Predicting 1-Year Mortality after Hip Fracture Surgery: An Evaluation of Multiple Machine Learning Approaches

Validation of the Charlson Comorbidity Index for the prediction of 30-day and 1-year mortality among patients who underwent hip fracture surgery

INTRODUCTION

The aim of our study was to validate the original Charlson Comorbidity Index (1987) (CCI) and adjusted CCI (2011) as a prediction model for 30-day and 1-year mortality after hip fracture surgery. The secondary aim of this study was to verify each variable of the CCI as a factor associated with 30-day and 1-year mortality.

METHODS

A prospective database of two-level II trauma teaching hospitals in the Netherlands was used. The original CCI from 1987 and the adjusted CCI were calculated based on medical history. To validate the original CCI and the adjusted CCI, the CCI was plotted against the observed 30-day and 1-year mortality, and the area under the curve (AUC) was calculated.

RESULTS

A total of 3523 patients were included in this cohort study. The mean of the original CCI in this cohort was 5.1 (SD ± 2.0) and 4.6 (SD ± 1.9) for the adjusted CCI. The AUCs of the prediction models were 0.674 and 0.696 for 30-day mortality for the original and adjusted CCIs, respectively. The AUCs for 1-year mortality were 0.705 and 0.717 for the original and adjusted CCIs, respectively.

CONCLUSIONS

A higher original and adjusted CCI is associated with a higher mortality rate. The AUC was relatively low for 30-day and 1-year mortality for both the original and adjusted CCIs compared to other prediction models for hip fracture patients in our cohort. The CCI is not recommended for the prediction of 30-day and 1-year mortality in hip fracture patients.

Systematic review of multivariable prognostic models for outcomes at least 30 days after hip fracture finds 18 mortality models but no nonmortality models warranting validation

OBJECTIVES

Prognostic models have the potential to aid clinical decision-making after hip fracture. This systematic review aimed to identify, critically appraise, and summarize multivariable prediction models for mortality or other long-term recovery outcomes occurring at least 30 days after hip fracture.

STUDY DESIGN AND SETTING

MEDLINE, Embase, Scopus, Web of Science, and CINAHL databases were searched up to May 2023. Studies were included that aimed to develop multivariable models to make predictions for individuals at least 30 days after hip fracture. Risk of bias (ROB) was dual-assessed using the Prediction model Risk Of Bias ASsessment Tool. Study and model details were extracted and summarized.

RESULTS

From 5571 records, 80 eligible studies were identified. They predicted mortality in n = 55 studies/81 models and nonmortality outcomes (mobility, function, residence, medical, and surgical complications) in n = 30 studies/45 models. Most (n = 46; 58%) studies were published since 2020. A quarter of studies (n = 19; 24%) reported using 'machine-learning methods', while the remainder used logistic regression (n = 54; 68%) and other statistical methods (n = 11; 14%) to build models. Overall, 15 studies (19%) presented 18 low ROB models, all predicting mortality. Common concerns were sample size, missing data handling, inadequate internal validation, and calibration assessment. Many studies with nonmortality outcomes (n = 11; 37%) had clear data complexities that were not correctly modeled.

CONCLUSION

This review has comprehensively summarized and appraised multivariable prediction models for long-term outcomes after hip fracture. Only 15 studies of 55 predicting mortality were rated as low ROB, warranting further development of their models. All studies predicting nonmortality outcomes were high or unclear ROB. Careful consideration is required for both the methods used and justification for developing further nonmortality prediction models for this clinical population.

Development and Internal Validation of a Multivariable Prediction Model for Mortality After Hip Fracture with Machine Learning Techniques

In order to estimate the likelihood of 1, 3, 6 and 12 month mortality in patients with hip fractures, we applied a variety of machine learning methods using readily available, preoperative data. We used prospectively collected data from a single university hospital in Copenhagen, Denmark for consecutive patients with hip fractures, aged 60 years and older, treated between September 2008 to September 2010 (n = 1186). Preoperative biochemical and anamnestic data were used as predictors and outcome was survival at 1, 3, 6 and 12 months after the fracture. After feature selection for each timepoint a stratified split was done (70/30) before training and validating Random Forest models, extreme gradient boosting (XGB) and Generalized Linear Models. We evaluated and compared each model using receiver operator characteristic (ROC), calibration slope and intercept, Spiegelhalter's z- test and Decision Curve Analysis. Using combinations of between 10 and 13 anamnestic and biochemical parameters we were able to successfully estimate the likelihood of mortality with an area under the curve on ROC curves of 0.79, 0.80, 0.79 and 0.81 for 1, 3, 6 and 12 month, respectively. The XGB was the overall best calibrated and most promising model. The XGB model most successfully estimated the likelihood of mortality postoperatively. An easy-to-use model could be helpful in perioperative decisions concerning level of care, focused research and information to patients. External validation is necessary before widespread use and is currently underway, an online tool has been developed for educational/experimental purposes ( https://hipfx.shinyapps.io/hipfx/ ).

