Archives of Gynecology and Obstetrics

, Volume 283, Issue 5, pp 981–988

Obesity decreases the chance to deliver spontaneously

Authors

    • Division of Perinatal Medicine, Department of Obstetrics and GynaecologySchleswig-Holstein University
  • Feriel Amari
    • Division of Perinatal Medicine, Department of Obstetrics and GynaecologySchleswig-Holstein University
  • Dörte W. Lüdders
    • Division of Perinatal Medicine, Department of Obstetrics and GynaecologySchleswig-Holstein University
  • Klaus Diedrich
    • Division of Perinatal Medicine, Department of Obstetrics and GynaecologySchleswig-Holstein University
  • Jan Weichert
    • Division of Perinatal Medicine, Department of Obstetrics and GynaecologySchleswig-Holstein University
Materno-fetal Medicine

DOI: 10.1007/s00404-010-1502-5

Cite this article as:
Beyer, D.A., Amari, F., Lüdders, D.W. et al. Arch Gynecol Obstet (2011) 283: 981. doi:10.1007/s00404-010-1502-5

Abstract

Purpose

To evaluate the impact of maternal obesity on labour, intrapartual assessment and delivery.

Methods

Retrospective cohort analysis of n = 11,681 deliveries supervised between 01 January 2000 and 31 December 2009. Results were analysed dividing the patients into two main groups according to their body mass index (BMI): group 1, control: BMI 18–24.9 and group 2 BMI, test >25. Subgroups were built: (0) BMI 25–29.9, (I) BMI 30–34.9, (II) BMI 35–39.9, (III) BMI >40. Exclusion criteria were defined as: delivery <37 + 0 weeks p.m., multiple pregnancy, comorbidity other than GDM, abnormal presentation, BMI <18.5, and incomplete data. The main outcome parameter was defined as secondary caesarean delivery rate and mode of delivery.

Results

N = 8,379 patients met the inclusion criteria and were divided in two groups: 1, n = 4,464 patients and 2, n = 3,915. Basic maternal characteristics including foetal vital parameters were equal in all groups. GDM occurred more frequently in obese patients (P < 0.001). For the main outcome parameter a significant decrease in the rate of spontaneous delivery between control/test groups (72–66%, P < 0.001) and control/I–III groups (72 vs. 50%, P < 0.001) could be observed. The rate of secondary c-section increased significantly according to a higher BMI (>40: OR 2.5, 95% CI 1.84–3.61, χ2P < 0.001). The groups showed no difference in the rate of injuries during delivery though foetal birth weight increased significantly with a higher BMI (3,412–3,681 g; P < 0.001).

Conclusion

Obesity decreases the chance to deliver spontaneously. Moreover, the obese patient suffers from a significantly longer trial of labour (7.9 vs. 9.5 h) and an elevated risk of surgical delivery.

Keywords

ObesityLabourMode of deliveryMaternal injuryFoetal outcome

Abbreviations

BMI

Body mass index

CI

Confidence interval

RRR

Relative risk reduction

OR

Odds ratio

GDM

Gestational diabetes mellitus

Introduction

Obesity in Europe has reached epidemic proportions. Current reports of the World Health Organization claim that 20% of the European population suffer from obesity [1, 2]. In the United States its prevalence has reached dramatic proportions. The National Health and Nutrition Examination Survey (NHANES) for 1999–2002 found approximately one-third of adult women in the US were obese [3]. Obesity is not only unnecessary and avoidable, but is also the root of many concomitant diseases, as shown in a published international meta-analysis [4, 5, 22]. Non-insulin dependent (type 2) diabetes, coronary heart disease, hypertension and stroke present only few of many possible medical conditions. Today we have learned that social as well as economical status combined with a restrictive attitude towards any kind of physical exercise may play an important role in its development. In addition, a specific genetic predisposition may contribute to its rise [6].

Modern obstetrics are challenged by obesity due to its higher rate of morbidity during pregnancy, labour and in childbed. Obesity also represents a problem in modern obstetric ultrasound diagnosis [7]. Published research is controversial concerning its risk evaluation. This may be explained by metrological differences and heterogeneous sample collectives. In a prospective and multicentre study of more than 16,000 patients, BMI 30–34.9 and BMI 35–39.9 obesity was associated with an increased risk of gestational hypertension, pre-eclampsia, gestational diabetes mellitus and foetal macrosomia when compared with a group of patients with a BMI less than 30 [8].

