VARICELLA DISEASE AND TRANSMISSION IN PEDIATRIC HOUSE OFFICERS

Abstract

Varicella-zoster virus (VZV) infection is primarily a benign, self-limited disease. For certain categories of patients, however, disease is more severe and may be life-threatening. These categories include immunocompromised individuals, infants born prematurely, children 13 years of age or older, patients with chronic cutaneous or pulmonary disorders and those receiving immunosuppressive doses of corticosteroids.1 High risk patients are frequently in the hospital setting where they come into contact with health care workers (HCWs), such as pediatric residents, who represent a contact source for infectious diseases including varicella. The current recommendation by the Committee on Infectious Diseases of the American Academy of Pediatrics for varicella-susceptible HCWs who are exposed to VZV is that they should be either furloughed or excused from patient contact from Days 8 to 21 after the onset of the rash in the index case; for those who have received varicella-zoster immunoglobulin or are immunocompromised, this interval should be extended until Day 28.1 Similar recommendations have been issued by the Centers for Disease Control and Prevention.2 Implementation of these guidelines is often impractical for HCWs such as pediatric residents who may be exposed repeatedly to varicella during their training. A history of previous varicella infection is considered indicative of immunity. For those with no such history serologic testing for the presence of VZV antibodies has been advocated.3-5 If positive they are considered immune and hence can be allowed to continue with patient care. This recommendation assumes that HCWs with positive VZV serology but negative history for varicella will not develop disease after exposure. To evaluate current recommendations related to pediatric residents who are varicella-susceptible, we examined our experience during a 5-year period in a university-based teaching program. This review documents the shortcomings of current recommendations and the need to emphasize administration of vaccine to HCWs with a negative history as well as negative serology for VZV. Methods. All pediatric residents who began training at Louisiana State University School of Medicine in New Orleans from July, 1991, to July, 1994, were included in this study. Pediatric residents in this program spend approximately two-thirds of their 3-year training at Children's Hospital, a 192-bed facility that contains a 23-bed neonatal intensive care unit and an 18-bed oncology unit. The other one-third of their time is spent at Charity/University Hospital, a facility with 40 general pediatric beds, a 25-bed neonatal intensive care unit and a pediatric emergency department which supports 81 000 visits each year. Before the beginning of their training the residents were asked about their history of previous varicella infection, and all had serologic evaluations of their VZV antibody status. All were followed prospectively for exposure to chickenpox or zoster during their residency training, and those who were negative either by history or by serologic screening were carefully monitored for development of chickenpox. The policy for work activity in exposed seronegative residents was that recommended by the American Academy of Pediatrics.1 VZV antibody titers were determined by the indirect immunofluorescence assay (IFA) (Electro-Nucleonics, Inc., Columbia, MD) using a dilution of 1:8 to define a protective antibody titer. Results. Fifty-eight residents (mean age, 25 years) began their training during the study period. Forty-nine (84%) reported a positive history occurring at a mean age of 11 years (range, 4 to 26 years), and 4 (7%) reported an uncertain history for previous varicella infection. All had positive VZV serologies by IFA and none developed varicella during their training. Five residents (9%) reported a negative history for previous varicella infection. Serologic testing was positive for VZV antibodies in two of them. The three residents with a combination of negative history and serology were instructed to avoid patient contact after a known exposure to varicella as recommended by the American Academy of Pediatrics.1 Restriction was necessary on eight occasions. All 5 residents with a negative history contracted varicella at a median of 8 (range, 3 to 23) months after starting their training including the two with positive serology. Only 1 developed the infection after an obvious exposure to a patient with varicella; the other 4 developed their infection with no recognizable source. Varicella occurred while all 5 were involved with patient care, 2 in the neonatal intensive care unit and 1 in the oncology unit. As a consequence 43 immunocompromised patients were exposed to varicella and required the administration of varicella-zoster immunoglobulin. Two patients subsequently developed varicella infection, 1 of whom, a severely premature neonate, died from complications of prematurity while infected. The residents missed a total of 41 days of work and recovery was complete in all. Discussion. Nine percent (5 of 58) of our residents beginning training were nonimmune to varicella as exhibited by their subsequent development of chickenpox. This prevalence of susceptibility is similar to what has been previously reported in adult populations.6 All 5 residents who developed varicella had negative histories for the disease yet 2 had detectable antibodies by IFA. IFA and enzyme-linked immunosorbent assay have a sensitivity of 90 to 97% and specificity of 80 to 88% as compared with the fluorescent antibody to membrane antigen assay.6, 7 The fluorescent antibody to membrane antigen assay is considered a more sensitive test for the detection of varicella antibodies, but it is not clear whether it is more specific.7, 8 There have been previous reports of varicella among HCWs serologically considered immune, the estimated false positive rate being 0.3%.9 Our experience indicates a much higher rate. The three residents with presumed nonimmune status to varicella had their rotations altered many times because of varicella exposure, yet all eventually became infected during their training. When disease occurred the event was unexpected and, hence, no timely exclusion from work had been instituted. All residents did well but they exposed immunocompromised patients to varicella, necessitating additional interventions and creating increased risk for a vulnerable population. Our experience suggests that a policy of routine exclusion of susceptible HCWs from patient contact after exposure to varicella, as suggested by various expert groups,1-5 is disrupting and inadequate for preventing spread of infection in the hospital setting. With the recent licensure of a varicella vaccine a policy emphasizing active immunization would seem more appropriate. The Committee on Infectious Diseases of the American Academy of Pediatrics has recommended that two doses of vaccine, 4 to 8 weeks apart, be given to healthy adolescents and young adults who have no history of varicella infection.10 The same document mentions that serologic testing is optional, but if positive the vaccine does not need to be given. Our findings suggest that vaccinating all residents with either a negative history for varicella or negative serology for VZV would be most prudent. Because the median time to contracting varicella infection in our study was 8 months, a policy of early vaccination is indicated. Andrew C. Oshiro, M.D.; Rodolfo E. Begue, M.D.; Russell W. Steele, M.D. Division of Infectious Diseases/Immunology Department of Pediatrics Louisiana State University School of Medicine and Children's Hospital New Orleans, LA