Analysis of machine learning algorithmic models for the prediction of vital status at six months after hip fracture in patients older than 74 years

BACKGROUND AND OBJECTIVE

The objective is to develop a model that predicts vital status six months after fracture as accurately as possible. For this purpose we will use five different data sources obtained through the National Hip Fracture Registry, the Health Management Unit and the Economic Management Department.

MATERIAL AND METHODS

The study population is a cohort of patients over 74 years of age who suffered a hip fracture between May 2020 and December 2022. A warehouse is created from five different data sources with the necessary variables. An analysis of missing values and outliers as well as unbalanced classes of the target variable («vital status») is performed. Fourteen different algorithmic models are trained with the training. The model with the best performance is selected and a fine tuning is performed. Finally, the performance of the selected model is analyzed with test data.

RESULTS

A data warehouse is created with 502 patients and 144 variables. The best performing model is Linear Regression. Sixteen of the 24 cases of deceased patients are classified as live, and 14 live patients are classified as deceased. A sensitivity of 31%, an accuracy of 34% and an area under the curve of 0.65 is achieved.

CONCLUSIONS

We have not been able to generate a model for the prediction of six-month survival in the current cohort. However, we believe that the method used for the generation of algorithms based on machine learning can serve as a reference for future works.

Can we predict 1-year functional outcomes and mortality following hip fracture in middle-aged and geriatric patients at time of admission?

This study's purpose is to determine if patients treated for hip fracture at highest risk for poor functional outcomes, shorter time to death, and death within 1-year can be predicted at the time of admission. We hypothesized that the Score for Trauma Triage in the Geriatric and Middle-Aged (STTGMA) tool can be used to predict risk of these variables. Between February 2019-July 2020, 544 patients ≥ 55-years-old were treated for hip fracture [AO/OTA 31A/B, 32A/C]. Each patient's demographics, functional status, and injury details were used to calculate their respective risk (STTGMA) score at time of admission. Patients were divided into risk quartiles by STTGMA score. Patients were contacted by phone to complete EuroQol-5 Dimension (EQ5D-3L) questionnaires on functional status. Comparative analyses were conducted on outcomes and EQ5D-3L questionnaire results. 439 patients (80.7%) had at least 1-year follow-up. 82 patients (18.7%) died within 1-year after hospitalization. Mean STTGMA score was 1.67% ± 4.49%. The highest-risk cohort experienced a 42x (p < 0.01) and 2.5x (p = 0.01) increased rate of 1-year mortality compared to the minimal- and low-risk groups respectively. The highest-risk cohort had the shortest time to death (p = 0.015). The highest-risk cohort had the lowest EQ5D index (p < 0.01) and VAS scores (p < 0.01) along with the highest rate of 30 day readmission (p < 0.01) and the longest length of stay (p < 0.01). The STTGMA tool provides important prognostic information for middle-aged and geriatric hip fracture patients that can help modulate care levels. This information is useful when counseling patients, their families, and caregivers on expected outcomes.

Risk prediction models of mortality after hip fracture surgery in older individuals: a systematic review

OBJECTIVE

This study aimed to critically assess existing risk prediction models for postoperative mortality in older individuals with hip fractures, with the objective of offering substantive insights for their clinical application.

DESIGN

A comprehensive search was conducted across prominent databases, including PubMed, Embase, Cochrane Library, SinoMed, CNKI, VIP, and Wanfang, spanning original articles in both Chinese and English up until 1 December 2023. Two researchers independently extracted pertinent research characteristics, such as predictors, model performance metrics, and modeling methodologies. Additionally, the bias risk and applicability of the incorporated risk prediction models were systematically evaluated using the Prediction Model Risk of Bias Assessment Tool (PROBAST).

