Traditionally open extensor tendon injuries in zones III to V (PIP to MP joints) have been treated with repair and immobilization in extension for 4 to 6 weeks. Early controlled motion protocols have been successfully used in zones VI and VII of the extensors. An early controlled mobilization protocol combined with strong repair for zones III to V extensor tendon lacerations was studied prospectively. From 1999 to 2003, 27 extensor tendon lacerations in 26 patients, mean age 34 years (range 14–70), were treated using dynamic extension splinting. Inclusion criteria were zone III to V, complete lacerations involving the extensor mechanism and possibly the dorsal capsule (without associated fractures or skin deficits) in patients without healing impairment. All injuries were treated in the emergency department with a core Kessler-Tajima suture and continuous epitendon suture. After an initial immobilization in a static splint ranging from 5 days (for zone V) to 3 weeks (for zone III), controlled mobilization was initiated with a dynamic splint that included only the injured finger. The patient was weaned off the dynamic splint 5 weeks after the initial trauma. The patients were treated in an outpatient basis and did not attend any formal physiotherapy program. The mean follow up was 16 months (range 10–24 months). No ruptures or boutoniere deformities were observed and no tenolysis was necessary. The mean TAM was 242deg for the fingers and 119deg for the thumbs. The mean grip and pinch strength averaged 85% and 88% that of the contralateral unaffected extremity. 77% of the patients achieved a good or excellent result in Miller’s classification. The mean loss of flexion was found to be greater than the mean extension deficit. The protocol described above was found to be safe, simple, functional, cost effective and reproducible for zone III to V simple extensor tendon injuries. Success is based on strong initial repair, close physician observation and a cooperative patient. The addition of physiotherapy for patients with flexion deficits in the period immediately after dynamic splinting may ameliorate results