At term, neonates with accelerated growth may be large for gestational age (LGA, birth weight of 90th percentile or more for gestational age), macrosomic (birth weight of at least 4,000 g, regardless of gestational age), or both.1 Maternal complications associated with delivery of a macrosomic neonate are known: prolonged labor, third or fourth degree laceration, cesarean delivery, venous thromboembolism, and postpartum hemorrhage. Likewise, adverse outcomes of neonates with weight of at least 4,000 g are recognized: shoulder dystocia, fractures, brachial plexus palsy, asphyxia, stillbirth, and neonatal or infant mortality.2–4 However, there is a knowledge gap regarding morbidity and mortality among nonmacrosomic LGA neonates—birth weight 90th percentile or greater but less than 4,000 g—at 37 weeks of gestation or more (term).
There are four reasons to have improved understanding of adverse outcomes of nonmacrosomic LGA. First, the rate of LGA (12.1%) is higher than the rate of macrosomia (9.2%).5,6 Second, detection of LGA is more likely than that of macrosomia.7,8 Third, a randomized clinical trial of more than 800 LGA fetuses at term indicates that short-term morbidity with LGA is reduced with induction at 37.0–38.6 weeks of gestation as compared with expectant management.9 Fourth, it is uncertain whether the morbidities associated with accelerated growth at term are confined to macrosomic LGA or if it extends to nonmacrosomic LGA.6,10
The primary objective of this analysis was to evaluate whether delivery after labor of a neonate with a birth weight 90% or greater for gestational age but less than 4,000 g at 37–39 weeks of gestation is associated with increased composite maternal and neonatal morbidity.
This was a population-based retrospective cohort study using the Period Linked Birth-Infant Death Data Files of the U.S. Vital Statistics data from 2011 to 2013 assembled by the National Center for Health Statistics and reported annually by the Centers for Disease Control and Prevention. These data, ascertained through birth certificates, comprised all live births in the United States between 2011 and 2013 and were linked to infant deaths within the first year. Our study sample was restricted to women of U.S. residency who delivered between 2011 and 2013; had a singleton, nonanomalous gestation between 37 and 39 weeks of gestation; experienced labor (defined as vaginal delivery or cesarean delivery after trial of labor); had a neonate who was appropriate for gestational age (AGA) or LGA (categorized according to Alexander et al)11 with birth weight less than 4,000 g; had diabetes status recorded; and had birth data recorded using the 2003 revised birth certificate. We excluded women at 40 weeks of gestation or later, because LGA at this gestational age is defined as greater than 4,000 g.11
The updated 2003 revised birth certificate has been incorporated gradually on a statewide basis and is the currently used birth certificate by the Centers for Disease Control and Prevention. Compared with the 1989 birth certificate version, the 2003 version contains more detailed obstetric, medical and demographic data.12 Maternal morbidity measures have been added to the data files since 2011; however, a few items were dropped in 2014. Therefore, our study used data from 2011 to 2013. The revised birth certificate was used by 36 states and Washington, DC, in 2011, 38 states and DC in 2012, and 41 states and D.C. in 2013, which represented 83%, 86%, and 90% of live births in the United States, respectively. Because the data are publicly available and do not contain direct personal identifiers, this study was exempt from review by the institutional review board at the McGovern Medical School at the University of Texas Health Science Center at Houston.
The 2003 revision of the birth certificate replaced the “clinical estimate of gestation” with the “obstetric estimate of gestation.” Additional information of the methods for this obstetric estimate of gestation is available in the Vital Statistics dataset guidelines.13 The obstetric estimate of gestation is reported in completed weeks (ie, 37 weeks of gestation includes deliveries from 37 0/7 weeks through 37 6/7 weeks). The obstetric estimate was selected rather than the clinical estimate because there is increasing evidence showing greater validity of the obstetric estimate compared with the last menstrual period–based estimate. The National Center for Health Statistics transitioned to the obstetric estimate as the standard for estimating the gestational age of a neonate in 2014.14
The main exposure variable was fetal growth (nonmacrosomic LGA vs AGA). The primary outcomes included composite maternal morbidity and composite neonatal morbidity. Composite maternal morbidity included any of the following: maternal transfusion, ruptured uterus, unplanned hysterectomy, admission to intensive care unit, or unplanned operating room procedure after delivery. Composite neonatal morbidity included any of the following: Apgar score less than 5 at 5 minutes, assisted ventilation required for more than 6 hours, seizure or serious neurologic dysfunction, significant birth injury (ie, skeletal fracture(s), peripheral nerve injury, and soft tissue or solid organ hemorrhage requiring intervention), or neonatal mortality (defined as death within 27 days).12 The determination of death within 27 days was made by linking birth and death certificates (which has the variable “age at death in days”).
