Search

Study designA retrospective, case-control study.ObjectiveWe aim to build a risk calculator predicting major perioperative complications after anterior cervical fusion. In addition, we aim to externally validate this calculator with an institutional cohort of patients who underwent anterior cervical discectomy and fusion (ACDF).Summary of background dataThe average age and proportion of patients with at least one comorbidity undergoing ACDF have increased in recent years. Given the increased morbidity and cost associated with perioperative complications and unplanned readmission, accurate risk stratification of patients undergoing ACDF is of great clinical utility.MethodsThis is a retrospective cohort study of adults who underwent anterior cervical fusion at any nonfederal California hospital between 2015 and 2017. The primary outcome was major perioperative complication or 30-day readmission. We built standard and ensemble machine learning models for risk prediction, assessing discrimination, and calibration. The best-performing model was validated on an external cohort comprised of consecutive adult patients who underwent ACDF at our institution between 2013 and 2020.ResultsA total of 23,184 patients were included in this study; there were 1886 cases of major complication or readmissions. The ensemble model was well calibrated and demonstrated an area under the receiver operating characteristic curve of 0.728. The variables most important for the ensemble model include male sex, medical comorbidities, history of complications, and teaching hospital status. The ensemble model was evaluated on the validation cohort (n=260) with an area under the receiver operating characteristic curve of 0.802. The ensemble algorithm was used to build a web-based risk calculator.ConclusionWe report derivation and external validation of an ensemble algorithm for prediction of major perioperative complications and 30-day readmission after anterior cervical fusion. This model has excellent discrimination and is well calibrated when tested on a contemporaneous external cohort of ACDF cases.

BackgroundC5 palsy is a common postoperative complication after cervical fusion and is associated with increased health care costs and diminished quality of life. Accurate prediction of C5 palsy may allow for appropriate preoperative counseling and risk stratification. We primarily aim to develop an algorithm for the prediction of C5 palsy after instrumented cervical fusion and identify novel features for risk prediction. Additionally, we aim to build a risk calculator to provide the risk of C5 palsy.MethodsWe identified adult patients who underwent instrumented cervical fusion at a tertiary care medical center between 2013 and 2020. The primary outcome was postoperative C5 palsy. We developed ensemble machine learning, standard machine learning, and logistic regression models predicting the risk of C5 palsy-assessing discrimination and calibration. Additionally, a web-based risk calculator was built with the best-performing model.ResultsA total of 1024 patients were included, with 52 cases of C5 palsy. The ensemble model was well-calibrated and demonstrated excellent discrimination with an area under the receiver-operating characteristic curve of 0.773. The following features were the most important for ensemble model performance: diabetes mellitus, bipolar disorder, C5 or C4 level, surgical approach, preoperative non-motor neurologic symptoms, degenerative disease, number of fused levels, and age.ConclusionsWe report a risk calculator that generates patient-specific C5 palsy risk after instrumented cervical fusion. Individualized risk prediction for patients may facilitate improved preoperative patient counseling and risk stratification as well as potential intraoperative mitigating measures. This tool may also aid in addressing potentially modifiable risk factors such as diabetes and obesity.

PurposePosterior cervical fusion is associated with increased rates of complications and readmission when compared to anterior fusion. Machine learning (ML) models for risk stratification of patients undergoing posterior cervical fusion remain limited. We aim to develop a novel ensemble ML algorithm for prediction of major perioperative complications and readmission after posterior cervical fusion and identify factors important to model performance.MethodsThis is a retrospective cohort study of adults who underwent posterior cervical fusion at non-federal California hospitals between 2015 and 2017. The primary outcome was readmission or major complication. We developed an ensemble model predicting complication risk using an automated ML framework. We compared performance with standard ML models and logistic regression (LR), ranking contribution of included variables to model performance.ResultsOf the included 6822 patients, 18.8% suffered a major complication or readmission. The ensemble model demonstrated slightly superior predictive performance compared to LR and standard ML models. The most important features to performance include sex, malignancy, pneumonia, stroke, and teaching hospital status. Seven of the ten most important features for the ensemble model were markedly less important for LR.ConclusionWe report an ensemble ML model for prediction of major complications and readmission after posterior cervical fusion with a modest risk prediction advantage compared to LR and benchmark ML models. Notably, the features most important to the ensemble are markedly different from those for LR, suggesting that advanced ML methods may identify novel prognostic factors for adverse outcomes after posterior cervical fusion.

