Reading a celiac biopsy report: Marsh grades and villous atrophy

Short answer

A celiac biopsy report describes what the pathologist saw in the small-bowel lining (villi, crypts and lymphocytes)2, and the European Society for the Study of Coeliac Disease (ESsCD) adult guideline says it should state a Marsh stage22. That guideline reads the stage together with blood tests9, and one pathology review says no single change on the biopsy is specific to celiac disease3.

Published by Gluten Verdict · updated · 8 sources, each linked below

Reviewed by Mark Peter, Registered dietitian (RD), on

Mustard-yellow waiting-room chair with a thin folder and coat on it.

Key points

  • One pathology review lists three elementary lesions, villous atrophy, crypt hyperplasia and more intraepithelial lymphocytes, and says none of them is specific to celiac disease2, 3.
  • ESsCD’s adult guideline says the modified Marsh classification remains the reference standard; the Corazza-Villanacci grades A, B1 and B2 are a simpler alternative5, 6.
  • Marsh 1 is non-specific: ESsCD cites it in 1.3%–6% of small-bowel biopsies, with causes from gluten-related disorders to Helicobacter pylori and drug reactions7, 8.
  • In adults, ESsCD says positive serology with Marsh II or III confirms celiac disease, while persistently positive IgA anti-TG2 serology with Marsh 0–I cannot establish a definitive diagnosis9, 10.
  • Sources word the stages differently, and ESsCD describes substantial variability between pathologists, particularly for mild or borderline changes11.

What the pathologist looks at: villi, crypts and lymphocytes

NIDDK says a pathologist examines the small-intestine tissue under a microscope for signs of celiac disease12. One pathology review describes the histological diagnosis as an integrated assessment of elementary lesions: increased intraepithelial lymphocytes, crypt hyperplasia and villous atrophy2.

Terms on a biopsy report, as the sources define them
Term on the reportWhat it describes
Villous atrophyA decrease in villous height and an altered crypt/villous ratio (normally 3:1), up to the total disappearance of villi13. Normal villi are digitiform, with villus height to crypt depth at 3:1 or more14.
Intraepithelial lymphocytes (IELs), “lymphocytosis”Lymphocytes sitting among the lining cells. 25 T-lymphocytes per 100 enterocytes is considered pathological and is also called “lymphocytosis”15. Most normal subjects have fewer than 25 per 100 epithelial cells16.
Crypt hyperplasiaExtension of the regenerative crypts, with more than one mitosis per crypt17.
Vi:Cr or Vh:Cr ratio (morphometry)ESsCD says measuring villous height against crypt depth and IEL density quantitatively has been proposed as a more objective method18, but that morphometry faces challenges in clinical adoption, including a lack of standardisation19.

The papers write the lymphocyte threshold in different ways: more than 25 per 100 epithelial cells16, 25 or more20, and “most often >25/100 cells”21. The Oslo comparison table gives ≥ 40 per 100 enterocytes on its Type 1 and Type 2 rows, and lists ≥25 per 100 on a row that sits beside Type 0 and Grade A30.

No one of these changes is specific to celiac disease: the review says the diagnosis rests on histological alterations accompanied by consistent clinical and serological data3.

What else a report may list: number of biopsies and where they came from

ESsCD says a pathology report should state the number of biopsies received, the quality of orientation, the Vi:Cr ratio, the IEL count and the Marsh stage22. Orientation matters because the guideline calls well-oriented duodenal biopsies essential23, and it notes that false negative biopsies can result from taking too few24.

On numbers, ESsCD recommends at least four biopsies from the distal duodenum plus two from the duodenal bulb25. The British Society of Gastroenterology, as summarised by Coeliac UK, advises at least 2 from the bulb (D1) and 4 from the second part (D2), in separate pathology pots to help interpretation26. Coeliac UK describes “ultra short coeliac disease” as positive serology with villous atrophy only in D127.

