Critical supplies control in the clinical laboratory

A critical supply is any product whose failure can change a test result: reagents, calibrators, control materials, diluents, solutions prepared in the laboratory and consumables involved in the measurement, such as cuvettes and dedicated tips. Brazil's ANVISA RDC 978/2025 defines in vitro diagnostic products as the reagents, calibrators, standards, controls, collection devices, materials and equipment used to analyse human biological material. [1]

Internal quality control (IQC) detects many problems with these products, but only once they are already on the bench. Supply management works before and around IQC: it selects reliable suppliers, records what arrived, prevents the use of expired or rejected lots and makes it possible to find which runs used a lot that was later recalled.

In Brazil, the mandatory reference is ANVISA RDC 978/2025. ISO 15189:2022 organizes the same topics for laboratories seeking accreditation. CLIA is the United States regulation: it does not bind Brazilian laboratories, but it is a good practical reference because it details labeling, expiration and lot checks. [1][5][7]

Article and clause numbers help you find the text; always read the full wording in the official source. RDC 978/2025 also defines an analytical run taking into account variables such as a change of reagent or calibrator lot. [1]

Think of a supply as a process with steps and evidence, not as an inventory item. Each step answers a question that an auditor, or your own team during an investigation, may ask.

1. Qualify the supplier and record the product: who may supply it and what is the current specification?
2. Receive and inspect: what arrived, in what condition and who checked it?
3. Accept the lot: does the new lot produce results equivalent to the previous lot?
4. Release and open: when did use begin and what is the in-use expiry after opening?
5. Monitor with IQC: did performance change after the switch?
6. Close and trace: when was the lot withdrawn, why, and which runs did it reach?

A rejected, suspended or recalled lot leaves the flow at any point. The record must show the block and remain available for the retention period.

For suppliers of consumables, RDC 978/2025 does not require a written contract: invoices and equivalent documents prove the purchase. That does not waive qualification. Where applicable, the laboratory must request proof of regulatory compliance, and both parties must ensure the products are fit for purpose. [1]

Define objective criteria and record every evaluation: product and establishment authorization, quality system, technical documentation, cold-chain transport, delivery time and support. ISO 15189:2022 asks for qualification, selection, performance evaluation and re-evaluation criteria, with records of the resulting actions. [7]

In the product record, register the regulatory authorization number and its expiry, the current version of the instructions for use, the measuring range of each analyte, the storage conditions and whether the lot requires acceptance before use. Keep versions: an older result must stay linked to the specification in force when it was issued. In Brazil, RDC 830/2023 covers the risk classification and authorization of these products. [2]

At receipt, record the product, lot, shipment, expiry date, quantity, invoice, transport temperature, packaging integrity and who checked it. RDC 978/2025 explicitly mentions the lot number, compliance of transport conditions and the receipt date. [1]

Keep lots that are not yet accepted, released lots and rejected or expired lots physically and documentarily separate. A shipment that arrived with a temperature excursion must stay in quarantine until it is evaluated. ISO 15189:2022 and the WHO quality management handbook describe this segregation and inventory control. [7][8]

Store products in temperature-monitored equipment compatible with the range stated by the manufacturer. When refrigeration is used, record the maximum, minimum and current temperatures, and plan how to preserve products during a power failure. [1]

There are two expiry dates. The manufacturer's applies to the unopened product. In-use stability starts when the product is opened or reconstituted and is usually shorter. Use whichever ends first. Once expired, RDC 978/2025 prohibits re-dating the product. [1]

For reagents prepared or aliquoted in the laboratory, the label must show name, concentration, lot, preparation date, person responsible, expiry, storage conditions and hazards. These preparations may not be sold or exchanged with other services. [1]

Different lots of the same reagent or calibrator can produce different results, and the effect can be clinically relevant even when the control stays within limits. The literature documents cases in which the change only appeared in patient results. [10]

The CLSI EP26 protocol guides the user evaluation: define the acceptance criterion before measuring, choose the number of samples based on that criterion and compare the current and new lots under the same conditions. [9]

Which samples to use

Prefer patient samples that cover the range of interest. Control materials may not be commutable: they can respond to a lot change differently from real samples and hide or exaggerate the difference. Use controls as a complement or when patient samples are not feasible, and record that limitation. [10]

How to decide

Choose the criterion before testing and record its source: a fraction of the total allowable error, the manufacturer's specification or another documented criterion. Calculate the difference for each pair, the mean difference and the dispersion. If the result does not meet the criterion, the lot is not used until there is a documented decision.

Difference (%) = 100 × (result with the new lot − result with the current lot) ÷ result with the current lot

The mean difference is +1.27%, with a standard deviation of 0.13%. With a hypothetical criterion of 2.3%, one third of a 7% total allowable error, the lot passes. If the same comparison showed +6.5% in every sample, as in a shipment that suffered a temperature excursion, the lot would be rejected and segregated. The values are illustrative and do not recommend a criterion for any analyte.

