Family Practice Vol. 20, No. 5, 607-611
© Oxford University Press 2003
Selections from Current Literature |
Gluten-free diets, coeliac disease and associated disorders
Department of Family Medicine, State University of New York at Stony Brook, Stony Brook, NY 11794-8461, USA. Correspondence to Lorraine Danowski; E-mail: ldanowski{at}notes.cc.sunysb.edu
Danowski L, Garguila Brand L and Connolly J. Gluten-free diets, coeliac disease and associated disorders. Family Practice 2003; 20: 607–611.
Received 7 May 2003; Accepted 19 May 2003.
Introduction
Coeliac sprue is a life-long inflammatory condition of the gastrointestinal tract that affects the small intestine of genetically susceptible individuals. This condition is a chronic malabsorptive disorder caused by exposure to dietary gluten. Villous atrophy, a lowering of the villous height to crypt depth ratio, an increase in intraepithelial lymphocytes and extensive surface cell damage and infiltration of the lamina propria with inflammatory cells are characteristic of the disease.1 Prompt improvement of nutrient absorption and healing of the intestinal mucosa is seen upon withdrawal of gluten from the diet.2
Dermatitis herpetiformis is considered an extraintestinal manifestation of coeliac sprue. This condition presents as a pruritic, blistering rash. Treament involves dapsone and a gluten-free diet, which, if strictly followed, may allow for withdrawal of the medication.1 In children, the onset of coeliac disease occurs within the first or third years of life after introduction of gluten into the diet. These children present with a classic syndrome of chronic diarrhoea, failure to thrive and abdominal distention.3 Atypical coeliac disease is seen in older children or adolescents, who display no overt features of malabsorption. In addition to recurrent abdominal pain, hypertransaminasaemia, stomatitis, arthralgia and defects in dental enamel, children may also experience depression, irritability or poor school performance.4
Adult presentation is increasingly common and can occur at any age. The prevalence varies widely in different countries. European studies show a prevalence rate of between 1:152 and 1:300. In the USA, this condition may affect 1:250.1
The endomysial antibody (EMA) test has become a simple reliable screening tool. This test has enabled accurate determination of prevalence.5 Silent coeliac disease has been found to be more frequent in patients with type I diabetes mellitus than in controls. First-degree relatives of diabetic patients also showed a higher incidence of coeliac disease. The higher prevalence of other autoimmune diseases in subjects with both coeliac disease and type I diabetes may be related to delayed diagnosis of coeliac disease.6 Evidence supports the fact that coeliac disease should be considered in cases of ‘unexplained’ female infertility and as a possible risk factor for an adverse pregnancy outcome.7
Martinelli P, Troncone R, Paparo F et al. Coeliac disease and unfavourable outcome of pregnancy. Gut2000; 46: 332–335.
Coeliac disease has been connected with adverse outcome of pregnancy. The goal of this study was to estimate the prevalence of untreated coeliac disease in women attending the obstetric/gynaecological department of a major city hospital, in an attempt to determine its effect on the reproductive status of these women. Serum collections from 845 pregnant woman were screened for EMAs. Patients with positive results were offered a small intestinal biopsy to confirm the diagnosis.
Twelve of 845 samples were positive. Three of these 12 women had a diagnosis of coeliac disease from childhood which was confirmed by biopsy. They did not report the disease on admission and had not received treatment for 10–25 years. This was a first pregnancy for all three women, and two had a breech presentation of the baby. None of the three reported gastrointestinal symptoms or were underweight.
Nine patients without a history of coeliac disease who had positive results underwent small intestine biopsy. The pathologist was blinded to serum results. In all cases, severe to total villous atrophy, crypt hyperplasia and lymphocyte infiltration were found. Of the newly diagnosed patients, three babies died, one had a breech presentation, one experienced pre-eclampsia, and premature deliveries were expected in two. An adverse neonatal outcome occurred in seven of the newly diagnosed cases: five small for gestational age and three preterm deliveries. Depressed haemoglobin was reported in four of the 12 newly diagnosed women with coeliac disease.
A control group was obtained by randomly selecting 206 women with negative results from the coeliac screening. Age was similar in both groups, but menarche occurred later in the patients with coeliac disease. A history of previous miscarriage was more common in coeliac patients than in controls. Two of the 206 controls had a stillbirth and another two infants suffered from severe perinatal disease; no babies died after birth. Three out of the 13 babies born to women with coeliac disease died during the first week of life. All 12 women with coeliac disease went to delivery. The mean gestation was similar in coeliac patients and controls, but 33% terminated before the 37th week in coeliac patients versus 11.6% in controls. Breech presentation occurred in 25% of the coeliac patients and 1.4% in controls. Mean birth weight was also lower in patients with coeliac disease than controls. In this study, 41% of the 12 patients diagnosed with coeliac disease had unfavourable outcome of pregnancy or low birth weight babies.
