Management of a Major Varicella Zoster Exposure in a Pediatric Oncology Population

Abstract

Management of the exposure of pediatric oncology patients to varicella zoster virus (VZV) is controversial. We report the exposure of 56 patients to a single child with chicken pox at a pediatric cancer housing facility and describe our strategic approach for their management. We reviewed the immune and clinical status of 56 children with cancer receiving ongoing treatment at Memorial Sloan Kettering Cancer Center (MSK) who, while living at a pediatric cancer housing facility, were exposed to the index patient. The management of patients exposed included:1)determination of immune status, 2)availability of vaccination history or VZV disease prophylaxis, 3)exposure status and subsequent isolation during the period of incubation, and4)VZV disease prophylaxis.In addition to the 56 patients exposed to the index case, 8 children with cancer treated at other facilities and 11 healthy siblings living in the facility were exposed.Of the 56 MSK patients, 21 were classified as immunosuppressed andreceived varicella zoster immune globulin (human), intravenous standard immune globulin, or acyclovir based on serostatus and immune function. The cohort was followed for 4 weeks after the exposure and no secondary infections were diagnosed. We performed risk assessment and created a management plan to control and prevent further exposure and development of disease. No secondary cases developed. This strategic approach could serve as a model for the management of VZV exposure for other pediatric oncology centers.

Introduction

The management of exposure of pediatric oncology patients to varicella zoster virus (VZV) is controversial.¹Often, exposure of such patients includes a small number of children which can be managed within normal operations of providing clinical care. At Memorial Sloan Kettering Cancer Center (MSK), we have established guidelines which are based on The Red Book that define specifically (1) what constitutes a significant exposure and (2) how one defines an immunocompromised child who is at risk for developing varicella.²

In this report, we describe the potential exposure of 56 patients and 11 siblings to a single child with overt chicken pox at a pediatric cancer housing facility and describe our strategic approach to exposure risk assessment and management. Our team was faced with a mass exposure to primary VZV in a population including immune suppressed patients. Clearly, this presented challenges beyond that of a single exposure. Rapid triage, treatment and exposure response were implemented to provide appropriate treatment and mitigate further exposure within a 48-hour time period.We learned the importanceof those assigned to respond to such exposures being available at any moment, including holidays as in this case.We followed the Red Book guidelines when pertinent to our case, and our institutional guidelines for specific areas of management.

Patients and Methods

Description of Index Case

The index patient was a fifteen-month-old immune competent female residing at a pediatric cancer housing facility while undergoing surgery for a benign central nervous system pathology at a different hospital.The patient had no prior history of chicken pox and no history of varicella vaccination. The patient developed a skin rash that led her mother to bring her to the emergency department (ED) at another hospital. She was discharged back to the pediatric cancer housing facility without specific diagnosis. On day 4 of the rash, she returned to the same ED and was diagnosed with classic varicella zoster including disseminated multi-stage macules, papules, and vesicles. The ED attending physician contacted the housing facility, as well as our center, to notify us of the varicella infection and possible exposure risks.

The patient was outgoing and active,often in the communal areas of the housing facility. Based on the history obtained from the housing facility staff, this child played for a significant amount of time with most of the children and families residing there, for the full 4-day period between ED visits. Based on this information, we assumed she had exposed most of the children residing at the housing facility from two days prior to the outbreak of her skin rash, until the day of the diagnosis of varicella zoster.

The index patient was transferred to a different housing facility and treated with acyclovir.

Exposure Setting

The setting includes an affiliated group pediatric housing facility that serves patients and families treated in 13 different hospitals including cancer centers. During treatment, pediatric patients, including patients with cancer and their (immunocompetent) family members reside at this 84-room housing facility for days to weeks. This facility has a playroom, a large kitchen, a dining room, and four communal area rooms. At the time of the presentation and diagnosis of the index case, there were no other children or adults at this facility with any evidence of chicken pox or shingles. The index patient traveled to the United States 17 days prior to her rash outbreak which may indicate an infection predated her arrival to the housing facility since the incubation period can range from 10–21 days for this viral infection.²

All of this transpired during Thanksgiving weekend. Upon notification of the varicella exposure, a team was immediately formed (figure 2) that included the pediatric ambulatory care center medical director, nurse leader, and administrative staff, as well as the infection control medical director and infection control manager. This team worked closely with the pediatric staff including physicians and nurse practitioners as well as the management team of the pediatric housing facility. The list of all patients who stayed at the housing facility for the five days of potential exposure was obtained from the management team of the housing facility within one day.

