THE JOURNAL OF INFECTIOUS DISEASES. VOL. 139, NO.5. MAY 1979 © 1979 by The University of Chicago. 0022-1899179/3905-0008$00.75
Primary Infection with Epstein-Barr Virus in Infants in the United States: Clinical and Serologic Observations From the Division of Virology, The Joseph Stokes, Jr. Research Institute of The Children's Hospital of Philadelphia, and the School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania
Gary Fleisher, Werner Henle, Gertrude Henle, Evelyne T. Lennette, and Robert
J. Biggar
Epstein-Barr virus (EBV) is the cause of infectious mononucleosis (1M) [lJ, which affects predominantly adolescents and young adults from relatively affluent backgrounds in developed countries [2J, However, typical cases confirmed by EBV-specific serology are seen occasionally in younger [3-5J or older individuals [6]. The EBVrelated antibody responses in 1M are similar at any age [7]. Titers of IgM and IgG antibodies to the viral capsid antigen (VCA) of EBV peak in the early acute phase. In the majority of pa-
tients with 1M, antibodies to the diffuse component of the EBV-induced early antigen also appear in the acute phase, whereas antibodies to the restricted component of early antigen become detectable rarely, and then only later in the course of protracted cases [8]. The IgM antibodies to VCA and antibodies to the diffuse component of early antigen disappear in one to two months, and the IgG antibodies to VCA decline to lower but persistent levels. Antibodies to the EBV-associated nuclear antigen (EBNA) emerge only weeks to months after .onset of 1M and then remain present for life. Over 90% of patients with 1M develop transient heterophil antibody responses [7J, which, with rare exception, are specific for the disease. In studies in Ghana of infants who were examined at monthly intervals, Biggar et al. [9, 10J found a seroconversion rate for antibodies to EBV of 81 % by the age of 21 months. None of these infants revealed signs of 1M. Despite monthly histories, physical examinations, counts of white blood cells, and differential counts, no specific clinical or hematological patterns emerged. The IgG antibody response to VCA was similar to that seen in 1M, as was that to EBNA. However, IgM antibodies to VCA ap-
Received for publication July 27, 1978, and in revised form October 26, 1978. This work was supported by research grant no. CA-04568 and contract no. NOI-CP-33272 from the National Cancel Institute, U.S. Public Health Service. Werner Henle is the recipient of career award no. 5-K6AI-22683 from the National Institutes of Health. Robert J. Biggar is the Project Officer (on site), Burkitt's Tumor Project, University of Ghana Medical School, Accra, Ghana, and the National Cancer Institute, Bethesda, Maryland. We thank the staff of the Emergency Room, The Children's Hospital of Philadelphia, for assistance in the collection of specimens, and Ms. Sheila Kelly for technical assistance. Please address requests for reprints to Dr. Werner Henle, The Children's Hospital of Philadelphia, 34th Street and Civic Center Boulevard, Philadelphia, Pennsylvania 19104.
553
Downloaded from http://jid.oxfordjournals.org/ at University of Illinois at Urbana-Champaign on September 7, 2015
Previous studies in Ghana had shown that primary infections with Epstein-Barr virus in infants under the age of two years remain silent and evoke antibody responses different from those seen in infectious mononucleosis. In order to determine whether or not these observations were limited to Africa, 80 American infants presenting with minor infectious complaints were studied serologically; 14 (17.5%) showed evidence of recent or current primary infections with Epstein-Barr virus. The clinical features of these 14 infants were similar to those of the other 66 and did not suggest a diagnosis of infectious mononucleosis, nor were there histories of a recent infectious mononucleosis-like illness. Ten (72%) had antibodies to the early antigen complex induced by Epstein-Barr virus; however, these antibodies were directed, as in the Ghanaian infants, against the restricted rather than the diffuse components, in contrast to the pattern generally observed in infectious mononucleosis. Possible reasons for the differences between the responses of infants and those of older individuals to primary infection with Epstein-Barr virus and to the early antigen complex are discussed.
