Forsythe, Brian

Abstract As artificial intelligence (AI) Large Language Models (LLM) like ChatGPT become more common in healthcare, patients increasingly use them to find medical information. While ChatGPT may support health literacy, its readability and accuracy compared to established resources remain unclear. The purpose of this study is to evaluate ChatGPT-4 responses on femoroacetabular impingement (FAI) and its surgical management compared to AAOS OrthoInfo content. We hypothesize that ChatGPT may deliver medically accurate information, but its readability may fall short when compared to OrthoInfo. Nine questions based on the OrthoInfo FAI page were submitted to ChatGPT-4, with and without a readability prompt. Topics included anatomy, pathology, cause, symptoms, workup, imaging, treatment, role of hip arthroscopy, and outcomes. Readability was assessed using validated indices. Accuracy was independently rated using a 4-point scale. Statistical comparisons were made using t-tests and ANOVA (P < 0.01 threshold). OrthoInfo content had a mean reading grade level of 8.0 and a Flesch Reading Ease score of 60.7. Unprompted ChatGPT responses were significantly less readable (grade level 16.2; Flesch score 21.7; P < 0.001). Prompting for readability improved ChatGPT outputs (grade level 10.6; Flesch score 58.2), making them comparable to OrthoInfo (P < 0.09). Accuracy was high across all sources, though OrthoInfo's response on FAI causes scored slightly lower (3.5 versus 4.0; P < 0.46). ChatGPT provides accurate information on FAI. However, without prompting, its complexity may hinder patient understanding. Prompting for readability enables ChatGPT to match established resources like OrthoInfo. Tailored prompting is key to using LLM effectively in patient education and promoting health literacy.

Background: Outcomes after arthroscopic rotator cuff repair (RCR) are frequently measured through clinically significant outcomes (CSOs) such as the minimal clinically important difference, the substantial clinical benefit, and the Patient Acceptable Symptom State. Global achievement of CSOs is challenging to predict. Purpose: To determine if unsupervised machine learning can identify distinct patient subgroups based on CSO achievement after elective arthroscopic RCR. Study Design: Case-control study; Level of evidence, 3. Methods: A prospectively collected database was analyzed to identify patients who underwent elective arthroscopic RCR from 2015 to 2017. Tear dimensions were measured on magnetic resonance imaging utilizing a validated technique. CSO achievements on the American Shoulder and Elbow Surgeons, the Single Assessment Numeric Evaluation, and the Constant-Murley subjective score at 2-year follow-up were calculated. An unsupervised random forest algorithm was utilized to develop and internally validate patient subgroups with significantly different rates of CSO achievement. Patient subgroup membership, along with a total of 30 demographic and clinical variables, as well as preoperative patient-reported outcomes, were incorporated into a stepwise multivariable logistic regression to identify factors predictive of optimal CSO achievement. Results: A total of 346 patients (192 male; mean ± SD age, 57.2 ± 9.1 years; body mass index, 30.1 ± 5.4 kg/m 2 ) were eligible for inclusion and followed for a mean of 3.8 years (range, 2.0-6.2 years) Of these, a total of 333 patients were partitioned by the random forest algorithm into 2 subgroups (stability, 0.16; connectivity: 180.8; Dunn: 0.16; silhouette: 0.05), with 176 patients in the optimal achievement subgroup and 157 patients in the suboptimal achievement subgroup. The 2 subgroups differed significantly (all P ≤ .004) in the likelihood of achievement of all CSOs. Stepwise multivariable logistic regression identified an increase of 1 mm in tear size in the sagittal dimension beyond 1.9 cm to predict a 10% increase in the probability of suboptimal achievement. Additional risk factors for suboptimal CSO achievement included increasing number of tendons involved (odds ratio [OR], 14.07; 95% CI, 4.50-44.02; P < .001), subscapularis involvement (OR, 8.67; 95% CI, 2.45-30.71; P = .01), and increased preoperative CMS score (OR, 1.11; 95% CI, 1.04-1.18; P = .001). Protective factors included performance of a subpectoral biceps tenodesis compared with biceps tenotomy (OR, 0.22; 95% CI, 0.05-0.92; P = .03). Conclusion: Clinically meaningful subgroups were uncovered using an unsupervised machine learning algorithm in patients undergoing arthroscopic RCR. Tear size, number of tendons involved, and subscapularis involvement were significant and additive predictors of suboptimal CSO achievement at 2-year minimum follow-up. Treatment of concurrent biceps pathology with tenodesis conferred 78% increased likelihood of CSO achievement compared with tenotomy.

