Clinicopathologic Significance of Immunohistochemical Fecal Occult
Blood Test in Subjects Receiving. Bidirectional Endoscopy

Hui-Hsiung Liu, Thomas W Huang1, Hsiao-Ling Chen1, Teh-Hong Wang2, Jaw-Town Lin2
Graduate Institute of Public Health, Taipei Medical College, 1Taipei Institute of Pathology and 2Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan.
Corresponding Author: Jaw-Town Lin, MD, PhD, Department of Internal Medicine
National Taiwan University Hospital, No. 7, Chun-Shan S. Rd., Taipei, Taiwan
Fax: +886 2 23947899, E-mail: jawtown@ha.mc.ntu.edu.tw.

ABSTRACT Background/Aims: Fecal occult blood test has been utilized to screen for lower gastrointestinal pathologies, such as colorectal cancer and polyps that bleed. Recent studies have revealed a relatively high frequency of upper gastrointestinal abnormalities in subjects with positive fecal occult blood by guaiacbased method. Although immunohistochemical tests of fecal occult blood were assumed to have greater diagnostic validity, the distribution of gastrointestinal pathology using such an examinations is not well established. This study aims to investigate the efficacy of immunohistochemical analysis of fecal occult blood in detecting upper and lower gastrointestinal lesions in asymptomatic individuals. Methodology: Subjects who underwent regular health checkups were enrolled if they received both esophagogastroscopic and colonoscopic examinations. Each subject was tested by an immunohistochemical fecal occult blood test. The fecal occult blood results were evaluated and correlated with lesions identified in endoscopic examinations. Results: In total 655 males and 722 females with age 46.2±12.1 years were enrolled. 287 cases (20.7%) had polypoid lesions of colon, including 6 colon cancers, 37 with polyps 31cm, 104 with polyp 5-9mm, and 140 with polyp <5mm. FOB was positive in 31 cases, of which 15 (15/31, 48.4%) were polypoid lesions of colon, 1 was colonic ulcer, 9 (29.0%) were active gastroduodenal ulcers but 6 (19.4%) had no significant lesions. The positive and negative predictive value for colon polyps was 48.4% and 80%, respectively. The sensitivity was 50% (3/6) for colon cancer and varied among polyps with different sizes: 16.2% (6/37) for polyps 31cm; 5.8% (6/104) for polyps 5-9mm and 0% (0/140) for polyps <5mm. Conclusions: A substantial portion of subjects (29%) with positive fecal occult blood reaction of immunohistochemical analysis but negative colonoscopy still needs esophagogastroscopic examination to disclose upper gastrointestinal lesions. Immunohistochemical determination of fecal occult blood remains imperfect for polypoid lesions of colon in view of its sensitivity and specificity.

 

INTRODUCTION

Colorectal cancer is an important health problem with increasing frequency in Western countries and Taiwan (1,2). Extensive evidence indicates that most colorectal cancers develop in a stepwise fashion from normal mucosa through adenomatous polyps to localized surgically curable malignancy, and eventually culminating in disseminating incurable disease (3). Four modalities are currently available for the evaluation of colorectal lesions. These include fecal occult blood (FOB) testing, sigmoidoscopy, barium enema, and colonoscopy. Among them, guaiac-based FOB test is commonly used for screening because it is rapid, simple, and inexpensive. Three randomized controlled trials have shown a statistically significant reduction of 15% to 33% colorectal cancer mortality rate after FOB screening (4-6). Therefore, the primary test adopted for colorectal cancer screening has been the guaiacimpregnated chemical test.

