R Adams1, R Ruffin, D Campbell. 1. Respiratory Medicine Unit, Queen Elizabeth Hospital, Adelaide, SA.
Abstract
BACKGROUND: A semi-quantitative index of lipid-laden macrophages on broncho-alveolar lavage (BAL) has been reported to be highly sensitive but only moderately specific for aspiration in adults. There has been little published literature evaluating this technique since the original report. AIMS: To assess the value of a lipid-laden macrophage index (LLMI) of greater than 100 to confirm the clinical diagnosis of aspiration in patients with abnormal radiological investigations. METHODS: Prospective evaluation of 80 adult patients with abnormal radiology was undertaken using BAL and the LLMI. A diagnosis of aspiration was made prior to bronchoscopy if the patient had one or more of: clinically witnessed aspiration, positive barium swallow or speech pathology assessment; or an upper gastrointestinal endoscopy showed severe reflux oesophagitis and the patient had a history consistent with aspiration. RESULTS: Eighteen patients were diagnosed with aspiration. Of these, 17 had an index > 100, and one had an index of 94 (mean 157, 99% CI 127-187, range 94-238). Of the other 62 subjects, seven had an index > 100 (mean 46, 99% CI 22-70, range 0-303). There was a significant difference between index scores for the two groups (p = 0.002). For aspiration, an index > 100 had a sensitivity of 94% and a negative predictive value of 98%. The specificity was 89%, with a positive predictive value of 71%. CONCLUSIONS: The LLMI is a sensitive indicator for aspiration causing radiological lung disease in adults. Its lack of specificity means it cannot be the sole means of diagnosis, but it can allow better targeting of other investigations in those patients in whom bronchoscopy is undertaken to investigate radiological abnormalities.
BACKGROUND: A semi-quantitative index of lipid-laden macrophages on broncho-alveolar lavage (BAL) has been reported to be highly sensitive but only moderately specific for aspiration in adults. There has been little published literature evaluating this technique since the original report. AIMS: To assess the value of a lipid-laden macrophage index (LLMI) of greater than 100 to confirm the clinical diagnosis of aspiration in patients with abnormal radiological investigations. METHODS: Prospective evaluation of 80 adult patients with abnormal radiology was undertaken using BAL and the LLMI. A diagnosis of aspiration was made prior to bronchoscopy if the patient had one or more of: clinically witnessed aspiration, positive barium swallow or speech pathology assessment; or an upper gastrointestinal endoscopy showed severe reflux oesophagitis and the patient had a history consistent with aspiration. RESULTS: Eighteen patients were diagnosed with aspiration. Of these, 17 had an index > 100, and one had an index of 94 (mean 157, 99% CI 127-187, range 94-238). Of the other 62 subjects, seven had an index > 100 (mean 46, 99% CI 22-70, range 0-303). There was a significant difference between index scores for the two groups (p = 0.002). For aspiration, an index > 100 had a sensitivity of 94% and a negative predictive value of 98%. The specificity was 89%, with a positive predictive value of 71%. CONCLUSIONS: The LLMI is a sensitive indicator for aspiration causing radiological lung disease in adults. Its lack of specificity means it cannot be the sole means of diagnosis, but it can allow better targeting of other investigations in those patients in whom bronchoscopy is undertaken to investigate radiological abnormalities.
Authors: M Woodhead; F Blasi; S Ewig; J Garau; G Huchon; M Ieven; A Ortqvist; T Schaberg; A Torres; G van der Heijden; R Read; T J M Verheij Journal: Clin Microbiol Infect Date: 2011-11 Impact factor: 8.067