Purpose To perform a prospective, double‐blind, randomized controlled trial to assess differences in integration and patient‐reported outcomes metrics (PROMs) of osteochondral allograft transplantation with and without bone marrow aspirate concentrate (BMAC) augmentation. Methods Patients (n = 36) undergoing osteochondral allograft transplantation of the knee were consented and enrolled in this prospective study. They were randomized to either iliac crest BMAC or sham incision groups and blinded to their allocation. Computerized tomography (CT) scans of the knee were obtained at 6 months postoperatively after the index transplantation and graded by the semiquantitative assessment CT osteochondral allograft system. PROMs, including the International Knee Documentation Committee and Knee injury and Osteoarthritis Outcome Score–Joint Replacement, were obtained at 6 months, 1 year, and 2 years postoperatively. Results On 6‐month postoperative CT scans, patients receiving BMAC‐treated grafts were more likely to have small cystic changes ( P = .01), with an associated trend toward reduction in large cyst formation ( P = .06), but equal osseous integration, graft signal density, and presence of discernible clefts and intra‐articular fragments. The BMAC group was less likely to undergo subsequent surgery for graft debridement or revision (5.3% vs 35.3%; P = .02). There were no significant differences in PROMs between the 2 groups preoperatively or postoperatively at 6 months, 1 year, or 2 years. Patients receiving BMAC trended toward a higher rate of achievement of Knee Documentation Committee and Knee injury and Osteoarthritis Outcome Score–Joint Replacement minimal clinically important difference (88% vs 55%; P = .076). Conclusions Patients receiving BMAC‐treated grafts were more likely to have small cystic changes and were less likely to undergo subsequent surgery for persistent or new symptoms after the index procedure. No difference in postoperative PROMs was shown at the 6‐month, 1‐year, and 2‐year follow‐ups between the 2 groups. Level of Evidence Level I, prospective, randomized controlled trial.

Purpose To assess the relation between tendon migration, as measured by radiostereometric analysis, and patient‐reported outcome measures (PROMs) after biceps tenodesis (BT); to determine the likelihood of achieving clinically significant outcomes (CSOs) after BT; and to identify factors that impact CSO achievement. Methods Patients undergoing arthroscopic suprapectoral or open subpectoral BT at a single, high‐volume academic medical center were prospectively enrolled. A tantalum bead sutured to the tenodesis construct was used as a radiopaque marker. Biceps tendon migration was measured on calibrated radiographs at 12 weeks postoperatively. PROMs (Constant‐Murley, Single Assessment Numeric Evaluation [SANE], and Patient‐Reported Outcomes Measurement Information System–Upper Extremity [PROMIS‐UE] scores) were collected preoperatively and at minimum 2‐year follow‐up. Results Of 115 patients enrolled, 94 (82%) were included (median age, 52 years; median body mass index, 31.4). At a mean follow‐up of 2.9 years, the median Constant‐Murley, SANE, and PROMIS‐UE scores were 33 (interquartile range [IQR], 26‐35), 90 (IQR, 80‐99), and 47 (IQR, 42‐58), respectively. Median tantalum bead migration was 6.5 mm (IQR, 1.8‐13.8 mm). There were significant correlations between migration and Constant‐Murley score ( r 2 = 0.222; β = –0.554 [95% confidence interval (CI), –1.027 to –0.081]; P = .022), SANE score ( r 2 = 0.238; β = –0.198 [95% CI, –0.337 to –0.058]; P = .006), and PROMIS‐UE score ( r 2 = 0.233; β = –0.406 [95% CI, –0.707 to –0.104]; P = .009). On univariable analysis, higher body mass index was associated with achievement of substantial clinical benefit (unadjusted odds ratio [OR], 1.078 [95% CI, 1.007 to 1.161]; P = .038). Greater bead migration was negatively associated with achievement of the minimal clinically important difference (unadjusted OR, 0.969 [95% CI, 0.943 to 0.993]; P = .014) and patient acceptable symptomatic state (unadjusted OR, 0.965 [95% CI, 0.937 to 0.989]; P = .008) on all 3 instruments. Conclusions A 1‐cm increase in post‐tenodesis biceps tendon migration was associated with a decrease in the Constant‐Murley, SANE, and PROMIS‐UE scores of 6, 2, and 4 points, respectively, at a mean of 2.9 years after surgery. Most patients achieved CSOs for these PROMs by latest follow‐up, and greater biceps tendon construct migration was negatively associated with the likelihood of CSO achievement. Level of Evidence Level IV, retrospective case series.

