Prevalence and predictors of positive tuberculin skin test results in a research laboratory

Latent tuberculosis was studied in a research laboratory. A prevalence of positive tuberculin skin test results (≥ 15mm) of 20% was found and the main predictors were place of birth in a foreign country with high prevalence of tuberculosis and a history of contact with patients with untreated active tuberculosis. Key-words: Latent tuberculosis. Prevalence. �isk factors. �esearch laboratory.

Latent tuberculosis after infection with Mycobacterium tuberculosis with no evidence of disease affects up to one third of the world population 5 18 .The frequency of a positive tuberculin skin test (TST) reaction as a diagnostic test for latent tuberculosis 13 14 in the general population in Asia is: India 41% 9 12 14 , Pakistan 40%, South Korea 87%, Cambodia 39% and Afghanistan 34%; in Africa: South Africa 38% and Zimbabwe 64% 5 , in Europe: the Netherlands 4% 1 , Germany 29% 4 , Italy 40% 16 and United Kingdom 27% 9 17 , and in America: the Pacific �egion 15%, Canada 17% and USA 22% 5 .We studied the frequency of positive TST results among laboratory members and epidemiological risk factors for positive tuberculin skin test results to identify subjects with a high likelihood of latent tuberculosis.
Study subjects.We evaluated 65 out of 100 laboratory members (scientists and technicians) who provided informed consent for their participation.No formal sample size calculation was performed.The exclusion criteria were medical illness (e.g.diabetes) or taking immunosuppressive drugs (e.g.corticosteroids).
Setting.The setting was an inner-city biomedical research laboratory in Berlin with a biosafety level 3 facility that included research with Mycobacterium tuberculosis.

Data collection.
During the study period between 2005 and 2007, standardized clinical and epidemiological data were collected by one of the authors (FHP) during an interview, followed by the tuberculin skin test.
Tuberculin skin testing.Quantitative tuberculin skin testing was performed by FHP by means of intracutaneous injection of 5TU of PPD (purified protein derivative) in 0.1ml of solution (Tubersol, Pasteur Aventis, Canada) into the volar aspect of the forearm (Mantoux test).Induration was read after 48 to 72 hours and recorded in mm by one trained examiner (FHP).

Study approval.
All probands provided informed consent and the study was approved by the institutional review board.
Fifty-four (83%) robands had a history of BCG (Bacille Calmette-Guérin) vaccination, eight (12%) had not had the BCG vaccination and three (6%) did not know.Eight (12%) probands had a history of contact with patients with untreated active tuberculosis (e.g. in workplace, school or home).Twentytwo (34%) probands had a positive history of travel to a country with high prevalence of tuberculosis (e.g.working as a health professional in Africa or living in India for at least one year).Fifty-two (80%) probands were native-born (in Germany) and 13 (20%) were foreign-born (one in �ussia, one in China, ten in India and one in the United Kingdom).Thirty-five (54%) probands had a history of working in a mycobacteriology laboratory with exposure to Mycobacterium tuberculosis.The thirteen (20%) laboratory members with a TST reaction ≥ 15mm (Table 2) had a mean age of 34 years (range 24 -49 years); five (38%) were male and eight (62%) were female.Twelve (92%) of them had a history of BCG vaccination.Three (23%) of them had a history of contact with patients with untreated active tuberculosis; ten (23%) had a positive travel history (see above); eight (62%) were foreign-born (see above); and nine (69%) had worked in a mycobacteriology laboratory with exposure to Mycobacterium tuberculosis.
The risk factors associated with a positive tuberculin skin reaction were therefore: place of birth in a foreign country; contact with patients with untreated active tuberculosis; positive history of travel to a country with high prevalence of tuberculosis; and history of working in a mycobacteriology laboratory.Based on the subjects' positive tuberculin skin reactions, a risk factor for infection with Mycobacterium tuberculosis other than a history of working in a mycobacteriology laboratory could be identified for all probands.On average, the worldwide frequency of positive tuberculin skin test reactions is 30%, and it can be as high as 80% in particular risk groups such as those that are in contact with patients with untreated active tuberculosis, or less than 20% in low-endemicity areas such as Canada or the Netherlands 5 9 18 .The total frequency of positive tuberculin skin reactions in this laboratory was 20%, and this is comparable with previous studies on positive tuberculin skin reactions in Germany and lower than the average of 30% worldwide (Table 1) 4 5 10 11 .

Table 2 -Frequency of positive tuberculin skin tests (TST) (> 15 mm) and clinical and epidemiological data. Explanation of terms: see
The most important associated factors for infection with Mycobacterium tuberculosis and for a positive TST reaction in this study were: place of birth in a foreign country with high prevalence of tuberculosis such as India and contact with patients with untreated active tuberculosis.Most probands were TST-positive if they were foreign-born in a country with high prevalence of tuberculosis or had histories of contact with patients with untreated active tuberculosis (Table 2).Thus, either of these criteria makes it possible to identify individuals who are likely to be TST-positive (i.e.PPD-positive), e.g. for the purpose of finding probands for research on latent tuberculosis using peripheral blood samples.
Another known risk factor for infection with Mycobacterium tuberculosis is to work in a laboratory with live Mycobacterium tuberculosis, e.g. with clinical samples from patients for diagnostic testing 2 3 7 8 15 .However, there have been few reports of airborne acquired active infection with Mycobacterium tuberculosis consequent to diagnostic mycobacteriology laboratory work over the last ten years in North America and Europe.The risk of infection in a biosafety level 3 facility is similar to the risk among the general population if standard precautions are followed 3 14 .Grist et al 7 studied evidence of tuberculosis acquired during work activities in more than 200 laboratories in the United Kingdom and only identified one case, in a mortuary technician 7 .Nonetheless, infection remains a possible complication of working with live Mycobacterium tuberculosis.Employees involved in the disposal of medical waste have been found to have an increased risk of infection with Mycobacterium tuberculosis 15 .
Before the general use of safety cabinets and face masks for handling Mycobacterium tuberculosis specimens became standard practice in industrialized countries over ten years ago, the risk among laboratory personnel of acquiring tuberculosis was markedly higher and was estimated to be between 100 and 200 times greater than among the general public 6 15 .
For the purpose of identifying latently infected individuals in the laboratory, two associated factors (place of birth in a foreign country with a high prevalence of tuberculosis and history of exposure to patients with untreated active tuberculosis) are useful preselection criteria before performing tuberculin skin testing.

Table 1 -Tuberculin skin test (TST) results and clinical and epidemiological data on study probands.
14T result interpretation: according to result and clinical risk factors;14contact with patients with untreated active tuberculosis: history of contact (e.g. in workplace, school or home); positive travel history: visit to a country with high prevalence of tuberculosis (e.g.working as a health professional in Africa or living in India); foreign-born: (one in �ussia, one in China, ten in India and one in United Kingdom).BCG: Bacille Calmette-Guérin, TST: tuberculin skin test.