Articles

Does C-section increase the rate of neonatal death?

It is a potentially devastating indictment of the rising C-section rate. Most midwifery and “natural” childbirth websites claim that elective C-section triples the rate of neonatal mortality. Mainstream web sites like Feministing.com, and newspapers like The New York Times have repeated the claim. There’s just one problem. It’s not true.

The claim originated with the paper Infant and Neonatal Mortality for Primary Cesarean and Vaginal Births to Women with “No Indicated Risk,” United States, 1998–2001 Birth Cohorts, MacDorman et al, Birth Volume 33 Page 175, September 2006. According to the authors:

Neonatal mortality rates were higher among infants delivered by cesarean section (1.77 per 1,000 live births) than for those delivered vaginally (0.62). The magnitude of this difference was reduced only moderately on statistical adjustment for demographic and medical factors, and when deaths due to congenital malformations and events with Apgar scores less than 4 were excluded. The cesarean/vaginal mortality differential was widespread, and not confined to a few causes of death. Conclusions: Understanding the causes of these differentials is important, given the rapid growth in the number of primary cesareans without a reported medical indication.

The implication, of course, is C-sections done without a medical indication raises the risk of neonatal death by a factor of three. The entire study hinges on a critical detail. Are women with “no indicated risk” really women who have no risk factors? The answer is a resounding no.

Since birth certificates are such an important source for research information, they have been repeatedly studied for accuracy. Birth certificates are highly accurate for administrative data like parents’ names or numerical data like weight or Apgar scores. It is well known, however, that they are highly inaccurate when it comes to listing complications.

How Well Do Birth Certificates Describe the Pregnancies They Report? The Washington State Experience with Low-Risk Pregnancies, Dobie et al report:

Conclusions: Because birth certificates significantly underestimated the complications of pregnancies, number of interventions, number of procedures, and prenatal visits, use of these data for health policy development or resource allocation should be tempered with caution.

The reporting of pre-existing maternal medical conditions and complications of pregnancy on birth certificates and in hospital discharge data,  M. Lydon-Rochelle,  et al. found:

Results Birth certificate and hospital discharge data combined had substantially higher true-positive fractions than did birth certificate data alone for cardiac disease (54% vs 29%), acute or chronic lung disease (24% vs 10%), gestational diabetes mellitus (93% vs 64%), established diabetes mellitus (97% vs 52%), active genital herpes (77% vs 38%), chronic hypertension (70% vs 47%), pregnancy-induced hypertension (74% vs 49%), renal disease (13% vs 2%), and placenta previa (70% vs 33%)… Conclusion In Washington, most medical conditions and complications of pregnancy that affect mothers are substantially underreported on birth certificates,…

In other words, for virtually every serious pregnancy complication, that information was missing from the birth certificate in more than half the cases.

Even a cursory look at the data showed that the authors assumptions were entirely unfounded. Women in the group characterized as planned C-sections for “no medical indication” had birth certificates that indicated that they had been in labor for hours before the C-section. Although the indications had been absent, it was clear that there must have been indications for the C-section.

In response to pointed criticism in the Letters to the Editor, the authors who had originally looked at births from the 1998-2001 cohort, now looked at births from the 1999-2002 cohort, performing the same analysis but applying an intention to treat methodology. The paper entitled Neonatal Mortality for Primary Cesarean and Vaginal Births to Low-Risk Women: Application of an “Intention-to-Treat” Model was published in February 2008. As the authors explained:

… an “intention-to-treat” methodology, a methodology commonly used in medical research… [E]mergency cesarean sections performed after a woman was in labor would be combined with vaginal births to create a “planned vaginal delivery” category since the original intention of the physician and the mother in both cases was presumably to deliver the infant vaginally. The “planned cesarean delivery” group would include only those deliveries where a cesarean section was performed without labor.

This analysis led to very different results:

… In the most conservative model, the adjusted odds ratio for neonatal mortality was 1.6 (95% CI 1.35–2.11) for cesareans with no labor complications or procedures, compared with planned vaginal deliveries. Conclusions: The finding that cesarean deliveries with no labor complications or procedures remained at a 69 percent higher risk of neonatal mortality than planned vaginal deliveries is important, given the rapid increase in the number of primary cesarean deliveries without a reported medical indication.

So now instead of claiming that C-sections increase the risk by a factor of 3, they are claiming that C-sections increase the risk of neonatal death by only half that amount. But the authors still do not address the primary flaw of the study. They really have no idea which C-sections were indicated and which were not. The difference is critical. If only 0.002% of the remaining birth certificates were missing risk data, there would be no difference in mortality in the two groups at all. Based on what we know about the reliability of birth certificate data, there is reason to believe that far more than 0.002% of birth certificates lack the relevant data.

The bottom line is that MacDorman and colleagues never showed that C-section increased the risk of neonatal death by any amount.. They demonstrated an entirely different principle: garbage in, garbage out. When you apply statistical analysis to erroneous data, you reach unsubstantiated, erroneous conclusions.

Posted in: Science and Medicine

Leave a Comment (38) ↓

38 thoughts on “Does C-section increase the rate of neonatal death?

