Abstract
Calcium can be measured as ionised (Ca-ionised) or albumin-adjusted total calcium (Ca-albumin). Current clinical guidelines predominantly utilise Ca-albumin, despite Ca-ionised being the gold standard. Discrepancies can occur between these measurement modalities and can lead to clinical dilemmas. It remains unclear how large these discrepancies are in older patients. This study investigated the discrepancies between Ca-ionised and Ca-albumin in geriatric patients.
This is an observational study of all geriatric patients (n = 876) in the Jeroen Bosch Hospital (January 2018 and January 2021) in whom both Ca-ionised and Ca-albumin were measured. Misclassification of calcaemic state (i.e. low, normal or high) was calculated (percentages), the measure of agreement was described using Cohen’s Kappa and for the continuous data Pearson’s correlation coefficient was used. Relevant categories of age and renal function were considered for effect modification effects and studied by interaction terms in a regression model.
In one-third of the measurements, there was a misclassification. Ca-albumin measurements failed to identify 28% of hypocalcaemia. In 3.5%, hypercalcemia based on Ca-albumin was not confirmed by Ca-ionised. The correlation coefficient between Ca-ionised and Ca-albumin was 0.743 (P = 0.01) and measure of agreement by Kappa was 0.213 (P < 0.001). In the oldest old (≥ 85 years) and patients with eGFR <30 ml/min/1.73 m2 ,the agreement by Kappa was lower, with values of 0.192 and 0.104, respectively.
There is a discrepancy between Ca-albumin and Ca-ionised in one-third of the geriatric patients, leading to clinical dilemmas. In the oldest old and patients with renal dysfunction, this problem is most pronounced.
Key Points
In one-third of geriatric patients, ionised calcium and albumin-adjusted calcium do not correspond.
This may present challenges in interpretation and management of patients with calcium disturbance.
This problem is the largest in the oldest old (≥ 85 years) and in patients with renal dysfunction.
Introduction and rationale
Older patients often undergo many homeostatic changes. Among these, a change in calcium concentration can occur due to several underlying processes in acute scenarios. For example, in acute scenarios, the kidneys might adjust their function to retain or excrete more calcium, influencing blood calcium levels. Also, changes in the absorption of calcium from the gastrointestinal tract can impact blood calcium levels. Lastly, inflammation may affect calcium homeostasis by influencing the activity of cells involved in bone remodelling and calcium regulation [1]. Patients with hypercalcemia can present with altered mental state, polydipsia, nausea, vomiting, constipation and even arrhythmias [1]. Conversely, hypocalcaemia can cause muscle cramps, spasms, neuropsychiatric changes as well as arrhythmias [1]. Given the potential life-threatening nature of these situations, accurate measurements of calcium levels are crucial for appropriate action.
To give some background on calcium in the human body, it should be noted that approximately 99% of total calcium in the body is present in the human skeleton. The remaining 1% circulates in plasma (Ca-total), of which about 40% is bound to proteins, predominantly albumin (Ca-albumin), 10% to small anions such as bicarbonate and phosphate and the remaining 50% is the free calcium fraction (Ca-ionised) [1]. As this free calcium fraction is the physiologically active form, measurement of Ca-ionised is considered gold standard [2]. Although the measurement of Ca-total is more commonly used in clinical practice, it can be influenced by protein levels and factors affecting protein binding, such as pH and free fatty acids. In patients with low albumin levels, Ca-albumin is used to correct the calcium concentration for hypoalbuminemia and to more accurately reflect calcaemic state.
Although Ca-ionised is considered the gold standard, in clinical practice, Ca-albumin is used frequently for practical reasons owing to their less demanding requirements in terms of costs and equipment. Also, current Dutch guidelines use Ca-albumin thresholds for acute calcium disturbances, and treatment advises are given based on Ca-albumin, not Ca-ionised [3].
In clinical practice, there can be large discrepancies between measurements of Ca-ionised and Ca-albumin, which can lead to a treatment dilemma. To give an example: in one patient, Ca-ionised was 1.15 mmol/l (reference: 1.12–1.32 mmol/l), whereas Ca-albumin was 2.82 mmol/l (reference: 2.15–2.60 mmol/l). According to Dutch guidelines, this patient is considered hypercalcaemic and requires calcium-lowering therapy. Ca-ionised, on the other hand, is reaching the lower limit and calcium-lowering therapy could cause hypocalcaemia. This discrepancy also causes a diagnostic dilemma when measuring parathyroid hormone (PTH). An elevated PTH can be interpreted as the patient being in a hyperparathyreoid state resulting in raised calcium levels (Ca-albumin), a pathological finding. On the contrary, an elevated PTH can be a reaction to low calcium levels (Ca-ionised), a physiological finding. These divergent treatment strategies could lead to hazardous situations if results are misinterpreted. Therefore, there is a clear need for clinicians to have well-defined guidelines for managing calcium disturbances.
