Much ado about “levator microtrauma”: Do epidurals protect the pelvic floor?

September 3rd, 2010 by avatar
Levator Ani

The levator ani is a muscle, not a clinical outcome.

Ok, ok, I’ve heard from enough of you about the study purportedly showing that epidurals protect the pelvic floor, I suppose it’s time to write up what I think about it.  First, a little about the study from the (overzealous) journal press release. I couldn’t resist doing a little bit of [line editing] on it:

Researchers from Australia undertook a prospective study of 488 women undergoing their first pregnancy between May 2005 and February 2008. The object of the study was [to determine predictors of delivery mode, but since they had enrolled all of those pregnant women they designed some other studies, too. In one, the researchers invited those 488 women to return for follow up at 3-4 months and retrospectively analyzed data on the 367 women who did] to determine if there are any risk factors during birth which may result in levator trauma including macrotrauma (large scale avulsion – tearing) and microtrauma (damage to the muscle tissues such as irreversible overdistention of the pelvic floor opening). Researchers believed that the findings from their study may help modify obstetric practice to help prevent levator injury…

No levator avulsion was recorded in the women who had a caesarean section. Levator avulsion was diagnosed in 13% of women who had a vaginal birth (9% of whom had had a vacuum delivery and 35% of whom had had a forceps delivery). Researchers found that forceps delivery was associated with a three to four-fold increase in levator avulsion. [They excluded the 13% of vaginal births in which levator avulsion was diagnosed and evaluated the rest of the women for "microtrauma". We put "microtrauma" in quotes because no one has ever defined or determined the prevalence of this "condition". The researchers invented it themselves! But anyway,...] Postbirth assessment showed that the longer the 2nd stage of labour, the higher the likelihood of microtrauma. Women who had an intrapartum epidural were found to have had a lower incidence of microtrauma. The researchers suggest that epidurals, because they relax the muscles through paralysis, may be beneficial in preventing levator trauma.

There are multiple problems with the press release and, for that matter, with the study itself. Christine Kent at Whole Woman Village Post does a nice job of reviewing some of them, including the fact that one of the study authors receives money from incontinence surgical device companies and ultrasound companies. But I’d like to focus on the use of “levator microtrauma” as the outcome reported.

Levator microtrauma is an example of a surrogate outcome (sometimes referred to as a surrogate endpoint). As defined by Temple (1995):

A surrogate endpoint of a clinical trial is a laboratory measurement or a physical sign used as a substitute for a clinically meaningful endpoint that measures directly how a patient feels, functions or survives. Changes induced by a therapy on a surrogate endpoint are expected to reflect changes in a clinically meaningful endpoint. [emphasis mine]

But as D’Agostino (2000) argues, some surrogate outcomes are extremely poor predictors of actual outcomes, and changing clinical practice based on studies that report only surrogate outcomes can be a major threat to patient safety if the therapy introduces other risks. So the questions we should ask ourselves when we see a study reporting a surrogate outcome are:

  • is the surrogate outcome a good predictor of a clinically important outcome (i.e., “how a patient feels, functions or survives”)?
  • does the treatment pose any excess risks over other alternatives to achieving that clinically important outcome?

In the case of “levator microtrauma,” there is absolutely no data whatsoever linking the author’s definition of microtrauma to pelvic organ prolapse or other important pelvic floor problems such as incontinence or sexual dysfunction. The aforementioned corporate-sponsored researcher showed in an earlier study that macrotrauma (aka levator avulsion) is an appropriate surrogate for pelvic organ prolapse, but remember that epidurals were not associated with macrotrauma in this study. Forceps deliveries were – and what’s the major modifiable risk factor for forceps delivery?  Epidurals!

But let’s say that microtrauma does lead to pelvic floor problems and that, therefore, epidural analgesia in labor may be a strategy for preventing those pelvic floor problems.  Is encouraging epidural analgesia in a woman who might otherwise forgo it the best strategy for preventing pelvic floor problems?  Maybe doing away with coached pushing, fundal pressure, episiotomy, and supine positioning might be the better strategy. Maybe postpartum exercises can help reverse changes associated with pregnancy and vaginal birth so they don’t turn into symptomatic pelvic floor problems.

Like other studies that report surrogate outcomes, this study is not useless.  It provides some data that can help us understand how epidurals affect pelvic floor muscle tone and strength and alter the process of vaginal birth, and I’ll be the first to say that we need more research in those areas. In other words, these data on a novel surrogate outcome can help us design more studies, not guide patient care.

