A-090 Impact of Changing Total Calcium Methodology on Observed Hypercalcemia Prevalence at an Academic Health System

Abstract

Abstract Background The UC Davis Health clinical chemistry laboratory implemented a new total laboratory automation (Roche Diagnostics, Indianapolis, IN) system in February 2022—replacing a prior system (Beckman Coulter, Brea, CA) that was in place since 2007. In August 2022, primary care and endocrinology providers contacted laboratory leadership to report an increase in patients with calcium values flagged as abnormally elevated, resulting in increased communications from patients to primary care providers and increased referrals to endocrinology. This change in testing methodology produced a shift in the upper reference interval limit from 10.5 mg/dL to 10.0 mg/dL. The new testing methodology determined total calcium photometrically vs by indirect potentiometry via the outgoing method. We hypothesize photometric-based total calcium measurements to be more sensitive. Methods We conducted a retrospective study to investigate the reported increase in hypercalcemic values—our electronic health system’s (EHR) EPIC’s self-service reporting tool (“SlicerDicer”) was utilized for data acquisition and visualization. Although the reference interval had already been verified prior to implementation, a “normal study” study was conducted to re-confirm manufacturer derived intervals. One hundred thirty-six samples from apparently healthy adults (age ≥ 18 years) were banked for this study. Parathyroid hormone (PTH) values were then measured to confirm levels were within normal limits. Additional observational data was also collected for PTH levels pre- and post-new automation line implementation. Quality control (QC) and proficiency testing (PT) data was also reviewed during this time. Results In the four months preceding implementation of the new automation line, 1211 patients had total calcium values flagged as elevated, compared to 5146 patients during a similar post implementation time period (302.8/month vs 1286.5/month), or an increase of 324.9%. During these same time periods, patients with abnormal PTH were observed at 270.8/month pre-implementation compared to 393/month post (45.2% increase). An increase in patients with concurrently performed calcium and PTH was also observed (30.3%; 583.8/month vs 760.8/month), though a greater increase was seen in patients with concurrently elevated calcium and abnormal PTH (294.5%; 22.8/month vs 89.8/month). One hundred twenty samples with normal PTH values were included in the total calcium normal study, which reported a mean (SD) value of 9.3 (0.33) mg/dL, confirming the manufacturer-derived intervals of 8.7–10 mg/dL. No biases, shifts, or trends were observed in PT or QC data in the four months preceding and post analyzer change. Conclusion Increased hypercalcemic events post-automation line implementation is likely not due to analytical error or inappropriate reference intervals, but rather due to increased sensitivity of methodology. Hypercalcemic cases determined by our new methodology were corroborated by corresponding abnormal PTH values. The perceived increased sensitivity could be due to known limitations of indirect potentiometry (i.e., lack of selectivity, difficulty detecting all forms of calcium, etc.). Further investigation is warranted to determine the overall impact of changing this one assay on the health system.