Thursday, February 12, 2009

Ultrasound Measurement of Cesarean Scar Thickness

A recent press release trumpeted the "new" findings of a Canadian researcher that cesarean scar thickness may predict the risk of uterine rupture in future pregnancies. Press releases stated it might be used to predict who should be encouraged to try for a Vaginal Birth After Cesarean (VBAC, pronounced "vee-back") or who should just schedule a repeat cesarean.

I was asked about this here on my blog, too, so I have prepared a detailed post about the pros and cons of this topic.

A word of warning--this post will be fairly technical and long. If you find that sort of thing boring, please feel free to skip this post. I promise, not all my blogging is like that. But sometimes it's important to get into details, and this is one of those times.

The Cliff Notes Version

In a nutshell, beware scientists with a self-promotion agenda.

Remember, this is the era of "Science By Press Release;" this particular press release coincided with a presentation on the topic at the Society for Maternal-Fetal Medicine's annual meeting in San Diego in January. The study hasn't been peer-reviewed or even published yet. It's a little premature to be drawing conclusions or altering policy from it.

Although the findings are interesting, caution should remain the byword in the use of ultrasound to predict the risk of uterine rupture. There are a number of very legitimate concerns about its accuracy, its utility, and the misuse of results that might occur with its widespread adoption.

Remember the take-away message:

Be cautious about using scar thickness to estimate uterine rupture risk. At this point, it deserves further study, but it should not be used to determine who should (and shouldn't) get a chance at Vaginal Birth After Cesarean.

A Little Background Information

First, a few explanations are in order. Birth professionals who read this blog will understand the terms being tossed around, but others may not. So let's take a moment to do a little explaining.

First, remember that one of the risks after a cesarean is that the scar from the incision may come apart in the next pregnancy (the risk is somewhat higher with labor but exists even without labor). This risk is very small, but potentially very serious. However, there are different degrees of separation that must be differentiated.

Terms for scar separation tend to be very inconsistently used, even by medical professionals, but generally fall into a few categories.

Markedly Thin Lower Uterine Segment

Sometimes the prior cesarean scar gets very thin but doesn't separate. This is usually called a "markedly thin" or "paper thin" lower uterine segment. Occasionally some sources will call it a "window" because they can 'see' the baby's hair or features, like a face pressed up against a nearly opaque window. However, "window" is a very inconsistently used term so other terminology is generally preferable.

It is unclear whether or not a "markedly thin" lower uterine segment (LUS) is risky. Many doctors assume it is a disaster about to happen, but there have been a number of women who have experienced this and still gone on to have a VBAC (with no rupture) later on---so obviously those LUS were stronger than the doctors thought.

On the other hand, in some cases, it might have been something about to happen---the problem is we just don't know for sure. Some thinning of the LUS is normal at the end of pregnancy and during labor; it is unclear whether at some point thinning becomes abnormal, and if so, at what point that happens. It is a matter of some disagreement.


Sometimes the scar actually comes open, like a zipper coming a bit unzipped, but the separation is mild and has very little bleeding. This is a "dehiscence" (although some sources call this a "window;" the inconsistency of that term is why it is best avoided).

One medical definition of dehiscence is:

Any defect in the preexisting cesarean scar with no maternal or fetal compromise

Although potentially serious, a dehiscence generally has a good outcome and is clinically fairly insignicant.

Uterine Rupture

Other times, the scar separation is more significant (a "uterine rupture" or UR) and the woman can have significant bleeding and difficulty recovering. A common medical definition of uterine rupture is:

A defect that involved the entire wall of the uterus, was symptomatic, and required operative intervention

A uterine rupture is always dangerous and needs attention, but many babies and mothers are fine with prompt treatment.

On the other hand, sometimes uterine rupture is truly catastrophic, especially if the placenta pulls away from the uterus during the rupture, depriving the baby of oxygen. The baby can sustain brain damage or even die, and the mother can lose her uterus or die too.