BACKGROUND: Modular knee arthrodesis (MKA) is a salvage treatment option for patients with challenging periprosthetic joint infections (PJI). The purpose of this study was to investigate the outcomes of patients who underwent MKA for PJI with a single technique and determine if specific factors are associated with MKA failure. METHODS: This was a retrospective review of 81 patients who underwent MKA at a single institution. Knee Society Scores were recorded before MKA and at the final follow-up (mean 52 months). Poisson regression was used to calculate rate ratios for MKA failure secondary to infection. RESULTS: The mean patient age was 67 years; most patients were McPherson B hosts (56.8%) and had type 3 extremities (53.1%), and all had a type III infection (chronic, >4 wks). Forty-six percent of patients had a prior explantation (59.5% failed 2-stage, 40.5% failed spacer). Staphylococcus epidermidis and Staphylococcus aureus were the most common organisms, 22.2% and 18.5%, respectively. Thirty percent of patients had at least one reoperation, excluding reimplantation (14.8% irrigation and debridement/wound closure, 9.9% MKA exchange, and 7.4% amputation). Of 82.7% of MKA patients with no evidence of infection, 82.1% (56 patients) underwent reimplantation endoprosthetic reconstruction, and 67.3% of these remained infection-free at the final follow-up. DISCUSSION: MKA is a salvage option for challenging PJI cases that may serve as definitive surgical management or as a bridge to endoprosthetic reconstruction for patients who have failed prior infection control procedures.

Purpose of reviewAnterior cruciate ligament (ACL) rupture is a common injury that has important clinical and economic implications. We aimed to review the literature to identify gender, racial and ethnic disparities in incidence, treatment, and outcomes of ACL injury.Recent findingsFemales are at increased risk for ACL injury compared to males. Intrinsic differences such as increased quadriceps angle and increased posterior tibial slope may be contributing factors. Despite lower rates of injury, males undergo ACL reconstruction (ACLR) more frequently. There is conflicting evidence regarding gender differences in graft failure and ACL revision rates, but males demonstrate higher return to sport (RTS) rates. Females report worse functional outcome scores and have worse biomechanical metrics following ACLR. Direct evidence of racial and ethnic disparities is limited, but present. White athletes have greater risk of ACL injury compared to Black athletes. Non-White and Spanish-speaking patients are less likely to undergo ACLR after ACL tear. Black and Hispanic youth have greater surgical delay to ACLR, increased risk for loss to clinical follow-up, and less physical therapy sessions, thereby leading to greater deficits in knee extensor strength during rehabilitation. Hispanic and Black patients also have greater risk for hospital admission after ACLR, though this disparity is improving. Females have higher rates of ACL injury with inconclusive evidence on anatomic predisposition and ACL failure rate differences between genders. Recent literature has suggested inferior RTS and functional outcomes following ACLR in females. Though there is limited and mixed data on incidence and outcome differences between races and ethnic groups, recent studies suggest there may be disparities in those who undergo ACLR and time to treatment.

