FOBT and DRE’s in the ED (June 2018)
Articles:
AIR: The Pharmacology of Airway Management (Part 1)
Questions:
For a Review Article
- How has reading the article changed your practice?
- What information had you believed in previously that were debunked by reading this article?
- What new information did you learn from reading this article?
- What are current areas of uncertainty on this topic that can be potential areas for research?
For an Original Research Article
- What are some strengths of the study?
- What are the limitations?
- What are the main outcomes of the study?
- Does reading this article change your practice? If so, how?
21 comments
Tom raised the question of the utility of FOBT in the ED for workup of GIB. I think it’s very much related to whether DRE’s are useful as well because the two procedures are often done at once.
.
Considering I’m probably the only proponent of ED anoscopy in our department, I’m obviously biased towards one camp. But I’m open to reviewing the evidence. I thought it’d be very much worthwhile to review the Gupta article Tom sent me, as well as the DRE review article, to see if they answer our question. Love to hear your thoughts.
Kessler and Bauer
I knew it! Pretty useless most of the time. Not useful in generalized abdominal pain and appendicitis. Not very helpful in differentiating colonic from more proximal bleeding. Somewhat helpful in assessing anal tone (can squeeze vs manometry) but surely this patient is going to be complaining of other issues. How helpful this is in the setting of trauma….??? The one situation where I do find it useful is when we are investigating for occult sources of anemia in a patient who does not report a hx of dark stools.
There are good literature to show routine use of DRE in trauma adds nothing. Same with workup of undifferentiated abd pain/appendicitis.
.
But be careful when you say “useless most of the time”. Because the “most of the time” varies in patient, context, case, and pathology, that it’s difficult to sum up in a single study. And some pathologies (hemorrhoids, fissures, abscess, masses, surprising BRBPR or melena) are SO OBVIOUS if you only look, and when you don’t, it’s really inexcusable to send them out without even trying.
1. I agree with Alex re: role of anoscopy and advocated and Dana delivered the new scope with light so you can see. This is underutilized in LGIB and results in patients with IH being admitted (which unless portal htn, anticoagulation, severe bleeding, anemia etc not require admission. Have seen abd CT and other tests done where definite visualization of active bleeding IH (with carefully done anoscopy, straining) would be diagnostic. Does not obviate follow up and referral when concern for other proximal site or need for colon ca screening. Or seeing definite proximal bleeding, signs of colitis guide dispo. .
2. Quick comments on DRE review. Of historical interest one of our first residents Barb Jackson at Carney surgery resident refused to come to ED until she did DRE on child with clinical appendicitis (she had done surg some years prior to EM) and she said useless test come see the pt.
Although the article title suggests a comprehensive review of the role of DRE, it is not complete. For example, does not discuss role or controversy in male UIT. Some time ago had missed prostatic abscess with poor outcome- role of DRE in male UTI, differentiating acute prostatitis vs prostatic abscess. Many refs still recommend DRE in male UTI as part of PE, “carefully” not to induce bacteremia. see https://www.aafp.org/afp/2016/0115/p114.html
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5678513/
3. Similarly, review of DRE and FOB does not address the usual clinical scenario and evidence for/against – syncope/presyncope in absence of postural vs, anemia, melena to diagnose occult GIB as cause.
Just some quick thoughts
Agree with Jim completely! Appreciate the shout out for anoscopy haha. See my long comment below on my thoughts on the 2 articles and overall application to our practice.
Kessler article.
Although I appreciate that the authors aim to argue against the routine use of the DRE, I worry that about the loss of an intern-level procedure! I wonder if this article (2012) was written in a time when the “knee-jerk” integration of the DRE was more widespread than it is today. For the most part, this year when I do a DRE (or am asked by my Attending to do a DRE) it is to answer the question of anal tone (spinal injury) or to look for alternative explanations (marked hemorrhoids). They discuss the notion of right-sided rectal tenderness in the appendicitis work up, which no one has mentioned to me this year (nor would I feel adept at identifying where in the rectum is most uncomfortable). I worry about the use of DRE-acquired FOBT in the evaluation of neoplasm-based bleeding (which they point out has lower sensitivity, specificity, positivity rate, and PPV), especially at BMC where many of our pts do not have access to screening colonoscopies…hence may represent an important source of bleeding. Having just come off Peds, I am pleased to say that the PEM attendings seem to agree with the article’s assessment that the DRE is potentially too traumatic to be clinically useful.