The risk of truly catastrophic uterine rupture is small, but if it happens to you, it's devastating. It is certainly something to be taken very seriously indeed.

Weighing the Risks

Although VBAC has a small but potentially very serious risk of uterine rupture, there are also risks from choosing repeat cesareans, including severe hemorrhage, anesthesia accidents, infection, hysterectomy, serious breathing problems for the baby, and maternal death. Furthermore, in future pregnancies, the risk of the placenta implanting abnormally rises significantly, and this can lead to prematurity, fetal death, maternal hysterectomy, and maternal death.

In many years of birth-related work in person and on the internet, I have known women and babies harmed both by uterine rupture during labor and by repeat cesarean and its complications. Let me assure you, I take both very seriously. I urge readers to have the utmost respect and compassion for all those who have their lives impacted by either.

You can debate the relative safety of VBAC vs. repeat cesareans for a long time, but the bottom line is that there is NO 100% "safe" choice after a cesarean. There is risk with trying for a VBAC, and there is risk from deciding on a repeat cesarean.

Although the risk of a severe complication with either choice is low, if it happens to you, it is overwhelming and devastating.

That's why it's so important to use cesareans only when truly needed.

The Allure of the Crystal Ball

In an ideal world, doctors would be able to figure out ahead of time who is most at risk for uterine rupture so those women could opt for a repeat cesarean, while the others could opt to try to VBAC if they wanted (called a "Trial of Labor" or TOL).

Much of the VBAC research in the last few years has been aimed at trying to determine the most important risk factors for uterine rupture. The problem is that a lot of research on UR is contradictory; there is no "smoking gun" study that shows a clear way to predict or avoid uterine rupture.

Even when a risk factor for UR is found, it's only a risk factor and not a true predictor; the vast majority of women with that risk factor will not experience a rupture even if they labor. If you insist on mandatory ERCS for all women with risk factor "X," the majority of those repeat cesareans will not have been necessary and will expose all those women to the considerable risks of repeat cesareans while preventing only a very few ruptures.

Coming up with a reasonable way to manage risk in women with a prior cesarean is one of the great dilemmas of obstetrics today. Frankly, the best solution is to prevent the first cesarean whenever possible, but with cesarean rates at around 1 in 3 of all childbearing women, this is not happening.

Failing primary prevention, the best course is to offer fully informed consent about risks and benefits of each choice, and then let the woman choose which option to pursue. In the end, the decision should be the woman's.

Measuring Cesarean Scar Thickness - The Bujold Study

Measuring cesarean scar thickness is one way doctors try to predict the risk of uterine rupture and know who would be the "best" candidates for a trial of labor and who might be at more risk.

Bujold's study, done in Quebec, measured scar thickness in 236 women. They found that a cut-off of 2.3 mm helped determine a group more at risk for UR in their study group.

Since the study has not even been published yet, it is difficult to evaluate. Here are some of the details given in the press release:

Bujold's study involved 236 pregnant women who had delivered previously by C-section but who planned a vaginal delivery. They used ultrasound to measure the lower part of the uterus, which correlates with scar thickness from the previous C- section, and then followed the women through their deliveries.

During labor and delivery, three of the women had a complete uterine rupture. In six, the scar reopened. Women who had uterine rupture had a very thin scar, Bujold said.

"We found the cutoff is probably 2.3 millimeters" in terms of scar thickness, he said. The average risk of rupture is about 1 percent, Bujold said, but in the study, "if you had a scar smaller than 2 mm, your risk of rupture [was] about 10 percent."

Please note, it's important to look at the strengths and weaknesses of this particular study....and there are several to look at.

The first problem here is that this study is small; the study group had only 236 women. You need a much larger study group than that to determine the significance of any particular risk factor on such a rare complication.

Second, the study has a somewhat higher-than-usual underlying rupture rate. Whenever a study reports a higher-than-usual complication rate, it's always important to dig deeper. A high UR rate suggests that some other factors may be at work.

