On Monday, March 22nd, the Food and Drug Administration did something that on its surface didn't make much sense: they recalled a rotavirus vaccine named RotaRix.

RotaRix, made from a weakened natural strain of human rotavirus, was shown to be safe and effective in pre-licensure trials of more than 60,000 infants. Sold in the United States since 2008, the vaccine has also been shown to be safe and effective post-licensure in millions of children. And rotavirus, which causes fever, vomiting, diarrhea, and dehydration in young children isn't a trivial disease in the U.S., causing about 250,000 emergency department visits, 70,000 hospitalization, and 60 deaths in the U.S. annually. In the developing world, the virus is a killer, accounting for about 2,000 deaths a day.

So why did the FDA withdraw the vaccine. They did it because they found genetic sequences suggesting that a pig virus, called porcine circovirus type 1 (PCV-1), was contaminating the cells used to make the vaccine. The detection of these genetic sequences was made possible by a highly sensitive, recently developed technique that had not previously been applied to vaccines. What made the withdrawal of the vaccine surprising was that PCV-1 has never been shown to cause disease in any mammalian species, including pigs. So the virus isn't a threat. Also, in fairness, the same technique should be applied to other items ingested by 6-month-olds, like Cheetos or fruit. Using this technique, one would invariably find genetic sequences of a variety of viruses, bacteria, fungi, and parasites, all evidence that we live in a microbe-rich environment. Although another rotavirus vaccine is available in the U.S., the FDA's action might cause a shortage of rotavirus vaccine -- in which case the FDA would have elevated a theoretical risk about a real risk and done harm.

It will be interesting to see where the FDA goes from here.