Figure 2.

Organizational Chart of Strategic Response to a major varicella zoster exposure. Summary figure depicts specific personnel and major interactions and responsibilities of personnel in relation to the response to a major varicella zoster virus exposure in a pediatric oncology setting. PACC, Pediatric Ambulatory Care Center; VZV, varicella zoster virus

Case Definitions

Exposure

Exposure via airborne route of varicella virus was defined as the physical presence of the index patient with overt chicken pox, or two days prior to rash, and the child or family member exposed in the same room for 15 minutes. However the time variable of the exposure was impossible to assess accurately and therefore, any physical presence with the index patient in the same room was considered an exposure. Non-exposure was defined as patients residing in the housing facility who were registered or remained in their room the entire time the index case was residing at the housing facility.

Exposed Patients

The exposed patients were categorized based on immune status and varicella serostatus.These included patients who were immune competent with known varicella seropositivity, patients who were immune competent and seronegative, and patients who were immune suppressed, regardless of their varicella serostatus.

There are no criteria for a strict definition of severe T-cell immunosuppression. Immunosuppression was therefore defined based on our institutional experience and guidelines as patients having had: (1) an allogeneic bone marrow transplant within 9 months, or (2) autologous bone marrow transplant within 6 months or (3) high dose immunosuppressive therapy within 6 months which included high doses of immunosuppressive agents.In our experience these were associated with a probable inability to mount a response to varicella exposure, regardless of serostatus.³

As there are also no strict definitions of immune competence in the oncology patient, we empirically determined it based on varicella serostatus with or without history of varicella infection or varicella vaccine, in healthy individuals, or in oncology patients who have not received proximate severely immunosuppressive treatment.Besides this definition of immune competence and suppression, all our procedures followed the Red Book recommendations.

All data were collected as part of infection control surveillance and outbreak control. This retrospective study was approved by the MSK Institutional Review Board.

Results

Risk Assessment

For our institution, the patients’ primary pediatric services were contacted, and asked to evaluate the exposure history of the specific patients and siblings on that initial list and separate them into three groups: 1) patients or siblings in the housing facility who were out of their rooms and in the facility were determined as exposed, 2) patients or siblings in the housing facility but staying in their room were determinednot exposed or 3) patients hospitalized, not physically present in housing facility during that same time period, were also not exposed.

Overall, 93 patients were identified as potentially exposed to VZV. After contact tracing, 18 children were considered not exposed, while 75 children were exposed, including 56 MSK patients, their 11 siblings, and 8 patients from 5 other institutions.

Triage, Determination of Isolation Status, and Treatment(Figure 1 and Table 1)

Figure 1.

Summary of strategic management of a varicella exposure in a pediatric oncology setting. VZV, varicella zoster virus

Table 1.

Management of Patients Post VZV Exposure

	Immune Status	Varicella serostatus/ Vaccine history	Post exposure VZV Prophylaxis		Post exposure Isolation
Patients
Group 1 N = 16	Immune competent	Positive	No intervention	N=15	Isolation: N=1
			IVIG elsewhere	N=1
Group 2 N = 12	Immune competent	Unknown status	Acyclovir	N=3	No Isolation
Group 3 N = 7	Immune competent	Negative	IVIG	N=2	Isolation: N=4
Group 4 N =10	Immune suppressed	Positive	VZIG	N=4	Isolation N=8 (2 pts followed elsewhere)
			IVIG	N=1
			Acyclovir	N=1
			No intervention	N=4
Group 5 N =11	Immune suppressed	Negative	VZIG	N=6	Isolation: N=11
			IVIG	N=5
Siblings
Group 6 N =11	Immune competent	7 vaccinated 4 not vaccinated	VZIG	N=1 (infant)	Isolation: N=2 (2 sibs followed elsewhere)
Other (othercenters)
Group 7 N =8	NA	NA	NA		NA

VZV, varicella zoster virus; IVIG, intravenous gamma globulin; VZIG, varicella zoster immunoglobulin; NA, not applicable.

