Reviewing the Newest Research on 

VBAC After Multiple Cesareans (VBAMC)

by Kmom

Copyright © 2005-7 Kmom@Vireday.Com. All rights reserved.

This FAQ last updated: June 2007



ACOG's Revised Guidelines on VBAMC

In July 2004, the American College of Obstetricians and Gynecologists released their revised VBAC guidelines.  In it, ACOG withdrew its previous support for a trial of labor in women with 2 prior cesareans.   They stated:

For women with 2 prior cesarean deliveries, only those with a prior vaginal delivery should be considered candidates for a spontaneous trial of labor.

This change of policy has led to many women being forced into needless repeat cesareans, ignoring the very real risks that further cesarean surgeries pose to mother, baby, and future babies.  It also ignores the mother’s fundamental right to informed refusal and to make her own healthcare choices.

Why the Change

ACOG justified their change in policy by citing a study (Caughey 1999) that found the risk for uterine rupture (UR) in women with 2 prior cesarean deliveries to be 3.7%.  After adjusting for confounding factors, the rupture risk was found to be nearly five times higher in women with 2 prior cesareans (VBA2Cs).  Thus, they felt justified in denying a trial of labor (TOL) to women with more than one prior cesarean.

However, by choosing to highlight only this study, they neglected the many other studies that found much lower risks for uterine rupture.  Furthermore, the Caughey study was extremely small (only 134 trials of labor after 2 cesareans), leading to the possibility that the risk was inflated due to inadequate numbers in the VBA2C group.  It is bad science to make sweeping policy changes on the basis of 134 patients; far larger groups are needed for policy-making decisions with such far-reaching health consequences.

If you look at other, larger Vaginal Birth After Multiple Cesarean studies, most have found a uterine rupture risk far smaller than the Caughey 1999 study, yet these studies were mostly ignored by the new guidelines.  For example, Asakura (1995) found a UR rate of 1.0% among 302 TOLs, a data set 3 times as large as the Caughey study.  Leung (1993) found a UR rate of 2.0% among 1,165 TOLs, a data set more than 8 times as large.  The largest study of VBAMC to date (Miller 1994) found a UR rate of 1.7% among 1,827 TOLs, a data set nearly 14 times as large as the Caughey study.

All three of these studies found rupture rates between 1-2% (about half that of Caughey 1999), and their larger data sets mean their results are far more statistically powerful.  Yet their results were ignored in favor of a trial of only 134 VBA2C patients.

Not all studies have found higher rupture rates.  Chattopadhyay (1994) found a 0.0% UR rate among 115 VBAMC TOLs (similar in size to Caughey’s study), while Phelan (1989) found a UR rate of 0.0% among 501 TOLs, a sample size nearly four times as large as the Caughey study group.  Yet these studies were also completely ignored in the new guidelines.

In its guidelines, ACOG does briefly mention the Asakura and Miller studies, but dismisses them immediately, noting that only the Caughey study controlled for potentially confounding variables.  Although controlling for confounding variables is desirable, it was a less common technique in older research.  Not controlling for variables does not invalidate this research---it just needs to be considered with the understanding the other factors may play a role too. In this situation, the confounding variable issue is just a smokescreen, used to justify dismissing inconvenient data. 

Another concern is why the Caughey study had rupture rates about twice the rate of all the other studies. When one study shows rates that are out of line with the rest of the literature, its results must be questioned. Yet this is completely unacknowledged by the new guidelines.

In summary, the new ACOG guidelines based its new ban on VBA2C almost entirely on the results of one study (which was extremely small and whose results were significantly out of line with other studies), while completely ignoring or dismissing other, larger studies on the topic. 

Selective consideration of the data is not good science, and major policy changes should not be placed on the shoulders of only 134 patients.  One must question whether the conclusions in the new guidelines were changed because of factors like medico-legal concerns instead. 

The Prior Vaginal Birth Factor

ACOG’s new guidelines do permit one exception to the VBA2C ban. Women who have had a prior vaginal birth are still considered candidates for a trial of labor after 2 prior cesareans.

This exception is based on another study from the same authors and database as the Caughey 1999 study.   In this study (Zelop 2000), the uterine rupture rate was found to be much lower in those who had already had a vaginal birth (2.5% vs. 3.9%).  Note that the 2.5% rate considered “acceptable” in these mothers was higher than in most of the other VBAMC studies whose 1-2% rupture rates were dismissed.

