GI Infections: GI Infections: Opportunists and Norovirus

0:00Welcome to Board Pearls. This is episode five of five, the last episode of the GI Infections chapter, in the Special Populations and Acute or Supportive Care module. This episode is the infections a weakened immune system lets in, and then the virus that spares no one. The first patient is a man with steroid-refractory ulcerative colitis. His case shows who gets cytomegalovirus colitis, why the biopsy site decides whether it's found, why a blood test isn't enough, and how it's treated, and a kidney recipient then shows how a resistance mutation names the next drug. The second patient is a man with advanced human immunodeficiency virus, HIV, infection. His case shows how Mycobacterium avium complex, MAC, is told apart from Whipple disease on the same biopsy, why MAC needs several drugs and for how long, and how the other opportunists on his list, microsporidia, Cyclospora, and Cystoisospora, are found and treated. The episode ends with an outbreak on a care-home floor, which shows why alcohol gel won't stop norovirus.

1:04Here's the first patient. A thirty-four-year-old man with ulcerative colitis, who weighs seventy kilograms, was admitted with acute severe colitis: ten bloody stools a day, fever, and a raised C-reactive protein, CRP. On day five of intravenous corticosteroids he isn't improving, and his team is discussing a biologic or a calcineurin inhibitor. Flexible sigmoidoscopy two days ago showed deep punched-out ulcers, and biopsies from the ulcer edges showed only active colitis.

1:34His steroid-refractory colitis with deep punched-out ulcers raises cytomegalovirus, CMV, colitis, reactivation of the virus in his inflamed colon. CMV is a herpesvirus that establishes lifelong latency after the first infection and reactivates under immunosuppression. CMV colitis emerges in solid organ transplant recipients, especially one to six months after transplant, when immunosuppression is most intense; in HIV infection with a CD4 count below one hundred, with disseminated disease mostly below fifty; in inflammatory bowel disease treated with steroids or biologics; and with chemotherapy or other long-term immunosuppression. The virus is already present in most people, and his high-dose steroids are the kind of suppression that releases it.

2:24His steroid failure is the setting where finding CMV matters most. CMV colitis causes bloody diarrhea, fever, abdominal pain, and weight loss, with deep punched-out or serpiginous ulcers, often in the right colon and terminal ileum. In acute severe ulcerative colitis that fails intravenous steroids, an adequate biopsy finds CMV in about a third of patients, and finding it changes the next step from escalating immunosuppression to antiviral therapy, with immunosuppression reduced where feasible. A biologic on top of undiagnosed CMV colitis would suppress the immune control he needs to clear it. Chapter fourteen covers the rest of steroid-refractory colitis.

3:09His first biopsies came from the wrong place. The diagnosis needs biopsies from the ulcer base, the floor of the ulcer, where CMV-infected fibroblasts and capillary endothelial cells sit in the submucosa. Biopsies from the ulcer edge or from normal-looking mucosa give false negatives, because the virus isn't concentrated there. So his edge biopsies can't exclude CMV, and he needs repeat sampling from the ulcer floor.

3:36His new biopsies need the right stains, written on the request. Standard hematoxylin and eosin shows large basophilic intranuclear inclusions with a clear halo, the owl's-eye Cowdry type B inclusion, along with enlarged cells and cytoplasmic inclusions. But it detects fewer than half of colonic cases, because inclusions can be sparse. CMV immunohistochemistry substantially improves sensitivity and is the recommended addition whenever CMV is in the differential, and tissue polymerase chain reaction, PCR, is a further option in specialized centers. His repeat ulcer-base biopsies show scattered owl's-eye inclusions and strongly positive immunohistochemistry. He has CMV colitis.

4:21His blood test helps but couldn't have decided it. Blood CMV PCR is the right first systemic test and supports systemic disease, but a positive result doesn't prove tissue invasion, and a negative result doesn't exclude colitis confined to the tissue. The tissue answers whether he has colitis, and the blood answers whether he has systemic disease.

