Antenatal Antecedents of Cognitive Impairment at 24 Months In Extremely Low Gestational Age Newborns

Abstract

BACKGROUND AND OBJECTIVES:

Extremely low gestational age neonates are more likely than term infants to develop cognitive impairment. Few studies have addressed antenatal risk factors of this condition. We identified antenatal antecedents of cognitive impairment determined by the Mental Development Index (MDI) portion of the Bayley Scales of Infant Development, Second Edition (BSID-II), at 24 months corrected age.

METHODS:

We studied a multicenter cohort of 921 infants born before 28 weeks of gestation during 2002 to 2004 and assessed their placentas for histologic characteristics and microorganisms. The mother was interviewed and her medical record was reviewed. At 24 months adjusted age, children were assessed with BSID-II. Multinomial logistic models were used to estimate odds ratios.

RESULTS:

A total of 103 infants (11%) had an MDI <55, and 99 infants (11%) had an MDI between 55 and 69. No associations were identified between organisms recovered from the placenta and developmental delay. Factors most strongly associated with MDI <55 were thrombosis of fetal vessels (OR 3.1; 95% confidence interval [CI] 1.2, 7.7), maternal BMI >30 (OR 2.0; 95% CI 1.1, 3.5), maternal education ≤12 years (OR 3.4; 95% CI 1.9, 6.2), nonwhite race (OR 2.2; 95% CI 1.3, 3.8), birth weight z score < −2 (OR 2.8; 95% CI 1.1, 6.9), and male gender (OR 2.7; 95% CI 1.6, 4.5).

CONCLUSIONS:

Antenatal factors, including thrombosis of fetal vessels in the placenta, severe fetal growth restriction, and maternal obesity, convey information about the risk of cognitive impairment among extremely premature newborns.

What's Known on this Subject:

Among extremely premature infants, survival has improved, but the rate of cognitive impairment has not. Impaired cognition is the most frequent developmental problem identified in survivors. Several antenatal factors have been associated with cognitive impairment, mostly related to social disadvantage.

What This Study Adds:

In addition to social disadvantage, antenatal characteristics associated with cognitive impairment include maternal obesity and thrombosis of fetal stem vessels. Prenatal infection and inflammation were not associated with impaired early cognitive function among extremely preterm infants.

Among extremely low gestational age newborns (ELGANs), the survival rate improved during the 1990s, but the rate of low scores on assessments of cognition did not.^1,2 Cognitive impairment is the most frequent developmental problem identified in extremely premature survivors.^3–5 Impaired early cognitive function, assessed with the Bayley Scales of Infant Development (BSID), is moderately predictive of cognitive function at school age.⁶

Antenatal factors associated with cognitive limitations in ELGANs include unmarried mother,⁷ minority race,⁵ fewer years of education,⁵ lack of antenatal steroid exposure,^5,8 maternal smoking during pregnancy,⁸ and preeclampsia.⁹ Most previous studies relied on samples defined by birth weight,^10–15 which distorts associations with factors that influence birth weight,¹⁶ whereas some relied on small samples,^9,11,17 and none focused on detailed data about prenatal factors.

METHODS

The ELGAN Study

The ELGAN study was designed to identify characteristics and exposures that increase the risk of structural and functional neurologic disorders in ELGANs.¹⁸ During 2002 to 2004, women delivering before 28 weeks of gestation at 1 of 14 participating institutions in 11 cities in 5 states were asked to enroll in the study. The enrollment and consent processes were approved by the individual institutional review boards.

Mothers were approached for consent either on antenatal admission or shortly after delivery, depending on clinical circumstance and institutional preference. A total of 1249 mothers of 1506 infants consented. Approximately 260 women were either missed or did not consent to participate.

Demographic and Pregnancy Variables

After delivery, a trained research nurse interviewed each mother in her native language by using a structured data collection and following procedures contained in a manual. The mother’s report of her own characteristics and exposures, as well as the sequence of events leading to preterm delivery was taken as truth, even when her medical record provided discrepant information.

Shortly after discharge, the research nurse reviewed the maternal chart by using a second structured data collection form to collect information about events after admission. The clinical circumstances that led to each maternal admission and ultimately to each preterm delivery were operationally defined.¹⁹ A course of antenatal adrenal corticosteroids given to enhance fetal lung maturation was considered complete if the gravida received 2 doses of betamethasone 24 hours apart and delivered at least 48 hours after the first dose, or if she received 4 doses of dexamethasone at 12-hour intervals. Neonatal data were collected from the newborn’s medical record.