Comment
These data imply that with minimal expense, 
US$100 per test, a negative outcome of pregnancy might be reduced or eliminated. In a case–control study, a comparison of 94 untreated and 31 treated coeliac women indicated that the relative risk of abortion was 8.90 times higher and the relative risk of a low birth weight baby was 5.84 times higher in untreated mothers. In the before– after study, 12 of the treated and untreated women were compared. Results indicated that adherence to a gluten-free diet reduced the relative risk of abortion by 9.18 times and the number of low birth weight babies from 29.4% to zero.12 The 12 cases that were identified are slightly higher than in the general population. The authors point out that this disease is more common in women than in men and accounts for this higher percentage. They also stated that the 12 cases had no overt signs of malnutrition and no one was underweight. This led to the conclusion that nutritional factors were probably not of major importance in the negative outcomes reported.
Follow-up was as follows: one was lost to follow-up; three had no further pregnancies and were complying with the gluten-free diet; eight patients were pregnant, with one still to deliver, and the other seven reached term. Six of these patients gave birth to healthy babies and the seventh patient gave birth to a baby with a major cardiac malformation. Notably, the seventh patient was the only subject on a gluten-containing diet.
Perhaps it is the dysregulation of the immune system that may account for an unfavourable outcome of pregnancy despite only minor nutritional abnormalities.8,9
This article did not report the length of gestation at the time of diagnosis or when the gluten-free diet was initiated for the 12 cases identified. Follow-up results were promising for a simple, inexpensive screening and adherence to a gluten-free diet.
Abdulkarim A, Burgart L, See J, Murray J. Etiology of nonresponsive celiac disease: results of a systematic approach. Am J Gastroenterol2002; 97: 2016–2021.[CrossRef][Medline]
Non-responsive coeliac disease (NCD) is a lack of response to a gluten-free diet (GFD) or recurrence of symptoms in a patient despite adherence to the diet after initial positive response to these restrictions. Clinical improvement is usually seen within the first few weeks of a GFD. Intestinal mucosa restoration may take up to 2 years. The goals of this study were to determine the aetiology of persistent and relapsing symptoms in coeliac patients referred to a tertiary care centre for evaluation of NCD and to characterize further those patients identified as having true refractory sprue (RS).
NCD was defined as persistence or recurrence of symptoms for up to a year, despite presumed adherence to a GFD. RS was defined as persistent symptoms and evidence for histological injury despite adherence to a GFD, as evaluated by a dietitian with clinical expertise, for up to a year. Fifty-five patients with a presumed diagnosis of NCD from a single referral centre underwent a systematic sequential evaluation which included a detailed dietary review, serological testing for coeliac disease, repeat small intestine and colonic biopsies, small intestine aspirates for quantitative culture, 72 h fecal fat measurement, small intestine radiographic studies and computed tomography (CT) body imaging. All tests were not completed on all patients if a cause was identified and treatment was successful. A comprehensive assessment of dietary compliance was the initial step of evaluation and was comprised of three parts: physician interview of the patient to determine their perspective of the GFD, interview and detailed evaluation by an expert dietitian about dietary compliance and serological tests (EMAs and gliadin antibodies).
Forty-nine cases were not diagnosed at the investigators’ facility. In thirty-two of these cases, the original biopsy slides were retrieved and reviewed by the same pathologist. The remaining cases confirmed diagnosis by original biopsy report, repeat biopsy, serological markers and response to GFD. Fifty-five patients were studied in total. After extensive evaluation, six patients proved to be negative for coeliac disease. Nine of the remaining 49 were identified as RS. Four of these patients had RS alone and the other five had RS in addition to another condition. Twenty-five patients had gluten contamination of their diet. The source of contamination was found to be commercially packaged corn or rice cereals that contained malted barley, or frequent dining out. Once the diet was strictly followed, symptoms resolved or improved in all 25 patients. Other causes for unresolved symptoms included: seven bacterial overgrowth; six pancreatic insufficiency; five lymphocytic colitis; four collagenous colitis; four irritable bowel syndrome; two ulcerative jejunitis; one lymphoma; one pancreatic cancer; one fructose intolerance; one T-cell receptor gene rearrangement; one tropical sprue; and one protein-losing enteropathy.
Seven out of the nine patients diagnosed with RS had improvement or remission of symptoms once the cause was identified, treated or eliminated. Two patients were lost to follow-up.