  1. windriven says:

    And interesting post Dr. Tuteur. But playing the devil’s advocate here, you have accused your colleagues of inadequately sorting and vetting their data noting that, “If only 0.002% of the remaining birth certificates were missing risk data, there would be no difference in mortality in the two groups at all.” But that “if” means that your observation is speculative. Isn’t it incumbent upon you or others who question the validity of the study to demonstrate rather than conjecture that the data is wrong?

    I am also curious why the peer review process didn’t expose these concerns about the data, especially after the controversy associated with the 2006 publication.

    It seems credible that cesarians would be no more risky to the baby, perhaps even less risky, than vaginals. There would seem to be plenty of interested parties who might fund research along these lines. So isn’t that ball in your (and others who believe as you do) court?

  2. edgar says:

    It is Amy’s big conspiracy theory all over again. Amy based on the data they had available, they did show that. I believe that you yourself are a big fan of birth and death certificates when it suits you, so why isn’t that garbage in, garbage out?

    I will be the first to admit that birth and death certificates both are flawed data, but it is the best data set we have. I think looking at these data sets are a great starting point for looking at some of these questions. The next step would be a matched prospective cohort.

    It is a good, low cost starting point from which to develop further studies.

    You seem to attach an awful lot weight to studies like these, instead of taking them for what they are.

  3. “Isn’t it incumbent upon you or others who question the validity of the study to demonstrate rather than conjecture that the data is wrong?”

    I already demonstrated it.

    In the paper, MacDorman explained that “no indicated risk” meant the absence of any of 30 different possible complications and risk factors. She explained that she used “no indicated risk” as a proxy for “no risk.” However, we know from multiple studies on the accuracy of birth certificates that the section on risks and complications leaves out important risks and complications as much as 50% of the time.

    “I am also curious why the peer review process didn’t expose these concerns about the data, especially after the controversy associated with the 2006 publication.”

    Lay people often view a scientific paper as the end of the vetting process. In truth, it is merely the beginning. Getting the paper published means that the editors of the journal thought that the paper was worthy of being included in the debate. It is up to the readers to determine whether the authors’ conclusions are reasonable and whether they are supported by the data that the authors presented. In this case they are not.

    The paper was published in the journal Birth, which functions as the journal of the “natural” childbirth community. Moreover, the authors of the paper are editors of the journal.

    The initial paper received serious criticism. The data allocation was so sloppy a number of C-sections for fetal distress were included in the group “no indicated risk”. Therefore, the authors went back and analyzed the data using an “intention to treat” methology.” Instead of simply relying on the what was written in the space set aside for risks, they looked at the whole birth certificate to determine from other data whether the C-section had been done for an indication that wasn’t mentioned.

    Looking at the entire birth certificate, instead of merely what was written in the space for risk, knocked their claimed risk of neonatal death in half. In order to draw any valid conclusions, though, you would need to look at the hospital records to determine the risk factors and complications, since we know that the birth certificates themselves are unreliable.

  4. edgar:

    “but it is the best data set we have”

    Not for the information that we need to study this issue. Hospital records provide far more information and are more reliable when it comes to the issue of risk factors and complications.

    “It is a good, low cost starting point from which to develop further studies.”

    No, the study is a piece of junk. The authors showed nothing.

  5. edgar says:

    Hospital records are not a data set. However, you are correct, it is a more robust methodology. But it is an expensive and time consuming process.
    Until you yourself refrain from quoting birth and death certificates, how in the world can castigate other for doing so? The fact of the matter is that NCHS data including birth and death certificates is used all the time for that very reason because, I repeat, it is the best dataset we have.

  6. edgar says:

    You seem to assume that missing information is only an issue in c-sections and not vaginal births. we do not know if there is a reporting difference between these two groups.

  7. edgar:

    “Until you yourself refrain from quoting birth and death certificates”

    As the research quoted above demonstrates, birth certificates are very reliable when it comes to administrative data, and that would include place of birth and attendant. They are very unreliable when it comes to risk. Therefore, if you are attempting to claim that C-sections done on women with no risk factors and no complications lead to an increased neonatal mortality rate, you must use a dataset that reports risk factors and complications accurately. Otherwise, you can’t draw any conclusions.

    “You seem to assume that missing information is only an issue in c-sections and not vaginal births. we do not know if there is a reporting difference between these two groups.”

    The bias is all in one direction. Risk factors and complications are missing from a significant proportion of birth certificates in births that had risk factors or complications. There is no evidence that false risk factors and complications were added when there were none present at the actual birth.

    Moreover, to the extent that risk factors and complications are distributed among women, those who had C-sections are more likely to have had complications and risk factors than those who had vaginal deliveries.

  8. overshoot says:

    Birth certificates?

    <insert obscenity>

    None of my three have any indication on their birth certificates of why they were delivered by C-section. No mention of the fact that their mother was preeclamptic (to the point of being hospitalized at 32 weeks to manage the condition for the first two), a primapara with twins for the first with attempt of labor after membrane rupture; an intended VBAC for the second (induction contraindicated) with placental insufficiency the second time around, placental calcification at 34 and 36 weeks for both pregnancies, etc.

    None of that. This is like trying to research treatment risks from newspaper accounts.

    Oh, yeah: all three are fine now.

