No abstract

This study expands a new competing-risks model for the prediction of a small-for-gestational-age (SGA) neonate using maternal demographic characteristics and medical history and second-trimester fetal biometry. This approach involves a joint prior distribution of gestational age at delivery and birth-weight Z-score, updated by the biomarkers' likelihood according to Bayes' theorem. Estimated fetal weight (EFW) was expressed conditionally to gestational age at delivery and birth-weight Z-score. The association between EFW and birth weight was steeper for earlier gestations. The prediction of SGA was better for increasing degree of prematurity and greater severity of smallness. What are the clinical implications of this work?A competing-risks model using maternal demographic characteristics and medical history and second-trimester fetal biometry provides effective risk stratification for a SGA neonate.

AbstractThe analyses of 18 biochemical parameters (alanine aminotransferase, albumin, aspartate aminotransferase, calcium, cholesterol, chloride, creatinine, iron, glucose, γ- glutamyl transferase, alkaline phosphatase, phosphorus, potassium, sodium, total protein, triglycerides, uric acid, and urea nitrogen) were performed for 166 healthy individuals and 108 patients with end-stage renal failure (ESRF). The application of cluster analysis proved that there were points of similarity among all 18 biochemical parameters that formed major groups; these groups corresponded to the authors’ assumption of the existence of several overall patterns of biochemical parameters that may be termed “enzyme-specific”; “general health indicator”; “major component excretion”; “blood-specific indicator”; and “protein-specific”. These patterns also appear in the subsets of males and females that were obtained by separation of the general dataset. In addition, the performance of factor analysis similarly proved the validity of this assumption. This projection and modelling method indicated the existence of seven latent factors, which explained 70.05% of the total variance in the system for healthy individuals and more than 72% of the total variance in the system for patients with ESRF. All these results support the probability that a general health indicator could be constructed by taking into account the existing classification groups in the list of biochemical parameters.

This study provides models for assessment and modeling of routinely used biochemical laboratory data, finding groups of similarity among clinical tests usually determined on HIs and ESRF patients, contributing in data mining and reducing costs.