Recommendations

Consensus on digestive endoscopy
Guidelines of the SFED
Indications for gastrointestinal biopsies during upper digestive endoscopy or screening of oeso-gastro-duodenal neoplasia

Head of project: D. Heresbach,
Working group: B. Napoléon, P Burtin, J.-C. Delchier
Reading group: E. Vaillant, D. Lamarque, R. Laugier and the Board of the Société Française d’Endoscopie Digestive (SFED)

Introduction

According to the annual survey ”2 jours d’endoscopie en France (2 days of endoscopy in France)”, 1.1 million of upper digestive endoscopies are performed each year, of which 70 % reveal an abnormality. Nevertheless, the observed abnormalities lead to biopsies in only 15 % of oesophageal lesions against 95 % of gastric lesions and 84 % of duodenal lesions. Carrying out biopsies on the gastrointestinal mucosa during upper digestive endoscopic exploration allows a macroscopic examination and a pathological analysis at the same time. Biopsies are essential when the endoscopic aspect of the mucosal lesion is insufficient to lead to a precise and complete diagnosis, which impacts therapy as well as surveillance rhythm or prognosis.

Objectives

Performing biopsies is a low-risk procedure, but its ”indirect” cost is not negligible. It is thus justified:

• to define this practice with recommendations in terms of indications and quality criteria;

• to specify when and how they must complete or be associated with mucosal chromendoscopy to target these biopsies.

Biopsy techniques

Gastrointestinal mucosal biopsies are carried out with disposable forceps as reuse of gastrointestinal endoscopic biopsy forceps is no longer allowed (AFSSAPS, Official Jounal of 2001, June 6^th). Use of large jaws forceps is preferred to small-sized forceps because of a better quality pathological analysis, although there is no significant improvement in diagnostic yield. The advantage of forceps enabling multiple biopsies in a single passage through the operating channel of the endoscope has not yet been established. Sampling by biopsy forceps should be targeted with precision. It should ideally be carried out perpendicularly to the mucosal surface. When the incidence of the forceps is particularly tangential, as is the case in the oesophagus, the highest biopsy quality can be obtained with a continuous succion while applying the forceps onto the mucosa. Presence of an endoscopic assistant is necessary when performing biopsies as the forceps is triggered by this assistant. The method of collecting samples is not codified, but it is recommended to collect tissue fragments in the jaws of the forceps with the help of a fine metallic forceps (tweezers); use of a needle, source of contamination by accidental jab, should be prohibited. Samples should immediately be put in a recipient containing a fixing agent, generally supplied by the pathological laboratory. Diluted and coloured formalin or alcohol without Bouin’s fluid should be preferred as they denature nucleic acids, even if some pathologists found that Bouin’s fluid enables an easy morphological analysis. The jar should immediately be labelled. In case of sampling on different anatomical sites, each jar is clearly identified by a number. As soon as endoscopy is finished, the operator fills in a request for a precise morbid anatomical examination by indicating at least the sampling site, with reference to jar numbers in case of multiple sites, macroscopic aspects and the objective of the endoscopy.

Contra-indications

Vascular lesions (oesogastric varices, angiodysplasia and angiomas) should not be biopsied, except dyschromic and iodo-negative areas in the inferior third of the oesophagus, if located on a varix before their treatment. In case of active gastrointestinal bleeding, the haemorrhagic lesion should not undergo biopsy at the site of haemorrhage but biopsies may be performed at a distance.

In case of acquired or induced coagulation disorders, arising from antithrombotics intake (anti-vitamin K, platelet anti-aggregants, heparins, etc.), it is not recommended to perform an antithrombotics relay or interruption (except if a recent INR is below 1.5 in case of anti-vitamin K treatment), uniquely for performing biopsies. Loop macrobiopsy or nasoendoscopy is, on the other hand, contra-indicated under platelet anti-aggregant, anti-vitamin K or heparin therapy and necessitates an interruption and relay according to recommendations and underlying thrombotic risk (cf. specific SFED recommendation guidelines).

