Archives of Gynecology and Obstetrics

, Volume 284, Issue 4, pp 827–836

Pregnancy, obesity, gestational weight gain, and parity as predictors of peripartum complications

Authors

    • Departments of Obstetrics and GynecologyUniversity of Arkansas for the Medical Sciences
  • Dorota A. Doherty
    • School of Women’s and Infants’ HealthUniversity of Western Australia
  • Suneet P. Chauhan
    • Eastern Virginia Medical School
  • Jennifer M. Klimpel
    • Naval Medical Center—Portsmouth
  • Shannon D. Huff
    • University of Mississippi Medical Center
  • John C. Morrison
    • University of Mississippi Medical Center
Materno-fetal Medicine

DOI: 10.1007/s00404-010-1754-0

Cite this article as:
Magann, E.F., Doherty, D.A., Chauhan, S.P. et al. Arch Gynecol Obstet (2011) 284: 827. doi:10.1007/s00404-010-1754-0

Abstract

Purpose

To determine if the best predictor of pregnancy complications is pre-pregnancy body mass index (BMI) alone or in combination with other factors.

Methods

BMI and peripartum outcomes of singleton pregnancies were evaluated. Recursive partitioning and logistic regression modeling was used.

Results

Of the 4,286 cohorts, 26% were obese (BMI >30 kg/m2) and, compared to cohorts with normal weight, at risk for wound infections (P < 0.001), and shoulder dystocia (P < 0.001). High-risk patients (15%; BMI >32.5, parity, pregnancy weight gain of 28 lb by 28 weeks) were at increased risk for wound infection (P < 0.001), endometritis (P < 0.001), shoulder dystocia (P = 0.001) and 5 min Apgar score <4 (P < 0.041) and at lower risk for pre-term delivery (P = 0.007).

Conclusions

Since BMI, parity, and weight gain until 28 weeks together provide better prediction of peripartum complications than BMI alone, these characteristics can be used to triage and refer patients.

Keywords

PregnancyMaternal obesityBody mass indexPregnancy complications

Introduction

Obesity is becoming more prevalent in obstetrics. It is currently estimated by the American College of OB-GYN that 38% of all pregnant women are overweight or obese based on pre-pregnancy weights. Maternal obesity has long been correlated with an increased risk of chronic hypertension and diabetes prior to pregnancy and adverse pregnancy outcomes including preeclampsia, gestational diabetes, fetal macrosomia, cesarean deliveries, postpartum endometritis, and a prolonged postpartum hospital stay [17]. The perinatal problems that have been identified with maternal obesity and pregnancy include an increased risk of neural tube defects, birth asphyxia, birth trauma, neonatal hypoglycemia, and stillbirth [811]. Two large studies from the Swedish Medical Birth Registry have linked a pre-pregnancy diagnosis of obesity and morbid obesity with late fetal death, preeclampsia, cesarean delivery, shoulder dystocia, and large for gestational age (LGA) neonates [12, 13].

In addition to pre-pregnancy body mass index (BMI), weight gain during pregnancy has also been correlated with adverse pregnancy outcomes. A systematic review article, which searched English publications from 1990 to 2007, noted that there is a strong association between maternal gestational weight gain and pre-term birth, total birth weight, LGA, and small for gestational age neonates. Furthermore, moderate evidence supported an association with cesarean delivery and weight retention at from 3 months to 3 years postpartum [14]. The reviewers’ recommended additional studies to fully understand the impact of weight gain during pregnancy and peripartum outcomes.

There were two objectives for this investigation. The first objective was to evaluate the National Institute of Health weight classification and its association with perinatal outcomes. The second objective was to determine if the pre-pregnancy BMI alone versus BMI in combination with other factors would improve the prediction of perinatal outcomes.

Materials and methods

This is an observational study of pregnant women attending the prenatal clinics of the University of Mississippi Medical Center in Jackson, MS and the Naval Medical Center—Portsmouth, Portsmouth, VA and delivering at those respective hospitals. Antepartum and intrapartum patient information was retrospectively collected following delivery. Data collection was done manually by three of the authors (EFM, JKM, and SDH) and completed by the patient’s hospital discharge except on neonates admitted to the neonatal intensive care unit (NICU) whose data collection was completed manually at the time of their hospital discharge or perinatal death. This study was approved by the IRB at the University of Mississippi Medical Center in Jackson, MS (IRB # 2005-012) and Naval Medical Center, Portsmouth, VA IRB/IACUC protocol (CIP #2005.0053).