Variable Definitions

The gestational age estimates were based on a hierarchy of the quality of available information. Most desirable were estimates based on the dates of embryo retrieval or intrauterine insemination or fetal ultrasound before the 14th week (62%). When these were not available, reliance was placed sequentially on a fetal ultrasound at 14 or more weeks (29%), last menstrual period without fetal ultrasound (7%), and gestational age recorded in the log of the NICU (1%). The birth weight z score is the number of standard deviations the infant’s birth weight is above or below the median weight of infants at the same gestational age in a standard data set.²⁰ The 6 initiators of preterm delivery, preterm labor, prelabor rupture of the fetal membranes, placental abruption, cervical insufficiency, preeclampsia, and delivery for fetal indications, are defined elsewhere.¹⁹

Placentas

All who biopsied the placenta were trained to ensure sterile technique. A piece of the chorion and underlying trophoblast tissue was removed under sterile conditions from the placenta at the midpoint of the longest distance between the cord insertion and the edge of the disk, and then placed in liquid nitrogen until transfer to a –80°C freezer. Eighty-two percent of the placentas were sampled within 1 hour of delivery. The microorganism recovery rate was not influenced by the interval between birth and sampling. The microbiologic procedures are described in detail elsewhere,²¹ as are the procedures for the preparation and reading of histologic slides.²²

Fetal thrombi were recognized when fibrin was seen attached to a vessel wall with ingrowth of endothelial cells. This degree of organization excludes acute thrombi, but there are no accepted criteria for assigning an exact age to these thrombi. Villous injury was not required for the diagnosis of thrombi. Not all thrombi were occlusive. Many, but not all thrombi, were associated with demonstrable villous injury, such as hemorrhage or sclerosis, but villous injury was not explicitly coded.

Level 3 severity of chorionic plate inflammation was defined as >20 neutrophils/20×. Grade 3 inflammation of the amnion was defined as numerous large or confluent foci, whereas grade 4 was defined as necrosis. Inflammation in the chorion/decidua was similarly, but separately, graded.

Inflammation in the umbilical cord was graded from 0 to 5. Grade 3 required neutrophils in perivascular Wharton jelly, grade 4 required panvasculitis and umbilical cord vasculitis extending deep into Wharton jelly, and grade 5 required a “halo lesion” (ring of precipitate in Wharton jelly encircling each vessel). Neutrophilic infiltration into fetal stem vessels in the chorionic plate required that neutrophils appeared to have migrated toward the amnionic cavity.

Developmental Assessment at 24 Months

Families were invited to bring their child for a developmental assessment close to the time when he or she would be 24 months corrected age. Fully 91% of children had this developmental assessment, which included a neurologic examination^19,23,24 and the BSID, second edition (BSID-II).²⁵ Of these children, 77% had their examination within the range of 23.5–27.9 months. All BSID-II assessments were age adjusted. Neurologic examiners were asked to rate the child on the Gross Motor Function Classification System (GMFCS)²⁶ separately from the neurologic examination.

Certified examiners administered and scored the BSID-II. Before testing, examiners were told the child's age. After completion of testing, they were told the gestational age so that the unadjusted Mental Development Index (MDI) and Psychomotor Development Index could be obtained. Only 2% of examiners indicated at the time of the examination that they had more than a limited amount of information about the child.

Cognitive impairment was defined as an MDI <70. An MDI <55 was considered severe cognitive impairment.

The child was classified as nontestable on a scale if his or her impairments prohibited standardized administration, or if >2 items were judged to be “not applicable.”

Data Analysis

The generalized form of the null hypothesis that we evaluated states that the risk of a low BSID-II score is not associated with any maternal, pregnancy, or delivery characteristic. To avoid attributing cognition and perception deficits to motor impairments, we limited our analyses to children who were able to walk independently (GMFCS <1), and who, therefore, were unlikely to have functionally important fine-motor impairments.

Because our outcomes of interest (MDI <55 and MDI between 55 and 69) are mutually exclusive and each is appropriately compared with the same referent group (MDI ≥70), we used maximum-likelihood multinomial (polytomous) logistic regression (Stata command: mlogit; Stata Corp, College Station, TX) to calculate odds ratios. We selected variables as potential confounders if identified in the literature or if in our data they were associated with both another relevant exposure and low MDI with probabilities ≤0.25.²⁷

To adjust for potential confounders, we created 2 individual conditional logistic multivariable models, each comparing children in 1 of the 2 abnormal outcome groups to the same referent group (ie, those with MDI ≥70). These models contained a hospital cluster term to account for the possibility that infants born at a particular hospital are more like each other than like infants born at other hospitals.

