Search | Search by Center | Search by Source | Keywords in Title
Dasinger EA, Graham LA, Wahl TS, Richman JS, Baker SJ, Hawn MT, Hernandez-Boussard T, Rosen AK, Mull HJ, Copeland LA, Whittle JC, Burns EA, Morris MS. Preoperative opioid use and postoperative pain associated with surgical readmissions. American journal of surgery. 2019 Nov 1; 218(5):828-835.
BACKGROUND: The extent of preoperative opioid utilization and the relationship with pain-related readmissions are not well understood. METHODS: VA Surgical Quality Improvement Program data on general, vascular, and orthopedic surgeries (2007-2014) were merged with pharmacy data to evaluate preoperative opioid use and pain-related readmissions. Opioid use in the 6-month preoperative period was categorized as none, infrequent, frequent, and daily. RESULTS: In the six-month preoperative period, 65.7% had no opioid use, 16.7% had infrequent use, 6.3% frequent use, and 11.4% were daily opioid users. Adjusted odds of pain-related readmission were higher for opioid-exposed groups vs the opioid-naïve group: infrequent (OR 1.17; 95% CI:1.04-1.31), frequent (OR 1.28; 95% CI:1.08-1.52), and daily (OR 1.49; 95% CI:1.27-1.74). Among preoperative opioid users, those with a pain-related readmission had higher daily preoperative oral morphine equivalents (mean 44.5 vs. 36.1, p? < 0.001). CONCLUSIONS: Patients using opioids preoperatively experienced higher rates of pain-related readmissions, which increased with frequency and dosage of opioid exposure.