Despite the abundant evidence for the efficacy of screening in reducing colorectal cancer mortality, certain caveats exist for guaiac-based FOB. It has a modest sensitivity and almost no sensitivity for the detection of the adenomatous polyps which are the precursors of most colorectal carcinomas (7). The emergence of cancers among subjects with a negative test suggested that patients cannot be reassured even after screening (8,9). Another potential problem is a considerable number of patients with positive FOB results and normal colonoscopic findings may have lesions detected during upper gastrointestinal endoscopy (10- 14). Recently, preliminary reports have revealed that an immunohistochemical FOB test has greater diagnostic validity for colorectal cancer than a chemical guaiac test (15). However, the contribution of detection of upper gastrointestinal lesions attributed to pos-itive immunohistochemical FOB is not well documented because few studies have subjected individuals to complete both esophagogatroscopic and colonoscopic examinations. We aimed to investigate the efficacy of immunohistochemical FOB test in the detection of clinically significant upper gastrointestinal and colorectal abnormalities in a group of subjects who received both esophagogastroscopy and colonoscopy, the so-called bidirectional endoscopy.

METHODOLOGY

From December 1997 to November 1999, a total of 2,892 subjects participated in a health checkup at our center. All individuals received FOB tests by an immunohistochemical method (OC-Hemodia, Eiken Chemical, Tokyo, Japan). The OC-Hemodia was a commercial latex agglutination test which was performed according to manufacturer's instructions by experienced personnel, who were unaware of endoscopic results. Stool specimens were collected before endoscopy. The decision to perform bidirectional endoscopy was voluntarily made by the attendants. Colonoscopy was performed followed by esophagogastroscopy. Exclusion criteria included: patients undergoing only unidirectional endoscopy; gross gastrointestinal bleeding, anal-rectal bleeding or gross blood on digital examination, previous history of a known gastrointestinal bleeding lesion, gastrointestinal cancer, previous gastrointestinal surgery, inflammatory bowel disease, and premenopausal females with iron deficiency anemia. Finally, 1,387 subjects fulfilling above criteria were recruited on a voluntary basis. Criteria were predetermined for accepting the endoscopic lesions as a source of occult gastrointestinal bleeding according to previous reports (10-14). In brief, ulcers, carcinomas, vascular lesions, large polyps (>1.0cm) or small polyps with hemorrhagic, friable or ulcerated mucosa were considered significant. Suspected malignancy was subsequently verified by histology. Individuals found to have cancer were transferred for adequate treatment. Those with negative FOB tests, together with those with positive FOB tests where no significant lesion was found following endoscopic investigation, were referred to undertake repeat screens at two yearly intervals. The endoscopic findings were then correlated with FOB results to calculate the sensitivity, positive predictive value and negative predictive value, especially for the polypoid lesions of colon.

RESULTS

Of 1,387 subjects studied, there were 655 males and 722 females with a mean age of 46.2!O12.1 years. The mean value of hemoglobin and hematocrit was 13.6!O2.1g/dL and 41.0!O3.6%, respectively. Polypoid lesions of colon were noted in 287 subjects (20.7%), including 6 colon cancers, 37 with polyps diameter 31cm, 104 with polyps size between 5 and 9mm, and 140 with polyps <5mm. The results of the immunohistochemical FOB test were positive in 31 (2.2%) individuals. The relationship between positive FOB results and significant endoscopic lesions is summarized in Table 1. Of the 31 subjects with positive FOB, 25 (80.6%) had significant gastrointestinal lesions. There were 15 (48.4%) with polypoid lesions of colon, 1 had colonic ulcer, 9 (29.1%) had active gastroduodenal ulcers, but 6 (19.4%) had no significant lesions. The calculated positive and negative predictive value for detecting colon polypoid lesion was 48.4% (15/31) and 80% (1084/1356), respectively. The sensitivity for detecting colon cancer was 50% (3/6) and varied with different sizes of polyps: 16.2% (6/37) for polyps 31cm; 5.8 (6/104) for polyps 5-9mm and 0% (0/140) for polyps <5mm.

TABLE 1 The Results of Immunohistochemical Fecal Occult Blood Test Correlated with Significant Endoscopic Lesions
Gastrointestinal lesion	Immunohistochemical fecal occult blood (FOB) test
	Positive (n=31)	Negative (n=1,356)
Colon cancer	3	3
Colon polyp
≧1cm	6	31
5-9mm	6	98
＜5mm	0	140
Colon ulcer	1	1
Duodenal ulcer	6	96
Gastric ulcer	3a	49b
Others	6	938
a: including one case subsequently proved to be gastric cancer. b: including one case subsequently proved to be gastric cancer and eight cases with combined duodenal ulcer.

