bs_bs_banner
doi:10.1111/jpc.12502
ORIGINAL ARTICLE
Congenital cytomegalovirus infection is associated with high maternal socio-economic status and corresponding low maternal cytomegalovirus seropositivity James Basha,1,2,* Jenna M Iwasenko,1,2,* Peter Robertson,3 Maria E Craig4 and William D Rawlinson1,2,5 1 Virology and 3Serology, Department of Microbiology, SEALS, Prince of Wales Hospital, Randwick, 2School of Medical Sciences, Faculty of Medicine, 4School of Women’s and Children’s Health, 5ACPS, SOMS and BABS, University of New South Wales, Kensington, New South Wales, Australia
Aims: Human cytomegalovirus (CMV) is the leading infectious cause of congenital infection in developed countries. Globally, CMV seropositivity has been associated with low socio-economic status (SES); however, Australian data are lacking. Therefore, we examined the association between SES and CMV seroprevalence in children and pregnant women. Methods: Three groups were examined: 1, a prospective cohort of Australian children aged 0–15 years (n = 220); 2, a clinic-based sample of pregnant women (n = 778); and 3, a case series of infants and children (n = 219) with symptomatic congenital CMV infection. SES was determined using a postcode-based score from the Australian Bureau of Statistics.Group 1 was recruited from endocrinology clinics and follow-up at Prince of Wales Hospital and Children’s Hospital at Westmead. Group 2 was recruited at the Royal Hospital for Women. Congenitally infected infants were identified through the Australian Paediatric Surveillance Unit. Results: CMV seroprevalence among all children was 20% (95% confidence interval (CI) 15–25%), and there was no association with SES (P = 0.58). Seroprevalence among pregnant women was 57% (53–60%), and higher rates of CMV seropositivity were associated with lower SES (P < 0.001). More congenital CMV cases were reported in the highest socio-economic groups (55%) than the lowest (9%) (P < 0.001). Conclusions: A marked socio-economic gradient in CMV seroprevalence is evident in Australian pregnant women and cases of congenital CMV but not in unselected Australian children. These findings highlight the importance of a community-wide approach to CMV awareness and the potential for hygienic measures to reduce the burden of congenital CMV in Australia. Key words:
Australia; congenital infection; cytomegalovirus; seroprevalence; socio-economic status.
What is already known on this topic
What this paper adds
1 The seroprevalence of human cytomegalovirus (CMV) is higher in low-SES or developing countries compared with developed (high socio-economic status (SES)) countries. 2 Congenital CMV is the most frequent infectious cause of congenital malformation. 3 Seropositivity rates for CMV in mothers are ~40% and increased in developing countries.
1 Increased CMV seroprevalence in lower SES groups is apparent even within the more limited social stratification of a wealthy, developed country such as Australia. 2 Although CMV seroprevalence is higher in low-SES women, most reported cases of symptomatic congenital CMV were from high-SES areas in Australia. 3 CMV seroprevalence among Australian children is 20% with no association with socioeconomic status.
Human cytomegalovirus (CMV) is the leading infectious cause of congenital infection in Australia1 affecting an estimated 0.64% (0.5–2.2%) of livebirths.2 Reactivation of latent CMV infection with subsequent fetal transmission precedes a larger number of cases of congenital CMV.3 However, the risk of intra-uterine
Correspondence: Professor William Rawlinson, Virology, Department of Microbiology, SEALS, Prince of Wales Hospital, Randwick, NSW 2031, Australia. Fax: +61 2 9382 8533; email: [email protected] *Joint first authors Conflict of interest: None declared. Accepted for publication 19 November 2013.