RESULTS

Within the purview of this investigation, a total of 21 studies were identified, constituting 21 original risk prediction models. The discriminatory capacity of the included risk prediction models, as denoted by the minimum and maximum areas under the subject operating characteristic curve, ranged from 0.710 to 0.964. Noteworthy predictors, recurrent across various models, included age, sex, comorbidities, and nutritional status. However, among the models assessed through the PROBAST framework, only one was deemed to exhibit a low risk of bias. Beyond this assessment, the principal limitations observed in risk prediction models pertain to deficiencies in data analysis, encompassing insufficient sample size and suboptimal handling of missing data.

CONCLUSION

Subsequent research endeavors should adopt more stringent experimental designs and employ advanced statistical methodologies in the construction of risk prediction models. Moreover, large-scale external validation studies are warranted to rigorously assess the generalizability and clinical utility of existing models, thereby enhancing their relevance as valuable clinical references.

Machine learning-based mortality prediction in hip fracture patients using biomarkers

The purpose of this retrospective study was to assess whether mortality following a hip fracture can be predicted by a machine learning model trained on basic blood and lab test data as well as basic demographic data. Additionally, the purpose was to identify the key variables most associated with 1-, 5-, and 10-year mortality and investigate their clinical significance. Input data included 3751 hip fracture patient records sourced from the Medical Information Mart for Intensive Care IV database, which provided records from in-hospital database systems at the Beth Israel Deaconess Medical Center. The 1-year mortality rate for all patients studied was 21% and for those aged 80+ was 29%. We assessed 10 different machine learning classification models, finding LightGBM to have the strongest 1-year mortality prediction performance, with accuracy of 81%, AUC of 0.79, sensitivity of 0.34, and specificity of 0.98 on the test set. The strongest-weighted features of the 1-year model included age, glucose, red blood cell distribution width, mean corpuscular hemoglobin concentration, white blood cells, urea nitrogen, prothrombin time, platelet count, calcium levels, and partial thromboplastin time. Most of these were also in the top 10 features of the LightGBM 5- and 10-year prediction models trained. Testing for these high-ranking biomarkers in new hip fracture patients can aid clinicians in assessing the likelihood of poor outcomes for hip fracture patients, and additional research can use these biomarkers to develop a mortality risk score.

Systematic review of machine-learning models in orthopaedic trauma

Aims

Machine-learning (ML) prediction models in orthopaedic trauma hold great promise in assisting clinicians in various tasks, such as personalized risk stratification. However, an overview of current applications and critical appraisal to peer-reviewed guidelines is lacking. The objectives of this study are to 1) provide an overview of current ML prediction models in orthopaedic trauma; 2) evaluate the completeness of reporting following the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement; and 3) assess the risk of bias following the Prediction model Risk Of Bias Assessment Tool (PROBAST) tool.

Methods

A systematic search screening 3,252 studies identified 45 ML-based prediction models in orthopaedic trauma up to January 2023. The TRIPOD statement assessed transparent reporting and the PROBAST tool the risk of bias.

Results

A total of 40 studies reported on training and internal validation; four studies performed both development and external validation, and one study performed only external validation. The most commonly reported outcomes were mortality (33%, 15/45) and length of hospital stay (9%, 4/45), and the majority of prediction models were developed in the hip fracture population (60%, 27/45). The overall median completeness for the TRIPOD statement was 62% (interquartile range 30 to 81%). The overall risk of bias in the PROBAST tool was low in 24% (11/45), high in 69% (31/45), and unclear in 7% (3/45) of the studies. High risk of bias was mainly due to analysis domain concerns including small datasets with low number of outcomes, complete-case analysis in case of missing data, and no reporting of performance measures.

Conclusion

The results of this study showed that despite a myriad of potential clinically useful applications, a substantial part of ML studies in orthopaedic trauma lack transparent reporting, and are at high risk of bias. These problems must be resolved by following established guidelines to instil confidence in ML models among patients and clinicians. Otherwise, there will remain a sizeable gap between the development of ML prediction models and their clinical application in our day-to-day orthopaedic trauma practice.

[Application and prospect of machine learning in orthopaedic trauma]

Objective

To review the current applications of machine learning in orthopaedic trauma and anticipate its future role in clinical practice.

Methods

A comprehensive literature review was conducted to assess the status of machine learning algorithms in orthopaedic trauma research, both nationally and internationally.

