Accuracy of the Physiological and Operative Severity Score for the enUmeration of Mortality and morbidity score and the Nottingham risk score in hip fracture patients in Sweden - A prospective observational study

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/ ).

Predictive value of the systemic immune-inflammation index on one-year mortality in geriatric hip fractures

BACKGROUND

Geriatric hip fractures are associated with a high incidence of mortality. This study examines the predictive value of the systemic immune-inflammation index (SII) on one-year mortality in elderly hip fracture patients.

METHODS

A single-center retrospective study was conducted between February 2017 and October 2020. Three hundred and eleven surgically treated consecutive hip fracture patients were included in the study. Admission, postoperative first day, and postoperative fifth-day SII values were calculated. The receiver operating characteristic (ROC) curve was used to calculate the cut-off values, and patients were divided into high and low groups according to these cut-off values. After univariate Cox regression analysis, significant factors were included in the multivariate Cox proportional hazards model to adjust the effect of covariates and explore independent predictive factors associated with mortality. Further subgroup analysis was performed to evaluate the accuracy of the results for different clinical and biological characteristics.

RESULTS

The mean age was 80.7 ± 8.0 years, and women made up the majority (67.8%) of the patients. The one-year mortality rate was 28.0%. After univariate and multivariate analyses, high postoperative fifth-day SII remained an independent predictor of one-year mortality (adjusted HR 2.16, 95% CI 1.38-3.38, p = 0.001). Older age, male gender, Charlson comorbidity index (CCI) ≥ 2, and hypoalbuminemia were found to be other independent predictors. The optimal cut-off value of the postoperative fifth-day SII was calculated at 1751.9 units (p < 0.001).

CONCLUSION

The postoperative fifth-day SII is a simple and useful inflammatory biomarker for predicting one-year mortality in patients with hip fracture.

Predictive Value of Machine Learning Models in Postoperative Mortality of Older Adults Patients with Hip Fracture: A Systematic Review and Meta-analysis

BACKGROUND

Some researchers have used machine learning to predict mortality in old patients with hip fracture, but its application value lacks an evidence-based basis. Hence, we conducted this meta-analysis to explore the predictive accuracy of machine learning for mortality in old patients with hip fracture.

METHODS

We systematically retrieved PubMed, Cochrane, Embase, and Web of Science for relevant studies published before July 15, 2022. The PROBAST assessment tool was used to assess the risk of bias in the included studies. A random-effects model was used for the meta-analysis of C-index, whereas a bivariate mixed-effects model was used for the meta-analysis of sensitivity and specificity. The meta-analysis was performed on R and Stata.

RESULTS

Eighteen studies were included, involving 8 machine learning models and 398,422 old patients undergoing hip joint surgery, of whom 60,457 died. According to the meta-analysis, the pooled C-index for machine learning models was 0.762 (95% CI: 0.691 ∼ 0.833) in the training set and 0.838 (95% CI: 0.783 ∼ 0.892) in the validation set, which is better than the C-index of the main clinical scale (Nottingham Hip Fracture Score), that is, 0.702 (95% CI: 0.681 ∼ 0.723). Among different machine learning models, ANN and Bayesian belief network had the best predictive performance.

CONCLUSION

Machine learning models are more accurate in predicting mortality in old patients after hip joint surgery than current mainstream clinical scoring systems. Subsequent research could focus on updating clinical scoring systems and improving their predictive performance by relying on machine learning models.

O-POSSUM and P-POSSUM as predictors of morbidity and mortality in older patients after hip fracture surgery: a meta-analysis

BACKGROUND

The POSSUM model has been widely used to predict morbidity and mortality after general surgery. Modified versions known as O-POSSUM and P-POSSUM have been used extensively in orthopedic surgery, but their accuracy is unclear. This systematic review evaluated the predictive value of these models in older patients with hip fractures.

METHODS

This study was performed and reported based on the "Preferred reporting items for systematic reviews and meta-analyses" guidelines. PubMed, Cochrane, EMBASE, and Web of Science were comprehensively searched for relevant studies, whose methodological quality was evaluated according to the "Methodological index for non-randomized studies" scale. Revman 5 was used to calculate weighted ratios of observed to expected morbidity or mortality.

RESULTS

The meta-analysis included 10 studies, of which nine (2549 patients) assessed the ability of O-POSSUM to predict postoperative morbidity, nine (3649 patients) assessed the ability of O-POSSUM to predict postoperative mortality, and four (1794 patients) assessed the ability of P-POSSUM to predict postoperative mortality. The corresponding weighted ratios of observed to expected morbidity or mortality were 0.84 (95% CI 0.70-1.00), 0.68 (95% CI 0.49-0.95), and 0.61 (95% CI 0.16-2.38).

CONCLUSIONS

While O-POSSUM shows reasonable accuracy in predicting postoperative morbidity in older patients with hip fractures, both P-POSSUM and O-POSSUM substantially overestimate postoperative mortality. The POSSUM model should be optimized further for this patient population.

Risk scoring models for patients with proximal femur fractures: Qualitative systematic review assessing 30-day mortality and ease of use

BACKGROUND

Hip fractures are a common orthopaedic injury affecting a particularly frail and vulnerable patient cohort. They are at risk of many complications, including prolonged length of stay and mortality. Efforts to identify those at high risk may be beneficial. Over 25 risk prediction models are published for patients with hip fractures.

