- Email: [email protected]
Small-intestinal involvement in familial adenomatous polyposis: evaluation by double-balloon endoscopy and intraoperative enteroscopy
ORIGINAL ARTICLE: Clinical Endoscopy
Small-intestinal involvement in familial adenomatous polyposis: evaluation by double-balloon endoscopy and intraoperative enteroscopy Takayuki Matsumoto, MD, Motohiro Esaki, MD, Ritsuko Yanaru-Fujisawa, MD, Tomohiko Moriyama, MD, Shinichiro Yada, MD, Shotaro Nakamura, MD, Takashi Yao, MD, Mitsuo Iida, MD Fukuoka, Japan
Background: Small-intestinal adenoma occurs in patients with familial adenomatous polyposis (FAP). Objectives: The aim was to analyze the diagnostic yield of a double-balloon endoscopy (DBE) and an intraoperative enteroscopy (IOE) for small-intestinal involvement in FAP. Patients: Forty-one patients with FAP. Interventions: We examined 12 patients with FAP by using oral DBE before a colectomy and 29 patients with FAP by using IOE. The incidence and the endoscopic findings of adenoma were compared between the 2 procedures. Phenotypes of FAP and genotypes of adenomatous polyposis coli (APC) were then compared between patients with small-intestinal adenomas and those without. The genotype was classified into a 5’ mutation (exons 1-14), a 3’ mutation (exon 15), and a negative mutation of APC. Main Outcome Measurement: The prevalence of adenoma. Results: A DBE detected small-intestinal adenomas in 9 of 12 patients (75%), as did an IOE in 15 of 29 patients (52%, P O.05). The adenomas occurred predominantly in the jejunum, with a configuration of diminutive polyps in 22 patients. In addition, a DBE detected nonpolypoid adenoma in a patient, and nodular, broad-based protrusion (advanced lesions) in 3 patients, whereas an IOE detected advanced lesions in a patient. Patients with smallintestinal adenoma had more severe duodenal adenomatosis than those patients without small-intestinal adenoma (P!.001). In cases in which APC was analyzed, the prevalence of small-intestinal adenoma was higher in patients with a 3’ mutation (100%) than in those with a 5’ mutation (44%) and with a negative mutation (42%, P!.02). Limitation: Not a prospective randomized study. Conclusions: A DBE is equal to an IOE for scrutiny of small-intestinal adenomas in FAP. There seems to be a genotype-jejunal phenotype correlation in FAP. (Gastrointest Endosc 2008;68:911-9.)
See CME section; p. 961. Copyright ª 2008 by the American Society for Gastrointestinal Endoscopy 0016-5107/$34.00 doi:10.1016/j.gie.2008.02.067
and an overtube.2-4 Because the latter procedure is characterized by 2 balloons attached at the tips of the enteroscope and the overtube, it is referred to as double-balloon endoscopy (DBE). Recent clinical trials showed that both VCE and DBE are best indicated for patients with hereditary GI polyposis.5-11 Our previous investigation showed that DBE was superior to VCE for the detection of diminutive polyps in patients with familial adenomatous polyposis (FAP) and Peutz-Jeghers syndrome.9 Wong et al10 recently reported that, in their patients with FAP, VCE identified fewer numbers of jejunal polyps than did push enteroscopy. These observations suggest that VCE may not be the choice of procedure for the scrutiny of small-intestinal polyposis.
www.giejournal.org
Volume 68, No. 5 : 2008 GASTROINTESTINAL ENDOSCOPY 911
It has become possible to observe an extensive area of the small intestine by endoscopy. Video capsule endoscopy (VCE) is a noninvasive procedure characterized by the application of a wireless capsule.1 Another procedure is an invasive method with a conventional enteroscope Abbreviations: APC, adenomatous polyposis coli; CHRPE, congenital hypertrophy of the retinal pigment epithelium; DBE, double-balloon endoscopy; FAP, familial adenomatous polyposis; IOE, intraoperative enteroscopy; PCR, polymerase chain reaction; PTT, protein truncation test; VCE, video capsule endoscopy.
Enteroscopy for familial adenomatous polyposis
More recently, Mo ¨ nkemu ¨ller et al11 detected jejunal adenomas by DBE in patients with FAP and found that jejunal neoplasm segregated in patients with mutations at exon 15 of the adenomatous polyposis coli gene (APC). However, in 1980s and 1990s, we and other investigators reported that intraoperative enteroscopy (IOE)12,13 and push enteroscopy detected a similar or even greater prevalence of jejunal adenomas in FAP.14,15 In the present retrospective study, we analyzed findings obtained by an IOE and a DBE in our patients with FAP to determine the practical value of a DBE in the disease and to correlate small-intestinal phenotypes with APC mutation.
Matsumoto et al
Capsule Summary What is already known on this topic d
Small-intestinal adenomas may occur in patients with familial adenomatous polyposis (FAP), but their prevalence varies, depending on the modality used for their detection.
What this study adds to our knowledge d
In a study of 41 patients with FAP, the prevalence of small-intestinal adenoma was 75% in patients examined by oral double-balloon endoscopy and 52% in those examined by intraoperative enteroscopy.
PATIENTS AND METHODS Patients During a period from 1990 to 2006, we made an established diagnosis of FAP in 122 patients from 90 family histories. The diagnostic criterion of FAP was a confirmation of more than a hundred colorectal adenomas. Among those patients, we performed an IOE in 29 patients at the time of subtotal colectomy with ileorectal anastomosis or a total proctocolectomy with ileoanal anastomosis during a period from 1990 until 2003. During the subsequent 3-year period from 2004 to 2006, we applied a DBE for 12 patients before surgery. These 41 patients were the subjects of the present investigation. Informed consent for an enteroscopy was obtained from each subject. Furthermore, the genetic analysis of APC was approved by the institutional review board at Kyushu University Hospital.
in polyp number (1-4 polyps, 1 point; 5-20 polyps, 2; O20 polyps, 3), polyp size (1-4 mm, 1; 5-10 mm, 2; O10 mm, 3), histologic type (tubular, 1; tubulovillous, 2; villous, 3), and dysplasia (mild, 1; moderate, 2; severe, 3). A score of 0 points was regarded as stage 0, 1-4 as stage I, 5-6 as stage II, 7-8 as stage III, and 9-12 as stage IV. The ampullary lesion was determined to be positive or negative according to the histologic findings. An epidermoid cyst was assessed by physical examination. Osteoma was examined by plain orthopantomography. Congenital hypertrophy of the retinal pigment epithelium (CHRPE) was screened by direct ophthalmoscopy and was regarded to be positive when hyperpigmented areas with well-defined margins were found. Desmoids were examined by physical examination and abdominal CT.
Assessment of clinical features Enteroscopy
The clinical record of each subject at the time of an enteroscopy was reviewed, and manifestations specific to FAP were retrospectively investigated. As we previously reported,16 the clinical manifestations were evaluated in all the patients at the time of the initial diagnosis. Colorectal polyposis was graded to be sparse (!200), intermediate (between 200 and 1000), or profuse (O1000) by the number of polyps under radiography or macroscopy of the colon. In 35 patients who were treated by surgery, the number of macroscopic polyps was directly counted in the resected specimen. In the remaining 6 patients who had not undergone surgery, colorectal polyps were counted in double-contrast barium enema examination. Upper-GI lesions were examined by chromoscopy with a side-viewing instrument and a forceps biopsy. We used the side-viewing instrument to precisely observe both the periampullary and gastric lesions.17-20 Fundic gland polyps and adenoma of the stomach, and duodenal adenoma were determined to be positive when the histology of the biopsy specimen was compatible with each lesion. The severity of duodenal adenomatosis was determined according to the staging system reported by Spigelman et al.21 The severity was scored by the sum of grades
Oral DBE was performed by an endoscopist (T.M.) with the Double-Balloon Enteroscopy System (Fujinon-Toshiba ES System Co, Tokyo, Japan). After an overnight fast, the patients were prepared by continuous intravenous infusion and were examined by an enteroscopy under a light sedation with IV midazolam coupled with 17.5 mg pethidine chloride. The procedure was carried out under fluoroscopy, as has been previously reported.22 In brief, an endoscopist controlled the enteroscope, and an assistant advanced the overtube. During the insertion of the enteroscope, the balloon of the overtube was inflated and attached to the small-intestinal wall. Alternatively, the balloon of the enteroscope was inflated and attached to the intestinal wall during the insertion of the overtube. The small intestine within the reach of the enteroscope was intermittently shortened by pulling back the overtube and the enteroscope with both balloons inflated and attached to the intestinal wall. When the enteroscope was unable to advance further, a plain abdominal radiograph was obtained with the patient in the supine position, and then the small intestine within the reach of the enteroscope was carefully explored. The depth of
912 GASTROINTESTINAL ENDOSCOPY Volume 68, No. 5 : 2008
www.giejournal.org
Matsumoto et al
exploration during an oral DBE was calculated according to the method reported by May et al.23 An IOE was performed by 1 of 3 endoscopists (M.I., R.F., T.M.) during the surgical procedures, as previously described.12 In brief, an enterotomy was made in the middle of the small intestine, through which a sterilized video endoscopy (SIF-2000; Olympus) was inserted. The enteroscope was initially inserted to the ligament of Treitz with a manual maneuver by the surgeons, and it was slowly pulled back to the enterotomy. The enteroscope was then manually inserted to the ileocecal valve and again pulled back to the enterotomy. In both a DBE and an IOE, a chromoscopy by spraying a 0.5% indigo-carmine solution and a forceps biopsy were performed when any endoscopic sign of adenoma was suggested.