 

DISCUSSION

Not all gastrointestinal abnormalities identified by endoscopic examinations may lead to positive reactions of FOB test. The positive results varied with test or the type and size of gastrointestinal lesions in different studies (10-14). Rockey et al. disclosed 36 to 64% lesions in the gastrointestinal tract which presented positive occult blood by guaiac-based method (14). A high detection rate of 80.6% (25/31) in this study may be ascribed to different criteria adopted and various methods of analyses. Additionally, results may also vary according to whether individuals with anemia or symptoms are included (16,17).

Even under meticulous investigation of gastrointestinal tract by bidirectional endoscopy, 19.4% (6/31) of the subjects with positive FOB tests still had no obvious significant abnormalities in this study. Despite the advantage of immunohistochemical testing which is designed to minimize false-positive reaction, our results showed that other conditions remained which can give positive FOB results. Such discrepancy may result from the methodology in collecting stool specimens and/or inexperienced laboratory work (18). Therefore, immunohistochemical FOB test as well as other occult blood tests cannot be considered conclusive for the detection of all gastrointestinal lesions.

In clinical practice, an asymptomatic individual with a positive FOB result is usually asked for a colonoscopic examination. However, upper gastrointestinal lesions may also result in a positive FOB result. Nevertheless, the relative contribution of upper gastrointestinal lesions to the frequency of positive immunohistochemical FOB testing in not well established because few studies have subjected individuals to bidirectional endoscopy (17). Earlier reports utilizing guaiac-based chemical tests revealed significant upper gastrointestinal abnormalities in 27-43% of patients with positive FOB and negative colonoscopy (10-14). Some studies have shown the proportion of upper gastrointestinal lesions was even more frequently identified than colonic lesions in patients with positive FOB if iron-deficiency anemia and active bleeding had been excluded (14). In agreement with previous studies, our result also disclosed that a substantial portion (9/31, 29%) of subjects with positive FOB tests had upper gastrointestinal lesions. Although colonic lesions (16/31,51.6%) were more frequently found than upper gastrointestinal lesions, our results, together with previous reports, implicated that positive FOB testing is not site-specific, irrespective of the analyzing methods.

The principal use of FOB tests is to screen for lower gastrointestinal pathologies, such as colorectal cancer and polyps that bleed. In the present study, the positive and negative predictive value for polypoid lesions of colon (including colon cancers) was 48.4% and 80%, respectively. The calculated sensitivity was modest (50%) for colon cancer but overall sensitivity was much lower (4.2%) for polyps. Moreover, our results showing the sensitivity of FOB test for polyps was related to the size which was consistent with previous reports (19,20). Their results also suggested that colon polyps less than 1cm in diameter seldom bleed to lead to positive reaction of FOB. It should be noted that the sensitivity of an immunohistochemical FOB test for colon cancer varies with the sampling frequency of stool. Nakama et al. have reported that sensitivities of immunochemical occult blood test for oneday, two-day, and three-day method were 67.9%, 88%, and 90.8%, respectively for colon cancers (21). To render an acceptable sensitivity for immunohistochemical FOB as a means of screening for colon cancer, testing should be performed on two or three samples.

In conclusion, our results disclosed the predictive value of FOB test in detecting gastrointestinal lesions. A substantial portion of subjects (29%) with positive FOB reaction with negative colonscopic results still needs an esophagogastroscopic examination. More cases are required to elucidate the true value of immunohistochemical occult blood test and bidirectional endoscopy in the detection of upper and lower gastrointestinal lesions.