Even so, once again the rupture rate in patients from this Caughey/Zelop/Shipp database far exceeds those in other studies, and no one is asking why. The issue here is less whether the uterine rupture rate is a reasonable risk in VBA2C mothers who have had a prior vaginal birth; the issue should be why the rate in this group is so out of line with that of other research, and why the research is being considered so selectively. 


VBAC After More Than Two Cesareans

In the past, while ACOG guidelines did not exactly advocate for VBAC after 3 or more cesareans, it did not ban them either.  The 1994 guidelines stated:

A woman who has had two or more previous cesarean deliveries with lower uterine segment incisions and who wishes to attempt vaginal birth should not be discouraged from doing so in the absence of contraindications

Although ACOG did not exactly endorse VBAMC then, it was left up to the judgment of the managing physicians and the birthing mother. Now, of course, that decision has been taken out of their hands. 

The risk of rupture in higher order VBACs is unfortunately largely unknown.  Higher order VBACs did occasionally happen in the past, but not in any kind of systematic, studied way.  Several studies have had small numbers of VBAMC TOLs, but the sample size was not large enough for any real conclusions.

There is only one substantial study of higher-order VBACs.  Miller (1994) had 1,827 TOLs in women with 2 or more cesareans.  Of these, there were 241 TOLs in women with 3 or more prior cesareans (VBA3+Cs).  Overall, the rupture rate was found to be 1.7% in all VBAMCs combined; 1.8% in VBA2Cs, but only 1.2% in VBA3+Cs.  This seems to contradict the theory that rupture risk rises linearly as the number of prior incisions rises.  However, these labors may just have been managed with more caution (i.e., less induction and augmentation), thus decreasing the risk of rupture.  Without more details, we cannot know.  But the 1.2% rupture risk in the VBA3+C group is close to the 1% “acceptable” rupture risk commonly cited for VBA1C.  Therefore, higher order VBACs should not be categorically denied either. 

Although ACOG does not currently endorse a TOL in most VBAMC mothers, some women are still managing to have higher order VBACs.  The highest-order VBAC documented in the medical literature in the past is a VBAC after 5 cesareans (Veridiano 1989).  The recent Landon study (2006) documented 104 TOLs in women with 3 or 4 cesareans.  There are anecdotal stories of VBA4C, VBA5C, even VBA7C births.

More data is urgently needed to properly judge the risks in higher order VBACs, but the data we have so far suggests that a trial of labor should not automatically be ruled out.  Furthermore, any consideration of possible uterine rupture risks in higher order VBACs must also be balanced against the substantial risks of continuing cesareans.


VBAMC Research Since The 2004 Guidelines

Since the revised VBAC guidelines came out, several new studies on VBAMC have been published. This new data calls out for the guidelines to be revised to permit VBAMC trials of labor once again. 

These new studies, like before, show a wide range of UR results.  For example, Garg (2004) found a rupture rate of 0.0% in a small study of 100 TOLs, while Lieberman (2004) found a much higher 3.0% rupture rate in a similarly sized study that had 99 TOLs after multiple prior cesareans.  Like the Caughey (1999) and Chattopadhyay (1994) studies, the small size of these studies raises the question of the reliability of their findings.  The results (either the 0% or the 3%) could well be skewed by the smallness of their data sets. 

On the other hand, three new VBAMC studies since 2004 are based on much larger data sets, and thus their conclusions are far more powerful.  

Note also that the results of these three large studies fall mostly within the 1-2% rate that was found in the largest pre-2004 data.  In fact, if only the most statistically powerful studies are considered from both time periods, the results show a uterine rupture risk largely between 1-2%. 

TABLE A:  Summary of the Largest VBAMC Studies

Study Name

Number of TOLs

Uterine Rupture Rate

Asakura 1995



Phelan 1989



Lin and Raynor 2004



Landon 2006 N=975 0.9%

Macones 2005



Leung 1993



Miller 1994




Although further research would still be desirable, the most statistically powerful studies then and now show a much lower risk level than the Caughey (1999) study that was used to justify banning most VBA2Cs.  This alone is reason enough to reconsider the ban, but further digging into the research shows that the risk may be even lower if only spontaneous labors are considered. 


Labor Interventions and Rupture Risk

The research figures presented so far represent the rupture risk in studies with high rates of induction or augmentation. This may be inflating the risk artificially.

