Victor Yu and Janet Stout, Legionnaires' disease experts, at Special Pathogens Laboratory in Uptown.
By Sean D. Hamill Pittsburgh Post-Gazette
When the VA Pittsburgh Health Care System first revealed on Nov. 16 that it had an outbreak of Legionnaires' disease that sickened four patients at its Oakland veterans hospital, it blamed a water treatment system that had been in use there since 1993.
In the last paragraph of its news release, the VA wrote: "Our existing copper-silver ionization system -- which reduces the presence of Legionella in water -- may not be as effective as previously thought, as is the case in other health systems using this method. Consequently, we are shifting to a chlorination system to ensure better control."
Those 45 words reignited a debate that has raged ever since Legionella, the often-deadly bacteria that causes Legionnaires', was first identified in 1977 after an outbreak at an American Legion convention in Philadelphia, and 30 years after Pittsburgh researchers first confirmed that it was transmitted through water systems: What is the best way to prevent the spread of Legionella?
As it so often is in scientific and commercial debate, the answer to the question depends on whom you ask.
Though the federal Centers for Disease Control and Prevention, which is working with the Pittsburgh VA on its outbreak, has maintained for at least 15 years that chlorination is the best prevention method in water systems, there is no unanimous answer in the broader scientific or water treatment community.
There are passionate, well-researched voices on multiple sides of the debate. Some believe in copper-silver, some in chlorination, some in chlorine-dioxide, others in monochloramine. Together these four methods are the dominant prevention technology in use today, researchers and industry officials say. All rely on releasing minute amounts of the chemicals into a building's water system to kill Legionella.
While Legionella has proved to be an amazingly persistent and resilient bacteria, it was not hard to find ways to kill it. The hard part has been killing it over the long term, and in old, corroded water pipes filled with biofilm that hides the bacteria, as the Pittsburgh VA discovered. Reaching agreement on what is the best way to do that has been even harder.
"The available data is still very limited and the various manufacturers all believe theirs is best," said Glenn Morris, director of the University of Florida's Emerging Pathogens Institute. "The type of national study to look at all these various systems has simply not been done."
William McCoy, who chairs the American Society of Heating, Refrigerating and Air-Conditioning Engineers committee, disagrees. The committee is in the midst of creating new standards for dealing with Legionnaires' in buildings.
"I'd say [the national study] has already been done," said Mr. McCoy, who is chief technology officer for Phigenics, a Naperville, Ill., water treatment company. "All four of [the dominant methods] have been shown to work properly if properly managed."
Dr. Morris, who headed a committee in 2000 in Maryland that studied the issue for more than a year and created state guidelines for hospitals to deal with Legionella, places a big part of the confusion that still reigns squarely on the CDC, which participated in Maryland's study and is the agency most look to for answers on questions like this.
"The CDC feels like it has all the answers," he said. "But there isn't any motivation on the part of the CDC to do more studies. They think they've done them."
The VA's outbreak
After the Pittsburgh VA realized in late October that it may have a Legionnaires outbreak, it reached out to the Allegheny County and the Pennsylvania health departments.
In response, the state asked the CDC to assist the VA in analyzing the problem, Pennsylvania Health Department spokeswoman Kait Gillis wrote in an email.
On Nov. 16, after the CDC completed its field work in Pittsburgh, the Pittsburgh VA reported it had four Legionnaires cases in which patients contracted the disease, a former of pneumonia, from inside its University Drive hospital in Oakland. All four patients recovered.
But six days later, on Thanksgiving day, the VA reported it had a fifth case. A day later that patient, William Nicklas, 87, of Hampton, a World War II veteran, died.
In addition, two other families say that men in their families -- John McChesney, 63, of Columbus, Pa., who died Oct. 23, and John Ciarolla, 83, of North Versailles, who died July 18, 2011 -- also died after contracting Legionnaires' following stays at University Drive. But it is not yet clear if they contracted it in the hospital or outside.
In an interview with the Pittsburgh Post-Gazette the week after Thanksgiving, CDC's Legionnaires' team explained its role in the investigation and, in the process, reiterated the CDC's position on the two hotly contested issues of what system is best and whether hospitals should test their water regularly for Legionella.
CDC's epidemic intelligence service officer Alicia Demirjian, who came to Pittsburgh in November to help the VA, said in a telephone interview: "We believe the best way to treat the water is chlorination. It seems like chlorination has been safe."
As for copper-silver ionization systems, like the one the Pittsburgh VA had in place since 1993, she said: "There's not that much information out there to say how effective it is in controlling Legionella."
At the time, though, Dr. Demirjian said she did not yet have enough evidence to say what went wrong with the Pittsburgh VA's copper-silver system, whether the system itself failed to work properly, whether it was never effective, whether it was poorly maintained or whether the outbreak was the result of the VA's new construction at the hospital.
New construction and renovation are frequently associated with Legionnaires outbreaks because during such projects, water systems can be turned off or not used for a while, allowing water to stagnate and Legionella bacteria to grow.
"The investigation is not over yet," Dr. Demirjian said.
Her colleague on CDC's Legionnaires team, Lauri Hicks, a medical epidemiologist, said during the same interview that what they do know is that the copper-silver system at the VA "clearly didn't work, whether it was functioning properly or not."
