Isolated cervical tuberculosis in patients with HIV infection.

OBJECTIVE: Tuberculosis isolated to the head and neck region is common in patients with HIV infection. However, the management of isolated head and neck tuberculosis has not been reported in the literature. This study was done to describe the characteristics of tuberculosis isolated to the head and neck region in patients infected with HIV and to detect differences in presentation and diagnostic management based on the status of HIV infection at presentation.

METHODS: A retrospective study was performed including 38 patients infected with HIV who were seen with tuberculosis isolated to the head and neck region at two tertiary care centers during a 10-year period. These patients were divided into two groups on the basis of the HIV status at presentation, which indirectly reflects the level of immunosuppression. Group 1 included 11 patients (29%) with AIDS at presentation. Group 2 included 27 patients (71%) with HIV infection but not AIDS.

RESULTS: The cervical lymphatics were the most common site for isolated head and neck tuberculosis (89%), with the supraclavicular nodes most often involved (53%). Extralymphatic involvement was less common (11%), but involved a variety of anatomic locations (skin, spinal cord, larynx, parotid). The presenting history and physical examination had a low sensitivity for tuberculosis in patients with HIV infection, mainly because of the presence of multiple confounding factors. Purified protein derivative testing was highly sensitive for tuberculosis in patients with HIV infection alone (61 %); however, its usefulness was diminished in patients with AIDS (14%; p=0.03). Fine-needle aspiration biopsy was 94% sensitive for diagnosing tuberculosis and was not affected by the status of HIV infection. Surgical biopsy was the gold standard for diagnosing tuberculosis but was associated with chronically draining fistulas in a significant number of cases (14%).

CONCLUSIONS: These data suggest that tuberculosis should be considered in the differential diagnosis of all head and neck lesions in patients infected with HIV, even in the absence of pulmonary involvement. Purified protein derivative testing should be done liberally in these patients, with realization that the sensitivity of purified protein derivative testing is reduced in patients with AIDS. Fine-needle aspiration biopsy should be the key diagnostic test in this patient population, with open surgical biopsy reserved for highly suspicious cases in which other measures were not diagnostic.