Indications Oesophageal biopsies Definitive indications for oesophageal biopsies are: Any abnormality of relief or dyschromatic pattern of the oesophageal mucosa. In patients at risk for squamous cell carcinoma of the oesophagus, the search for abnormalities should be facilitated by chromoendoscopy with Lugol® a 2 % iodine-potassium solution,; only iodo-negative areas would thus undergo biopsy.It must be used in patients having:– a personal history of head and neck carcinoma or previous oesophageal squamous carcinoma;– a Plummer-Vinson syndrome (sideropenic hypochromic anaemia and annular stenosis of the upper third of the oesophagus);– a history of ingestion of caustic product, especially after a delay > 15 years whatever accompanying oesophageal stenosis or symptoms;– a more than 15-year past history of achalasia;– a genetic predisposition revealed by specific syndromes ie, palmoplantar keratosis or tylosis in its form with late onset between 5 and 15 years;– an history of alcohol and/or tobacco intoxication.The surveillance rhythm is currently being subject to debate, but may rely upon indications mentioned in Table 1. In case of ectopic oeso-gastric mucosal junction (Z line) it is necessary to perform biopsies to confirm the presence of intestinal metaplasia which defines Barrett’s oesophagus (BE) (prevalence : 1 – 3 % in the general adult population) whereas it frequency reaches up to 5 – 10 % during repeat upper digestive endoscopies for gastro-esophageal reflux disease (GERD).– biopsies are first performed on abnormalities of relief or dyschomic area and their location should be well described (height and quadrant) within Barrett’s oesophagus. Then, systematic sampling according to the Seattle protocol is performed to complete the targeted biopsies: One biopsy per quadrant every 2 cm in height, in separated jars ; the cartography is reproduced on a diagram (example: BE 10 cm high = 20 biopsies); endoscopic pathological analysis of biopsies and length of BE measured by endoscopy lead to present surveillance recommendations. According to lesions observed (intestinal metaplasia, low- or high-grade dysplasia), this surveillance rhythm is detailed in a specific SFED guideline. In case of gastro-oesophageal reflux, systematic biopsies of the mucosal junction (Z line) are performed to screen intestinal metaplasia of the cardia, equivalent to an ”ultra short” BE without endoscopic abnormality. However, this indication that has not been validated yet, since estimation of the risk of cancer of the cardia remains controversial in this case. It is also accepted:a) to perform biopsies:– on all elevated tumoral lesions of the oesophageal mucosa (papilloma, carcinoma);– on all mucosal lesion of uncertain nature, especially in immunodepressed patients;– on all esophageal ulceration, in the squamous cell mucosa as well as in the glandular mucosa; – on all glandular heterotopic foci in the upper oesophagus in case of ulceration or associated oesophageal symptoms; – systematically, especially in the upper third of the oesophagus and in the absence of visible lesion, in case of herpetiformis dermatitis and in case of oesophageal symptoms without obvious lesions or with discrete mucosal modifications, suggestive of oesinophilic oesophagitis (at least five stepped biopsies and presence of 15 oesinophils per field showing important swelling on at least one biopsy).b) biopsies are not required:– on sub-mucosal tumours: superficial biopsy with forceps is not contributive in this case but it would be necessary to perform a real sub-mucosal biopsy or fine needle aspiration to obtain a relevant pathological analysis. In fact, this type of biopsy is achievable only after incision of the mucosa for a deep sampling, which increases the risk of bleeding or even perforation due to its blind nature. It is thus preferable, when a confirmation diagnosis is absolutely necessary, to perform an EUS-guided biopsy; – on vascular lesions; – on typical peptic lesions at the time of the diagnosis ; in the case they should be sampled, biopsies would be performed during a repeat upper digestive endoscopy after proton pump inhibitor therapy. Currently, there is no validated technique enabling a precise targeting of biopsies on areas exclusively at risk for dysplasia: among the techniques daily accessible, chromoendoscopy with methylene blue or 2 % acetic acid enables identifying areas of intestinal metaplasia, and sometimes of high-grade dysplasia, only if it is performed during a zoom endoscopy . Electronic techniques for processing images using narrow spectral bands (FICE or NBI systems) also enable identifying areas of intestinal metaplasia and detecting high-grade dysplasia in a manner identical to chromoendoscopic staining. None of these methods detect low-grade dysplasia. If their diagnostic yield for lesions of high-grade dysplasia or carcinoma is identical to that of stepped biopsies in some expert centres, the absence of inter-observer