Marsh stages in plain terms

Coeliac UK says a healthcare professional will classify the damage on the Marsh scale1, and ESsCD says the modified Marsh classification remains the reference standard5. The Oslo definitions paper notes that three versions are in use: Marsh, Marsh-Oberhuber and Corazza28.

Marsh stages: Coeliac UK’s wording beside the classification papers
Stage on the reportCoeliac UK’s plain wordingThe classification papers
Marsh 0The lining of the intestine is normal and it is unlikely that the person has coeliac disease29.Type 0, “Normal”, in the Oslo comparison table30.
Marsh 1Increased number of lymphocytes, with normal villi31.Type 1, infiltrative: villi within normal limits and IELs above 25 per 100 epithelial cells32.
Marsh 2Increased lymphocytes; the depressions in the lining are deeper than normal, but villi length is normal33.Type 2, hyperplastic: as type 1, plus hyperplasia of the glandular elements32.
Marsh 3The villi are becoming flattened34.Type 3, destructive: varying degrees of villous atrophy with crypt hyperplasia and increased IELs32.
Marsh 3a, 3b, 3c—Oberhuber’s split of type 3: the pathology review words it mild, moderate and total villous atrophy35; ESsCD words it partial, subtotal and total36.
Marsh 4The villi are completely flattened37.In the Oslo table, a Type 4 row appears only in the Marsh-Oberhuber column, described as an atrophic hypoplastic lesion with flat mucosa, normal crypt height and no inflammation30.

So the same label can be worded differently from one source to the next, as the 3a–3c and Marsh 4 rows show.

The pathology review notes that the original Marsh type 3 groups mild, moderate and severe atrophy in a single category38, which the 3a–3c split separates35. ESsCD says the 3A–3C sub-classifications describe mucosal damage but have limited clinical relevance in routine practice and do not significantly influence treatment decisions or outcomes, though they may help monitor inflammatory activity39. It also notes that Marsh staging compresses a continuum of injury into discrete stages40.

The simplified grading: Corazza-Villanacci A, B1 and B2

The pathology review says Corazza and Villanacci proposed a version meant to simplify and standardize the work of pathologists, with two categories: non-atrophic (grade A) and atrophic (grade B)6.

Corazza-Villanacci grades and the Marsh types they sit beside
GradeWhat it describesMarsh type beside it
ANormal villi with a pathological increase in IELs41.Type 1 in the review’s figure legend42; in the Oslo table, Grade A sits beside Type 0, Type 1 and Type 2 rows30.
B1Villus/crypt ratio less than 3:1, with a pathological increase in IELs43.Types 3A–3B, mild to moderate atrophy42.
B2Villi no longer identifiable, with a pathological increase in IELs44.Type 3C, severe atrophy42.

One 2007 study, whose authors include Corazza and Villanacci, who proposed the simplified system6, sent 60 biopsy slides to six pathologists, who were not given any clinical information, and had each grade them with both systems45. Mean kappa agreement was 0.35 (fair) for Marsh-Oberhuber and 0.55 (moderate) for the simplified system46. ESsCD’s adult guideline, for its part, encourages pathologists to use the modified Marsh classification for uniformity47.

Mild changes and what else causes them

ESsCD calls Marsh-I histology non-specific and cites it in 1.3%–6% of small-bowel biopsies7. Among the aetiologies it reports are gluten-related disorders, wheat allergy, Helicobacter pylori infection and drug-related reactions8. For Marsh I with negative serology, it says celiac disease is unlikely and other causes should be explored48.

The pathology review lists entities with lymphocytosis but no villous atrophy: several medications, Helicobacter pylori gastritis, duodenal parasitic infestations and autoimmune conditions49. The Oslo definitions add that other diseases may cause small-intestinal inflammation without qualifying as celiac disease50.

ESsCD also describes substantial variability between pathologists, particularly with mild or borderline changes11, and encourages a second opinion in uncertain cases51.

