Objective To assess the midterm follow-up outcomes of total hip arthroplasty (THA) for the treatment of patients with juvenile-onset ankylosing spondylitis (JAS). Methods The clinical data of 81 patients (127 hips) with JAS (age≤16 years, JAS group) and 267 patients (391 hips) with adult onset ankylosing spondylitis (AAS) (age>16 years, AAS group) between January 2004 and March 2018 were retrospectively analysed. The baseline demographics, clinical, radiographic, and laboratory parameters were collected. Before operation and at last follow-up, the overall disease activity [Bath ankylosing spondylitis disease activity index (BASDAI)] and function status [Bath ankylosing spondylitis functional index (BASFI)], hip subjective score [Harris hip score (HHS)] and objective score [12-item short form health survey (SF-12), including physical component score (PCS) and mental component score (MCS)], and patient satisfaction for THA were reviewed. The major orthopedic complications, including periprosthetic infection, dislocation, periprosthetic fractures, and poor incision healing, were also recorded during the follow-up period. Results The comparison of preoperative baseline parameters showed that the body mass, body mass index, age of onset, age of surgery, disease duration, and the proportion of combined smoking history in the JAS group were significantly lower than those in the AAS group (P<0.05), the proportion of bilateral surgeries, proportion of uveitis, proportion of combined family history, C-reactive protein, albumin, and preoperative BASFI were significantly higher than those in the AAS group (P<0.05). Both groups were followed up. The follow-up time in the JAS group was 29-199 months, with an average of 113 months; in the AAS group was 35-199 months, with an average of 98 months. Incisions in both groups healed by first intention. During the follow-up period, there were 1 case of periprosthetic fracture, 1 case of dislocation, and 1 case of ceramic fragmentation in the JAS group, 1 case of periprosthetic infection and 6 cases of periprosthetic fracture in the AAS group. There was no significant difference in the incidence of complications between the two groups (P>0.05). At last follow-up, the BASDAI, BASFI, SF-12 MCS, SF-12 PCS, and HHS score of the two groups were significantly improved when compared with those before operation (P<0.05); but there was no significan difference in the difference of the above parameters before and after operation and the patient satisfaction between the two groups (P>0.05). Conclusion The midterm follow-up outcomes of THA for the treatment of JAS patients were reliable. A low age at disease onset did not exert a significant negative effect on THA reconstruction for the treatment of ankylosing spondylitis.