Intended for healthcare professionals


Controlling chickenpox in hospitals

BMJ 1997; 314 doi: (Published 04 January 1997) Cite this as: BMJ 1997;314:4

Vaccination may be the way forward

  1. E M Jones, Senior registrar in microbiologya,
  2. D S Reeves, Consultant microbiologista
  1. a Southmead Health Services NHS Trust, Westbury-on-Trym, Bristol BS10 5NB

    Chickenpox seems to be affecting more older people.1 2 3 More adult patients are being seen in hospital. The reasons for this are not clear, although Ross and Lantos have suggested that the explanation may simply be the increasing number of immigrants from tropical countries, where fewer adults have immunity to chickenpox.4 Whatever the reasons for the trend, more adult patients with chickenpox are being seen in hospitals. Chickenpox is highly contagious from two or three days before the rash appears until the lesions crust. Fluid from the vesicles of shingles is also infectious. Chickenpox is potentially a very serious illness in adults, in pregnancy, and in patients who are immunosuppressed (including those taking corticosteroids5). Because the disease is transmissible before the rash appears it poses particular problems for the tracing and subsequent management of patients' contacts and also staff contacts.

    The new edition of Immunisation against Infectious Disease 19966 recommends that people at increased risk of severe varicella zoster infection who are exposed to chickenpox or herpes zoster should be tested for antibodies to varicella zoster. If they have no detectable antibodies, human varicella zoster immunoglobulin should be issued. Given within 10 days of exposure, this can at least attenuate the illness in contacts. However, the immunoglobulin is scarce, and acyclovir may therefore have to be given on occasion. Neither alternative necessarily prevents the disease entirely.

    Staff contacts who are not immune must be identified because they could transmit the disease to vulnerable patients while incubating the disease themselves. Current standard practice is that non-immune staff are excluded from work from the 10th to the 21st days after exposure to avoid transmission. Prophylactic acyclovir has been used to allow the member of staff to remain at work,7 but this carries a risk of transmission if the rash breaks through despite the acyclovir.

    These infection control measures may seem straightforward, but in practice tracing contacts and obtaining blood samples for antibody testing can be time consuming and expensive. Even the most rigorous efforts will inevitably miss some contacts, either because people are unaware of the risks or unaware that they have been in contact with the disease. Secondary cases may then occur with further risk of transmission. The measures required to contain outbreaks may be costly: 166 person days of work were lost in a chickenpox outbreak in Brisbane, Australia,8 and 82 incidents of nosocomial chickenpox in a 30 month period in Britain resulted in 12 secondary cases, £20 000 worth of VZIG being used, susceptible staff being moved from contact with patients, and elective admissions being restricted to those with a history of chickenpox.2

    The rising incidence of chickenpox in our hospitals and the continued risk posed by inpatients with herpes zoster have led to calls for a targeted vaccination policy. A vaccine against chickenpox has been available for over 20 years. Developed and used extensively in Japan, it has also been licensed in some European countries, and in March 1995 the United States Food and Drug Administration approved its use in persons aged 12 months or over who have not had varicella.9 Paradoxically, the guidelines for use of the vaccine endorsed by the American Academy of Pediatrics specifically state that it should not be given to children with immunosuppressive disorders. This contrasts with the targeted approach in European countries, where the vaccine is specifically given to such children. The United States authorities delayed approval of the vaccine because of two concerns: that the immunity might decline with age and so lead to a large increase in adult infections and that reactivation of vaccine virus could result in herpes zoster in later life. Similar concerns have also been expressed in Britain,10 where doubts have also been voiced about universal immunisation on the grounds of a low perceived benefit for the individual child.

    Britain has yet to license a varicella vaccine for any indication, though it is available on a compassionate, named patient, basis from Smith Kline Beecham and Pasteur Mérieux MSD.Immunisation against Infectious Diseases 1996 now goes as far as suggesting that immunisation should be considered for immunosuppressed patients at long term risk of chickenpox. However, the data on the immunogenicity, safety, and efficacy of the vaccine collected for over 20 years have shown that its long term efficacy seems to be good.11 Studies of use of the vaccine in children with renal transplants have also been favourable.12 Almost certainly, therefore, lives could be saved by use of the vaccine in certain groups of children or adults for whom chickenpox could be a fatal illness.

    Vaccination of non-immune hospital staff would protect both patients and staff as well as simplify infection control measures, but this suggestion is more controversial. More studies are needed since most of the data so far available are from experience with children. The safety and efficacy of the vaccine in adults needs to be confirmed, and information is also needed on the incidence of mild, but possibly infectious disease after vaccination. Nevertheless, the available data support the targeted vaccination of patients and health workers who are at risk. This would reduce the risk of the serious threat of severe varicella zoster infection in immunosuppressed patients. Such a policy would also reduce the costs of controlling outbreaks and days lost from work.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.