  9. windriven says:

    Dr. Tuteur, I don’t want to belabor this but:

    Isn’t it incumbent upon you or others who question the validity of the study to demonstrate rather than conjecture that the data is wrong?

    “I already demonstrated it. ”

    With all due respect, I disagree. I think you have raised material questions about the validity but I would not agree that you have demonstrated the data to be invalid.

    “Lay people often view a scientific paper as the end of the vetting process. ”

    I am not a physician but I am a scientist and am familiar with the process that scientific papers endure on their way to being published – at least in the field of physics. And while it is true that peer review is not a perfect filter, my puzzlement remains given the questions raised regarding the data set in the 2006 paper. It is disturbing that the 2008 paper appeared in a journal edited by the authors – especially after the ruckus attending the 2006 paper. Is it common in medicine for researchers to publish in journals they edit?

  10. In the January 2008 issue of Obstetrics and Gynecology contains another paper that adds to the mountain of evidence demonstrating that it is impossible to determine C-section risk factors merely by looking at birth certificates. According to Cesarean Delivery Among Women With Low-Risk Pregnancies: A Comparison of Birth Certificates and Hospital Discharge Data:

    “Among 40,932 women with primary cesarean deliveries and no risk indicated on the birth certificate, 35,761 (87.4%) had a risk identified in the hospital discharge data. The overall agreement between data sources on the presence of any risk indicator was low (κ=0.18). Among primary cesarean deliveries, the percentage without indicated risk was 58.3% when using birth certificate data alone and 3.9% when using hospital discharge data in combination with the birth certificate.

    CONCLUSION: Using birth certificate information alone overestimated the proportion of women who had no-indicated-risk cesarean deliveries in Georgia. Evidence of many indications for cesarean delivery can be found only in the hospital discharge data. The construct of no indicated risk as determined from birth certificates should be interpreted with caution, and the use of linked data should be considered whenever possible.”

    It is inappropriate to base a study on data known to be wrong a substantial portion of the time.

  11. JerryM says:

    windriven, she demonstrated it clearly. It’s like the climatologists using the temperature data of the last 10 years to show there is no global warming. Given the wrong data, any claim can be supported.

    It’s a shame the peer review process is more concerned with how the data is used instead of what data is used, but I guess it’s inherent in the process – they’re not experts on what data is available, only on what the proper way of handing data is.

  12. windriven says:

    @ Jerry

    Perhaps what constitutes ‘demonstrated … clearly’ is different in medicine than in physics. In science as I’ve studied it, casting doubt on a data set proving a data set invalid. I’m not trying to be a jerk here but those of us who are committed to science hold woo-meisters accountable for their imperfect assertions. Should we not strive for precision ourselves? Or do we get to operate by relaxed rules because we’re on the ‘right’ side?

    And I would be circumspect about using the paradigm of global warming to argue scientific rectitude given the recent disclosures about apparent egregious misconduct at the CRU at East Anglia.

  13. windriven says:

    I’m sorry but I used a mathematical shorthand for inequality in my response to Jerry. It was not recognized when I posted and apparently deleted. The sentence should read:

    In science as I’ve studied it, casting doubt on a data set does not equal proving a data set invalid.

    Perhaps someone clever in these things can tell me how to get opposing less than and greater than symbols to display.

  14. windriven says:

    @ Dr. Tuteur

    Thank you, doctor. Given your citation from Obstetrics and Gynecology, your point is well taken. I hope you accept my earlier skepticism in the spirit in which it was intended.

    JT

  15. “I hope you accept my earlier skepticism in the spirit in which it was intended.”

    I do.

  16. Basiorana says:

    As far as I can tell, all this analysis is proving is that there is no evidence that C-section raises neonatal mortality, but it doesn’t prove that the statement is untrue, or that C-sections are safer or as safe as vaginal birth. It just says that there’s no evidence either way, and at worst it’s a lower difference than the original study reported. Yet you seem to be extrapolating that this means that the claim is itself untrue– when in fact it is simply unproven.

    Their conclusion is obviously unsubstantiated, but is it fair to call it erroneus until there is evidence to the contrary?

  17. Basiorana:

    “As far as I can tell, all this analysis is proving is that there is no evidence that C-section raises neonatal mortality, but it doesn’t prove that the statement is untrue, or that C-sections are safer or as safe as vaginal birth.”

    The primary purpose of C-section is to lower the neonatal mortality rate, so it would be rather startling if it raised the neonatal mortality rate. That doesn’t mean that it couldn’t happen. It means that the claim that C-sections increase neonatal mortality is an extraordinary claim and therefore requires extraordinary evidence, not to mention a biological basis.

    There are literally thousands of studies showing that C-sections lower the neonatal mortality rate in a wide variety of situations and settings. There is copious national and international data confirming the life-saving nature of C-sections.

    So this study purports to overturn everything we know about C-sections, but it is based on data that is profoundly flawed (and that the authors almost certainly knew was profoundly flawed), and is therefore useless. Since the authors must have known about the flaws in birth certificates, it was disingenuous at best and deliberately misleading at worst to base an entire study on the assumption that “no indicated risk” means “no risk.”

    Finding that this study is useless does not prove that C-sections are life saving. That claim is justified by many other sources of data.

