IN OCTOBER 2003,

Kristi Williams' four-year-old son, Christian, was diagnosed with leukemia. Normally, the prognosis would have been poor, but as fate would have it, Williams was already busy creating a potentially powerful cancer fighter: Christian's brother.

Williams was five months pregnant. Christian's medical team explained to her that by saving the stem cells from her newborn's umbilical cord blood, there was a good chance that those blood-forming cells could be used to help cure her older child. In a best-case scenario, healthy stem cells could be transplanted, replacing Christian's unhealthy ones.

Noah was born in January 2004, and the Williamses banked his blood with Cord Blood Registry, a private bank based in San Bruno, Calif. It was a perfect match. In March 2004, Christian received the transplant. Today, he's a healthy six-year-old who just started first grade.

Christian's experience would seem to make an airtight argument for banking umbilical cord blood. But while the procedure holds promise for some, the price tag is high. So high, in fact, that for many parents, the choice not to do it makes much more sense.

The Appeal
Stem cells from umbilical cord blood can be used to treat a variety of childhood and in some cases, adult illnesses, ranging from certain types of cancer to rare blood disorders. The most common uses for banked stems cells are in treatments for leukemia and other blood cancers and sickle cell anemia. In most cases, the stem cells are used on a sick sibling, but in rare instances the stem cells can be used on the donor patient. And unlike fetal stem cells, there's nothing controversial about saving cord blood, since the child isn't affected.

The first stem-cell transplants were performed during the late 1980s. The New York Blood Center, which has the oldest and largest public cord-blood bank program, started collecting donated cord blood in 1992. Private cord-blood banks soon followed. As of September 2004, 280,000 mothers had paid to bank their baby's cord blood privately, while nearly 80,000 mothers had donated cord blood.

The Cost
Private cord-blood banking is expensive. It usually costs about $1,500 to $2,000 upfront to save the blood, and $100 or more a year to store it.

And many of the private banks are for-profit institutions that aggressively market their services. Expecting parents are bombarded with brochures at their doctor's office and in maternity shops. Baby magazines are full of advertisements. As one brochure puts it: "If you don't save your baby's cord blood today, you may be passing up the best medical treatment option for tomorrow."

Such heavy-handed pitches anger some doctors. "The whole field should be carefully evaluated from an advertising standpoint," says Dr. Bertram Lubin, director of medical research at Children's Hospital Oakland Research Institute in Oakland, Calif. "[People] are not given scientifically accurate information." He argues that while many of the claims in the brochures aren't entirely false, parents need a broader understanding of the issues, and shouldn't make a decision based on marketing material alone.

Not Right for Most

The American College of Obstetricians and Gynecologists and the American Academy of Pediatrics don't endorse private cord-blood banking for future use unless, as in Williams' case, there's a family history of a treatable illness. The organizations say the odds of a child ever needing his own cord blood by age 18 are as low as one in 200,000. Private cord-blood banks put the likelihood of use by age 21 at one in 2,700, or less than 0.04%.

And even if your child is diagnosed with one of the illnesses that stem cells can help treat, there's still a good chance that the child won't be able to use his or her own cells. Stem cells from a patient's own cord blood aren't used to treat genetic diseases, and are rarely used to treat leukemia, since there's a chance that those cells could develop cancer once again, says Children's Hospital Oakland Research Institute's Lubin. (Private cord-blood banks, such as the Cord Blood Registry, argue this point. They say there have been cases when a doctor has used a patient's own stem cells to treat some cancers successfully.)

And a sibling's cord blood isn't a guaranteed match. There's only a 25% chance that a sibling will be an exact match, and another 25% chance the cells will match well enough to be considered for transplantation, says Dr. Philip Lowry, associate professor at the University of Massachusetts Memorial Medical Center. Half the time, a transplantation isn't even possible.

Public facilities actually offer better odds for some patients. There's a donor network of public facilities across the country. Caucasian families of European descent have a 70% to 80% of finding a close-enough match, says Dr. Robert Krance, the director of the Pediatric Stem Cell Transplant Program at the Texas Children's Hospital. And while the chances are much lower for minorities, the medical community is actively recruiting more donors from other groups.

And recent legislation passed by Congress and expected to go to the floor of the Senate this fall would authorize the collection of 150,000 units for a national public bank, says Patrick Thompson, a spokesman for the National Marrow Donor Program. Once that happens, the likelihood that a patient will find a match across all races and ethnicities should increase greatly, he says.

"There are lots of things going on, and families shouldn't think that if they don't save their baby's cord blood and their child develops cancer he will be deprived of their only salvation," Krance says. "That's not the case."

Who Should Bank It?
All of this being said, parents who have a history of certain genetic, hematologic and oncologic diseases should at least discuss the idea of saving their baby's cord blood with their doctor, according to advice from the American Pregnancy Association. Families at high risk for such diseases might find that banking is a smart thing to do.

For most, however, the decision is largely a financial one. If money is no object, clearly it couldn't hurt to bank the blood for future unknown medical uses. But the typical middle-class family probably can't afford that kind of insurance policy. Given the small chance that the cells will ever be used, UMass's Lowry recommends foregoing the blood bank and saving that money in a health savings account for other health-related needs.