Anyone who has even a basic understanding of clinical research should recognize that we need much more data before we can say that epidurals may help prevent future pelvic floor prolapse.  Oh wait, the editor-in-chief of a major international obstetric journal just said exactly that! In the headline of a major press release!



Temple RJ. A regulatory authority’s opinion about surrogate endpoints. Clinical Measurement in Drug Evaluation. Edited by Nimmo WS, Tucker GT. New York: Wiley; 1995.

Uncategorized , , , , ,

  1. September 3rd, 2010 at 12:52 | #1

    Thanks for this! I had three friends link to that study from facebook saying “See? epidural’s AREN’T bad!” I agree with your *headdesk* sentiment. The study defies all logic.

  2. avatar
    Louise Roth
    September 3rd, 2010 at 13:52 | #2

    The whole premise of this finding seems wrong on its face. Isn’t immobility and the supine position for the second stage the biggest contributor (after forceps, of course) to pelvic floor damage? And don’t epidurals nearly always result in immobility and supine delivery? Of course, hospitals do have other ways to get women to lay on their backs in bed, but it seems that the relevant comparison would be pushing position.

  3. September 3rd, 2010 at 13:52 | #3

    Once again you’ve utterly misrepresented a study. There are so many things wrong with your assessment that it’s difficult to know where to begin.

    Let’s start with the most egregious. You claim:

    “there is absolutely no data whatsoever linking the author’s definition of microtrauma to pelvic organ prolapse …”

    That is flat out false.

    The authors were attempting to determine why vaginal delivery increases the risk of pelvic organ prolapse and related symptoms in older women. Some women sustain visible damage to the levator ani muscles (macrotrauma) resulting in a gap between the muscles that the pelvic organs can fall through. However, many women who have no evidence of mactrotrauma go on to develop pelvic organ prolapse. The mechanism seems to be damage to the substance of the muscle itself (microtrauma) that leads to weakening and stretching of the muscles. That can also lead to widening the gap between the muscles, a weakening that may only be noticeable when the intraabdominal pressure is increased, such as when the woman coughs or sneezes.

    Muscle microtrauma can occur in any muscle, is widely mentioned in the scientific literature and has been mentioned in connection with the levator ani muscles in previous scientific papers. No one knows exactly how the levator ani muscles weaken in the aftermath of childbirth, but microtrauma is a very plausible explanation.

    Levator ani microtrauma is not a surrogate outcome since the study was designed specifically to look at all possible ways that the levator may be damaged by vaginal delivery.

    The authors then looked at the factors that seemed to be associated with macro and micro trauma to the levator muscles. Operative vaginal delivery seemed to be the biggest risk factor. The authors noted in passing that epidural anesthesia appeared to be protective. It was a tangential observation, mentioned along with all the other associations.

    The study is not about epidurals, has nothing to do with epidurals and makes no specific claims about epidurals. In your zeal to dismiss the possibility that epidurals might have any beneficial effects, you’ve trashed a study that mentioned epidurals only in passing. You’ve made accusations that are completely false. Levator microtrauma is real, accepted and potentially involved in pelvic organ prolapse. Microtrauma is not a surrogate endpoint, but a real and meaningful endpoint.

  4. September 3rd, 2010 at 14:00 | #4

    The gap between the study and the headlines was ridiculous. Thanks for clearing it up, Romano.

  5. avatar
    Christie B
    September 3rd, 2010 at 14:13 | #5

    Thanks for this analysis! Urinary incontinence is a scary thing so stories like this are powerful.
    I think the best (recent) historical example of misuse of unvalidated surrogate endpoints guiding clinical practice is in the Term Breech Trial, where the surrogate endpoints seemed to indicate additional neonatal morbidity with planned attempts at vaginal birth. However, follow-up with the actual clinical indicators showed no difference. However, clinical practice changed significantly and 10 years of experience and training with vaginal breech delivery was lost in the US.

  6. September 4th, 2010 at 10:16 | #6

    I’m the senior author of the study discussed by you. After 25 years of research in this field it still depresses me how excited people get when it comes to research that may affect the choices made by women in childbirth. There is way too much ideology and zealotry out there for a rational discussion. Amy Romano, you seem to intuitively know what’s right- saying: “Maybe doing away with coached pushing, fundal pressure, episiotomy, and supine positioning might be the better strategy. ” How do you know? Where is the data?

