Multivariable Risk Modelling and Survival Analysis with Machine Learning in SARS-CoV-2 Infection.

AIM: To evaluate the performance of a machine learning model based on demographic variables, blood tests, pre-existing comorbidities, and computed tomography(CT)-based radiomic features to predict critical outcome in patients with acute respiratory syndrome coronavirus 2 (SARS-CoV-2). METHODS: We retrospectively enrolled 694 SARS-CoV-2-positive patients. Clinical and demographic data were extracted from clinical records. Radiomic data were extracted from CT. Patients were randomized to the training (80%, n = 556) or test (20%, n = 138) dataset. The training set was used to define the association between severity of disease and comorbidities, laboratory tests, demographic, and CT-based radiomic variables, and to implement a risk-prediction model. The model was evaluated using the C statistic and Brier scores. The test set was used to assess model prediction performance. RESULTS: Patients who died (n = 157) were predominantly male (66%) over the age of 50 with median (range) C-reactive protein (CRP) = 5 [1, 37] mg/dL, lactate dehydrogenase (LDH) = 494 [141, 3631] U/I, and D-dimer = 6.006 [168, 152.015] ng/mL. Surviving patients (n = 537) had median (range) CRP = 3 [0, 27] mg/dL, LDH = 484 [78, 3.745] U/I, and D-dimer = 1.133 [96, 55.660] ng/mL. The strongest risk factors were D-dimer, age, and cardiovascular disease. The model implemented using the variables identified using the LASSO Cox regression analysis classified 90% of non-survivors as high-risk individuals in the testing dataset. In this sample, the estimated median survival in the high-risk group was 9 days (95% CI; 9-37), while the low-risk group did not reach the median survival of 50% (p < 0.001). CONCLUSIONS: A machine learning model based on combined data available on the first days of hospitalization (demographics, CT-radiomics, comorbidities, and blood biomarkers), can identify SARS-CoV-2 patients at risk of serious illness and death.

Glutathione S-transferase T1- and M1-null genotypes and coronary artery disease risk in patients with Type 2 diabetes mellitus.

INTRODUCTION: Since long-term exposure to oxidative stress is strongly implicated in the pathogenesis of diabetic complications, polymorphic genes of detoxifying enzymes must be involved in the development of coronary artery disease (CAD). We assessed the potential glutathione S-transferase (GST) gene-gene (GSTM1(null)-GSTT1(null)) and gene-smoking interactions on the development of CAD in patients with Type 2 diabetes. MATERIALS & METHODS: In a case-only design, we enrolled 231 patients with Type 2 diabetes (147 male, 66.1 +/- 9.7 years) referred to our institute for coronary angiography investigation. CAD was diagnosed if there was over 50% obstruction of one or more major vessels. RESULTS: Coronary angiography revealed significant CAD in 184 patients (80%). Male gender (p < 0.001), smoking habits (p = 0.003) and GSTT1(null) genotype (p = 0.003) were significantly correlated with the increasing extent of the coronary atherosclerosis. Case-only analysis revealed that patients with both M(null)-T(null) genotypes had the highest risk for 3-vessel CAD compared with patients who express both GST genes (odds ratio: 3.1; 95% confidence interval: 1.0-10.3, p = 0.04). A nearly threefold interaction existed between cigarette smoking and M(null)-T(null) genotypes (odds ratio: 2.9, 95% confidence interval: 1.7-7.8, p = 0.03). A significant interaction between M(null)-T(null) genotypes and smoking was also observed on the increasing number of coronary vessels that were diseased (chi(2) = 14.0; p = 0.03). CONCLUSION: These data suggest that polymorphisms in GSTM1 and GSTT1 genes are risk factors for CAD in Type 2 diabetic patients, especially among smokers. These genetic markers may permit the targeting of preventive and early intervention on high-risk patients to reduce their cardiovascular risk.

Effect of fluoride application on shear bond strength of brackets bonded with a resin-modified glass-ionomer.

PURPOSE: The purpose of this study was to assess the effect of fluoride application at 3 different steps of the bonding process on the shear bond strength and bond failure site of a resin-modified glass ionomer cement. MATERIAL: Sixty stainless steel brackets were bonded to bovine incisors with Fuji Ortho LC (GC Europe, Leuven, Belgium) under 4 different enamel conditions: (1) uncontaminated enamel, (2) enamel precleaned with fluoride-containing prophylaxis paste, (3) 1.1% acidulated phosphate fluoride (APF) gel applied immediately before conditioning, and (4) 1.1% APF gel applied immediately before bonding. After bonding, all samples were stored in distilled water for 24 hours and subsequently tested in shear mode on a universal testing machine. RESULTS: No significant differences were found between groups 1 (uncontaminated enamel) and 2 (fluoride application during prophylaxis). Both groups showed significantly greater shear bond strength values than groups 3 (fluoride application before conditioning) and 4 (fluoride application before bonding). Groups 3 and 4 did not differ significantly. Moreover, no significant differences in debond locations were found among the 4 groups. CONCLUSIONS: Fluoride application during initial prophylaxis does not affect the bond strength values of Fuji Ortho LC, whereas it significantly lowers bond strength values when applied before both conditioning and bonding.

Anthropometric indices of school children and familiar risk factors.

BACKGROUND: Overweight in adolescence predicts adverse health effects in adulthood. We carried out a primary school health program and assessed children's growth and body composition. METHODS: Were screened 869 (448 M, 421 F) primary school children: height, weight, four skinfolds, and four circumferences were measured. A family-reported questionnaire was used to determine family composition, history, and lifestyle. RESULTS: Age was 118 +/- 5 months, BMI 18 +/- 3 kg/m(2). No difference by gender was observed as for BMI or blood pressure. Girls had higher skinfold thickness at the biceps (BCF), triceps (TCF), subscapular (SSF), and suprailiac (SIF) areas (P < 0.001), hip and thigh circumferences (P < 0.01), body fat percentage (P < 0.001). Boys had higher waist circumference (P < 0.01), waist/thigh ratio, and conicity index (P < 0.001). Offspring BMI was correlated with birth weight (P < 0.05), parental BMI and scholarship level (P < 0.001), children blood pressure (P < 0.001), and hours per day spent in television viewing (P < 0.01). Family history for diabetes was associated with higher BMI, SSF, waist circumference (P < 0.05), and upper thigh (P < 0.01). Family history for hypertension was associated with higher SSF/TCF ratio (P < 0.05). CONCLUSIONS: Three of 869 children had BMI >30 kg/m(2) (2 boys and 1 girl), 33 had BMI >25 kg/m(2) (17 boys and 16 girls). The percentages of children who could be considered overweight (BMI >/=95(th) percentile of age- and sex-specific NHANES I reference data) were boys, 10.0%, and girls, 9.3%. Anthropometric and anamnestic data on child and family yield more accurate estimates of risk profile: fat distribution seems relevant for metabolic and cardiovascular disorders.