In our last post, we responded to several questions from a reader named Kate concerning the implications of low PAPP-A levels detected during the course of first trimester screening for aneuploidy. Not addressed in our last post were her two other comments: Is there a correlation with celiac disease and placental health or low PAPP-A levels? I have read reports of celiac disease causing many of same risk factors associated with low PAPP-A.
To begin to address these issues, it is important to discuss first what celiac disease is to better understand how it might have an impact on pregnancy. Celiac disease (CD) is the result of intolerance of the gastrointestinal tract to gluten. Gluten is a protein contained in wheat, barley, and rye and is therefore present in many cereals, pasta, flour, breads and as an additive in many processed foods, medicines, vitamins, and even lip balm. As the consequence of an abnormal immune response (with production of antibodies such as antigliadin type IgA and IgG and IgA antitransglutaminase), individuals with CD damage their own intestinal villi that are necessary for the absorption of nutrients into the blood stream. Thus celiac disease can result in not only gastrointestinal symptoms, but malabsorption and nutritional deficiencies that can have significant consequences on the long-term health of an individual if CD is not recognized and treated. However, the complications related to CD can be dramatically reduced by strict adherence to a “gluten-free” dietary regimen.
The tendency to develop CD appears to be genetically inherited but there is a great variability in age of onset, severity of symptoms, and consequences of the condition even within families. Fasano and colleagues (Arch Int Med 2003;163:268-92) estimated that in the U.S., about 1 in 133 people have CD. Rodrigo reported (World J Gastroenterol 2006;12:6585–6593) that 4 to 12 percent of an affected person’s first-degree relatives will also have the disease. Gastrointestinal symptoms of CD include abdominal pain and bloating, chronic diarrhea, pale, foul-smelling and fatty stool, occasionally vomiting and constipation, and weight loss. Even in the absence of significant gastrointestinal complaints, individuals with CD may develop conditions such as iron deficiency anemia, osteoporosis and bone pain or arthritis, chronic fatigue, depression, anxiety, numbness and tingling in the hands and feet, oral ulcers, and even seizures. Dermatitis herpetiformis, an itchy, blistering skin rash, affects 15 to 25 percent of individuals with celiac disease. CD has even been implicated in menstrual irregularity, infertility, and recurrent early pregnancy loss. There is also an association of CD with chronic liver disease and intestinal cancer.
With all that said and done, let’s begin to look at the consequences of celiac disease for pregnancy. Nørgård and colleagues (Am J Gastroenterol 1999;94:2435-40) reviewed birth outcomes in Danish women with CD between 1977 and 1992. They found that before women were first treated for CD there was an increased risk of low birthweight (odds ratio [OR] = 2.6, 95% CI = 1.3-5.5) and intrauterine growth retardation (OR = 3.4, 95% CI = 1.6-7.2). Pregnant women who were given dietary counseling before delivery had no increased risk of low birthweight and no babies with intrauterine growth retardation. Therefore, treatment of women with CD is important in the prevention of fetal growth retardation. Similarly, Ludvigsson and colleagues (Gastroenterology 2005;129:454-63) compared pregnancy outcomes in 1149 offspring of women diagnosed and treated for CD prior to birth and 929 offspring to women diagnosed after delivery. They found that “undiagnosed CD was associated with an increased risk of intrauterine growth retardation (OR = 1.62; 95% CI: 1.22-2.15), low birth weight (OR = 2.13; 95% CI: 1.66-2.75), very low birth weight (OR = 2.45; 95% CI: 1.35-4.43), preterm birth (OR = 1.71; 95% CI: 1.35-2.17), and caesarean section (OR = 1.82; 95% CI: 1.27-2.60). In contrast, a diagnosis of CD made before the birth was not associated with these adverse fetal outcomes.” Again, the risks of CD, at least with regard to fetal growth, can often be dramatically reduced by identification and treatment of women with CD.
