Macrosomic births in the United States: Determinants,

outcomes, and proposed grades of risk

Sheree L. Boulet, MPH, Greg R. Alexander, RS, MPH, ScD, Hamisu M. Salihu, MD, PhD, and
MaryAnn Pass, MD, MPH
Birmingham, Ala

OBJECTIVE: We describe maternal risk factors for macrosomia and assess birth weight categories to deter-
mine predictive thresholds of adverse outcomes.
STUDY DESIGN: We analyzed linked live birth and infant death cohort files from 1995 to 1997 for the United
States with the use of selected term (37-44 weeks of gestation) single live births to mothers who were US
residents. We compared macrosomic infants (4000-4499 g, 4500-4999 g, and >5000 g infants) with a normo-
somic control group of infants who weighed 3000 to 3999 g.
RESULTS: Maternal risk factors for macrosomia included nonsmoking, advanced age, married, diabetes
mellitus, hypertension, and previous macrosomic infant or pregnancy loss. The risks of labor complications,
birth injuries, and newborn morbidity rose with each gradation of macrosomic birth weight. Infant mortality
rates increased significantly among infants weighing >5000 g.
CONCLUSION: Although a definition of macrosomia as >4000 g (grade 1) may be useful for the identifica-
tion of increased risks of labor and newborn complications, >4500 g (grade 2) may be more predictive of
neonatal morbidity, and >5000 g (grade 3) may be a better indicator of infant mortality risk. (Am J Obstet Gy-
necol 2003;188:1372-8.)

Key words: Macrosomia, birth weight, infant death, cesarean delivery, diabetes mellitus, prenatal
care, birth injury

Fetal macrosomia, varyingly characterized as birth
weight of >4000, >4500, or >5000 g, is associated with nu-
merous perinatal and maternal complications. 1-7 Past re-
search indicates that the incidence of macrosomia in the
United States has been increasing over time, most likely
because of the rising rates of maternal diabetes mellitus
and/or obesity.8-13 Macrosomic infants are at elevated
risk of shoulder dystocia, brachial plexus injury, skeletal
injuries, meconium aspiration, perinatal asphyxia, hypo-
glycemia, and fetal death.1-5,7,9 Maternal complications
are often related to cephalopelvic disproportion and in-
clude prolonged labor, labor augmentation with oxy-
tocin, cesarean delivery, postpartum hemorrhage,
infection, third- and fourth-degree lacerations, throm-
boembolic events, and anesthetic accidents.1-6,9,14
From the Department of Maternal and Child Health, School of Public
Health, University of Alabama at Birmingham.
Supported in part by the Department of Health and Human Services, the
Health Resources and Services Administration, and the Maternal and
Child Health Bureau (grant No. 5T76MC00008).
Received for publication August 14, 2002; revised November 6, 2002;
accepted January 23, 2003.
Reprint requests: Greg R. Alexander, RS, MPH, ScD, Department of Ma-
ternal and Child Health, School of Public Health, University of Al-
abama at Birmingham, 320-A Ryals Bldg, 1665 University Blvd,
Birmingham, AL 35294-0022. E-mail: alexandg@uab.edu
© 2003, Mosby, Inc. All rights reserved.
0002-9378/2003 $30.00 + 0
doi:10.1067/mob.2003.302
Risk characteristics that increase the probability of the
delivery of a macrosomic infant include maternal obesity,
multiparity, previous macrosomic infant, male fetus, ma-
ternal birth weight, paternal birth weight, maternal dia-
betes mellitus, postdatism, prepregnancy weight/height,
body mass index, advanced maternal age, ethnicity, ex-
cessive maternal weight gain, gestational hypertension,
preeclampsia, prolonged labor, increased interpregnancy
interval, and marital status.1-3,9-11,15-20 There is substantial
variation in the literature regarding the strength of the
association between these risk characteristics and macro-
somia. In fact, few pregnancies with these risk factors will
actually result in a large baby, even when several risk fac-
tors are present in a single patient.15
The establishment of a widely accepted, standard defi-
nition for macrosomia has been problematic. Fetal macro-
somia is defined by the American College of Obstetricians
and Gynecologists as infants with an absolute birth weight
of >4500 g, irrespective of gestational age or other demo-
graphic variables.9 Nevertheless, considerable variation
in the definition (eg, 4000, 4100, 4500, and 4536 g) per-
sists in the literature, with some researchers using differ-
ent cutoffs for diabetic and nondiabetic patients.11 In
addition to hindering comparisons among studies, the
use of prophylactic cesarean delivery for suspected
macrosomia has been questioned because of the lack of a
universally accepted definition for macrosomia, along