I look forward to reading the more updated article, and discussing with Steve McGuire (the biggest opponent I have found to date).
Gupta article.
I love the way that they did their intervention – gathering the stakeholders, consolidating the GI faculty, an educational intervention and then finally restriction. I also appreciate their discussion of disinvestment. They make a great point about doing tests because another team will want them. On numerous occasions I have admitted pts and the receiving team will ask about FOBT. Although I now feel increasingly convinced that I don’t need to be doing these tests, I can relate to the feeling of obligation that you have to perform these tests before bringing in the pt (ironically if, as the article points out, the inpt team doesn’t even want or need those tests). I think I will have to do some more reading into the data that they cite before I feel comfortable routinely pushing back on this test.
It’s important to separate out the FOBT and the DRE. Which one do you think is not useful?
.
I think the gupta article makes a case of eliminating FOBT but say nothing of DRE. Be careful of extrapolating limiting FOBT to DRE because of the hassle bias. After all, the two procedures often come together. It’s the finger in the bottom that gets tiring sometimes.
.
See my long comment below.
Kessler et al.
Like Andy, I am increasingly worried that future generations of residents will lose the ability to turn to their interns and ask them if they performed a DRE and give them judging looks if they did not. At least there is still manual disimpaction. From reading this article, I learned that the DRE was or still is being used to assess for appendicitis. This was the first I heard of this, and given that it seems to correlate poorly to appendicitis, I will not make it a part of my practice. I agree that DREs may be more traumatic in children and adolescents than some of our older patients. In terms of anal tone, my assessments are mostly 3 speed- they have it, they don’t have it, and maybe it is decreased. I think for these type of assessments, a DRE may give me enough information without being very specific. Unless there is frank melena or bright red blood on DRE, it is difficult to interpret the significance of guaiac positive stools, as evidenced by this article. I still think it is worth looking for hemorrhoids or fissures during an exam, especially because a small amount of rectal bleeding can look significant on toilet paper and in the toilet bowl. In a sick or potentially unstable patient with a reported history of bloody stools, I think an inspection of the stool is important and a DRE can give some information about whether this is likely upper GI or lower GI bleeding.
Very thoughtful analysis
Gupta et al.
I really enjoyed reading this paper. While the FOBT discussion was important, I found the discussion of the challenges of de-adoption and the steps the team took to create trust and buy-in to be the most valuable. A model like this can and should be applied to other tests/interventions/protocols that are no longer evidence-based or deemed useful. It is difficult to create a change in culture without systematically recruiting stakeholders, providing education, and reinforcing your message. In our hospital, when I admit any patient with a suspected GI bleed, I am inevitably asked what the guaiac results were. This is because this is the way things have been done (and many of us have been taught to do DREs in medical school). From the article, I feel reassured that a DRE will likely not be a useful tool in evaluating anemia, especially since the majority of patients without frank GI bleeding will not get a colonoscopy/endoscopy as an inpatient. Also worth noting the importance of delineating between types of anemia before going down that route. I would like to learn more about the sensitivity and specificity of outpatient FOBT- for example, the ones for colon cancer screening currently in use. Also, I would like to learn more about the correlation between positive FOBTs and incidence of clinical significant findings on colonoscopy.
Kessler et al.
I’m glad this is generating a discussion. I have never done a rectal exam when evaluating a patient for appendicitis and I never will. Aside from making the patient less likely to present to the doctor when they are unwell there is not benefit to performing it to aid in diagnosis of non-specific abdominal pain.
I also wonder about the utility of performing this exam in the trauma room. Every patient that I have encountered who was later found to have cauda equina has had normal or equivocal anal tone. I feel that the presence of anal tone as evaluated by MDs should not give us a false sense of reassurance when considering this diagnosis.
I also worry that as fewer and fewer of these exams are performed these days, the sensitivity of MDs to detect reduced anal tone will also diminish. Our “muscle memory” eventually wanes without repetition and I fear that if we’re still relying on this outdated exam we may begin to miss more of these potentially catastrophic diagnoses.
sorry that was meant to be my response to the Gupta et al paper.
Kessler et al.
