Castle, Joshua

Purpose To develop a predictive machine learning model to identify prognostic factors for continued opioid prescriptions after arthroscopic meniscus surgery. Methods Patients undergoing arthroscopic meniscal surgery, such as meniscus debridement, repair, or revision at a single institution from 2013 to 2017 were retrospectively followed up to 1 year postoperatively. Procedural details were recorded, including concomitant procedures, primary versus revision, and whether a partial debridement or a repair was performed. Intraoperative arthritis severity was measured using the Outerbridge Classification. The number of opioid prescriptions in each month was recorded. Primary analysis used was the multivariate Cox‐Regression model. We then created a naïve Bayesian model, a machine learning classifier that uses Bayes’ theorem with an assumption of independence between variables. Results A total of 581 patients were reviewed. Postoperative opioid refills occurred in 98 patients (16.9%). Multivariate logistic modeling was used; independent risk factors for opioid refills included male sex, larger body mass index, and chronic preoperative opioid use, while meniscus resection demonstrated decreased likelihood of refills. Concomitant procedures, revision procedures, and presence of arthritis graded by the Outerbridge classification were not significant predictors of postoperative opioid refills. The naïve Bayesian model for extended postoperative opioid use demonstrated good fit with our cohort with an area under the curve of 0.79, sensitivity of 94.5%, positive predictive value (PPV) of 83%, and a detection rate of 78.2%. The two most important features in the model were preoperative opioid use and male sex. Conclusion After arthroscopic meniscus surgery, preoperative opioid consumption and male sex were the most significant predictors for sustained opioid use beyond 1 month postoperatively. Intraoperative arthritis was not an independent risk factor for continued refills. A machine learning algorithm performed with high accuracy, although with a high false positive rate, to function as a screening tool to identify patients filling additional narcotic prescriptions after surgery. Level of Evidence III, retrospective comparative study.

Background: Recent studies have suggested increased rates of lower extremity (LE) musculoskeletal injury after a diagnosed concussion, although significant heterogeneity exists. Purpose: To examine the current body of research and determine whether there is an increased risk for LE musculoskeletal injury after a concussion and to identify populations at an increased risk. Study Design: Systematic review; Level of evidence, 3. Methods: A systematic review of current literature using MEDLINE and PubMed databases was performed. Keywords included concussion, athlete, lower extremity injury, and return to sport. Inclusion criteria required original research articles written in the English language examining the rate of LE injuries after a diagnosed concussion. Results: A total of 13 studies involving 4349 athletes (88.1% male and 11.9% female; mean age, 19.8 years) met inclusion criteria. Athletes were classified as high school (46.1%), collegiate (17.0%), or professional (36.9%). Of the 13 studies, 4 demonstrated an increased risk of LE injury within 90 days of a diagnosed concussion (odds ratio [OR], 3.44; 95% CI, 2.99-4.42), and 6 revealed an elevated risk of injury within 1 year of concussion (OR, 1.85; 95% CI, 1.73-2.84). Increased risk was seen in professional (OR, 2.49; 95% CI, 2.40-2.72) and collegiate (OR, 2.00; 95% CI, 1.96-2.16) athletes compared with high school athletes (OR, 0.97; 95% CI, 0.89-1.05). A stepwise increase in risk of sustaining an LE injury was observed with multiple concussions, with increasing risk observed from ≥2 (OR, 2.29; 95% CI, 1.85-2.83) to ≥3 (OR, 2.86; 95% CI, 2.36-3.48) career concussions. Conclusion: An increased incidence of LE injuries was observed at 90 days and 1 year after the diagnosis of a concussion. Higher levels of competition, such as at the collegiate and professional levels, resulted in an increased risk of sustaining a subsequent LE injury after a diagnosed concussion. These results suggest an at-risk population who may benefit from injury prevention methods after a concussion. Future studies should focus on identifying which injuries are most common, during what time period athletes are most vulnerable, and methods to prevent injury after return to sports.

