Table 1.
«Normal»/ Preinfiltrative Marsh 0 |
Infiltrative Marsh I |
Infiltrative/ Hyperplastic Marsh II |
«Flat»/ Mosaic Marsh III |
«Flat»/ Unresponsive Marsh IV |
|
---|---|---|---|---|---|
Disorder | |||||
Gluten hypersensitivitya | + | + | + | + | + |
Tropical sprueb | +/- | + | + | + | - |
Chronic diarrhoea/Marasmusc | +/- | + | + | + | - |
Giardiasis/infectionsd | + | + | + | + | - |
GVHDe | - | + | + | + | - |
Food antigensf
|
- - - - |
+ - + - |
+ - + - |
+ - + + |
- - - - |
Transport and enzyme disordersg
|
+ + + |
- - - |
- - - |
- - - |
- - - |
Immunodeficienciesh
|
+ | + | +/- | - | - |
IBDi | + | + | + | - | - |
Drugs (NSAIDs)j | + | + | + | +/- | - |
Neonatal enteropathiesk
|
+ + + |
+/- - - |
+/- - - |
+/- - - |
- - - |
(i) These gluten-induced and other hypersensitivity reactions lead to alterations in villous shape, ultimately being involved in the hyperplastic remodelling of mucosa into mosaic plateaux. Villi do not undergo atrophy, however, and neither is the mucosa subject to any atrophic process, since structural recovery ensues following use of a gluten-free diet:
(ii) The minimal change mucosal lesions (Marsh I and II) cannot be non-specific, because they represent specific host responses to identified inciting antigens:
(iii) The subdivision of Marsh III (a, b, c) has been shown in various independent studies to have no practical value: this misinterpretation results from the failure to recognise the elevated mosaic plateaux which amalgamate villi into these lozenge-shaped blocks of tissue. It follows [see Note (i]) that attempted sub-classification of the severe Marsh III lesion are futile, and would be better to be abandoned:
(iv) Immunodeficiency’s, Inflammatory bowel diseases (IBD), particularly Crohn’s disease and drugs may cause intraepithelial lymphocytosis similar to the Marsh I and II lesions while flat mucosa is only rarely encountered.