Ultrasound of metacarpophalangeal joints is a sensitive and reliable endpoint for drug therapies in rheumatoid arthritis: results of a randomized, two-center placebo-controlled study

Abstract

Introduction: We aimed to investigate the sensitivity and reliability of two-dimensional ultrasonographic endpoints at the metacarpophalageal joints (MCPJs) and their potential to provide an early and objective indication of a therapeutic response to treatment intervention in rheumatoid arthritis (RA).

Methods: A randomized, double-blind, parallel-group, two-center, placebo-controlled trial investigated the effect on ultrasonographic measures of synovitis of repeat dose oral prednisone, 15 mg or 7.5 mg, each compared to placebo, in consecutive two-week studies; there were 18 subjects in a 1:1 ratio and 27 subjects in a 2:1 ratio, respectively. All subjects met the 1987 American College of Rheumatology criteria for the diagnosis of RA, were ≥18 years-old with RA disease duration ≥6 months, and had a Disease Activity Score 28 based on C-reactive protein (DAS28(CRP)) ≥3.2. Subjects underwent high-frequency (gray-scale) and power Doppler ultrasonography at Days 1 (baseline), 2, 8 and 15 in the dorsal transverse and longitudinal planes of all 10 MCPJs to obtain summated scores of quantitative and semi-quantitative measures of synovial thickness as well as vascularity. The primary endpoint was the summated score of power Doppler area measured quantitatively in all 10 MCPJs in the transverse plane at Day 15. Clinical efficacy was assessed at the same time points by DAS28(CRP).

Results: All randomized subjects completed the trial. The comparison between daily 15 mg prednisone and placebo at Day 15 yielded a statistically significant treatment effect (effect size = 1.17, P = 0.013) in change from baseline in the primary endpoint, but borderline for prednisone 7.5 mg daily versus placebo (effect size = 0.61, P = 0.071). A significant treatment effect for DAS28(CRP) was only observed at Day 15 in the prednisone 15 mg group (effect size = 0.95, P = 0.032). However, significant treatment effects at all time points for a variety of ultrasound (US) endpoints were detected with both prednisone doses; the largest observed effect size = 2.33. Combining US endpoints with DAS28(CRP) improved the registration of significant treatment effects. The parallel scan inter-reader reliability of summated 10 MCPJ scores were good to excellent (ICC values >0.61) for the majority of US measures.

Conclusions: Ultrasonography of MCPJs is an early, reliable indicator of therapeutic response in RA with potential to reduce patient numbers and length of trials designed to give preliminary indications of efficacy.

Trial registration: Clinicaltrials.gov identifier: NCT00746512.

Figure 1

Illustration of ultrasonographic scanning in the longitudinal and transverse plane using a splint to standardize image acquisition. The four columns contain the semi-quantitative scales with scores from 0 to 4: 0 representing the lack of PD signal and 4 being severe PD signal; 0 representing no synovial thickening and 4 being severe synovial thickening. The last row demonstrates the region of interest (ROI) for quantitative analysis. Images were cropped for clarity. LONG: longitudinal, TRANS: transverse, ST: synovial thickening, VASC: vascularity, Quant ROI: region of interest for quantitative analysis, STA: synovial thickness area, PDA: power Doppler area.

Figure 2

Within-panel comparisons (panel A and B) of US and DAS28(CRP) endpoint responsiveness. Effect size and 90% CI are presented. Power Doppler measures: PDA: Power Doppler Area; continuous measure of PD signal within the synovium; VASCi: Vascular Index (0 to 4 scale). Gray scale US measure of synovial swelling: STA: Synovial Thickness Area; area bounded by synovium; STi: Synovial Thickness Index (0 to 4 scale). ^* Significant effect: P <0.05. CI, confidence interval.