Cervical Incomptence and Cerclage - 4 - Diagnosis

As pointed out in our previous posts on this subject, one of the greatest frustrations of cervical incompetence has been to establish its diagnosis in a timely and reliable fashion. Since multiple factors can contribute to cervical incompetence, and the patient’s own response may not be consistent during different pregnancies, cervical incompetence must not be viewed as a ‘disease’, so much as a ‘syndrome’ with a common final pathway. And, being a ‘syndrome’ in this case makes it easier to understand why there is not a ‘diagnostic test’ for cervical incompetence – indeed all we really have are several different methods of screening for the condition.

I divide screening into several different categories. The first is the patient’s obstetrical history – the classic approach upon which we relied during my training. If the patient presents for consultation with a history of premature delivery or midtrimester pregnancy loss, it is important to characterize the events surrounding the delivery in as much detail as possible from both the patient’s recollections and the medical records. For example, some of the questions asked include: gestational age at delivery; presence or absence of painful contractions prior to presentation and the duration of the same; cervical status at presentation (long, short, firm, soft, degree of effacement); amount of amniotic fluid (too little, too much); condition of the membranes (intact, ruptured, or ballooning in the vagina); evidence of prior or concurrent urinary tract infection, sexually transmitted infection or bacterial vaginosis; other pregnancy-related or previous medical conditions (including history of infertility and known Mullerian anomalies) and type of medical therapy; size of the baby and appropriateness for gestational age; course of the labor itself (short, long, use of oxytocin and why); outcome of the baby and neonatal complications if the baby was not stillborn; pathologic evaluation of the placenta and baby; laboratory studies obtained at the time of admission (e.g., CBC, urinalysis and culture); and, if this was not a first pregnancy, comparison of this pregnancy to the others with regard to these factors and intervening procedures (e.g., traumatic vaginal delivery, D&C, cervical conization, myomectomy). The goal of this line of questioning is to try to separate the possible diagnosis of cervical incompetence from other causes that may or may not put another pregnancy at risk for premature delivery. The patient with cervical incompetence is much more likely to have a history of relatively painless labor, presenting with advanced cervical dilation and effacement (out of proportion to the labor), followed by a rapid delivery. If she has had previous pregnancies, there is also a tendency for sequentially earlier deliveries with cervical incompetence.

The problem with the ‘historical approach’ to screening is that it often meant a patient lost one or more pregnancies before a ‘diagnosis’ of cervical incompetence was established and if the pertinent historical information could not be accurately obtained, the diagnosis might be further delayed. To address that concern, over the years, I have put together my own little laundry list, based on both experience and, where available, the scientific literature, of women who should be observed more carefully, even during a first pregnancy, for evidence of cervical incompetence. Included among those risk factors are those mentioned in our last post: family history of premature delivery and cervical incompetence; maternal obesity; congenital uterine abnormalities; uterine fibroids; previous uterine surgery (e.g., D&C, removal of uterine septum, myomectomy); previous cervical surgery; history of maternal infertility and assisted reproductive techniques for conception; multiple gestation; recurrent vaginal and urinary tract infections; connective tissue disorders (e.g., Ehlers-Danlos syndrome and Marfan’s syndrome) and perhaps collagen vascular diseases; polycystic ovary syndrome/insulin resistance; and perhaps diabetes. In addition to predisposing risk factors, one should remain alert to those that develop during pregnancy and also maintain a high index of suspicion related to certain patient complaints such as increasing pelvic pressure or excessive vaginal discharge, both of which may accompany premature cervical effacement.

The major contributor today to cervical evaluation in both categories above, and the primary means of detecting cervical incompetence in the presence of new ‘signs and symptoms’ or, fortuitously, during other assessment, has been ultrasound. I am not going to make any effort to review the literature that debates the value of ultrasound in this regard. In the most general sense, cervical length in midtrimester is inversely proportion to the risk for preterm delivery; and in women at greatest risk for preterm delivery, the course of their cervical changes and the events surrounding delivery often are consistent with the traditional history we attribute to cervical incompetence.

