High Aspartate Aminotransferase/Alanine Aminotransferase Ratio May Be Associated with All-Cause Mortality in the Elderly: A Retrospective Cohort Study Using Artificial Intelligence and Conventional Analysis

Deep learning from head CT scans to predict elevated intracranial pressure

BACKGROUND AND PURPOSE

Elevated intracranial pressure (ICP) resulting from severe head injury or stroke poses a risk of secondary brain injury that requires neurosurgical intervention. However, currently available noninvasive monitoring techniques for predicting ICP are not sufficiently advanced. We aimed to develop a minimally invasive ICP prediction model using simple CT images to prevent secondary brain injury caused by elevated ICP.

METHODS

We used the following three methods to determine the presence or absence of elevated ICP using midbrain-level CT images: (1) a deep learning model created using the Python (PY) programming language; (2) a model based on cistern narrowing and scaling of brainstem deformities and presence of hydrocephalus, analyzed using the statistical tool Prediction One (PO); and (3) identification of ICP by senior residents (SRs). We compared the accuracy of the validation and test data using fivefold cross-validation and visualized or quantified the areas of interest in the models.

RESULTS

The accuracy of the validation data for the PY, PO, and SR methods was 83.68% (83.42%-85.13%), 85.71% (73.81%-88.10%), and 66.67% (55.96%-72.62%), respectively. Significant differences in accuracy were observed between the PY and SR methods. Test data accuracy was 77.27% (70.45%-77.2%), 84.09% (75.00%-85.23%), and 61.36% (56.82%-68.18%), respectively.

CONCLUSIONS

Overall, the outcomes suggest that these newly developed models may be valuable tools for the rapid and accurate detection of elevated ICP in clinical practice. These models can easily be applied to other sites, as a single CT image at the midbrain level can provide a highly accurate diagnosis.

Prognostic Biomarkers of Mortality in Older Patients Without Cancer in Home Healthcare

Introduction The number of patients without cancer who receive home healthcare is increasing; however, prognostic prediction is challenging among them. This study aimed to investigate the prognostic value of generic biomarkers for mortality in patients without cancer who receive home healthcare. Materials and methods The multicenter retrospective cohort study included 114 older patients without cancer, of which 12 (10.5%) died during the study period. The median (interquartile range (IQR)) of the study observation period was 181 (49-293) days. Generic biomarkers included hemoglobin (Hb), albumin (Alb), C-reactive protein (CRP), estimated glomerular filtration rate (eGFR), aspartate aminotransferase (AST), and alanine aminotransferase (ALT). A multivariate-adjusted Cox proportional hazard model on all-cause mortality was used to calculate hazard ratio (HR) and 95% confidence interval (95% CI) for each biomarker. The cut-off values of each biomarker were calculated by receiver operating characteristic curve analysis. The performance of cut-off values was evaluated by time-dependent area under the curves (AUCs). Results The median (IQR) of AST was 13 (10-21) U/L. The biomarkers significantly predictive of mortality were Hb (fully adjusted HR: 0.41; 95% Cl: 0.25 - 0.70), Alb (HR: 0.41; 95% Cl: 0.02 - 0.69), and AST (HR: 1.09; 95% Cl: 1.00 - 1.18), along with male sex (HR: 4.07; 95% Cl: 1.15 - 14.35). The AUC of a cut-off value of AST (> 31 U/L) at 360 days was 0.72 (95% CI 0.71 - 0.72; p < 0.01), which outperformed the AUCs for Hb and Alb. Conclusion AST, in addition to Hb and Alb, may be useful for predicting the prognosis of older patients without cancer, who had a normal-to-mild increased level of AST, in home healthcare settings. Larger-sample and longer follow-up studies will be warranted.

Serum mAST/ALT ratio had high predictive value for adverse outcome of severe fever with thrombocytopenia syndrome with severe condition

BACKGROUND

Severe fever with thrombocytopenia syndrome (SFTS) usually demonstrates multi-organ injury with a high mortality rate. This study aimed to investigate associations of serum aspartate/alanine aminotransferase (AST)/ALT, cytosolic AST (cAST)/ALT and mitochondrial AST (mAST)/ALT ratios with the prognosis of SFTS patients.

METHODS

A total of 355 confirmed SFTS patients were included. Clinical and laboratory data were compared between survivors and nonsurvivors. Logistic regression analysis was used to assess the independent risk factors for fatality in all patients and those admitted to the intensive care unit (ICU). The predictive values of the risk factors and constructed risk models were evaluated.