The association between maternal obesity, excessive gestational weight gain and increased foetal birth weight in non-diabetic women has been well described [912]. In addition, Ben-Haroush et al. [13] identified maternal obesity as major risk factor for large-for-gestational-age infants in pregnancies complicated by gestational diabetes.

The aim of this study was to evaluate the relative contribution of maternal weight on the mode of delivery, especially the secondary caesarean delivery rate.

Methods

The current study is a retrospective cohort analysis of all patients who attended the delivery unit of Schleswig–Holstein University Hospital in Lübeck between 01 January 2000 and 31 December 2009. The university centre for maternal and foetal health serves as a reference centre for the German Federal Land of Schleswig–Holstein, the western part of Mecklenburg–Vorpommern and a great part of the northern German coastal region with more than 4 million inhabitants. Both, Schleswig–Holstein and Mecklenburg–Vorpommern can be characterized as rural in the majority, with a predominantly Caucasian population.

Exclusion criteria were defined as: preterm delivery <37 + 0 weeks of gestation p.m., confirmed multiple pregnancy, maternal and foetal comorbidity except for gestational diabetes mellitus (GDM), presentation other than cephalic presentation, history of multiple abdominal operation except for caesarean delivery and repeat caesarean delivery, maternal BMI <18.5 and incomplete data.

All eligible patients were divided into two major groups according to their body mass index. Group 1, called the “control” group, was characterized by body mass index 18.5–24.9 whereas group 2, the “test” group, was defined as BMI >25. We then divided group 2 according to the degree of obesity: 0, BMI 25–29.9; I, BMI 30–34.9; II, BMI 35–39.9 and III, BMI >40.

Primary outcome was defined as mode of delivery focussing on the secondary caesarean delivery rate.

Time of labour, the degree of maternal injury and foetal outcome was further analysed. Foetal outcome was identified by either a specially trained obstetrician or paediatrist from the neonatal intensive care unit of the University hospital.

Every delivery was accompanied by a midwife and a specially trained obstetrician who took responsibility for the course and outcome of labour.

With regard to the mode of delivery, it differed between spontaneous, vaginal operative and caesarean delivery. Vaginal operative delivery included forceps and vacuum extraction and was specified according to the foetal position which could be situated in mid or outlet position. Caesarean delivery was analysed according to primary and secondary caesarean section. Additionally, repeat or re-repeat caesarean deliveries were documented.

Injuries were divided according to the extent of vaginal, perineal and cervical lacerations. Vaginal lacerations were defined as injuries of the vagina, infiltrating deeply into the mucosa and reaching more than 2 cm over the surface. Cervical lesions were recorded on their clinical relevance: non-bleeding small tears were excluded. Perineal lacerations were defined as: first degree, superficial tear which involves the vaginal mucosa and/or perineal skin; second degree, extends into the fascia and muscles that surround the vagina; third degree, extends into or through the external anal sphincter muscle and fourth degree, extends into anorectal lumen and thus involved disruption of both the external and internal anal sphincter.

All data were extracted from the electronic database of our centre [PIA Fetal Database™ (GE, USA)].

Patients’ characteristics were analysed according to maternal and foetal data.

Statistical analysis was performed using SPSS statistical package version 17.0 for windows. Analysis included Mann–Whitney test for continuous data, χ2 test for categorical data and correlation studies. Differences were considered significant when P < 0. 05.

Variables were checked for best fit using the Hosmer–Lemeshow test. Multiple logistic regression analysis was performed in order to construct a model to assess the independence of the influence of body mass index X1, induction of labour X2, maternal age X3, parity X4, previous c-section rate X5 and softening X6 on the primary outcome Y = secondary caesarean delivery in question.

Adjusted odds ratios (PRs) were determined by using Mantel–Haenszel technique. Estimates of 95% confidence intervals (95% CI) were made with a test-based method, based on the Mantel–Haenszel χ2 [14].

