Hee University Medical Center, Hyeon Seok Hwang, MD, Kyung Hwan Jeong, MD, Keimyung University Dongsan Medical Center, Yaerim Kim, MD, Kyubok Jin, MD, Sungeun Park, RN, Pusan National University Hospital, Hyo Jin Kim, MD, Seunghee Ji, RN, Inje University Busan Paik Hospital, Yunmi Kim, MD, Yeong Hoon Kim, MD, Youngeun Oh, RN, CHA Medical School Bundang CHA Medical Center, So-Young Lee, MD, Dong-Ho Yang, MD, Seoul National University Hospital, Kook-Hwan Oh, MD, Suin Lee, RN, Chonnam National University Hospital, Chang Seong Kim, MD, Eun Hui Bae, MD, Soo Wan Kim, MD, Jaegyeong Jeong, RN, Hallym University Kangnam Sacred Heart Hospital, Hayne Cho Park, MD, Young-Ki Lee, MD, Hyejung Seo, RN, Hanyang University Seoul Hospital, Jun Sung Park, MD, Gheun-Ho Kim, MD, Veterans Health Service Medical Center, Dong-Young Lee, MD.
Contributor Information
Ji Eun Kim, Department of Internal Medicine, Korea University Guro Hospital, Seoul, Korea; Department of Internal Medicine, Korea University College of Medicine, Seoul, Korea.
Donghyuk Kang, Department of Internal Medicine, Bucheon St Mary’s Hospital, The Catholic University College of Medicine, Seoul, Korea.
Seung Min Song, Department of Internal Medicine, Korea University Guro Hospital, Seoul, Korea.
Min Woo Kang, Department of Internal Medicine, Korea University Guro Hospital, Seoul, Korea.
Hyo Jin Kim, Department of Internal Medicine, Korea University Guro Hospital, Seoul, Korea.
Eun Jung Cho, Department of Internal Medicine, Korea University Guro Hospital, Seoul, Korea.
Shin Young Ahn, Department of Internal Medicine, Korea University Guro Hospital, Seoul, Korea; Department of Internal Medicine, Korea University College of Medicine, Seoul, Korea.
Gang Jee Ko, Department of Internal Medicine, Korea University Guro Hospital, Seoul, Korea; Department of Internal Medicine, Korea University College of Medicine, Seoul, Korea.
Young Joo Kwon, Department of Internal Medicine, Korea University Guro Hospital, Seoul, Korea; Department of Internal Medicine, Korea University College of Medicine, Seoul, Korea.
the O.R.C.H.E.S.T.R.A. Study Investigators:
Young Joo Kwon, Shin Young Ahn, Gang Jee Ko, Ji Eun Kim, Eun Jung Cho, Donghyuk Kang, So Yeon Bae, Ji Yeon Baek, Ji Yong Jung, Yeonsoo Lee, Yu Ah Hong, Leeji Bae, Bum Soon Choi, Tae Hyun Ban, Yujung Choi, Sun Ae Yoon, Eunhwa Ju, Jong Hyun Jhee, Hyeong Cheon Park, Suin Kang, Yang Gyun Kim, Ju-Young Moon, Sang Ho Lee, Yunmi Go, Hyeon Seok Hwang, Kyung Hwan Jeong, Yaerim Kim, Kyubok Jin, Sungeun Park, Hyo Jin Kim, Seunghee Ji, Yunmi Kim, Yeong Hoon Kim, Youngeun Oh,, So-Young Lee, Dong-Ho Yang, Kook-Hwan Oh, Suin Lee, Chang Seong Kim, Eun Hui Bae, Soo Wan Kim, Jaegyeong Jeong, Hayne Cho Park, Young-Ki Lee, Hyejung Seo, Jun Sung Park, Gheun-Ho Kim, and Dong-Young Lee
FUNDING
This study was supported by a cooperative research fund from the Korean Society of Nephrology (Kyowa Kirin Korea (2018) and DKSH Korea) (Grant numbers I1803531 and I1803861). The funding organizations had no role in the design or conduct of the study; collection, management, analysis, or interpretation of the data; or preparation, review, or approval of the manuscript.
AUTHORS’ CONTRIBUTIONS
Study concept and design: Y.J.K.; formal analysis: J.E.K., M.W.K.; methodology and data collection: J.E.K., D.K., S.M.S., M.W.K., H.J.K., E.J.C., S.Y.A., G.J.K., Y.J.K., and the ORCHESTRA Study Investigators; drafting of the manuscript: J.E.K.; critical revision of the manuscript for important intellectual content: all authors; supervision: Y.J.K. Each author contributed important intellectual content during manuscript drafting or revision and agrees to be personally accountable for the individual’s own contributions and to ensure that questions pertaining to the accuracy or integrity of any portion of the work, even one in which the author was not directly involved, are appropriately investigated and resolved, including with documentation in the literature if appropriate.
DATA AVAILABILITY STATEMENT
We confirm that the data supporting the findings of this study are available within the article and its Supplementary Materials. The raw data are not publicly available because they contain personal and clinical information that could compromise participant privacy.
