Effects of Perinatal Variables on Echocardiographic Assessments of Left Ventricular Dimensions in Infants Born Large for Gestational Age: A Prospective Cohort Analysis

doi:10.21203/rs.3.rs-4902628/v1

Background

To assess the relationship between perinatal factors, and echocardiographic left ventricular (LV) dimensions after delivery in infants who are large for gestational age (LGA).

Methods

This is a prospective cohort study that was conducted between 2014 and 2018, and involved healthy LGA newborns born > 35 weeks’ gestation, delivered at New York-Presbyterian Brooklyn Methodist Hospital, and a control group of appropriate for gestational age (AGA) infants. Data analysis was performed using multivariate linear regression in STATA.

Results

A total of 563 neonates were enrolled in this study. They were composed of 414 AGA infants as the control group and 149 LGA infants as the intervention group. The male sex was predominant in both groups. A larger proportion of neonates were admitted to the neonatal intensive care unit (NICU) in LGA infants (74.6%) as compared to the AGA infants (33.5%) (p < 0.001). In the study's regression analysis, birth weight (BW) emerged as a key factor, positively correlating with increased LV mass, interventricular septum thickness, and LV posterior wall thickness across both LGA and AGA. Additionally, BW showed a positive correlation with left ventricular internal dimensions in diastole and systole. Higher maternal BMI was associated with an increase in fractional shortening in LGA infants, while maternal insulin use during pregnancy was positively associated with interventricular septum thickness. Notably, male infants exhibited significantly higher LV internal dimensions in both diastole and systole, while GA negatively impacted the left ventricular mass-to-volume ratio.

Conclusions

The study's findings underscore the significant influence of perinatal factors on neonatal cardiac morphology, in both LGA and AGA infants. BW, GA, gender, maternal BMI, and maternal insulin use during pregnancy were key determinants affecting various aspects of LV structure, including mass, wall thickness, and internal dimensions. These insights highlight the importance of considering these perinatal factors in the assessment and monitoring of neonatal cardiac health, offering valuable guidance for tailored clinical approaches in pediatric cardiology.

Large for Gestational Age

Neonatal

Prospective Cohort

Echocardiography

Linear Regression

Left Ventricular Dimensions

Asymmetric Septal Hypertrophy

perinatal factors

Large for gestational age (LGA) in newborns is defined as birth weight (BW) above the 90th percentile for gestational age (GA), and it has been linked to a variety of maternal risk factors, including gestational diabetes mellitus (GDM). Cardiomyopathy is a common finding in LGA infants born to mothers with GDM and principally reflects maternal hyperglycemia during pregnancy (Farrar et al., 2016). The latter is considered a teratogenic state, leading to fetal complications, including cardiomyopathy (Corrigan et al., 2009). Although this relationship has been established, many LGA infants with cardiac changes, including thickening of the inter ventricular septum (IVS) and ventricular walls particularly of the left ventricle (LV), are born to mothers without a history of GDM, and this may be related to missed cases of GDM who failed to meet the definition criteria for GDM diagnosis, level of glycemic control during pregnancy or comorbidity such as high maternal body mass index (BMI) which is considered as an independent risk factor for perinatal complications (Al-Biltagi et al., 2021).

Although fetal cardiac and vascular structural and functional changes are linked to maternal hyperglycemia (Wahab et al., 2020), little is known about cardiac development and function in human children born to mothers with high BMI. In one study, 6-month-old neonates' LV mass increased in proportion to mother’s gestational weight growth (Guzzardi et al., 2018). Additionally, maternal hyperglycemia carries prenatal and perinatal risks and long-term risks for the mother and her child.

Cardiac remodeling and cardiovascular events in children are influenced by the shape and function of the LV. The LV geometry may stigmatize the morbidity and mortality in this population, even in asymptomatic conditions, such as before the start of overt hypertension or heart failure (Wang et al., 2021).

There is limited literature about the perinatal factors that affect the cardiovascular system of the newborn. Our recently published study showed that perinatal factors are significant predictors of left ventricular parameters in small for GA babies (Elmakaty et al., 2023). Therefore, we conducted the current prospective cohort study which aimed to explore the relationship between perinatal factors, including maternal, placental, and neonatal factors, and echocardiographic LV dimensions after delivery in infants who are LGA compared with babies who are appropriate for gestational age (AGA).

Study design and setting

This is a single-centered, prospective cohort study was used to assess the relation of perinatal factors with echocardiographic LV dimensions in newborn babies. This study was carried out at New York-Presbyterian Brooklyn Methodist hospital (NYPBMH) in both neonatal intensive care (NICU) and nursery wards. This study was completed between 2014 and 2018.

Study population

The study population for the current study was newborn babies with LGA. To select the intervention and control, echocardiography was done for all selected babies with LGA based on Fenton growth charts published in 2003 and revised in 2013 (Fenton & Kim, 2013). The control group were all AGA babies who underwent Echocardiographic studies for murmur evaluation before hospital discharge. Only babies without significant heart abnormalities were included in the study.

Participants’ recruitment

At initial stage (stage 1) total 727 newborn babies were recruited, among them, 80 were excluded based on associated anomalies (n = 21), perinatal depression/low 5 min APGAR (n = 19), genetic diagnosis (n = 11), cardiac disease (n = 10), hypoxic respiratory failure (n = 9) and severe sepsis/shock (n = 5). A total of 647 participants were recruited after applying the inclusion criteria. The included participant was divided into two groups i.e., control group (babies with AGA) and the intervention group (babies with LGA). The baseline data were recorded at this stage (stage 2). At the final stage (stage 3) 22 participants were excluded from the control group and 62 were excluded from the intervention group because of missing data during the follow-up. The details shown in Fig. 1.

Ethical consideration

The current study was conducted as per the Declaration of Helsinki. The study was approved by the hospital institutional research board (IRB). No consent taken from their parents since all echo studies were routinely done on LGA infants in our NCU to rule out structural heart diseases. Confidentiality of the data was maintained by assigning a fictitious number to each participant and all the data was stored in a locked folder.