This review doesn’t really change things for me. I have never uses the DRE as part of my standard examination for undifferentiated abdominal pain and I’m not about to start. For occult bleeding and anal tone, I’m equally unconvinced, as so much of the important information that leads you towards observation in hospital or advanced imaging is obtained on history. Subsequently, I can’t imagine not obtaining advanced imaging studies in patient with certain high risk injuries/complaints, even with a benign DRE. Finally, since when is the DRE for appendicitis even a thing? This might be the first time I’ve heard that since starting medical school.
Overall, the review seems limited in its focus. Frankly, a negative FOBT won’t stop me from doing a q4-6h HCT in the context of a symptomatic GIB by history, so why do it. We shouldn’t be reassured by the positivity or negativity of a test with multiple false positives.
I would be interested in seeing information on more objectively useful aspects of the DRE than those posed above… I would like to see information on how well ED docs can predict prostatitis/prostatic abscesses, for instance.
Gupta et al.
Strengths: Large sample size (32k), longitudinal
Limitations: Single institution,
Outcomes: Reduced cost, institutional behavior change, clarified indications for endoscopy/colonoscopy evaluation, patient safety
Does it change my practice? I think so, or at least I want it to. If a patient comes in with a history of days – weeks of dark stools, it is possible they have an occult GI bleed. If the Hgb is stable now and in 6 hours, I’m not sure what a FOBT actually does for me. If a patient has a massive GI bleed, what does my testing anything do for me? If a patient is hypotensive with a GIB, haven’t I already called GI when the patient rolls in, or at least considered angiography and IR intervention if I presume lower?
Gupta et al.
Starts by saying FOBT is used to cancer screening, and isn’t good for it. WHO CARES? We don’t use it for that reason in the ED.
.
Obvious endoscopy has “revolutionized our approach”, but we don’t do that in EM. And our GI department is less aggressive about scoping quickly than other places I’ve worked. So FOBT is the best we have, is it any good?
.
The sensitivity and specificity for FOBT is suboptimal. I concede that adding FOBT is probably of less utility. But that DOESN’T necessarily mean we skip the whole DRE.
.
Kesseler et al.
.
It doesn’t answer our question of whether we should do it because it looked for only 3 reasons in which we would consider it. There are potentially many more (fissures, abscesses, hemorrhoids vs. GIB? etc?)
.
We know DRE is useless for dx of abdominal pain. We use it for rectal tone with some success. And who cares about “colonic lesions” aka cancer in terms of FOBT, we care about GIB.
.
Just the cases off the top of my head in which DRE or FOBT changed my management:
– Sent home elderly with coumadin with reported BRBPR with neg CTA and labs with completely brown stool guaiac neg. I f/u ed up on the patient and she did well. MANY would admit any elderly pt with BRBPR, much less on coumadin.
– Middle aged woman with BRBPR. DRE showed red-purple material, guaiac neg. Turned out she ate beets. Maybe we should just ask about beets, or every single food/drug/vitamin that turns poop red… but that imo takes longer than a DRE FOBT
– HOW MANY PATIENTS SAY THEIR STOOL IS DARK? you rectalize, find out it’s not, and guaiac neg, and discharge them? If we don’t look, should we admit ALL OF THEM for endoscopy? or apply UGIB decision rules for all of them maybe?
– Sick trauma 1 patient, don’t look good. Not sure what is going on, maybe tachy or hypotensive. A quick DRE can rule out large GIB quickly.
– How often you hear about pts with BRBPR, you do a rectal and see a bleeding hemorrhoid and send them on their way. Should we admit all for colonoscopy? Because there are no decision rules for admission vs. discharge of LGIB.
– On a similar note, we often defer pelvic exams too. I think for specific indications like VB in pregnancy who are stable, not bleeding a lot, Judy Linden has shown that perhaps it could be deferred. However, cervical cancer / trauma etc can be missed when you don’t do pelvics in non preg patients as I have seen. These are RARE, but if you didn’t even look, I can’t imagine that holding up well in a case review.
.
The issue is that these cases are so diverse that no study can pick it up. The two studies report low utility in specific situations, but there are many more in which we have no data. And that’s when experience comes in.
.
The Gupta study doesn’t answer our question because it looks at FOBT ONLY, which is really a minor step after doing the DRE. THe labor and hassle bias comes from the DRE, not the FOBT.
.
We really need a RCT looking at people who get DRE and those who don’t, and see a list of the cases in which management were changes and what happened, in additional the % of management change.
.