Whenever possible, approval should be given by someone other than the person who performed the comparison. This segregation of duties is good management practice, not an explicit requirement of RDC 978/2025.

For single-use products, RDC 978/2025 requires IQC at least with every new lot, every shipment and as the manufacturer instructs. This frequency can only be reduced with technical criteria, initial training provided by the supplier and certificates from lot-by-lot certification programs run by accredited laboratories, kept available to the health authority. [1]

• Declare when the new lot starts being used, with date, time, affected controls and person responsible. Without this record you cannot separate before from after.
• Follow the instructions for use regarding calibration: many manufacturers require calibration or calibration verification when the reagent lot changes. Record the evidence.
• Do not recalculate the control mean and standard deviation just because the reagent changed. The limits represent expected performance; a lot change that shifts the control is exactly what IQC must reveal.
• If the limits need revising, do it formally, with the reason, statistical data, person responsible and approver.
• Follow the first runs after the switch: compare the observed mean with the target and investigate shifts, even if no Westgard rule was violated.
• Do not mix components from different lots of the same kit unless the manufacturer allows it. [5]

A reagent lot change is different from a control material lot change. In the second case, the laboratory establishes the mean and standard deviation of the new material; in the first, the control material is the tool that reveals the effect of the new reagent. [6]

In Brazil, technical complaints and adverse events associated with regulated products must be reported in Notivisa. The laboratory must also investigate the occurrence, take measures to prevent recurrence and cooperate with field actions, which are the responsibility of the manufacturer or registration holder. [1][3][4]

When a manufacturer recalls a lot, the immediate question is: which runs and which results used that lot? Answering requires each run to have its lot recorded at the time of analysis, not reconstructed later from the team's memory. With that link, the laboratory defines the period, evaluates the potentially affected results and documents the decision.

1. Suspend use and segregate the lot.
2. Record the incident with a description, the notice reference, the action taken and the person responsible.
3. Identify the runs and controls that used the lot.
4. Report where applicable and follow the field action to completion.
5. Declare the replacement lot before resuming routine work.

Keep all these records for the minimum of 5 years required by RDC 978/2025. [1]

Indicators turn records into management decisions. Review them in the periodic IQC review and when re-evaluating suppliers.

Interpret the proportion of nonconformities associated with lot changes with caution: it depends on how many changes occurred and on the period. Use it to prioritize investigations, not to rank suppliers without analysis.

On the control panel, the focus remains IQC. Supply data sits on a dedicated screen for each control, reached through a link that flags when something needs checking.

To see the workflow with your laboratory's data, request a demo.

1. List critical supplies by test and instrument, starting with the highest-risk ones.
2. Define qualification criteria and evaluate current suppliers.
3. Standardize receipt records and the segregation of lots not yet accepted.
4. Write the acceptance protocol: samples, number of pairs, criterion and approver.
5. Record when each lot starts and stops being used and link it to the runs.
6. Establish the response to incidents and recalls, including reporting.
7. Review indicators and suppliers periodically and keep records for 5 years.

To explore related topics, see also:

• RDC 978/2025: what changes in internal quality control
• How to choose allowable total error
• Control material in IQC
• Westgard rules and control behavior
• Analytical sigma: practical laboratory QC guide

Sources consulted on September 26, 2026. Numerical examples and roadmaps are for teaching. For the ISO and CLSI standards, the public description of the content was consulted; full application requires access to the corresponding document.

1. ANVISA (Brazil). Resolution RDC No. 978 of June 6, 2025. Services that perform clinical laboratory tests. In Portuguese.
2. ANVISA (Brazil). Resolution RDC No. 830 of December 6, 2023. In vitro diagnostic medical devices. In Portuguese.
3. ANVISA (Brazil). Resolution RDC No. 67 of December 21, 2009. Technovigilance. In Portuguese.
4. ANVISA (Brazil). Resolution RDC No. 551 of August 30, 2021. Field actions for medical devices. In Portuguese.
5. HHS/CMS. 42 CFR §493.1252 — Test systems, equipment, instruments, reagents, materials, and supplies (CLIA).
6. HHS/CMS. 42 CFR §493.1256 — Control procedures (CLIA).
7. ISO 15189:2022. Medical laboratories — Requirements for quality and competence. Clauses 6.6 and 6.8.
8. World Health Organization. Laboratory quality management system: handbook. 2011.
9. CLSI. EP26: User Evaluation of Acceptability of a Reagent Lot Change.
10. Thompson S, Chesher D. Lot-to-lot variation. Clin Biochem Rev. 2018;39(2):51–60.