Fleisher et al.
554
Materials and Methods
Selection of patients. Eighty infants under the age of two years presenting to the outpatient department of The Children's Hospital of Philadelphia were studied, as approved by the hospital's Committee on Protection of Human Subjects. They were enrolled consecutively if they had a complaint of possible infectious etiology. There were equal numbers of males and females; 75% of the infants were black and the remainder white. The majority of families were of low socioeconomic status, as is the population served by the outpatient department. All infants had been delivered at term without complications and were judged to be normal at birth. None had a history of major illness, episodes compatible with 1M, or recurrent infections indicative of immunodeficiency. During the visit, a complete physical examination was performed, with particular attention to the pharynx, lymph nodes, spleen, and liver. Hematologic studies. Differential leukocyte counts were ~btained for all infants. Wrightstained blood smears were evaluated for atypical Iym phocytes. Heterophil antibody tests. The immune adherence hemagglutination test [11] and the ox-
cell hemolysin test [12] were employed to assay heterophil antibodies. A titer of >1:40 was taken as evidence of an 1M-specific heterophil antibody response. EB V serologic studies. 19M and 19G antibodies to VCA and 19G antibodies to the restricted and diffuse components of the EBV-induced early antigen complex were titrated by indirect immunofluorescence [7, 13]. Antibodies to EBNA were detected by anticomplement immunofluorescence [14]. Results
Results of the serologic tests for EBV infections in the 80 infants, by age, are shown in table 1. Four of the eight infants less than six months of age had maternal 19G antibody to VCA at titers of 1: 10-1 :40. Protection from EBV infection appears to persist for two to three months after maternal antibodies are no longer detectable [9]. Overall, 60 infants (75%) had no serologic evidence of EBV infection, six (7.5%) had antibody patterns indicating past infection, and 14 (17.5%) had evidence of recent or current infection. The number of infants with past EBV infection increased from none of eight at three to six months of age to nine of 18 at 19-24 months of age. No race-associated differences were apparent. Among the 14 children with serologically
Table 1. Incidence of serologic evidence of no, past, recent, and current primary infections with Epstein-Barr virus in infants under two years of age. Age in months (no.} 3-6 (8) 7-12(34) 13-18 (20) 19-24 (18) Total (80)
No. (%) with indicated status of infection* None
Past
Recent
Current
8(100)t 31 (91) 12 (60) 9 (50)
0 0 2 (10) 4 (22)
0 1 (3) 3 (15) 2 (11)
0 2 (6) 3 (15) 3 (17)
60 (75)
6 (7.5)
6 (7.5)
8 (10)
'None = no antibodies to Epstein-Barr virus; past = no IgM antibody to viral capsid antigen (VCA) but moderate titers of IgG antibody to VCA and antibody to Epstein-Barr virus-associated nuclear antigen (EBNA); recent = no IgM antibody to VCA, high titer of IgG antibody to VCA, and absence or low titer of antibody to EBNA; and current = presence of IgM antibody to VCA, high titer of IgG antibody to VCA, and no antibody to EBNA. tFour (50%) with maternal IgG antibodies to VCA.
Downloaded from http://jid.oxfordjournals.org/ at University of Illinois at Urbana-Champaign on September 7, 2015
peared in reduced titer. Most strikingly, the majority of the infants developed antibodies to the restricted component of the early antigen complex instead of the diffuse component; this pattern, the opposite of that seen in 1M, was previously undescribed. Heterophil antibody responses were either minimal or not detectable. The present study was undertaken to provide comparative data on primary infections with EBV in infants in the United States. Since it would have been difficult to arrange for serial monthly bleedings of a large group of American infants during the first two years of life, consecutively chosen infants were studied clinically, hematologically, and serologically on single occasions. As a result, several recent or current primary infections with EBV were identified on the basis of serologic data. As will be described here, the clinical, hematological, and serologic data on these American infants indicate that their response to primary EBV infections was similar to that of the Ghanaian infants.
