Metformin Use During Conception and Pregnancy

The following recent query requested my opinion regarding the safety of metformin during the periconceptual period and throughout pregnancy. Although there are not many large or comprehensive studies addressing these concerns, the bulk of the data available to us is encouraging...

Dr. T,

Quick opinion if you don't mind. As you may recall, I miscarried on 9/12. I have since seen my PCP for a regular check-up. He prescribed me Metformin….he believes based on my history, weight, blood work and family history, my body may have issues with the breakdown of sugars (i.e., type 2 diabetes but I'm not diagnosed with that). He said that it also may have some positive side effects for me including weight loss and assistance in helping me to conceive (although that doesn't appear to be a problem since I WAS able to get pregnant even though I miscarried). He says it is completely safe.

I have read mixed things online about Metformin and potential effects on babies. Namely that no known birth defects have been caused from it but that there are not many studies either. Additionally, I have read some things about it potentially causing miscarriage.

Could you please give me your opinion on this? I would like to take it as I have for a week now, because I physically feel better. I'm very scared of the effect it may have of a pregnancy and if I were to stop taking it during my pregnancy (as my doctor said this is elective as he feels it would benefit me but is not imperative for me to take). Do you know of any potential miscarriage issues with this prescription?

Thanks again for EVERYTHING!

Christe


Women with insulin resistance are at increased risk for hyperinsulinemia, type 2 diabetes, polycystic ovary syndrome (PCOS), and hyperandrogenism (increased levels of ‘male hormones’). They also are at risk for reduced fertility secondary to ovulatory dysfunction and a suboptimal hormonal milieu that may impair conception, implantation, and placentation. Pregnancy complications include higher rates of miscarriage, gestational diabetes, hypertensive disorders, preterm delivery and operative deliveries, excessive maternal weight gain, fetal macrosomia as well as growth restriction, and admission of their babies to neonatal intensive care units for a variety of reasons (Boomsa, et al., Semin Reprod Med. 2008;26:72-84). In my own experience, they also appear to be at increased risk for complications related to cervical insufficiency.

Metformin is an oral hypoglycemic (blood sugar lowering) agent whose primary affect seems to be mediated through its ability to reduce insulin resistance, thereby leading to a reduction in blood glucose and insulin levels. Metformin has also been found to have other beneficial affects, some of which appear to be independent of its hypoglycemic activity. Included among these are its effects on lipids, inflammation, hemostasis, and endothelial cell and platelet function (Anfossi G, et al. Curr Vasc Pharmacol. 2010 Jan 1. [Epub ahead of print]; Matsumoto T, et al., Am J Physiol Heart Circ Physiol. 2008;295:H1165-H1176).

In women with PCOS, “reduction of hyperinsulinemia with metformin and diet is associated not only with improvement of the biochemical endocrinopathy, but, commonly, with restoration of menstrual cycles and fertility (Goldenberg, et al, Minerva Ginecol. 2008;60:63-75).” When used in infertile women with PCOS in combination with clomiphene citrate, an ovulation-inducing drug, metformin was shown to improve improve conception rates and, perhaps, live-birth rates compared to either drug alone (Legro, et al., N Engl J Med. 2007;356:551-66). In a recent small study of 66 women with PCOS who were clomiphene resistant and underwent in vitro fertilization, those who “received metformin (until conception) showed a significantly higher number of good quality embryos and implantation rate when compared with the placebo controls (Qublan, et al., J Obstet Gynaecol. 2009;29:651-5).” They were also found to undergo fewer spontaneous abortions in early pregnancy.

Very few studies have been done in which metformin therapy has been continued throughout the pregnancy, but in those that have, the results have been encouraging. Khattab and colleagues (Gynecol Endocrinol. 2006;22:680-4) studied 200 nondiabetic women who took metformin while undergoing assisted reproduction, of which 80 stopped the drug once they conceived and 120 continued it throughout pregnancy. Demographically, both groups were similar. Miscarriage rates “in the metformin group were 11.6% compared with 36.3% in the control group (p < 0.0001; odds ratio = 0.23, 95% confidence interval 0.11-0.42).” Similarly, Nawaz and colleagues (J Obstet Gynaecol Res. 2008;34:832-7), found that “In women with PCOS, continuous use of metformin during pregnancy significantly reduced the rate of miscarriage, gestational diabetes requiring insulin treatment and fetal growth restriction.” Furthermore, no significant congenital anomaly, intrauterine death or stillbirth in any of the woman who took metformin during in this study.

