Risk Factors and a Nomogram Model Establishment for Postoperative Delirium in Elderly Patients Undergoing Arthroplasty Surgery: A Single-Center Retrospective Study

Predicting the Risk of Postoperative Delirium in Elderly Patients Undergoing Hip Arthroplasty: Development and Assessment of a Novel Nomogram

OBJECTIVE

To construct and internally validate a nomogram that predicts the likelihood of postoperative delirium in a cohort of elderly individuals undergoing hip arthroplasty.

METHODS

Data for a total of 681 elderly patients underwent hip arthroplasty were retrospectively collected and divided into a model (n = 477) and a validation cohort (n = 204) according to the principle of 7:3 distribution temporally. The assessment of postoperative cognitive function was conducted through the utilization of The Confusion Assessment Method (CAM). The nomogram model for postoperative cognitive impairments was established by a combination of Lasso regression and logistic regression. The receiver operating characteristic (ROC) curve, calibration plot, and decision curve analysis (DCA) were used to evaluate the performance.

RESULTS

The nomogram utilized various predictors, including age, body mass index (BMI), education, preoperative Barthel Index, preoperative hemoglobin level, history of diabetes, and history of cerebrovascular disease, to forecast the likelihood of postoperative delirium in patients. The area under the ROC curves (AUC) for the nomogram, incorporating the aforementioned predictors, was 0.836 (95% CI: 0.797-0.875) for the training set and 0.817 (95% CI: 0.755-0.880) for the validation set. The calibration curves for both sets indicated a good agreement between the nomogram's predictions and the actual probabilities.

CONCLUSION

The use of this novel nomogram can help clinicians predict the likelihood of delirium after hip arthroplasty in elderly patients and help prevent and manage it in advance.

Risk factors and prediction model of delirium in elderly patients after hip arthroplasty

Objective

To analyze the risk factors of delirium in elderly patients after hip arthroplasty and to construct a prediction model.

Methods

Clinical data of 248 elderly patients who underwent hip arthroplasty in the Department of Traumatology and Orthopedics at Wuhan Fourth Hospital were retrospectively collected from November 2021 to February 2023. Logistic regression analysis was used to identify the risk factors of delirium after hip arthroplasty, and a nomogram prediction model was constructed using the RMS package of R4.1.2 software. The accuracy and stability of the model was evaluated based on the Hosmer-Lemeshow goodness-of-fit test and the receiver operating characteristic (ROC) curve.

Results

Age, nighttime sleep, anesthesia method, intraoperative blood loss, hypoxemia, and C-reactive protein (CRP) level were all risk factors of delirium after the hip arthroplasty (P<0.05). These factors were used to construct a nomogram prediction model that was internally validated using the Bootstrap method. The prediction model had the area under ROC curve (AUC) of 0.980 (95% CI: 0.964-0.996), indicating that it has certain predictive value for postoperative delirium. When the optimal cut off value was selected, the sensitivity and specificity were 92.7% and 92.3%, respectively, indicating that the prediction model is effective.

Conclusions

Age, short nighttime sleep, general anesthesia, high intraoperative blood loss, hypoxemia, and high CRP levels are independent risk factors for delirium after hip arthroplasty. The nomogram prediction model constructed based on these risk factors can effectively predict delirium in elderly patients after hip arthroplasty.

Nomogram for predicting the risk of postoperative delirium in elderly patients undergoing orthopedic surgery

OBJECTIVE

To retrospectively analyze the risk factors for postoperative delirium (POD) after orthopedic surgery in elderly patients and establish an individualized nomogram to predict the risk of POD.

METHODS

The data of 1011 patients who underwent orthopedic surgery from January 2019 to January 2022 were retrospectively analyzed. Univariate and multivariate logistic analyses were used to screen for independent risk factors. Stepwise regression was conducted to screen risk factors to construct a nomogram to predict the risk of POD after orthopedic surgery in elderly individuals, and nomogram validation analyses were performed.

