iDEA: Drexel E-repository and Archives >
Drexel Theses and Dissertations >
Drexel Theses and Dissertations >
New preterm infant growth curves influence of gender and race on birth size
Please use this identifier to cite or link to this item:
|Title: ||New preterm infant growth curves influence of gender and race on birth size|
|Authors: ||Groveman, Sue A.|
Premature infants -- Growth
|Issue Date: ||12-Sep-2008 |
Adequate fetal and postnatal growth is crucial to the health of infants and significant to their future adult health. Preterm infant growth curves are used in the assessment of fetal and postnatal growth. Current growth curves, such as the Lubchenc curves, are limited by older homogenous data sets, smaller sample sizes, varying age ranges, combined-gender curves and/or disparate data sources. Evaluation of these curves was needed to determine whether a new set of curves with updated high-risk percentile classification cutoffs for preterm infants was needed. The presence of birth size differences between genders and racial groups also was important to identify and explore.
In the first specific aim, smoothed growth curves were created for males and females for weight, length and head circumference using a large data set (Pediatrix Medical Group). The original data included 391,681 infants (56% male; 52% white, 23% Hispanic, 15% black, 10% other) ranging from 22 to 42 weeks gestation. Infants of nonsingleton pregnancies or with ambiguous gender, congenital anomalies or physiologically improbable growth measurements were excluded. The curves were fit to the data using the LMS method. The goodness-of-fit was assessed using worm plots, z-scores, and visual inspection. The curves were validated through the evaluation of z-scores anddistribution of infants between the high-risk percentile-for-age classifications. The new curves were compared to the Lubchenco curves visually, at selected points, and via the percentages of small-for-gestational-age (SGA) and large-for-gestational-age (LGA) infants.
The second specific aim investigated size differences between males and females and among racial groups. Gender differences in birth size were compared via ANOVAs, overlaid curves, specific points on the new Pediatrix curves, and distribution of infants between the high-risk percentile-for-age classifications. Racial differences in birth size were compared via ANOVAs, z-score comparisons, distribution of infants between the high-risk percentile-for-age classifications, and odds ratios. Several maternal characteristics were examined to explain the differences found in birth size by racial group (logistic regression). The sickness level of the infants was investigated to help explain group differences.
For the first specific aim, the new Pediatrix percentiles were found to have smaller measurements than the Lubchenco curves until about 30 weeks, were somewhat similar between about 30 and 36 weeks, and were larger after 36 weeks. Overall the Lubchenco curves misclassified 15% of males and 10% of females as SGA, appropriatefor- gestational age (AGA), or LGA according to the new Pediatrix curves.
For the second specific aim, males were found to be significantly larger than females in weight, length, and head circumference. A higher percentage of black infants were born earlier and were significantly smaller than Hispanic and white infants. Relative to the medians, depending on gestational age, the z-scores of black infants placed up to19 percentiles below white infants. At 32-41 weeks, black infants had two to three times the risk of white infants of being born SGA. Controlling for race, preeclampsia/eclampsia was a consistent predictor of SGA for weight, length and head circumference at all age groups and smoking was a predictor for infants born at 32-41 weeks. The prevalence of being sick (APGAR at 1 minute < 3) at birth was greater in black infants and those classified as SGA for weight.
Accurate preterm growth curves are crucial to the assessment of growth status and therefore are vital to the health of infants. This study found strong evidence for the replacement of the Lubchenco growth curves with contemporary gender-specific curves. This study also found significant differences in birth size (weight, length, and head circumference) between male and female infants and among the racial groups that warrant further investigation.|
|Appears in Collections:||Drexel Theses and Dissertations|
Items in iDEA are protected by copyright, with all rights reserved, unless otherwise indicated.