However, evaluation of the cervix involves more than simply measuring the length (or what’s left of the length) of the endocervical canal. Other factors to consider are the changes in length over time, the configuration of the internal os and the endocervical canal, and the presence of and degree of distention by membranes in the endocervical canal. In the patient who is being followed for possible cervical incompetence by history, or who has risk factors for the same, serial evaluation may be necessary and, in my practice, preferable. Most women with cervical incompetence will develop visible changes in the configuration and length of the endocervical canal between 16 and 24 weeks. And, with serial follow-up, one is more likely to detect changes at a time when intervention can still provide a favorable outcome for the pregnancy.

One of the questions that has been raised is whether or not ALL pregnant women should be routinely assessed for cervical incompetence in midtrimester. I can tell you that over the past 6 years in my own experience, at least one-third of the cases of cervical incompetence necessitating placement of ‘rescue’ cerclages have been in women between 18-24 weeks with no previously recognized risk factors in whom advanced cervical changes were detected during the course of a routine sonogram to assess fetal growth and anatomy usually scheduled for that time in pregnancy.

In most cases, a ‘suspicious’ cervix was first noted during transabdominal assessment and then confirmed transvaginally. When one considers the safety of ultrasound and the speed with which a transvaginal sonogram can be performed, the epidemic of obesity (a risk factor itself) that precludes adequate transabdominal assessment of the cervix, and the potential savings to the healthcare system by preventing the delivery of even a few 23-26 week babies annually at any institution, I think the cost of including cervical assessment during the routine pregnancy ultrasound examination that has become standard at 18-22 weeks, the additional cost of the transvaginal ultrasound could be readily justified…

Labels: cerclage, cervical incompetence

Permalink | 7 Comments| Email Post

Post your comment

Cervical Incompetence and Cerclage - 3 - Significance of the Internal Cervical Os

The incidence of cervical incompetence in the general population is really not known and probably varies by population subgroups. Estimates range between 1 in 100 and 1 in 2000. My sense over the years (considering the number of cerclages I have placed in midtrimester) is that it is more common than widely recognized and that the higher estimate is probably more accurate (We have almost 6000 deliveries per year at our institution and there is not a week that goes by when I do not place at least one ‘rescue cerclage’ in midtrimester). It is also my sense that the incidence appears to be increasing – at least in the populations of pregnant women with which I work on a regular basis. As mentioned in my last post, there are probably innate, acquired, and iatrogenic factors that contribute to the ‘syndrome’ of cervical incompetence. And, many of these factors overlap with those that are commonly recognized as risk factors for premature labor and delivery.

Some of the potential risk factors for cervical incompetence include past obstetrical and family history of premature delivery and cervical incompetence, maternal obesity, congenital uterine abnormalities, uterine fibroids, previous uterine surgery (e.g., D&C, removal of uterine septum, myomectomy), previous cervical surgery, multiple gestation, traumatic vaginal delivery, recurrent vaginal infections, connective tissue disorders (e.g., Ehlers-Danlos syndrome and Marfan's syndrome), polycystic ovary syndrome/insulin resistance, and perhaps diabetes. There is debate that a ‘short cervix’ by itself may increase the risk for cervical insufficiency, but if the cervical connective tissue is normal and there is no loss of integrity at the internal cervical os, this seems to be a weak call in many instances.

Whether there is an underlying connective tissue abnormality of the cervix, pathologic activation of the cascade of biochemical events that leads to remodeling of the cervical connective tissue, or simply a congenital or acquired anatomical abnormality of the internal cervical os, the most pathognmonic and sentinel event occurring with cervical incompetence is progressive loss of integrity at the internal cervical os. Cervical change that anticipates labor starts from the inside and progresses outward. Premature cervical ‘remodeling’ can lead to this loss of integrity at the internal os, but it is also often seen in conjunction with congenital uterine malformations (Mullerian defects) that affect midline structures (such as the cervix) and as the consequence of trauma during operative procedures, the most common being D&C. I emphasize this point, because I am a firm believer in the laws of physics with regard to cervical incompetence and its progression, even prior to the connective tissue remodeling (‘ripening’) that will eventually occur once the cervix has been stretched from within.

