In a previous screening study, 16% of patients with psoriasis had IgA and/or IgG antibodies to gliadin (AGA). The aim of the present study was to evaluate the effect of a gluten-free diet (GFD) in 33 AGA-positive and six AGA-negative psoriasis patients. Of the 33 AGA-positive patients, two had IgA antibodies to endomysium (EmA) and 15 an increased number of lymphocytes in the duodenal epithelium, but in some this increase was slight. Two patients had villous atrophy. A 3-month period on a GFD was followed by 3 months on the patient's ordinary diet. The severity of psoriasis was evaluated with the psoriasis area and severity index (PASI). The examining dermatologists were unaware of the EmA and duodenal biopsy results throughout the study. Thirty of the 33 patients with AGA completed the GFD period, after which they showed a highly significant decrease in mean PASI. This included a significant decrease in the 16 AGA-positive patients with normal routine histology in duodenal biopsy specimens. The AGA-negative patients were not improved. After GFD, the AGA values were lower in 82% of those who improved. There was a highly significant decrease in serum eosinophil cationic protein in patients with elevated AGA. When the ordinary diet was resumed, the psoriasis deteriorated in 18 of the 30 patients with AGA who had completed the GFD period. In conclusion, psoriasis patients with raised AGA might improve on a GFD even if they have no EmA or if the increase in duodenal intraepithelial lymphocytes is slight or seemingly absent.

The occurrence of EG2-positive (EG2+) eosinophils and IgE in biopsy specimens of duodenal mucosa and skin from 39 psoriasis patients was studied, with emphasis on the relation to serum eosinophil cationic protein (ECP), serum IgE and the presence or absence of serum IgA and IgG antigliadin antibodies. Psoriasis patients had significantly elevated serum levels of ECP even after exclusion of five of 37 sera which were Phadiatop positive. The elevated serum ECP was not associated with the presence of IgA or IgG antibodies to gliadin. After exclusion of Phadiatop positive sera the serum IgE values did not differ from those of a group of healthy blood donors. Patients with psoriasis had a pronounced increase of EG2+ cells in their duodenal stroma. Patients without antibodies to gliadin tended to have even more EG2+ cells than those with such antibodies and those with increased duodenal intraepithelial lymphocytes. IgE+ cells were present in most duodenal specimens, and in some specimens there were > 100 IgE+ cells/section. The number of EG2+ cells was increased in lesional skin and, in some patients, also in non-involved skin, but there was a more pronounced increase in EG2 reactivity in the duodenal than in the skin specimens. IgE reactivity was increased both in non-involved and involved skin and was significantly related to the number of IgE-positive cells in the duodenal stroma. The results of this study indicate that the gastrointestinal tract and the eosinophil granulocyte might be involved in psoriasis in a hitherto unknown way.

The function of the fixed macrophage system in 18 psoriasis patients was evaluated by measuring the elimination rate of injected autologous erythrocytes coated with iC3b or IgG. The mean half-life of iC3b-coated erythrocytes was significantly prolonged in patients with psoriasis compared with healthy controls (4.7 +/- 0.8 vs 2.7 +/- 0.2 min, P = 0.01). There was also a decrease in the total number of cells eliminated from the circulation (2.5 +/- 0.2 x 10(8) vs 3.3 +/- 0.2 x 10(8), P = 0.01). There was an even more pronounced increase in the half-life of IgG-coated erythrocytes (85 +/- 18 vs 20 +/- 5 min, P < 0.001), with normal values in only 5 of 15 patients, and 4 of these 5 patients were receiving systemic treatment. The slow elimination was interpreted as being caused by primary or secondary defects in receptor function rather than by blocking of the receptors by immune complexes, since patients with psoriasis show normal levels of circulating immune complexes. Further studies are needed to elucidate the nature of these defects.

In a screening study concerning IgA and IgG antibodies to gliadin (IgA AGA and IgG AGA, respectively) in psoriasis, raised levels of IgA and AGA were found to be more common than in a reference group. To determine whether elevated AGA levels were associated with an increased number of intraepithelial lymphocytes, 33 patients with IgA AGA (n = 28) or IgG AGA (n = 5) values above 90% of the reference values (>50 units/ml IgA AGA and <12 units/ml IgG AGA) underwent gastroduodenoscopy and duodenal biopsy in a prospective study. For comparison, six patients with low levels of both IgA AGA and IgG AGA were included. Five biopsy specimens were taken in each patient. Paraffin-embedded specimens were examined with regard to the degree of intraepithelial lymphocyte infiltration, and scored from 0 to 3. Biopsy specimens with a score of 0 had one mononuclear cell or less per four epithelial cells. The specimens were also examined with regard to the presence of intraepithelial CD3(+) T lymphocytes and gamma/delta(+) T lymphocytes.In the six patients with low IgA AGA and low IgG AGA, the biopsy score was 0. Fourteen of the 33 patients with raised AGA had a score of greater than or equal to 1; of these, 12 had raised IgA AGA and two had slightly raised IgG AGA. Two of the patients with raised IgA AGA had partial villous atrophy, but the majority had normal villous architecture. There was a significant correlation both between the biopsy score and the number of intraepithelial CD3(+) cells and between the score and the number of intraepithelial gamma/delta(+) positive T lymphocytes.The serum IgA AGA levels were significantly correlated with the duodenal biopsy score, the number of intraepithelial gamma/delta(+) T lymphocytes, and the number of CD3(+) intraepithelial T lymphocytes. Most patients had no, or only mild, gastrointestinal symptoms. Of the 14 patients with biopsy scores greater than or equal to 1, seven had severe psoriasis and five moderately severe psoriasis, whereas only two had mild psoriasis. There was no relationship between the duodenal score and haemoglobin, folate, whole blood selenium or serum zinc levels.Some of these patients improve on a gluten-free diet, but it is still too early to draw any definite conclusions concerning the type of relationship between the skin lesions, the increased number of intraepithelial lymphocytes in the duodenal mucosa and gluten hypersensitivity.

It was recently observed that in six patients with psoriasis and one with palmoplantar pustulosis, with newly discovered gluten intolerance, a gluten-free diet had a remarkable effect on the skin lesions. This prompted us to undertake a screening investigation to discover whether increased levels of serum antibodies to gliadin are more common in patients with psoriasis than in healthy persons. IgA and IgG antibodies to gliadin (IgA AGA and IgG AGA) were quantified by a micro-ELISA method.Out of 302 patients with psoriasis, 16% (18 females, 31 males) showed serum IgA AGA levels above the 90th percentile value (51 u/ml) of the reference group. This tendency was even more marked when the proportion of patients with values > 70 u/ml was compared with the corresponding proportion of 99 reference subjects. Thus, 3% of the reference subjects but 7.9% of the patients had values > 70 u/ml. The corresponding figures for men were 1.6% and 8.9%, respectively. Men with psoriasis had a significantly higher mean IgA AGA than the male reference group. The means based on logarithmic values of the individual IgA AGA values were significantly higher in the psoriatic groups than in the reference groups. Although the mean level of IgG AGA was not increased in the psoriasis group, there was a correlation between the values for IgA AGA and IgG AGA.The serum concentrations of IgG, IgA and IgM were also measured. In the male patients, the mean IgA value was significantly increased. Women in whom IgA AGA was elevated also showed a significantly increased mean IgA. There was no correlation between the IgA AGA and the IgA values. The IgG levels were significantly elevated in both men and women with psoriasis, whereas the IgM levels did not differ from those in the reference group.Studies on the clinical relevance of the findings in this report, both with regard to the activity of the psoriasis and to the presence of intestinal abnormalities, are in progress.