The response to pericapsular soft tissue and pelvic realignment therapy may be partially predicted by the relevant factors influencing the program's response of the candidates with hip osteoarthritis for joint replacement.

Objective: To explore the risk factors for discontinuation of pericapsular soft tissue and pelvic realignment (PSTP-R) therapy derived from Shiatsu in the candidates with osteoarthritis for total hip replacement (THR) (i.e., candidates for total hip replacement) treated from 2017 to 2020, and to identify the effect modifiers of PSTP-R therapy for patients who continued therapy for 6 months.

Methods: Exploratory analyses of data from the clinical trial for PSTP-R therapy in a prospective observational study in which the PSTR exercises for 6 months improved Harris Hip Score (HHS) in 193 patients with an HHS below 60 points in two study centers, even in 130 patients with complete loss of cartilage on radiography among those 193 ones. The risk factors for the discontinuation of PSTP-R therapy and the effect modifiers for PSTP-R therapy were explored statistically.

Results: The risk of discontinuation of PSTP-R therapy increased as the frequency of buttock pain at baseline increased, and was mitigated as the baseline opening angle of the hip according to the modified Patrick's test increased. Cartilage loss on radiography was not a risk factor for withdrawal from PSPT-R therapy. Among patients who continued PSTP-R therapy for 6 months, a lower Kellgren-Lawrence grade at baseline was associated with an increase in the total score of HHS.

Conclusion: Buttock pain at baseline was most associated with discontinuation of PSPT-R therapy. The patients that can improve with PSPT-R therapy should be selected to avoid inappropriate surgery by the detailed analysis of relevant clinical factors influencing the response for this program as well as image findings such as cartilage loss on radiography at baseline.

Trial registration: Registered in the Clinical Trials Registry on 20^th July 2017 (UMIN000028277).