Differences in the maternal characteristics stratified by fetal growth status were examined using chi-square tests for categorical variables. The rates of composite maternal morbidity and composite neonatal morbidity were reported as the number of cases per 1,000 live births. We used multivariable Poisson regression models with robust error variance to examine the association between fetal growth status and the risk of composite maternal and neonatal morbidity outcomes, while adjusting for maternal age (younger than 20 years, 20–34 years, 35 years or older), race and ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, non-Hispanic other, unknown), maternal education (less than high school, high school, more than high school, unknown), marital status (married, not married), primiparous (yes, no, unknown), prepregnancy body mass index (BMI [calculated as weight in kilograms divided by height in meters squared], underweight [less than 18.5], normal weight [18.5–24.9], overweight [25–29.9], obesity class I [30.0–34.9], obesity class II [35.0–39.9], obesity class III [40 or greater], unknown), prenatal care (yes, no, unknown), cigarette use during pregnancy (yes, no, unknown), diabetic status (no diabetes, pregestational diabetes, gestational diabetes), hypertensive disorder (yes, no), gestational age, neonatal sex (male, female), and delivery year (2011, 2012, 2013). We repeated the same analysis within strata of diabetic groups. The results were presented as adjusted relative risk (aRR) with 95% CI. Those missing data for maternal race and ethnicity, maternal education, parity, prepregnancy BMI, prenatal care, and cigarette use during pregnancy were categorized and analyzed as an “unknown” group.
We conducted a sensitivity analysis to ascertain whether the associations of adverse maternal composite morbidity persisted after excluding maternal transfusion. Because the birth certificate data did not collect information regarding the number of units given in a blood transfusion, the reasons for doing the sensitivity analysis were as follows: 1) transfusion of 1–2 units may be falsely recorded,15 2) the decision to transfuse less than 4 units may be subjective and is not necessarily a “severe” morbidity,16,17 and 3) we desired to be congruent with other publications on the topic.18 All statistical analyses were conducted using SAS 9.4 and STATA 15.
From 2011 to 2013, there were 11,871,286 live births in the United States recorded into the national database, and 10,465,727 live birth records (88.2%) used the 2003 revised birth certificate. Of this group, 3,917,831 live birth records (33.0%) met inclusion criteria. Nonmacrosomic LGA neonates accounted for 1.3% (n=50,630) of these live births (Fig. 1).
Women who gave birth to a nonmacrosomic LGA neonate were more likely to be older (35 years of age or older), be non-Hispanic white, have more than high school education, be married, be overweight or obese, and have pregestational or gestational diabetes. They were also more likely to have hypertensive disorders and cesarean delivery. However, women with an AGA singleton live birth were more likely to be primiparous and smoke during pregnancy (Table 1).
The rate of composite maternal morbidity was 53% higher among women who delivered a nonmacrosomic but LGA neonate (6.27/1,000 live births) than among those who delivered an AGA neonate (4.09/1,000 live births, P<.001). The rate of maternal transfusion, ruptured uterus, unplanned hysterectomy, ICU admission, and unplanned operating room procedure after delivery were all significantly higher among women who delivered a nonmacrosomic LGA neonate (Table 2).
The multivariable adjusted analysis showed that the risk of composite maternal morbidity was higher among women who delivered nonmacrosomic LGA neonates (aRR 1.40, 95% CI 1.25–1.56). Increased risk of composite maternal morbidity for women who delivered nonmacrosomic LGA neonates was found in those without diabetes (aRR 1.39, 95% CI 1.22–1.57) and those with gestational diabetes (aRR 1.50, 95% CI 1.12–2.00). However, there was no significant difference in the risk of composite maternal morbidity for women who delivered nonmacrosomic LGA vs AGA neonates among those with pregestational diabetes (aRR 1.25, 95% CI 0.72–2.15; Table 3).
The rate of composite neonatal morbidity was 83% higher among nonmacrosomic LGA neonates than among AGA neonates (11.09 vs 6.07/1,000 live births, P<.001, aRR 1.47 [95% CI 1.35–1.60]). The rates of individual neonatal morbidities (Apgar score less than 5 at 5 minutes, assisted ventilation required for 6 hours, seizure or serious neurologic dysfunction, significant birth injury, and neonatal mortality) were also significantly higher for nonmacrosomic LGA neonates (Table 4).
Our data suggests a higher risk of composite neonatal morbidity with nonmacrosomic LGA neonates as compared with AGA neonates, with an overall aRR of 1.47 (95% CI 1.35–1.60). Analyses stratified by maternal diabetes and sensitivity analysis by whether transfusion was included in the maternal composite were consistent with the overall analysis (Appendices 1 and 2, available online at http://links.lww.com/AOG/B378).