Comment
The investigators conclude that in 10% of cases, the original diagnosis was incorrect. Gluten contamination was the most frequent cause of NRD despite dietary counselling. Gluten contamination should be considered first in patients with suspected RS. They also found that despite the sensitivity for gliadin antibody and high sensitivity and specificity of EMAs, five of the 25 patients with known gluten contamination had no detectable antibodies. This may further complicate the issue of compliance with the GFD.
The RS was a small group and tended to be older than the NCD group. Twelve patients received steroids for the treatment of RS. Only three of these patients, however, had RS and the rest were treated for a variety of diseases in which steroids may have been potentially harmful.
Meloni GF, Dessole S, Vargiu N, Tomasi PA, Musumeci S. The prevalence of coeliac disease in infertility Hum Reprod1999; 14: 2759–2761.
Sardinia is an island where the prevalence of coeliac disease in the general population is high, 10.6 per 1000.10 This region was selected by investigators to correlate silent coeliac disease with infertility. The study group was comprised of 99 women and their partners who were examined for infertility. The following protocol was followed to determine the aetiology of the infertility. The presence of ovulation was detected by serial pelvic ultrasonography and serial serum progesterone measurements during the mid-luteal phase. Endocrine status included determination of pituitary gonadotrophins, prolactin, androstenedione, testosterone and thyroid hormones. Tubal patency and uterine morphology were also evaluated, and diagnostic laparoscopy was performed as indicated. Post-coital testing was performed to screen for cervical factors. Male partners underwent semen analysis and testicular sonography. The infertility was considered unexplained if no obvious cause could be found after evaluation. Venous blood samples were obtained from all study participants. These samples were tested for antigliadin antibody (AGA) assays for IgA and IgG. All samples that tested positive for IgA- and/or IgG-class AGA were tested for IgA-class anti-EMAs. Subjects who tested positive for at least two of the three markers underwent assessment of serum ferritin, folate and vitamin B12 status. Additionally, these latter subjects underwent jejunal biopsy, and women with confirmed coeliac disease were advised to start a gluten-free diet. A control group for the determination of the prevalence of silent coeliac disease was obtained from a previous screening conducted on school children from the same geographical area.10
None of the 99 female subjects had abnormal levels of IgA-class AGA. Four women had IgG-class AGAs and were EMA positive. Three of these women had histological features of coeliac disease on jejunal biopsy. Two of the newly diagnosed coeliac patients belonged to the group of patients with unexplained infertility. The patient who tested positive for IgG, AGA and EMA, but had a normal jejunal biopsy, also belonged to the unexplained infertility group. This patient was considered to have latent coeliac disease. The other woman with coeliac disease was the partner of a man with azoospermia. One male partner was positive for histological features of coeliac disease which is not different from that of the general population.
Comment
In this study, the observed frequency of histologically confirmed silent coeliac disease is 3.03% in the female group. This represents a 3-fold higher prevalence than in the general population of this region. These data support the hypothesis that silent coeliac disease may represent a risk factor for infertility. The underlying mechanisms remain unclear. No study subject displayed overt malabsorption, folic acid or vitamin B12 deficiency which could have an adverse effect on fertility. One subject proved to be iron deficient. None of the newly diagnosed coeliac patients displayed the usual delayed menarche. Approximately half the male subjects in this study had abnormalities of sperm morphology and motility, but only one was diagnosed with coeliac disease. This is similar to the incidence in the general population. Female subjects appear to be more affected by coeliac disease than their male counterparts; why this happens remains unknown.
The investigators used a group of schoolchildren as their control group for prevalence. This group may not represent a similar sample population. Although adult patients may be diagnosed at any age, there appears to be a bimodal peak noted in the 30s and 40s for women.11 The mean age of the women was 33.04 with a range of 26–45 years.
The authors point out that further screening studies are needed in other populations of infertile women to establish whether the prevalence of coeliac disease depends on a genetically determined predisposition.
Kemppainen T, Kroger H, Janatuninen E et al. Osteoporosis in adult patients with celiac disease. Bone1999; 24: 249–255.[Medline]
The investigators of this cross-sectional study looked at the data on the severity of coeliac disease and bone mineral density (BMD) of 77 coeliac patients, and BMD results were compared with control subjects matched for age, gender and menopausal status. The study group was comprised of seven woman who had not reached remission despite adherence to a gluten-free diet for at least a year, and 19 men and 23 women in remission. Each coeliac patient was matched with two age, gender and menopausal status controls from a separate study of volunteers. The anthropometric and BMD data of each coeliac patient were compared with two control subjects. A questionnaire was used to obtain information on drug and calcium supplements, symptoms and menopausal status. Each participant also underwent a physical examination. Coeliac subjects completed four day food records which were checked by a nutritionist. Height, weight, body mass index (BMI) calculations and upper endoscopy were performed, with duodenal biopsies to verify diagnosis and degree of severity. Endoscopies were not carried out on subjects if villous atrophy had been verified within the last 6 months via duodenal biopsies. BMD was measured at the spine, left femoral neck, trochanter and Ward’s triangle.