  18. Zoe237 says:

    Touche on the global warming example! Anyway…

    Those are valid criticisms of this study, as in many analyses of birth certificate data. The authors also describe the problem in the discussion section of the paper. They attempt to correct for it, but acknowledge that it is impossible to do so completely. Therefore, the answer to the Tuteur’s title question is, thus far, is “we don’t know, but there is reason to ask the question.”

    It seems to me that the main risks of cesarean sections would be to the mother, although I have heard reports of respiratory problems in neonates correlated with c-section. The current fad in childbirth and parenting is that mothers who consider risks to their own bodies in addition to their babies are “selfish.”

    Any thoughts on these studies (I realize the last two are not directly comparable)? I wonder if another issue is that some elective c-sections are done too early, e.g. 37-38 weeks, that it could account for some of the discrepancy possible erronously attributed to c-section itself.

    http://www.ncbi.nlm.nih.gov/pubmed/11675055

    2001 Oct 13;358(9289):1203-7.
    Early maternal and neonatal morbidity associated with operative delivery in second stage of labour: a cohort study.
    Murphy DJ, Liebling RE, Verity L, Swingler R, Patel R.

    Women undergoing caesarean section were more likely to have a major haemorrhage (>1 L; 2.8, 1.1-7.6) and extended hospital stay (>/=6 days; 3.5, 1.6-7.6) than those with vaginal delivery. Babies delivered by caesarean section were more likely to require admission for intensive care (2.6, 1.2-6.0) but less likely to have trauma (0.4, 0.2-0.7) than babies delivered by forceps. Overall neonatal morbidity was low, but a few babies in each group had serious complications (serious trauma, eight vs three; sepsis, six vs 13; and jaundice, ten vs 12 after vaginal delivery and caesarean section, respectively). Major haemorrhage was less likely after delivery by a skilled obstetrician (0.5, 0.3-0.9). INTERPRETATION: The data lend support to an aim to deliver women vaginally, unless there are clear signs of cephalopelvic disproportion, and underline the importance of skilled obstetricians supervising complex operative deliveries.

    http://www.ncbi.nlm.nih.gov/pubmed/17977819?ordinalpos=1&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_SingleItemSupl.Pubmed_Discovery_RA&linkpos=2&log$=relatedarticles&logdbfrom=pubmed

    Maternal and neonatal individual risks and benefits associated with caesarean delivery: multicentre prospective study.
    Villar J, Carroli G, Zavaleta N, Donner A, Wojdyla D, Faundes A, Velazco A, Bataglia V, Langer A, Narváez A, Valladares E, Shah A, Campodónico L, Romero M, Reynoso S, de Pádua KS, Giordano D, Kublickas M, Acosta A; World Health Organization 2005 Global Survey on Maternal and Perinatal Health Research Group.

    CONCLUSIONS: Caesarean delivery independently reduces overall risk in breech presentations and risk of intrapartum fetal death in cephalic presentations but increases the risk of severe maternal and neonatal morbidity and mortality in cephalic presentations.

    http://www.ncbi.nlm.nih.gov/pubmed/18077440?itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_RVDocSum&ordinalpos=6

    BMJ. 2008 Jan 12;336(7635):85-7. Epub 2007 Dec 11.
    Risk of respiratory morbidity in term infants delivered by elective caesarean section: cohort study.
    Hansen AK, Wisborg K, Uldbjerg N, Henriksen TB.

    CONCLUSION: Compared with newborns delivered vaginally or by emergency caesarean sections, those delivered by elective caesarean section around term have an increased risk of overall and serious respiratory morbidity. The relative risk increased with decreasing gestational age.

    http://www.ncbi.nlm.nih.gov/pubmed/17296957?itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_RVDocSum&ordinalpos=21

    CMAJ. 2007 Feb 13;176(4):455-60.
    Maternal mortality and severe morbidity associated with low-risk planned cesarean delivery versus planned vaginal delivery at term.
    Liu S, Liston RM, Joseph KS, Heaman M, Sauve R, Kramer MS; Maternal Health Study Group of the Canadian Perinatal Surveillance System.

    INTERPRETATION: Although the absolute difference is small, the risks of severe maternal morbidity associated with planned cesarean delivery are higher than those associated with planned vaginal delivery. These risks should be considered by women contemplating an elective cesarean delivery and by their physicians.

  19. Zoe237 says:

    “The primary purpose of C-section is to lower the neonatal mortality rate, so it would be rather startling if it raised the neonatal mortality rate. That doesn’t mean that it couldn’t happen. It means that the claim that C-sections increase neonatal mortality is an extraordinary claim and therefore requires extraordinary evidence, not to mention a biological basis”

    I was under the impression that the issue was elective c-sections, not necessary ones. I would think that the burden of proof for unnecessary surgery (or surgery without “indicated risk” as in the title of the paper) would be on those advocating it or defending its practice. Is it biologically plausible that an elective surgery with few benefits in the case of absence of indication could have some risks?

  20. Plonit says:

    The primary purpose of C-section is to lower the neonatal mortality rate, so it would be rather startling if it raised the neonatal mortality rate. That doesn’t mean that it couldn’t happen. It means that the claim that C-sections increase neonatal mortality is an extraordinary claim and therefore requires extraordinary evidence, not to mention a biological basis.