A large body of research shows that induction and augmentation (using artificial drugs to start or strengthen labor) significantly increase the risk of rupture. This is particularly true if multiple types of induction drugs are used, or if the mother has never had a prior vaginal birth.

Uterine rupture risk is often quoted as being around 1% in a TOL, but that averages together both induced/augmented labors and spontaneous labors. (In VBAC labors that are spontaneous, rupture rates usually hover around 0.5%, or half of a percent.) In this study, the rate of actual ruptures was 3/236, or about 1.3%. Therefore, in all likelihood, factors like induction and augmentation are strongly at work here too. We don't yet know if they controlled for those factors.

It would be interesting to know the details of the cases of the 3 ruptures and the 6 dehiscences, to find out how many involved induced or augmented labor. In many studies, the majority of ruptures and dehiscences involve artificial strengthening of labor contractions.

The other big variable here was partly addressed by Bujold's presentation; the type of suture repair done. This is another giant controversy in VBAC these days; one-layer vs. two-layer repair of the uterus.

Bujold has done research in the past showing that a one-layer repair of the uterus strongly increased the risk of rupture in later pregnancies. However, other studies have not found similar results. It all depends on the study you look at.

A further variable not accounted for in most one-layer vs. two-layer debates is the TYPE of suture material. Bujold and his team used a different type of suture material than the one-layer studies done in the U.S.; his group's higher rate of rupture may simply have to do with the TYPE of suture material rather than the number of layers used. Alas, there has yet to be a definitive study on this topic that controls adequately for other factors.

In this study (as reported here), Bujold found that the combination of a single-layer repair and a scar thickness less than 2.3 mm strongly increased the chances of uterine rupture (21.8 times the risk). That's a very strong increase of risk, which definitely deserves further study.

But again, other studies on single-layer sutures have not found the same level of risk with single-layer suturing. Would other studies measuring scar thickness in single-layer sutured mothers find a similar increase? We just don't know. In addition, we don't know if the type of suture material, pattern of stitch, and induction/augmentation status were controlled for.

This is far from a definitive study on this topic and should be taken with a large grain of salt. Its results merely call for further study, not changes in hospital policy.

The Problems with Studies Measuring Cesarean Scar Thickness

There are a number of problems with studies that have been done using cesarean scar thickness as a predictor of uterine rupture.

The first major problem with this is the issue of inter-observer variation. The type of measuring done here is fairly subjective, particularly between different observers and different methods. Transvaginal ultrasound seems to be more accurate than abdominal ultrasound, yet often abdominal ultrasound is what's being used. Should a woman's chance at even trying for a VBAC rest on data that can vary significantly depending on who (and what) is doing the measuring?

As noted above, another significant problem with these studies are their small sizes. From a discussion of the topic on the VBAC Facts blog:

Where we do draw the line at what is “thick enough?” This is where studies come into play. There are several studies that focus on measuring uterine thickness via ultrasound on women with prior cesareans...but none of them are large enough to make any definitive decisions.

When looking at something like uterine rupture that happens about half of a percent of the time, you need to include thousands of test subjects in order to get an accurate assessment of the frequency of the occurrence. We just don’t have that here.

These are interesting preliminary studies that should be duplicated using thousands of women. If there is a way to accurately predict which scars will rupture, this is important information to have, but there is currently insufficient evidence available.

The largest study on uterine scar thickness is the main original one (Rozenberg et al, Lancet, 1996). It was the largest by far with 642 women, but even that falls considerably short of the thousands needed to have the power to determine the statistical significance of a particular risk factor on a rare occurrence.

Third and most importantly, each study finds a different cut-off spot where the risk for rupture increases and becomes "too much."

The original Rozenberg study from 1996 found that a cut-off of 3.5 mm was most useful in determining when risk went up; the recent Bujold study found that a cut-off of 2.3 was the most useful. That's a pretty fair descrepancy. How do you reconcile the difference?

