The identification of small-intestinal lesions in untreated celiac disease (CD) rests principally on mucosal biopsy. Traditional method for this procedure has been blind suction biopsy of the duodenum (1) and the duodenum and jejunum (2). The mucosal pathology in CD remains the gold standard for diagnosis as set out by ESPGAN in 1970 (3) and recently revised (4). Although clinical assessment and serological and biochemical markers are indispensable to the diagnostic evaluation of celiac-disease, they are not always pathognomonic. However, villous atrophy and crypt hypertrophy with chronic inflammatory cell infiltration of the upper small-intestinal mucosa are indispensible criteria of CD whose absence rules out this diagnosis. With the advent of upper gastrointestinal endoscopy, the duodenum and occasionally the first part of the jejunum can be directly visualized and multiple biopsies can be obtained. Many endoscopists favor this technique for the diagnosis of CD, especially in adults. There have been few reports of endoscopic small-bowel biopsies in children, despite the importance of diagnosing celiac disease in infancy so that treatment can be rapidly instituted. This report evaluates the relative advantages of the two available techniques—blind suction biopsy and endoscopy biopsy of the mucosa of the duodenum or jejunum. BLIND SUCTION MUCOSAL BIOPSY In medical centers without endoscopic facilities, pediatric biopsy tubes can be used with safety (5), at low cost, and at any age. Once the child is lightly sedated, intubation and progression of the tube can be followed with the use of fluoroscopy or spot radiographic films. Among the disadvantages of this method are the discomfort experienced by older children with inadequate sedation, the length of time required for the tube to be placed correctly (10 min-1.5 h), the necessity of employing some kind of radiation, and sometimes (in 1% of patients) the failure to guide the tube past the pylorus. Its great advantages are that the capsule can be placed with ease in the jejunum and that the specimens taken (one or more) usually include full-thickness mucosa—i.e., muscularis mucosae—with good preservation of the villous epithelium. The specimens can then be carefully orientated on mesh or filter paper before fixation and cut histologically as vertical sections. Preservation of the delicate surface epithelial layer in the villous atrophy of celiac disease is of immense value and also assists in evaluating the numbers and types of intraepithelial lymphocytes. The technique is also useful in enzyme assays for dissacharidase deficiencies in cd. ENDOSCOPIC MUCOSAL BIOPSIES Using pediatric-size gastroscopes with a diameter of 7.9-9.0 mm and 2-mm channels along with biopsy forceps (FB 21K) with an open cup of 1.7 mm, an experienced pediatric endoscopist can easily obtain small-bowel mucosal specimens (6-8). Children aged 1 month and older can be intubated. For older children (1 year and older), the adult gastroscope with its larger biopsy forceps can be used (FB-13K, FB-25K, or FB-3K) (9). Sullivan et al. (8) reported 83% success in obtaining adequate jejunal biopsies with their method of combined endoscopic capsule biopsy. Sedation varied from chloral hydrate 50 mg/kg for the youngest (<12 months) to diazepam 0.1-0.2 mg/kg i.v. and pethidine 1 mg/kg i.v. (9) to ketamine hydrochloride 6 mg/kg i.m. (10) or i.v. (2 mg/kg) with addition of hyoscine (0.01 mg/kg and diazepam 0.1 mg/kg) (8). Barakat et al. (7) gave no sedation to children under 24 months of age and diazepam 0.2-0.4 mg/kg i.v. alone for older children. Thus, with better, though often heavier, sedation of the children for endoscopy and less time taken to reach the duodenum for biopsy (1.5-10 min), both patients and physicians may prefer this method to blind suction biopsy. Inspection of the esophageal, gastric, pyloric, and duodenal or jejunal mucosa during passage of the endoscope is of great importance and is one advantage of the endoscopy method. For example, Oderda et al. (6) found a significant increase in peptic esophagitis and intestinal metaplasia in untreated CD patients compared to those on a glutenfree diet. This group also reported more frequent damage to the duodenal mucosa after endoscopy in untreated celiac patients, whereas Helicobacter pylori-associated duodenitis was less frequent. Failure to reach the second part of the duodenum or beyond and complications were reported by users of both techniques. Most specimens were taken from below the middle of the second part of the