AIM

The primary aim of this study was to assess the performance of predictor scores in predicting 30-day mortality. The secondary aim was to assess the ease of use of these systems.

METHODS

A qualitative systematic review was performed. A search was conducted on online databases, including PubMed, CINAHL, Clinical Trials.gov, Cochrane, DARE, EMBASE, SCOPUS, and Web of Science.. The terms fragility hip fractures and risk prediction models were utilised while performing the search. These were then expanded using Boolean operators and similar terms. Search results were imported to Covidence. Primary observational studies using one or more hip fracture mortality prediction models and 30-day mortality as an outcome were included. Systematic reviews and studies on specific patient groups defined other medical conditions (e.g. COVID positive or dialysis patients) were excluded.

RESULTS

3,101 studies were screened following duplicate removal. 34 papers were included in the review, in which 23 scoring systems were reported. Six of these were pre-operative and reported in multiple studies. Most demonstrated appropriate fit and fair discrimination. Five of the six pre-operative scoring systems examined, displayed appropriate ease of use, allowing risk calculation at the time of admission.

CONCLUSION

Nottingham Hip Fracture Score remains the most extensive reported scoring system and performs fair overall with AUROCs of 0.64-0.80 and good fit in calibration across all studies. However, new systems utilise many similar predictors. There is a need for the standardisation of publications on scoring systems to allow further systematic review and meta-analyses.

Validation and updating of the Older Person's Emergency Risk Assessment (OPERA) score to predict outcomes for hip fracture patients

INTRODUCTION

Hip fractures are associated with significant morbidity and mortality in older people. Accurate risk stratification is important for planning of care, informed decision-making and communication with patients and relatives. The Older Persons' Emergency Risk Assessment (OPERA) score is a risk stratification score for older people admitted to hospital. Our aims were to validate OPERA in hip fracture patients, update the score and compare performance with the Nottingham Hip Fracture Score (NHFS).

METHODS

This dual-centre 3-year observational study (2016-2018) included acutely admitted hip fracture patients managed surgically aged ⩾65 years. The primary outcome was 30-day mortality. Secondary outcomes included residence at 120 days and 1-year mortality. Model performance was assessed using area under the curve (AUC) analysis and Brier scores (discrimination) and calibration curves. The OPERA score was updated using regression analysis with additional independent predictors and validated using bootstrap analysis.

RESULTS

2142 patients (median age 86 [80-91] years) were included with a 30-day mortality of 5.2% and a 1-year mortality of 31.4%. 30-day mortality AUC for OPERA was 0.75 (95% CI, 0.73-0.77) and for NHFS 0.68 (0.65-0.70). For 1-year mortality AUC for OPERA was 0.74 (0.73-0.75) and for NHFS 0.70 (0.69-0.71). The OPERA Score was updated to Hip-OPERA, including ASA grade. Hip-OPERA demonstrated an AUC for 30-day mortality of 0.77 (0.73-0.81) and an AUC for 1-year mortality of 0.76 (0.75-0.77). AUC for new residential care status at 120 days was 0.79 (0.78-0.80).

CONCLUSIONS

Hip-OPERA demonstrated superior discrimination to the NHFS and OPERA for 30-day mortality, 1-year mortality and residence at 120 days following hip fracture. External validation is desirable.

Validation of the Nottingham Hip Fracture Score in Predicting Postoperative Outcomes Following Hip Fracture Surgery

OBJECTIVE

Although several prognostic models have been developed for patients who underwent hip fracture surgery, their preoperative performance was insufficiently validated. We aimed to verify the effectiveness of the Nottingham Hip Fracture Score (NHFS) for predicting postoperative outcomes following hip fracture surgery.

METHODS

This was a single-center and retrospective analysis. A total of 702 elderly patients with hip fractures (age ≥ 65 years old) who received treatment in our hospital from June 2020 to August 2021 were selected as the research participants. They were divided into the survival group and the death group based on their survival 30 days after surgery. The multivariate logistic regression model was used to identify the independent risk factors for the 30-day mortality after surgery. The NHFS and American Society of Anaesthesiologists (ASA) grades were used to construct these models, and a receiver operating characteristic curve was plotted to assess their diagnostic significance. A correlation analysis was performed between NHFS and length of hospitalization and mobility 3 months after surgery.

RESULTS

There were significant differences in the age, albumin level, NHFS, and ASA grade between both groups (p < 0.05). The length of hospitalization in the death group was longer than the survival group (p < 0.05). The perioperative blood transfusion and postoperative ICU transfer rates in the death group were higher than in the survival group (p < 0.05). The death group's incidence of pulmonary infections, urinary tract infections, cardiovascular events, pressure ulcers, stress ulcers with bleeding, and intestinal obstruction was higher than the survival group (p < 0.05). The NHFS and ASA III were independent risk factors for the 30-day mortality after surgery, regardless of age and albumin level (p < 0.05). The area under the curve (AUC) of the NHFS and ASA grade for predicting the 30-day mortality after surgery was 0.791 (95% confidence interval [CI] 0.709-0.873, p < 0.05) and 0.621 (95% CI 0.477-0.764, p > 0.05), respectively. The NHFS positively correlated with hospitalization length and mobility grade 3 months after surgery (p < 0.05).

CONCLUSION

The NHFS demonstrated a better predictive performance than the ASA score for the 30-day mortality after surgery and positively correlated with the hospitalization length and postoperative activity limitation in elderly patients with hip fractures.

