Selecting pathologic measures of placental inflammation that affect pregnancy and childhood health is largely empirical. We aimed to systematically select several core inflammation-related placental measures to construct a novel placental inflammatory evaluation criterion with a high prognostic relevance to child morbidity.
We used data from the US Collaborative Perinatal Project (1959-1976), a longitudinal birth cohort study that recruited women during pregnancy and followed the children until 7 years of age. Bootstrap resampling, least absolute shrinkage and selection operator, and receiver-operator curve were used to select placental pathologic measures that were closely related to child morbidity to form a placental inflammatory composite index.
Twenty-six candidate placental inflammation-related measures were ranked based on their close association with adverse neonatal outcomes. The top five placental measures were: (i) neutrophilic infiltration in umbilical artery; (ii) placental weight-birthweight ratio; (iii) necrosis in decidua capsularis; (iv) bacterial colony in epithelium of amnion; and (v) opacity of membranes and fetal surface. Several composite indexes were constructed. A five-measure composite index that had the highest prognostic relevance was chosen. Compared with subjects without any of the five abnormal measures, those with any lesion ranging from 1 to 5 had a 1.2- to 4.6-fold risk of adverse child outcomes, respectively.
Our composite index is simple, evidence-based, and has predictive value for child morbidity. It may be used as a novel placental inflammatory evaluation criterion.
8 Figures and Tables
Figure 1 Study subjects flow chart.
Table 1 Definitions of placental lesions related to intrauterine infection, the Collaborative Perinatal Project, 1959–1976
Table 2 Definitions of adverse outcomes used to develop evaluation systems, the Collaborative Perinatal Project, 1959–1976
Figure 2 Association between the number of (•) areas under the curve (AUC) > 0.7 and the number of measures in placental index.
Table 3 Number of times that the top 15 measures were retained by LASSO procedure using six adverse outcomes, the Collaborative Perinatal Project, 1959–1976
Figure 3 Nonlinear relation between the points of the five-measure evaluation system and the probability of composite disease. (•) Probability. (•) Upper confidence interval (CL). (•) Lower CL.
Table 4 Detailed combinations of various placental composite indexes constructed by various placental measures, the Collaborative Perinatal Project, 1959–1976
Table 6 Risk of adverse childhood morbidity by scoring of five-measure composite index, the Collaborative Perinatal Project, 1959–1976
Download Full PDF Version (Non-Commercial Use)