Results

The rapid advancement of computer data processing and the growing convergence of medicine and industry have led to the widespread utilization of artificial intelligence in healthcare. Currently, machine learning plays a significant role in orthopaedic trauma, demonstrating high performance and accuracy in various areas including fracture image recognition, diagnosis stratification, clinical decision-making, evaluation, perioperative considerations, and prognostic risk prediction. Nevertheless, challenges persist in the development and clinical implementation of machine learning. These include limited database samples, model interpretation difficulties, and universality and individualisation variations.

Conclusion

The expansion of clinical sample sizes and enhancements in algorithm performance hold significant promise for the extensive application of machine learning in supporting orthopaedic trauma diagnosis, guiding decision-making, devising individualized medical strategies, and optimizing the allocation of clinical resources.

Predictors of outcomes in geriatric patients with moderate traumatic brain injury after ground level falls

Introduction

The elderly population constitutes one of the fastest-growing demographic groups globally. Within this population, mild to moderate traumatic brain injuries (TBI) resulting from ground level falls (GLFs) are prevalent and pose significant challenges. Between 50 and 80% of TBIs in older individuals are due to GLFs. These incidents result in more severe outcomes and extended recovery periods for the elderly, even when controlling for injury severity. Given the increasing incidence of such injuries it becomes essential to identify the key factors that predict complications and in-hospital mortality. Therefore, the aim of this study was to pinpoint the top predictors of complications and in-hospital mortality in geriatric patients who have experienced a moderate TBI following a GLF.

Methods

Data were obtained from the American College of Surgeons' Trauma Quality Improvement Program database. A moderate TBI was defined as a head AIS ≤ 3 with a Glasgow Coma Scale (GCS) 9-13, and an AIS ≤ 2 in all other body regions. Potential predictors of complications and in-hospital mortality were included in a logistic regression model and ranked using the permutation importance method.

Results

A total of 7,489 patients with a moderate TBI were included in the final analyses. 6.5% suffered a complication and 6.2% died prior to discharge. The top five predictors of complications were the need for neurosurgical intervention, the Revised Cardiac Risk Index, coagulopathy, the spine abbreviated injury severity scale (AIS), and the injury severity score. The top five predictors of mortality were head AIS, age, GCS on admission, the need for neurosurgical intervention, and chronic obstructive pulmonary disease.

Conclusion

When predicting both complications and in-hospital mortality in geriatric patients who have suffered a moderate traumatic brain injury after a ground level fall, the most important factors to consider are the need for neurosurgical intervention, cardiac risk, and measures of injury severity. This may allow for better identification of at-risk patients, and at the same time resulting in a more equitable allocation of resources.

Machine learning applications for the prediction of extended length of stay in geriatric hip fracture patients

BACKGROUND

Geriatric hip fractures are one of the most common fractures in elderly individuals, and prolonged hospital stays increase the risk of death and complications. Machine learning (ML) has become prevalent in clinical data processing and predictive models. This study aims to develop ML models for predicting extended length of stay (eLOS) among geriatric patients with hip fractures and to identify the associated risk factors.

AIM

To develop ML models for predicting the eLOS among geriatric patients with hip fractures, identify associated risk factors, and compare the performance of each model.

METHODS

A retrospective study was conducted at a single orthopaedic trauma centre, enrolling all patients who underwent hip fracture surgery between January 2018 and December 2022. The study collected various patient characteristics, encompassing demographic data, general health status, injury-related data, laboratory examinations, surgery-related data, and length of stay. Features that exhibited significant differences in univariate analysis were integrated into the ML model establishment and subsequently cross-verified. The study compared the performance of the ML models and determined the risk factors for eLOS.

RESULTS

The study included 763 patients, with 380 experiencing eLOS. Among the models, the decision tree, random forest, and extreme Gradient Boosting models demonstrated the most robust performance. Notably, the artificial neural network model also exhibited impressive results. After cross-validation, the support vector machine and logistic regression models demonstrated superior performance. Predictors for eLOS included delayed surgery, D-dimer level, American Society of Anaesthesiologists (ASA) classification, type of surgery, and sex.

CONCLUSION

ML proved to be highly accurate in predicting the eLOS for geriatric patients with hip fractures. The identified key risk factors were delayed surgery, D-dimer level, ASA classification, type of surgery, and sex. This valuable information can aid clinicians in allocating resources more efficiently to meet patient demand effectively.