Results

A total of n = 11,681 patients and deliveries were analysed. All in all, n = 8,379 deliveries met the criteria for eligibility and could be divided in groups as described above. Figure 1 demonstrates the patients’ flow chart of the trial. The comparison of maternal basic characteristics between the control and test groups showed no difference except for a significantly higher rate of gestational diabetes mellitus and BMI in all test groups (GDM 3.2 vs. 7.8% and BMI 22.1 vs. 30.2; P < 0.001), see also Table 1. The comparison of duration of labour until delivery showed no difference for both main groups 1 + 2. However, a significant prolongation of labour trial was observed comparing control group with the different obese groups (I–III). The time frame reached from 7.9 h in the control group to 9.5 h in the group of patients with a BMI >40 (P < 0.001). Gestational weeks until delivery were equal for all groups: 39 weeks and 2 days, respectively.
https://static-content.springer.com/image/art%3A10.1007%2Fs00404-010-1502-5/MediaObjects/404_2010_1502_Fig1_HTML.gif
Fig. 1

Patient flow chart

Table 1

Patients’ characteristics, mean values, standard deviation (SD) and percentages

 

BMI 18–24.9 (N = 4,464)

BMI >25 (N = 3,915)

BMI 25–29.9 (N = 2,309)

BMI 30–34.9 (N = 1,016)

BMI 35–39.9 (N = 407)

BMI >40 (N = 183)

Maternal

 Age (years)

30.6

30.5

30.7

30.4

30.0

30.3

 Height (cm)

167.8

167.2

167.5

166.7

167.0

166.8

 Weight (kg)

61.6

79.3

71.8

82.7

96.6

117.7

 BMI, mean (SD)

22.1 (1.6)

30.2 (1.4)

27.1 (1.4)

32 (1.5)

37 (1.4)

44.7 (4.6)

 Weight gain (kg)

15.1

14.6

15.1

14.3

13.4

13.9

 GDM (%)

145 (3.25)

304 (7.8)

118 (5.1)

100 (9.8)

53 (13.0)

33 (18.0)

 Duration of birth [h] (SD)***

7.9

7.8

7.6

7.8

8.2

9.5

Foetal

 Ratio male/female

2,215/2,246

2,015/1,836

1,193/1,112

514/448

218/185

90/91

 Birth weight [g] (SD)***

3,413 (499)

3,535 (506)

3,508 (515)

3,560 (509)

3,561 (562)

3,682 (571)

 Weight <2.500 g (%)

130 (2.9)

77 (2.0)

41 (1.8)

20 (2.0)

12 (2.9)

4 (2.2)

 Foetal length (cm)

51.9

52.4

52.3

52.5

52.4

52.9

 Head circumference (cm)

34.8

35.0

35.0

35.1

35.0

35.2

 APGAR 1

8.7

8.7

8.7

8.7

8.6

8.6

  5

9.7

9.7

9.7

9.7

9.6

9.6

  10

9.9

9.9

9.9

9.9

9.9

9.8

 Arterial cord pH

7.3

7.3

7.3

7.3

7.3

7.3

 Gestational weeks

39.3 + 3.0

39.3 + 3.0

39.3 + 3.0

39.4 + 3.1

39.2 + 3.2

39.2 + 2.7

*** P < 0.001

With regard to basic foetal parameters, no differences could be observed between the groups except for a higher birth weight according to an increased maternal body mass index (control: 3,413 g vs. test: 3,535 g). Moreover, the comparison between the control group and subgroups revealed significant differences in foetal birth weight which reached its peak in the group of patients at BMI >40 with 3,682 g compared to 3,412 g in the control group (P < 0.001). However, similar male/female newborn-rates could be seen in all groups (see Table 1).

Table 2 illustrates the results of the multiple logistic regression analysis. Body mass index, induction of labour and maternal age show P values with (P < 0.001). The mode of delivery is shown in Table 3. With regard to the rate of vaginal operative delivery, no inter-group difference could be observed. Significant differences were found in the rate of spontaneous delivery and secondary caesarean section in favour of the control group (spontaneous delivery: control 72% vs. test 66%; secondary caesarean delivery: control 12.7% vs. test 16.1%; P < 0.001). Nevertheless, the rate of secondary section increased over the groups from 12.7% in the control group to 27.3% of patients with a BMI >40 (OR 2.5, 95% CI 1.84–3.61, χ2P < 0.001), see Fig. 2 and Table 4. The relative risk reduction for secondary c-section was 31.5% in the group of patients with a BMI 25–29. 9 and climbed to 68.8% for patients with a BMI >40 (RRR in I, 43%; II, 54%). Table 4 demonstrates the different odds ratio and 95% CI according to the BMI. Primary caesarean delivery and repeat caesarean delivery showed no significant inter-group differences. Additionally, a stable rate of 34% of patients undergoing peridural anaesthetics could be seen throughout the groups (Table 3).
https://static-content.springer.com/image/art%3A10.1007%2Fs00404-010-1502-5/MediaObjects/404_2010_1502_Fig2_HTML.gif
Fig. 2

