PCOGS Papers

The association between birthweight 4000 g or greater and perinatal outcomes in patients with and without gestational diabetes mellitus

www. AJOG.org

Tania F. Esakoff, MD; Yvonne W. Cheng, MD, MPH; Teresa N. Sparks; Aaron B. Caughey, MD, PhD
OBJECTIVE: The objective of the study was to examine the association

between birthweight of 4000 g or greater and perinatal outcomes in women with and without gestational diabetes mellitus (GDM).
STUDY DESIGN: This was a retrospective cohort study of 36,241 singleton pregnancies stratied by the diagnosis of GDM, with presence or absence of birthweight of 4000 g or greater. Outcomes examined included neonatal hyperbilirubinemia, hypoglycemia, respiratory distress syndrome (RDS), shoulder dystocia, and Erbs palsy. 2 tests and multivariable logistic regression analyses were used to control for confounders. RESULTS: In women with GDM, neonates with a birthweight of 4000 g

RDS (4.0% vs 1.5%; P .03), shoulder dystocia (10.5% vs 1.6%; P .001), and Erbs palsy (2.6% vs 0.2%; P .001). Even without GDM, these outcomes occurred more frequently in infants with birthweight of 4000 g or greater. GDM increases the odds of adverse outcomes associated with birthweight of 4000 g or greater, particularly shoulder dystocia (adjusted odds ratios [aORs], 16.4 [GDM] vs 9.6 [non-GDM] and Erbs palsy (aORs, 41.9 [GDM] vs 6.7 [non-GDM]).
CONCLUSION: Birthweight of 4000 g or greater is associated with a

higher incidence of adverse perinatal outcomes such that neonatal providers should be alerted. Key words: birthweight of 4000 g or greater, gestational diabetes, perinatal outcomes

or greater, compared with those with a birthweight of less than 4000 g, had higher frequencies of hypoglycemia (5.3% vs 2.6%; P .04),

Cite this article as: Esakoff TF, Cheng YW, Sparks TN, et al. The association between birthweight 4000 g or greater and perinatal outcomes in patients with and without gestational diabetes mellitus. Am J Obstet Gynecol 2009;200:672.e1-672.e4.

here is a preponderance of evidence in the literature that suggests macrosomia (birthweight of 4000 g) is associated with adverse perinatal outcomes. The risk of maternal complications, such as prolonged labor, cesarean delivery, postpartum hemorrhage, infection, third- and fourth-degree lacerations, and thromFrom the Department of Obstetrics, Gynecology, and Reproductive Sciences, University of California, San Francisco, School of Medicine, San Francisco, CA.
Presented at the 75th Annual Meeting of the Pacic Coast Obstetrical and Gynecological Society, Victoria, BC, Canada, Oct. 15-19, 2008. Received Sept. 15, 2008; revised Dec. 12, 2008; accepted Feb. 26, 2009. Reprints not available from the authors. A.B.C. is supported by the Robert Wood Johnson Foundation as a physician faculty scholar, Grant RWJF-61535. 0002-9378/free 2009 Mosby, Inc. All rights reserved. doi: 10.1016/j.ajog.2009.02.035

For Editors Commentary, see Table of Contents

boembolic events, appears to increase in the setting of macrosomia.1-6 In addition, macrosomia is also associated with adverse neonatal outcomes; these include stillbirth, neonatal mortality secondary to birth asphyxia, shoulder dystocia, birth injury, and meconium aspiration syndrome.1-3,6-8 Whereas these associations have been reported, what is less clear is the causal relationship between increased birthweight and these outcomes and whether these relationships are modied by the presence of other risk factors. Do all macrosomic fetuses experience the same risks or does genetic predisposition toward greater birthweight or the intrauterine environment alter both the short- and long-term consequences of macrosomia? Although the majority of macrosomic babies are born to nondiabetic mothers,9 gestational diabetes mellitus (GDM) remains a well-established risk factor. Multiple studies have shown that GDM by itself, even in the absence of macrosomia, predisposes a patient to an in-

