Effect of Treatment of Mild Gestational Diabetes on Long-Term Maternal Outcomes

Abstract

Objective

The main purpose of this article is to evaluate whether identification and treatment of women with mild gestational diabetes mellitus (GDM) during pregnancy affects subsequent maternal body mass index (BMI), anthropometry, metabolic syndrome, and risk of diabetes.

Study Design

This is a follow-up study of women who participated in a randomized controlled treatment trial for mild GDM. Women were enrolled between 5 and 10 years after their index pregnancy. Participants underwent blood pressure, height, weight, and anthropometric measurements by trained nursing personnel using a standardized approach. A nurse-assisted questionnaire regarding screening and treatment of diabetes or hypercholesterolemia, diet, and physical activity was completed. Laboratory evaluation included fasting serum glucose, fasting insulin, oral glucose tolerance test, and a lipid panel. Subsequent diabetes, metabolic syndrome, obesity, and adiposity in those diagnosed with mild GDM and randomized to nutritional counseling and medical therapy (treated) were compared with those who underwent routine pregnancy management (untreated). Multivariable analyses were performed adjusting for race/ethnicity and years between randomization and follow-up visit.

Results

Four-hundred fifty-seven women with mild GDM during the index pregnancy were included in this analysis (243 treated; 214 untreated) and evaluated at a median 7 years after their index pregnancy. Baseline and follow-up characteristics were similar between treatment groups. Frequency of diabetes (9.2 vs. 8.5%, p =0.80), metabolic syndrome (32.2 vs. 34.3%, p =0.63), as well as adjusted mean values of homeostasis model assessment for insulin resistance (2.5 vs. 2.3, p =0.11) and BMI (29.4 vs. 29.1 kg/ m², p =0.67) were also not different.

Conclusion

Identification and treatment of women with mild GDM during pregnancy had no discernible impact on subsequent diabetes, metabolic syndrome, or obesity 7 years after delivery.

The original impetus behind the diagnosis of gestational diabetes mellitus (GDM) was to establish maternal risk for subsequent development of diabetes.¹ In the majority of women with GDM, hyperglycemia resolves immediately postpartum, and such women have a 17 to 63% risk of developing diabetes within 5 to 16 years after the index pregnancy.² Women with a history of GDM have a sevenfold increased risk of developing diabetes in the future.³ Consequently, the American Diabetes Association recommends follow-up testing in women with a history of GDM no less than every 3 years, or sooner if impaired glucose tolerance is detected immediately after delivery.⁴ A potential maternal benefit of the diagnosis, treatment, and surveillance after pregnancy is prevention of or delay in the future onset of overt diabetes. Indeed, pharmacological or lifestyle interventions in individuals at high risk of developing diabetes have been shown to be beneficial.⁵

Women with a history of GDM have lower pancreatic β-cell function and higher insulin resistance within years of their initial diagnosis.⁶ Additional risk factors for attenuated β-cell function include weight gain and increased visceral fat volume with rising levels of C-reactive protein and falling levels of adiponectin.^7,8 Therefore, treatment of women with GDM during pregnancy and adoption of positive lifestyle changes reinforced over the course of prenatal care could plausibly result in improved β-cell function and reduced diagnosis of diabetes in the years following the diagnosis. However, future risk of diabetes is not the only long-term maternal risk associated with GDM. It has also been suggested that glucose intolerance identified during pregnancy is associated with latent metabolic syndrome and risk of cardiovascular disease.^9–11 This is true even in women with milder degrees of glucose intolerance. For example, the prevalence of metabolic syndrome 3 months after delivery in women with a history of GDM was twofold higher when compared with women with normal glucose tolerance.¹² These data suggest that identification of women with GDM also provides an opportunity for early intervention to modify the risk of subsequent metabolic syndrome and cardiovascular disease.

The purpose of this follow-up cohort analysis of women identified with mild GDM and who were enrolled in the Eunice Kennedy Shriver National Institute of Child Health and Human Development Maternal-Fetal Medicine Units (MFMU) Network’s multicenter randomized controlled treatment trial¹³ was to determine whether identification and treatment of such women decrease the risk of diabetes and subsequent metabolic syndrome or obesity up to 10 years after diagnosis when compared with women with mild GDM who were not treated during pregnancy.

