Total Serum Calcium Spike-Recovery Experiments to Investigate Bones of Contention in Clinical Laboratories

BACKGROUND: To help assess patient safety risks associated with ongoing differences in total calcium assays following bias shifts, we performed spike-recovery experiments within our clinical chemistry external quality assessment (EQA) schemes to evaluate bias and recovery against a reference measurement procedure (RMP). METHODS: Aqueous calcium chloride dihydrate was spiked (0.30/0.60 mmol/L; 1.20/2.40 mg/dL) into pooled human serum. Bias was assessed against RMP-assigned targets and recovery against added spikes. Minimum analytical performance specifications (APS) for standard measurement uncertainty (1.36%) defined acceptable bias/recovery. RESULTS: Commercially available analytical principles across analytical platforms did not achieve adequate bias against RMP-assigned targets and recovery against added spikes for total calcium in 3 EQA specimens. Roche Cobas (5-nitro-5'-methyl-1,2-bis(o-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid [NM-BAPTA]), Roche Cobas Pro (NM-BAPTA), and Beckman AU (Arsenazo III) methods exhibited better bias and long-term stability. The Abbott Alinity (Arsenazo III) method exhibited notable negative bias (-2.33% to -3.10%), apparently coinciding with the introduction of Abbott Calcium2. Negative bias was present to a lesser extent in the Abbott Architect (Arsenazo III). The Siemens Atellica (cresolphthalein complexone) assay exhibited notable proportional over recovery (110.2% to 112.0%). There was some positive bias in the Siemens Atellica (Arsenazo III) and QuidelOrtho (dry slide/sensor) methods. CONCLUSIONS: Differences across analytical principles/platforms and difficulties achieving minimum APS for total calcium were shown within our EQA schemes, notably negative bias for Abbott Alinity and Architect (Arsenazo III), over-recovery for Siemens Atellica (cresolphthalein complexone) and positive bias for Siemens Atellica (Arsenazo III). Understanding the impact on clinical decision-making and patient outcomes is imperative for managing any patient safety risks.