Application of artificial intelligence in trauma orthopedics: Limitation and prospects

The varieties and capabilities of artificial intelligence and machine learning in orthopedic surgery are extensively expanding. One promising method is neural networks, emphasizing big data and computer-based learning systems to develop a statistical fracture-detecting model. It derives patterns and rules from outstanding amounts of data to analyze the probabilities of different outcomes using new sets of similar data. The sensitivity and specificity of machine learning in detecting fractures vary from previous studies. AI may be most promising in the diagnosis of less-obvious fractures that are more commonly missed. Future studies are necessary to develop more accurate and effective detection models that can be used clinically.

Use of the Shizuoka Hip Fracture Prognostic Score (SHiPS) to Predict Long-Term Mortality in Patients With Hip Fracture in Japan: A Cohort Study Using the Shizuoka Kokuho Database

Hip fractures are common in patients of advanced age and are associated with excess mortality. Rapid and accurate prediction of the prognosis using information that can be easily obtained before surgery would be advantageous to clinical management. We performed a population-based retrospective cohort study using an 8.5-year Japanese claims database (April 2012-September 2020) to develop and validate a predictive model for long-term mortality after hip fracture. The study included 43,529 patients (34,499 [79.3%] women) aged ≥65 years with first-onset hip fracture. During the observation period, 43% of the patients died. Cox regression analysis identified the following prognostic predictors: sex, age, fracture site, nursing care certification, and several comorbidities (any malignancy, renal disease, congestive heart failure, chronic pulmonary disease, liver disease, metastatic solid tumor, and deficiency anemia). We then developed a scoring system called the Shizuoka Hip Fracture Prognostic Score (SHiPS); this system was established by scoring based on each hazard ratio and classifying the degree of mortality risk into four categories based on decision tree analysis. The area under the receiver operating characteristic (ROC) curve (AUC) (95% confidence interval [CI]) of 1-year, 3-year, and 5-year mortality based on the SHiPS was 0.718 (95% CI, 0.706-0.729), 0.736 (95% CI, 0.728-0.745), and 0.758 (95% CI, 0.747-0.769), respectively, indicating good predictive performance of the SHiPS for as long as 5 years after fracture onset. Even when the SHiPS was individually applied to patients with or without surgery after fracture, the prediction performance by the AUC was >0.7. These results indicate that the SHiPS can predict long-term mortality using preoperative information regardless of whether surgery is performed after hip fracture.

Artificial Intelligence for Hip Fracture Detection and Outcome Prediction: A Systematic Review and Meta-analysis

IMPORTANCE

Artificial intelligence (AI) enables powerful models for establishment of clinical diagnostic and prognostic tools for hip fractures; however the performance and potential impact of these newly developed algorithms are currently unknown.

OBJECTIVE

To evaluate the performance of AI algorithms designed to diagnose hip fractures on radiographs and predict postoperative clinical outcomes following hip fracture surgery relative to current practices.

DATA SOURCES

A systematic review of the literature was performed using the MEDLINE, Embase, and Cochrane Library databases for all articles published from database inception to January 23, 2023. A manual reference search of included articles was also undertaken to identify any additional relevant articles.

STUDY SELECTION

Studies developing machine learning (ML) models for the diagnosis of hip fractures from hip or pelvic radiographs or to predict any postoperative patient outcome following hip fracture surgery were included.

DATA EXTRACTION AND SYNTHESIS

This study followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses and was registered with PROSPERO. Eligible full-text articles were evaluated and relevant data extracted independently using a template data extraction form. For studies that predicted postoperative outcomes, the performance of traditional predictive statistical models, either multivariable logistic or linear regression, was recorded and compared with the performance of the best ML model on the same out-of-sample data set.

MAIN OUTCOMES AND MEASURES

Diagnostic accuracy of AI models was compared with the diagnostic accuracy of expert clinicians using odds ratios (ORs) with 95% CIs. Areas under the curve for postoperative outcome prediction between traditional statistical models (multivariable linear or logistic regression) and ML models were compared.

RESULTS

Of 39 studies that met all criteria and were included in this analysis, 18 (46.2%) used AI models to diagnose hip fractures on plain radiographs and 21 (53.8%) used AI models to predict patient outcomes following hip fracture surgery. A total of 39 598 plain radiographs and 714 939 hip fractures were used for training, validating, and testing ML models specific to diagnosis and postoperative outcome prediction, respectively. Mortality and length of hospital stay were the most predicted outcomes. On pooled data analysis, compared with clinicians, the OR for diagnostic error of ML models was 0.79 (95% CI, 0.48-1.31; P = .36; I2 = 60%) for hip fracture radiographs. For the ML models, the mean (SD) sensitivity was 89.3% (8.5%), specificity was 87.5% (9.9%), and F1 score was 0.90 (0.06). The mean area under the curve for mortality prediction was 0.84 with ML models compared with 0.79 for alternative controls (P = .09).

