Machine Learning Algorithms Predict Prolonged Opioid Use in Opioid-Naïve Primary Hip Arthroscopy Patients

Declining Postoperative 90-Day Opioid Prescriptions From 2010 to 2021 Following Hip Arthroscopy for Femoroacetabular Impingement Syndrome

Purpose

To analyze postoperative opioid prescriptions after hip arthroscopy for femoroacetabular impingement syndrome (FAIS) in a large, opioid-naive population and to evaluate factors associated with receiving more opioids.

Methods

Opioid-naive adult patients who underwent hip arthroscopy for FAIS were queried in the 2010 to 2022 PearlDiver Mariner 161 national administrative database. Exclusion criteria included patients with a history of chronic pain and patients who received opioid prescriptions more than 30 days before surgery. Patient variables were extracted: age, sex, and Elixhauser Comorbidity Index. Ninety-day postoperative opioid prescriptions (by total morphine milligram equivalents [MMEs]) were assessed with multivariate linear regression. Ninety-day postoperative opioid prescriptions from 2011 to 2021 were assessed.

Results

Of 27,079 patients with postoperative opioid prescriptions identified, a mean ± standard deviation of 347.6 ± 729.2 MMEs (40 tablets of 5 mg oxycodone) were prescribed per patient, with a mean of 1.6 prescriptions filled per patient within 90 days following surgery. Seventy-five percent of patients filled fewer than 600 MMEs, but a small subset filled more than 2,000 MMEs. Multivariate analysis revealed that, compared to patients in the age 30- to 39-year group, those aged 20 to 29 years received fewer MMEs (Δ = -72.5, P < .017). Compared to those with an Elixhauser Comorbidity Index of 2 or under, those >2 were prescribed more MMEs (Δ = 52.5, P < .017). Sex did not correlate with the postoperative MMEs prescribed. From 2011 to 2021, a 58.2% decrease in the 90-day mean MMEs prescribed was noted per patient (P < .017).

Conclusions

Fewer postoperative MMEs were filled following FAIS hip arthroscopy for patients in their 20s relative to those in their 30s, as well as for those with lower comorbidity burden. Patient sex was not associated with differences in postoperative MMEs prescribed. The amount of mean MMEs prescribed per patient decreased from 2011 to 2021.

Clinical Relevance

This study provides information about the typical amount of narcotics required after surgery. This is increasingly useful information, as surgeons/clinicians continue to try to minimize the role of narcotics in postoperative recovery.

A Review of Leveraging Artificial Intelligence to Predict Persistent Postoperative Opioid Use and Opioid Use Disorder and its Ethical Considerations

PURPOSE OF REVIEW

Artificial intelligence (AI) offers a new frontier for aiding in the management of both acute and chronic pain, which may potentially transform opioid prescribing practices and addiction prevention strategies. In this review paper, not only do we discuss some of the current literature around predicting various opioid-related outcomes, but we also briefly point out the next steps to improve trustworthiness of these AI models prior to real-time use in clinical workflow.

RECENT FINDINGS

Machine learning-based predictive models for identifying risk for persistent postoperative opioid use have been reported for spine surgery, knee arthroplasty, hip arthroplasty, arthroscopic joint surgery, outpatient surgery, and mixed surgical populations. Several machine learning-based models have been described to predict an individual's propensity for opioid use disorder and opioid overdose. Natural language processing and large language model approaches have been described to detect opioid use disorder and persistent postsurgical opioid use from clinical notes. AI holds significant promise in enhancing the management of acute and chronic opioids, which may offer tools to help optimize dosing, predict addiction risks, and personalize pain management strategies. By harnessing the power of AI, healthcare providers can potentially improve patient outcomes, reduce the burden of opioid addiction, and contribute to solving the opioid crisis.

