Epidemiology of varicella-zoster virus infection: Chickenpox

INTRODUCTION — Varicella-zoster virus (VZV) infection causes two clinically distinct forms of disease: varicella (chickenpox) and herpes zoster (shingles). Primary VZV infection results in the diffuse vesicular rash of varicella, or chickenpox.

The epidemiology of varicella has changed dramatically since the introduction of the varicella vaccine in 1995. In the United States, routine childhood immunization has reduced disease incidence, complications, hospital admissions, and deaths in children and in the general population, indicating strong herd immunity. Similar immunization programs have been adopted by several other countries, including Uruguay, Germany, Taiwan, Canada, and Australia [1].

This topic will address the morbidity and mortality of varicella infection prior to and after immunization. The clinical manifestations and management of varicella are discussed elsewhere. (See "Clinical features of varicella-zoster virus infection: Chickenpox" and "Treatment of varicella (chickenpox) infection" and "Vaccination for the prevention of chickenpox (primary varicella infection)" and "Prevention and control of varicella-zoster virus in hospitals" and "Varicella-zoster virus infection in pregnancy".)

EPIDEMIOLOGY OF VARICELLA PRIOR TO VACCINE

Incidence of varicella — Varicella occurs throughout the year in temperate regions, but the incidence typically peaks in the months of March through May [2]. According to national seroprevalence data from the pre-vaccine era, greater than 95 percent of persons in the United States acquired varicella before 20 years of age, and fewer than 2 percent of adults were susceptible to infection [3-6]. Prior to 1995 the Centers for Disease Control and Prevention (CDC) estimated the yearly incidence of chickenpox in the United States at approximately four million cases, with nearly 11,000 admissions and 100 deaths [7].

A chart review of over 250,000 members of a health maintenance organization (HMO) was conducted from 1990 to 1992 to assess the epidemiology of varicella and its complications [8]. The following results were noted:

●A total of 5686 potential incident cases in this HMO population were identified during this period; using age-specific predictive values, the estimated number of incident cases was 4884 [8].

●Seventy-six percent of cases occurred in persons less than 10 years of age. The incidence rates per 100,000 person-years in each age group were as follow: 5234 for ages 0 to 4 years; 4132 for ages 5 to 9 years; 1404 for ages 10 to 14 years; 610 for ages 15 to 19 years; and 175 for ages >20 years [8]. These rates continued to decline for older age groups.

In tropical countries, varicella occurs mainly among young adults [9]. Several studies have confirmed primary VZV infections in these patients following intense exposure to cases in military installations, hospitals, and day care centers [3,10-12]. Seroepidemiologic surveys have confirmed that >20 percent of military recruits enlisting from outside the 50 US states were susceptible to VZV [12].

The following studies illustrate the high overall attack rates that can be seen in susceptible populations in military and school settings [11,13]:

●An epidemiologic investigation of two varicella outbreaks in a military training center documented that 42 percent of 810 adult recruits from Puerto Rico were seronegative for VZV; there was an overall attack rate of 30 percent in the initial varicella outbreak (estimated to be 71 percent in susceptible individuals) [11].

●In a 2004 varicella school outbreak reported in China, 138 of 1407 (9.8 percent) of students developed primary varicella [13]. Many of these children had not been vaccinated, since immunization was not required.

Varicella-related hospitalizations and mortality — Although children were most commonly affected by varicella, adults and infants less than one year of age were overrepresented among those who developed complicated disease, with high rates of mortality [14-16]. For example, adults over the age of 20 years accounted for less than 5 percent of varicella cases but 55 percent of varicella-related deaths (CDC unpublished data 1997) [17]. In another study, adults with chickenpox had a 25-fold higher risk of complications compared to children [18]. A review of age-specific data for varicella and encephalitis in the United States from 1972 to 1978 demonstrated that persons >20 years of age accounted for fewer than two percent of varicella cases but 12 and 28 percent of varicella encephalitis and deaths, respectively [15].