Influence of time to surgery on clinical outcomes in elderly hip fracture patients : an assessment of surgical postponement due to non-medical reasons

AIMS

Factors associated with high mortality rates in geriatric hip fracture patients are frequently unmodifiable. Time to surgery, however, might be a modifiable factor of interest to optimize clinical outcomes after hip fracture surgery. This study aims to determine the influence of postponement of surgery due to non-medical reasons on clinical outcomes in acute hip fracture surgery.

METHODS

This observational cohort study enrolled consecutively admitted patients with a proximal femoral fracture, for which surgery was performed between 1 January 2018 and 11 January 2021 in two level II trauma teaching hospitals. Patients with medical indications to postpone surgery were excluded. A total of 1,803 patients were included, of whom 1,428 had surgery < 24 hours and 375 had surgery ≥ 24 hours after admission.

RESULTS

Prolonged total length of stay was found when surgery was performed ≥ 24 hours (median 6 days (interquartile range (IQR) 4 to 9) vs 7 days (IQR 5 to 10); p = 0.001) after admission. No differences in postoperative length of hospital stay nor in 30-day mortality rates were found. In subgroup analysis for time frames of 12 hours each, pressure sores and urinary tract infections were diagnosed more frequently when time to surgery increased.

CONCLUSION

Longer time to surgery due to non-medical reasons was associated with a higher incidence of postoperative pressure sores and urinary tract infections when time to surgery was more than 48 hours after admission. No association was found between time to surgery and 30-day mortality rates or postoperative length of hospital stay.Cite this article: Bone Joint J 2022;104-B(12):1369-1378.

Inter-rater agreement in pPOSSUM scores of geriatric trauma patients: a prospective evaluation

PURPOSE

Risk prediction models are widely used in the perioperative setting to identify high-risk patients who may benefit from additional care and to aid clinical decision-making. pPOSSUM is such a prediction model, however, little is known about the inter-rater agreement when scoring subjective parameters. This study assessed the inter-rater agreement between clinicians of different specialties and work-level when scoring 30 clinical case reports of geriatric hip fracture patients with pPOSSUM.

METHODS

Eighteen clinicians of the department of Surgery (three specialists, four residents), Anaesthesiology (four specialists, two residents) and Emergency Medicine (three specialists, two residents) who were familiar with the pPOSSUM scoring system were asked to calculate the scores. The kappa statistic and the statistical method of Fleiss were used to analyse inter-rater agreement.

RESULTS

The response rate was 100%. Among surgeons, Anaesthesiologists and Emergency department doctors (ED), the overall mean kappa values were 0.42, 0.08 and 0.20, respectively. Among surgery, anaesthesiology and ED residents the overall mean kappa values were 0.21, 0.33 and 0.37, respectively. Within the department of Surgery, Anaesthesiology and Emergency Medicine the overall mean kappa values were 0.23, 0.12 and 0.22, respectively. An overall mean kappa value of 0.19 was seen among all specialists. All residents had an overall mean kappa value of 0.21 and all clinicians had an overall mean kappa value of 0.21.

CONCLUSION

The overall inter-rater agreement of clinicians and interdisciplinary agreement when scoring geriatric hip fracture patients with pPOSSUM was low and prone to subjectivity in our study. A higher work-experience level did not lead to better agreement. When pPOSSUM is calculated without clinical assessment by the same clinician, caution is advised to prevent over-reliance on the pPOSSUM risk prediction model.

LEVEL OF EVIDENCE

III.

Derivation and external validation of a 30-day mortality risk prediction model for older patients having emergency general surgery

BACKGROUND

Older people (≥65 yr) are at increased risk of morbidity and mortality after emergency general surgery. Risk prediction models are needed to guide decision making in this high-risk population. Existing models have substantial limitations and lack external validation, potentially limiting their applicability in clinical use. We aimed to derive and validate, both internally and externally, a multivariable model to predict 30-day mortality risk in older patients undergoing emergency general surgery.

METHODS

After protocol publication, we used the National Surgical Quality Improvement Program (NSQIP) database (2012-6; estimated to contain 90% data from the USA and 10% from Canada) to derive and internally validate a model to predict 30-day mortality for older people having emergency general surgery using logistic regression with elastic net regularisation. Internal validation was done with 10-fold cross-validation. External validation was done using a temporally separate health administrative database exclusively from Ontario, Canada.

RESULTS

Overall, 6012 (12.0%) of the 50 221 patients died within 30 days. The model demonstrated strong discrimination (area under the curve [AUC]=0.871) and calibration across the spectrum of observed and predicted risks. Ten-fold internal cross-validation demonstrated minimal optimism (AUC=0.851, optimism 0.019 [standard deviation=0.06]) with excellent calibration. External validation demonstrated lower discrimination (AUC=0.700) and degraded calibration.

CONCLUSION

A multivariable mortality risk prediction model was strongly discriminative and well calibrated internally. However, poor external validation suggests the model may not be generalisable to non-NSQIP data and hospitals. The findings highlight the importance of external validation before clinical application of risk models.

Application of POSSUM and P-POSSUM scores in the risk assessment of elderly hip fracture surgery: systematic review and meta-analysis

Background

Since Mohamed et al. analyzed 2326 orthopedic cases in 2002 and believed that the POSSUM formula can be directly used to predict postoperative morbidity and mortality in orthopedic patients, applications of POSSUM and P-POSSUM scores in the hip fracture surgery have been mostly reported in the field of orthopedics, but there are still some inconsistencies in the related reports.