Materials and Methods

Between February 2012 and September 2013, we conducted a follow-up study of the women who participated in the MFMU Network’s randomized clinical trial (RCT) for mild GDM.¹³ Mild GDM was defined as a fasting glucose of less than 95 mg/dL but two of three timed measurements that exceeded established thresholds following a 100-g oral glucose tolerance test (OGTT): 1 hour, 180 mg/dL: 2 hours, 155 mg/dL: and 3 hours, 140 mg/dL. Women with mild GDM who consented were randomly assigned to treatment with formal nutritional counseling and diet therapy with insulin, if required, or usual prenatal care.¹³ Results of the glucose tolerance tests were shared with women blinded to their diagnosis after delivery. They were advised to have their fasting glucose checked annually and to notify their obstetrician of their diagnosis in the event of another pregnancy. Eligibility for the follow-up study included enrollment in the RCT at a center still participating in the MFMU Network at the time of the follow-up study (12 of 16 centers; 94% of the original RCT patients). Mothers were contacted between 5 and 10 years after the index pregnancy. With institutional review board approval, each center retrieved contact information from the original study charts and, together with web-based location software, attempted to locate every woman from the RCT enrolled at their center. Women agreeing to participate in this follow-up study were scheduled for a clinic visit and asked to fast for 8 hours or more prior to their visit.

At the study visit, trained nursing personnel obtained height and weight measurements using a hospital grade scale and stationary stadiometer. Blood pressure measurements were performed using a standardized methodology which included two measurements separated by 1 minute using either a manual or digital sphygmomanometer after 10 minutes in the seated position without legs crossed. Waist and hip circumferences were measured by research nurses who had undergone standardized instruction in these measurement techniques.¹⁴ In a similar fashion, measurements of the triceps, subscapular, suprailiac, and thigh skinfolds were obtained using Harpenden calipers. A nurse-assisted questionnaire that included questions on subsequent health history, diet, and physical activity was completed. Laboratory evaluation included a fasting serum glucose, fasting insulin, and lipid panel including measurements of triglycerides and cholesterol. A 75-g oral glucose tolerance test was performed in women who were not being treated for diabetes. Blood samples were collected and sent to Northwest Lipid Metabolism and Diabetes Research Laboratories for analysis.

Outcomes of interest for this follow-up study were subsequent diabetes, metabolic syndrome, obesity, and adiposity in the mother. Diabetes was diagnosed in women either currently treated for diabetes or who met American Diabetes Association diagnostic criteria for an abnormal 75-g glucose tolerance test.⁴ Homeostasis model assessment for insulin resistance (HOMA-IR) was calculated as fasting glucose (mmol/L) x fasting insulin (μU/mL)/22.5.¹⁵ The diagnosis of metabolic syndrome was based on criteria of both the American Heart Association and the National Heart Lung Blood Institute. Metabolic syndrome was diagnosed when three or more of the following five criteria were met: (1) a waist circumference greater than 88 cm, (2) serum triglycerides150 mg/dL or greater or current treatment for hyperlipidemia, (3) high-density lipoprotein (HDL) cholesterol less than 50 mg/dL, (4) a systolic blood pressure of 130 mm Hg or greater or a diastolic blood pressure 85 mm Hg or greater or current treatment for hypertension, and (5) a fasting serum glucose of 100 mg/dL or more or current treatment for diabetes (oral agent or insulin).¹⁶ Body mass index (BMI; kg/m²) was evaluated as a continuous outcome, as well as a categorical outcome according to the World Health Organization (WHO) obesity classification system, and compared according to treatment group during the index pregnancy. Additionally, maternal body circumference measurements and skinfold thicknesses were compared as continuous measures according to treatment.

Chi-square, Fisher’s exact, and Wilcoxon rank sum tests were used to assess differences between treatment groups in baseline and follow-up characteristics. Diabetes and metabolic syndrome were evaluated as binomial outcomes using log-binomial regression to estimate adjusted relative risks (RR) and 95% confidence intervals (CI). BMI, HOMA-IR, skin-folds (thigh, tricep, subscapular, suprailiac), and circumferences (waist, hip) were evaluated as continuous outcomes using least-square means general linear regression to estimate adjusted means and 95% CI. Continuous variables were log-transformed when appropriate. Multivariable models were adjusted for race/ethnicity and years between randomization and the follow-up visit. SAS software (SAS Institute, Cary, NC) was used for the analyses. All tests were two-tailed and p < 0.05 was used to define statistical significance without adjustment for multiple comparisons. No imputation for missing data was performed.