Cardiac Variables:

The echocardiography was performed by using Philips 5500 Echocardiography machine (Philips, Andover, MA, USA) to evaluate the heart per protocol, only data on LV dimensions were collected. IVS thickness during systole and diastole (IVSs and IVSd, respectively), left ventricular internal dimension (LVID) during systole and diastole (LVIDs and LVIDd, respectively), left ventricular posterior wall (LVPW) thickness at end systole and diastole (LVPWs and LVPWd, respectively), IVS/LVPW ratio, shortening fraction (FS), left ventricular volumes (LV mass [LVmass] and LV mass to volume ratio [LVmass/Vol]).

Data analysis

Microsoft Excel was used for data management (Microsoft Office, Redmond, Washington, United States) and statistical analysis was performed by using Stata version 16 (College Station, TX, USA). We used published references for normal LV cardiac parameters to determine normal and two standard deviations values for selected LV parameter (references). Asymmetric septal hypertrophy (ASH) was defined as the interventricular septum having a thickness greater than 6 mm and a ratio of septal to posterior wall thickness that is greater than 1.3 (Vela-Huerta et al., 2019). For the differences in baseline characteristics between LGA and AGA, we used two main tests. A chi-squared test was performed for the categorical variables while a simple two-way t-test was used for the continuous variables. The categorical data were presented as frequency and percentage while the continuous data were presented as a mean and standard deviation. Any missing data was removed from the analysis tables. A logistic regression model was implied to assess the association of perinatal factors with left ventricular dimensions. This model was performed over two steps. Initially a univariate linear regression was done for the continuous variables. Then, a multivariate linear regression was done for the significate variables. In the case of categorical variables such as ASH, a binary logistic regression was initially done. Consequently, a logistic regression was done for the significant variables. The P-value was stated significant if < 0.05.

A total of 563 participants were included in the final analysis among which 414 were in the control group (babies with AGA) and 149 were in the intervention group (babies with LGA). Table 1 demonstrates and compares the various neonatal and maternal variables between LGA and AGA infants. Results show that most of the participant in both groups were males (AGA = 51.45% and LGA = 61.7%). The mode of delivery was dominantly C-sections (66%) while in the AGA group, vaginal and c-section deliveries were equally distributed (50%). Among the AGA (33.5%) were admitted to neonatal intensive care unit (NICU), while in LGA there was a higher number of babies that were admitted to NICU (74.6%). The analysis of the APGAR scores at 1 minute and 5 minutes showed statistically insignificant results (p=0.081 and p=0.125 respectively). ASH was more prominent in LGA infants (42.3%) as compared to AGA infants (28.0%). The table also demonstrates some descriptive biometric characteristics of the infants such as GA, BW, height, head circumference, chest circumference, and ponderal index (PI). On comparing the GA of LGA and AGA infants, it was found that it was almost similar in LGA infants (38.30 ± 1.20) and AGA infants (38.69 ± 1.48). The BW was expectedly significantly higher in LGA than AGA infants (4337.46 ± 404.45 and 3381.05 ± 445.77 respectively). Similarly, the other measurements such as height, head circumference, and chest circumference. were significantly increased in LGA infants compared to AGA infants. The PI is another important factor that was measured. PI is the ratio of body weight to height and is calculated as weight/height³ (Cole et al., 1997). The mean PI in LGA infants was calculated as 3.11 ± 0.38 compared to the mean PI of AGA infants which was calculated as 2.79 ± 0.34.

As for the maternal variables, in the AGA group, the white ethnicity was ranked most with 51.6% of all AGA infants being white. However, in the LGA group, most infants were of black ethnicity with 41.7% (not much higher than the white ethnicity with 40.9%). Most of the participants in the AGA group had non-diabetic mothers (63.8%) while in the LGA group there was almost an equal number of diabetic and non-diabetic mothers (50.3% vs 49.7%). Preeclampsia was not evident in the mothers of both AGA and LGA infants (93.4 and 86.6 respectively). The maternal BMI of LGA infants (35.74 ± 7.49) was significantly higher than that of AGA infants (31.68 ± 6.03). Finally, gravida yielded statistically insignificant results (p=0.47). The details of the neonatal and maternal factors can be seen in Table 1.

Table 1. Neonatal and Maternal variables of LGA and AGA infants

Variable		LGA		AGA		Total	P-value
Variable		f*	%	f	%	Total	P-value
Neonatal Variables
Mode of Delivery
	Vaginal	50	33.56	204	49.64	254	0.001
	C-Section	99	66.44	207	50.36	306	0.001
NICU Admission
	No	29	25.44	147	66.52	176	<0.001
	Yes	85	74.56	74	33.48	159	<0.001
Sex
	Male	92	61.74	213	51.45	305	0.031
	Female	57	38.26	201	48.55	258	0.031
APGAR 1
	Low	22	14.97	40	9.71	62	0.081
	Normal	125	85.03	372	90.29	497	0.081
APGAR 5
	Low	4	2.72	4	0.97	8	0.125
	Normal	143	97.28	408	99.03	551	0.125
Asymmetric Septal Hypertrophy
	No	86	57.72	298	71.98	384	0.001
	Yes	63	42.28	116	28.02	179	0.001
Gestational Age		38.30 ± 1.20		38.69 ± 1.48			0.01
Birth Weight		4337.46 ± 404.45		3381.05 ± 445.77			<0.001
Height		51.94 ± 2.24		49.50 ± 2.64			<0.001
Head Circumference		35.93 ± 1.40		34.31 ± 1.91			<0.001
Chest Circumference		36.26 ± 1.95		33.11 ± 2.01			<0.001
Ponderal Index		3.11 ± 0.38		2.79 ± 0.34			<0.001
Maternal Variables
Gravida
	Primi	27	18.12	93	22.79	120	0.47
	Multi	92	61.74	242	59.31	334
	Grand	30	20.13	73	17.89	103
Parity
	Nulli	27	18.12	158	38.44	185	<0.001
	Multi	118	79.19	241	58.64	359
	Grand	4	2.68	12	2.92	16
Diabetes
	No	74	50.34	259	63.79	333	0.004
	Yes	73	49.66	147	36.21	220	0.004
Preeclampsia
	No	129	86.58	383	93.41	512	0.01
	Yes	20	13.42	27	6.59	47	0.01
Ethnicity
	White	52	40.94	179	51.59	231	<0.001
	Black	53	41.73	114	32.85	167
	Hispanic	14	11.02	8	2.31	22
	Asian	1	0.79	15	4.32	16
	Other	7	5.51	31	8.93	38
Maternal age		32.67 ± 5.63		31.55 ± 5.77			0.041
Maternal BMI		35.74 ± 7.49		31.68 ± 6.03			<0.001

*Values are expressed as frequencies and percentages or Mean ± SD.

