Machine Learning Accurately Predicts Short-Term Outcomes Following Open Reduction and Internal Fixation of Ankle Fractures.

Ankle fractures are common orthopedic injuries with favorable outcomes when managed with open reduction and internal fixation (ORIF). Several patient-related risk factors may contribute to poor short-term outcomes, and machine learning may be a valuable tool for predicting outcomes. The objective of this study was to evaluate machine-learning algorithms for accurately predicting short-term outcomes after ORIF for ankle fractures. The Nationwide Inpatient Sample and Nationwide Readmissions Database were queried for adult patients ≥18 years old who underwent ORIF of an ankle fracture during 2013 or 2014. Morbidity and mortality, length of stay >3 days, and 30-day all-cause readmission were the outcomes of interest. Two machine-learning models were created to identify patient and hospital characteristics associated with the 3 outcomes. The machine learning models were evaluated using confusion matrices and receiver operating characteristic area under the curve values. A total of 16,501 cases were drawn from the Nationwide Inpatient Sample and used to assess morbidity and mortality and length of stay >3 days, and 33,504 cases were drawn from the Nationwide Readmissions Database to assess 30-day readmission. Older age, Medicaid, Medicare, deficiency anemia, congestive heart failure, chronic lung disease, diabetes, hypertension, and renal failure were the variables associated with a statistically significant increased risk of developing all 3 adverse events. Logistic regression and gradient boosting had similar area under the curve values for each outcome, but gradient boosting was more accurate and more specific for predicting each outcome. Our results suggest that several comorbidities may be associated with adverse short-term outcomes after ORIF of ankle fractures, and that machine learning can accurately predict these outcomes.

Comparing 30-day all-cause readmission rates between tibiotalar fusion and total ankle replacement.

BACKGROUND: End-stage ankle arthritis is a debilitating condition that negatively impacts patient quality of life. Tibiotalar fusion and total ankle replacement are treatment options for managing ankle arthritis. Few studies have examined short term readmission rates of these two procedures. The objective of this study was compare all-cause 30-day readmission rates between patients undergoing tibiotalar fusion vs. total ankle replacement. METHODS: This study queried the Nationwide Readmission Database (NRD) from 2013-2014 and used international classification of disease, 9th revision (ICD-9) procedure codes to identify all patients who underwent a tibiotalar fusion or a total ankle replacement. Comorbidities, insurance status, hospital characteristics, and readmission rates were statistically compared between the two cohorts. Risk factors were then identified for 30-day readmission. RESULTS: A total of 5660 patients were analyzed with 2667 in the tibiotalar fusion cohort and 2993 in the total ankle replacement cohort. Univariate analysis revealed that the readmission rate after tibiotalar fusion (4.4%) was statistically greater than after total ankle replacement (1.4%). Multivariable regression analysis indicated that deficiency anemia (OR 2.18), coagulopathy (OR 3.51), renal failure (OR 2.83), other insurance relative to private (OR 3.40), and tibiotalar fusion (OR 2.51) were all statistically significant independent risk factors for having a readmission within 30-days. CONCLUSIONS: These findings suggest that during the short-term period following discharge from the hospital, patients who received a tibiotalar fusion are more likely to experience a 30-day readmission. These findings are important for decision making when a surgeon encounters a patient with end stage ankle arthritis. LEVEL OF EVIDENCE: Level III, cohort study.

Identifying Risk Factors for 30-Day Readmissions After Triple Arthrodesis Surgery.