Certain populations have been identified where these aforementioned issues are particularly relevant. In one study of Slomp et al., it was shown that in an intensive care population (ICU) population, Ca-albumin measurement could lead to false-positive hypercalcemia and false-negative hypocalcaemia [4]. Another study concluded that Ca-ionised is better than Ca-albumin to detect hypercalcemia in patients with multiple myeloma [5]. In older patients, only old studies were found on this specific dilemma [6, 7].
In the older patients, several factors can contribute to calcium imbalances. For instance, underlying conditions like heart failure can alter water balance, and pH levels may fluctuate depending on the underlying cause. Also, older patients are likely to use a large number of medications that can possibly influence acid–base homeostasis. Coupled with their generally lower resilience, the need for precise calcium measurements is particularly crucial in this population.
The objective of this study was to investigate the potentially important difference between the two calcium measurement modalities and to verify whether observations from clinical practice could be confirmed or not. First, to quantify the magnitude of the discrepancies between Ca-ionised and Ca-albumin, and second to discern whether certain factors influence this relationship.
Methods
Study design
This is an observational retrospective cohort study, in which discrepancies between Ca-albumin and Ca-ionised were studied in geriatric patients (age ≥ 70 years). To study this, concentrations of Ca-albumin and Ca-ionised had to be measured within one patient in the same venous blood sample.
Participants and setting
Geriatric patients were included, with inclusion criteria: (1) were treated by a geriatrician; (2) who underwent laboratory testing of Ca-total, albumin and Ca-ionised within one venous sample measured and (3) aged ≥70 years. In the case of repeated measurements of a single patient, only the first measurement was included. Exclusion criteria were: (1) patients with known renal replacement therapy (dialysis); (2) patients who signed an opt out in the hospital system for use of data for research goals. Patients of ≥85 years old and with renal dysfunction (eGFR ≤30 m ml/min/1.73 m2) were considered most vulnerable to homeostatic disbalance.
The study period was from January 2018 to January 2021. Geriatric patients, both inpatients and outpatients, from the Jeroen Bosch Hospital, a teaching hospital in the Netherlands, were included. Figure 1 shows the inclusion of the patients.
Flowchart of inclusion.
Variables and data collection procedure
Data on the main variables, namely Ca-total, albumin and Ca-ionised were extracted from the laboratory information system (LIS). Figure 1 shows the steps that were taken before analysis.
From the included patients, additional data on patients’ characteristics namely age and sex were collected by the same system. Additional laboratory data related to this study question were also collected, namely pH, PTH, creatinine and estimated glomerular filtration rate (eGFR), the potential confounders or effect modifiers.
For Ca-ionised, Ca-total and albumin, from which Ca-albumin was calculated, the same venous blood sample was used. Data on pH were included at the same time as the venous blood draw was taken, but collected from venous blood gas or arterial blood gas samples. For PTH measurements of the included patients, blood draws within 14 days of the Ca-total, albumin and Ca-ionised collection were used to include PTH. PTH was included for secondary analysis. No patients signed an opt-out declaration. All these data were transferred to statistical package for social sciences (SPSS) version 27 for analysis (IBM Corp. Released 2017. IBM SPSS Statistics for Windows, Version 27.0. Armonk, NY: IBM Corp.)
Laboratory methods
Ca-total and Ca-ionised concentrations were measured by the department of Clinical Chemistry of the Jeroen Bosch Hospital as part of routine standard diagnostic procedure/patient care. Ca-total was measured based on the Arsezano III complex formation, and measured biochromatically at 658 nm and 694 nm, using Chemistry XPT analysers (Siemens Healthineers). Ca-ionised blood gas analysis was performed on the Rapidpoint 500 (Siemens Healthineers) using an Ag/AgCl electrode.
