An Artificial Intelligence Approach to Predicting Unplanned Intubation Following Anterior Cervical Discectomy and Fusion.

STUDY DESIGN: Level III retrospective database study. OBJECTIVES: The purpose of this study is to determine if machine learning algorithms are effective in predicting unplanned intubation following anterior cervical discectomy and fusion (ACDF). METHODS: The National Surgical Quality Initiative Program (NSQIP) was queried to select patients who had undergone ACDF. Machine learning analysis was conducted in Python and multivariate regression analysis was conducted in R. C-Statistics area under the curve (AUC) and prediction accuracy were used to measure the classifier's effectiveness in distinguishing cases. RESULTS: In total, 54 502 patients met the study criteria. Of these patients, .51% underwent an unplanned re-intubation. Machine learning algorithms accurately classified between 72%-100% of the test cases with AUC values of between .52-.77. Multivariable regression indicated that the number of levels fused, male sex, COPD, American Society of Anesthesiologists (ASA) > 2, increased operating time, Age > 65, pre-operative weight loss, dialysis, and disseminated cancer were associated with increased risk of unplanned intubation. CONCLUSIONS: The models presented here achieved high accuracy in predicting risk factors for re-intubation following ACDF surgery. Machine learning analysis may be useful in identifying patients who are at a higher risk of unplanned post-operative re-intubation and their treatment plans can be modified to prophylactically prevent respiratory compromise and consequently unplanned re-intubation.

Graft Choice for Anterior Cruciate Ligament Reconstruction in Women Aged 25 Years and Younger: A Systematic Review.

CONTEXT: Although anterior cruciate ligament (ACL) tears are relatively common in athletic populations, few studies have systematically reviewed graft choice in young women. OBJECTIVE: To quantitatively and qualitatively examine reported outcomes for graft choice in women aged 25 years and younger undergoing primary ACL reconstruction. DATA SOURCE: A systematic review was performed using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. An electronic search in the PubMed (includes MEDLINE) and EMBASE databases was completed using a combination of key terms. STUDY SELECTION: Studies were included if they reported graft choice outcomes in women aged 25 years and younger. STUDY DESIGN: Systematic review. LEVEL OF EVIDENCE: Level 4. DATA EXTRACTION: The following information was extracted: title, author, year of publication, number of female patients and age, graft type, follow-up, and patient-reported outcome measures. The following outcome scores were identified as being reported or not reported by each study: graft failure, contralateral ACL (CACL) rupture, IKDC (International Knee Documentation Committee), graft survival (Kaplan-Meier), Lysholm, Tegner, KT-1000, kneeling pain, return to sport, and Lachman. RESULTS: Of 1170 identified articles, 16 met inclusion criteria, reporting on 1385 female patients aged 25 years and younger. Comparison of 655 bone-patellar tendon-bone (BPTB) versus 525 hamstring tendon (HT) autografts showed significant differences in mean failure rate between BPTB autografts (6.13% ± 2.58%) and HT autografts (17.35% ± 8.19%), P = 0.001. No statistically significant differences in CACL failure rates were found between BPTB autografts and HT autografts (P = 0.25). Pooled results for IKDC were possible in 3 of the HT autograft studies, showing a mean score of 88.31 (95% CI 83.53-93.08). Pooled Lysholm score results were possible in 2 of the HT autograft studies, showing a mean score of 93.46 (95% CI 91.90-95.01). CONCLUSION: In female patients aged 25 years and younger, BPTB autografts showed significantly less graft failure compared with HT autografts. However, BPTB autografts had comparable patient-reported outcomes compared with HT autografts with the available data. The overall state of evidence for graft choice in female patients aged 25 years and younger is low. Future studies should report statistics by age and sex to allow for further analysis of graft choice for this specific population that is known to be more vulnerable to ACL injury.

Epidemiology of rugby-related fractures in high school- and college-aged players in the United States: an analysis of the 1999-2018 NEISS database.

OBJECTIVES: Rugby is a high-impact collision sport with identical competition rules by sex. The aim of this study was to analyze the trend of rugby-related fractures by body site, sex, and age in amateur athletes from 1999 to 2018. METHODS: The National Electronic Injury Surveillance System (NEISS) database was queried to characterize rugby-related injuries from 1999 to 2018 in patients aged 14-23 years old. National injury estimates were calculated using sample weights. Chi-square analysis and one-way ANOVA were performed to compare categorical variables. RESULTS: Out of a total of 43,722 weighted cases of rugby-related fractures over the 19-year period, 70.9% were among high school- and college-aged males and females between the ages of 14-23 years (N = 30,996). Males constituted 79.3% of cases whereas females composed 20.7% of the cases. The proportion of upper extremity fractures was similar in both males and females, yet facial fractures were significantly more common among males than females (27.9% vs. 14.6%, P < 0.001). Among facial fractures, nasal fractures represented 74.4% of the sample, yet facial fractures only resulted in hospital admittance in 1.3% of cases. Lower extremity fractures were more likely to be severe, with 11.3% of LE cases being admitted to the hospital. CONCLUSION: Rugby players in the US diagnosed in the ED with fractures consisted largely of male, high school- and college-aged athletes. Males and females experienced upper and lower-extremity fractures at comparable rates, yet lower extremity injuries were more likely to be admitted to a hospital. Men were significantly more likely to experience a facial fracture in which a majority were nasal fractures.

