Development of a postoperative delirium risk scoring tool using data from the Australian and New Zealand Hip Fracture Registry: an analysis of 6672 patients 2017-2018

Development and validation of a new drug-focused predictive risk score for postoperative delirium in orthopaedic and trauma surgery patients

BACKGROUND

Postoperative delirium (POD) is the most common complication following surgery in elderly patients. During pharmacist-led medication reconciliation (PhMR), a predictive risk score considering delirium risk-increasing drugs and other available risk factors could help to identify risk patients.

METHODS

Orthopaedic and trauma surgery patients aged ≥ 18 years with PhMR were included in a retrospective observational single-centre study 03/2022-10/2022. The study cohort was randomly split into a development and a validation cohort (6:4 ratio). POD was assessed through the 4 A's test (4AT), delirium diagnosis, and chart review. Potential risk factors available at PhMR were tested via univariable analysis. Significant variables were added to a multivariable logistic regression model. Based on the regression coefficients, a risk score for POD including delirium risk-increasing drugs (DRD score) was established.

RESULTS

POD occurred in 42/328 (12.8%) and 30/218 (13.8%) patients in the development and validation cohorts, respectively. Of the seven evaluated risk factors, four were ultimately tested in a multivariable logistic regression model. The final DRD score included age (66-75 years, 2 points; > 75 years, 3 points), renal impairment (eGFR < 60 ml/min/1.73m2, 1 point), anticholinergic burden (ACB-score ≥ 3, 1 point), and delirium risk-increasing drugs (n ≥ 2; 2 points). Patients with ≥ 4 points were classified as having a high risk for POD. The areas under the receiver operating characteristic curve of the risk score model were 0.89 and 0.81 for the development and the validation cohorts, respectively.

CONCLUSION

The DRD score is a predictive risk score assessable during PhMR and can identify patients at risk for POD. Specific preventive measures concerning drug therapy safety and non-pharmacological actions should be implemented for identified risk patients.

A novel risk assessment tool for postoperative delirium in vascular surgery: The stress model (Siena posTopeRative dElirium in vaScular Surgery)

OBJECTIVE

Postoperative delirium (POD) is a common complication with a high health-related impact. The creation of a model (Siena posTopeRative dElirium in vaScular Surgery) to identify high-risk patients with consecutive prompt diagnosis and correct management.

METHODS

This is an observational retrospective study to evaluate POD incidence in patients who underwent elective vascular surgery procedures between 2018 and 2020. POD was detected using CAM and defined as the onset of an acute confusional state, clinically manifesting as a disturbed state of consciousness, cognitive dysfunction, or alteration in perception and behavior. The total population was divided in the development and validation subsamples. Multivariable logistic regression analysis was performed, identifying variables related to the occurrence of POD. An additive score was created and the STRESS score was internally validated using the Validation subgroup.

RESULTS

A total of 1067 patients were enrolled. POD occurred in 111 cases (10.4%). Multivariable logistic regression analysis for POD occurrence revealed as significant predictors: age>75 years, CKD, dyslipidaemia, psychiatric disease, CAD, hospitalization in the previous month, preoperative NLR >3.59, preoperative Hb < 12 g/dl, preoperative Barthel score <75, major amputation, CLTI revascularization, general anesthesia, and postoperative urinary catheter. These variables were used to create the STRESS score. The model was applied to both development and validation subgroups; AUC was respectively 0.7079 (p < .0001) and 0.7270 (p < .0001).

CONCLUSION

The STRESS score has a good predictive potentiality for POD occurrence in elective vascular surgery procedures. However, implementation and external validation are needed to be correctly used in everyday clinical practice.

Risk prediction models for postoperative delirium in elderly patients with fragility hip fracture: A systematic review and critical appraisal

BACKGROUND

Elderly patients with fragility hip fracture continue to experience a high rate of postoperative delirium (POD), which has a significant negative impact on prognosis and imposes a huge economic burden. A number of risk prediction models have been developed to detect POD early. However, the risk of bias and clinical applicability of the models remain unclear. The aim of this study was to systematically evaluate risk prediction models for POD.