To support the observations in humans and, perhaps, to provide a mechanism of action, Luchetti and colleagues (J Steroid Biochem Mol Biol. 2008;111:200-7) found in mouse studies that hyperandrogenization, such as that which occurs in PCOS, induces embryo resorption in early pregnancy and that this is correlated with reduced production of progesterone-induced blocking factor (PIBF) and increased production of cyclooxygenase-2 (COX-2) - the overall effect of these changes creating a pro-inflammatory environment. Coincident treatment with metformin is able to reverse such changes and prevent early pregnancy loss in this animal model. To further support the overall beneficial effect of metformin in human pregnancy being the result of its overall anti-inflammatory properties, Orio and colleagues (Eur J Endocrinol. 2007;157:69-73) found in nonpregnant PCOS women that metformin treatment significantly reduced WBC count and C-reactive protein (CRP), reduced androgens, reduced low density lipids, and increased high-density lipids – all contributing to a reduction in the “proinflammatory” status of those PCOS women receiving metformin.

Finally, to answer our reader’s final concerns, in all the studies we have reviewed, in no instance did taking metformin, either during conception or throughout any time frame of pregnancy, appear to have a serious deleterious affect on the babies. Although the studies have been small, there does not appear to be a greater risk for spontaneous abortion, later pregnancy loss, or congenital anomalies (Goldenberg, et al., Minerva Ginecol. 2008;60:63-75; Nawaz, et al., J Obstet Gynaecol Res. 2008;34:832-7; Qublan, et al., J Obstet Gynaecol. 2009;29:651-5; Elizur, et al., Fertil Steril. 2008;89:1595-602; Bolton, et al., Eur J Pediatr. 2009;168:203-6; Ekpebegh, et al., Diabet Med. 2007;24:253-8). Furthermore, Bolton and colleagues (Eur J Pediatr. 2009;168:203-6) have reported that metformin is actually associated with beneficial effects of fewer growth restricted (< 10th percentile) and macrosomic (> 90th percentile babies) and fewer cases of neonatal hypoglycemia requiring glucose infusion.

Labels: diabetes, insulin resistance, metformin

Permalink | 12 Comments| Email Post

Post your comment

Assessing Fetal Lung Maturity - 2

The fetal lungs are the last organ system to “mature” so that survival outside the womb is possible. Maturity involves several components. First, there must be sufficient surface area within the lung to allow sufficient exchange of gases (oxygen in and carbon dioxide out) to support metabolic functions. This is accomplished by millions of small sacs called alveoli that give the lungs a sponge-like appearance. Second, the alveoli must develop to the point that the inner lining of cells (epithelial cells) that come in contact with inspired air are very thin – gas exchange can only occur over a short distance between the blood vessels in the alveoli and the air that fills the alveoli. Third, the alveoli must be able to remain open so that the air can get into them and gas exchange can take place. The first two events are generally quite complete by about 32-34 weeks, however, the third is the most essential component from that point on and it is the focus of our fetal lung maturity testing as we shall explain.

Alveoli are like little bubbles. The laws of physics predict that because of the high ratio of surface tension to volume of little bubbles, their tendency is to collapse. To prevent this from happening, certain cells in the lungs – the type II alveolar cells – begin to excrete chemicals that can reduce the surface tension in the alveoli. These chemicals are called ‘surfactants’ and they are a complex combination of phospholipids and apoproteins. When sufficient surfactants are produced that the alveoli can remain open to function, the fetal lungs are considered ‘mature’. Prior to this time, the baby is at risk for developing respiratory distress syndrome (RDS). RDS occurs in about 1% of all pregnancies and it can have serious short- and long-term consequences, involving both the lungs and other organs, that can extend beyond the neonatal period in its most severe forms.

When babies are very premature, RDS is the result of a combination of both alveolar epithelial cell immaturity (the lining cells have not yet thinned out) and a deficiency of surfactants. Later in pregnancy (“near term”), severe RDS can sometimes also occur but at this point it is usually the result of insufficient surfactants alone – but the end result can be just as devastating. Certain medical conditions can delay surfactant production in babies, the most common being maternal diabetes with poor blood sugar control and isoimmunization (such as Rh-disease). Large babies (macrosomic) are also at greater risk for RDS even if maternal diabetes is not contributing to the fetal size. Babies who have developed heart failure (hydrops fetalis) for any number of reasons can also have a relative deficiency of surfactants (as well as pulmonary edema). For reasons that are not entirely clear, babies delivered by cesarean section, particularly, when this is done prior to the onset of labor, are also at increased risk for respiratory difficulties.

The amniotic fluid is mostly fetal urine, but it is mixed with effluent from the fetal lungs as well. The fluid from the fetal lungs contains surfactants and other indicators of type II alveolar cell activity. When we perform an amniocentesis to assess fetal lung maturity, we are looking for direct and/or indirect evidence to support the presumption that there is adequate surfactant present to minimize the risk for RDS. One should realize that all of the tests we will discuss have some degree of “false positivity” – an indicator suggesting lung maturity, but the baby still develops respiratory problems – however, there is enough experience with their use that a “mature” value with a test result generally means the baby has at least a 95% chance of not developing RDS.