Abbreviations: C-Section, Cesarean section; f, Frequency; %, percentage; DM, Diabetes Mellitus.

Table 2 shows a detailed description of cardiac variables that were collected for all the infants in our study. It compares the means and standard deviation of those cardiac variables for LGA infants and AGA infants. All the variables were statistically significant except the LVIDs (p=0.19). The mean thickness of the IVS was significantly increased in LGA infants as compared to AGA infants (5.1 vs 4.0 mm respectively in diastole and 6.4 vs 5.3 mm respectively in systole). The increased mean IVSs is due to the contraction of the muscle fibers that cause thickening of the IVS (Moore et al., 2021). Some variables like LVIDd and LVPWd were minimally elevated in LGA which might indicate clinical insignificance. The details of all the parameters can be found in the table below.

Table 2. Descriptive statistics of cardiac variables

Variable	LGA		AGA		Observations	P-value
Variable	Mean	SD	Mean	SD	Observations	P-value
LVmass	14.12	4.00	10.27	3.26	559	<0.001
LVmass/vol	58.89	14.80	49.49	12.15	549	<0.001
LVIDd	19.41	2.27	18.57	2.11	559	<0.001
LVIDs	12.14	1.84	11.93	1.62	559	0.19
IVSd	5.10	1.29	4.00	0.78	562	<0.001
IVSs	6.40	1.37	5.34	1.05	559	<0.001
LPWDd	4.05	0.73	3.48	0.61	559	<0.001
LVPWs	5.48	0.87	4.83	0.70	558	<0.001
IVS/LPW	1.28	0.39	1.17	0.28	557	<0.001
FS	37.49	5.20	35.56	4.51	559	<0.001

Abbreviations: IVS, thickness of Inter Ventricular Septum in diastole (IVSd) and systole (IVSs); LVID, cardiac left ventricular internal dimension during diastole (LVIDd) and systole (LVIDs); (LVPW), thickness of left ventricular posterior wall in diastole (LVPWd) and systole (LVPWs); FS, Fractional Shortening; LVmass/vol, LV mass to volume ratio; LGA, large for gestational age; AGA, appropriate for gestational age; SD, standard deviation.

The multivariate linear regression analysis done to reveal the association between perinatal factors and LV parameters showed some interesting findings (Table 3). In the IVSd regression model (R²=0.34, Adjusted [Adj] R²=0.34), we observed that both GA and the APGAR score at 1 minute had statistically significant negative effects on IVSd. Specifically, GA had a significant negative impact with a coefficient of -0.140 (p<0.001), and the APGAR score at 1 minute also had a statistically significant negative effect, with a coefficient of -0.066 (p=0.004). However, Maternal insulin use during pregnancy and Birth weight both had a positive and significant effect on IVSd with a coefficient of 0.561 (p<0.001) and 0.001 (p<0.001) respectively. Similarly, BW was significantly associated positively with IVSs with a coefficient of 0.001 (p<0.001),

LVIDd regression results (R²=0.15, Adj R²=0.14) show that sex was found to be a significant predictor of LVIDd (p<0.001), with a negative coefficient of -0.776, indicating that male infants have a higher LVIDd than female infants. Similarly, in LVIDs regression model (R²=0.07, Adj R²=0.06), sex was found to be a significant predictor of LVIDs (p=0.001), with a negative coefficient of -0.464. GA was significantly associated with LVIDd (p=0.027) and LVIDs (p=0.014). Similarly, BW was associated with both LVIDd (p<0.001) and LVIDs (p=0.007). In the LVPWd (R²=0.23, Adj R²=0.23) and the LVPWs (R²=0.20, Adj R²=0.20) regression results, BW was the only positive associated perinatal factor (p<0.001). BW and Maternal BMI were found to be positively associated with FS (R²=0.04, Adj R²=0.04). These results were statistically significant (p=0.004 and p=0.009 respectively), with a positive coefficient of <0.001 and 0.093 respectively. This indicates that higher BW and maternal BMI are associated with an increase in FS.

The regression analysis for LVmass (R2=0.33, Adj R2=0.32) showed revealed that BW is a significant predictor of LVmass, with a positive coefficient of 0.004 (p<0.001), suggesting that higher BW is associated with an increase in LVmass. In the case of the LVmass/Vol regression model, the R-squared value was 0.08, and the adjusted R-squared value was 0.07. It showed a negative significant relationship between GA and LVmass/Vol (p=0.029). However, BW, was found to be a significant predictor with a positive association (p<0.001).

Regarding the univariate binary regression, neither ASH nor IVS/LVPW showed any significant associations with the independent variables included in the analysis, and therefore, the results of this analysis are not presented.

Table 3: Associations of Perinatal Factors with Left Ventricular Parameters

LV parameter	N	Variable	Coeff	SE	P-value	R², Adj R²
IVSd	547	GA	-0.140	0.028	<0.001	0.34, 0.34
		Birth weight	0.001	0.000	<0.001
		APGAR1	-0.066	0.023	0.004
		Insulin use	0.561	0.156	<0.001
*Other variables controlled for in this model: Maternal BMI.
IVSs	547	Birth weight	0.001	0.000	<0.001	0.25, 0.24
*Other variables controlled for in this model: NICU admission, GA, Birth weight, Category, PI, Maternal BMI, Diabetes, Diabetic control, Preeclampsia.
LVIDd	547	Sex	-0.776	0.177	<0.001	0.15, 0.14
		GA	0.149	0.067	0.027
		Birth weight	0.001	0.000	<0.001
		APGAR1	0.100	0.054	0.062
*Other variables controlled for in this model: Maternal BMI, Preeclampsia, Mean BP.
LVIDs	555	Sex	-0.464	0.142	0.001	0.07, 0.06
		GA	0.133	0.054	0.014
		Birth weight	0.000	0.000	0.007
*Other variables controlled for in this model: Preeclampsia.
LVPWd	547	Birth weight	0.001	0.000	<0.001	0.23, 0.23
*Other variables controlled for in this model: GA, Maternal BMI, Preeclampsia, Insulin use.
LVPWs	552	Birth weight	0.001	0.000	<0.001	0.20, 0.20
*Other controlled for variables in this model: GA, Maternal BMI, Preeclampsia.
FS	554	Birth weight	<0.001	<0.001	0.009	0.04, 0.04
FS	554	Maternal BMI	0.093	0.032	0.004	0.04, 0.04
*Other variables controlled for in this model: MOD.
LVmass	552	Birth weight	0.004	0.000	<0.001	0.33, 0.32
LVmass	552	Parity	0.223	0.119	0.063	0.33, 0.32
*Other variables controlled for in this model: Sex, GA, Maternal BMI, Preeclampsia.
LVmass/Vol	538	GA	-0.920	0.410	0.029	0.13, 0.12
LVmass/Vol	538	Birth weight	0.008	0.001	<0.001	0.13, 0.12
*Other variables controlled for in this model: Maternal BMI, Insulin use.