368
transmission of CMV is significantly higher after primary maternal infection, with a 32% risk of transmission to the fetus.2 There is a significant global variation in the seroprevalence of CMV, with seropositivity ranging from 40% to 100% among different populations (Fig. 1). CMV seropositivity is associated with older age, female sex, low socio-economic status (SES), ethnic minorities and low-resource settings.4–6 Socio-economic disparities in the rates of CMV seropositivity appear early in life and become greater with increasing age, although the age at which the effects of socio-economic disparities become evident is unclear.5,7 In Australia, the population weighted rate of CMV seropositivity is 57%, and higher seroprevalence is associated with older age6 and with pregnancy.4 At present, there are no data reported
Journal of Paediatrics and Child Health 50 (2014) 368–372 © 2014 The Authors Journal of Paediatrics and Child Health © 2014 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
J Basha et al.
Cytomegalovirus and socio-economic status
(b) Australian pregnant women
90%
90% CMV seroprevalence (%)
CMV seroprevalence (%)
80% 70% 60% 50% 40% 30% 20% 10%
80% 70% 60% 50% 40% 30% 20% 10%
0%
0% 1–7 (n = 82)
8 (n = 30)
9 (n = 33)
10 (n = 75)
(c) Australian congenital CMV cases Reported cases of congenital CMV (%)
(a) Australian children
1–7 8 9 10 (n = 69) (n = 137) (n = 131) (n = 451)
Socioeconomic Decile
Socioeconomic Decile
45% 40% 35% 30% 25% 20% 15% 10% 5% 0% 1–2
3–4
5–6
7–8
9–10
Socioeconomic Decile
Fig. 1 Distribution of cytomegalovirus (CMV) seroprevalence by socio-economic decile. (a) CMV seroprevalence in Australian children (group 1, n = 220) (1–7 vs. 8–10, P = 0.58); (b) CMV seroprevalence in Australian pregnant women (group 2, n = 788) (1–7 vs. 8–10, P < 0.001); (c) cases of congenital CMV (group 3, n = 219) (1–2 vs. 8–10, P < 0.001).
regarding the influence of socio-economic factors on CMV serostatus and congenital CMV infection within the Australian population. Furthermore, serostatus and subsequent CMV reactivation3 make the relationship between population serostatus, socio-economic factors and rates of congenital CMV infection of interest. The aims of this study were to determine the prevalence of CMV seropositivity in Australian infants, children and pregnant women, examining the association between congenital CMV infection and SES.
Methods Patient populations Three groups were examined: 1, a prospective cohort of Australian children aged 0–15 years (n = 220) collected between 2000 and 2007; 2, a clinic-based sample of pregnant women (n = 778) collected between 2002 and 2004; and 3, a case series of infants and children (n = 219) with symptomatic congenital CMV infection collected between 1999 and 2010. Group 1 was from a prospective cohort of 220 Australian children with a family history of type 1 diabetes mellitus. Blood was collected at birth, and participants were tested regularly for CMV serostatus over a 7-year follow-up period. CMV serostatus at the most recent visit for each participant was used for analysis. Group 2 (n = 778), included 600 pregnant women described in Munro et al.4 and an additional 178 women recruited from the identical population since publication. The group 3 case series consisted of 219 symptomatic cases of congenital CMV reported to the Australian Paediatric Surveillance Unit (APSU) between January 1999 and August 2010 using definitions outlined by Srikanthan et al.8 Ethical approval was obtained from the Human Research Ethics Committee of the Children’s Hospital Westmead (09/CHW/27) and Northern Hospital Network (09/G/116; 02/085).
CMV was defined as a positive immunoglobulin G (IgG) serology result. CMV IgG was measured using either the Abbott Axsym (Abbott Diagnostics, Lake Forest, IL, USA; samples prior to June 2008) or the Abbott Architect (samples from June 2008 onwards).
SES classification Individual SES was classified using the Australian Bureau of Statistics Socioeconomic Index for Areas (SEIFA) 2006 Index of Relative Socioeconomic Advantage and Disadvantage9 which uses 21 variables including income, education, employment, occupation and housing.10 The score for each postal area is created by adding together the weighted characteristics of these variables for the area; the scores are then ranked and classified into deciles. The lowest 10% of SES advantage postal areas is represented by decile 1 through to the highest 10%, which represents decile 10 or the most advantaged SES. We defined low SES as SEIFA deciles 1–7 and high SES as 8–10, with deciles 1–7 combined for analysis. Exclusion criteria for patients in the analysis included incomplete postcode or serology data.
