Ultrasound biometry evolves in the management of diabetes in pregnancy

Diabetes mellitus is one of the most common medical complications in pregnancy. In the USA, depending upon the diagnostic criteria used, 135 000–200 000 women annually develop gestational diabetes mellitus (GDM) adding to the number of women whose pregnancy is affected by either Type 1 or Type 2 diabetes. The rate of Type 2 diabetes has increased significantly (approximately 33%) compounded with the presumed parallel risk for obesity1, 2. A GDM diagnosis is associated with a 30–70% increased risk of developing the disease in future pregnancies, especially in obese women with advanced maternal age who required insulin in the index pregnancy. Women with GDM often develop Type 2 diabetes in later life3-6. Maternal hyperglycemia and the resultant fetal hyperinsulinemia are central to the pathophysiology of diabetic complications of pregnancy. For GDM7-9 and pregestational diabetes mellitus (PGDM), the complications include an increase in perinatal mortality, congenital malformations, deviant fetal growth (macrosomia and growth restriction), metabolic complications, birth trauma and the resultant increase in neonatal intensive care unit admission10-12. Diabetes-related complications can be decreased for both pregnant and non-pregnant patients with intensified therapy that results in improved glycemic control13, 14. The foundation for the use of ultrasound for the pregnant diabetic woman is early identification of congenital malformations and recognition of deviant fetal growth. In the case of fetal malformations, ultrasound technology will enhance well-timed options for the mother and, if needed, safer pregnancy termination. In addition, ultrasound provides an instrument for evaluation of deviant fetal growth and fetal weight estimation for timely delivery. This editorial explores the means by which ultrasound can be helpful to the clinician managing a diabetic pregnancy. Pregnancies compromised by PGDM (Types 1 and 2) have a 3–4-fold increased incidence of congenital anomalies compared to the general obstetric population. The frequency of congenital anomalies among diabetic offspring at birth is estimated at 6–10%. The prevalence of malformations appears to increase with the degree of diabetic severity (class)15-17. The results of a population-based study18 suggested a strong association between maternal diabetes and the risk for malformations of the central nervous system (CNS) and cardiovascular system and structural malformations of the fetal skeleton, face, urogenital system and gastrointestinal tract. Therefore, perinatal deaths as a result of congenital malformations account for approximately 40–50% in infants of diabetic mothers18-20. Congenital heart disease accounts for approximately 50% of all diabetes-related major congenital anomalies. The incidence of cardiac anomalies in the infants of diabetic mothers is 27/1000 compared to 8/1000 in the general population. The most common cardiac anomalies seen in fetuses of diabetic mothers are transposition of the great vessels, ventricular septal defects and coarctation of the aorta16, 20. In addition to congenital anomalies, fetuses and infants of diabetic mothers are at risk for accelerated myocardial growth. The cardiomyopathy is mainly related to poor glycemic control. Even in temporary glycemic elevation, cardiac hypertrophy and diastolic abnormalities can develop21. In adequately controlled glycemia, cardiac growth and ventricular diastolic filling are normal21-24. Some authors have suggested that myocardial thickening progresses during the second half of pregnancy even in fetuses of well-controlled diabetic mothers but that the observed hypertrophy is generally mild and does not affect age-related changes in fetal cardiac function23, 24. Diabetic cardiomyopathy occurs in about 30% of the newborns of diabetic mothers. Furthermore, thrombosis in fetal veins and alterations in the vascularization of the placental villi have been reported in association with cardiomyopathy25-27. It is characterized by disproportional thickening of the septum, resulting in a transient hypertrophic subaortic stenosis with left ventricular outflow obstruction and congestive heart failure21, 28. Septal hypertrophy and cardiomegaly do not necessarily impair cardiac function. As a result, infants born with septal or myocardial hypertrophy may be completely asymptomatic. Typically the transient myocardial hypertrophy is resolved within 2–4 weeks after delivery and certainly within 6–12 months of postnatal life in both GDM and PGDM29. The question remains whether an echocardiographic examination should be routinely performed in the fetus of a diabetic mother. Although there is generally complete resolution of the cardiomegaly, this examination helps to identify the quality of glycemic control and differentiates between the pathological and the constitutional macrosomic infant. It will enhance timely initiation of therapy and assist in selecting the mode of delivery for these large infants. It is recommended that this test be performed between 20 and 23 gestational weeks in conjunction with the anomaly scan and repeated at 37–38 weeks in anticipation of time and mode of delivery. The incidence of neural tube defect is approximately 20/1000 compared with 1–2/1000 in the non-diabetic obstetric population. A thorough examination of the spine is often difficult when the fetus is active or in an inappropriate position. An important factor during the evaluation is to demonstrate the integrity of the skin over the spine. Anencephaly occurs in about 1/200 diabetic pregnancies. Microcephaly and the caudal regression syndrome are other CNS lesions associated with diabetes in pregnancy. Caudal regression syndrome (CRS) is typically symmetrical sacrococcygeal or lumbosacrococcygeal agenesis, of variable extent, usually accompanied by multiple musculoskeletal abnormalities of both the pelvis and legs. The abnormality probably results from a