Notes · 3 October 2026 · 5 min read

Reagent lot changeover: a checklist for Chemistry

A new reagent lot is a change to the method. The eight steps from the delivery dock to first patient use, and the record each one leaves for an assessor.

A reagent lot changeover is done properly when five facts can be shown afterwards: the lot was identified at receipt, it was kept apart until accepted, it was verified against the outgoing lot before patient use, someone signed the acceptance, and the date it went into use is recorded. ISO 15189:2022 clause 6.6 asks for all five. The checklist below puts them in the order the work happens.

At the door

  1. Read the lot number off the delivery note and the box, and compare them. A lot that differs from the one in use is a new lot, whatever the product name says. Record date received, lot, expiry, quantity, supplier, delivery-note number and condition.
  2. Check it against the acceptance register before it reaches the shelf. The register says whether this lot has been tested, is in testing, or has never been seen. Never seen means testing has to be started, not assumed.
  3. Keep it apart. A lot that is not yet accepted goes on a separate shelf or carries a clear label. Clause 6.6.4 asks that unaccepted stock is segregated from accepted stock.

On the bench

  1. Run the new lot beside the current one. Same day, same analyser, quality control at two or more levels and a small set of patient samples across the reportable range. Where the calibrator lot is changing too, change one thing at a time if stock allows, so a shift can be attributed.
  2. Compare against criteria set in advance. The laboratory decides the allowable difference between lots for each assay before the comparison is run. A difference judged after the numbers are in is not an acceptance criterion.
  3. Sign it. Testing complete is not the same as accepted. The record needs a name and a date, and the lot stays apart until it has them.

Into service

  1. Record the Date in Use on the day. This is the date an investigation will need months later, when an EQA result or a run of QC has moved and someone asks which lot was on the analyser. A date reconstructed from memory is the weakest link in most lot records.
  2. Watch the first runs. Internal QC in the days after the change is the last check. A step in QC that lines up with the Date in Use is a lot effect until shown otherwise.

Where changeovers go wrong

Rarely at the bench. The comparison is usually done well. The failures are at the edges: a new lot shelved as if it were the old one because the product name matched, a lot used over a weekend while its sign-off sat in a tray, and a Date in Use written in a week later from a guess. Each of those is a gap between two records kept in two places by two people.

How LotWarden handles the edges

LotWarden does not run the comparison or set the criteria. It closes the gaps around it. At Receive a delivery it reads the lot from the scanned docket and, in Chemistry, the Lot acceptance cross-check gives each lot a verdict against the register: accepted, awaiting sign-off, testing in progress, rejected or not on register. Lots needing attention go onto an investigation worklist with the action required. Date in Use records the first-use date through Excel, signed and dated. It warns when QC acceptance is not recorded and refuses a lot whose QC is recorded as not acceptable. Lot history then shows the whole sequence for any lot in one place.

Plainly answered

How many samples are needed to verify a new reagent lot?

ISO 15189:2022 does not set a number. The laboratory sets it per assay in its own procedure, based on how much lot-to-lot variation the assay is known to show and how small a shift matters clinically. Many laboratories use quality control at two or more levels with a small set of patient samples across the range, and more for assays with a history of lot effects.

Does a new shipment of the same lot need testing again?

Clause 6.6.3 refers to each new lot or shipment. A new shipment of a lot already accepted has been through different transport and storage, so the laboratory should decide in its procedure what check it applies. For most products that is a check of condition on receipt and quality control at first use rather than a full comparison.

What if the new lot fails the comparison?

The lot is rejected, kept apart and labelled, and the supplier is contacted. The record should show the results, the decision, who made it and what happened to the stock. If the current lot is running low, that is a supply problem to raise early, not a reason to release the new lot.

Why does the Date in Use matter so much?

Because every later investigation depends on it. When results shift, the first question is what changed and when. A lot record with an accurate first-use date answers that in seconds. Without it, the laboratory is matching QC charts to delivery dates and guessing.