Varicella

Objectives:

• Estimate the incidence of varicella in an Italian paediatric population followed by family paediatricians
• Study the natural history of varicella, particularly possibile differences between primary and secondary infections and its complications.
• Describe the therapeutic options used to treat varicella, particularly the use of acyclovir
• Quantify the number of work days lost by parents looking after children with varicella

The cohort which was studied is of about 90,000 children up to 14 years of age, followed by 102 family paediatricians using the programme JB95 in their clinical practice. 
The information on the children with varicella followed by these doctors was collected on purpose designed forms on JB95. The forms were filled out at the moment of the diagnosis (recruitment) and during follow up visits. Once completed the forms were sent by modem to a central server at 
Società Servizi Telematici di Padova. 
The study started in October 1997 and finished in May 1998. 
During the study period 5786 forms were collected about 2958 cases of varicella. 2637 (89%) of children were followed up until complete clinical resolution of the illness.

The regional distribution of the cases was not homogeneous and most children (65%) were recruited in two regions (Veneto e Marche). However, as the Italian paediatric population is homogeneous as regards the health care it receives (99% children studied were followed by the national health service) we do not think this uneven distribution influences our results. Particularly as our study was not aiming to identify regional variations in the health care received by children with varicella. 

In most studies published to date, the incidence of varcella is calculated on the total population without taking into consideration that the susceptibility to the illness varies according to acquired specific immunity. As a consequence the real incidence of varicella is underestimated, as the population at risk is not the total population but only those who have not yet had varicella. Using a mathematical model we calculated the real incidence rate of varicella on the numbers of children susceptible as identified by the model. Assuming a 10% correction factor (see text) the overall incidence of varicella in the study period was 6.2% (CI 95%: 6.0-6.4). The incidence varies for the different age groups and is at its highest in children aged 4 (11.6%; IC 95%: 10.7-12.5). Using the same correction factor, we find that at 15 years of age 92% children has had varicella.

The average age at which children got varicella was 4.59 years (± 2.59), independent of sex. 672/2958 children (22%) got varicella from somebody in the same household and were classified as secondary cases. Secondary cases are more common in children less than one year of age (p<0.001). 152 children (5.1%) had an associated chronic illness. 

At onset the clinical symptomatology was characteristic of varicella and in general moderate. In 147 children it presented with more severe symptoms (severe headache, greater than 100 vesicles, fever above 38C and/or complications). The severity of illness at onset is significantly associated with secondary infection (p= 0.011) and age > 10 anni (p<0.001). On the other hand, severity of symptoms at onset is not associated with concurrent chronic illness (p=0.096). 

The mean duration of illness, defined as disappearance of symptoms and absence of new vesicles was 8.5 days 
In children with secondary infection, with concurrent chronic illness and who had more severe symptoms at onset, the duration of illness was significantly longer (p=0.014, p=0.013, p<0.001). Age of onset did not seem to be associated with duration of illness. 

125 children (4.2%) had 130 complications during the study period. In most cases these were severe skin infections, upper respiratory tract infections or conjunctivitis. The risk of complications was not associated with the type of infection, age or the concurrence of chronic illness. 

9 children were hospitalised for complications. The reasons were skin infections (4 cases), pneumonia (3 cases) e ataxia (2 cases). The risk of hospitalisation was higher in children aged less than 1 year (p=0.048) and in cases of secondary infection (p=0.0012). 

3664 prescriptions were given to 2637 children. The most commonly prescribed drugs were oral antihistamines (58.2%), oral acyclovir (ACV) (23.5%) and anxiolytics (7.2%). The severity at onset and age (> 10 years) were significantly associated with pharmacological treatment. 

586 of the 2637 children (22.2%) followed up were treated with oral ACV,413 started ACV at diagnosis while 173 children started the drug during follow-up. The presence of complications at onset is the only parameter independently associated with starting treatment. 
The use of ACV is more common in children aged less than 1 year or more than 10 years (p<0.001). It is also associated with secondary infection (p<0.001) and severity at onset (p<0.001). 

Multivariate analysis did not highlight any association between the use of ACV and length of treatment after adjusting for severity, type of infection, age and region of residence. 
The duration of illness does not differ in children starting treatment at diagnosis compared to those commencing treatment during follow-up. 
However, as this is not a randomised trial it is not possible to make any comment on the efficacy of ACV. 

1983 working days were lost by parents of 300 parents (11.4%) due to the varicella. The average number of working days lost was 6.6. 
We did not find any association between the age of the child, type of infection or presence of concurrent chronic illness and number of working days lost. The use of ACV was also not associated with a reduction in working days lost.