RESULTS

A total of 1200 infants survived to 24 months and 85% of surviving study participants were assessed with BSID-II and GMFCS (Fig 1). After excluding those with impaired gross motor function (GMFCS ≥1), 921 were included in the analyses described in this article. Placenta microbiology and histology were available for 85% (n = 783) of these 921 infants.

FIGURE 1.

Study participants who contributed data to the present analysis among the 1506 newborns enrolled in the Extremely Low Gestational Age Newborn Study.

Maternal Prepregnancy Characteristics

Infants whose mother was black, single, or eligible for public insurance were at increased risk of MDI <55, as were infants whose mother had less than a high school equivalent education, or a prepregnancy BMI of >30 (Table 1).

TABLE 1.

Risks of MDI Scores Associated With Social and Demographic Characteristics of the Mother

Maternal Characteristics		<55	55–69	≥70	Row N
Racial identity	White	7	8	85	545
	Black	20	15	65	242
	Other	13	16	71	119
Hispanic	Yes	17	13	70	108
	No	11	10	79	808
Age	<21	14	12	74	125
	21–35	11	11	78	618
	>35	10	10	80	178
Years of education	<12	28	12	60	142
	12 (high school)	11	15	75	232
	>12 to <16	8	10	80	201
	16 (college)	7	6	86	176
	>16	4	8	88	138
Marital status	Single	16	14	70	379
	Married	8	9	83	542
Self and/or partner	Yes	10	11	79	800
support	No	20	11	69	101
Public insurance	Yes	17	15	69	336
	No	8	9	86	566
Prepregnancy BMI	<18.5	9	6	85	66
	18.5 to <25	9	11	80	452
	25 to <30	11	9	80	183
	>30	16	14	70	184
Maximum number of infants		103	99	719	921
Percent		11	11	78

These are row percents.

Pregnancy Characteristics and Exposures

Of all the factors listed in Table 2, only assisted conception was associated with a decreased risk of MDI <55.

TABLE 2.

Risks of MDI Scores Associated With Pregnancy Characteristics and Exposures During Pregnancy

Pregnancy Characteristics		<55	55–69	≥70	Row N
Smoking prepregnancy	Yes	14	12	75	212
	No	11	11	79	685
Smoking during pregnancy	Yes	12	15	73	117
	No	11	10	79	781
Passive smoking	Yes	14	9	77	212
	No	11	11	78	683
Years since last pregnancy	<1	8	15	77	107
	1–2	14	15	71	144
	2+	15	12	73	267
Conception assistance	Yes	6	8	87	197
	No	13	12	75	698
No. of prenatal visits	10+	9	14	77	619
	< 10	12	10	78	273
Vaginal bleeding	Yes	9	11	80	361
during 1st 12 wk	No	13	11	76	534
Vaginal bleeding	Yes	8	11	81	261
after 12 wk	No	13	11	77	634
Fever during pregnancy	Yes	9	12	79	58
	No	11	11	78	837
Vaginal/cervical infection	Yes	14	14	72	113
	No	11	10	79	782
Urinary tract infections	Yes	8	13	79	134
	No	12	10	75	761
Highest white blood cell counta	>20 K	9	4	88	187
	≤20 K	12	13	75	715
Any medication	Yes	12	11	78	786
	No	9	13	78	108
Aspirin	Yes	15	11	71	46
	No	11	11	78	846
Nonsteroidal anti-inflammatory drugs	Yes	9	12	79	57
	No	12	11	78	834
Acetaminophen	Yes	9	9	82	453
	No	14	12	74	438
Antibiotic	Yes	11	13	77	265
	No	12	10	78	627
Maximum no. of infants		103	99	719	921
Percent		11	11	78

These are row percents.

^aWithin the interval from before delivery to 48 h after delivery.

Delivery Characteristics

The highest risks of MDI <55 were found with shorter duration of labor and when infants were delivered for fetal indications or for maternal preeclampsia, and the lowest risk was found when delivery was due to placental abruption (Table 3).

TABLE 3.