Information collected on each pregnancy included maternal demographics, the presence of chronic diseases (pre-existing diabetes or chronic hypertension), development of a urinary tract infection, gestational diabetes and/or preeclampsia, induction of labor and indication for that induction, mode of delivery, shoulder dystocia at delivery, fetal weight and gestational age at delivery, Apgar scores, umbilical cord pH, development of chorioamnionitis/postpartum endometritis/wound infection and perinatal outcome. The neonatal birth weights were classified as LGA or small for gestational age based on the population established percentiles of >90 and <10, respectively, of the Alexander’s national reference for fetal growth [15]. The women were weighed and their BMI was calculated on their first prenatal visit, at or within 2 weeks of their 28 week visit, and when they were admitted to labor and delivery. All of the prenatal visits, when the initial weight was obtained and the BMI calculated were in the first trimester of pregnancy. Gestational ages were confirmed by ultrasound. The women were classified into groups based on their BMI at their first prenatal visit. The grouping was based on the Institute of Medicine and National Research Council Committees’ recent update of their guidelines on maternal weight gain during pregnancy [16]. The women were classified into four groups: underweight (BMI, weight in kg/height {m}2) <18.5; normal, 18.5–24.9; overweight, 25–29.9, and; obesity, ≥30 kg/m2 based on their BMI at the time of their first prenatal visit. In addition to the risk categories defined with BMI at the first antenatal visit, an alternative risk categorization was developed based on a combination of maternal factors as well as the alternative BMI groupings. The pregnancy risk groups, based on BMI alone and composite pregnancy risk groups were both correlated with antepartum, intrapartum, and neonatal outcomes.

Medians and interquartile ranges (IQR) were used to summarize continuous data, and frequency distributions were used to summarize categorical data. Univariable comparisons of outcomes between different pregnancy risk groups were conducted using Kruskal–Wallis non-parametric analysis of variance for continuous data and χ2 test for categorical data. Univariable and multivariable logistic regression analysis was implemented to compare the effects of risk group categories on the pregnancy outcomes, with the odds ratios (OR) and their 95% confidence intervals (CI) used to summarize the effects sizes. Risk groups based on BMI alone and risk groups derived using BMI together with other risk factors were evaluated using logistic regression modeling.

Recursive partitioning modeling useful for identification of nonlinear and complex interactions, that is an alternative to logistic regression modeling, was implemented to derive a binary classification tree for allocation of risk categories for adverse pregnancy outcomes (RPART, S-Plus). Derivation of the alterative risk groups based on recursive partitioning split the population into different subgroups of pregnancies as different as possible with respect to the response variable, the occurrence of at least one of the following adverse outcomes: preeclampsia, gestational diabetes, induction of labor, cesarean delivery, PPH, post-term delivery, endometritis, wound infection, neonate born LGA, perinatal death, and neural tube defects. Candidate maternal characteristics considered in recursive partitioning along with BMI included ethnicity, age, parity, gravidity, weight gain in pregnancy from first visit to 28 weeks, pre-existing hypertension and diabetes. Several of the initial risk subgroups derived using recursive partitioning were subsequently merged using the stepwise logistic regression, when no statistically significant differences between the rates of adverse pregnancy outcome were found between these subgroups. The final four risk groups retained for further analysis were based on a combination of parity (0 vs. >0), age (<30 vs. ≥30), BMI (<25, 25–32.5, and ≥32.5), and weight gain until 28 pregnancy weeks (<28 lb vs. ≥28 lb). All hypothesis tests were two-sided and P values <0.05 were considered statistically significant. SPSS (version 15, Chicago, IL) and S-Plus (MathSoft Inc., Cambridge, MA) statistical software was used for data analysis.

Results

Over the 18 months of the study (January 2007–July 2008), 4,500 women met the inclusion criteria for this study. Median maternal age was 24 years (IQR 21–29, range 13–44) and median parity 1 (IQR 0–1, range 0–10). The majority of the women were Caucasian (2,064, 45.9%), followed by African-American ethnicity (1,860, 41.3%), Hispanic (362, 8%), and other ethnicities (182, 4%). The pre-pregnancy BMI was normal (18.5–24.9) in 1,966 women, underweight in 274 (BMI <18.5), overweight (BMI 25–29.9) in 1,073, and obese (BMI >30) in 1,183. Maternal and pregnancy characteristics by BMI risk group are presented in Table 1. Statistically significant differences between the BMI risk groups were evident for age, ethnicity, gravidity, parity, and weight gain in pregnancy until 28 pregnancy weeks (all P < 0.001). Rates of pre-existing hypertension and diabetes significantly increased with increasing BMI (both P < 0.001). Gestational age at delivery and birth weight significantly increased with higher BMI (both P < 0.001). Pregnancy loss before 20 weeks occurred in 10 pregnancies (0.2%). Relative to pregnancies with normal BMI (0.1% of pregnancy losses), pregnancies with BMI >30 (OR 10.02, 95% CI 1.20–83.31, P = 0.033, 0.5% of pregnancy losses) were at an increased risk of early pregnancy loss while the risk in pregnancies with BMI 25–30 was comparable to the risk among pregnancies with normal BMI (OR 1.83, 95% CI 0.12–29.34, P = 0.668, 0.1% of pregnancy losses).
Table 1

Patient characteristics stratified by the pre-pregnancy BMI

Characteristic

Pre-pregnancy BMI

P value

Normal (n = 1,966)

Underweight (n = 278)

Overweight (n = 1,073)

Obese (n = 1,183)

Agea

24 (21–28)

22 (20–25)

24 (21–28)

26 (22–30)

<0.001

Caucasian

1,001 (50.9)

101 (36.3)

522 (48.6)

440 (37.2)

<0.001

African-American

669 (34.0)

104 (37.4)

416 (38.8)

671 (56.7)

 

Hispanic

191 (9.7)

41 (14.7)

88 (8.2)

