3D printing application in rehabilitation services

Rehabilitation covers a wide range of interventions that restore or compensate for lost function. A cane, a wheelchair tray insert, a modified cup holder, a grip sleeve for a walker. These are technical aids. They compensate for a limitation but carry no direct therapeutic claim. Orthoses, prostheses and anything that loads or aligns a limb are medical devices, and they fall under regulatory control for design, traceability and material documentation.

That distinction is the first thing to settle before a printer is switched on. A technical aid can be iterated quickly, printed in a clinic, fitted the same week. A medical device needs a documented material batch, a controlled print file and a paper trail that survives an audit. The hardware is similar. The process discipline is not.

Practically, we see the 3D printing application in rehabilitation work best when a part is low-load, patient-specific and changes shape between users. When the same part must carry a patient's weight for years, or mate to a machined joint, printing is often the wrong answer even if it fits on the bed.

• 1
  Technical aidCompensates for function. No therapeutic claim. Fast iteration is acceptable.
• 2
  Medical deviceTreats or aligns. Needs material traceability, controlled files, audit records.
• 3
  Hybrid partsPrinted interface plus machined load path. Common in modern orthotics.

FDM is the workhorse for technical aids. A 0.4 mm nozzle lays down layers around 0.2 mm, so a cuff or tray insert takes a few hours. The strengths are obvious: cheap machine time, easy color and material swaps, and parts you can sand, drill or heat-form after printing. The weaknesses matter too. Layer lines run in one plane, so a thin FDM bracket splits along the grain when pulled. Keep FDM parts in compression or in low-load contact with the body.

SLS prints in a powder bed with no support structures. That makes it the better choice for complex geometry: a wrist splint with internal lattice, a socket with variable wall thickness, a part that has to wrap around a joint. Nylon PA12 comes out strong and slightly flexible, and the surface is uniformly matte. The trade-off is cost per build and a rougher finish that sometimes needs sealing for hygiene.

SLA and DLP give the finest detail and the smoothest surface. That suits ear moulds, dental-adjacent work and thin shells where fit tolerance is tight. The catch is the resin. Most standard resins are brittle, and many are not rated for skin contact over months. Use biocompatible resins only when the part is classed as a device and the resin data sheet supports it.

• 1
  FDMCuffs, trays, holders, grips. Cheap, repairable, anisotropic. Keep loads off layer lines.
• 2
  SLSSplints, sockets, lattices, wraps. No supports, strong in all directions, matte finish.
• 3
  SLA / DLPEar moulds, thin shells, fine fit. Check resin skin-contact rating before use.

PLA is fine for a fitting check. It is not fine for a part a patient uses every day. It creeps under sustained load, softens in a hot car and degrades with repeated cleaning. Use it for form trials, then move on.

PETG and ABS give better toughness and heat resistance at low cost. TPU is the one to reach for when the part touches skin or has to flex, such as a strap pad or a bumper on a frame. Print TPU slow, in thicker layers, and expect stringing.

For load-bearing printed parts, PA12 and PA11 from SLS are the usual answer. They take impact, resist fatigue and clean up well. PEEK and PEKK are the high end: they tolerate autoclave cycles and aggressive disinfectants, and they cost accordingly. If a part must hold a patient's weight through a machined pivot, aluminium 6061-T6 or 316L stainless is still the safer call. We machine those to ±0.005 mm and can match surface finish from Ra 0.8–1.6 μm up to Ra 0.2–0.8 μm on sealing faces.

• 1
  Avoid PLACreeps under load, softens with heat, poor for repeated disinfection.
• 2
  TPU for skin contactFlexible, grippy, forgiving. Print slow with thick layers.
• 3
  PA12 for structureImpact and fatigue resistant. Standard for SLS splints and sockets.
• 4
  PEEK for disinfectionAutoclave-stable, chemically resistant, high cost.

Most rehab devices are not purely printed or purely machined. A prosthetic socket is usually printed for the interface and machined for the adapter that bolts to the pylon. The printed section is shaped to the residual limb; the machined section carries the load and the thread. Splitting the part this way keeps cost down and keeps the critical dimensions in metal.

The same logic applies to wheelchair mounts, orthotic hinges and any bracket that sees a bolted joint. Printed threads strip. Printed press fits relax. Machined aluminium and stainless hold both. We run 5-axis and mill-turn centers for these small metal components, and the printed shell can come from the same project file so the two halves line up.

There is also a repair argument. A clinic can reprint a broken cuff overnight. It cannot reprint a failed hinge. Designing the wear part as the printed one and the structural part as the machined one makes field replacement realistic without a full rebuild.

• 1
  PrintBody interface, shell, pad, cover, grip, low-load bracket.
• 2
  MachineThreads, pivots, press fits, bolted joints, load-bearing plates.
• 3
  Keep one datumDesign printed and machined sections from the same coordinate system.

A workable sequence starts with scanning, not printing. A handheld or phone-based scanner captures the limb or the contact surface, and the mesh is cleaned and thickened to a printable shell. Skipping this step and modelling from measurements alone costs more fitting time later.

Then comes load thinking. Mark where the part touches the body, where it sees force and where it must flex. That decides wall thickness and process. A 2 mm PA12 shell is fine for a forearm splint. The same shell under a heel strike is not.

Print a fitting check first in cheap material, confirm the contact points, then run the final version in the service material. Keep the scan, the print file and the material batch together if the part is a medical device. That record is what makes a reprint a controlled reprint rather than a new part.

For clinics that need the metal side as well, the usual path is a quick DFM review, then machining. We return a quotation and free DFM analysis within 12 hours, and production can start within 24 hours once the file is frozen.

• 1
  Scan firstCapture the contact surface. Clean the mesh before thickening.
• 2
  Fit checkPrint in cheap material, verify contact, then cut the final part.
• 3
  Record keepingScan, print file and material batch travel together for device parts.