Let me elaborate on this in the way I explain it to patients: The internal cervical os is supposed to remain closed and basically contiguous with the rest of the inner surface of the uterus – distributing the weight of the baby and fluid over that entire surface area. If the internal os has been damaged and is not closed, or if other factors decrease the resistance at that anatomic site allowing it to begin to open, the tendency will be to simply continue to open, and as that progresses, it actually takes less pressure to continue that process with time. We are talking simple physics here, the same principle that holds when we drive a wedge into a log to split it. It’s hard to get started and then eventually it just gives way.

All women who wear spiked heels should know the reason for this. If you concentrate all the weight of your body, say 150 lbs, on a heel that measures say .2 x .2 inches that translates into a weight of 3750 lbs per square inch! That’s why spiked heels can punch holes in flooring. If we apply that same thinking to pregnancy and the entire volume of the uterus is focused on a weak point, a defective and slightly opened internal cervical os, the tendency is to enlarge that opening and eventually squeeze the membranes into the cervical canal – an event we call cervical funneling – just like squeezing a water balloon through a small opening. The important thing to recognize is that we CAN detect loss of integrity at the internal os during pregnancy by ultrasound and evaluate the configuration of the cervical canal as it begins to distend with membranes and amniotic fluid BEFORE this is readily apparent by clinical symptoms or obvious changes evaluable by vaginal examination of the patient. And this sometimes gives us an opportunity to intervene in a way that can prolong and preserve a pregnancy that might have otherwise been lost or accompanied by extreme prematurity and its attendant morbidity....

Labels: cerclage, cervical incompetence

Permalink | 10 Comments| Email Post

Post your comment

Cervical Incompetence and Cerclage - 2 - An Overview of Cervical Structure

In the nonpregnant state, the cervix comprises the lower one-third of the uterus and connects the uterine cavity to the vagina. The overall length of the cervix is highly variable but generally falls in the range of 2.5 to 5.0 cm. Only about one-half to one-third of the cervix extends into the vagina and this is called the portio vaginalis. The portion of the cervix that opens into the vagina is called the external cervical os; the uppermost portion that opens into the uterus is called the internal cervical os. As we will emphasize in later discussion, the internal cervical os is a key player in the etiology, pathogenesis, and diagnosis of cervical incompetence. The portion of the cervix that connects the external os and internal os contains the endocervical canal and this narrow tube separates the relatively unsterile environment of the vagina from the uterine cavity. The endocervical canal is lined by a single layer of mucous producing cells and the chemicals and immunoglobulins that are secreted into the canal provide the major barrier to ascending infection by potential pathogenic microorganisms from the vagina and ectocervix.

Although the cervix is contiguous with the body of the uterus, it is structurally different from the uterus in several key aspects. Both the uterine wall (the myometrium) and the cervix contain smooth muscle and fibrous connective tissue, but there is a much greater percentage of the connective tissue in the cervix than in the myometrium. The uterus is “designed” to contract and, when the time is right, eventually push the baby out, while the role of the cervix, under normal circumstances, is to keep the baby inside until it is mature enough to survive in the cold cruel world outside the womb.

The fibrous connective tissue of the cervix is mostly composed of types I and II collagen, elastin, and proteoglycans. The collagen is heavily ‘cross-linked’ and this imbues the cervix with a tremendous resistance, again under normal circumstances, to stretching and ‘softening’ until the biochemical cascade that progresses to labor ensues. At that point, the cervix is capable of undergoing a remarkably rapid transformation from a structure that has the consistency of a rubber eraser to the soft, compliant, elastic structure that will permit the relatively easy passage of the baby from the uterus and into the birth canal – a transformation that results from the remodeling (uncross-linking) of the collagen and the extracellular matrix.

The cervix, then, is a very dynamic organ that always has the innate ‘potential’ to change from barrier to facilitator of uterine evacuation. In cervical incompetence, some women probably do not have a normal percentage or configuration of the fibrous connective tissue, others may be overly sensitive, due to genetic susceptibility or metabolic abnormalities, to stimuli that would ordinarily not lead the “normal" cervix to undergo the transformations associated with premature cervical ripening, others may be exposed to higher levels of factors that can initiate cervical ripening independently of the innate pathways, and others may have congenital or iatrogenic diminution of normal ‘integrity’ at the internal os that leads to subclinical cervical changes that eventually result in activation of the innate ‘ripening’ pathways or to ascending infection that can promote more rapid cervical change and/or premature uterine contractions. Then of course, there may be individuals who have more than one of these predisposing factors...