creased risk of undesirable perinatal outcomes. These include intrauterine fetal demise, neonatal death, shoulder dystocia, and preeclampsia.10-12 Although previous studies have attempted to address the question of adverse outcomes associated with either macrosomia or GDM, there is a paucity of data focusing on the effects of both together. In particular, because women with GDM are more likely to have macrosomic neonates because of intrauterine effects of hyperglycemia whereas women without GDM may have macrosomic neonates because of a genetic predisposition, the neonatal outcomes in these 2 settings may differ. Additionally, although several different denitions of macrosomia at term gestation exist, the American College of Obstetricians and Gynecologists has generally dened macrosomia as a birthweight or estimated fetal weight greater than either 4000 or 4500 g, irrespective of gestational age.13 Unfortunately, estimated fetal weight often does not reliably predict macrosomia, and to date, the majority of studies have focused on outcomes of babies with birthweights of

672.e1

American Journal of Obstetrics & Gynecology JUNE 2009

www.AJOG.org
4500 g or greater with less information on outcomes when birthweight is 4000 g or greater. We set out to answer these questions by comparing the perinatal outcomes of patients with gestational diabetes and no macrosomia, gestational diabetes and macrosomia, no gestational diabetes and no macrosomia, and patients with macrosomia but not gestational diabetes using a birthweight cutoff of 4000 g or greater.

PCOGS Papers
TABLE 1

Prevalence of perinatal outcomes in patients without gestational diabetes stratied by presence vs absence of macrosomia
Outcome Hyperbilirubinemia Hypoglycemia RDS Absence of GDM Bwt < 4000 g, % (n) 9.1% (2109) 1.2% (269) 1.2% (283) Absence of GDM Bwt > 4000 g, % (n) 7.6% (243) 2.4% (77) 1.7% (55) 0.7% (21) P value .23 .001 .02 .001 .001

.............................................................................................................................................................................................................................................. a a .............................................................................................................................................................................................................................................. a a .............................................................................................................................................................................................................................................. a a

Shoulder dystocia

.............................................................................................................................................................................................................................................. a a

0.9% (205)

6.0% (190)

Brachial plexus injury

a

0.1% (29)

..............................................................................................................................................................................................................................................

Bwt, birthweight; GDM, gestational diabetes mellitus; RDS, respiratory distress syndrome.

M ATERIALS AND M ETHODS

This was a retrospective cohort study of women with singleton pregnancies with and without a diagnosis of gestational diabetes mellitus who delivered at the University of California, San Francisco (UCSF). Institutional review board approval was obtained from the Committee on Human Research. Inclusion criteria were women with singleton pregnancy at term who received prenatal care and delivered at UCSF between 1982 and 2006, including women with and without gestational diabetes mellitus. During the study period, all women were screened for GDM using the 1 hour, 50 g glucose load (GLT) between 24 and 28 weeks gestation if they did not have risk factors of GDM. In women considered high risk, early screening at rst prenatal visit or during the rst trimester was performed; if screened negative, they then underwent repeat screening between 24 and 28 weeks gestation. Gestational diabetes mellitus was diagnosed when a patient had an abnormal 1 hour GLT of greater than 140 mg/dL followed by an abnormal 3 hour, 100 g, glucose tolerance test during which 2 of the 4 values were found to be abnormal according to the National Diabetes Data Group thresholds (105, 190, 165, and 145 mg/dL for fasting, 1, 2, and 3 hours after glucose load, respectively) between 1982 and 2001. From 2002 to 2006, the diagnostic criteria of GDM were based on that of Carpenter and Coustan thresholds (95, 180, 155, and 140 mg/dL for fasting, 1, 2, and 3 hours after glucose load, respectively).

Results are statistically signicant. Esakoff. Birthweight 4000 g or greater and perinatal outcomes in GDM. Am J Obstet Gynecol 2009.