EBV Infection in Infants
Table 2.
555
Clinical findings in relation to serology of Epstein-Barr virus. No. (%) with indicated serology
Clinical findings
Past infection (n = 6)
48 15 8 1 7
(80) (25) (13) (2) (12)
3 (50) 3 (33) I (17) 0 0
0 15 29 9 1 6
(25) (48) (15) (2) (10)
0 2 (33) 2 (33) I (17) 0 1 (17)
documented recent or current infection with EBV, 10 (71%) had fever, three (21%) had pharyngeal inflammation, and none had significant lymphadenopathy (table 2). The diagnosis assigned on discharge from the outpatient department was pharyngitis in three of these infants (21%), upper respiratory tract infection in six (43%), otitis media in four (29%), and gastroenteritis in one (7%). On a clinical basis none of the children was suspected of having 1M at the time blood was drawn, and none had a recent history of 1M-like illness. Three of these 14 children with recent or current primary EBV infections had 2%-3% atypical lymphocytes on their peripheral blood smears. Similarly, 51 (77%) of the 66 children without recent or current EBV infections had fever, 18 (27%) had pharyngeal inflammation, nine (14%) had significant anterior cervical lymphadenopathy, and one (1.5%) had hepatosplenomegaly. The diagnosis on discharge was pharyngitis in 17 (27%) of these infants, upper respiratory tract infection in 31 (37%), otitis media in 10 (15%), and gastroenteritis in seven (11 %). Thirteen children had atypical lymphocyte counts of 1%-6%. The serologic patterns of the 14 infants with recent or current infections are presented in table 3. These data were collected on a single occasion, which allows the possibility of missing a transient preceding (lgM antibody to VCA) or subsequent (antibody to early antigen complex) response, a problem not encountered in a primary
Recent or current infection (n = 14) 10 (71) 3 (21) 0 0 0 0 3 6 4 0 1
(21) (43) (29) (7)
Total (n
= 80)
61 20 9 1 7
(76) (25) (11) (1) (9)
0 20 37 14 1 8
(25) (46) (18) (I) (10)
EBV infection with defined onset of illness. Three of the 14 infants had blood drawn sufficiently early in the course of infection to have titers of IgM antibody to VCA of I: 10-1 :40. Titers of IgG antibody to VCA ranged from I :40 to I: 1,280 and were ~l :320 in nine cases. None of the infants had antibodies to the diffuse component of the early antigen complex, but 10 had antibodies to the restricted component at titers ranging from 1:10 to 1:80. Patients no. 6 and no. I n, who had no antibodies to the early antigen complex, had blood drawn too early in the course of the infection for such a response to be expected. The titers of antibody to EBNA were I :40 in either the immune adherence hemagglutination test or the ox-cell hemolysin test. Discussion
Earlier studies have demonstrated an increasing incidence of EBV seropositivity with age, at rates inversely related to socioeconomic status [15-18]. Of the infants between 19 and 24 months of age tested in this study, 50% were seropositive, whereas 8JC1o of the infants studied in Ghana had antibodies to EBV at 21 months. This geographic difference has been previously documented and presumably stems from increased personal contact under conditions of crowding and/or poor hygiene.
Downloaded from http://jid.oxfordjournals.org/ at University of Illinois at Urbana-Champaign on September 7, 2015
Physical signs Fever Pharyngeal inflammation Adenopathy Hepatosplenomegaly Rash Diagnosis Infectious mononucleosis Pharyngitis Upper respiratory tract infection a titis media Adenitis Gastroen teri tis
No infection (n = 60)
Fleisher et al.
556
Table 3.
Titers of Epstein-Barr virus-specific and heterophil antibody in infants with recent or current infections.
Infant no.