In one study the cut-off is 3.5 mm, in another it's 2.5, 2.3, 2.0, 1.6, 1.5 or 1.0 mm. There are studies to support each of those cut-offs. Which cut-off do you choose to use?

Here's a list of several scar thickness studies in women with prior cesareans. Which one do you trust in?

Study and Cut-off Where Risk Went Up

Rozenberg, 1996 - 3.5 mm (n=642 women)
Qureshi, 1997 - 2.0 mm (n= 43 women)
Montanari, 1999 - 3.5 mm (n= 61 women)
Asakura, 2000 - 1.6 mm (n=186 women)
Suzuki, 2000 - 2.0 mm (n= 39 women)
Gotoh, 2000 - 2.0 mm (n=348 women)
Sen, 2004 - 2.5 mm (n= 71 women)
Cheung, 2005 - 1.5 mm (n=102 women)
Bujold, 2009 - 2.3 mm (n=236 women)

Other Experts Express Concern

The reason that most doctors are not doing this ultrasound scar measurement routinely already (despite the concept having been around for more than 10 years) is because they recognize the weaknesses of it.

In one 2003 survey, only 16% of Canadian doctors were using ultrasound to predict rupture risks. Doctors know the varying cut-off results means that it's not a very reliable method of determining risk.

These concerns were reflected by other doctors who commented on Bujold's press release.

Dr. Shoshana Haberman, director of perinatal testing services at Maimonides Medical Center in Brooklyn, N.Y., said she has been doing this measurement on women with previous C-sections for a few years. And while the new study results are interesting, she said, the prediction method is not yet definitive.

"We need more data -- that's the bottom line," Haberman said. "We need more data to decide the cutoff."

The ultrasound measure is also operator-specific, she added, so it could vary from person to person.

A Strong Potential for Misuse

In the press release, Bujold states that the study should be used to encourage more women to VBAC, given the known increase in risk with each repeat cesarean.

"There is a growing concern about the increase in cesarean births because there is a body of evidence showing that they are associated with higher rates of maternal and infant complications," said Emmanuel Bujold, M.D., with the Department of Obstetrics & Gynecology, Faculty of Medicine, Universite Laval, Quebec. "There are far fewer complications to the mother and infants as a result of a vaginal birth," he continued, "So it is important to determine when a patient with a history of prior cesarean section can have a vaginal birth safely."

However, although this statement sounds well-intentioned, it is disingenuous. Most of the time, these sorts of cut-offs are being used to DENY women access to VBACs, not encourage them. VBACs are extremely hard to come by these days in many areas; this will only be used as ammunition against them, not encouragement for them.

At best you could make a case for using scar thickness measurements as a way to strongly increase the trial of labor rate for women over some random cut-off, but it simply can't be used as a way to "guarantee" no rupture in a VBAC attempt. Nor does it guarantee a definite rupture in women who labor below the cut-off. There are cases of women rupturing above even the 3.5 mm cut-off, and many cases of women who have not ruptured below the arbitrary cut-offs set in these studies.

Alas, it's just not that simple, and no one has rupture-specific psychic powers. It would be wonderful if this were THE key to avoiding uterine rupture and encouraging more VBACs, but it is not.


There are too many problems yet for doctors to start doing universal ultrasound measurements of cesarean scar thickness and using them to determine who is "allowed" to have a trial of labor and who is "required" to have a repeat cesarean.

You might make a case for using this data to counsel women more closely about their possible risks, as long as you mentioned the strengths and weaknesses of the studies about it, and as long as the ultimate choice was up to the parents. Or it might be used for deciding who needs the most careful monitoring during labor after cesarean.

However, in reality, it's going to be used to DENY women the right to decide for themselves, either by requiring mandatory repeat cesareans in women whose scar thickness falls below an arbitrary cut-off, or by using the data to scare women out of considering a VBAC (without sufficient mention of the weakness of the data).

If a woman falls into a group that might be at increased risk for uterine rupture, the logical thing to do is to counsel those parents about their possibly increased risk from a TOL, as well as the possible current and future risks from repeat cesareans.