Calcium
Ca-ionised was considered normal between 1.12 and 1.32 mmol/l; Ca-albumin was considered normal between 2.15 and 2.60 mmol/l. Ca-albumin was calculated using a local hospital formula (Ca-albumin = Ca-total + (34—albumin) × 0.016, based on albumin concentrations measured by bromocresol purple).
Albumin was analysed using Colorimetric assay using bromochresol purpol (Siemens Atellica).
Renal function
Creatinine was analysed using enzymatic colorimetric assay (Siemens Atellica), and eGFR was calculated using Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) [8].
PTH was analysed using two-site sandwich immunoassay, a third generation PTH assay (Siemens Atellica).
Ca-ionised is the physiologically active form and was considered the gold standard in our study [9].
Quantitative variables
All continuous data were used as such in the analyses but for better understanding categories were also made for reporting the population characteristics. Renal function was classified into four categories: ≤ 30 ml/min/1.73 m2; 31–60 ml/min/1.73 m2; 61–90 ml/min/1.73 m2 and > 90 ml/min/1.73 m2. Age was segmented into two categories: 70–84 years and ≥ 85 years old, also known as ‘the oldest old’ [10].
Next, categories were made of Ca-ionised and Ca-albumin in line with clinical interpretation of the data. For Ca-ionised, hypocalcaemia was classified as <1.12 mmol/l and hypercalcemia as >1.32 mmol/l. For Ca-albumin, hypocalcaemia was classified as <2.15 mmol/l and hypercalcemia as >2.60 mmol/l. Values between were labelled as normal range.
Power calculation
Based on the pathophysiology, there should be no difference between Ca-ionised and Ca-albumin, as it should reflect the same electrolyte, but based on our hypotheses there is a difference. Therefore, differences in mean were used as starting point for power calculation. Sample size was calculated by an a priori matched paired t-test. An effect size (Cohen-D) of 0.2 was considered clinically relevant. Considering that beta = 20% and alfa = 0.05, an estimated number of 199 patients was needed to find such a clinically relevant difference, as calculated by G*power [11]. To enhance power for our secondary analyses, we used a period of 3 years, which exceeded the number of patients that was pre-calculated.
Statistical methods
Descriptives are given for the patients’ characteristics, including mean ± SD and % per category for age, and % for sex. Mean ± SD was given for all laboratory results, while renal function and calcium were also calculated as % per category.
The primary objective focused on examining the discrepancies between Ca-ionised and Ca-albumin measurements.
First analyses of categories were made. We calculated the percentages of two scenarios:
1) Agreement of calcaemic state;
2) Misclassification of calcaemic state.
Agreement for the different subcategories was measured by Cohen’s Kappa.
To further quantify this relationship, correlation coefficient on the continuous data between Ca-ionised and Ca-albumin was calculated using Pearson’s Correlation. R2 was calculated as effect size, representing the proportion of the variance in one measure that can be predicted from the other measure [12]. It should be noted that very high correlations should be expected (>0.9) since both measures are taken within one patient, from one venous blood sample, of one electrolyte of interest. In clinical practice, both measures are interchangeable for that reason.
For the secondary aim, effect modification was tested for age (cut of 85 years) and eGFR (cut of 30 ml/min/1.73 m2) based on clinically relevant scores, to test whether the results are the same or different in a subgroup of more vulnerable patients who might have larger homeostatic disbalance. Effect modification effects were studied by a regression model, and if interaction terms were significant (P < 0.10), results were reported in strata. Although pH and PTH might also be of interest, the missing values of these variables were too large to further analyse on effect modification. Only 244 pH measures and 58 PTH measures were available, with the risk of selection bias considered as too high.
SPSS version 27 was used for all analyses. Significance was set at P < 0.05.
Ethical considerations
This study was approved by the Dutch local medical ethics committee (registration number NW2021–92). As this was a retrospective observational laboratory study, no informed consent was necessary as approved by the ethical review board. All identifying variables were removed from the database to ensure patient privacy, and databases were secured and accessible for the investigators only.
Results
Participants
A total of 1,024 samples were extracted from LIS that measured Ca-total, albumin and Ca-ionised in patients treated by a geriatrician between 1 January 2018 and 1 January 2021 (Figure 1). Of these, 876 met all inclusion criteria and were included for further analyses.
Descriptive data
Patient characteristics are shown in Table 1. A total of 876 patients were analysed, 45% male, mean age 84 years with 51% in the oldest old category (≥ 85 years).