Labels: cervical incompetence, cervical insufficiency; premature labor

Permalink | 4 Comments| Email Post

Post your comment

Cervical Incompetence and Cerclage - 1 - An Introduction

I have had several readers request information about cervical incompetence and cerclage. These are interrelated subjects in which I have had a keen interest and much personal experience over the years, but they have also been the topics of considerable controversy in our professional literature. And, except for some brief references to these in previous posts, I have chosen until now to avoid a more detailed discussion. The reasons for that are not so much apprehension about facing my peers (since this is not a peer review journal and I would put my experiences and successes up against anyone’s), as an uncertainty as to how to address the subject in a way that would be informative and relevant to a wide range readers as well as objective in its presentation.

Cervical incompetence, or cervical insufficiency, relates to premature, progressive cervical change that subsequently results in preterm delivery and/or pregnancy loss. Most purists will also add the admonition that such change must occur in the absence of uterine activity (i.e., painless cervical dilation and effacement) – a point that was emphasized in my own training and with which I now take major issue. The problems with the study of cervical incompetence are that, like preterm labor, the underlying causes and pathology can be multifactorial, the diagnostic criteria are not consistent, and treatment regimens, including the value and timing of cervical cerclage, are varied and difficult to subject to critical, randomized research because of the vastly different experiences and skill sets of physicians and the sensitive issues related to what is in the balance, specifically, the life of the baby, under circumstances when patient and provider have their backs against the wall and the pervasive attitude is often: “Well, what have we got to lose by trying?” “If we don’t do something, the baby will be lost anyway or delivered so early that he/she is at great risk for mortality or life-long morbidity secondary to prematurity.”

In recent years, the debates on all fronts have come to the fore. The primary reason for this is that prior to the widespread application of ultrasound to the evaluation of the cervix in midtrimester, the diagnosis of cervical incompetence was usually a retrospective one – made only after the preterm delivery and/or loss of one or more babies under circumstances that met the “definition” of cervical incompetence. And, believe me, when the purists got involved and the strict criteria were applied, the debate was often quite heated – “She had some contractions with that pregnancy, so it must have been preterm labor and not cervical incompetence.” “She came in with a fever, so it must have been chorioamnionitis (infection of the membranes) and could not have been cervical incompetence.” “She delivered her last baby at 35 weeks, so how could she have an incompetent cervix.” Anyway, you get the picture. The problem with all these sorts of comments is that they tend to ignore the possibility of “What came first, the chicken or the egg!”

However, now we have a lot of experience with evaluation of the cervix by ultrasound, and some of us have a very low threshold for performing the same under circumstances wherein the patient may be ‘at risk’ for cervical incompetence or when there appears to be premature cervical changes picked up as an incidental finding at the time of a routine ultrasound done at 18-20 weeks for the assessment of fetal anatomy. And, as a result, and as I will detail in subsequent posts, we are now in the position of more often detecting what may be either significant (advanced) premature cervical changes that permit late (“rescue”), but timely, intervention with even a first pregnancy as well as more subtle changes that may or may not indicate the patient is truly at risk for preterm delivery or pregnancy loss. It is the latter group that currently should be approachable by randomized research to clarify the most sensible approach to therapy, but even that is proving more difficult with time.

In this series, I will have several goals: We will provide a basic understanding of cervical structure and illustrate the differences between the cervix and the uterus; we will briefly discuss the biochemical changes that occur in the cervix coincident with cervical change and the factors that might contribute to those events prematurely; we will point out risk factors for cervical incompetence; we will discuss assessment of the cervix by ultrasound; and we will discuss treatment options under various conditions, including my own approach to cerclage – focusing on that approach under pregnancy “rescue” conditions…so, stay tuned!

