Markers of Kidney Function in Early Childhood and Association With Maternal Comorbidity

Abstract

Importance: Kidney functional capacity is low at birth but doubles during the first 2 weeks of life and reaches near-adult levels at age 1 to 2 years. Existing reference intervals for markers of kidney function in newborns are mostly based on preterm newborns, newborns with illness, or small cohorts of term newborns, and the consequences of maternal comorbidities for newborn kidney function are sparsely described.

Objective: To establish robust reference intervals for creatinine and urea in healthy children in early childhood and to assess whether maternal comorbidity is associated with newborn creatinine and urea concentrations.

Design, setting, and participants: This multicenter, prospective, population-based cohort study assessed data and umbilical cord blood samples from participants in the Copenhagen Baby Heart Study (CBHS) who were born between April 1, 2016, and October 31, 2018, and venous blood samples from a subsample of CBHS participants who were enrolled in the COMPARE study between May 3, 2017, and November 4, 2018. Cord blood samples of 13 354 newborns from the CBHS and corresponding venous blood samples of 444 of those newborns from the COMPARE study were included. Blood samples were collected at birth, age 2 months, and age 14 to 16 months, with follow-up completed on February 12, 2020. Healthy nonadmitted term newborns from maternity wards at 3 hospitals in the Capital Region of Denmark were included.

Exposures: Maternal comorbidity.

Main outcomes and measures: Creatinine and urea concentrations.

Results: Among 13 354 newborns in the CBHS cohort, characteristics of 12 938 children were stratified by sex and gestational age (GA). Of those, 6567 children (50.8%) were male; 5259 children (40.6%) were born at 37 to 39 weeks' GA, and 7679 children (59.4%) were born at 40 to 42 weeks' GA. Compared with children born at 40 to 42 weeks' GA, those born at 37 to 39 weeks' GA had lower birth weight, Apgar scores at 5 minutes, placental weight, and placental-fetal weight ratio. Children born at 37 to 39 weeks' GA vs those born at 40 to 42 weeks' GA were more frequently small for GA at birth and more likely to have placental insufficiency and exposure to maternal preeclampsia, maternal diabetes, maternal kidney disease, and maternal hypertension. Among children born at 37 to 39 weeks' GA, reference intervals were 0.54 to 1.08 mg/dL for creatinine and 5.32 to 14.67 mg/dL for urea; among children born at 40 to 42 weeks' GA, reference intervals were 0.57 to 1.19 mg/dL for creatinine and 5.60 to 14.85 mg/dL for urea. At birth, multifactorially adjusted odds ratios among children exposed to preeclampsia were 9.40 (95% CI, 1.68-52.54) for a venous creatinine concentration higher than the upper reference limit, 4.29 (95% CI, 1.32-13.93) for a venous creatinine concentration higher than the 90th percentile, and 3.10 (95% CI, 1.14-8.46) for a venous creatinine concentration higher than the 80th percentile.

Conclusions and relevance: In this study, improved reference intervals for creatinine and urea concentrations were generated. Preeclampsia was associated with an increased risk of high newborn creatinine concentrations, suggesting that newborns of mothers with preeclampsia need closer observation of their kidney function.

Figure 1.. Creatinine and Urea Concentrations in Maternal Venous, Umbilical Cord, and Newborn Venous Blood at Birth

The area under the density curve represents probability and therefore sums to 1. The scale of the y-axis in each density plot was determined by the range of the x-axis and ranges from 1 to infinity. P values were derived from unpaired t tests comparing maternal venous blood with newborn venous blood and umbilical cord blood. Paired t tests were used to compare newborn venous blood with umbilical cord blood. Equations used in the linear regression modeling of newborn venous blood as a function of umbilical cord blood, stratified by gestational age, are provided in the Results section under Comparison of Umbilical Cord Blood, Maternal Venous Blood, and Newborn Venous Blood at Birth.

Figure 2.. Development of Creatinine and Urea Concentrations in Venous Blood in Early Childhood

The area under the density curve represents probability and therefore sums to 1. The scale of the y-axis in each density plot was determined by the range of the x-axis and ranges from 1 to infinity. P values were derived from unpaired t tests.

Figure 3.. Association of Maternal Comorbidities With Risk of High Neonatal Creatinine Concentrations

Odds ratios (ORs) of a newborn having an umbilical cord blood creatinine concentration higher than the upper reference limit (0.98 mg/dL) as a function of maternal comorbidities. The ORs, 95% CIs, and P values were derived from logistic regression analysis. Individuals represent the number of newborns with and without recorded maternal comorbidity. Events represent the number of newborns with a creatinine concentration higher than the upper reference limit. In the multifactorially adjusted model, ORs were adjusted for maternal age at birth, child’s sex, gestational age, birth weight, birth length, multiple births, and placental-fetal weight ratio higher or lower than the median. BMI indicates body mass index (calculated as weight in kilograms divided by height in meters squared).