Results

As shown in ►Fig. 1, of the 905 eligible women with mild GDM from the original trial, 666 (74%) were contacted. Of these, 483 consented and participated in the follow-up study. Twenty-six women were excluded from this analysis because they were pregnant at the time of follow-up. Thus, 457 women (50% of the original study cohort) were included in this analysis (243 treated; 214 untreated). Compared with nonparticipant women, the women included in this analysis were similar in age, parity, gestational age at enrollment, and OGTT results during the index pregnancy, but were less often Hispanic (55 vs. 61%) and had a slightly lower mean 50-g glucose loading-test result (156 vs. 160 mg/dL). A total of 430 women (94%) agreed to glucose tolerance testing and had blood drawn for cholesterol and lipid profile during their follow-up visit. These women were included in the analysis of metabolic syndrome. Maternal baseline characteristics from the index pregnancy and selected follow-up characteristics are shown in ►Table 1. Maternal age, years since randomization, race, parity, BMI, and number of subsequent pregnancies were no different between the treatment groups. Glucose testing results during the index pregnancy were also similar between groups.

Fig. 1.

Screening and enrollment flow diagram. GDM, gestational diabetes mellitus.

Table 1.

Baseline and follow-up maternal characteristics, in treated and untreated women

Characteristic	Treated n = 243	Untreated n = 214	p-Valuea
Baseline (index pregnancy)
Age (y)	29 (26−33)	29 (25−33)	0.33
Race/Ethnicity			0.41
White	82 (33.7)	58 (27.1)
Black	26 (10.7)	22 (10.3)
Hispanic	128 (52.7)	125 (58.4)
Other	7 (2.9)	9 (4.2)
Primigravida	49 (20.2)	53 (24.8)	0.24
BMI (kg/m2) prepregnancy	25.9 (22.9−29.4)	25.7 (22.6−28.9)	0.62
BMI (kg/m2) at enrollment	29.7 (26.3−33.2)	29.7 (27.0−33.0)	0.67
Gestational age enrollment	29.0 (27.7−30.1)	28.9 (27.9−30.1)	0.89
50 mg glucose challenge (mg/dL)	155 (145−170)	157 (145−170)	0.94
100 g glucose tolerance test (mg/dL)
Fasting result	88 (84−91)	88 (83–91)	0.35
1-hour result	190 (181−203)	194 (185−203)	0.05
2-hour result	170 (160−182)	171 (160−182)	0.84
3-hour result	144 (120−155)	141 (114−156)	0.23
Treated with insulin	21 (8.8)	0 (0.0)	<0.001
Follow-up
Years between randomization and follow-up	7 (6−8)	7 (6−8)	0.54
Age at follow-up (y)	36 (33−40)	36 (32−40)	0.47
Additional pregnancies	1 (0−1)	0 (0–1)	0.47
Additional pregnancies, categories			0.50
0	118 (48.6)	111 (51.9)
1	80 (32.9)	72 (33.6)
2 or more	45 (18.5)	31 (14.5)
Weight gain since index pregnancy (follow-up weight—prepregnancy weight; kg)	6.3 (2.0−11.3)	5.8 (0.9−11.8)	0.63
Months breastfed index child	4 (0−12)	4 (0−9)	0.40

Abbreviation: BMI, body mass index.

Note: Data are median (interquartile range) or n (%) unless otherwise specified.

^aBased on chi-square test or Wilcoxon rank sum test.

Follow-up glucose testing results and reported diabetes treatment are listed in ►Table 2. The likelihood of subsequent overt diabetes diagnosis was the same between women with mild GDM who were treated during pregnancy and those who received routine prenatal care (9.2 vs. 8.5%, p =0.80). This remained true after adjustment for race/ethnicity and years between randomization and follow-up (RR = 1.10, 95% CI: 0.60–2.03). There was also no difference between treatment groups in rates of impaired fasting glucose defined as fasting glucose of 100 mg/dL or greater (29.1 vs. 27.1%, p = 0.66). Of note, the majority of diabetes diagnoses in this cohort (76%) occurred during testing for this follow-up study. According to participant questionnaire responses (data not shown), only 90 women, 27% of those treated compared with 11% in those untreated, recalled undergoing postpartum follow-up testing for subsequent diabetes.

Table 2.

