Episiotomy and perineal tears:
cause and effect
Online posting: May 9, 1997
Published in print: August 15, 1997 CMAJ 1997;157:365
Re: Episiotomy and severe perineal trauma: of science and
fiction, by Dr. Michael E. Helewa, CMAJ
1997;156:811-3 [full text / résumé]
See response from: M.R. Helewa
I would like to react to the editorial concerning our article
"Association between median episiotomy and severe perineal
lacerations in primiparous women" (CMAJ
1997;156:797-802 [full text / résumé]). Although I appreciated the fact that Helewa
recognized the original nature of our study, I must disagree with
him about the causal relation between median episiotomy and
severe perineal lacerations.
First, the classification of research designs used by Helewa
is incorrect. There is no such thing as a cross-sectional study
when evaluating the association between episiotomy and perineal
tears. Almost all studies on the topic are true cohort studies,
meaning that women either exposed to episiotomy or not are
followed until birth to assess the incidence of perineal tears.
Although in most of these studies involve retrospective data
collection, this does not invalidate the direction of the study.
The distinction between cross-sectional and cohort studies is
important when assessing a causal relation. The temporal
principle (that cause must precede effect) can be supported by a
cohort study but not by a cross-sectional (prevalence) study.
Second, Helewa mixes apples and bananas, namely median and
mediolateral episiotomy. Most of the studies he refers to that
imply that episiotomy is not associated with severe lacerations
concern mediolateral episiotomy. In this context, his statement
is correct: these studies showed either no association or a very
small one. By contrast, he missed 9 out of 11 of the studies
cited in our article that deal specifically with median
episiotomy and severe perineal lacerations in primiparous women;
all of which show a strong association. I challenge him to find a
single study that does not show an association between median
episiotomy and third- and fourth-degree tears.
I agree that a randomized controlled trial is the best design
to demonstrate causality, but it is not the sine qua non.
Is there any randomized trial in humans showing that smoking
causes lung cancer?
Michel Labrecque, MD, MSc
Department of Family Medicine
Laval University
Quebec, Que.
michel.labrecque@mfa
.ulaval.ca
CMAJ email
CMAJ Readers' Forum