Population characteristics.
| Total patients | n = 876 (100%) | |
|---|---|---|
| Sex | Male | 45.0% |
| Age | Mean (SD) | 84 (6.3) |
| < 85, years | 427 (48.8%) | |
| ≥ 85, years | 449 (51.2%) | |
| Laboratory results | Mean (SD) | |
| Ca-total, mmol/l | 2.35 (0.18) | |
| Albumin, g/l | 33.3 (5.0) | |
| Ca-ionised, mmol/l | 1.15 (0.09) | |
| Ca-albumin, mmol/l | 2.37 (0.17) | |
| pH (n = 244) | 7.43 (0.07) | |
| PTH (n = 58), pmol/l | 11.2 (12.2) | |
| eGFR by CKD-EPI (n = 856) | Mean (SD) | 57.9 (22.8) |
| ≤ 30 ml/min/1.73 m2 | 14.3% | |
| 31–60 ml/min/1.73 m2 | 36.6% | |
| 61–90 ml/min/1.73 m2 | 44.2% | |
| > 90 ml/min/1.73 m2 | 5.0% | |
| % Hypocalcaemia | ||
| Ca-ionised | 32.2% | |
| Ca-albumin | 5.0% | |
| % Hypercalcaemia | ||
| Ca-ionised | 3.3% | |
| Ca-albumin | 6.4% |
| Total patients | n = 876 (100%) | |
|---|---|---|
| Sex | Male | 45.0% |
| Age | Mean (SD) | 84 (6.3) |
| < 85, years | 427 (48.8%) | |
| ≥ 85, years | 449 (51.2%) | |
| Laboratory results | Mean (SD) | |
| Ca-total, mmol/l | 2.35 (0.18) | |
| Albumin, g/l | 33.3 (5.0) | |
| Ca-ionised, mmol/l | 1.15 (0.09) | |
| Ca-albumin, mmol/l | 2.37 (0.17) | |
| pH (n = 244) | 7.43 (0.07) | |
| PTH (n = 58), pmol/l | 11.2 (12.2) | |
| eGFR by CKD-EPI (n = 856) | Mean (SD) | 57.9 (22.8) |
| ≤ 30 ml/min/1.73 m2 | 14.3% | |
| 31–60 ml/min/1.73 m2 | 36.6% | |
| 61–90 ml/min/1.73 m2 | 44.2% | |
| > 90 ml/min/1.73 m2 | 5.0% | |
| % Hypocalcaemia | ||
| Ca-ionised | 32.2% | |
| Ca-albumin | 5.0% | |
| % Hypercalcaemia | ||
| Ca-ionised | 3.3% | |
| Ca-albumin | 6.4% |
SD, standard deviation; Ca-total, total calcium; Ca-ionised, ionised calcium; Ca-albumin, albumin-adjusted calcium
Population characteristics.
| Total patients | n = 876 (100%) | |
|---|---|---|
| Sex | Male | 45.0% |
| Age | Mean (SD) | 84 (6.3) |
| < 85, years | 427 (48.8%) | |
| ≥ 85, years | 449 (51.2%) | |
| Laboratory results | Mean (SD) | |
| Ca-total, mmol/l | 2.35 (0.18) | |
| Albumin, g/l | 33.3 (5.0) | |
| Ca-ionised, mmol/l | 1.15 (0.09) | |
| Ca-albumin, mmol/l | 2.37 (0.17) | |
| pH (n = 244) | 7.43 (0.07) | |
| PTH (n = 58), pmol/l | 11.2 (12.2) | |
| eGFR by CKD-EPI (n = 856) | Mean (SD) | 57.9 (22.8) |
| ≤ 30 ml/min/1.73 m2 | 14.3% | |
| 31–60 ml/min/1.73 m2 | 36.6% | |
| 61–90 ml/min/1.73 m2 | 44.2% | |
| > 90 ml/min/1.73 m2 | 5.0% | |
| % Hypocalcaemia | ||
| Ca-ionised | 32.2% | |
| Ca-albumin | 5.0% | |
| % Hypercalcaemia | ||
| Ca-ionised | 3.3% | |
| Ca-albumin | 6.4% |
| Total patients | n = 876 (100%) | |
|---|---|---|
| Sex | Male | 45.0% |
| Age | Mean (SD) | 84 (6.3) |
| < 85, years | 427 (48.8%) | |
| ≥ 85, years | 449 (51.2%) | |
| Laboratory results | Mean (SD) | |
| Ca-total, mmol/l | 2.35 (0.18) | |
| Albumin, g/l | 33.3 (5.0) | |
| Ca-ionised, mmol/l | 1.15 (0.09) | |
| Ca-albumin, mmol/l | 2.37 (0.17) | |
| pH (n = 244) | 7.43 (0.07) | |
| PTH (n = 58), pmol/l | 11.2 (12.2) | |
| eGFR by CKD-EPI (n = 856) | Mean (SD) | 57.9 (22.8) |
| ≤ 30 ml/min/1.73 m2 | 14.3% | |
| 31–60 ml/min/1.73 m2 | 36.6% | |