duodenum (ampulla of Vater), which is the area most easily reached with the endoscope, or the third part of the duodenum. By endoscopy, macroscopic markers of the duodenal mucosa indicative of untreated CD, such as reduction in the number or loss of Kerkring folds (11), mosaic pattern, scalloped folds or visibility of the underlying blood vessels, have been described in adult CD, but they have not been reported in children. HISTOLOGICAL INTERPRETATION Since histopathological assessment of the intestinal mucosa is of such great importance for the diagnosis of celiac disease, the two biopsy methods have to be critically assessed from the following two aspects: (a) the legitimacy of regarding the duodenal mucosa as representative of and comparable to upper jejunal morphology, and (b) the mucosal preservation and adequacy of the tissue obtained by either technique. The duodenal mucosa distal to the ampulla of Vater has shorter and more leaflike villi than those of the jejunum (12). Villous height ranges from 200-300 μm and <200 μm over Brunner's glands if they extend to the mucosa (the majority of Brunner's glands are found in the submucosa). Morphometric measurements of jejunal villi beyond the ligament of Treitz range 320-570 μm in height and 85-140 μm in width (13). The histological interpretation of duodenal biopsy specimens taken endoscopically can be problematic. We found only two studies in children comparing endoscopic biopsies with blind suction biopsies, both without morphometry. These reports claimed to show “adequate” histology on endoscopic biopsy. In our view, the first study (7) showed one biopsy specimen that was distinctly shorter than the claimed normal villi (their Fig. 2a); another illustration, offered as an example of subtotal villous atrophy, showed an oblique section of the mucosa without muscularis mucosae (their Fig. 3a). In the other study (9), an endoscopic biopsy specimen of the duodenum representing full-thickness mucosa and vertical orientation showed an example of subtotal villous atrophy (their Fig. 18) that could be interpreted as “duodenitis” because of the presence of short, stumpy villi with well-preserved enterocytes. The elongated crypts contained cells of goblet or mucoid type, which are uncharacteristic of intestinal crypt cells. Their Fig. 2 showed unequivocal distortion of the overlying villi by Brunner's glands, which the authors claimed did not interfere with the histological assessment. Overall, not enough information is available on the relative quality of endoscopic and suction biopsies in children. Although Barakat et al. and Granot et al. (7,9) claimed to have obtained equally good results with both methods, Scott and Jenkins (14) found otherwise. They studied 93 patients, nine of whom were endoscoped with the Olympus GIF-P2 and were presumably children. Of the 36 biopsy specimens obtained thus, only nine were suitable for histology, and in four of the nine patients no histological assessment was possible. By comparison, 76% of suction biopsy specimens were suitable for histological interpretation. In other reports, presumably in adults and using an adult endoscope with larger forceps (15-17), endoscopic biopsies produced results that were equal or superior to suction biopsies (96% versus 85%). Because of the controversial nature of the diagnostic value of the two methods, we conducted a random small histological pilot study of 12 children's biopsy specimens, which were viewed blindly by one of us. We found the histological quality good or adequate for diagnosis in six of seven suction biopsy specimens but only one of five endoscopic specimens. The specimens were squashed or fragmented in one of seven suction and in four of five endoscopic biopsies. Brunner glands were noted in one of seven suction and in four of five endoscopic biopsies. The orientation was vertical or partly vertical in six of seven suction and in three of five endoscopic specimens, leaving two of the five endoscopic specimens inadequate for assessment because of oblique sectioning. No difference between the two techniques was noted in regard to the presence or absence of the muscularis mucosae (present in five of seven suction and three of five endoscopic specimens). These findings raise questions regarding the quality of endoscopic biopsies for diagnostic purposes and prompted us to conduct a much larger study, which has recently been concluded.
No takes yet. Share an insight, caveat, or question.
Branski et al. (1996) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: