What is the Prague C & M classification?
The Prague C & M classification is a standardized endoscopic reporting system for the extent of columnar-appearing mucosa in the distal esophagus. It was developed by the International Working Group for the Classification of Oesophagitis (IWGCO) and published by Sharma and colleagues to replace inconsistent descriptors such as “short-segment” or “long-segment” Barrett’s esophagus without numeric precision.
Prague notation uses two integers measured in whole centimeters above the gastroesophageal (GE) junction:
- C (circumferential extent): the proximal margin of the circumferential columnar-appearing segment, measured as the distance in cm from the GE junction to the highest point where columnar mucosa encircles the esophageal lumen.
- M (maximal extent): the proximal margin of any columnar-appearing mucosa (circumferential segment or non-circumferential tongue), measured as the distance in cm from the GE junction to the highest point of columnar mucosa anywhere in the esophagus.
Results are reported as C# M# (for example, C2 M5). By definition, M is always greater than or equal to C. When C is less than M, the difference (M minus C) represents the vertical extent of non-circumferential columnar mucosa (“tongues”) above the circumferential segment.
Prague C & M describes endoscopic appearance and extent. It does not, by itself, diagnose Barrett’s esophagus, which requires histologic intestinal metaplasia on biopsy in the appropriate clinical context.
Clinical context: Barrett’s esophagus
Barrett’s esophagus is a metaplastic condition in which the normal squamous lining of the distal esophagus is replaced by columnar epithelium, most often in the setting of chronic gastroesophageal reflux disease (GERD). Endoscopically, Barrett’s mucosa typically appears salmon-pink or velvety compared with pale squamous mucosa, though appearance alone is insufficient for diagnosis.
Contemporary consensus defines endoscopic Barrett’s esophagus as more than 1 cm of columnar-appearing mucosa above the GE junction with intestinal metaplasia confirmed on biopsy. Segments measuring 1 cm or less are often termed ultrashort or “indefinite for Barrett’s” until histology confirms intestinal metaplasia; Prague M values of 1 or less therefore require particular care in labeling and surveillance planning.
Barrett’s esophagus is the major risk factor for esophageal adenocarcinoma. Risk stratification and surveillance intervals depend on segment length, presence and grade of dysplasia, family history, and current society guidelines, not on Prague values alone. Accurate C and M documentation nonetheless improves communication between endoscopists, pathologists, and surgeons, allows comparison across serial examinations, and supports research registries and quality metrics.
Identifying the GE junction
Correct Prague measurements depend on a reproducible GE junction landmark. The GE junction is defined as the most proximal extent of the gastric mucosal folds (the “Z-line” or squamocolumnar junction as seen endoscopically), not the diaphragmatic hiatus.
When a hiatus hernia is present, the diaphragmatic impression can sit several centimeters distal to the true GE junction. Using the hiatus as the reference point systematically overestimates C and M and can misclassify segment length. In hernia patients, identify the Z-line at the tops of gastric folds even if it appears displaced above the diaphragmatic pinch.
Additional landmarks that help orientation include the gastric rugal pattern, the transition from squamous to columnar mucosa, and (when used) narrow-band imaging or acetic acid chromoendoscopy to delineate abnormal mucosa. If the junction is obscured by inflammation, stricture, or post-treatment change, document uncertainty and consider repeat assessment after acid suppression or healing.
How to measure C and M during endoscopy
Measurements are taken with the endoscope withdrawn to straighten the esophagus, using the endoscopic ruler, open biopsy forceps of known width, or calibrated marks on the shaft. Measurements are recorded to the nearest whole centimeter in standard Prague reporting.
Step-by-step technique
- Landmark the GE junction at the proximal limit of gastric mucosal folds, not the diaphragmatic impression.
- Assess circumferential columnar mucosa. Determine the highest level at which columnar-appearing mucosa completely encircles the lumen. Measure the distance from the GE junction to that level; this distance is C. If no circumferential columnar mucosa is present, C = 0.
- Assess maximal columnar mucosa. Identify the most proximal tongue or island of columnar-appearing mucosa anywhere in the esophagus. Measure from the GE junction to that point; this distance is M. If no columnar mucosa is seen above the GE junction, M = 0.
- Verify internal consistency: M must be greater than or equal to C. If a tongue extends above a circumferential segment, M will exceed C.
- Record Prague notation as C# M# (for example, C3 M6) in the procedure report alongside a diagram or photo when feasible.