REFERENCES

1. Van Custem E, Peeters M, Verslype C, Filez L, Haustermans K, Janssens J: The medical treatment of colon cancer: actual status and new development. Hepatogastroenterology 1999; 46:709-716.
2. Chen HH, Yen MF, Lai MS, et al: Evaluation of a selective screening for colorectal carcinoma: the Taiwan multicenter cancer screening project. Cancer 1999; 86:1116-1128.
3. Fearon ER, Vogelstein B: A genetic model for colorectal tumorigenesis. Cell 1990; 61:759-767.
4. Mandel JS, Bond JH, Church TR, et al: Reducing mortality from colorectal cancer by screening for fecal occult blood. Minnesota colon cancer control study. N Engl J Med 1993; 328:1365-1371.
5. Hardcastle JD, Chamberlain JO, Robinson MH, et al: Randomised controlled trial of faecal-occult-blood screening for colorectal cancer. Lancet 1996; 348:1472-1477.
6. Kronborg O, Fenger C, Olsen J, Jorgensen OD, Sondergaard O: Randomised study of screening for colorectal cancer with faecal-occult-blood test. Lancet 1996; 348:1467- 1471.
7. Ahlquist DA, Wieand HS, Moertel CG, et al: Accuracy of fecal occult blood screening for colorectal neoplasma. A prospective study using Hemoccult and HemoQuant test. JAMA 1993; 269:1262-1267.
8. Bouvier V, Launoy G, Herbert C, Lefevre H, Maurel J, Gignoux M: Colorectal cancer after a negative Hemoccult IIR test and programme sensitivity after a first round of screening: the experience of the development of calvados (France). Br J Cancer 1999; 81:305-309.
9. Cauffman JG, Forsyth RA, Lapsys FX, Clark V: Fiveyear follow-up of asymptomatic patients with negative occult blood test results in whom clinically significant colorectal pathology was detected. Am J Gastroenterol 1997; 92:2036-2039.
10. Zuckerman G, Benitez J: A prospective study of bidirectional endoscopy (colonoscopy and upper endoscopy) in the evaluation of patients with occult gastrointestinal bleeding. Am J Gastroenterol 1992; 87:62-66.
11. Hsia PC, Al-Kawas FH: Yield of upper endoscopy in the evaluation of asymptomatic patients with Hemoccult-positive stool after a negative colonoscopy. Am J Gastroenterol 1992; 87:1571-1574.
12. Geller AJ, Kolts BE, Achem SR, Wears R: The high frequency of upper gastrointestinal pathology in patients with fecal occult blood and colon polyps. Am J Gastroenterol 1993; 88:1184-1187.
13. Chen YK, Gladden DR, Kestenbaum DJ, Collen MJ: In there a role for upper gastrointestinal endoscopy in the evaluation of patients with occult-blood-positive stool and negative colonoscopy? Am J Gastoenterol 1993; 88:2026- 2029.
14. Rockey DC, Koch J, Cello JP, Sanders LL, McQuaid K: Relative frequency of upper gastrointestinal and colonic lesions in patients with positive occult-blood tests. N Engl J Med 1998; 339:153-159.
15. St John DJB, Young GP, Alexeyeff MA, et al: Evaluation of new occult blood tests for detection of colorectal neoplasia. Gastroenterology 1993; 104:1661-1668.
16. Itzkowitz SH: Heads or tails in a positive occult blood test. Lancet 1998; 352:1490.
17. Joosten E, Ghesquiere B, Linthoudt H, et al: Upper and lower gastrointestinal evaluation of elderly inpatients who are iron deficient. Am J Med 1999; 107:24-29.
18. Edwards JB: Improving fecal blood testing. Gastroenterology 1998; 114:226-227.
19. Nakama H, Fattah A, Zhang B, Kamijo N Fujimori K, Miyata K: Detection rate of immunochemical fecal occult blood test for colorectal adenomatous polyps with severe dysplasia. J Gastroenterol 1997; 32:492-495.
20. Fattah A, Nakama H, Kamijo N, Fujimori K, Zhang B: Colorectal adenomatous polyps detected by immunohistochemical occult blood screening. Hepatogastroenterology 1998; 45:712-716.
21. Nakama H, Kamijo N, Fujimo K, Fattah A, Zhang B: Relationship between fecal sampling times and sensitivity and specificity of immunohistochemical fecal occult blood tests for colorectal cancer: a comparative study. Dis Colon Rectum 1997; 40:781-784.