Age (months) 15 10 24 24 12 12 8 17 16 12
11 18 18 23
IgM-VCA
Primary Infection with Epstein-Barr Virus in Infants in the United States: Clinical and Serologic Observations From the Division of Virology, The Joseph Stokes, Jr. Research Institute of The Children's Hospital of Philadelphia, and the School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania
Gary Fleisher, Werner Henle, Gertrude Henle, Evelyne T. Lennette, and Robert
J. Biggar
Epstein-Barr virus (EBV) is the cause of infectious mononucleosis (1M) [lJ, which affects predominantly adolescents and young adults from relatively affluent backgrounds in developed countries [2J, However, typical cases confirmed by EBV-specific serology are seen occasionally in younger [3-5J or older individuals [6]. The EBVrelated antibody responses in 1M are similar at any age [7]. Titers of IgM and IgG antibodies to the viral capsid antigen (VCA) of EBV peak in the early acute phase. In the majority of pa-
tients with 1M, antibodies to the diffuse component of the EBV-induced early antigen also appear in the acute phase, whereas antibodies to the restricted component of early antigen become detectable rarely, and then only later in the course of protracted cases [8]. The IgM antibodies to VCA and antibodies to the diffuse component of early antigen disappear in one to two months, and the IgG antibodies to VCA decline to lower but persistent levels. Antibodies to the EBV-associated nuclear antigen (EBNA) emerge only weeks to months after .onset of 1M and then remain present for life. Over 90% of patients with 1M develop transient heterophil antibody responses [7J, which, with rare exception, are specific for the disease. In studies in Ghana of infants who were examined at monthly intervals, Biggar et al. [9, 10J found a seroconversion rate for antibodies to EBV of 81 % by the age of 21 months. None of these infants revealed signs of 1M. Despite monthly histories, physical examinations, counts of white blood cells, and differential counts, no specific clinical or hematological patterns emerged. The IgG antibody response to VCA was similar to that seen in 1M, as was that to EBNA. However, IgM antibodies to VCA ap-
Received for publication July 27, 1978, and in revised form October 26, 1978. This work was supported by research grant no. CA-04568 and contract no. NOI-CP-33272 from the National Cancel Institute, U.S. Public Health Service. Werner Henle is the recipient of career award no. 5-K6AI-22683 from the National Institutes of Health. Robert J. Biggar is the Project Officer (on site), Burkitt's Tumor Project, University of Ghana Medical School, Accra, Ghana, and the National Cancer Institute, Bethesda, Maryland. We thank the staff of the Emergency Room, The Children's Hospital of Philadelphia, for assistance in the collection of specimens, and Ms. Sheila Kelly for technical assistance. Please address requests for reprints to Dr. Werner Henle, The Children's Hospital of Philadelphia, 34th Street and Civic Center Boulevard, Philadelphia, Pennsylvania 19104.
553
Downloaded from http://jid.oxfordjournals.org/ at University of Illinois at Urbana-Champaign on September 7, 2015
Previous studies in Ghana had shown that primary infections with Epstein-Barr virus in infants under the age of two years remain silent and evoke antibody responses different from those seen in infectious mononucleosis. In order to determine whether or not these observations were limited to Africa, 80 American infants presenting with minor infectious complaints were studied serologically; 14 (17.5%) showed evidence of recent or current primary infections with Epstein-Barr virus. The clinical features of these 14 infants were similar to those of the other 66 and did not suggest a diagnosis of infectious mononucleosis, nor were there histories of a recent infectious mononucleosis-like illness. Ten (72%) had antibodies to the early antigen complex induced by Epstein-Barr virus; however, these antibodies were directed, as in the Ghanaian infants, against the restricted rather than the diffuse components, in contrast to the pattern generally observed in infectious mononucleosis. Possible reasons for the differences between the responses of infants and those of older individuals to primary infection with Epstein-Barr virus and to the early antigen complex are discussed.