1372
 Volume 188, Number 5 Boulet et al 1373
Am J Obstet Gynecol

Table I. Maternal characteristics by birth weight category for single live births to mothers who were US residents
(1995-1997)
Category (%)

Control, Grade 1, Grade 2, Grade 3,

Characteristic 3000-3999 g (No.) 4000-4499 g (No.) 4500-4999 g (No.) >5000 g (No.)

Race of mother
White 62.6 (44,614,159) 71.9 (691,207) 73.4 (116,141) 69.7 (11,896)
Black 12.9 (917,258) 7.6 (73,389) 7.0 (11,019) 8.5 (1,447)
Hispanic 18.1 (1,288,253) 15.4 (147,940) 14.7 (23,211) 16.1 (2,748)
American Indian 1.0 (67,435) 1.1 (10,708) 1.3 (2,095) 1.7 (283)
Samoan 0.05 (2,646) 0.07 (627) 0.08 (133) 0.2 (29)
Other 5.5 (390,478) 3.9 (37,596) 1.3 (5,648) 3.9 (662)
Married 70.1 (4,994,417) 78.1 (751,176) 80.1 (126,793) 79.0 (13,486)
Maternal age (y)
10-17 4.6 (330,293) 2.5 (23,975) 1.7 (2,660) 1.2 (199)
18-34 83.6 (5,958,498) 82.8 (796,571) 81.2 (128,511) 78.4 (13,375)
35-49 11.8 (838,697) 14.7 (140,914) 7.1 (27,074) 20.5 (3,490)
Educational attainment
High 46.0 (3,235,791) 52.4 (498,254) 53.2 (83,201) 50.8 (8,547)
Average 35.2 (2,474,116) 32.9 (312,326) 32.8 (51,253) 33.6 (5,660)
Low 18.8 (1,324,654) 14.7 (139,610) 14.0 (21,947) 15.6 (2,629)
Parity
Primiparous 41.4 (2,936,922) 34.0 (324,830) 29.9 (47,052) 26.5 (4,499)
High parity for age 2.5 (179,584) 2.8 (26,477) 3.7 (5,851) 5.6 (947)
Previous macrosomic birth 0.9 (60,528) 3.7 (35,575) 7.3 (11,524) 11.6 (1,972)
Previous pregnancy loss 2.5 (177,347) 2.7 (25,639) 2.9 (4,596) 3.2 (552)
Maternal complications
Diabetes mellitus 2.3 (162,300) 3.7 (35,322) 6.2 (9,727) 11.8 (2016)
Hypertension 3.3 (232,551) 3.3 (31,721) 3.9 (6,207) 4.9 (838)
Smoking during pregnancy 9.6 (682,147) 5.5 (52,821) 4.5 (7,072) 4.0 (689)

All maternal characteristics differed significantly (P < .05, 2) by birth weight categories.