defect in the mid-posterior axis mesoderm of the embryo occurring before the fourth week postconception. Although rare, CRS is 200 times more prevalent in patients with PGDM compared to 0.2–1% in infants born to non-diabetic women. It should be noted that CRS is still a relatively rare condition that complicates 1/200 to 1/500 diabetic pregnancies30-32. However, this anomaly cannot be considered pathognomonic for diabetes since it occurs in other conditions as well. Approximately 16–22% of CRS cases are associated with maternal diabetes. Ultrasound diagnosis of CRS is possible as early as the end of the first trimester and is even more accurate with the use of transvaginal ultrasound. Characteristic ultrasound images demonstrate abrupt interruption of the spine at dorsal, lumbar or sacral levels and abnormal aspect of the lower limbs with the femora fixed in a typical ‘V’ formation as a result of external rotation of the hip joints30-32. The most common gastrointestinal disorders associated with diabetes are imperforate anus, small bowel atresia and small left colon syndrome19. Some authors have attributed imperforate anus to a variation of the caudal regression. The diagnosis is usually accompanied by the presence of other abnormalities. In earlier studies, several authors reported an association between hyperglycemia and an increased risk and incidence of congenital anomalies33-35. This concept has been supported by both clinical and laboratory studies. In one study33, when glycosylated hemoglobin (HbA1c) levels were above 8.5%, there was a 22.5% incidence of fetal anomalies; between 1.0 and 8.5%, anomalies occurred in 5.1% of pregnancies; and below 6.9%, no anomalies were observed. The majority of recent studies failed to demonstrate a reduction in congenital malformation and perinatal mortality rates10-12. Several centers in which women with PGDM received preconception care and achieved established levels of glycemic control reported anomaly rates similar to that of the general obstetric population36-40. Others have suggested that normalizing the rate of congenital malformations is difficult and most likely unachievable. Some 50% of all pregnancies are unplanned. Thus, half of the women with PGDM will not benefit from preconception counseling. Of the group that could benefit from preconception care, approximately 50–60% can successfully achieve targeted levels of glycemic control. Therefore, substantial numbers of diabetic women will be exposed to hyperglycemia during organogenesis41-44. Ideally, primary prevention utilizing preconception glycemic control would suffice to eliminate the majority of diabetes-induced congenital anomalies. However, since this scenario is currently unrealistic, secondary prevention that utilizes appropriate ultrasound technology and timely examination becomes the major driving force for addressing fetal congenital malformations. An additional confounder is the 40–60% of gestational and Type 2 diabetic women who are obese. Maternal obesity increases the rate of suboptimal ultrasound visualization for fetal cardiac and craniospinal structures45, 46. These women are at greater risk for suboptimal visualization during the ultrasound examination. Controversy exists as to the optimum time to perform the first ultrasound examination and how many subsequent examinations are needed for identification of fetal malformations47, 48. During the late 1980s and early 1990s, the transvaginal approach was developed and used to recognize and define anatomical structures and organs in early pregnancy49-53. Since that time, several authors have reported improved visualization of fetal organs as early as 10–16 weeks of gestation54-58. Fetal evaluation cannot be wholly completed during the first anatomical scan. For example, the CNS continues to develop throughout pregnancy. The gyri and the sulci develop slowly and the cerebellum and the corpus callosum show a significant difference between early and subsequent examinations at 20–22 weeks' gestation59, 60. It has been demonstrated that the fetal heart can already be evaluated at the end of the first trimester. However, at 22 weeks' gestation, the heart measurements are 69–73% larger, which of course allows for a more accurate diagnosis54, 55, 61-63. Thus, it seems that an early scan at 10–14 weeks of gestation permits exclusion of major malformations and an additional scan at 20–22 weeks is advisable in order to rule out previously unrecognized malformations. Using the early and the 20–22-week scans as points of reference for anomaly screening by expert sonographers will result in identification of the majority of malformations. Three malformations that remain difficult to identify and, therefore, may be missed are small ventricular septal defects, imperforate anus and cleft palate without cleft lip. The rationale for early screening gained further support with the development and recognition of effective screening for trisomy 21 with a combination of maternal age, fetal nuchal translucency and maternal serum free beta-human chorionic gonadotropin and pregnancy-associated plasma protein-A concentrations at 11 to 13 + 6 weeks of For a rate of the rate of trisomy 21 this is approximately This is an over the 30% rate achieved by maternal age or the rate both maternal age and serum the use of an early anatomical scan accurate in general and more in the pregnancy compromised by diabetes and related The for early gestational in diabetic pregnancies is alterations in fetal growth are gestational women with advanced diabetes are to delivery and of delivery are all gestational age and to fetal of the have been considered the most for gestational as a is one of the most accurate the week of The of in diabetic pregnancies was when the authors reported that of cases were more 6 below the for Some of these fetuses with growth were later as Others failed to the Therefore, a fetus in an abnormal metabolic may be at increased risk for growth as as there is no that an early growth