However, because CD is the result of an aberration of the immune response, reminiscent of that seen in autoimmune conditions, the questions remain, could this abnormal immune response also contribute to poor pregnancy outcome in ways other than impairing fetal growth and what are the mechanisms of the impairment in growth and perhaps other suboptimal pregnancy outcomes in women with CD? It is well-known that people with CD are at greater risk for having or developing other autoimmune conditions such as type 1 diabetes, autoimmune thyroid, liver and adrenal disease, and other autoimmune conditions such as rheumatoid arthritis and Sjogren’s syndrome. Conceivably, autoantibodies causing these conditions could cross the placenta to the baby and cause complications. Indeed, in the case of Sjogren’s syndrome, this is often associated with the production of antibodies (anti-Ro (SS-A) and anti-La (SS-B)) that can cause congenital heart block and cardiac malformations.
Although an association of CD with systemic lupus erythematosus and antiphospholipid antibody syndrome has not been widely recognized, there have been very recent case reports of the same (Gupta, et al., Rheumatol Int. 2008;28:1179-80; Jorge, et al., Rev Esp Enferm Dig 2008;100:102-3). Furthermore, the presence of anticardiolipin antibodies have been demonstrated in some patients with CD (Karoui, et al., Dig Dis Sci. 2007;52:1096-100) and CD antibodies have been demonstrated in some patients with antiphospholipid syndrome (Shamir, et al., Lupus 2004;13:214)! If this holds true, there is a subpopulation of pregnant women with CD who will be at increased risk for thromboembolic complications and, perhaps, abnormalities of placentation that could lead to intrauterine growth restriction, pregnancy-induced hyperetensive disorders, premature delivery, and even ‘unexplained’ fetal death. And, to answer Kate’s question, when this is associated with an abnormality of placentation, there is a very good chance that first trimester PAPP-A levels will be on the low side – indeed, perhaps we should be screening for CD in women with low PAPP-A and also we should consider screening of women with known CD for PAPP-A to possibly help identify that subpopulation of women at risk for the pregnancy complications we have noted above.
Diagnosis of CD is usually made by the detection of anti-tissue transglutaminase (tTGA) and anti-endomysial (EMA) antibodies in the blood (and confirmed by duodenal biopsy). Anti-gliadin antibodies can also be demonstrated in many individuals with CD. The question then arises, could these CD-associated antibodies cause damage to the placenta and/or the fetus and compromise pregnancy outcome? Robinson and colleagues (Placenta 2006;27:148-57) have demonstrated that tissue transglutaminase (tTG) protein and mRNA are expressed in stromal cells and trophoblasts in first trimester and at term, with higher levels later in pregnancy and suggested that this could be a target for CD-associated antibodies. To support this hypothesis, Bustos and colleagues (Am J Reprod Immunol 2006;55:201-7) have shown that the prevalence of positive antibodies for antigliadin type IgA and IgG and IgA antitransglutaminase in women with recurrent pregnancy loss RPL was significantly higher than controls (P < 0.04).
In a very interesting article, Hadziselimovic and colleagues (Fetal Pediatr Pathol 2007;26:125-34) looked at the amount of gliadin that was present in term placentas of compliant (women who maintained a gluten-free diet) and noncompliant CD patients and correlated that with placental damage to the extravillous trophoblasts and fetal birth weight. They described the extravillous trophoblasts in the noncompliant women as being “overloaded with gliadin” whereas there was moderate to no gliadin present in the controls. “The weight of newborns was lower if extravillous trophoblasts were loaded with gliadin (-2.24SD) (p = 0.004)” and there was increased apoptosis (cell death) of the trophoblasts in the placentas of the noncompliant women.
In conclusion, there appear to be various mechanisms by which CD might deleteriously affect pregnancy outcome, and perhaps others we have not yet explored, such as specific nutrient deficiencies that might lead to certain fetal anomalies. The ‘autoimmune’ component of CD seems to play a major role in terms of fetal growth impairment as pregnancy progresses and perhaps by increasing the risk for early abnormalites of placentation that could contribute to both early pregnancy loss and later pregnancy complications. The reassuring point we can glean from the literature regarding CD and pregnancy is that strict adherence to a gluten-free diet appears to overcome many of the risk factors that have been correlated with CD. I wish that could be true for some of the other major pregnancy complications we face on a regular basis!