Randomized studies evaluating a placebo response rate using saline control injections
| Study | Number of patients | Study design | Type of blocks (placebo arm) | Placebo response or false-positive rate* | Comments |
| Revel et al 45 | 80–38 who received placebo | Randomized, parallel group | IA with 1 mL saline | 18% placebo response | Response defined as >75% relief. |
| Cohen et al 18 | 229–47 who received placebo | Randomized, three-arm parallel group | MBB with 0.5 mL saline | 30% placebo response | Response defined as >50% relief. |
| Lord et al 209 | 50 patients with whiplash | Randomized, three-period crossover | Cervical MBB with 0.5 mL bupivacaine, lidocaine and saline | 40% placebo response, false-positive rate (concordant relief with LA blocks but positive response to placebo) 19% | Response defined as longer pain relief (complete or profound) with bupivacaine than lidocaine, and a negative response to saline. False-negative rate (positive but discordant response to LA injections but negative response to placebo) 32%. Sensitivity 54%, specificity 88%. |
| Dias da Rocha et al 173 | 104 | Patient-blinded, one-way crossover | Lumbar MBB with 0.5 mL saline, and if negative, lidocaine | 16% placebo response | Response defined as ≥50% relief. All patients received saline and if negative, lidocaine MBB after 10 min. |
| Schütz et al 146 | 60 | Single-blind, three-way crossover | Bilateral lumbar IA injections with 1.5 mL of mepivacaine (verum), saline (placebo) or extra-articular needle placement without injection (sham) | Sham and placebo groups had 20% and 38% response rates, respectively, at 1 hour | Response defined as >2-point decrease in back pain. Non-inferiority shown between placebo and verum injection. |
Injections in lumbar spine unless specified.
*False-positive rate based on positive response to verum and negative response to sham injection.
IA, intra-articular; LA, local anesthetic; MBB, medial branch block.