Abstract
The development of equations to calculate adjusted calcium have traditionally used creatinine ≤200 µmol/L as inclusion criteria but excluded patients with diseases affecting calcium metabolism. 1 Since most laboratories are routinely reporting estimated glomerular filtration rate (eGFR) on creatinine requests from primary care, we sought to explore if the inclusion of eGFR in adjusted calcium equation derivation would improve accuracy in the primary care setting. Use of a creatinine value of ≤200 µmol/L in the inclusion criteria will likely include some patients with chronic kidney disease (CKD) stage 3–5 and who may have metabolic bone disease with disturbed calcium metabolism.
We previously showed that an adjustment equation derived for the primary care population, only correctly identified calcium status (judged by ionized calcium) in 91% of the participants. 2 One possible explanation was the inclusion of some CKD patients. Therefore, we hypothesize that in a primary care population, an equation derived using an additional inclusion criteria of eGFR >60 mL/min/1.73 m2 could outperform an equation derived using traditional Payne’s criteria.
Retrospective biochemical data for the calcium adjustment equation derivation were extracted from the laboratory database for 12 primary care Trusts for up to a three-month period in 2016. The mathematical equation derivation for the eGFR >60 population followed the same method that was described in Jassam et al. 3 With the exception of one, all 11 trusts calculated eGFR by the CKD-EPI equation. Primary care data were filtered first according to Payne’s criteria and then to exclude patients with eGFR< 60 mL/min/1.73 m2. We derived two equations for each of the 12 primary care data-sets; one equation using Payne’s criteria and the second using the modified criteria.
The performance of the equations for one site (Roche 1, Table 1) was evaluated against ionized calcium, in which both equations predicted the correct calcium status in 91% of 123 healthy individuals.
Albumin, calcium and creatinine means in the original Payne’s criteria ≤200 µmol/L and the modified criteria (eGFR> 60 mL/min/1.73 m2).
BCG: bromocresol green; BCP: bromocresol purple.
The outcome of comparing both equations for all 12 data-sets was a correlation equation of Y = 1.00 X + 0.00, P values range from 0.306 to 0.9946. Comparing creatinine, calcium and albumin means for the two patient populations showed a statistically significant difference in creatinine but not in total calcium. There was a minor but statistically significant difference in albumin (Table 1).
It is known that acid-base imbalance alters the albumin regression coefficient. 4 The agreement in the equations suggests that using a creatinine value of ≤200 µmol/L is effective in excluding CKD patients with acid-base disturbance. This is evident by a small change in n value post the application of eGFR criteria and the lack of clinically significant change in calcium and albumin means. Therefore, it is safe to conclude that Payne’s criteria remain the best approach for derivation of an adjusted calcium equation in a primary care setting until further evidence becomes available.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Ionised calcium analyser was loaned by Roche Diagnostics. The authors received no financial support for the research,authorship, and/or publication of this article.
Ethical approval
Not applicable.
Guarantor
NJ.
Contributorship
NJ planned, analysed the data and wrote the first draft. NJ, DT and JHB drafted and reviewed the article. All authors read and agreed the final article.