Maternal diabetes diagnosis 5 to 10 years after diagnosis of mild gestational diabetes, in treated and untreated women

Diabetes diagnosis, and its components	Treated n = 243	Untreated n = 214	p-Valuea	Adjusted relative risk (95% CI)b
Diabetes diagnosis	21 (9.2)	17 (8.5)	0.80	1.10 (0.60−2.03)
75 g glucose tolerance test results
Fasting glucose (mg/dL)	94 (89−101)	93 (88−100)	0.35
Fasting glucose ≥ 126 mg/dL	7 (3.1)	6 (3.0)	0.97
2-hour glucose (mg/dL)	122 (102−159)	123 (96−151)	0.26
2-hour glucose ≥ 200 mg/dL	15 (7.0)	13 (6.8)	0.93
Treated with oral agent	3 (1.3)	3 (1.4)	1.00
Treated with insulin	1 (0.4)	2 (0.9)	0.60

Abbreviation: CI, confidence interval.

Note: Data are median (interquartile range) or n (%) unless otherwise specified.

^aBased on chi-square test, Fisher’s exact test, or Wilcoxon rank sum test.

^bAdjusted for race-ethnicity and years between randomization and follow-up visit.

As shown in ►Table 3, identificationof metabolic syndrome among the 430 women who agreed to a blood draw was no different between treatment groups (32.2 vs. 34.3%, p =0.63). This remained true after consideration of race/ethnicity and the number of years since diagnosis of mild GDM (RR 0.95, 95% CI: 0.73–1.25). Further, there was no impact of treatment during pregnancy on any of the constituents of metabolic syndrome up to 10 years after the original diagnosis. The diagnosis of metabolic syndrome was most commonly based on waist circumference and HDL cholesterol criteria. As with the diagnosis of diabetes, history of treatment for hyperlipidemia or hypertension accounted for a minority of those meeting each of those constituent criteria.

Table 3.

Metabolic syndrome 5 to 10 years after diagnosis of mild gestational diabetes, in treated and untreated women

Metabolic syndrome, and its components	Treated n = 243	Untreated n = 214	p-Valuea	Adjusted relative risk (95% CI)b
Metabolic syndrome	73 (32.2)	69 (34.3)	0.63	0.95 (0.73–1.25)
Waist circumference > 88 cm	148 (60.9)	127 (59.4)	0.73
Elevated triglycerides or treatment	70 (30.8)	50 (25.0)	0.18
Triglycerides ≥ 150 mg/dL	68 (30.0)	48 (24.1)	0.18
Medication for hyperlipidemia	3 (1.2)	4 (1.9)	0.71
HDL-C < 50 mg/dL	133 (58.6)	124 (62.3)	0.43
Elevated blood pressure or treatment	45 (18.5)	39 (18.2)	0.94
Blood pressure ≥ 130 mm Hg systolic or ≥ 88 mm Hg diastolic	38 (15.6)	29 (13.6)	0.53
Medication for hypertension	12 (4.9)	15 (7.0)	0.35
Elevated fasting glucose or treatment	70 (30.6)	57 (28.4)	0.62
Fasting glucose ≥ 100 mg/dL	66 (29.1)	54 (27.1)	0.66
Treated with oral agent or insulin	4 (1.7)	5 (2.4)	0.74

Abbreviations: CI, confidence interval; HDL-C, high-density lipoprotein cholesterol.

Note: Data are n (%).

^aBased on chi-square test, Fisher’s exact test, or Wilcoxon rank sum test.

^bAdjusted for race/ethnicity and years between randomization and follow-up visit.

►Table 4 shows that mean adjusted HOMA-IR (2.5 vs. 2.3, p =0.11) and mean adjusted maternal BMI (29.4 vs. 29.1 kg/ m², p =0.67) at the time of follow-up were no different between treatment groups. When analyzed according to the WHO obesity classification system, there was also no difference between each BMI category (see ►Fig. 2). As shown, the majority of women in both groups were either overweight or obese (defined by a BMI of 25 or greater) at follow-up. Also shown in ►Table 4, measurement of a variety of skinfold thicknesses or hip and waist circumferences failed to reveal any differences in maternal anthropometric measurements according to treatment group.

Table 4.