BMD results were reported as absolute values, and the percentage variation from the mean normal value was calculated using values from control subjects. The subjects were divided into tertiles based on the BMD at the lumbar spine and femoral neck to examine the relationship of calcium intake and serum biochemical values to BMD. Serum calcium, alkaline phosphatase, parathyroid hormone (PTH), 25-hydroxyvitamin D, cross-linked C-terminal telopeptide of human type I collagen (ICTP) and C-terminal extension peptide of type I procollagen (PICP) were measured.
Height and weight for female coeliac and control subjects showed no significant difference. Post-menopausal coeliac subjects showed a lower mean menopausal age than controls. Male coeliac subjects weighed less and were shorter than controls. The BMD of the lumbar spine, femoral neck and trochanter region were significantly lower in female celiac subjects than controls. Male coeliac subjects showed significantly lower BMD of the lumbar spine and femoral neck than controls. BMDs of newly diagnosed and previously diagnosed patients showed no differences at any sites in both sexes.
World Health Organization (WHO) criteria was used to classify patients with osteoporosis. Twenty-six percent of all coeliac subjects were shown to have osteoporosis at the lumbar spine compared with 5% in the control group. The presence of osteoporosis at the femoral neck was rare in both groups. Osteopenia at the lumbar spine in women was related to the severity of the disease. Osteoporosis at the lumbar spine was found more often in patients with previously diagnosed disease not in remission than those newly diagnosed or in remission. The presence of osteoporosis at the femoral neck also followed this same trend. In men, osteoporosis at the lumbar spine was found more often in newly diagnosed coeliac patients than in coeliac patients in remission. None of the men had osteoporosis at the femoral neck. Male patients experienced osteopenia at the lumbar spine in 39% of patients in remission, 63% with newly diagnosed coeliac disease and 14% in controls. Osteopenia at the femoral neck was found in 26% of patients in remission, 67% of newly diagnosed coeliac patients and 5% of the control population. BMDs were lower at the lumbar spine, femoral neck, trochanter and Ward’s triangle for age in the coeliac patients.
Low 25-hydroxyvitamin D concentrations were typically found in coeliac patients; 64% of the men and 71% of women showed abnormally low values, and some were pathologically depressed values. Five out of seven patients with elevated PTH values also had low 25-hydroxyvitamin D values. Ten percent of women and 14% of men showed low serum calcium values. Fourteen female coeliac subjects displayed elevated alkaline phosphatase levels. In male subjects, the serum calcium and phosphorus values were related to the severity of the disease. Elevated PTH levels were only found in newly diagnosed female patients. Men in the lowest tertile of BMD at the lumbar spine had lower serum 25-hydoxyvitamin D and higher serum PTH levels compared with men in the other tertiles. Women displayed no significant differences in biochemical measurements between tertiles of BMD. Regression analysis was used to determine if an association between calcium intake, age, weight, height, menopausal status or other putative factors and BMD of lumbar spine and femoral neck could be shown. In male subjects, a clear association between BMD and age, weight and serum 25-hydroxyvitamin D values at the lumbar spine was shown. Low BMD values were related to age and weight at the femoral neck. In female subjects, post-menopausal status was the main determinant factor of low BMD at the lumbar spine and femoral neck. Weight and depressed serum vitamin D values were associated with BMD at the femoral neck in female subjects. No association was found between calcium intake and BMD in either sex.
Comment
The results of this study show that many coeliac patients are prone to reduced BMD at the lumbar spine and femoral neck. Women with coeliac disease not in remission and newly diagnosed males had osteopenia and osteoporosis more often than controls. The investigators state that a slight bias may have occurred in this study as control subjects with osteoporotic fractures were excluded from the control group. In this study, untreated patients did not have a significantly lower mean BMD when compared with treated patients.
The authors felt that the pathogenesis of bone mineral loss in coeliac patients may not be the same for all patients. A defect in the vitamin D transporter protein present in coeliac patients may account for the low values observed in this study. Four day diet records were checked by a nutritionist, but the level of compliance with the diet was not considered in the data collection.
Despite the fact that calcium and BMD were not related, 31% of patients did show depressed calcium intake, which was attributed to secondary lactose intolerance. Restriction of milk and milk products should be re-evaluated once the recovery of villous atrophy has been shown.