Statistics Descriptive statistics are reported as median and interquartile range (IQR) with CMV seroprevalence reported as percentage (95% confidence interval (CI)). Fisher’s exact and χ2-tests were used to examine association between groups, and results were reported as odds ratios (95% CI). Prism V5.03 (GraphPad, San Diego, CA, USA) and Dimension Research Confidence Interval for Proportion Calculator (Dimension Research Inc., DuPage County, IL, USA) were used for statistical analysis. Results with α < 0.05 were considered statistically significant.
Testing
Results
Serology was performed by the South Eastern Area Laboratory Services at Prince of Wales Hospital, Sydney. Seropositivity for
CMV seroprevalence in children at a median age of 4 years (IQR 2.8–5.1) was 20% (n = 44/220, 95% CI 15–25%). Seropositivity
Journal of Paediatrics and Child Health 50 (2014) 368–372 © 2014 The Authors Journal of Paediatrics and Child Health © 2014 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
369
Cytomegalovirus and socio-economic status
J Basha et al.
for CMV was higher in children aged >8 years (33%) compared with younger children (18%, P = 0.05). There was no association between CMV seropositivity rates and socio-economic deciles for this group of children (low SES 22% in vs. high SES 19%, P = 0.58) (Fig. 1A). Overall, 488/788 of pregnant Australian women were CMV IgG positive, giving an overall seroprevalence of 57% (95% CI 53–60%). Seropositivity was lower in those from a high SES (SES deciles 8–10) compared with a women in the lower seven deciles, with an odds ratio of 1.7 (95% CI 1.2–2.2, P < 0.001) (Fig. 1B). In the case series of congenital CMV in neonates Australiawide between 1999 and 2010, a larger proportion of congenital CMV cases were from high socio-economic deciles. Nine per cent (20/219) of congenital CMV cases were reported from socio-
economic deciles 1–2 and 55% (121/219) from socio-economic deciles 8–10 (Fig. 1C). The distribution of congenital CMV cases across deciles is significantly different to that expected because of the population distribution across the deciles (P < 0.001) where 44% of the population is in deciles 8–10. The prevalence of CMV seropositivity in our Australian populations, compared with data from other countries, shows CMV seroprevalence in the Australian populations studied is consistent with other high-SES countries (Fig. 2).
Discussion This analysis of three different population groups demonstrates that in Australia, higher SES is associated with lower
Fig. 2 Cytomegalovirus (CMV) seroprevalence in pregnant women by region and socioeconomic status in this study (black bars) and previously published studies (grey bars).11–23
370
Journal of Paediatrics and Child Health 50 (2014) 368–372 © 2014 The Authors Journal of Paediatrics and Child Health © 2014 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
J Basha et al.
