Mycobacterium tuberculosis and polymorphonuclear pleural effusion: incidence and clinical pointers.

BACKGROUND

Delayed diagnosis and treatment of a polymorphonuclear cell (PMN)-predominant pleural effusion due to Mycobacterium tuberculosis (MTB) are associated with poor outcome and the risk of tuberculosis transmission. We investigated the clinical differences of PMN-predominant pleural effusion due to MTB or other microorganisms.

METHODS

From January 2000 to April 2007, a total of 354 patients with tuberculous pleurisy were identified. Among them, 39 (11.0%) adults had PMN-predominant pleural effusion (MTB group). Their clinical characteristics were compared with the 117 age-/gender-matched controls (1:3) selected from 715 patients with PMN-predominant pleural effusion due to other microorganisms.

RESULTS

Among patients with PMN-predominant septic pleural effusion, 5.2% were due to MTB. The in-hospital mortality rate in the MTB group was 36%, similar to that of the control group. Sputum samples were culture-positive for MTB in 41%. Weight loss (p=0.006), initial leukocyte count <or=11,000/microL (p=0.007), and poor clinical response to empirical antibiotics in the first 3 days (p=0.002) were independent factors suggestive of tuberculous pleurisy. A shift toward mononuclear cell predominance of pleural effusions within 1 week was significantly associated with tuberculous pleurisy. In the MTB group, if anti-tuberculous treatment was started more than 14 days after the initial visit, there was a worse prognosis (p=0.034). Among those with delayed treatment, 96.2% had finding(s) suggestive of tuberculous pleurisy.

CONCLUSIONS

A high index of clinical suspicion can identify MTB in about 5.2% of patients presenting with PMN-predominant septic pleural effusions. Awareness of the clinical pointers can lead to early diagnosis and improved clinical outcome.