Dr. Hicks said a building's operations staff has to be diligent to keep the disinfectant in the water at the right levels "no matter what kind of system you're dealing with" -- copper-silver or chlorination system.
As for testing water regularly for Legionella, Dr. Demirjian said the CDC basically recommends: "Not for Legionella, unless there are outbreaks."
Dr. Hicks said: "The challenging thing here is with sampling. If you find [your water] isn't clear of Legionella, what to do about it if you find it becomes the issue. It's not like we can go around eradicating Legionella from every water supply."
"There are many examples of how testing water proactively can be problematic," she said before pointing out that "the VA was sampling for quite some time" before its outbreak.
Asked more recently if these and other statements are the CDC's official position, CDC spokesman Thomas Skinner would not comment and simply sent a Web link to a CDC report, "Guidelines for Preventing Health-Care-Acquired Pneumonia, 2003."
That report restated the same recommendations it made in a 1997 paper that chlorination was the best method and that it had no recommendation on regularly testing hospital water for Legionella.
As to other treatment methods, the 9-year-old CDC paper concludes: "Alternative methods for controlling and eradicating Legionella spp. in water systems, such as treatment of water with chlorine dioxide, heavy metal ions (i.e. copper-silver ions), ozone, or ultraviolet light have limited the growth of Legionella spp. under laboratory or operating conditions. However, more data are needed regarding the long-term efficacy of these methods."
Pittsburgh researchers controversy
In 2003, the same year the CDC last put out its official position on how to deal with Legionella in hospitals, Victor Yu and Janet Stout, researchers who then worked at the Pittsburgh VA, published a paper in the Infection Control and Hospital Epidemiology journal.
In it, they tracked the long-term impact of use of copper-silver ionization systems in controlling Legionella in 16 hospitals across the country -- including six in the Pittsburgh area: Children's of Pittsburgh, St. Clair, West Penn, UPMC Mercy, Magee-Womens of UPMC and UPMC St. Margaret.
Five of those hospitals had previously used chlorination and had problems with it, they reported. But the 16 hospitals, which all had used copper-silver for five to 11 years each, reported just one case of Legionnaires while using that method.
It was because of that study that they began full-heartedly recommending copper-silver to clients in most cases.
As excited as they were about the paper, it fell with a thud at the CDC's offices in Atlanta.
"The CDC ignored [the paper] as if it never occurred," said Dr. Yu, who, along with Dr. Stout, left the VA after a dispute with management in 2006 and 2007, respectively.
The CDC would not comment about its opinion of the paper.
But Tim Keane, owner of Legionella Risk Management, a consulting company in Chalfont in southeastern Pennsylvania, believes the reason the CDC didn't place weight on Dr. Yu and Dr. Stout's paper is that it shows not that copper-silver worked, but that it failed.
"That paper written by Stout and Yu is the best example of why not to use copper-silver," he said. "They say that almost half those hospitals had around 30 percent positive Legionella samples. And they say it works. According to me, it's a failure. I want results around 5 percent" positive Legionella samples at water sources.
Mr. Keane's disagreement with Dr. Yu and Dr. Stout isn't just scientific, however. He, like many in the industry, say these doctors and other researchers and consultants have a bias because they benefit by advocating for one system or another.
"I know [Dr. Yu and Dr. Stout] have done seminars advocating for copper-silver," said Mr. Keane, who advocates for the use of chlorine-dioxide systems.
Dr. Yu and Dr. Stout said they do often recommend copper-silver to clients who ask their advice because they believe it has been proven effective, but they don't sell the systems.
"And we're now doing the same research on chlorine-dioxide" that they did with copper-silver, said Dr. Stout. She and Dr. Yu now work in their own, private lab, Special Pathogens Laboratory in Pittsburgh. "And we're close to recommending the use of chlorine-dioxide more widely."
They don't recommend the use of chlorination, as the CDC does, because of research and experience that regular use of higher levels of chlorine in water pipes can result in corrosion that can be costly to repair.
They have recommended the use of chlorine-dioxide to a few clients. And last year they worked with an Italian company to install what is believed to be the first commercial use of monochloramine in the United States, at UPMC Mercy.
Monochloramine is used in about 25 percent of municipal water plants as a replacement for chlorine, but, because it has to be mixed on site and can be volatile, hospitals have been reluctant to use it.
After a year of use, the monochloramine system has worked well and "looks promising" for more general use, said Dr. Yu, who, with Dr. Stout, was part of the team that concluded in 1982 that Legionella is transmitted through water systems. "Many people are watching this trial."
Joe Hannigan, president of Klenzoid, a water treatment company in Philadelphia, is the North American distributor for the monochloramine system that was installed at UPMC Mercy.
He has long believed that chlorine-dioxide was the best way to prevent the spread of Legionella and worked with Dr. Yu and Dr. Stout to install one of those systems at Geisinger Medical Center in Danville, Montour County.
He doesn't like to use copper-silver because it is not yet fully approved by the U.S. Environmental Protection Agency.
But if the debate over what is the best way to deal with Legionella was reignited because of the deaths of as many as three veterans, perhaps that is the silver lining to come out of the episode at the Pittsburgh VA, said Dr. Morris, of the University of Florida.
"If this could lead to the questions being revisited and initiation of some national studies that could draw real conclusions, that would be the best thing possible," he said. "I remain optimistic."