Risks of MDI Scores Associated With Delivery Characteristics

Delivery Characteristics		<55	55–69	≥70	Row N
Number of fetuses	1	12	10	78	609
	2+	9	12	79	312
Antenatal steroids	Complete course	12	11	77	593
	Partial course	9	12	79	234
	None	11	9	81	94
Magnesium	For tocolysis	11	10	80	503
	For PE	15	12	74	121
	No	10	13	77	288
Cesarean delivery	Yes	10	10	80	618
	No	14	12	74	303
Initiator of delivery	PTL	9	11	80	414
	pPROM	14	9	76	202
	Preeclampsia	16	11	74	121
	Abrupt	5	11	84	101
	Cervical insufficiency	15	10	75	48
	Fetal indication	17	20	63	35
Duration of labor	0 h	13	12	74	227
	≤12 h	17	9	74	211
	>12 h	8	11	82	483
Duration of ROM	<1 h	11	11	78	539
	1–24 h	10	10	79	144
	>24 h	12	11	78	238
Maximum no. of infants		103	99	719	921
Percent		11	11	78

These are row percents. preeclampsia (PE),; prolonged preterm rupture of membranes (pPROM),; preterm labor (PTL),; rupture of membranes (ROM).

Placental Cultures and Histologic Findings

Placental inflammation and presence of microorganisms were not associated with low MDI. The only placental finding associated with low MDI was thrombosis of fetal stem vessels; however, there were only 40 infants with this exposure (Tables 4 and 5).

TABLE 4.

Risks of MDI Scores Associated With Organisms Cultured From the Placenta

Organisms		<55	55–69	≥70	Row N
Total sample
No. of organisms isolated	0	11	10	79	435
	1	10	11	78	201
	2+	11	11	78	198
Any anaerobe	Yes	8	10	81	224
	No	12	11	77	610
Any aerobe	Yes	13	11	76	260
	No	10	11	79	574
Any Mycoplasma	Yes	15	12	72	85
	No	10	11	79	749
Skin organismsa	Yes	10	11	79	154
	No	11	11	78	680
Vaginal organismsb	Yes	10	12	78	132
	No	11	11	78	702
Maximum no. of infants		91	90	653	834
Percent		11	11	78

Cesarean delivery sample
No. of organisms isolated	0	11	11	78	333
	1	8	9	83	142
	2+	10	9	81	81
Any anaerobe	Yes	5	10	84	116
	No	11	10	79	440
Any aerobe	Yes	12	7	81	129
	No	9	11	80	427
Any Mycoplasma	Yes	10	10	80	41
	No	10	10	80	515
Skin organismsa	Yes	7	12	81	69
	No	10	10	80	487
Vaginal organismsb	Yes	7	11	82	55
	No	10	10	80	501
Maximum no. of infants		55	57	444	556
Percent		10	10	80

These are row percents.

^aCorynebacterium sp, Propionebacterium sp, Staphylococcus sp.

^bPrevotella bivia, Lactobacillus sp, Peptostreptococcus magnus, Gardnerella vaginalis.

TABLE 5.

Risks of MDI Scores Associated With Histologic Characteristics of the Placenta, Total Sample

Histologic Characteristic		< 55	55–69	≥70	Row N
Inflammation chorionic platea	Yes	9	12	79	159
	No	11	11	78	679
Inflammation external	Yes	12	10	78	295
membranesb	No	11	12	78	547
Neutrophil infiltration fetal stem	Yes	12	10	78	209
vessels	No	10	11	78	621
Umbilical cord vasculitisc	Yes	11	13	76	137
	No	11	11	79	682
Thrombosis of fetal stem vessels	Yes	23	5	73	40
	No	10	11	79	789
Infarct	Yes	9	15	75	142
	No	11	10	78	705
Increased syncytial knots	Yes	12	14	74	174
	No	11	10	79	676
Decidual hemorrhage/	Yes	11	12	77	129
Fibrin deposition	No	11	10	78	708
Maximum no. of placentas		94	94	669	857
Percent		11	11	78

These are row percents.

^aStage 3 and severity 3.

^bGrades 3 and 4.

^cGrades 3, 4, and 5.

Infant Characteristics

The risk of MDI <55 decreased with increasing gestational age, birth weight, birth weight z score, and head circumference z score. Boys were also at moderately higher risk than girls of MDI <55 (Table 6).

TABLE 6.