Methods

The electronic library was searched for all literature that met the purpose from its inception to 2021. Relative risk (RR) was selected to evaluate whether the model could be used to assess the risk of surgery in patients with elderly hip fractures. Finally, sensitivity analyses and subgroup analyses were performed.

Results

Thirteen studies were finally included, including 9 retrospective and 4 prospective studies.The morbidity analysis includes 11 studies, and the result was RR = 1.07 (95% CI 0.93–1.24), The mortality analysis includes 11 studies on POSSUM and 5 studies on P-POSSUM. The results of mortality by POSSUM and by P-POSSUM were RR = 1.93 (95% CI 1.21–3.08) and RR = 1.15 (95% CI 0.89–1.50), respectively. POSSUM had more accuracy to predict mortality for sample < 200 subgroup(RR = 2.45; 95% CI 0.71–8.42) than sample > 200 subgroup(RR = 1.59; 95% CI 1.06–2.40), and in the subgroup of hip fractures that did not distinguish between specific fracture types(RR = 1.69, 95% CI 0.87–3.32) than intertrochanteric neck fracture subgroup(RR = 5.04, 95% CI 1.07–23.75) and femoral femoral fracture subgroup(RR = 1.43,95% CI 1.10–1.84).

Conclusion

POSSUM can be used to predict morbidity in elderly hip fractures. The P-POSSUM was more accurate in predicting mortality in elderly hip fracture patients compared to the POSSUM, whose predictive value for mortality was influenced by the sample size and type of fracture studied. In addition, we believe that appropriate improvements to the POSSUM system are needed to address the characteristics of orthopedic surgery.

Validation of the Nottingham Hip Fracture Score (NHFS) for the prediction of 30-day mortality in a Swedish cohort of hip fractures

BACKGROUND

Hip fracture is a common osteoporotic fracture with great morbidity and mortality. The utility of ASA classification is limited, as most patients are ≥ASA 3. A reliable predictor of mortality risk could support decision-making. We aimed to evaluate Nottingham hip fracture score (NHFS) for the prediction of 30-day mortality and then to recalibrate the formula converting NHFS to risk of 30-day mortality.

METHODS

All patients >60 years with surgically treated hip fracture surgery during 2015-16 were assessed. Data was extracted manually from routinely collected clinical data in registry and medical records. Discriminative performance of NHFS and ASA was assessed with C-statistics. The conversion formula from NHFS to risk of 30-day mortality was recalibrated using logistic binominal regression. Observed vs expected ratios of 30-day mortality were compared with the 2012 NHFS-formula and recalibration was performed in a split dataset.

RESULTS

1864 patients were included, with 213 deaths within 30 days. C-statistic were 0.64 for NHFS and 0.62 for ASA. Comparing expected values from the 2012-revision with our observed deaths gave a ratio of 1.37. Relating predicted levels of 30-day mortality based on 70% of our cohort vs. 30% test portion of our Swedish dataset gave a ratio of 0.97.

DISCUSSION

NHFS underestimated mortality in our cohort and showed poor discrimination. Revision of the formula based on a split dataset improved calibration. We suggest NHFS to be routinely implemented to support clinical judgements, expand preoperative assessment and escalate intraoperative monitoring.

A new preoperative risk score for predicting mortality of elderly hip fracture patients: an external validation study

BACKGROUND

Hip fractures are common in the elderly and have a high risk of mortality. Several risk prediction models for mortality of hip fracture have been developed, but most of them are difficult to apply accurately in clinical practice.

AIMS

The objective of the present study was to perform an external validation of a new published preoperative risk score for predicting mortality.

METHODS

We carried out a retrospective cohort study from January 2014 to December 2018 for elderly hip fracture patients discharged from a orthopedic center in China. The preoperative risk score was calculated for each patient, and further divided into two groups: low-risk group (score < 24 points) and high-risk group (score ≥ 24 points) using the receiver operating characteristic (ROC) curve. The outcome was 30-day, 6-month and 1-year all-cause mortality, and the relationship between the risk score and mortality was assessed by univariate and multivariate Cox proportional hazard models. The area under the curve (AUC), Hosmer-Lemeshow test and calibration plots were used to test the discrimination and calibration.

RESULTS

A total of 460 consecutive patients were included in the study, and high-risk score was an independent risk factor for 30-day mortality [Hazard ratio (HR) 6.70; 95% Confidence interval (CI) 1.82-24.69; p = 0.004], 6-month mortality (HR 2.94; 95% CI 1.68-5.17; p < 0.001) and 1-year mortality (HR 3.30; 95% CI 2.09-5.20; p < 0.001). Also, each point increase in the risk score resulted in a 11% increase in 30-day mortality (HR 1.11; 95% CI 1.07-1.16; p < 0.001), 6% increase in 6-month mortality (HR 1.06; 95% CI 1.04-1.09; p < 0.001), and 5% increase in 1-year mortality (HR 1.05; 95% CI 1.03-1.07; p < 0.001). Moreover, the risk score had an AUC of 0.89 (95% CI 0.80-0.98) for 30-day mortality, 0.77 (95% CI 0.70-0.83) for 6-month mortality, and 0.76 (95% CI 0.70-0.81) for 1-year mortality. Calibration plots showed a good calibration between observed and predicted mortality, which was also demonstrated by the Hosmer-Lemeshow test.

