Sniffing Out TB

An African rat helps detect tuberculosis in Tanzania, prompting the Mozambique government to pursue a similar project.

Oct 18, 2012
Jef Akst

African pouched rat, WikimediaThe Belgian NGO Apopo is on the hunt for the life-threatening bacterium that causes tuberculosis (TB), and it is finding it using an unlikely helper—the African giant pouched rat, so named for its paunchy cheek pouches. After a bit of training, the rats can identify a distinctive TB smell—caused by organic compounds the bacterium produces in infected samples—in a fraction of a second, tackling as many as 1,680 samples per day, The Hindu reported. To put that in perspective, a trained lab tech can screen about 40 samples per day, a number that takes a pouched rat just 7 minutes.

The organization has been using the rats as a secondary screening method in hospitals, screening more than 97,000 samples and upping detection rates by as much as 23 percent. The success of the project caught the attention of the Mozambique government, which has asked Apopo to set up a similar screening effort in the country. “We are in the process of setting up a laboratory and training handlers,” Apopo representative Tess Tewelde told The Hindu. “The program will kick in by December.”

The projects add to the growing number of animals being used to detect disease. Dogs, for example, are being trained to identify and even differentiate among a variety of human cancers. Bees have also showed promising abilities in recognizing TB volatiles.

And the applications for such impressive sniffers aren’t limited to medicine. Apopo is also teaching pouched rats to detect landmines, for example. “We have 40 rats in mining operations in Mozambique,” Tewelde told The Hindu. “So far we have returned over 5.5 million square metres of land to local population and have detected 2,500 mines, 2,200 unexploded ordinances and over 12,500 small arms ammunitions in Mozambique. In 30 minutes, our rats can cover an area that can take a fast [human] de-miner over two days.”

(Hat tip to GenomeWeb.)