Explore This Issue
ACEP Now: Vol 38 – No 08 – August 2019ILLUSTRATION: Chris Whissen PHOTOS: shutterstock.com
If a test result does not impact management of a patient—or, worse, is misleading—should the test even be performed? Scrutiny about the guaiac-based fecal occult blood test and its role in the emergency department is long overdue. By carefully assessing this question, some hospitals have correctly concluded that it is time to retire fecal occult blood testing (FOBT).
The FOBT legend begins in the annals of screening for lower gastrointestinal malignancy. Small amounts of heme present in a stool sample react with a hydrogen peroxide–based developer to oxidize guaiac-infused paper, resulting in a blue color.1 Small amounts of blood, potentially from an otherwise asymptomatic early malignancy, may trigger appropriate downstream screening. Ultimately, population-based cancer screening using FOBT was popularized in the 1990s after several trials demonstrated mortality reductions.2 While we do not perform cancer screening in the emergency department, we may take an interest in the presence or absence of blood in the stool. Thus, this inexpensive, readily available test became part of the armamentarium of the emergency department.
Problems with FOBT
Unfortunately, guaiac-based FOBT is not a good test. When used for cancer screening, the sensitivity for malignancy—using heme positivity as a surrogate—is quite poor. Estimates for sensitivity of a single sample range from 15 to 30 percent.3 Because of these limitations, the typical collection procedure involves providing the patient with three separate cards upon which to collect samples from six different stools. These cumbersome procedures have coaxed the movement from guaiac-based kits toward immunochemical assays, whose sensitivity is sufficient with only one sample. Worse still with respect to ED practice, both sensitivity and specificity of guaiac-based FOBT for malignancy decrease with samples collected by digital rectal exam compared to those collected from spontaneously passed stool.4
Meanwhile, the list of sources of false-positive guaiac-based FOBT results is extensive. Nongastrointestinal blood, such as epistaxis, may produce a positive result unrelated to a clinically important etiology. Many foods can confound results of the guaiac-based FOBT, including meat products containing nonhuman heme and vegetables containing peroxidases, such as broccoli. When guaiac-based FOBT is used in outpatient screening, patients are instructed to abstain from such foods, which is not an option in the emergency department. The subjective nature of judging the color-based outcome on the card also leads to both false positives and false negatives.
These data do not generalize well to the emergency department. In the first place, the FOBT is designed to detect blood from an occult lower intestinal source rather than clinically important upper gastrointestinal bleeding. Understandably, as we are using this test outside its defined scope, we have little data with which to accurately describe its sensitivity and specificity. One small study evaluated three different stool tests for lower gastrointestinal bleeding in patients with anemia and known upper gastrointestinal lesions, and the guaiac-based test was positive in only 11 of 42 patients.5 The same study evaluated patients with quantities of ingested blood up to 15 mL; only one of 12 study subjects resulted a positive guaiac-based test. In a study evaluating 2,796 patients undergoing an immunochemical fecal blood test, a substantially more sensitive assay, the authors report advantages for detection of lower gastrointestinal tract lesions but no difference in presence of upper gastrointestinal lesions.6
With this excess of data throwing the utility of FOBT into question, the onus ought to be on those who would support its use in clinical settings to demonstrate its value. Hospitals that retrospectively evaluated the effect of guaiac-based FOBT on downstream testing have found profound mismatch between test results and follow-up endoscopy.7 This finding is consistent with surveys of the perceived value of the test across the spectrum of general practitioners as compared with gastroenterologists—a scant minority of gastroenterologists reported FOBT as appropriate for use in the emergency department, and the majority do not rely on the test to alter management.8
Eliminating the Test
In response, several hospitals have specifically removed the test either from use in the emergency department or the entire hospital.9,10 Parkland Hospital in Dallas, for example, retrospectively evaluated their practice patterns and identified FOBT as a low-value intervention used outside its appropriate scope. After a marginally successful initial attempt at reducing its use through educational interventions, the hospital simply eliminated the test. The discontinuation was led by the gastroenterology group, justifying their de-adoption by focused dissemination of cases in which the FOBT result had misled physicians. For example, there had been a colonic neoplasm thought to have been missed due to a false-negative FOBT collected by digital rectal examination. In many cases, unnecessary upper endoscopies were performed in response to false-positive results.