Enteroscopy for familial adenomatous polyposis
TABLE 1. Phenotypic features of the subjects
Clinical variables Age (median) (y)
DBE group IOE group (n Z 12) (n Z 29) 42 (range 20-66)
P
28 (range 12-61)
.04
Sex (women/men)
5/7
14/15
NS
Family history of FAP (þ/–)
7/5
12/17
NS
Severity (profuse/ intermediate/scattered)
2/4/6
10/14/5
.04
Invasive cancer (þ/–)
2/10
9/20
NS
Fundic gland polyposis (þ/–)
6/6
15/14
NS
Gastric adenoma (þ/–)
5/7
15/14
NS
Gastric cancer (þ/–)
0/12
0/29
NS
Score 0 (stage 0)
3
3
NS
Score 1w4 (stage I)
1
5
Score 5,6 (stage II)
3
12
Score 7,8 (stage III)
2
7
Score 9-12 (stage IV)
3
2
Ampullary adenoma (þ/–)
5/6
16/13
NS
CHRPE (þ/–)y
5/6
16/8
NS
Osteoma (þ/–)
7/5
20/9
NS
Epidermoid cyst (þ/–)
4/8
6/23
NS
Desmoid (þ/–)
1/11
3/26
NS
Colorectal lesions
Gastric lesions
Histology Biopsy specimens obtained by a gastroduodenoscopy, DBE, and IOE were fixed in 5% formalin, embedded in paraffin, and cut in slices approximately 5-mm thick. The slices were stained with hematoxylin and eosin, and observed by a pathologist (T.Y.) who was informed that the specimens were from subjects with FAP but was not informed of other information. In the specimens positive for epithelial neoplasia, the histologic phenotype and the category were graded according to the Vienna classification.24,25 The latter was divided into noninvasive low-grade neoplasia, noninvasive high-grade neoplasia, and invasive neoplasia.
APC analysis In 30 patients from whom informed consent for genetic analysis was obtained, APC was screened by the protein truncation test (PTT), as was previously described.16 In brief, genomic DNA and messenger RNA were obtained from peripheral leukocytes, and the latter was reversetranscribed to complementary DNA. Complementary DNA was amplified with 2 primer pairs by polymerase chain reaction (PCR) for proximal segments of APC (segment A, codon 2-479; segment B, 348-758). DNA was amplified with 4 primer pairs for distal segments of APC (segment C, codon 658-1283; segment D, 1099-1700; segment E, 1547-2246; segment F, 2123-2843). The PCR products were applied to rabbit reticulocyte lysate and translated into protein incorporation with 38S-methionine. The synthesized protein was electrophoresed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and full-length or truncated protein was determined. Based on the results of PTT, the APC mutation was classified into proximal mutation (segment A-B, exons 1-14), distal mutation (segment B-F, exon 15), and negative mutation.
Duodenal adenomatosis*
Extraintestinal lesions
NS, Not significant; þ, present; –, absent. *Assessed according to the classification reported by Spigelman et al (Ref. 21). ySix patients (1 patient in the DBE group and 5 patients in the IOE group) were not examined.
ability test or the c2 test when appropriate. Probabilities !.05 were considered to be statistically significant.
RESULTS Demographic data
Parametric data were expressed as median (range) and were compared between 2 groups by using the MannWhitney test. Nonparametric data were expressed as percentage and were compared by the Fisher exact prob-
Demographic data between the DBE group and the IOE group are compared in Table 1. Although the sex and the rate of proband for FAP were not different between the 2 groups, the age at the time of an enteroscopy was significantly older in the DBE group than in the IOE group (42 years [range 20-66 years] vs 28 years [12-61 years], P ! .05). In contrast, the scattered type of
www.giejournal.org
Volume 68, No. 5 : 2008 GASTROINTESTINAL ENDOSCOPY 913
Statistical analysis
Enteroscopy for familial adenomatous polyposis
colorectal adenomatosis was more frequent in the DBE group than in the IOE group (50% vs 17%, P ! .05). Although the incidence of invasive colorectal cancer at the time of diagnosis was higher in the IOE group (45%) than in the DBE group (20%), the difference did not reach a statistical significance. The prevalence of extracolonic lesions is also compared in Table 1. Fundic gland polyposis was found in 21 of the overall 41 patients (51%), and gastric adenoma was found in 20 patients (49%). The prevalence of those gastric lesions was not different between the 2 groups. The prevalence of duodenal (75% vs 90%) and ampullary adenoma (50% vs 55%) was not different between the DBE group and the IOE group. In addition, the severity of duodenal adenomatosis was not different between the DBE and IOE groups. Similarly, the prevalence of CHRPE (45% vs 67%) and osteoma (58% vs 69%) was not different between the 2 groups.
Enteroscopic and histologic diagnosis
Matsumoto et al
TABLE 2. Comparison of enteroscopic diagnosis between the DBE group and the IOE group DBE group, IOE group, no. (%) no. (%) (n Z 12) (n Z 29) P Prevalence of adenoma
9 (75)
15 (52)
NS
Jejunum
9 (75)
14 (48)
NS
Ileum
2 (17)
3 (10)
NS
!5
9 (75)
23 (79)
NS
R5
3 (25)
6 (21)
Diminutive polyp
7 (58)
15 (52)
NS
Nonpolypoid lesion
1 (8)
0 (0)
NS
Advanced lesion
3 (25)
1 (3)
.07
Site of adenoma
No. adenoma
Endoscopic configuration
In the DBE group, the time for examination ranged from 58 to 108 minutes, with a medium of 78 minutes, and the calculated length of insertion from the ligament of Treitz ranged from 150 to 450 cm, with a medium of 320 cm. Comparisons of enteroscopic and histologic diagnoses between the DBE and IOE groups are shown in Table 2. The overall prevalence of small-intestinal adenoma was 75% (9 of 12 patients) in the DBE group and 52% (15 of 29 patients) in the IOE group. The prevalence determined by the 2 procedures was not statistically different (P O .05). In both DBE and IOE groups, the prevalence of adenoma was higher in the jejunum than in the ileum. The number of identified adenoma was 5 or more in 25% of patients in the DBE group and in 21% of the IOE group. In both DBE and IOE groups, sessile, diminutive, and whitish polyps were the most frequent enteroscopic features of the small-intestinal adenoma (Fig. 1A). In 3 patients in the DBE group (25%) and in a patient in the IOE group (3%), nodular, broad-based protrusions were identified (Fig. 1C). Furthermore, an adenoma in a patient in the DBE group was recognized as a sharply demarcated depression (Fig. 1E). Histologic examination showed that most of the lesions were tubular adenomas (Fig. 1B, D, and F), although there was a tubulovillous adenoma in a patient in the IOE group. All the lesions were diagnosed as noninvasive low-grade neoplasia, without any sign of high-grade or invasive neoplasia. Among patients in the DBE group, 4 patients were subsequently examined by anal DBE, 4 patients were examined by small-bowel radiography plus retrograde ileoscopy, and 3 patients were examined by VCE. These procedures did not identify any small-intestinal polyp in the lower part of the ileum. The remaining patient was not examined further.
Twenty-four patients were determined to have smallintestinal adenoma by a DBE or an IOE. A comparison of phenotypes between patients with small-intestinal adenoma and those without is shown in Table 3. As shown in the table, there were trends toward higher incidence in small-intestinal adenoma in men, in patients with profuse or intermediate polyposis, and in patients with fundic gland polyposis. However, the correlation between those phenotypes and small-intestinal adenoma was insignificant. Although not statistically significant, there was also a trend toward a higher incidence of small-intestinal adenoma in patients with duodenal adenoma than those without (96% vs 71%, P Z .07). Furthermore, the score for duodenal adenomatosis was significantly higher in the former (score 0-11, 6.5) than in the latter (score 0-6, 4), P !.001). As shown in Figure 2, the stage of duodenal adenomatosis was more advanced in patients with smallintestinal adenoma than those without (P ! .001). The prevalence of CHRPE, epidermoid cysts, and desmoid was not different between patients with small-intestinal adenoma and those without. However,
914 GASTROINTESTINAL ENDOSCOPY Volume 68, No. 5 : 2008
www.giejournal.org
Histologic type
NS
Tubular only
9 (75)
14 (48)
Tubular and tubulovillous
0 (0)
1 (3)
Category
NS
Noninvasive low grade
9 (75)
15 (52)
Noninvasive high grade
0 (0)
0 (0)
Invasive
0 (0)
0 (0)
NS, Not significant.
