Determining the Association of Preoperative Controlled Prescription Medication Use With Outcomes After Total Knee Arthroplasty

Prescription drug abuse is a well-documented problem. In 2019 alone, there were nearly 71 000 drug overdose–related deaths recorded in the United States, and many of these overdoses involved the use of prescription opioids, stimulants, or sedatives.¹ To combat this issue, many states have established prescription drug monitoring programs and instituted mandatory checks of the resulting databases prior to prescribing drugs with abuse potential. These databases provide a patient-specific controlled substance prescription history. This includes information on medications prescribed, dose and duration of each prescription, date of prescription(s), prescriber(s), and pharmacy. The study by Emara et al² uses NarxCare, an overdose risk score (ORS) was developed to aggregate these measures into a composite score to aid prescribers and pharmacists in identifying abuse potential in patients receiving controlled substances.

As opioid and controlled substance use is prevalent in patients undergoing pain-driven surgical procedures, such as total knee arthroplasty (TKA), it is important to recognize how these medications may be associated with postoperative outcomes. It is with this in mind that the study by Emara et al² investigated the ORS as a potential risk stratification tool for patients undergoing TKA. In this retrospective review of 4326 patients, higher ORSs were associated with greater odds of increased health care resource use in a dose-dependent manner. This included increased health care use in the form of increased length of stay, nonhome discharge destination, 90-day emergency department visits, and hospital readmission. Additionally, Emara et al² found that rates of reoperation were higher among patients with an ORS of 500 or greater (out of a total possible score of 999), compared with patients with an ORS of 0 (ie, those who were prescription drug–naive). Perhaps most importantly, Emara et al² found that an ORS of 300 or greater was a useful clinical threshold for identifying patients at a higher risk of deleterious outcomes. Emara et al² concluded that surgeons could use ORS to identify patients at high risk and that these patients should be alerted to their elevated risk.

The findings of this study² add to and expand on previous literature that has identified preoperative opioid and benzodiazepine use as risk factors for increased health care resource use, complications, and revision surgery after TKA.^3,4 However, the study by Emara et al² addresses many of the shortcomings of prior work. For instance, past studies specific to the association of preoperative opioid use with postoperative outcomes have defined opioid use binarily (ie, opioid users or nonusers) or categorically by the number of opioid prescriptions received in the preoperative period (eg, 0, 1, or >2 prescriptions).⁴ The same issue exists in prior work evaluating the associations of benzodiazepine use with outcomes.³ These qualitative measures lack granularity and fail to capture the severity, temporality, and quantity of prescription drug use. They also ignore the potential influence of the combination of multiple controlled medications. The ORS, and consequently the study by Emara et al,² were able to overcome these deficiencies by combining a complex composite of information into a simple and clinically useful score.

Given the extremely high incidence of hip and knee arthroplasty performed in the United States,⁵ outcome optimization and cost containment remain critical considerations. These can be accomplished, in part, through proper patient selection and preoperative patient optimization. Preoperative patient optimization is not a novel concept in total joint arthroplasty. Orthopedic surgeons routinely recommend weight loss to patients in high-risk body mass index categories and frequently require smoking cessation prior to proceeding with operative interventions in an effort to optimize patient outcomes.⁶ While Emara et al² caution against the use of ORS as a metric to indicate surgical ineligibility, patients with an ORS of 300 or greater should be alerted of their increased risk profile. Further study will be needed to determine whether an elevated ORS represents a modifiable risk factor, and until that time, no definitive clinical recommendations can be made. However, the data presented by Emara et al² suggest that achieving preoperative controlled substance freedom may not be necessary.

While Emara et al² did not investigate whether the ORS represents a modifiable risk metric, some limited evidence has suggested that opioid cessation prior to TKA may be associated with improved outcomes.⁷ Future prospective studies should assess the modifiability of opioid, sedative, and stimulant use (and consequently the ORS) as risk factors associated with worse postoperative outcomes. Additionally, there is opportunity for further study of the ORS and how it may be associated with other outcomes of interest in patients undergoing hip or knee arthroplasty. For instance, the association between the ORS and patient-reported outcome measures, as well as arthroplasty-specific surgical complications, such as prosthetic joint infection, deep vein thrombosis, and wound complications, should be investigated. Furthermore, there is continued need for surgeon awareness and education on the risks associated with patients receiving preoperative controlled prescription medications. While there is clearly still work to be done in the realm of controlled prescription medication use and its associations with outcomes after arthroplasty, the ORS appears to offer a useful, quantifiable metric. This score could prove to be clinically useful while serving as a helpful academic metric for future research.