Patterns of columnar mucosa
Four endoscopic patterns are commonly described from the relationship between C and M:
| Pattern | Prague criteria | Description |
|---|---|---|
| None | C0 M0 | No columnar-appearing mucosa above the GE junction by Prague measurement. |
| Tongue-only (non-circumferential) | C0 M>0 | Columnar mucosa present only as tongues or islands without a circumferential segment (for example, C0 M2). |
| Fully circumferential | C = M > 0 | Columnar mucosa is circumferential to its maximal extent with no tongue above the circumferential margin (for example, C4 M4). |
| Circumferential with tongue(s) | C < M | A circumferential segment extends to C, with non-circumferential columnar mucosa reaching M. Tongue component = M minus C (for example, C2 M5 has a 3 cm tongue above a 2 cm circumferential segment). |
Segment length categories (based on M)
Although Prague notation is numeric, clinicians still map maximal extent (M) to traditional segment-length terms for surveillance discussions and historical comparability:
| Category | Maximal extent (M) | Clinical note |
|---|---|---|
| No measurable segment | M = 0 | No columnar mucosa above the GE junction on endoscopy. |
| Ultrashort | M < 1 cm (after rounding, often M0 or M1 with subcentimeter raw measurements) | Requires histologic confirmation of intestinal metaplasia before labeling as Barrett’s esophagus; inter-observer variability is highest for very short segments. |
| Short-segment | M ≥ 1 cm and < 3 cm | Meets length threshold for endoscopic Barrett’s when intestinal metaplasia is present on biopsy. |
| Long-segment | M ≥ 3 cm | Associated with higher cancer risk in population studies; surveillance and dysplasia management follow guideline-based pathways. |
Prague C and M values should be reported even when segment labels are used, because numeric extent captures tongues and circumferential components more precisely than verbal descriptors alone.
Worked examples
- C0 M0: Normal squamous lining to the GE junction; no columnar mucosa for Prague documentation.
- C0 M2: Two centimeter tongue of columnar mucosa without circumferential involvement; tongue-only pattern.
- C3 M3: Three centimeter fully circumferential columnar segment; no tongue above the circumferential margin.
- C2 M5: Circumferential columnar mucosa to 2 cm with tongues extending to 5 cm; tongue component 3 cm. Often classified as long-segment by M.
Relationship to histologic diagnosis and biopsy strategy
Prague measurements apply to columnar-appearing mucosa, which may include gastric-type columnar metaplasia, inflammation-related appearance, or post-treatment mucosa. Barrett’s esophagus specifically requires intestinal metaplasia (goblet cells) on histology.
When Barrett’s is suspected, current practice emphasizes adequate sampling:
- Obtain at least eight biopsies from the Barrett’s segment (four-quadrant biopsies every 1 to 2 cm is a widely taught protocol) to maximize detection of dysplasia and confirm intestinal metaplasia.
- Target visible abnormalities: perform endoscopic mucosal resection (EMR) or directed biopsies of nodules, ulcers, or irregular mucosa because focal lesions carry higher dysplasia yield than random biopsies alone.
- For segments 1 cm or less, confirm intestinal metaplasia before assigning a Barrett’s diagnosis; Prague alone cannot establish the diagnosis.
After radiofrequency ablation, endoscopic resection, or anti-reflux surgery, Prague remeasurement on surveillance endoscopy documents residual or recurrent columnar mucosa and should be interpreted together with pathology.
Surveillance and management context
Surveillance intervals for nondysplastic Barrett’s esophagus, workup of low- or high-grade dysplasia, and consideration of endoscopic eradication therapy are governed by professional society guidelines and local expertise. Prague C and M support those pathways by providing:
- A baseline extent at index diagnosis for future comparison.
- Clear documentation when segment length changes on follow-up (progression, regression, or tongues developing above a circumferential segment).
- Standardized data for multidisciplinary tumor boards and registry reporting.
Dysplasia grade, visible lesions, patient comorbidities, and shared decision-making dominate treatment choices. A patient with C1 M2 and indefinite dysplasia on random biopsies requires different management than a patient with C6 M8 and high-grade dysplasia in a visible nodule, even though both have measurable columnar mucosa.
Inter-observer variability and quality assurance
Prague classification has good reproducibility for longer segments but lower agreement for segments 5 cm or shorter, particularly at the GE junction in the setting of hiatus hernia, erosive esophagitis, or incomplete distension. Quality improvement strategies include:
- Photographing the GE junction and the proximal margin of columnar mucosa with the measurement reference visible.
- Using a second observer or blinded review for research and registry submissions.
- Deferring definitive extent reporting when mucosa is heavily inflamed until repeat endoscopy on optimized acid suppression.
- Training new endoscopists on hernia cases with explicit instruction to avoid the diaphragmatic impression as the GE junction.
Important limitations
- Appearance is not histology: Prague documents endoscopic extent only; intestinal metaplasia and dysplasia require biopsy.
- Rounding: Standard reporting uses whole centimeters; subcentimeter differences near 1 cm can change ultrashort versus short-segment categorization after rounding.
- Post-treatment esophagus: Scarring, neosquamous epithelium, and islands of buried columnar mucosa may not be fully captured by surface extent alone.
- Not a cancer risk calculator: Cancer risk integrates dysplasia, segment length, demographics, and other factors beyond C and M.
- Does not replace guidelines: Surveillance and intervention decisions follow current gastroenterology society recommendations and pathology results.
Using this CalcMD calculator
Enter the circumferential extent (C) and maximal extent (M) in centimeters as measured above the GE junction. The tool validates that values are non-negative, that M is greater than or equal to C, and that entries do not exceed 30 cm. It rounds to the nearest whole centimeter, outputs standard Prague notation (C# M#), describes the mucosal pattern (none, tongue-only, fully circumferential, or circumferential with tongue), calculates tongue extent when applicable, assigns a segment-length label from M, and displays measurement reminders aligned with published Prague criteria.
Use the output for education, structured procedure documentation, and teaching. It does not replace histology, pathology review, or clinical judgment.