Small-intestinal phenotype and other phenotypes
Matsumoto et al
Enteroscopy for familial adenomatous polyposis
Figure 1. Endoscopic and histologic findings of adenomas found in patients with FAP. A, A diminutive adenoma found in the jejunum of a 48-year-old man; a DBE shows a small, whitish polyp. B, On histologic examination, the lesion (A) was composed of epithelial cells with enlarged nuclei with obvious nucleoli; the findings are compatible with noninvasive low-grade tubular adenoma. (H&E, orig. mag. 200). C, An advanced adenoma in the jejunum of a 48-year-old woman; a DBE with chromoscopy shows a large, broad-based protrusion with a nodular surface. D, A biopsy specimen (C) contained tubules of noninvasive low-grade adenoma (H&E, orig. mag. 200). E, A nonpolypoid adenoma in the jejunum of a 38-year-old man; a DBE with chromoscopy shows a small, distinct, depressed lesion in the jejunum. A biopsy specimen from the lesion (F) contains crowding tubules of noninvasive low-grade adenoma (H&E, orig. mag. 200).
www.giejournal.org
Volume 68, No. 5 : 2008 GASTROINTESTINAL ENDOSCOPY 915
Enteroscopy for familial adenomatous polyposis
Matsumoto et al
TABLE 3. Comparison of phenotype of FAP between patients with small-intestinal adenoma and those without small-intestinal adenoma
Clinical variables
With small-intestinal adenoma (n Z 24)
Without small-intestinal adenoma (n Z 17)
P
Age (median) (y)
28 (range 13-57)
34 (range 12-66)
NS
Sex (women/men)
8/16
11/6
.06
Family history of FAP (þ/–)
14/10
5/12
NS
Severity (profuse to intermediate/scattered)
20/4
10/7
.08
Invasive cancer (þ/–)
6/18
6/11
NS
Fundic gland polyposis (þ/–)
15/9
5/12
.06
Gastric adenoma (þ/–)
14/10
6/11
NS
Gastric cancer (þ/–)
0/24
0/17
NS
Duodenal adenomatosis (stage 0/I-IV)*
23/1
12/5
.07
Ampullary adenoma (þ/–)
15/9
8/7
NS
CHRPE (þ/–)y
15/8
6/6
NS
Osteoma (þ/–)
20/4
7/10
.008
Epidermoid cyst (þ/–)
8/16
2/15
NS
Desmoid (þ/–)
1/23
3/14
NS
Colorectal lesions
Gastric lesions
Duodenal lesions
Extraintestinal lesions
NS, Not significant. *Assessed according to the classification reported by Spigelman et al (Ref. 21). ySix patients (1 patient in the DBE group and 5 patients in the IOE group) were not examined.
the prevalence of osteoma was significantly higher in the former patients than in the latter (83% vs 41%, P Z .008).
Genotype and small-intestinal phenotype PTT determined 9 patients to have a 5’ mutation (exons 1-14) and 9 patients to have a 3’ mutation (exon 15) of APC. However, PTT did not identify an APC mutation in the remaining 12 patients. A correlation between APC genotype and small-intestinal phenotype is shown in Table 4. As shown in the table, the prevalence of small-intestinal adenoma was the highest in a 3’ mutation with a value of 100% followed by a 5’ mutation (44%) and a negative mutation (42%). The prevalence of small-intestinal adenoma was significantly different among the 3 groups (P ! .02).
with advanced adenomas found by a DBE was treated by a local resection during the subsequent proctocolectomy. We planned to treat the advanced adenomas in the other 2 patients by endoscopic mucosal resection during the forthcoming proctocolectomy. We decided to follow-up the remaining 20 patients with diminutive small-intestinal adenomas. Five of the patients were examined by a repeated DBE during a period ranging from 2 to 4 years after the initial enteroscopy. The follow-up examination revealed that the size and the number of small-intestinal adenomas remained unchanged.
DISCUSSION
An advanced adenoma found by an IEO was removed by local resection during the surgery. Similarly, a patient
The results of our retrospective study indicated that the prevalence of small-intestinal adenoma was slightly but insignificantly higher in patients with FAP examined by a DBE than in those examined by an IOE, and that the
916 GASTROINTESTINAL ENDOSCOPY Volume 68, No. 5 : 2008
www.giejournal.org
Follow-up data
Matsumoto et al
Enteroscopy for familial adenomatous polyposis
TABLE 4. Comparison of APC genotype between patients with small-intestinal adenoma and those without small-intestinal adenoma* With smallintestinal adenoma, no. (%)y
Without small-intestinal adenoma, no. (%)y
Total
50 mutation (exons 1-14)
4 (44)
5 (56)
9
30 mutation (exon 15)
9 (0)
0 (0)
9
Negative mutation
5 (42)
7 (58)
12
18
12
30
APC genotype
Total
*The prevalence of small-intestinal adenoma is significantly different between the groups (P ! .02). yData in parentheses refer to percentage in each genotype.
small-intestinal phenotype of FAP was closely associated with osteoma, the severity of duodenal adenoma, and the germline mutation in exon 15 of APC. Even though we could not compare the diagnostic value of a DBE and an IOE in identical individuals, the higher prevalence of the smallintestinal adenoma in patients examined by a DBE may be a consequence of scrutiny achieved by the procedure. With the development of novel enteroscopy systems, small-intestinal involvement in patients with FAP was rigorously examined in various clinical trials. The prevalence of small-intestinal polyps reported in those investigations, however, ranges widely according to the procedures applied or to the number of subjects examined.6-8 In particular, Caspari et al8 found diminutive small-intestinal polyps in all of their 4 patients with FAP, whereas Burke et al,6 who examined the largest number of subjects among the VCE studies, detected small-intestinal polyps in 9 of 15 patients. In another investigation, by Schulmann et al,5 a combination of push enteroscopy and VCE detected small-intestinal adenoma in 17 of 29 patients with FAP (59%). Of note is that, in our enteroscopic survey with enrolling the largest number of patients with FAP to date, 59% of the patients were proven to have small-intestinal adenoma. It thus seems probable that the actual prevalence of small-intestinal adenoma in FAP is approximately 60% in both Western countries and Japan.
Schulmann et al5 reported that there was a considerable discordance in the number of small-intestinal polyps between push enteroscopy and VCE in patients with FAP. More recently, Wong et al10 analyzed VCE findings in 32 patients with FAP and with small-intestinal adenomas proven by push enteroscopy and found that the latter procedure identified a greater number of adenomas. However, Mo ¨ nkemu ¨ller et al11 performed a DBE for 9 patients with colectomized FAP and found small-intestinal adenoma in 6 patients. Because the performance of an oral DBE was shown to be superior to that of push enteroscopy15 and because adenomas occur predominantly in the jejunum of FAP,12-15 it seems to be reasonable to presume that the value of a DBE for FAP is equivalent to that of a combination of push enteroscopy and VCE. Our results further suggested oral DBE to be alternative to an IOE for the assessment of small-intestinal involvement in FAP.12,13 As has been reported in duodenal adenomatosis,17,20,21 the predominant endoscopic findings in small-intestinal adenomas in our patients were diminutive, whitish polyps or larger, broad-based lesions. Although there have been sparse descriptions of advanced jejunal adenomas in ¨nkemu FAP,15,26 Mo ¨ller et al11 recently detected such lesions by DBE in 2 of their 9 patients with FAP. Furthermore, we could identify in a patient, by a DBE, an adenoma characterized by a sharply demarcated depression. Similar nonpolypoid adenomas were demonstrated in the colon and in the duodenum of patients with FAP.27,28 In addition, analyses of the proliferative activity of those nonpolypoid lesions indicated that they are presumably precursors of protruding adenoma.28,29 The nonpolypoid adenoma as found in one of our patients with FAP thus suggests the heterogeneity in configuration of early adenoma even in the small intestine. Since the identification of APC, genotype-phenotype correlations have been shown in some clinical
www.giejournal.org
Volume 68, No. 5 : 2008 GASTROINTESTINAL ENDOSCOPY 917
Figure 2. A comparison of the stage of duodenal adenomatosis between patients with small-intestinal adenoma and those without the adenoma.
Enteroscopy for familial adenomatous polyposis
Matsumoto et al
manifestations of FAP. In our previous study, the prevalence of duodenal adenoma was 100% in patients with mutations in exons 10-15 of APC, whereas the value was less than 50% in other patients. A similar trend was found in another Japanese survey by Enomoto et al30 In a Swedish survey of 180 patients, Bjo ¨ rk et al31 found that severe duodenal adenomatosis and duodenal cancer in FAP were closely associated with APC mutations distal to codon 1051 of exon 15. Furthermore, cases of severe duodenal adenomatosis with APC mutations at the distal site of exon 15 (codons 1552 and 1556) were reported.32,33 These observations suggested a genotype-duodenal phenotype correlation in patients with FAP. We could find a positive correlation between the stage of duodenal adenomatosis and the presence of smallintestinal adenoma in the present investigation. There was also a close association of jejunal adenoma with mutations at exon 15 of APC. An identical observation was recently noted in 9 German patients with FAP examined by DBE.11 Furthermore, the positive correlation between small-intestinal adenoma and osteoma in our patients seems to support the genotype-jejunal phenotype in FAP, because the latter manifestation was shown to be related to mutations in the 3’ part of APC.34,35 In a large survey by Heinimann et al,36 however, there was not any genotypephenotype correlation in small-intestinal polyps. The insignificant result is probably a consequence of a low sensitivity in the diagnostic tool for the small intestine, because the prevalence of the small-intestinal polyp in that investigation was only 7%. Because the duodenum and the jejunum are consecutive organs with identical anatomical characteristics, it seems reasonable to conclude that the occurrence of adenomas in these 2 areas share a similar trend in the genotypic event with respect to APC. It has been suggested that genetic testing for APC is a necessary procedure not only for the early diagnosis of FAP but also for the prediction of the early colorectal cancer in families with the established diagnosis of the disease.30,34-36 From our results, it seems to be reasonable to regard the genetic testing as a hallmark for the surveillance of the small-intestinal involvement in patients with FAP. From this point of view, it is necessary for enteroscopists to accumulate more data with respect to the natural history of small-intestinal lesions in the disease. Because we could confirm a close association between the stage duodenal adenoma and small-intestinal lesions, patients with FAP and with verified duodenal adenoma seem to be practically the best candidates for enteroscopy. There seem to be some limitations in the interpretation of our results. In the first place, the incidence of smallintestinal adenoma may have been overestimated in this study, because the prevalence of extracolonic manifestations, especially the duodenal lesions, was apparently high in our patients. However, such a provisional severe phenotype in our patients seems to be a consequence of the scrutiny in the assessment, and it does not seem
to have biased our observation, because a similar trend in the high prevalence or the severe phenotype of the manifestations with regard to the scrutiny was also reported in other clinical investigations.20,30,31,37,38 In the second place, the difference in the specification of the instrument between a DBE and an IOE may have biased our results. However, because the view-angle (120 ) and the resolution were not obviously different between the 2 video endoscopes, these specifications do not seem to have contributed greatly to our observations. In conclusion, our results indicated that an oral DBE was equal to an IOE for the detection of small-intestinal adenoma in patients with an established diagnosis of FAP and that small-intestinal adenoma in the disease manifested not only a diminutive polyp but also a broad-based advanced adenoma or even a sharply demarcated depression in its configuration. In addition, the prevalence of small-intestinal adenoma was associated with the severity of duodenal adenomatosis and with mutations in exon 15 of APC. These observations suggest a DBE to be a necessary procedure for scrutiny for patients with FAP and that there is a close genotype-jejunal phenotype correlation in FAP.
