[Esophagitis in patients with acquired human immunodeficiency syndrome: an histological and immunohistochemistry study]

BACKGROUND

Almost all patients with acquired immunodeficiency virus syndrome will have gastrointestinal symptoms during the course of their illness. The high prevalence and complications of esophagitis are well documented.

AIM

Graduate esophagitis; identify microorganisms like Candida sp, cytomegalovirus, herpesvirus and mycobacteria; identify by immunohistochemical staining viral agents cytomegalovirus, herpesvirus I, herpesvirus II, Epstein-Barr Virus, human papilloma virus and human immunodeficiency virus; verify how immunohistochemistry changes the profile of esophagitis; verify the association between the histological and endoscopical findings; verify the relevance of the number of fragments studied in the characterization of the histological agents.

METHODS

We studied retrospectively esophageal biopsies in 227 patients with acquired immunodeficiency virus syndrome using hematoxylin and eosin, PAS (periodic acid of Schiff), Groccott and Ziehl-Nielsen stains and immunoperoxidase stains to detect opportunistic agents. Endoscopic aspects were studied.

RESULTS

The non-specific esophagitis grade III, in the inferior third of the esophagus, was the most frequent type. Candida sp was the most frequent agent, followed by viruses cytomegalovirus, herpesvirus and mycobacteria. The presence of plaque and ulceration suggested the diagnosis of esophageal candidiasis and cytomegalovirus esophagitis. Immunohistochemical allowed the characterization of cytomegalovirus and of herpesvirus in those cases where other techniques could not achieve it, furthermore the cytomegalovirus was also found in histological normal cases, making the use of this technique advisable in routine diagnosis. The herpesvirus I was not found isolated but associated to herpesvirus II. We have not found immunoreactivity for the Epstein-Barr virus and the human immunodeficiency virus. The number of fragments does not seem to influence the detection of the etiologic agent.

CONCLUSION

The endoscopic findings of plaques or ulcers are associated with candidiasis or cytomegalovirus esophagitis. Immunohistochemisty improved the diagnosis of viral infections. It is possible to detect cytomegalovirus infections in endoscopic and histologic normal cases.

Human immunodeficiency virus and endoscopy: Experience of a general hospital in Singapore

INTRODUCTION

As the human immunodeficiency virus (HIV) pandemic continues, demand on the endoscopic services will also increase. The aim of the present study was to characterize the characteristics and endoscopic findings of HIV-infected Asian patients.

METHODS

Patients with HIV/AIDS who had endoscopy from January 1997 to September 2002 were identified and retrospectively reviewed.

RESULTS

There were 125 patients (male, 111; mean age: 43.2 +/- 11 years; mean CD4 count 63 +/- 85/mm(3), range, 1-342/mm(3)) who underwent 181 procedures (upper gastrointestinal endoscopy [UGIE], n = 141; lower gastrointestinal endoscopy [LGIE], n = 40). The median time from diagnosis to endoscopy was 240 days (-660 to 4680 days). For UGIE the main findings were candidiasis, 23.1%; cytomegalovirus esophagitis/ulcers, 11.2%; duodenal ulcers, 8.4%; gastric ulcers, 7.0%; portal-hypertensive-related changes, 6.3%; idiopathic esophageal ulcers, 3.5%; herpes simplex esophagitis/ulcers, 3.5%; and tuberculosis, 2.1%. The CD4 counts were significantly lower in those with opportunistic infections (P = 0.004) but there was no difference between significant and non-significant findings (P = 0.191). For LGIE 35% had endoscopic colitis, 80% of which were non-specific. Significant findings were ileal tuberculosis, n = 3; colon cancer, n = 1; and colonic fistula, n = 1. Ten patients had undergone 16 procedures (UGIE, n = 13; LGIE, n = 3) prior to the diagnosis of HIV/AIDS. They were all male Chinese patients, with positive contact with commercial sex workers (CSW) and had lymphopenia at time of endoscopy. The median time from endoscopy to diagnosis was 180 days (range, 1-660 days). There were no significant differences in age (P = 0.512) and CD4 count at diagnosis (P = 0.066) between patients who had endoscopy before and after diagnosis of HIV/AIDS. Four procedures led to the suspicion of HIV.

CONCLUSIONS

Endoscopic findings of symptomatic HIV Asian patients are comparable to those of the West. Universal precautions should always be exercised in all procedures to avoid transmission of disease.