Having provided this information as background, the actual tests commonly used to evaluate fetal lung maturity will be discussed in our next post….

Labels: amniocentesis, diabetes, fetal lung maturity, fetal macrosomia, Rh-isoimmunization, surfactants

Permalink | 5 Comments| Email Post

Post your comment

Amniotic Fluid - 8 - Evaluation and Management of Polyhydramnios

In the last couple of posts we have reviewed causes and complications related to excessive amniotic fluid, otherwise known as polyhydramnios or, simply hydramnios. Although 50-60% of cases of hydramnios are idiopathic (no identifiable cause) and about 90% of cases are mild to moderate, about 10% are severe and these latter are the more likely to be accompanied by considerable fetal and neonatal morbidity and mortality secondary to an underlying fetal cause of the hydramnios – chromosomal abnormality, congenital anomaly, fetal anemia, inborn error, or congenital infection. Evaluation of the pregnancy with hydramnios, therefore, focuses primarily on these concerns.

The first step in any evaluation of hydramnios is to take a detailed medical, obstetrical, and family history and to review the course, medical complications, and basic laboratory studies performed to date in the current pregnancy. It is not necessary to cover the extent of such a discussion herein but examples of pertinent information include: the current pregnancy being a multiple gestation; a previous pregnancy accompanied by hydramnios, fetal macrosomia, or diabetes (and the outcome of that pregnancy); known maternal diabetes; known Rh- or other isoimmunization; history of blood transfusion; hemorrhage or trauma during the current pregnancy; history of known/suspected exposure to parvovirus B19 (Fifth’s disease) or illness accompanied by fever and/or rash during the current pregnancy; exposure to young children at home or in the workplace; family history of inborn errors of metabolism or congenital birth defects – particularly, cardiac, gastrointestinal, and neural tube, and neuromuscular disorders; past or family history of aneuploidy or recurrent pregnancy loss; advanced maternal age; report of decreased fetal movement; maternal history of medications and nonprescription (licit and illicit) drug use during the pregnancy. It is also important to get some feel for the onset of the hydramnios related to timing in pregnancy (e.g., gestational age when noticed; slow onset vs. rapid onset) and maternal signs and symptoms of disease and cardiorespiratory compromise.

The next step is to perform a thorough, high resolution ultrasound examination. In this, the degree of hydramnios should be documented objectively by a four-quadrant amniotic fluid index (AFI). This will be valuable as a ‘baseline’ for comparison during subsequent ultrasound evaluations of the pregnancy. Fetal growth should be assessed to determine if the baby is abnormally large or growth-restricted for the gestational age of the pregnancy – either of which might help narrow down the differential diagnosis. A detailed anatomical survey of the baby should include: central nervous system and spine; face and facial midline structures; neck; thorax; heart and rhythm; diaphragm; gastrointestinal tract; genitourinary tract; and, extremities. In addition, it is important to document whether or not the baby appears to have normal movement – flexion and extension – of the extremities since, if this is not present, it might suggest an underlying neuromuscular disorder. Evidence of fetal hydrops (indicative of fetal anemia or high-output cardiac failure) should also be sought.

If there is a twin (or higher order multiple) gestation, it is important not only to assess fetal growth and anatomy, but to determine chorionicity of the twins (i.e., dichorionic-diamnionic; monochorionic-diamnionic; or, monochorionic-monoamnionic) and if there is any significant discordance for growth or amniotic fluid surrounding the babies. Twin pregnancies are at higher risk for fetal anomalies, chromosomal abnormalities, abnormalities of placentation, and in monochorionic twins, a condition called ‘twin-to-twin transfusion syndrome (TTTS)’ (a discussion of which will be reserved for another post at another time).

A critical step in the evaluation of the pregnancy complicated by hydramnios (as it was in that complicated by oligohydramnios) is performing Doppler flow velocimetry (DFV) studies. These should be done at least on the fetal umbilical and middle cerebral arteries (MCA) and should be considered for the fetal ductus venosus and umbilical vein and the maternal uterine arteries. The goals of DFV under these circumstances are to ascertain if there is any difficulty perfusing the placenta (increased resistance indices) from either the fetal or maternal side; assess whether there is any evidence of fetal blood flow redistribution (“cranial sparing”) related to relative ‘placental insufficiency’ (decreased resistance to blood flow in the MCA); if there is increased peak systolic velocity (PSV) of blood flow in the MCA which would be suggestive of significant fetal anemia; or if there is evidence of fetal cardiac decompensation (abnormal wave forms – increased resistance or pulsatility - in the fetal ductus venosus or umbilical vein). DFV is a critical evaluation in the monochorionic twin pregnancy, especially if there is discordance for growth and/or amniotic fluid, that might help differentiate simple intrauterine growth restriction, or hydramnios related to aneuploidy or fetal anomalies, from TTTS.