918 GASTROINTESTINAL ENDOSCOPY Volume 68, No. 5 : 2008
www.giejournal.org
DISCLOSURE The authors report that there are no disclosures relevant to this publication.
REFERENCES 1. Iddan G, Meron G, Glukhovsky A. Wireless capsule endoscopy. Nature 2000;405:417. 2. Yamamoto H, Sekine Y, Sato Y, et al. Total enteroscopy with a nonsurgical steerable double-balloon method. Gastrointest Endosc 2001;53:216-20. 3. Yamamoto H, Kita H, Sunada K, et al. Clinical outcomes of double-balloon endoscopy for the diagnosis and treatment of small-intestinal diseases. Clin Gastroenterol Hepatol 2004;2:1010-6. 4. May A, Nachbar L, Wardak A, et al. Double-balloon enteroscopy: preliminary experience in patients with obscure gastrointestinal bleeding or chronic abdominal pain. Endoscopy 2003;35:985-91. 5. Schulmann K, Hollerbach S, Kraus K, et al. Feasibility and diagnostic utility of video capsule endoscopy for the detection of small bowel polyps in patients with hereditary polyposis syndromes. Am J Gastroenterol 2005;100:27-37. 6. Burke CA, Santisi J, Church J, et al. The utility of capsule endoscopy small bowel surveillance in patients with polyposis. Am J Gastroenterol 2005;100:1498-502. 7. Mata A, Llach J, Castells A, et al. A prospective trial comparing wireless capsule endoscopy and barium contrast series for small-bowel surveillance in hereditary GI polyposis syndromes. Gastrointest Endosc 2005;61:721-5. 8. Caspari R, von Falkenhausen M, Krautmacher C, et al. Comparison of capsule endoscopy and magnetic resonance imaging for the detection of polyps of the small intestine in patients with familial adenomatous polyposis or with Peutz-Jeghers syndrome. Endoscopy 2004;36:1054-9. 9. Matsumoto T, Esaki M, Moriyama T, et al. Comparison of capsule endoscopy and enteroscopy with double-balloon method in patients with obscure bleeding and polyposis. Endoscopy 2005;37:827-35.
Matsumoto et al 10. Wong RF, Tuteja AK, Haslem DS, et al. Video capsule endoscopy compared with standard endoscopy for the evaluation of small-bowel polyps in persons with familial adenomatous polyposis (with video). Gastrointest Endosc 2006;64:530-7. 11. Mo¨nkemu¨ller K, Fry LC, Ebert M, et al. Feasibility of double-balloon enteroscopy-assisted chromoendoscopy of the small bowel in patients with familial adenomatous polyposis. Endoscopy 2007;39: 52-7. 12. Iida M, Yao T, Ohsato K, et al. Diagnostic value of intraoperative fiberscopy for small-intestinal polyps in familial adenomatous polyposis. Endoscopy 1980;12:161-5. 13. Rodriguez-Bigas MA, Penetrante RB, Herrera L, et al. Intraoperative small bowel enteroscopy in familial adenomatous and familial juvenile polyposis. Gastrointest Endosc 1995;42:560-4. 14. Iida M, Matsui T, Itoh H, et al. The value of push-type jejunal endoscopy in familial adenomatous coli/Gardner’s syndrome. Am J Gastroenterol 1990;85:1346-8. 15. Bertoni G, Sassatelli R, Tansini P, et al. Jejunal polyps in familial adenomatous polyposis assessed by push-type endoscopy. J Clin Gastroenterol 1993;17:343-7. 16. Matsumoto T, Iida M, Kobori Y, et al. Genetic predisposition to clinical manifestations in familial adenomatous polyposis with special reference to duodenal lesions. Am J Gastroenterol 2002;97:180-5. 17. Iida M, Yao T, Itoh H, et al. Natural history of duodenal lesions in Japanese patients with familial adenomatosis coli (Gardner’s syndrome). Gastroenterology 1989;96:1301-6. 18. Iida M, Yao T, Itoh H, et al. Natural history of gastric adenomas in patients with familial adenomatosis coli/Gardner’s syndrome. Cancer 1988;61:605-11. 19. Iida M, Yao T, Itoh H, et al. Natural history of fundic gland polyposis in patients with familial adenomatous polyposis/Gardner’s syndrome. Gastroenterology 1985;89:1021-5. 20. Burke CA, Beck GJ, Church JM, et al. The natural history of untreated duodenal and ampullary adenomas in patients with familial adenomatous polyposis followed in an endoscopic surveillance program. Gastrointest Endosc 1999;49:358-64. 21. Spigelman AD, Williams CB, Talbot IC, et al. Upper gastrointestinal cancer in patients with familial adenomatous polyposis. Lancet 1989;2: 783-5. 22. Matsumoto T, Moriyama T, Esaki M, et al. Performance of antegrade double-balloon enteroscopy: comparison with push-enteroscopy. Gastrointest Endosc 2005;62:392-8. 23. May A, Nachbar L, Schneider M, et al. Push-and-pull enteroscopy using the double-balloon technique: method of assessing depth of insertion and training of the enteroscopy technique using the Erlangen EndoTrainer. Endoscopy 2005;37:66-70. 24. Schelemper RJ, Riddell RH, Kato Y, et al. The Vienna classification of gastrointestinal epithelial neoplasia. Gut 2000;47:251-5. 25. Rubio CA, Nesi G, Messerini L, et al. The Vienna classification applied to colorectal adenomas. J Gastroenterol Hepatol 2006;21:1697-703. 26. Ishida H, Iwama T, Inokuma S, et al. APC gene mutations in a jejunal adenoma causing intussusception in a patient with familial adenomatous polyposis. J Gastroenterol 2002;37:1057-61.
www.giejournal.org
Enteroscopy for familial adenomatous polyposis
27. Iida M, Aoyagi K, Fujimura Y, et al. Nonpolypopid adenomas of the duodenum in patients with familial adenomatous polyposis (Gardner’s syndrome). Gastrointest Endosc 1996;44:305-8. 28. Kubota O, Kino I. Depressed adenomas of the colon in familial adenomatous polyposis: histology, immunohistochemical detection of proliferating cell nuclear antigen (PCNA), and analysis of the background mucosa. Am J Surg Pathol 1995;19:318-27. 29. Matsumoto T, Iida M, Nakamura S, et al. Depressed adenoma of the duodenum in patients with familial adenomatous polyposis: endoscopic and immunohistochemical features. Cancer 1999;86:1414-20. 30. Enomoto M, Konishi M, Iwama T, et al. The relationship between frequencies of extracolonic manifestations and the position of APC germline mutation in patients with familial adenomatous polyposis. Jpn J Clin Oncol 2000;30:82-8. ˚ kerbrant H, Iselius L, et al. Periampullary adenomas and ad31. Bjo¨rk J, A enocarcinomas in familial adenomatous polyposis: cumulative risks and APC gene mutations. Gastroenterology 2001;121:1127-35. 32. Leggett BA, Young JP, Biden K, et al. Severe upper gastrointestinal polyposis associated with sparse colonic polyposis in a familial adenomatous polyposis family with an APC mutation at codon 1520. Gut 1997;41:518-21. 33. Matsumoto T, Iida M, Kobori Y, et al. Progressive duodenal adenomatosis in a familial adenomatous polyposis pedigree with APC mutation at codon 1556. Dis Colon Rectum 2002;45:229-33. 34. Wallis YL, Morton DG, McKeown CM, et al. Molecular analysis of the APC gene in 205 families: extended genotype-phenotype correlations in FAP and evidence for the role of APC amino acid changes in colorectal cancer predisposition. J Med Genet 1999;36:14-20. 35. Giardiello FM, Petersen GM, Piantadosi S, et al. APC gene mutations and extraintestinal phenotype of familial adenomatous polyposis. Gut 1997;40:521-5. 36. Heinimann K, Mu¨llhaupt B, Weber W, et al. Phenotypic differences in familial adenomatous polyposis based on APC gene mutation status. Gut 1998;43:675-9. 37. Heiskanen I, Kellokumpu I, Ja¨rvinen H. Management of duodenal adenoma in 98 patients with familial adenomatous polyposis. Endoscopy 1999;31:412-6. 38. Wallace MH, Frayling IM, Clark SK, et al. Attenuated adenomatous polyposis coli: the role of ascertainment bias through failure to dye-spray at colonoscopy. Dis Colon Rectum 1999;42:1078-80.
Received June 10, 2007. Accepted February 18, 2008. Current affiliations: Department of Medicine and Clinical Science (T.M., M.E, R.Y.-F., T.M., S.Y., S.N., M.I.), and Department of Anatomic Pathology (T.Y., M.D.), Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan. Reprint requests: Takayuki Matsumoto, Department of Medicine and Clinical Science, Graduate School of Medical Sciences, Kyushu University, Maidashi 3-1-1, Higashi-ku, Fukuoka 812-8582, Japan.