    +++++++++++++++

    Hmmm, an extraordinary claim that requires extraordinary evidence, huh?

    Do any claims about CS vis-a-vis risks or benefits to mothers and babies really fall into the category of “extraordinary” in the sense used by skeptics (following Carl Sagan)? The claims made for homeopathy are extraordinary in exactly this sense, because they violate so much else of what we already know about the physical world.

    “The primary purpose of C-section is to lower the neonatal mortality rate, so it would be rather startling if it raised the neonatal mortality rate.”

    Now this I do find a classic piece of non-logical thinking. You seem to be arguing that it is our good intentions in using a procedure dictates the outcome of that procedure. cf. “The primary purpose of shave and enema was to lower the incidence of puerperal fever, so it would be rather startling if it raised the incidence of puerperal fever”.

    The claim that CS for a particular indication, or without any indication, lowers or raises neonatal mortality (or any other of the usual outcome measures) is an ordinary claim that requires ordinary evidence. MacDorman’s paper may not provide even that ordinary evidence, but let’s not get carried away. No one is talking about little green men from outer space here, but only the risks and benefits in populations of a medical procedure.

  21. Zoe273,

    Let’s look first at the studies that claim to show that elective C-section is associated with increased neonatal or maternal mortality.

    The Villar study is yet another example of how evidence and conclusions can be misrepresented. My critique of this study was published by the BMJ as a rapid response and in the print edition.

    Villar et al. claim:

    “Women undergoing caesarean deliveries, either intrapartum or elective … had double the risk for severe maternal morbidity and mortality (including death, hysterectomy, blood transfusion, and admission to intensive care) and up to five times the risk of a postpartum infection compared with women undergoing vaginal delivery.”

    Based on the conclusion of the study itself, you might be forgiven for thinking that the study showed that elective C-sections resulted in higher maternal or neonatal mortality or morbidity. It doesn’t show that. Why? Because the authors chose to call all non-emergency cesareans (regardless of the medical indication) “elective”. They never even looked at maternal request cesareans or medically unindicated cesareans.

    The use of the term “elective” in this study is misleading. The study compares women who had vaginal deliveries with women who had medically indicated cesarean sections, both non-emergent and emergent. It never investigated elective cesareans and therefore it reaches no conclusions about elective cesareans.

    According to Table 1, among the 13,208 “elective” cesareans: repeat C -section 46.1%; breech 14.2%; pre-existing medical conditions 9%; complication of current pregnancy 40.2%.

    As the authors acknowledge:

    “… [this] group had higher risk in terms of women with previous complicated pregnancies or perinatal outcomes, problems related to current pregnancy, and being referred from other institutions for delivery…”

    This study show that cesarean section reduced the risk of neonatal death, particularly for breech babies. There was a slightly increased risk of maternal death in women undergoing emergency cesarean, but there was no significant difference in maternal death between vaginal delivery and non- emergent medically indicated cesarean. No conclusions can be reached about cesareans done without a medical indication.

    The Liu study is misleading as well.

    “Although the absolute difference is small, the risks of severe maternal morbidity associated with planned cesarean delivery are higher than those associated with planned vaginal delivery. These risks should be considered by women contemplating an elective cesarean delivery and by their physicians.”

    The authors are looking at morbidity because there was no difference in maternal mortality. Indeed there were 2 deaths in the vaginal delivery group and 0 in the C-section group.

    Furthermore, the authors have a very strange definition of “severe morbidity.” Because they include wound hematoma and disruption, they artificially inflate the “severe morbidity” of C-sections.

    The study also had several other relevant findings. C-section decreased the risk of neonatal death. So we are talking about the increase in maternal morbidity from a procedure that is shown to save neonatal lives.

    Second, the risk of maternal morbidity was substantially higher in women who had emergency, as opposed to elective, C-sections. It was safe to plan a C-section for breech (both for baby and mother) than to require one emergently.

  22. Plonit:

    “You seem to be arguing that it is our good intentions in using a procedure dictates the outcome of that procedure.”

    I’m not talking about intentions. I’m talking about the scientific evidence.We know from copious scientific evidence, national and international statistics that C-sections save neonatal lives. Therefore, it would be rather startlingly to find that they didn’t. And, indeed, no one has yet found that C-sections at term increase neonatal mortality. The MacDorman paper did not show that, and no other paper shows that.

  23. edgar says:

    Amy, I maintain my point, the paper is very clear about it’s data limitations.

    Perhaps you are the one without an understanding of the publication process. I can’t imagine going to NIH or whomever for funding and asking to for $ to do a study to see if c-section without any risk factor changed the mortality rate without having some sort of info to back up preliminary info.
    THAT is the purpose of peer review and studies, to have blocks to build on.
    As I said, it is a cheap and easy way to get prelim information from which to build upon.

  24. edgar:

    “the paper is very clear about it’s data limitations.”

    No, the paper is clear about the theoretical limitations of the data. It is silent on the the fact that the data is already known to be so corrupted as to be useless in this setting.

  25. Plonit says:

    We know from copious scientific evidence, national and international statistics that C-sections save neonatal lives. Therefore, it would be rather startlingly to find that they didn’t.

    +++++++++++

    This is a meaningless statement.