The management strategy for patients exposed included the following: identification of immune status, availability of vaccination history or VZV serostatus, the exposure status, and thereafter the need for isolation precautions and VZV disease prophylaxis. Patients that went back to their primary local home/institution for treatment were not part of the decisions for post exposure prophylaxis and isolation.

The varicella serostatus, immune function,and recent immunosuppressive treatment were assessed on all 75 patients and 11 siblings exposed over the 24–48 hours following the notification by a team of approximately 20 clinical staff members.

The management of the 11 siblings was done on the same principle as that of the patients, and therefore they are included in the entire cohort as patients. Of note, all siblings were considered immune competent, and 7 siblings were previously vaccinated while 4 were not. All exposed patients and siblings were discharged home from the housing facility. One infant sibling, five months of age was thought to be at higher risk for VZV infection due to age. Maternal varicella history was unknown.

Patients were stratified into 5 categories based on immune status and varicella serostatus or history of varicella vaccine (Table 1). The management of these 56 patients included: (1) VZV Immune globulin or IVIG based on availability, (2) acyclovir, or (3) no intervention. Patients who were at risk for contracting VZV were placed on isolation if they needed hospital care.

Family education

Immediate notification and education to the families regarding the chicken pox exposure was a priority. The pediatric teams called each patient or family that was exposed, and an informational letter was sent to all families residing at the pediatric cancer housing facility.

Discussion

This study describes the expeditious institutional response to a sustained exposure of many children with cancer to a child with primary varicella who was living in the same residential setting. The index case had a generalized VZV rash for one week prior to the actual diagnosis, raising concern that this child might have exposed a large number of pediatric oncology patients in a housing facility. The questions we faced included: How do we define an exposure? How do we define whether patients are immune to VZV, when they may be immunosuppressed from their oncological treatment? To whom do we provide prophylaxis against VZV and how do we do so? And lastly, who do we isolate post exposure and for how long?

Based on the Red Book guidelines, our infection control and epidemiology department created and implemented institution-specific guidelines (Table 2) that include a definition of exposures.⁴Normally for the typical exposures, we do ask each possibly exposed patient about duration and location of the exposure. However, in this specific case, the number of patients involved in this exposure over an entire week, in a housing facility was very difficult to assess.As it was thought that the index case child was in frequent, broad contact with other children, we had to make an empiric decision that anyone who was present in the housing facility during that week was considered exposed, in this case, 75 of 93 children identified as being in the facility.

Table 2.

Center-specific definitions of VZV exposure

Infectious Process and Route of Transmission	Body area site of rash	Geographic area	Time in area	Rule for seronegative exposed
Chicken Pox AIRBORNE AND CONTACT ROUTE	Any area	-CONFINED SPACE Household contact, Same inpatient room, PACC waiting area, Playroom -FACE TO FACE CONTACT	15 minutes Regardless of duration	Review visit/exposures for 2 days prior to rash onset Post exposure with VZIG or acyclovir for all seronegative individuals at clinician’s discretion Isolation – day 8–28**
Disseminated zoster Definition: >2 dermatomes above and below zoster >50 vesicular lesions >Localized zoster with clinician suspecting early dissemination AIRBORNE AND CONTACT ROUTE	Any area	Same as above	15 minutes	Exposure same day as rash appearance Post exposure with VZIG or acyclovir for all seronegative individuals at clinician’s discretion (see above) Isolation – day 8–28**
Varicella zoster-localized in immunosuppressed patient/HCW CONTACT ROUTE	Face/exposed area of body	Household contact, Same inpatient room, PACC waiting area, Playroom	Intimate contact with positive patient or HCW with presumed contact with skin lesions (i.e. hugging/touching)	Exposure same day as rash appearance Post exposure with VZIG or acyclovir for all seronegative individuals at clinician’s discretion (see above) Isolation – day 8–28**
Varicella zoster-localized in immunosuppressed patient/HCW covered by clothing	Local site covered by clothing	Household contact, Same inpatient room, PACC waiting area, Playroom	No exposure	No follow-up
Rash after varicella or shingles vaccine	Any area	Any area	No exposure	No follow-up

^**if VZIG not given, this duration is 8–21 days.