    And how do you know what ‘pelvic floor damage’ is? By all means do check my website if you really want to know:


    It seems you’re interested in those issues, and good on you for that. I’d be happy to answer any questions you may have, and I promise not to be prejudiced in any way. We all want the same: healthy mums and healthy babies. Just try and avoid the zealotry please.

    And good on you for being suspicious about industry links. I’ve written about these issues (Dietz HP. Bias in research and conflict of interest: why should we care? Int Urogynecol J 2007; 18 (3): 241-243) and share your concerns. However, this study was not sponsored by industry, and I assure you that I do not currently receive any money or other benefits from any device or pharmaceutical company.

    And thanks, Amy Tuteur, for your contribution. I agree with everything you said.

    All the best

    Prof HP Dietz MD PhD

  7. September 4th, 2010 at 14:46 | #7

    What an honor to have the author of the study write to you. That is pretty cool. I respect that you left his and Dr Tuteur’s comments and criticism up. I am hoping you make a statement correcting your representation of the study so that we don’t spread false information since all of us are simply promoting science and patient safety through patient education. You have a large audience and I would not want them to get the wrong idea. I like this paper because it was simply thought provoking. I do a lot of pelvic reconstruction surgery. More papers like this looking at the basic science underlying pelvic floor damage will help us improve our understanding and surgical techniques in the future. That is how I will apply the paper in practice. It didn’t really speak to me about labor management.

  8. September 4th, 2010 at 15:07 | #8

    It’s always an honor when researchers come to this blog to discuss their research, so thank you Dr. Dietz. As I said in this post, my issue was more with the BJOG press release and related media attention than with the study itself. I think the study is useful, but that the headline in the press release that epidurals may be protective is premature based on this data. I assume you would agree with that statement. I worry, too, that since we know epidurals are associated with operative vaginal deliveries and prolonged second stage and appear to be associated with OP fetal position, too, and all of these are associated with pelvic floor problems, recommending epidurals as a strategy to reduce pelvic floor trauma based on this study may actually turn out to do more harm to the pelvic floor than good.

    Where are the data on the alternative strategies I suggested? Fundal pressure is rarely documented and when it is long-term studies reporting its effect are unlikely, so we don’t have much data. But we know it is highly associated with anal sphincter injury, especially when used in combination with episiotomy. See the link in my post for more info on that. For coached pushing, see Schaffer’s work http://www.ncbi.nlm.nih.gov/pubmed/15902179. It also relies on surrogate outcomes, but the mechanism whereby a longer duration of more forceful valsalva pushing might lead to muscle injury and weakness is pretty clear (which stregthens the possibility that a surrogate outcome may be a good predictor of clinical outcomes). As for supine positioning, I don’t believe there is any data that supports or refutes an association with pelvic floor strength and function, because maternal position during second stage and the moment of birth itself isn’t reliably or consistently documented and long-term studies that look at outcomes (surrogate or clinical) relative to birth position are nonexistent as far as I know. The Cochrane reviewers who looked at position in the second stage in women without epidurals found no data on long-term outcomes but reported a small decrease in instrumental vaginal births as well as a reduction in second stage duration and use of episiotomy. Seems plausible that there may be a protective effect, then, on the pelvic floor. Like I said, though, I don’t know for sure, which is why I said “maybe”.

    Another issue about advising epidurals to protect the pelvic floors is that there are many different types of epidurals, different ways of managing labors with epidurals, and lots of other covariables. Like you said in your paper, you have no data on the duration of active pushing versus laboring down in the women in your study (do you know if laboring down is the usual practice in the setting?). There are still many places where women are valsalva pushing from the moment they’re 10 centimeters with an epidural and instrumental delivery is done routinely if the baby isn’t out after 2 hours. In that setting is an epidural good for the pelvic floor? Probably not, but for a woman (or her provider, anestesiologist, etc) who has only read the headlines or the study abstract, they may well conclude the opposite.

    Yes, I care about pelvic floor outcomes in women and like I said I think this data may prove useful for understanding pelvic floor issues and for that matter epidurals – especially if other research follows up on some of clues and inferences in your data. But I honestly think the media attention, which started with the BJOG release, should be considered malpractice.

    As for Dr. Tuteur’s assertion that microtrauma is not a surrogate outcome, I stand by my assertion that it is, using the definition that tells us that surrogate endpoints are substitutes for “how a patient feels, functions, or survives.” And the idea that epidurals were mentioned only “in passing” in the study is absurd. Why was it in the abstract, keywords, 12 times in the manuscript itself including a whole paragraph devoted to it in the discussion, in the headline of the journal press release, and in the headline of countless media stories? Far more women have read or heard in the past week that epidurals protect the pelvic floor than will ever read this post or any other criticism of the study or the media management of it. I saw a comment on a newspaper web article yesterday from a first time mom saying she is fearful of giving birth and ending up peeing in her pants and now after reading the story is more inclined to get an epidural. Should she not be exposed to another point of view about the strength and usefulness of that data, or told of other low- to no-risk strategies that *might* (or in some cases *will*) protect her pelvic floor?

