Refractory coeliac disease: a window between coeliac disease and enteropathy associated T cell lymphoma

The treatment of coeliac disease (CD) is straightforward and simple: life-long adherence to a gluten-free diet. However, in a small subgroup of patients, the clinical and histological abnormalities persist or recur. This non-responsiveness leaves a poorly understood syndrome known as refractory coeliac disease (RCD). A specific definition of RCD is lacking in the literature. We speculate that RCD may appear in a subgroup of coeliacs with persisting histologic abnormalities. In all patients screened for RCD we look for DQ2 and DQ8. In non-DQ2/DQ8 patients we reconsider the diagnosis of CD and of auto-immune enteropathy. Most of the patients referred to us because of suspicion of RCD are affected by other diseases. Probably the commonest cause of non-responsiveness is continued gluten intake. Exocrine pancreas insufficiency, hyperthyroid disease, collagenous colitis are other common explanations. RCD and enteropathy-associated T cell lymphomas (EATL) can be distinguished by intra-epithelial lymphocyte phenotyping and TCR-gamma gene rearrangements. In RCD, an unexplained sustained stimulation of T cell cytotoxic activity is present. Immunosuppressive treatment might moderate this. Cyclosporine has been reported as a resounding success in case reports; however, our results were disappointing. We suggest azathioprine and steroids in RCD without aberrant T-lymphocytes in their mucosa. However, in RCD with aberrant T-lymphocytes we suggest chemotherapy. As the prognosis of EATLs is extremely poor the early detection of RCD with aberrant T cells is crucial.

Gluten challenge in borderline gluten-sensitive enteropathy

OBJECTIVES

In patients with signs and symptoms of malabsorption, suggestive of gluten-sensitive enteropathy, small intestinal biopsies sometimes only reveal infiltration of lymphocytes into the mucosal epithelium. This infiltrative lesion (Marsh I) is not a definite proof for gluten-sensitive enteropathy. However, in the present study, we aimed to show that a subgroup of these patients could ultimately be identified as being gluten sensitive.

METHODS

A total of 38 patients with a Marsh I lesion were subjected to a gluten challenge comprising 30 g of gluten added daily to a normal gluten-containing diet for 8 wk. Before and after the challenge, small intestinal biopsies were taken, and symptoms and signs of malabsorption were scored.

RESULTS

In 12 patients we demonstrated a significant change in mucosal histopathology, i.e., subtotal villous atrophy (Marsh IIIB, n = 1), partial villous atrophy (Marsh 3A, n = 6) or infiltrative-crypthyperplastic lesions (Marsh II, n = 5). In the other 26 patients, the small intestinal mucosa remained unchanged. After initiation of a gluten-free diet, follow-up small intestinal biopsies in 12 patients who initially had progressive mucosal pathology after gluten challenge showed normalization of mucosal pathology in seven cases, regression to a Marsh I lesion in four, and to a Marsh II lesion in one. Symptom relief was seen in all 12 patients. Ten of 26 patients without histological response to the gluten challenge were motivated to adhere to a gluten-free diet. Follow-up biopsies revealed unchanged Marsh I lesions in eight patients and normalization (Marsh 0) in two patients. Three patients had follow-up biopsies while on a normal diet. All had unchanged Marsh I lesions.

CONCLUSIONS

In the present study we demonstrated that a gluten challenge might be useful in identifying patients as being sensitive to gluten if initial small intestinal biopsies reveal only minor abnormalities.

Gluten challenge in borderline gluten-sensitive enteropathy

OBJECTIVES

In patients with signs and symptoms of malabsorption, suggestive of gluten-sensitive enteropathy, small intestinal biopsies sometimes only reveal infiltration of lymphocytes into the mucosal epithelium. This infiltrative lesion (Marsh I) is not a definite proof for gluten-sensitive enteropathy. However, in the present study, we aimed to show that a subgroup of these patients could ultimately be identified as being gluten sensitive.

METHODS

A total of 38 patients with a Marsh I lesion were subjected to a gluten challenge comprising 30 g of gluten added daily to a normal gluten-containing diet for 8 wk. Before and after the challenge, small intestinal biopsies were taken, and symptoms and signs of malabsorption were scored.

RESULTS

In 12 patients we demonstrated a significant change in mucosal histopathology, i.e., subtotal villous atrophy (Marsh IIIB, n = 1), partial villous atrophy (Marsh 3A, n = 6) or infiltrative-crypthyperplastic lesions (Marsh II, n = 5). In the other 26 patients, the small intestinal mucosa remained unchanged. After initiation of a gluten-free diet, follow-up small intestinal biopsies in 12 patients who initially had progressive mucosal pathology after gluten challenge showed normalization of mucosal pathology in seven cases, regression to a Marsh I lesion in four, and to a Marsh II lesion in one. Symptom relief was seen in all 12 patients. Ten of 26 patients without histological response to the gluten challenge were motivated to adhere to a gluten-free diet. Follow-up biopsies revealed unchanged Marsh I lesions in eight patients and normalization (Marsh 0) in two patients. Three patients had follow-up biopsies while on a normal diet. All had unchanged Marsh I lesions.