with the relative inaccuracy of fetal weight estimates at
higher birth weights and the controversy regarding the
efficacy of interventions in reducing adverse outcomes
that are associated with macrosomia.4,15,20-27
The purpose of this investigation, which used recent
national birth data, is to describe maternal risk factors for
macrosomia and to assess birth weight categories to de-
termine thresholds that are most predictive of adverse
outcomes, including labor and delivery complications,
neonatal morbidity, and infant death. Although it is
widely accepted that high birth weights are associated
with increased neonatal mortality rates, the birth weight
cutoffs that signal significant increases in general mor-
bidity and death have yet to be determined.
Material and methods
The data used for this study were derived from the Na-
tional Center for Health Statistics 1995 through 1997 US
linked live birth/infant death files.28 Term (37-44 weeks
of gestation), single live births to mothers who were resi-
dents of the United States were selected for analysis. We
further selected birth weights that were >3000 g. After
these selections, the sample included 7,127,529 normo-
somic births (3000-3999 g) and three groups of macro-
somic births (961,467 births [4000-4499 g], 158,247
births [4500-4999 g], and 17,065 births [>5000 g]). Be-
cause of the large sample size, group comparisons can be
anticipated to reveal statistically significant differences,
even for very modest differences that may not be of clini-
cal importance.
Maternal risk factors that were selected for this study
included marital status, maternal age, educational attain-
ment, parity, previous macrosomic birth, previous preg-
nancy loss, diabetes mellitus, hypertension, and smoking
while pregnant. Maternal educational attainment was
classified as high if the mother reported S12 years of ed-
ucation and as low if the mother reported <12 years of
education. Adolescents were classified as having high ed-
ucational attainment if they were S2 years above their
grade level for age. Conversely, low educational attain-
ment for adolescents was considered to be S2 years
below expected grade level for age. High parity for age
was considered to be 1 births for adolescents, S3 previ-
ous births for mothers 18 to 21 years old, S4 previous
births for mothers 22 to 24 years old, and S5 previous
births for mothers S25 years old. The Revised-Graduated
Index of Prenatal Care Utilization (R-GINDEX) was used
to measure adequacy of prenatal care use. 29 This index
assesses the adequacy of prenatal care use on the basis of
three variables (the trimester that prenatal care began,
the number of visits, and the gestational age of the infant
at birth) and is recommended for use in the assessment
1374 Boulet et al May 2003
Am J Obstet Gynecol

Table II. Prenatal care, obstetric procedure, obstetric complication, birth outcome, and mortality rate by birth weight
category
Category (%)

Control, Grade 1, Grade 2, Grade 3,

Variable 3000-3999 g (No.) 4000-4499 g (No.) 4500-4999 g (No.) >5000 g (No.)

Prenatal intensive care 5.9 (417,051) 6.4 (61,313) 7.2 (11,462) 9.1 (1,554)
Obstetric complication
Meconium staining 6.1 (432,503) 7.1 (67,740) 7.4 (11,627) 8.2 (1,404)
Excessive bleeding 0.5 (34,850) 0.7 (6,888) 0.9 (1,402) 1.1 (184)
Prolonged labor 0.9 (64,325) 1.2 (11,641) 1.3 (2,111) 1.5 (251)
Dysfunctional labor 2.8 (201,492) 4.3 (41,289) 5.2 (8,174) 5.6 (947)
Cephalopelvic disproportion 2.3 (166,533) 5.3 (50,723) 8.3 (13,118) 11.1 (1897)
Breech presentation 2.7 (189,084) 2.6 (24,912) 3.2 (5,007) 4.0 (674)
Obstetric procedure
Fetal monitoring 82.1 (5,853,780) 81.8 (786,298) 80.5 (127,424) 78.3 (13,364)
Ultrasound scan 98.6 (7,026,355) 98.8 (950,132) 98.9 (156,531) 99.0 (16,899)
Induced labor 17.5 (1,249,384) 22.4 (215,265) 23.4 (36,991) 20.3 (3,456)
Cesarean delivery 18.0 (1,279,370) 25.5 (245,166) 35.6 (56,406) 50.6 (8,629)
Birth outcome
Apgar score S6 (5 min) 0.6 (35,590) 0.7 (5,592) 1.0 (1,257) 2.2 (285)
Apgar score S3 (5 min) 0.1 (5,767) 0.1 (968) 0.2 (246) 0.5 (71)
Assisted ventilation <30 min 1.7 (121,307) 2.1 (19,888) 2.5 (4,027) 3.3 (556)
Assisted ventilation S30 min 0.3 (20,026) 0.3 (3,173) 0.5 (850) 1.3 (215)
Birth injuries 0.3 (18,347) 0.5 (5,004) 0.8 (1,337) 1.3 (218)
Hyaline membrane disease 0.2 (12,974) 0.2 (2,004) 0.4 (551) 0.8 (133)
Meconium aspiration 0.2 (16,736) 0.3 (2861) 0.4 (603) 0.6 (105)
Mortality rate
Hebdomadal (<7 d) 0.40 (2,863) 0.37 (360) 0.44 (70) 1.41 (24)
Neonatal (<28 d) 0.71 (5,044) 0.60 (579) 0.71 (113) 1.93 (33)
Postneonatal (S28 d) 1.44 (10,312) 1.00 (961) 1.08 (171) 2.05 (35)
Infant (<1 y) 2.16 (15,356) 1.60 (1,540) 1.80 (284) 3.98 (68)