The results of this population-based study suggest that although nonmacrosomic LGA occurs in about 1% of singleton neonates at 37–39 weeks of gestation exposed to labor, they and their mothers are at increased risk for adverse outcomes. Women who delivered nonmacrosomic LGA neonates had more than a 50% increased rate of composite maternal morbidity compared with those who delivered AGA neonates, and nonmacrosomic LGA neonates had more than an 80% rate of composite neonatal morbidity. When stratified by maternal diabetic status, composite maternal morbidity was significantly higher for those without diabetes and those with gestational diabetes but not pregestational diabetes.
In spite of the association of adverse outcomes of nonmacrosomic LGA at 37–39 weeks of gestation, the management of such pregnancies is problematic for five reasons. First, most accelerated growth is unrecognized as being LGA before birth8,19 and may not benefit from interventions. Second, estimation of birth weight to determine whether the fetus is LGA may increase the risk of cesarean delivery.20,21 Third, though a randomized clinical trial involving upwards of 6,000 low-risk nulliparous women reported maternal and neonatal benefits of elective induction at 39 weeks of gestation, the improved outcomes may not be applicable to pregnancies with LGA. Additionally, the trial does not address how to avert complications with nonmacrosomic LGA at 37 and 38 weeks of gestation and among parous women.22 Fourth, although meta-analyses of randomized trials focused on pregnancies with suspected LGA (ultrasonographic estimated fetal weight 90% or greater for gestational age at term) or macrosomia recommend delivery at 38 weeks of gestation to avert complications such as shoulder dystocia and fracture,23,24 the American College of Obstetricians and Gynecologists Practice Bulletin on fetal macrosomia does not recommend induction for suspicion of accelerated growth.1 Fifth, a planned randomized clinical trial to assess whether induction at 39 weeks of gestation for suspected LGA, as compared with expectant management, decreases long-term neonatal morbidity is daunting. Among neonates with LGA, the rate of adverse outcomes associated with sequelae is about 1%, and to decrease this rate by one-third, more than 29,000 women would need to be randomized (alpha=0.05; beta=0.2; power=0.8). Notwithstanding these problems with LGA at term, we recommend compliance with American College of Obstetricians and Gynecologists recommendations of expectant management, at least until 39 weeks of gestation.
There are several strengths to our study. Our report differs from other prior studies on the topic in that we excluded women who had macrosomic neonates, stratified our analysis by diabetic status, and provided data both on maternal and neonatal morbidity outcomes.2–6 The large sample size permitted us to compare uncommon outcomes, often linked with long-term sequelae, and to adjust for known confounders. We also performed sensitivity analysis.
The limitations of our analysis, however, need to be acknowledged. Because our study used data from the U.S. Vital Statistics datasets, a variety of clinical information (eg, postpartum hemorrhage, meconium aspiration syndrome) was unavailable for analysis. Also, we used the 2003 revised birth certificate. The revised birth certificate represented 83%, 86%, and 90% of U.S. live births in 2011, 2012, and 2013, respectively. Therefore, our findings may not be generalizable to the whole U.S. population. Nomograms for categorizing appropriate compared with accelerated growth are also from similar datasets and we accepted the assigned gestational age when categorizing neonates as being small for gestational age, AGA, or LGA. As with our analysis, prior reports on adverse outcomes with macrosomic neonates have also used Vital Statistics data.2,3 We excluded women who did not labor and those at 40 or more weeks of gestation. Thus, our findings may not be applicable to women who have scheduled cesarean delivery or those at least 40 weeks of gestation. The categorization of neonates being nonmacrosomic LGA compared with AGA was based on actual birth weight, and the dataset does not permit determination of whether the clinicians were aware of the accelerated growth, which may influence the management of labor.19,20 Though we adjusted for several confounders, we could not adjust for unmeasured confounders like obstetric history, suspected LGA, medication use before and during pregnancy, and the type of hospital where the delivery occurred. Lastly, the adverse outcomes are uncommon and interventions to mitigate them may be difficult.
In conclusion, after labor and delivery at 37–39 weeks of gestation, delivery of a neonate with a birth weight 90% or greater for gestational age but less than 4,000 g is associated with increased composite maternal and neonatal morbidity.
1. Fetal macrosomia. Practice Bulletin No. 173. American College of Obstetricians and Gynecologists. Obstet Gynecol 2016;128:e195–209.
2. Boulet SL, Alexander GR, Salihu HM, Pass M. Macrosomic births in the United States: determinants, outcomes, and proposed grades of risk. Am J Obstet Gynecol 2003;188:1372–8.
3. Zhang X, Decker A, Platt RW, Kramer MS. How big is too big? The perinatal consequences of fetal macrosomia. Am J Obstet Gynecol 2008;198:517.e1–6.
