When — not if: CDC stepping up plans for global flu pandemic

ICPs tapped for critical roles in meeting the threat

Spurred by the "near-miss" of a global influenza pandemic in Hong Kong in 1997 — which caused fatal infections in otherwise healthy people and was transmitted to at least two health care workers — public health officials are nearing completion of a pandemic plan for the United States that includes new infection control guidelines for medical settings.

Developed by advisors and officials at the National Vaccine Program Office (NVPO) of the Centers for Disease Control and Prevention, the draft guidelines obtained by Hospital Infection Control outline specific infection control measures for health care facilities during an influenza pandemic. (See recommendations, p. 77.) The risk of transmission in a pandemic will be high because there will be little, if any, immunity within the population during the months it takes to develop and implement a vaccine. As opposed to the seasonal antigenic "drift" of flu strains, a pandemic could occur if an influenza A virus took a more dramatic "shift" and mutated into a virus that could spread rapidly without vaccine or existing immunity in its path. (See related story, p. 75.) While the 1918 global influenza pandemic has taken on an almost mythical status as an infectious disease, pandemics also occurred in 1957 and 1968.

The new CDC draft guidelines estimate that the next flu pandemic will cause between 89,000 and 207,000 deaths in the United States alone. In addition, U.S. facilities will be beset by between 314,000 and 733,000 hospitalizations and between 18 million and 42 million outpatient visits.

Though pandemic discussions often take on the tone of "what-if" bioterrorism scenarios, epidemiologists regard the emergence of a pandemic strain of influenza as inevitable.

"Almost everyone who works in influenza feels that there will be another pandemic," says Carolyn Bridges, MD, a medical epidemiologist in the CDC influenza branch. "It is just a matter of when."

In response to such pandemic threats, the CDC and its advisors and working groups have mapped out six sections of the overall plan, including three that are now in the comment and review phase. In addition to the infection control guidelines highlighted in this issue, pandemic draft plans also have been completed on health care resource management and use of nontraditional settings for delivery of medical care. The three other guidelines in development include allocation of a limited vaccine supply, triage and care of patients who develop pneumonia during a flu pandemic, and guidelines for use of antiviral drugs. "Those will be [released], we hope, in the next three to six months," says Martin Myers, MD, acting director of the NVPO. The six documents will likely be compiled into one plan eventually, he says. There also is a state planning guideline and other pandemic materials on the NVPO Web site, but the aforementioned drafts had not been posted there as this issue went to press. (See editor’s note at the end of this article for information on obtaining the drafts and submitting comments.)

As a part of pandemic planning, communities and health care organizations will need to have in place special infection control guidelines that take into account the likelihood that a high proportion of the population will be affected and that secondary infections are a major source of morbidity and mortality related to influenza virus infection, the draft warns. With an increased number of people seeking care, resources traditionally used for infection control may be in short supply. In that regard, the CDC is interested in comments from ICPs on the feasibility and appropriateness of its draft pandemic measures.

"The infection control practitioner is going to be a critical component of a pandemic response," Myers says. "We are asking for [ICP] input." For example, the CDC draft questions whether using masks for infection control during a pandemic will become somewhat impractical. There are unresolved questions about who among staff, patients, and community members will be infected and how long one would have to be masked to prevent exposures. "In the middle of a pandemic, masks in a hospital may not make a whole lot of sense," Myers says. "They still make sense for other purposes, but they may not make a whole lot of sense for trying to contain influenza. And we don’t think masks out in the community will work [either]. These will be major issues, and we would really like [ICP] input as to whether these things are practical."

Photographs from the 1918 influenza pandemic show people wearing surgical masks in public, but Myers notes, "I don’t think anybody thinks it had any effect at all, yet you could be arrested in San Francisco if you didn’t have your mask on. We’re trying to come up with guidelines so that we avoid some of those [situations]." While concurring that the issue does raise some questions of practicality, an ICP who reviewed the draft for HIC says masking of workers treating known flu patients or cohorts would still make sense.

"Masks for droplet precautions are a very basic standard, which [the guidelines] point out," says Patti Grant, RN, BSN, MS, CIC, director of infection control at RHD Memorial Medical Center in Dallas. "If nothing else, wearing a [standard surgical] mask would protect two of the mucous membranes that we have on our face from frequent touching. A lot of times, we touch our nose or our mouth without even thinking about it. I guess if you run out of masks, you run out of masks. But until that happens, I don’t see why you would not make that part of the mandatory protocol."

The draft also notes that during a pandemic, when hospitals are likely to be overwhelmed and many staff sickened, it may be necessary to adopt policies, for example, allowing workers with flu to care for flu patients. "You have to weigh the risks and benefits of patient care," Grant says. "It really needs to be a very flexible policy for infection control and employee health so they can balance the risk of sick health care workers with the risk of minimal [staffing] of health care workers."

While concurring that infection control will play a critical role in implementing such policies during a pandemic, Grant says, many experienced practitioners fortunately are already familiar with the basic principles outlined in the guidelines (e.g., cohorting and use of barrier precautions). Still, it’s important to have a document that complements and centralizes existing infection control knowledge prior to the chaos of a pandemic. "If this were to happen, God forbid, next winter — before we have the whole plan ready to go — the majority of [ICPs] would know how to deal with it," she says.

Many flu experts now think the world dodged a pandemic bullet in late 1997 in Hong Kong, where 18 people became the first known humans infected with the avian H5N1 influenza A strain. While the primary cases were thought to be infected by exposure to poultry, researchers recently confirmed that secondary transmission occurred to two health care workers treating infected patients.¹ (See related story, p. 79.)

