University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Management Recommendations > Odynophagia

Ward 86 Management Recommendations

Managing an HIV Patient with Odynophagia, Dysphagia, or Gastroesophageal Reflux Disease

updated August 2018

Contributors: Mark A. Jacobson, MD
Annie Luetkemeyer, MD
Dan Wlodarczyk, MD

Approach to Diagnosis

  • Many HIV patients with odynophagia or dysphagia have gastroesophageal reflux disease (GERD) and will respond to antacid treatment. However, the provider must be aware that there are important drug-drug interactions between acid-blocking medications (proton-pump inhibitors, H2 blockers, and cation-based antacids) and certain antiretroviral medications (see box below). Among patients with an absolute CD4 T-cell count of <350 cells/µL, candidal esophagitis may occur, even in the absence of oral thrush.

  • Among those with a CD4 count of <100 cells/µL, cytomegalovirus (CMV) esophagitis, herpes simplex virus (HSV) esophagitis, or giant esophageal aphthous ulcers may occur.

  • As in HIV-uninfected patients, persistent dysphagia also may be caused by an esophageal stricture or motility disorder, achalasia, or an esophageal, gastric, or mediastinal malignancy. Because the risk of solid tumors, as well as lymphoma, is significantly increased in HIV-infected patients, even those virologically suppressed with antiretroviral therapy (ART), progressive dysphagia with anorexia and weight loss should trigger a workup for malignancy.

  • Also, as in HIV-uninfected patients, acute odynophagia may be caused by pharyngitis due to gonorrhea, chlamydia, or Streptococcus pyogenes infection.
  • Multiple medications can cause or contribute to odynophagia or dysphagia. A through review of the patient's medications should be done. Important classes include antipsychotics and neuroleptics, anticonvulsants, antihistamines, antidepressants, bisphosphonates, potassium, and nonsteroidal antiinflammatory drugs.

Candidal esophagitis

  • This condition commonly presents with odynophagia, and occasionally presents with dysphagia or substernal chest pain.

  • Usually, but not always, it is associated with oropharyngeal thrush.

  • Endoscopy is indicated only for those patients who do not respond to empiric antifungal treatment.

  • Fluconazole-refractory candidal esophagitis may occur in some patients.

    1. Most such cases are in patients with a CD4 count of <200 cells/µL who are not receiving ART.

    2. The cause may be a fluconazole-resistant strain of Candida albicans, particularly in those who have had substantial prior exposure to fluconazole, or a non-albicans Candida species with inherent decreased susceptibility (eg, C glabrata) or outright resistance (eg, C krusei) to fluconazole. In some cases of C glabrata infection, resistance can be overcome by higher doses of fluconazole. In such cases that are suspected clinically, based on fluconazole treatment failure, an empiric trial of oral itraconazole solution is often effective.(1) However, itraconazole can have important drug interactions with antiretroviral medications, which should be reviewed prior to initiation.

    3. Before treating with other alternative antifungals (eg, intravenous caspofungin, oral posaconazole, or oral voriconazole), we recommend obtaining microbiologic documentation of fluconazole-resistant Candida from an oral swab (if patient has thrush) or an esophageal brushing obtained by endoscopy.

CMV esophagitis

  • This condition occurs primarily in patients with a CD4 count of <50 cells/µL and typically presents with odynophagia.

  • Diagnosis requires an endoscopic biopsy that demonstrates the presence of intracellular CMV inclusions on histopathology. CMV culture of esophageal brushings is not clinically useful in diagnosing any form of opportunistic CMV gastrointestinal disease because of poor specificity and sensitivity.(2,3) Tests for plasma CMV viremia have poor sensitivity and specificity for all forms of CMV end-organ disease.

  • Treatment is oral valganciclovir 900 mg BID for 3 weeks, then 900 mg QD for 3 weeks. In patients unable to take oral medication, we start treatment with intravenous ganciclovir 5 mg/kg every 12 hours until the patient can tolerate swallowing pills.

Giant aphthous ulcers

  • This condition occurs almost exclusively in patients with a CD4 count of <100 cells/µL.

  • This is a diagnosis of exclusion that requires endoscopic biopsy to rule out other causes.

  • Treatment options include intralesional steroid injection, a short course of oral prednisone (eg, 1 mg/kd/day for 5 days), or thalidomide.(4)

Herpes simplex esophagitis

  • This is very rarely reported in association with HIV and occurs almost exclusively in patients with a CD4 count of <100 cells/µL.

  • It requires endoscopic biopsy to diagnose. Multinucleated giant cells on biopsy or a positive culture from an esophageal brushing can confirm the diagnosis.

  • We suggest treatment with oral valacyclovir 1 gm BID or acyclovir 400 mg 5 times daily, or intravenous acyclovir 5 mg/kg every 8 hours for patients unable to swallow pills.

  • Oral viscous lidocaine may provide symptomatic relief.