CONCLUSION

Our present study findings indicated that the preoperative risk score was an accurate mortality risk assessment tool for elderly hip fracture patients, regardless of short- and long-term follow-up.

Does preoperative hemodynamic preconditioning improve morbidity and mortality after traumatic hip fracture in geriatric patients? A retrospective cohort study

INTRODUCTION

Dehydration is a major problem in the older population with traumatic hip fractures (THF). A preoperative hemodynamic preconditioning (PHP) protocol may help in achieving hemodynamic stability to ensure adequate perfusion and oxygenation using only clinical parameters to assess cardiovascular performance.

MATERIALS AND METHODS

A single-centre retrospective study in geriatric trauma patients was conducted in a Level 1 Trauma Centre in Switzerland. Patients over the age of 70 with THFs and with Portsmouth Physiological and Operative Severity Score for the enUmeration of Mortality (P-POSSUM) scores ≥ 5% who underwent surgical treatment between February 2015 and October 2017 were included. It was hypothesized that patients whose hemodynamic stability was optimised before surgery would have fewer complications and reduced mortality postoperatively. Primary outcomes were complications and mortality. Secondary outcomes were hospital length of stay (HLOS) and place of discharge.

RESULTS

100 patients were included in the PHP group and 79 patients were included in the non-PHP group. The median age was 86.5 (82-90) in the PHP group and 86 (82-90) in the non-PHP group. Patients who had been treated according to the PHP protocol showed a significant reduction in mortality at 30 days (p = 0.02). The PHP group showed an 8.1 and 3.5% reduced mortality at 90 days and at 1 year, respectively. The PHP group showed an 11.7% reduction of patients with complicated courses. No significant differences were seen in HLOS and discharge disposition.

CONCLUSIONS

The PHP group showed a significant reduction in short-term mortality, a reduction in long-term mortality, and a reduction in the number of patients with complicated courses. The PHP protocol is a safe, strictly regulated, non-invasive fluid resuscitation protocol for the optimization of geriatric patients with a THF that requires minimal effort.

LEVEL OF EVIDENCE

Level III, therapeutic.

Validation of the Nottingham Hip Fracture Score (NHFS) in Patients with Hip Fracture: A Prospective Cohort Study in the Netherlands

Introduction

The Nottingham Hip Fracture Score (NHFS) was developed to predict 30-day mortality for patients with hip fracture. This study aimed to validate the NHFS in a cohort with sufficient statistical power.

Methods

Data were extracted from a prospective hip-fracture database (FAMMI). Patients were included between January 1, 2018 and January 11, 2021. All consecutively admitted patients ≥18 years of age with a hip fracture (ie, femoral neck fracture, intertrochanteric fracture, and subtrochanteric fracture) were included. Mann–Whitney’s U values were calculated to find potential miscalibration of the NHFS formula. Discrimination evaluation was performed using the concordance statistic as an equivalent to area under the receiver-operating curve.

Results

In total, 2,458 patients were included. Mean age was 80±12 years, and 66% were women (n=1,631). Median NHFS was 5 (4–6) and overall 30-day mortality 7.9% (n=195). Overall goodness of fit was tested with Pearson’s ?² (11.8, df 10; P=0.297). No statistically significant signs of miscalibration were found (Mann–Whitney U, P=0.08). Discrimination was tested with area under the receiver- operating curve, which was 72.1% (95% CI 68.7%–75.4%). However, observed 30-day mortality in our population of hip-fracture patients was slightly higher than the NHFS prediction.

Conclusion

The NHFS seemed to predict 30-day mortality with reasonable accuracy for patients with a hip fracture in a population within the Netherlands.

Predictive Values of Preoperative Characteristics for 30-Day Mortality in Traumatic Hip Fracture Patients

Hip fracture patients have a high risk of mortality after surgery, with 30-day postoperative rates as high as 10%. This study aimed to explore the predictive ability of preoperative characteristics in traumatic hip fracture patients as they relate to 30-day postoperative mortality using readily available variables in clinical practice. All adult patients who underwent primary emergency hip fracture surgery in Sweden between 2008 and 2017 were included in the analysis. Associations between the possible predictors and 30-day mortality was performed using a multivariate logistic regression (LR) model; the bidirectional stepwise method was used for variable selection. An LR model and convolutional neural network (CNN) were then fitted for prediction. The relative importance of individual predictors was evaluated using the permutation importance and Gini importance. A total of 134,915 traumatic hip fracture patients were included in the study. The CNN and LR models displayed an acceptable predictive ability for predicting 30-day postoperative mortality using a test dataset, displaying an area under the ROC curve (AUC) of as high as 0.76. The variables with the highest importance in prediction were age, sex, hypertension, dementia, American Society of Anesthesiologists (ASA) classification, and the Revised Cardiac Risk Index (RCRI). Both the CNN and LR models achieved an acceptable performance in identifying patients at risk of mortality 30 days after hip fracture surgery. The most important variables for prediction, based on the variables used in the current study are age, hypertension, dementia, sex, ASA classification, and RCRI.