and large-scale reproducibility has not yet allowed them to replace stepped biopsy protocols. Gastric biopsies Definitive indications for oesophageal biopsies are: Biopsy carried out on a lesion identified by endoscopy any tumoral lesion concerning the gastric mucosa; biopsies of the tumour should be associated with biopsies of the surrounding mucosa so as to ensure that possible adjoining dysplasia areas are not left out; any gastric ulcer; biopsy protocols should include biopsy on the fundus and a minimum of eight biopsies carried out especially on the borders of the ulcer, spread on the whole circumference of the lesion; enlarged folds, so as not to miss a lymphoma or gastric linitis. It is then mandatory to perform either multiple deep biopsies by repeating the sampling on the same location or to perform a macro-biopsy of a fold with a diathermic loop. Biopsies in search of Helicobacter pylori (Hp) They should concern areas of apparently normal gastric mucosa at the level of the antrum and fundus on the small and large flexure. They are mandatory in case of gastric or duodenal ulcers and are recommended (in the absence of contra-indication), although debatable, in case of a NSAID induced lesions in which prolonged treatment with Proton Pump Inhibitor is envisaged. They are recommended in cases of personal history of operated gastric cancer if the search has not been carried out before gastric surgery or in case of first degree or multiple family history of gastric cancer. Performing at least two antral biopsies, two fundic biopsies and one biopsy at the angle of the small flexure enables the search for Hp infection and the correct evaluation of intestinal atrophy and metaplasia at the same time. Search and surveillance of pre-neoplastic lesions of the stomach Two types of situations increasing the risk of gastric cancer or pre-cancerous lesions may be distinguished: those for which an association is certain and where the importance of an endoscopic surveillance is suggested by several trials. It concernsa) personal history of intestinal metaplasia with low- or high-grade dysplasia to:– confirm the diagnosis;– establish a cartography of the lesions through serial biopsies;– search for a potential relationship with an H. pylori infection. Monitoring is motivated by an incidence of gastric cancer of the order of 0.6 – 6 % according to the degree of dysplasia without clear recommendation of a follow-up rhythm based on evidence based data. Expert opinion agree for a follow-up at 3 to 5 years intervals; in cases of ”confirmed” high-grade dysplasia, endoscopic or surgical treatment might be discussed.b) gastric polypsThe rule is to perform biopsies on the polyps to specify their nature that cannot always be determined by their endoscopic shape and mucosal pattern. A resection procedure by polypectomy or mucosectomy is recommended in case of adenoma and for hyperplastic polyps > 5 mm. Fundic gland polyps may be left on site if they are off the context of a familial adenomatous polyposis. When gastric polyps diagnosis is made within the scope of a familial adenomatous polyposis, an endoscopic follow-up of the stomach is recommended after resection at 1 year and then every 5 years; large fundic gland polyps (> 1 cm) should be removed and the small ones monitored through perendoscopic biopsies every 3 years.c) personal history of gastric resection for gastric carcinoma : no recommendation of a rhythm based on scientific data. those for which a search is justified despite the impact of surveillance is still debated:a) chronic atrophic gastritis with fundic intestinal metaplasia : monitoring is debated since the incidence of gastric cancer may reach 0.25 % to 1 %.b) HNPCC syndrome. c) history of gastrectomy for gastric or duodenal ulcer 15 to 20 years after surgery.d) Biermer’s anaemia. e) first-degree family history of gastric cancer. When upper digestive endoscopy is performed in these latter cases, at least two antral biopsies, two fundic biopsies and one biopsy at the ungulus of the small flexure enables the correct evaluation of intestinal atrophy and metaplasia at the same time as well as to detect the presence of H. pylori. Dyspepsia During the course of dyspepsia with normal oeso-gastro-duodenal endoscopy, biopsies usually detect the presence of H. pylori and preneoplastic lesions as mucosal atrophy type or intestinal metaplasia. Performing at least two antral biopsies, two fundic biopsies and one biopsy at the angle of the small flexure enables the search for infection by H. pylori and the correct evaluation of intestinal atrophy and metaplasia at the same time. The advantage of this approach is associated with the fact that eradication of the bacteria has a weak but established role in the control of symptoms and that, in addition, it would stop the progression of possible lesions with low or exceptional neoplastic risk. However, in the absence of clear dyspeptic symptoms there is no indication to recommend