    Transplants save lives. Blood transfusions save lives. Antibiotics save lives. Chemotherapy saves lives…..

    but in a subset of people having transplants, blood transfusions, antibiotics, chemotherapy etc…*with no medical indication* it would not be at all startling to find that the risks of these interventions outweighed the benefits (since there are no defined benefits).

    Now, MacDorman et al may be barking up the wrong tree with “no indicated risk” = “no medical indication”. But supposing they were able to identify those caesareans that were done without medical indication, would it be startling to find risks outweighing benefits?

  26. Plonit:

    “would it be startling to find risks outweighing benefits?”

    As a matter of fact, it would be startling to find that C-section increases neonatal mortality even if there is no medical indication for that C-section. That’s because childbirth is inherently dangerous. It has an intrinsic rate of neonatal death quite apart from any medical indications or risk factors. C-section transfers the bulk of the risk from baby to mother. As a general matter, C-sections are good for babies.

    Fetal trauma in term pregnancy. Am J Obstet Gynecol 2007;197:499.e1-499.e7 is a comprehensive study of over 100,000 consecutive deliveries of singleton term neonates in the vertex position with no congenital anomalies:

    “The potential study population in Nova Scotia for the 14-year period (1988-2001) was 153,209. Twenty-two percent of the population was excluded based on the inclusion criteria of singleton (1%), term (7%), no major anomalies (3%), no fetal deaths (0.1%), and in vertex presentation(11%) to give the actual study population of 119,432.”

    Deliveries were grouped as follows:

    “… mode of delivery was defined as vaginal or cesarean, and method of delivery was defined as spontaneous vaginal, assisted vaginal (vacuum, forceps), … and cesarean delivery (with and without labor)…”

    The definitions of trauma were:

    “Fetal trauma was considered major trauma if 1 or more of the following were present: depressed skull fracture, intracranial hemorrhage, or brachial plexus palsy. Minor trauma was considered if 1 or more of the following were present: linear skull fracture, other fractures, facial palsy, or cephalhematoma.”

    The results of the study showed that the risk of major and minor trauma was dramatically reduced by Cesarean section, particularly if the Cesarean was performed before labor began. If the risk of major trauma in a vaginal delivery is described as 1.0, the relative risk of major trauma is reduced by 82% by a C-section done in labor (RR 0.18), and reduced by 88% by a C-section done before labor begins (RR 0.12)… The same results also apply to minor trauma. As compared to a spontaneous vaginal delivery, the risk of minor trauma is lowered most (by 89%) for a C-section without labor (RR 0.11) and raised most by vaccum delivery (RR 5.49).

    And that’s just traumatic injuries. The risk of anoxic brain damage is also reduced if babies do not have to endure labor.

  27. Calli Arcale says:

    windriven:

    Perhaps someone clever in these things can tell me how to get opposing less than and greater than symbols to display.

    The reason they’re not showing is because they have special meaning in HTML, the language in which this page is written. Your browser removes them when parsing and then rendering the page.

    To tell the browser not to do that, the most reliable method is to use the HTML codes for the characters.

    Example:
    less than: < or <
    greater than: > or >

    Rendered:
    less than: < or <
    greater than: > or >

    I’m hoping I did that right and the example will be visible. If not, I’ll try again.

  28. Calli Arcale says:

    Drat; it rended the examples. Okay, take the spaces out and you get the codes. ;-) (Don’t forget to include the semicolon at the end; it signals to the browser that the end of the code has been reached.)

    Less than:
    & # 6 0 ;
    or
    & l t;

    Greater than:
    & # 6 2 ;
    & g t ;

  29. Zoe237 says:

    Thank you for your thoughts. I don’t have a lot of time to respond these days, but the study you referenced (“Fetal Trauma in Pregnancy”) does not “prove” that ELECTIVE c-section increases or decreases the rate of neonatal death.

    (Amy Tuteur, M.D.):”As a matter of fact, it would be startling to find that C-section increases neonatal mortality even if there is no medical indication for that C-section. That’s because childbirth is inherently dangerous. It has an intrinsic rate of neonatal death quite apart from any medical indications or risk factors. C-section transfers the bulk of the risk from baby to mother. As a general matter, C-sections are good for babies.”

    Nor does it prove that elective c-sections are “good for babies.” The first study I quoted (comparing instrumental vaginal delivery to cesarean section) also found that cesarean section decreased the rate of physical trauma, but increased the rate of NICU admissions. Why would this be? Fetal trauma is hardly the only measure of outcomes.

