Vaccinate all susceptible hospital workers for chicken pox

CDC recommends new varicella vaccine for HCWs

Deciding that the benefits of chicken pox immunization outweigh the risk of using a live virus vaccine in hospitals, the Centers for Disease Control and Prevention in Atlanta is recommending vaccinating all health care workers susceptible to varicella zoster virus (VZV).1

The recommendations represent the first statement by the CDC Advisory Committee on Immunization Practices (ACIP) on the Varivax vaccine, which is manufactured by Merck and Company, Inc. in West Point, PA. Licensed in March 1995, the vaccine is approved by the U.S. Food and Drug Administration in Rockville, MD, for use in healthy people at least a year old.

A highly contagious disease caused by VZV, chicken pox causes a systemic infection that usually results in lifetime immunity. Though most cases occur in childhood, susceptible adolescents, adults, and immunocompromised people usually have more severe disease than children and are at higher risk for complications.

While some infection control professionals have already adopted policies regarding the immunization of health care workers, others have been awaiting the ACIP recommendations because use of the vaccine raises some difficult issues in health care settings. There are questions and caveats about use of a vaccine that may not confer immunity in some 10% to 30% of health care workers. (See related story, p. 112.) There is also some risk that vaccinated health care workers could transmit the vaccine strain of the virus to at-risk patients, such as those with HIV infection or otherwise immunocompromised.

Another unresolved issue is when long-term immunity may wane in those vaccinated, and whether booster doses would then be appropriate. To address such issues, and determine the need for future changes in the vaccine, post-marketing surveillance is being conducted by both the manufacturer and the CDC.

In the interim -- citing the greater risk of nosocomial transmission of wild viral strains and the possibility of ensuing severe VZV infections in immunocompromised patients -- the CDC committee recommended vaccination of "all susceptible hospital personnel." Immuno-compromised people and pregnant women should not be vaccinated. Women should avoid becoming pregnant within one month of vaccination, ACIP recommended. (See recommendations, p. 113.)

Despite the lingering questions and policy issues, the ACIP endorsement gives ICPs a formidable new weapon against chaotic hospital outbreaks of VZV, says Willian Schaffner, MD, ACIP liaison committee member representing the American Hospital Association, and chairman of the department of preventive medicine at Vanderbilt University School of Medicine in Nashville, TN.

"Nosocomial [VZV] transmission is a matter that vexes every hospital that takes care of children through adolescents," he says. "We have introductions and threatened introductions every year."

A VZV immunization program is already under way at the university hospital, with occupational health staff immunizing all new, susceptible employees, he adds. A priority among current employees is immunizing those in high-risk areas for VZV transmission, including emergency rooms and pediatric wards, he says.

"The easiest strategy is to ask people whether they have had varicella," Schaffner notes. "If they say they have, that has a high predictive value so you don't have to test those people. Then you just test all of the negatives and the 'don't knows.' You will discover about 60% of those people are genuinely immune, and then you just vaccinate the susceptibles."

Even though the vaccine is not 100% effective, widespread use should eliminate many situations where an infected health care worker has contact with other workers and patients throughout the hospital before varicella infection is suspected. Such situations have historically sent infection control and employee health services scrambling to identify exposed susceptibles, particularly other health care workers who may need to be furloughed before they infect others.

"Varicella introductions in a hospital create a terrific amount of turbulence and consume an enormous amount of time on the part of occupational health, infection control, and the laboratory," Schaffner says. "Varicella vaccine has the potential to reduce that terrifically."

The vaccine is composed of the Oka strain of live, attenuated VZV originally isolated in the early 1970s from the vesicular fluid of a child with natural varicella in Japan. The limited data available suggest that vaccinated people have a small but real risk for transmitting the vaccine virus for a few weeks after immunization, and that risk increases in vaccinees who develop a varicella-like rash, ACIP warned.2-4 In clinical trials of the vaccine, non-localized rash with a median of five lesions developed in 5.5% of immunized adolescents and adults.

"On the basis of available data, the risk appears to be minimal, and the benefits of vaccinating susceptible health care workers outweigh this potential risk," ACIP concluded. "As a safeguard, institutions may wish to consider precautions for personnel in whom rash develops. ... Vaccinees in whom vaccine-related rash develops, particularly health care workers and household contacts of immunocompromised people should avoid contact with susceptible people who are at high risk for severe complications."

Such a "vaccine-rash" policy has already been implemented at Santa Clara Valley Medical Center in San Jose, CA, says Donna Haiduven, RN, MSN, CIC, infection control supervisor at the center.

"If [a vaccinated health care worker] gets lesions they are sent to employee health," she says. "If it's one lesion localized at the site they are not restricted. If it is more than one lesion that would be considered a generalized 'rash,' and they would be sent home."

One lesion at the site is considered a local reaction to the vaccine rather than a potentially contagious rash, so the employee would be able to continue working with the lesion site covered. The policy is working well at the hospital and is preferable to removing all vaccine recipients from work for a post-immunization period that could go out beyond two months, she says.

