25. Perioperative complications associated with epidural injections followed by cervical decompression and fusion

Abstract

<h3>BACKGROUND CONTEXT</h3> Cervical epidural steroid injections (CESIs) are widely used to alleviate pain caused by degenerative diseases of the cervical spine. However, some data suggest a potential link between CESIs and infection rates following fusion surgeries. No prior studies have assessed the association between CESI and other postoperative outcomes such as hematomas, revision rates, or wound complications following anterior cervical discectomy and fusion (ACDF) and posterior cervical decompression and fusion (PCDF). Therefore, using a national dataset, we aimed to explore whether an association exists between preoperative CESIs and a variety of serious complications following ACDF and PCDF surgery. <h3>METHODS</h3> In this retrospective cohort study, we queried the IBM MarketScan national claims dataset (after IRB approval) and included adult patients who had undergone inpatient or outpatient ACDF or PCDF surgery due to disc herniation, stenosis, radiculopathy, spondylosis, or myelopathy from January 2014 to September 2017. Current procedural terminology codes were used to identify patients who received a CESI within one year prior to their surgery. Multivariable logistic regression measured the association between preoperative CESI and postoperative rates of 90-day infection, hematoma, wound complications, revision surgery, or combined complications (the former first three complications). We report odds ratios (OR) and 95% confidence intervals (CI). <h3>RESULTS</h3> Among 44,777 ACDF patients, 31.2% (n=13,950) received a preoperative CESI. For PCDF, 3,284 patients underwent the surgery of which 22.3% (n=734) received a preoperative CESI. The median time from most recent CESI to ACDF surgery is 76 days (interquartile range 41-144 days) and 95 days (interquartile range 49-176 days) for PCDF. Among ACDF patients, preoperative CESI was associated with lower odds of developing a hematoma following ACDF (OR 0.60; 95% CI 0.37-0.99; p=0.0453); no significant associations existed for the other outcomes including infection (OR 0.98; 95% CI 0.77-1.26; p=0.8846), revision surgery (OR 0.94; 95% CI 0.71-1.23; p=0.6300), wound complications (OR 0.92; 95% CI 0.71-1.17; p=0.4865) or combined complications (OR 0.88; 95% CI 0.73-1.07; p=0.1904). In the PCDF cohort, no significant associations existed between preoperative CESI and post-PCDF infection (OR 1.26; 95% CI 0.82-1.95; p=0.2962), revision surgery (OR 1.96; 95% CI 0.91-4.18; p=0.0842), wound complications (OR 1.11; 95% CI 0.71-1.72; p=0.6608) or combined complications (OR 1.14; 95% CI 0.78-1.68; p=0.4929). <h3>CONCLUSIONS</h3> Our study is the first to assess a broad complication profile that includes hematoma rates, infection rates, revision rates, and wound complications following ACDF and PCDF surgeries. We found no association between preoperative CESI and the aforementioned complications following ACDF and PCDF which supports first-hand clinical outcomes. Future work will aim to delineate the role of timing of CESI before ACDF and PCDF surgery, and the impact of quantity of CESIs on complication risks. <h3>FDA DEVICE/DRUG STATUS</h3> This abstract does not discuss or include any applicable devices or drugs.