falled below the trial condition. Cohort studies of HIV infected tuberculosis patients in Thailand has shown completion rate of less than 50% which was much less than WHO recommended rate of 85%. Directly observed therapy is the best strategy to combat non- compliance problem of treatment.
High mortality rate after treatment even with short course regimen was seen in HIV infected tuberculosis patients. The mortality rate was between 14-44%. In Thailand, this mortality rate was between 5-25% and the highest rate came from the study in advanced AIDS patients. In non HIV infected tuberculosis patients, the mortality rate after short course regimen was usually less than 5%. Interestingly, in the study with 44% of mortality rate from San Francisco, only 6% of patients died directly because of tuberculosis. Factors which associated with decreased survival were disseminated disease, intrathoracic adenopathy , oral candidiasis and CD4 T lymphocyte depletion. One study has correlated level CD4 T lymphocyte with survival of HIV infected tuberculosis patients and found that with CD4 T lymphocyte less than 50 cells/microlitre, mean survival was 6 months as compare to 60 months in patients with CD4 T lymphocyte more than 300 cells/microlitre.
Adverse drug reactions from antituberculous drugs were observed more frequently in HIV infected tuberculosis patients The well known drug, Thiacetazone, was prohibited to use in HIV infected patients because of unacceptable high rate of adverse drug reactions. And some reactions were severe and fatal such Steven-Johnson syndrome or exfoliative dermatitis. Adverse drug reactions from other drugs were also reported to be higher in HIV infected patients in Thailand, we have adverse drug reactions rate of 4-18%. These reactions would effect the compliance of patients so comprehensive management of these reactions would improve patients' compliance.
Drug resistance is another problem in HIV infected tuberculosis patients. Most of the studies have shown the higher rate of drug resistant rate in HIV infected than non-HIV infected patients especially multidrug resistant tuberculosis.
Clinical features of tuberculosis in HIV infected patients.
Co-infection of HIV in tuberculosis patient has great effect on the clinical features and diagnosis in individual patients. Studies from the beginning of HIV epidemic have shown that clinical presentations, yields of sputum examination and chest radiographs in HIV infected tuberculous patients were varied and depended on immune status of patients. In patients of less immune deficiency state, tuberculosis presented with usual tuberculosis with signs and symptoms of productive cough, fever and weight lost. But in advanced immunodeficiency status, patients had more prolonged fever and extrapulmonary tuberculosis could be found more frequently. Stigmata of advanced immunodeticiency status could be seen in these patients such as oral candidiasis, herpes,zoster wasting syndrome. For physical examination these clinical signs should be looked for in order to determine the HIV infection in each tuberculosis patients
Microscopic sputum examination for diagnosis of pulmonary tuberculosis in HIV infected patient was another point of controversy. In early studies of HIV infected tuberculosis patients, some reports have mentioned that in HIV infected tuberculosis patients, microscopic sputum examination had less sensitivity and specificity and doubted that this technique was still useful as a standard measure for diagnosis or not. In contrast, some later studies has reported a better yield of microscopic sputum examination in HIV infected tuberculosis patients. In general, because microscopic sputum examination by smeared and stained method is a cheap, easy, rapid and do not need a sophisticated instrument, the lower sensitivity and specificity in HIV infected tuberculosis could be compromised. Sputum examination by smeared and stained method is still the recommended standard diagnostic technique for HIV infected tuberculous patients. If culture for mycobacteria is available, it should be added diagnostic technique and could not be substitution of microscopic sputum examination. Interestingly, two studies has correlated level of CD4 T lymphocyte with yield of sputum examination and found than even at very low level of CD4 T lymphocyte count, less than 50-100 cells/microlitre. Microscopic sputum examination could give a high sensitivity and higher than in the patients with CD4 T lymphocyte count more than 300 cells/microlitre. These findings confirmed the benefit to use microscopic sputum examination as diagnostic technique in HIV infected tuberculosis patients. For extrapulmonary tuberculosis such as lymphadenitis, aspiration and microscopic examination gave better results than sputum examination. Histopathologic examination may be necessary for extrapulmonary tuberculosis which aspiration could not be done such as colonic tuberculosis, intraabdominal tuberculous lymphadenitis, etc.
Chest radiographs of tuberculosis in HIV infected patients followed the same situation as sputum examination and clinical presentations. The so-called "atypical features" in chest radiographs could be seen more frequently in HIV infected patients. These atypical patterns are mid and lower lung fields infiltrations, miliary diseases, intrathoracic lymphadenopathy, pleural effusion and normal chest radiographs.
It is important to keep in mind that when look after HIV infected patients with pulmonary infections. Individual patient may have more than one opportunistic infection like pulmonary tuberculosis with pneumocystis carinii pneumoniae because of immunocompromised status of patients. Some pulmonary opportunistic infections could give clinical features which mimic pulmonary tuberculosis such as pulmonary nocardiosis, pulmonary rhodococosis, pulmonary cryptococosis or aspergillosis. These infections should be looked for in HIV infected patients with clinical features of tuberculosis but sputum examinations were negative for tubercle bacilli.
Treatment of tuberculosis in HIV infected patients
The response to treatment of tuberculosis in HIV infected patients is as good as in non-HIV infected patients with standard short course regimen of 6 months. This standard regimen has an intensive phase of isoniazid rifampicin, pyrazinamide and ethambutol for 2 months and continuation phase of isoniazid, rifampicin for 4 months. Conventional standard 12 months regimen was not recommended because it was not cost-effectiveness. Thiacetazone must not be used because of unacceptable adverse drug reactions. The problem of tuberculosis treatment in HIV infected patient is the optimal duration of treatment. Extended continuation phase to 7 months had ever been recommended in order to reduce the relapse rate in the United States. And eventhough most of the studies of short course regimen for tuberculosis in HIV infected patients have shown a higher ralapsed rate in HIV infected group but was not high enough to change the duration of treatment in the situation of restrained resources. The resources which will be used to decrease relapsed rate should provide for treatment of new infectious tuberculosis patients.