systematic upper digestive endoscopy as a screening test to research Hp infection or pre-neoplastic lesion. Gastropathies Gastropathies are characterised by the presence of endoscopically visible mucosal lesions without patholologic inflammation. Except for gastropathies induced by portal hypertension where biopsies are contra-indicated, biopsies are always useful to confirm a diagnosis for gastritis or gastric disease and to eventually detect an infection by H. pylori even in the presence of an apparently normal mucosa. The biopsy protocol always includes two fundic biopsies, two antral biopsies and one biopsy at the antro-fundic transition zone. Culture of H. pylori In case the treatment of H. pylori is a failure, it is recommended to carry out a culture with a antibiogram before establishing a new line of treatment. In such a case, two fundic biopsies and two antral biopsies should be performed for bacteriological tests. Samples should be placed in physiological saline serum and rapidly dispatched to a centre specialised in the culture of the bacterium. Molecular screening methods of mutations associated with resistance against antibiotics would soon enable obtaining rapid results from samples dispatched immediately and without any particular precautions. Etiologic evaluation of a chronic colitis Within the scope of the evaluation of a chronic colitis or inflammatory bowel disease, in particular lymphocytic or collagenous colitis, celiac disease, or Crohn disease, multiple biopsies should also be performed in healthy areas and stepped at the esphageal, gastric and duodenal levels for the search of a focal foveolar gastritis characterised by a lymphocytic and macrophagic infiltrate, particularly in H. pylori-negative patients. This gastritis is more common at the antral level than at the fundic level without any specific macroscopic translation or in the form of tiny erythematous bands. Sub-mucosal tumours Their main advantage is to confirm the sub-mucosal character of the tumour as seen through endoscopy, since the biopsies generally look normal. Thus EUS-guided FNA is preferable when a confirmation of the diagnosis of the nature of a sub-mucosal tumour is absolutely necessary. These biopsies are not necessary in case of typical aberrant pancreas. Anaemia Even in the case of iron deficiency anaemia, fundic biopsies may enable identifying an atrophic gastritis. Macrocytosis may be masked by an association to martial deficiency and induced by malabsorption of iron as a consequence of the hypochlorhydria. Duodenal biopsies Definitive indications for duodenal biopsies are:– anaemia with iron deficiency without known cause, whatever the age of the patient (the yield of biopsies is not decreased in elderly patients, particularly past 65;– folate deficiencies: duodenal biopsies need to be associated with gastric biopsies;– chronic diarrhoea : associated with colonic and ileal biopsies;– suspicion of celiac disease, even if the diagnosis is set by positive circulating antibodies. at initial endosocopy, biopsies should be multiple in the first duodenum and more distal in the third duodenum; villous atrophy is maximal in the distal duodenum whereas associated signs (increase in intra-epithelial lymphocytes, cryptic hyperplasia) are more frequent at the level of the proximal duodenum. The number of biopsies to be performed is at minimum 4 for an optimal diagnosis (two distal and two proximal);– assessment of the response to a gluten-free diet during the course of celiac disease, one year after the implementation of the diet; – resistance to a well conducted gluten-free diet: search for a type I or type II refractory sprue (in this case, request for a CD3 and CD8 immunomarking on the usual biopsies and make samples for frozen biopsies for a molecular or cytometry test of clonality of lymphocytes on biopsies); – suspicion of specific parasites (giardiasis, strongyloidiasis);– villous pattern lesions, before deciding on the type of treatment, particularly endoscopic, or establishing a score for lesions within the scope of a familial adenomatous polyposis with, in case of a normal duodenum or low score for lesions (Spigelman grade I or II), a recommendation for endoscopic duodenal surveillance every 2 years;– suspicion of Crohn’s disease or chronic colitis without proof of its diagnosis, where duodenal biopsies should be performed. The presence of granuloma is not constant and varies from 5 % to 15 % according to the series. However, granulomas are more frequently identified in the duodenum, even in the duodenal bulb, rather than in the stomach. Indications under evaluated are:– systematic duodenal biopsies during upper digestive endoscopy whatever the indication: according to some expert centers, this policy may multiply, the endoscopic prevalence of celiac disease by 4 (1.0 – 3.9 %) and of giardiasis by 5 (0.8 – 0.45 %) by increasing the number of duodenal biopsies by 65 %;– regular hypertrophic papilla of Vater.

