- Design
- Systematic review and random-effects meta-analysis of randomised controlled trials with post hoc subgroup analysis and meta-regression, searched to June 2026
- Population
- 10 randomised trials of perioperative intravenous lidocaine versus placebo or standard care in adult spine surgery; 655 patients in the primary pain analysis
- Primary outcome
- Postoperative pain intensity at 24 hours, with opioid consumption, nausea and vomiting and length of stay as secondary outcomes
- Effect
- Pain -0.83 (95% CI -1.36 to -0.30, I-squared 89%, prediction interval -2.69 to 1.02); opioids -11.64 mg morphine equivalents (-16.14 to -7.14); fusion -1.23 (-1.81 to -0.64) vs decompression -0.20 (-0.61 to 0.21), subgroup p=0.005
Intravenous lidocaine is a standard opioid-sparing adjunct with an evidence base built by pooling every kind of spine operation together. This meta-analysis, searching to June 2026, asked whether that pooling has been hiding a procedure-specific effect across ten randomised trials.
Overall, lidocaine reduced 24-hour pain by 0.83 points (95% CI -1.36 to -0.30, moderate certainty) and opioid consumption by 11.64 mg of intravenous morphine equivalents (-16.14 to -7.14). Heterogeneity was high at 89%, and the prediction interval ran from -2.69 to +1.02 - meaning a future trial could plausibly find lidocaine slightly worse than placebo.
The subgroup analysis explains why. After instrumented fusion or complex spine surgery the reduction was -1.23 points (-1.81 to -0.64), above the minimal clinically important difference of 1.0. After decompression it was -0.20 (-0.61 to 0.21), which is nothing. The test for subgroup difference was significant at p=0.005, and baseline pain severity explained about 53% of the heterogeneity - a coherent story in which lidocaine helps where there is more pain to relieve. Infusion rate did not modify the effect at all, which argues against chasing higher doses. No effect on nausea and vomiting or length of stay.
The subgroup analysis was post hoc and exploratory, so this is a strong hypothesis rather than a settled fact. But it is a biologically sensible one, it is supported by the meta-regression on baseline pain, and it points the same way as clinical experience.
The practical use is in deciding where to spend the infusion pump and the monitoring. Use lidocaine for instrumented fusion and complex reconstruction; do not add it routinely to a single-level decompression, where it costs setup and vigilance for no measurable analgesia.
- Use perioperative intravenous lidocaine for instrumented fusion and complex spine surgery
- Do not add it routinely to simple decompression - the effect was 0.2 points
- Do not increase the infusion rate to chase effect; rate did not modify the result
- Expect no benefit for nausea, vomiting or length of stay
- The subgroup finding is post hoc - treat it as a strong hypothesis, not a settled rule
The statistics, in plain English
A mean difference of -0.83 points on a 10-point scale is below the 1.0-point threshold usually considered noticeable, so the overall result is statistically significant and clinically marginal - which is exactly what averaging a real effect with a null one produces. I-squared of 89% means the trials disagreed strongly, and the prediction interval (-2.69 to +1.02) is the honest guide to what the next trial might find: it includes no benefit. Post hoc subgroup analyses are the classic source of spurious findings, but this one is supported by a meta-regression showing baseline pain severity explains about half the heterogeneity, which is independent corroboration rather than the same claim restated.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for anaesthesiology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free