The specific types of complications among patients with varicella also tend to vary by age, with bacterial infections occurring in most children [19] and pneumonitis being more common in adults. A review of 613 varicella-related hospital admissions in England from 1968 through 1993 demonstrated that 23 percent of adults developed varicella pneumonitis [19]. Furthermore, the risk of pneumonitis was six times higher in smokers than nonsmokers.

EPIDEMIOLOGY OF DISEASE AFTER INTRODUCTION OF VARICELLA VACCINE

Incidence of infection — The Varicella Activity Surveillance Project (VASP) was established in 1995 as a cooperative agreement between the Centers for Disease Control and Prevention and two cities in the United States (Antelope Valley, CA and Philadelphia, PA) to monitor disease incidence after the introduction of vaccine. At the start of the national varicella vaccination program, the Advisory Committee on Immunization Practices (ACIP) recommended one dose of varicella vaccine for children 12 to 18 months of age, older susceptible children (19 months of age through 12 years of age), and high-risk groups [14]. (See "Measles, mumps, and rubella immunization in adults".)

This recommendation was followed by substantial declines in the incidence of varicella and related hospitalizations within the first five years of the vaccination program; in three surveillance sites, the incidence of varicella declined by greater than 70 percent from 1995 to 2000 with vaccination rates ranging from 74 to 84 percent [20]. Further follow-up surveillance confirmed and extended these original findings:

●By 2005, overall varicella incidence in active surveillance sites had declined by 90 percent [21]. This decline in incidence occurred in the context of rapid uptake of vaccination coverage among young children (>92 percent) supplemented by catch-up vaccination among older children.

●In a Connecticut surveillance program, the number of cases of varicella plateaued from 2001 to 2005 with an increasing number of cases occurring in children who were previously immunized with only one dose of vaccine. Notably, disease that did occur was milder amongst vaccinated children than unvaccinated children [22].

●A similar observation of milder disease in vaccinated adults has also been reported in a 10-year study of more than 17,000 patients with varicella [23]. Adults had a twofold (95% CI 1.8-2.3) higher risk of developing a complication and a sixfold (95% CI 4.0-9.7) higher risk of being hospitalized than children. Adolescents had illness severity intermediate between the severity in children and adults.

●In this same study [23], age-specific incidence rates significantly declined for both children and adults; among children 0 to 14 years of age, the incidence declined by 90 percent, while in adults the incidence declined by 74 percent, despite low rates of vaccination. These findings highlight the benefits of "herd immunity"; this term refers to protection against infection that occurs indirectly for unvaccinated persons when others in the community are immunized.

Further refinement of the vaccination program occurred in 2006, when the ACIP recommended a second dose of varicella vaccine because of a number of varicella outbreaks occurring in highly vaccinated populations of schoolchildren [24]. After the implementation of the two-dose schedule, the incidence of varicella decreased by 85 percent (from 25.4 per 100,000 population during 2005-2006 to 3.9 per 100,000 population during 2013-2014) in states reporting data to the National Notifiable Diseases Surveillance System [25].The largest declines were seen in children and adolescents aged 5 to 14 years.

These aggregate data also support the use of vaccine in individuals who have been exposed to varicella. In one study, 68 percent of the adults had been exposed in the household [23]. Post-exposure prophylaxis with varicella vaccine can prevent or modify disease if given between three and five days of exposure. (See "Post-exposure prophylaxis against varicella-zoster virus infection", section on 'Who is eligible?'.)

Outbreaks — The epidemiology of varicella outbreaks (defined as ≥5 varicella cases epidemiologically linked to a common setting that occurred within one incubation period) have significantly decreased in number, size, and duration with the introduction of varicella vaccine [26]:

●The number of outbreaks declined from 236 in the period from 1995 to 1998 to 46 in the period from 2002 to 2005

●The median number of cases per outbreak decreased from 15 to 9

●The duration of these outbreaks declined from 45 to 30 days.

Varicella-related hospitalizations — Following implementation of the US varicella vaccine program in 1995, rates of varicella-related hospitalizations and complications have significantly decreased [20,27,28]. Data from the Varicella Activity Surveillance Project were used to compare rates of hospitalization and rates of complications among patients hospitalized with varicella from 1995 to 2005 [27]. For this project, the vaccination periods were defined as early (1995 to 1998) middle (1999 to 2001) and late (2002 to 2005). The study found the following:

●A total of 26,290 varicella cases were reported from 1995 to 2005. Of these cases, 170 patients (6.47 per 1000 varicella cases) required hospitalization.

