WASHINGTON -- Pet food contaminated with an industrial chemical might have sickened or killed 39,000 cats and dogs nationwide, based on an extrapolation from data released yesterday by one of the nation's largest chains of veterinary hospitals.
Banfield, The Pet Hospital, reported that an analysis of its database, compiled from records collected by its more than 615 veterinary hospitals, suggests that three out of every 10,000 cats and dogs that ate the pet food contaminated with melamine developed kidney failure. There are an estimated 60 million dogs and 70 million cats in the United States, according to the American Veterinary Medical Association.
The hospital chain cared for 1 million dogs and cats during the three months when the more than 100 brands of now-recalled contaminated pet food were sold. It saw 284 extra cases of kidney failure among cats during that period, or a roughly 30 percent increase, when compared with background rates.
"It has meaning, when you see a peak like that. We see so many pets here, and it coincided with the recall period," said veterinarian Hugh Lewis, who oversees the mining of Banfield's database to do clinical studies. The chain continues to share its data with the Food and Drug Administration.
FDA officials have said the database compiled by the huge veterinary practice would probably provide the most authoritative picture of the harm done by the tainted food.
In central Ohio, no confirmed cases of pet poisonings have been reported, although some cases are suspected.
From its findings, Banfield officials calculated an incidence rate of .03 percent for pets, although there was no discernible uptick among dogs. That suggests the contamination was overwhelmingly toxic to cats, Lewis said. That is in line with what other experts have said.
At least six pet-food companies have recalled products made with imported Chinese wheat gluten tainted with the chemical. The recall involved about 1 percent of the overall U.S. pet food supply.
Measuring the tainted food's impact on animal health has proved an elusive goal. Previous estimates have ranged from the FDA's admittedly low tally of roughly 16 confirmed deaths to the more than 3,000 unconfirmed cases logged by one Web site.
"On a percentage basis, it's not breathtaking, but unfortunately it's a number that, if it was your pet that was affected, it's too high," veterinarian Nancy Zimmerman, Banfield's senior medical adviser, said of the newly estimated incidence rate.
In another estimate yesterday, the founder of a veterinary group said 5,000 to 10,000 pets might have fallen ill from eating the contaminated food, and 1,000 to 2,000 might have died.
The estimate was based on a Veterinary Information Network survey of 1,400 veterinarians among its 30,000 members. About one-third reported at least one case, said Paul Pion, the network's founder. He cautioned that a final, definitive tally isn't possible, and that even his estimate could be halved -- or doubled.
Banfield's veterinarians treat an estimated 6 percent of the nation's cats and dogs. After the first recall was announced, the chain beefed up its software to allow those veterinarians to plug in extra epidemiological information to help track cases, Zimmerman said.
The new template allowed vets to log what a sick pet had eaten, any symptoms its owner might have noticed, the results of a physical examination, any urine and blood test results and other observations.