Why patient specimens, not just QC

CLSI EP26, User Evaluation of Acceptability of a Reagent Lot Change (2nd ed., 2022), gives laboratories a patient-sample-based statistical protocol for deciding whether a candidate reagent lot is acceptable before it replaces the current lot. The protocol is built around one laboratory-defined number: the critical difference (CD), the largest lot-to-lot difference the laboratory judges medically acceptable for that analyte and concentration range. There is no universal CD; each laboratory sets it per assay, and it can be tighter near a medical decision point than elsewhere in the range.

Commutability is the property that makes a QC or calibrator material behave, on a given method, the way a patient specimen with the same analyte concentration would behave. When a control or calibrator material is noncommutable, a lot change can shift the reading on that processed material by a different amount, or in a different direction, than it shifts a real plasma or serum specimen. Reference: Miller, Myers, and Rej, writing on why commutability matters, identify commutability failure as a recognized source of discordance between method-comparison results built on processed materials and results built on patient specimens. That is the reason patient specimens remain the reference standard for judging whether a lot effect is real.

General stock and expiration control applies across laboratory materials. Whether a lot change needs verification is assay-, material-, manufacturer-, and procedure-specific; do not treat a chemistry reagent comparison as a universal protocol for every reagent, calibrator, control, antibody, test strip, disk, cuvette, culture medium, or stain.

The regulatory floor sits underneath the EP26 patient-sample protocol, not in place of it. Under 42 CFR 493.1256(d), a laboratory testing quantitative nonwaived analytes must run two different control concentrations at least once each day of patient testing, with a defined mean and standard deviation for each control lot. Under 42 CFR 493.1256, controls must also be run and must meet acceptability criteria before patient testing resumes after a complete reagent change, major preventive maintenance, or replacement of a critical instrument part. QC is required and it is a real check, but the opening scenario shows what it does not check: a matrix-dependent bias that only shows up in a real patient specimen.

QC verifies the analyzer and reagent system are behaving as expected on control material; a patient-sample comparison verifies the new lot agrees with the old lot on the material that actually matters, the patient's own specimen.