Dr. Margules should be congratulated on his recent letter discussing a study of 510 patients with fluoroscopically guided injections into the hip (1). He states that one of the reasons that corticosteroid injections are not commonly used in hip osteoarthritis (OA) is the technical difficulty in delivering such injections. As a consequence, corticosteroid injection is omitted from current treatment guidelines, although the same procedure is advised for the treatment of knee OA (2).
Ultrasonography (US), a safe, noninvasive procedure, may represent a credible alternative to fluoroscopy, especially owing to the lack of radiation exposure associated with the former procedure. In addition, US is able to detect features of OA such as joint space narrowing and osteophyte formation that may help predict the response to treatment. It can also localize collections of fluid within the joint, allowing guided aspiration if infection is a concern. In addition, US can be used to check the accuracy of the injection without the use of contrast and ionizing radiation, as is the case with fluoroscopy.
We now report a prospective study of US-guided injections into the hip joints of patients with severe hip OA, which was undertaken in order to determine the predictive value of US and radiography features. All patients fulfilled the American College of Rheumatology criteria for hip OA (3), had ongoing pain and disability despite use of nonsteroidal antiinflammatory drugs and analgesics, and were on the waiting list for a total hip replacement (THR). Informed consent was obtained, and the study was approved by the local research ethics committee.
The baseline assessment included documentation of analgesic requirements and use of a patient 100-mm visual analog scale (VAS) for pain when walking. Radiographs of the affected hip were scored as mild, moderate, or severe, and US was performed using a standardized technique (4). The hip joint was then injected with 40 mg of triamcinolone and 2 ml of 1% lidocaine under US guidance; all injections were accurately placed.
Followup visits were performed at 2, 6, and 12 weeks when the response to corticosteroid injection (defined as a reduction of >15 mm in the VAS pain score for walking) was noted. Baseline findings on radiography and US were analyzed for any features that might predict the response to treatment.
Eleven patients were recruited, 7 of whom were female; the mean age of the patients was 63 years (range 53–72 years). The mean VAS score for pain at baseline was 78 mm. Six patients had severe OA changes on radiography, and the remaining 5 had moderate OA. Eight patients had an effusion on US (4 with moderate and 4 with severe radiographic OA). Five patients had anterior osteophytes on US (3 with severe and 2 with moderate radiographic OA).
Overall, 6 patients (55%) described a response at 2 weeks, compared with 4 (33%) of 11 patients and 3 (38%) of 8 patients at 6 and 12 weeks, respectively. No patient had improvement in range of motion (ROM) or the Western Ontario and McMaster Universities Osteoarthritis Index (5). There were no complications such as joint infection.
Of the 6 responders at 2 weeks, 2 had severe radiographic OA, but both had US effusion, compared with 4 who had moderate OA (3 of whom had US effusion). At 6 weeks, 3 of 4 responders had moderate radiographic OA, all with US effusion, and these 3 continued to benefit from therapy up to week 12. All 5 patients with an effusion but without osteophytes on US responded at week 2, with 4 (80%) of 5 patients and 3 (75%) of 4 patients continuing to respond at 6 and 12 weeks, respectively. Only 1 of 5 patients (20%) with osteophytes on US responded at 2 weeks.
In this study, severe radiographic OA and osteophytes on US were both (not surprisingly) associated with a poor response to treatment (20%), whereas patients with an effusion but no osteophytes on US had the greatest chance of a sustained response (75%), suggesting that this may be a predictor of favorable response.
In this study, the response to injection was less than that described by Margules (1). However, all of our patients were awaiting THR, suggesting greater disease severity, as highlighted by the lack of improvement in ROM and function. Whether the use of a potentially more potent corticosteroid such as triamcinolone hexacetonide prolongs response (6) is unclear.
Our preliminary evidence suggests that certain patients with end-stage OA who are awaiting THR may benefit from corticosteroid injection, and radiographic and US findings may be used to help predict that response. In addition, US represents a superior means of accurately delivering corticosteroid into the hip joint. A large, randomized controlled trial is required to further investigate these preliminary observations.