While the health care workers had mild infections, the threat of human transmission increases the likelihood of co-infection and subsequent mutation with a common circulating flu virus. That scenario is what flu experts say could lead to a highly transmissible, deadly flu strain. Indeed, many now see the H5N1 incident as a near-disaster, as the virus was stopped in part by a massive poultry slaughter right before typical seasonal flu strains began circulating that year.

H5N1 remains out there’

"I think it is entirely possible that it could have mushroomed, particularly with the influenza season approaching in Hong Kong and the increased possibility of co-infection with the H5 and the human influenza viruses," Bridges says. No additional human cases have been detected with H5N1, but public health officials are wary of its re-emergence. "You can’t get rid of H5N1 in the environment because it is in the wild bird population," she says. "It remains out there."

Noting that the H5N1 situation was a "near miss" in terms of a global pandemic, another influenza expert says a disturbing aspect of the case is that many of those infected had severe or fatal infections despite the absence of influenza risk factors. Indeed, of the six deaths, some occurred in otherwise healthy adults, a situation reminiscent of accounts of the 1918 pandemic. Overall, five of the 18 infections occurred in adults ranging in age from 19 to 37 years old.

"They weren’t high-risk. That’s [like] 1918, " says Arnold Monto, MD, professor of epidemiology in the school of public health at the University of Michigan in Ann Arbor. Although not wanting to criticize the ongoing efforts of public health officials, Monto questions whether adequate support and priority have been given to pandemic planning in the wake of the H5N1 avian flu incident. "To me, it’s astonishing," he says. "I thought if there was any good that would come out of the Hong Kong situation, it would be that we would all be aware of the extreme dangers of a pandemic. Even though it is totally unpredictable when this is going to happen, it really will happen at some point. And we had better be prepared."

Moreover, another avian influenza A virus in Hong Kong — H9N2 — also infected humans for the first time last year.² Two people were infected, both of whom survived after hospitalization. Five additional cases were reported from mainland China but have not been confirmed by the CDC. The CDC did not find any additional transmission in Hong Kong after sending a team over in the spring of 1999. "Some of the internal genes of the H9N2 virus are similar to those from the H5N1 virus," Bridges says. "They have some similar properties, and we know there is a lot of H9 in the bird population and possibly in some domestic poultry. So it is not just H5 we have to worry about. There is no human immunity for this one either."

With ongoing surveillance a cornerstone of pandemic planning, the CDC has distributed test kits to its sentinel labs and those working with the World Health Organization that include reagents for detection of both H5N1 and H9N2 strains. "There are also labs working on vaccine for these viruses," Bridges says. "There are no plans to incorporate [that] into a vaccine that we recommend distributing and inoculating people with, but [we need] to have the technology ready to develop [vaccine] in case these viruses come back in force."

How long would it take to produce and distribute vaccine from the first appearance of a pandemic strain? "I think the logistics would be extremely difficult," Bridges concedes. "In the best-case scenario, around six to eight months for the U.S. population, to get it distributed and into people’s arms. Antivirals would probably end playing a very large role, for treatment and prophylaxis."

Four influenza-specific antiviral agents are currently licensed in the United States. Amantadine and rimantadine are approved for treatment and prophylaxis of influenza A infections. Zanamivir and oseltamivir are approved for the treatment of influenza and have activity against both influenza A and B viruses; however, these drugs are not currently approved for prophylaxis, the draft guidelines explain. Amantadine and rimantadine have been shown to prevent infection in 70% to 90% of patients when used prophylactically. "Vaccine would be preferable," Bridges says. "All of the antivirals are associated with some side effects, and if you needed to protect someone throughout the whole season, you would have to be taking prophylaxis for a long time. And that may not be feasible, depending on the antiviral drug supply. Vaccine obviously would be the best bet."

In that regard, CDC is working with the National Institutes of Health and the Food and Drug Administration to develop a library of "seed" viruses to expedite vaccine development should a pandemic strain emerge. "One of the differences between a pandemic and another natural disaster is that you will get a warning — anywhere from weeks to months," Myers says. "If we already had a seed virus, it would shorten the time to make the new vaccine."

However, during a pandemic, it is likely that the virus will be identified in the United States before enough vaccine for the entire population has been produced. If that is the case, it will be necessary to prioritize distribution of vaccine in such a way as to reduce morbidity, mortality, and social disruption, the draft states. Guidelines for determining priority groups for vaccination in the situation of an initial vaccine shortage are being established, but will likely include health care workers essential to patient care and the running of a health care institution as well as other essential community workers and those at high risk for medical complications.

Health care institutions should consider in advance who are the most essential staff and develop plans to immunize them in accordance with state and local pandemic plans, the CDC draft recommends.

"Health care workers would be a very important group," Bridges says. "Obviously, morticians are generally fairly high on people’s list as well. In 1918 they had a very difficult time keeping up with coffins and burials."

(Editor’s note: The three draft pandemic flu guidelines mentioned in this story have been posted on the American Health Consultants Web site at www.ahcpub. com/online.html. Click on "Hot Topics" and then "Breaking News." Infection control professionals can comment on these proposed guidelines to the CDC via e-mail by going to www.cdc.gov/od/nvpo/. Click on "Enter NVPO" and then on "Contact Us.")

References

1. Bridges CB, Katz JM, Wing HS, et al. Risk of influenza A (H5N1) infection among health care workers exposed to patients with influenza A (H5N1), Hong Kong. J Infect Dis 2000; 181:344-388.

2. Peiris M, Yuen KY, Leung CW, et al. Human infection with influenza H9N2 (Research letters). Lancet 1999; 354:916-917.