In women with only one prior scar, the risks of uterine rupture are much lower with totally spontaneous labor (Zelop 1999).  So when deciding policy about whether a trial of labor in women with 2 prior cesareans is feasible, the crucial questions become:

The wide disparity in rupture rates in VBAMC studies (0.0%-3.7%) suggests that differences in labor management may have been important. Leung (1993), for example, found that aggressively augmenting VBAC mothers in early irregular labor strongly increased the risk of uterine rupture. Had these mothers been allowed to labor spontaneously, the rupture rate in this study might have been much lower. 

Personal correspondence with the authors of the Caughey (1999) study reveals that all 5 ruptures in the study were found in the group that was augmented; the rate of rupture in the unaugmented, uninduced group was 0.0%.  However, the numbers in this group are very small.

The trend towards more ruptures in intervention groups was confirmed in Macones (2005) as well; 16 of the 19 uterine ruptures in the VBA2C trial of labor group occurred in labors that were induced or augmented.  If the 65% of the participants who were augmented/induced in that study had been allowed to labor spontaneously instead, the overall rupture rate in that study (1.8%) would likely have been much lower. 

In Lin and Raynor (2004), the risk of rupture in spontaneous labor was doubled between the VBA1C group and the VBA2C group (0.4% vs. 0.8%), while the risk for VBAMC rupture in induced labor was nearly tripled (1.0% vs. 2.7%).  

Although more study is needed, preliminary research suggests that inducing or aggressively augmenting women with multiple prior scars may disproportionately increase their risk of rupture compared to women with only one prior scar. Because most of the VBAMC research so far does not separate out these factors, the risk for spontaneous VBAMCs may well be far less than it appears in the literature.


Spontaneous Labor Rupture Risks Lower

Unfortunately, there are limited studies on the risk for uterine rupture in totally spontaneous labor VBAMCs.  Inductions and augmentation are such an integral part of most hospital VBACs that little attempt has been made to discern their risks from those of totally spontaneous VBACs.  However, we now finally have some data with which to gauge the true risk associated with spontaneous VBAMCs.

In Lin and Raynor (2004), 596 women with two or more prior cesareans had a trial of labor. There were 6 ruptures, for a total VBAMC rupture rate of 1.0%. This is a relatively low rate, probably because 88% (n=523) of the mothers labored spontaneously.  This smaller spontaneous-labor VBAMC group had a rupture rate of only 0.8%.  This is well within the 1% rupture rate considered “reasonable” for VBA1C mothers.  Furthermore, the study does not address whether women in the spontaneous labor group were augmented with pitocin; many probably were.  Therefore, the rate for uninduced, unaugmented VBAMC labor in that study may have been even lower. 

There are two other studies that may bear some light on spontaneous VBAMC labors, one encouraging and one not. Lieberman (2004) studied VBAC labors that were planned for birth centers and found increased risk for VBAMC mothers.  They had records for 99 trials of labor in women with 2 or 3 prior cesareans; there were 3 ruptures in this group, for a much higher rupture rate of 3.0%. Presumably, all of these women should have been in spontaneous labor because they were in a birth center. On the other hand, there are many “natural” ways to induce/augment labor, and there have been case reports of castor oil, mechanical inductions, etc. being associated with rupture. Also, like Caughey (1999), the VBAMC arm of this study is very small (only 99 TOLs); chance can easily throw off the results in such small groups.  The results cannot be dismissed, but they should be viewed with caution. 

On the other hand, the results of Macones (2005) were more encouraging.  It did not directly address the question of uterine rupture rates in spontaneous vs. induced/augmented labor.  However, by using the statistics that are provided in the study, approximate rates can be deduced.  There were 3 ruptures in approximately 379 spontaneous VBAMC trials of labor, for a uterine rupture rate of 0.8%.  This is exactly the same rate as the Lin and Raynor 2004 study. 

So we have data from three studies on the rate of rupture in spontaneous labor in women with 2 or more prior cesareans.  One very small study (n=99) found a rupture rate of 3.0%, while two much larger studies (n=523, n=379) found rupture rates of 0.8%.  Although more research specific to spontaneous VBAMC labor needs to be done, the more statistically powerful studies should be regarded as more reliable for now and suggest that a careful reconsideration of the ban on VBAMC is in order. 