INTRODUCTION: The two historically dominant surgical options for displaced geriatric proximal humerus (PHFx) fractures are open reduction internal fixation (ORIF) and hemiarthroplasty (HA). However, shoulder arthroplasty (SA), predominantly in the form of reverse total shoulder arthroplasty (RTSA), has emerged as an attractive treatment option. We aim to compare the utilization trends, complications, and costs associated with surgical management of geriatric proximal humerus fractures (PHFs) between 2010 and 2019. We hypothesized that 1) the proportion of patients undergoing SA would increase over time, 2) the short-term complication rate in patients undergoing SA would decline over time, and 3) hospital related costs would decline for SA patients over time. PATIENTS AND METHODS: The National Inpatient Sample was queried from 2010 to 2019 to identify all PHFx in patients aged 65 or older that underwent ORIF, SA, or HA. Multivariable regression was used to evaluate differences between fixation methods regarding health care utilization metrics, hospital costs, and index hospital complications. The primary outcome of interest was the method of surgical management utilized in the treatment of geriatric PHFs, and secondary outcomes of interest included hospitalization cost, length of stay (LOS), discharge destination and index hospitalization complications. RESULTS: A total of 105 886 geriatric patients that underwent surgical management of PHFx were identified. While the proportion undergoing ORIF decreased from 59% to 29%, the proportion undergoing SA increased from 9% to 67%. Hospital costs decreased over time for patients treated with SA and increased for those treated with ORIF. Compared to ORIF, SA was associated with higher cost, decreased length of stay, and lower mortality and complication rates. CONCLUSION: Over the last decade, SA has become the most common surgical treatment modality performed for geriatric PHFx. Index hospital complications are reduced in SA patients compared to ORIF patients, driven largely by a lower rate of blood transfusion. Although costs are decreasing and average length of stay is now lower in SA patients compared to ORIF patients, SA remains associated with higher hospital costs overall.

BackgroundModular knee arthrodesis (MKU) is a salvage treatment for recurrent periprosthetic joint infection (PJI) or PJI associated with notable bone loss. Reimplantation endoprosthetic reconstruction (REI) is an option in patients with MKU who have PJI clearance but are not satisfied with pain or functional outcomes with MKU. The purpose of this study was to evaluate the outcomes of MKU to REI conversion.MethodsThis was a single-center retrospective cohort study of 56 patients who underwent MKU to REI from 2010 to 2019. All patients were staged according to the McPherson staging system. An infecting organism was documented based on pre-MKU aspiration or intraoperative cultures at the time of MKU. Rate ratios were calculated for relevant patient factors. Rate ratios were calculated using Poisson regression with a log link.ResultsThe mean REI patient age was 67 years, most of the patients were McPherson B hosts (62.5%) with a type 2 (46.4%) or type 3 (51.8%) limb score, and all PJI were chronic. The most common infecting organisms at the time of MKU were Staphylococcus epidermidis (23.2%) and Staphylococcus aureus (23.2%, MSSA 14.3%, MRSA 8.9%). The mean time from MKU to REI was 220 days. An 8.9% REI index hospitalization complication rate and a 21.4% overall complication rate (excluding reinfection) were observed. Sixty-seven percent of the patients remained infection-free at an average follow-up of 37 months, among those there was 96.4% implant survivorship. No notable association was observed between index PJI organism or McPherson staging and REI failure secondary to PJI.DiscussionApproximately two thirds of patients who undergo conversion from MKU to REI have infection-free survival at the midterm follow-up. An index infecting organism and a McPherson host type do not seem to be markedly associated with reinfection risk. These findings help guide expectations of PJI MKU conversion to REI.

BackgroundGiven the significant cost and morbidity of patients undergoing lumbar fusion, accurate preoperative risk-stratification would be of great utility. We aim to develop a machine learning model for prediction of major complications and readmission after lumbar fusion. We also aim to identify the factors most important to performance of each tested model.MethodsWe identified 38,788 adult patients who underwent lumbar fusion at any California hospital between 2015 and 2017. The primary outcome was major perioperative complication or readmission within 30 days. We build logistic regression and advanced machine learning models: XGBoost, AdaBoost, Gradient Boosting, and Random Forest. Discrimination and calibration were assessed using area under the receiver operating characteristic curve and Brier score, respectively.ResultsThere were 4470 major complications (11.5%). The XGBoost algorithm demonstrates the highest discrimination of the machine learning models, outperforming regression. The variables most important to XGBoost performance include angina pectoris, metastatic cancer, teaching hospital status, history of concussion, comorbidity burden, and workers' compensation insurance. Teaching hospital status and concussion history were not found to be important for regression.ConclusionsWe report a machine learning algorithm for prediction of major complications and readmission after lumbar fusion that outperforms logistic regression. Notably, the predictors most important for XGBoost differed from those for regression. The superior performance of XGBoost may be due to the ability of advanced machine learning methods to capture relationships between variables that regression is unable to detect. This tool may identify and address potentially modifiable risk factors, helping risk-stratify patients and decrease complication rates.