CONCLUSIONS AND RELEVANCE

The findings of this systematic review and meta-analysis suggest that the potential applications of AI to aid with diagnosis from hip radiographs are promising. The performance of AI in diagnosing hip fractures was comparable with that of expert radiologists and surgeons. However, current implementations of AI for outcome prediction do not seem to provide substantial benefit over traditional multivariable predictive statistics.

Mode of anesthesia is not associated with outcomes following emergency hip fracture surgery: a population-level cohort study

Background

Hip fractures often occur in frail patients with several comorbidities. In those undergoing emergency surgery, determining the optimal anesthesia modality may be challenging, with equipoise concerning outcomes following either spinal or general anesthesia. In this study, we investigated the association between mode of anesthesia and postoperative morbidity and mortality with subgroup analyses.

Methods

This is a retrospective study using all consecutive adult patients who underwent emergency hip fracture surgery in Orebro County, Sweden, between 2013 and 2017. Patients were extracted from the Swedish National Hip Fracture Registry, and their electronic medical records were reviewed. The association between the type of anesthesia and 30-day and 90-day postoperative mortality, as well as in-hospital severe complications (Clavien-Dindo classification ≥3a), was analyzed using Poisson regression models with robust SEs, while the association with 1-year mortality was analyzed using Cox proportional hazards models. All analyses were adjusted for potential confounders.

Results

A total of 2437 hip fracture cases were included in the study, of whom 60% received spinal anesthesia. There was no statistically significant difference in the risk of 30-day postoperative mortality (adjusted incident rate ratio (IRR) (95% CI): 0.99 (0.72 to 1.36), p=0.952), 90-day postoperative mortality (adjusted IRR (95% CI): 0.88 (0.70 to 1.11), p=0.281), 1-year postoperative mortality (adjusted HR (95% CI): 0.98 (0.83 to 1.15), p=0.773), or in-hospital severe complications (adjusted IRR (95% CI): 1.24 (0.85 to 1.82), p=0.273), when comparing general and spinal anesthesia.

Conclusions

Mode of anesthesia during emergency hip fracture surgery was not associated with an increased risk of postoperative mortality or in-hospital severe complications in the study population or any of the investigated subgroups.

Level of evidence: Therapeutic/Care Management, level III

A New Nomogram Model for Predicting 1-Year All-Cause Mortality After Hip Arthroplasty in Nonagenarians With Hip Fractures: A 20-Year Period Retrospective Cohort Study

Background

China has become an ageing society and as it continues to age, it will face an increasing number of hip fractures in nonagenarians. However, few preoperative assessment tools to determine the postoperative mortality risk in nonagenarians with hip fracture were available. The aim of this study was to identify all-cause mortality risk factors after hip arthroplasty in nonagenarians with hip fractures and to establish a new nomogram model to optimize the individualized hip arthroplasty in nonagenarians with hip fractures.

Methods

We retrospectively studied 246 consecutive nonagenarians diagnosed with hip fracture from August 2002 to February 2021 at our center. During the follow-up, 203 nonagenarians with a median age of 91.9 years treated with hip arthroplasty were included, of which 136 were females and 67 were males, and 43 nonagenarians were excluded (40 underwent internal fixation and 3 were lost to follow-up). The full cohort was randomly divided into training (50%) and validation (50%) sets. The potential predictive factors for 1-year all-cause mortality after hip arthroplasty were assessed by univariate and multivariate COX proportional hazards regression on the training set, and then, a new nomogram model was established and evaluated by concordance index (C-index) and calibration curves.

Results

After analyzing 44 perioperative variables including demographic characteristics, vital signs, surgical data, laboratory tests, we identified that age-adjusted Charlson Comorbidity Index (aCCI) (p = 0.042), American Society of Anesthesiologists (ASA) classification (p = 0.007), Urea (p = 0.028), serum Ca2+ (p = 0.011), postoperative hemoglobin (p = 0.024) were significant predictors for 1-year all-cause mortality after hip arthroplasty in the training set. The nomogram showed a robust discrimination, with a C-index of 0.71 (95%CIs, 0.68–0.78). The calibration curves for 1-year all-cause mortality showed optimal agreement between the probability as predicted by the nomogram and the actual probability in training and validation sets.