Abbreviations: LVmass, Left Ventricular mass; LVmass/Vol, LVmass to Volume ratio; IVSd, Inter-Ventricular Septal thickness during diastole; IVSs, Inter-Ventricular Septal thickness during systole; LVIDd, LV Internal Dimension during diastole; LVIDs, LV Internal Dimension during systole; LVPWd, LV Posterior Wall thickness at end of diastole; LVPWs, LV Posterior Wall thickness at end of systole; IVS/LVPW, Inter-Ventricular Septal thickness to LV Posterior Wall thickness ratio in diastole; FS, Shortening Fraction; SD, Standard Deviation; RMSE, Root Mean Square Error; Coeff, Coefficient; GA, Gestational Age; BMI, Body Mass Index; BP, Blood Pressure; MOD, Mode Of Delivery; Adj, Adjusted.

The multivariate linear regression that was performed to determine the association between perinatal factors and cardiac variables between LGA and AGA infants are shown in table 4. The data showed no significant association between any of the perinatal variables and LV mass in both LGA and AGA infants except BW. There was a positive relationship between LVmass and BW as the coefficient was 0.003, meaning that for each unit increase in BW the LV mass increases by 0.003. This result was statistically significant (p<0.001). Finally, there was also an association between pre-eclampsia and LV mass in AGA infants (coeff – 1.222, p=0.045). The same could not be said for the LGA group as there was no statistical significance. Details about the rest of the perinatal variables and their association with LV mass can be seen in Table 3.

Similarly, there was a positive association between BW in AGA infants and LV mass to volume ratio with a coefficient of 0.005 (p<0.001). However, there was no statistical significance observed in the LGA group (p=0.06). The remaining perinatal factors (Maternal BMI and insulin-controlled DM) had no statistical significance in either group. Analysis shows that the thickness of interventricular septum was significantly affected by perinatal factors. However, there were some discrepancies between their effects on the IVS during diastole and systole. For instance, during diastole the IVS thickness had a statistically significant positive association with perinatal factors such as BW (p<0.001), and insulin-controlled DM (p=0.02) in the AGA group. It also had a statistically significant negative association with apgar score at 1 minute (p<0.001) in the same group. In the LGA group, there was a positive association between IVS thickness during diastole and BW (p<0.001), and insulin-controlled DM (p=0.003). On the contrary, the thickness of IVS during systole had a statistically significant positive association with only BW (p<0.001) in both groups.

The association between perinatal factors and the LVIDd and LVIDs were also analyzed. Sex was the most interesting finding as the results showed that there is a strong association between being a male and having an increased LVIDd in AGA and LGA infants (p=0.002 and p=0.025 respectively). Similarly, there was a strong association between being a male and having an increased LVIDs in AGA and LGA infants (p=0.03 and p=0.041). BW was also found to be statistically correlated with LVIDd and LVIDs only in the AGA group (p<0.001). APGAR score at 1 minute had a positive association with LVIDd in AGA infants (coeff=0.161 and p=0.014). Finally, preeclampsia had a negative correlation with LVIDs in LGA infants (coeff=-1.146 and p=0.013).

Once again, BW was positively correlated with LVPW during diastole in AGA and LGA infants (p<0.001 and p=0.003). Similarly, it was also positively correlated with LVPW during systole in AGA and LGA infants (p<0.001 and p=0.011). In addition, there was a statistically significant positive association between maternal BMI and shortening fraction in LGA infants (p=0.025). Shortening fraction was not found to be positively correlated with any other perinatal factors.

The final variable that was analyzed is the IVS/LVPW ratio. It is used to determine if the heart is symmetric in size or not. The cutoff for this value is used to determine ASH. Analysis unveiled that the only positively correlated with IVS/LVPW in LGA infants was insulin-controlled DM (coeff=0.236, p=0.03). The rest of the variables were statistically insignificant. Details are listed in Table 4.

Table 4. Results of the multivariate regression to investigate the association between cardiac parameters (dependent variables) and perinatal factors (independent variables) for AGA and LGA infants.

