U. of Wis. quietly scraps risky lab equipment
By RYAN J. FOLEY
ASSOCIATED PRESS WRITER
MADISON, Wis. -- The University of Wisconsin-Madison has quietly
decided to stop manufacturing its signature aerosol chambers used for
researching infectious disease, which were involved in a few dangerous
lab accidents nationwide, including one in Seattle in 2004.
The College of Engineering is shutting down the business after an
internal audit found it was poorly managed and carried the potential for
huge liability costs in the event the chambers failed, exposing
researchers to toxic agents.
"Like any mechanical thing, it has seals and gaskets. Those seals and
gaskets can fail and then that would release these toxic agents to human
exposure. That's the risk," UW-Madison internal auditor Ed Ruotsinoja
said in an interview.
His report was obtained by The Associated Press under the open records
law.
For 30 years, the university made so-called Madison chambers after they
were pioneered in one of its labs in the 1970s. Over time, about 25
universities and companies worldwide purchased them.
The chambers are used to infect animals such as mice and guinea pigs
with everything from tuberculosis to anthrax, replicating the conditions
under which the pathogens would enter human lungs.
The federal government encouraged their use after the 2001 terrorist
attacks to develop vaccines and treatments for biological agents. But as
the research increased, accidents exposed some scientists to dangerous
substances and caught the attention of safety regulators and government
watchdogs.
Three researchers in Seattle tested positive for tuberculosis in 2004
after they were exposed in a lab. A state of Washington investigation
blamed their company for not requiring workers to wear enough safety
equipment but the chamber they were using also had cracked seals that
caused leaks, records show.
According to the Washington investigation, researchers at Colorado
State also reported leaks in their chamber but none of their employees
were infected. Then in 2006, a researcher at Texas A&M University was
infected with Brucella while she cleaned out the Madison chamber and was
sick for weeks.
Edward Hammond, former director of the Sunshine Project, a watchdog
group, uncovered those problems through public records requests. He had
long questioned the safety of the chambers and criticized the university
for failing to notify the public and customers of problems.
"I'm shedding no tears for the Madison chamber," Hammond said. "Its
track record did not look good."
The college's associate dean, Steven Cramer, said he was not aware of
any manufacturing defects associated with the chamber and that any
problems stemmed from operator error.
He said he instructed workers not to fill any more orders for the
product last spring after receiving the audit. Their shop is putting the
final touches on the last chamber, which could be shipped in coming
weeks.
"From the audit, it apeared to me there was potentially more risk than
the revenue benefit from producing the chambers," he said.
He said two employees, including the former manager of the shops, are
being let go as part of a reorganization designed to provide more
educational opportunities for students.
The business was not lucrative - it made a modest profit - but it gave
the university some prestige. From the Bronx to Bangalore, many of those
studying toxic agents used a Madison chamber.
The College of Engineering asked for the internal review of the
business just as demand was growing because of a boost in federal and
private money for bioterrorism-related work.
The review questioned whether the business could survive in an
environment of "increased opportunity and risk of field failures"
without major changes to limit the university's liability. It found
accounting problems and "a material weakness in managing the risks of
the external business sales."
Ruotsinoja said the college was selling the $40,000 chambers without
warranties requiring buyers to do routine maintenance that would have
limited liability. In at least one instance, the college inadvertently
filled a purchase order that required it to accept liability for any
problems, he said.
"Given that the Madison Chamber is used for the testing of infectious
agents, this poses a significant risk," his report warned.
Ruotsinoja said UW officials tended to fix problems as they popped up
rather than develop routine maintenance plans to prevent them.
His report recommended the college consider patenting the product and
trademarking its name and then licensing production to a company. But
the university's intellectual property arm decided not to take those
steps after failing to find a company interested and concluding a patent
wasn't possible, Cramer said.
That came as surprising and unwelcome news to researchers who depend on
the chambers. It also means they will be forced to maintain the chambers
on their own without help from the university, said Cramer, who denied
there was any attempt to keep the decision quiet.
Still, a prominent researcher and a retired professor who helped
pioneer the equipment in the 1970s both said they had not been notified
of the move.
"That's news to me," said Donald Smith, the retired medical school
professor whose lab is credited with inventing the chamber. "If you just
look at the investigators that are using the chamber now, I think they'd
be very unhappy to learn that all of a sudden this is not going to be
built anymore."
He said he had visited the university in recent months and was assured
chamber production continued, which was technically true since the final
one is being built.
David McMurray, a Texas A&M researcher who has used chambers for
decades in his tuberculosis research, said he was disappointed by the
decision.
"They've really been an incredibly valuable resource," he said. "I'm
deeply concerned that somehow, if the University of Wisconsin is no
longer going to underwrite this type of activity, they are able to find
a company willing to make the chamber and provide it to scientists who
need it." U. of Wis. quietly scraps risky lab equipment