42 (3.6)

 

Other ethnicities

91 (4.6)

28 (10.1)

40 (3.7)

23 (1.9)

 

Gravidity

2 (1–3)

2 (1–3)

2 (1–3)

2 (2–3)

<0.001

Parity

1 (0–1)

1 (0–1)

1 (0–1)

1 (0–2)

<0.001

Nulliparous

925 (47.0)

133 (47.8)

468 (43.6)

351 (29.7)

<0.001

Weight gain 0–28 weeks (pounds/week)a

0.71 (0.50–1.00)

0.75 (0.54–0.93)

0.75 (0.50–1.07)

0.61 (0.35–0.89)

<0.001

Weight gain 28 weeks–delivery (pounds/week)a

0.91 (0.62–1.38)

0.92 (0.62–1.33)

0.94 (0.61–1.47)

0.94 (0.57–1.43)

0.391

Pre-existing hypertension

43 (2.2)

4 (1.4)

33 (3.1)

175 (14.8)

<0.001

Pre-existing diabetes

13 (0.7)

3 (1.1)

13 (1.2)

68 (5.7)

<0.001

Gestational age at delivery (weeks)a

39.1 (38.0–40.1)

38.7 (36.6–39.9)

39.3 (38.3–40.1)

39.0 (37.6–39.9)

<0.001

Birth weight (g)a

3,270 (2,915–3,595)

2,978 (2,388–3,374)

3,395 (3,075–3,755)

3,370 (2,916–3,780)

<0.001

Underweight = BMI <18.5; Normal weight = BMI 18.50–25; Overweight = BMI 25–30; Obese = BMI >30

aN (%) or medians and interquartile ranges (Q1Q3) are shown

Univariable and adjusted effects of BMI categories on antenatal and intrapartum pregnancy outcomes are shown in Table 2. Compared to women with a normal BMI, obese women were more likely to develop preeclampsia (OR 2.53, 95% CI 1.94–3.29, P < 0.001), urinary tract infection (OR 1.80, 95% CI 1.38–2.34, P = 0.001), gestational diabetes (OR 4.29, 95% CI 3.05–6.02, P < 0.001), require induction of labor (OR 1.69, 95% CI 1.34–2.12, P < 0.001), and cesarean delivery (OR 1.67, 95% CI 1.41–1.98). Compared to women with normal BMI, overweight women were more likely to develop gestational diabetes (OR 2.16, 95% CI 1.49–3.13, P < 0.001), cesarean delivery (OR 1.40, 95% CI 1.18–1.66, P < 0.001).
Table 2

Antenatal and intrapartum outcomes according to the pre-pregnancy BMI category

Outcome (n = 4,490)a

n (%)b

Unadjusted OR

P value

Adjusted OR

P value

OR

95% CI

OR

95% CI

Pregnancy loss

10 (0.2)

      

 Normal weight

1 (0.1)

1.00

     

 Overweight

1 (0.1)

1.83

(0.12–29.34)

0.668

1.83

(0.12–29.34)

0.668

 Obese

6 (0.5)

10.02

(1.20–83.31)

0.033

10.02

(1.20–83.31)

0.033

Preeclampsia

392 (8.7)

      

 Normal weight

116 (5.9)

1.00

  

1.00

  

 Overweight

65 (6.1)

1.01

(0.74–1.39)

0.939

0.99

(0.72–1.37)

0.967

 Obese

192 (16.3)

3.09

(2.42–3.95)

<0.001

2.53

(1.94–3.29)

<0.001

UTI

348 (7.8)

      

 Normal weight

130 (6.6)

1.00

  

1.00

  

 Overweight

81 (7.6)

1.15

(0.87–1.54)

0.329

1.18

(0.88–1.57)

0.273

 Obese

122 (10.4)

1.63

(1.26–2.12)

<0.001

1.80

(1.38–2.34)

<0.001

Gestational diabetes

248 (5.5)

      

 Normal weight

56 (2.8)

1.00

  

1.00

  

 Overweight

64 (6.0)

2.13

(1.47–3.08)

<0.001

2.16

(1.49–3.13)

<0.001

 Obese

122 (10.4)

3.91

(2.82–5.42)

<0.001

4.29

(3.05–6.02)

<0.001

Pre-term delivery

690 (15.4)

      

 Normal weight

278 (14.2)

1.00

     

 Overweight

122 (11.4)

0.78

(0.63–0.98)

0.036

0.75

(0.59–0.94)

0.013

 Obese

218 (18.6)

1.39

(1.14–1.69)

0.001

0.96

(0.78–1.19)

0.705

Induction of labor

815 (18.2)

      

 Normal weight

278 (14.1)

1.00

  

1.00

  

 Overweight

186 (17.4)

1.28

(1.04–1.56)

0.019

1.24

(0.98–1.57)

0.078

 Obese

302 (25.7)

2.10

(1.75–2.52)

<0.001

1.69

(1.34–2.12)

<0.001

CS delivery

549 (29.2)

      

 Normal weight

462 (23.5)

1.00

  

1.00

  

 Overweight

329 (30.7)

1.45

(1.23–1.71)

<0.001

1.40

(1.18–1.66)

<0.001

 Obese

461 (39.2)

