We studied the long-term results of a prospectively se lected group of 24 patients with 12 anteroinferior and 12 posterior glenoid labral lesions; all patients had func tional instability but none had ligamentous detachment. After arthroscopic debridement, patients involved in throwing sports were not allowed to return to full athletic activity until full strength of the external rotators was achieved and documented on isokinetic evaluation. Fol lowup was 36 to 72 months with an average of 48 months. Follow-up isokinetic evaluation revealed an av erage +4.4% and +8.6% concentric strength and -4.3% and -0.4% eccentric strength of the operated shoulder compared with the uninvolved shoulder at 90 and 180 deg/sec, respectively. Long-term good or ex cellent results were achieved in 21 patients, and 16 were functioning at their preinjury level of sports activi ties. Sixty-two percent of baseball pitchers were unim paired in pitching. The average University of California Los Angeles shoulder rating score was 31 of 35 (11 excellent, 10 good, and 3 poor) and the average Rowe- Zarins ratings scale was 85 of 100 (8 excellent, 13 good, and 3 poor). These results justify an initial arthroscopic debridement of anteroinferior or posterior labral flap tears rather than capsulorrhaphy when there is no gross instability or Bankart lesion. © 1995, SAGE Publications. All rights reserved.