PACC, pediatric ambulatory care center; VZIG, varicella zoster immunoglobulin; HCW, health care worker

The most complex aspect of how to manage varicella zoster exposure in a pediatric oncology and hematopoietic stem cell transplant (HSCT) recipient patient population is the question of who to administer prophylaxis to:do we follow the VZV serostatus, do we check the general T-cell function, or do we check the actual T-cell in-vitro response to VZV?The immune response to VZV continues to be poorly understood,^{5, 6} and therefore, its assessment and testing remains difficult. It is clear that for seronegative patients, the risk of developing varicella infection is high. It is also clear that chemotherapy and immunotherapy can affect both T-cell and B-cell function,i.e. humoral immunity.⁷If this outbreak had occurred more recently patients treated with anti-B cell antibodies such as rituximab, or CD19-CAR T-cells would have been additional candidates for prophylaxis.

Based on data from several reports of VZV exposure or outbreaks in the oncology setting, where seropositive patients developed varicella infection post exposure, we cannot solely base our prophylaxis on VZV serostatus. Moreover, in our personal experience over many years, at least two seropositive patients who were immunosuppressed and declined prophylaxis developed a varicella infection. Our institution created empiric guidelines that constitute our standard of care for the management of patients exposed to VZV (Table 3). The recommendations for prophylaxis are for immunocompetent, VZV seronegative patients or immunosuppressed patients regardless of their VZV serostatus. Based on our personal experience and institutional guidelines, patients were considered immune suppressed if (1) patients were within 0–9 months post allogeneic HSCT, (2) patients were 0-6 months post autologous HSCT, (3) patients had immunosuppressive chemotherapy such as high dose cyclophosphamide, ifosfamide, or temozolomide or immunosuppressive agents such as alemtuzumab, rituximab, or high dose steroids in the preceding 0–6 months. In our patient population, all 11 patients who were seronegative and immunosuppressed received prophylaxis with immune globulin, whereas 6 of the 10 patients who were seropositive and immunosuppressed received prophylaxis with immune globulin or acyclovir; the other 4 patients were followed elsewhere, and the decision was not to administer prophylaxis to them (Table 1 Group 4). None of the 21 immunosuppressed patients developed secondary varicella.

Table 3.

Center-specific recommendations for Post VZV exposure prophylaxis

IMMUNE STATUS	NO PRIOR HISTORY OF CHICKEN POX OR VZV VACCINE	VZV seronegative (with or without History of VZV Vaccine)	HISTORY OF CHICKEN POX OR VZV VACCINE	VZV seropositive (with or without History of Chicken Pox or VZV Vaccine)
Immune Competent	Isolation day 8–21 post exposure - May consider Acyclovir OR  if not contraindication for vaccine, administer within 3–5 days of exposure		No action necessary
Immune Suppressed	- Administer VZIG up to 10 days post exposure - If > 10 days post exposure or cannot administer VZIG, antiviral prophylaxis with Acyclovir can be given 7–10 days post exposure - Isolation day 8–28 post exposure
Immune Status Unclear	- Administer VZIG up to 10 days post exposure - If > 10 days post exposure or cannot administer VZIG, antiviral prophylaxis with Acyclovir can be given 7–10 days post exposure - Isolation day 8–28 post exposure		To be assessed on a case-by-case basis with Infectious Disease Service
VZV Immune Globulin Not Given	- Isolation day 8–21 post exposure	- Isolation day 8–21 post exposure	No action necessary

VZV, varicella zoster virus; VZIG, varicella zoster immunoglobulin

Outbreaks of varicella infection have been described in pediatric oncology patients^8–10 including separate reports of 6, 7 and 12 patients. Kavallotis et al. describe the exposure of 79 pediatric oncology patients to VZV with 6 immunosuppressed patients who received immune globulin prophylaxis, but still developed varicella. Adler et al. report on an outbreak of 7 patients following the exposure of 82 families at a pediatric group housing facility, and the need for formal recommendations on the management of exposed individuals. Lastly, Manistarski et al. describe an outbreak of 12 patients with breakthrough varicella following an exposure, with a conclusion that“a positive IgG serology did not confer protection after exposure to varicella.”