Exclusion criteria were multifetal gestations and pregestational diabetes mellitus (type 1 or type 2 diabetes mellitus), noncephalic presentation, cesarean delivery without a trial of labor, and pregnancies complicated by congenital abnormalities. The primary predictor examined was a birthweight of 4000 g or greater. This predictor was examined in women with and without the diagnosis of GDM. The outcomes of interest were hyperbilirubinemia, hypoglycemia, respiratory distress syndrome (RDS), shoulder dystocia, and brachial plexus injury. These were dened using standard denitions. Hyperbilirubinemia was dened as total serum bilirubin greater than 5 mg/dL. Hypoglycemia was dened as blood glucose less than 35 mg/dL or plasma glucose less than 40 mg/dL. Neonatal RDS was diagnosed by the presence of at least 2 of the following 3 criteria: (1) evidence of respiratory compromise (tachypnea, retractions, and/or nasal aring) shortly after delivery and a persistent oxygen requirement for more than 24 hours, (2) administration of exogenous pulmonary surfactant, and/or (3) radiographic evidence (atelectasis, air bronchograms, and a diffuse reticulogranular inltrate) of neonatal pulmonary hyaline membrane disease as diagnosed by an attending pediatric radiologist or neonatologist. Shoulder dystocia was dened as a prolonged head-to-body delivery time and/or the use of obstetric maneuvers. Brachial plexus injury was dened as short-term

paralysis of the arm thought to be secondary to birth trauma as determined by the pediatricians and neonatologists. The 2 test was used to compare the dichotomous outcomes and P .05 was designated to indicate statistical signicance. Multivariable logistic regression analyses were performed to control for potential confounders. The effect estimates were expressed as adjusted odds ratios. Covariates included in the multivariable regression model were gestational age, maternal age, insurance status, marital status, race/ethnicity, body mass index, fetal position, and year of delivery. Statistical analysis was performed using STATA version 9.0 (Stata Corp, College Station, TX).

R ESULTS
In women who did not receive a diagnosis of gestational diabetes, a birthweight of 4000 g or greater was associated with a higher incidence of adverse perinatal outcomes. In nondiabetic patients, the incidence of neonatal hypoglycemia was 2-fold higher in the presence of birthweight of 4000 g or greater compared with birthweight less than 4000 g (P .001), and the incidence of shoulder dystocia and brachial plexus injury was approximately 6- and 7-fold higher, respectively (P .001; Table 1). With the presence of both birthweight of 4000 g or greater and GDM, the incidence of neonatal hypoglycemia was 2 times higher than in the group with GDM but birthweight less than 4000 g (P .04). However, the presence of birth672.e2

JUNE 2009 American Journal of Obstetrics & Gynecology

PCOGS Papers
TABLE 2

www.AJOG.org
creases this risk even further. When both birthweight of 4000 g or greater and GDM are present, the effect estimates of these outcomes appear to be more than additive. It is still unknown why some outcomes, such as shoulder dystocia and Erbs palsy, carry the greatest risk increase compared with the other outcomes. It may be due to not only the increased birthweight but also its anthropometric distribution. How can the dramatic differences in neonatal outcomes be explained? In particular, it is of interest whether these increased risks are due, in part, to differences in fetal anthropometrics. Perhaps there exists a relationship between fat deposition and shoulder dystocia or Erbs palsy such that if some of these characteristics could be identied with prenatal ultrasound, including measurements of fetal visceral fat and abdominal circumference, such information might improve patient counseling prior to delivery. Undoubtedly there is evidence that fetal abdominal subcutaneous tissue thickness at term can be used to predict the likelihood of operative vaginal and cesarean delivery.14 Furthermore, there are data to support the use of ultrasound for estimating fetal weight composition15 as well as data that the sonographic fetal weight composition differs in patients with and without diabetes.16 If some predictions can be made, issues of the optimal mode and timing of delivery could potentially be addressed. However, such prenatal identication of fetal weight and anthropometrics is still experimental and caution for making current recommendations based on such information should be advised. Whereas anthropometrics may be 1 key to differences in perinatal outcomes associated with macrosomia, these differences may fundamentally be related to causality. There is some evidence in the literature to suggest that macrosomia has etiologic contributions from both genetic predetermination and the intrauterine environment.17 One can hypothesize that if a fetus is genetically meant to be large, then it is less likely to exhibit abnormal metabolic