  • If refractory to above treatment, repeat endoscopy and send viral culture of lesion brushing or biopsy to test for acyclovir resistance.

Initial Empiric Treatment of Odynophagia/Dysphagia

  • There have been no randomized trials of empiric treatment strategies for HIV-infected patients with odynophagia or dysphagia. Our approach is directed by the patient's absolute CD4 T-cell count and results of the oral examination.

  • In patients with a CD4 count of <350 cells/µL who have evidence of thrush, we treat odynophagia empirically with fluconazole 200 mg/daily. If there is no improvement after 1 week, we will add a proton-pump inhibitor (PPI) in order to treat GERD. (See box below for H2 antacid alternatives to use instead of a PPI if the patient is on atazanavir- or rilpivirine-based ART.) If there is no improvement after another week, we refer the patient for endoscopic gastroduodenoscopy (EGD).

  • In those with a CD4 count of <350 cells/µL who have no evidence of thrush, we treat odynophagia empirically with a combination of fluconazole 200 mg/daily and a PPI regimen. (See box below for H2 antacid alternatives to use instead of a PPI if the patient is on atazanavir- or rilpivirine-based ART.) If there is no improvement after 1 week, we refer the patient for EGD.

  • Of note, there is an increase in the risk of C difficile and enteric infections in people who take proton-pump inhibitor (PPI) antacids, which cause a reduction in gastric acid, a natural defense against infectious agents. These drugs are commonly used for gastrointestinal complaints and are available without prescription. They also are associated with community-acquired pneumonia and health care-associated pneumonia, malabsorption of iron, calcium and magnesium, and chronic kidney disease. There are multiple drug interactions between PPIs and antiretrovirals as well. The lowest dosage and the shortest duration of treatment should be used for appropriate indications.

Potential Interactions between Acid-Blocking Medications and ARVs

There are important drug-drug interactions between acid-blocking medications and certain ARVs. Any acid-blocking medications may inhibit absorption of atazanavir and rilpivirine. Cation-based antacids such as calcium carbonate, aluminum hydroxide, and magnesium hydroxide (including TUMS, Maalox, Mylanta, and Rolaids) may chelate integrase inhibitors and decrease their plasma levels. Certain dosing precautions are recommended, and some combinations are contraindicated.

  • Proton-pump inhibitors (PPIs) are contraindicated with atazanavir or rilpivirine.

  • Long-acting H2 blocker antacid drugs (eg, famotidine 40 mg once daily) can be coadministered with boosted atazanavir or rilpivirine if dosed appropriately and spaced appropriately:

    • Atazanavir: Dosing of atazanavir and H2 blockers is complex; in general, atazanavir should be boosted by ritonavir or cobicistat and given simultaneously with the H2 blocker or at least 10 hours after.
    • Rilpivirine: H2 blockers should be given at least 4 hours after or 12 hours before rilpivirine dosing.
    • We recommend that H2 blockers not be given twice a day with these two ARVs.
  • Cation-based antacids (eg, calcium carbonate, aluminum or magnesium hydroxide) may interfere with absorption of atazanavir and rilpivirine, as well as integrase inhibitors.

    • Atazanavir: Antacids should be given at least 2 hours before or at least 2 hours after atazanavir.
    • Rilpivirine: Antacids should be given at least 2 hours before or at least 4 hours after rilpivirine.
    • Dolutegravir: Can be taken simultaneously with a calcium antacid if the two agents are taken with a meal. Dolutegravir must be taken at least 2 hours before or at least 6 hours after an aluminum- or magnesium-containing antacid.
    • Elvitegravir: We do not recommend coadministration.
    • Raltegravir: May be taken simultaneously with calcium carbonate antacids. Should not be given with aluminum- or magnesium-containing hydroxide antacids.
    • Bictegravir: Should be taken on an empty stomach and at least 2 hours before any aluminum, calcium and/or magnesium containing antacids, buffered medications, or sucralfate.


  1. Saag MS, Fessel WJ, Kaufman CA, et al. Treatment of fluconazole-refractory oropharyngeal candidiasis with itraconazole oral solution in HIV-positive patients. AIDS Res Hum Retroviruses. 1999 Nov 1;15(16):1413-7.
  2. Wilcox CM, Diehl DL, Cello JP, et al. Cytomegalovirus esophagitis in patients with AIDS. A clinical, endoscopic, and pathologic correlation. Ann Intern Med. 1990 Oct 15;113(8):589-93.
  3. Laine L, Bonacini M, Sattler F, et al. Cytomegalovirus and Candida esophagitis in patients with AIDS. J Acquir Immune Defic Syndr. 1992;5(6);605-9.
  4. Jacobson JM, Spritzler J, Fox L, et al. Thalidomide for the treatment of esophageal aphthous ulcers in patients with human immunodeficiency virus infection. National Institute of Allergy and Infectious Disease AIDS Clinical Trials Group. J Infect Dis. 1999 Jul;180(1):61-7.