All outcome variables differed significantly (P < .05, 2) by birth weight categories; Cochran-Armitage O2 trend test was significant for
all comparisons at P < .01 (trend tests for mortality rates were limited to 4000 g to >5000 g birth weight categories).

of birth weight– and gestational age–related birth out-
comes.29 Gestational age in completed weeks was com-
puted from the interval between the date of the last
menstrual period (LMP) and the date of birth. For
records that were missing the day of LMP, a gestational
age value was input when there was valid data for month
and year of LMP. Where the LMP was unknown or in-
compatible with birth weight, the clinical estimate of ges-
tation was used if it was consistent with birth weight.30
The 2 test was used to test for significant differences in
the proportions of maternal demographic risk character-
istics, prenatal care use, obstetric complications and pro-
cedures, birth outcomes, and infant death among the
birth weight categories. Positive monotonic relationships
between advancing birth weight categories and increas-
ing risks of adverse birth outcomes were tested with the
use of the Cochran-Armitage 2 trend test. Multiple lo-
gistic regression was then used to calculate odds ratios
and 95% CIs for the independent association of macro-
somic birth weight categories on an array of outcome
measures that included obstetric complications, obstetric
procedures, intensive use of prenatal care, and infant
death and morbidity, after controlling for maternal risk
factors (such as age, marital status, education, prenatal
care use, parity, previous macrosomic birth, previous
pregnancy loss, maternal diabetes mellitus, hypertension,
smoking, alcohol use, and gestational age). Macrosomic
birth weight categories (4000-4499 g, 4500-4999 g, >5000 g)
were entered into the logistic models as dummy variables,
and normal weight infants (3000-3999 g) were used as the
reference group.
Results
Table I shows maternal characteristics by birth weight
category. A number of maternal risk factors occurred
with higher frequency among the macrosomic categories
compared with the normosomic categories. Among race
of mother categories, white, American Indian, and
Samoan mothers were disproportionately overrepre-
sented in the macrosomic groups. Mothers of macro-
somic infants were significantly more likely than those of
normal birth weight infants to be married and older
(S35 years old), and less likely to be <18 years old and
primiparous. Significantly higher proportions of mothers
with a higher educational level, a high parity for age, a pre-
vious pregnancy loss, or a previous macrosomic birth were
found in the macrosomic groups. Mothers of macrosomic
infants were more likely than mothers of normosomic in-
Volume 188, Number 5 Boulet et al 1375
Am J Obstet Gynecol

