Table H.4.1GRADE findings for risks of planned CS compared with planned vaginal birth for women with an uncomplicated pregnancy and no previous CS (maternal outcomes)

Quality assessmentSummary of findings
Number of womenEffect
No. of studiesDesignLimitationsInconsistencyIndirectnessImprecisionOther considerations – Percentage of unplanned CS in planned vaginal birth groupPlanned CSPlanned vaginal birthRelative (95% CI)AbsoluteQuality
Maternal death
1 study

(Deneux-Tharaux et al., 2006)
observational studyserious1no serious inconsistencyserious1no serious imprecisionOf maternal deaths occurring in the planned vaginal birth group 13/49 (26.5%) were women who gave birth by unplanned CS9/737*

(cases/controls)
49/9133*
(cases/controls)
OR 2.28*

(1.11 to 4.65)
Not calculable (NC)Very low
1 study

(Dahlgren et al. 2009)
observational studyno serious limitationsno serious inconsistencyserious2,3no serious imprecision5580/38021

(14.7%)
0/1046

(0%)
0/38021

(0%)
NCNCVery Low
1 study

(Liu et al., 2007)
observational studyno serious limitationsno serious inconsistencyserious2no serious imprecision187,978/2,292,420

(8.2%)
0/4676641/2,292,420

(0.02 per 1000)
NC1.8 fewer per 100,000 birth*

(from 2 fewer to 6 more)
Very Low
Perineal and abdominal pain (during birth) (range of scores: 0-10; Better indicated by lower values)
1 study

(Schindl et al., 2003)
observational studyserious4,5no serious inconsistencyno serious indirectnessno serious imprecision93/903

(10.3%)
Median score = 1.0*

(n = 44)
Median score = 7.3*

(n = 903)
Significantly lower for planned CSDifference 6.3 lower*Very Low
Perineal and abdominal pain (3 days postpartum) (range of scores: 0-10; Better indicated by lower values)
1 study

(Schindl et al., 2003)
observational studyserious5no serious inconsistencyno serious indirectnessno serious imprecision93/903

(10.3%)
Median score = 4.5*Median score = 5.2*Significantly lower for planned CSDifference 0.7 lower*Very Low
Perineal and abdominal pain (4 months postpartum) (range of scores: 0-10; Better indicated by lower values)
1 study

(Schindl et al., 2003)
observational studyserious5,7no serious inconsistencyno serious indirectnessno serious imprecision93/903

(10.3%)
Median score = 0.0*Median score = 0.17*No significant difference between groupsDifference 0.17 lower*Very Low
Injury to bladder/ureter
1 study

(Dahlgren et al., 2009)
observational studyno serious limitationsno serious inconsistencyserious2,3no serious imprecision5580/38021

(14.7%)
0/1046

(0%)
53/38021

(0.14%)
NC1 fewer per 1000*

(from 2 fewer to 2 more)
Very Low
Injury to cervix
1 study

(Dahlgren et al., 2009)
observational studyno serious limitationsno serious inconsistencyserious2,3no serious imprecision5580/38021

(14.7%)
0/1046

(0%)
108/38021

(0.28%)
NC3 fewer per 1000*

(from 3 fewer to 1 more)
Very Low
Injury to vagina
1 study

(Dahlgren et al., 2009)
observational studyno serious limitationsno serious inconsistencyserious2,3no serious imprecision5580/38021

(14.7%)
0/1046

(0%)
213/38021

(0.56%)
NC6 fewer per 1000*

(from 6 fewer to 2 fewer)
Very Low
Iatrogenic surgical injury
1 study

(Dahlgren et al., 2009)
observational studyno serious limitationsno serious inconsistencyserious2,3no serious imprecision5580/38021

(14.7%)
0/1046

(0%)
27/38021

(0.07%)
NC7 fewer per 10,000

(from 10 fewer to 30 more)*
Very Low
Hysterectomy
1 study

(Geller et al., 2010a)
observational studyno serious limitationsno serious inconsistencyno serious indirectnessno serious imprecision1340/3868

(35%)
0.6%0.1%p = 0.135 more per 1000Low
1 study

(Dahlgren et al., 2009)
observational studyno serious limitationsno serious inconsistencyserious2,3serious65580/38021

(14.7%)
1/1046

(0.1%)
4/38021

(0.01%)
RR 9.09*

(1.36 to 60.33)
1 more per 1000*

(from 0 more to 5 more)
Very Low
1 study

(Liu et al., 2007)
observational studyno serious limitationsno serious inconsistencyserious2no serious imprecision187,978/2,292,420

(8.2%)
27/46,766*

(0.6 per 1000)
367/2,292,420*

(0.2 per 1000)
RR 3.60*

(2.44 to 5.31)
41 more per 100,000*

(from 23.6 more to 68 more)
Very Low
Hysterectomy due to postpartum haemorrhage
1 study