Volume 68, No. 5 : 2008 GASTROINTESTINAL ENDOSCOPY 919
Small-intestinal involvement in familial adenomatous polyposis: evaluation by double-balloon endoscopy and intraoperative enteroscopy Takayuki Matsumoto, MD, Motohiro Esaki, MD, Ritsuko Yanaru-Fujisawa, MD, Tomohiko Moriyama, MD, Shinichiro Yada, MD, Shotaro Nakamura, MD, Takashi Yao, MD, Mitsuo Iida, MD Fukuoka, Japan
Background: Small-intestinal adenoma occurs in patients with familial adenomatous polyposis (FAP). Objectives: The aim was to analyze the diagnostic yield of a double-balloon endoscopy (DBE) and an intraoperative enteroscopy (IOE) for small-intestinal involvement in FAP. Patients: Forty-one patients with FAP. Interventions: We examined 12 patients with FAP by using oral DBE before a colectomy and 29 patients with FAP by using IOE. The incidence and the endoscopic findings of adenoma were compared between the 2 procedures. Phenotypes of FAP and genotypes of adenomatous polyposis coli (APC) were then compared between patients with small-intestinal adenomas and those without. The genotype was classified into a 5’ mutation (exons 1-14), a 3’ mutation (exon 15), and a negative mutation of APC. Main Outcome Measurement: The prevalence of adenoma. Results: A DBE detected small-intestinal adenomas in 9 of 12 patients (75%), as did an IOE in 15 of 29 patients (52%, P O.05). The adenomas occurred predominantly in the jejunum, with a configuration of diminutive polyps in 22 patients. In addition, a DBE detected nonpolypoid adenoma in a patient, and nodular, broad-based protrusion (advanced lesions) in 3 patients, whereas an IOE detected advanced lesions in a patient. Patients with smallintestinal adenoma had more severe duodenal adenomatosis than those patients without small-intestinal adenoma (P!.001). In cases in which APC was analyzed, the prevalence of small-intestinal adenoma was higher in patients with a 3’ mutation (100%) than in those with a 5’ mutation (44%) and with a negative mutation (42%, P!.02). Limitation: Not a prospective randomized study. Conclusions: A DBE is equal to an IOE for scrutiny of small-intestinal adenomas in FAP. There seems to be a genotype-jejunal phenotype correlation in FAP. (Gastrointest Endosc 2008;68:911-9.)
See CME section; p. 961. Copyright ª 2008 by the American Society for Gastrointestinal Endoscopy 0016-5107/$34.00 doi:10.1016/j.gie.2008.02.067
and an overtube.2-4 Because the latter procedure is characterized by 2 balloons attached at the tips of the enteroscope and the overtube, it is referred to as double-balloon endoscopy (DBE). Recent clinical trials showed that both VCE and DBE are best indicated for patients with hereditary GI polyposis.5-11 Our previous investigation showed that DBE was superior to VCE for the detection of diminutive polyps in patients with familial adenomatous polyposis (FAP) and Peutz-Jeghers syndrome.9 Wong et al10 recently reported that, in their patients with FAP, VCE identified fewer numbers of jejunal polyps than did push enteroscopy. These observations suggest that VCE may not be the choice of procedure for the scrutiny of small-intestinal polyposis.
www.giejournal.org
Volume 68, No. 5 : 2008 GASTROINTESTINAL ENDOSCOPY 911
It has become possible to observe an extensive area of the small intestine by endoscopy. Video capsule endoscopy (VCE) is a noninvasive procedure characterized by the application of a wireless capsule.1 Another procedure is an invasive method with a conventional enteroscope Abbreviations: APC, adenomatous polyposis coli; CHRPE, congenital hypertrophy of the retinal pigment epithelium; DBE, double-balloon endoscopy; FAP, familial adenomatous polyposis; IOE, intraoperative enteroscopy; PCR, polymerase chain reaction; PTT, protein truncation test; VCE, video capsule endoscopy.
Enteroscopy for familial adenomatous polyposis
More recently, Mo ¨ nkemu ¨ller et al11 detected jejunal adenomas by DBE in patients with FAP and found that jejunal neoplasm segregated in patients with mutations at exon 15 of the adenomatous polyposis coli gene (APC). However, in 1980s and 1990s, we and other investigators reported that intraoperative enteroscopy (IOE)12,13 and push enteroscopy detected a similar or even greater prevalence of jejunal adenomas in FAP.14,15 In the present retrospective study, we analyzed findings obtained by an IOE and a DBE in our patients with FAP to determine the practical value of a DBE in the disease and to correlate small-intestinal phenotypes with APC mutation.
Matsumoto et al
Capsule Summary What is already known on this topic d
Small-intestinal adenomas may occur in patients with familial adenomatous polyposis (FAP), but their prevalence varies, depending on the modality used for their detection.
What this study adds to our knowledge d
In a study of 41 patients with FAP, the prevalence of small-intestinal adenoma was 75% in patients examined by oral double-balloon endoscopy and 52% in those examined by intraoperative enteroscopy.
PATIENTS AND METHODS Patients During a period from 1990 to 2006, we made an established diagnosis of FAP in 122 patients from 90 family histories. The diagnostic criterion of FAP was a confirmation of more than a hundred colorectal adenomas. Among those patients, we performed an IOE in 29 patients at the time of subtotal colectomy with ileorectal anastomosis or a total proctocolectomy with ileoanal anastomosis during a period from 1990 until 2003. During the subsequent 3-year period from 2004 to 2006, we applied a DBE for 12 patients before surgery. These 41 patients were the subjects of the present investigation. Informed consent for an enteroscopy was obtained from each subject. Furthermore, the genetic analysis of APC was approved by the institutional review board at Kyushu University Hospital.
in polyp number (1-4 polyps, 1 point; 5-20 polyps, 2; O20 polyps, 3), polyp size (1-4 mm, 1; 5-10 mm, 2; O10 mm, 3), histologic type (tubular, 1; tubulovillous, 2; villous, 3), and dysplasia (mild, 1; moderate, 2; severe, 3). A score of 0 points was regarded as stage 0, 1-4 as stage I, 5-6 as stage II, 7-8 as stage III, and 9-12 as stage IV. The ampullary lesion was determined to be positive or negative according to the histologic findings. An epidermoid cyst was assessed by physical examination. Osteoma was examined by plain orthopantomography. Congenital hypertrophy of the retinal pigment epithelium (CHRPE) was screened by direct ophthalmoscopy and was regarded to be positive when hyperpigmented areas with well-defined margins were found. Desmoids were examined by physical examination and abdominal CT.
Assessment of clinical features Enteroscopy
The clinical record of each subject at the time of an enteroscopy was reviewed, and manifestations specific to FAP were retrospectively investigated. As we previously reported,16 the clinical manifestations were evaluated in all the patients at the time of the initial diagnosis. Colorectal polyposis was graded to be sparse (!200), intermediate (between 200 and 1000), or profuse (O1000) by the number of polyps under radiography or macroscopy of the colon. In 35 patients who were treated by surgery, the number of macroscopic polyps was directly counted in the resected specimen. In the remaining 6 patients who had not undergone surgery, colorectal polyps were counted in double-contrast barium enema examination. Upper-GI lesions were examined by chromoscopy with a side-viewing instrument and a forceps biopsy. We used the side-viewing instrument to precisely observe both the periampullary and gastric lesions.17-20 Fundic gland polyps and adenoma of the stomach, and duodenal adenoma were determined to be positive when the histology of the biopsy specimen was compatible with each lesion. The severity of duodenal adenomatosis was determined according to the staging system reported by Spigelman et al.21 The severity was scored by the sum of grades
Oral DBE was performed by an endoscopist (T.M.) with the Double-Balloon Enteroscopy System (Fujinon-Toshiba ES System Co, Tokyo, Japan). After an overnight fast, the patients were prepared by continuous intravenous infusion and were examined by an enteroscopy under a light sedation with IV midazolam coupled with 17.5 mg pethidine chloride. The procedure was carried out under fluoroscopy, as has been previously reported.22 In brief, an endoscopist controlled the enteroscope, and an assistant advanced the overtube. During the insertion of the enteroscope, the balloon of the overtube was inflated and attached to the small-intestinal wall. Alternatively, the balloon of the enteroscope was inflated and attached to the intestinal wall during the insertion of the overtube. The small intestine within the reach of the enteroscope was intermittently shortened by pulling back the overtube and the enteroscope with both balloons inflated and attached to the intestinal wall. When the enteroscope was unable to advance further, a plain abdominal radiograph was obtained with the patient in the supine position, and then the small intestine within the reach of the enteroscope was carefully explored. The depth of
912 GASTROINTESTINAL ENDOSCOPY Volume 68, No. 5 : 2008
www.giejournal.org
Matsumoto et al
exploration during an oral DBE was calculated according to the method reported by May et al.23 An IOE was performed by 1 of 3 endoscopists (M.I., R.F., T.M.) during the surgical procedures, as previously described.12 In brief, an enterotomy was made in the middle of the small intestine, through which a sterilized video endoscopy (SIF-2000; Olympus) was inserted. The enteroscope was initially inserted to the ligament of Treitz with a manual maneuver by the surgeons, and it was slowly pulled back to the enterotomy. The enteroscope was then manually inserted to the ileocecal valve and again pulled back to the enterotomy. In both a DBE and an IOE, a chromoscopy by spraying a 0.5% indigo-carmine solution and a forceps biopsy were performed when any endoscopic sign of adenoma was suggested.
