Clinical and cost-effectiveness of ultrasound-guided intra-articular corticosteroid and local anaesthetic injection for hip osteoarthritis: a randomised controlled trial (HIT)

Abstract

Purpose: Evidence of the effectiveness of intra-articular corticosteroid injection for hip osteoarthritis (OA) is limited. The HIT trial compared the clinical and cost-effectiveness of an ultrasound-guided intra-articular hip injection (USGI) of 40mg triamcinolone acetonide and 4ml 1% lidocaine hydrochloride combined with best current treatment (BCT) with (i) BCT alone (primary objective) and (ii) an USGI of 5ml 1% lidocaine only combined with BCT (EudraCT: 2014-003412-37). Methods: This was a pragmatic, three-parallel group, single-blind, randomised controlled trial in adults with moderate to severe painful hip OA recruited from community musculoskeletal services and primary care. Participants were randomised equally to: (1) BCT alone, (2) BCT plus USGI triamcinolone/lidocaine, or (3) BCT plus USGI lidocaine only. BCT comprised written information and individualised advice regarding weight loss, exercise, footwear, walking aids and analgesia. Outcomes were collected postally at 2 weeks, 2, 4 and 6 months. The primary outcome was self-reported current hip pain intensity (0-10 numeric rating scale (NRS)) over 6 months (repeated measures analysis). A wide range of secondary outcomes were measured at each follow-up point including pain and physical function (Western Ontario and McMaster University Arthritis Index, WOMAC), pain self-efficacy, participant’s global impression of change, general health (EQ-5D-5L) and healthcare utilisation. Participants and clinicians were not blind to allocation to BCT alone or injection, however participants randomised to either injection arm were blind to the nature of injection. The research nurse involved in data collection and study statistician were blind to treatment allocation. 204 participants were required to detect a minimum difference of 1 point in mean pain NRS score between arms (1) and (2) with 80% power (5% two-tailed significance level, 15% loss to follow-up). Analysis was by intention-to-treat. To determine the cost-effectiveness of BCT plus USGI triamcinolone/lidocaine compared with BCT alone, a cost-utility analysis was conducted, adopting an NHS perspective. QALYs were calculated over a period of 6 months using the area under the curve approach. Results: 199 participants were recruited (43% male, mean age 63 years), 67 to arm (1) and 66 each to arms (2) and (3). Primary outcome completion rates were 95% at 2 weeks, 94% at 2 months, 90% at 4 months, and 89% at 6 months. Greater mean improvement in hip pain intensity (0-10 NRS) over 6 months was seen with BCT plus USGI triamcinolone/lidocaine compared with BCT alone: -1.43 (95% CI -2.15, -0.72). Greater mean improvement in pain intensity was seen at 2 weeks (-3.17; -4.06, -2.28,) and 2 months (-1.81; -2.71, -0.92,), but not at 4 months (-0.86; -1.78, 0.05) or 6 months (0.12; -0.80, 1.04). Participants treated with BCT plus USGI triamcinolone/lidocaine compared with BCT alone had greater mean improvement in function (WOMAC-F -5.47; (-9.41, -1.53)) and pain self-efficacy (5.87; 2.30, 9.45) over 6 months. More participants reported being completely recovered/much better at 2 months with BCT plus USGI triamcinolone/lidocaine than with BCT alone (45.4% v 6.9%; relative risk=6.7; 2.5, 17.9). There was no statistically significant difference in hip pain intensity over 6 months between BCT plus USGI triamcinolone/lidocaine compared with BCT plus USGI lidocaine only (-0.52; -1.21, 0.18). However, a statistically significant difference was seen in favour of the group receiving USGI triamcinolone/lidocaine in comparison to USGI lidocaine alone for a range of secondary outcome measures, over 6 months (e.g. pain self-efficacy, function, and work presenteeism). There was one possible treatment-related serious adverse event: a participant with no signs of infection at randomisation died from endocarditis four months after USGI triamcinolone/lidocaine. BCT plus USGI triamcinolone/lidocaine was less costly (mean cost difference per participant £-161.59) and associated with significantly higher quality-adjusted life-years (QALYs) than BCT only over 6 months (mean difference 0.0477 (0.0257, 0.0699). Increased number of visits to NHS consultants and hip-related surgery in the BCT only group accounted for the majority of cost difference. Conclusions: USGI triamcinolone/lidocaine combined with BCT leads to greater improvements in pain and function over 6 months in adults with hip OA than BCT alone, and was highly cost-effective. There was no significant difference in hip pain intensity over 6 months between the groups receiving USGI triamcinolone/lidocaine and USGI/lidocaine only raising the possibility of a degree of placebo effect.