  9. September 4th, 2010 at 15:26 | #9

    @Jeff Livingston, MD
    Thanks for your comment. I’m not sure what you think I misrepresented. I stated, “there is absolutely no data whatsoever linking the author’s definition of microtrauma to pelvic organ prolapse or other important pelvic floor problems such as incontinence or sexual dysfunction” This is relevant to how useful a surrogate endpoint microtrauma will turn out to be. I think you and I see this similarly – that this was a useful study for showing the prevalence of this type of muscle injury and proposing a definition based on serial ultrasound measurements that can be used to diagnose microtrauma. It seems the next study should be to use that definition of microtrauma and see if it correlates with clinical outcomes such as pelvic organ prolapse, incontinence, etc. As far as I can tell, no one has done that, which is why I made that statement. Another way it might be useful is to measure the effects of treatments for muscle injury (not my area of expertise.)

    If it’s another thing in my post that you want me to clear up, let me know. But as of now I stand by everything I wrote.

  10. avatar
    September 4th, 2010 at 15:30 | #10

    To anyone with a background in normal physiological birth, studies like this are just depressing because it is not that they are comparing “normal” birth to birth with an epidural… they are comparing normal *interventive and managed birth* in a hospital setting, with normal interventive and managed birth with an epidural. To me all this study suggests is that when birthing in a managed hospital scenario, it is possible that an epidural may result in less damage overall.

    Call me when there is a study comparing epidural use to self-directed natural pushing in woman-led positions, and I’ll get excited.

  11. September 4th, 2010 at 15:51 | #11

    A key characteristic of science, as opposed to pseudoscience, is that all possible outcomes are allowed. That’s why “intelligent design” is not science; there is no evidence that would lead it’s advocates to announce that there is no “intelligent designer.” The conclusion is predetermined.

    This analysis of a scientific paper veers perilously close to pseudosciece. It starts with an unalterable conclusion – under no circumstances can epidurals be beneficial. It then precedes to trash the study, making demonstrably false claims about microtrauma and negative insinuations about the authors’ integrity, for the SOLE purpose of insuring that no one credits the observation that epidurals may be protective. Finally, with absolutely no data, it asserts that the current recommendations of NCB advocates would surely be protectve.

    Frankly, I think you owe the authors of the study an apology and a correction. You deliberately mischaracterized their work and impugned their integrity to fufill a private agenda of demonizing epidurals.

  12. September 4th, 2010 at 18:14 | #12

    I’m not going to comment on the research because I have not read it. However, I can theoretically see how an epidural could relax the pelvic floor muscles and allow greater stretch for those women who do achieve a vaginal birth with an epidural. Unfortunately most women with or without an epidural are subjected to a number of interventions that increase the risk of perineal trauma during birth. I conducted an extensive literature review for my PhD and in summary practices associated with increased risk of perineal trauma are: directed pushing; the use of analgesia and anaesthesia; an instrumental birth; hands on techniques; squatting, supine or lithotomy position. If you want the references and full literature review you can download it here:http://www.box.net/files/0/f/0#/files/0/f/0/1/f_474591782
    There is also a study by Professor Hannah Dahlen in press which found an increased risk of perineal trauma associated with: obstetrician care vs mw care; land birth vs waterbirth. I bet the media do not pick up on this research when it is published. Can you imagine the headline “Waterbirth with a midwife protects the pelvic floor!” – no chance.
    What we need is further research into physiological birth (ie. birth where women are left to get on with it). In addition it would be nice if the media reported research in a non-biased and non-sensationalist way so that women could have access to all information (from both sides of the birth divide). It would be even nicer if midwives and obstetricians could work together to provide the best possible maternity care for women. If we shared knowledge and respect we could revolutionise maternity care and outcomes.