is associated with infants. diagnosis of fetal in diabetic pregnancies is an important clinical and of diabetic mothers are at risk for birth trauma not of but of greater growth in to the and are complications associated with and increase in frequency with birth However, to the diagnosis of still remains significantly This can be by studies small to of glycemic prevalence of in a rates of maternal obesity in a and fetal as as in the of diagnostic criteria used to define diabetes. In the of of fetal weight above the from to with from to The for fetal weight from to and from to diagnostic currently in clinical use in all of both and to the of adequately the affected infant. However, of the of in these not use but to further that will enhance of the that a currently does not that can define the compromised by abnormal to use the technology will both the mother and fetus of the that is currently an estimated fetal weight by several measurements of the fetus are and these are a that to fetal to and large for gestational age and is a that of between the and However, these often to identify deviant fetal growth in pregnancies compromised by diabetes. For example, is a fetus whose weight was to be at the and to the of diabetes to be at the at the time of this fetus not more compromised the fetus whose weight was at the and at the The majority of fetuses at of the the and above the are small or These do not the small or large from the Therefore, the clinician to in addition to fetal weight clinical as of glycemic control and fetal studies when and the presence or of or The the clinical the diagnostic the greater is there are for fetal but there is a of and in all birth The of the to account for in fetuses and in the of fetuses of diabetic mothers may for the in However, most are at gestational age The + and + are the most used for However, it should be noted that the of the diabetic mother has a significantly from that of the general population. that of fetal weight at in Type 1 diabetic women could be in by as as when fetal weight is compared to Therefore, a fetus at birth could have an of as as or as as An was to to fetal In if is greater of newborns an and would more patients with Type 1 diabetes that estimation of weight utilizing and was as accurate as more of fetuses were estimated within of birth An of was required for the to be that the fetal weight would be above approximately The of for the and the clinical of these In a to the use of to delivery in diabetic it was that of and of general used this technology for clinical Diabetes appears to affect the more the or Thus, this is in either macrosomic or The fetal is the most affected by in fetal and the of the therefore, be as the for An association between large fetal between and weeks of gestation with in fetuses of diabetic mothers has been However, measurements generally not the weeks of pregnancy. The one to the and the other a of accelerated growth. accelerated growth more as demonstrated by measurements at Several studies have this The use of with the of may enhance the of in the fetus of a diabetic mother. It was that when was greater the was in approximately of results were reported by a of 2 above the and a the to but the for increased to studies the of to an the risk for a fetus was For an this risk was For fetuses the risk for was In the an was a diagnostic test at weeks' The for a macrosomic was and the In addition, it was suggested that a of during early achieved a of and of The identification of fetal was with the of by the and during the trimester. measurements 13 at were significantly more of infants when compared with or as It has been that is more in and in infants with a large group of suggested evaluation of in the of abnormal fetal growth. greater in but However, this approach has not been by and has not gained In infants of women with controlled diabetes there is of the majority of the organs but not of the weight of and has been demonstrated in the infants of diabetic mothers an increase in of this it was suggested that be used to fetal for fetal It was that the increase in was as early as the week of gestation and more with increased of pregnancy. Furthermore, not remain when were throughout pregnancy. This approach may an early fetal in addition to maternal of for initiation of therapy in the pregnancy affected by approximately of birth it accounts for about 50% of the However, the of fetal in the used in may account for of fetal Several have been used to and and and and was at the of the in the position. These were compared in studies to and for and weight at delivery. for in the mother and include of and Of the is the to perform and is to the ultrasound has a for the diagnosis of the macrosomic In of estimated are within of birth weight and the are within of birth This will affect to of fetal and delivery. Some of to be a poor of fetal in the non-diabetic pregnancy to fetal weight in to Others that achieved and In addition, that fetal will be a more accurate and in and diabetic Fetal and measurements have been to be in diabetic may the of fetal and the to between constitutional and diabetic macrosomic of for of the and The for the measurements of fetal and the was were most in by the of an was as and is one of the most complications in with have been to identify the fetus at risk for this A association was between fetal weight and However, diabetic patients have a greater risk for at weight compared to non-diabetic Furthermore, increased risk for is already when fetal weight is in the birth not are infants of diabetic mothers at significantly increased risk for but of cases of in these pregnancies at birth at or greater The in an is usually to fetal weight for selecting cases that can be with a risk of birth trauma cases in which the risk of trauma is to delivery by Therefore, when fetal weight from to delivery is even when there is the for maternal and medical For all the above there to be to identify fetal growth abnormalities in fetuses of