Industries / Health / Medical devices / Prosthetics
Sockets and fit
The socket is the custom-shaped shell that holds the residual limb, the part of the limb left after amputation, and carries the body's weight into the artificial limb; fitting it so that it stays on, spreads pressure and copes with a limb that swells and shrinks is the core of the trade.
7
Milestones achieved, from 1949 to 2025.
2
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1949
Suction sockets go routine
A suction socket holds a thigh prosthesis on by air pressure through a one-way valve, with no straps around the waist. Dubois D. Parmelee of New York patented the idea in 1863, English fitters used it in the 1920s and then let it lapse, and it spread in Germany only after a workable valve came in 1932. After a US Army commission saw German results in 1946, the University of California, Berkeley fitted 52 people in 10 places by September 1947, and 20 one-week schools trained 250 surgeons and 200 prosthetists. With records of more than 500 cases by October 1949, the committee recommended general use and the Veterans Administration made the socket routine, as Chester Haddan and Atha Thomas reported in 1954.
LessonAn idea patented in 1863 took hold only when a field trial and training schools taught ordinary fitters to use it.
Credit
Dubois D. Parmelee; the University of California, Berkeley, Prosthetic Devices Research Project; the Advisory Committee on Artificial Limbs of the US National Research Council
1959
Weight on the tendon
The patellar-tendon-bearing (PTB) socket for below-knee amputation carries much of the body's weight on the tendon below the kneecap, in a close-fitting socket held on by a cuff above the knee instead of side joints and a leather thigh corset. The University of California Biomechanics Laboratory issued its manual in November 1959. The Veterans Administration Prosthetics Center in New York then fitted 53 men with it between 1959 and 1960, and in 14 of them the limb shrank so much in the first one to three months that new sockets had to be made.
LessonA closer fit made the limb's changes in size matter more; a quarter of the men needed a new socket within three months.
Credit
University of California Biomechanics Laboratory, whose 1961 report on the socket is by Charles W. Radcliffe and J. Foort; Frank A. Witteck of the VA Prosthetics Center for the field study
1985
Thigh sockets bear on bone
The total-contact quadrilateral socket was the standard for above-knee amputation from the 1960s. In the early 1980s prosthetists published a family of designs, known generically as ischial containment sockets, that direct the sideways forces of walking onto the pelvic bone near the ischium, the bone a person sits on, rather than onto soft tissue; Ivan Long and John Sabolich each published theirs in 1985. The 1992 Atlas of Limb Prosthetics credits the origin to Long and further development to Sabolich, Thomas Guth, Daniel Shamp and Christopher Hoyt, while noting that the quadrilateral socket was still the most prescribed.
Credit
Ivan Long; John Sabolich and colleagues
1986
Silicone liners roll on
A roll-on liner is a silicone sleeve that is turned inside out and rolled onto the limb, where it holds the soft tissue in place and is then joined to the socket, for example by a pin lock. Össur Kristinsson of Össur in Reykjavik wrote that his prefabricated ICEROSS (Icelandic Roll On Silicone Socket) had been in use in Iceland since early 1986, after custom silicone sockets he described in 1984. By 1997 it was common enough in British practice for I. McCurdie, R. Hanspal and a colleague to survey 85 doctors and senior prosthetists on when to prescribe it, with an 85% response, while noting that "very little research" lay behind its use.
LessonThe liner spread on clinical experience well ahead of trials.
Credit
Össur Kristinsson, Össur hf
2001
A vacuum holds limb volume
Limbs lose fluid and shrink during the day, which loosens the socket. W. J. Board, G. M. Street and C. Caspers drew a vacuum of 78 kPa below atmospheric pressure on a suction socket and measured 10 below-knee users before and after a 30-minute walk: limbs lost 6.5% of their volume with ordinary suction and gained 3.7% with the vacuum, and X-rays showed 4 mm less pistoning, the up-and-down slip of the limb in the socket. From 1 January 2003 the US HCPCS billing code list has carried a code, L5781, for a vacuum pump for residual limb volume management.
Credit
W. J. Board, G. M. Street and C. Caspers
2020
Legs anchored to bone
An osseointegrated implant replaces the socket: a titanium fixture is placed in the thigh bone, and the leg attaches to a connector that comes out through the skin. Clinical development began in 1990, and the US Food and Drug Administration (FDA) allowed Integrum's OPRA Implant System under a humanitarian exemption in 2015, then gave full premarket approval on 18 December 2020 after a study of 65 people. In Rickard Brånemark's five-year follow-up of 51 patients treated at one centre, 92% of fixtures survived but only 45% of patients went five years without a revision, 11 had deep infections and 15 had mechanical complications.
LessonTaking away the socket took away its skin problems but added infection and breakage, and the authors called the rise in both by year five concerning.
Credit
Rickard Brånemark and colleagues; Integrum AB, Mölndal, Sweden
2025
Adjustable sockets on trial
Adjustable-volume sockets let the wearer tighten or loosen the socket during the day with dials, straps or moving panels. In a randomized trial at two clinics and two laboratories, D. H. Gates and colleagues at the University of Michigan gave 29 above-knee users three commercial designs for four weeks each: the Infinite Socket (LIM Innovations), the CJ Sail (CJ Socket Technologies) and the Quatro (Quorum). Many sockets were given up while the wearers got used to them, 41% of CJ, 25% of Infinite and 24% of Quatro fittings; where they worked, the CJ and Quatro improved comfort and utility scores over a laminated socket, and a 2026 report from the same trial found no difference in balance or falls.