Once a comprehensive ultrasound has been completed, a discussion should be held with the patient about what else can be done at this time, diagnostically and therapeutically, if indicated. Again, a detailed discussion of this is beyond the purpose of our post today, but some examples are as follows depending on the findings: 1) If the baby is growth-restricted and/or has visible abnormalities (major structural or subtle), an amniocentesis should be offered for fetal chromosomal studies and congenital infection, particularly, for cytomegalovirus (CMV). 2) Growth restriction with hydramnios and abnormal resistance to fetal placental-perfusion by umbilical DFV carries about a 50% chance of aneuploidy, even in the absence of visible abnormalities, so fetal karyotype should be encouraged with this combination of findings as well; 3) If there is increased PSV (> 1.5 MoM) in the fetal MCA, even in the absence of hydrops fetalis, then the baby may need to be evaluated for significant anemia – best done by percutaneous umbilical blood sampling (PUBS) with preparations made for coincident transfusion. This becomes even more critical if the baby already has hydrops; 4) If a fetal arrhythmia has been identified, medical therapy should be attempted to correct this condition; 5) If a twin gestation is present and there appears to be TTTS, then the patient should be counseled and offered a referral to one of the few sites in the country with the expertise to handle this condition.

As a routine part of maternal evaluation, especially if no readily apparent cause of the hydramnios is identified by ultrasound, I will frequently recommend the following: blood type and antibody screen; thyroid studies; a full 3-hour glucose tolerance test (unless the patient has already been diagnosed with diabetes); serologic testing for evidence of recent CMV or Parvovirus B19 infection and consider screening for toxoplasmosis and syphilis. If a woman is a known diabetic, I will include a hemoglobin A1C level and make efforts to optimize her diabetic control.

If a correctible cause for the hydramnios, such as fetal anemia, has not been identified and/or there are significant risks to the pregnancy because of the hydramnios itself, especially, if the pregnancy is less than 30 weeks and there is premature labor to contend with, or the mother has developed cardiorespiratory compromise secondary to massive hydramnios, there are limited options for management. Acute management of maternal cardiorespiratory decompensation may require amnioreduction. This is an amniocentesis procedure in which a large bore needle/catheter is inserted into the uterus and the fluid slowly drained until the AFI is in a ‘normal’ range of 10-20 cm. The most common risks to this procedure are rupture of membranes, premature labor, and placental abruption if the fluid is decompressed too rapidly. Unfortunately, since under normal circumstances, amniotic fluid volume is replaced daily, the fluid will often reaccumulate within 48-72 hours, necessitating repetitive procedures. Under these circumstances, the risk of the previously noted complications, as well as of infection, increase further.

As an adjunct to amnioreduction, or if the situation is not so acute, another option is to use potent prostaglandin synthetase inhibitors that have the effect of decreasing fetal urine production (and, hence, amniotic fluid) and may also decrease uterine contractions that usually accompany hydramnios, thereby, decreasing the risk of premature labor. Indomethacin has had the widest experience in this regard and is relatively safe for both mother and baby. After an initial loading dose of 100 mg, I will frequently place the patient on 25-50 mg of indomethacin every 6 hours. It usually takes at least 4 days (sometimes much longer) to get any response to this regimen. Once indomethacin has been started, it is important to monitor both amniotic fluid and the fetal ductus arteriosus which can constrict in response to the drug and is a primary means of maintaining the “fetal circulation” (bypassing the lungs and allowing proper distribution of well-oxygenated blood throughout the body) while the baby is in utero. One must be especially careful about using indomethacin in women who have underlying kidney problems, cardiac disease, long-standing diabetes, hypertensive disorders, pregestational and pregnancy-related preeclampsia, or evidence of infection because if their renal output also drops significantly, they can be pushed into congestive heart failure.

Another prostaglandin inhibitor that has also been tried, and with which I must admit limited experience, is sulindac (usually dosed at 200 mg every 12 hours). Sulindac has greater selectivity for the cyclooxygenase 2 (COX-2) enzyme and appears to be capable of reducing fetal urine output with less of an effect on the ductus arteriosus, although its effect on the fetal kidneys is also less than that of indomethacin. It may be safer to use later in gestation than indomethacin which I will usually stop at 32 weeks (and no later than 34 weeks) gestation. The risk of premature delivery is so high with severe hydramnios requiring amnioreduction and/or prostaglandin inhibitor therapy that I often couple their use with a course of corticosteroids to accelerate fetal lung maturation in the event that delivery occurs or becomes necessary.

