LGA / Total Births (%)*
LOS Days *
Male
AVE GA Weeks
TABLE 1 : LGA Infant and Maternal Data
Cesarean Section *
LGA Infant Data Z Score **
Apgar 1 min
Apgar 5 min
NICU
Need IV
PC 495 / 6611 ( 7.5 ) 6 ± 11 63 % 38 ± 3 50 % 1.83 ± 0.65 7 ± 2 8 ± 1 28 % 42 % 50 % 100 % C 411 / 3815 ( 10.8 ) 5 ± 8 61 % 38 ± 2 42 % 1.65 ± 0.71 7 ± 2 8 ± 2 28 % 43 % 44 % 99 %
BMI **
BMI > 30
Hypoglycemia
Hypertension
Preeclampsia , HELLP % all hypertension
DM
LGA Maternal Data
GDM % all DM **
Type 1 % all DM
Type 2 % all DM
Survive
Tobacco ** |
Cannabis * |
Opioid |
COVID-19 |
|
|
|
+ |
PC 38.33 ± 8.81 81 % 30 % 13 % 27 % 44.4 % 29.6 % 26 % 7 % 1 % 2 % 0 C 37.61 ± 8.42 82 % 35 % 14 % 27 % 55.6 % 22.2 % 22.2 % 9 % 3 % 1 % 8 %
* p < 0.05 ; ** p < 0.01 ; LGA : large for gestational age ; PC : Pre-COVID period , January 1 , 2016-February 29 , 2020 ; LOS : length of stay ; AVE GA : Average gestational age ; NICU : Neonatal intensive care unit ; BMI : Body mass index ; HELLP : syndrome of hemolysis , elevated liver functions , low platelets ; DM : diabetes mellitus ; GDM : gestational diabetes mellitus
or SGA ( birth weight < 10th percentile for gestational age ) on the problem list in the electronic medical record ( Epic Systems , Verona , WI ). A single neonatologist maintains the problem lists for all newborns admitted to our hospital to assure consistent and accurate data . We used Cogito Slicer- Dicer , which is an Epic-built , self-service reporting tool with access to key criteria such as ICD-9 and ICD-10 codes to identify the total number of babies born and to filter which babies were diagnosed as LGA or SGA ( Epic Systems , Verona , WI ). Additional maternal and infant co-morbidities were filtered using this method . Anthropometric data were also examined by investigators to verify the accuracy of LGA and SGA diagnoses .
SAMPLE SIZE AND STATISTICAL ANALYSIS
All newborns meeting eligibility for this study were incorporated into the analyses . A total of 10,426 infants were born at our institution during the review period . This number of infants provided more than enough subjects to reach 80 % power to detect a medium effect size between groups based on the group ( PC vs C ) at the p = 0.05 significance level .
Descriptive analyses were conducted to characterize the PC and C cohorts detailed in Tables 1-2 and to explore the normative distribution of all information collected . Differences in LGA , SGA , and select maternal and infant outcomes were examined using a series of chi-square analyses . Statistical significance was established at the p < 0.05 level . SPSS version 27.0 was used to complete the analyses .
RESULTS
A total of 10,426 infants were born at our institution during the review period : 6,611 ( 63.4 %) during the PC period and 3,815 ( 36.6 %) during the C period . For the entire observation period ( PC + C ) there were 906 LGA newborns and 648 SGA newborns . More than half of the overall sample were male 908 ( 58 %), and 731 ( 48 %) were born by cesarean delivery . The mean length of stay for all LGA , and SGA newborns was 9 ± 20.4 days . The mean gestational age was 37 ± 2.9 weeks and the mean birthweight for all newborns from both groups was 3.2 ± 1.1 kg .
Of all pregnant mothers admitted to our children ’ s hospital during the PC period , 40 % ( 9280 / 22778 ) had a BMI of greater than 30 . Of those admitted during the C period , 46 % ( 6344 / 13778 ) of these pregnant women had a BMI greater than 30 ( p < 0.05 ). Nearly three-quarters of mothers delivering LGA or SGA infants during the C cohort ( 401 ; 74 %) had a BMI > 30 . This was significantly higher than mothers delivering LGA or SGA infants ( 660 ; 65 %) in the PC cohort ( p < 0.05 ). Maternal substance use was significantly higher ( p < 0.01 ) for tobacco , cannabis , and opioids in the C cohort . Twenty-six percent of mothers in the C cohort reported tobacco use ( 168 ; 26 %). This was a significantly higher proportion of individuals using tobacco in the PC cohort ( 181 ; 20 %) Similarly , maternal cannabis use was significantly higher among the C cohort ( 71 ; 11 %) compared to the PC cohort ( 50 ; 5.5 %). Finally , maternal opioid use was significantly higher in the C ( 91 ; 14 %) compared to the PC ( 41 ; 4.5 %) cohorts .
LGA NEWBORN AND MATERNAL CHARACTERISTICS BY COHORT
Table 1 provides select maternal and newborn characteristics among LGA cohorts ( PC vs C ). Significantly more newborns were born LGA in the C cohort ( 411 , 10.8 %) compared to LGA newborns reported in the PC cohort ( 495 , 7.5 %) ( p < 0.05 ). LGA comparisons between the PC and C cohorts revealed a statistically significant decrease in the proportion of cesarean births ( p < 0.05 ). The proportion of all types of maternal diabetes mellitus ( DM ) was similar for both groups but the proportion of gestational diabetes was significantly higher in the C cohort ( p < 0.01 ). Type 1 and type 2 DM were lower among mothers in the C cohort , though not statistically significant .
West Virginia Medical Journal • March 2023 • 31