A quality assessment of YouTube content on shoulder instability.

OBJECTIVES: The Internet is a widely used resource for patients seeking health information, yet little editing or regulations are imposed on posted material. We sought to assess the quality and accuracy of information presented on shoulder instability on the online video platform YouTube. We hypothesize that YouTube videos concerning shoulder instability will be of little quality, accuracy, and reliability. METHODS: The first 50 YouTube videos resulting from the keyword query 'shoulder instability' were analyzed. The Journal of American Medical Association (JAMA) benchmark criteria (score range, 0-4) was used to assess video accuracy and reliability, and the Global Quality Score (GQS; score range, 0-5) was used to assess the quality of the video's educational content along with a generated Shoulder-Specific Score (SSS). RESULTS: The 50 videos observed collectively had 5,007,486 views, with the mean number of views being 100,149.72 ± 227,218.04. Of all videos observed, 32% were from a medical source and 56% had content relating to pathology information. The mean JAMA score was 2.84 ± 0.74, with the highest scores coming from academic sources. The mean GQS and SSS scores were 2.68 ± 0.84 and 5.30 ± 3.78. The mean GQS score was highest in videos from medical sources (3.3 ± 0.8) and videos about surgical technique/approach (3.2 ± 1.1). Advertisements were negative predictors of the JAMA score (ß = -0.324, P = 0.014), and academic (ß = 0.322, P = 0.015) and physician sources (ß = 0.356, P = 0.008) were positive predictors. CONCLUSION: YouTube videos on shoulder instability are of low quality and accuracy and are not reliable. Care providers should be aware of the overall low quality of information available on YouTube regarding shoulder instability.

Supervised Machine Learning for Predicting Length of Stay After Lumbar Arthrodesis: A Comprehensive Artificial Intelligence Approach.

INTRODUCTION: Few studies have evaluated the utility of machine learning techniques to predict and classify outcomes, such as length of stay (LOS), for lumbar fusion patients. Six supervised machine learning algorithms may be able to predict and classify whether a patient will experience a short or long hospital LOS after lumbar fusion surgery with a high degree of accuracy. METHODS: Data were obtained from the National Surgical Quality Improvement Program between 2009 and 2018. Demographic and comorbidity information was collected for patients who underwent anterior, anterolateral, or lateral transverse process technique arthrodesis procedure; anterior lumbar interbody fusion (ALIF); posterior, posterolateral, or lateral transverse process technique arthrodesis procedure; posterior lumbar interbody fusion/transforaminal lumbar interbody fusion (PLIF/TLIF); and posterior fusion procedure posterior spine fusion (PSF). Machine learning algorithmic analyses were done with the scikit-learn package in Python on a high-performance computing cluster. In the total sample, 85% of patients were used for training the models, whereas the remaining patients were used for testing the models. C-statistic area under the curve and prediction accuracy (PA) were calculated for each of the models to determine their accuracy in correctly classifying the test cases. RESULTS: In total, 12,915 ALIF patients, 27,212 PLIF/TLIF patients, and 23,406 PSF patients were included in the algorithmic analyses. The patient factors most strongly associated with LOS were sex, ethnicity, dialysis, and disseminated cancer. The machine learning algorithms yielded area under the curve values of between 0.673 and 0.752 (PA: 69.6% to 80.1%) for ALIF, 0.673 and 0.729 (PA: 66.0% to 81.3%) for PLIF/TLIF, and 0.698 and 0.749 (PA: 69.9% to 80.4%) for PSF. CONCLUSION: Machine learning classification algorithms were able to accurately predict long LOS for ALIF, PLIF/TLIF, and PSF patients. Supervised machine learning algorithms may be useful in clinical and administrative settings. These data may additionally help inform predictive analytic models and assist in setting patient expectations. LEVEL III: Diagnostic study, retrospective cohort study.

Foundational Health for Runners: Is it the Key to Minimizing Injury?

BACKGROUND: Injury rates in runners are as high as 80%. Here, we focus on the concept of foundational health including sleep, recovery, nutrition, stress and physical health and how it can reduce injuries. METHODS: The literature was reviewed to find papers linking running injuries and athletic performance to the foundational health topics discussed. RESULTS: There are many factors that can improve athletic performance and reduce injuries in runners other than the often-discussed topics: training philosophies, footwear, and running form. This paper shows how a multidisciplinary approach including education on sleep, rest, stress, nutrition, strength, and mobility all can improve performance and reduce injuries. CONCLUSIONS: The care and management of an injured runner is multifactorial and the treatment should be as well. By optimizing foundational health, the sports medicine professional will not only reduce injury risk, but also improve performance and overall health.