Risks of Repeat Cesareans

No discussion of whether VBAMC should be allowed is complete without consideration of the risks of elective repeat cesareans.  ACOG’s VBAC guidelines barely mention the risks of multiple repeat cesareans. This is a grievous omission because any decision between VBAC and elective repeat cesarean must consider all the possible risks of both choices.  Instead the guidelines reveal their authors’ bias by concentrating almost entirely on the risks of VBAC.

The risks of cesareans include not only immediate operative complications but also fetal complications and long-term maternal morbidity. Even an elective cesarean increases the risk of maternal death, hemorrhage, thromboembolic events, hysterectomy, intestinal complications, anesthesia complications, adhesions, and long-term pain.  Infants of elective cesareans also experience increased risk from iatrogenic prematurity, fetal lacerations, and significant fetal respiratory morbidity.  If the mother goes on to have more children, there are significant risks to future pregnancies, including ectopic pregnancy, placenta previa, placenta accreta, and placental abruptions.  These are life-threatening to both mother and future babies. 

Macones (2005) notes that his study only examines the short-term consequences of VBAMC vs. repeat cesarean.  He points out:

As the number of prior cesareans increase in subsequent pregnancies, surgical complications increase, as do rates of placenta previa and accreta.  Thus there may be long-term implications to consider in strategies that include multiple repeat cesareans (especially for women planning large families).

If women are forced into mandatory repeat cesareans on a large-scale basis, the complications will start accumulating.  With enough repeat cesareans, even rare complications like maternal death will rise; women will die.  Others will lose their fertility due to hysterectomy from hemorrhage or placenta accreta.  Still others will live with life-long pain and disability from internal scar tissue.   

The maternal morbidity associated with forced repeat cesareans may extend throughout a woman’s entire reproductive life, and will increase with each successive cesarean.  Because of this, both the risks of VBAMC and multiple elective repeat cesareans must be considered equally in any policy decisions.  The current VBAMC guidelines do not do this. 



Before 2004, ACOG supported a trial of labor after at least 2 prior cesareans.  After 2004, it did not.  This decision has far-reaching implications for maternal-child health.  The question is what was behind this big reversal in policy. 

In all likelihood, there were two main factors in ACOG’s decision.  First, ACOG’s committee chose to center all its attention on a small study of limited statistical power that found a much higher rate of rupture than expected.  Since one of the authors of that small study (Dr. Caroline Zelop) was a primary consultant during the revision of the VBAC guidelines, her study was given strong consideration.  In doing so, the committee conveniently ignored a number of larger studies that found a much lower risk for rupture.  Such selective consideration of the data is bad science and should not be the basis for such sweeping changes of policy.

Second, in the current litigious climate, doctors have gotten even more skittish about VBACs.  Since most OBs are not familiar with the wide spectrum of VBAMC data and the small Caughey study was heavily publicized, many perceive VBAMC as “too risky.” Because ACOG is at heart a trade union to protect the interests of its members, revised guidelines have now been adopted that give its members legal grounds to justify avoiding VBAMCs.

VBAMC was not suddenly discovered to be “unsafe” or far more dangerous than previously believed. Although the Caughey study suggests caution, the preponderance of data shows that VBAMC is still a reasonable choice.   This is underscored by the fact that all of the VBA2C studies state that choosing a trial of labor after 2 prior cesareans is a reasonable choice, given appropriate counseling and informed consent. Even the Caughey study concludes, “A trial of labor in patients who have undergone two previous cesareans appears to be a reasonable consideration.”

The problem does not lie with new VBAMC research. That is largely positive. The problem is that the political climate for VBACs has changed so strongly in the last few years that few doctors are willing to attend VBACs after one cesarean, let alone after two or more cesareans.   They are looking for any excuse to limit the VBACs they attend, and these new VBAC guidelines give them the ammunition they need. 

The research shows that, although the rupture risk probably is increased somewhat in VBA2C mothers over VBA1C mothers, this risk is still well within reasonable limits, particularly in women with spontaneous labors.  Furthermore, any rupture risk must be balanced against the significant morbidity associated with further surgeries.  When all the risks are weighed, VBAMC continues to be a reasonable choice for well-informed and motivated women.

This is recognized in the journal discussion by Dr. Macones after his VBA2C study.  He states:

The recent ACOG Practice Bulletin on VBAC…suggests that for women with 2 prior cesareans, only those with a prior vaginal delivery should be allowed a trial of labor.  Our data do support that a prior vaginal delivery is protective: however, the absolute risk even in those without a prior vaginal delivery is quite small. Thus, it seems reasonable to consider VBAC in those with 2 prior cesareans with no prior vaginal delivery, especially if they go into labor spontaneously.