  13. September 4th, 2010 at 19:02 | #13

    @Amy Tuteur, MD
    A key characteristic of science, as opposed to pseudoscience, is that all possible outcomes are allowed. That’s why “intelligent design” is not science; there is no evidence that would lead it’s advocates to announce that there is no “intelligent designer.” The conclusion is predetermined.
    I am LOLing at this!! Your blindness on this point is staggering. I’m not trying to derail the comment thread, and I do hope that I don’t, but I had to take this up. What you say of “intelligent design”, and what you have previously said of creationism is exactly what you should also say about the belief in evolution. I will not derail the thread but cannot allow this point to pass unchallenged.

    There is no evidence that would lead the advocates of evolution to announce that blind, purposeless chance did not bring about this universe and everything that is in it.

  14. September 4th, 2010 at 20:29 | #14

    ” I conducted an extensive literature review for my PhD and in summary practices associated with increased risk of perineal trauma”

    You are talking about damage to the superficial tissues of the perineum. We are talking about damage to the deep musculature, an entirely different issue.

  15. September 4th, 2010 at 20:36 | #15

    @Amy Tuteur, MD
    The media does not make this distinction in it’s reporting. I’m sure women are interested in all areas of their perineum and how to avoid damage. Why report one study and not the others?

  16. September 4th, 2010 at 20:55 | #16

    @Amy Tuteur, MD
    This analysis of a scientific paper veers perilously close to pseudosciece. It starts with an unalterable conclusion – under no circumstances can epidurals be beneficial.
    On the contrary, Amy Romano admits the possibility that an epidural may be protective, but wonders aloud if it is the best protection of pelvic floor trauma, when you consider that epidurals increase the likelihood of the need for forceps. In fact, I know that most NCB advocates will say (perhaps not every time they use the word “epidural,” but with some regularity) that they know of instances in which an epidural was either beneficial or necessary; including many times in which an epidural was the key in a woman ending up having a vaginal birth, instead of a C-section. But it still doesn’t change the fact that epidurals have downsides and negative side effects. Nor does it change the fact that epidurals are associated with greater use of forceps, which definitely leads to greater pelvic trauma than an unassisted vaginal birth.

    In a similar fashion, proponents of circumcision can point to certain studies that seem to demonstrate that cutting off part of a penis may lead to a reduction in the transmission of HIV/AIDS; yet intactivists can show that condom usage and monogamy (or abstinence) are far better methods of reducing the spread of diseases including HIV/AIDS, and that even if circumcision provides some protection against the transmission of AIDS to or from the man, that it still leaves a man very vulnerable to AIDS, when compared to intercourse using a condom.

  17. September 4th, 2010 at 20:58 | #17

    “The media does not make this distinction in it’s reporting.”

    Regardless of what the media does, you, as a midwife, should be cognizant of the difference.

  18. September 4th, 2010 at 21:26 | #18

    @Amy Tuteur, MD

    I am thanks.

  19. September 4th, 2010 at 23:29 | #19

    I have read the post, the criticisms of the post, and your replies, Ms. Romano.

    I don’t have anything to add, as assessing studies is not my gifting. I can say that I appreciate how respectful your tone always is, and how much you simply make sense. I’m reasonably certain you don’t think epidurals are “bad.” I certainly didn’t take it that way…I understood you to be saying that all the hooplah around this was unjustified. From all I can gather from this discussion, this is basically a study that tells us we should study this more…

    Do I have that right?

  20. avatar
    Kerri Bundy
    September 5th, 2010 at 12:05 | #20

    I talked to a nursing Ph.D. candidate who’d dissertation was research on women and kegels. 96 woman, early 20′s to mid 60′s none could kegel properly. Some had been instructed by physicians, some had read about it, some had never heard of it. This candidate also quoted a doctor as saying “I can talk any woman into surgery.” There is no meaningful reason for anyone but the woman to know how to do a proper kegel.
    This relates to the pelvic floor discussion in that we cannot work/exercise, protect, recover what we do not know. All women need to know their bodies better.

  21. September 5th, 2010 at 19:55 | #21

    Nice. my thoughts exactly!@Krista

  22. September 5th, 2010 at 20:07 | #22

    >> There is no evidence that would lead the advocates of evolution to announce that blind, purposeless chance did not bring about this universe and everything that is in it.

    I can’t quite decipher this, but it sounds like you are saying that evolution has as little evidence for it as creation. If so, that would be a staggeringly incorrect statement, as there is really no question that the variation of species on this earth is due to evolution. From phylogeny to DNA evidence to fossil record, everything is absolutely consistent with evolution of species over the millions of years that life has existed on this planet.

