Perinatal factors contributing to chronic kidney disease in a cohort of Japanese children with very low birth weight

Abstract

Background: Developmental programming of chronic kidney disease (CKD) in young adults is linked to preterm birth and intrauterine growth restriction (IUGR). Which confers a higher risk of progression to chronic kidney damage in children with very low birth weight (VLBW; born weighing < 1500 g): prematurity or IUGR?

Methods: This is a national historical cohort study of children with VLBW cared for in perinatal medical centers in Japan. Predictive factors included three latent variables (prematurity, IUGR, stress during neonatal period) and eight observed variables (gestational age, birth weight Z-score, maternal age, duration of treatment with antibiotics and diuretics, maternal smoking, late-onset circulatory collapse, kidney dysfunction) during the perinatal period. The primary endpoint was estimated glomerular filtration rate (eGFR) at age ≥ 3 years. A structural equation model was used to examine the pathologic constitution.

Results: The 446 children with VLBW included 253 boys and 193 girls, of mean age 5.8 ± 2.6 years and mean eGFR 111.7 ml/min/1.73 m² at last encounter. Pathway analyses showed intrauterine malnutrition (β = 0.85) contributed more to chronic kidney damage than stress during the neonatal period (β = - 0.19) and prematurity (β = 0.12), and kidney dysfunction and late-onset circulatory collapse were important observed variables in stress during the neonatal period.

Conclusions: IUGR was more harmful to future kidneys of VLBW neonates. Neonatal kidney dysfunction and late-onset circulatory collapse were important risk factors for subsequent CKD development. This emphasizes the need for obstetricians to monitor for fetal growth restriction and neonatologists to minimize neonatal stress to prevent CKD in later life.

Keywords: Chronic kidney disease; Intrauterine growth restriction; Maternal smoking; Neonatal events; Prematurity; Very low birth weight.

Fig. 1

Flowchart illustrating the inclusion/exclusion of individuals in the study. The second questionnaire identified 675 children at 15 institutions. Of these, 446 children, 253 boys and 193 girls, with VLBW fulfilled the eligibility criteria and were included in this study

Fig. 2

Results of the structural equation model with eGFR as a dependent variable. The model fit in the pathway analyses was adequate (CFI 0.959, RMSEA 0.041, and CMIN/df 1.734). The pathways were determined using exploratory methods, in addition to the standardized regression coefficients. IUGR (β = 0.85) contributed more to childhood kidney damage than did stress during the neonatal period (β = -0.19) and prematurity (β = 0.12). Gestational age may be an indicator of prematurity; the Z-score of birth weight, maternal age, and maternal smoking may be indicators of IUGR; and antibiotic dosing periods, diuretic dosing periods, late-onset circulatory collapse, and neonatal kidney dysfunction may be indicators of neonatal stress. The definition of a Z-score is the position of a raw score in terms of its distance from the mean, measured in standard deviation units.