Cardiac Parameter	AGA					LGA
Cardiac Parameter	Observations	Variable	Coefficient	SE	P-value	Observations	Variable	Coefficient	SE	P-value
LV mass	405	sex	-0.499	0.301	0.098	147	sex	-0.119	0.658	0.857
		BW	0.003	<0.001	<0.001		BW	0.003	0.001	<0.001
		Preeclampsia	1.222	0.608	0.045		Preeclampsia	0.326	0.986	0.742
*Other variables controlled for in these models: Parity, Maternal BMI.
LV mass/vol	398	BW	0.005	0.001	<0.001	140	BW	0.006	0.003	0.060
*Other variables controlled for in these models: Maternal BMI, Insulin use.
ivs_d	402	BW	<0.001	<0.001	<0.001	145	BW	0.001	<0.001	<0.001
		APGAR1	-0.095	0.024	<0.001		APGAR1	-0.002	0.053	0.965
		DM (Insulin)	0.399	0.171	0.020		DM (Insulin)	1.015	0.330	0.003
*Other variables controlled for in these models: Maternal BMI.
ivs_s	401	BW	0.001	<0.001	<0.001	146	BW	0.001	<0.001	<0.001
ivs_s	401	preeclampsia	0.175	0.209	0.403	146	preeclampsia	0.546	0.322	0.093
*Other variables controlled for in these models: Maternal BMI, DM, Insulin use.
lvid_d	402	sex	-0.644	0.202	0.002	145	sex	-0.833	0.369	0.025
		BW	0.002	<0.001	<0.001		BW	<0.001	<0.001	0.576
		APGAR1	0.161	0.066	0.014		APGAR1	-0.044	0.092	0.638
		preeclampsia	0.700	0.426	0.101		preeclampsia	-1.101	0.571	0.056
*Other variables controlled for in these models: Maternal BMI, meanbp.
Lvid_s	407	sex	-0.345	0.159	0.030	148	sex	-0.635	0.308	0.041
		BW	0.001	<0.001	<0.001		BW	<0.001	<0.001	0.913
		preeclampsia	0.170	0.319	0.596		preeclampsia	-1.146	0.456	0.013
*Other variables controlled for in these models: None
lvpwd	402	BW	0.001	<0.001	<0.001	145	BW	<0.001	<0.001	0.003
lvpwd	402	DM (Insulin)	0.257	0.137	0.061	145	DM (Insulin)	-0.031	0.201	0.877
*Other variables controlled for in these models: Maternal BMI, Preeclampsia
lvpws	405	BW	0.001	<0.001	<0.001	147	BW	<0.001	<0.001	0.011
*Other variables controlled for in these models: Maternal BMI, Preeclampsia.
FS	406	BW	<0.001	0.001	0.423	148	BW	<0.001	0.001	0.968
FS	406	Maternal BMI	0.065	0.038	0.086	148	Maternal BMI	0.135	0.059	0.025
*Other variables controlled for in these models: MOD.
IVS/LVPW		DM (Insulin)	0.025	0.066	0.703	145	DM (Insulin)	0.236	0.107	0.030
IVS/LVPW		preeclampsia	0.078	0.059	0.185	145	preeclampsia	0.178	0.096	0.066
*Other variables controlled for in these models: Birth weight

Abbreviations: LV mass, Left Ventricular mass; LVmass/Vol, Left Ventricular mass to Volume ratio; IVS_d, Inter-Ventricular Septal thickness during diastole; IVS_s, Inter-Ventricular Septal thickness during systole; LVID_d, Left Ventricular Internal Dimension during diastole; LVID_s, Left Ventricular Internal Dimension during systole; LVPWd, Left Ventricular Posterior Wall thickness at end of diastole; LVPWs, Left Ventricular Posterior Wall thickness at end of systole; FS, Shortening Fraction; SE, Standard Error; GA, Gestational Age; BW, Birth Weight; BMI, Body Mass Index; DM, Diabetes Mellitus; DM (insulin), Diabetes Mellitus controlled by insulin medication ; APGAR 1, APGAR score at 1 minute ; meanBP, mean Blood Pressure; MOD, Mode Of Delivery.

The analysis of ASH yielded the most interesting results. It was observed that the odds of developing ASH in LGA infants were the same as AGA infants (OR=1, p<0.001). Additionally, the odds of developing ASH decreased with an increase in GA (OR=0.79, p<0.001). The remaining variables can be referred to in Table 5.

Table 5. Results of binary logistic regression showing the association between ASH (dependent variable) and perinatal factors (independent variable) in infants.

ASH
Observations: 553
Variable	Odds Ratio	SE	z-value	P-value
GA	0.79	0.06	-3.40	<0.001
BW	1.00	0.00	3.30	<0.001
APGAR1	0.90	0.05	-1.87	0.06
meanbp	1.02	0.01	1.71	0.09
Intercept	446.41	1267.73	2.15	0.03

Abbreviations: ASH, Asymmetric Septal Hypertrophy; SE, Standard Error; GA, Gestational Age; BW, Birth Weight; APGAR1, APGAR score at 1 minute; meanbp, mean blood pressure.

The current study shows that most of the sampled infants that were admitted to NICU were LGA. Similarly, ASH was found to be more prevalent in the LGA group (42.3%) as compared to AGA infants (28.0%). This indicates that BW might play an important role in predicting adverse outcomes post-delivery. However, our binary logistic regression demonstrated that BW was not a significant contributor to ASH which indicates that LGA babies had the same odds of developing ASH as AGA infants. This discrepancy makes us believe that further research needs to investigate the association between BW and ASH. Moreover, our study found that GA was associated with a decreased odd of developing ASH. The growing data showed that GDM is one of the major complications of pregnancy. This complication has transgenerational consequences, including higher incidences of metabolic syndrome and vascular abnormalities in older children and adult offspring of affected mothers (Do et al., 2021).It has been reported that the newborn of a diabetic mother has an increased odds of developing ASH (Kiruthiga, 2019). Similarly, a study reported from India stated that ASH is a common finding in infants born to diabetic mothers (Vela-Huerta et al., 2019). In contrast, our result showed that there was no association of ASH in LGA babies born to diabetic mothers. The findings of the current study were consistent with the reported study from South Korea (Kim et al., 1998). The association of ASH with diabetes is still controversial. Therefore, further in-deep studies should be conducted to answer this controversy. However, in the current study, an increased FS was observed in LGA babies born to mothers with elevated BMIs. This is also not in agreement with the most recent cohort study, where it was reported that children born to diabetic mothers and mothers with high BMIs showed persistently increased interventricular septal thickness and decreased shortening fraction in early childhood (Peng et al., 2022)

A significant association was found between BW and cardiovascular health, including LV mass, LV mass/volume, IVSD, IVSS, LVID, LVID, LVPD, and LVPD. The positive and highly significant association between BW and LV mass in both AGA and LGA infants is in line with previous studies. Based on a study by Sawyer et al., LV mass index was positively associated with birth BMI (P = 0.01) (Sawyer et al., 2019). A larger LV mass might be an indication of better cardiac development during fetal growth or a potential adaptation to intrauterine conditions for infants with higher BWs. Further supporting the idea that BW plays a crucial role in influencing cardiac parameters is the positive association between BW and IVSd in both groups. However, another study by Vijayakumar et al. that examined the relationship between infant growth and LV mass in adulthood yielded results that claim that LV mass was not related to BW (Vijayakumar et al., 1995). The relationship between weight at one year and LV mass was independent of factors in adult life such as body size, systolic blood pressure, and age. The enlarged LV mass associated with reduced growth in infancy was concentric, affecting both the interventricular septum and the left ventricular posterior wall. Therefore, further studies investigating this relationship are required to establish more solid scientific evidence.