Rigid flatfoot deformity is a debilitating condition that can be managed by triple arthrodesis surgery. Triple arthrodesis has the potential to restore health-related quality of life, but it is also associated with several complications. Few studies have examined the 30-day readmission rates after triple arthrodesis. The objective of this study was to investigate risk factors for 30-day all-cause readmissions after triple arthrodesis. The nationwide readmission database was queried from 2013. By using International Classification of Disease, Ninth Revision, procedure codes, all triple arthrodesis procedures were identified. Demographic factors, comorbidities, insurance status, and hospital characteristics were statistically compared between patients who experienced a 30-day readmission and those who did not. Multivariable logistic regression was used to identify independent risk factors for 30-day readmission. Overall, 1916 triple arthrodesis cases were identified. The overall 30-day readmission rate after triple arthrodesis was 4.6%. Univariate analysis revealed a statistically higher proportion of patients with electrolyte abnormalities (13.8% vs 4.6%; p < .01) in the patients who were readmitted within 30 days compared with those who were not. Multivariable analysis demonstrated Medicaid insurance, relative to private insurance, as the only statistically significant predictor of 30-day readmission with an odds ratio of 4.43 (p < .05). These results suggest that patients of lower socioeconomic status may be at a greater risk for development of a short-term readmission after triple arthrodesis surgery. These findings are important for surgeon and patient communication, counseling, and postoperative care when choosing to pursue triple arthrodesis surgery.

Recovery room radiographs after total hip arthroplasty: tradition vs utility?

Routine inpatient radiographs after total hip arthroplasty can be taken in the recovery room immediately after surgery or in the radiology suite later in the hospital stay. In a review of 632 consecutive recovery room series, we found that 17% of series were inadequate to detect technical issues. We identified technical issues on 12 series (1.9%) and technical issues that impacted inpatient management on 2 series (0.3%). One of these 2 was a dislocation that was detected clinically before imaging. The other was a medial penetration of an acetabular screw that probably did not require the immediate revision that it received. Findings suggest that the single routine inpatient series should be taken in the radiology suite, rather than in the recovery room.

Contemporary management of symptomatic lumbar disc herniations.

Lumbar disc herniations are common clinical entities that may cause lumbar-related symptoms. The spectrum of treatment options is geared toward a patient's clinical presentation and ranges from nothing to surgical intervention. Many lumbar disc herniations cause no significant symptoms. In studies of asymptomatic individuals who have never experienced lumbar-related symptoms, 30% have been reported to have major abnormality on magnetic resonance imaging. The mainstay of treatment of patients with symptomatic disc herniations is accepted to be nonoperative (as long as there are no acute or progressive neurologic deficits); this includes medications, physical therapy, and potentially lumbar injection. For patients with symptomatic disc herniations who fail to respond appropriately to conservative measures, surgical intervention may be considered. For this population, lumbar discectomy is considered to be a good option.

Whiplash causes increased laxity of cervical capsular ligament.

BACKGROUND: Previous clinical studies have identified the cervical facet joint, including the capsular ligaments, as sources of pain in whiplash patients. The goal of this study was to determine whether whiplash caused increased capsular ligament laxity by applying quasi-static loading to whiplash-exposed and control capsular ligaments. METHODS: A total of 66 capsular ligament specimens (C2/3 to C7/T1) were prepared from 12 cervical spines (6 whiplash-exposed and 6 control). The whiplash-exposed spines had been previously rear impacted at a maximum peak T1 horizontal acceleration of 8 g. Capsular ligaments were elongated at 1mm/s in increments of 0.05 mm until a tensile force of 5 N was achieved and subsequently returned to neutral position. Four pre-conditioning cycles were performed and data from the load phase of the fifth cycle were used for subsequent analyses. Ligament elongation was computed at tensile forces of 0, 0.25, 0.5, 0.75, 1.0, 2.5, and 5.0 N. Two factor, non-repeated measures ANOVA (P<0.05) was performed to determine significant differences in the average ligament elongation at tensile forces of 0 and 5 N between the whiplash-exposed and control groups and between spinal levels. FINDINGS: Average elongation of the whiplash-exposed capsular ligaments was significantly greater than that of the control ligaments at tensile forces of 0 and 5 N. No significant differences between spinal levels were observed. INTERPRETATION: Capsular ligament injuries, in the form of increased laxity, may be one component perpetuating chronic pain and clinical instability in whiplash patients.

Dynamic mechanical properties of intact human cervical spine ligaments.

