Application of Machine Learning Algorithms to Predict Acute Kidney Injury in Elderly Orthopedic Postoperative Patients

Potential Predictors for Deterioration of Renal Function After Transfusion

BACKGROUND

Transfusion of packed red blood cells (pRBCs) is still associated with risks. This study aims to determine whether renal function deterioration in the context of individual transfusions in individual patients can be predicted using machine learning. Recipient and donor characteristics linked to increased risk are identified.

METHODS

This study was registered at ClinicalTrials.gov (NCT05466370) and was conducted after local ethics committee approval. We evaluated 3366 transfusion episodes from a university hospital between October 31, 2016, and August 31, 2020. Random forest models were tuned and trained via Python auto-sklearn package to predict acute kidney injury (AKI). The models included recipients' and donors' demographic parameters and laboratory values, donor questionnaire results, and the age of the pRBCs. Bootstrapping on the test dataset was used to calculate the means and standard deviations of various performance metrics.

RESULTS

AKI as defined by a modified Kidney Disease Improving Global Outcomes (KDIGO) criterion developed after 17.4% transfusion episodes (base rate). AKI could be predicted with an area under the curve of the receiver operating characteristic (AUC-ROC) of 0.73 ± 0.02. The negative (NPV) and positive (PPV) predictive values were 0.90 ± 0.02 and 0.32 ± 0.03, respectively. Feature importance and relative risk analyses revealed that donor features were far less important than recipient features for predicting posttransfusion AKI.

CONCLUSIONS

Surprisingly, only the recipients' characteristics played a decisive role in AKI prediction. Based on this result, we speculate that the selection of a specific pRBC may have less influence than recipient characteristics.

Analysis of Baseline Characteristics, Length of Stay, Cost of Care, Complications and Subgroup Analysis of Patients Undergoing Total Ankle Arthroplasty-A Large Database Study

Although total ankle arthroplasty (TAA) is becoming a progressively common procedure with a reported 10-fold increase in its prevalence over the past 2 decades; there is still limited large-scale data regarding its overall outcome. Using the National Inpatient Sample (NIS) database, patients who underwent TAA between 2016 and 2019 were identified (ICD-10 CMP code). Data regarding demographic details, co-morbidities, geographic locations of procedure, hospital stay, expenditure incurred, and complications encountered were analyzed. Additionally, a comprehensive subgroup analysis was performed to evaluate the impact of multiple preoperative variables (including gender, diabetes, obesity, CKD and tobacco abuse) on the patient outcome. Overall, 5087 patients (mean age: 65.1 years, 54% males, 85% Caucasians, 75% from large metropolitan regions) underwent TAA. Eighty eight percent of patients were discharged to home; and the mean length of hospital stay and hospital-related expenditure were 1.7 ± 1.41 days and $92,304.5 ± 50,794.1, respectively. The overall complication rate was 8.39% {commonest medical complications: anemia [131 (2.6%) patients) and acute renal failure [37 (0.7%) patients]; commonest local complication: periprosthetic mechanical adversities [90 (1.7%) patients]}. Female and CKD patients demonstrated significantly higher risks of medical (female: p = .003; CKD: p < .001) and surgical (female: p = .005; CKD: p < .019) complications; while obesity substantially enhanced the risk of medical adversities (p < .001). Based on our study, we could conclude that the rates of TAA in the United States are on the rise, especially in regions with population greater than 250,000. TAA is a safe procedure with relatively low complication rates. The complications and hospital-associated expenditure seem to vary between different patient subgroups.

Construction and Validation of a Nomogram for Predicting Acute Kidney Injury After Hip Fracture Surgery

Background

Acute kidney injury (AKI), characterized by sudden impairment of kidney function, is an uncommon complication following hip fracture surgery that is associated with increased morbidity and mortality. We constructed a nomogram to stratify patients according to risk of AKI after hip fracture surgery to guide clinicians in the implementation of timely interventions.

Methods

Patients who received hip fracture surgery from January 2015 to December 2021 were retrospectively identified and divided into a training set (n=448, surgery from January 2015 to December 2019) and a validation set (n=200, surgery from January 2020 to December 2021). Univariate and multivariate logistic regression were used to identify risk factors for AKI after surgery in the training set. A nomogram was constructed based the risk factors for AKI, and was evaluated by receiver operating characteristic (ROC) analysis, calibration curves, and decision curve analysis (DCA).

Results

The mean age was 82.0±6.22 years-old and the prevalence of post-surgical AKI was 13.3%. Age, American Society of Anesthesiologists (ASA) score, the preexistence of chronic kidney disease (CKD), cemented surgery and the decrease of hemoglobin on the first day after surgery were identified as independent risk factors of AKI after hip fracture surgery, and a predictive nomogram was established based on the multivariable model. The predictive nomogram had good discrimination ability (training set: AUC: 0.784, 95% CI: 0.720-0.848; validation set: AUC: 0.804, 95% CI: 0.704-0.903), and showed good validation ability and clinical usefulness based on a calibration plot and decision curve analysis.

Conclusion

A nomogram that incorporated five risk factors including age, ASA score, preexisting CKD, cemented surgery and the decrease of hemoglobin on the first day after surgery had good predictive performance and discrimination. Use of our results for early stratification and intervention has the potential to improve the outcomes of patients receiving hip fracture surgery. Future large, multicenter cohorts are needed to verify the model's performance.

Artificial Intelligence and Machine Learning in Perioperative Acute Kidney Injury

Acute kidney injury (AKI) is a common complication after a surgery, especially in cardiac and aortic procedures, and has a significant impact on morbidity and mortality. Early identification of high-risk patients and providing effective prevention and therapeutic approach are the main strategies for reducing the possibility of perioperative AKI. Consequently, several risk-prediction models and risk assessment scores have been developed for the prediction of perioperative AKI. However, a majority of these risk scores are only derived from preoperative data while the intraoperative time-series monitoring data such as heart rate and blood pressure were not included. Moreover, the complexity of the pathophysiology of AKI, as well as its nonlinear and heterogeneous nature, imposes limitations on the use of linear statistical techniques. The development of clinical medicine's digitization, the widespread availability of electronic medical records, and the increase in the use of continuous monitoring have generated vast quantities of data. Machine learning has recently shown promise as a method for automatically integrating large amounts of data in predicting the risk of perioperative outcomes. In this article, we discussed the development, limitations of existing work, and the potential future direction of models using machine learning techniques to predict AKI after a surgery.