CONFLICT OF INTEREST STATEMENT
Y.J.K. received research grants (grant numbers I1803531) administered by the Korean Society of Nephrology with funding from Kyowa Kirin Korea and DKSH Korea Ltd). S.Y.A. received research grants (grant number I1803861) from Kyowa Kirin Korea and DKSH Korea Ltd. The authors declare that there are no other relevant financial relationships or conflicts of interest to disclose.
References
- 1. Ketteler M, Block GA, Evenepoel P et al. Executive summary of the 2017 KDIGO Chronic Kidney Disease-Mineral and Bone Disorder (CKD-MBD) Guideline Update: what’s changed and why it matters. Kidney Int. 2017;92:26–36. 10.1016/j.kint.2017.04.006 [DOI] [PubMed] [Google Scholar]
- 2. Gauci C, Moranne O, Fouqueray B et al. Pitfalls of measuring total blood calcium in patients with CKD. J Am Soc Nephrol. 2008;19:1592–8. 10.1681/ASN.2007040449 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Kaku Y, Ookawara S, Miyazawa H et al. New method for the approximation of corrected calcium concentrations in chronic kidney disease patients. Ther Apher Dial. 2016;20:46–52. 10.1111/1744-9987.12351 [DOI] [PubMed] [Google Scholar]
- 4. Yamaguchi S, Hamano T, Doi Y et al. Hidden hypocalcemia as a risk factor for cardiovascular events and all-cause mortality among patients undergoing incident hemodialysis. Sci Rep. 2020;10:4418. 10.1038/s41598-020-61459-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Ahn SY, Ko GJ, Hwang HS et al. Understanding the Korean Dialysis Cohort for Mineral, Vascular Calcification, and Fracture (ORCHESTRA) Study: design, method, and baseline characteristics. Kidney Blood Press Res. 2024;49:326–35. [DOI] [PubMed] [Google Scholar]
- 6. Burtis CA, Ashwood ER, Tietz NW. Tietz Textbook of Clinical Chemistry. 6th edn. St., Louis, MO: W.B. Saunders: 2018;. [Google Scholar]
- 7. Payne RB, Little AJ, Williams RB et al. Interpretation of serum calcium in patients with abnormal serum proteins. BMJ. 1973;4:643–6. 10.1136/bmj.4.5893.643 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Goransson LG, Skadberg O, Bergrem H. Albumin-corrected or ionized calcium in renal failure? What to measure?. Nephrol Dial Transplant. 2005;20:2126–9. 10.1093/ndt/gfh988 [DOI] [PubMed] [Google Scholar]
- 9. Law MM, Smith JD, Schneider HG et al. Misclassification of calcium status in end-stage kidney disease using albumin-adjusted calcium levels. Nephrology. 2021;26:725–32. 10.1111/nep.13910 [DOI] [PubMed] [Google Scholar]
- 10. Rivara MB, Ravel V, Kalantar-Zadeh K et al. Uncorrected and albumin-corrected calcium, phosphorus, and mortality in patients undergoing maintenance dialysis. J Am Soc Nephrol. 2015;26:1671–81. 10.1681/ASN.2014050472 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Obi Y, Mehrotra R, Rivara MB et al. Hidden hypercalcemia and mortality risk in incident hemodialysis patients. J Clin Endocrinol Metabol. 2016;101:2440–9. 10.1210/jc.2016-1369 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Gromadzinski L, Januszko-Giergielewicz B, Pruszczyk P. Hypocalcemia is related to left ventricular diastolic dysfunction in patients with chronic kidney disease. J Cardiol. 2014;63:198–204. 10.1016/j.jjcc.2013.08.003 [DOI] [PubMed] [Google Scholar]
- 13. Connor TB, Rosen BL, Blaustein MP et al. Hypocalcemia precipitating congestive heart failure. N Engl J Med. 1982;307:869–72. 10.1056/NEJM198209303071407 [DOI] [PubMed] [Google Scholar]
- 14. Newman DB, Fidahussein SS, Kashiwagi DT et al. Reversible cardiac dysfunction associated with hypocalcemia: a systematic review and meta-analysis of individual patient data. Heart Fail Rev. 2014;19:199–205. 10.1007/s10741-013-9371-1 [DOI] [PubMed] [Google Scholar]
- 15. Mozos I. Laboratory markers of ventricular arrhythmia risk in renal failure. Biomed Res Int. 2014;2014:1. 10.1155/2014/509204 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Kidney Disease: Improving Global Outcomes CKDMBDWG. KDIGO clinical practice guideline for the diagnosis, evaluation, prevention, and treatment of chronic kidney disease-mineral and bone disorder (CKD-MBD). Kidney Int Suppl. 2009;:S1–130. [DOI] [PubMed] [Google Scholar]
- 17. 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;67:1179–87. 10.1111/j.1523-1755.2005.00185.x [DOI] [PubMed] [Google Scholar]
- 18. 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;52:519–30. 10.1053/j.ajkd.2008.03.020 [DOI] [PubMed] [Google Scholar]
- 19. Block GA, Klassen PS, Lazarus JM et al. Mineral metabolism, mortality, and morbidity in maintenance hemodialysis. J Am Soc Nephrol. 2004;15:2208–18. 10.1097/01.ASN.0000133041.27682.A2 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
We confirm that the data supporting the findings of this study are available within the article and its Supplementary Materials. The raw data are not publicly available because they contain personal and clinical information that could compromise participant privacy.