In closing, I would like to mention only one other caution about hydramnios that is often over-looked with regard to my last statement in the paragraph above. If hydramnios is present and associated with diabetes and/or fetal macrosomia, fetal lung maturation may be delayed as much as 2-3 weeks as the result of hyperinsulinemia in the baby. Hyperinsulinemia suppresses the development of lung surfactants and one last study that should be considered, and is highly recommended, prior to the elective delivery of baby because of hydramnios, or macrosomia, is an amniocentesis to assess fetal lung maturity, especially if the planned delivery is by cesarean section.

Well, this concludes our series on amniotic fluid. As I said at the outset, evaluation of amniotic fluid is an important part of every pregnancy and understanding the causes, complications, and management of the pregnancy with abnormalities of amniotic fluid is a daily part of my routine. I have tried to make our discussions digestible for the nonclinician as well as a valuable overview for the primary care professional involved in the care of women during pregnancy and hope that we have accomplished that here! Thanks for reading!
Dr T

Labels: CMV, diabetes, Doppler flow velocimetry, fetal macrosomia, polyhydramnios

Permalink | 10 Comments| Email Post

Post your comment

Amniotic Fluid - 7 - Complications Related to Polyhydramnios

When excessive amniotic fluid (polyhydramnios or, simply, hydramnios) is present, there are increased risks of complications for both mother and baby. Some of the risks to the baby are obvious if there is an identifiable etiology for the hydramnios, such as maternal diabetes, multiple gestation, congenital malformation, chromosomal abnormality, severe fetal anemia secondary to isoimmunization or Parvovirus B19, neuromuscular disorders, or congenital infection. Indeed, past reviews confirm the risk for poor outcomes when an etiology is found. For example, Stoll and colleagues (Community Genet 1999;2:36-42) identified 290 cases of polyhydramnios in 225,669 consecutive pregnancies and diagnosed congenital malformations prenatally in 44.5% of the cases. Among these, 10.3% of the infants were stillborn, 41% had more than one malformation, 14.5% had a chromosomal abnormality.

Similarly, Biggio and colleagues (Obstet Gynecol 1999;94:773-7) compared 370 women with singleton pregnancies beyond 20 weeks' gestation and hydramnios with 36,426 controls who had normal amniotic fluid volumes. “The perinatal mortality rate in all women with hydramnios was 49 per 1000 births, compared with 14 per 1000 births in the control group (P < .001). Women with hydramnios had 25 times more anomalies than controls (8.4% versus 0.3%; P < .001)…the cesarean rate was three times higher in women with hydramnios compared with controls (47.0% versus 16.4%; P < .001).” Interestingly, in their study, the increased risks were concentrated in the nondiabetic women with hydramnios.

However, as we mentioned previously, 50-60% of hydramnios is idiopathic (without an identifiable cause). So the question remains, are there increased risks to the baby if no identifiable etiology for the hydramnios is found? In other words, does the excessive fluid alone seem to contribute to or be associated with poor perinatal outcome. The scientific literature would indicate that it does. For example, Magann and colleagues (Obstet Gynecol Surv 2007;62:795-802) recently presented an extensive review dating back more than 50 years and found that idiopathic hydramnios was linked “to fetal macrosomia (in the absence of diagnosed maternal diabetes), an increase in the risk of adverse pregnancy outcomes, and a 2- to 5-fold increase in the risk of perinatal mortality.” So, what are some of the pregnancy risks, irrespective of the cause of the excessive amniotic fluid.

Common risks secondary to overdistention of the uterus include abdominal pain, premature labor and delivery, and premature rupture of membranes. There is also an increased risk of uterine rupture, although this is rare in the absence of a previous cesarean delivery or other operative uterine procedure. In the presence of severe hydramnios, especially in a woman of small stature, overdistention of the uterus can put so much pressure on the mother’s diaphragm that she has difficulty breathing in ANY position and maternal cardiorespiratory decompensation may occur under these circumstances.

Often under these circumstances, placental perfusion is also reduced, the baby develops relative placental insufficiency, and as a consequence of the baby’s (and probably the placenta’s) unhappiness, the mother develops preeclampsia. Doppler flow studies have shown a greater incidence of fetal blood flow ‘redistribution’ (an indirect indicator of ‘placental insufficiency’) in the presence of hydramnios and this is most likely due to the excessive pressure on the umbilical vessels and the placenta itself resulting in decreased fetal perfusion. Indeed, any fetal condition associated with hydramnios that places the baby in a ‘distressed’ situation, particularly, severe fetal anemia and other causes of hydrops fetalis, increases the risk for maternal preeclampsia.

