The difference was small in absolute terms
Adults who received an epidural spinal injection before posterior lumbar interbody fusion, or PLIF, had more infections recorded within 120 days of surgery than patients without an injection, according to a retrospective cohort study. Infection occurred in 3.17% of injection patients and 2.32% of the no-injection group. That is an absolute difference of 0.85 percentage points, although the finding describes an association rather than proving the injection caused the infections.
The statistical signal was strong, but the absolute gap was modest. The study reported an odds ratio of 1.38, with a 95% confidence interval from 1.22 to 1.53 and a p-value below 0.0001. An odds ratio compares the odds of an outcome between two groups, so the figure captures a relative difference, while the observed infection rates show the smaller gap in actual percentage points. The authors said that small difference might not be clinically significant and reported a number needed to harm of 118.
The analysis also examined two other complications. Dural leaks occurred in 0.21% of people with ESI and 0.19% of the no-injection group, a difference the analysis did not find statistically significant, with p = 0.8269. Nerve injury was recorded in 0.16% of the ESI group and 0.09% of the comparison group. The reported odds ratio was 1.75, with a 95% interval of 1.04 to 2.93 and p = 0.0472. Nerve injury remained rare in both groups.
The closer the injection, the higher the infection rate
The relationship with timing was more pronounced for infection. Compared with the 2.32% no-injection rate, infections occurred in 2.98% of patients whose most recent ESI was 3 years to 1 year before PLIF, 3.17% when it was 1 year to 6 months before surgery, 3.45% at 6 months to 6 weeks, and 3.71% within 6 weeks of surgery. The corresponding odds ratios were 1.29, 1.37, 1.50 and 1.62. The reported intervals for those estimates were 1.10 to 1.52, 1.05 to 1.80, 1.20 to 1.89 and 1.08 to 2.44, respectively.
The records cannot settle cause or timing
Researchers drew on a retrospective cohort of 172,422 adults aged 18 to 85 who underwent PLIF between 2016 and 2022. Of them, 9,740, or 5.6%, had preoperative ESI. The comparison group was PLIF patients with no injection in the preceding three years. The study counted ESI exposure up to three years before surgery and excluded complications that occurred more than 120 days after PLIF.
Investigators used chi-squared tests on the cohort pooled across years and summarized categorical outcomes with odds ratios and 95% confidence intervals. The analysis used PearlDiver's R package for statistical procedures. Because the design was retrospective, its findings show associations in the data, not that ESI caused infection, dural leak or nerve injury.
That limitation matters because the authors could not account for comorbidities such as diabetes, smoking or immunosuppression. They also noted that people who received ESI may have had more severe or persistent symptoms. Those differences create possible confounding and selection bias, meaning the observed association could partly reflect who received injections rather than the injection itself.
The timing result should not be read as a tested minimum waiting period. The study found the strongest infection association closest to surgery, but the authors said the optimal interval remains uncertain. A time-dependent relationship between ESI and nerve injury could not be demonstrated because nerve-injury incidence was low. Separate analyses by number of injections were not possible because few patients received multiple injections.
The authors suggested that ESIs may remain safe in many contexts, while acknowledging that the small absolute infection difference might not be clinically significant. They reported no financial support for the research, authorship or publication and declared no potential conflicts of interest.
Paper data and sources
Original title: Effect of Preoperative Epidural Spinal Injections on Complications After Posterior Lumbar Interbody Fusion.
Authors: Dardiny Saint-Clair, Varun Gopinatth, Bethlehem Dawit et al.
Journal/Repository: Global spine journal
Status: Peer-reviewed
First online: 2026-08-21
DOI: 10.1177/21925682261481449
Original paper