With a little help from my friends: What gastrointestinal pathologists need to know from gastroenterologists

Pathologists are physicians!

I know, it can be easy to forget this. Pathologists aren’t always that visible. Many do not interact with patients, and some only infrequently interact with other physicians. You may not even know where the pathology department at your institution is, and if you do, you may not have visited it. It can be easy to visualize pathology as a “black box”: specimen goes in, diagnosis comes out, simple as that.

None of this changes the fact that pathologists are indeed physicians. Pathologists went to medical school. They understand disease and diagnosis just as much as other physicians. Sending them a specimen is a consultation. Much like an internist may have referred a patient to you, you are essentially "referring” the patient to a pathologist.

With this in mind, let’s consider how gastroenterologists and gastrointestinal pathologists can improve communication for the good of our patients. If an internist sent you a patient and provided absolutely no information, it would be frustrating. If an internist sent many patients to you and simply said “Rule out H. pylori” for every single patient, that also leaves room for improvement. Communication between pathologists and clinicians has not received very much attention,¹ but we can use some basic concepts as a start to improving dialogue.

Provide key clinical information

As I already alluded to, communicating what you actually need to know about a patient is very helpful. Writing “rule out celiac” for every single duodenal biopsy is not helpful – trust me, we are already keeping celiac in mind. But if you have a patient who genuinely may have celiac, stating it just that one time alerts the pathologist that this particular patient is at high risk for the disease, and perhaps extra scrutiny is warranted. This also applies to H. pylori (we’re already looking for the organisms) and Barrett’s esophagus (we’re already looking for the goblet cells).

Similarly, writing a note with the specimens sent to pathology allows you to communicate important information. Are you worried about graft-versus-host disease? Let the pathologist know the patient had a transplant. Concerned about immune checkpoint inhibitor colitis? Let the pathologist know what medications the patient is taking. Performing surveillance on a rectal scar where a neuroendocrine tumor was previously found? Please write “h/o rectal NET, surveillance.” While the pathologist often (but not always) has access to the patient’s chart, a simple guiding note will ensure the pathologist understands what you are thinking and what truly concerns you about each of your shared patients.

Obtain adequate tissue

Appropriate sampling is key. For mucosal biopsies, more tissue is usually better. Obviously, tissue is sometimes hard to obtain and it’s important not to cause too much mucosal damage, but if a pathologist gets a biopsy consisting of half a millimeter of tissue, it’s going to be harder to interpret than one with several millimeters of tissue. The disease process in question might also go unsampled. In a similar vein – if you suspect a submucosal lesion, performing a biopsy just of the overlying mucosa is unlikely to provide the answer you seek.

Gastric polyps sometimes arise due to various forms of gastritis,² so if you sample a gastric polyp, please also send a separate biopsy of the background mucosa. Eosinophilic esophagitis and gastroesophageal reflux disease can look similar microscopically, but the former usually involves the entire esophagus and the latter usually doesn’t, so if both are in your differential, please send samples from the proximal, mid, and distal esophagus.³ Finally, while some specimens require cautery to obtain, the resultant tissue damage can make specimens hard to interpret. Whenever possible, be gentle!

Separate and label specimens when appropriate

On the subject of specimen handling – If you take several specimens and it’s important to know what changes are present in each specific piece of tissue, please submit them in separate jars. Placing all the tissue in the same jar may make it impossible to tell what tissue came from where.

One example is colon polyps, which can fragment during the journey from a patient’s colon to a glass slide. If you need a specific diagnosis for each polyp, they should remain separate. Otherwise, the pathologist may have “three polyps” but 10 tissue fragments, 4 of which are adenomatous and 6 of which are not, making it impossible to tell how many of those endoscopic polyps were truly adenomas (One? Two? All three?). This concept also applies to esophageal biopsies – submitting those distal, mid, and proximal biopsies in the same jar makes it hard to tell the true distribution of eosinophilic disease. Another example is gastric biopsies taken via the Sydney protocol. If gastric intestinal metaplasia is a concern, submitting each sample in a separate jar makes it easier to tell exactly how widespread the metaplasia should be.

