Learning Objectives: After reading this article, the physician should be able to:
- Describe the role of therapeutic iron supplementation on the result of stool guaiac testing
- Discuss the common clinical conditions that may produce a false positive or false negative guaiac test.
- Explain the possible effects of acute iron overdose on guaiac stool testing.
The standard guaiac reaction is a bedside chemical test universally used to determine the presence of blood in gastric aspirate, emesis material, or fecal specimens. The actual chemical reaction that is manifested on the Hemoccult testing card (and many others) is the oxidation of impregnated gum guaiac to a blue quinone compound.
This reaction is initiated by a developing solution containing hydrogen peroxide, using blood as a peroxidase catalyst to initiate the reaction. Blood is not the only peroxidase that can catalyze this reaction, and the most common substances producing false positive tests — next to a benign cause of bleeding, such as hemorrhoids — are of dietary origin. A variety of vegetables contain natural peroxidases that perform the same function as does occult blood in initiating the blue color reaction on the test card. If the patient is a vitamin user, large doses of Vitamin C (1–2 g/day) will have enough of an antioxidant effect to render true bleeding as being falsely negative with guaiac testing.
The Gastroccult test utilizes the same oxidation/peroxidase principles, but it is specially buffered. If the gastric aspirate has a pH of less than 3, the guaiac test used on a stool sample may be falsely negative. Although occult blood testing has gained its greatest popularity for cancer screening, every emergency physician must be cognizant of the intricacies and vagaries of occult testing of stool and gastric aspirate.
This month's column concludes the discussion by focusing on specific parameters of false negative and false positive reactions, with particular emphasis on the effect of iron supplementation, specifics on how the sample is collected, and other factors that affect the clinical use of this test.
An Investigation of the Effects of Oral Iron Supplementation on In Vivo Hemoccult Stool Testing, Andersen GD, et al, Am J Gastroenterology, 1990;85(5):558
This is a prospective randomized double-blind crossover study examining the effects of oral iron ingestion on the results of fecal occult blood testing in healthy volunteers. It was designed to test the age-old question of whether the ingestion of iron tablets results in false positive results on stool testing for gastrointestinal pathology, especially colon cancer. The authors note that this question has been studied extensively in the past, but there have been conflicting results. The authors believe that their study was better designed because it controlled for variables in the diet, collection method, concurrent disease, menstrual contamination, and the interpretation technique.
One hundred healthy asymptomatic volunteers, ages 19 to 40 (mean age 24), were selected to be included in this study. Individuals were screened for prior history of peptic ulcer disease, bowel disease, hemorrhoids, anal fissures, or any past or current abdominal complaints. The subjects were placed on a restricted diet during the entire period to avoid aspirin, NSAIDs, and any supplements containing vitamin C. The participants also were instructed to avoid foods that were known to be high in peroxidase content, including red meats and various raw vegetables. The dietary restrictions were particularly stringent for four days prior to collecting the stool samples. Women in the group were tested at a time when they were not menstruating.
Subjects were given ferrous gluconate (300 mg tablet) in a schedule of two tablets in the morning, two tablets at noon, and one in the evening. Each tablet contained 35 mg of elemental iron. Another group received ferrous sulfate tablets (325 mg) given three times a day. Each of these tablets contains 65 mg of elemental iron. The control group received a placebo.
Stool samples were obtained prior to the initiation of the iron tablets and at the end of each two week iron tablet/placebo period. The stool was collected with a sling device that was draped into the toilet bowl. Stool was not obtained by a digital rectal examination. The stool was tested with a Hemoccult II card. All samples were read within 48 hours of voiding. Samples were not rehydrated prior to interpretation. Two separate physicians simultaneously interpreted the reading results to eliminate reader bias. Of the 100 participants, 78 completed the study.
The results demonstrated that no false positive stool testing occurred secondary to iron supplementation. There was one false positive, and that was in a woman who was not taking iron and had a positive sample one hour before the onset of menses. The authors reviewed the prior literature, and emphasized the difficulty in standardizing occult blood testing due to dietary issues, storage, collection, and interpretation techniques. The authors believe that in 1990, the Hemoccult II card was the most reliable guaiac-based test. The Hemoccult Sensa test has been chemically manipulated to increase the color intensity and stability of the test specimen. The authors conclude that the therapeutic use of ferrous sulfate or ferrous gluconate does not produce false positive testing by standard guaiac methods. The conclusion is that a clinician should suspect a pathologic cause for any positive Hemoccult test in patients who are taking iron supplements.