References

1 Ribeiro U, Posner MC, Safatle-Ribeiro AV, Reynolds JC. Risk factors for squamous cell carcinoma of the oesophagus. Br J Surg 1996; 83: 1174 – 1185

2 Gehanno P. Cancers associated with cancer of the esophagus. Rev Prat 1999; 49: 1177 – 1179

3 Brucher BLDM, Stein HJ, Bartels H et al. Achalasia and esophageal cancer: incidence, prevalence, and prognosis. World J Surg 2001; 25: 745 – 749

4 Risk JM, Mills HS, Garde J et al. The tylosis oesophageal cancer (TOC) locus: more than just a familial cancer gene. Dis Eosophagus 1999; 12: 173 – 176

5 Dikshit RP, Boffetta P, Bouchardy C et al. Risk factors for the development of second primary tumors among men after laryngeal and hypopharyngeal carcinoma. Cancer 2005; 103: 2326 – 2333

6 Dubuc J, Seyrig JA, Barbier JP et al. Endoscopic screening for esophageal squamous-cell carcinoma in high-risk patients: a prospective study conducted in 62 French endoscopy centers. Endoscopy 2006; 38: 690 – 695

7 ASGE guideline: the role of endoscopy in the surveillance of premalignant conditions of the upper GI tract. Gastrointest Endosc 2006; 63: 570 – 580

8 Allum WH, Griffin SM, Watson A, Colin-Jones D on behalf of the Association of Upper Gastrointestinal Surgeons of Great Britain and Ireland, the British Society of Gastroenterology, and the British Association of Surgical Oncology. Guidelines for the management of oesophageal and gastric cancer. Gut 2002; 50 (Suppl. V): v1 – v23

Table 1 Predisposition and risk factors for oesophageal squamous cell carcinoma (OSCC) which may justify gastroscopic screening and surveillance with iodine-potassium iodide (i/i or Lugol) colouration. OSCC predisposing factors Relative risk/population Incidence or prevalenceof OSCC Time lag for OSCC onset/risk factor and average age of onset Recommendation for Upper digestive endoscopy Head and neck carcinomas × 3 Frequency:– Synchronous OSCC: 2 – 14 %– Metachronous OSCC: 30 % 10 years/diagnostic head and neck carcinomas Every year for 10 years, then at a less frequent rhythm (every 3 or 5 years?) Achalasia × 15 to × 120 Prevalence: 0.4 – 9 % 20 years/diagnostic achalasiaMean age: 48 – 71 15 years/diagnostic of achalasiaRhythm? Intake of caustic product × 100 Incidence: 2.3 – 6.2 % 30 years/intake of caustic productMean age: 35 – 50 15 years/diagnostic of caustic-induced lesionsRhythm? Tylosis – Incidence at 45 years:– 14 % after 5-year monitoring– 95 % cumulated at 65 years of age Mean age: 30 Rhythm?

Corresponding author:
D. Heresbach
Commission recommandation de la SFED
Service des Maladies de l’Appareil Digestif, CHU Pontchaillou,
F-35033 Rennes Cedex 9, France
[e-mail]