  23. September 5th, 2010 at 20:14 | #23

    Amy – I’m pretty surprised at your post here. It seems like a lot of your followers and fans got wind of this article and wanted to see it trashed, and you did it to please them. I have to agree with Amy T that your criticism of this article is not justified. The idea that they are using surrogate outcomes is incorrect. The point of the paper is the identification of microtrauma. That is the outcome they are measuring. If it were a surrogate, they would be making claims about epidurals and some other downstream outcome, which they are not.

    The idea of microtrauma of muscles is a well documented idea, and is the basis of much if not all muscle remodeling. The soreness one feels after working out is the results of this microtrauma, for example.

    To me it feels like you started out with the desire to deny the results of this study, rather than looking at it from an objective point of view.

  24. avatar
    September 5th, 2010 at 21:26 | #24

    Amy, thanks for taking this up. I do hope this study leads them into further studies. Unfortunately the conjecture about the possibility of epidural protecting the pelvic floor seems to have been hijacked by the media, and is what potential mothers are reading. These simplistic conclusions don’t usually end up carrying much weight in six months’ time. This conclusion does not appear to be the main aim of the research.

    I have read their responses on their Blog. Academic discussion so easily turns into mud-slinging. It is not possible for an ordinary woman like me to critically appraise either side. Unfortunately, this rhetorical nitpicking overshadows the benefits of La Maze’s methods, which are almost totally at odds with the degree of intervention these obstetricians are talking about. They are trying to perfect intervention, and work out ways of repairing the damage they cause, which is the only moral path for them. At least they admit that there is a lot they do not know, but at the same time they seem to be saying that Lamaze knows nothing, which is clearly untrue. This discredits them in my eyes.

    I have been fortunate to benefit greatly from Christine Kent’s work with non-surgical management of pelvic organ prolapse so the quest for better surgical repairs is a bit like the quest for a better designed machine gun, as far as I am concerned – totally irrelevant.

    I can read. I have education. I have internet access. These factors did not prevent my first labour being fully-managed. I will never know how medically necessary it all was. These factors have given me a means of living an energetic and normal life despite the prolapses. It is the millions of women throughout the world who do not have these advantages who are vulnerable to saying yes to all these fixes which may improve their lives in the short term, but multiply the risks to their bodies when they start the surgical repairs treadmill.

    My second and third labours and births were much less medicalised, thanks to Lamaze and his followers. Macrotrauma? Microtrauma? Phooey! To me, it doesn’t matter what caused it. What is important is how women like me respond to it.

  25. September 5th, 2010 at 21:48 | #25

    @Nicholas Fogelson, MD
    I’m saying that evolution is not falsifiable. There is no evidence that would lead its advocates to announce that evolution is not the cause of everything we see today or in the past. You bring up phylogeny and the similarity of DNA as a proof of evolution; I say it is a proof of a common Designer. You bring up the fossil record, I say it is merely millions of dead things buried in rock layers laid down by water all over the earth. Charles Darwin believed that his lack of proof for his hypothesis of evolution was due to the fossil record not being very well studied. He said that there should be innumerable transitional forms, and that if they were not found, then it would show that he was incorrect. They have not been found, and nobody has a reasonable explanation of finding them… yet it is not seen as disproving evolution.

    Now, here I am derailing the thread. Sorry.

    Dr. Fogelson, Dr. Amy, or anyone else, if you really want to take this up, you can go to this post I just published and have at it. I won’t think anything negative if you don’t feel like taking me up on it; I really don’t feel like the time either, but we can discuss it if you have the time and inclination.

  26. avatar
    September 6th, 2010 at 01:52 | #26

    Here’s a quick primer on the falsifiability of evolution, Kathy:


    Is there no micro or macro trauma (in general, obviously not lavator) associated with cesarean section? Does microtrauma lead to some sort of clinical problem? Where’s the evidence? If there is no proof that microtrauma leads to health problems, what’s the point in even looking at it? The “downstream outcome” Dr. Fogelson mentions is prolapse. Are the authors claiming that epidurals may prevent future pelvic organ prolapse, because that’s what the press release says:

    BJOG release: The use of epidurals may help prevent future pelvic organ prolapse

    Why bother to look at microtrauma without the assumption that it leads to permament problems? Here’s what the study says. Why the disconnect? Yikes.

    “Intrapartum epidural appeared to have a protective effect (P = 0.03; OR 0.42; 95% CI 0.19–0.93).

    Conclusion Levator trauma at the time of first delivery is associated with vaginal delivery, forceps and a longer second stage. Epidural pain relief may exert a protective effect.”

    An honest obstetrician would admit that the press release, and the RCOG overreached from the study itself. That’s not just the media.

