Faculty Bibliography

BACKGROUND:Annotating free-text clinical notes into structured data is critical for future large-scale data analysis in institutional and national orthopaedic registries. In total hip arthroplasty (THA), classifying implant fixation, use of technology, and especially surgical approach are particularly difficult for classical machine-learning techniques. In this pilot, we evaluated the feasibility of GPT-4 to capture and justify these common elements in THA operative notes using a custom few-shot learning prompt. METHODS:The GPT-4 was trained with a few-shot learning approach using plain language descriptions of various fixations, technologies, and approaches, along with examples from gold-standard operative notes-four for fixation, 11 for technology, and 13 for surgical approach. The test set comprised 240 unique notes (60 for fixation, 90 for technology, and 120 for approach) from primary THAs performed by 38 surgeons at a single institution (November 2011 to March 2024). The GPT-4's output was compared against manual chart reviews for accuracy. The quality of clinical justifications was assessed using Flesch-Kincaid Grade Level (FKGL) scores for readability, self-BLEU scores for logical diversity, and character-level sequence matches with original notes. RESULTS:The GPT-4 classified fixation, technology, and approach with an overall accuracy of 100, 98.9, and 97.5%, respectively. The model also provided justifications for classifications with average FKGL scores of 17.9, 16.2, and 24.4 for fixation, technology, and approach, respectively, and average self-BLEU scores of < 0.1 each. Justifications had character-level sequence matches of 87.6, 89.2, and 96.5%, respectively, with direct note citations for fixation, technology, and approach. CONCLUSION/CONCLUSIONS:Applying GPT-4 with a custom few-shot prompt to THA operative notes demonstrated excellent performance in capturing fixation, technology, and approach methods. Moreover, the model's ability to cite details from the original notes is critical for model validation before widespread adoption, exhibiting a promising alternative to manual chart review for clinical data capture.

BACKGROUND:There is a growing number of patients who undergo multiple primary hip and knee joint arthroplasties during their lifetime. Whether patients who have multiple replaced joints are at an increased long-term risk of periprosthetic joint infection (PJI) is not known. The purpose of this study was to compare rates of PJI in patients who have more than one primary arthroplasty. METHODS:We reviewed 36,129 patients who underwent primary total joint arthroplasty at a single institution from 2011 to 2024. Patients were categorized as having one to four primary hip or knee arthroplasties. The PJI incidence was compared using Chi-square testing and binary logistic regressions, and multivariate models adjusted for sex, body mass index, diabetes, renal disease, smoking status, and Charlson Comorbidity Index (CCI). Sub-analyses compared patients who had one versus two, three, and four arthroplasties. RESULTS:When comparing patients who had one, two, three, or four primary joint arthroplasties, there was no significant difference in the rates of PJI between groups (P = 0.112). Multivariate analyses showed no statistically significant association between the number of arthroplasties and PJI (adjusted odds ratio (OR) for two, three, and four arthroplasties versus one: 1.34, 95% confidence interval (CI) 1.02 to 1.74, P = 0.083; 1.98, 95% CI 0.77 to 4.12, P = 0.105; 1.57, 95% CI 0.09 to 7.24, P = 0.657, respectively). Sub-analyses comparing one versus three and one versus four arthroplasties showed no significant differences. CONCLUSION/CONCLUSIONS:In this single-institution cohort, additional primary hip or knee arthroplasties did not appear to substantially increase PJI risk. These findings suggest a potential trend that requires confirmation with larger, prospective, multicenter, or registry-based studies. Nevertheless, these results provide preliminary evidence to inform patient counseling and guide future research on the risks of multiple arthroplasties.

PURPOSE/OBJECTIVE:To investigate the incidence and risk factors for dressing-induced allergic contact dermatitis (DIACD) following total hip and knee arthroplasty (THA and TKA, respectively) across different dressings and sealants. METHODS:A retrospective review was conducted of patients who underwent primary, elective THA or TKA between 2019 and 2024 with ≥ 90 days of follow-up. Incidences of DIACD were identified by reviewing medical records for "allergy" diagnoses and use of antihistamines or corticosteroids within 30 days postoperatively. Patient characteristics, prior exposure, treatment, dressing type, and allergy history were analyzed. RESULTS:A total of 61 (0.3%) of the 23,396 investigated patients developed a DIACD on average 12.2 ± 7.3 days postoperatively. Overall, 41% had a preoperative allergy (excluding seasonal), and 55.7% were treated with topical or low-dose oral antihistamines and corticosteroids. The majority (41%) of the DIACD involved mesh-adhesive dressings, and a liquid skin adhesive (2-octyl cyanoacrylate) was also used in 41% of cases, often in combination with the primary dressing. Of the 61 DIACD patients, 24 (39.3%) had previously undergone THA or TKA, and nearly half of these (n = 11, 45.8%) had been exposed to the same dressing without prior occurrence of DIACD. DIACD patients were significantly more likely to have undergone TKA (73.8 vs. 58.3%, p = 0.015) and to have never smoked (75.4 vs. 58.4%, p = 0.014). The effect sizes of these findings were negligible (Cramer's V = 0.016 and 0.019, respectively). CONCLUSIONS:The incidence of DIACD following joint arthroplasty is low (0.3%) but remains a frustrating complication, primarily occurring two weeks postoperatively, with mesh-adhesive dressings most frequently implicated. Patients with prior exposure to dressings, those undergoing TKA, and non-smokers are at higher risk. Identifying at-risk patients can guide dressing selection and application.