| 61–90 ml/min/1.73 m2 | 44.2% | |
| > 90 ml/min/1.73 m2 | 5.0% | |
| % Hypocalcaemia | ||
| Ca-ionised | 32.2% | |
| Ca-albumin | 5.0% | |
| % Hypercalcaemia | ||
| Ca-ionised | 3.3% | |
| Ca-albumin | 6.4% |
SD, standard deviation; Ca-total, total calcium; Ca-ionised, ionised calcium; Ca-albumin, albumin-adjusted calcium
Main results
In 67.6% of the calcium measurements, Ca-albumin corresponded with Ca-ionized, whereas in 32.4% of the cases, the measurements indicated a different calcaemic state and therefore a different diagnosis (Figure 2). Agreement by Cohen’s Kappa was rather low (0.213).
Correlation of ionised calcium and albumin-adjusted calcium in a geriatric population. Ca-ionised, ionised calcium; Ca-albumin, albumin-adjusted calcium. Statistics: Agreement by Cohen’s Kappa 0.213 (P < 0.001).
The correlation coefficient between the continuous data on Ca-ionised and Ca-albumin was 0.743 (P = 0.01) with an R2 of 55.2%, indicating that one measure contributes to 55.2% of the other measure.
Differences in subgroups
For age and renal function, interaction terms were significant in the analyses (both P < 0.001), and therefore data for subgroups are shown.
In the subgroup of the oldest old (≥ 85 years), measure of agreement of Ca-albumin and Ca-ionised is worse (Kappa 0.192 (P < 0.001) in ≥ 85 years versus Kappa 0.233 (P < 0.001) in < 85 years) although the percentage of misclassification is similar (33% in aged ≥ 85 years versus 32% in aged < 85 years, Figure 3).
Correlation of ionised calcium and albumin-adjusted calcium in patients <85 years and ≥ 85 years. Ca-ionised, ionised calcium; Ca-albumin, albumin-adjusted calcium. Statistics: Agreement by Cohen’s Kappa 0.233 (P < 0.001) in patients <85 years and 0.192 (P < 0.001) in patients ≥85 years.
In patients with eGFR ≤ 30 ml/min/1.73 m2, measure of agreement between Ca-albumin and Ca-ionised was lower with Kappa 0.104 (P < 0.001) versus Kappa 0.196 (P < 0.001) in patients with eGFR > 30 ml/min/1.73 m2, and the percentage of misclassification is much larger (45% in renal dysfunction versus 31% in eGFR > 30 ml/min/1.73 m2, Figure 4).
Correlation of ionised calcium and albumin-adjusted calcium in patients with estimated glomeral filtration rate ≤ 30 ml/min/1.73 m2 and > 30 ml/min/1.73 m2. Abbreviations: Ca-ionised, ionised calcium; Ca-albumin, albumin-adjusted calcium. Statistics: Agreement by Cohen’s Kappa 0.104 (P < 0.001) in patients with eGFR ≤ 30 ml/min/1.73 m2 and 0.196 (P < 0.001) in patients with eGFR > 30 mL/min/1.73 m2.
Discussion
In this study, we showed that in a cohort of geriatric patients in the Netherlands, there is a discrepancy in different calcium measurements, with rather low agreement between Ca-albumin and Ca-ionised. In approximately one-third of the cases, Ca-albumin resulted in an incorrect diagnosis and could therefore cause a clinical dilemma. In the oldest old (≥ 85 years old) and patients with renal dysfunction (eGFR ≤ 30 ml/min/1.73 m2), disagreement is larger.