Fleisher et al.
554
Materials and Methods
Selection of patients. Eighty infants under the age of two years presenting to the outpatient department of The Children's Hospital of Philadelphia were studied, as approved by the hospital's Committee on Protection of Human Subjects. They were enrolled consecutively if they had a complaint of possible infectious etiology. There were equal numbers of males and females; 75% of the infants were black and the remainder white. The majority of families were of low socioeconomic status, as is the population served by the outpatient department. All infants had been delivered at term without complications and were judged to be normal at birth. None had a history of major illness, episodes compatible with 1M, or recurrent infections indicative of immunodeficiency. During the visit, a complete physical examination was performed, with particular attention to the pharynx, lymph nodes, spleen, and liver. Hematologic studies. Differential leukocyte counts were ~btained for all infants. Wrightstained blood smears were evaluated for atypical Iym phocytes. Heterophil antibody tests. The immune adherence hemagglutination test [11] and the ox-
cell hemolysin test [12] were employed to assay heterophil antibodies. A titer of >1:40 was taken as evidence of an 1M-specific heterophil antibody response. EB V serologic studies. 19M and 19G antibodies to VCA and 19G antibodies to the restricted and diffuse components of the EBV-induced early antigen complex were titrated by indirect immunofluorescence [7, 13]. Antibodies to EBNA were detected by anticomplement immunofluorescence [14]. Results
Results of the serologic tests for EBV infections in the 80 infants, by age, are shown in table 1. Four of the eight infants less than six months of age had maternal 19G antibody to VCA at titers of 1: 10-1 :40. Protection from EBV infection appears to persist for two to three months after maternal antibodies are no longer detectable [9]. Overall, 60 infants (75%) had no serologic evidence of EBV infection, six (7.5%) had antibody patterns indicating past infection, and 14 (17.5%) had evidence of recent or current infection. The number of infants with past EBV infection increased from none of eight at three to six months of age to nine of 18 at 19-24 months of age. No race-associated differences were apparent. Among the 14 children with serologically
Table 1. Incidence of serologic evidence of no, past, recent, and current primary infections with Epstein-Barr virus in infants under two years of age. Age in months (no.} 3-6 (8) 7-12(34) 13-18 (20) 19-24 (18) Total (80)
No. (%) with indicated status of infection* None
Past
Recent
Current
8(100)t 31 (91) 12 (60) 9 (50)
0 0 2 (10) 4 (22)
0 1 (3) 3 (15) 2 (11)
0 2 (6) 3 (15) 3 (17)
60 (75)
6 (7.5)
6 (7.5)
8 (10)
'None = no antibodies to Epstein-Barr virus; past = no IgM antibody to viral capsid antigen (VCA) but moderate titers of IgG antibody to VCA and antibody to Epstein-Barr virus-associated nuclear antigen (EBNA); recent = no IgM antibody to VCA, high titer of IgG antibody to VCA, and absence or low titer of antibody to EBNA; and current = presence of IgM antibody to VCA, high titer of IgG antibody to VCA, and no antibody to EBNA. tFour (50%) with maternal IgG antibodies to VCA.
Downloaded from http://jid.oxfordjournals.org/ at University of Illinois at Urbana-Champaign on September 7, 2015
peared in reduced titer. Most strikingly, the majority of the infants developed antibodies to the restricted component of the early antigen complex instead of the diffuse component; this pattern, the opposite of that seen in 1M, was previously undescribed. Heterophil antibody responses were either minimal or not detectable. The present study was undertaken to provide comparative data on primary infections with EBV in infants in the United States. Since it would have been difficult to arrange for serial monthly bleedings of a large group of American infants during the first two years of life, consecutively chosen infants were studied clinically, hematologically, and serologically on single occasions. As a result, several recent or current primary infections with EBV were identified on the basis of serologic data. As will be described here, the clinical, hematological, and serologic data on these American infants indicate that their response to primary EBV infections was similar to that of the Ghanaian infants.