Prevalence of perinatal outcomes in patients with gestational diabetes stratied by presence vs absence of macrosomia
Outcome Hyperbilirubinemia Hypoglycemia RDS Presence of GDM Bwt < 4000 g, % (n) 10.4% (93) 2.6% (23) 1.5% (13) 1.6% (14) 0.2% (2) Presence of GDM Bwt > 4000 g, % (n) 13.2% (20) 5.3% (8) 4.0% (6) 2.6% (4) P value .90 .04 .03

.............................................................................................................................................................................................................................................. a a .............................................................................................................................................................................................................................................. a a .............................................................................................................................................................................................................................................. a a

Shoulder dystocia

.............................................................................................................................................................................................................................................. a a

10.5% (16)

.001 .001

Brachial plexus injury

a

..............................................................................................................................................................................................................................................

Bwt, birthweight; GDM, gestational diabetes mellitus; RDS, respiratory distress syndrome. Results are statistically signicant. Esakoff. Birthweight 4000 g or greater and perinatal outcomes in GDM. Am J Obstet Gynecol 2009.

weight of 4000 g or greater in addition to GDM conferred a nearly 10-fold increase in the incidence of both shoulder dystocia and brachial plexus injury compared with the subgroup of women with GDM but birthweight less than 4000 g (P .001; Table 2). To further examine the effect of having a birthweight of 4000 g or greater in women with and without GDM, we controlled for potential confounding using multivariable logistic regression and examined the association of having a birthweight of 4000 g or greater with neonatal outcomes stratied by the diagnosis of GDM (Table 3). Although there were essentially no differences in the odds of hyperbilirubinemia and hypoglycemia, there are large differences in the odds of RDS (adjusted odds ratio [aOR], 3.10; 95% conTABLE 3

dence interval [CI], 1.11-8.65 in GDM vs aOR, 1.54; 95% CI, 1.02-2.33 in nonGDM), shoulder dystocia (aOR, 16.45; 95% CI, 6.71-40.33 in GDM vs aOR, 9.62; 95% CI, 7.38-12.54 in non-GDM), and brachial plexus injury (aOR, 41.89; 95% CI, 4.05-433.64 in GDM vs aOR, 6.65; 95% CI, 2.90-15.27 in non-GDM). Of note, when we stratied the analyses by year (1982-2001 and 2002-2006), there were minimal differences in our ndings (data not shown).

C OMMENT
In this large cohort, we report that not only does birthweight of 4000 g or greater increase the prevalence of adverse perinatal outcomes such as hypoglycemia, RDS, shoulder dystocia, and Erbs palsy but also that GDM status in-

The odds of adverse perinatal outcomes in infants weighing > 4000 g as compared with infants weighing < 4000 g in women with and without gestational diabetes
Outcome Hyperbilirubinemia Hypoglycemia RDS Absence of GDM/with Bwt > 4000 g, aOR (95% CI)a 0.90 (0.74-1.09) 2.04 (1.42-2.92) Presence of GDM/presence of Bwt > 4000 g, aOR (95% CI)b 1.60 (0.93-2.74) 2.60 (1.05-6.45) 3.10 (1.11-8.65)

.............................................................................................................................................................................................................................................. c c .............................................................................................................................................................................................................................................. c c .............................................................................................................................................................................................................................................. c c

1.54 (1.02-2.33)