(Liu et al., 2007)
observational studyno serious limitationsno serious inconsistencyserious2no serious imprecision187,978/2,292,420

(8.2%)
12/46766

(0.3 per 1000)
254/2,292,420

(0.1 per 1000)
RR 2.31*

(1.30 to 4.09)
14 more per 100,000

(from 3 more to 33 more)
Very Low
Deep vein thrombosis
1 study

(Dahlgren et al., 2009)
observational studyno serious limitationsno serious inconsistencyserious2,3no serious imprecision5580/38021

(14.7%)
0/1046

(0.0%)
3/38021

(0.007%)
NC0.7 fewer per 1000*

(from 0.2 fewer to 4 more)
Very Low
1 study

(Liu et al., 2007)
observational studyno serious limitationsno serious inconsistencyserious2no serious imprecision187,978/2,292,420

(8.2%)
28/46,766

(0.6 per 1000)
623/2,292,420

(0.3 per 1000)
RR 2.20*

(1.51 to 3.20)
32 more per 100,000*

(from 14 more to 59 more)
Very Low
Pulmonary embolism
1 study

(Dahlgren et al., 2009)
observational studyno serious limitationsno serious inconsistencyserious2,3no serious imprecision5580/38021

(14.7%)
0/1046

(0%)
1/38021

(0.002%)
NC0.2 fewer per 1000*

(from 0.1 fewer to 4 more)
Very Low
Blood transfusion
1 study

(Geller et al., 2010a)
observational studyno serious limitationsno serious inconsistencyno serious indirectnessserious61340/3868

(35%)
3/180

(1.7%)
74/3868

(1.9%)
OR 0.87

(0.27 to 2.78)
2 fewer per 1000

(from 14 fewer to 34 more)
Very Low
1 study

(Dahlgren et al., 2009)
observational studyno serious limitationsno serious inconsistencyserious2,3serious65580/38021

(14.7%)
3/1046

(0.3%)
123/38021

(0.3%)
RR 0.89*

(0.20 to 3.99)
0 fewer per 1000*

(from 2 fewer to 5 more)
Very Low
1 study

(Allen et al., 2006)
observational studyno serious limitationsno serious inconsistencyserious3serious61480/17714

(8.3%)
2/721

(0.3%)
73/17714

(0.4%)
RR 0.7*

(0.2 to 2.7)
1 fewer per 1000*

(from 2 fewer to 5 more)
Very Low
1 study

(Liu et al., 2007)
observational studyno serious limitationsno serious inconsistencyserious2no serious imprecision187,978/2,292,420

(8.2%)
11/46,766

(0.2 per 1000)
1500/2,292,420

(0.7 per 1000)
RR 0.20*
(0.20 to 0.64)
41 fewer per 100,000*

(from 53 fewer to 23 fewer)
Very Low
Early postpartum haemorrhage
1 study

(Geller et al., 2010a)
observational studyno serious limitationsno serious inconsistencyno serious indirectnessno serious imprecision1340/3868

(35%)
2/180

(1.1%)
231/3868

(6.0%)
OR 0.23**

(0.06 to 0.94)
49 fewer per 1000

(from 4 fewer to 56 fewer)
Low
1 study

(Allen et al., 2006)
observational studyno serious limitationsno serious inconsistencyserious3no serious imprecision1480/17714

(8.3%)
28/721

(3.9%)
1098/17714

(6.2%)
RR 0.06*

(0.4 to 0.9)
23 fewer per 1000*

(from 35 fewer to 6 fewer)
Very Low
Infection (wound and postpartum)
1 study

(Dahlgren et al., 2009)
observational studyno serious limitationsno serious inconsistencyserious2,3serious65580/38021

(14.7%)
11/1046

(1.1%)
293/38021

(0.8%)
RR 1.36*

(0.75 to 2.4)
3 more per 1000*

(from 2 fewer to 11 more)
Very Low
1 study

(Liu et al., 2007)
observational studyno serious limitationsno serious inconsistencyserious2no serious imprecision187,978/2,292,420

(8.2%)
281/46766

(6.0 per 1000)
4833/2,292,420

(2.1 per 1000)
RR 2.85*

(2.52 to 3.21)
390 more per 100,000*

(from 323 more to 464 more)
Very Low
Infection (wound)
1 study

(Geller et al., 2010a)
observational studyno serious limitationsno serious inconsistencyno serious indirectnessno serious imprecision1340/3868

(35%)
0.01%0.00%p = 1.01 more per 10,000Low
1 study

(Allen et al., 2006)
observational studyno serious limitationsno serious inconsistencyserious3serious61480/17714

(8.3%)
11/721

(1.5%)
157/17714

(0.9%)
RR 1.7*

(0.9 to 3.2)
6 more per 1000*

(from 1 fewer to 19 more)
Very Low
Anaesthetic complication
1 study

(Dahlgren et al., 2009)
observational studyno serious limitationsno serious inconsistencyserious2,3serious65580/38021