fants to have diabetes mellitus and hypertension and less

likely to have smoked during the index pregnancy.
Table II shows prenatal care, obstetric procedures, ob-
stetric complications, birth outcomes, and mortality rates
by birth weight category and reveals that, although in-
creasing birth weight is associated with unfavorable birth
outcomes, the vast majority of macrosomic infants (even
those >5000 g) are not reported to experience an adverse
outcome or death. Mothers in all categories of macroso-
mia were more likely to report intensive use of prenatal
care than the nonmacrosomic mothers. The proportion
of all obstetric complications (meconium staining, exces-
sive bleeding, prolonged labor, dysfunctional labor,
cephalopelvic disproportion, breech presentation)
showed a progressive and significant increase among the
macrosomic birth weight categories. The frequency of
cephalopelvic disproportion demonstrated the most dra-
matic increase with the complication being reported for
2.3% of infants who weighed 3000 to 3999 g and 11.1% of
infants who weighed >5000 g. Among the obstetric proce-
dures, the proportion of cesarean deliveries increased
substantially across birth weight categories, with more
than one half of all infants weighing >5000 g delivered by
cesarean section, compared with only 18% of normo-
somic infants. Labor induction increased among the
4000- to 4499-g and 4500- to 4999-g groups and then de-
creased slightly among the >5000-g group yet still re-
mained higher than the normosomic infants. The
frequency of adverse birth outcomes increased slightly
over the birth weight categories, with the >4500-g infants
showing the most significant differences. For example,
the proportion of infants who were affected by hyaline
membrane disease was 0.4% in the 4500- to 4999-g cate-
gory and 0.8% in the S5000-g category compared with
0.2% for the 3000- to 3999-g and 4000- to 4499-g cate-
gories. Finally, the mortality rates either declined or
showed very little change for the 4000- to 4499-g and
4500- to 4999-g categories compared with the 3000- to
3999-g category. However, all mortality rates increased
substantially for the >5000-g category, as evidenced by the
change in the infant mortality rate from 2.16 (3000-3999
g) to 3.98 (>5000 g).
Table III shows the results of the logistic regression
analysis of obstetric procedures, obstetric complications,
birth outcomes, and deaths. The normosomic group was
used as the reference group. After maternal sociodemo-
graphic risk factors were controlled, macrosomic infants
were far more likely to have cephalopelvic disproportion
compared with normosomic infants (4000- to 4499-g in-
fants: odds ratio, 2.42 [95% CI, 2.40, 2.45]; 4500- to 4999-
g infants: odds ratio, 4.07 [95% CI, 3.99, 4.15]; and
>5000-g infants: odds ratio, 5.79 [95% CI: 5.50, 6.08]). All
categories of macrosomic infants were at greater risk of
being delivered by cesarean delivery compared with the
reference group. Birth injuries were also more likely
Figure. Increased risk of adverse outcomes by macrosomia
grade. Open bars, Grade 1 (4000-4499 g); gray bars, grade 2 (4500-
4999 g); black bars, grade 3 (5000+ g).
among all macrosomic groups, with the 4000- to 4499-g
group having an almost 2-fold risk compared with nor-
mosomic infants. The risk of other adverse birth out-
comes (such as hyaline membrane disease, meconium
aspiration, birth asphyxia, and poor Apgar scores) were
elevated moderately among the 4000- to 4499-g category
yet were increased substantially among the higher birth
weight categories. Increased risk of death, compared with
normal-birth-weight infants, was only evident in the
>5000-g category. Infants in this category were 2.69 times
as likely as nonmacrosomic infants to die during the
neonatal period.
The Figure shows the stepwise increase in risk of ad-
verse outcomes by macrosomic categories defined as
grade 1 (4000-4499 g), grade 2 (4500-4999 g), and grade
3 (>5000 g). Over this 3-year study period, macrosomic
births (if defined by the above birth weight categories,
without regard to their gestational age) represented ap-
proximately 10.5% of the reported single live births to
mothers who were US residents (approximately 8.7%
grade 1, 1.5% grade 2, and 0.2% grade 3).
Comment
Many of the risk factors for fetal macrosomia that were
identified in this investigation (eg, white race of mother,
advanced maternal age, previous macrosomic birth, high
educational attainment, married, maternal diabetes mel-
litus, hypertension, multiparity, and nonsmoker) have
been recognized in the literature.1-3,9-11 In this study, pre-
vious macrosomic birth and diabetes mellitus were the
strongest risk factors for macrosomia, with the magnitude
1376 Boulet et al May 2003
Am J Obstet Gynecol

Table III. Logistic regression analyses of obstetric procedure, obstetric complication, birth outcome, and death by birth
weight category
Category