Association Between High-Sensitivity Troponin T on Admission and Organ Dysfunction During Hospitalization in Patients Aged 80 Years and Older with Hip Fracture: A Single-Centered Prospective Cohort Study

Background

Prognostic evaluation of elderly patients with hip fracture is an issue that has been highly concerned by clinicians. Only a few studies have focused on organ dysfunction after hip fracture in the elderly. This study aimed to investigate the association between high-sensitivity troponin T (hs-TnT) at admission and organ dysfunction during hospitalization in elderly patients with hip fracture.

Methods

We enrolled 168 patients with hip fracture who were aged 80 years and older at Geriatric Orthopaedic Center of Sichuan Provincial Orthopedic Hospital between January 2020 and August 2020. Baseline characteristics, perioperative information, and short-term clinical outcomes were analyzed.

Results

Of the 208 patients admitted during the study period, 168 met the inclusion criteria; of these, 91 (54.2%) had higher hs-TnT than the 99th percentile in the normal population. After adjustment for confounders, elevated hs-TnT was independently associated with multiple organ dysfunction syndrome in the elderly (MODSE) (adjusted OR, 5.76; 95% CI, 1.74–19.10; P = 0.004), heart dysfunction (adjusted OR, 7.48; 95% CI, 2.17–25.82; P = 0.001), MODS severity score > 3 (adjusted OR, 5.22; 95% CI, 1.32–20.60; P = 0.018), and length of hospital stay > 14 days (adjusted OR, 2.38; 95% CI, 1.05–5.36; P = 0.037).

Conclusion

Increased hs-TnT on admission is an independent risk factor for MODSE after hip fracture in patients aged 80 years and older. Effective measures should be applied to avoid progression of MODSE from pre-failure stage to failure stage.

Systemic immune-inflammation index independently predicts poor survival of older adults with hip fracture: a prospective cohort study

Background

The systemic immune-inflammation index (SII), based on peripheral platelet, neutrophil and lymphocyte counts, has been proven to be a promising prognostic indicator in various diseases. Hip fracture is a common injury among the older adults, and has become a global public health problem with high mortality and disability rates. However, the relationship between SII and the prognosis of hip fracture is not yet well-known. The aim of the this study was to explore the predictive value of SII in older adults with hip fracture undergoing surgery.

Methods

This was a prospective cohort study performed from January 2014 to December 2018 at a orthopaedic center, China. The SII was calculated as platelet×neutrophil/lymphocyte counts. Univariable and multivariable Cox proportional hazard models were used to assess the association between SII and all-cause mortality.

Results

A total of 290 older adults with hip fracture were included, and the mean (SD) age was 77.6 (8.6) years, and 189 (65.2%) were female. The median (IQR) SII was 759.4 (519.0–1128.7) × 10⁹/L. After a median follow-up time of 33.4 months, 13 (4.5%), 26 (9.0%) and 54 (18.6%) patients died within the 30-day, 1-year and last follow-up, respectively. Multivariable Cox analysis revealed that each increase of 100 units of SII was associated with a 8% increased hazard of death at 1-year follow-up (HR = 1.08, 95% CI: 1.01–1.17, p = 0.033), and 9% increased hazard of death at last follow-up (HR = 1.09, 95% CI: 1.03–1.15, p = 0.003).

Conclusions

SII is associated with poor all-cause mortality in older adults with hip fracture undergoing surgery, and deserves further investigation and application in clinical practice.

The management of displaced femoral neck fractures: a narrative review

This article serves to review the existing clinical guidelines, and highlight the most recent medical and surgical recommendations, for the management of displaced femoral neck fractures (FNFs). It stresses the need for multi-disciplinary intervention to potentially improve mortality rates, limit adverse events and prevent further economic liability.Globally, the incidence of FNFs continues to rise as the general population ages and becomes more active. The annual number of FNFs is expected to exceed six million by 2050. The increased burden of FNFs exacerbates the demand on all services associated with treating these injuries.The management of FNFs may serve as an indicator of the quality of care of the geriatric population. However, despite escalating health costs, a significant 30-day and one-year mortality rate, increased rate of peri-operative adverse events and sub-optimal functional clinical outcomes, continued controversy exists over optimal patient care.Much debate exists over the type of surgery, implant selection and peri-operative clinical care and rehabilitation. FNF care models, systematized clinical pathways, formal geriatrics consultation and specialized wards within an established interdisciplinary care framework may improve outcomes, mitigate adverse events and limit unnecessary costs. Cite this article: EFORT Open Rev 2021;6:139-144. DOI: 10.1302/2058-5241.6.200036.

The Ability of the Nottingham Hip Fracture Score to Predict Mobility, Length of Stay and Mortality in Hospital, and Discharge Destination in Patients Admitted with a Hip Fracture