    Multiple other studies have found that respiratory morbidity, length of hospital stay, and NICU admission were significantly increased by elective c-sections. Some of these are no doubt plagued by the problems that you mentioned with the first one… i.e. looking at birth certificates and being unable to exclude cesarean sections performed WITH indicated risk. And the issue actually can go either way, which is why the “intention to treat” model should be used.

    http://journals.lww.com/greenjournal/Fulltext/2009/06000/Neonatal_Outcomes_After_Elective_Cesarean_Delivery.7.aspx

    In a 2006 review conducted by the Department of Health and Human Services: Agency for Healthcare Research and Quality, the reviewers detailed the following evidence:

    “Neonatal mortality: Higher risk for “cesarean” than for spontaneous vaginal delivery; no controls for underlying maternal or neonatal
    indications for cesarean (III weak)

    Respiratory morbidity: Higher risk with cesarean; risk drops with advancing gestational age; no study evaluated meconium aspiration
    syndrome by mode of delivery (II moderate)

    Intracranial hemorrhage: No difference between prelabor cesarean and spontaneous vaginal delivery; higher risk for assisted
    vaginal deliveries and cesarean deliveries in labor than
    for spontaneous vaginal delivery (III- weak)

    Facial nerve injury: No difference between vacuum or prelabor cesarean delivery and spontaneous vaginal delivery; higher risk for
    forceps and combined vacuum and forceps than for either a vaginal or cesarean delivery (III-weak)

    Brachial plexus injury: Lower risk for all cesareans than for spontaneous vaginal delivery; higher risk for vacuum, forceps, and combined vacuum and forceps delivery than for spontaneous
    vaginal delivery (III-weak)

    Fetal lacerations Lower risk for elective cesarean than for unplanned
    Cesarean (III- weak)

    Neonatal length of stay Higher risk of longer hospital stay with elective compared
    with vaginal delivery (III- weak)”

    In their view, the only moderate evidence is for the risk of respiratory morbidity. All others are weak or inconclusive.

    http://www.ahrq.gov/downloads/pub/evidence/pdf/cesarean/cesarreq.pdf

    Viswanathan M, Visco AG, Hartmann K, Wechter, ME, Gartlehner G, Wu JM, Palmieri R, Funk
    MJ, Lux, LJ, Swinson T, Lohr KN. Cesarean Delivery on Maternal Request. Evidence
    Report/Technology Assessment No. 133.

    Now, this information may be out of date (2006). I welcome newer studies that may answer the question in the title of the blog post, particularly ones that show no risk for respiratory distress in neonates delivered by elective cesarean section.

    (Amy Tuteur, MD): “It means that the claim that C-sections increase neonatal mortality is an extraordinary claim and therefore requires extraordinary evidence, not to mention a biological basis.”

    Here is a biologically plausible explanation of why elective cesarean sections may be risky to infants:

    Semin Perinatol. 2006 Oct;30(5):296-304.

    Respiratory transition in infants delivered by cesarean section.

    http://www.ncbi.nlm.nih.gov/pubmed/17011402?ordinalpos=1&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_SingleItemSupl.Pubmed_Discovery_RA&linkpos=2&log$=relatedreviews&logdbfrom=pubmed

    There is also the concern of primary elective cesarean section risk to future children, including the risk of placenta previa and accreta. This has been documented.

    Furthermore, the American College of Obstetricians and Gynecologists state in their ethics committee platform on patient choice that the burden of proof is on those who wish to change existing practice… “i.e. the replacement of usual care in labor with major surgical procedure” IOW, we need proof that elective cesarean section decreases neonatal and maternal mortality and morbidity.

    http://www.acog.org/from_home/publications/ethics/co395.pdf

    http://www.acog.org/from_home/publications/press_releases/nr05-09-06-1.cfm

    You state that “websites claim that elective C-section triples the rate of neonatal mortality… there’s only one problem: it’s not true” and that “…c-sections are good for babies.” Whether ELECTIVE c-sections decrease or increase neonatal mortaility has not been shown either way. This is hardly an issue of “extraordinary claim; extraordinary evidence.” The fact that you would even claim so conveys your extremist views on intervention in childbirth.

  30. Zoe237 says:

    I should add, in the absence of reliable information on neonatal mortality, morbidity, whether by fetal trauma, NICU stays, or respiratory distress, certainly becomes relevant.

    (I believe it is relevant either way, as is maternal mortality and morbidity, but evidently not to others).

  31. Zoe237,

    The study I cited is one of many that demonstrate the benefits of C-section. I quoted it as an example, not as proof of my claim.

    The bottom line is that childbirth is inherently dangerous. Like reproduction throughout the animal kingdom in general and reproduction in humans specifically, it has a high wastage rate. Millions of ova are never fertilized, billions of sperm never reach an ovum, 20% of established pregnancies end in miscarriage, and birth itself has an intrinsic death rate.

    Childbirth is inherently dangerous to both babies and mothers, but is much more dangerous to babies. Indeed, the single most dangerous day of the 18 years of childhood is the day of birth.

    There is no way to eliminate the danger of childbirth, but there are ways to reduce it and there are ways to shift more of the risk to the mother in exchange for less risk to the baby. C-section is less dangerous for babies than vaginal delivery. That doesn’t mean that every baby will benefit from a C-section; many will be fine without it, but this is the basis of our rising C-section rate. Many babies who will die during a vaginal delivery will be saved by C-section. No babies who would die during C-section will be saved by a vaginal delivery.

    The neonatal mortality statistics are clear on this point. That’s why researchers have switched to looking at morbidity. It is certainly true that babies accidently delivered prematurely by C-section have an increased risk of TTN (transient tachypnea of the newborn). The problem though, is the prematurity, not the C-section.

    Moreover, TTN, as its name indicates, is a transient, relatively benign issue. When the biggest indictment against C-section is TTN, we’ve accepted that C-sections are extremely safe for babies.

  32. Zoe237,

    Let’s look at the studies you cited.