Spontaneous delivery and secondary c-section according to BMI

Table 2

Multiple logistic regression analysis

 

Coefficient (β)

Standard error

Wald χ2

P value

Odds ratio

95% CI

BMI

0.31

0.08

15.72

0.000

0.73

0.63–0.85

Induction of labour

2.04

0.19

119.77

0.000

7.71

5.35–11.12

Maternal age

0.06

0.01

79.98

0.000

1.06

1.05–1.08

Parity

−0.28

0.82

0.12

0.733

1.32

0.27–6.54

Previous c-section rate

1.48

1.25

1.41

0.236

0.23

0.02–2.62

Softening

0.00

0.31

0.00

0.992

1.00

0.54–1.85

Intercept

−0.83

0.36

5.25

0.022

0.44

 
Table 3

Mode of delivery

 

BMI 18–24.9 (N = 4,464)

BMI >25 (N = 3,915)

BMI 25–29.9 (N = 2,309)

BMI 30–34.9 (N = 1,016)

BMI 35–39.9 (N = 407)

BMI >40 (N = 183)

Spontaneous delivery, N (%)***

3,213 (72.0)

2,582 (66.0)

1,614 (69.9)

653 (64.3)

223 (54.8)

92 (50.3)

Vaginal operative delivery, N (%)

192 (4.3)

122 (3.1)

75 (4.6)

34 (3.4)

10 (2.5)

3 (1.6)

 Mid

72 (1.6)

45 (1.1)

26 (1.1)

15 (1.5)

3 (1.0)

1 (0.5)

 Outlet

120 (2.7)

77 (2.0)

49 (2.1)

19 (1.9)

7 (1.7)

2 (1.1)

Caesarean delivery, N (%)

 Primary

222 (5.0)

215 (5.5)

110 (4.8)

64 (6.3)

30 (7.4)

11 (6.0)

 Secondary***

568 (12.7)

632 (16.1)

329 (14.2)

169 (16. 6)

84 (20.6)

50 (27.3)

 Repeat

150 (5.0)

150 (3.8)

72 (5.4)

39 (6.0)

25 (9.6)

14 (8.7)

 Re-repeat

35 (0.9)

58 (1.5)

37 (1.6)

15 (1.6)

1 (0.5)

5 (3.2)

 PDA

1,538 (34.5)

1,347 (34.4)

802 (34.7)

331 (32.6)

143 (35.1)

71 (38.8)

*** P < 0,001, χ2

Table 4

Labour and delivery complications among singleton pregnancies

 

Control

BMI 18–24.9 (N = 4,464)

Test

BMI >25 (N = 3,915)

0

BMI 25–29.9 (N = 2,309)

I

BMI 30–34.9 (N = 1,016)

II

BMI 35–39.9 (N = 407)

III

BMI >40 (N = 183)

 

OR (95% CI)

P

OR (95% CI)

P

OR (95% CI)

P

OR (95% CI)

P

OR (95% CI)

P

Vaginal operative delivery

0.72 (0.57, 0.90)

0.005

0.75 (0.57, 0.98)

0.035

0.77 (0.53, 1.12)

0.168

0.56 (0.29, 1.07)

0.078

0.36 (0.12, 1.15)

0.084

Primary caesarean delivery

1,11 (0.92, 1.35)

0.287

0.97 (0.76, 1.21)

0.706

1.29 (0.96, 1.71)

0.087

1.52 (1.02, 2.26)

0.038

1.22 (0.66, 2.29)

0.529

Secondary caesarean delivery

1.32 (1.16, 1.49)

0.000

1.14 (0.99, 1.32)

0.006

1.37 (1.14, 1.65)

0.000

1.74 (1.38, 2.30)

0.000

2.58 (1.84, 3.61)