(14.7%)
4/1046

(0.4%)
117/38021

(0.3%)
RR 1.24*

(0.34 to 4.59)
1 more per 1000*

(from 2 fewer to 11 more)
Very Low
1 study

(Liu et al., 2007)
observational studyno serious limitationsno serious inconsistencyserious2no serious imprecision187,978/2,292,420

(8.2%)
247/46,766

(5.3 per 1000)
4793/2,292,420

(2.1 per 1000)
RR 2.5*

(2.22 to 2.86)
319 more per 100,000*

(from 257 more to 389 more)
Very Low
Intraoperative trauma
1 study

(Allen et al., 2006)
observational studyno serious limitationsno serious inconsistencyserious3Serious61480/17714

(8.3%)
1/721

(0.1%)
51/17714

(0.3%)
RR 0.5

(0.1 to 3.5)
1 fewer per 1000

(from 3 fewer to 7 more)
Very Low
Uterine rupture
1 study

(Liu et al., 2007)
observational studyno serious limitationsno serious inconsistencyserious2no serious imprecision187978/2,292,420

(8.2%)
7/46,766

(0.2 per 1000)
660/2,292,420

(0.3 per 1000)
RR 0.51*

(0.25 to 1.07)
13 fewer per 100,000*

(from 22 fewer to 2.2 more)
Very Low
Length of hospital stay (mean/days)
1 study

(Geller et al., 2010a)
observational studyno serious limitationsno serious inconsistencyno serious indirectnessno serious imprecision1340/3868

(35%)
3.2

(SD 0.7)
2.6

(SD 1.1)
Mean difference 1.58

(1.27 to 2.17)
Absolute mean difference 0.6 days longerLow
1 study

(Liu et al., 2007)
observational studyno serious limitationsno serious inconsistencyserious2no serious imprecision187,978/2,292,420

(8.2%)
3.96 days

(SD 1.36)
2.56 days

(SD 1.36)
Adjusted mean difference 1.47

(1.46 to 1.49)
Absolute mean difference 1.4 days longerVery Low
Assisted ventilation or intubations
1 study

(Liu et al., 2007)
observational studyno serious limitationsno serious inconsistencyserious2serious6187978/2,292,420

(8.2%)
6/46766

(0.1 per 1000)
133/2,292,420

(0.05 per 1000)
RR 2.21*

(0.99 to 4.90)
7 more per 100,000*

(from 0 fewer to 22 more)
Very Low
Acute renal failure
1 study

(Liu et al., 2007)
observational studyno serious limitationsno serious inconsistencyserious2serious6187978/2,292,420

(8.2%)
2/46,766

(0.04 per 1000)
45/2,292,420

(0.01 per 1000)
RR 2.17*

(0.58 to 8.14)
2 more per 100,000*

(from 9 fewer to 13 more)
Very Low
Cardiac arrest
1 study

(Liu et al., 2007)
observational studyno serious limitationsno serious inconsistencyserious2no serious imprecision187978/2,292,420

(8.2%)
89/46,766

(1.9 per 1000)
887/2,292,420

(0.3 per 1000)
RR 4.91*

(3.95 to 6.11)
151 more per 100,000

(from 115 more to 195 more)
Very Low
Obstetric shock
1 study

(Liu et al., 2007)
observational studyno serious limitationsno serious inconsistencyserious2no serious imprecision187978/2,292,420

(8.2%)
3/46,766

(0.06 per 1000)
435/2,292,420

(0.18 per 1000)
RR 0.33

(0.11 to 0.99)
12 fewer per 100,000*

(from 17 fewer to 0.1 fewer)
Very Low
1

Potential bias in selection criteria. Women with life threatening morbidity were excluded from cases but this information was not always available for controls. Cases are selected over a 5 year period; controls are selected from one week within that 5 year period.

2

Some indirectness regarding the population - all participants in planned CS group had breech presentation.

3

Women undergoing induction of labour were excluded

4

Not clear if the tools were validated

5

The statistical analysis poorly presented (as line graphs)

6

95% confidence interval includes both no effect and the upper or lower confidence limit crosses the minimal important difference (MID) of 25% in either direction

7

The questionnaires response rate was low (23.9%)

*

Calculated by NCC–WCH technical team

**

OR adjusted for age, race, gestational age and prolonged rupture of membranes

From: Appendix H, GRADE tables

Cover of Caesarean Section
Caesarean Section.
NICE Clinical Guidelines, No. 132.
National Collaborating Centre for Women's and Children's Health (UK).
London: RCOG Press; 2011 Nov.
Copyright © 2011, National Collaborating Centre for Women's and Children's Health.

No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK [www.cla.co.uk]. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.