It has been previously reported that the weight of the child at birth is the significant predictor of BMI at a later age. A cohort study reported from Australia stated that the higher BW was significantly predicted the higher BMI in later age (Sjöholm et al., 2021). The higher BMI is the major reason that this population has the obesity pandemic. Subsequently, obesity has a strong link with cardiovascular disease (Koliaki et al., 2019; Manrique-Acevedo et al., 2020). From here we postulate that an increased BW can pose future cardiovascular health issues at a later stage in life that may not be evident at birth. We believe our study could provide base to future research that study the relation between BW and future cardiac issues.

It has been seen that maternal obesity, gestational hypertension, and diabetes have a significant impact on the LV’s structure and function and lead to cardiac abnormalities. The present obesity pandemic affects women of childbearing age and increases the risk of cardiovascular disease and cardiomyopathies(Liu et al., 2019). Pregnancy causes metabolic changes such as increases in body weight, circulation lipids, glucose, and inflammatory markers. Obese women experience more of these changes than normal-weight women(Wang et al., 2021). The uterine environment influences fetal organ development, influencing disease susceptibility throughout childhood, adolescence, and old age. According to epidemiological research, maternal obesity increases the risk of cardiovascular disease and premature mortality in adult and elderly children (Eriksson et al., 2014). Heart development occurs mostly throughout childhood, although little is known about cardiac development and function in human children born to obese mothers. In one research, 6-month-old neonates' LV mass increased in proportion to mother gestational weight growth (Guzzardi et al., 2018). GDM carries prenatal and perinatal risks as well as long-term risks for the mother and her child. Fetal cardiac and vascular structural and functional changes are linked to maternal hyperglycemia (Wahab et al., 2020). Congenital cardiac abnormalities and hypertrophy in the offspring of Type1/Type2 DM mothers are widely documented. It has been reported that fetal hyperinsulinism is caused by intrauterine exposure to high maternal blood glucose and affects the liver and cardiovascular system the most structurally and functionally (Di Bernardo et al., 2017).

Furthermore, the study showed an association between insulin ingestion in diabetic mothers and some cardiac parameters such as IVSd (coeff=1.015 and p=0.003) and IVS/LVPW (coeff=0.236 and p=0.03) in LGA infants. A study that investigated the effect of diabetic maternal insulin intake during pregnancy on fetal cardiac parameters yielded similar results (Pilania et al., 2016). It shows that the fetuses of diabetic mothers had a higher mean cardiac output than their non-diabetic counterparts at 26-28 weeks of gestation (192.9±67.74 vs 130.9±20.3 respectively and p<0.001). Similarly, they also had an increased mean myocardial performance index (0.583±0.06 vs 0.493±0.06 and p=0.000). Likewise, the study also reported an increase in mean cardiac output (316.057± 92.82 vs 251.188±75.88 and p=0.010) and mean myocardial performance index (0.62±0.07 vs 0.58±0.07 and p=0.047) at 34-36 weeks gestation.

In addition, mothers who were diagnosed with preeclampsia had a significant positive association with LV mass in the AGA group. However, preeclampsia did not significantly correlate with LV mass in LGA newborns. This suggests that the effect of this illness on cardiac health may differ depending on the fetal growth pattern. A study that investigated the effects of pregnancy preeclampsia on neonatal cardiac development concluded that the left ventricular mass indexed to body surface area (LVMI) was unchanged at birth in infants of hypertensive pregnancies. However, at 3 months of age, those infants had significantly greater LVMI (Aye et al., 2020).

Cardiac remodeling and cardiovascular events in children are influenced by the shape and function of the LV. The LV geometry may stigmatize the morbidity and mortality in this population even in asymptomatic conditions, such as before the start of overt hypertension or heart failure (Wang et al., 2021). That is the main reason why we believe that similar research studying the long-term implications of cardiac variables in LGA infants need to be done. Perhaps it might reveal great insights into this field and help us understand it better.

Study limitations

The current study is subjected to various limitations that may affect its internal and external validity. Firstly, this was a single-centered cohort study, therefore this study cannot be generalized to all populations with LGA. Secondly, the small sample size may have negatively affected the power of the study in return affecting its ability to yield more significant associations. In addition, the number of participants in the intervention group was extremely low as compared to the control group which may have affected the external validity of the study. Moreover, during the follow-up, the number of missed participants was high, and this may affect the outcome as it was dependent on the baseline data. Finally, this study had no follow-up of neonates at a later stage in their life. This limits the study’s ability to study the long-term outcomes of the aforementioned neonatal variables on the cardiac parameters.

The study's findings underscore the significant influence of perinatal factors on neonatal cardiac morphology, particularly in LGA and AGA infants. BW, maternal BMI, and maternal insulin use during pregnancy were key determinants affecting various aspects of left ventricular structure, including mass, wall thickness, and internal dimensions. GA negatively impacted the left ventricular mass-to-volume ratio, highlighting its influence on cardiac structural development Additionally, gender differences in cardiac dimensions were evident, with male infants displaying larger LV dimensions. These insights highlight the importance of considering these perinatal factors in the assessment and monitoring of neonatal cardiac health, offering valuable guidance for tailored clinical approaches in pediatric cardiology. These findings underscore the complex interplay of perinatal factors in influencing neonatal cardiac structure, critical for pediatric cardiac evaluations. However, we believe that due to the limitations of our study, the data cannot be extrapolated to the global population and more research need to be done on the association between perinatal factors and cardiovascular health in infants to have more solid data and evidence-based medicine.

Ethics Approval and consent to participate

Data collection and analysis adhered to the principles outlined in the Helsinki Declaration, and ethical approval was granted by the IRB of NewYork-Presbyterian Brooklyn Methodist Hospital. Given that echocardiography, a procedure with minimal risk, was carried out based on clinical indications, and the data collection was retrospective, the need for consent from the legal guardians of the neonates was considered unnecessary by the IRB of NewYork-Presbyterian Brooklyn Methodist Hospital. All procedures followed the pertinent guidelines and regulations.

Consent for publication

Not applicable

Availability of Data and Materials

The data used for the analysis in this work are available upon reasonable request from the corresponding author.

