Journal of Arthritis

Osteoarthritis (OA) of the hip is a frequent cause of pain, disability and has significant social and economic impact [1,2]. Conservative treatment of OA includes activity modification, weight loss, physical therapy, non-steroidal anti-inflammatory medications (NSAIDs), Acetaminophen, physical therapy as well as intra-articular injections [3-7].

Intra-articular injections of the hip are typically performed for diagnostic and/or therapeutic reasons. For diagnostic purposes, an intra-articular anesthetic can be beneficial to distinguish symptomatic intra-articular pathology from potential extra-articular sources of hip pain [8-10]. Intra-articular cortisone injection is also commonly used in patients with osteoarthritis of the hip for therapeutic reasons. Studies, however, have demonstrated uncertain efficacy, [11,12] modest improvement [13] insignificant benefit [14,15] and an unpredictable duration of beneficial pain relief [16]. Other studies have demonstrated good efficacy, with 1–2 months pain relief [17-19] While intra-articular corticosteroid injections are described as a low-risk treatment option for those patients with osteoarthritis of the hip, with the need for repeated injections studies have raised concerns of chondrotoxicity [20,21].

Viscosupplementation by intra-articular injections of hyaluronic acid (HA) has shown beneficial effects in the treatment of OA [22]. HA reduces pain associated with OA by several mechanisms, including inhibition of tissue nociceptors, stimulation of endogenous HA, direct anti-inflammatory effects, and inhibition of matrix metalloproteinase activity [23]. Histological and magnetic resonance studies have also demonstrated a chondro-protective effect [23,24]. While a vastc majority of the literature has focused on HA in knee OA, there has recently been an increase in the number of studies on HA in the hip joint demonstrating its efficacy [1,6,25-32].

The combination of HA and cortisone has been demonstrated to be safe and effective in the knee [33]. Ozturk et al. published their results of a randomized controlled trial of 40 patients with knee osteoarthritis who received either an injection of HA alone or HA plus steroid (triamcinolone) [33]. They reported no progression of osteoarthritis in either group at 1 year follow-up. However, they found that pain decreased more rapidly and to lower levels with the combination injection compared to HA alone at 1 year follow-up. Grecomoro et al. published similar results in a randomized controlled trial comparing pain relief in knee osteoarthritis patients receiving either HA or combination HA and steroid injections [34]. They also found faster pain relief and lower VAS pain scores in the combination injection group. There have been no studies on the use of a combination of HA and cortisone on the hip. The purpose of the present study was to determine the efficacy of a combination of intraarticular HA and cortisone when compared to cortisone alone in the treatment of symptomatic hip OA.

This study was approved by the Institutional Review Board at our institution. All hip injections carried out from January 2009 to December 2014 were retrospectively reviewed. Inclusion criteria for the study included the following: (1) a diagnosis of symptomatic hip OA and (2) fluoroscopic guided intra-articular injection of corticosteroid. Exclusion criteria included the following: (1) intraarticular injection performed outside of our institution and (2) prior ipsilateral hip surgery. These were then matched in a 1:1 ratio with respect to age, gender, BMI and diagnosis with patients who underwent fluoroscopic guided intra-articular injection of corticosteroid and HA.

Pain severity was evaluated using the numerical rating scale (NRS), which measures pain intensity on a scale from 0 to 10 [28,35]. We considered an absolute change of 2 points on the NRS score, clinically significant pain relief, as previously reported [36,37]. Post-injection pain was recorded as the lowest pain rating in the 24-48 hours following injection. We also recorded the overall duration of pain relief obtained from the intra-articular injection.

Radiographic assessment including AP, false-profile, Dunn and cross-table lateral radiographic images of the involved hip was obtained prior to the injection. The degree of osteoarthritis was determined for each symptomatic hip utilizing the Tonnis classification [34,38]. The grade for each hip was determined based on the individual radiographic view with the highest grade [34,38].

All intra-articular injections were performed by the senior author under fluoroscopic guidance in the operating room under conscious sedation. Following a surgical time-out to confirm the correct patient and operative site, fluoroscopy was utilized to localize the hip joint. Five mL of 1% lidocaine was injected into the skin, subcutaneous and deeper tissue on a tract down to the joint. A 22 gauge needle was placed into the joint and an air arthrogram was performed to confirm intra-articular placement of the needle. Two mL of cortisone (40 mg/mL Kenalog, Bristol-Meyers Squibb, New York, NY) was then injected with or without 2 mL of HA (Supartz, Bioventus, Durham, NC). The hip was then taken through a range of motion to allow the injectant to coat the entirety of the articular surface. The procedure was performed in an outpatient setting with instructions to return to full unrestricted activity as tolerated after completion.

All data were summarized and reported using mean (range) for continuous, median (range) for ordinal variables, and count (percentage) for categorical variables, unless otherwise specified. Patient characteristics and pain outcomes were compared between patients receiving cortisone and a combination of cortisone with HA using nonparametric Wilcoxon rank-sum tests for variables measured on a continuous or ordinal scale, and chi-square tests for categorical variables. Pain ratings were evaluated and compared using Friedman’s test (a nonparametric analogue to repeated measures ANOVA) and the associated rank- sum multiple comparisons test to identify pair-wise differences while maintaining the overall type-I error rate at 0.05. All statistical tests were two-sided, and p values <0.05 were considered statistically significant. All analysis was conducted using SPSS version 15.0 (SPSS Inc. Chicago, IL).