2.09

(1.79–2.45)

<0.001

1.67

(1.41–1.98)

<0.001

CS for FD

337 (7.5)

      

 Normal weight

144 (7.3)

1.00

  

1.00

  

 Overweight

77 (7.2)

0.99

(0.74–1.32)

0.925

0.96

(0.72–1.28)

0.777

 Obese

94 (8.0)

1.09

(0.83–1.43)

0.518

0.90

(0.67–1.20)

0.465

Simultaneous covariates used for obtain adjusted OR are listed individually for each outcome

CS cesarean delivery, FD fetal distress, UTI urinary tract infection, GDM gestational diabetes mellitus, LGA large for gestational age

Pregnancy loss—no adjustments were made due to sample size limitations; preeclampsia—African-American ethnicity, nulliparity, pre-existing chronic hypertension; UTI—nulliparity; GDM—advancing maternal age, nulliparity, Hispanic ethnicity, ethnicities other than Caucasian, African-American, Hispanic, pre-existing chronic hypertension; pre-term delivery—maternal age, African-American ethnicity, preeclampsia; induction of labor—nulliparity, pre-existing chronic hypertension, preeclampsia, post-term gestation; CS—maternal age, nulliparity, African-American ethnicity, pre-existing diabetes, preeclampsia, antenatally diagnosed LGA fetus; CS for FD—maternal age, nulliparity, African-American ethnicity, pre-existing diabetes, preeclampsia, antenatally diagnosed LGA fetus

aTen cases of early pregnancy loss were excluded

bUnderweight cases are not included but % in each category have been retained

Effects of BMI groups on postpartum and neonatal outcomes are summarized in Table 3. Compared to women with normal BMI, women with BMI >30 were at higher risk of postpartum hemorrhage (OR 1.86, 95% CI 1.25–2.78, P = 0.002), wound infection (OR 9.79, 95% CI 3.83–24.99, P < 0.001), shoulder dystocia (OR 3.8, 95% CI 2.19–6.58, P < 0.001), meconium-stained amniotic fluid (OR 3.49, 95% CI 1.41–8.53, P = 0.007), and LGA infants (OR 3.10, 95% CI 2.32–4.15, P < 0.001). Obese and overweight women were at lower risk of small for gestational age infants (OR 0.67, 95% CI 0.48–0.93, P = 0.015 and OR 0.63, 95% CI 0.44–0.90, P = 0.010, respectively). Low Apgar scores at 5 min, low pH and perinatal mortality were not significantly associated with BMI. Maternal age and nulliparity were common maternal factors that altered risk of the individual adverse outcomes (Tables 2, 3).
Table 3

Postpartum and neonatal outcomes according to the pre-pregnancy BMI category

Outcome

N (%)

Unadjusted OR

P value

Adjusted OR

P value

OR

95% CI

OR

95% CI

PPH

166 (3.7)

      

 Normal weight

52 (2.6)

1.00

  

1.00

  

 Overweight

46 (4.3)

1.65

(1.10–2.47)

0.015

1.42

(0.94–2.15)

0.095

 Obese

63 (5.4)

2.09

(1.44–3.04)

<0.001

1.86

(1.25–2.78)

0.002

Endometritis

309 (6.9)

      

 Normal weight

131 (6.7)

1.00

  

1.00

  

 Overweight

56 (5.2)

0.77

(0.56–1.06)

0.111

0.65

(0.46–0.91)

0.011

 Obese

116 (9.9)

1.52

(1.17–1.97)

<0.001

1.31

(0.98–1.76)

0.065

Wound infection

63 (4.8)

      

 Normal weight

5 (1.1)

1.00

  

1.00

  

 Overweight

4 (1.2)

1.12

(0.30–4.22)

0.862

0.95

(0.25–3.60)

0.944

 Obese

54 (11.7)

12.13

(4.80–30.61)

<0.001

9.79

(3.83–24.99)

<0.001

Shoulder dystocia

77 (2.4)

      

 Normal weight

22 (1.5)

1.00

  

1.00

  

 Overweight

13 (1.8)

1.16

(0.57–2.37)

0.688

0.97

(0.47–2.00)

0.928

 Obese

39 (5.4)

4.07

(2.38–6.98)

<0.001

3.80

(2.19–6.58)

<0.001

Meconium Aspiration

30 (0.7)

      

 Normal weight

8 (0.4)

1.00

  

1.00

  

 Overweight

8 (0.7)

1.83

(0.67–4.90)

0.227

1.90

(0.70–5.11)

0.206

 Obese

13 (1.1)

2.73

(1.13–6.61)

0.026

3.47

(1.41–8.53)

0.007

5 min Apgar ≤4

296 (6.6)

      

 Normal weight

115 (5.9)

1.00

  

1.00

  

 Overweight

52 (4.9)

0.82

(0.58–1.15)

0.241

0.85

(0.58–1.25)

0.413

 Obese

102 (8.7)

1.53

(1.16–2.01)

0.003

1.05

(0.74–1.48)

0.799

pH <7.1

131 (2.9)

      

 Normal weight

55 (2.8)

1.00

  

1.00

  

 Overweight

16 (1.5)

0.82

(0.58–1.15)