4:41His treatment is intravenous ganciclovir, five milligrams per kilogram every twelve hours, adjusted for kidney function, for fourteen to twenty-one days of induction, then oral valganciclovir, nine hundred milligrams twice daily, until clinical and virologic resolution. At seventy kilograms, that's three hundred fifty milligrams every twelve hours. His immunosuppression is reduced as far as his colitis allows, because restored CD8 T-cell control is what clears CMV for good: the drug suppresses the virus, and the immune system clears it. Over a week, his stools and bleeding improve, and he moves to valganciclovir until both his symptoms and the virus resolve.

5:21Treatment can fail when the virus becomes resistant. A fifty-eight-year-old woman three months after a kidney transplant, inside the highest-risk window, has bloody diarrhea, fever, and deep right-colon ulcers, with CMV colitis confirmed on ulcer-base immunohistochemistry. After two weeks of intravenous ganciclovir, her viral load keeps rising and her symptoms persist, and genotyping shows a UL97 mutation.

5:47Her mutation predicts which drugs still work, because each drug depends on a different viral enzyme. Ganciclovir is a prodrug that must be phosphorylated by the viral UL97 kinase to become active, so a UL97 mutation blocks its activation. Foscarnet, a pyrophosphate analog, inhibits the viral UL54 DNA polymerase directly without needing activation, so it stays active. Maribavir, four hundred milligrams by mouth twice daily, inhibits the UL97 kinase itself rather than needing it for activation, so it also works in UL97-resistant disease, and it causes less myelosuppression than ganciclovir because it doesn't depend on phosphorylation in host cells. So she's treated with foscarnet or maribavir, with her immunosuppression reduced as far as her graft allows.

6:38A mutation in UL54 leaves fewer options. It alters the polymerase itself, causing resistance to both ganciclovir and foscarnet. A kinase mutation spares the polymerase drug, and a polymerase mutation hits both. Letermovir, a CMV terminase inhibitor, isn't an option for her, because it's used only for prophylaxis, in hematopoietic cell transplant and high-risk kidney transplant, and not to treat active or resistant CMV. Chapter thirty-four covers transplant prophylaxis and donor-recipient mismatch, and chapter five covers CMV esophagitis alongside herpes simplex and Candida.

7:16Now the second patient. A forty-one-year-old man with HIV stopped antiretroviral therapy two years ago and has a CD4 count of thirty. For two months he's had fever, night sweats, weight loss, abdominal pain, and chronic diarrhea, with hepatosplenomegaly and enlarged abdominal lymph nodes on computed tomography, CT. Upper endoscopy shows pale yellow duodenal nodules with blunted villi, and biopsy shows lamina propria packed with periodic acid-Schiff, PAS, positive macrophages.

7:49His PAS-positive macrophages mean either disseminated MAC infection or Whipple disease, and his CD4 count of thirty is what puts MAC first. Advanced immunocompromise, from HIV with a low CD4 count, transplant, or chronic steroids, lets in organisms a normal immune system contains, including MAC, microsporidia, and Cyclospora, and for these, the choice of stain decides whether the diagnosis is made, so the pathologist is told what's suspected. Disseminated MAC occurs below a CD4 count of fifty, with fever, night sweats, weight loss, abdominal pain, chronic diarrhea, hepatosplenomegaly, lymphadenopathy, and pale yellow duodenal nodules with blunted villi and macrophage-packed lamina propria, which is his picture.

8:36An acid-fast stain separates the two. MAC-laden macrophages are PAS-positive and also acid-fast on Ziehl-Neelsen or Kinyoun staining, while Whipple macrophages are PAS-positive but not acid-fast. His acid-fast stain is positive, so he has MAC. Stopping at the PAS stain would have led to the wrong treatment.

8:58His MAC needs several drugs at once. The regimen is a macrolide, clarithromycin five hundred milligrams twice daily or azithromycin five hundred to six hundred milligrams daily, plus ethambutol fifteen milligrams per kilogram daily, with rifabutin three hundred milligrams daily added in advanced disease or when macrolide resistance is suspected. MAC lives inside macrophages, where macrolides concentrate, so a macrolide is the core of treatment, but a macrolide alone quickly selects resistance among the organisms that survive. Ethambutol, which blocks synthesis of the mycobacterial cell-wall arabinogalactan, kills those survivors before resistance develops, and rifabutin, an RNA polymerase inhibitor, adds a third mechanism.