Plus, why would we change management if we didn’t take the opportunity to obtain a potentially useful data point?
.
We won’t know unless we look! You DON’T need to DRE for abd pain/appendicitis, and probably in routine trauma because there are specific studies to show low utility. But the issue is too complex and the use too diverse to be eliminated. Maybe you don’t need the FOBT part. But DRE is still crucial part of my practice… along with anoscopy :)
Thoughts on the Gupta article (prior to reading any other comments): For the most part, I agree with the article about in-hospital FOBT. I’ve seen a lot of clinicians on the inpatient services at BMC use it as a test for GI bleeding to assess for drops in hematocrit as part of an anemia work up, and very rarely are the MDs doing an actual rectal exam (that I’ve seen). If they have a suspicion for a bleed, DRE would definitely be indicated and I think a FOBT would be pretty useless, considering the sensitivity of 1 FOBT is quoted to be as low as 30% (per uptodate, my access to pubmed through BU has lapsed somehow). In my mind, in order for GI bleeding to meet criteria for admission or cause a significant hemoglobin drop, I would need to see frank blood or melena on my DRE or in the stool.
I could see clinicians arguing that a negative FOBT would be helpful in the inpatient work up of a microcytic anemia, but again the sensitivity is poor, and the next test would probably be a colonoscopy anyway if they didn’t have active bleeding or frank blood.
I can’t think of a situation in the ED where a positive or negative FOBT would change my management. I would more likely decide based off my DRE, vitals, and clinical picture.
The Kessler article was a little ridiculous. I don’t know who routinely does rectal exams on people with undifferentiated abdominal pain, and who is submitting young children to the physical and psychological trauma of having a finger stuck up their rectum in the area of modern imaging. I feel like this was an answer to a question that wasn’t being asked.
I think the only really useful part of this study would have been an analysis of DRE results that changed management in people with abdominal pain that were DRE positive for frank blood or melena. That didn’t appear to be a part of the data set, and they mention it changing management in 7% of patients. Looking through the table, they actually excluded patients in the article review that had reported GI bleeding. It would have been interesting to see how many of those patients had blood or not, but again I don’t think that particular point is controversial.
I do think there is a role for the DRE in neurological or trauma cases, and it has a pretty good sensitivity for testing tone according to this study. Again, I don’t think that’s a controversial topic, but I could be wrong, I’m only an intern.
The intern is spot on!
– DRE is probably a very useful exam but I fear that many of us new trainees don’t know how to do it adequately and don’t know how to interpret what we find. Speaking just for myself I received instruction on how to do a DRE as 1st year med student and the session was more focused on how to make the pt comfortable and not on how to gather data for clinical use. Since then I have only ever used the exam to answer two questions, do I see (not feel) hemorrhoids and is the FOBT positive/negative.
– RE Kessler’s article, I appreciate the authors claim that FOBT may be clinically futile can increase false positive rate and lead to more testing. I also found it interesting that from the 3 articles comparing methods of stool collections there was essentially no difference in finding colon pathology, meaning we don’t always have to collect stool via DRE if the question is blood or no blood. My other take aways from Kessler is DRE is useful to assess rectal tone but not to diagnose emergent abdominal pathologies.
– Re Gupta article, like the Kessler’s article, argues for the poor utility of the FOBT as it doesn’t change management b/c it its negative w/t positive hx pt still gets endoscopic investigations. So if we are to change our practice, the change needs to come from GI (the colon experts) and not us because it is the in patient teams that ask us time and again “what was guaiac?”.
The Kessler article seemed like a bit of a surprise to me. Though my time in the ED is minimal right now (1 month into intern year), I honestly haven’t even once considered doing a DRE for my abdominal pain patients. If they are female I’m more likely to do a pelvic exam. The article made me feel better about this, especially with the finding that it altered management 7% of the time and diagnostic harm was just as likely as diagnostic help.
It seems in terms of method of obtaining, DRE might be better than sampling spontaneously passed stool? But it seems there isn’t sufficient data to make this a hard and fast rule and it depends on the suspicion or complaint. I haven’t done any traumas yet, but now I’ll take note that asking the patient to squeeze seems to tell us more than just resting tone.
The Gupta article makes me question if I’ll really ever get FOBT, but after seeing Alex’s comment above, I see there are definitely some uses that I haven’t run into personally yet, but anticipate I will eventually.