Grade I, 4000-4499 g Grade II, 4500-4999 g Grade III, >5000 g

Variable Odds ratio 95% CI Odds ratio 95% CI Odds ratio 95% CI

Prenatal intensive care 1.04 1.03-1.05* 1.15 1.12-1.17* 1.38 1.30-1.46*

Obstetric complication
Meconium staining 1.18 1.17-1.19* 1.22 1.21-1.25* 1.35 1.28-1.42*
Excessive bleeding 1.41 1.37-1.45* 1.65 1.56-1.74* 1.86 1.61-2.16*
Prolonged labor 1.38 1.35-1.41* 1.55 1.48-1.62* 1.76 1.55-1.99*
Dysfunctional labor 1.56 1.54-1.58* 1.89 1.85-1.93* 1.99 1.86-2.13*
Cephalopelvic disproportion 2.42 2.40-2.45* 4.07 3.99-4.15* 5.79 5.50-6.08*
Breech presentation 1.02 1.01-1.04† 1.26 1.23-1.30* 1.55 1.43-1.67*
Obstetric procedure
Induced labor 1.20 1.20-1.21* 1.19 1.17-1.20* 0.93 0.89-0.96*
Cesarean delivery 1.62 1.61-1.63* 2.61 2.58-2.64* 4.68 4.54-4.83*
Birth outcome
5-min Apgar score S6 1.21 1.17-1.24* 1.65 1.56-1.75* 3.49 3.10-3.94*
5-min Apgar score S3 1.30 1.21-1.39* 2.01 1.76-2.29* 5.20 4.09-6.62*
Assisted ventilation S30 min 1.19 1.14-1.23* 1.85 1.73-1.99* 3.96 3.45-4.55*
Assisted ventilation <30 min 1.21 1.19-1.23* 1.46 1.43-1.51* 1.81 1.66-1.97*
Birth injuries 1.99 1.92-2.05* 3.14 2.96-3.32* 4.53 3.95-5.19*
Hyaline membrane disease 1.16 1.10-1.22* 1.84 1.68-2.01* 3.70 3.11-4.40*
Meconium aspiration 1.28 1.23-1.34* 1.65 1.52-1.79* 2.61 2.15-3.16*
Deaths
Neonatal 0.87 0.80-0.96† 1.00 0.83-1.21 2.69 1.91-3.80*
Postneonatal 0.79 0.74-0.85* 0.86 0.74-1.00 1.62 1.16-2.26†
Infant 0.82 0.78-0.86* 0.91 0.80-1.02 2.01 1.58-2.55*

All logistic regression models include measures of maternal race, age, education, marital status, prenatal care use, parity, previous
macrosomic birth, previous pregnancy loss, maternal diabetes mellitus, hypertension, smoking, alcohol use, and gestational age; the ref-
erence group was 3000-3999 g.
*P <.0001.
†P < .005.