    It’s certainly plausible that epidurals relax the pelvic floor enough to prevent damange. It’s also plausible, and falsifiable, that epidurals could cause an increase in forceps deliveries, and have the opposite effect. That puts the discussion firmly in the realm of science rather than pseudoscience. Unlike creationism vs evolution. Creationism/intelligent design is an argument based on faith rather than evidence.

  27. September 6th, 2010 at 07:02 | #27

    First, no more talk of creationism or intelligent design on this blog. Please follow Kathy’s link if you want to get involved with that. I will delete or edit any further comments that weigh in on that “debate”.

    Nicholas Fogelson, MD :
    Amy – I’m pretty surprised at your post here. It seems like a lot of your followers and fans got wind of this article and wanted to see it trashed, and you did it to please them. … To me it feels like you started out with the desire to deny the results of this study, rather than looking at it from an objective point of view.

    No, I shared with my readers a degree of skepticism about the study’s findings because of what I’ve said here already many times – there is ample evidence that epidurals are associated with outcomes such as instrumental delivery, OP delivery, prolonged second stage, and excess perineal trauma that all seem like they would, if anything, weaken the pelvic floor. Any time a new scientific finding is published that refutes conventional wisdom or other science, it deserves scrutiny. I did not set out to trash epidurals or this study because it dared to say something good about epidurals. I set out to sort out what was behind the disconnect – a real protective mechanism or science that may be flawed or limited? Or some other factor? I admit openly to being “biased” at the beginning of my quest, but I was biased because of the large body of *science* showing that epidurals have real, measurable adverse effects, including several that have potential significance for pelvic floor strength and function. They also have a very significant benefit – excellent pain relief, which is why I would never judge a woman for weighing the pros and cons and choosing one. But let’s do our best to give her solid evidence that helps her understand the likelihood of the real outcomes she cares about (peeing in her pants, needing future surgery for prolapse, etc.) Surrogate outcomes, by definition, don’t do that. More specifically, this novel surrogate outcome (a method of ultrasound measurement defined and described for the first time by these authors) does not and cannot do that.

    And while I don’t argue with your assertion that the authors were looking specifically at microtrauma and so it was appropriate for them to measure microtrauma, it is still a surrogate endpoint for pelvic floor function and strength. And the press release made claims about pelvic floor function and strength which is why it is appropriate to ponder whether the ultrasound findings described in the study are a useful surrogate.

    My main key message I was trying to get across by writing this post was this: please use caution when applying research on surrogate outcomes to patient care. History shows it can backfire.

    And let’s talk for a moment about bias. The profession of obstetrics has offered women three strategies for protecting the pelvic floor: routine episiotomy, elective primary c-section, and now epidurals. Oops, after decades of blind faith and routine use it turned out that episiotomy isn’t effective. So cross that one off the list. That leaves two options. And they have some things in common: they give doctors (rather than women) more control over the birth process, doctors and hospitals get major revenue from their use, and they have documented risks so any real or perceived pelvic floor benefit is a trade-off from the woman’s perspective. Also, some women have pelvic floor problems despite their use – so they’re no panacea. The idea that no one but obstetricians (who understand the science of muscle physiology better than lay people or midwives) may offer any ideas as to how to prevent pelvic floor problems when OBs are offering only risky and costly interventions is absurd and elitist. In any area of health there are prevention strategies that don’t involve doctors. In fact, most of the best prevention strategies don’t involve doctors, but no one gets paid for them so they don’t get used, except by those few engaged, motivated, health-conscious people. And, because industry funds so much of our research and there are few devices or machines that help with prevention, research on low-tech or behavioral prevention strategies is underfunded. Any one of us may have our individual biases and manage them well or poorly, but clearly our system is biased toward offering women doctor-controlled high-tech interventions over women-led, low-tech options.

    Happy Labor Day, everyone! May every woman have an opportunity to labor in health and safety!

  28. September 6th, 2010 at 08:43 | #28


    Four different doctors have already explained to you that you are wrong in your invocation of surrogate endpoints and you won’t admit it and apparently don’t understand it, so let me try again:

    There is a difference between basic research and clinical research. Basic research, as the name implies, is research on a scientific phenomenon that may or may not have clinical implications. So, for example, when Alexander Fleming noted and then researched a mold that killed the bacteria in his petri dishes, he was doing basic research. He announced that the mold made a substance that killed bacteria (an “antibiotic”) and speculated that the substance (named penicillin) might kill bacteria that were making people ill.