9:43His treatment continues until his immune system recovers, defined as a CD4 count above one hundred for at least six months, because killing the last intracellular organisms requires T-cell activation of macrophages that antibiotics can't replace. He restarts antiretroviral therapy with his MAC regimen, and his antibiotics stop on that CD4 criterion, not when his fever settles. Chapter thirty-four covers why primary MAC prophylaxis is no longer routine once effective antiretroviral therapy starts.

10:12His CD4 count also leaves him open to microsporidia, which would explain chronic watery diarrhea with weight loss when a biopsy is unrevealing. They infect patients with HIV at low CD4 counts and transplant recipients. Enterocytozoon bieneusi is the most common species, and Encephalitozoon intestinalis the other. Their spores are only one to two micrometers, so diagnosis needs a chromotrope or modified trichrome stool stain, the modified Weber stain, or PCR, which a routine stool exam won't provide unless it's asked for.

10:45The two species respond to different drugs because of their tubulin. Albendazole works by binding beta-tubulin and blocking microtubule formation. Encephalitozoon intestinalis has a tubulin albendazole binds well, and it's treated with albendazole four hundred milligrams twice daily for twenty-one days. Enterocytozoon bieneusi has a tubulin variant albendazole binds poorly, so fumagillin, which inhibits methionine aminopeptidase two, is its alternative, though it has limited United States availability. In HIV, antiretroviral therapy is the main treatment, because immune recovery clears the infection where the antiparasitic drugs fall short.

11:22Two protozoa on his list respond to trimethoprim-sulfamethoxazole. The first is Cyclospora, a coccidian that causes self-limited diarrhea in immunocompetent hosts and prolonged diarrhea in HIV. It classically comes from imported summer produce, raspberries, basil, cilantro, or snow peas, or from travel to Nepal or Peru, so a traveler home from Nepal with weeks of watery diarrhea is a typical case. Modified acid-fast stool staining shows eight-to-ten-micrometer oocysts, confirmed by their autofluorescence under ultraviolet light. Treatment is one double-strength trimethoprim-sulfamethoxazole tablet twice daily for seven to ten days, longer in HIV. Cyclospora must make its own folate and has no salvage pathway, so blocking two steps of folate synthesis kills it. Nitazoxanide and metronidazole act on anaerobic protozoa and don't work against coccidia.

12:19The second is Cystoisospora belli, the other coccidian, which causes chronic diarrhea mostly in immunosuppressed patients but also as chronic travelers' diarrhea in healthy hosts, and is most common in Haiti and other resource-limited regions. Modified acid-fast staining shows large oocysts, ten by twenty micrometers, and it's the only protozoan that causes peripheral eosinophilia, with Charcot-Leyden crystals in the stool, so eosinophilia with chronic diarrhea in a patient from Haiti prompts a modified acid-fast stain. Treatment is one double-strength trimethoprim-sulfamethoxazole tablet twice daily for ten days, longer in HIV, followed by secondary prophylaxis below a CD4 count of two hundred, the level at which CD4 control of intracellular protozoa fails, the same threshold as for Pneumocystis. With a CD4 count of thirty, he'd need prophylaxis after treatment until his count recovered above two hundred.

13:14The last infection doesn't depend on the immune system at all. On the long-term care floor of the woman from episode one, a dozen residents and several staff develop abrupt vomiting and watery non-bloody diarrhea with low-grade fever one winter weekend, a day or two after a staff member worked while sick, and most recover within two days.

13:35They have norovirus gastroenteritis. Norovirus is the leading cause of community and outbreak gastroenteritis in the United States and worldwide, year-round with a cold-weather peak. It causes abrupt vomiting and watery diarrhea twelve to forty-eight hours after exposure, lasting one to two days, with vomiting more prominent than diarrhea. Humans are the only host, and it spreads through ill or recently ill food handlers and through aerosolized vomitus.

