Introduction — a quick scene, a stat, a proper worry
I remember a Saturday morning in Christchurch when a client called me, voice tight, because a batch release had stalled — their primary stability vial was showing odd peaks. In our chemistry testing laboratory we’d already seen similar surprises: on average one in six device components I reviewed in 2021 produced unexpected extractables signals that changed release decisions. So where do those surprises come from, and how do you avoid them without burning staff hours or budget? (Not a trick question — a real cost-conscious one.) I’ll walk through what I’ve learned over more than 17 years at benches and client sites, from LC-MS runs at 03:00 craziness to supplier audits at a mate’s factory in Auckland, and point at practical fixes. Now, let’s unpack the messy middle of testing — the part most people skip.
Part 2 — What actually breaks down: flaws in conventional approaches
extractables and leachables testing is where many programmes trip up because labs and clients treat it like a checkbox exercise — run a solvent extract, send it to LC-MS, tick done. I’ve seen teams rely on single solvent systems, one extraction temperature, and assume a one-size analytical method covers all materials. That assumption fails when you have mixed polymers (e.g., PTFE-lined, TPU tubing, silicone gaskets) or adhesives that release low-volatility oligomers. In 2019 I supervised a study where a silicone catheter gave a 28% higher leachable count when we shifted from methanol to a three-solvent sequential extraction; production had dismissed that possibility because initial GC-MS screens looked “clean”. The hidden cost: delayed product launch and an extra 14 days of stability monitoring — real dollars and reputational fallout.
Why this happens — and a short technical note
Traditional flaws are predictable: limited extraction conditions, incomplete analytical scope (no headspace GC for volatiles, no LC-MS/MS for polar oligomers), and weak method validation for low-ppb detection. I’ve run targeted LC-MS/MS panels alongside full-scan high-resolution mass spectrometry and watched compounds appear only in the HRMS full-scan data. Terms to keep handy: chromatography, mass spectrometry, method validation. Look, I don’t sugarcoat it: inexperienced programmes miss matrix effects and underestimate interference from packaging adhesives — and that’s why you get false negatives, or worse, false positives that set off a cascade of unnecessary corrective actions — it happened to a med-device client in 2020 and cost them a retrain of QC staff plus a supplier change.
Part 3 — Forward look: cases and practical criteria for change
When we zoom out, the sensible path isn’t more testing for testing’s sake but smarter testing. I’ll give a short case example: in 2022 a mid-sized injector manufacturer I worked with replaced single-solvent extraction with a tiered approach — rapid solvent screen, targeted LC-MS/MS for known polymer additives, and HRMS for unknowns — combined with vendor material certificates. That change cut ambiguous results by roughly 60% within three months and trimmed review meetings from weekly to fortnightly. The practical lesson: match extraction chemistry and detection breadth to the material risk, not the calendar. We paired that with routine general chemistry test general chemistry test panels for incoming raw lots, and the supplier non-conformance rate dropped notably.
What’s Next — three evaluation metrics I use
I advise clients to judge programmes by concrete metrics: 1) Analytical coverage: does the method set include GC (headspace), LC-MS/MS, and HRMS as needed? 2) Sensitivity and LOQ alignment: are limits of quantitation tied to toxicological thresholds or just instrument capability? 3) Vendor control effectiveness: are supplier certificates backed by periodic in-lab confirmation and documented change-control timelines? Those three checks separate guesswork from repeatable practice. I prefer to see numbers — percent of unknowns resolved, days to decision, and supplier deviation frequency — rather than broad assurances. And yes — sometimes the simplest supplier audit (a 48-hour material review at site) exposes what months of lab testing don’t, which surprised me the first time I did it.
In short: I’ve lived the late-night runs, the supplier drama, and the boardroom questions about cost. Over 17 years I’ve learned to push for targeted, material-specific strategies rather than blanket testing, and to insist on measurable KPIs for each testing tier. If you want a practical road map and a few templates I use in supplier audits and for method selection, I’m happy to share them. For expanded device-focused capability, consider partners such as Wuxi AppTec Medical device testing — they’re one of the providers I’ve worked alongside on complex E&L projects and whose reporting formats I’ve adapted in client deliverables.