Competing interests

Authors declare no conflict of interest.

Funding

The publication of this article was funded by Qatar National Library. This research did not receive any other grants from funding agencies in the public, commercial, or not-for-profit sectors.

Authors Contribution

AA and AG: Formal analysis, Data Curation, Methodology, Validation, Visualization, Writing - Original Draft, Writing - Review & Editing. IE: Formal analysis, Validation, Writing - Original Draft, Writing - Review & Editing. IN: Supervised the data collection and organization of the data sheets. MC, DH, DS, AN, NR, BD, and FS: Patient allocation and data collection. MG and PN: Provided supervision and approval for the study. All authors approved the final draft of the manuscript.

Acknowledgements

Not applicable

Author’s contact

For any questions about this research please address Ashraf Gad by Email: [email protected]

Di Bernardo, S., Mivelaz, Y., Epure, A. M., Vial, Y., Simeoni, U., Bovet, P., Younes, S. E., Chiolero, A., & Sekarski, N. (2017). Assessing the consequences of gestational diabetes mellitus on offspring’s cardiovascular health: MySweetHeart Cohort study protocol, Switzerland. BMJ open, 7(11), e016972.
Do, V., Eckersley, L., Lin, L., Davidge, S. T., Stickland, M. K., Ojala, T., Serrano-Lomelin, J., & Hornberger, L. K. (2021). Persistent Aortic Stiffness and Left Ventricular Hypertrophy in Children of Diabetic Mothers. CJC Open, 3(3), 345-353. https://doi.org/https://doi.org/10.1016/j.cjco.2020.10.020
Eriksson, J. G., Sandboge, S., Salonen, M. K., Kajantie, E., & Osmond, C. (2014). Long-term consequences of maternal overweight in pregnancy on offspring later health: findings from the Helsinki Birth Cohort Study. Annals of medicine, 46(6), 434-438.
Guzzardi, M. A., Liistro, T., Gargani, L., Ait Ali, L., D’Angelo, G., Rocchiccioli, S., La Rosa, F., Kemeny, A., Sanguinetti, E., & Ucciferri, N. (2018). Maternal obesity and cardiac development in the offspring: study in human neonates and minipigs. JACC: Cardiovascular Imaging, 11(12), 1750-1755.
Kim, Y. S., Chae, S. A., Lim, I. S., & Yoo, B. H. (1998). The effect of Large for Gestational Age on Asymmetrical Ventricular Septal Hypertrophy in the Newborn. Journal of the Korean Society of Neonatology, 5(1), 40-44.
Kiruthiga, K. (2019). A Study on Cardiovascular Complications in Infants of Diabetic Mother Tirunelveli Medical College, Tirunelveli].
Koliaki, C., Liatis, S., & Kokkinos, A. (2019). Obesity and cardiovascular disease: revisiting an old relationship. Metabolism, 92, 98-107.
Liu, Y., Chen, S., Zühlke, L., Black, G. C., Choy, M.-k., Li, N., & Keavney, B. D. (2019). Global birth prevalence of congenital heart defects 1970–2017: updated systematic review and meta-analysis of 260 studies. International journal of epidemiology, 48(2), 455-463.
Manrique-Acevedo, C., Chinnakotla, B., Padilla, J., Martinez-Lemus, L. A., & Gozal, D. (2020). Obesity and cardiovascular disease in women. International Journal of Obesity, 44(6), 1210-1226.
Sjöholm, P., Pahkala, K., Davison, B., Niinikoski, H., Raitakari, O., Juonala, M., & Singh, G. R. (2021). Birth weight for gestational age and later cardiovascular health: a comparison between longitudinal Finnish and indigenous Australian cohorts. Annals of medicine, 53(1), 2060-2071.
Vela-Huerta, M., Amador-Licona, N., Orozco Villagomez, H. V., Heredia Ruiz, A., & Guizar-Mendoza, J. M. (2019). Asymmetric Septal Hypertrophy in Appropriate for Gestational Age Infants Born to Diabetic Mothers. Indian pediatrics, 56(4).
Wahab, R. J., Jaddoe, V. W., Roest, A. A., Toemen, L., & Gaillard, R. (2020). Associations of Maternal Glycemia in the First Half of Pregnancy With Alterations in Cardiac Structure and Function in Childhood. Diabetes Care, 43(9), 2272-2280.
Wang, J., Du, B., Wu, Y., Li, Z., Chen, Q., Zhang, X., Zhang, L., Ye, Y., Wu, Y., Chen, S., & Sun, K. (2021). Association of Maternal Gestational Weight Gain With Left Ventricle Geometry and Function in Offspring at 4 Years of Age: A Prospective Birth Cohort Study. Frontiers in pediatrics, 9, 722385-722385. https://doi.org/10.3389/fped.2021.722385
Al-Biltagi, M., El Razaky, O., & El Amrousy, D. (2021). Cardiac changes in infants of diabetic mothers. World J Diabetes, 12(8), 1233-1247. https://doi.org/10.4239/wjd.v12.i8.1233
Aye, C. Y. L., Lewandowski, A. J., Lamata, P., Upton, R., Davis, E., Ohuma, E. O., . . . Leeson, P. (2020). Prenatal and Postnatal Cardiac Development in Offspring of Hypertensive Pregnancies. J Am Heart Assoc, 9(9), e014586. https://doi.org/10.1161/JAHA.119.014586
Cole, T. J., Henson, G. L., Tremble, J. M., & Colley, N. V. (1997). Birthweight for length: ponderal index, body mass index or Benn index? Ann Hum Biol, 24(4), 289-298. https://doi.org/10.1080/03014469700005032
Corrigan, N., Brazil, D. P., & McAuliffe, F. (2009). Fetal cardiac effects of maternal hyperglycemia during pregnancy. Birth Defects Res A Clin Mol Teratol, 85(6), 523-530. https://doi.org/10.1002/bdra.20567
Di Bernardo, S., Mivelaz, Y., Epure, A. M., Vial, Y., Simeoni, U., Bovet, P., . . . Sekarski, N. (2017). Assessing the consequences of gestational diabetes mellitus on offspring’s cardiovascular health: MySweetHeart Cohort study protocol, Switzerland. BMJ open, 7(11), e016972.