RESULTS

The logistic regression results showed that age (≥ 75 years old vs. < 75 years old; odds ratio (OR) = 2.889; 95% confidence interval (CI), 1.149, 7.264), sex (male vs. female, OR = 2.368; 95% CI, 1.066, 5.261), and preoperative cognitive impairment (yes vs. no, OR = 13.587; 95% CI, 4.360, 42.338) were independent risk factors for POD in elderly patients who underwent orthopedic surgery (P < 0.05). A nomogram was constructed using 7 risk factors, i.e., age, American Society of Anesthesiologists (ASA) classification, sex, preoperative hemoglobin (Hb), preoperative pulmonary disease, cognitive impairment, and intraoperative infusion volume. The area under the curve (AUC) showed good discrimination (0.867), the slope of the calibration curve was 1.0, and the optimal net benefit of the nomogram from the decision curve analysis (DCA) was 0.01-0.58.

CONCLUSION

This study used 7 risk factors to construct a nomogram to predict the risk of POD after major orthopedic surgery in elderly individuals, and the nomogram had good discrimination ability, accuracy, and clinical practicability.

Predicting postoperative delirium after hip arthroplasty for elderly patients using machine learning

BACKGROUND

Postoperative delirium (POD) is a common and severe complication in elderly hip-arthroplasty patients.

AIM

This study aims to develop and validate a machine learning (ML) model that determines essential features related to POD and predicts POD for elderly hip-arthroplasty patients.

METHODS

The electronic record data of elderly patients who received hip-arthroplasty surgery between January 2017 and April 2021 were enrolled as the dataset. The Confusion Assessment Method (CAM) was administered to the patients during their perioperative period. The feature section method was employed as a filter to determine leading features. The classical machine learning algorithms were trained in cross-validation processing, and the model with the best performance was built in predicting the POD. Metrics of the area under the curve (AUC), accuracy (ACC), sensitivity, specificity, and F1-score were calculated to evaluate the predictive performance.

RESULTS

476 Arthroplasty elderly patients with general anesthesia were included in this study, and the final model combined feature selection method mutual information (MI) and linear binary classifier using logistic regression (LR) achieved an encouraging performance (AUC = 0.94, ACC = 0.88, sensitivity = 0.85, specificity = 0.90, F1-score = 0.87) on a balanced test dataset.

CONCLUSION

The model could predict POD with satisfying accuracy and reveal important features of suffering POD such as age, Cystatin C, GFR, CHE, CRP, LDH, monocyte count, history of mental illness or psychotropic drug use and intraoperative blood loss. Proper preoperative interventions for these factors could reduce the incidence of POD among elderly patients.

Comparison of logistic regression and machine learning methods for predicting postoperative delirium in elderly patients: A retrospective study

AIMS

To compare the performance of logistic regression and machine learning methods in predicting postoperative delirium (POD) in elderly patients.

METHOD

This was a retrospective study of perioperative medical data from patients undergoing non-cardiac and non-neurology surgery over 65 years old from January 2014 to August 2019. Forty-six perioperative variables were used to predict POD. A traditional logistic regression and five machine learning models (Random Forest, GBM, AdaBoost, XGBoost, and a stacking ensemble model) were compared by the area under the receiver operating characteristic curve (AUC-ROC), sensitivity, specificity, and precision.

RESULTS

In total, 29,756 patients were enrolled, and the incidence of POD was 3.22% after variable screening. AUCs were 0.783 (0.765-0.8) for the logistic regression method, 0.78 for random forest, 0.76 for GBM, 0.74 for AdaBoost, 0.73 for XGBoost, and 0.77 for the stacking ensemble model. The respective sensitivities for the 6 aforementioned models were 74.2%, 72.2%, 76.8%, 63.6%, 71.6%, and 67.4%. The respective specificities for the 6 aforementioned models were 70.7%, 99.8%, 96.5%, 98.8%, 96.5%, and 96.1%. The respective precision values for the 6 aforementioned models were 7.8%, 52.3%, 55.6%, 57%, 54.5%, and 56.4%.