While a simple, sensitive, and specific test for upper gastrointestinal bleeding would certainly be of value, for want of such test, we should not mischaracterize and rely upon a poor one. It is absolutely time to retire—as my institution has—the guaiac-based FOBT from emergency department and inpatient use.
The opinions expressed herein are solely those of Dr. Radecki and do not necessarily reflect those of his employer or academic affiliates.
References
- Carroll MR, Seaman HE, Halloran HP. Tests and investigations for colorectal cancer screening. Clinl Biochem. 2014;47(10-11):921-939.
- Lieberman DA, Rex DK, Winawer SJ, et al. Guidelines for colonoscopy surveillance after screening and polypectomy: a consensus update by the US Multi-Society Task Force on Colorectal Cancer. Gastroenterology. 2012;143(3):844-857.
- Tinmouth J, Lansdorp-Vogelaar I, Allison JE. Faecal immunochemical tests versus guaiac faecal occult blood tests: what clinicians and colorectal cancer screening programme organisers need to know. Gut. 2015;64(8):1327-1337.
- Nakama H, Fattah AS, Zhang B, et al. Digital rectal examination sampling of stool is less predictive of significant colorectal pathology than stool passed spontaneously. Eur J Gastroenterol Hepatol. 2000;12(11):1235-1238.
- Harewood GC, McConnell JP, Harrington JJ, et al. Detection of occult upper gastrointestinal tract bleeding: performance differences in fecal occult blood tests. Mayo Clin Proc. 2002;77(1):23-28.
- Chiang TH, Lee YC, Tu CH, et al. Performance of the immunochemical fecal occult blood test in predicting lesions in the lower gastrointestinal tract. CMAJ. 2011;183(13):1474-1481.
- Sharma VK, Komanduri S, Nayyar S, et al. An audit of the utility of in-patient fecal occult blood testing. Am J Gastroenterol. 2001;96(4):1256-1260.
- Ip S, Sokoro AA, Buchel A, et al. Use of fecal occult blood test in hospitalized patients: survey of physicians practicing in a large central Canadian health region and Canadian gastroenterologists. Can J Gastroenterol. 2013;27(12):711-716.
- Cleveland NJ, Yaron M, Ginde AA. The effect of removal of point-of-care fecal occult blood testing on performance of digital rectal examinations in the emergency department. Ann Emerg Med. 2010;56(2):135-141.
- Gupta A, Tang Z, Agrawal D. Eliminating in-hospital fecal occult blood testing: our experience with disinvestment. Am J Med. 2018;131(7):760-763.
Pages: 1 2 3 | Multi-Page
7 Responses to “It’s Time to Abandon Fecal Occult Blood Testing in the Emergency Department”
September 15, 2019
LarryPseudo science here. This test has saved me and my patients hundreds of times over the years. Really wish this article was better vetted before publication.
September 15, 2019
Chris WiesnerThe sensitivity of FOBT for malignancy (which is dependent on whether the malignancy is bleeding in addition to actual FOBT performance) is irrelevant for ED performance and value.
You misreported the study results on serial ingestion of blood and subsequent detection when you said “only one of 12 study subjects resulted a positive guaiac-based test”. Each of the 10 patients (not 12) had three serial stool samples. After ingesting 15 ml of blood the guaiac-based test was positive in at least one stool sample in six out of 10 subjects (60%). However, in only one patient was the guaiac-based test positive in all three of that patient’s stool samples. The study abstract doesn’t detail which samples were positive (it would be interesting to know if the timing mattered — i.e., were all the positives in samples collected later).
You wrote that one study “found profound mismatch between test results and follow-up endoscopy”. The study you cite (reference #7) evaluated the use of FOBT in hospital inpatients to detect colonic malignancies. Only 41% of patients with a positive FOBT were even referred to GI and colonoscopy to evaluate for malignancy was performed in 28%. How does a study that doesn’t even include upper endoscopy and is in an entirely irrelevant patient population have relevance to ED use? And saying “follow-up endoscopy” rather than “follow-up colonoscopy” is, bluntly, misleading.