of risk rising as the grade of macrosomia increased. This
study demonstrated that mothers of macrosomic infants
were more likely than mothers of normosomic infants to
have a previous pregnancy loss and to receive intensive
prenatal care and a cesarean delivery. The likelihood of
obstetric complications and the occurrence of poor birth
outcomes increased markedly with heavy birth weight.
Grade 3 macrosomic infants had significantly higher
mortality rates compared with normosomic infants, yet
grades 1 and 2 exhibited a lower risk of infant death,
which possibly reflects the role of nutrition, weight gain,
or other uncontrolled factors that influence both heavier
birth weights and lower mortality risk.
The findings of this study are in general agreement
with numerous others that have found macrosomia to be
associated with maternal and neonatal morbidity and
death.1-7,9,18 It has been reported that the incidence of
macrosomia has been increasing across the United States
and in other developed countries.8-13 Subsequent analysis
of national data is needed to confirm the continuation of
this trend. Further investigation of the relationship be-
tween the prevalence of documented risk factors (such as
maternal obesity and diabetes mellitus) and changes in
the incidence of macrosomia is also warranted to deter-
mine whether the prevalence of these risk factors ade-
quately explains the rising incidence rates. The effective-
ness of clinical interventions for the treatment of
suspected macrosomia among mothers who have specific
risk factors (such as diabetes mellitus, excessive weight
gain, and obesity) should also be assessed.
The limitations of this study include the finite set of
risk factors that were available on vital records for investi-
gation. Furthermore, vital record data have been well
noted for possible miscoding and/or misclassification
problems and, in particular, for a lack of precision in the
reported diagnoses of medical conditions and morbidi-
ties. There is little evidence to suggest that these prob-
lems vary by the birth weight categories used in this study
and would bias severely the relationships that were ob-
served between the grades of macrosomia and birth out-
comes. Nevertheless, the interpretation of these results
should incorporate an understanding of the limits of the
data set and should explore the potential validity of atyp-
ical findings. For example, hyaline membrane disease is a
neonatal complication that is associated strongly with
preterm birth as a consequence of immature lungs. In
this study of term births, the occurrence of hyaline mem-
brane disease and its increasing prevalence with higher
Volume 188, Number 5
Am J Obstet Gynecol
grades of macrosomia may more likely be attributable to
medical conditions other than prematurity. Diabetic ba-
bies may be born with an appropriate birth weight or
even macrosomia and also have hyaline membrane dis-
ease because of immature lung functions. Further analy-
sis of our data showed a strong correlation between
maternal diabetes mellitus and the presence of hyaline
membrane disease. Hence, the occurrence of hyaline
membrane disease among macrosomic babies in the
study population is probably due to morbidities (such as
diabetic condition) in the mother.
To our knowledge, there are no similar published stud-
ies that use recent national data to examine adverse out-
comes that are associated with macrosomia across a
number of different birth weight thresholds. This study
provides a conceptual framework for the establishment of
progressive birth weight categories of macrosomia that
are based on the risk of increased health care use, mor-
bidity, and death. Our data suggest that the incorporation
of a three-step definition for fetal macrosomia may be
more clinically useful than a single birth weight cutoff
value. Thus, grade 1 macrosomia may be viewed as a
threshold indicator for an increased risk of delivery com-
plications and can alert the clinician to the need for more
intensive intrapartum treatment. Along with represent-
ing a further increment in the risk of labor and delivery
complications, grade 2 infants often have a greater num-
ber of obstetric procedures and a greater likelihood of
adverse morbidity outcomes. Women at risk for delivery
of a grade 2 macrosomic infant may require enhanced
prenatal treatment. Finally, grade 3 macrosomia repre-
sents significant newborn mortality risks.
It should be noted that these broadly aggregated 500-g
birth weight categories or “grades” of macrosomia are
based on considerations of practicality and ease of use in
clinical, research, and policy situations rather than on a
more detailed analysis of the specific birth weight values
that would demark changing levels of risk optimally. This
study’s more easily recollected classification strategy was
selected therefore as a heuristic approach to facilitate
standard reporting and to promote the conceptualization
of macrosomia as a multilevel condition, with distinctive
variations in its degree of risk with specific adverse health
care use, morbidity, and mortality outcomes. Neverthe-
less, further research toward the generation of a macro-
somia risk score for a variety of adverse outcomes is
warranted.
In summary, the risk factors for macrosomia that are re-
ported in this study are largely consistent with the current
literature. We demonstrated that adverse outcomes (such
as labor complications, use of obstetric procedures,
neonatal morbidity, and infant death) differ across vary-
ing birth weight thresholds. Thus, although defining
macrosomia as >4000 g (grade 1) may be useful for the
identification of the increased risks of labor and newborn
Boulet et al 1377
complications, a definition of >4500 g (grade 2) may be
more predictive of neonatal morbidity, and >5000 g
(grade 3) may be a better indicator of infant death risk.
The proposed three-step definition for macrosomia in-
corporates easily remembered birth weight cutoff values
that may facilitate both the assessment of risk and the pre-
diction of outcome.
REFERENCES
1. Boyd ME, Usher RH, McClean FH. Fetal macrosomia: predic-
tion, risks, and proposed management. Obstet Gynecol 1983;
61:715-22.
2. Spellacy WN, Miller S, Winegar A, Petersen PQ. Macrosomia:
maternal characteristics and infant complications. Obstet Gy-
necol 1985;66:158-61.
3. Meshari AA, DeSilva S, Rahman I. Fetal macrosomia: maternal
risks and fetal outcome. Int J Gynecol Obstet 1990;32:215-22.
4. Lipscomb KR, Gregory K, Shaw K. The outcome of macrosomic
infants weighing at least 4500 grams: Los Angeles County + Uni-
versity of Southern California experience. Obstet Gynecol
1995;85:558-64.
5. Wollschlaeger K, Nieder J, Köppe I, Härtlein K. A study of fetal
macrosomia. Arch Gynecol Obstet 1999;263:51-5.
6. Ferber A. Maternal complications of fetal macrosomia. Clin Ob-
stet Gynecol 2000;43:335-9.
7. Grassi AE, Giuliano MA. The neonate with macrosomia. Clin
Obstet Gynecol 2000;43:340-8.
8. Johar R, Rayburn W, Weir D, Eggert L. Birth weights in term in-
fants: a 50-year perspective. J Reprod Med 1988;33:813-6.
9. ACOG practice bulletin No. 22. Washington (DC): American
College of Obstetricians and Gynecologists; 2000.
10. Brunskill AJ, Rossing MA, Connell FA, Daling J. Antecedents of
macrosomia. Paediatr Perinat Epidemiol 1991;5:392-401.
11. Langer O. Fetal macrosomia: etiologic factors. Clin Obstet Gy-
necol 2000;43:283-97.
12. Baker P, Lever P, Gorton E. An increasing incidence of fetal
macrosomia. J Obstet Gynecol 1992;12:281.
13. Ørskou J, Kesmodel U, Henriksen TB, Secher NJ. An increasing
proportion of infants weighing more than 4000 grams at birth.
Acta Obstet Gynecol Scand 2001;80:931-6.
14. Yawn BP, Wollan P, McKeon K, Field CS. Temporal changes in
rates and reasons for medical induction of term labor, 1980-
1986. Am J Obstet Gynecol 2001;184:611-9.
15. Sacks DA, Chen W. Estimating fetal weight in the management
of macrosomia. Obstet Gynecol Surv 2000;55:229-39.
16. Parks DG, Ziel HK. Macrosomia: a proposed indication for pri-
mary cesarean section. Obstet Gynecol 1978;52:407-9.
17. Davis R, Woelk G, Mueller BA, Daling J. The role of previous
birth weight on risk for macrosomia in a subsequent birth. Epi-
demiology 1995;6:607-11.
18. Bérard J, Dufour D, Vinatier D, Subtil D, Vanderstichèle S, Mon-
nier JC, et al. Fetal macrosomia: risk factors and outcome: a
study of the outcome concerning 100 cases >4500g. Eur J Obstet
Gynecol Reprod Biol 1998;77:51-9.
19. Mocanu EV, Greene RA, Byrne BM, Turner MJ. Obstetric and
neonatal outcome of babies weighing more than 4.5 kg: an analy-
sis by parity. Eur J Obstet Gynecol Reprod Biol 2000;92:229-33.
20. Xiong X, Demianczuk NN, Buekens P, Saunders L, Duncan MB.
Association of preeclampsia with high birth weight for gesta-
tional age. Am J Obstet Gynecol 2000;183:148-55.
21. Rouse DJ, Owen J. Sonography, suspected macrosomia, and pro-
phylactic cesarean: a limited partnership. Clin Obstet Gynecol
2000;43:326-34.
22. O’Reilly-Green C, Divon M. Sonographic and clinical methods
in the diagnosis of macrosomia. Clin Obstet Gynecol 2000;
43:309-20.
23. Sokol RJ, Chik L, Dombrowski MP, Zador IE. Correctly identifying
the macrosomic fetus. Am J Obstet Gynecol 2000;182:1489-95.
1378 Boulet et al May 2003
Am J Obstet Gynecol