CONCLUSIONS

In the present study we demonstrated that a gluten challenge might be useful in identifying patients as being sensitive to gluten if initial small intestinal biopsies reveal only minor abnormalities.

Glutathione S-transferases in small intestinal mucosa of patients with coeliac disease

Patients with villous atrophy due to coeliac disease have an increased risk of developing small intestinal malignancies. Intestinal glutathione (GSH) and glutathione S-transferases (GST) are involved in the protection against carcinogenesis. The aim of this study was to evaluate GSH content and GST enzyme activity in small intestinal mucosa of untreated coeliacs compared to controls. We evaluated GSH content and GST enzyme activity, including the levels of GST classes alpha, mu, pi and theta, in small intestinal biopsies of untreated coeliacs (flat mucosa, Marsh IIIC, n = 12) compared to normal subjects (n = 23). Next, we evaluated GSH and GST's in coeliacs in remission (Marsh 0 - I, n = 11), coeliacs with persisting villous atrophy while on a gluten-free diet (partial villous atrophy, Marsh IIIA (n = 5); subtotal villous atrophy, Marsh IIIB (n = 6)) and patients with infiltrative / crypt-hyperplastic Marsh II lesions (n = 4). Total GST enzyme activity and content of GSTalpha are markedly suppressed in Marsh IIIC lesions compared to controls (resp. 220 +/- 79 vs. 464 +/- 189 nmol / mg protein*min (P < 0.001) and 2.79 +/- 2.46 vs. 6.47 +/- 2.29 mg / mg protein (P < 0.001)). In coeliacs in remission these levels normalized. Total GST enzyme activity and GSTalpha levels are proportionately lowered according to the degree of mucosal pathology in Marsh II, IIIA and IIIB. (Spearman's sigma correlation coefficient for total GST, -0.596, P < 0.001; GSTalpha, -0.620, P < 0.001). GSTmu, pi and theta and GSH levels are not significantly different in the selected study groups of mucosal pathology compared to controls. Total GST enzyme activity and content of GSTalpha in small intestinal mucosa are significantly lower in untreated coeliac disease compared to controls. In Marsh II, IIIA and IIIB, GST enzyme activity and GSTalpha content are proportionally lower according to the degree of mucosal pathology. Normal values are seen in coeliacs in remission. This correlation between coeliac disease and a suppressed GSH / GST detoxification system may explain in part the carcinogenic risk in untreated coeliac disease.

Cyclosporin in the treatment of adults with refractory coeliac disease--an open pilot study

AIM

To evaluate the effect of cyclosporin treatment on clinical and histological parameters in adult patients with refractory coeliac disease.

METHODS

Thirteen patients were treated with oral cyclosporin for 2 months, aiming at serum levels of 100-200 ng/mL. Seven extended medication intake up to a maximum of 12 months. Before and after treatment, clinical parameters were monitored and small intestinal biopsies taken. Ten of 13 patients were typed for HLA-DQA1 and -DQB1 alleles.

RESULTS

Eight of 13 patients responded histologically to cyclosporin treatment. Normalization of villi was demonstrated in five patients, three after prolonged treatment. Eight patients reported a clinical response, of whom six had concomitant histological improvement. No serious side-effects of cyclosporin were noticed. Nine of 10 patients who were immunogenetically typed carried the coeliac disease associated serologic DQ2 markers, one carried neither DQ2 nor DQ8 markers.

CONCLUSION

In our study group of 13 adult refractory coeliac disease patients, cyclosporin in therapeutic doses induced a histological improvement in eight patients (61%), in five of whom (38%) normalization of villi was demonstrated. Thus, we believe that cyclosporin is a therapeutic option in refractory coeliac disease, although we could not confirm earlier reports of unconditional successful treatment.

Argon plasma coagulation in flexible gastrointestinal endoscopy: pilot experiences

BACKGROUND AND STUDY AIMS

In argon plasma coagulation (APC), high-frequency energy is transmitted to tissue by ionized gas, thus reducing contact with the tissue to a minimum. Successful endoscopic APC was initially reported in the palliative treatment of gastrointestinal neoplasms. The main objectives in this pilot study were to evaluate the treatment indications, efficacy and safety of the use of APC.

PATIENTS AND METHODS

Between September 1994 and January 1996, APC was used to treat 125 patients with various forms of gastrointestinal pathology.

