SESSION TYPE: Respiratory Infections Posters I
PRESENTED ON: Wednesday, October 24, 2012 at 01:30 PM - 02:30 PM
PURPOSE: In most centers, despite the low yield, bronchoalveolar lavage (BAL) is routinely sent for smear and culture for mycobacterial organisms (AFB) on all bronchoscopic procedures, with Mycobacterium tuberculosis (Mtb) and Mycobacterium avium-intracellulare (MAC) most commonly isolated. We aimed to devise an approach that would reduce the number of AFB specimens sent, without missing patients who might have benefited. This would reduce unnecessary testing and medical expenses.
METHODS: A retrospective study, conducted at Winthrop-University Hospital, a 591-bed teaching hospital in suburban Nassau County, adjacent to New York City, serving a population of diverse ancestry including some foreign-born patients. The rate of tuberculosis per 100,000 population was 2.8 in Nassau County compared to 12.3 in neighboring Queens County (2009, NY State Department of Health). The study included all patients who had bronchoscopy from 1/1/10-12/31/11. A two-tailed p value was calculated by Fisher’s exact test to compare AFB-positivity based on clinical-radiologic pattern.
RESULTS: The study included 459 patients; 369 had regular bronchoscopy, and 90 had EBUS bronchoscopy. 352 patients had BAL sent for AFB stain and culture, none of which were positive for Mtb, while 10 had MAC identified. The incidence of MAC was 2.2% (10/459). When patients had clinical-radiologic features of MAC, 15.4% (8/52) grew MAC, while those who had findings suggesting malignancy only grew MAC in 0.8% (2/244) (p<0.0001). However, 20% (2/10) of patients who were positive for MAC from BAL were not suspected for this diagnosis pre-bronchoscopy.
CONCLUSIONS: The epidemiological incidence rates of TB and MAC vary among regions. Relatively few patients undergoing bronchoscopy had AFB positive BAL at our institution, located in a relatively low incidence area. MAC from BAL was more common in patients with cough and tree-in-bud nodules on CT imaging. Comparatively those suspected to have malignancy based on a lung nodule or mass had a very low incidence of MAC. Thus a specific algorithm is needed at each hospital which is tailored to their incidence.
CLINICAL IMPLICATIONS: A combined clinical-radiologic-demographic algorithm could reduce the number of BAL AFB studies without missing many unsuspected cases of Mtb or MAC, and reducing unnecessary cost.
DISCLOSURE: The following authors have nothing to disclose: Bala Prakash, Shweta Upadhyay, Steve Salzman
No Product/Research Disclosure InformationWinthrop University Hospital, Mineola, NY