At this point, it should NOT be used for denying a woman the right to try to VBAC. Yet you know that's how it's going to be used. In fact, research shows that it HAS been used this way already, despite the many weaknesses of existing research.

Informed consent, yes.......coerced surgery, no.

Before this becomes standard of care, there needs to be a LOT of very large, multi-center, randomized, double-blind studies, using ultrasound measurements from multiple observers, and controlled for other factors like suture type/material, induction/augmention, etc.

Frankly, right now, measuring scar thickness is just another way to prevent or scare women from having a VBAC. The intent behind the investigation may be reasonable, but the pratical usage will not be.

Until there's a lot more study on this topic, it is not an accurate way to assess the risk for potential uterine rupture, and it should not be used to determine who should not be "permitted" to try for a VBAC.

Right now, it's just another data dredge and publicity ploy, rather than a really tested and true way of assessing future risk.

As Gretchen Humphries, advocacy director of the International Cesarean Awareness Network says, "It isn't anywhere close to clinically useful and we all know it'll get misused."

Just remember the take-away message:

Be cautious about using scar thickness to estimate uterine rupture risk. At this point, it deserves further study, but it should not be used to determine who should (and shouldn't) get a chance at Vaginal Birth After Cesarean.


Gretchen Humphries, Advocacy Director, ICAN said...

Excellent summary.

I would also add that in much of the research that comes from Bujold, et al., the overall uterine rupture rate runs higher than the generally expected rate (generated from decades of research on this topic, from a multitude of hospitals). I've often wondered what it is about this hospital and these researchers that elevates the overall uterine rupture does make me think twice about the research that this group publishes.

Anonymous said...

I have a family member who just had a scheduled repeat c/s. She traveled across the state to find a "VBAC specialist", who, in the last few weeks of her pregnancy, informed her that her scar was "very thin" and that her baby weighed over 10 lbs, and that she had a 100% chance of rupture and that either the baby or the mother would die.

The baby weighed less than 9 lbs (they were assured that this doctor could predict the baby's weight to within one ounce, because he's so precise and careful with his ultrasounds).

With the baby's weight being so far off from what the doctor had estimeted, I can't help but suspect that his estimation of the scar from ultrasounds was also incorrect. And then there's the fact that there's no good research showing us what the scar thickness does for us anyway.

Another unnecessarian.

Anonymous said...

I still think I'd get it done for myself just to know how thick I am. I've had one inverse T and two normal vaginal births before that. I would like to VBAC, but I want to know as much as I can. Thank you for the ideas you present as they gave me something to chew on.

Miss Momma said...

I know this is an old post of yours but it is very interesting and extremely pesonal to me. I had a 10 c dehiscense that was found upon incision. It was a full thickness separation and the only thing keeping baby in place was the peritoneum. This c-section was after a 24 hour TOL for a vba2c attempt. The OB stated that there was very little blood and it looked as though it had happened days prior to labor. I am so perplexed as to how a dehiscence and a rupture can have such different results with such similar descriptions of the uterine findings. In each case the uterus comes apart. In the case of a dehiscense there are no adverse affects, but in the case of a rupture the problems can be devastating for mother and baby. Is there ANY information out there describing the difference between the two and why one might happen and not the other? Any and all information on the subject would be greatly appreciated.

Anonymous said...

I think this article is misleading and does not help women make a choice. It is dogmatic and ignores the dangers of VBAC. Of course there are women that have a safe VBAC with thin uterus scars and women with thik scars rupture. But if the scar is thin, then the likelyhood of rupture appears to be higher. And even if these studies are not statistically signficant (hence the different results for the cut off), they all seem to point towards one direction: more risk with thinner scars. If this is all the information we have, it is normal to be risk averse when the scar is very thin. But of course there is a grey zone, where it is very difficult to make an informed decision and you just have to trust the experience of your midwife or doctor.

Well-Rounded Mama said...