CONCLUSIONS

The optimal application of the logistic regression model could provide quick and convenient POD risk identification to help improve the perioperative management of surgical patients because of its better sensitivity, fewer variables, and easier interpretability than the machine learning model.

Risk factors for postoperative delirium following total hip or knee arthroplasty: A meta-analysis

Objectives

The purpose of this study was to identify risk factors for delirium after total joint arthroplasty (TJA) and provide theoretical guidance for reducing the incidence of delirium after TJA.

Methods

The protocol for this meta-analysis is registered with PROSPERO (CRD42020170031). We searched PubMed, the Cochrane Library and Embase for observational studies on risk factors for delirium after TJA. Review Manager 5.3 was used to calculate the relative risk (RR) or standard mean difference (SMD) of potential risk factors related to TJA. STATA 14.0 was used for quantitative publication bias evaluation.

Results

In total, 25 studies including 3,767,761 patients from 9 countries were included. Old age has been widely recognized as a risk factor for delirium. Our results showed that the main risk factors for delirium after TJA were patient factors (alcohol abuse: RR = 1.63; length of education: SMD = −0.93; and MMSE score: SMD = −0.39), comorbidities (hypertension: RR = 1.26; diabetes mellitus: RR = 1.67; myocardial infarction: RR = 17.75; congestive heart failure: RR = 2.54; dementia: RR = 17.75; renal disease: RR = 2.98; history of stroke: RR = 4.83; and history of mental illness: RR = 2.36), surgical factors (transfusion: RR = 1.53; general anesthesia: RR = 1.10; pre-operative albumin: SMD = −0.38; pre-operative hemoglobin: SMD = −0.29; post-operative hemoglobin: SMD = −0.24; total blood loss: SMD = 0.15; duration of surgery: SMD = 0.29; and duration of hospitalization: SMD = 2.00) and drug factors (benzodiazepine use: RR = 2.14; ACEI use: RR = 1.52; and beta-blocker use: RR = 1.62).

Conclusions

Multiple risk factors were associated with delirium after TJA. These results may help doctors predict the occurrence of delirium after surgery and determine the correct treatment.

Systematic review registration

https://www.crd.york.ac.uk/prospero/, identifier: CRD42020170031.

Lower preoperative serum uric acid level may be a risk factor for postoperative delirium in older patients undergoing hip fracture surgery: a matched retrospective case-control study

Background

Postoperative delirium (POD) is a common complication after hip fracture surgery that is associated with various short- and long-term outcomes. The mechanism of POD may be associated with the oxidative stress process. Uric acid has been shown to provide a neuroprotective effect in various neurodegenerative diseases through its antioxidant properties. However, it is unclear whether lower preoperative serum uric acid levels are associated with the development of POD after hip fracture surgery. Therefore, this study assessed the association of lower preoperative uric acid levels in patients with POD during hospitalization.

Methods

This is a matched retrospective case-control study that included 96 older patients (≥65 y) who underwent hip fracture surgery. POD was diagnosed using the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition. Patients diagnosed with POD (cases) were matched 1:1 with patients without POD (controls) on the basis of age, sex, and anesthesia type. The relationship between preoperative uric acid and POD was analyzed by multivariable analysis.

Results

The POD and non-POD groups each had 48 patients. In the univariate analysis, lower log preoperative serum uric acid value, higher neutrophil-to-lymphocyte ratio, and cerebrovascular disease were more likely in patients with POD than in those with no POD. Multivariable conditional logistic regression analysis showed that lower log preoperative serum uric acid (adjusted odds ratio [aOR], 0.028; confidence interval [CI], 0.001–0.844; p = 0.040), higher neutrophil-to-lymphocyte ratio (aOR, 1.314; 95% CI, 1.053–1.638; p = 0.015), and increased surgery duration (aOR, 1.034; 95% CI, 1.004–1.065; p = 0.024) were associated with increased risk of POD.

Conclusions

Lower preoperative serum uric acid levels may be an independent risk factor for POD after adjustment for possible confounding factors. However, large prospective studies are needed to confirm this finding.