Increasing Chronic Opioid Usage Despite Reduced Prescriptions After Vascular Surgery

Abstract

Perioperative narcotics confer an increased risk of chronic opioid use (COU), but optimal prescribing in vascular surgery is undefined. The study goal was to identify current practices in perioperative opioid prescription and COU in opioid-naïve patients undergoing vascular surgery. The Optum Clinformatics Data Mart is a deidentified commercial and Medicare Advantage claims database with beneficiaries across the United States. From 2004 to 2018, adult patients undergoing vascular surgery within a 3-year continuous enrollment period were queried. Common procedural terminology codes for amputation, thoracic and abdominal endovascular aortic repair, arteriovenous fistula, carotid endarterectomy, supra- and infra-inguinal bypass, open abdominal aortic repair, peripheral endovascular intervention, and venous procedures were used. Opioid-related definitions were derived from prior literature: patients were considered opioid-naïve if they did not fill a narcotic prescription within 1 year to 14 days prior to surgery. Perioperative prescriptions were those filled between 14 days prior and 7 days after surgery, and chronic opioid use was defined as filling 10 prescriptions or a 120-day supply within the postoperative year. Morphine milligram equivalents (MME) were calculated, with perioperative totals exceeding 50 MME/day as an indicator of increased overdose risk. Multivariate logistic regression was performed to identify patient factors associated with COU. Of 119,491 patients, 66,637 (53.3%) were female, with a mean age of 64.0 ± 14.2 years. Procedures were most commonly venous (44.6%), followed by peripheral endovascular interventions (14.7%) and carotid endarterectomy (14.0%). Perioperative MME/day varied significantly by procedure (P < .0001), ranging from 40.0 in supra-inguinal bypass to 23.8 MME/day in venous procedures (Fig 1A). Rates of postoperative COU varied significantly by procedure (P < .0001) and were highest following amputation (7.62%), peripheral endovascular intervention (6.95%), and infra-inguinal bypass (6.94%) (Fig 1A). Although the percentage of perioperative prescriptions greater than 50 MME/day decreased from 2004 to 2018 (27.11% to 13.98%; P < .0001), rates of COU increased (2% to 5.8%; P < .0001) (Fig 1B). In a multivariate logistic regression for factors associated with chronic opioid use, increased age (P < .001), prior substance abuse (P < .001), and a history of diabetes (P = .002) were independently associated with COU. Supra-inguinal bypass, infra-inguinal bypass, and amputations are associated with increased perioperative opioid prescriptions and postoperative COU. Despite reductions in perioperative opioid prescriptions, the incidence of COU following vascular surgery in opioid-naïve patients has increased in the last 15 years. Comorbidities including increased age, prior substance abuse, and diabetic history are independently associated with COU risk.