The authors conclude that complete reversal of metabolic bone disease may not be possible in coeliac subjects. Prevention and management of osteoporosis in these patients is needed to restore the intestinal mucosa to normal so that calcium and vitamin D absorption can take place. Untreated or poorly treated coeliac subjects seem to be at increased risk for the development of osteoporosis.
Conclusions
It may be argued that if patients do not present with symptoms of malabsorption or malnutrition why would testing for coeliac disease be necessary. Simply because it can be? The anti-EMA test is safe and simple and has an acceptable sensitivity and specificity. But what situations warrant testing?
The literature reviewed here shows that with minimal expense and compliance with a gluten-free diet, certain adverse consequences of this disease may be eliminated or reduced. Compliance with the diet may enable couples who were unable to conceive to become fertile. Negative outcomes of pregnancy may be minimized/eliminated simply by following a gluten-free diet.
A closer look at diet composition is necessary in cases of refractory coeliac disease as inadvertent gluten contamination is most often the culprit. Once dietary compliance is established, other causes of symptoms should be addressed. Incorrect diagnosis may cause continued symptoms, needless suffering by patients and use of medications that are potentially harmful.
Metabolic bone disease appears to be a more complex issue. BMDs of coeliac subjects were lower at the lumbar spine and femoral neck in both sexes and also lower in females at the trochanter region. However, newly diagnosed and previously diagnosed subjects showed no differences of BMD at any sites. Subjects with previously diagnosed disease or disease of longer duration would be expected to have lower BMDs. This was not true of the reviewed study. The precise relationship between bone disease and coeliac disease remains to be elucidated.
Routine screening is not practical or necessary; however, existent data imply a role for screening distinct populations. The diet requires a high level of motivation and knowledge about gluten-free products and is an ongoing process. Beneficial results are seen rapidly when gluten is removed from the diet of affected individuals.
Some practical points remain. Is the diet palatable and easy to follow? Yes; the diet can be appealing and some excellent products have been developed. There are a wide variety of products and cookbooks on the market; many can be shipped to your home. Some companies will send a free sample pack to patients so they can taste test the products before purchase. Websites are available to obtain reliable information on brand-specific products available for coeliac patients. Not all products will be acceptable to all patients. However, not all patients respond well to all medications.
Is it more costly? Yes; a loaf of tapioca bread costs US$4.39 and coconut macaroons US$7.75 per package. Some of the rice pasta and other products are available in a regular supermarket and may be more reasonably priced. Below are helpful websites for patients getting started on a gluten-free diet.
References
1 American Gastroenterological Association. Medical Position Statement: celiac sprue. Gastroenterology 2001; 120:1522–1525.[Medline]
2 Abdulkarim A, Burgart L, See J, Murray J. Etiology of nonresponsive celiac disease: results of a systematic approach. Am J Gastroenterol 2002; 97:2016–2021.[CrossRef][Medline]
3 Nehra V. New clinical issues in celiac disease. Gastroenterol Clin North Am 1998; 27:453–465.[CrossRef][Medline]
4 Farrell R, Kelly C. Current concepts: celiac sprue. N Engl J Med 2002; 346:180–188.
5 James M, Scott B. Coeliac disease: the cause of the various associated disorders? Eur J Gastroenterol Hepatol 2001; 13:1119–1121.[CrossRef][ISI][Medline]
6 Not T, Tommasini A, Tonini G et al. Undiagnosed celiac disease and risk of auto immune disorders in subjects with type I diabetes mellitus. Diabetologia 2001; 44:151–155.[CrossRef][ISI][Medline]
7 Stazi A, Mantovani A. A risk factor for female fertility and pregnancy: celiac disease. Gynecol Endocrinol 2000; 14:454–463.[Medline]
8 Sher KS, Mayberry JF. Female fertility, obstetrics and gynaecological history in coeliac disease. Digestion 1994; 55:243–246.[Medline]
9 Stewart K, Willoughby JM. Postnatal presentation of celiac disease. Br Med J 1988; 297:1245.[Medline]
10 Meloni GF, Dore A, Fanciulli G et al. Subclinical celiac disease in schoolchildren from northern Sardinia. Lancet 37: 353.
11 Swinson C, Levi J. Is coeliac disease underdiagnosed? Br Med J 1980; 281:1258–1260.[Medline]
12 Ciacci C, Cirillo M, Auriemma G, Di Dato G, Sabbatini F, Mazzacca G. Celiac disease and pregnancy outcome. Am J Gastroenterol 1996; 91:718–722.[ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||