On the kickoff call, ask for the Schedule of Accreditation covering ISO 10993-12 sample preparation plus EU 10/2011 D2 simulant migration by GC-MS, with LOQ at or below 1.8 mg/kg for adipates. The reply tells you whether the lab can deliver a defensible DOA result for medical-grade flexible PVC. Four blocks decide every go/no-go: Scope, Simulant, Sensitivity, Contract.
Scope: Verify the Lab’s Schedule of Accreditation, Not the Certificate
ISO 17025 is method-specific, not blanket. A lab can hold a valid ISO 17025 certificate and still not be accredited for the test your medical-PVC sample requires. The certificate is the cover page; the Schedule of Accreditation is the contract.
Read the Schedule at three depths:
- Standard + edition — ISO 10993-5:2009 listed, not just “ISO 10993”; ISO 10993-12 explicitly named
- Method within the standard — GC-MS spelled out, not assumed under a generic “extractables” line
- Parameter range — measurement range covers the headroom you need below the DOA SML
Obtain the Schedule directly from the accreditation body (A2LA, NVLAP, UKAS, IAS, CNAS), not the lab. MedTech expert Ran Chen reports an FDA-flagged case where an ISO 17025-accredited lab’s Schedule did not list the extractables technique. The work was competent, the paperwork did not match it, and the flag landed at the method level.
Three red flags disqualify on this block. The phrase “ISO 17025 certified” — accreditation is granted, not purchased. Accreditation held at a sister facility rather than the testing site.
A fourth: advanced techniques (ICP-MS, LC/HR-MS, GC-MS) named in marketing but absent from the Schedule. For US 510(k) work, an active ASCA Summary Test Report shortens review — verify status against the FDA registry, since recognition has been withdrawn from non-compliant labs.
Simulant: Confirm the Lab Runs EU 10/2011 D2 (or Full D2e + D2i Substitution)
The overall migration ceiling under EU 10/2011 is 10 mg/dm² (or 60 mg/kg) into simulant D2 — olive oil. The DOA-specific migration limit (SML) for adipate esters in medical-grade PVC sits at 18 mg/kg. Two questions decide whether the lab has read Annex III:
- Does the lab run D2 (olive oil) directly? D2 is the reference fatty-food simulant. Capability is the baseline yes/no — smaller labs decline because olive-oil work is messier than alcohol-based methods.
- If D2 is substituted, do they run D2e (95% ethanol) AND D2i (isooctane) in parallel? Annex III allows substitution only when D2 is technically infeasible, and requires both — not D2i alone. “D2i only” signals the lab has not read 10/2011 closely.
- What temperature ceiling do they apply for D2i runs? Contact temperature is capped at 60 °C, and D2i can overestimate migration with polyolefins — confirm if your device sees sterilization or warm-line infusion.
As a 2026 benchmark, a European accredited lab posts overall migration into D2 at LOQ 3 mg/dm², 6-8 weeks, €352-942 per sample.
Sensitivity: Choose a Lab With GC-MS LOQ Below 1.8 mg/kg for Adipate Esters
The arithmetic is the qualifying question. The DOA SML is 18 mg/kg, so for a defensible result you want ten times headroom below it — GC-MS LOQ at or below 1.8 mg/kg for adipate esters. A “below LOQ” reading with LOQ at 5 mg/kg tells you nothing.
The device could be at 4.9 mg/kg (passing) or 0.4 mg/kg (passing by 45×), and a notified body will ask. Modern GC-MS methods sit roughly 1000× below SML on standard adipate work, so 1.8 mg/kg is not aggressive. See the detection methods for DOA migration walkthrough for the instrument-side math.
ISO 10993-12 extraction conditions are an active method debate. A 2022 round-robin in Regulatory Toxicology and Pharmacology questioned their adequacy. Ask how the lab handles extraction ratio (3 or 6 cm² per mL) and temperature (37, 50, or 70 °C) for your geometry.
DOA also elutes near other adipate esters and phthalate residues. The lab should describe co-elution mitigation — selected-ion monitoring, isotope-dilution internal standard, column choice — in 30 seconds. If they cannot, they have not run the test enough times.
Contract: Choose a Lab That Locks in Retention, Raw Data, and Audit Rights
Four clauses must be in writing before the first sample ships:
- Raw chromatogram delivery for every batch — reviewers ask, and summary-only creates a delay loop during review
- Lot retention minimum 12 months, 36 months for implantable devices; clock starts at test completion
- Audit observation rights for your QA team on-site with reasonable notice
- Written change-order procedure for any deviation, signed by your QA and the lab’s technical manager before execution
Set three turnaround bands rather than one: rush 5-7 days, standard 10-15 days, regulatory submission 4-6 weeks with method validation. Pricing in the €350-950 per-sample band is normal for accredited EU work in 2026.
A medical-grade DOA supplier — Bastone among them — should provide a sample certificate of analysis the lab can cross-reference against simulant results.
The testing vendor landscape consolidated through late 2024 and early 2025, so re-verify the Schedule after any ownership change. Pre-PO verification for supplier claims sits in our note on verifying a phthalate-free supplier claim before the PO.
The One Question That Beats All the Others
If you can ask only one question on the call, ask for the Schedule of Accreditation from the accreditation body — not the lab — and read it against your test method line by line. Every failure mode in lab selection traces back to a Schedule that did not cover the work; every successful submission, to one that did. The certificate sells; the Schedule defends.