Opioid Utilization Following Lumbar Arthrodesis

Trends and Factors Associated With Long-Term Use

Piyush Kalakoti, MD; Nathan R. Hendrickson, MD; Nicholas A. Bedard, MD; Andrew J. Pugely, MD

Disclosures

Spine. 2018;43(17):1208-1216. 

In This Article

Results

Overall, 26,553 eligible patients (OU: 58.3%) undergoing lumbar arthrodesis (ALIF: 8.5%; P/TLIF: 43.8%; PLF: 41.5%; ALIF+PLF: 6.2%) between 2007 and 2015 were registered with the database. Most patients were elderly ≥ 50 years of age (n = 24,558; 92.5%) and were female (n = 15,624; 58.9%). Differences in characteristics between ON and OU groups are presented in Table 1.

Postoperative trends in the narcotic prescription filling rates were calculated up to 1 year following lumbar arthrodesis in both OU and ON groups and compared with average preoperative filling rate of 28.5% (Figure 1). At 1 month, 60.2% ONs and 82.9% OUs had a filled opioid prescription. At 3 months, rates of prescription opioids declined significantly to 13.9% in ONs versus 53.8% in OUs, while plateauing at 6 to 12-month postoperative period (ON: 8.4–9.6%; OU: 42.1–45.3%). At 1 year following lumbar arthrodesis, a significantly higher proportion of OUs (42.4%) remain on narcotics than ON patients (8.6%), translating to an absolute relative risk (RR) of using narcotics for OU at 4.92 (95% CI: 4.35–5.55; Table 2). Subgroup analysis trending postoperative narcotic utilization rates for lumbar surgical approaches is depicted in Table 3. At 1 year, the absolute risk of using narcotics was highest in patients undergoing ALIF + PLF (RR: 5.5, 95% CI: 4.19–7.18) followed by PLF (RR: 5.1, 95% CI: 4.58–5.63), P/TLIF (RR: 4.33–5.21, 95% CI: 4.19–7.18), and ALIF (RR: 4.5, 95% CI: 3.6–5.4; Table 3).

Figure 1.

Percentage of OU and ON filling opioid prescriptions following lumbar spinal arthrodesis plotted by monthly intervals. Average preoperative monthly prescription refill rate was calculated based upon average number of OU refilling prescriptions each month over 3 months before the index surgery.

In an unadjusted analysis, patients with psychiatric comorbidities (depression/anxiety), fibromyalgia, lower back pain, morbid obesity, alcohol or drug abuse, and preoperative narcotic use were associated with prolonged narcotic use following ALIF, P/TLIF, and PLF (Table 4). In multivariable models controlled for confounders (exposure variables), we noted preoperative opioid use as a significant driver of 1-year narcotic use following ALIF (OR: 7.86; 95% CI: 6.57–9.46; P < 0.001) along with younger age (≤50 years) (OR: 1.24; P = 0.035), male gender (OR:1.17; P = 0.048), depression/anxiety (OR: 1.29; P = 0.008), fibromyalgia (OR: 1.26; P < 0.001), and history of drug abuse (OR: 2.15; P < 0.001) (Figure 2). Likewise, we noted preoperative opioid use as a major driver of 1-year narcotic use following P/TLIF (OR: 4.62; 95% CI: 3.84–5.62; P < 0.001) and PLF (OR: 7.18; 95% CI: 6.50–7.95; P < 0.001; Figures 3 and 4). Common factors increasing likelihood of narcotic use following P/TLIF or PLF at 1 year are preoperative psychiatric comorbidity (depression/anxiety), fibromyalgia, morbid obesity, and drug dependence. Interestingly, younger patients had a lower likelihood of narcotic use following P/TLIF (OR: 0.79; P < 0.050), whereas patients with chronic lower back pain (OR: 1.41; P < 0.001) and bladder/bowel symptomatology (OR: 1.31; P < 0.001) are likely at higher odds (Figure 3). On the contrary, younger age (≤50 years; OR: 1.81; P < 0.001) and alcohol abusers (OR:1.57; P < 0.001) were associated with higher likelihood of narcotic use following PLF (Figure 4). All models depicted good calibration (P > 0.10) (Online Supplement: Figure S2, http://links.lww.com/BRS/B362), and model applicability can be explored in our proposed app (http://neuro-risk.com/opiod-use/; Online Supplement: Interactive app, http://links.lww.com/BRS/B362).

Figure 2.

Multivariable analysis demonstrating factors associated with postoperative opioid use at 12 months following ALIF. The odds ratios alongside 95% confidence interval for significant factors are depicted with forest plots on the right.

Figure 3.

Multivariable analysis demonstrating factors associated with postoperative opioid use at 12 months following P/TLIF. The odds ratios alongside 95% confidence interval for significant factors are depicted with forest plots on the right.

Figure 4.

Multivariable analysis demonstrating factors associated with postoperative opioid use at 12 months following PLF. The odds ratios alongside 95% confidence interval for significant factors are depicted with forest plots on the right.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....