Summary

Frozen hip, an uncommon cause of hip pain, should be suspected in patients with pain that is not proportional to their radiological findings, and in patients in whom a marked decrease in external rotation is encountered in the setting of a non-arthritic hip. Conservative management is recommended as the initial treatment, and this study sets out to assess the effectiveness of this approach.

Abstract

Introduction

Adhesive capsulitis is a well-known entity in the shoulder. However, it may also affect other joints including the hip (aka frozen hip). There are few reports in the literature, most of which focus on surgical treatment. The goal of this study is to report clinical results of a consecutive group of patients with frozen hip to ascertain the success rate of non-operative and operative treatment.

Methods

A retrospective chart review was performed of patients seen between January 2010 and July 2016 in a university sports medicine practice with a hip emphasis. The diagnosis of frozen hip was established when a patient presented with hip joint pain associated with loss of internal rotation (IR) in addition to external rotation (ER) loss (>10o side-to-side ER difference). Standard AP and lateral radiographs excluded gross signs of FAI, dysplasia and osteoarthritis. Pain relief of at least 50% with image guided intra-articular injection confirmed an intra-articular pain generator. Patients were initially treated with physical therapy (PT) to improve range of motion (ROM) by joint mobilization, traction and stretching, followed by strengthening after near symmetric ROM was achieved.

Results

Sixteen patients (14 females – 87.5%) with an average age of 43. years + 8.7 were studied. Mean ROM was (affected side/non-affected side): Flexion 115.3o+21.8 / 125.6o+17.0, IR 17.6o+15.9 / 22.6o+13.6, ER 27.1o+17.3 / 55.3o+14.9. The mean side-to-side ROM difference was Flexion 10.3o +13.7, IR 5.0o+12.2 and ER 27.1o+17.3. Three patients (18.8%) had a prior history of frozen shoulder, 2 of these 3 patients reported a traumatic onset of pain and one of these 2 patients presented as a non-insulin dependent diabetic. One patient (6.3%) had Crohn’s disease. The other 12 patients (75.0%) reported neither co-morbidities nor trauma, thus were considered idiopathic. Six patients (37.5%) underwent surgery after failing an average 11 months (range, 1 -32) of PT. Surgery consisted of manipulation under anesthesia, arthroscopic synovectomy and labral treatment. At a mean follow-up of 46.2 months + 29.9,14 (87.5%) patients reported satisfactory improvement of their pain. One patient had a hip arthroscopy recently after failing 32 months of conservative treatment, and one surgical patient still presented pain on final follow-up.

Discussion/Conclusion
Frozen hip has many similar clinical characteristics to those of the shoulder, such as nonspecific and painful global loss of range of motion. Frozen hip should be considered in the differential diagnosis of non-arthritic hip pain, especially in patients presenting with pain that is not proportional to their radiological findings and in patients in whom a marked decrease in ER is encountered, and otherwise normal / non-arthritic radiographs.
Joint mobilization, traction and stretching appear effective in gaining ROM and decreasing pain in the frozen hip, similar to what is found in the shoulder.
In conclusion, more than half the patients in this series were successfully treated with formal rehabilitation alone, while 37.5% of patients ultimately came to surgery for painful frozen hip. Conservative treatment should be considered for the first line of treatment for frozen hips.