Laboratory tests as short-term correlates of stroke

A model for predicting postoperative persistent acute kidney injury (AKI) in AKI after cardiac surgery patients with normal baseline renal function

BACKGROUND

Persistent acute kidney injury (AKI) after cardiac surgery is not uncommon and linked to poor outcomes.

HYPOTHESIS

The purpose was to develop a model for predicting postoperative persistent AKI in patients with normal baseline renal function who experienced AKI after cardiac surgery.

METHODS

Data from 5368 patients with normal renal function at baseline who experienced AKI after cardiopulmonary bypass cardiac surgery in our hospital were retrospectively evaluated. Among them, 3768 patients were randomly assigned to develop the model, while the remaining patients were used to validate the model. The new model was developed using logistic regression with variables selected using least absolute shrinkage and selection operator regression.

RESULTS

The incidence of persistent AKI was 50.6% in the development group. Nine variables were selected for the model, including age, hypertension, diabetes, coronary heart disease, cardiopulmonary bypass time, AKI stage at initial diagnosis after cardiac surgery, postoperative serum magnesium level of <0.8 mmol/L, postoperative duration of mechanical ventilation, and postoperative intra-aortic balloon pump use. The model's performance was good in the validation group. The area under the receiver operating characteristic curve was 0.761 (95% confidence interval: 0.737-0.784). Observations and predictions from the model agreed well in the calibration plot. The model was also clinically useful based on decision curve analysis.

CONCLUSIONS

It is feasible by using the model to identify persistent AKI after cardiac surgery in patients with normal baseline renal function who experienced postoperative AKI, which may aid in patient stratification and individualized precision treatment strategy.

Predicting Risk of Stroke From Lab Tests Using Machine Learning Algorithms: Development and Evaluation of Prediction Models

Background

Stroke, a cerebrovascular disease, is one of the major causes of death. It causes significant health and financial burdens for both patients and health care systems. One of the important risk factors for stroke is health-related behavior, which is becoming an increasingly important focus of prevention. Many machine learning models have been built to predict the risk of stroke or to automatically diagnose stroke, using predictors such as lifestyle factors or radiological imaging. However, there have been no models built using data from lab tests.

Objective

The aim of this study was to apply computational methods using machine learning techniques to predict stroke from lab test data.

Methods

We used the National Health and Nutrition Examination Survey data sets with three different data selection methods (ie, without data resampling, with data imputation, and with data resampling) to develop predictive models. We used four machine learning classifiers and six performance measures to evaluate the performance of the models.

Results

We found that accurate and sensitive machine learning models can be created to predict stroke from lab test data. Our results show that the data resampling approach performed the best compared to the other two data selection techniques. Prediction with the random forest algorithm, which was the best algorithm tested, achieved an accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and area under the curve of 0.96, 0.97, 0.96, 0.75, 0.99, and 0.97, respectively, when all of the attributes were used.

Conclusions

The predictive model, built using data from lab tests, was easy to use and had high accuracy. In future studies, we aim to use data that reflect different types of stroke and to explore the data to build a prediction model for each type.

Causes of mortality in patients after first-ever stroke: A retrospective population-based study

BACKGROUND

Stroke is the third most common cause of death in developed countries and it is the most common cause of disability in the adult population of Iran. In this study, we aimed to evaluate the effects of age, sex, and other predisposing risk factors on mortality after stroke.

METHODS

We studied 1572 patients with first-ever stroke during a 7-year period from January 2008 to December 2014. Patients' medical records including demographic information, past medical history, physical examination, and laboratory testing were reviewed. We analyzed the correlation of qualitative and quantitative variables with sex and mortality.

RESULTS

Of all patients, 252 (16%) died during the hospital stay and of the remaining 1320 patients, 453 (34.3%) died during the follow-up period. There was no significant correlation between mortality and sex (p = .508). Descriptively, the number of women was higher in all age groups except in the age group 55-64 years. No significant correlation was observed between sex and age group (p = .748). We also observed a significant association between age group and mortality (p < .001). Hypertension is the most prevalent disease in both men and women. Higher levels of creatinine, urea, fasting blood sugar, neutrophils, cholesterol, and LDL significantly increase and higher levels of lymphocytes, platelets, RBCs, hemoglobin, and triglyceride significantly decrease the mortality.

CONCLUSION

There are no sex differences in mortality after first-ever stroke. Elderly patients need more support and attention due to greater stroke mortality. Complete blood count, lipid profile and blood levels of urea, creatinine, and fasting blood sugar may be useful in predicting mortality after first-ever stroke.

Stroke Risk in Atrial Fibrillation: Beyond the CHA2DS2-VASc Score

PURPOSE OF REVIEW

Stroke prevention is the cornerstone of atrial fibrillation (AF) management and the anticoagulation decision is currently based on CHA₂DS₂-VASc risk score. We discuss several novel risk factors besides those included in CHA₂DS₂-VASc score and alternative models for stroke prediction.

RECENT FINDINGS

Several clinical markers including obstructive sleep apnea and renal failure, laboratory markers like brain natriuretic peptide, imaging criteria including left atrial appendage morphology, spontaneous echo contrast, and coronary artery calcium score may predict stroke in AF patients. Addition of African American ethnicity to CHA₂DS₂-VASc score also improves stroke prediction in AF. Finally, novel models including TIMI-AF score, ATRIA score, and GARFIELD-AF scores have potential roles in risk stratification for stroke. While CHA₂DS₂-VASc score is the currently recommended risk stratification model for stroke prediction in AF, use of additional clinical, laboratory, imaging markers, ethnicity, and novel stroke prediction models may further assist in decision to anticoagulate the AF patient for stroke prevention.

Prognostic Value of Inflammatory and Cardiovascular Biomarkers for Prediction of 90-Day All-Cause Mortality after Acute Ischemic Stroke—Results from the Linz Stroke Unit Study

BACKGROUND

Early outcome prediction after acute ischemic stroke is of great interest. The aim of our study was to evaluate the prognostic value of blood biomarkers in patients with acute ischemic stroke.

METHODS

We measured interleukin-6 (IL-6), d-dimer, amino-terminal pro–B-type natriuretic peptide (NT-proBNP), high-sensitivity cardiac troponin T, and soluble ST2 plasma concentrations within 24 h after admission to our stroke unit in 721 consecutive acute ischemic stroke patients. End point was 90-day all-cause mortality.

RESULTS

During follow-up 81 patients died (11%). In univariate Cox proportional hazards regression analyses with the biochemical markers dichotomized according to median values, all baseline blood biomarkers were strong prognostic markers. However, in the multivariate analysis after adjustment for several clinical variables and the NIH Stroke Scale (NIHSS), only NIHSS &gt;3 [risk ratio (RR) 7.87, 95% CI, 3.61–17.16; P &lt; 0.001], IL-6 &gt; 7 pg/mL (RR 4.09, 95% CI, 2.02–8.29; P &lt; 0.001), and NT-proBNP &gt;447 ng/L (RR 4.88, 95% CI, 2.41–9.88; P &lt; 0.001) remained independent predictors. Using a simple multimarker approach combining these 3 complementary markers, we demonstrated that patients with increased NIHSS, IL-6, and NT-proBNP had the poorest outcome with a mortality rate of 38%, whereas no patient with negative readings for all 3 markers died during follow-up.

CONCLUSIONS

In this large cohort of patients with acute ischemic stroke, IL-6 and NT-proBNP at admission were strong and independent prognostic markers for 90-day all-cause mortality, and provided complementary prognostic information to the routinely used stroke severity score NIHSS.