METHODS

CNKI, WanFang DATA, Vip Database, SinoMed, PubMed, Web of Science, Embase, and the Cochrane Library were searched for studies published by July 2023. The literature was screened independently by two investigators. The Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modeling Studies (CHARMS) and the Prediction Model Risk of Bias Assessment Tool (PROBAST) were respectively used for data extraction, risk of bias, and applicability assessment.

RESULTS

A total of 16 studies on the construction of POD risk prediction models were included. The area under the ROC curve of the models ranges from 0.670 to 0.957. The most common predictors of POD included age, history of dementia, history of delirium, ASA classification, preoperative waiting time, and preoperative albumin level. All models had a high risk of bias, mainly due to inadequate sample size, inappropriate handling of missing data, a lack of model performance evaluation, and overfitting of the models.

CONCLUSIONS

Overall, risk prediction models for POD in fragility hip fracture patients are still in the development stage. The majority of POD prediction models have substantial bias risks, attributable primarily to poor reporting of analysis and evaluation of model performance. In future research, it is recommended to conduct validation of the models or develop localized prediction models with demonstrated high performance, with the aim of benefiting POD screening.

SYSTEMATIC REVIEW REGISTRATION

PROSPERO CRD42023449153.

Prediction models for postoperative delirium in elderly patients with machine-learning algorithms and SHapley Additive exPlanations

Postoperative delirium (POD) is a common and severe complication in elderly patients with hip fractures. Identifying high-risk patients with POD can help improve the outcome of patients with hip fractures. We conducted a retrospective study on elderly patients (≥65 years of age) who underwent orthopedic surgery with hip fracture between January 2014 and August 2019. Conventional logistic regression and five machine-learning algorithms were used to construct prediction models of POD. A nomogram for POD prediction was built with the logistic regression method. The area under the receiver operating characteristic curve (AUC-ROC), accuracy, sensitivity, and precision were calculated to evaluate different models. Feature importance of individuals was interpreted using Shapley Additive Explanations (SHAP). About 797 patients were enrolled in the study, with the incidence of POD at 9.28% (74/797). The age, renal insufficiency, chronic obstructive pulmonary disease (COPD), use of antipsychotics, lactate dehydrogenase (LDH), and C-reactive protein are used to build a nomogram for POD with an AUC of 0.71. The AUCs of five machine-learning models are 0.81 (Random Forest), 0.80 (GBM), 0.68 (AdaBoost), 0.77 (XGBoost), and 0.70 (SVM). The sensitivities of the six models range from 68.8% (logistic regression and SVM) to 91.9% (Random Forest). The precisions of the six machine-learning models range from 18.3% (logistic regression) to 67.8% (SVM). Six prediction models of POD in patients with hip fractures were constructed using logistic regression and five machine-learning algorithms. The application of machine-learning algorithms could provide convenient POD risk stratification to benefit elderly hip fracture patients.

Risk prediction models for postoperative delirium in elderly patients with hip fracture: a systematic review

Objectives

To systematically evaluate the risk prediction models for postoperative delirium in older adult hip fracture patients.

Methods

Risk prediction models for postoperative delirium in older adult hip fracture patients were collected from the Cochrane Library, PubMed, Web of Science, and Ovid via the internet, covering studies from the establishment of the databases to March 15, 2023. Two researchers independently screened the literature, extracted data, and used Stata 13.0 for meta-analysis of predictive factors and the Prediction Model Risk of Bias Assessment Tool (PROBAST) to evaluate the risk prediction models for postoperative delirium in older adult hip fracture patients, evaluated the predictive performance.

Results

This analysis included eight studies. Six studies used internal validation to assess the predictive models, while one combined both internal and external validation. The Area Under Curve (AUC) for the models ranged from 0.67 to 0.79. The most common predictors were preoperative dementia or dementia history (OR = 3.123, 95% CI 2.108-4.626, p < 0.001), American Society of Anesthesiologists (ASA) classification (OR = 2.343, 95% CI 1.146-4.789, p < 0.05), and age (OR = 1.615, 95% CI 1.387-1.880, p < 0.001). This meta-analysis shows that these were independent risk factors for postoperative delirium in older adult patients with hip fracture.