Machine Learning Algorithms Can Be Reliably Leveraged to Identify Patients at High Risk of Prolonged Postoperative Opioid Use Following Orthopedic Surgery: A Systematic Review

Background: Machine learning (ML) has emerged as a method to determine patient-specific risk for prolonged postoperative opioid use after orthopedic procedures. Purpose: We sought to analyze the efficacy and validity of ML algorithms in identifying patients who are at high risk for prolonged opioid use following orthopedic procedures. Methods: PubMed, EMBASE, and Web of Science Core Collection databases were queried for articles published prior to August 2021 for articles applying ML to predict prolonged postoperative opioid use following orthopedic surgeries. Features pertaining to patient demographics, surgical procedures, and ML algorithm performance were analyzed. Results: Ten studies met inclusion criteria: 4 spine, 3 knee, and 3 hip. Studies reported postoperative opioid use over 30 to 365 days and varied in defining prolonged use. Prolonged postsurgical opioid use frequency ranged from 4.3% to 40.9%. C-statistics for spine studies ranged from 0.70 to 0.81; for knee studies, 0.75 to 0.77; and for hip studies, 0.71 to 0.77. Brier scores for spine studies ranged from 0.039 to 0.076; for knee, 0.01 to 0.124; and for hip, 0.052 to 0.21. Seven articles reported calibration intercept (range: -0.02 to 0.16) and calibration slope (range: 0.88 to 1.08). Nine articles included a decision curve analysis. No investigations performed external validation. Thematic predictors of prolonged postoperative opioid use were preoperative opioid, benzodiazepine, or antidepressant use and extremes of age depending on procedure population. Conclusions: This systematic review found that ML algorithms created to predict risk for prolonged postoperative opioid use in orthopedic surgery patients demonstrate good discriminatory performance. The frequency and predictive features of prolonged postoperative opioid use identified were consistent with existing literature, although algorithms remain limited by a lack of external validation and imperfect adherence to predictive modeling guidelines.

Personalized Medicine Using Predictive Analytics: A Machine Learning-Based Prognostic Model for Patients Undergoing Hip Arthroscopy

BACKGROUND

Personalized medicine models to predict outcomes of orthopaedic surgery are scarce. Many have required data that are only available postoperatively, mitigating their usefulness in preoperative decision making.

PURPOSE

To establish a method for predictive modeling to enable individualized prognostication and shared decision making based on preoperative patient factors using data from a prospective hip preservation registry.

STUDY DESIGN

Cohort study (Prognosis); Level of evidence, 2.

METHODS

Preoperative data of 2415 patients undergoing hip arthroscopy for femoroacetabular impingement syndrome between February 2008 and November 2017 were retrospectively analyzed. Two machine-learning analyses were evaluated: Tree-structured survival analysis (TSSA) and Cox proportional hazards modeling for predicting time to event and for computing hazard ratios for survivorship after hip arthroscopy. The Fine-Gray model was similarly used for repeat hip arthroscopy. Variables were selected for inclusion using the minimum Akaike Information Criterion (AIC). The stepwise selection was used for the Cox and Fine-Gray models. A web-based calculator was created based on the final models.

RESULTS

Prognostic models were successfully created using Cox proportional hazards modeling and Fine-Gray models for survivorship and repeat hip arthroscopy, respectively. The Harrell C-statistics of the Cox model calculators for survivorship after hip arthroscopy and the Fine-Gray model for repeat hip arthroscopy were 0.848 and 0.662, respectively. Using the AIC, 13 preoperative variables were identified as predictors of survivorship, and 6 variables were identified as predictors for repeat hip arthroscopy. In contrast, the TSSA model performed poorly, resulting in a Harrell C-statistic <0.6, rendering it inaccurate and not interpretable. A web-based calculator was created based on these models.

CONCLUSION

This study successfully created an institution-specific machine learning-based prognostic model for predictive analytics in patients undergoing hip arthroscopy. Prognostic models to predict survivorship and the need for repeat surgeries were both adapted into web-based tools to assist the physician with shared decision making. This prognostic model may be useful at other institutions after performing external validation. Additionally, this study may serve as proof of concept for a methodology for the development of patient-specific prognostic models. This methodology may be used to create other predictive analytics models in different realms of orthopaedic surgery, contributing to the evolution from evidence-based medicine to personalized medicine.