Purpose To investigate which factors predispose patients for prolonged opioid use after medial patellofemoral ligament (MPFL) reconstruction. Methods A retrospective review of all patients who underwent MPFL reconstruction at a single institution between January 2013 and June 2020 was conducted. Opioid consumption before and after surgery was recorded and confirmed using Michigan Automated Prescriptions System monitoring program. Patients were classified into preoperative opioid users and nonusers. Risk factors for continued opioid use were assessed by collecting patient demographic variables, psychiatric history, number of previous patellar dislocations, and operative factors. Results A total of 102 patients were included during the time frame of interest. Patients were on average 21.6 ± 8.5 years old with a mean body mass index of 28.2 ± 7.9. Thirty patients (29.0%) sustained >10 dislocations preoperatively. Preoperative opioid use was present in 13 (12.7%) patients. Greater than 10 dislocations (odds ratio [OR] 5.00, 95% confidence interval [CI] 1.12‐20.92) and psychiatric history (OR 3.33, 95% CI, 1.2‐9.1; P = .016) significantly predicted opioid refills the first month after surgery. Risk factors for opioid refills at 2 to 12 months postoperatively included smoking (OR 4.50, 95% CI 1.13‐17.96), preoperative opioid use (OR 7.32, 95% CI 1.88‐28.47), psychiatric disorder (OR 3.77, 95% CI 2.3‐6.2; P < .001), age >30 years (OR 7.03, 95% CI 3.63‐13.61; P < .001), and obesity (OR 2.68, 95% CI 1.40‐5.14; P = .002). Compared with Outerbridge 0, a greater percentage of patients with Outerbridge 1 or 2 and 3 or 4 continued using opioids 2 to 12 months after surgery (OR 3.06, 95% CI 1.33‐7.02; P = .006 and OR 2.86, 95% CI 1.24‐6.59; P = .010, respectively). Conclusions For patients undergoing MPFL reconstruction, preoperative opioid use, cartilage damage, age >30 years, smoking history, body mass index >30, and history of psychiatric disorder were found to be significantly associated with prolonged opioid use after surgery. Postoperative opioid refills in this cohort declined after 1 month. Level of Evidence Level III, retrospective cohort study.

Purpose To assess the effects of surgical treatment of athletic pubalgia (AP) on game use and performance metrics in National Basketball Association (NBA) players. Methods A retrospective review of all NBA players who underwent surgical management for AP from 1996 to 2018 was performed. A matched control group was created for comparison. The index period was defined as the entire NBA season in which surgery occurred, including the corresponding offseason. Player demographics, use (games played, games started, and minutes per game) and performance (player efficiency rating) metrics were collected for all players. Statistical analysis was performed to compare data before and after return to play. Results Thirty players with a history of surgical management for AP were included in the final analysis. Following surgery for AP, NBA players were found to have a return to play (RTP) rate of 90.91% (30/33). The average RTP following surgery was 4.73 ± 2.62 months. Compared with control athletes, athletes in the AP group played significantly fewer seasons postinjury (4.17 ± 2.70 vs 5.49 ± 3.04 seasons, respectively; P = .02). During the first year following RTP, NBA players experienced significant reductions in game use and performance, both when compared with the year prior and matched control athletes ( P < .05). At 3‐year follow‐up, players continued to demonstrate significant reductions in game use (minutes per game, P < .05) but not performance. Conclusions Following surgical treatment of AP, NBA players demonstrated a high RTP rate, but shortened career. A short‐term reduction in game use and performance metrics was found the year of return following surgery. However, 3‐year follow‐up performance metrics normalized when compared with healthy controls. Study Design Level III; retrospective case‐control study.

Purpose To investigate the demographic factors that influence time to respond (TTR), time to completion (TTC), and response rate when using a text messaging‐based system and to determine the feasibility and applicability of mobile messaging‐based services for collection of patient‐reported outcomes among orthopaedic sports medicine patients. Methods On the day of surgery, patient mobile phone number was collected and the automated mobile messaging service (MOSIO, Seattle, WA) messaged patients for 10 ``days postoperatively. Patient visual analog scale (VAS) scores were collected 3 times daily, side effects were asked each evening, and Patient‐Reported Outcomes Measurement Information System (PROMIS) Pain Interference (PI) Short Form was collected on postoperative day 3 and 7. Results A total of 177 patients were enrolled in the study. The overall response rate to the survey questions was 75.0%. For all patients, the average TTR of questions was 35.09 ± 12.57 minutes. The TTC was 2.75 ± 3.56 minutes for PROMIS‐PI, 3.51 ± 1.26 minutes for VAS, and 3.80 ± 6.87 for side‐effect questions. When patients were stratified into age groups, the youngest group, 16 to 32 years, had the greatest response rate of 85.2% and patients in the 49 to 59 years group had the lowest response rate of 68.4% and 69.1%, respectively ( P < .001). There was no significant difference in the TTR or TTC for VAS, PROMIS‐PI, or side‐effect questions when patients were stratified by age or sex groups ( P > .05). Conclusions Collectively, all age groups successfully achieved a mean response rate of 75%; however, significantly lower response rates were observed for patients >49 years old. Differences in age and sex did not impact the overall TTR or TTC for VAS, PROMIS‐PI, or side‐effect questions. Mobile‐based applications present as an emerging opportunity to track postoperative outcome scores and reduce clinic survey load. Level of Evidence Case series, level of evidence IV.