Shoulder dystocia

.............................................................................................................................................................................................................................................. c c

9.62 (7.38-12.54) 6.65 (2.90-15.27)

16.45 (6.71-40.33)

Brachial plexus injury

41.89 (4.05-433.64)

..............................................................................................................................................................................................................................................

aOR, adjusted odds ratio; Bwt, birthweight; CI, condence interval; GDM, gestational diabetes mellitus; RDS, respiratory distress syndrome.
a

The reference group is neonates born to women without GDM who weighed 4000 g; b The reference group is neonates born to women with GDM who weighed 4000 g; c Results are statistically signicant. Esakoff. Birthweight 4000 g or greater and perinatal outcomes in GDM. Am J Obstet Gynecol 2009.

672.e3

American Journal of Obstetrics & Gynecology JUNE 2009

www.AJOG.org
derangements such as hypoglycemia or hyperbilirubinemia that are frequently associated with intrauterine hyperglycemia. It is also possible that if the fetus is destined to be large because of genetic programming, then perhaps the mother is more likely to be able to deliver it without birth injury. This is supported by the observation that when macrosomia is present, women of tall stature are less likely to have birth injury associated with the macrosomic fetus.18 By contrast, studies in rhesus monkeys have shown that when a hyperglycemic intrauterine environment is present, the offspring are more likely to have metabolic derangements19 and excessive weight gain.20 In humans, intrauterine hyperglycemia also appears to cause both short-term neonatal effects such as hypoglycemia21 and to affect long-term metabolic function because the offspring of mothers with intrauterine hyperglycemia are at much higher risk of developing the metabolic syndrome as children.22 Thus, it appears that the potential etiology of macrosomia does indeed have an impact on the risks associated with greater birthweight. The strengths of our study are that it consists of a large and diverse patient population that is potentially generalizable. The large sample size also provides sufcient statistical power to detect differences between the comparison groups; however, there are some limitations as well. We studied only the perinatal outcomes that we were powered to detect; however, there are other rare outcomes that may be associated with the presence of gestational diabetes and macrosomia that we did not have sufcient statistical power to examine, including intrauterine fetal demise and neonatal acidemia. Even despite our adequate power, with several of the rare outcomes, we had wide condence intervals. Thus, although we are certain there is an association between birthweight of 4000 g or greater and these outcomes, it is difcult to pinpoint the exact quantitative effect. As with all observational studies, our ndings are prone to confounding bias. We attempted to control for such confounding using statistical techniques; however, there may be residual confounding or interactions that we could not observe or adjust for using statistical models. Despite these limitations, our ndings support that birthweight of 4000 g or greater is associated with neonatal morbidity and the risks further increase in the setting of gestational diabetes. Such neonates should be screened by pediatricians for hypoglycemia and unrecognized Erbs palsy. Consideration should also be given to screening women in this setting for type 2 diabetes in the primary care setting. Further research needs to improve our ability to prenatally characterize fetal weight and anthropometrics, potentially through the use of newer modalities such as 3-dimensional ultrasound and fetal magnetic resonance imaging. However, we must also be cognizant of the patient anxiety, medical intervention, and economic burdens that false-positive results might induce. f
REFERENCES
1. Stotland NE, Caughey AB, Breed EM, Escobar GJ. Risk factors and obstetric complications associated with macrosomia. Int J Gynaecol Obstet 2004;87:220-6. 2. Boyd ME, Usher RH, McClean FH. Fetal macrosomia: prediction, risks, and proposed management. Obstet Gynecol 1983;61:715-22. 3. Meshari AA, DeSilva S, Rahman I. Fetal macrosomia: prediction, risks and fetal outcome. Int J Gynaecol Obstet 1990;22:215-22. 4. Handa VL, Danielsen BH, Gilbert WM. Obstetric anal sphincter lacerations. Obstet Gynecol 2001;98:225-30. 5. Gregory KD, Henry OA, Ramicone E, Chan LS, Platt LD. Maternal and infant complications in high and normal weight infants by method of delivery. Obstet Gynecol 1998;92:507-13. 6. Oral E, Cagdas A, Gezer A, Kaleli S, Aydinli K, Ocer F. Perinatal and maternal outcomes of fetal macrosomia. Eur J Gynecol Reprod Biol 2001;99:167-71. 7. 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. 8. Raio L, Ghezzi F, Di Naro E, et al. Perinatal outcome of fetuses with a birth weight greater than 4500 g: an analysis of 3356 cases. Eur J Gynecol Reprod Biol 2003;109:160-5. 9. Cunningham FG, Leveno KL, Bloom SL, Hauth JC, Gilstrap LC, Wenstrom KD. Fetal growth disorders. In: Cunningham F, ed. Wil-