Risks of MDI Scores Associated With Characteristics of the Infants

Characteristics of the Infant		<55	55–69	≥70	Row N
Gender	Male	15	13	72	472
	Female	8	8	84	449
Gestational	23–24	15	11	74	163
Age, wk	25–26	12	10	78	435
	27	8	11	80	323
Birth weight, g	≤750	14	13	73	318
	751–1000	11	10	79	421
	>1000	5	10	85	182
Birth weight	< –2	22	14	68	51
z scorea	≥ –2, < –1	15	12	73	123
	≥ –1	10	10	80	747
Head	< –2	21	10	70	73
circumference	≥ –2, < –1	14	14	72	207
z scorea	≥ –1	9	18	81	609
Maximum number of infants		103	99	719	921
Percent		11	11	78

These are row percents.

^aBirth weight and head circumference z scores based on Yudkin et al²⁰ standard.

Multivariable Analyses

In multivariable models that included the total sample, heightened risk of MDI <55 was predicted by nonwhite race, maternal education <12 years, maternal BMI >30, male gender, and birth weight z score >2 SDs below the mean. When the sample was limited to children for whom we had information about histologic characteristics, the same variables were predictive of MDI <55, whereas thrombosis of fetal vessels provided additional information about increased risk (odds ratio 3.1, 95% confidence interval [CI] 1.2, 7.7). For both total sample and the histology-limited sample, the only significant risk factors for MDI 55 to 69 were male and nonwhite race. Multinomial models without a hospital cluster term provided results that were very similar to those provided by individual logistic regression models that included a hospital cluster term (Table 7).

TABLE 7.

Multivariable Analysis of Variables Associated With Low MDI

Variables	<55	55–69
Total sample (n = 921)
Gestational age 23–24 wk	1.9 (0.97, 3.6)	1.0 (0.5, 1.9)
Gestational age 25–26 wk	1.2 (0.7, 2.1)	0.8 (0.5, 1.3)
Single mother	1.3 (0.7, 2.2)	1.2 (0.7, 2.1)
BMI >30	1.9 (1.1, 3.1)a	1.3 (0.9, 2.3)
Vaginal/cervical infection	1.2 (0.6, 2.2)	1.4 (0.8, 2.6)
Cesarean delivery	0.8 (0.5, 1.2)	0.8 (0.5, 1.4)
BWZ < –2	2.2 (1.5, 7.5)a	1.9 (0.7, 4.8)
Male	2.5 (1.6, 4.1)a	2.0 (1.3, 3.2)a
Nonwhite race	2.3 (1.4, 3.8)a	2.1 (1.3, 3.5)a
Mother's education <12 y	2.8 (1.6, 4.9)a	0.9 (0.5, 1.8)
Mother's education >16 y	0.8 (0.4, 1.5)	0.7 (0.4, 1.2)
Total histology/BSID sample (n = 857)
Gestational age 23–24 wk	1.9 (0.9, 3.9)	1.1 (0.6, 2.1)
Gestational age 25–26 wk	1.1 (0.6, 2.0)	0.8 (0.4, 1.3)
Single mother	1.2 (0.7, 2.2)	1.2 (0.7, 2.0)
BMI >30	2.0 (1.1, 3.5)a	1.2 (0.7, 2.1)
Vaginal/cervical infection	1.2 (0.6, 2.4)	1.4 (0.7, 2.7)
Cesarean delivery	0.9 (0.5, 1.5)	0.9 (0.5, 1.5)
BWZ < –2	2.8 (1.1, 6.9)a	1.9 (0.7, 5.0)
Male	2.7 (1.6, 4.5)a	2.1 (1.3, 3.4)a
Thrombosis fetal vessels	3.1 (1.2, 7.7)a	0.6 (0.1, 2.5)
Increased syncytial knots	1.4 (0.8, 2.5)	1.5 (0.9, 2.5)
Nonwhite race	2.2 (1.3, 3.8)a	2.2 (1.3, 3.7)a
Mother's education <12 y	3.4 (1.9, 6.2)a	1.1 (0.6, 2.2)
Mother's education >16 y	0.7 (0.4, 1.4)	0.7 (0.4, 1.3)

Risk ratios and 95% CIs. The referent group is infants with MDI ≥70. BWZ, birthweight z score.

^ap < .05.

Discussion

Three maternal attributes (obesity, nonwhite race, and maternal education) and 1 placental finding (thrombosis of fetal stem vessels) were associated with impaired early cognitive function in extremely preterm infants. Neonatal predictors of such impairment were birth weight z score < –2 (ie, below the third centile) and male gender. In general, stronger associations were found for MDI <55 than for MDI 55 to 69.