Comment: The issue of oral iron preparations causing a false positive guaiac test on a stool sample has been debated, and some textbooks still say that this issue is controversial. The theory has been that a positive test is due to blood loss from GI mucosal irritation from the iron, or a positive reaction is produced by the iron itself. Neither condition is clinically correct, and the results of this study are crystal clear for all physicians: Do not attribute a positive stool test of occult blood to the therapeutic use of iron tablets. (See also New Engl J Med 1989;320:1500 and Dig Dis Sci 1998;33:172.)
In addition, a positive test for occult blood cannot be automatically attributed to the presence of an anal fissure, menstrual blood, or hemorrhoids. Such benign local bleeding may well be the source of the positive test, but even the patient with bulging hemorrhoids who is found to have blood in the stool requires additional evaluation to prove that there is no other GI pathology. It is reasonable to reassure the patient who reports bright red rectal bleeding that the source is very unlikely that large bleeding hemorrhoid, but they should be discharged with the strong written advice that a colonoscopy is in order once the hemorrhoids are resolved.
It is clear that the therapeutic use of iron supplements in normal doses with normal GI transit time will not create a falsely positive guaiac test. However, the issue is not as straightforward as first meets the eye, and there are important caveats that should be noted by the emergency physician. While it is true that standard therapeutic use of normal doses of ferrous sulfate or ferrous gluconate tablets do not produce a guaiac positive stool under normal circumstances, iron pills themselves can initiate the reaction.
If you take a ground up iron pill and place it on a guaiac card (see photograph), a blue color will form, suggesting the presence of blood. In the presence of rapid GI transit or particularly in the presence of iron overdose, the iron pill itself may produce a false positive test for blood. Most patients suffering from acute iron overdose will experience significant gastrointestinal distress, manifested by vomiting and profuse diarrhea. If the ingested iron tablets are present in the stool or gastric contents, bedside testing of actual particles of an iron tablet will signify that blood is present.
While a hemorrhagic gastroenteritis is certainly a possibility with acute iron poisoning, merely finding a guaiac positive test of stool/gastric material in the presence of acute iron ingestion does not equate with gastrointestinal bleeding. Most textbooks fail to quote this phenomenon, and strongly state that iron itself does not produce a false positive Hemoccult reading. However, this is not technically correct under the circumstances of rapid GI transient time or vomiting immediately after the ingestion of iron tablets. I could not find mention of this observation in any textbooks, in the package inserts for Hemoccult testing, or even in review articles demonstrating the effect of iron tablets on stool guaiac testing. In short, the books and review articles are incorrect or at least incomplete and potentially misleading. One should not, however, attribute a positive reaction to the presence of iron tablet fragments until other conditions are ruled out.
The clinician also may be led astray by a false positive guaiac reaction if errant blood makes its way into the specimen. This would be an issue after a vigorous digital rectal examination, or if the patient happens to have internal or external hemorrhoids or anal fissure. When performing mass screenings for occult malignancies (recommended for all patients over age 50), it is suggested that stool samples be collected from toilet paper, from a sling draped in the toilet, or by colonoscopy, not by a digital rectal examination (Can J Gastroenterol 2001;15(4):227).
For cancer screening, three separate stool samples are obtained, and two different areas of the stool are tested. In the ED, this is clearly impossible and usually only one sample, obtained by a digital examination, is tested. In general, compared with a digitally derived source, spontaneously passed stool is preferred for cancer screening, and this method has consistently been shown to have better predictive value (Eur Gastrol Hepatal 2000;12(11):1235). Because some authors have not been able to find a significantly different diagnostic yield between collection methods (Arch Intern Med 1991;151(11):2180) for the purpose of the ED exam, a positive test should be considered a positive test.
When it comes to cancer screening, the digital sample lacks the appropriate statistical sensitivity to be considered a mandate for a full GI work-up for presumed occult bleeding. However, if a few falsely positive examinations for occult blood in the ED only prompt a colonoscopy even though the patient only had hemorrhoids, that may not be such a bad idea. ED patients are a select group anyway, and are often unlikely to have a family doctor willing to do three stool guaiac tests to screen for cancer.
Of course, if there is significant skin irritation, maceration, or any other lesion in the area of stool sampling, extraneous blood will be detected. Another potential for a false positive Hemoccult test would be contamination of the sample by iodine, specifically the omnipresent Betadine. Even a trace of Betadine on a Hemoccult card will produce a strong blue color reaction in the absence of blood. Because physicians often start an IV or do some other form of wound care or skin prep (central line, chest tube, Foley catheter) while wearing gloves, and under some circumstances, such as for a multiple trauma workup, one may perform a rectal exam with that same gloved hand without changing it. In such cases, the clinician may wrongly suspect GI bleeding. Obviously, prepping the perineal area with Betadine will produce the same false positive test. A common scenario would be using Betadine to prep the patient for a Foley catheter, allowing a few drops of the antiseptic to contaminate the rectal area.