CMV seroprevalence among pregnant women, and a higher maternal SES is associated with a greater risk of symptomatic congenital CMV (Fig. 1). This study confirms that the association between higher CMV seroprevalence and low SES seen at a global population level11–21 is reflected even within the more limited social stratification of a wealthy developed country (Australia). In a global context, CMV seropositivity among pregnant Australian women is comparable to the middle to high socio-economic groups of other high-resource countries11,18,20,21 (Fig. 2). The finding that most reported symptomatic cases of congenital CMV infection are in the population with the lowest maternal CMV seroprevalence and high maternal SES suggests that primary maternal infection is an important factor in the population assessed through the APSU programme (Fig. 1C). This agrees with studies which have found that primary maternal infection with CMV, that is CMV seroconversion during pregnancy, is more likely to result in intra-uterine transmission of CMV.2,24 This differs from recent data estimating significantly more cases of congenital CMV (75%) result from non-primary CMV infection.3 However, this discrepancy most likely results from the method of case ascertainment used in the APSU programme, which involves recognition by a paediatrician of congenital CMV in young infants, likely biasing cases to those most severely affected.1,4 In addition, Wang et al.3 identify that although primary maternal infection has a much higher rate of transplacental transmission than non-primary infection, the majority of pregnancies occurring in CMV seropositive women and that pregnancy represents a relatively short period of time in which to acquire primary infection. We observed no variation in rates of CMV seropositivity in children of different SES (Fig. 1A); however, there was an expected trend that CMV seropositivity was higher in older (>8 years) than younger children (Fig. 1A). The data suggest that at a young age (
doi:10.1111/jpc.12502
ORIGINAL ARTICLE
Congenital cytomegalovirus infection is associated with high maternal socio-economic status and corresponding low maternal cytomegalovirus seropositivity James Basha,1,2,* Jenna M Iwasenko,1,2,* Peter Robertson,3 Maria E Craig4 and William D Rawlinson1,2,5 1 Virology and 3Serology, Department of Microbiology, SEALS, Prince of Wales Hospital, Randwick, 2School of Medical Sciences, Faculty of Medicine, 4School of Women’s and Children’s Health, 5ACPS, SOMS and BABS, University of New South Wales, Kensington, New South Wales, Australia
Aims: Human cytomegalovirus (CMV) is the leading infectious cause of congenital infection in developed countries. Globally, CMV seropositivity has been associated with low socio-economic status (SES); however, Australian data are lacking. Therefore, we examined the association between SES and CMV seroprevalence in children and pregnant women. Methods: Three groups were examined: 1, a prospective cohort of Australian children aged 0–15 years (n = 220); 2, a clinic-based sample of pregnant women (n = 778); and 3, a case series of infants and children (n = 219) with symptomatic congenital CMV infection. SES was determined using a postcode-based score from the Australian Bureau of Statistics.Group 1 was recruited from endocrinology clinics and follow-up at Prince of Wales Hospital and Children’s Hospital at Westmead. Group 2 was recruited at the Royal Hospital for Women. Congenitally infected infants were identified through the Australian Paediatric Surveillance Unit. Results: CMV seroprevalence among all children was 20% (95% confidence interval (CI) 15–25%), and there was no association with SES (P = 0.58). Seroprevalence among pregnant women was 57% (53–60%), and higher rates of CMV seropositivity were associated with lower SES (P < 0.001). More congenital CMV cases were reported in the highest socio-economic groups (55%) than the lowest (9%) (P < 0.001). Conclusions: A marked socio-economic gradient in CMV seroprevalence is evident in Australian pregnant women and cases of congenital CMV but not in unselected Australian children. These findings highlight the importance of a community-wide approach to CMV awareness and the potential for hygienic measures to reduce the burden of congenital CMV in Australia. Key words:
Australia; congenital infection; cytomegalovirus; seroprevalence; socio-economic status.
What is already known on this topic
What this paper adds
1 The seroprevalence of human cytomegalovirus (CMV) is higher in low-SES or developing countries compared with developed (high socio-economic status (SES)) countries. 2 Congenital CMV is the most frequent infectious cause of congenital malformation. 3 Seropositivity rates for CMV in mothers are ~40% and increased in developing countries.
1 Increased CMV seroprevalence in lower SES groups is apparent even within the more limited social stratification of a wealthy, developed country such as Australia. 2 Although CMV seroprevalence is higher in low-SES women, most reported cases of symptomatic congenital CMV were from high-SES areas in Australia. 3 CMV seroprevalence among Australian children is 20% with no association with socioeconomic status.
Human cytomegalovirus (CMV) is the leading infectious cause of congenital infection in Australia1 affecting an estimated 0.64% (0.5–2.2%) of livebirths.2 Reactivation of latent CMV infection with subsequent fetal transmission precedes a larger number of cases of congenital CMV.3 However, the risk of intra-uterine
Correspondence: Professor William Rawlinson, Virology, Department of Microbiology, SEALS, Prince of Wales Hospital, Randwick, NSW 2031, Australia. Fax: +61 2 9382 8533; email: [email protected] *Joint first authors Conflict of interest: None declared. Accepted for publication 19 November 2013.
