Europe’s Pursuit of a Better Transhumeral Residual Limb

Home > Articles > Europe’s Pursuit of a Better Transhumeral Residual Limb
By Phil Stevens, MEd, CPO, FAAOP

The typical transhumeral residual limb presents a challenge to prosthetic fitting and rehabilitation. The cylindrical shape and conical dimensions of the residual humerus, and the lack of distal insertions of the residual muscle bellies make prosthetic suspension problematic and allow the prosthesis to rotate around the long axis of the limb. The result can be an unstable connection between the limb and the socket.

Clinicians often attempt to stabilize the connection by using aggressive proximal socket coverage as well as straps and harnesses that cross the shoulder girdle to the contralateral shoulder or chest wall, which limit the mobility of the affected shoulder joint. A transhumeral residual limb that permits a rotationally stable socket without undue restriction of shoulder motion would be ideal. For decades, several creative European surgeons have pursued this objective. This article presents their techniques of surgically improving the transhumeral residual limb to permit better prosthetic function at this challenging amputation level.


In 1971, Ernst Marquardt, MD, an orthopedic surgeon with Heidelberg University, Germany, introduced the concept of a distal angulation osteotomy in transhumeral- amputative care. The initial case was performed on the right humerus of a child with severe, quadrilateral limb deficiency. Twelve months after the initial procedure, Marquardt was encouraged by the functional and radiologic observations and repeated the procedure on the child's left arm, noting the boy's subsequent ability to use prostheses with good bimanual coordination.1 At the time of Marquardt's first English publication, he had developed two angulation osteotomy procedures for children and one for adults.1

At its essence, the angulation osteotomy is performed by resecting a ventral wedge along the distal aspect of the residual humerus. The location and depth of the wedge depends on individual patient presentation, with a more distal location in children and a more proximal location in adults, because a greater abundance of distal soft tissues requires a longer bony extension distal to the osteotomy. In general, the wedge is created 3-5cm from the distal end of the humerus in children, and 6-7cm in adults. When length is of lesser concern, a 70-90 degree ventral wedge is removed to the depth of the dorsal periosteum, allowing the bone to be reattached at an angle of approximately 110 degrees.1 When length sparing is desirable, the resected ventral wedge spans half of the humeral cross section with a dorsal transection spanning the other half. The resultant 110 degree angulation thus preserves additional humeral length.1

In describing post-surgery prosthetic rehabilitation, Marquardt clarified that the shoulder region of the limb was left uncovered, permitting full shoulder motion. Using what he describes as an open splint construction, the distal ventral aspects of the limbs bore the prostheses' weight along the angulated bony segments. The distal dorsal aspects of the limbs were left open to preserve sensory input in those areas. These elements were framed medially and laterally by outside-locking hinges that supported the distal prosthetic forearms.1

With time and additional clinical experience, a limitation of the procedure was observed in the tendency of the angulated bone to return to its original, unbent orientation. Between 1972 and 1989, 61 patients with long transhumeral residual limbs underwent angulation osteotomies at the Heidelberg University Orthopedic Hospital. The procedure was used to improve function in patients who had undergone transhumeral amputations and in children with a risk of terminal osseous overgrowth. Follow-up observations were available for 31 patients with 43 angulation osteotomies. Of the ten adults, the resultant angle of the osteotomies had not straightened. However, among the 33 angulation osteotomies in children, one had straightened within six months, seven within 12 months, and a further 12 had straightened up to 24 months post-surgery.2 Thus, the ultimate viability of the procedure appears to depend on patient age and the remaining humeral growth. However, Marquardt and his collaborating authors restate in this follow-up publication that the angulation osteotomy is "the only surgical technique that improves humeral stump function, providing a rotation-stable humeral prosthesis and a free-moving shoulder joint."2

Incidentally, angulation osteotomy, with minor variation, has been advocated in print as recently as 2015.3 The authors cite the continued value of the osteotomy procedure in providing a lever arm for both prosthetic suspension and supplemental rotational control, advocating a longer distal bone segment of 6-8cm, an angle of 70 degrees, and a posterior fixation plate to support the revised bone shape over time.3