BACKGROUND/UNASSIGNED:Technological advancements in total hip arthroplasty (THA), including robotic-assisted (RA-THA) and navigation-assisted (NA-THA) techniques, aim to improve outcomes. However, impact on recovery timing remains unclear. This study examined whether these technologies reduce the time to reach the minimal clinically important difference (MCID) on the Hip Disability and Osteoarthritis Outcome Score for Joint Replacement compared with conventional THA. METHODS/UNASSIGNED:This retrospective study analyzed osteoarthritic THA patients (01/2020-04/2023) who completed preoperative and postoperative Hip Disability and Osteoarthritis Outcome Score for Joint Replacement questionnaires. The exclusion criteria included bilateral procedures or revision within 1 year. MCID was defined using anchor-based (23 points) and distribution-based thresholds (7.6 points). Multivariable interval-censored accelerated failure time models assessed time to MCID. RESULTS/UNASSIGNED:= .140). CONCLUSIONS/UNASSIGNED:Anchor-based MCID demonstrated comparable recovery times across RA, NA, and conventional THA, suggesting no patient-perceived advantage with technology. Distribution-based thresholds indicated RA-THA achieved faster statistically significant improvement, though the relevance remains uncertain.

BACKGROUND:Patients who undergo primary total knee arthroplasty (TKA) may return to work at variable times following surgery, the timeline for which is partly affected by the physical intensity of their occupation. The purpose of this study was to evaluate patient satisfaction and limitations when returning to work following TKA. METHODS:This retrospective review surveyed patients undergoing primary TKA between June 2011 and January 2022, with at least 1 year of follow up, regarding return to work. Of the 914 respondents, 507 (55.5 %) worked preoperatively and were stratified into high intensity (HI) (i.e., laborer, construction), standard intensity (SI) (i.e., walking, climbing stairs), or low intensity (LI) (i.e., sedentary desk jobs) groups. Baseline characteristics and survey responses were compared across groups. Among those who worked preoperatively, 35 (6.9 %), 213 (42 %), and 259 (51.1 %) were in the HI, SI, and LI groups, respectively. RESULTS:Of the 507 patients who worked prior to TKA, 447 (88.2 %) returned to work after surgery and 60 (11.8 %) did not. The HI group was comprised of more young males and more smokers than the SI and LI groups. In the LI group, 30 % returned within 1 month following surgery and an additional 44 % within 2 months. Similarly in the SI group, 11 % returned to work in less than 1 month with an additional 39 % returning within 2 months. In the HI group, 4 % returned within the first month and additional 42 % returned within 2 months. HI workers were more commonly hindered in their return (HI: 30.8 %, SI: 23.1 %, LI: 7.7 %), 'moderately declined' in their work ability (HI: 23.1 %, SI: 9.7 %, 3.0 %), and 'very unsatisfied' with their return (HI: 11.5 %, SI: 10.8 %, LI: 8.1 %). CONCLUSION/CONCLUSIONS:TKA leads to improvements in work function and satisfaction across all intensity levels, but HI work is associated with longer recovery times and comparatively lower return-to-work satisfaction compared to SI and LI groups.

