Every modern medical device biocompatibility submission rests on chemistry. Specifically, it rests on extractables and leachables studies that quantify what can migrate from a device into a patient under intended clinical use. Regulators no longer accept a generic materials declaration or a legacy animal study alone; they want defensible E&L data, analysed under ISO 10993-18, interpreted under ISO 10993-17, and reported in a way that lets reviewers follow the argument from raw chromatogram to risk conclusion. This guide explains how to design extractables and leachables studies that pass scrutiny on the first submission.
Why extractables and leachables studies matter
Patients rarely ingest a medical device, but they are continuously exposed to whatever migrates out of it during contact. A vascular catheter, a hip implant, an insulin pump, each has a potential chemical-release profile that regulators must be convinced is tolerable. These studies provide the quantitative evidence behind that conviction. They bridge the gap between “we used food-grade polymer” and “this patient will not be harmed”.
Since the 2020 revision of ISO 10993-18 and the 2023 revision of ISO 10993-17, the methodology has tightened considerably. Extraction study design, identification confidence, and threshold derivation are all areas where FDA and notified bodies have issued more deficiency letters than at any point in the past decade. Notified bodies in the EU, under MDR Annex II, now expect the E&L dossier to feed directly into the clinical evaluation report, and FDA reviewers increasingly request raw chromatographic data rather than summary tables.
Extractables versus leachables: the distinction that matters
An extractable is any substance that can be removed from the device by a solvent under exaggerated conditions. A leachable is a subset: a substance that actually migrates into the patient under clinical use. Extractables studies define the chemical universe; leachables studies confirm which members of that universe are real-world relevant.
Designing ISO 10993-18-compliant extractables and leachables studies
Step 1: Select extraction solvents
ISO 10993-18 requires at least three solvents of different polarity to cover the extractables space: a polar (water or saline), a non-polar (hexane), and a mid-polarity solvent (isopropanol). For devices with specific clinical exposure, additional physiologically relevant simulants may be added.
Step 2: Choose exaggerated extraction conditions
The extraction temperature and time must exceed clinical conditions by a defined safety factor. A common choice is 50°C for 72 hours for extended-contact devices, though the standard allows for higher or lower severity if justified. E&L protocols for implants often use multiple extraction cycles to demonstrate exhaustion.
Step 3: Set the Analytical Evaluation Threshold
The AET is the concentration below which an extractable can be dismissed without toxicological evaluation. It is derived from the Dose-Based Threshold and the clinical dose regime. AAMI TIR 106 is the industry reference for AET derivation; its application requires care because different analytical techniques yield different technique-specific AETs.
Step 4: Deploy a multi-technique analytical strategy
A credible extractables profile requires orthogonal techniques: HS-GC-MS and GC-MS for volatile organic compounds (VOC) and semi-volatile organic compounds (SVOC), respectively, LC-MS (ideally with both positive and negative ESI) for non-volatile organic compounds (NVOC), ICP-MS for metals, and IC for counterions. Single-technique screens leave gaps that reviewers invariably identify.
Step 5: Identify every extractable above AET
Per industry current standards, identification confidence falls as Levels 1 to 4, from fully characterised (authentic standard match) to tentative class assignment or unknowns.
When a leachables study is required on top of extractables
Not every device needs a leachables study. ISO 10993-17 allows the extractables profile to stand alone when clinical exposure conditions are milder than the exaggerated extraction and the margins of safety are robust. Leachables studies are expected when: extractables approach AET or PDE limits, the patient population is vulnerable (paediatric, pregnant), or cohorts of concerns may be present in the device (as part of raw materials or processing steps).
A well-designed leachables protocol uses clinically realistic simulants and durations. The same orthogonal analytical strategy used for extractables is applied, scaled to the lower expected concentrations. FDA’s biocompatibility guidance is explicit that leachables must be reported alongside the toxicological risk assessment that interprets them.
How regulators interpret the extractables dataset
An E&L report is read through a risk lens, not a chemistry lens. Reviewers ask three questions in sequence. First, is the chemical universe bounded: did the extraction exhaust the device, did the analytical suite cover the full polarity and volatility space, and are the identification confidence levels transparent? Second, are exposures realistic: was patient body weight, contact duration, and route appropriate to the clinical indication, and were worst-case assumptions preserved where uncertainty exists? Third, is the safety margin defensible: does the Margin of Safety hold under a sensitivity analysis, and are conservative defaults applied for compounds without compound-specific toxicology data?
A submission that answers these three questions on the first reading sails through review. A submission that leaves any of them unanswered receives a deficiency letter, and the review clock restarts. For that reason, our toxicology team writes the risk narrative in parallel with the analytical chemistry team, so that every extractable above AET has a ready toxicological context before the package leaves the laboratory. Building the narrative late is the single most common cause of E&L-driven submission delays we see on second-opinion engagements.
Reporting extractables and leachables studies for regulatory submission
A good E&L report narrates, not just tabulates. It describes the device, the intended clinical use, the chosen extraction conditions and their justification, the analytical methods and their validation data, the identification approach, and the raw-data archive. Every identified extractable appears with its AET-normalised concentration, its identification confidence level, and a cross-reference to the toxicological evaluation in the TRA. Our Toxicology Monographs frequently serve as the evaluation reference for thousands of compounds that show up in E&L workstreams, providing tolerable intake derivations with published regulatory citations.
Common pitfalls in E&L programmes
Pitfall 1: Under-exaggerated extraction. Clinical-equivalent extraction misses extractables that realistic variation might release. The margin of exaggeration must be justified.
Pitfall 2: Insufficient analytical coverage. Studies designed to use only one analytical tool such as GC-MS lead to lack of coverage in the screening extractable work.
Pitfall 3: Unknown extractables left unresolved. Regulators do not accept “not in library” as a dismissal. An unknown above AET must be identified, characterised with orthogonal data, or conservatively evaluated.
Pitfall 4: No leachables study when one was needed. Relying on extractables for a case where cohorts of concerns compounds may be present, for example, invites an automatic deficiency letter. Our ISO 10993-17 TRA guide lays out the decision criteria.
Pitfall 5: Missing AET per technique. The AET is technique-dependent; applying a single number across all analytical methods masks compounds that are actually above their relevant threshold.
Pitfall 6: Treating suppliers as a black box. Raw-material formulation changes are a silent invalidator of an E&L dossier. A supplier-change log, tied to the risk assessment, is essential for products with a long commercial life.
How ToxLibrary supports your E&L programme
Robust extractables and leachables studies are technically demanding, expensive, and central to market approval. If your team is scoping an E&L study, interpreting a finished dataset, or building the TRA that follows, we can help. Our toxicologists have supported E&L programmes for implants, drug-delivery devices, and combination products across FDA, MDR, and PMDA submissions. Reach out: a defensible chemistry dossier is less expensive and a better investment than a resubmission.