About ten years ago, building upon the fundamental principles of the angulation osteotomy in remodeling the rigid support structures of the residual humerus to increase prosthetic comfort and function in individuals with transhumeral amputations, researchers from Trondheim, Norway, developed an implantable internal prosthesis that might help support and stabilize the external prosthesis.4 In the introduction to their approach, the authors discuss the stabilizing benefits of the humeral condyles in patients with elbow disarticulations. However, the functional and cosmetic results of prosthetic rehabilitation at this amputation level are limited by the overall bulk and mechanical limitations of prostheses with outside-locking hinges. Here the authors cite a novel surgical approach reported by two Brazilian authors in which the stabilizing benefits of the humeral condyles were preserved by resecting 3cm of humeral bone length just proximal to the condyles. This shortened the overall length of the residual limb, permitting use of a more functional, lower-profile prosthetic elbow, while retaining the benefits of condylar stabilization.5 To gain these benefits, the Norwegian team developed a titanium implant, the Humerus-T-Prosthesis (HTP), that would mimic the shape and function of the humeral condyles during prosthetic fittings.4

Prior to any clinical applications, the HTP was cemented into the distal intramedullary cavity of a human cadaver humerus and stress tested to determine axial and torsional load limits of the fixation. Encouragingly, axial displacement of the HTP did not occur until a load of 2,700 Newtons (just over 600 pounds) was applied. Rotation occurred at a torsional load of 52.5Nm (just over 40 foot-pounds).4 However, this rotation fully recovered when the implant was unloaded, with no rotational migration.

Following this testing, the HTP was surgically implanted in three subjects, all of whom began subsequent prosthetic rehabilitation three to four weeks post-surgery. In each case, the external prosthesis was successfully stabilized by the implant, allowing the trim lines to terminate inferior to the deltoid. This allowed for increased shoulder range of motion (ROM) within the prosthesis. For example, in the first patient, active shoulder abduction increased from 90 to 150 degrees. Active internal and external rotation went from being absent to 60 and 50 degrees, respectively. Combined shoulder flexion and extension ROM increased from 155 to 215 degrees.4 Similar gains were documented by the second and third subjects.

While the subjects experienced some initial pain over the newly formed condyles during the early stages of prosthetic fitting, this was ameliorated with subsequent design changes to the external prostheses, and the first two patients became full-time prosthetic users 24 months after the procedure. The third subject experienced an unrelated traumatic fracture to his ipsilateral scapula and was still limited in his use of the prosthesis at the time of the report. Additional pain considerations were reported using pre-operative and 12-month post-operative Visual Analog Scale (VAS) scores regarding phantom limb pain and neck and contralateral shoulder pain. As reasonably expected, the VAS scores for phantom limb pain were largely unchanged. However, dramatic improvements were reported in the VAS scores for neck and shoulder pain, with Patient One's pain decreasing from 4.5 (where 10 equates to the worst imaginable pain and 0 equates to the best imaginable) to 1 on the 11-point VAS, and Patient Two's pain decreasing even more dramatically, from 8.8 to 0.2.4 These improvements can be reasonably attributed to the revised prosthesis and the translation of the prosthetic load onto the residual limb rather than across the upper torso to the contralateral shoulder girdle.

At the time of the publication, in 2006, the authors reported that an additional four subjects had been managed with a second-generation HTP. Unfortunately, there has been no subsequent publication on this approach.



Building upon the principles of Marquardt and the Norwegian approach, an Austrian team recently published their initial findings on a modified T-shaped surgical implant "designed to achieve better force distribution and to resemble the anatomical shape of the humeral condyles."6 The implant is called the subcutaneous implant-supported attachment (SISA), and the authors report their initial observations of two transhumeral patients managed with this approach.

