Search

PURPOSE: To determine the etiologies of esophageal symptoms in human immunodeficiency virus (HIV)-infected patients failing antifungal treatment. METHODS: Between August 1, 1990 and December 31, 1994, all HIV-infected patients seen at a large inner-city hospital who had esophageal complaints despite being on antifungal therapy were prospectively evaluated for the cause of symptoms. Thus, the population studied included patients given empiric antifungal therapy for esophageal symptoms and patients who developed symptoms while on long-term antifungal therapy. Endoscopy was performed in all patients. The cause of symptoms was determined by the clinical, endoscopic, and pathologic findings, and follow-up after treatment. RESULTS: Over the 53-month study period, patients failing empiric antifungal therapy were identified. The majority (77%) of these patients had esophageal ulcers; 25 patients had idiopathic ulcers and 24 had cytomegalovirus. In 2 patients, Candida was present with other causes of ulcerative esophagitis. Candida esophagitis alone was diagnosed in only 3 patients. No endoscopic abnormalities were observed in 14 patients (19%). An additional patients developed esophageal symptoms while receiving antifungal therapy; endoscopic findings in these patients included ulceration in 16 (67%), Candida esophagitis alone in 2, and normal in 6. Empirically treated patients in whom odynophagia was not the only symptom, those with dysphagia alone, and those with a CD4 count >100/mm3 were less likely to have an endoscopic diagnosis. CONCLUSIONS: Esophageal ulceration is the most common cause of esophageal symptoms in HIV-infected patients failing empiric antifungal therapy and those developing symptoms while receiving antifungal agents. Given these findings, endoscopy should be the test of choice for these nonresponders, rather than escalating the dose of antifungal agent, adding other empiric treatments, or performing barium esophagography. [C] 1996 by Excerpta Medica, Inc. Am J Med. 1996;101:599-604

PURPOSE: Although cytomegalovirus (CMV) esophagitis is an important complication of acquired immunodeficiency syndrome, there has been little study specifically addressing the response to currently available antiviral therapy, relapse rate without; maintenance therapy, and long-term outcome. PATIENTS AND METHODS: Over a 45-month period, 44 patients with CMV esophagitis established endoscopically and histopathologically were prospectively identified from among all human immunodeficiency virus (HIV)-infected patients undergoing endoscopy. Induction therapy consisted of intravenous ganciclovir at 10 mg/kg per day for approximately 14 days. Foscarnet was given at 60 mg/kg every 8 hours for nonresponders to ganciclovir. RESULTS: Of these patients, 35 completed induction ganciclovir therapy, resulting in a complete response in 17 (49%) and a partial response in 10 (29%), yielding a 77% overall response rate. Seven of 8 nonresponders were subsequently treated with foscarnet, with a clinical response seen in 5 patients. In the 18 eventual complete responders to ganciclovir or foscarnet followed up without maintenance therapy, 7 (39%) relapsed at a median of 4 months (range 2 to 18 months). In all cases, relapse was manifested by recurrent odynophagia. Reinduction ganciclovir therapy yielded a complete response in 1 patient and a partial response in 2, and induction foscarnet treatment resulted in a complete response in the other treated patients. During long-term follow-up, 1 complete responder developed CMV colitis with concurrent retinitis, and 4 other patients developed retinitis. The median survival after diagnosis was 8.2 months, although survival for greater than 1 year was seen in 4 patients. No patient died as a direct result of esophageal disease, although ulcer-related bleeding may have contributed to death in 2 patients with end-stage liver diseases and hepatic encephalopathy. CONCLUSIONS: CMV esophagitis has a favorable response to induction ganciclovir therapy, and a long-term remission may occur after induction therapy alone. Despite the favorable response to ganciclovir therapy, the long-term survival is poor, reflecting the severe immunodeficiency of these patients.