The Nottingham Hip Fracture Score (NHFS) has been developed for predicting 30-day and 1-year mortality after hip fracture. We hypothesise that NHFS may also predict other adverse events. Data from 666 patients (190 men, 476 women), aged 60.2-103.4 years, admitted with a hip fracture to a single centre from 1/10/2015 and 7/12/2017 were analysed. The ability of NHFS to predict mobility within 1 day after surgery, length of stay (LOS) find mortality, and discharge destination was evaluated by receiver operating characteristic curves and two-graph plots. The area under the curve (95% confidence interval [CI]) for predicting mortality was 67.4% (58.4-76.4%), prolonged LOS was 59.0% (54.0-64.0%), discharge to residential/nursing care was 62.3% (54.0-71.5%), and any two of failure to mobilise, prolonged LOS or discharge to residential/nursing care was 64.8% (59.0-70.6%). NHFS thresholds at 4 and 7 corresponding to the lower and upper limits of intermediate range where sensitivity and specificity equal 90% were identified for mortality and prolonged LOS, and 4 and 6 for discharge to residential/nursing care, which were used to create three risk categories. Compared with the low risk group (NHFS = 0-4), the high risk group (NHFS = 7-10 or 6-10) had increased risk of in-patient mortality: rates = 2.0% versus 7.1%, OR (95% CI) = 3.8 (1.5-9.9), failure to mobilise within 1 day of surgery: rates = 18.9% versus 28.3%, OR = 1.7 (1.0-2.8), prolonged LOS (> 17 days): rates = 20.3% versus 33.9%, OR = 2.2 (1.3-3.3), discharge to residential/nursing care: rates = 4.5% vs 12.3%, OR = 3.0 (1.4-6.4), and any two of failure to mobilise, prolonged LOS or discharge to residential/nursing care: rates = 10.5% versus 28.6%, 3.4 (95% CI 1.9-6.0), and stayed 4.1 days (1.5-6.7 days) longer in hospital. High NHFS associates with increased risk of mortality, prolonged LOS and discharge to residential/nursing care, lending further support for its use to identify adverse events.

Prospective cohort protocol examining the perioperative indicators for complications and early mortality following hip fracture surgery in the frail patient

INTRODUCTION

The primary aim is to validate earlier suggested risk factors and to find new associated risk factors for (30-day) mortality after a hip fracture in the frail population. The secondary aim is to determine the factors associated with perioperative complications. At last we want to develop and validate a more specific 30-day mortality prediction tool compared with the Nottingham Hip Fracture Score. The 30-day mortality prediction can help inform surgical risk and guide shared decision-making among patients, family and physicians.

METHODS AND ANALYSIS

The study is designed as a prospective multicentre cohort study within the area of Rotterdam, the Netherlands starting from January 2018. All patients over 65 years of age, with an acute proximal hip fracture, are included. Treatment of patients will be by standard practice of care using the latest national and international guidelines. Inclusion will be continued at least until January 2021 and including at least 2500 patients. In this large cohort we hope to have sufficient strength and quality to identify risk factors of 30-day mortality and to compare them to known risk factors in literature. Moreover, we plan to develop and validate a 30-day mortality prediction tool, which identifies patients with a high probability of 30-day mortality.

ETHICS AND DISSEMINATION

Ethical approval for this protocol was given by the Ethics Committee of the Maasstad Hospital (TWOR). Patient data are stored anonymously using the Castor data management system. No external funding is used for this study. Results will be published in peer-reviewed publications and at international conferences.

TRIAL REGISTRATION NUMBER

NL8313.

Novel biomarkers for prediction of outcome in hip fracture patients-An exploratory study

BACKGROUND

Little is known about the value of biomarkers for prognostication in hip fracture patients. The main objective of the present study was to assess if biomarkers add useful information to an existing risk score for prediction of 30-day mortality in patients suffering from out of hospital hip fractures.

METHODS

In a prospective observational single centre study, association between plasma concentration of ninety-two biomarkers at admission and 30-day mortality was analysed using logistic regression adjusted for risk factors included in Nottingham Hip Fracture Score (NHFS). Biomarkers associated with the outcome in the adjusted analysis were further evaluated by calculating the net reclassification improvement (NRI) and the change in area under the receiver operating characteristics curve (AUC) relative to the NHFS.

RESULTS

997 patients were included. Sixty-two patients died within 30 days (6.2%). Eleven biomarkers were associated with 30-day mortality in adjusted analysis. Of these biomarkers Growth Differentiation Factor-15 (GDF-15) had NRI for the primary outcome (12.1%; 95% CI: 1.2-23.3) and Carbohydrate Antigen 125 (CA-125) improved the AUC relative to NHFS (improvement: 0.05; 95% CI: 0.01-0.10, P = .027). Both CA-125 and GDF-15 improved the AUC for a composite outcome of 30-day mortality and cardiovascular complications.

CONCLUSIONS

Adding GDF-15 or CA-125 to the Nottingham Hip Fracture Score improves the discrimination with regard to predicting 30-day mortality and may help to identify a subgroup of hip fracture patients with a particularly poor prognosis. The value of these biomarkers should be explored in further studies to confirm clinical utility.

The Sernbo score as a predictor of 1-year mortality after hip fracture: a registry study on 55,716 patients

Purpose

Patients sustaining a hip fracture have a high mortality rate during the first postoperative year and the Sernbo score may stratify patients into a high, intermediate and low risk of death during this period. We assessed its predictive properties on patients from the National Swedish Hip Fracture Register.

Patients and methods

55,716 hip fracture patients, 69% women older than 65 years at surgery (registered between 2010 and 2015) with complete Sernbo scores and mortality data were studied. Receiver-operating characteristics analyses (ROC) were used. Validation of Sernbo score was performed.

Results

The overall 1-year mortality rate was 26%—and 17%, 27.4% and 55.6% in the low, intermediate and high-risk groups, respectively. The ROC analysis indicated a predictive ability of the Sernbo score, with an AUC of 0.69 (CI 0.68–0.69).

Conclusion

In this registry-based study, the easy-to-use Sernbo scoring system proved to be appropriate and useful way to identify hip fracture patients with a high-risk mortality during the first postoperative year.