Maternal HOMA-IR, BMI, and anthropometric measurements 5 to 10 years after diagnosis of mild gestational diabetes, in treated and untreated women

Measurement	Adjusted mean (95% CI)b		p-Valueb
	Treated n = 243	Untreated n = 214
HOMA-IRa	2.5 (2.3, 2.8)	2.3 (2.0, 2.5)	0.11
BMIa (kg/m2)	29.4 (28.5, 30.2)	29.1 (28.2, 30.1)	0.67
Skinfolds (mm)
Thigha	31.3 (29.3, 33.5)	30.8 (28.7, 33.0)	0.68
Triceps	27.8 (26.5, 29.0)	26.8 (25.5, 28.2)	0.25
Subscapular	27.9 (26.5, 29.2)	27.2 (25.8, 28.7)	0.48
Suprailiac	26.5 (25.2, 27.9)	25.6 (24.1, 27.0)	0.27
Circumference (cm)
Waist	96.2 (94.3, 98.1)	94.2 (92.2, 96.3)	0.12
Hipa	108.3 (106.6, 110.0)	106.9 (105.1, 108.7)	0.19

Abbreviations: BMI, body mass index; CI, confidence interval; HOMA-IR, homeostasis model assessment for insulin resistance.

^aLog values back transformed.

^bBased on least squares means general linear models and adjusted for race-ethnicity and years between randomization and follow-up visit.

Fig. 2.

Percent of participants in each category of body mass index, by treatment group.

Conclusion

There are several important findings from this analysis of long-term outcomes in women diagnosed with mild GDM. First, fewer than 1 in 10 women with mild GDM developed overt diabetes within a median 7 years of their diagnosis and the majority of these were not diagnosed until the time of this study protocol. Importantly, treatment during pregnancy did not modify this risk. Next, metabolic syndrome was identified in one-third of these women during follow-up, but again, treatment during pregnancy did not affect its frequency. Finally, treatment of mild GDM during pregnancy did not reduce the likelihood of maternal obesity later in life nor affect any maternal anthropometric measurements up to 10 years after the index pregnancy. When taken together, while treatment of mild GDM has been shown to improve immediate pregnancy outcomes,¹⁵ long-term benefits of treatment during pregnancy are not apparent in women 5 to 10 years after delivery.

Women identified with GDM likely have reduced insulin secretion or chronic insulin resistance before pregnancy and are therefore already at increased risk of developing diabetes later in life.^6,17,18 In contrast to an expected rate between 17 and 63%, however, the rate of development of diabetes in this cohort of women was less than 10% within 10 years of the index pregnancy.¹⁷ This is perhaps not surprising considering that these women were at the mild end of the diagnostic spectrum.¹⁸ Conversely, almost one-third of women had evidence of impaired glucose tolerance based on fasting glucose values. That said, the development of glucose intolerance and subsequent diabetes has been more strongly linked to maternal weight and weight gain subsequent to the diagnosis of GDM.^19,20 In this study, the majority of women gained more than 5 kg since pregnancy and were considered overweight or obese at follow-up.

Similar to the rate of impaired glucose tolerance in this cohort, approximately one-third had evidence of metabolic syndrome. The likelihood of metabolic syndrome is increased even in women previously identified with mild glucose intolerance not meeting criteria for the diagnosis of GDM.¹⁴ As with diabetes though, other cardiometabolic risk factors like weight gain and obesity make the diagnosis of metabolic syndrome more likely. While maternal metabolic profile remains linked to the diagnosis of GDM after adjustment for BMI,²¹ studies of the differential impact of glucose intolerance or maternal weight in development of hypertension or cardiovascular disease indicate maternal weight is the stronger of these two factors.²² Our results for metabolic syndrome, which is heavily based on waist circumference, suggest that treatment of GDM during pregnancy alone is insufficient to impact its future onset. The effect of treatment during pregnancy on subsequent BMI, HOMA-IR, or adiposity was also negligible. Ongoing efforts to mitigate the additive risk of GDM and obesity through dietary, lifestyle, and physical activity after the pregnancy are likely necessary. Indeed, it has been shown that the most potent reductions in risk of developing subsequent diabetes, and by inference metabolic syndrome, have been achieved through intensive and ongoing lifestyle modification.⁵

Like other investigators, we found that a large percentage of women in this trial did not recall undergoing postpartum follow-up testing for subsequent diabetes.^23,24 A majority (76%) of the women with diabetes at follow-up were identified during the testing for this study. Such low rates of postpartum follow-up in these women at risk of diabetes, metabolic dysfunction, and cardiovascular disease have prompted calls for increased awareness through professional and public health campaigns focused on women with a history of GDM.²⁵ The American College of Obstetricians and Gynecologists recently emphasized the importance of postpartum glucose screening in optimizing care of women with gestational diabetes after delivery.²⁶ Our results indicate that women with a history of mild GDM are at increased risk of diabetes, albeit lower than expected. Unfortunately, however, dietary therapy during pregnancy (and even insulin when indicated) did not translate into reductions in subsequent diabetes or metabolic syndrome. These findings are also consistent with other reports that obesity and weight gain are potent cofactors in the development of diabetes and metabolic syndrome.

