Article Text

Varicella: to vaccinate or not to vaccinate?
  1. ANNE GERSHON, Columbia University College of Physicians and Surgeons
  1. 630 West 168th Street
  2. New York, USA

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Despite the fact that live attenuated varicella vaccine was developed about 25 years ago,1 many questions remain about its use, the major, recurring one being whether it is worthwhile. Varicella is often perceived as a mild ailment. However, not only is varicella not necessarily a benign disease, but it is usually not possible to predict in advance whether an individual will develop a severe illness. In the United States, where varicella vaccine has been licensed for the past two years, utilisation rates vary across the country. The Centers for Disease Control and Prevention (CDC) has been alarmed by recent reports of deaths from varicella in healthy unvaccinated young women as well as in unvaccinated children. Currently in the United States, varicella is the leading cause of deaths that are preventable by vaccination.2 3

    Complications of varicella include those described in this issue by Jaeggi and colleagues,4 and involve the central nervous system, bacterial superinfections, and severe infections in immunocompromised patients and adults. Although they concluded that the rate of complications in the population studied was low, one cannot assume that this will continue to be the case. For example, in the United States, invasive group A β haemolytic streptococcal infections as a complication of varicella have been recognised as a mounting problem for less than a decade.5 Although these potentially life threatening bacterial superinfections are not now a problem in Switzerland, they might emerge in the future.

    An additional illness that is not always perceived as a complication of varicella is zoster. Zoster occurs only in individuals who have previously had primary infection with varicella zoster virus (VZV), either as varicella or following vaccination; only these individuals may harbour latent VZV with the potential to reactivate and cause zoster. Importantly, it has now been demonstrated in two immunised populations, children with leukaemia6 and renal transplant patients,7 that zoster occurs five to seven times less frequently after vaccination than after natural varicella. It is possible that the decreased frequency of zoster in vaccinated people is either because of a lower rate of latent VZV infection after vaccination than after natural disease or a decreased ability of the vaccine virus to reactivate in dorsal root ganglia or both. Thus the goals of varicella immunisation are to prevent complications of VZV infection: encephalitis, ataxia, invasive group A streptococcal superinfections, varicella in immunocompromised children, pregnant women and other adults, and to prevent zoster. Only by widespread use of the vaccine can these goals be accomplished. Thus I take issue with Jaeggi et al, who propose immunisation only on a selective basis. This has been the approach in Japan, where only about 20% of children are immunised and there has been little impact of the vaccine on the population.8 Selective immunisation may also increase the incidence of varicella in adults, an occurrence that is highly undesirable.9

    The duration of immunity is often cited as an unknown with regard to varicella vaccine. Following the introduction of any new vaccine, it is not possible to measure the duration of vaccine induced immunity until a lifetime has gone by. However, the record for varicella vaccine has been admirable. Small studies from Japan have indicated immunity persisting even after approximately 20 years.10 Recent studies from the United States indicate persistence of immunity in several hundred children for up to 10 years following immunisation.11 Additional studies on the duration of immunity are in progress.

    Varicella vaccine seems to provide a lower degree of protection than some other live vaccines such as that against measles. In part, this may result from a greater scrutiny of varicella vaccine recipients in an age when seroconversions after vaccination have been evaluated, as well as from the stringent efficacy trials requiring the vaccine to prove effective even after intense household exposure to unvaccinated family members with full blown varicella. It appears that the rate of complete protection from varicella vaccine is about 90%.12 Risks for breakthrough varicella are described by Lim et al in this issue.13 These authors confirmed the importance of the immunising dose of this vaccine. Others have also found that the chance of developing a breakthrough infection is increased in people who manifest lower titres of antibodies in the weeks following immunisation, a phenomenon that is related directly to the titre of virus in the immunising dose.14 15 The very best efficacy was shown in children who received a dose of vaccine about five times higher than that in currently licensed varicella vaccines.16 Therefore, better protection might be afforded by use of a higher titred varicella vaccine. VZV is notoriously difficult to propagate in vitro, so development of a higher titred vaccine that is economically feasible is a recognisably difficult task, but one worth pursuing.

    Another risk factor for breakthrough varicella identified by Limet al is immunisation at less than 1 year of age.13 This observation is in keeping with that of Teradaet al who showed that the cell mediated immune response to VZV is age dependent.17 Young infants with varicella develop lower cell mediated immunity responses than older children. Thus, the recommendation to withhold varicella vaccination until after the first birthday, as is done in the United States, seems worthwhile. This recommendation may also be important, in that children who experience varicella either in utero or in the first year of life are at increased risk of developing zoster in the first few years of childhood.18

    I believe that in developed countries we should aim to immunise all children over the age of 1 year, as well as adults who are susceptible to varicella, to control the disease and its complications. This approach may not be sensible in developing countries where more serious infections such as measles and bacterial meningitis are not yet under control. However, there are arguments for the use of varicella vaccine in some developing regions where the incidence of varicella in adulthood is high, such as Southeast Asia. The physician must always remember the admonition of Hippocrates: “first do no harm”. In consideration of varicella vaccine in developed countries, the benefits continue to outweigh the mainly theoretical risks. I believe the day has arrived when not immunising will cause more potential harm than immunisation.