Comparative evaluation of inductively coupled plasma mass spectrometry and a phosphatase method for lithium...

The objective of this study was to compare lithium results obtained by inductively coupled plasma mass spectrometry (ICP-MS) and a phosphatase method in paired patient serum samples and to summarize the available analytical evidence from aqueous-standard verification and routine quality-control records. A total of 171 paired serum samples were included in all method-comparison analyses, comprising 141 non-hemolyzed/non-chylous samples, 17 hemolyzed samples, and 13 chylous samples. ICP-MS analytical response was assessed using traceable aqueous lithium standards, while imprecision was summarized descriptively from routine quality-control records. Matrix-matched trueness verification, spike-recovery, standard-addition, and other serum-specific validation procedures were not performed. Although the two methods showed a strong linear association (r = 0.989), Deming regression demonstrated a constant positive bias of the phosphatase method relative to ICP-MS. At the four predefined concentrations of 0.5, 1.0, 1.2, and 2.0 mmol/L, the predicted phosphatase-method results exceeded the upper bounds of the corresponding acceptable predicted-result intervals. Manufacturer-reported LOD and LOQ values were lower for ICP-MS, but study-specific verification was not performed. These findings identify clinically relevant between-method bias under the analytical conditions of this study but do not constitute full matrix-specific validation of ICP-MS for serum lithium measurement. Prospective matrix-matched validation is required before broader clinical implementation.