14:00It spreads so well because of four features: an infectious dose of about twenty particles; shedding that starts before symptoms and continues for weeks afterward; persistence on surfaces, in food, and in water across a wide range of temperatures; and many strains without cross-protective immunity, so people can be reinfected. Classic settings are cruise ships, daycare centers, nursing homes, hospitals, cafeterias, and raw oysters. Prominent vomiting, little fever, and a short course point to norovirus, while bloody or mucoid stool, higher fever, and a longer course point to invasive bacteria, as episode three explains.

14:40The residents need supportive care, and the floor needs outbreak control. Treatment is oral rehydration, intravenous fluids for severe dehydration, and a bland diet as tolerated, with no antiviral. Outbreak control relies on hand-washing with soap and water, because alcohol sanitizers work less well against norovirus than against most viruses, and on bleach-based cleaning, because the same physical persistence that drives outbreaks defeats alcohol. Her floor's staff wash with soap and water and clean with bleach, and her own diarrhea is watched beyond a couple of days, because C. difficile remains possible in her.

15:21The other viral causes are found on multiplex stool PCR. Routine rotavirus vaccination has reduced severe rotavirus disease by about eighty-five percent in vaccinated populations, though rotavirus still matters in unvaccinated and immunocompromised people. Adenovirus, sapovirus, and astrovirus complete the viral differential, and they're self-limited and managed supportively.

15:45Here are the traps the boards set in these patients. Escalating immunosuppression in steroid-refractory ulcerative colitis without looking for CMV. Biopsying the ulcer edge instead of the base. Relying on standard staining without immunohistochemistry. Excluding CMV colitis on a negative blood PCR, or diagnosing it on a positive blood PCR alone. Giving ganciclovir without reducing immunosuppression, or without adjusting for kidney function. Continuing ganciclovir in UL97 resistance instead of switching to foscarnet or maribavir. Using foscarnet for a UL54 mutation. And treating active CMV with letermovir.

16:25For the other opportunists, the traps are these. Calling PAS-positive macrophages Whipple disease without an acid-fast stain. Treating MAC with a macrolide alone, or stopping before the CD4 count has been above one hundred for six months. Missing microsporidia without a trichrome stain, or using albendazole for Enterocytozoon. Forgetting that antiretroviral therapy is the main treatment in HIV. Using metronidazole or nitazoxanide for Cyclospora or Cystoisospora. Forgetting Cystoisospora prophylaxis below a CD4 count of two hundred. And relying on alcohol gel in a norovirus outbreak, or treating norovirus with antibiotics.

17:07So, start to finish. CMV colitis is suspected early after transplant, in HIV below a CD4 count of one hundred, with chemotherapy, and above all in steroid-refractory acute severe ulcerative colitis, where it's found in about a third. The ulcer base is biopsied with CMV immunohistochemistry, because standard staining misses many cases and blood PCR neither proves nor excludes colitis. Treatment is intravenous ganciclovir for fourteen to twenty-one days, then valganciclovir, with immunosuppression reduced. UL97 resistance responds to foscarnet or maribavir, UL54 resists both ganciclovir and foscarnet, and letermovir only prevents CMV.

17:52Disseminated MAC, below a CD4 count of fifty, gives PAS-positive macrophages that are acid-fast, unlike Whipple disease. It's treated with a macrolide plus ethambutol, rifabutin in advanced disease, and antiretroviral therapy, until the CD4 count has been above one hundred for six months. Microsporidia need a trichrome stain, and Encephalitozoon responds to albendazole while Enterocytozoon doesn't. Cyclospora and Cystoisospora are acid-fast and treated with trimethoprim-sulfamethoxazole. Norovirus causes brief, vomiting-predominant outbreaks, treated supportively and controlled with soap and water and bleach.

18:32That completes chapter thirty-three. Chapter thirty-four is gastrointestinal disease in the immunocompromised host.

18:39For the full chapter, the practice vignettes, and the topic-tagged question bank, head to board pearls dot com. You'll find the rest of the series on Apple Podcasts, Spotify, or wherever you listen to podcasts. That brings us to the end of episode five of five of chapter thirty-three, and I'll see you in the next one.