Pain and Functional Outcomes After Sacroiliac Joint Injection with Anesthetic and Corticosteroid at Six Months, Stratified by Anesthetic Response and Physical Exam Maneuvers

Abstract

Abstract Objective To evaluate sacroiliac joint (SIJ) injection outcomes with local anesthetic and corticosteroid. Design Prospective cohort. Setting Single academic medical center. Methods Thirty-four patients referred for SIJ injection with a clinical diagnosis of SIJ pain underwent injections with 1:1 mixture of 2% lidocaine and triamcinolone 40 mg/mL. Pain provocation physical exam (PE) maneuvers were recorded immediately before and after injection. Outcome measures at two to four weeks and six months included pain numeric rating scale (NRS) and Oswestry Disability Index (ODI). Results For the analysis of outcomes by the overall group (not stratified by PE and/or anesthetic block), a 58.8% (95% confidence interval [CI] = +/-16.5%) ≥2 NRS reduction, a 32.4% (95% CI = +/-15.7%) ≥50% NRS reduction, and a 38.2% (95% CI = +/-16.3%) ≥30% ODI reduction were observed at two to four weeks, with similar improvements at six months. Outcomes stratified based on pre-injection PE did not reveal significant differences at either time point. The stratification based on the presence of 100% postinjection anesthetic response demonstrated a significant difference at two to four weeks for ≥50% NRS improvement. The true positive/true negative group (TP/TN) stratification demonstrated a significant difference for ≥50% NRS improvement at two to four weeks, whereas six-month outcomes for TP/TN demonstrated significant differences for ≥50% NRS and ≥30% ODI improvement. An increased injection response was observed with stratification of patients more likely to have true SIJ pain (i.e., TP), with TP/TN stratification demonstrating a 75% (95% CI = +/-30.0%) ≥2 NRS improvement and a 62.5% (95% CI = +/-33.5%) improvement of ≥50% NRS and ≥30% ODI for the TP group at two to four weeks, with similar results at six months. Conclusions SIJ steroid injection based on referral clinical diagnosis is unlikely to demonstrate true injection efficacy, and more specific selection criteria are warranted.