    His research on penicillin reported bacteria kill rates. That was not a “surrogate endpoint” for lives saved, because he was looking at the actual action of penicillin. The fact that he speculated that penicillin might save lives does not make the bacteria kill rates a “surrogate endpoint.”

    Similarly Shek and Dietz are doing basic research on levator trauma that may or may not have clinical implications. They were looking at what happens to the levator muscles in the wake of childbirth. They found that there could be macro trauma (such as avulsion) and microtrauma. They also reported on factors that increased or decreased the risk of these changes. They were not researching prolapse. They did not clan that muscle injury leads to or is the cause of prolapse; they merely speculated that it might be related.

    Interestingly, you gave the authors a pass on their claim that forceps increases the risk of levator trauma. You could have made an impassioned argument that forceps might be beneficial and that the authors used a “surrogate endpoint” of levator trauma to unfairly malign forceps. You ignored that entirely.

    So let me ask you directly: Do you believe that the authors observation that forceps increases levator trauma should be ignored because it is a “surrogate endpoint”?

  29. September 6th, 2010 at 11:43 | #29

    Please come to my blog post.

  30. avatar
    Sheila Rhodes
    September 7th, 2010 at 11:04 | #30

    And you certainly can have pelvic floor damage and incontinence problems without EVER having been pregnant, let alone birthed a baby.
    @Amy Romano

  31. September 8th, 2010 at 07:22 | #31

    Not having read the study yet, I wonder what kind of care the women without epidurals received. With all due respect, most doctors I have worked with do not know how to support a woman in natural labor. If the women in this study who had a drug free birth were like most of the women I work with, and received no help or care to work through a natural birth, this could make a difference.

    For example, if the women who received no epidural, were encouraged to push the way women with an epidural were, I would not be surprised if they were to recieve more perineal trauma. It makes sense to me that women who feel the urge to push and are encouraged to push as hard as women who have epidruals, are putting a lot more force behind that push.

    On the other hand, if they have a competent practitioner that knows and understands how to support a physiological birth, they may push a different way that might help to preserve the perineum better. (This is of course not proven by research, but it would be interesting to see this done).

    In the case of this study, I’m guessing, it is comparing a group of women who receive epidurals and are cared for my doctors that know how to care for them well, to a group of women who don’t receive epidurals and are cared for providers that don’t know how to best support them in their pushing efforts.

    Something that makes me think that the outcomes would be different if these two groups of women were treated differently, are the homebirth studies done. Whatever you want to say about the safety of homebirth, I think that most studies show that there is less perineal truama in a homebirth-obviously these are done without epidurals by providers that know and understand how to support woman in natural labor.

    But I do think this should also make us, as natural birth advocates, look at this and ask ourselves the question, “why would an epidural help protect the perineum”? If it does, how can be best help those who wish to not have an epidural, receive the same degree of protection? If this research does pan out, then we should be looking at how best to help women achieve the same results who choose to go without drugs.

    Just a thought, and before Amy jumps down my throat, yes more research would need to be done on this hypothesis, and as I have not read all the research on homebirth or this paper, these are just my own thoughts and opinions, and recognize them as such. Also, I am not trying to debate the effectiveness or safety of homebirth….does that rule out everything Amy? I guess I’ll see:)

  32. September 8th, 2010 at 13:46 | #32

    Cutting to the chase. There is an excellent meta-analysis indicating that immediate directed pushing is the culprit, even with an epidural!

    See: Brancato et al. A Meta-Analysis of Passive Descent Versus Immediate Pushing in Nulliparous Women With Epidural Analgesia in the Second Stage of Labor. JOGNN, 37 , 4-12; 2008. DOI: 10.1111/J.1552-6909.2007.00205.x

    More back up evidence…

    Less Pelvic Floor Damage Associated With Uncoached Than Coached Pushing During Labor, AUGS/SGS 2004 Joint Scientific Meeting: Abstract 14. Presented July 30, 2004.

    “…no evidence that bearing down during contractions helps either the mother or the child…” Bloom et al. The American Journal of Obstetrics & Gynecology (Jan. 2006).

  33. September 9th, 2010 at 13:25 | #33

    “There is an excellent meta-analysis indicating that immediate directed pushing is the culprit, even with an epidural!”

    That study doesn’t even address the issue of levator damage or pelvic organ prolapse, let alone identify the culprit.

  1. September 5th, 2010 at 00:46 | #1
  2. September 13th, 2010 at 08:05 | #2
  3. November 14th, 2010 at 14:04 | #3