The California Department of Public Health notified the Alaska Division of Public Health (ADPH) about this suspected MDR TB transmission in Alaska. ADPH reviewed case records and identified an Africa-born patient (AK1) with MDR TB and positive sputum smear results who had been employed in seafood production at the time of his 2006 diagnosis. During contact investigation for the case, ADPH evaluated 3 roommates with previously positive tuberculin skin test results. ADPH did not expand the contact investigation because many workplace contacts were no longer employed at the facility and were unreachable. No persons identified in the initial contact investigation were subsequently identified as outbreak case-patients. A 2010 review of employer records confirmed that all 4 case-patients had been employed in the same facility during AK1’s infectious period (Figure). Because contact information for other workers was unknown, no further investigation could be pursued.

Initial genotyping results (spoligotyping and 12-locus mycobacterial interspersed repetitive units–variable number of tandem repeats [MIRU-VNTR] analysis) demonstrated that AK1’s isolate had a genotype that differed from those in California and Washington at 1 MIRU-VNTR locus (4). To confirm these results and further evaluate the relatedness of the isolates, 24-locus MIRU-VNTR (5) and IS6110-based restriction fragment length polymorphism analyses (6) were conducted and showed exact matches among the 4 cases by spoligotype, 24-locus MIRU-VNTR, and restriction fragment length polymorphism (Table 1). The single-locus difference in initial and subsequent genotype results of AK1 was determined to be a laboratory error.

The 4 case-patients worked in a seafood production facility in Alaska during the summer of 2006. The facility included multiple large buildings with high ceilings and open areas with production lines. Patients AK1, CA1, and CA4 completed follow-up interviews about their activities in Alaska. All 3 reported working in the same building; 2 also worked in a second building. They lived in 3 different apartments that were in 2 different buildings and ate and socialized primarily in each dormitory’s cafeteria. Case-patients reported working up to 12 hours per day, 7 days per week; they did not report any common activities outside of work. No other links were identified among the case-patients.

Contact investigation of the secondary cases (CA1, CA4, and WA1) was conducted, and 47 (96%) of 49 contacts were fully evaluated. Of these 47 contacts, 2 had active TB disease (CA2 and CA3, US-born children with negative culture results), and 30 (64%) had latent TB infection (LTBI); 28 began treatment for MDR LTBI (Table 2). Review of these contact investigations determined that reopening and expanding the investigations would not be productive.

To facilitate prompt investigation, the Centers for Disease Control and Prevention is now actively monitoring the NTGS database for new cases matching the outbreak genotype. NTGS surveillance identified 1 additional case in Washington State during 2010. This case-patient had never worked in seafood production and had not been identified during the contact investigation of WA1. However, his cousin was WA1’s roommate, and he later moved into WA1’s apartment after WA1 moved out, so unrecognized contact could have occurred.

This outbreak underscores the importance of considering TB transmission in nonresidential settings. Current TB-control guidelines emphasize the need to identify and assess the risk for transmission at all possible sites, including workplaces (8). Local health departments should weigh the probable yield of expanded and worksite investigations relative to other TB-control activities. If worksite investigations are pursued, clarifying employers’ responsibilities for funding and supporting those investigations might help mobilization of the substantial resources typically required.

Industries that employ large numbers of foreign-born workers from countries with a high TB incidence might encounter TB among their employees. In seafood production facilities, where those workers live and work together, transmission risk is likely increased. Interventions to identify TB cases more quickly include employee and employer education regarding TB symptoms and institution of a cough alert program to ensure access to clinical evaluation of a persistent cough (9,10). TB control programs should consider the possibility of domestic TB transmission even among foreign-born patients, particularly if the patients have lived or worked in crowded conditions with other persons at higher risk of having TB.

Because initial genotype results indicated an exact match among only 3 of the 4 cases, an additional conclusion of this outbreak is that epidemiologic links are possible among case-patients with closely related genotypes. However, resources to explore epidemiologic links among patients with nonmatching genotypes should be used judiciously. Discussion with the genotyping laboratory and retesting are important first steps when epidemiologic links are suspected among patients with closely related genotypes.