Background: Studies suggest that similar clinical results are achieved via arthroscopic and open biceps tenodesis (BT) techniques. Purpose: To quantify the postoperative migration of the BT construct between arthroscopic suprapectoral BT (ASPBT) and open subpectoral BT (OSPBT) techniques via interference screw (IS) or single-suture suture anchor (SSSA) fixation using radiostereometric analysis. Study Design: Cohort study; Level of evidence, 2. Methods: Distal migration of the biceps tendon after OSPBT with a polyetheretherketone IS, OSPBT with 1 SSSA, ASPBT with polyetheretherketone IS, and ASPBT with 2 SSSAs was measured prospectively. Patients with symptomatic biceps tendinopathy and preoperative patient-reported outcome measures (PROMs) including Constant-Murley subjective, Single Assessment Numeric Evaluation, or Patient-Reported Outcomes Measurement Information System–Upper Extremity scores were included. A tantalum bead was sutured on the proximal end of the long head of the biceps tendon before fixation of tendon tissue. Anteroposterior radiographs were performed immediately postoperatively, at 1 week, and at 3 months. Bead migration was measured, and preoperative PROMs were compared with those at latest follow-up. Results: Of 115 patients, 94 (82%) were available for final follow-up. IS fixation yielded the least tendon migration with no difference between the open and arthroscopic approaches (4.31 vs 5.04 mm; P = .70). Fixation with 1 suture anchor demonstrated significantly greater migration than that achieved with an IS at both 1 week (6.47 vs 0.1 mm, 6.47 vs 1.75 mm, P < .001;) and 3 months (14.76 vs 4.31 mm, 14.76 vs 5.04 mm, P < .001) postoperatively. Two-suture anchor fixation yielded significantly greater migration than IS fixation at 1 week (7.02 vs 0.1 mm, P < .001; 7.02 vs 1.75 mm, P = .003) but not 3 months postoperatively (8.06 vs 4.31 mm, P = .10; 8.06 vs 5.04 mm, P = .07). Four patients with suture anchor fixation (3 patients in the OSPBT 1 SSSA group, 9.4%, and 1 patient in the ASPBT 2 SSSAs group, 3.8%) developed a Popeye deformity, whereas no Popeye deformities occurred in the IS groups. Mean 3-month bead migration in patients with and without a Popeye deformity was 60.8 and 11.2 mm, respectively ( P < .0001). PROMs did not differ among groups at final follow-up. Conclusion: Interference screw fixation yielded the least tendon migration whether achieved arthroscopically or open. The available data indicated that fixation with 1 SSSA but not 2 SSSAs resulted in significantly greater migration than that achieved with an IS. Despite variations in tendon migration, PROMs were similar among all groups. When SSSAs are used, tendon migration may be minimized by using ≥2 anchors.