Several VZV prophylaxis studies have been published by investigators from the United Kingdom. Fisher et al.reviewed the prevention of varicella in children with malignancies looking at the post-exposure prophylaxis with standard VZIG (10 studies) versus oral acyclovir (11 studies) as used in the United Kingdom and Japan, and strongly recommended randomized studies for post exposure prophylaxis comparing these two approaches.¹¹ A follow-up study byBeckingham et al.¹² reported on a VZV exposure of 25 pediatric oncology patients including 19 seropositive patients and 6 seronegative patients. Only the 6 seronegative children received VZV IgG prophylaxis, and all received acyclovir. None of the 19 seropositive patients, and 2 of the 6 seronegative patients prophylaxed with acyclovir developed evidence of secondary varicella. In a follow-up study,¹³ the authors reported on such a trial randomizing VZIG and acyclovir. After a 13-month period, over 482 patients were screened, 32 patients registered and only 3 patients randomized following a VZV exposure; all 3 patients were randomized to receive acyclovir, and none had breakthrough varicella. Giving the limited recruitment of the pilot trial, the investigators stopped the trial early.

Costa et al.¹⁴ described a retrospective study on the management of pediatric oncology patients following VZV exposure in Canada. The authors reviewed data on 51 patients including 19 patients treated with VZIG, 15 patients treated with acyclovir, and 18 patients who did not receive prophylaxis. The time to prophylaxis varied among patients. Six of the 17 patients who received VZIG, two of the 15 patients who received acyclovir, and 3 patients in the non-prophylaxis group developed varicella.

The timing of the immune globulin prophylaxis is also important. A review of 2 ½ years of data from Norway identified 130 children with malignant disorders or other immune conditions who received zoster immunoglobulin and whose data was available. Only 2.9% of patients who received VZIG within 3 days of the exposure developed varicella, compared to 37.5% when patients received VZIG within 4–5 days from exposure and 50% of patients who received VZIG ≥ 5 days from exposure.¹⁵ This study was important in the recommendation of the early VZIG post-exposure prophylaxis to occur within 72 hours following exposure.

Whether or not to isolate post-exposure in the final question we faced, but there are well documented recommendations from the American Academy of Pediatrics Red Book. The isolation period at the time of this exposure was day +10 to +21 post exposure if no prophylaxis is given, and day +10 to +28 post exposure for patients who receive prophylaxis. This post exposure isolation is a universal standard of care which we followed for our patient population. It is notable that the day +10 of beginning of incubation has changed to day +8 today (table 2).⁴

We understand there are several limitations to our manuscript. We cannot be certain that our intervention prevented secondary cases as no control group could be included. Another limitation of our study is the inability to truly ascertain by history who was indeed exposed to VZV, as, due to this large patient population, we had to assume that all children living in the housing facility for the long duration of time the index patient had active VZV infection were exposed.

In summary, we rapidly responded and triaged the exposure of 56 children with cancer to a child with primary varicella who lived with them in a housing facility for children and their families. We were able to strategically plan for its management of the exposure as one team including all the different disciplines with communication between the disciplines (figure 2). We saw no secondary cases following this exposure of 56 children which represents a significant achievement, especially with published reports often showing secondary VZV cases. This study is important in the way that it gives an example as to how a departmental operation can be strategized and include the leadership and the different parties involved in order to address a major infectious disease exposure; It also shows how this operation can reconcile different aspects of guidelines, including those of the Red Book, the Department Standard Operating Procedure together with the guidance of the Infection Control department and the expertise in immunology of the Pediatric Oncology and Bone Marrow Transplant staff.

Data Availability statement:

The authors confirm that the data supporting the findings of this study are available within the article.