Japan's blood tests missed HIV in at least one case. Credit: EPA Photo/EPA

Even as the list of emerging pathogens continues to grow, governments are struggling to keep their blood supplies clear of the most pressing known threats. The latest scare comes from Japan, where health officials in December announced that a patient had become infected with HIV after receiving a blood transfusion.

“We were shocked,” says Kazunari Tanaka, a spokesman for the Japanese health ministry's Blood and Blood Products Division. “We want to take blood safety to a higher level.”

Although the donor's blood had tested negative for HIV in May 2003, the case was discovered when a more sensitive test in November revealed his blood is HIV-positive. In December, the recipient tested positive for an HIV strain identical to that found in the donor.

The current case is the first since a new, more sensitive, system—the nucleic acid test (NAT)—was put in place in 1999 as a second line of detection. Because the NAT amplifies viral DNA, it can detect small amounts of virus. Of the 315 samples that have tested positive for HIV since 2000, the NAT detected the virus in 8 samples that had cleared other tests. Although several other nations, including the US and many European countries, use the NAT to screen for HIV, Japan is the only country to use it to screen samples for both hepatitis B and C.

Despite the NAT's promise, a Japanese health ministry investigation reported last summer that blood tainted with HIV and other viruses might have slipped through the system. Soon after, Japan approved its first law regulating blood supply. In December, the ministry said that hepatitis C–infected blood drawn in 2000 had been given to at least one patient.

There is some question over whether such cases are underreported, but the recent case of HIV and one possible case of hepatitis C are the only ones in some 8 million transfusions over the last four years, maintains Tanaka. By comparison, the risk for HIV transmission through transfusion is 1 in 3 million in the US, and one in several thousand in China and Thailand.

Japan uses the NAT on a pool of 50 samples of donor blood; the US screens pools of 16 samples. If a batch tests positive, samples are tested individually, but the more samples pooled together, the less likely tests are to detect viruses.

Testing individual samples is ideal, “but both cost and logistics are problems for single-unit testing,” says Harvey Klein, head of the Department of Transfusion Medicine at the US National Institutes of Health. Technologies being developed at the American Red Cross might solve these problems, he adds.

The Japanese Red Cross, which now manages Japan's blood supply, insists that diluting samples by pooling only 20 samples would still not have picked up the virus—but it has not done those tests. Under orders from the health ministry, however, the organization plans to begin using 20-sample pools within six months.

The fact that every virus requires a different test makes comprehensive testing a costly proposition. Each new pathogen, such as the West Nile virus, adds to the burden, says Tanaka. The ideal solution, he says, would be to kill or maim pathogens in one fell swoop.

One promising alternative is adding chemicals such as amotosalen, which damages bacterial and viral DNA without harming red blood cells, platelets and plasma. The treatment, known as pathogen inactivation, would also be effective against emerging pathogens, but adding chemicals to blood could have unanticipated long-term effects. The method is already in use for platelets in Norway, but no adverse reactions have yet been seen.

Japan is especially anxious to find a comprehensive solution to the problem, as it is the only developed nation with a rising HIV infection rate (see page 215). Between 1987 and 2003, the proportion of donor samples testing positive for the virus shot up from 0.13% to 1.55%.