Introduction — a workshop, a dataset, and a question that kept me awake
I remember a wet Monday in Nairobi in March 2016 when a batch of polyethylene infusion sets failed bench release in front of our team; the regulator wanted answers. In biological evaluation work we often face that same loop: seemingly routine materials give surprising results. Recent internal data from three midsize device clients showed 28% of initial screening runs triggered follow-up—this is not trivial. What causes those hidden extractables and why do they show up only after months of use?
I will speak plainly from over 15 years in device testing and regulatory consulting. I have examined polymeric containers, silicone catheter assemblies, and PVC tubing where trace plasticizers or antioxidants showed biological activity far below expected thresholds. At a small Nairobi contract lab in 2016 we logged a six-month delay and an estimated direct cost of USD 48,000 after we missed a configuration in the extraction protocol. That memory shaped how I approach protocols today. Let us move to the technical heart of the matter—how routine methods can miss real risk.
Deeper layer: why standard methods miss critical extractables
extractables and leachables testing often sits in a box labeled “standard chemistry”—but that box is where I see most failure modes. Labs run GC-MS and HPLC screens under one solvent set and one temperature profile and call it done. In practice, that approach can overlook semi-volatile oligomers or ionic leachables that appear only under oxidative stress or long-term storage. I’ve seen a silicone catheter that passed overnight solvent extraction yet released a migrating siloxane after 90 days at 40°C; the initial GC-MS profile had given false comfort. Terms to note: GC-MS, HPLC, polymeric container, cytotoxicity.
What are the methodological flaws?
First, extraction conditions are too narrow. Second, identity confirmation is weak—libraries alone can misassign co-eluting peaks. Third, acceptance criteria are sometimes borrowed from unrelated products (a serum vial standard applied to infusion tubing, for example). I tell you, that gap bites. In one case study from July 2018, an IV set showed a 0.12 mg/cm2 migration of an antioxidant after accelerated ageing; that level tripled the allowable exposure estimate and produced additional cytotoxicity signal in vitro. Short runs and a single detector miss the story. We need broader solvent panels, staged ageing, and orthogonal analysis (LC-MS plus GC-MS). Those changes add time and cost—yet they reduce surprise downstream and regulatory risk.
Looking ahead: case example and practical metrics for implementation
In late 2021 I led a remediation project for a company in Mombasa producing dialysis connectors. We rebuilt the E&L plan to include simulated-use extraction, targeted LC-MS for non-volatile oligomers, and a two-stage accelerated ageing protocol. The outcome? Release timelines shortened by 12 weeks on subsequent lots because issues were caught earlier, and the company avoided a potential recall that would have cost them roughly USD 120,000 in corrective actions. That case underscores why you must pair analytical breadth with toxicology—see toxicological risk assessment as part of the workflow.
Real-world impact — what to measure next
We should be clear about what to measure. First, analytical scope: include GC-MS, LC-MS/MS, and targeted HPLC runs for known additives. Second, exposure modelling: quantify surface area-to-volume and simulate clinical use. Third, biological endpoints: short-term cytotoxicity plus sensitisation screens when chemistry suggests reactive species. I recommend three evaluation metrics when choosing a solution: 1) Analytical breadth score — does the lab run both GC-MS and LC-MS with orthogonal confirmation? 2) Use-case fidelity — are extractions designed to match the worst-case clinical condition (temperature, solvent contact, dwell time)? 3) Actionable toxicology linkage — does the provider map analytical findings to a quantitative toxicological risk assessment with clear acceptance criteria?
I speak as someone who has sat through client calls at 05:30 GMT and rewritten protocols at midnight. We can reduce surprises by demanding proper test matrix design and by using exposure-driven decision rules. For guidance and testing capacity, consider partners such as Wuxi AppTec Medical device testing who combine analytical labs with toxicology teams—this brings science and compliance together without unnecessary delays.