Abandoning VBAMC for legal expediency unconscionably forces mothers into unwanted and unnecessary surgery, exposing them and their infants to a high risk of both short-term and long-term complications.  This is a long way from the Hippocratic Oath of “First, do no harm.”  ACOG needs to de-politicize its research reviews, put the safety of mothers and babies ABOVE the litigious concerns of its members, and respect the right of each mother to make an informed decision about her mode of birth.

Vaginal Birth After Multiple Cesareans remains a reasonable choice for women who desire a VBAC, and should continue to be supported by healthcare providers.



Phelan JP et al.  Twice a cesarean, always a cesarean?  Obstetrics and Gynecology  February 1989; 73(2):161-5.  

Veridiano NP.  Vaginal delivery after cesarean section.  International Journal of Gynaecology and Obstetrics  August 1989;29(4):307-11.  

Flamm BL et al.  Vaginal birth after cesarean delivery: results of a 5-year multicenter collaborative study.  Obstetrics and Gynecology  November 1990;76(5 Pt 1):750-754.

Leung AS et al.  Risk factors associated with uterine rupture during trial of labor after cesarean delivery: a case-control study.  American Journal of Obstetrics and Gynecology  1993;168:1358-63.  

ACOG Committee Opinion.  Vaginal delivery after a previous cesarean birth.  #143, October 1994.

Chattopadhyay SK et al.  Planned vaginal delivery after two previous cesarean sections.  British Journal of Obstetrics and Gynaecology  June 1994;101(6):498-500.  

Miller DA et al. Vaginal birth after cesarean: a 10-year experience. Obstetrics and Gynecology August 1994;84(2):255-8. 

Asakura H and Myers, SA.  More than one previous cesarean delivery: a 5-year experience with 435 patients. Obstetrics and Gynecology  June 1995;85(6):924-9.

Davies GAL et al.  Vaginal birth after cesarean: physicians' perceptions and practice. Journal of Reproductive Medicine  July 1996;41:515-520.

Caughey AB et al.  Rate of uterine rupture during a trial of labor in women with one or two prior cesarean deliveries. American Journal of Obstetrics and Gynecology  October 1999;181(4):872-6.

Zelop CM et al. Uterine rupture during induced or augmented labor in gravid women with one prior cesarean delivery. American Journal of Obstetrics and Gynecology  October 1999;181(4):882-6.

Zelop CM et al. Effect of previous vaginal delivery on the risk of uterine rupture during a subsequent trial of labor.  American Journal of Obstetrics and Gynecology  2000;183(5):1184-6.

Spaans WA et al. Trial of labour after two or three previous caesarean sections. Eur J Obstet Gynecol Reprod Biol  Sept 10, 2003;110(1):16-9.

Mankuta DD et al.  Vaginal birth after cesarean section: trial of labor or repeat cesarean section?  A decision analysis. American Journal of Obstetrics and Gynecology  Sept 2003;189(3):714-9. 

ACOG Practice Bulletin: Clinical Management Guidelines for Obstetrician-Gynecologists.  Vaginal birth after previous cesarean delivery. #54, July 2004. 

Garg V and Ekuma-Nkama E.  Vaginal birth following two cesarean deliveries---are the risks exaggerated?  Ann Saudi Med  July-Aug 2004;24(4):276-9. 

Lieberman E et al. Results of the national study of vaginal birth after cesarean in birth centers. Obstet Gynecol  November 2004;104(5):933-42.

Lin C and Raynor D.  Risk of uterine rupture in labor induction of patients with prior cesarean section: an inner city hospital experience. American Journal of Obstetrics and Gynecology  2004;190:1476-8. 

Macones GA et al. Obstetric outcomes in women with two prior cesarean deliveries: is vaginal birth after cesarean delivery a viable option? American Journal of Obstetrics and Gynecology  2005;192:1223-9.

Landon, MB et al.  Risk of uterine rupture with a trial of labor in women with multiple and single prior cesarean delivery.  Obstetrics and Gynecology.  July 2006;108(1):12-20.


Copyright © 2005-7 All rights reserved. No portion of this work may be reproduced or sold, either by itself or as part of a larger work, without the express written permission of the author; this restriction covers all publication media, electrical, chemical, mechanical or other such as may arise over time.

[ Back to Kmom Area ]