A rusty fish-scale crystal in an ulcer is a medication, not a curiosity. Swanson showed that sevelamer (Renvela, Renagel) looks broad, curved, and two-tone on H&E. Sodium polystyrene sulfonate (Kayexalate, SPS) looks narrower and violet, and published injury is often transmural necrosis. Cholestyramine is bright orange and usually sits in the lumen without injury. Put the resin on the requisition. Do not call every ulcer sevelamer colitis because a crystal is nearby.

Experienced teaching points

Clinical Pearls

Tell pathology the binder and the Kayexalate. Swanson had zero sevelamer histories on the requisitions.
Sevelamer: broad, irregular fish scales, eosinophilic to rusty brown on H&E, violet on PAS/D. Crushed tablets look the same.
SPS: narrower rectangular fish scales, violet on H&E, magenta on PAS/D. Published GI injury is often colonic transmural necrosis. Sorbitol is not required for the association.
Cholestyramine: bright orange, no internal fish scales, typically not associated with mucosal injury.
Crystals in injured mucosa are a clue, not automatic proof that the resin caused the hole. Stop or switch with nephrology when the mucosa is actually injured.

Clinical Bottom Line

Resin	H&E / PAS-D	What to do with it
Sevelamer (Renvela, Renagel)	Broad, curved, irregularly spaced fish scales. Eosinophilic to rusty brown on H&E. Violet on PAS/D. Non-polarizable. Crushed tablets match the biopsy.	CKD phosphate binder. Crystals often sit in eroded or ulcerated mucosa (colon more than esophagus or small bowel). Causality is suspected, not automatic. Discuss stopping or switching with nephrology if the mucosa is injured and no better cause is sitting there.
Sodium polystyrene sulfonate (Kayexalate, SPS)	Narrower, more rectangular fish scales. Violet on H&E. Magenta on PAS/D.	Hyperkalemia resin. Published GI injury is often colonic transmural necrosis, with or without sorbitol. Mortality in compiled case series is high because those are the cases that get written up. Stop SPS. Get surgery involved if the abdomen is surgical.
Cholestyramine	Bright orange on H&E. No internal fish scales. Gray or hot pink on PAS/D.	Usually a luminal passenger, not an injury pattern. Do not blame the ulcer on it.

Why does the ulcer have fish scales?

Because a non-absorbable resin went through the gut and crystallized. Swanson, Limketkai, Liu, and colleagues (Am J Surg Pathol 2013, PMID 24061514) first named sevelamer crystals in the GI tract: 15 specimens from 7 CKD patients, mostly colon, with ulceration, inflammatory polyps, ischemia, or necrosis in the background. Only one mucosa was normal. They crushed a Renvela tablet and processed it. The histology matched. None of the requisitions mentioned sevelamer. That is the practical failure: the med list never reached the microscope.

Do not treat that paper as proof that sevelamer always cuts mucosa. The authors said further study was needed for a causal role. Yuste and colleagues (Clin Kidney J 2017, PMID 28852493) added cases and a review: bleeding is the usual symptom, intestine is the usual site, diabetes and prior abdominal surgery show up often, and dose does not track lesion severity cleanly. Crystals can also sit in people with almost no symptoms. Association is real. Automatic 'sevelamer colitis' on every rusty crystal is not.

How do I tell sevelamer from SPS?

Color and scale geometry, then the indication.

Sevelamer: two-tone yellow-brown to rusty, broad curved scales, PAS/D violet. The patient is usually on a phosphate binder.
SPS: violet on H&E, magenta on PAS/D, narrower rectangular scales. The patient was treated for hyperkalemia. Harel's systematic review of published SPS GI events (Am J Med 2013, PMID 23321430) found the colon in about three quarters of cases and transmural necrosis in about two thirds of those reports, including preparations without sorbitol. That is selected sick patients, not a population incidence. It is still enough to stop treating SPS injury as a sorbitol-only myth.
Cholestyramine: orange, no mosaic, typically incidental.

Vegetable material, dystrophic calcium, and ink can fake a scale pattern. Ask for PAS/D. Do not argue from a single pale shard on one level.

What should I do on the list tonight?

Read the med list before you biopsy a dialysis-patient ulcer, polyp, or unexplained bleed. Sevelamer, SPS, calcium polystyrene sulfonate, and bile-acid binders all belong on the card you send with the jar.
Photograph. Biopsy the edge and the debris, not just pretty nearby mucosa.
If the abdomen is peritoneal, do not wait for PAS/D. SPS-associated necrosis is a surgical disease when it is transmural.
If pathology calls sevelamer crystals in injured mucosa, call nephrology the same day. Switch the binder if the gut is the problem and phosphorus can be managed another way. Do not promise that stopping the drug will heal a perforation that already happened.
If pathology calls SPS crystals, stop SPS even if someone swears the prep had no sorbitol.

This page is resin identification. NSAID ulcers, CMV in transplant, and ischemic colitis are separate differential lists. Ischemic colon without crystals lives on the colon-ischemia page.

Selected references

Last reviewed September 20, 2026. Written for clinicians biopsying ulcers in CKD patients, and for the pathologist who was not told the phosphate binder or the Kayexalate.

For your teaching file

Save this article as a PDF

Drop your email and we'll open a print-ready version you can save as a PDF — and you'll start getting our weekly GI endoscopy newsletter.

← All articles