Indeed, the very first obstetrical patient I ever saw die (30 years ago) had a baby with hydrops secondary to severe maternal Rh-isoimmunization and polyhydramnios. An attempt was made to transfuse the baby in utero and afterwards she was sent to the antepartum unit for monitoring. I noticed her blood pressure was elevated and checked her urine to also find 4+ proteinuria. I remember notifying her attending physician ( I was a second year resident at the time) that she appeared to be developing severe preeclampsia and was brushed off that this was simply the ‘stress of the procedure that she had just been through.’ When I came in to round on her the next morning, she was not in her bed and when I asked if she had been discharged, I was told that she had had a hypertensive crisis in the middle of the night, a cerebrovascular accident, and could not be resuscitated. The occurrence of severe maternal preeclampsia in the presence of fetal hydrops has come to be known as “mirror syndrome” in which the mother’s condition reflects (and is probably driven by) the dire fetal condition (Vidaeff, et al. J Reprod Med 2002;47:770-4). Needless to say, there are some things one NEVER forgets!

Hydramnios can also cause several complications related to the onset and course of labor. Too much fluid often leads to lack of ‘engagement’ of the fetal head in the pelvis and/or an unstable fetal lie (breech or transverse). This can be a special problem when the membranes rupture (spontaneously or artificially) because if there is no ‘presenting part’ obstructing the cervix, the umbilical cord can suddenly prolapse with the gush of fluid through the cervix into the vagina turning a relatively uncomplicated situation into an emergency. Acute release of the fluid and decompression of the uterus can also cause sudden separation of the placenta (placental abruption) from the uterine wall. Stretching of the uterine muscle (myometrium) can also result in abnormal labor patterns secondary to poor contractility (myometrial dysfunction) and at times can result in poor contraction (involution) of the uterus following delivery, a situation that is usually accompanied by post-partum hemorrhage. All of these complications contribute to the increased rate of cesarean deliveries in pregnancies with hydramnios and the increased rate of maternal and fetal complications.

One other complication which occurs frequently (and is often not thought about) in the presence of hydramnios, particularly if this is associated with diabetes or simply, with fetal macrosomia, is immaturity of fetal lung development. As we have pointed out in earlier posts, late preterm (near-term) elective delivery of a baby just because it is “too big” can have tragic consequences. It is not unusual for macrosomic babies to have a 2-3 week lag in the functional ability of their lungs at birth because excessive insulin production (hyperinsulinemia) that often accompanies macrosomia can delay the production of the lung surfactants that reduce surface tension in the alveoli and are necessary for expansion of these so that oxygen exchange can occur normally. There is nothing sadder than seeing a 10 lb baby of a diabetic mother laying in the neonatal intensive care unit struggling to survive with severe respiratory distress syndrome and persistent fetal circulation as a consequence of an elective (often cesarean) delivery.

Having discussed some of the more common complications of polyhydramnios, in our next (and final!?!) post on the topic of amniotic fluid, we will address the evaluation and management of the pregnancy with too much amniotic fluid…

Labels: diabetes, hydramnios, polyhydramnios, Rh-isoimmunization

Permalink | 14 Comments| Email Post

Post your comment

Diabetes in Pregnancy - 11 - Hemoglobin A1c and Diabetic Control

In our last post, we discussed the preconceptional evaluation of the woman with pregestational diabetes – this evaluation pertains to women with either type 1 or type 2 diabetes. As part of the evaluation, we mentioned screening for ‘hemoglobin A1c’ and one of my local readers asked me to explain that in more detail. She is, herself, a nurse, and I appreciate her reminding me that sometimes I get ahead of myself without providing adequate explanations to our readers, even those who are medical professionals themselves. I confess this is a common fault of our profession, so let me see what I can do to offer an explanation of what it is and why we suggest including it in the evaluation of diabetic women…

Hemoglobin is contained in red blood cells and is the oxygen-carrying component of red blood cells and, because of its iron content, it is that which makes blood appear red. In individuals without genetic abnormalities of their hemoglobin, hemoglobin A (which stands for ‘Adult’), is the predominant type of hemoglobin found in the red blood cells. In normal (nondiabetic) individuals, more than 92% of the hemoglobin A is simply hemoglobin A. However, a small percentage of the hemoglobin A molecules have sugar (glucose) that is attached to them. These ‘glycosylated’ hemoglobin molecules are called ‘hemoglobin A1c’.

The percentage of hemoglobin A that actually becomes hemoglobin A1c depends on the blood glucose levels over the long-term, generally six to eight weeks (although some believe that 3-4 weeks is more accurate). The higher the average blood glucose levels, the higher the percentage of hemoglobin A1c. Once a hemoglobin molecule becomes ‘glycosylated’, it stays that way for the life of the red blood cell (about 120 days) (Bunn, et al., Biochem Biophys Res Commu. 1975;67:103–9). Short-term fluctuations of blood sugar, such as those associated with meals, do not have much effect on the concentrations of hemoglobin A1c, unless of course, they are abnormal and contribute to an unusually high range overall. For these reasons, levels of hemoglobin A1c are a direct reflection of blood glucose control over time.

