© 2003 American Thoracic Society
Delay in diagnosis among hospitalized patients with active tuberculosis—predictors and outcomesTo the Editor:We read with interest the study performed by Greenaway and colleagues (1). The authors conclude that as experience with in-hospital tuberculosis (TB) decreased, the likelihood of poor outcomes increased. The overall mortality in their report was 12%. However, the death rate rose to 40–50% for patients admitted to the intensive care unit (ICU). These observations confirm earlier reports suggesting that a delay in diagnosis and institution of appropriate antitubercular therapy are important predictors of mortality for patients with pulmonary TB (2). Patients admitted to the ICU with pulmonary TB deserve special consideration. In a study of hospitalized patients, malnutrition was an important risk for a poor outcome (3). In this study, 51% of the patients admitted to the ICU died. In another study, the mortality rate of TB patients with respiratory failure necessitating ICU admission and mechanical ventilatory support was 60%, despite appropriate four drug antitubercular therapy. Nutritional status, as measured by serum albumin concentration and hemoglobin, was an important predictor of survival (4). Nutritional status may, therefore, be an important prognostic factor that has been largely overlooked by the medical community. Early and aggressive attention to improving the patient's nutritional status may, therefore, be an important intervention in decreasing the mortality of TB in the ICU. Standard dose antitubercular therapy works well in the majority of patients. However, therapeutic drug monitoring has been advocated in patients slow to respond to standard antitubercular treatment, in patients with end-stage liver or renal disease, and in HIV infection (5). In addition, recent reports have described patients without markers for poor outcome who nevertheless responded poorly to antitubercular therapy (6). The adverse outcome was attributed to subtherapeutic drug levels. Although several mechanisms have been proposed to explain the low serum levels of antitubercular drugs, the nutritional status of the patient may be a contributing factor. Thus, drug therapy and nutritional balance appear to be interrelated aspects of TB infection therapy that contribute to patient outcome. Therapeutic antitubercular drug monitoring should be considered on a case-by-case basis in the TB patient in the ICU and could allow for better control of their therapy. Whereas other variables in the treatment of TB are predetermined, recognition of the potential benefits of nutritional support and therapeutic drug monitoring are interventions that the clinician can modify, and may reduce mortality from TB in the ICU.
East Tennessee State University Johnson City, Tennessee REFERENCES
From the Authors: Case series of patients with tuberculosis requiring intensive care and mechanical ventilation have reported mortality rates ranging from 33–100% (1–3). These patients often have advanced disease (i.e., miliary disease) and have had delayed diagnosis and treatment. In one study, nutritional status was an important predictor of survival in patients with tuberculosis and respiratory failure (4). In our study, delayed treatment was highly associated with late but not with early ICU admission (5). A total of 45 (11%) patients were admitted to the ICU at some point during hospitalization. Of these patients, 35 were admitted directly or transferred to an ICU within 4 days of hospitalization (early ICU admission), of whom 14 (40%) died. When compared to the 376 patients never admitted to the ICU, the demographic and clinical characteristics of these 35 patients were not significantly different, except they were slightly older. Although only 31% were treated or isolated within the first 24 hours (compared to 57% among those never admitted to the ICU), the occurrence of treatment delay exceeding one week was similar (26 versus 25%). Ten patients were admitted late to the ICU after an average of 20 days of hospitalization (median 18, range 7–52 days), of whom 50% died. Compared with the 35 patients admitted early to the ICU, or patients never admitted to the ICU, those admitted late to the ICU were more likely to be older and female, not to have had cough, and to have been smear negative. In contrast to those admitted early to the ICU, they were very likely to have had treatment delayed. This suggests that lack of recognition of TB in these patients leads to clinical deterioration to the point of requiring ICU admission. Based on our data, we believe it more likely that delayed treatment was responsible for ICU admission and mortality, than patients' nutritional status, or absorption of tuberculous medication.
Montreal Chest Institute Montreal, Quebec, Canada REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