Enteroscopy for familial adenomatous polyposis
TABLE 1. Phenotypic features of the subjects
Clinical variables Age (median) (y)
DBE group IOE group (n Z 12) (n Z 29) 42 (range 20-66)
P
28 (range 12-61)
.04
Sex (women/men)
5/7
14/15
NS
Family history of FAP (þ/–)
7/5
12/17
NS
Severity (profuse/ intermediate/scattered)
2/4/6
10/14/5
.04
Invasive cancer (þ/–)
2/10
9/20
NS
Fundic gland polyposis (þ/–)
6/6
15/14
NS
Gastric adenoma (þ/–)
5/7
15/14
NS
Gastric cancer (þ/–)
0/12
0/29
NS
Score 0 (stage 0)
3
3
NS
Score 1w4 (stage I)
1
5
Score 5,6 (stage II)
3
12
Score 7,8 (stage III)
2
7
Score 9-12 (stage IV)
3
2
Ampullary adenoma (þ/–)
5/6
16/13
NS
CHRPE (þ/–)y
5/6
16/8
NS
Osteoma (þ/–)
7/5
20/9
NS
Epidermoid cyst (þ/–)
4/8
6/23
NS
Desmoid (þ/–)
1/11
3/26
NS
Colorectal lesions
Gastric lesions
Histology Biopsy specimens obtained by a gastroduodenoscopy, DBE, and IOE were fixed in 5% formalin, embedded in paraffin, and cut in slices approximately 5-mm thick. The slices were stained with hematoxylin and eosin, and observed by a pathologist (T.Y.) who was informed that the specimens were from subjects with FAP but was not informed of other information. In the specimens positive for epithelial neoplasia, the histologic phenotype and the category were graded according to the Vienna classification.24,25 The latter was divided into noninvasive low-grade neoplasia, noninvasive high-grade neoplasia, and invasive neoplasia.
APC analysis In 30 patients from whom informed consent for genetic analysis was obtained, APC was screened by the protein truncation test (PTT), as was previously described.16 In brief, genomic DNA and messenger RNA were obtained from peripheral leukocytes, and the latter was reversetranscribed to complementary DNA. Complementary DNA was amplified with 2 primer pairs by polymerase chain reaction (PCR) for proximal segments of APC (segment A, codon 2-479; segment B, 348-758). DNA was amplified with 4 primer pairs for distal segments of APC (segment C, codon 658-1283; segment D, 1099-1700; segment E, 1547-2246; segment F, 2123-2843). The PCR products were applied to rabbit reticulocyte lysate and translated into protein incorporation with 38S-methionine. The synthesized protein was electrophoresed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and full-length or truncated protein was determined. Based on the results of PTT, the APC mutation was classified into proximal mutation (segment A-B, exons 1-14), distal mutation (segment B-F, exon 15), and negative mutation.
Duodenal adenomatosis*
Extraintestinal lesions
NS, Not significant; þ, present; –, absent. *Assessed according to the classification reported by Spigelman et al (Ref. 21). ySix patients (1 patient in the DBE group and 5 patients in the IOE group) were not examined.
ability test or the c2 test when appropriate. Probabilities !.05 were considered to be statistically significant.
RESULTS Demographic data
Parametric data were expressed as median (range) and were compared between 2 groups by using the MannWhitney test. Nonparametric data were expressed as percentage and were compared by the Fisher exact prob-
Demographic data between the DBE group and the IOE group are compared in Table 1. Although the sex and the rate of proband for FAP were not different between the 2 groups, the age at the time of an enteroscopy was significantly older in the DBE group than in the IOE group (42 years [range 20-66 years] vs 28 years [12-61 years], P ! .05). In contrast, the scattered type of
www.giejournal.org
Volume 68, No. 5 : 2008 GASTROINTESTINAL ENDOSCOPY 913
Statistical analysis
Enteroscopy for familial adenomatous polyposis
colorectal adenomatosis was more frequent in the DBE group than in the IOE group (50% vs 17%, P ! .05). Although the incidence of invasive colorectal cancer at the time of diagnosis was higher in the IOE group (45%) than in the DBE group (20%), the difference did not reach a statistical significance. The prevalence of extracolonic lesions is also compared in Table 1. Fundic gland polyposis was found in 21 of the overall 41 patients (51%), and gastric adenoma was found in 20 patients (49%). The prevalence of those gastric lesions was not different between the 2 groups. The prevalence of duodenal (75% vs 90%) and ampullary adenoma (50% vs 55%) was not different between the DBE group and the IOE group. In addition, the severity of duodenal adenomatosis was not different between the DBE and IOE groups. Similarly, the prevalence of CHRPE (45% vs 67%) and osteoma (58% vs 69%) was not different between the 2 groups.
Enteroscopic and histologic diagnosis
Matsumoto et al
TABLE 2. Comparison of enteroscopic diagnosis between the DBE group and the IOE group DBE group, IOE group, no. (%) no. (%) (n Z 12) (n Z 29) P Prevalence of adenoma
9 (75)
15 (52)
NS
Jejunum
9 (75)
14 (48)
NS
Ileum
2 (17)
3 (10)
NS
!5
9 (75)
23 (79)
NS
R5
3 (25)
6 (21)
Diminutive polyp
7 (58)
15 (52)
NS
Nonpolypoid lesion
1 (8)
0 (0)
NS
Advanced lesion
3 (25)
1 (3)
.07
Site of adenoma
No. adenoma
Endoscopic configuration
In the DBE group, the time for examination ranged from 58 to 108 minutes, with a medium of 78 minutes, and the calculated length of insertion from the ligament of Treitz ranged from 150 to 450 cm, with a medium of 320 cm. Comparisons of enteroscopic and histologic diagnoses between the DBE and IOE groups are shown in Table 2. The overall prevalence of small-intestinal adenoma was 75% (9 of 12 patients) in the DBE group and 52% (15 of 29 patients) in the IOE group. The prevalence determined by the 2 procedures was not statistically different (P O .05). In both DBE and IOE groups, the prevalence of adenoma was higher in the jejunum than in the ileum. The number of identified adenoma was 5 or more in 25% of patients in the DBE group and in 21% of the IOE group. In both DBE and IOE groups, sessile, diminutive, and whitish polyps were the most frequent enteroscopic features of the small-intestinal adenoma (Fig. 1A). In 3 patients in the DBE group (25%) and in a patient in the IOE group (3%), nodular, broad-based protrusions were identified (Fig. 1C). Furthermore, an adenoma in a patient in the DBE group was recognized as a sharply demarcated depression (Fig. 1E). Histologic examination showed that most of the lesions were tubular adenomas (Fig. 1B, D, and F), although there was a tubulovillous adenoma in a patient in the IOE group. All the lesions were diagnosed as noninvasive low-grade neoplasia, without any sign of high-grade or invasive neoplasia. Among patients in the DBE group, 4 patients were subsequently examined by anal DBE, 4 patients were examined by small-bowel radiography plus retrograde ileoscopy, and 3 patients were examined by VCE. These procedures did not identify any small-intestinal polyp in the lower part of the ileum. The remaining patient was not examined further.
Twenty-four patients were determined to have smallintestinal adenoma by a DBE or an IOE. A comparison of phenotypes between patients with small-intestinal adenoma and those without is shown in Table 3. As shown in the table, there were trends toward higher incidence in small-intestinal adenoma in men, in patients with profuse or intermediate polyposis, and in patients with fundic gland polyposis. However, the correlation between those phenotypes and small-intestinal adenoma was insignificant. Although not statistically significant, there was also a trend toward a higher incidence of small-intestinal adenoma in patients with duodenal adenoma than those without (96% vs 71%, P Z .07). Furthermore, the score for duodenal adenomatosis was significantly higher in the former (score 0-11, 6.5) than in the latter (score 0-6, 4), P !.001). As shown in Figure 2, the stage of duodenal adenomatosis was more advanced in patients with smallintestinal adenoma than those without (P ! .001). The prevalence of CHRPE, epidermoid cysts, and desmoid was not different between patients with small-intestinal adenoma and those without. However,
914 GASTROINTESTINAL ENDOSCOPY Volume 68, No. 5 : 2008
www.giejournal.org
Histologic type
NS
Tubular only
9 (75)
14 (48)
Tubular and tubulovillous
0 (0)
1 (3)
Category
NS
Noninvasive low grade
9 (75)
15 (52)
Noninvasive high grade
0 (0)
0 (0)
Invasive
0 (0)
0 (0)
NS, Not significant.