When the villi are flat

The review says confounders that typically cause villous atrophy include olmesartan and other angiotensin receptor blockers, various immunomodulatory drugs, common variable immunodeficiency, autoimmune enteropathy, Whipple disease and tropical sprue52. ESsCD defines seronegative villous atrophy as malabsorption with negative serology and villous atrophy on biopsy53, and its differential table groups the other causes under infections, medicines, malignancies, immunodeficiency, autoimmune disorders and more54.

Why the biopsy and blood tests are read together

NIDDK says doctors typically diagnose celiac disease with blood tests and biopsies of the small intestine55, and the ACG guideline abstract says intestinal biopsy is required in most patients to confirm the diagnosis56. The Oslo definitions say celiac antibodies strongly support the diagnosis but by themselves are not confirmatory4.

How ESsCD’s adult guideline reads blood test and biopsy together
Blood test (serology)BiopsyWhat the guideline says
PositiveMarsh II or IIIConfirms the diagnosis of celiac disease9.
Persistently positive IgA anti-TG2Normal architecture (Marsh 0–I)A definitive diagnosis cannot be established; with HLA-DQ2 or DQ8 it may be classified as potential celiac disease10.
NegativeMarsh ICeliac disease is unlikely; other causes should be explored48.
NegativeVillous atrophyAfter other causes are excluded, the guideline says the diagnosis should rely on the clinical and histological response to a gluten-free diet in people with HLA-DQ2 or DQ857.

Some people never have a biopsy report. ESsCD conditionally recommends a no-biopsy approach for adults under 45 whose IgA anti-TG2 is at least 10 times the upper limit of normal, confirmed on a second blood sample58. The ACG abstract suggests a nonbiopsy strategy for selected children59, and Coeliac UK says some adults and children can be diagnosed from blood tests alone60.

On timing, NIDDK says doctors do not recommend starting a gluten-free diet before diagnostic testing, because it can affect test results61, and the Oslo definitions say biopsies must be taken while the patient is on a gluten-containing diet62.

Normal-looking biopsy, positive antibodies: potential celiac disease

The Oslo definitions recommend “potential” celiac disease for people with normal small-intestinal mucosa who are at increased risk because of positive serology63. Coeliac UK describes it as repeatedly high antibody tests and the associated HLA genetics, with normal or minimal changes (Marsh 0 and 1), once other conditions are ruled out64.

ESsCD says potential celiac disease may remain stable, regress or progress to overt celiac disease65. It also lists conditions that can give a false-positive serology, among them hypergammaglobulinemia, autoimmune diseases, chronic liver disease and enteric infections66, and the Oslo paper notes that the group is hard to define because the biopsies taken to exclude active disease vary in adequacy67.

Healing and follow-up biopsies

ESsCD defines mucosal healing by what a follow-up biopsy shows: normal mucosa counts as healing, and persisting Marsh III is called “persistent villous atrophy”68. The pathology review calls the Corazza-Villanacci system helpful for assessing mucosal damage and the response to a gluten-free diet at follow-up69.

ESsCD adds that the 3A–3C sub-stages may help in monitoring, particularly for severe enteropathy such as Marsh 3C, which it says carries a higher risk of incomplete histological recovery and may benefit from closer follow-up70.

Questions people ask

Is Marsh 3 the same as celiac disease?

Not by itself. ESsCD says positive celiac-specific serology with Marsh II or III confirms the diagnosis in adults9, and villous atrophy has other causes, from infections and medicines to immunodeficiency54.

Does the letter after Marsh 3 change what happens next?

ESsCD says the 3A–3C sub-classifications have limited clinical relevance in routine practice and do not significantly influence treatment decisions or outcomes39.

Can two pathologists grade the same biopsy differently?

Yes. ESsCD describes substantial interobserver variability, particularly for mild or borderline changes11, and encourages a second opinion when the picture is uncertain51.