0.241

0.56

(0.30–1.07)

0.079

 Obese

46 (3.9)

1.53

(1.16–2.01)

0.003

0.81

(0.50–1.32)

0.403

SGA

272 (6.1)

      

 Normal weight

127 (6.5)

1.00

  

1.00

  

 Overweight

45 (4.2)

0.65

(0.45–0.92)

0.014

0.63

(0.44–0.90)

0.010

 Obese

67 (5.7)

0.89

(0.65–1.21)

0.447

0.67

(0.48–0.93)

0.015

LGA

335 (7.5)

      

 Normal weight

91 (4.6)

1.00

  

1.00

  

 Overweight

101 (9.4)

2.14

(1.60–2.88)

<0.001

2.18

(1.62–2.93)

<0.001

 Obese

138 (11.7)

2.74

(2.08–3.61)

<0.001

3.10

(2.32–4.15)

<0.001

Perinatal death

80 (1.8)

      

 Normal weight

34 (1.7)

1.00

  

1.00

  

 Overweight

10 (0.9)

0.53

(0.26–1.08)

0.082

0.85

(0.36–1.99)

0.705

 Obese

29 (2.5)

1.43

(0.87–2.37)

0.158

1.26

(0.66–2.38)

0.487

Simultaneous covariates used for obtain adjusted OR are listed individually for each outcome

PPH postpartum hemorrhage, SGA small for gestational age, LGA large for gestational age, pH acidity or basicity of an umbilical artery blood sample

PPH—ethnicity other than Caucasian, African-American or Hispanic, nulliparity, pre-existing diabetes, cesarean delivery, macrosomia diagnosed during pregnancy, gestational age <37 completed weeks (less PPH); endometritis—Hispanic ethnicity, nulliparity, cesarean delivery, pre-existing diabetes, large for GA baby, urinary tract infection in pregnancy, gestational age at delivery (less with advancing gestation); wound infection—GDM, and large for gestational age baby and urinary tract infection during pregnancy; shoulder dystocia—LGA fetus and gestational at delivery; meconium—maternal age (less for older women), ethnicities other than Caucasian, African-American and Hispanic, post-term delivery, and IUGR fetus; 5 min Apgar ≤4—maternal age, African-American ethnicity, nullparity, pre-existing hypertension (less likely if occurs), pre-existing diabetes, induction, gestational age at delivery (less with advancing gestation), delivery occurring post-term, cesarean delivery, being born SGA or LGA, meconium and shoulder dystocia; pH <7.1—African-American ethnicity, nulliparity, pre-existing diabetes, induction, cesarean delivery, gestational age at delivery (lower risk with advancing gestation), being born SGA or LGA, meconium; SGA—African-American ethnicity, nulliparity, preeclampsia; LGA—African-American ethnicity (lowers the likelihood of LGA), pre-existing diabetes, GDM and preeclampsia (lowers the likelihood of LGA); perinatal death—gestational age at delivery, SGA fetus, post-term delivery

Recursive partitioning indicated alternative BMI thresholds of <25, 25–32.5, and >32.5 as indicative of a risk for adverse pregnancy outcome. Additional simultaneous risk factors were age with thresholds of <30 and ≥30, nulliparity and weight gain <28 lb until 28 pregnancy weeks (Table 4). Women who delivered before 28 pregnancy weeks were categorized using the combination of age, BMI, and parity alone. Compared to the BMI based risk groupings, the categorization that includes age, parity, and weight gain in addition to a modified cutoff for BMI leads to an improved separation of pregnancies with at least one adverse outcome among preeclampsia, gestational diabetes, labor induction, cesarean delivery, postpartum hemorrhage, endometritis, LGA neonate, perinatal death, and neural tube defects (Table 4).
Table 4

Alternative risk categorizations using BMI alone and a combination maternal factors in addition to BMI

Risk group (n = 4,490)a

n (%)

At least one complication (n = 2,308, 51.4%)b

Description

BMI

 18.5–25

1,965 (43.8)

845 (43.0%)

Normal BMI

 <18.5

276 (6.1)

116 (42.0%)

Underweight

 25–30

1,072 (23.9)

566 (52.8%)

Overweight

 30+

1,177 (26.1)

781 (66.4%)

Obesity

BMI + other factors

 Lowest risk

677 (15.1)

238 (35.2%)

P > 0, age <30, BMI <25, wt gain <28 lb

 Low risk

2,080 (46.3)

952 (45.8%)

P > 0, age <30, BMI <25, wt gain >28 lb;

P > 0, age >30, BMI < 25;

P > 0, BMI 25–32.5, wt gain <28 lb;

P = 0, age <30, BMI <25;

P = 0, age >30, BMI <25, wt gain <28 lb

 Moderate risk

1,070 (23.8)

622 (58.1%)

P > 0, BMI 25–32.5, wt gain > 28 lb;

P > 0, age <30, BMI > 32.5;

P = 0, age <30, BMI <25, wt gain <28 lb;

P = 0, BMI 25–32.5, wt gain <28 lb

 High risk

663 (14.8)

496 (74.8%)

P > 0, BMI > 32.5;

P = 0, BMI 25–32.5, wt gain >28 lb;