PCOGS Papers
liams obstetrics. 21st ed. New York: Mc-Graw Hill; 2001:758. 10. Casson IF, Clarke CA, Howard CV, et al. Outcomes of pregnancy in insulin dependent diabetic women: results of a ve year population cohort study. Br Med J 1997;315:275-8. 11. Hawthorne G, Robson S, Ryall EA, Sen D, Roberts SH, Ward Platt MP. Prospective population based survey of outcome of pregnancy in diabetic women: results of the Northern Diabetic Pregnancy Audit, 1994. Br Med J 1997; 315:279-81. 12. Kwik M, Seeho SK, Smith C, McElduff A, Morris JM. Outcomes of pregnancies affected by impaired glucose tolerance. Diabetes Res Clin Pract 2007;77:263-8. 13. American College of Obstetricians and Gynecologists. Clinical management guidelines for obstetrician-gynecologists. Fetal macrosomia. ACOG practice bulletin no. 22. American College of Obstetricians and Gynecologists; Washington, DC: 2000. 14. Assimakopoulos E, Zafrakas M, Garmiris P, et al. Fetal abdominal subcutaneous tissue thickness measured by ultrasound at term is associated with birth weight and mode of delivery. Clin Exp Obstet Gynecol 2007;34: 171-4. 15. Crane SS, Avallone DA, Thomas AJ, Catalano PM. Sonographic estimation of fetal body composition with gestational diabetes mellitus at term. Obstet Gynecol 1996;88:849-54. 16. Parretti E, Carignani L, Cioni R, et al. Sonographic evaluation of fetal growth and body composition in women with different degrees of normal glucose metabolism. Diabetes Care 2003;26:2741-8. 17. Sacks DA, Liu AI, Wolde-Tsadik G, Amini SB, Huston-Presley L, Catalano PM. What proportion of birth weight is attributable to maternal glucose among infants of diabetic women? Am J Obstet Gynecol 2006;194:501-7. 18. Gudmundsson S, Henningsson AC, Lindqvist P. Correlation of birth injury with maternal height and birthweight. BJOG 2005;112:764-7. 19. Susa JB, Gruppuso PA, Widness JA, et al. Chronic hyperinsulinemia in the rhesus monkey: effects of physiologic hyperinsulinemia on fetal substrates, hormones and hepatic enzymes. Am J Obstet Gynecol 1984;15:150:415-20. 20. Susa JB, Widness JA, Hintz R, Liu F, Sehgal P, Schwartz R. Somatomedins and insulin in diabetic pregnancies: effects on fetal macrosomia in the human and rhesus monkey. J Clin Endocrinol Metab 1984;58:1099-105. 21. HAPO Study Cooperative Research Group, Metzger BE, Lowe LP, et al. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med 2008;358:1991-2002. 22. Boney CM, Verma A, Tucker R, Vohr BR. Metabolic syndrome in childhood: association with birth weight, maternal obesity, and gestational diabetes mellitus. Pediatrics 2005;115: e290-6.

JUNE 2009 American Journal of Obstetrics & Gynecology

672.e4