Evidence

No randomized trials have compared STT arthrodesis with distal scaphoid excision (Level I). No prospective comparative studies (Level II) or retrospective comparative studies (Level III) exist either. What is available to make a treatment decision is a collection of case series with few patients in each. Most of the reports on STT arthrodesis combine results from patients who had a fusion for scapholunate instability or other problems with patients who had the fusion for primary STT arthritis. This amalgamation of reported data also makes interpretation of outcome difficult.

In his monumental review of 800 triscaphe fusions, Dr. Watson¹ does identify 98 patients whose arthrodesis was for primary STT arthritis. The case series review states that 62% of the patients were examined in the office and the others had data pulled from chart reviews. The mean follow-up was 3.4 years. Patients were immobilized for 3 weeks in a long arm cast, followed by 3 weeks in a short arm cast. Pain was graded by the patients as mild, moderate, or severe. The flexion extension arc was 85% and 80%, respectively, of the contralateral side. Grip strength was 77% of the other side. The results with regard to pain were not broken out independently for STT arthritis versus other indications for STT fusion. The rate of nonunion across all indications for STT arthrodesis was only 4%.

In a review of eight patients with STT arthritis, Srinivasan and Matthews² found four were pain free, three had pain with certain activities, and one had constant pain. The flexion extension arc averaged 115 degrees compared with 124 degrees on the uninjured side. One of the eight had a nonunion. Follow-up averaged 4 years.

Only one case series describes results of distal scaphoid excision specifically for STT arthritis.³ Garcia-Elias and colleagues³ reported on 21 patients with a mean follow-up of 29 months. Patients were immobilized for 2 to 3 weeks in a short arm splint. The preoperative visual analog scale (VAS) pain score was 7.5, and the postoperative score was 0.6, with 13 having no pain. Grip improved from 57% of the contralateral side preoperatively to 83% of the contralateral side after surgery. Wrist flexion averaged 57 degrees and wrist extension averaged 61 degrees. The radiolunate angle increased from 9 degrees before surgery to 17 degrees after surgery.

Evidence

No randomized trials have compared PRC with a four-corner fusion (Level I). No prospective comparative studies exist (Level II). Four retrospective comparative studies (Level III) have been reported, only one of which has a methodology that would minimize patient selection bias and the results of which might therefore be valid.^4–7

The highest quality study, by Cohen and Kozin⁴ in 2001, retrospectively examined 2 cohorts of 19 patients at different institutions, which performed exclusively either PRC or four-corner fusions. Importantly, there were no preoperative differences with regard to age, sex, stage of arthritis, or preoperative pain or function. All patients were stage II, except one four-corner patient who was stage III. The most significant limitation is that the follow-up averaged 19 months for the PRC group and 28 months for the four-corner group. Typically, these procedures are performed on patients with the hope of decades of postoperative functionality, thus any conclusion based on 2 years of follow-up is limited. Nevertheless, it avoids the selection bias in other comparative studies where patients had one or the other procedure based on surgeon preference.

The authors found no statistical difference in range of motion, grip strength, pain relief, or patient satisfaction. A power analysis was not included in the study. The flexion-extension arc was 81 degrees in PRC and 80 degrees in four-corner fusion. Grip strength was 71% for PRC versus 79% for four-corner fusion. VAS pain scores at follow-up were 1.4 for PRC and 1.2 for four-corner fusion. Five of the PRC patients and fourof the four-corner patients took nonsteroidal antiinflammatory drugs or analgesics for wrist pain at final follow-up. Three patients in each group changed jobs or retired because of wrist pain and function.

One nonunion patient was in the arthrodesis cohort, which had a successful repeat arthrodesis. One patient with PRC had persistent pain and went on to a total wrist fusion. Radiographic analysis of the arthrodesis group demonstrated no evidence of progression of arthritic changes at the radiolunate articulation. Although three patients in the PRC group showed narrowing or sclerosis at the radiocapitate articulation, these findings did not correlate with pain or function.

Wyrick and coworkers’⁵ retrospective review of 17 patients with four-corner and 10 patients with PRC treatment with a 27-month mean follow-up showed better grip strength, range of motion, and pain relief with a PRC.⁵ There were three nonunions in the arthrodesis group, only one of which was symptomatic. However, five of the patients with four-corner fusion required revision surgery, whereas none of the patients with PRC required revision.

