Impact of Preoperative Opioid Use on Postoperative Patient-reported Outcomes in Lumbar Spine Surgery Patients.

STUDY DESIGN This was a retrospective cohort study. OBJECTIVE Investigate the impact of preoperative opioid use on postoperative Patient-Reported Outcomes Measurement Information System (PROMIS) physical function (PF) and pain interference (PI) scores in patients undergoing elective spine surgery. BACKGROUND DATA The PROMIS has demonstrated reliability and validity in conditions such as lumbar stenosis, disc herniation, and cervical spondylosis. Although previous studies have identified the negative impact of preoperative opioid use on legacy patient-reported outcome measures following lumbar spine surgery, no study to date has utilized PROMIS computer adaptive tests. METHODS Consecutive patients who underwent lumbar spine surgery at a single institution between 2014 and 2016 completed PROMIS PF and PI scores at baseline preoperatively and at 3, 12, and 24 months postoperatively. Preoperative opioid use was defined as >1 month before surgery. Univariate and linear mixed model multivariate analysis was performed to evaluate for correlation of preoperative opioid use, as well as patient risk factors, with postoperative PROMIS PI and PF scores at each time point. RESULTS Ninety-one patients met inclusion criteria with PROMIS scores at every time point. A total of 36 (39.6%) patients self-reported taking opioids at the time of surgery. Mean duration of opioid use among opioid users was 6.5±7.4 months. Patients taking preoperative opioids had significantly less improvement at all time points out to 24 months. At 24 months, patients in the nonopioid group had mean PI improvement of -13.0±14.2 versus -4.9±15.4 in the opioid group (P=0.014). The mean postoperative improvement in the opioid group did not achieve minimally clinically important difference (MCID) of 8 at any time point. CONCLUSIONS Patients who do not use opioids preoperatively show significant postsurgical improvement in PI scores compared with patients who use preoperative opioids. Mean improvement in PROMIS PI scores failed to meet an MCID of 8 in opioid users, whereas mean improvement exceeded this MCID in opioid naive patients. The results of this study help elucidate the deleterious impact of opioids, allowing surgeons to better set patient expectations. LEVEL OF EVIDENCE Level III.