Labels: cerclage, cervical insufficiency; premature labor, incompetent cervix

Permalink | 6 Comments| Email Post

Post your comment

Assessing Fetal Lung Maturity - 4 - TDx-FLM, Lamellar Body Count, and Foam Stability Index

In our last post, we described the two earliest tests that were introduced in the 1970’s to assess fetal lung maturity – the lecithin/sphingomyelin (L/S) ratio and detection of phosphatidylyglycerol. Originally, both of these tests were performed by thin-layer chromatography (TLC), a technique that required considerable expertise and was both time-consuming and expensive to perform. In the last 20 years, other tests have become available that are simpler, less costly, and just as reliable in their prediction of fetal lung maturity, although both the L/S ratio and PG tests are still widely used.

The next test that saw a rise in use in the early 1990’s was the TDx-FLM assay. This test employs fluorescence polarization to measure the ratio of surfactant to albumin (S/A) in uncentrifuged amniotic fluid. It is simple, quantitative, can be performed in 60 minutes or less and is highly reproducible (Steinfeld, et al., Obstet Gynecol 1992;79:460-4) . The threshold for fetal lung maturity using the most recent version of the test is set at > 55mg surfactant/gram albumin. TDx-FLM results correlate well with the L/S ratio and PG and with clinical outcomes (Obstet Gynecol 1993;82:1004-8). Blood or meconium in the amniotic fluid can interfere with the accuracy of the test results but, generally, will not result in immature results being erroneously reported as ‘mature’. PG can be used under these conditions as a back-up study. As in L/S ratio determinations, the standard threshold may not be quite as reliable in the woman who is diabetic, particularly, if she has a large baby. Livingston and colleagues (Obstet Gynecol. 1995;86:826-9) have shown that values > 70mg/g were not associated with respiratory distress syndrome in infants of diabetic mothers, even if PG was absent.

Another simple, rapid, and accurate test that can be performed on amniotic fluid to assess fetal lung maturity is the lamellar body (LB) count. LB are made by the type II cells in the alveoli that produce surfactant and, indeed, they are the packages in which surfactant is stored. They are released into the amniotic fluid from the lungs in increasing quantities that parallel surfactant production. LB are about the size of blood platelets so they can be quickly counted in the same laboratory machines that are used to perform a complete blood count (Dubin, Am J Clin Pathol 1992;97:836-49). It is recommended that the assay be done on uncentrifuged amniotic fluid and LB counts in the range of 30,000-40,000/microliter correlate well with fetal lung maturity (Ghidini, et al., Arch Gynecol Obstet 2005;271:325-8). There is evidence to support using a higher threshold of 50,000 in diabetic women. Because of their similarity in size to platelets, amniotic fluid contaminated with blood may interfere with the accuracy of the LB count, resulting in both falsely elevated LB counts, initially, and lower counts over time.

The final test I want to mention is one that actually measures the functional capacity of the surfactants that are present in the amniotic fluid. This test is called the foam stability index (FSI). It too is simple, inexpensive, and can be rapidly performed. I believe it is elegant in terms of both its relative simplicity and its predictive value. If you recall from our first post on this subject, I described alveoli as little “bubbles” that were at the mercy of the physics of surface tension and surfactants as chemicals that helped reduce surface tension so that the alveoli would not so readily collapse. The FSI assay takes advantage of this principle.

The FSI assay takes centrifuged amniotic fluid and adds a fixed volume (0.5 mL) to tubes containing serial concentrations of ethanol to produce final ethanol concentrations ranging between 44 and 50%. The tubes are then shaken for 30 seconds and the tube with the highest ethanol concentration that is able to retain a stable ring of bubbles at the air/fluid interface is read as the FSI result. RDS rarely occurs when the FSI is 47 or greater. Again, the simple beauty of the test is in its ability to measure the function, not the individual components, of the surfactants present in the amniotic fluid.

There are other approaches to the assessment of fetal lung maturity, but these are the most commonly available and widely used. It is generally accepted that in the nondiabetic pregnancy, any of these amniotic fluid tests that meets its threshold for ‘maturity’ meets the current standard of care for assessment of fetal lung maturity. Rather than performing several of these studies, most providers will now first perform a rapid test (e.g., TDx-FLM, FSI, or LB count) and only resort to the L/S ratio and/or PG studies if the initial test is immature. Another option would be to perform a second rapid test before doing the L/S ratio or PG since these studies may need to be sent off to an outside laboratory, delaying the final result.