Conclusion

Research on the risk prediction models for postoperative delirium in older adult hip fracture patients is still in the developmental stage. The predictive performance of some of the established models achieve expectation and the applicable risk of all models is low, but there are also problems such as high risk of bias and lack of external validation. Medical professionals should select existing models and validate and optimize them with large samples from multiple centers according to their actual situation. It is more recommended to carry out a large sample of prospective studies to build prediction models.

Systematic review registration

The protocol for this systematic review was published in the International Prospective Register of Systematic Reviews (PROSPERO) under the registered number CRD42022365258.

Evaluating the performance of machine learning methods for risk estimation of delirium in patients hospitalized from the emergency department

INTRODUCTION

Delirium is a cerebral dysfunction seen commonly in the acute care setting. It is associated with increased mortality and morbidity and is frequently missed in the emergency department (ED) and inpatient care by clinical gestalt alone. Identifying those at risk of delirium may help prioritize screening and interventions in the hospital setting.

OBJECTIVE

Our objective was to leverage electronic health records to identify a clinically valuable risk estimation model for prevalent delirium in patients being transferred from the ED to inpatient units.

METHODS

This was a retrospective cohort study to develop and validate a risk model to detect delirium using patient data available from prior visits and ED encounter. Electronic health records were extracted for patients hospitalized from the ED between January 1, 2014, and December 31, 2020. Eligible patients were aged 65 or older, admitted to an inpatient unit from the emergency department, and had at least one DOSS assessment or CAM-ICU recorded within 72 h of hospitalization. Six machine learning models were developed to estimate the risk of delirium using clinical variables including demographic features, physiological measurements, medications administered, lab results, and diagnoses.

RESULTS

A total of 28,531 patients met the inclusion criteria with 8057 (28.4%) having a positive delirium screening within the outcome observation period. Machine learning models were compared using the area under the receiver operating curve (AUC). The gradient boosted machine achieved the best performance with an AUC of 0.839 (95% CI, 0.837-0.841). At a 90% sensitivity threshold, this model achieved a specificity of 53.5% (95% CI 53.0%-54.0%) a positive predictive value of 43.5% (95% CI 43.2%-43.9%), and a negative predictive value of 93.1% (95% CI 93.1%-93.2%). A random forest model and L1-penalized logistic regression also demonstrated notable performance with AUCs of 0.837 (95% CI, 0.835-0.838) and 0.831 (95% CI, 0.830-0.833) respectively.

CONCLUSION

This study demonstrated the use of machine learning algorithms to identify a combination of variables that enables an estimation of risk of positive delirium screens early in hospitalization to develop prevention or management protocols.

Postoperative delirium in geriatric patients with hip fractures

Postoperative delirium (POD) is a frequent complication in geriatric patients with hip fractures, which is linked to poorer functional recovery, longer hospital stays, and higher short-and long-term mortality. Patients with increased age, preoperative cognitive impairment, comorbidities, perioperative polypharmacy, and delayed surgery are more prone to develop POD after hip fracture surgery. In this narrative review, we outlined the latest findings on postoperative delirium in geriatric patients with hip fractures, focusing on its pathophysiology, diagnosis, prevention, and treatment. Perioperative risk prediction, avoidance of certain medications, and orthogeriatric comprehensive care are all examples of effective interventions. Choices of anesthesia technique may not be associated with a significant difference in the incidence of postoperative delirium in geriatric patients with hip fractures. There are few pharmaceutical measures available for POD treatment. Dexmedetomidine and multimodal analgesia may be effective for managing postoperative delirium, and adverse complications should be considered when using antipsychotics. In conclusion, perioperative risk intervention based on orthogeriatric comprehensive care is the most effective strategy for preventing postoperative delirium in geriatric patients with hip fractures.