Our finding that male gender was predictive of impaired early cognitive function agrees with most^5,28 but not all²⁹ studies of extremely preterm infants.^28,30,31 The mechanisms for this association are not well understood. Boys born prematurely have a greater risk of morbidities, such as cerebral white matter damage and chronic lung disease, and gender differences in brain structure and response to injury might also be important.³²

In our multivariable regression models, fetal growth restriction is a predictor of an MDI <55, while not predicting a less severely low MDI. Others, too, have found that growth-restricted infants are at increased risk of cognitive limitations.^33,34 One explanation invokes placental dysfunction, which can restrict placental transfer of substrates for brain development.³⁵ An alternative explanation is that fetal growth restriction increases the risk of neonatal complications^36,37 that are more proximally related to cognitive impairment.³⁸

Children who experience a major postnatal infection (including bacteremia and necrotizing enterocolitis) are more likely than others to have cerebral white matter injury³⁹ and cognitive impairment.^15,17 Perinatal inflammation, as evidenced by chorioamnionitis, is also a risk factor for white matter injury and cerebral palsy.⁴⁰ We, however, did not find any evidence that antenatal inflammation predicts cognitive impairment.

The only histologic characteristic of the placenta associated with cognitive impairment was thrombosis of fetal stem vessels, which is considered a lesion of uteroplacental circulation and has been associated with adverse neonatal outcomes.⁴¹ In addition to stillbirth,⁴² perinatal liver disease,⁴³ and intrauterine growth restriction in monochorionic twins,⁴⁴ fetal vessel thrombosis has been linked to adverse neurologic outcomes, including neonatal encephalopathy⁴⁵ and neonatal stroke.⁴⁶ In the ELGAN cohort, it has been associated with ventriculomegaly of the offspring,⁴⁷ which, in turn, is associated with low MDI.⁴

One antecedent that we identified, maternal obesity, is of particular interest because it is potentially modifiable and its prevalence is increasing.⁴⁸ Although maternal obesity is predictive of higher maternal blood levels of C-reactive protein⁴⁹ and interleukin-6,⁵⁰ we found no indirect evidence to support a link between maternal inflammation, as indicated by placental findings and initiators of preterm delivery, and subsequent impaired infant cognitive function. Maternal obesity is associated with multiple complications of pregnancy⁵¹ and with perinatal mortality.^52–55 Perhaps most relevant to our study is the association with fetal growth restriction.^19,56 In addition, obesity is linked to chronic illnesses^57,58 and social disadvantage,⁵⁹ which might affect the mother-infant interaction and the resources available to enhance the home environment.

The association of assisted conception with a low MDI, seen in univariable analyses, was no longer significant in multivariable analyses. The most likely explanation is that the risk information provided by the assisted conception variable is more strongly conveyed by 2 other correlates of socioeconomic status: race and education.

We are not aware of any other large prospective studies relating early cognitive functioning in extremely preterm infants to placental findings or initiators of preterm delivery. A potential limitation of this study is that MDI <55 is only moderately predictive of cognitive impairment at school age.⁶⁰ The strengths of this prospective study include the large number of infants, selection of participants based on gestational age rather than birth weight, and evaluations by individuals who were not aware of infants’ medical or social histories.

This study has implications for efforts to prevent cognitive impairment among extremely premature infants. One antecedent identified here, maternal obesity, is potentially modifiable during the preconception period and pregnancy. The prevalence of another antecedent, fetal growth restriction, could be decreased by interventions to prevent severe preeclampsia, which occurred in 13% of our sample of extremely preterm infants.

To what extent the cognitive adversities associated with fetal growth restriction reflect postnatal phenomena (especially those associated with chronic lung disease) remains unclear. Postnatal disorders appear to be involved in explaining some of the relationship between chronic lung disease and impaired development.⁷ If the occurrence of these disorders can be minimized, then there is hope that the influence of fetal growth restriction on cognitive deficiencies can be reduced.

CONCLUSIONS

Antenatal characteristics associated with impaired mental development include maternal BMI >30, nonwhite race, maternal education, thrombosis of fetal stem vessels, and male gender. Our findings do not support an important role of prenatal infection or inflammation in the development of impaired early cognitive function among extremely preterm infants.

Glossary

BSID-II

Bayley Scales of Infant Development, second edition

CI

confidence interval

ELGAN

extremely low gestational age newborn

GMFCS

Gross Motor Function Classification System

MDI

Mental Development Index

OR

odds ratio