Tomaino and researchers’⁶ retrospective review examined 15 patients with PRC and 9 patients with arthrodesis at an average of 5.5 years. Mean postoperative pain and grip scores were comparable. PRC preserved a greater arc of motion, 77 versus 52 degrees for arthrodesis. Three patients with PRC treatment were unsatisfied, and one went on to a wrist fusion.

Krakauer and coauthors’ retrospective review⁷ compares results of mostly stage II PRC patients with mostly stage III four-corner patients and is therefore not believed to be helpful in providing a valid comparison of the two procedures.

Evidence

No randomized trials have compared the surgical treatments for osteoarthritis of the DRUJ (Level I). No prospective comparative studies exist (Level II). One retrospective comparative study exists (Level III), which has serious methodologic flaws, as well as numerous case series (Level IV) with which to evaluate these procedures.

Minami and colleagues⁸ retrospectively compare the results of 20 Darrach resections with 25 SK and 16 hemiresections for DRUJ osteoarthritis. The follow-up was impressively a minimum of 5 years and a mean of 8 of 11 years. The authors originally performed Darrach resections on all patients with symptomatic DRUJ osteoarthritis. Later, they performed SK procedures if the triangular fibrocartilage complex (TFCC) could not be reconstructed or there was positive ulnar variance of greater than 5 mm. If the TFCC was intact or reconstructible, a hemiresection with interposition (e.g., Bowers) was performed. Unfortunately, the Darrach resection group was older (68 versus 53 SK or 60 hemiresections), and 75% were arthritic changes caused by prior distal radius trauma, whereas only about 25% of the SK and hemiresection group conditions were due to secondary osteoarthritis and 75% were due to primary or idiopathic osteoarthritis. No ordinal preoperative pain scoring is available with which to compare postoperative pain results. These points make any comparison of postoperative pain, grip, or range of motion questionable. Of note, 40% of patients who had undergone Darrach resections had no postoperative pain, 20% had slight pain, 25% had moderate pain, and 15% had severe pain. Grip strength decreased in 65% of patients with a Darrach resection.

Watson and Gabuzda⁹ reviewed results from 32 patients with matched hemiresection without interposition for post-traumatic osteoarthritis of the DRUJ. Two patients in the series had Madelung’s deformity, and three had previously undergone a Darrach procedure. Twenty-one were examined personally and 11 by telephone. Follow-up averaged 51 months. Of the 21 patients examined, 6 experienced residual pain at the DRUJ. Of the 11 questioned by telephone, 5 had no pain, 3 had mild pain, 2 had moderate pain, and 1 had severe pain. Revisions for radioulnar impingement were performed on three patients.

Van Schoonhoven and coworkers¹⁰ evaluated 36 patients with matched hemiresection with interposition for post-traumatic osteoarthritis of the DRUJ with an average follow-up of 34 months. Preoperative pain scores measured 7.8, whereas postoperative scores were 3.9. Grip strength improved from 40% to 64% of the uninjured side. In 21 patients, radioulnar impingement was observed and required revision in 5 patients.

Lamey and Fernandez¹¹ report on 18 patients with post-traumatic osteoarthritic changes of the DRUJ treated with a modified SK procedure. Follow-up averaged 4 years. Supination improved dramatically from 16 degrees before surgery to 76 degrees after surgery, and pronation improved from 42 to 81 degrees. Grip strength went from 36% of the unaffected side to 73% of the unaffected side. Fifteen patients reported no pain in the area of the DRUJ, and three reported mild pain with rotation. Eleven patients had repeat operations, primarily for removal of hardware. The ulnar stump was reported as stable in 16 patients.

The use of ulnar head prosthesis for salvage of failed distal ulnar resections has been reported. Results of ulnar head prosthetics for salvage or primary indications has been promising. For example, Schecker and Severo¹² reported on 23 patients, but with a mean follow-up of only 15 months. All 23 patients were reported to be free of pain with a normal pronation/supination arc. Eight of the 13 patients who were out of work because of DRUJ disability returned to work. Implant longevity, given the short follow-up of most of these case series on prosthetic options, continues to be a concern.