Association between multidimensional prognostic index (MPI) and pre-operative delirium in older patients with hip fracture

Pre-operative delirium may cause delay in surgical intervention in older patients hospitalized for hip fracture. Also it has been associated with higher risk of post-surgical complications and worst functional outcomes. Aim of this retrospective cohort study was to evaluate whether the multidimensional prognostic index (MPI) at hospital admission was associated with pre-operative delirium in older individuals with hip fracture who are deemed to require surgical intervention. Consecutive older patients (≥ 65 years) with hip fracture underwent a comprehensive geriatric assessment to calculate the MPI at hospital admission. According to previously established cut-offs, MPI was expressed in three grades, i.e. MPI-1 (low-risk), MPI-2 (moderate-risk) and MPI-3 (high risk of mortality). Pre-operative delirium was assessed using the four 'A's Test. Out of 244 older patients who underwent surgery for hip fracture, 104 subjects (43%) received a diagnosis of delirium. Overall, the incidence of delirium before surgery was significantly higher in patients with more severe MPI score at admission. Higher MPI grade (MPI-3) was independently associated with higher risk of pre-operative delirium (OR 2.45, CI 1.21-4.96). Therefore, the MPI at hospital admission might help in early identification of older patients with hip fracture at risk for pre-operative delirium.

Risk factors for postoperative delirium in geriatric patients with hip fracture: A systematic review and meta-analysis

Objectives

This systematic review and meta-analysis was conducted to identify the potential risk factors for postoperative delirium in geriatric patients with hip fracture.

Methods

PubMed, EMBASE, and Cochrane Library were searched from inception until December 31st, 2021. A combined searching strategy of subject words and free words was adopted. Studies involving risk factors for postoperative delirium in elderly patients undergoing hip fracture surgeries were reviewed. Qualities of included studies were assessed using the Newcastle–Ottawa Scale. Data were pooled and a meta-analysis was performed using Review Manager 5.3.

Results

A total of 37 studies were included. The following risk factors were significant: advanced age (per year increase) (OR: 1.05, 95% CI 1.04–1.07), age>80 years (OR: 2.26, 95% CI 1.47–3.47), male (OR: 1.53, 95% CI 1.37–1.70), preoperative cognitive impairment (OR:3.20, 95% CI 2.12–4.83), preoperative dementia (OR: 2.74, 95% CI 2.18–3.45), preoperative delirium (OR: 9.23, 95% CI 8.26–10.32), diabetes (OR: 1.18, 95% CI 1.05–1.33), preoperative functional dependence (OR: 1.31, 95% CI 1.11–1.56), ASA level (per level increase) (OR: 1.63, 95% CI 1.04–2.57), ASA level≥3(OR: 1.76, 95% CI 1.39–2.24), low albumin (OR: 3.30, 95% CI 1.44–7.55), medical comorbidities (OR: 1.15, 95% CI 1.06–1.25), Parkinson's disease (OR: 4.17, 95% CI 1.68–10.31) and surgery delay>48 h (OR: 1.90, 95% CI 1.36–2.65).

Conclusions

Clinicians should be alert to patients with those risk factors. To identify the risk factors more precisely, more research studies with larger sample size and better design should be conducted.

Is delirium associated with negative outcomes in older patients with hip fracture: analysis of the 4904 patients 2017-2018 from the Australian and New Zealand hip fracture registry

AIM

To determine associations between delirium and health outcomes using the Australia and New Zealand population-based hip fracture patient registry (ANZHFR).

METHODS

We performed a retrospective cohort study using data from the ANZHFR among hip-fracture surgery patients admitted to and discharged from hospital between 1 January 2017 and 31 December 2018.

RESULTS

Of the 4904 patients with complete data and included in the analysis, 1789 (36.5%) experienced delirium during their hospital stay. Patients with delirium also had a higher rate of in-hospital mortality (adjusted HR = 1.76; 95% CI = 1.24, 2.49; P < 0.001), a higher rate of long-term mortality (adjusted HR = 1.30; 95% CI = 1.15, 1.48; P < 0.001) and a higher odds of discharge to an aged care facility (adjusted OR = 1.24; 95% CI = 1.04, 1.48; P = 0.019).

CONCLUSION

A high rate of postoperative delirium exists among Australian and New Zealand hip fracture patients. Rates of hospital mortality, length of hospital stay and discharge to residential aged care are considerably worse in these patients.