4. Ecker JL, Greenberg JA, Norwitz ER, Nadel AS, Repke JT. Birth weight as a predictor of brachial plexus injury. Obstet Gynecol 1997;89:643–7.
5. Chauhan SP, Grobman WA, Gherman RA, Chauhan VB, Chang G, Magann EF, et al. Suspicion and treatment of the macrosomic fetus: a review. Am J Obstet Gynecol 2005;193:332–46.
6. Mendez-Figueroa H, Truong VTT, Pedroza C, Chauhan SP. Large for gestational age infants and adverse outcomes among uncomplicated pregnancies at term. Am J Perinatol 2017;34:655–62.
7. Chauhan SP, Parker D, Shields D, Sanderson M, Cole JH, Scardo JA. Sonographic estimate of birth weight among high-risk patients: feasibility and factors influencing accuracy. Am J Obstet Gynecol 2006;195:601–6.
8. Heywood RE, Magann EF, Rich DL, Chauhan SP. The detection of macrosomia at a teaching hospital. Am J Perinatol 2009;26:165–8.
9. Boulvain M, Senat MV, Perrotin F, Winer N, Beucher G, Subtil D, et al. Groupe de Recherche en Obstétrique et Gynécologie (GROG). Induction of labour versus expectant management for large-for-date fetuses: a randomised controlled trial. Lancet 2015;385:2600–5.
10. Chauhan SP, Rice MM, Grobman WA, Bailit J, Reddy UM, Wapner RJ, et al. Neonatal morbidity of small- and large-for-gestational-age neonates born at term in uncomplicated pregnancies. Obstet Gynecol 2017;130:511–19.
11. Alexander GR, Kogan MD, Himes JH. 1994-1996 US singleton birth weight percentiles for gestational age by race, Hispanic origin, and gender. Matern Child Health J 1999;3:225–31.
12. Osterman MJ, Martin JA, Mathews TJ, Hamilton BE. Expanded data from the new birth certificate, 2008. Natl Vital Stat Rep 2011;59:1–28.
13. Division of Vital Statistics. Guide to completing the facility worksheets for the certificate of live birth and report of fetal death (2003 revision). Available at https://www.cdc.gov/nchs/data/dvs/GuidetoCompleteFacilityWks.pdf
. Retrieved November 2, 2018.
14. Martin JA, Osterman MJK, Kirmeyer SE, Gregory ECW. Measuring gestational age in vital statistics data: transitioning to the obstetric estimate. Natl vital Stat Rep 2015;64:1–20.
15. Main EK, Abreo A, McNulty J, Gilbert W, McNally C, Poeltler D, et al. Measuring severe maternal morbidity: validation of potential measures. Am J Obstet Gynecol 2016;214:643.e1–10.
16. Callaghan WM, Grobman WA, Kilpatrick SJ, Main EK, DʼAlton M. Facility-based identification of women with severe maternal morbidity: it is time to start. Obstet Gynecol 2014;123:978–81.
17. Kilpatrick SJ, Berg C, Bernstein P, Bingham D, Delgado A, Callaghan WM, et al. Standardized severe maternal morbidity review: rationale and process. Obstet Gynecol 2014;124:361–6.
18. Chen HY, Grobman WA, Blackwell SC, Chauhan SP. Neonatal and maternal morbidity among low-risk, Nulliparous women at 39–41 weeks of gestation. Obstet Gynecol 2019;133:1–10.
19. Sparks TN, Cheng YW, McLaughlin B, Esakoff TF, Caughey AB. Fundal height: a useful screening tool for fetal growth? J Matern Fetal Neonatal Med 2011;24:708–12.
20. Froehlich RJ, Sandoval G, Bailit JL, Grobman WA, Reddy UM, Wapner RJ, et al. Association of recorded estimated fetal weight and cesarean delivery in attempted vaginal delivery at term. Obstet Gynecol 2016;128:487–94.
21. Dude AM, Grobman WA, Yee LM. Association between sonographic estimated fetal weight and the risk of cesarean delivery among Nulliparous women with diabetes in pregnancy. Am J Perinatol 2018;35:1297–302.
22. Grobman WA, Rice MM, Reddy UM, Tita ATN, Silver RM, Mallett G, et al. Labor induction versus expectant management in low-risk nulliparous women. N Engl J Med 2018;379:513–23.
23. Magro-Malosso ER, Saccone G, Chen M, Navathe R, Di Tommaso M, Berghella V. Induction of labour for suspected macrosomia at term in non-diabetic women: a systematic review and meta-analysis of randomized controlled trials. BJOG 2017;124:414–21.
24. Boulvain M, Irion O, Dowswell T, Thornton JG. Induction of labour at or near term for suspected fetal macrosomia. The Cochrane Database of Systematic Review 2016, Issue 5. Art No.: CD000938. DOI: 10.1002/14651858.CD000938.pub2.