Postexposure Varicella Vaccination in Siblings of Children With Active Varicella

A live attenuated varicella vaccine was approved by the Food and Drug Administration on March 17, 1995. The use of the vaccine for postexposure prophylaxis is currently not recommended by the manufacturer in the package insert nor is it listed as an approved use by the Committee on Infectious Diseases of the American Academy of Pediatrics (Red Book Committee).1 The use of postexposure varicella vaccination has been shown to be effective if given within 3 days of exposure in several studies done in the 1970s and 1980s.2-10 However, since these studies were done the vaccine manufacturing process has been changed to increase the viability, stability and yield of the virus.11 No studies have been done with the currently available Oka/Merck vaccine for postexposure vaccination. We evaluated the efficacy of postexposure varicella vaccination in siblings of children presenting with active varicella. Methods. We offered varicella vaccination to all family members with a negative history of varicella whenever a patient presented with active varicella. Serologic evidence for prior varicella was done in adults with a negative history of varicella, varicella serologic evaluation was not done in any of the siblings of the children with active varicella. Parents counted the number of lesions on their children on the fifth day of the rash. Parents were contacted by telephone several times between 10 and 30 days after vaccination to look for any lesions in the vaccinated family members. If any lesions developed they were counted on the fifth day of the rash. Parents were encouraged to have the vaccinated siblings evaluated by one of the investigators if they were not sure whether a lesion was consistent with chickenpox. Statistical analysis was done using the Wilcoxon signed rank test. Seven pairs were used. When more than one vaccinated sibling was associated with an index case, the sibling with the greatest number of lesions was used for statistical analysis. Results. Sixteen household members without a prior history of chickenpox of 12 children with active varicella were given postexposure varicella vaccination. Three of the 16 household members were parents. All 3 of the parents had serologic evidence of prior varicella infection and were excluded from the study. Three of the remaining 13 vaccinated children could not be contacted by telephone or did not return telephone messages and were excluded from the study. This left 10 vaccinated children and 7 index cases for study. The ages of the vaccinated siblings were 14 months to 12 years (mean age, 5.7 years). The ages of the index cases were 4 to 15 years (mean age, 8.4 years). The mean number of lesions in the 7 index cases and the 10 vaccinated siblings were 245 and 13.4, respectively (P = 0.018). Five of the 10 vaccinated siblings did not develop lesions whereas 5 developed between 5 and 83 lesions that occurred 12 to 13 days after vaccination in all 5 patients. Lesions were not tested as to whether they were wild-type or vaccine strain virus. Table 1 lists the ages and the number of lesions counted on the fifth day of the rash in each patient.TABLE 1: Age, sex and number of lesions in the index cases and their vaccinated siblings Discussion. There have been several published reports on the use of postexposure varicella vaccination.2-10 However, since the manufacturing process of the varicella vaccine has changed,11 none of these studies used the currently available vaccine, which was approved by the Food and Drug Administration in March, 1995. Several reports have documented the effectiveness of postexposure varicella vaccination in hospitalized children after exposure on a hospital ward.2-6 In the mid-1970s, Asano et al.7 used Oka strain varicella vaccine as postexposure prophylaxis in family contacts of children with clinical varicella or zoster. Eighteen susceptible contacts were vaccinated (most within 3 days of exposure) and 19 contacts remained unvaccinated and served as controls. Clinical varicella developed in all of the controls and none of the vaccinated contacts developed lesions. Subsequently Asano et al.8 in 1982 reported dose-response data from studies using postexposure vaccination in 45 household contacts of patients with varicella. None of the 30 children who received the higher dose [between 800 and 15 000 plaque-forming units (pfu)] of the vaccine within 3 days of exposure developed lesions whereas 6 of 10 children who received a lower dose (300 to 600 pfu) developed mild varicella. Children who were vaccinated 5 days after exposure developed typical cases of varicella despite receiving high doses of the vaccine. The vaccine used currently in the United States has ∼1500 pfu.1 In 1984 Naganuma et al.9 vaccinated 40 children after household exposure and found that 30 did not develop lesions, 6 had a mild rash and 4 had typical varicella. In 1984 Arbeter et al.10 summarized the results of a double blind, placebo-controlled trial of Oka/Merck varicella vaccine (4350 pfu) administered as postexposure prophylaxis. Twenty-six susceptible siblings of children with active varicella were vaccinated within 5 days of the onset of lesions in the index case; 13 received varicella vaccine and 13 received placebo. In the vaccinated group 9 of 13 did not develop lesions and 4 developed mild cases of varicella. Twelve of the 13 in the placebo group developed typical varicella. Among the 10 children who received the vaccine within 3 days of exposure, only one developed lesions. This study concluded that varicella vaccine was 67% effective in preventing disease and 100% effective in modifying disease. However, the vaccine used was a higher dose than the currently available vaccine. We demonstrated that 5 of 10 susceptible siblings did not develop lesions when they were vaccinated within 3 days of the appearance of rash in the index case. The other 5 vaccinated children developed mild disease with only 1 child developing >20 lesions. Although we did not have an unvaccinated control group in our study, prior reports have demonstrated an attack rate of 87% among susceptible household contacts.12 In addition we did not test the skin lesions to determine whether they were wild-type varicella virus or vaccine-strain virus. However, given that all lesions occurred 13 to 14 days after onset in their siblings, and prior studies have shown that only 7 to 8% of vaccinated children develop vaccine-associated lesions with only a median of 2 to 5 lesions,1 all 5 subjects who developed lesions in this study probably had wild-type virus. We believe that postexposure varicella vaccination is effective in preventing or modifying disease in siblings of children with active disease. We recommend that varicella vaccination be considered in all siblings of children with active varicella who have a negative history of varicella disease or vaccination. Mark B. Salzman, M.D. Cass Garcia, L.V.N. Department of Pediatrics; Kaiser Permanente West Los Angeles Medical Center; Los Angeles, CA