Background: Since the time of Paul Harrington numerous implants have been introduced for correction of scoliosis, but none are ideal. Newer devices are very expensive, and in our country some patients cannot afford them. Objectives: The aim of this study was to compare the results of the Harrington rod (HR) device and the newer Cotrel-Dubousset (CD) device in treatment of adolescent idiopathic scoliosis (AIS). Materials and Methods: A retrospective review assessed patients with AIS admitted for spinal curve correction treated with HR (n = 120) and CD devices (n = 138) between October 1988 to April 2001 at the Shafa Yahyaeian Hospital, Tehran, Iran. We extracted information from the patient’s file and radiographs before, after and two years post-operation. Results: The mean age of patients was 16.7 ± 2.5 years. There was no statistically significant difference between the two groups regarding gender, age, curve before surgery, and percentage of flexibility. The mean curvature was 70 ± 20.7 in the HR and 64.81 ± 19.4 in the CD group before surgery (P = 0.09); and the mean curvature was 40 ± 16.3 and 26.58 ± 15.37 in HR and CD groups respectively after surgery (P = 0.156). The mean curvature was 47.2 ± 15.9 in HR and 31.2 ± 15.4 in CD groups at two years follow-up (P = 0.156). Conclusions: Results of many studies have shown no significant impairment in long-term quality of life and function in patients treated with Harrington rods. According to previously performed studies and the current study, surgical correction with Harrington rods seem to be comparable with the newer more expensive CD device. Although there is no doubt that the preference is to use newer devices in view of some disadvantages of HR, but this does not preclude using it for patients that cannot afford the newer devices.