Limitations of this analysis include its relatively small size and the limited recall information available between delivery and follow-up evaluation. Also, only half of the original cohort was evaluated 7 to 9 years after the index pregnancy. Those women who were included in this analysis were less often Hispanic and had a slightly lower mean 50-g glucose loading-test result. Despite its small size and these differences though, this cohort allows a unique opportunity to evaluate the long-term implications of identification and treatmentof mild GDM during pregnancy. Another weakness is that we have an incomplete picture of what happened during the interval between delivery and follow-up evaluation. However, according toparticipantquestionnaire responses, weknow that recall of the GDM diagnosis (90 vs. 56% in the treated and untreated groups, respectively) and recall of diabetes testing after delivery were positively linked to treatment during pregnancy. Despite this, there was no difference in weight gain after pregnancy. Although women in the entire study cohort eventually knew of their diagnosis after delivery, responses of those women treated during pregnancy imply a greater impact of the diagnosis than in those not treated during pregnancy. Unfortunately, however, this did not translate into any measurable benefit with regard to weight gain, obesity, metabolic dysfunction, or subsequent diagnosis of diabetes.

In summary, this follow-up study of women enrolled in a randomized treatment trial for mild GDM reveals fewer women than expected develop overt diabetes later in life. Treatment of those women during pregnancy, despite indirect evidence of a significant impact on diabetes follow-up, had no discernible effects on likelihood of subsequent diabetes or metabolic syndrome an average of 7 years after pregnancy. Treatment during pregnancy also did not have any long-lasting impact on exercise regimen or adiposity as determined by anthropometric measurement years after the diagnosis. Conversely, pregnancy outcomes in women with mild GDM have been shown to be improved with treatment due to less excessive infant size and consequent shoulder dystocia.¹³ In light of our findings, we believe the benefits of the diagnosis and treatment of mild GDM during pregnancy are primarily limited to immediate obstetric and neonatal outcomes rather than longer-term maternal outcomes such as subsequent diabetes or metabolic syndrome.

Appendix A

In addition to the authors, other members of the Eunice Kennedy Shriver National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network are as follows:

Universityof Texas Southwestern Medical Center, Dallas, TX-L. Moseley, J. Price, A. Sias, K. Gonzales, Y. Delira

The Ohio State University, Columbus, OH -F. Johnson, S. Wylie, D. Habash, S. Heintzman, E. Nini, J. Iams, C. Durnwald University of Utah Health Sciences Center, Salt Lake City, UT -K. Hill, M. Thompson, A. Sowles, G. Anderson (Intermountain Healthcare)

Columbia University, New York, New York -S. Bousleiman, M. Talucci, V. Carmona, I. Quezada, A. Ranzini (St. Peter’s University Hospital), M. Lake (St. Peter’s University Hospital),

S. Davis (St. Peter’s University Hospital), M. Hoffman (Christiana Care), S. Lynch (Christiana Care), J. Benson (Christiana Care), C. Kitto (Christiana Care), L. Plante (Drexel U.), C. Tocci (Drexel U.), Y. Williams (Drexel U.)

Brown University, Providence, RI -D. Allard, B. Anderson,

K. Pereda, E. Hipolito, J. McNamara

University of Alabama at Birmingham, Birmingham, AL - S. Harris, A. Tita, A. McClain, J. Sheppard

University of North Carolina at Chapel Hill, Chapel Hill, NC- K. Clark, B. Eucker, S. Timlin, K. Pena, T. Varney

MetroHealth Medical Center-Case Western Reserve University, Cleveland, OH - W. Dalton, C. Milluzzi, P. Catalano, B. Mercer University of Texas Medical Branch, Galveston, TX -A. Salazar,

Acosta, S. Bouse, G. Hankins, S. Jain

Northwestern University, Chicago, IL -G. Mallet, M. Ramos-Brinson, C. Collins, L. Stein, M. Dinsmoor (NorthShore HealthSystems-Evanston Hospital)

University of Texas Health Science Center at Houston- Children’s Memorial Hermann Hospital, Houston, TX - F. Ortiz,

Sibai, B. Rech, L. Garcia

The George Washington University Biostatistics Center, Washington, DC - E. Thom, L. Doherty, L. Mele, T. Spangler

Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, MD - C. Spong, S. Tolivaisa