Excess Opioid Medication and Variation in Prescribing Patterns Following Common Orthopaedic Procedures

Abstract

Background: Postoperative pain management in orthopaedic surgery accounts for a substantial portion of opioid medications prescribed in the United States. Understanding prescribing habits and patient utilization of these medications following a surgical procedure is critical to establishing appropriate prescribing protocols that effectively control pain while minimizing unused opioid distribution. We evaluated prescribing habits and patient utilization following elective orthopaedic surgical procedures to identify ways of improving postoperative opioid-prescribing practices.

Methods: We performed a review of prescribing data of 1,199 procedures and gathered telephone survey results from 557 patients to determine the number of opioid pills prescribed postoperatively and the number of unused pills. The data were collected from adult patients who underwent 1 of the 5 most common elective orthopaedic procedures at our institution in fiscal year 2015: total hip arthroplasty, total knee arthroplasty, endoscopic carpal tunnel release, arthroscopic rotator cuff repair, or lumbar decompression. We converted all dosages to opioid equivalents of oxycodone 5 mg and performed analyses of prescribing patterns, patient utilization, and patient disposal of unused opioids.

Results: Prescribing patterns following the 5 orthopaedic procedures showed wide variation. The median numbers of oxycodone 5-mg equivalent opioid pills prescribed upon discharge were 90 pills (range, 20 to 330 pills) for total hip arthroplasty, 90 pills (range, 10 to 200 pills) for total knee arthroplasty, 20 pills (range, 0 to 168 pills) for endoscopic carpal tunnel release, 80 pills (range, 18 to 100 pills) for arthroscopic rotator cuff repair, and 80 pills (range, 10 to 270 pills) for lumbar decompression. Thirty-seven percent of patients overall requested and received at least 1 refill. The mean number of total pills prescribed (and standard deviation) including refills was 113.6 ± 75.7 for total hip arthroplasty, 176.4 ± 108.0 for total knee arthroplasty, 24.3 ± 29.0 for carpal tunnel release, 98.2 ± 59.6 for rotator cuff repair, and 107.4 ± 64.4 for lumbar decompression. Participants reported unused opioid medication in 61% of cases. During the study year, >43,000 unused opioid pills were prescribed. Forty-one percent of patients reported appropriate disposal of unused opioid pills.

Conclusions: Prescribing patterns vary widely, and a large amount of opioid medications remains unused following elective orthopaedic surgical procedures. Effective prescribing protocols are needed to limit this source of potential abuse and opioid diversion within the community.

Fig. 1

Study flow diagram demonstrating inclusion in prescribing data review, inclusion in telephone survey, attrition from lack of telephone contact, and exclusion at each stage of the study. THA = total hip arthroplasty, TKA = total knee arthroplasty, CTR = endoscopic carpal tunnel release, RCR = arthroscopic rotator cuff repair, LD = lumbar decompression (with or without arthrodesis), EMR = electronic medical record, and FY = fiscal year.

Fig. 2

Box plots of median (middle line in box) and interquartile range (box) of total number of opioid pills prescribed (blue) and total pills taken (red) following orthopaedic procedures in FY2015. The whiskers indicate the range. All pills are shown as an oxycodone 5-mg equivalent. THA = total hip arthroplasty, TKA = total knee arthroplasty, RCR = arthroscopic rotator cuff repair, CTR = endoscopic carpal tunnel release, and LD = lumbar decompression (with or without arthrodesis).

Fig. 3

Bar graph of rates for postoperative opioid refill prescriptions (red) and patient satisfaction with postoperative pain control (blue) as denoted by a response of “satisfied” or “very satisfied” on telephone survey. TKA = total hip arthroplasty, THA = total knee arthroplasty, RCR = arthroscopic rotator cuff repair, CTR = endoscopic carpal tunnel release, and LD = lumbar decompression (with or without arthrodesis).

Fig. 4

Stacked bar graph showing how patients disposed of unused opioid pills (by procedure). Less than half of respondents reported appropriate disposal of unused pills in every surgical procedure. Blue denotes appropriate disposal (per the FDA), gray denotes patients who still have remaining pills, orange denotes patients who reported not knowing where the unused pills were, and red denotes “other.” THA = total hip arthroplasty, TKA = total knee arthroplasty, RCR = arthroscopic rotator cuff repair, CTR = endoscopic carpal tunnel release, and LD = lumbar decompression (with or without arthrodesis).

Fig. 5

Histograms by procedure showing the distribution of opioid pills taken. The orange, dashed vertical line indicates the median value of the initial postoperative prescription. Note that, for each procedure except total knee arthroplasty (TKA), approximately half of the patients took fewer than the median initial number prescribed. THA = total hip arthroplasty, RCR = arthroscopic rotator cuff repair, CTR = endoscopic carpal tunnel release, and LD = lumbar decompression (with or without arthrodesis).