0.000

Anal sphincter laceration

0.84 (0.55, 1.29)

0.431

0.97 (0.60, 1.60)

0.899

0.88 (0.44, 1.74)

0.707

0.22 (0.03, 1.58)

0.131

0.49 (0.07, 3.53)

0.475

GDM

2.51 (2.05, 3.07)

0.000

1.60 (1.25, 2.06)

0.000

3.25 (2.50, 4.24)

0.000

4.46 (3.20, 6.22)

0.000

6.55 (4.34, 9.89)

0.000

Shoulder dystocia

1.71 (0.48, 6.07)

0.405

No case

 

5.25 (1.48, 20.61)

0.000

5.51 (1.01, 30.15)

0.000

No case

 

Stillbirth

2.28 (0.86, 6.09)

0.099

1.61 (0.49, 5.29)

0.431

2.94 (0.83, 10.43)

0.096

5.52 (1.38, 22.14)

0.016

No case

 
Lacerations were seen in all groups and showed no significant differences. First degree perineal lacerations were seen most frequently in all groups ranging from 19.3% in the test group to 21% in the control group. This was followed by vaginal lacerations which occurred in 17.3% of the cases in the control and 16.2% in the test group. Second degree perineal lacerations were seen in a frequency of 7–9% of the cases. Third degree, fourth degree and cervical lacerations were less frequent with a rate reaching up to 1%. Shoulder dystocia occurred in 4 patients (0.1%) of the control group and 6 patients (0.2%) of group 2, but did not show a significant increase. When comparing different groups a significant increase could be proven for control versus BMI 30–34.9 group with OR 5.5 (95% CI 1.2–20.6, χ2P = 0.014). In addition when the rate of stillbirths was analysed the rate ranged from 0.1% in the control group to 0.7% in group II with P = 0. 016. Table 5 illustrates the inter-group distribution of the injuries and complications during delivery.
Table 5

Injuries and complications

 

BMI 18–24.9 (N = 4,464)

Control

BMI >25 (N = 3,915) Group 2

BMI 25–29.9 (N = 2,309)

BMI 30–34.9 (N = 1,016)

BMI 35–39.9 (N = 407)

BMI >40 (N = 183)

Laceration, N (%)

 First degree

860 (19.3)

820 (20.9)

503 (21.8)

219 (21.6)

68 (16.7)

30 (16.4)

 Second degree

328 (7.3)

310 (7.9)

172 (7.4)

100 (9.8)

33 (8.1)

5 (2.7)

 Third degree

50 (1.1)

37 (0.9)

25 (1.1)

10 (1.0)

1 (0.2)

1 (0.5)

 Fourth degree

5 (0.1)

2 (0.1)

2 (0.1)

Cervix injury, N (%)

43 (1.0)

47 (1.2)

32 (1.4)

11 (1.1)

3 (0.7)

1 (0.5)

Vaginal laceration, N (%)

771 (17.3)

633 (16.2)

399 (17.3)

155 (15.3)

57 (14.0)

22 (12.0)

Shoulder dystocia, N (%)

4 (0.1)

6 (0.2)

5 (0.5)

2 (0.5)

 

Stillbirth, N (%)

6 (0.1)

12 (0.3)

5 (0.2)

4 (0.4)

3 (0.7)

– no case

Discussion

Numerous trials have been conducted focussing on different aspect of obesity in pregnancy, during labour and in childbed. Obesity during pregnancy was proven to be a risk factor for maternal and foetal health. Nevertheless, the variety of different outcome parameters of these trials hardly allows a general and superficial comparison. Additionally parameters other than obesity were found to compromise mother and child during labour and delivery. Cedergren [15] identified the ethnic origin as important risk factor for complications of obese patients during labour and delivery. In a large analytic study including 2,967 deliveries Hornemann et al. [16] identified advanced age as a risk factor for higher grade perineal lacerations in 1.7%. In a systematic review of the subject Langenveld et al. analysed 11 trials including 2,461 deliveries for delivery before 34 weeks of gestation due to maternal comorbidity. He described a pooled risk of a delivery before 34 weeks due to a recurrence of hypertension, pre-eclampsia, or HELLP was 7.8% (95% CI 6.7–9.0) [17].