Labels: fetal lung maturity, FSI, lamellar body count, LB, TDx-FLM

Permalink | 8 Comments| Email Post

Post your comment

Assessing Fetal Lung Maturity - 3 - L/S Ratios and Phosphatidylglycerol

In our last two posts we explained the rationale for assessing fetal lung maturity – minimizing the risks for fetal complications related to respiratory distress syndrome (RDS). This has become increasingly important as we try to reduce the morbidity associated with iatrogenic preterm delivery secondary to elective induction of labor and cesarean sections, particularly, when the mother has an uncertain date of conception, late entry to prenatal care, or medical conditions, such as diabetes, that can result in a delay in fetal lung maturation – specifically, in the production of surfactants. The only method currently available for assessing fetal lung maturity is sampling of amniotic fluid by the invasive procedure called amniocentesis. However, there are several different ‘tests’ that can be performed once that fluid is obtained. In this post, let us look at the first two that were introduced in the 1970’s.

In 1971, Gluck and colleagues (Am J Obstet Gynecol 1971;109:440-45) described the first test that became widely used to assess fetal lung maturity – the lecithin/sphingomyelin (L/S) ratio. The rationale for the test is as follows: sphingomyelin is present in amniotic fluid in relatively constant amounts during the third trimester and is not contained in fluids produced by the fetal lungs; whereas, lecithin, which is produced by the lungs, begins to increase in concentration midway through the third trimester. Using a fairly complicated procedure called thin-layer chromatography (TLC), the amounts of both substances can be assessed. The ratio of lecithin to sphingomyelin is used to compare the two because that removes the variable of amniotic fluid volume that can affect the actual concentrations of these substances. Gluck and colleagues found that an L/S ratio of 2.0 or more predicted the absence of RDS in about 98% of neonates; ratios of 1.5 to 1.9 were associated with a 50% risk of RDS; ratios below 1.5 carried a risk of 73%.

It is clear from these results that not all babies with an ‘immature’ L/S ratio will develop RDS, but this test alone does not tell us which babies will not. Furthermore, we have learned through the years that certain pregnancies, particularly those associated with maternal diabetes and/or fetal macrosomia (big babies), may be at greater risk for RDS even with low ‘mature’ L/S ratios – indeed under these clinical circumstances, some providers prefer to see an L/S ratio equal to or greater than 2.5. The presence of maternal blood can falsely lower the L/S ratio and the presence of fetal meconium (feces) in the amniotic fluid can falsely elevate the L/S ratio.

The next test, introduced in the mid-1970’s by Hallman and colleagues (Am J Obstet Gynecol 1976;125:613-17), was assessment of amniotic fluid for the presence of phosphatidylglycerol (PG). These investigators found that PG production by the lungs did not begin to increase until several weeks after the rise in lecithin – and usually was not detectable in significant amounts before 36 weeks’ gestation. PG appears to be involved in the distribution of ‘surfactants’ over the surface of the alveoli. Its presence then, by both timing of production and action, has been found to be associated with a high likelihood of fetal lung maturity. It too was originally assayed by TLC and was simply reported as “present” or “absent” without any quantitative determination.

In recent years, slide agglutination assays have simplified the assessment of PG with a “positive” threshold set at greater than 2%. Like the L/S ratio, the presence of PG also predicts the absence of RDS in 95% or more of cases, but again, its absence does not necessarily indicate the baby will develop RDS. Many providers prefer to see the presence of PG in those conditions associated with delayed fetal lung maturity before proceeding with elective delivery, particularly when there are low ‘mature’ L/S ratios. Furthermore, since detection of PG is not affected by the presence of blood or meconium, it can be used to assess fetal lung maturity on samples of amniotic fluid obtained vaginally in cases of premature rupture of membranes. Measurement of L/S ratios under these circumstances is less reliable unless a very clean specimen is obtained.

These two tests for fetal lung maturity were the first to be widely employed and their value has withstood the test of time. In the next post we will conclude our discussion of fetal lung maturity assessment with several other tests that have been introduced more recently...

Labels: lecithin/sphingomyelin ratio; L/S; phosphatidlyglycerol; PG; fetal lung maturity

Permalink | 0 Comments| Email Post

Post your comment