LessonAdjustment helped some wearers but did not change falls; 78% fell or stumbled at least once in a three-week period in at least one socket.
Credit
D. H. Gates, N. J. Rosenblatt and colleagues, University of Michigan
Watch for
A socket that resizes itself
Joan Sanders's lab at the University of Washington has built a below-knee socket with motor-driven panels and sensors that measure the gap between liner and socket wall, so a controller can resize it during walking. In a 2025 test of military tasks with 11 service members and veterans, the automatic mode adjusted the socket more often and over a wider range, and scored lower on the team's socket fit measure, than phone-app or locked modes, but the change in comfort score did not differ between modes. A trial of about 100 people is recruiting, with its main results due in March 2027 by the trial record.
Achieved whena group not involved in its development reports daily-life results, or the socket is sold and its users are followed.
Claimed by
Joan Sanders's group, University of Washington, in a 2025 paper; trial record updated May 2026
Sensors that track fit
A team at the Scuola Superiore Sant'Anna in Pisa and the University of Twente built a sensor for above-knee sockets from a 1 mm silicone layer filled with magnetic particles, 70% by weight, and a small coil. On the bench it detected changes in the liner-to-socket distance from 0 to 7.00 mm and compression up to 50%; the authors say it must still be validated in prosthesis users.
Achieved whenthe sensor is tested inside sockets worn by prosthesis users and its readings are checked against an independent measure of fit.
Claimed by
F. Andrei, A. Menciassi, L. Paternò and colleagues, July 2026
Open
01
Adjustable sockets on test limbs
A bench comparison of the ways sold to manage limb volume, such as adjustable panels and dials, added socks and elevated vacuum, on an artificial below-knee limb whose volume can be changed by plus or minus 5% while a machine applies walking-like loads. Measure pistoning, interface pressure and what happens when the socket is deliberately over- or under-tightened.
The briefCredit · unclaimed02
Fuse limits for bone-anchored legs
Bench measurements of when the fail-safe devices used with bone-anchored legs let go. These are mechanical fuses fitted between the leg and the implant to protect the bone in a fall. Load them in twisting, in bending in two planes, in pulling and in combinations, and plot where each one releases against a published estimate of the loads that break the thigh bone.
The briefCredit · unclaimed
Open 01 · the brief
Adjustable sockets on test limbs
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A bench comparison of the ways sold to manage limb volume, such as adjustable panels and dials, added socks and elevated vacuum, on an artificial below-knee limb whose volume can be changed by plus or minus 5% while a machine applies walking-like loads. Measure pistoning, interface pressure and what happens when the socket is deliberately over- or under-tightened.
Why it is still open
A 2023 review of 35 research designs and 16 commercial adjustable sockets found that the link between design and comfort or function is rarely investigated, that many lack safeguards against over- or under-tightening, and that only 17.6% of clinical studies assessed a commercial socket. A 2025 University of Toronto mock limb that changes volume was tested inside one rigid socket; its authors note that socket stiffness and load change the readings and list testing the changing limb under load as work still to do.
Built with what exists
The Toronto mock limb, made from urethane of two hardnesses with water bladders and an Arduino-driven syringe pump, for under CAD 400 by its authors' count, with design files on GitHub (Decisionics/BiofidelicLimb); commercial adjustable systems such as Click Medical's RevoFit2, sold through prosthetic distributors; standard and elevated-vacuum suspension; thin-film force sensors; a linear actuator or a university materials test frame for cyclic loading; a laser displacement sensor for pistoning.
First step
Build the Toronto limb from its published materials list and repeat its plus or minus 2.5% volume test inside one adjustable socket.
Done looks like
At least four volume-management methods each run through 1,000 load cycles at minus 5%, zero and plus 5% volume, with pistoning in millimetres, peak pressures and the results of an over- and under-tightening test, all published with raw data.
Open 02 · the brief
Fuse limits for bone-anchored legs
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
Bench measurements of when the fail-safe devices used with bone-anchored legs let go. These are mechanical fuses fitted between the leg and the implant to protect the bone in a fall. Load them in twisting, in bending in two planes, in pulling and in combinations, and plot where each one releases against a published estimate of the loads that break the thigh bone.
Why it is still open
A 2020 workshop of makers, clinicians and standards staff found that each fail-safe device is built to different requirements, that there is no clear set of standards, and backed forming a group to propose a new working group under ISO/TC 168, the international committee for prosthetics and orthotics standards. A 2024 University of Melbourne study says most devices on the market release only under twisting, a few also under front-to-back bending, and none under all six load directions, and that thigh bone fractures occur in 6.3% of these femurs within four years.
Built with what exists
Devices named in the Melbourne study, the OPRA Axor (Sweden) and the BADAL X LUCI Connector (OTN, the Netherlands), bought through a clinic or lent by their makers; an axial-torsion materials test frame; a six-axis load cell; a machined dummy connector in place of a real implant; the six-direction failure envelope published by the Melbourne team.
First step
Write a draft test method from the questions listed in the 2020 workshop report and run twisting tests on one device in a borrowed test frame.
Done looks like
Release loads for at least two device models, five units each, in at least five load directions and two combined loads, plotted against the femur envelope and published with a draft test method sent to ISO/TC 168.