Implant-grade TPU: what the 52-week implantation data actually shows

Why long-term implant is the hardest gate

A material that stays in the body for years has to answer three questions that a short-contact material never faces: does it hold its mechanical properties while it is there, does the tissue around it stay quiet, and can the supplier guarantee the same material five years from now.

Most of that difficulty is not polymer chemistry. It is evidence. A long-term implantation study runs for years, and a device maker cannot repeat one for every candidate material — which is why both the biological data and the registration route end up deciding the project.

The certification baseline

In January 2024 the NLMD range completed full medical-grade TPU certification across 70A–85D — the whole hardness span from soft catheter durometer to rigid structural parts. That made Nilun one of only two suppliers worldwide holding a full-range medical TPU certification at the time.

Full range matters more than it sounds. A device programme that starts with a soft shaft and ends with a rigid hub would otherwise need two suppliers, two material files and two change-control processes.

What is actually registered

Five implant-grade polycarbonate grades are filed under China's national master file (MAF) system: NLMD6575A, NLMD6585A and NLMD6595A on the aliphatic side, plus NLMD8585A and NLMD8555D on the aromatic side.

For a device maker, a MAF is the routing that shortens registration: the material data sits in a filed dossier that the device submission can reference, without the compounder disclosing the formulation itself. It is a reference path, not a certificate, and it does not remove the device-level biological evaluation.

The 52-week implantation data

The 95A grade in the implant range was taken through subchronic systemic toxicity testing and a 52-week intramuscular implantation study at a CNAS-accredited laboratory (accreditation L13260, report WT252114-1). Irritation index was read at three points:

Why a falling curve beats a single low number

A single passing score at one time point tells you the sample that was tested was acceptable. A curve that falls 2.0 to 0.8 to 0.0 tells you the tissue is actively settling — the capsule is forming and quietening rather than progressively reacting to the material.

That is the shape a long-term implant material has to show, because a device left in place for years is not judged on week 13. It is judged on what the surrounding tissue looks like at year one and beyond.

Batch traceability and supply continuity

Biological data is worth little if the next lot behaves differently. Production runs on an Emerson DCS control system with imported compounding equipment and a self-developed fluidised bed, with start-up data captured into the quality traceability system; design capacity is 60,000 t/yr.

For a device customer the practical question is not peak capacity but continuity — whether the grade they qualified in 2026 is the same grade available in 2031, under a change-control agreement rather than a best effort.

What this evidence does not prove

It does not prove that a finished device is safe. Material-level biocompatibility is one input; the device still needs its own ISO 10993 evaluation covering geometry, surface area, contact duration, sterilisation residues and leachables from every other component in the assembly.

It also does not substitute for the customer's own NMPA, FDA or EU MDR submission. What it does is remove one of the longest and least controllable items from that submission — the material dossier — and replace it with a reference to a filed master file.

Grade performance, irradiation status and filing scope should be confirmed against the current TDS, the biocompatibility reports and the registration documents before any design freeze.