Tuberculosis outbreak linked to pubs in South Wales

Eurosurveillance weekly releases 2006 > Volume 11 / Issue 9

J Vidal-Alaball (Josep.Vidal-Alaball@nphs.wales.nhs.uk), S Hayes, R Jones
National Public Health Service for Wales, Health Protection Team, Mid and West Wales Region, Swansea, United Kingdom

On 10 February 2004, the specialist respiratory nurse for a hospital in a city in South Wales alerted the local health authorities that Mycobacterium tuberculosis infections in a particular area of the city appeared to be higher than expected in a relatively young age group. All the cases were in patients who reported visiting a particular pub (bar) regularly.
As a result of this alert, an outbreak investigation team compiled an initial list of six possible associated cases in five men and one woman who were aged between 25 and 55 years and all lived in the city. Initial interviews established that there were many shared social links between the patients, who were part of a network of regular drinkers at four pubs in a suburb of the city. The interviews were difficult because of the complexity of social relations and reluctance by some of the patients to provide information about their social and sexual life. All reported regularly visiting a certain pub (Pub A) in the area. By October 2004, two further linked cases had been diagnosed.
To determine whether there were any other linked tuberculosis cases, all local pubs mentioned by cases were visited, surrounding hospitals were contacted, letters were sent to primary care physicians, and primary care tuberculosis prescribing data collected from primary care, was analysed to detect unusual patterns. No other cases were identified as a result.
During 2005, two further cases of M. tuberculosis infection were diagnosed in the same area of the city, also in people who had links with Pub A. Tuberculosis typing (VNTR method) showed that the cases had identical profiles to cases 1, 2, 3, 5 and 6. Both patients were already known to the outbreak investigation team: case 8 had been named but not identified during the initial investigations and case 9 had been screened and initially cleared, but later confirmed through fine needle aspirate of the lung. No new social contacts were identified from these last two cases.
Contact tracing established that 2/20 household contacts (10%) and 3/18 social contacts screened (17%) required prophylactic antibiotics.
Genetic typing of all cultures was undertaken. PCR-based variable number tandem repeat (VNTR) method for tuberculosis typing was performed. Cases 1, 2, 3, 5 and 6 had identical profiles. An earlier case in a patient who had presented in August 2002 was not microbiologically related to other cases. This case was therefore not considered to be part of the outbreak, and is not included in the Table. No genetic information was available for case 4, as the diagnosis was made while the case was travelling in Australia or for case 7 as this was a clinical diagnosis with a negative culture.
In 2006, the local health authorities have continued to liaise with primary care physicians, respiratory nurses and chest physicians, but no further related cases have been identified.

Conclusion
This outbreak of nine cases included seven cases with identical microbiological
profiles, one case that was diagnosed on clinical grounds and one case that was diagnosed outside the United Kingdom. All nine patients were local residents of white ethnicity (unusual in patients in the UK) who reported a history of drinking in Pub A. Tuberculosis typing allowed one earlier case to be excluded from the outbreak.
In this particular outbreak it has been very difficult to take reliable medical histories due to a combination of poor history telling on the part of the patients and the complex social context in which the outbreak took place. Other authors have previously reported that conventional contact tracing has been insufficient in tuberculosis outbreaks linked to pubs [1]. Without genetic typing it would have been very difficult to establish links between cases in this outbreak. With tuberculosis typing (VNTR method) now routinely available in Wales for all positive cultures, we may see further cases linked to this outbreak in the future.

Acknowledgments:Dr Brendan Mason (Regional Epidemiologist), Mrs Cheryl Owen (Respiratory Nurse), Mrs Nicola John (Local Public Health Director), Wales Centre for Mycobacteria

References:
Diel R, Meywald-Walter K, Gottschalk R, Rusch-Gerdes S, Niemann S. Ongoing outbreak of tuberculosis in a low-incidence community: a molecular-epidemiological evaluation. Int J Tuberc Lung Dis 2004; 8(7):855-61.