The aim of this study was to identify the influence of maternal obesity on the mode of delivery focussing on the secondary c-section rate. Excluding significant confounding parameters the eligibility criteria had to be identified very carefully. This explains the reason why the study focuses on singleton and non-preterm pregnancies excluding maternal and foetal pathologic findings. Moreover, the groups showed equal basic characteristics which encouraged a critical inter-group comparison. Whether the BMI groups were examined conjointly or compared with each other, a significant increase in gestational diabetes mellitus was observed with the increase of BMI in our study. This confirms Aydin et al. [18] who evaluated data of 9,112 pregnancies. Additionally, obesity is accompanied by maternal insulin resistance and foetal hyperinsulinaemia even in the absence of maternal diabetes mellitus [19].

The coherence between foetal hyperinsulinaemia and macrosomy is well known.

The analysed cohort showed a significant increase of foetal birth weight according to a higher body mass index which could be seen throughout the groups. This was also reflected in the duration of labour which reached its peak at 9.5 h in the group of patients with a body mass index >40 (P < 0.001).

Nevertheless, this finding highlights the main question: Does obesity increase the risk of operative delivery? With regard to previously published papers, different and contradictory findings are described. In a study conducted by Jensen et al. in 1998, based on 4,258 patients, women with a BMI value between 20 and 25 were taken as a reference population and compared to obese patients. A significant correlation between the rate of caesarean section and BMI could not be determined for the different groups [20]. However, Vahratian et al. [21] reported a moderate association between maternal pre-pregnancy obesity and an unexpected caesarean delivery. Nevertheless, Credergren published in 2004 a large prospective population based cohort study including n = 3,480 women with morbid obesity (BMI >40), n = 12,698 patients with a BMI 35.1–40 and a control group with a BMI ranging from 19.8 to 26. Among other findings, Cedergren [15] could identify an increased risk for caesarean delivery (OR 2.69; 95% CI 2.49, 2.90), shoulder dystocia (OR 3.14, 95% CI 1.86, 5.31) and instrumental delivery (OR 1.34, 95% CI 1.16, 1.56) for the obese groups compared with the normal weight control group. When comparing the different trials, it soon becomes clear that different group definitions of obesity according to the body mass index severely complicate any comparison. Therefore, we adjusted using the definition of obesity published by the World Health Organization creating the groups described above [1, 22]. When answering the question of this trial, we too came to the same conclusion as Cedergren. The main groups showed a significant decrease in the rate of spontaneous delivery in the case of the heavy-weight patients and this trend was observed throughout the groups. Furthermore, the data revealed another significant finding. According to the higher BMI the rate of secondary section increased significantly up to 27% in the group of patients with a BMI >40 as shown in Table 3. When comparing previously published papers, we have to understand that there were no consistent definitions used by different authors concerning the mode of delivery. Many authors simply analyse the rate of caesarean section without differentiating between primary and secondary caesarean sections, and do not mention the possibility of repeat operative delivery [15, 18, 23]. Therefore, this study aimed at the crucial differentiation of the various delivery modes as well as possible injuries. Interestingly enough, it was not possible to detect differences in the rate of primary section. This might be explained by the fact that overweight patients, in principle, wish to experience a spontaneous delivery like many non-obese women. Unfortunately, a great part of this group ends up with a secondary section (see Table 3). Furthermore, obese patients may present a greater challenge for anaesthetic interventions during labour and delivery as different trials could show [24, 25]. However, the rate of peridural anaesthetics was stable throughout the groups at a rate of 34%.

Voldner et al. analysed the outcome of delivery of n = 535 women in 2009. He stated a significant higher injury rate in obese patients than among normal weight patients [26]. In comparison, when evaluating n = 8,379 patients we could not prove higher injury rates among the obese cohort. In addition, different degrees of perineal lacerations as well as vaginal and cervical injuries were evaluated carefully.

In conclusion, the results of this trial show a strong association between a longer length of labour and a higher rate of secondary caesarean section with an increasing body mass index. On the other hand, a decreasing spontaneous delivery rate according to increasing body mass index could be found. The rate of delivery injuries showed no significant inter-group difference. The results confirm other studies that modern obstetrics are more and more challenged by obese patients which present a high-risk group.

Conflict of interest statement

There are no conflicts of interest for any of the authors.

Copyright information

© Springer-Verlag 2010