More in Celiac diagnosis, testing and gluten sensitivity

Sources

  • [1, 29, 31, 33, 34, 37, 60, 64] Coeliac UK, Testing for Coeliac Disease: Blood Tests and Biopsy, checked Nov 17, 2026.
  • [2, 3, 6, 13, 14, 15, 16, 17, 32, 35, 38, 41, 42, 43, 44, 49, 52, 69] Pathologica (Italian Society of Pathology), via PubMed Central, Celiac disease: histology, differential diagnosis, complications. A practical approach (Villanacci et al., 2020), checked Nov 17, 2026.

    Read on 2026-11-17: under “Pathological intestinal mucosa” the review says the histological diagnosis of celiac disease consists of an integrated assessment of elementary lesions, and lists increased intraepithelial T-lymphocytes, crypt hyperplasia and villous atrophy.

    Read on 2026-11-17: the review lists Type 1 or infiltrative lesion (villi within normal morphological limits; IEL greater than 25/100 epithelial cells), Type 2 or hyperplastic lesion (villi within normal limits and increased IEL, as in type 1, plus hyperplasia of the glandular elements) and Type 3 or destructive lesion (varying degrees of villous atrophy with hyperplasia of glandular crypts and increased IEL).

    Read on 2026-11-17: the review says Oberhuber et al. divided the Marsh type 3 lesion into three subgroups and lists 3a as mild villous atrophy, 3b as moderate villous atrophy and 3c as total villous atrophy, each with a pathological increase of IEL.

    Read on 2026-11-17: the legend of Figure 1 pairs “Type 1 - Grade A lesion”, “Type 3A-3B - Grade B1” (mild to moderate villous atrophy) and “Type 3C - Grade B2” (severe villous atrophy).

  • [4, 21, 28, 30, 50, 62, 63, 67] Gut (BMJ), via PubMed Central, The Oslo definitions for coeliac disease and related terms (Ludvigsson et al., 2013), checked Nov 17, 2026.

    Read on 2026-11-17: Table 2, “Comparison of histopathological classifications”, has three classification columns (Marsh, Marsh-Oberhuber, Corazza). The row “Normal” reads Type 0, Type 0, Normal. The row for normal architecture and increased intraepithelial lymphocytes ≥25/100 enterocytes reads Type 0, Type 0, Grade A; the row for normal architecture with ≥ 40/100 enterocytes reads Type 1, Type 1, Grade A and, with crypt hyperplasia, Type 2, Type 2, Grade A. The last row, “Atrophic hypoplastic lesion: flat mucosa, normal crypt height, no inflammation with normal intraepithelial lymphocyte counts”, reads No equivalent under Marsh, Type 4 under Marsh-Oberhuber and No equivalent under Corazza.

  • [5, 7, 8, 9, 10, 11, 18, 19, 20, 22, 23, 24, 25, 36, 39, 40, 47, 48, 51, 53, 54, 57, 58, 65, 66, 68, 70] United European Gastroenterology Journal (European Society for the Study of Coeliac Disease), via PubMed Central, ESsCD 2025 Updated Guidelines on the Diagnosis and Management of Coeliac Disease in Adults. Part 1: Diagnostic Approach, checked Nov 17, 2026.

    Read on 2026-11-17: the table headed “The differential diagnosis of non-coeliac villous atrophy” lists infections; medicines and treatments; malignancies and lymphomas; other pathologies (Crohn’s disease, collagenous sprue, amyloidosis); immunodeficiency; autoimmune disorders; graft-related disorders; and idiopathic cases.

  • [12, 55, 61] National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), Diagnosis of Celiac Disease, checked Nov 17, 2026.
  • [26, 27] Coeliac UK, How to test adults (healthcare professionals), checked Nov 17, 2026.
  • [45, 46] Clinical Gastroenterology and Hepatology, via PubMed, Comparison of the interobserver reproducibility with different histologic criteria used in celiac disease (Corazza et al., 2007), abstract, checked Nov 17, 2026.
  • [56, 59] American Journal of Gastroenterology (American College of Gastroenterology), via PubMed, American College of Gastroenterology Guidelines Update: Diagnosis and Management of Celiac Disease (Rubio-Tapia et al., 2023), abstract, checked Nov 17, 2026.