From January 2009 to December 2014, 438 hip injections were performed at our institution. After applying inclusion and exclusion criteria, 119 patients were included in the cortisone group. Fifty-five were male (46%) and 64 were female (54%). Mean patient age was 50.5 ± 10.2 years (range 36–75). Mean patient BMI was 25.1 ± 3.4 (19.5-35.7) and mean duration of follow-up was 26.4 ± 15.4 months (range 6-66). These were then matched in a 1:1 ratio with respect to age, gender, BMI and diagnosis with 119 patients who received a combined HA and cortisone injection (combined group). Fifty-five were male (46%) and 64 were female (54%). Mean patient age was 48.7 ± 9.8 years (range 34-72). Mean patient BMI was 24.7 ± 3.6 (20.3-38.9).

The median pre-operative NRS score for the C group was 7 (5-9) and the post-operative score was 2 (0-3), with the difference being statistically significant, p=0.0003. The median pre-operative NRS score for the HA group was 7 (5-10) and the post-operative score was 1(0-4), with the difference being statistically significant, p=0.00004 (Figures 1-3). The mean duration of pain relief was 32.3 ± 8.6 (3-72) days for the C group and 128.6 ± 20.3 (2-184) days for the HA group, with the difference also being statistically significant, p=0.000001 (Figure 4).

Figure 1: Median pre- and post- injection NRS scores for cortisone group.

Figure 2: Median pre- and post-injection NRS scores for combined group.

Figure 3: Comparison of post-injection NRS scores between cortisone and combined groups.

Figure 4: Comparison in duration of pain relief between cortisone and combined groups.

There were no significant differences between the two groups with respect to age, BMI, duration of follow-up, pre-operative or postoperative pain scores (p=0.412, p=0.483, 0.0579, p=0.792, p=0.195 respectively) (Table 1).

Table 1: Comparison of demographic variables between cortisone and combined groups.

This is the first study to evaluate the efficacy of a combination of intra-articular HA and cortisone when compared to cortisone alone in the treatment of symptomatic hip OA. Although we found no statistically significant difference in post-injection pain scores, there was a significant increase in the duration of pain relief for the group that received both HA and cortisone.

Several studies have demonstrated the efficacy of intra-articular cortisone in hip OA. Flanagan et al., Micu et al., Atchia et al., all demonstrated about one month of relief [14,17,18]. Robinson et al. were able to demonstrate up to 6 weeks of relief [19]. Results from the present study compare favorably with the literature demonstrating just over 4 weeks of relief for the cortisone group alone.

The efficacy of HA in hip osteoarthritis has also been previously studied. Qvistgard et al., demonstrated about 10 weeks of relief while Milgore et al., and Conrosier et al. both demonstrated close to 12 weeks of relief [6,27]. Spitzer et al., demonstrated almost 16 weeks of relief with course of 3 weekly injections [31]. The present study was able to obtain over 16 weeks of relief from a single injection of a combination of HA and cortisone.

There have been numerous studies of sequential injections and the demonstration of a dose-response curve for HA in the knee with a consensus of 3 week weekly injections of HA providing the maximum benefit [2,12,30,32,38]. In the United States, HA is not currently FDA approved in the hip making its use “off-label.” This combined with the use of fluoroscopy and OR anesthesia makes it both practically and economically unfeasible to provide serial injections currently. While no such studies are available for the hip, further studies are necessary to determine this dose-response relationship in the hip.

The present study does have several limitations. Retrospective studies introduce inherent selection, recall and treatment biases. In addition the patient population was small and heterogeneous in nature making comparisons difficult even with a control group. Additionally there were no functional scores available for this group of patients. This has been rectified and we have modified our practice to begin collecting these scores prospectively. Additionally the strengths of the study are that the study is adequately powered, a control group was employed allowing for accurate comparisons between groups, and a single surgeon utilizing a single technique performed all of the injections.

The implementation of combination HA and cortisone injection compared to cortisone alone could have several positive impacts in the treatment of hip osteoarthritis. In theory, this new treatment may lead to less doctor’s appointments and a decreased need for repeat injections as the pain relief is longer. This treatment may also have a secondary economic benefit due to patients not having to miss as many days of work. Finally, our hope is that the increased duration of pain relief from these injections may allow patients to delay surgical intervention, including hip replacement, for their condition.

This is the first study to evaluate the efficacy of a combination of intra-articular hyaluronic acid and cortisone in hip OA. While no difference in post injection pain relief when compared to cortisone alone was noted, there was a significant difference in the duration of pain relief with the combination providing more than 3 additional months of relief. Due to the retrospective nature of the study, further research is needed to determine if these results bear out in doubleblinded prospective randomized controlled trials. Other additional areas of study include establishing a dose response curve and determining if the combination also confers improvement in functional scores.