Electronic supplementary material

The online version of this article (10.1007/s00068-020-01375-4) contains supplementary material, which is available to authorized users.

Hip fracture mortality: Predictive models

The preoperative estimation of the risk of mortality after a hip fracture is very useful to plan time of surgery and perioperative care, inform patients and families about the prognosis and allows comparisons between different units. Different models have been developed to stratify mortality risk, but they show heterogeneity in terms of type of population and variables included, monitoring the time and statistical methods used, which makes it difficult to establish comparisons between them. The vast majority of them are awaiting external validation in populations different from those in which they were originally proposed. So far, the Nottingham Hip Fracture Score (NHFS) and the Orthopaedic Physiological and Operative Severity Score for the enUmeration of Mortality and Morbidity (O-POSSUM) are the most commonly used models. The NHFS is simpler and faster to apply, and by not including intraoperative variables can be used at the time of admission.

Validation of the Nottingham Hip Fracture Score (NHFS) to predict 30-day mortality in patients with an intracapsular hip fracture

BACKGROUND

The Nottingham Hip Fracture Score (NHFS) was developed to predict 30-day mortality following a fracture of the hip. While the NHFS has been validated in three hip fracture populations within Great Britain, these studies make no distinction between the type of fracture and surgery. Literature 'however' shows an increased risk for mortality after a hemi-arthroplasty following an intra-capsular hip fracture. To verify whether the mortality after an intra-capsular hip fracture is higher compared to the predicted mortality score according to the NHFS, a validation of the NHFS in patients with a hemi-arthroplasty after an intra-capsular hip fracture was performed.

METHODS

The NHFS was calculated for consecutive patients presenting with an intra-capsular fracture of the hip in two level II trauma teaching hospitals between 1 January 2011 and 1 May 2016. The observed 30-day mortality was compared with that predicted by the NHFS using several validation statistics.

RESULTS

A total of 901 patients were included in the present study. Mean age in the patients was 83 years (SD 8) and 623 (68%) of the patients were female. Almost 60% of the patients had an ASA-score (American Society of Anaesthesiologists [ASA]) of≥3 and overall 30-day mortality was 9.5% (n=86). The median NHFS was 5, and there was no significant change in median NHFS over the past 5 years. The mortality rate in the studied population of hemi-arthroplasty patients was significantly higher than mortality rates predicted by the NHFS (p=0.022, Pearson's Chi-squared test).

CONCLUSIONS

Findings suggest that for a patient with a hemi-arthroplasty following an intra-capsular hip fracture, there could be an underestimation for the 30-day mortality rate following the NHFS prediction model.

LEVEL OF EVIDENCE

Prognostic Level III, retrospective cohort study.

Higher 30-day mortality associated with the use of intramedullary nails compared with sliding hip screws for the treatment of trochanteric hip fractures: a prospective national registry study

AIMS

The aim of this study was to investigate the association between the type of operation used to treat a trochanteric fracture of the hip and 30-day mortality.

PATIENTS AND METHODS

Data on 82 990 patients from the National Hip Fracture Database were analyzed using generalized linear models with incremental case-mix adjustment for patient, non-surgical and surgical characteristics, and socioeconomic factors.

RESULTS

The use of short and long intramedullary nails was associated with an increase in 30-day mortality (adjusted odds ratio (OR) 1.125, 95% confidence interval (CI) 1.040 to 1.218; p = 0.004) compared with the use of sliding hip screws (12.5% increase). If this were causative, it would represent 98 excess deaths over the four-year period of the study and one excess death would be caused by treating 112 patients with an intramedullary nail rather than a sliding hip screw.

CONCLUSION

There is a 12.5% increase in the risk of 30-day mortality associated with the use of an intramedullary nail compared with a sliding hip screw in the treatment of a trochanteric fractures of the hip.

Accuracy of the Physiological and Operative Severity Score for the enUmeration of Mortality and morbidity score and the Nottingham risk score in hip fracture patients in Sweden - A prospective observational study

BACKGROUND

Little is known about accuracy of common risk prediction scores in elderly patients suffering from hip fractures. The objective of this study was to investigate accuracy of the Physiological and Operative Severity Score for the enUmeration of Mortality and morbidity (POSSUM) score, Portsmouth-POSSUM (P-POSSUM) score and the Nottingham Hip Fracture Score (NHFS) for prediction of mortality and morbidity in this patient group.

METHODS

This was a prospective single centre observational study on 997 patients suffering out-of-hospital cervical, trochanteric or subtrochanteric fracture of the neck of the femur. Calibration and discrimination was assessed by calculating the ratio of observed to expected events (O:E) and areas under receiver operating characteristics curves (ROC).

RESULTS

The 30-day mortality was 6.2% and complications, as defined by POSSUM, occurred in 41% of the patients. Overall O:E ratios for POSSUM, P-POSSUM and NHFS scores for 30-day mortality were 0.90, 0.98, and 0.79 respectively. The models underestimated mortality in the lower risk bands and overestimated mortality in the higher risk bands. In contrast, POSSUM predicted morbidity well with O:E ratios close to unity in most risk bands. The areas under the ROC curves for the scoring systems was 0.60-0.67.

CONCLUSION

The POSSUM score and NHFS show moderate calibration and poor discrimination in this cohort. The results suggest that mortality and morbidity in hip fracture patients are largely dependent on factors that are not included in these scores.