EBV Infection in Infants
Table 2.
555
Clinical findings in relation to serology of Epstein-Barr virus. No. (%) with indicated serology
Clinical findings
Past infection (n = 6)
48 15 8 1 7
(80) (25) (13) (2) (12)
3 (50) 3 (33) I (17) 0 0
0 15 29 9 1 6
(25) (48) (15) (2) (10)
0 2 (33) 2 (33) I (17) 0 1 (17)
documented recent or current infection with EBV, 10 (71%) had fever, three (21%) had pharyngeal inflammation, and none had significant lymphadenopathy (table 2). The diagnosis assigned on discharge from the outpatient department was pharyngitis in three of these infants (21%), upper respiratory tract infection in six (43%), otitis media in four (29%), and gastroenteritis in one (7%). On a clinical basis none of the children was suspected of having 1M at the time blood was drawn, and none had a recent history of 1M-like illness. Three of these 14 children with recent or current primary EBV infections had 2%-3% atypical lymphocytes on their peripheral blood smears. Similarly, 51 (77%) of the 66 children without recent or current EBV infections had fever, 18 (27%) had pharyngeal inflammation, nine (14%) had significant anterior cervical lymphadenopathy, and one (1.5%) had hepatosplenomegaly. The diagnosis on discharge was pharyngitis in 17 (27%) of these infants, upper respiratory tract infection in 31 (37%), otitis media in 10 (15%), and gastroenteritis in seven (11 %). Thirteen children had atypical lymphocyte counts of 1%-6%. The serologic patterns of the 14 infants with recent or current infections are presented in table 3. These data were collected on a single occasion, which allows the possibility of missing a transient preceding (lgM antibody to VCA) or subsequent (antibody to early antigen complex) response, a problem not encountered in a primary
Recent or current infection (n = 14) 10 (71) 3 (21) 0 0 0 0 3 6 4 0 1
(21) (43) (29) (7)
Total (n
= 80)
61 20 9 1 7
(76) (25) (11) (1) (9)
0 20 37 14 1 8
(25) (46) (18) (I) (10)
EBV infection with defined onset of illness. Three of the 14 infants had blood drawn sufficiently early in the course of infection to have titers of IgM antibody to VCA of I: 10-1 :40. Titers of IgG antibody to VCA ranged from I :40 to I: 1,280 and were ~l :320 in nine cases. None of the infants had antibodies to the diffuse component of the early antigen complex, but 10 had antibodies to the restricted component at titers ranging from 1:10 to 1:80. Patients no. 6 and no. I n, who had no antibodies to the early antigen complex, had blood drawn too early in the course of the infection for such a response to be expected. The titers of antibody to EBNA were I :40 in either the immune adherence hemagglutination test or the ox-cell hemolysin test. Discussion
Earlier studies have demonstrated an increasing incidence of EBV seropositivity with age, at rates inversely related to socioeconomic status [15-18]. Of the infants between 19 and 24 months of age tested in this study, 50% were seropositive, whereas 8JC1o of the infants studied in Ghana had antibodies to EBV at 21 months. This geographic difference has been previously documented and presumably stems from increased personal contact under conditions of crowding and/or poor hygiene.
Downloaded from http://jid.oxfordjournals.org/ at University of Illinois at Urbana-Champaign on September 7, 2015
Physical signs Fever Pharyngeal inflammation Adenopathy Hepatosplenomegaly Rash Diagnosis Infectious mononucleosis Pharyngitis Upper respiratory tract infection a titis media Adenitis Gastroen teri tis
No infection (n = 60)
Fleisher et al.
556
Table 3.
Titers of Epstein-Barr virus-specific and heterophil antibody in infants with recent or current infections.
Infant no.
Age (months) 15 10 24 24 12 12 8 17 16 12
11 18 18 23
IgM-VCA