Small-intestinal phenotype and other phenotypes
Matsumoto et al
Enteroscopy for familial adenomatous polyposis
Figure 1. Endoscopic and histologic findings of adenomas found in patients with FAP. A, A diminutive adenoma found in the jejunum of a 48-year-old man; a DBE shows a small, whitish polyp. B, On histologic examination, the lesion (A) was composed of epithelial cells with enlarged nuclei with obvious nucleoli; the findings are compatible with noninvasive low-grade tubular adenoma. (H&E, orig. mag. 200). C, An advanced adenoma in the jejunum of a 48-year-old woman; a DBE with chromoscopy shows a large, broad-based protrusion with a nodular surface. D, A biopsy specimen (C) contained tubules of noninvasive low-grade adenoma (H&E, orig. mag. 200). E, A nonpolypoid adenoma in the jejunum of a 38-year-old man; a DBE with chromoscopy shows a small, distinct, depressed lesion in the jejunum. A biopsy specimen from the lesion (F) contains crowding tubules of noninvasive low-grade adenoma (H&E, orig. mag. 200).
www.giejournal.org
Volume 68, No. 5 : 2008 GASTROINTESTINAL ENDOSCOPY 915
Enteroscopy for familial adenomatous polyposis
Matsumoto et al
TABLE 3. Comparison of phenotype of FAP between patients with small-intestinal adenoma and those without small-intestinal adenoma
Clinical variables
With small-intestinal adenoma (n Z 24)
Without small-intestinal adenoma (n Z 17)
P
Age (median) (y)
28 (range 13-57)
34 (range 12-66)
NS
Sex (women/men)
8/16
11/6
.06
Family history of FAP (þ/–)
14/10
5/12
NS
Severity (profuse to intermediate/scattered)
20/4
10/7
.08
Invasive cancer (þ/–)
6/18
6/11
NS
Fundic gland polyposis (þ/–)
15/9
5/12
.06
Gastric adenoma (þ/–)
14/10
6/11
NS
Gastric cancer (þ/–)
0/24
0/17
NS
Duodenal adenomatosis (stage 0/I-IV)*
23/1
12/5
.07
Ampullary adenoma (þ/–)
15/9
8/7
NS
CHRPE (þ/–)y
15/8
6/6
NS
Osteoma (þ/–)
20/4
7/10
.008
Epidermoid cyst (þ/–)
8/16
2/15
NS
Desmoid (þ/–)
1/23
3/14
NS
Colorectal lesions
Gastric lesions
Duodenal lesions
Extraintestinal lesions
NS, Not significant. *Assessed according to the classification reported by Spigelman et al (Ref. 21). ySix patients (1 patient in the DBE group and 5 patients in the IOE group) were not examined.
the prevalence of osteoma was significantly higher in the former patients than in the latter (83% vs 41%, P Z .008).
Genotype and small-intestinal phenotype PTT determined 9 patients to have a 5’ mutation (exons 1-14) and 9 patients to have a 3’ mutation (exon 15) of APC. However, PTT did not identify an APC mutation in the remaining 12 patients. A correlation between APC genotype and small-intestinal phenotype is shown in Table 4. As shown in the table, the prevalence of small-intestinal adenoma was the highest in a 3’ mutation with a value of 100% followed by a 5’ mutation (44%) and a negative mutation (42%). The prevalence of small-intestinal adenoma was significantly different among the 3 groups (P ! .02).
with advanced adenomas found by a DBE was treated by a local resection during the subsequent proctocolectomy. We planned to treat the advanced adenomas in the other 2 patients by endoscopic mucosal resection during the forthcoming proctocolectomy. We decided to follow-up the remaining 20 patients with diminutive small-intestinal adenomas. Five of the patients were examined by a repeated DBE during a period ranging from 2 to 4 years after the initial enteroscopy. The follow-up examination revealed that the size and the number of small-intestinal adenomas remained unchanged.
DISCUSSION
An advanced adenoma found by an IEO was removed by local resection during the surgery. Similarly, a patient
The results of our retrospective study indicated that the prevalence of small-intestinal adenoma was slightly but insignificantly higher in patients with FAP examined by a DBE than in those examined by an IOE, and that the
916 GASTROINTESTINAL ENDOSCOPY Volume 68, No. 5 : 2008
www.giejournal.org
Follow-up data
Matsumoto et al
Enteroscopy for familial adenomatous polyposis
TABLE 4. Comparison of APC genotype between patients with small-intestinal adenoma and those without small-intestinal adenoma* With smallintestinal adenoma, no. (%)y
Without small-intestinal adenoma, no. (%)y
Total
50 mutation (exons 1-14)
4 (44)
5 (56)
9
30 mutation (exon 15)
9 (0)
0 (0)
9
Negative mutation
5 (42)
7 (58)
12
18
12
30
APC genotype
Total
*The prevalence of small-intestinal adenoma is significantly different between the groups (P ! .02). yData in parentheses refer to percentage in each genotype.
small-intestinal phenotype of FAP was closely associated with osteoma, the severity of duodenal adenoma, and the germline mutation in exon 15 of APC. Even though we could not compare the diagnostic value of a DBE and an IOE in identical individuals, the higher prevalence of the smallintestinal adenoma in patients examined by a DBE may be a consequence of scrutiny achieved by the procedure. With the development of novel enteroscopy systems, small-intestinal involvement in patients with FAP was rigorously examined in various clinical trials. The prevalence of small-intestinal polyps reported in those investigations, however, ranges widely according to the procedures applied or to the number of subjects examined.6-8 In particular, Caspari et al8 found diminutive small-intestinal polyps in all of their 4 patients with FAP, whereas Burke et al,6 who examined the largest number of subjects among the VCE studies, detected small-intestinal polyps in 9 of 15 patients. In another investigation, by Schulmann et al,5 a combination of push enteroscopy and VCE detected small-intestinal adenoma in 17 of 29 patients with FAP (59%). Of note is that, in our enteroscopic survey with enrolling the largest number of patients with FAP to date, 59% of the patients were proven to have small-intestinal adenoma. It thus seems probable that the actual prevalence of small-intestinal adenoma in FAP is approximately 60% in both Western countries and Japan.
Schulmann et al5 reported that there was a considerable discordance in the number of small-intestinal polyps between push enteroscopy and VCE in patients with FAP. More recently, Wong et al10 analyzed VCE findings in 32 patients with FAP and with small-intestinal adenomas proven by push enteroscopy and found that the latter procedure identified a greater number of adenomas. However, Mo ¨ nkemu ¨ller et al11 performed a DBE for 9 patients with colectomized FAP and found small-intestinal adenoma in 6 patients. Because the performance of an oral DBE was shown to be superior to that of push enteroscopy15 and because adenomas occur predominantly in the jejunum of FAP,12-15 it seems to be reasonable to presume that the value of a DBE for FAP is equivalent to that of a combination of push enteroscopy and VCE. Our results further suggested oral DBE to be alternative to an IOE for the assessment of small-intestinal involvement in FAP.12,13 As has been reported in duodenal adenomatosis,17,20,21 the predominant endoscopic findings in small-intestinal adenomas in our patients were diminutive, whitish polyps or larger, broad-based lesions. Although there have been sparse descriptions of advanced jejunal adenomas in ¨nkemu FAP,15,26 Mo ¨ller et al11 recently detected such lesions by DBE in 2 of their 9 patients with FAP. Furthermore, we could identify in a patient, by a DBE, an adenoma characterized by a sharply demarcated depression. Similar nonpolypoid adenomas were demonstrated in the colon and in the duodenum of patients with FAP.27,28 In addition, analyses of the proliferative activity of those nonpolypoid lesions indicated that they are presumably precursors of protruding adenoma.28,29 The nonpolypoid adenoma as found in one of our patients with FAP thus suggests the heterogeneity in configuration of early adenoma even in the small intestine. Since the identification of APC, genotype-phenotype correlations have been shown in some clinical
www.giejournal.org
Volume 68, No. 5 : 2008 GASTROINTESTINAL ENDOSCOPY 917
Figure 2. A comparison of the stage of duodenal adenomatosis between patients with small-intestinal adenoma and those without the adenoma.
Enteroscopy for familial adenomatous polyposis
Matsumoto et al
manifestations of FAP. In our previous study, the prevalence of duodenal adenoma was 100% in patients with mutations in exons 10-15 of APC, whereas the value was less than 50% in other patients. A similar trend was found in another Japanese survey by Enomoto et al30 In a Swedish survey of 180 patients, Bjo ¨ rk et al31 found that severe duodenal adenomatosis and duodenal cancer in FAP were closely associated with APC mutations distal to codon 1051 of exon 15. Furthermore, cases of severe duodenal adenomatosis with APC mutations at the distal site of exon 15 (codons 1552 and 1556) were reported.32,33 These observations suggested a genotype-duodenal phenotype correlation in patients with FAP. We could find a positive correlation between the stage of duodenal adenomatosis and the presence of smallintestinal adenoma in the present investigation. There was also a close association of jejunal adenoma with mutations at exon 15 of APC. An identical observation was recently noted in 9 German patients with FAP examined by DBE.11 Furthermore, the positive correlation between small-intestinal adenoma and osteoma in our patients seems to support the genotype-jejunal phenotype in FAP, because the latter manifestation was shown to be related to mutations in the 3’ part of APC.34,35 In a large survey by Heinimann et al,36 however, there was not any genotypephenotype correlation in small-intestinal polyps. The insignificant result is probably a consequence of a low sensitivity in the diagnostic tool for the small intestine, because the prevalence of the small-intestinal polyp in that investigation was only 7%. Because the duodenum and the jejunum are consecutive organs with identical anatomical characteristics, it seems reasonable to conclude that the occurrence of adenomas in these 2 areas share a similar trend in the genotypic event with respect to APC. It has been suggested that genetic testing for APC is a necessary procedure not only for the early diagnosis of FAP but also for the prediction of the early colorectal cancer in families with the established diagnosis of the disease.30,34-36 From our results, it seems to be reasonable to regard the genetic testing as a hallmark for the surveillance of the small-intestinal involvement in patients with FAP. From this point of view, it is necessary for enteroscopists to accumulate more data with respect to the natural history of small-intestinal lesions in the disease. Because we could confirm a close association between the stage duodenal adenoma and small-intestinal lesions, patients with FAP and with verified duodenal adenoma seem to be practically the best candidates for enteroscopy. There seem to be some limitations in the interpretation of our results. In the first place, the incidence of smallintestinal adenoma may have been overestimated in this study, because the prevalence of extracolonic manifestations, especially the duodenal lesions, was apparently high in our patients. However, such a provisional severe phenotype in our patients seems to be a consequence of the scrutiny in the assessment, and it does not seem
to have biased our observation, because a similar trend in the high prevalence or the severe phenotype of the manifestations with regard to the scrutiny was also reported in other clinical investigations.20,30,31,37,38 In the second place, the difference in the specification of the instrument between a DBE and an IOE may have biased our results. However, because the view-angle (120 ) and the resolution were not obviously different between the 2 video endoscopes, these specifications do not seem to have contributed greatly to our observations. In conclusion, our results indicated that an oral DBE was equal to an IOE for the detection of small-intestinal adenoma in patients with an established diagnosis of FAP and that small-intestinal adenoma in the disease manifested not only a diminutive polyp but also a broad-based advanced adenoma or even a sharply demarcated depression in its configuration. In addition, the prevalence of small-intestinal adenoma was associated with the severity of duodenal adenomatosis and with mutations in exon 15 of APC. These observations suggest a DBE to be a necessary procedure for scrutiny for patients with FAP and that there is a close genotype-jejunal phenotype correlation in FAP.