In healthy individuals, hemoglobin A1c comprises less than 7% of the total hemoglobin that is present. Conditions that can falsely elevate levels of hemoglobin A1c include kidney failure, hypertriglyceridemia, and folate and vitamin B12 deficiencies that are accompanied by slower rates of red blood cell turnover. Conditions that cause more rapid turnover of red blood cells, such as blood loss, sickle cell disease, or glucose-6-phosphate dehydrogenase (G6PD) deficiency, can falsely decrease levels.

Rahbar and colleagues (Biochem Biophys Res Commun 1969; 36: 838–43) first identified the presence of elevated levels of hemoglobin A1c in diabetics and others have correlated the levels with degree of blood glucose control (Koenig, et al., N Eng. J Med 1976; 295: 417-20). To give some perspective on this, average blood sugars of 90 mg/dL are associated with hemoglobin A1c concentrations of about 5%, blood sugars of 120 mg/dL with 6%, 150 mg/dL with 75, 180 mg/dL with 8%, and so on. The International Diabetes Federation and American College of Endocrinology suggest that ‘normal’ levels of hemoglobin A1c be considered values below 6.5%.

As pointed out in our previous post, poor diabetic control, as reflected in elevated levels of hemoglobin A1c during the time of embryogenesis, is associated with both early pregnancy loss and congenital anomalies. For example, as shown by Miller and colleagues (New Engl Med J 1981;304:1331-1334) a hemoglobin A1c level >8.5% confers a risk of birth defects of approximately 22% versus 3.4% in women with A1c levels <8.5%. On the flip side, normal hemoglobin A1c levels during this critical time in fetal development, even in long-term diabetics with other complications, are generally accompanied by a risk for birth defects close to that of the nondiabetic population.

It is important to note, however, that it is not the hemoglobin A1c level that causes the birth defects, it is the poor diabetic control (high maternal blood sugars). In other words, if a diabetic comes in for preconceptional counseling and has a hemoglobin A1c of 11.5% (very bad), quickly normalizes her blood sugar control according to our strict standards, then gets pregnant a week later and continues good control of her blood sugars during the period of her baby’s embryogenesis, the hemoglobin A1c level may be high, but the baby’s risk should be minimal. On the other hand, if she says “I have normal blood sugars” and she gets to 14 weeks and her hemoglobin A1c is 11.5%, check her glucometer and be on the look out for major congenital malformations!

Labels: diabetes, hemoglobin A1c

Permalink | 0 Comments| Email Post

Post your comment

Diabetes in Pregnancy - 2 - Glucose Metabolism and Insulin Action

To continue the discussion of diabetes in pregnancy, let’s first start with a very simplified review of glucose metabolism and the role of insulin so that we understand what the furor is all about….


Glucose is one of the primary substrates for the production of energy in cells throughout our body, especially, in the brain, muscles, and kidneys. The two main sources of glucose are dietary carbohydrates, primarily, starches (large “sugar” polymers, or complex carbohydrates) from fruits, vegetables, and grains, and disaccharides (made up of only two “sugar” molecules), such as lactose from milk and sucrose from refined sugars. Through the digestion of these carbohydrates in the gut, they are first broken down to monosaccharides (single “sugar” molecules) before they can be absorbed. Once they cross the gut wall, these monosaccharides are transported via the blood to the liver where they are all converted into glucose molecules (also a monosaccharide). The other monosaccharides cannot be used for ‘fuel’ until they are converted into glucose, therefore, the liver’s role is pivotal in the distribution of this primary source of energy to other body tissues.

Once glucose has gotten into cells, several things can happen to it. First, if the cell needs ‘energy’, glucose can be broken down by a series of enzymes that ultimately results in the production of two molecules of pyruvate. (Pyruvate is the substrate for many important reactions that are not necessary for our discussion herein). As a consequence of the production of pyruvate from glucose, there is also the concomitant net production of two molecules of ATP (adenosine triphosphate) which is the common source of ‘fuel’ for most of the metabolic processes in the cell. Secondly, if there is more glucose than the cell needs, glucose can be converted to glycogen (a polymer of glucose) and stored in that form. Only the liver and muscles have the capability to make and store glycogen (interestingly, the brain does not!) but they are limited in how much they can store. Glycogen can rapidly be converted back to glucose when the need arises. Thirdly, if there is an excess beyond the capacity of the liver and muscles to store glycogen, glucose can be converted into ‘fatty acids.’ Fatty acids are stored in fat (adipose) tissues as triglycerides.

So, where does insulin fit into this picture? Well, none of the events above can occur unless glucose can actually get into the cell in the first place. Among insulin’s many functions, its primary role for sake of our discussion is to enhance the ability of cells to get glucose from the blood into the cell. Under normal circumstances, insulin secretion by the pancreatic β-cells in the islets of Langerhans is proportional to the amount of glucose in the blood within a very tight ‘normal’ range for blood glucose levels. (Actual production of insulin within the β-cells is under the influence of many different hormones, including several pregnancy-related hormones). Once insulin is secreted, it is transported in the blood and binds to specific receptors on the cell membrane. Binding of insulin to the cell membrane results in signals that activate several different metabolic processes (specific to the cell type to which the insulin has bound).