Conclusion

A novel nomogram model integrating 5 independent predictive variables were established and validated. It can effectively predict 1-year all-cause mortality after hip arthroplasty in nonagenarians with hip fracture and lead to a more optimized and rational therapeutic choice.

Development and internal validation of a machine-learning-developed model for predicting 1-year mortality after fragility hip fracture

Background

Fragility hip fracture increases morbidity and mortality in older adult patients, especially within the first year. Identification of patients at high risk of death facilitates modification of associated perioperative factors that can reduce mortality. Various machine learning algorithms have been developed and are widely used in healthcare research, particularly for mortality prediction. This study aimed to develop and internally validate 7 machine learning models to predict 1-year mortality after fragility hip fracture.

Methods

This retrospective study included patients with fragility hip fractures from a single center (Siriraj Hospital, Bangkok, Thailand) from July 2016 to October 2018. A total of 492 patients were enrolled. They were randomly categorized into a training group (344 cases, 70%) or a testing group (148 cases, 30%). Various machine learning techniques were used: the Gradient Boosting Classifier (GB), Random Forests Classifier (RF), Artificial Neural Network Classifier (ANN), Logistic Regression Classifier (LR), Naive Bayes Classifier (NB), Support Vector Machine Classifier (SVM), and K-Nearest Neighbors Classifier (KNN). All models were internally validated by evaluating their performance and the area under a receiver operating characteristic curve (AUC).

Results

For the testing dataset, the accuracies were GB model = 0.93, RF model = 0.95, ANN model = 0.94, LR model = 0.91, NB model = 0.89, SVM model = 0.90, and KNN model = 0.90. All models achieved high AUCs that ranged between 0.81 and 0.99. The RF model also provided a negative predictive value of 0.96, a positive predictive value of 0.93, a specificity of 0.99, and a sensitivity of 0.68.

Conclusions

Our machine learning approach facilitated the successful development of an accurate model to predict 1-year mortality after fragility hip fracture. Several machine learning algorithms (eg, Gradient Boosting and Random Forest) had the potential to provide high predictive performance based on the clinical parameters of each patient. The web application is available at www.hipprediction.com. External validation in a larger group of patients or in different hospital settings is warranted to evaluate the clinical utility of this tool.

Trial registration

Thai Clinical Trials Registry (22 February 2021; reg. no. TCTR20210222003).

Supplementary Information

The online version contains supplementary material available at 10.1186/s12877-022-03152-x.

Dementia is a surrogate for frailty in hip fracture mortality prediction

Purpose

Among hip fracture patients both dementia and frailty are particularly prevalent. The aim of the current study was to determine if dementia functions as a surrogate for frailty, or if it confers additional information as a comorbidity when predicting postoperative mortality after a hip fracture.

Methods

All adult patients who suffered a traumatic hip fracture in Sweden between January 1, 2008 and December 31, 2017 were considered for inclusion. Pathological fractures, non-operatively treated fractures, reoperations, and patients missing data were excluded. Logistic regression (LR) models were fitted, one including and one excluding measurements of frailty, with postoperative mortality as the response variable. The primary outcome of interest was 30-day postoperative mortality. The relative importance for all variables was determined using the permutation importance. New LR models were constructed using the top ten most important variables. The area under the receiver-operating characteristic curve (AUC) was used to compare the predictive ability of these models.

Results

121,305 patients were included in the study. Initially, dementia was among the top ten most important variables for predicting 30-day mortality. When measurements of frailty were included, dementia was replaced in relative importance by the ability to walk alone outdoors and institutionalization. There was no significant difference in the predictive ability of the models fitted using the top ten most important variables when comparing those that included [AUC for 30-day mortality (95% CI): 0.82 (0.81–0.82)] and excluded [AUC for 30-day mortality (95% CI): 0.81 (0.80–0.81)] measurements of frailty.

Conclusion

Dementia functions as a surrogate for frailty when predicting mortality up to one year after hip fracture surgery. The presence of dementia in a patient without frailty does not appreciably contribute to the prediction of postoperative mortality.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00068-022-01960-9.