    The first study,Neonatal Outcomes After Elective Cesarean Delivery, is not about elective primary C-sections. It is a comparison of VBAC to elective repeat C-sections. There were no deaths in any group.

    The second paper, Cesarean Delivery on Maternal Request, is a position paper prepared by a private contractor for the Agency for Healthcare Research and Quality. It is not a scientific paper and it is not a government guideline. It is merely one submission to an overall review.

    The paper is a systematic review of potentially related information since, as the authors acknowledge, there are no studies of Cesarean on maternal request. The authors conclude “The evidence is significantly limited by its minimal relevance …” I’m not sure exactly why you cited it.

    The other citations you offered (one is actually a press release, not a paper at all) don’t provide any new or different data.

    My personal feeling on C-sections on maternal request is that they are inappropriate. I never did them when I was practicing because, in my judgment, medical procedures should only be reserved for medical indications. I did not do social inductions, either, for the same reason.

    I note with some amusement, though, how “natural” childbirth and homebirth advocates have tied themselves into knots over this issue. They argue forcefully that a woman’s autonomy justifies any reproductive choice, even those (like homebirth and waterbirth) that are known to be dangerous and even fatal for babies. Yet women’s autonomy seems to be respected only if the choices are approved. When a woman requests a C-section for no other reason than because she wants one, suddenly her judgment can no longer be trusted.

    You can’t have it both ways. If autonomy justifies a woman’s right to risk her baby’s life at a homebirth, it justifies her desire for a C-section on maternal request.

  33. desiree says:

    when i was first researching VBAC vs. repeat c/s for the birth of my second child, i read through as many papers on VBAC vs. repeat c/s as i could find. i can’t recall ever seeing a mortality rate as high as MacDorman found in the c/s group, which led me to really doubt those results. it also led me to wonder whether it would be ok, research-wise, to do a cochrane-review style analysis of the VBAC vs. ERCS papers to get a handle on the neonatal mortality rate of the c/s groups. i mean, you have all that data on low risk moms, is there anything wrong with using it for such a closely related purpose?

    secondly, i’ve heard the claims that c/s can raise the risk of infection in the first year because the baby gets less antibodies from the mother. is there any research that looks at outcomes in the first year of life?

  34. “it also led me to wonder whether it would be ok, research-wise, to do a cochrane-review style analysis of the VBAC vs. ERCS papers to get a handle on the neonatal mortality rate of the c/s groups. i mean, you have all that data on low risk moms, is there anything wrong with using it for such a closely related purpose?”

    Repeat C-sections differ from primary C-sections in important ways, and VBAC differs from vaginal delivery in important ways, so a direct comparison is not appropriate. For example, VBAC has a much larger risk of uterine rupture than the risk of rupture of an unscarred uterus making the calculation of risks and benefits very different.

    “i’ve heard the claims that c/s can raise the risk of infection in the first year because the baby gets less antibodies from the mother. is there any research that looks at outcomes in the first year of life?”

    There’s no difference in transmission of antibodies between vaginal delivery and C-section because the transmission occurs during pregnancy, not during labor.

    Perhaps you are referring to the claim that the gut flora of babies delivered by C-section differs from that of babies delivered vaginally. There is some evidence that the gut flora differs in the first year and claims have been made about allergic diseases and other disease, but there is no definitive evidence to support these claims of disease.

  35. Calli Arcale says:

    I read about a study a couple of years ago in which researchers analyzed the poo of infants. (Their own, as it happens, as their own infants provided a very convenient source.) Obviously there are blinding issues, but they were just curious how the fecal flora would change over the course of a year. They found that in all the infants (some breastfed, some bottlefed, delivered in a variety of circumstances), their gut flora changed almost completely by their first birthday. In other words, it may not matter particularly if an infant doesn’t go through the birth canal. Obviously, the babies are picking up new flora all the time in their environment anyway.

    What it comes down to, I think, is that it’s much easier to pick up fecal flora from our environment than we like to believe.

  36. Calli Arcale says:

    Addendum: I forgot one detail. The scientists found that the gut flora changed almost completely not just once but *several times*.

    Ancedote: my two children were born by c-section and turned out just fine. ;-) (Indeed, their birth certificates would not give you a clue as to why they were born that way. Yet neither was a planned c-section. It was decided after arrival at the hospital, due to medical indications. I have no doubt whatsoever that this very major confounder is overlooked when people try to count the number of “unnecessary” c-sections.)

  37. “I have no doubt whatsoever that this very major confounder is overlooked when people try to count the number of “unnecessary” c-sections.”

    What is particularly distressing is that the authors had every reason to know that “no indicated risk” did not mean “no risk” yet they implied that it did. Both MacDorman and her co-author Declercq have publicly criticized the current C-section rate and have injected that criticism into their science. For example, MacDorman’s recent paper on premature birth implied that the C-section rate was responsible for an increase in deaths of premature infants. They presented no evidence to support that, and, indeed, there is no evidence to support that.

    Unfortunately, ideology often trumps facts in discussions and even scientific papers about contemporary issues in obstetrics. This further erodes trust between patients and doctors.

  38. diana says:

    I would suggest to replicate this study here in italy: in some cities, nearly 60% of deliveries are done using C-section..

Comments are closed.