RESULTS

Mean age and biochemical parameters were significantly greater in nonsurvivors than in survivors. In ICU patients, the three ratios, high-sensitivity troponin I (hsTnI), creatine kinase (CK), lactate dehydrogenase (LDH) and α-hydroxybutyrate dehydrogenase (α-HBDH) were elevated markedly in nonsurvivors than in survivors. Multivariate logistic regression analysis showed that age, three ratios and α-HBDH were independent risk factors for mortality in all patients. Only the three ratios were independent risk factors for death in ICU patients. Risk Models (M1, M2 and M3) and simplified models (sMs) containing the three ratios respectively had comparatively high predictive values for fatality in all patients with area under ROC curves (AUCs) > 0.85. In ICU patients, mAST/ALT ratio had the highest predictive value, sensitivity and odds ratio (OR) for mortality among three ratios.

CONCLUSION

AST/ALT, cAST/ALT and mAST/ALT ratios were associated with unfavorable clinical outcome of SFTS. The prognostic value of mAST/ALT ratio was higher in severe cases.

Prognostic value of De Ritis ratio in patients with acute myocardial infarction

BACKGROUND

The association between aspartate aminotransferase to alanine aminotransferase ratio (De Ritis ratio) and prognosis of patients with acute myocardial infarction (AMI) remains poorly investigated.

METHODS

This study included 3000 patients with AMI undergoing percutaneous coronary intervention. Patients were categorized in groups according to tertiles of the De Ritis ratio: tertile 1 (De Ritis ratio < 1.11; 905 patients), tertile 2 (De Ritis ratio 1.11 to 1.95; 1003 patients) and tertile 3 (De Ritis ratio > 1.95; 1002 patients). The primary endpoint was 3-year all-cause mortality.

RESULTS

At 3 years, all-cause deaths occurred in 487 patients: 119 deaths (13.2%) in patients of 1st tertile, 164 deaths (17.8%) in patients of 2nd tertile and 204 deaths (21.9%) in patients of 3rd tertile of the De Ritis ratio (adjusted hazard ratio [HR] = 1.16, 95% confidence interval [CI] 1.02 to 1.31], P = 0.023 per unit increment in the logarithmic scale of the De Ritis ratio); cardiac deaths occurred in 328 patients: 76 deaths (8.2%) in patients of 1st tertile, 110 deaths (12.0%) in patients of 2nd tertile and 142 deaths (15.4%) in patients of 3rd tertile of the De Ritis ratio (adjusted HR = 1.20 [1.04-1.40], P = 0.014 per unit increment in the logarithmic scale of De Ritis ratio). The C-statistic of the multivariable model(s) with baseline data without and with De Ritis ratio was 0.822 [0.805-0.839] and 0.823 [0.805-0.840], (P = 0.419) for all-cause mortality and 0.831[0.811-0.852] and 0.832 [0.811-0.853], P = 0.621) for cardiac mortality.

CONCLUSIONS

In patients with AMI, elevated De Ritis ratio was associated with increased risk of 3-year mortality.

Elevated All-Cause Mortality among Overweight Older People: AI Predicts a High Normal Weight Is Optimal

It has been proposed that being overweight may provide an advantage with respect to mortality in older people, although this has not been investigated fully. Therefore, to confirm that and elucidate the underlying mechanism, we investigated mortality in older people using explainable artificial intelligence (AI) with the gradient-boosting algorithm XGboost. Baseline body mass indexes (BMIs) of 5699 people (79.3 ± 3.9 years) were evaluated to determine the relationship with all-cause mortality over eight years. In the unadjusted model, the first negative (protective) BMI range for mortality was 25.9−28.4 kg/m2. However, in the adjusted cross-validation model, this range was 22.7−23.6 kg/m2; the second and third negative BMI ranges were then 25.8−28.2 and 24.6−25.8 kg/m2, respectively. Conversely, the first advancing BMI range was 12.8−18.7 kg/m2, which did not vary across conditions with high feature importance. Actual and predicted mortality rates in participants aged <90 years showed a negative-linear or L-shaped relationship with BMI, whereas predicted mortality rates in men aged ≥90 years showed a blunt U-shaped relationship. In conclusion, AI predicted that being overweight may not be an optimal condition with regard to all-cause mortality in older adults. Instead, it may be that a high normal weight is optimal, though this may vary according to the age and sex.