P = 0, BMI >32.5

aExcluding 10 early pregnancy losses

bComplications include: preeclampsia, gestational diabetes, labor induction, cesarean delivery, postpartum hemorrhage, endometritis, post dates, LGA fetus, perinatal death, and neural tube defects (n = 5)

Summaries of the univariable and adjusted effects of risk groupings on the likelihood of adverse antenatal and intrapartum outcomes are summarized in Table 5, and the effects on the postpartum and neonatal outcomes are shown in Table 6. BMI with maternal demographics and maternal weight gain to 28 weeks identified additional peripartum complications that were not identified by the pre-pregnancy BMI alone.
Table 5

Antenatal and intrapartum outcomes according to the risk category defined using a combination of parity (nulliparous vs. parous), age (<30 vs. ≥30), pre-pregnancy BMI (<25, 25–32.5, ≥32.5) and weight gain until 28 weeks gestational age (<28 vs. ≥28 lb)

Outcome

N (%)

Unadjusted OR

P value

Adjusted OR

P value

OR

95% CI

OR

95% CI

Preeclampsia

392 (8.7)

      

 Lowest risk

38 (5.6)

1.00

  

1.00

  

 Low risk

119 (5.7)

1.01

(0.69–1.47)

0.954

1.05

(0.72–1.53)

0.800

 Moderate risk

94 (8.8)

1.60

(1.08–2.34)

0.019

1.42

(0.96–2.11)

0.082

 High risk

141 (21.3)

4.51

(3.10–6.58)

<0.001

3.44

(2.34–5.07)

<0.001

UTI

348 (7.8)

      

 Lowest risk

28 (4.1)

1.00

  

1.00

  

 Low risk

141 (6.9)

1.72

(1.14–2.61)

0.010

1.72

(1.14–2.61)

0.010

 Moderate risk

99 (9.3)

2.36

(1.54–3.64)

<0.001

2.36

(1.54–3.64)

<0.001

 High risk

77 (11.6)

3.05

(1.95–4.76)

<0.001

3.05

(1.95–4.76)

<0.001

Gestational diabetes

248 (5.5)

      

 Lowest risk

7 (1.0)

1.00

  

1.00

  

 Low risk

89 (4.3)

4.23

(1.97–9.29)

<0.001

4.19

(1.93–9.09)

<0.001

 Moderate risk

82 (7.7)

7.71

(3.54–16.80)

<0.001

7.58

(3.48–16.52)

<0.001

 High risk

70 (10.6)

11.24

(5.13–24.62)

<0.001

10.94

(4.98–24.01)

0.044

Pre-term delivery

690 (15.4)

      

 Lowest risk

142 (21.0)

1.00

     

 Low risk

272 (13.1)

0.56

(0.44–0.70)

<0.001

0.58

(0.46–0.74)

<0.001

 Moderate risk

134 (12.5)

0.54

(0.41–0.69)

<0.001

0.46

(0.35–0.61)

<0.001

 High risk

142 (21.6)

1.02

(0.79–1.33)

0.863

0.67

(0.51–0.90)

0.007

Induction of labor

815 (18.2)

      

 Lowest risk

72 (10.6)

1.00

  

1.00

  

 Low risk

207 (14.8)

1.45

(1.11–1.91)

0.007

1.61

(1.17–2.20)

0.003

 Moderate risk

210 (19.6)

2.05

(1.54–2.72)

<0.001

2.03

(1.47–2.87)

<0.001

 High risk

226 (34.1)

4.38

(3.27–5.87)

<0.001

3.24

(2.22–4.70)

<0.001

CS delivery

549 (29.2)

      

 Lowest risk

124 (19.3)

1.00

  

1.00

  

 Low risk

486 (24.1)

1.25

(1.01–1.55)

0.042

1.31

(1.05–1.62)

0.017

 Moderate risk

368 (34.6)

2.06

(1.65–2.59)

<0.001

2.02

(1.60–2.54)

<0.001

 High risk

285 (45.5)

3.28

(2.57–4.19)

<0.001

2.79

(2.17–3.38)

<0.001

CS for FD

337 (7.5)

      

 Lowest risk

46 (6.8)

1.00

  

1.00

  

 Low risk

135 (6.5)

0.99

(0.70–1.41)

0.955

1.07

(0.75–1.54)

0.696

 Moderate risk

86 (8.0)

1.25

(0.86–1.82)

0.247

1.27

(0.87–1.89)

0.216

 High risk

70 (10.6)

1.69

(1.14–2.51)

0.009

1.45

(0.97–2.17)

0.072

Simultaneous covariates used for obtain adjusted OR are listed individually for each outcome

CS cesarean delivery, FD fetal distress, UTI urinary tract infection

Preeclampsia—African-American ethnicity, pre-existing chronic hypertension; UTI—no adjustments; GDM—Hispanic ethnicity, ethnicity other than Caucasian or African-American or Hispanic; pre-term delivery—African-American ethnicity, pre-existing chronic hypertension, preeclampsia; induction of labor—pre-existing chronic hypertension, preeclampsia, post-term, pre-term GA; CS—African-American ethnicity, pre-existing diabetes, preeclampsia, macrosomia diagnosed antenatally, gestational age <37 completed weeks; Cs for FD—African-American ethnicity, induction of labor, gestational age <37 completed weeks