As mentioned before, Ca-albumin has been used to predict calcaemic state mainly for practical and financial reasons, but in recent years technical improvement in laboratories has made measurement of Ca-ionised more accessible. Since both measurements are now more easily available, the clinical dilemma has resurfaced. In two very old studies of Sorva et al., this discrepancy in a geriatric population has been described in the early 90s [6, 7]. And despite more accurate ways of adjusting Ca-total for albumin, the correlation has merely improved since then. In addition, literature suggests that Ca-total more precisely captures the alterations in albumin binding associated with changes in albumin concentration and the consistent presence of unbound free calcium ions than Ca-albumin does [13]. For clinical practice, these results are worrisome when realising guidelines still use Ca-albumin thresholds and therefore can be interpreted and managed wrongly, possibly harming the patient. Therefore, we plead to adjust guidelines by using Ca-ionised for acute calcium disturbances for the older patient in order to prevent incorrect handling of calcium findings.
It is important to consider how these differences can exist from a technical point of view. First, there are different methods to measure albumin. The measured albumin concentration is dependent on the type of laboratory assay that is used (bromocresol green versus bromocresol purple), with a difference of up to 3–6 g/l [14]. Therefore, using the standard Payne formula (based on bromocresol green) for calculating the Ca-albumin will lead to erroneous results. It is essential that Ca-albumin is calculated using a formula based on the local albumin measurement method. Even so, as is shown in a previous study done in the ICU, several formulas were developed, but none were accurate enough to predict the correct calcium concentration and resulted in a substantial percentage of misclassification [4]. An alternative formula for Payne’s formula was developed for the very old hospitalised patients, but was an insufficient surrogate for Ca-ionised in a Finnish geriatric population [15]. Also in our study, we used a locally adjusted formula but still misclassified one-third of calcium findings. Second, classification of different calcaemic states, namely hypocalcaemia and hypercalcemia, hinges on the use of established reference intervals. However, it is important to recognise that these reference values for both Ca-ionised and Ca-total are not universally standardised and could be suboptimal. Lower reference limits for Ca-ionised range from 1.08 mmol/l to 1.15 mmol/l per laboratory, and Ca-total from 2.15 mmol/l to 2.20 mmol/l. Even a slight deviation of reference intervals has considerable influence on the classification of calcaemic states. Therefore, it should be noted that misclassification might be confusing, but not necessarily clinically relevant. As calcium is a continuous variable, borderline values will be classified into one calcaemic state, depending on reference intervals, but will not always bear large diagnostic or therapeutic consequences. Third, when interpreting the results, the analytical variation should be taken into consideration for Ca-ionised and Ca-total, and for Ca-albumin also the analytical variation of albumin is essential. For the measurement of Ca-total, the analytical variation is larger than the biological variation. This substantial analytical variation could be a considerable contributor to the suboptimal correlation coefficient and the misclassifications of calcaemic states.
Although our study confirms observations from clinical practice in an observational design, our results should be interpreted in the light of some limitations. First, it is single centered and as mentioned, each hospital and laboratory uses different types of laboratory assays and reference intervals, which means all findings should be considered in the light of the described reference intervals and analytical methods. The discrepancies and disagreements we found, however, were larger than the variance of measure, considering both measures were taken within one patient at the same time. Our results might therefore be either an over- or underestimation, and replication studies in other centres are recommended. To follow on, our studied population included only a geriatric population. Although this was our population of interest as geriatric patients are frequently underrepresented [16], it is unclear to what extent the observed discrepancies are also prevalent in other age categories or in non-frail older patients. Possible effect modifiers that were not studied are albumin, pH and PTH because of the influence on calcium binding and calcium levels. Due to epidemiologic reasons, we decided not to continue with this secondary analysis because of a high risk of selection bias. This would be an interesting question for future research.
Conclusion
In a cohort of geriatric patients, there is a discrepancy between the measurement of calcium with Ca-ionised and Ca-albumin, resulting in one-third of incorrect diagnoses when using Ca-albumin with risk of underdiagnosing hypocalcaemia and overdiagnosing hypercalcemia. In the oldest old and in patients with renal dysfunction, this discrepancy is more substantial. Current guidelines often use Ca-albumin, which may lead to difficulties in interpretation and management of calcium disturbances. Although these differences are at least confusing for the clinician, not all small differences are probably clinically relevant. Further studies are needed on the clinical implication of this finding.
Declaration of Conflicts of Interest
None.
Declaration of Sources of Funding
None.
Comments