Purpose To construct an algorithm to optimize clinical outcomes in subacromial impingement based on current, high‐level evidence. Methods A systematic review of all clinical trials on subacromial impingement published from 1999 to 2020 was performed. Demographic, clinical, range of motion (ROM), and patient‐reported outcome measure (PROM) data were collected. Interventions were compared via arm‐based Bayesian network meta‐analysis in a random‐effects model and treatments ranked via surface under the cumulative ranking curves with respect to 3 domains: pain, PROMs, and ROM. Results A total of 35 studies comprising 3,643 shoulders (42% female, age 50 ± 5 years) were included. Arthroscopic decompression with acromioplasty ranked much greater than arthroscopic decompression alone for pain relief and PROM improvement, but the difference in absolute PROMs was not statistically significant. Corticosteroid injection (CSI) alone demonstrated inferior outcomes across all 3 domains (pain, PROMs, and ROM) with low cumulative rankings. Physical therapy (PT) with CSI demonstrated moderate‐to‐excellent clinical improvement across all 3 domains whereas PT alone demonstrated excellent ROM and low‐moderate outcomes in pain and PROM domains. PT with nonsteroidal anti‐inflammatory drugs or alternative therapies ranked highly for PROM outcomes and moderate for pain and ROM domains. Finally, platelet‐rich plasma injections demonstrated moderate outcomes for pain, forward flexion, and abduction with very low‐ranking outcomes for PROMs and external rotation. Conclusions Arthroscopic decompression with acromioplasty and PT demonstrated superior outcomes whereas CSI demonstrated poor outcomes in all 3 domains (pain, PROMs, and ROM). For patients with significant symptoms, the authors recommend PT with CSI as a first‐line treatment, followed by acromioplasty and PT if conservative treatment fails. For patients with symptoms limited to 1 to 2 domains, the authors recommend a shared decision‐making approach focusing on treatment rankings within domains pertinent to individual patient symptomatology. Level of Evidence I, systematic review and network meta‐analysis of Level I studies.

Purpose To develop a machine learning algorithm to predict total charges after ambulatory hip arthroscopy and create a risk‐adjusted payment model based on patient comorbidities. Methods A retrospective review of the New York State Ambulatory Surgery and Services database was performed to identify patients who underwent elective hip arthroscopy between 2015 and 2016. Features included in initial models consisted of patient characteristics, medical comorbidities, and procedure‐specific variables. Models were generated to predict total charges using 5 algorithms. Model performance was assessed by the root‐mean‐square error, root‐mean‐square logarithmic error, and coefficient of determination. Global variable importance and partial dependence curves were constructed to show the impact of each input feature on total charges. For performance benchmarking, the best candidate model was compared with a multivariate linear regression using the same input features. Results A total of 5,121 patients were included. The median cost after hip arthroscopy was $19,720 (interquartile range, $12,399‐$26,439). The gradient‐boosted ensemble model showed the best performance (root‐mean‐square error, $3,800 [95% confidence interval, $3,700‐$3,900]; logarithmic root‐mean‐square error, 0.249 [95% confidence interval, 0.24‐0.26]; R 2 = 0.73). Major cost drivers included total hours in facility less than 12 or more than 15, longer procedure time, performance of a labral repair, age younger than 30 years, Elixhauser Comorbidity Index (ECI) of 1 or greater, African American race, residence in extreme urban and rural areas, and higher household and neighborhood income. Conclusions The gradient‐boosted ensemble model effectively predicted total charges after hip arthroscopy. Few modifiable variables were identified other than anesthesia type; nonmodifiable drivers of total charges included duration of care less than 12 hours or more than 15 hours, operating room time more than 100 minutes, age younger than 30 years, performance of a labral repair, and ECI greater than 0. Stratification of patients based on the ECI highlighted the increased financial risk borne by physicians via flat reimbursement schedules given variable degrees of comorbidities. Level of Evidence Level III, retrospective cohort study.

Purpose To develop and internally validate a machine‐learning algorithm to reliably predict cost after anterior cruciate ligament reconstruction (ACLR). Methods A retrospective review of the New York State Ambulatory Surgery and Services database was performed to identify patients who underwent elective ACLR from 2015 to 2016. Features included in initial models consisted of patient characteristics (age, sex, insurance status, income, medical comorbidities as classified by the Clinical Classifications Software diagnosis code) as well as intraoperative variables (type of anesthesia and procedure‐specific factors). Models were generated to predict total charges using 4 algorithms: random forest, extreme gradient boost, elastic net penalized regression, and support vector machines with radial kernels. Training was performed with 10‐fold cross‐validation followed by internal validation via 0.632 bootstrapping. Model discriminative performance was assessed by area under the receiver operating characteristic curve, calibration, and the Brier score. Decision curve analysis was performed to demonstrate the net benefit of using the final model in practice. Results In total, 7,311 patients undergoing ambulatory ACLR were included. The random forest model demonstrated the best performance assessed via internal validation (area under the curve = 0.85), calibration, and the Brier score (0.208). Cost incurred was influenced by anesthesia type, operating room time, and number of chronic comorbidities. Decision curve analysis revealed a net benefit for use of the random forest model and the model was integrated into a web‐based open‐access application. Conclusions The random forest model predicted cost after ambulatory ACLR using a large, statewide database with good performance. The top variables found to predict increased charges were general anesthesia, operating room time, meniscal repair, self‐pay insurance, patient neighborhood characteristics, and number of chronic conditions. Level of Evidence III, retrospective cohort study.