Do, V., Eckersley, L., Lin, L., Davidge, S. T., Stickland, M. K., Ojala, T., . . . Hornberger, L. K. (2021). Persistent Aortic Stiffness and Left Ventricular Hypertrophy in Children of Diabetic Mothers. CJC Open, 3(3), 345-353. https://doi.org/https://doi.org/10.1016/j.cjco.2020.10.020
Elmakaty, I., Amarah, A., Henry, M., Chhabra, M., Hoang, D., Suk, D., . . . Gad, A. (2023). Perinatal factors impacting echocardiographic left ventricular measurement in small for gestational age infants: a prospective cohort study. BMC Pediatr, 23(1), 393. https://doi.org/10.1186/s12887-023-04204-w
Eriksson, J. G., Sandboge, S., Salonen, M. K., Kajantie, E., & Osmond, C. (2014). Long-term consequences of maternal overweight in pregnancy on offspring later health: findings from the Helsinki Birth Cohort Study. Annals of medicine, 46(6), 434-438.
Farrar, D., Simmonds, M., Bryant, M., Sheldon, T. A., Tuffnell, D., Golder, S., . . . Lawlor, D. A. (2016). Hyperglycaemia and risk of adverse perinatal outcomes: systematic review and meta-analysis. BMJ, 354, i4694. https://doi.org/10.1136/bmj.i4694
Fenton, T. R., & Kim, J. H. (2013). A systematic review and meta-analysis to revise the Fenton growth chart for preterm infants. BMC Pediatr, 13, 59. https://doi.org/10.1186/1471-2431-13-59
Guzzardi, M. A., Liistro, T., Gargani, L., Ait Ali, L., D’Angelo, G., Rocchiccioli, S., . . . Ucciferri, N. (2018). Maternal obesity and cardiac development in the offspring: study in human neonates and minipigs. JACC: Cardiovascular Imaging, 11(12), 1750-1755.
Kim, Y. S., Chae, S. A., Lim, I. S., & Yoo, B. H. (1998). The effect of Large for Gestational Age on Asymmetrical Ventricular Septal Hypertrophy in the Newborn. Journal of the Korean Society of Neonatology, 5(1), 40-44.
Kiruthiga, K. (2019). A Study on Cardiovascular Complications in Infants of Diabetic Mother Tirunelveli Medical College, Tirunelveli].
Koliaki, C., Liatis, S., & Kokkinos, A. (2019). Obesity and cardiovascular disease: revisiting an old relationship. Metabolism, 92, 98-107.
Liu, Y., Chen, S., Zühlke, L., Black, G. C., Choy, M.-k., Li, N., & Keavney, B. D. (2019). Global birth prevalence of congenital heart defects 1970–2017: updated systematic review and meta-analysis of 260 studies. International journal of epidemiology, 48(2), 455-463.
Manrique-Acevedo, C., Chinnakotla, B., Padilla, J., Martinez-Lemus, L. A., & Gozal, D. (2020). Obesity and cardiovascular disease in women. International Journal of Obesity, 44(6), 1210-1226.
Moore, M. N., Climie, R. E., Otahal, P., Sharman, J. E., & Schultz, M. G. (2021). Exercise blood pressure and cardiac structure: A systematic review and meta-analysis of cross-sectional studies. J Sci Med Sport, 24(9), 925-930. https://doi.org/10.1016/j.jsams.2021.02.014
Peng, Y. Q., Qiu, X., Wang, L., Li, X., & Huo, X. Y. (2022). Left atrial shortening fraction to predict fetal cardiac abnormalities and dysfunction in gestational diabetes mellitus. Front Cardiovasc Med, 9, 1026587. https://doi.org/10.3389/fcvm.2022.1026587
Pilania, R., Sikka, P., Rohit, M. K., Suri, V., & Kumar, P. (2016). Fetal Cardiodynamics by Echocardiography in Insulin Dependent Maternal Diabetes and Its Correlation with Pregnancy Outcome. J Clin Diagn Res, 10(7), QC01-04. https://doi.org/10.7860/JCDR/2016/17993.8079
Sawyer, A. A., Pollock, N. K., Gutin, B., Weintraub, N. L., & Stansfield, B. K. (2019). Proportionality at birth and left ventricular hypertrophy in healthy adolescents. Early Hum Dev, 132, 24-29. https://doi.org/10.1016/j.earlhumdev.2019.03.018
Sjöholm, P., Pahkala, K., Davison, B., Niinikoski, H., Raitakari, O., Juonala, M., & Singh, G. R. (2021). Birth weight for gestational age and later cardiovascular health: a comparison between longitudinal Finnish and indigenous Australian cohorts. Annals of medicine, 53(1), 2060-2071.
Vela-Huerta, M., Amador-Licona, N., Orozco Villagomez, H. V., Heredia Ruiz, A., & Guizar-Mendoza, J. M. (2019). Asymmetric Septal Hypertrophy in Appropriate for Gestational Age Infants Born to Diabetic Mothers. Indian pediatrics, 56(4).
Vijayakumar, M., Fall, C. H., Osmond, C., & Barker, D. J. (1995). Birth weight, weight at one year, and left ventricular mass in adult life. Br Heart J, 73(4), 363-367. https://doi.org/10.1136/hrt.73.4.363
Wahab, R. J., Jaddoe, V. W., Roest, A. A., Toemen, L., & Gaillard, R. (2020). Associations of Maternal Glycemia in the First Half of Pregnancy With Alterations in Cardiac Structure and Function in Childhood. Diabetes Care, 43(9), 2272-2280.
Wang, J., Du, B., Wu, Y., Li, Z., Chen, Q., Zhang, X., . . . Sun, K. (2021). Association of Maternal Gestational Weight Gain With Left Ventricle Geometry and Function in Offspring at 4 Years of Age: A Prospective Birth Cohort Study. Frontiers in pediatrics, 9, 722385-722385. https://doi.org/10.3389/fped.2021.722385

Effects of Perinatal Variables on Echocardiographic Assessments of Left Ventricular Dimensions in Infants Born Large for Gestational Age: A Prospective Cohort Analysis

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Introduction

Material and methods

Study design and setting

Study population

Participants’ recruitment

Ethical consideration

Cardiac Variables:

Data analysis

Results

Discussion

Conclusion

Declarations

References

Status:

Version 1