BACKGROUND:Obesity is a known risk factor for complications after total hip arthroplasty (THA), with societal guidelines recommending surgical delay for patients above body mass index (BMI) targets. Consequently, patients are motivated to reach BMI targets before the office visit, discussing surgical booking. Our study investigates BMI fluctuations between surgical booking and the surgery date and whether these fluctuations have implications for peri- and postoperative outcomes. METHODS:We retrospectively reviewed our institutional database for elective, primary, unilateral THA from 2015 to 2024 with a minimum 90-day follow-up. The cohort was stratified into three groups by percent BMI change from booking date to THA date: Group 1, decrease in BMI; Group 2, 0 to 5% increase in BMI; and Group 3, > 5% increase in BMI. Baseline demographic factors, as well as peri- and postoperative outcomes, were compared. A multivariate regression analysis evaluated risk factors for interval change in BMI. We reviewed 10,400 THA patients who had an average 62-day delay between booking and surgery dates, where 44.2, 42.2, and 13.6% were allocated to Groups 1, 2, and 3, respectively. RESULTS:Operative time, length of stay, and discharge to a rehab facility were significantly higher in Group 3 compared to the others. There were no significant differences seen in 90-day outcomes, including emergency department (ED) visits, readmissions, or revision rates. Multivariate regression analyses identified that elevated BMI at the time of surgery predicted increased septic revision incidence (OR [odds ratio]: 1.1, P < 0.001). Surgical delay and BMI change between booking and surgery did not influence all-cause and septic revision rates. CONCLUSION/CONCLUSIONS:Our study is the first to evaluate preoperative BMI fluctuations between the booking date and surgery date. We found that most patients (55.8%) gained weight, with 13.6% increasing > 5% BMI. Importantly, these weight changes do not impact short-term complications or revision rates, comforting THA surgeons who see interval weight gain on the day of surgery.

BACKGROUND:Return to sports is a concern for many patients undergoing total hip arthroplasty (THA). As younger, active patients increasingly undergo THA, identifying factors that influence athletic recovery is critical. Although prior work has reported favorable outcomes, large-scale studies comparing return rates across both surgical approaches and sport types remain limited. METHODS:We retrospectively analyzed 1,115 athletically participant THA patients and compared rates of returning to sports up to one year postoperatively. Survey data captured sport type, participation at four time points, frequency, exertion, and perceived recovery. Patients were stratified by surgical approach: posterior (PA) (n = 519), anterior (AA) (n = 556), and lateral (LA) (n = 50). RESULTS:Preoperatively, 45.2% of patients were active in at least one sport versus 43.7% at one year postoperatively; 73.2% of preoperative athletes returned, and 23.3% of those previously inactive took up sports. Return-to-sport rates were 68.1, 77.0, and 81.8% for PA, AA, and LA, respectively (Chi-square = 3.42, P = 0.181). There were no significant differences between AA and PA, and the lateral approach was not statistically comparable due to a smaller sample size. Low-impact sports had significantly higher return rates than high-impact sports (72.4 versus 50.0%; P < 0.001). At survey completion (> one year postoperatively), satisfaction rates were high across all sports, and activity restrictions were reported by about one in five patients, regardless of approach. Sport-specific outcomes showed that most patients maintained or improved performance. CONCLUSION/CONCLUSIONS:Patients undergoing THA can expect high return rates and satisfaction with athletic endeavors. Surgical approach generally does not affect return-to-sport outcomes after THA. Most patients, regardless of approach, resumed or exceeded preoperative activity. In addition to high return rates among previously active patients, many previously inactive patients participated in sports after THA, highlighting the role of THA in enabling sport resumption and participation.

BACKGROUND:Sleep impairment following total hip arthroplasty (THA) is common and may decrease patient satisfaction and early recovery. Standardized postoperative recommendations for sleep disturbance have not been established. We aimed to assess whether melatonin use improves sleep quality and quantity in the acute period following THA. METHODS:Patients undergoing primary, elective THA between July 2021 and March 2024 were prospectively enrolled and randomized to receive either five mg of melatonin or a placebo nightly for 14 days postoperatively. Participants recorded nightly pain scores on the visual analog scale (VAS), the number of hours slept, and the number of nighttime awakenings in a sleep diary. Sleep disturbance was assessed preoperatively and on postoperative day (POD) 14 using the patient-reported outcome measurement information system sleep disturbance (PROMIS-SD) form. Epworth Sleepiness Scores (ESS) were collected to assess sleep quality and were the primary outcome of this study. Of the 139 patients who completed the study protocol, there were 64 patients in the placebo group and 75 patients in the melatonin group. RESULTS:Both groups demonstrated comparable postoperative ESS (melatonin: 6.0 ± 4.0; placebo: 6.8 ± 4.5, P = 0.35). Melatonin patients experienced significantly more hours slept on POD2 (6.5 ± 1.7; 5.7 ± 2.4, P = 0.017) and averaged over POD one to three (6.1 ± 1.6; 5.7 ± 2.0, P = 0.14), although this was not statistically significant. Fewer nighttime awakenings in the melatonin group were observed on POD two (2.7 ± 1.5; 3.1 ± 2.0, P = 0.28), although this was not statistically significant. The melatonin group demonstrated significantly lower postoperative PROMIS-SD scores (52.5 ± 9.3; 56.3 ± 9.2, P = 0.040). CONCLUSION/CONCLUSIONS:Melatonin may not improve overall postoperative sleep quality following THA as measured by the ESS. Melatonin may promote sleep duration in the POD one to three period, although potential benefits wane after POD three. Melatonin is safe and can be considered for THA patients experiencing early postoperative sleep disturbance.