Table 6

Postpartum and neonatal outcomes according to the risk category defined using a combination of parity (nulliparous vs. parous), age (<30 vs. ≥30), pre-pregnancy BMI (<25, 25–32.5, ≥32.5) and weight gain until 28 weeks gestational age (<28 vs. ≥28 lb)

Outcome

N (%)

Unadjusted OR

P value

Adjusted OR

P value

OR

95% CI

OR

95% CI

PPH

166 (3.7)

      

 Lowest risk

11 (1.6)

1.00

  

1.00

  

 Low risk

51 (2.5)

1.52

(0.79–2.94)

0.211

1.35

(0.70–2.63)

0.363

 Moderate risk

50 (4.7)

2.95

(1.52–5.71)

0.001

2.35

(1.20–4.60)

0.013

 High risk

54 (8.1)

5.37

(2.78–10.36)

<0.001

4.31

(2.20–8.47)

<0.001

Endometritis

309 (6.9)

      

 Lowest risk

16 (2.4)

1.00

  

1.00

  

 Low risk

117 (5.6)

2.46

(1.45–4.18)

0.001

2.28

(1.34–3.89)

0.002

 Moderate risk

86 (8.0)

3.60

(2.09–6.19)

<0.001

2.70

(1.55–4.69)

<0.001

 High risk

90 (13.6)

6.46

(3.75–11.13)

<0.001

4.18

(2.39–7.32)

<0.001

Wound infection

63 (4.8)

      

 Lowest risk

2 (1.4)

1.00

  

1.00

  

 Low risk

3 (0.6)

0.41

(0.07–2.48)

0.333

0.36

(0.06–2.21)

0.271

 Moderate risk

24 (6.5)

4.77

(1.11–20.44)

0.036

3.46

(0.79–15.12)

0.099

 High risk

34 (11.3)

8.76

(2.07–36.99)

0.003

5.62

(1.31–24.21)

0.020

Shoulder dystocia

77 (2.4)

      

 Lowest risk

7 (1.3)

1.00

  

1.00

  

 Low risk

30 (1.9)

1.47

(0.64–3.38)

0.358

1.39

(0.61–3.20)

0.390

 Moderate risk

18 (2.6)

2.00

(0.83–4.82)

0.123

1.63

(0.67–3.98)

0.280

 High risk

22 (6.1)

4.97

(2.10–11.75)

<0.001

4.36

(1.82–10.44)

0.001

Meconium

30 (0.7)

      

 Lowest risk

1 (0.1)

1.00

  

1.00

  

 Low risk

15 (0.7)

4.91

(0.65–37.24)

0.124

4.87

(0.64–37.03)

0.126

 Moderate risk

8 (0.7)

5.09

(0.64–40.81)

0.125

5.16

(0.64–41.52)

0.123

 High risk

6 (0.9)

6.17

(0.74–51.42)

0.092

5.94

(0.71–49.57)

0.100

5 min Apgar ≤4

296 (6.6)

      

 Lowest risk

39 (5.8)

1.00

  

1.00

  

 Low risk

119 (5.7)

0.99

(0.68–1.44)

0.968

1.02

(0.67–1.53)

0.908

 Moderate risk

53 (5.0)

0.85

(0.56–1.300)

0.452

0.72

(0.46–1.15)

0.169

 High risk

85 (12.8)

2.40

(1.61–3.56)

<0.001

1.59

(1.02–2.48)

0.041

pH <7.1

131 (2.9)

      

 Lowest risk

24 (3.5)

1.00

  

1.00

  

 Low risk

47 (2.3)

0.63

(0.38–1.04)

0.068

0.82

(0.47–1.42)

0.478

 Moderate risk

23 (2.1)

0.60

(0.33–1.06)

0.078

0.64

(0.34–1.23)

0.179

 High risk

37 (5.6)

1.61

(0.95–2.71)

0.077

0.98

(0.54–1.79)

0.948

SGA

272 (6.1)

      

 Lowest risk

54 (8.0)

1.00

  

1.00

  

 Low risk

127 (6.1)

0.76

(0.54–1.06)

0.104

0.88

(0.62–1.23)

0.445

 Moderate risk

45 (4.2)

0.51

(0.34–0.77)

0.001

0.53

(0.35–0.81)

0.003

 High risk

46 (6.9)

0.88

(0.58–1.32)

0.524

0.69

(0.45–1.06)

0.086

LGA

335 (7.5)

      

 Lowest risk

24 (3.5)

1.00

  

1.00

  

 Low risk

112 (5.4)

1.54

(0.99–2.43)

0.057

1.29

(0.82–2.03)

0.276

 Moderate risk

118 (11.0)

3.35

(2.14–5.26)

<0.001

2.89

(1.83–4.56)

<0.001

 High risk

81 (12.2)

3.78

(2.36–6.04)

<0.001

3.35

(2.07–5.41)

<0.001

Perinatal death

80 (1.8)

      

 Lowest risk

13 (1.9)

1.00

  

1.00

  

 Low risk

37 (1.8)

0.92

(0.49–1.75)

0.809

1.48

(0.75–2.91)

