Biomarkers of cholesterol homeostasis in a clinical laboratory database sample comprising 667,718 patients.

BACKGROUND: Circulating noncholesterol sterols/stanols (NCS) are used in clinical lipidology as surrogate measures of cholesterol synthesis and absorption, where they can be valuable tools in assessing cholesterol metabolism and personalizing therapies in patients with dyslipidemia. OBJECTIVES: To describe the distributions of plasma NCS concentrations and inter-NCS correlations in a large cohort of American patients constituting a clinical laboratory database, and to investigate the relationship between circulating NCS, age, sex, and apolipoprotein E (APOE) genotype. METHODS: A total of 667,718 patient blood samples submitted for testing to Health Diagnostic Laboratory, Inc. (Richmond, VA) were analyzed for cholesterol absorption markers (sitosterol, campesterol, and cholestanol) and one cholesterol synthesis marker (desmosterol). NCS percentiles were determined, along with intermarker correlations (Pearson's R). Analysis of variance was used to assess the effect of age and sex on NCS level, and to evaluate the relationship between cholesterol synthesis/absorption status and APOE genotype in a subset of 336,866 patients. RESULTS: Mean NCS concentrations were: sitosterol, 2.45 µg/mL; campesterol, 3.3 µg/mL; cholestanol, 2.92 µg/mL; and desmosterol 0.99 µg/mL. The correlations between each NCS and its ratio to total cholesterol ranged from 0.72 (cholestanol) to 0.94 (desmosterol). NCS levels were significantly affected by age and sex (P < .0001), and prevalence of cholesterol hyperabsorption was higher in APOE ε4 allele carriers compared with the other APOE genotypes. CONCLUSIONS: We have described sample distributions of NCS biomarkers and characterized their relationship to age, sex, and APOE genotype. These data may facilitate research into altered cholesterol homeostasis and human disease, and help physicians optimize lipid-lowering therapies.

Discordance between apolipoprotein B and low-density lipoprotein particle number is associated with insulin resistance in clinical practice.

BACKGROUND: Discordance between measures of atherogenic lipoprotein particle number (apolipoprotein B [ApoB] and low-density lipoprotein [LDL] particle number by nuclear magnetic resonance spectroscopy [LDL-PNMR]) is not well understood. Appropriate treatment considerations in such cases are unclear. OBJECTIVES: To assess discordance between apoB determined by immunoassay and LDL-PNMR in routine clinical practice, and to characterize biomarker profiles and other clinical characteristics of patients identified as discordant. METHODS: Two retrospective cohorts were evaluated. First, 412,013 patients with laboratory testing performed by Health Diagnostic Laboratory, Inc., as part of routine care; and second, 1411 consecutive patients presenting for risk assessment/reduction at 6 US outpatient clinics. Discordance was quantified as a percentile difference (LDL-PNMR percentile - apoB percentile) and attainment of percentile cutpoints (LDL-PNMR ≥ 1073 nmol/L or apoB ≥ 69 mg/dL). A wide range of cardiovascular risk factors were compared. RESULTS: ApoB and LDL-PNMR values were highly correlated (R(2) = 0.79), although substantial discordance was observed. Similar numbers of patients were identified as at-risk by LDL-PNMR when apoB levels were < 69 mg/dL (5%-6%) and by apoB values when LDL-PNMR was < 1073 nmol/L (6%-7%). Discordance (LDL-PNMR > apoB) was associated with insulin resistance, smaller LDL particle size, increased systemic inflammation, and low circulating levels of "traditional" lipids, whereas discordance (apoB > LDL-PNMR) was associated with larger LDL particle size, and elevated levels of lipoprotein(a) and lipoprotein-associated phospholipase A2 (Lp-PLA2). CONCLUSION: Discordance between apoB and LDL-PNMR in routine clinical practice is more widespread than currently recognized and may be associated with insulin resistance.

Comprehensive biomarker testing of glycemia, insulin resistance, and beta cell function has greater sensitivity to detect diabetes risk than fasting glucose and HbA1c and is associated with improved glycemic control in clinical practice.

Blood-based biomarker testing of insulin resistance (IR) and beta cell dysfunction may identify diabetes risk earlier than current glycemia-based approaches. This retrospective cohort study assessed 1,687 US patients at risk for cardiovascular disease (CVD) under routine clinical care with a comprehensive panel of 19 biomarkers and derived factors related to IR, beta cell function, and glycemic control. The mean age was 53 ± 15, 42 % were male, and 25 % had glycemic indicators consistent with prediabetes. An additional 45 % of the patients who had normal glycemic indicators were identified with IR or beta cell abnormalities. After 5.3 months of median follow-up, significantly more patients had improved than worsened their glycemic status in the prediabetic category (35 vs. 9 %; P < 0.0001) and in the "high normal" category (HbA1c values of 5.5-5.6; 56 vs. 18 %, p < 0.0001). Biomarker testing can identify IR early, enable and inform treatment, and improve glycemic control in a high proportion of patients.

An automated dental caries detection and scoring system for optical images of tooth occlusal surface.

Dental caries are one of the most prevalent chronic diseases. The management of dental caries demands detection of carious lesions at early stages. This study aims to design an automated system to detect and score caries lesions based on optical images of the occlusal tooth surface according to the International Caries Detection and Assessment System (ICDAS) guidelines. The system detects the tooth boundaries and irregular regions, and extracts 77 features from each image. These features include statistical measures of color space, grayscale image, as well as Wavelet Transform and Fourier Transform based features. Used in this study were 88 occlusal surface photographs of extracted teeth examined and scored by ICDAS experts. Seven ICDAS codes which show the different stages in caries development were collapsed into three classes: score 0, scores 1 and 2, and scores 3 to 6. The system shows accuracy of 86.3%, specificity of 91.7%, and sensitivity of 83.0% in ten-fold cross validation in classification of the tooth images. While the system needs further improvement and validation using larger datasets, it presents promising potential for clinical diagnostics with high accuracy and minimal cost. This is a notable advantage over existing systems requiring expensive imaging and external hardware.