●Decreases in varicella-related hospitalization rates occurred in both older and younger age groups from the early to combined middle/late periods: among those <20 years of age, rates decreased by 77 percent; among adults >20 years of age, rates decreased by 60 percent.

●Logistic regression analyses demonstrated that risk of hospitalization was independently associated with being <5 or >15 years of age, not being vaccinated, and being immunocompromised. Those <1 year of age and >20 years of age had the highest risk of varicella-related hospitalization (adjusted ORs 7.0 [95% CI 4.1-12.1] and 7.6 [95% CI 4.5-12.6], respectively).

●The rate of complications during the early vaccination period decreased by 70 to 90 percent to the middle/later vaccination periods. Most of these complications were skin and soft tissue infections and pneumonitis.

Of note, most of these hospitalizations were preventable through implementation of existing vaccination policy recommendations.

Overall mortality — National death records from 1990 to 2001 were reviewed to assess the effect of the vaccination program on mortality associated with varicella infection [29]:

●The rate of death due to varicella fluctuated from 1990 through 1998 and then declined sharply.

●This decline was observed in all age groups under 50 years, with the greatest reduction (92 percent) among children one to four years of age.

Overall declines in mortality have also been noted elsewhere; in one surveillance system in Massachusetts between 1998 and 2003, the overall varicella incidence declined by 79 percent [21]; for deaths in which varicella was listed as the underlying cause, rates of mortality decreased by 67 percent from an annual average of 105 deaths during 1990 to 1994 to 35 deaths during 1999 to 2001.

While there has been a substantial decline in varicella-related deaths since vaccine licensure [29], the CDC surveillance program reported six varicella-related deaths in 2007 [30]; five of these deaths occurred in adults aged 23 to 78 years.

Childhood mortality — Approximately two years after the introduction of vaccine, varicella-related deaths were still being reported. A review of varicella-related deaths in children in the United States from 1997 attempted to identify factors associated with poor outcome [31]:

●Ninety percent of children who died did not have identifiable risk factors for severe varicella.

●The most frequent varicella-related complications in children were secondary bacterial infections and pneumonia.

Adult mortality — Varicella remains a more severe disease among adults [20,32], although incidence rates and morbidity are declining in the era of varicella vaccine [23].

FUTURE CONCERNS — Because increasing use of varicella vaccine will lead to an expected decrease in circulating wild-type VZV, unimmunized children may acquire infection at an older age, when they are more susceptible to severe infection. Mathematical models predict that if immunization rates in children are >90 percent, a greater proportion of varicella infections will occur at older ages; however, overall disease burden will decline [22]. This emphasizes the need for high levels of immunization among children to prevent late-onset disease.

SUMMARY AND RECOMMENDATIONS

●Epidemiology prior to vaccination  

•Prior to the introduction of the varicella vaccine in 1995, the Centers for Disease Control and Prevention (CDC) estimated the yearly incidence of varicella (chickenpox) in the United States at approximately four million cases with nearly 11,000 admissions and 100 deaths. (See 'Incidence of varicella' above.)

•Although children were most commonly affected by varicella, adults and infants less than one year of age were overrepresented among those who developed complicated disease with high rates of mortality. (See 'Varicella-related hospitalizations and mortality' above.)

●Epidemiology after vaccination – The epidemiology of varicella has changed dramatically since the introduction of the varicella vaccine in 1995.

•One decade after the introduction of the varicella vaccine, the overall incidence of varicella in active surveillance sites declined by 90 percent. (See 'Epidemiology of disease after introduction of varicella vaccine' above.)

•The number and size of outbreaks also decreased since the initiation of the vaccination initiative in the United States. (See 'Outbreaks' above.)

•The rate of complications from varicella infection declined dramatically after the introduction of the vaccine; most complications were skin and soft tissue infections and pneumonitis. (See 'Varicella-related hospitalizations' above.)