The HTP was truly T-shaped, with an emphasis on the horizontal projections of the medial and lateral structures. The Austrian team determined that this shape only provided about 1,000mm2 of axial loading surface. If the narrow extensions of the HTP were modified to a butterfly shape, this surface area could be increased by a factor ranging from 2 to 3.5, depending upon the implant size. These principles were used to design the SISA implant, with resultant loadbearing surfaces ranging from 2,000- 3,700mm2 in five modular sizes.6

The two initial subjects selected for the SISA study had different histories with common limitations. The first was an established prosthetic user who had experienced a traumatic transhumeral amputation 23 years prior. The second was substantially less experienced, having incurred a traumatic transhumeral amputation about nine months earlier. Both were using myoelectric prostheses with straps and harnesses anchored around the neck and contralateral shoulder girdle for suspension and stabilization, with both subjects citing pain, discomfort, and skin irritation related to the suspension harnessing.

Prior to surgical implantation of the SISA units, objective measures of prosthetic function, shoulder mobility, and pain were documented. These were repeated six to 18 months after implantation. Prosthetic rehabilitation began four months after the implant surgery, with both subjects using an Ottobock Dynamic Arm Elbow, electric wrist rotator, and SensorHand Speed.6

As with the HTP, active ROM with an external prosthesis improved appreciably, though it was reported much differently. The authors measured combined ROM across shoulder abduction and adduction, flexion and extension, and internal and external rotation. These measurements were taken without the prosthesis, with the prosthesis prior to the implant, and with a very different, less obtrusive prosthesis following the surgical implant. For the first patient, the average joint restriction within the three measured planes prior to the SISA implant was 43 degrees. Following implantation and the new socket, this average restriction was reduced to 9 degrees.6 For the second patient, the improvement was more striking, with the average mobility restriction while wearing a prosthesis reduced from 62 to 2.5 degrees.6

In addition, pre- and post-surgical prosthetic function were assessed using the Southampton Hand Assessment Procedure (SHAP), a clinically validated function test originally developed to assess the effectiveness of upper-limb prostheses. It is a series of timed item performance tasks in which a score of 100 points indicates normal hand function. For the two subjects, SHAP scores improved from 8 to 14 and from 1 to 32, respectively.


The limitations associated with upper-limb prostheses continue to be analyzed and creatively addressed. In addition to improvements in sensory input and myoelectric control through implanted electrodes ("Improving the Human-machine Interface: Revolutionary Advancements in Upper-limb Prostheses," The O&P EDGE, January 2017), innovative teams from across Europe have developed and applied novel surgical solutions to address the challenge of socket suspension and stabilization. While none of these solutions have reached mainstream utilization, they demonstrate a series of progressive improvements toward the objective of improved prosthetic function following upper-limb amputation.

Phil Stevens, MEd, CPO, FAAOP, is in clinical practice with Hanger Clinic, Salt Lake City. He can be contacted at .


    1. Marquardt, E. and G. Neff. 1974. The angulation osteotomy of above-elbow stumps. Clinical Orthopaedics and Related Research 104:232-8.
    2. Neusel, E., M. Traub, K. Blasius, and E. Marquardt. 1997. Results of humeral stump angulation osteotomy. Archives of Orthopaedic and Trauma Surgery 116 (5):263-5.
    3. Cheesborough, J. E., L. H. Smith, T. A. Kuiken, and G. A. Dumanian. 2015. Targeted muscle reinnervation and advanced prosthetic arms. Seminars in Plastic Surgery 29 (1): 62-72.
    4. Witsø, E., T. Kristensen, P. Benum, et al. 2006. Improved comfort and function of arm prosthesis after implantation of a Humerus-T-Prosthesis in trans-humeral amputees. Prosthetics and Orthotics International 30 (3):270-8.
    5. de Luccia, N. D. and H. L. Marino. 2000. Fitting of electronic elbow on an elbow disarticulated patient by means of a new surgical technique. Prosthetics and Orthotics International 24(3):247-251.
    6. Salminger, S., A. Gradischar, R. Skiera, et al. 2016. Attachment of upper arm prostheses with a subcutaneous osseointegrated implant in transhumeral amputees. Prosthetics and Orthotics International, DOI: 10.1177/0309364616665732.