One of the messages that insulin sends to the cell by binding to it is to increase the number of plasma membrane glucose transporters, or GLUTs. GLUTs are stored in the cell and are ‘recruited’ to move to the cell membrane by the action of insulin. Once there, they facilitate the transport of glucose from the blood into the cell. GLUTs are continuously being ‘turned over’ and they only hang around in the cell membrane as long as enough insulin is bound to the cell to keep them there. Different GLUTs with different affinities for glucose, characteristically, reside in different tissues: GLUT1 is present in most tissues, GLUT2 is present in liver and pancreatic β-cells, GLUT3 is found in the brain, and GLUT4 is found in the heart, adipose tissue and skeletal muscle. In addition to enhancing glucose transport, insulin stimulates glycogen and fat production, increases amino acid transport into cells, promotes the transcription of specific gene products, and stimulates growth, DNA synthesis, and cell replication.

In our next post, we will discuss how pregnancy alters the ‘normal’ state of affairs with regard to insulin production and action…

Labels: diabetes, insulin

Permalink | 5 Comments| Email Post

Post your comment

Diabetes in Pregnancy -1 - Overview of an Epidemic

Today we admitted a 25 year old woman to the hospital that has Class R/F diabetes and is 9 weeks pregnant. I will explain what ‘Class R/F’ means later on but, simply put, it’s not good because both her kidneys and her eyes have suffered severe damage from long-standing poorly-controlled diabetes. The outcome of pregnancies under these circumstances is usually suboptimal for both mother and baby. Anyway, the reason for her admission was to try to get her diabetes under control. She is the third such patient we have cared for in the past month. I will return to her story in a later post, but she has given me the incentive to begin a long overdue series on diabetes in pregnancy, and so we begin….

Diabetes mellitus is a group of conditions characterized by elevated levels of blood glucose (sugar) that results from inadequate amounts of insulin production, defects in insulin action (e.g., ‘resistance’ to its action), or both. Despite the advances that have been made in the management of diabetes over the years, we are now confronting a worldwide epidemic of diabetes that is being promulgated by the comparably worldwide epidemic of obesity. The World Health Organization has estimated that, barring some remarkable changes in lifestyle, eating, and exercise habits, over the next 20 years there will be a 35% increase in the prevalence of diabetes. To put the magnitude of this problem in perspective, according to International Diabetes Federation statistics, in 1985, an estimated 30 million people worldwide had diabetes; in 2000, a little over a decade later, the figure had risen to over 150 million; by 2025, the figure is expected to rise to 380 million. Quite frankly, that may well be a conservative estimate.

In the U.S. between 1980 and 2005, the number of individuals diagnosed with diabetes increased from 5.6 million to 15.8 million. That reflects an increase in crude prevalence of 120% over that time period. Today, 5-6% of all Americans know they have diabetes and it is estimated that one-third or more of individuals who have the condition, currently do not even know it. Although we generally attribute a large percentage of the increase in prevalence of diabetes to the aging population, the mean age of onset of disease is falling and many young women in the child-bearing years are at increased risk for developing diabetes in pregnancy or for having undiagnosed ‘pregestational’ before they conceive. The dangers of this will be addressed later. In the state of South Carolina where I work, more than 8% of the population has documented disease. The condition is more common in Black women and they tend to develop it at younger ages. Diabetes also appears to be rapidly increasing in the rapidly expanding Hispanic population.

Overall, nearly 5% of pregnancies are complicated by diabetes. Ninety percent of these women will have ‘gestational’ diabetes that is not present early in the pregnancy, but develops by late second or early third trimester. Reasons for this will be discussed later as well. Routine screening for diabetes in pregnancy is a ‘standard of care’ and detects 85-90% women with the condition by 28 weeks gestation. Whether pregestational or gestational, diabetes is associated with increased risks for both fetal and maternal morbidity. Although the risks can be reduced almost to that of the nondiabetic population by good control of blood glucose levels, many pregnant women are unable or unwilling to devote the time and effort necessary to achieve that control and, even more sadly, many providers are also either unable, unwilling, or do not have the time or resources it takes to provide the educational support and ongoing care necessary to optimize pregnancy outcome.

In the next several posts, we will focus our discussion on insulin and its action, the physiological changes that occur during pregnancy that help to ‘bring out’ the disease, and on ‘gestational diabetes’ and its diagnosis and management.

Labels: diabetes, gestational diabetes

Permalink | 0 Comments| Email Post

Post your comment