368
transmission of CMV is significantly higher after primary maternal infection, with a 32% risk of transmission to the fetus.2 There is a significant global variation in the seroprevalence of CMV, with seropositivity ranging from 40% to 100% among different populations (Fig. 1). CMV seropositivity is associated with older age, female sex, low socio-economic status (SES), ethnic minorities and low-resource settings.4–6 Socio-economic disparities in the rates of CMV seropositivity appear early in life and become greater with increasing age, although the age at which the effects of socio-economic disparities become evident is unclear.5,7 In Australia, the population weighted rate of CMV seropositivity is 57%, and higher seroprevalence is associated with older age6 and with pregnancy.4 At present, there are no data reported
Journal of Paediatrics and Child Health 50 (2014) 368–372 © 2014 The Authors Journal of Paediatrics and Child Health © 2014 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
J Basha et al.
Cytomegalovirus and socio-economic status
(b) Australian pregnant women
90%
90% CMV seroprevalence (%)
CMV seroprevalence (%)
80% 70% 60% 50% 40% 30% 20% 10%
80% 70% 60% 50% 40% 30% 20% 10%
0%
0% 1–7 (n = 82)
8 (n = 30)
9 (n = 33)
10 (n = 75)
(c) Australian congenital CMV cases Reported cases of congenital CMV (%)
(a) Australian children
1–7 8 9 10 (n = 69) (n = 137) (n = 131) (n = 451)
Socioeconomic Decile
Socioeconomic Decile
45% 40% 35% 30% 25% 20% 15% 10% 5% 0% 1–2
3–4
5–6
7–8
9–10
Socioeconomic Decile
Fig. 1 Distribution of cytomegalovirus (CMV) seroprevalence by socio-economic decile. (a) CMV seroprevalence in Australian children (group 1, n = 220) (1–7 vs. 8–10, P = 0.58); (b) CMV seroprevalence in Australian pregnant women (group 2, n = 788) (1–7 vs. 8–10, P < 0.001); (c) cases of congenital CMV (group 3, n = 219) (1–2 vs. 8–10, P < 0.001).
regarding the influence of socio-economic factors on CMV serostatus and congenital CMV infection within the Australian population. Furthermore, serostatus and subsequent CMV reactivation3 make the relationship between population serostatus, socio-economic factors and rates of congenital CMV infection of interest. The aims of this study were to determine the prevalence of CMV seropositivity in Australian infants, children and pregnant women, examining the association between congenital CMV infection and SES.
Methods Patient populations Three groups were examined: 1, a prospective cohort of Australian children aged 0–15 years (n = 220) collected between 2000 and 2007; 2, a clinic-based sample of pregnant women (n = 778) collected between 2002 and 2004; and 3, a case series of infants and children (n = 219) with symptomatic congenital CMV infection collected between 1999 and 2010. Group 1 was from a prospective cohort of 220 Australian children with a family history of type 1 diabetes mellitus. Blood was collected at birth, and participants were tested regularly for CMV serostatus over a 7-year follow-up period. CMV serostatus at the most recent visit for each participant was used for analysis. Group 2 (n = 778), included 600 pregnant women described in Munro et al.4 and an additional 178 women recruited from the identical population since publication. The group 3 case series consisted of 219 symptomatic cases of congenital CMV reported to the Australian Paediatric Surveillance Unit (APSU) between January 1999 and August 2010 using definitions outlined by Srikanthan et al.8 Ethical approval was obtained from the Human Research Ethics Committee of the Children’s Hospital Westmead (09/CHW/27) and Northern Hospital Network (09/G/116; 02/085).
CMV was defined as a positive immunoglobulin G (IgG) serology result. CMV IgG was measured using either the Abbott Axsym (Abbott Diagnostics, Lake Forest, IL, USA; samples prior to June 2008) or the Abbott Architect (samples from June 2008 onwards).