24. Öçer F, Kaleli S, Budak E, Oral E. Fetal weight estimation and
prediction of fetal macrosomia in non-diabetic pregnant
women. Eur J Obstet Gynecol Reprod Biol 1999;83:47-52.
25. Weeks JW, Pitman T, Spinnato JA II. Fetal macrosomia: does an-
tenatal prediction affect delivery route and outcome? Am J Ob-
stet Gynecol 1995;173:1215-9.
26. Kolderup LB, Laros RK, Musci TJ. Incidence of persistent birth
injury in macrosomic infants: association with mode of delivery.
Am J Obstet Gynecol 1997;177:37-41.
27. Rouse DJ, Owen J, Goldenberg RL, Cliver SP. The effectiveness
and costs of elective cesarean delivery for fetal macrosomia di-
agnosed by ultrasound. JAMA 1996;276:1480-6.
28. National Center for Health Statistics. 1995, 1996, 1997 Birth Co-
hort Linked Birth/Infant Death Data Sets, NCHS CD-ROM Se-
ries 20, Numbers 12a, 14a, 15a, ASCII Version, US Dept Health
and Human Services, Centers for Disease Control and Preven-
tion. Available from: www.cdc.gov/nchs.
29. Alexander GR, Kotelchuck M. Quantifying the adequacy of pre-
natal care: a comparison of indices. Public Health Rep 1996;
111:408-18.
30. Ventura SJ, Martin JA, Curtin SC, Mathews TJ. Report of the na-
tality statistics, 1995 (monthly vital statistics report: vol 45, no
11, supp 2). Hyattsville (MD): National Center for Health Sta-
tistics; 1997.

Correction
In the letter by Sharma and El-Refaey, the second author’s name was misspelled. The cor-
rect citation is Sharma S, El-Refaey H. The use of misoprostol as a cervical ripening agent in
operative hysteroscopy. Am J Obstet Gynecol 2003;188:297-8.