Miss Momma, my heart goes out to you. I'm glad you and your baby are all right; thank goodness. But I understand how puzzling all of this can be to try and process.

I'm not sure I have any definitive answers for you though. I'm not sure even the experts know why some scar separations are relatively benign and others so catastrophic.

A lot has to do with other factors, such as whether the baby has protruded into the mother's abdomen after a true rupture. Outcomes are much worse if this happens. This didn't happen to you; you had one thin layer left. Even had that layer not been there, it might not have happened; not all rupture result in the baby protruding into the mother's abdomen. But it's definitely a risk factor for poorer outcome.

The main risk factor for poor outcome with rupture is what happens with the placenta. In some ruptures, the placenta abrupts (pulls away from the uterus) and of course this disrupts the baby's oxygen supply etc. This is when rupture becomes truly catastrophic and babies die. Fortunately, this happens only in a very small percentage of true ruptures, and not at all with a dehiscence. So while a potentially very serious risk, it happens only to a very very small percentage.

Why do some women experience a dehiscence and it doesn't progress to a full rupture? No one knows. I can't prove it, but I think it often has something to do with placentation. I think a lot of the worst ruptures are associated with abnormal attachment issues.

Another factor is whether the uterus is subjected to extra strong contractions, as with induction and augmentation. I think forces beyond the norm for labor stress the uterus more strongly and some scars give way that might not have otherwise.

And an unpredictable factor is how the mother healed previously, what kind of stitch/suture materials her doctors used and how her body responded to that.

I'm sorry I don't have any real definitive answers for you. The bottom line is that it's a bit of a mystery to the experts too. They just don't know why some people have one outcome and others have a different one.

I'm so glad you and your baby are okay.

Well-Rounded Mama said...

Anonymous, I don't think the article ignores the potential risks of VBAC at all; it clearly delineates them and how serious a rupture can be. I take that very seriously indeed, having had beloved friends who have experienced ruptures.

However, I do think that critics ignore how serious the complications with a repeat cesarean can be. I have good friends who have experienced extremely serious complications with this as well. I recommend you review Silver 2006, with lots of data on the risks associated with accumulating cesareans. Too often, the risks of VBAC are emphasized while the risks of repeat cesareans are conveniently ignored.

I agree that it's interesting and a little alarming that there is a trend towards more risk with thinner scars, and I can understand how this might make care providers hesitant or at least nervous. OTOH, the potential of misuse here is very high. Interoperator differences mean that the measurements might vary greatly from one person to another, and studies can't even agree on a common cutoff for increased risk. Nor are the studies large enough to be definitive. So there's potential for a LOT of women to be forced into repeat cesareans they don't really need if these measurements become standard measures by which access to a TOL is decided.

I think the information on thin scars can be used to counsel the woman about her potential risks, but as a tool to determine who "gets" the chance to VBAC and who does not, it is far too ill-defined and unreliable, and the potential for abuse is too high.

In an ideal world, the woman and her provider evaluate this risk together and decide their course of action from there. However, I disagree that a woman should close her eyes and blindly trust her provider's judgment; too many providers these days use ANY excuse not to attend a TOL and ignore the harms from repeat cesareans. Far too often it's the provider's own benefit that is topmost in their minds, not the woman's. Sad but very true for some providers these days.

This information should be just that...information used to discuss potential risk (and benefits) of each choice; just another tool for the purposes of decision-making. It should not be used as a blunt cudgel to take women's decision-making powers away from them. And that's the way it's being used in some places. That I cannot and will never agree with.

H R J said...

I've had three c-sections and dehissed with my last. I tried to vbac after the first one, and even went into labor with the third one, but the labor stopped and when I delivered via c-section a week later, I dehissed on the operating table. I found this post because I was looking for more information on what they will look for when they check the scar when they deliver the fourth baby due in June. What my surgeon told me is that what really determines the strength of the scar is not the scar itself but the tissue next to the scar. It would seem that there could be some cases in which a thick scar could actually be a risk factor in a rupture.