BACKGROUND CONTEXT: Most previous studies have investigated ligament mechanical properties at slow elongation rates of less than 25 mm/s. PURPOSE: To determine the tensile mechanical properties, at a fast elongation rate, of intact human cervical anterior and posterior longitudinal, capsular, and interspinous and supraspinous ligaments, middle-third disc, and ligamentum flavum. STUDY DESIGN/SETTING: In vitro biomechanical study. METHODS: A total of 97 intact bone-ligament-bone specimens (C2-C3 to C7-T1) were prepared from six cervical spines (average age: 80.6 years, range, 71 to 92 years) and were elongated to complete rupture at an average (SD) peak rate of 723 (106) mm/s using a custom-built apparatus. Nonlinear force versus elongation curves were plotted and peak force, peak elongation, peak energy, and stiffness were statistically compared (p<.05) among ligaments. A mathematical model was developed to determine the quasi-static physiological ligament elongation. RESULTS: Highest average peak force, up to 244.4 and 220.0 N in the ligamentum flavum and capsular ligament, respectively, were significantly greater than in the anterior longitudinal ligament and middle-third disc. Highest peak elongation reached 5.9 mm in the intraspinous and supraspinous ligaments, significantly greater than in the middle-third disc. Highest peak energy of 0.57 J was attained in the capsular ligament, significantly greater than in the anterior longitudinal ligament and middle-third disc. Average stiffness was generally greatest in the ligamentum flavum and least in the intraspinous and supraspinous ligaments. For all ligaments, peak elongation was greater than average physiological elongation computed using the mathematical model. CONCLUSIONS: Comparison of the present results with previously reported data indicated that high-speed elongation may cause cervical ligaments to fail at a higher peak force and smaller peak elongation and they may be stiffer and absorb less energy, as compared with a slow elongation rate. These comparisons may be useful to clinicians for diagnosing cervical ligament injuries based upon the specific trauma.

Neck ligament strength is decreased following whiplash trauma.

BACKGROUND: Previous clinical studies have documented successful neck pain relief in whiplash patients using nerve block and radiofrequency ablation of facet joint afferents, including capsular ligament nerves. No previous study has documented injuries to the neck ligaments as determined by altered dynamic mechanical properties due to whiplash. The goal of the present study was to determine the dynamic mechanical properties of whiplash-exposed human cervical spine ligaments. Additionally, the present data were compared to previously reported control data. The ligaments included the anterior and posterior longitudinal, capsular, and interspinous and supraspinous ligaments, middle-third disc, and ligamentum flavum. METHODS: A total of 98 bone-ligament-bone specimens (C2-C3 to C7-T1) were prepared from six cervical spines following 3.5, 5, 6.5, and 8 g rear impacts and pre- and post-impact flexibility testing. The specimens were elongated to failure at a peak rate of 725 (SD 95) mm/s. Failure force, elongation, and energy absorbed, as well as stiffness were determined. The mechanical properties were statistically compared among ligaments, and to the control data (significance level: P < 0.05; trend: P < 0.1). The average physiological ligament elongation was determined using a mathematical model. RESULTS: For all whiplash-exposed ligaments, the average failure elongation exceeded the average physiological elongation. The highest average failure force of 204.6 N was observed in the ligamentum flavum, significantly greater than in middle-third disc and interspinous and supraspinous ligaments. The highest average failure elongation of 4.9 mm was observed in the interspinous and supraspinous ligaments, significantly greater than in the anterior longitudinal ligament, middle-third disc, and ligamentum flavum. The average energy absorbed ranged from 0.04 J by the middle-third disc to 0.44 J by the capsular ligament. The ligamentum flavum was the stiffest ligament, while the interspinous and supraspinous ligaments were most flexible. The whiplash-exposed ligaments had significantly lower (P = 0.036) failure force, 149.4 vs. 186.0 N, and a trend (P = 0.078) towards less energy absorption capacity, 308.6 vs. 397.0 J, as compared to the control data. CONCLUSION: The present decreases in neck ligament strength due to whiplash provide support for the ligament-injury hypothesis of whiplash syndrome.