My gastoenterologists generally ask about, are interested in, and act on guaiac test results in the appropriate clinical situation.
Certainly the performance characteristics as well as false positives and negatives are an issue. But no test is perfect, and the limited noninvasive surrogates — like the BUN/Cr ratio — surely aren’t better. Bringing up issues like not being able to withhold certain foods or the possibility of epistaxis being a confounder is silly. We are not screening every patient with a rectal exam and FOBT — the test should be used in appropriately selected patients in which case it can provide useful (if imperfect) clinical information.
FOBT testing can help distinguish between true melena and black stool caused by confounders like iron supplements. It can help confirm whether reported hematemesis at home was truly due to UGIB. It can infrequently be lifesaving in cases of unexplained syncope due to an occult GI bleed that later worsens. I’m aware of two malpractice suits that the MD settled because of syncope followed by death at home due to GI bleeding when a rectal exam and FOBT were not done.
September 15, 2019
Brian LevyOf course the gastroenterologists don’t like the test. They don’t like it because they would prefer that every single patient get sent for a colonoscopy that they will get to charge a fee for performing. The difficulty is that in many environments (where I practice it may take six months or a year for a patient to be seen for a scope, and the GI docs ho-hum in disbelief when they are told about cases, leaving the ER doc with the full burden of medicolegal risk. No, FOBT isn’t entirely sensitive, but in my experience, a positive is a genuine positive in almost every case.
September 30, 2019
Ryan Patrick RadeckiI appreciate the concerns of the various correspondents. It is indeed reasonable to raise concerns regarding the generalizability of the presented evidence with regard to FOBT and malignancy. However, to do so inadvertently supports the overall assertion of the piece regarding lack of evidence describing its validity as a test for Emergency Department use.
Need for clarifications are correctly illuminated in the citation regarding ingestion of blood, in which, ultimately, after ingesting 15mL of blood for three days, 6 of 10 subjects had at least one positive guaiac-based test. However, only 3 of 10 demonstrated two or more positive guaiac-based results, and only 1 of 10 demonstrated positive results on all three. Stated otherwise, it remains 20 of 30 samples demonstrated negative guaiac-based results while ingesting 15mL of blood daily.
Regarding the citation of the audit of inpatient use of FOBT, it is not misleading to characterize the results as a “profound mismatch”. Delving into the particulars of low-quality retrospective evidence provides easy targets for criticism, while this article simply observes the infrequency with which FOBT results were acted upon. Furthermore, when a subsequent gastrointestinal investigation was performed, it was more frequently an upper endoscopy than a colonoscopy, implying FOBT was not being performed primarily as colorectal cancer screening.
Anecdotal observations aside, it is clear relying upon a guaiac-based test will result in missed opportunities for investigation owing to its poor sensitivity. Despite its appealing convenience and generally accurate positive results, it should not be part of the routine evaluation for UGI bleeding.
February 29, 2020
TommyThe guaiac test is looking for the presence of peroxidase,
Peroxidase is found in many elements besides blood, making the test almost or completely useless in the clinical setting.
April 4, 2021
Kevin Bower, MD FACEPAlong these same lines of thinking, a normal H/H in an acute GI bleed would be thought of as unnecessary, and a low H/H could lead to an emergent endoscopy. The FOBT is only one piece of a puzzle that ER doctors use to decide if the patient has an acute medical condition. The value of a single FOBT has to be taken in context, and not as an absolute test that will determine further investigation.
June 15, 2023
Joseph WoodI appreciate Dr. Radecki’s thoughtful article. However, point-of-care testing looking for small amounts of occult GI blood are not really on point in emergency practice. EPs occasionally are asked to evaluate a chief complaint of black stools. As there are multiple causes of black stools, a point of care test that can reliably identify, or rule out, blood as the cause of the black stool is helpful and may help avoid some further testing. The real question then, is what’s the sensitivity and specificity of point-of-care testing of black stools for blood.