0.256

 Moderate risk

4 (0.4)

0.19

(0.06–0.59)

0.004

0.17

(0.05–0.61)

0.007

 High risk

26 (3.9)

2.08

(1.06–4.08)

0.033

1.94

(0.95–3.97)

0.068

Simultaneous covariates used for obtain adjusted OR are listed individually for each outcome

PPH postpartum hemorrhage, SGA small for gestational age, LGA large for gestational age

PPH—Hispanic ethnicity, ethnicity other than Caucasian, African-American and Hispanic, cesarean delivery, LGA fetus, gestational age at delivery <37 completed weeks (less PPH); endometritis—Hispanic ethnicity, urinary tract infection during pregnancy, induction of labor, cesarean delivery, LGA fetus; wound infection—GDM, preeclampsia, LGA fetus, urinary tract infection during pregnancy; shoulder dystocia—LGA fetus, gestational at delivery; meconium—delivery post-term, and SGA fetus; 5 min Apgar ≤4—African-American ethnicity, pre-existing hypertension, pre-existing diabetes, induction of labor, cesarean delivery, SGA fetus, LGA fetus, meconium, shoulder dystocia; pH <7.1—African-American ethnicity, pre-existing diabetes, induction of labor, cesarean delivery, gestational age <37 completed weeks, SGA fetus, LGA fetus, meconium, shoulder dystocia; SGA—African-American ethnicity, preeclampsia, delivery pre-term; LGA—African-American ethnicity (lowers the likelihood of LGA), pre-existing diabetes, GDM, delivery post-term, delivery pre-term (lowers the likelihood of LGA); perinatal death—African-American ethnicity, pre-term delivery, meconium

Discussion

In our study, obese women compared to women with a normal BMI were more likely to develop preeclampsia, gestational diabetes, have an indicated induction of labor, cesarean delivery, wound infection, shoulder dystocia, meconium-stained amniotic fluid, LGA neonates, and decreased risk of SGA neonates. Multiple investigations have recognized the increased risk of preeclampsia [13], gestational diabetes [7, 17], labor induction [17], cesarean delivery [7, 13, 1719], shoulder dystocia [13], meconium aspiration [13], LGA neonates [13, 17, 18], and that maternal obesity protects against having a SGA neonates [12]. These pregnancy complications that are connected with an increasing maternal BMI are associated with a greater use of health care services (inpatient and outpatient) including more prenatal tests, more obstetric ultrasound examinations, more prenatal visits with physicians, more medications dispensed, more telephone calls to providers and longer hospital stays [20].

In our study, excessive weight gain prior to 28 weeks was identified as an additional factor in the identification of pregnancies with peripartum complications. In women with normal pre-pregnancy BMI, a maternal weight gain of greater than 35 lb, has been linked with a decreased risk of a SGA neonate, and an increased risk for preeclampsia, failed induction, cesarean delivery, and LGA neonate [21].

In the evaluation of risk factors for pregnancy and perinatal complications, we observed that by combining pre-pregnancy BMI with maternal demographics and weight gain until 28 weeks, additional peripartum complications were identified that were not recognized with the BMI alone. Patients with BMI of at least 32.5 kg/m2 or nulliparous women with BMI of 25–32.5 kg/m2 who gain 28 lb by 28 weeks, are at a much higher risk for induction of labor, cesarean delivery, wound infection and shoulder dystocia, endometritis and low Apgar score <4 at 5 min. Previously Chen et al. reported that compared to newborns of normal weight, the risk of neonates with Apgar of 4–6 at 5 min is significantly higher among obese (BMI ≥30 kg/m2) parturients. However, even with a sample size of over 58,000 women and their newborns, the investigators could not validate the correlation of an increasing BMI and a low Apgar score of 0–3. Using our definition of a high-risk pregnancy, it is feasible to identify neonates with low Apgar scores (0–3) at 5 min. We acknowledge that our definition of a high-risk pregnancy that utilizes BMI, parity, and weight gain needs to be validated in other populations. This study provides guidance to health care practitioners who practice in hospitals which might not have access to additional resources (personnel and equipment) to mange complicated pregnancies or their neonates in order that a timely referral to a tertiary center could be undertaken.

There are limitations of this study that should be acknowledged. Though we evaluated known complications linked with obesity in pregnancy, we did not evaluate other potential confounding variables that may contribute to postpartum complications. Our cohorts were from tertiary centers and the findings may not be applicable to non-tertiary centers. The strengths of this study is that the information was prospectively collected following delivery, involved a large sample size (4,500 women), and comprehensively evaluated maternal characteristic(s) along with an increasing pre-pregnancy BMI that impact pregnancy outcomes. Additional studies are needed to demonstrate the obstetric interventions that can decrease the morbidity associated with obesity.

In summary, this study suggest that antepartum, intrapartum, and perinatal complications linked with an increasing BMI and that compared to BMI alone, the combination of BMI, parity, and weight gain during pregnancy provides additional information about peripartum complications. The predictive power of this new classification (BMI, parity, and weight gain) needs to be cross-validated in future studies before it is ready for widespread clinical application.

Conflict of interest

We declare that we have no conflict of interest.

Copyright information

© Springer-Verlag 2010