Clinical Orthopaedics and Related Research ®

A Publication of The Association of Bone and Joint Surgeons ®

High Survivorship and Few Complications With Cementless Total Wrist Arthroplasty at a Mean Followup of 9 Years

Joseph A. Gil MD, Robin N. Kamal MD, Eugene Cone MD, Arnold-Peter C. Weiss MD



Total wrist arthroplasty (TWA) has been described as traditionally being performed with fixation in the radius and carpus with cement. The TWA implant used in our series has been associated with promising results in studies with up to 6 years followup; however, studies evaluating survivorship, pain, and function with this implant are limited.


1) To report ROM and pain scores after wrist reconstruction with cementless fourth-generation TWA at a mean followup of 9 years (range, 4.8–14.7 years). (2) To report complications of a cementless fourth-generation TWA and the cumulative probability of not undergoing a revision at a mean followup of 9 years.


This is a retrospective case series of 69 patients who were treated for pancarpal wrist arthritis between 2002 and 2014. Of those, 31 had inflammatory arthritis (rheumatoid arthritis [n = 29], juvenile rheumatoid arthritis [n = 1], and psoriatic arthritis [n = 1]); all of these patients received TWA with the cementless implant studied in this investigation. Another 38 patients had osteoarthritis or posttraumatic arthritis; in this subgroup, 28 patients were 65 years or younger, and all underwent wrist fusion (none were offered TWA). Ten patients with osteoarthritis were older than 65 years and all were offered TWA; of those, eight underwent TWA, and two declined the procedure and instead preferred and underwent total wrist arthrodesis. The mean age of the 39 patients who had TWA was 56 ± 8.9 years (range, 31–78 years) at the time of surgery; 36 were women and three were men. The patients who underwent TWA were seen at a minimum of 4 years (mean, 9 years; range, 4–15 years), and all had been examined in 2016 as part of this study except for one patient who died 9 years after surgery. The dominant wrist was involved in 60% (25) of the patients. All patients were immobilized for 4 weeks postoperatively and then underwent hand therapy for 4 to 6 weeks. Pain and ROM were gathered before surgery as part of clinical care, and were measured again at latest followup; at latest followup, radiographs were analyzed (by the senior author) for evidence of loosening, defined as any implant migration compared with any previous radiograph with evidence of periimplant osteolysis and bone resorption. Subjective pain score was assessed by a verbal pain scale (0–10) and ROM was measured with a goniometer. Complications were determined by chart review and final examination. Kaplan Meier survival analysis was performed to estimate the cumulative probability of not undergoing a revision.


The mean preoperative active ROM was 34± 18° flexion and 36° ± 18° extension. Postoperatively, the mean active ROM was 37° ± 14° flexion and 29° ± 13° extension. The mean difference between the preoperative pain score (8.6 ± 1.2) and postoperative pain score (0.4 ± 0.8) was 8.1 ± 1.9 (p < 0.001). Implant loosening occurred in three (7.7%) patients. No other complications occurred in this series. Kaplan-Meier survivorship analysis estimated the cumulative probability of remaining free from revision as 78% (95% CI, 62%–91%) at 15 years.


Cementless fourth-generation TWA improves pain while generally preserving the preoperative arc of motion. The cumulative probability of remaining free from revision at 14.7 years after the index procedure is 77.7% (95% CI, 62.0%–91.4%). Future studies should compare alternative approaches for patients with endstage wrist arthritis; such evaluations—which might compare TWA implants, or TWAs with arthrodesis—will almost certainly need to be multicenter, as the problem is relatively uncommon.

Level of Evidence

Level IV, therapeutic study.

Back to top