RESULTS

For local palliative treatment, APC was successfully used alongside snare loop coagulation, dilation, stenting and/or radiotherapy to treat the following conditions: carcinoma of the esophagus: 15 patients, mean number of treatment sessions (MTS) 3.3; gastric carcinoma: 10 patients, MTS 4.9; rectosigmoid carcinoma: seven patients, MTS 2.7; carcinoma of the papilla of Vater: two patients, MTS 1.5. Repeated treatment was also effective for tubulovillous adenoma of the rectum (20 patients, MTS 2.5), stomach (three patients, MTS 2.0), duodenum (two patients, MTS 1.5) and papilla of Vater (two patients, MTS 3.0). In addition, APC proved helpful in coagulating the remaining tissue and achieving hemostasis after polypectomy in the colon (18 patients, MTS 1.2) and in endoscopic treatment of Zenker's diverticulum, for coagulation of the tissue bridge and hemostasis (31 patients, MTS 2.5). Finally, APC was helpful in coagulation of multiple gastric polyps (one patient, one session), hemostasis in superficial ulceration of the duodenal bulb (one patient, one session), after dilation of benign stenoses of anastomoses in the esophagus (one patient, one session) and colon (one patient, one session) and for vascular malformations in the colon (three patients, MTS 1.3), duodenum (one patient, one session), antrum (one patient, two sessions), and watermelon stomach (six patients, MTS 2.8). We recognized signs of perforation in six patients after treatment of Zenker's diverticulum (n = 3), polypectomy in the colon (n = 2) and coagulation of angiodysplasia in the cecum. Laparotomy was carried out in two patients; in one, a perforation was sutured, and in the other no focus of leakage was seen. All six patients recovered without further complications. No complications were observed in any other patients.

CONCLUSIONS

These initial experiences indicate that APC seems to be effective in a number of indications, and relatively safe. Objective evaluation, a longer follow-up period, and comparative trials with other treatment modalities should follow.

Venous versus arterialised venous blood for assessment of blood glucose levels during glucose clamping: comparison in healthy men

The aim of this study was to investigate the influence of the arteriovenous (A-V) gradient in blood glucose concentrations at low and high insulin levels on the determination of glucose requirements during glucose clamping in 9 healthy, insulin sensitive, male volunteers. In a random order two clamps were performed, once using arterialised venous blood (A Clamp, mean pO2 = 11.5 +/- 0.36 kPa, 86 +/- 2.7 mmHg), and once using venous blood (V clamp, mean pO2 = 7.9 +/- 0.21 kPa, 59 +/- 1.6 mmHg). Insulin levels were maintained at 48 +/- 2.4 mU/l from 0-180 min and at 1054 +/- 114 mU/l from 180-360 min. Elevation of insulin levels caused a significant rise of the A-V gradient: from 0.3 +/- 0.1 to 0.5 +/- 0.1 mmol/l (p < 0.05) and from 0.2 +/- 0.1 to 0.3 +/- 0.1 mmol/l (p < 0.05) during the A and V clamps, respectively. Despite these A-V glucose gradients no significant differences were found for the glucose requirements during the last 30 min of each period of insulin infusion between the A and V clamps: 43.70 +/- 3.4 vs 44.8 +/- 2.8 mumol.kg-1.min-1 during the low insulin level and 77.3 +/- 5.0 vs 76.2 +/- 3.4 mumol.kg-1.min-1 during the high insulin level. We conclude that the A-V glucose gradient, even at high insulin levels, does not influence the assessment of glucose requirements to a measurable extent, allowing the use of the simpler technique of taking venous rather than arterialised venous blood for the measurements of glucose levels during glucose clamping.

The absorption kinetics of subcutaneously infused U 40 and U 100 insulin in diabetics

The influence of U 40 and U 100 insulin on the insulin kinetics during continuous subcutaneous insulin infusion (CSII) was investigated in eight insulin-dependent diabetics. CSII was started at t = -1.5 h with a bolus of 2 IU at the individual treatment infusion rate (22-35 IU/24 h), followed at t = 0 h by a premeal dose of 6 IU. Insulin delivery was interrupted for 120 min at t = 4 h and reinitiated with a loading dose of 2 IU for another 2 h. The incremental peak plasma free insulin levels and the time to reach these levels after the premeal insulin dose were not significantly different: 17 +/- 2 vs. 16 +/- 2 mU/l and 144 +/- 22 vs. 129 +/- 9 min for U 40 and U 100 insulin, respectively. During the 2 h interruption of insulin delivery plasma free insulin levels declined by 5.0 +/- 1.6 and 7.2 +/- 1.3 mU/l (NS) for U 40 and U 100 insulin, respectively. The blood glucose profiles were virtually identical throughout the studies for either insulin concentration. These results indicate that varying the insulin concentration in the range between U 40 and U 100 does not affect the insulin kinetics during CSII in insulin-dependent diabetics.