Risk Factors for Increased Consumption of Narcotics After Hip Arthroscopy

A Prospective, Randomized Control Trial

Ryan S. Selley, MD; Matthew J. Hartwell, MD; Bejan A. Alvandi, MD; Michael A. Terry, MD; Vehniah K. Tjong, MD


J Am Acad Orthop Surg. 2021;29(12):527-536. 

In This Article


Arthroscopic procedures in the hip are increasing in frequency; establishing the appropriate postoperative narcotic prescription quantities that provide adequate analgesia while limiting excess narcotic distribution is essential.[8] The aim of this study was to investigate whether randomizing the opioid prescription size in patients undergoing hip arthroscopy would affect total consumption while also identifying risk factors for increased opioid use. In this cohort, randomizing patients to either 30 or 60 tablets of hydrocodone/acetaminophen 10/325 did not have a significant effect on total tablets consumed or postoperative analgesia. Furthermore, 80% of patients required 14 tablets or less, and pain requirements of 94% of patients were satisfied with 30 tablets. However, several risk factors for increased opioid use were identified: opioid or muscle relaxant use within 1 year of surgery and lower preoperative iHOT-12 scores.

Physician overprescription can be directly linked to opioid abuse. Daubresse et al highlighted the dramatic increase in opioid prescriptions in recent years, and the work of Kelly et al and others has shown that even exposure to prescription narcotics under appropriate circumstances can increase the risk of future abuse.[19–21] In addition, patients with excess narcotics often divert their painkillers to friends and family.[11] Orthopedic surgeons in particular represent one of the major providers of narcotics, prescribing the third highest amounts behind pain specialists and primary care physicians.[7,22] Efforts have been made to decrease overprescription in orthopaedic surgery by accurately quantifying the amount of opioids used by patients after common procedures and establishing education and prescribing guidelines.[7,23–26] Stepan et al[25] noted a significant reduction in opioid prescription sizes after development of institutional guidelines for arthroscopic procedures of the hip, shoulder, and knee. In contrast to these findings, this study did not find a correlation between prescription size and consumption. Predictably, smaller prescription sizes did result in less unused tablets available in the community.

Identifying patients at risk of prolonged or increased use can help predict the quantity of opioids to prescribe and potential modifiable risk factors. Preoperative opioid use in hip arthroscopy has been linked to worse postoperative pain and prolonged postoperative narcotic use.[26,27] As patients enrolled in this study by design were not using opioids at the time of surgery, this study was not able to assess this factor. However, it was noted in this cohort that any narcotic use within a year preoperatively independently increased the average number of tablets used postoperatively by 12.6. In a retrospective insurance database study, other risk factors identified in the literature for prolonged narcotic use after hip arthroscopy include utilization of preoperative anxiolytic and muscle relaxant medications, substance use or abuse, morbid obesity, and back pain.[27] The present study found that preoperative muscle relaxant use independently increased postoperative opioid use by 22.5 tablets on average, and we did not find a correlation with any of the above risk factors and increased use.[27] Several other risk factors for increased pain after hip arthroscopy have been identified. Psychological distress has been noted to affect postoperative pain; Aoki et al found that patients who scored higher in this trait required 40% more intraoperative opioid and were twice as likely to request postoperative nerve blockade.[28] Tan et al[29] noted that use of infusion pressure higher than 80 mm Hg, femoral osteochondroplasty, and labral repair during hip arthroscopy significantly increased postoperative pain scores. Last, intra-abdominal fluid extravasation as measured by ultrasonography trended toward increased opioid use in the postanesthesia care unit in a study by Haskins et al.[30] To our knowledge, no studies have analyzed risk factors for increased opioid use as it pertains to preoperative outcome testing; in this cohort, lower scores on the iHOT-12 significantly predicted postoperative opioid use as each point decrease correlated with 0.25 more tablets being consumed.

Although many risk factors may not be modifiable, multimodal analgesia has demonstrated promising effects on postoperative pain control. A study comparing intra-articular pain cocktail injection and no injection noted significant reduction in postanesthesia care unit opioid requirements.[17] Likewise, preprocedure celecoxib use has demonstrated decreased postanesthesia care unit pain scores and earlier discharge time.[16] Our institutional multimodal pain management protocol for patients undergoing hip arthroscopy includes preprocedure celecoxib and acetaminophen, intraprocedure intra-articular pain cocktail injection at the end of the case, and the above-outlined postoperative prescriptions. We have not pursued the use of perioperative nerve blockade because of the reported risk of falls associated with its use.[31,32]

Information gleaned in this study and other similar studies at our institution has helped to decrease the total amount of MME given to patients after procedures without compromising pain control. Furthermore, since the inception of this study, institutional guidelines have been created to further aid in reducing the number of opioids prescribed as was previously customary. Future work is being performed by this group using preoperative information to help predict postoperative opioid requirements and individually tailor prescriptions based on preoperative demographics and questionnaires such as the iHOT-12.

The primary limitation of this study is reliance on patient-reported narcotic consumption. It is conceivable that patients miscounted or were dishonest in their account of tablets consumed or changed their behavior knowing that opioid use was being monitored. In addition, at the inception of this study there were no reported values for opioid use in hip arthroscopy with which we could perform a power analysis. As such, a post hoc analysis was performed that indicated we would need a total of 298 patient to detect a difference between our groups; thus, this study was underpowered. Furthermore, the patient demographics of this study likely are not representative of all patients undergoing hip arthroscopy as we are located in an urban area and have a high proportion (85%) of college educated patients. This factor could impact opioid use though we did not detect a difference in this cohort. The use of subjective surveys introduced intrinsic patient bias. Last, no patients in this study were actively using opioid pain medications preoperatively, which is not representative of all patients undergoing hip arthroscopy and an identified risk factor for higher postoperative narcotic use in hip arthroscopy.[26]