Modified Extensor Pollicis Longus Rerouting Technique for Boutonniere Deformity of the Thumb in Rheumatoid Arthritis

PurposeTo assess the outcomes of a modified extensor pollicis longus (EPL) rerouting technique for boutonniere deformity of the thumb in patients with rheumatoid arthritis.MethodsA total of 21 thumbs in 18 patients with a mean age of 63 years were retrospectively analyzed after an average follow-up period of 3.2 years. The preoperative deformities were classified as either mild (5 thumbs) or moderate (16 thumbs). After either metacarpophalangeal (MCP) joint synovectomy or implant arthroplasty, the ulnarly dislocated EPL tendon was reduced dorsally and sutured to the dorsal base of the proximal phalanx. If the interphalangeal (IP) joint extended with manual traction on the proximal portion of the extensor pollicis brevis tendon, no further treatment was considered. If the IP joint did not extend with this maneuver, the insertion of the extensor pollicis brevis tendon was dissected and transferred to the distal portion of the EPL tendon.ResultsThe average MCP joint extensor lag improved from 62° (range, 32° to 85°) before surgery to 17° (range, active extension 12° to extensor lag 70°) at the final follow-up (P < .05), whereas average MCP joint flexion decreased from 83° (range, 52° to 95°) to 68° (range, 30° to 90°) (P < .05). Hyperextension at the IP joint was improved from 30° (range, 10° to 50°) before surgery to an average extensor lag of 2° (range, extensor lag 24° to hyperextension 20°) at the final follow-up. The average combined MCP and IP motion did not significantly change. The boutonniere deformity was improved in 18 of 21 thumbs. The 3 failures all had moderate-stage deformity prior to treatment.ConclusionsA modified EPL rerouting technique provided satisfactory results together with a low risk of IP joint extension loss.Type of study/level of evidenceTherapeutic IV.