918 GASTROINTESTINAL ENDOSCOPY Volume 68, No. 5 : 2008
www.giejournal.org
DISCLOSURE The authors report that there are no disclosures relevant to this publication.
REFERENCES 1. Iddan G, Meron G, Glukhovsky A. Wireless capsule endoscopy. Nature 2000;405:417. 2. Yamamoto H, Sekine Y, Sato Y, et al. Total enteroscopy with a nonsurgical steerable double-balloon method. Gastrointest Endosc 2001;53:216-20. 3. Yamamoto H, Kita H, Sunada K, et al. Clinical outcomes of double-balloon endoscopy for the diagnosis and treatment of small-intestinal diseases. Clin Gastroenterol Hepatol 2004;2:1010-6. 4. May A, Nachbar L, Wardak A, et al. Double-balloon enteroscopy: preliminary experience in patients with obscure gastrointestinal bleeding or chronic abdominal pain. Endoscopy 2003;35:985-91. 5. Schulmann K, Hollerbach S, Kraus K, et al. Feasibility and diagnostic utility of video capsule endoscopy for the detection of small bowel polyps in patients with hereditary polyposis syndromes. Am J Gastroenterol 2005;100:27-37. 6. Burke CA, Santisi J, Church J, et al. The utility of capsule endoscopy small bowel surveillance in patients with polyposis. Am J Gastroenterol 2005;100:1498-502. 7. Mata A, Llach J, Castells A, et al. A prospective trial comparing wireless capsule endoscopy and barium contrast series for small-bowel surveillance in hereditary GI polyposis syndromes. Gastrointest Endosc 2005;61:721-5. 8. Caspari R, von Falkenhausen M, Krautmacher C, et al. Comparison of capsule endoscopy and magnetic resonance imaging for the detection of polyps of the small intestine in patients with familial adenomatous polyposis or with Peutz-Jeghers syndrome. Endoscopy 2004;36:1054-9. 9. Matsumoto T, Esaki M, Moriyama T, et al. Comparison of capsule endoscopy and enteroscopy with double-balloon method in patients with obscure bleeding and polyposis. Endoscopy 2005;37:827-35.
Matsumoto et al 10. Wong RF, Tuteja AK, Haslem DS, et al. Video capsule endoscopy compared with standard endoscopy for the evaluation of small-bowel polyps in persons with familial adenomatous polyposis (with video). Gastrointest Endosc 2006;64:530-7. 11. Mo¨nkemu¨ller K, Fry LC, Ebert M, et al. Feasibility of double-balloon enteroscopy-assisted chromoendoscopy of the small bowel in patients with familial adenomatous polyposis. Endoscopy 2007;39: 52-7. 12. Iida M, Yao T, Ohsato K, et al. Diagnostic value of intraoperative fiberscopy for small-intestinal polyps in familial adenomatous polyposis. Endoscopy 1980;12:161-5. 13. Rodriguez-Bigas MA, Penetrante RB, Herrera L, et al. Intraoperative small bowel enteroscopy in familial adenomatous and familial juvenile polyposis. Gastrointest Endosc 1995;42:560-4. 14. Iida M, Matsui T, Itoh H, et al. The value of push-type jejunal endoscopy in familial adenomatous coli/Gardner’s syndrome. Am J Gastroenterol 1990;85:1346-8. 15. Bertoni G, Sassatelli R, Tansini P, et al. Jejunal polyps in familial adenomatous polyposis assessed by push-type endoscopy. J Clin Gastroenterol 1993;17:343-7. 16. Matsumoto T, Iida M, Kobori Y, et al. Genetic predisposition to clinical manifestations in familial adenomatous polyposis with special reference to duodenal lesions. Am J Gastroenterol 2002;97:180-5. 17. Iida M, Yao T, Itoh H, et al. Natural history of duodenal lesions in Japanese patients with familial adenomatosis coli (Gardner’s syndrome). Gastroenterology 1989;96:1301-6. 18. Iida M, Yao T, Itoh H, et al. Natural history of gastric adenomas in patients with familial adenomatosis coli/Gardner’s syndrome. Cancer 1988;61:605-11. 19. Iida M, Yao T, Itoh H, et al. Natural history of fundic gland polyposis in patients with familial adenomatous polyposis/Gardner’s syndrome. Gastroenterology 1985;89:1021-5. 20. Burke CA, Beck GJ, Church JM, et al. The natural history of untreated duodenal and ampullary adenomas in patients with familial adenomatous polyposis followed in an endoscopic surveillance program. Gastrointest Endosc 1999;49:358-64. 21. Spigelman AD, Williams CB, Talbot IC, et al. Upper gastrointestinal cancer in patients with familial adenomatous polyposis. Lancet 1989;2: 783-5. 22. Matsumoto T, Moriyama T, Esaki M, et al. Performance of antegrade double-balloon enteroscopy: comparison with push-enteroscopy. Gastrointest Endosc 2005;62:392-8. 23. May A, Nachbar L, Schneider M, et al. Push-and-pull enteroscopy using the double-balloon technique: method of assessing depth of insertion and training of the enteroscopy technique using the Erlangen EndoTrainer. Endoscopy 2005;37:66-70. 24. Schelemper RJ, Riddell RH, Kato Y, et al. The Vienna classification of gastrointestinal epithelial neoplasia. Gut 2000;47:251-5. 25. Rubio CA, Nesi G, Messerini L, et al. The Vienna classification applied to colorectal adenomas. J Gastroenterol Hepatol 2006;21:1697-703. 26. Ishida H, Iwama T, Inokuma S, et al. APC gene mutations in a jejunal adenoma causing intussusception in a patient with familial adenomatous polyposis. J Gastroenterol 2002;37:1057-61.
www.giejournal.org
Enteroscopy for familial adenomatous polyposis
27. Iida M, Aoyagi K, Fujimura Y, et al. Nonpolypopid adenomas of the duodenum in patients with familial adenomatous polyposis (Gardner’s syndrome). Gastrointest Endosc 1996;44:305-8. 28. Kubota O, Kino I. Depressed adenomas of the colon in familial adenomatous polyposis: histology, immunohistochemical detection of proliferating cell nuclear antigen (PCNA), and analysis of the background mucosa. Am J Surg Pathol 1995;19:318-27. 29. Matsumoto T, Iida M, Nakamura S, et al. Depressed adenoma of the duodenum in patients with familial adenomatous polyposis: endoscopic and immunohistochemical features. Cancer 1999;86:1414-20. 30. Enomoto M, Konishi M, Iwama T, et al. The relationship between frequencies of extracolonic manifestations and the position of APC germline mutation in patients with familial adenomatous polyposis. Jpn J Clin Oncol 2000;30:82-8. ˚ kerbrant H, Iselius L, et al. Periampullary adenomas and ad31. Bjo¨rk J, A enocarcinomas in familial adenomatous polyposis: cumulative risks and APC gene mutations. Gastroenterology 2001;121:1127-35. 32. Leggett BA, Young JP, Biden K, et al. Severe upper gastrointestinal polyposis associated with sparse colonic polyposis in a familial adenomatous polyposis family with an APC mutation at codon 1520. Gut 1997;41:518-21. 33. Matsumoto T, Iida M, Kobori Y, et al. Progressive duodenal adenomatosis in a familial adenomatous polyposis pedigree with APC mutation at codon 1556. Dis Colon Rectum 2002;45:229-33. 34. Wallis YL, Morton DG, McKeown CM, et al. Molecular analysis of the APC gene in 205 families: extended genotype-phenotype correlations in FAP and evidence for the role of APC amino acid changes in colorectal cancer predisposition. J Med Genet 1999;36:14-20. 35. Giardiello FM, Petersen GM, Piantadosi S, et al. APC gene mutations and extraintestinal phenotype of familial adenomatous polyposis. Gut 1997;40:521-5. 36. Heinimann K, Mu¨llhaupt B, Weber W, et al. Phenotypic differences in familial adenomatous polyposis based on APC gene mutation status. Gut 1998;43:675-9. 37. Heiskanen I, Kellokumpu I, Ja¨rvinen H. Management of duodenal adenoma in 98 patients with familial adenomatous polyposis. Endoscopy 1999;31:412-6. 38. Wallace MH, Frayling IM, Clark SK, et al. Attenuated adenomatous polyposis coli: the role of ascertainment bias through failure to dye-spray at colonoscopy. Dis Colon Rectum 1999;42:1078-80.
Received June 10, 2007. Accepted February 18, 2008. Current affiliations: Department of Medicine and Clinical Science (T.M., M.E, R.Y.-F., T.M., S.Y., S.N., M.I.), and Department of Anatomic Pathology (T.Y., M.D.), Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan. Reprint requests: Takayuki Matsumoto, Department of Medicine and Clinical Science, Graduate School of Medical Sciences, Kyushu University, Maidashi 3-1-1, Higashi-ku, Fukuoka 812-8582, Japan.
Volume 68, No. 5 : 2008 GASTROINTESTINAL ENDOSCOPY 919