"We consider the post-vaccination period to be from the first dose until one month after the second dose," she says. "The ACIP recommendation is one dose and then after four to eight weeks a second dose. So that post-vaccination period could be quite long."

The question of managing vaccinated workers is just one of many raised in implementing such programs, and some ICPs may prefer to simply restrict vaccinees during the immediate post-vaccination period from working with immunocompromised patients or others at risk of severe complications should they become infected, she notes. (See key questions, p. 110.)

In recommending across-the-board immunizations of health care workers, ACIP takes a stronger stance on use of the vaccine than that recommended in the package labeling guidelines issued by Merck and approved by the FDA. In response to questions on the issue by Hospital Infection Control, Merck provided package insert information last revised on March 29, 1996 that states the following:

"Post-marketing experience suggests that transmission of vaccine virus may occur rarely between healthy vaccinees who develop a varicella-like rash and healthy susceptible contacts. Transmission of vaccine virus from vaccinees without a varicella-like rash has been reported but has not been confirmed. Therefore, vaccine recipients should attempt to avoid, whenever possible, close association with susceptible high-risk individuals for up to six weeks. In circumstances where contact with high-risk individuals is unavoidable, the potential risk of transmission of vaccine virus should be weighed against the risk of acquiring and transmitting natural varicella virus."

Inquiries to Merck did not result in any additional information on the label warning, but an FDA official clarified that the language was not designed to specifically caution against vaccinated health care workers treating susceptible immunocompromised patients for up to six weeks post-immunization.

"It's not directed at them specifically," says Lorrie Harrison, spokeswoman in the FDA Center for Biologic Evaluation and Research, which regulates vaccines at the agency. "It basically just says you have to weigh the benefits and risks and don't come in contact with folks who are at increased risk."

The FDA cannot comment specifically on the ACIP recommendations, but it is not uncommon for labeling stipulations to be somewhat out of sync with advisory committee recommendations, she adds.

"Unfortunately, it is not in harmony all of the time," she says.

Schaffner expressed disappointment at the "extremely conservative" labeling language and notes the difference in emphasis among the CDC and FDA is regrettable.

"[Merck] has highlighted the potential risk of transmission in the settings such as hospitals, where there might be immunocompromised individuals," he says. "That's clearly to minimize their own legal liability."

The difference in emphasis and language between ACIP and the FDA comes at time when the committee hopes to harmonize the package inserts with the ACIP statements, he adds.

"Here comes a brand new vaccine that is very useful, and it has a major glitch -- a major difference in emphasis in the package insert and the ACIP statement," he says. "Most of us on the ACIP think that is very unfortunate."

The essence of the difference appears to be that the label language allows for the possibility of transmission from vaccinees who do not develop a rash, while ACIP suggests using the appearance of rash in the immunized as a trigger for policy interventions. In recommending health care workers' immunization without some blanket stipulation for furloughs from high-risk patient care, ACIP reasoned that the risk is very low, and any infections that occur with the vaccine virus will be substantially less severe than that caused by the "wild" virus.

"You would expect an exceedingly mild disease in the recipient, and even then, that infection could be treated with acyclovir," Schaffner says. "The essential point is that ACIP believes that the risk of transmission is exceedingly low. The importance of eliminating transmission in the health care setting takes precedence over the exceedingly small risk of [vaccine-related] transmission."

Despite the ACIP statement encouraging use of varicella vaccine, the warning in the package insert may give hospitals pause in implementing programs, he concedes.

"The package insert clearly has been a barrier to hospital programs until now and I think it will still be a relative break on the system, but with the appearance of the ACIP statement I would hope that now a number of hospitals will reconsider and gather the resolve to initiate their own immunization programs."

The policy at Vanderbilt calls for vaccinated employees who develop a varicella-like rash to report to occupational health, where they will be evaluated and furloughed if necessary, he says.

"We are ready to accept the occasional -- we really think it is going to be very rare -- case of vaccine-associated varicella," he says. "We are prepared to deal with that and we don't think it is a big deal."

References

1. Centers for Disease Control and Prevention. Prevention of varicella: Recommendations of the Advisory Committee on Immunization Practices. MMWR 1996; 45(No. RR-11):1-36.

2. Weibel RE, Neff BJ, Kuter BJ, et al. Live attenuated varicella vaccine: efficacy trail in healthy children. N Engl J Med 1984; 310:1409-1415.

3. Diaz PS, Au D, Smith S. Lack of transmission of the live attenuated varicella vaccine virus to immunocompromised children after immunization of their siblings. Pediatrics 1991; 87:166-170.

4. Tsolia M, Gershon AA, Steinberg SP, et al. The National Institute of Allergy and Infectious Diseases Varicella Vaccine Collaborative Study Group. Live attenuated varicella vaccine: Evidence that the virus is attenuated and the importance of skin lesions in transmission of varicella-zoster virus. J Pediatr 1990; 116:185-189. *