On the topic of gastric biopsies, please label jars with the sample site, as it can be important for the pathologist to know where exactly the tissue originated. This is particularly useful in atrophic gastritis, which makes the oxyntic mucosa resemble antral mucosa. As a result, a biopsy just labeled “stomach” may lead to avoidable errors, such as an antralized portion of fundic mucosa being mistaken for a simply inflamed piece of antral mucosa.

Understand the limitations of pathology

Please be aware when the pathologist is only going to be able to help you so much. This often comes up in the context of biopsies for colitis. Unfortunately, most colitides (including inflammatory bowel disease) have overlapping features under the microscope.⁴ We can give a differential, but the biopsy is unlikely to provide an answer in the absence of clinical context. There are also diseases that simply do not have microscopic correlates. One example is mast cell activation syndrome. The mast cells in the gastrointestinal tract may be behaving pathologically, but they have the same appearance, quantity, and distribution under the microscope as in people without the syndrome.⁵ As much as pathologists would like to provide a definitive answer, performing histochemical or immunohistochemical stains to highlight them does not change this fact. This also applies when a case is challenging microscopically. Maybe it’s due to that cautery, or the tissue got contorted during processing, or the disease process was under-sampled. For any number of reasons, a biopsy can be difficult to interpret. Pathologists don’t want to give an equivocal answer if they can avoid it. If cancer is there, we want to say “positive for malignancy” rather than “atypical and suspicious,” but there are times where the latter simply can’t be avoided. This may necessitate a conversation and/or repeat sampling.

Contact us
Along those lines, I have saved the most crucial tip for last – feel free to get in touch! We also want what’s best for our patients and we are happy to discuss an unusual or challenging scenario with you. Send us an email, give us a call. Even better, set up a time to come by the pathology department, and we can show you the microscope slides and discuss everything together, in person. This also applies when doubt or uncertainty may exist upon reviewing a pathology report. Studies have shown that pathologists and clinicians often communicate differently, and this can lead to different interpretations of pathology reports.^6-8 Most pathologists do try very hard to make their findings and their meaning clear in the reports they issue, but miscommunication can still happen. The best way to clear that up is to get in touch with your colleague – the pathologist.

There’s a lot that pathologists can do, but there’s also a lot we are unable to do, especially without the clinical context in mind. The clearer a picture we have of the patient’s situation and the gastroenterologist’s needs, the better our interpretation and report will be. As with any other physician you are consulting regarding a patient, it’s always helpful when key information is provided and extraneous information is included. Also, like with other physicians, please reach out any time we can do more to help!

Dr. Gonzalez is Professor of Pathology and Laboratory Medicine at Emory University Hospital, Atlanta, Georgia. He has no relevant conflicts. His X account is @RaulSGonzalezMD.

References

1. Lehr HA, Bosman FT. Communication skills in diagnostic pathology. Virchows Arch. 2016 Jan. doi: 10.1007/s00428-015-1848-y.

2. Vos S, et al. Gastric Epithelial Polyps: When to Ponder, When to Panic. Surg Pathol Clin. 2020 Sep. doi: 10.1016/j.path.2020.05.004.

3. Odze RD. Pathology of eosinophilic esophagitis: what the clinician needs to know. Am J Gastroenterol. 2009 Feb. doi: 10.1038/ajg.2008.40.

4. Cerilli LA, Greenson JK. The differential diagnosis of colitis in endoscopic biopsy specimens: a review article. Arch Pathol Lab Med. 2012 Aug. doi: 10.5858/arpa.2012-0205-RA. PMID: 22849731.

5. Panarelli NC, et al. What Is the Value of Counting Mast Cells in Gastrointestinal Mucosal Biopsies? Mod Pathol. 2023 Feb. doi: 10.1016/j.modpat.2022.100005.

6. Powsner SM, et al. Clinicians are from Mars and pathologists are from Venus. Arch Pathol Lab Med. 2000 Jul. doi: 10.5858/2000-124-1040-CAFMAP.

7. Mirham L, et al. Addressing the Diagnostic Miscommunication in Pathology. Am J Clin Pathol. 2021 Sep. doi: 10.1093/ajcp/aqab014.

8. Battah K, et al. Discrepancies between pathologists and clinicians in interpretation of diagnostic uncertainty in surgical pathology reports. Sci Rep. 2025 Jul. doi: 10.1038/s41598-025-11032-8.