Purpose The purposes of this study were to identify patient characteristics and risk factors for overnight admission following outpatient hip arthroscopy and to develop a machine learning algorithm that can effectively identify patients requiring admission following elective hip arthroscopy. Methods A retrospective review of a prospectively collected national surgical outcomes database was performed to identify patients who underwent elective outpatient hip arthroscopy from 2006 to 2018. Patients admitted overnight postoperatively were identified as those with length of stay of 1 or more days. Models were generated using random forest (RF), extreme gradient boosting (XGBoost), adaptive boosting (AdaBoost), elastic net penalized logistic regression, and an additional model was produced as a weighted ensemble of the four final algorithms. Results Overall, 1,276 patients were included. The median age was 43 years, and 64.2% (819) were female. Of the included patients, 109 (8.5%) required an overnight stay following elective outpatient hip arthroscopy. The most important factors for inpatient admission were increasing operative time, general anesthesia, age extremes, male gender, greater body mass index (BMI), American Society of Anesthesiologists classification >1, and the following preoperative lab values outside of normal ranges: sodium, platelet count, hematocrit, and leukocyte count. The ensemble model achieved the best performance based on discrimination assessed via internal validation (area under the curve = .71), calibration, and decision curve analysis. The model was integrated into a Web‐based open‐access application able to provide both personalized predictions and explanations. Conclusion A machine learning algorithm developed based on preoperative features identified increasing operative time, age extremes, greater BMI, sodium, hematocrit, platelets, and leukocyte count as the most important variables associated with inpatient admission with fair validity.

Purpose To compare the performance of the Patient‐Reported Outcomes Measurement Information System (PROMIS) physical function (PF), pain interference (PIF), and depression computer adaptive tests (CAT) relative to legacy instruments in patients undergoing meniscal surgery. Methods Patients scheduled to undergo meniscal surgery completed legacy knee function PROMs (International Knee Documentation Committee [IKDC], Knee Injury and Osteoarthritis Outcome Score [KOOS] subscores), Marx Activity Rating Scale (MARS), Veterans‐Rand 12 (VR12), Short Form 12 (SF12), and the Brief Resilience Scale (BRS) alongside PROMIS PF, PIF, and Depression preoperatively. Spearman rank correlations were calculated, and score distributions were examined for floor and ceiling effects. Results 152 patients (46.6 ± 14.9 years, 67.1% male) completed PROMs for appropriate inclusion. PROMIS PF yielded high‐moderate to high correlations with the IKDC and KOOS subscales ( r = 0.61 to 0.73), demonstrating similar performance to the IKDC. PROMIS PIF demonstrated moderately high‐moderate to high correlations with the IKDC, KOOS subscales, VR‐12 Physical Component Score (PCS), and SF12 PCS ( r = 0.62 to 0.71), performing comparably to KOOS Pain ( r = 0.55 to 0.92). PROMIS Depression demonstrated moderate to high‐moderate correlations with the mental health legacies ( r = 0.46 to 0.66). Significant ceiling effects were observed for MARS (n = 29, 18.8%), and significant floor effects were exhibited by PROMIS Depression (n = 38, 25%) and MARS (n = 27, 17.6%). Conclusion The PROMIS PF, PIF CAT, and Depression instruments exhibit comparable performance profiles relative to legacy knee PROMs. PROMIS PF and PIF demonstrated no floor and ceiling effects, whereas PROMIS Depression exhibited a significant relative floor effect. PROMIS PF and PIF may be appropriately used to establish functional baselines preoperatively. Level of Evidence IV, diagnostic case series.