Intended for healthcare professionals

Education And Debate

No reason not to treat

BMJ 1995; 310 doi: (Published 14 January 1995) Cite this as: BMJ 1995;310:109
  1. Henry H Balfour, professor of laboratory medicine and pathologya
  1. a University of Minnesota Medical School, Minneapolis, Minnesota, USA
  1. Correspondence to: Box 437 UMHC, 15–144 PWB Harvard Street at East. RiverParkway, Minneapolis, Minnesota, USA 55455.

    Acyclovir has been shown to shorten the course of chickenpox in otherwise healthy children by 25–33% according to three placebo controlled trials in the United States.1 2 3 All clinical events monitored were favourably affected by acyclovir, with the most important differences being reductions in degree and duration of fever, constitutional illness score, and number of residual lesions present at the 28 day follow up examination. No appreciable adverse events were reported. These studies also identified two subsets of patients particularly suitable for acyclovir therapy: adolescents and children who acquired chickenpox from a family member.

    Need for treatment

    The main controversy regarding use of acyclovir is not its clinical usefulness, which is scientifically incontestable, but whether treatment is needed for a relatively benign disease. To fully appreciate the impact of acyclovir on chickenpox you need to make daily visits to treated subjects and controls, as our research nurse clinicians did. In the three studies, which included 979 children and adolescents, there were nearly 6000 clinic and home visits. Such daily observations made it evident that the “misery index” of chickenpox for the patient and family was rapidly attenuated by antiviral drugs. It is difficult to integrate such firsthand observations into a two dimensional Kaplan-Meier plot.

    Is treatment of chickenpox practical? The reported trials have enrolled children only during the first 24 hours of rash, and antiviral drugs begun later has not been studied for children. Treatment may need to be started within 24 hours because a placebo controlled study of naval recruits who were enrolled as late as 72 hours after onset of rash concluded that acyclovir was beneficial only for adults whose treatment was started during the first 24 hours.4 Parents cannot always recognise the index case of chickenpox in the family soon enough to notify a doctor within 24 hours. With knowledge of the incubation period, however, they can identify secondary cases, which is actually more important because secondary cases are more severe.

    Problems of treatment

    Cost and cumbersome dosing are valid arguments against using acyclovir. Cost varies tremendously from country to country. Because of its relatively poor bioavailability, oral acyclovir must be given four times per day, which is inconvenient and reduces compliance.

    A theoretical argument against treatment of chickenpox is that widespread use of acyclovir will lead to viral resistance. This concern has not been completely allayed, but short term treatment with acyclovir for herpesviruses in otherwise healthy patients has not resulted in emergence of resistant viral strains.5 6

    Another concern is that treatment of chickenpox will blunt the active immune response, leaving the patient susceptible to a second case of chickenpox at an older age, when the likelihood of complications is greater. But three papers have now shown durability of immunity after acyclovir treatment.1 7 8


    Some have argued that if treatment is good, prophylaxis would be better. I am opposed to prophylaxis against chickenpox using acyclovir for the following reasons. Firstly, we do not have any controlled data showing that acyclovir can prevent chickenpox. Secondly, unless antibody titres are carefully followed, it is uncertain whether acyclovir prophylaxis has resulted in a subclinical case of chickenpox or whether the exposure was trivial. Without that knowledge the patient will need to receive prophylaxis after every subsequent exposure. As the patient grows older, the likelihood of chickenpox with complications increases. Thirdly, data from the first controlled treatment trial in Minnesota indicated that a short course of acyclovir did not reduce antibody titres measured one year after chickenpox.1 Another study indicated that the same holds true for varicella zoster virus specific cell mediated immunity.8 It would therefore be better to allow chickenpox to manifest itself andthen treat immediately with acyclovir, knowing that there will be a durable immune response. Prophylaxis could leave the patient with imperfect immunity and susceptible to chickenpox later in life when the disease can be more serious.

    The final factor in the “to treat or not to treat” equation is the difficulty of predicting who will have a relatively severe case of chickenpox at the onset of disease. We searched for predictors of disease severity in our placebo controlled trials and identified only age and case order in the family. Added to this conundrum is the fact that even mild immunosuppressive therapy has been associated with visceral dissemination of chickenpox.9 Since acyclovir given orally is virtually devoid of side effects, why not treat chickenpox with acyclovir now?


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    View Abstract