Presentation Authors: Dima Raskolnikov*, Steven Ngo, Sarah Holt, Chinonyerem Okoro, John Gore, Seattle, WA
Introduction: The over-prescribing of opioids following urological surgery places patients at risk for opioid overuse and dependency. However, no guidelines exist to help urologists consistently prescribe appropriate quantities of pain medication. We sought to characterize the variation in opioid prescribing habits at time of discharge following nephrectomy.
Methods: We performed a retrospective review of patients who underwent partial or radical nephrectomy between November 2016 and May 2018 at an academic medical center. We reviewed patient demographic characteristics including age, sex, travel distance to hospital, marital status, race, employment, and insurance status, Charlson Comorbidity Index (CCI), and pre-operative opioid use. Surgeries were categorized as open or laparoscopic/robotic. Inpatient daily opioid use was recorded in oral morphine equivalents (OME). Discharge opioid prescriptions were recorded in OMEs and categorized by both post-graduate year (PGY) of prescriber and discharge day.
Results: We identified 173 eligible patients, 23 (13%) of whom used opioids pre-operatively. A laparoscopic/robotic approach was used in 111 (64%) cases. Patients used a daily median 22.5 OME during their hospitalization, weaning to median 15 OME on day of discharge. All but 2 patients were prescribed opioids at discharge, with median 315 OME per prescription. Amount of opioid used on the last hospital day explained 34% of the variance in the amount prescribed at discharge (p < 0.01, Figure). On multivariate logistic regression, opioid use pre-operatively, hospital days, PGY-level of prescriber, year of surgery, open vs. minimally invasive approach, and insurance type were associated with discharge OMEs (R2=0.49, p < 0.05). Within this model, age, sex, CCI, and day of discharge did not explain variation in OME at discharge. Correlations between median inpatient OME and discharge OME were similar.
Conclusions: Inpatient opioid use poorly predicts discharge opioid prescriptions, accounting for less than 50% of the variance between prescriptions. Patients were discharged with a median OME 20 times the amount used on their last hospital day. Systems are needed to help minimize variability in opioid prescribing practices and reduce the overall quantity prescribed.
Source of Funding: This work was supported in part by the Urology Care Foundation Residency Research Award and Russell Scott, Jr., MD Urology Research Fund.