Predictors of adverse events after percutaneous pedicle screws fixation in patients with single-segment thoracolumbar burst fractures

BACKGROUND

Percutaneous pedicle screw fixation (PPSF) is the primary approach for single-segment thoracolumbar burst fractures (TLBF). The healing angle at the thoracolumbar junction is one of the most significant criteria for evaluating the efficacy of PPSF. Therefore, the purpose of this study was to analyze the predictors associated with the poor postoperative alignment of the thoracolumbar region from routine variables using a support vector machine (SVM) model.

METHODS

We retrospectively analyzed patients with TLBF operated at our academic institute between March 1, 2014 and December 31, 2019. Stepwise logistic regression analysis was performed to assess potential statistical differences between all clinical and radiological variables and the adverse events. Based on multivariate logistic results, a series of independent risk factors were fed into the SVM model. Meanwhile, the feature importance of radiologic outcome for each parameter was explored. The predictive performance of the SVM classifier was evaluated using the area under the receiver operating characteristic curve (AUC), accuracy (ACC) and confusion matrices with 10-fold cross-validation, respectively.

RESULTS

In the recruited 150 TLBFs, unfavorable radiological outcomes were observed in 53 patients (35.33%). The relationship between osteoporosis (p = 0.036), preoperative Cobb angle (p = 0.001), immediate postoperative Cobb angle (p = 0.029), surgically corrected Cobb angle (p = 0.001), intervertebral disc injury (Score 2 p = 0.001, Score 3 p = 0.001), interpedicular distance (IPD) (p = 0.001), vertebral body compression rate (VBCR) (p = 0.010) and adverse events was confirmed by univariate regression. Thereafter, independent risk factors including preoperative Cobb angle, the disc status and IPD and independent protective factors surgical correction angle were identified by multivariable logistic regression. The established SVM classifier demonstrated favorable predictive performance with the best AUC = 0.93, average AUC = 0.88, and average ACC = 0.87. The variables associated with radiological outcomes, in order of correlation strength, were intervertebral disc injury (42%), surgically corrected Cobb angle (25%), preoperative Cobb angle (18%), and IPD (15%). The confusion matrix reveals the classification results of the discriminant analysis.

CONCLUSIONS

Critical radiographic indicators and surgical purposes were confirmed to be associated with an unfavorable radiographic outcome of TLBF. This SVM model demonstrated good predictive ability for endpoints in terms of adverse events in patients after PPSF surgery.

Association between General Anesthesia and the Occurrence of Cerebrovascular Accidents in Hip Fracture Patients

Background

General anesthesia is an important factor leading to postoperative complications, and cerebrovascular accidents take the first place in the causes of postoperative death. Therefore, it is extremely important to explore the correlation between general anesthesia and the occurrence of cerebrovascular accidents in hip fracture patients.

Objective

To explore the association between general anesthesia and the occurrence of cerebrovascular accidents in hip fracture patients.

Methods

The data of 240 hip fracture patients treated in our hospital from February 2017 to February 2021 were retrospectively analyzed, and the patients were divided into the general anesthesia group (n = 120) and nongeneral anesthesia group (n = 120) according to whether or not they received general anesthesia, so as to compare their incidence rate of cerebrovascular accidents between the two groups, record their hemodynamic changes, and analyze the association between various risk factors under general anesthesia and the occurrence of cerebrovascular accidents.

Results

No statistical differences in patients' general information such as age and gender between the two groups were observed (P > 0.05); compared with the nongeneral anesthesia group, the incidence rate of cerebrovascular accidents was significantly higher in the general anesthesia group (P < 0.001); between the two groups, the heart rates and mean arterial pressure (MAP) at 15 min after anesthesia, at the time of skin incision, and 15 min before the end of surgery were significantly different (P < 0.05); according to the multiple logistic regression analysis, general anesthesia was a risk factor affecting the occurrence of cerebrovascular accidents in hip fracture patients, and under general anesthesia, age ≥80 years, BMI ≥23 kg/m², types of anesthetic drugs ≥4, intraoperative blood pressure ≥140 mmHg, and intraoperative heart rate ≥80 bpm were also regarded as the risk factors.

Conclusion

General anesthesia is a risk factor affecting the occurrence of cerebrovascular accidents in hip fracture patients, and under general anesthesia, age ≥80 years, BMI ≥23 kg/m², types of anesthetic drugs ≥4, intraoperative blood pressure ≥140 mmHg, and intraoperative heart rate ≥80 bpm will further increase the possibility of cerebrovascular accidents.