diabetic mothers. studies the of for prevention of do not include in the that will significantly affect this The increased risk of is by between macrosomic infants of diabetic and non-diabetic delivery of the fetus at risk by of and when the is between and will by the incidence of in diabetic studies to identify the fetus at risk for the development of in The a index by the from the the suggested the difference between and have not been used of in to to the of previously and of and in addition to an adequately are the foundation of the growth of the developing fetus conditions as diabetes in pregnancy and associated with with this growth the result is deviant fetal growth. In the diabetic fetal growth complications are the result of is the for fetal the growth rate of the fetus to weeks' gestation is of insulin In the fetal insulin are between and weeks' It is late in gestation that the major of insulin and accelerated fetal growth and This is with of abnormal growth levels of maternal were not achieved in diabetic fetal growth in a However, the for of glycemic control may to an The majority of studies use a poor of the glycemic in pregnancy since it to define the appropriate that will In addition, the time from fetal weight estimation to delivery from to In one a was mainly to the This in many studies to account for the deviant growth in the diabetic infant. The fetus approximately The may result in an additional weight in the already large of diabetic patients can be with greater and of the between fetal complications and measurements of macrosomic and there is a of the and the of of fetal growth in the pregnant diabetic weeks the growth of the fetal and a similar to that of The of abnormal growth is after these and is to the and In one accelerated fetal growth was noted between and weeks' of Type 1 diabetic ultrasound examinations in the of growth of at week in fetuses to be at abnormal fetal growth in pregnant diabetic mothers were and growth throughout the trimester. Fetal weight estimation was considered with an rate of The results of this support the of both early and late accelerated growth in fetuses of gestational and pregestational diabetic women. These are from the growth of macrosomic fetuses of non-diabetic to develop growth during the second trimester in pregnancies in order to fetal weight at delivery have been The were small in In was for fetuses and for fetuses in all these to about the It is to that when the is poor in pregnancies, the to fetal weight by a compromised conditions in the second trimester is is a large in and Therefore, the of maternal weight and the presence of diabetes in pregnancy an regression would significantly maternal disease are used to the for therapy in pregnancies affected by diabetes. Several the use of between and weeks' gestation to if the fetus is at risk for and as the for the initiation of reported that patients with an and plasma with a rate of delivery of an of in to a rate of for with In patients with plasma between and measurements of the fetal fetuses at risk for accelerated growth in reported similar However, when targeted levels of glycemic control were not there was a rate of in patients with above and below the of Several to be a ultrasound early in the trimester as effective as In clinical when abnormality was ultrasound examination the a a growth A of of and risk of measurements in of and of with a risk of A ultrasound is more a ultrasound examination for fetal growth and of fetal growth and clinical for should not be in that the abnormal diabetic fetus are and of which assist in between the and large In addition, clinical as glycemic and obesity should be in the to a delivery the is to the of The use and of ultrasound is currently Some studies that is support this are the technology or are the of the with in diabetes is Several studies to fetal weight with ultrasound. However, this technology has a of it can fetal structures but for most of gestation, to the The can of fetal structures as fetal and for fetal weight during the trimester. In this ultrasound was reported to be to the major in diabetes in pregnancy is fetal anomalies. It was reported that ultrasound and more identify fetal abnormalities ultrasound. Furthermore, was more accurate ultrasound. These can be to ultrasound and may of the diagnosis of It will enhance of the severity of fetal anomalies, and cardiac anomalies, especially and congenital heart In diabetic ultrasound has a significant rate in the diagnosis of fetal abnormalities and remains in the recognition of fetal This should be by and who should not have about the of in this The evaluation of the fetus of a diabetic mother should include in the first trimester a transvaginal ultrasound examination to rule out congenital abnormalities and measurements for A ultrasound examination for congenital malformations to be performed at approximately weeks' fetal weight and cardiac evaluation will enhance identification of the large or small infants. During the measurements to be performed in order to assist in the of the and the of deviant fetal growth. Ultrasound in the of development of the fetus of the pregnancy affected by diabetes is a major in medical However, the Since ultrasound has a of care, to in that the of obstetric ultrasound upon the and of the the As it the of of quality to which means that who to perform ultrasound a of technology may be and be However, that the majority of the technology are a small of diagnostic as In of the of the as the metabolic with the and of the of these to enhance have ultrasound to of as a diagnostic are needed in order to and the time, be of more should be As has for more great to who will there support for The approach in ultrasound to a combination of and In the pregnancy compromised by the is required to that a great of In this to when but not the targeted to account of the of the fetus in diabetic pregnancy are established the inappropriate for and fetuses currently in use will of the metabolic that this