SES classification Individual SES was classified using the Australian Bureau of Statistics Socioeconomic Index for Areas (SEIFA) 2006 Index of Relative Socioeconomic Advantage and Disadvantage9 which uses 21 variables including income, education, employment, occupation and housing.10 The score for each postal area is created by adding together the weighted characteristics of these variables for the area; the scores are then ranked and classified into deciles. The lowest 10% of SES advantage postal areas is represented by decile 1 through to the highest 10%, which represents decile 10 or the most advantaged SES. We defined low SES as SEIFA deciles 1–7 and high SES as 8–10, with deciles 1–7 combined for analysis. Exclusion criteria for patients in the analysis included incomplete postcode or serology data.
Statistics Descriptive statistics are reported as median and interquartile range (IQR) with CMV seroprevalence reported as percentage (95% confidence interval (CI)). Fisher’s exact and χ2-tests were used to examine association between groups, and results were reported as odds ratios (95% CI). Prism V5.03 (GraphPad, San Diego, CA, USA) and Dimension Research Confidence Interval for Proportion Calculator (Dimension Research Inc., DuPage County, IL, USA) were used for statistical analysis. Results with α < 0.05 were considered statistically significant.
Testing
Results
Serology was performed by the South Eastern Area Laboratory Services at Prince of Wales Hospital, Sydney. Seropositivity for
CMV seroprevalence in children at a median age of 4 years (IQR 2.8–5.1) was 20% (n = 44/220, 95% CI 15–25%). Seropositivity
Journal of Paediatrics and Child Health 50 (2014) 368–372 © 2014 The Authors Journal of Paediatrics and Child Health © 2014 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
369
Cytomegalovirus and socio-economic status
J Basha et al.
for CMV was higher in children aged >8 years (33%) compared with younger children (18%, P = 0.05). There was no association between CMV seropositivity rates and socio-economic deciles for this group of children (low SES 22% in vs. high SES 19%, P = 0.58) (Fig. 1A). Overall, 488/788 of pregnant Australian women were CMV IgG positive, giving an overall seroprevalence of 57% (95% CI 53–60%). Seropositivity was lower in those from a high SES (SES deciles 8–10) compared with a women in the lower seven deciles, with an odds ratio of 1.7 (95% CI 1.2–2.2, P < 0.001) (Fig. 1B). In the case series of congenital CMV in neonates Australiawide between 1999 and 2010, a larger proportion of congenital CMV cases were from high socio-economic deciles. Nine per cent (20/219) of congenital CMV cases were reported from socio-
economic deciles 1–2 and 55% (121/219) from socio-economic deciles 8–10 (Fig. 1C). The distribution of congenital CMV cases across deciles is significantly different to that expected because of the population distribution across the deciles (P < 0.001) where 44% of the population is in deciles 8–10. The prevalence of CMV seropositivity in our Australian populations, compared with data from other countries, shows CMV seroprevalence in the Australian populations studied is consistent with other high-SES countries (Fig. 2).
Discussion This analysis of three different population groups demonstrates that in Australia, higher SES is associated with lower
Fig. 2 Cytomegalovirus (CMV) seroprevalence in pregnant women by region and socioeconomic status in this study (black bars) and previously published studies (grey bars).11–23
370
Journal of Paediatrics and Child Health 50 (2014) 368–372 © 2014 The Authors Journal of Paediatrics and Child Health © 2014 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
J Basha et al.
CMV seroprevalence among pregnant women, and a higher maternal SES is associated with a greater risk of symptomatic congenital CMV (Fig. 1). This study confirms that the association between higher CMV seroprevalence and low SES seen at a global population level11–21 is reflected even within the more limited social stratification of a wealthy developed country (Australia). In a global context, CMV seropositivity among pregnant Australian women is comparable to the middle to high socio-economic groups of other high-resource countries11,18,20,21 (Fig. 2). The finding that most reported symptomatic cases of congenital CMV infection are in the population with the lowest maternal CMV seroprevalence and high maternal SES suggests that primary maternal infection is an important factor in the population assessed through the APSU programme (Fig. 1C). This agrees with studies which have found that primary maternal infection with CMV, that is CMV seroconversion during pregnancy, is more likely to result in intra-uterine transmission of CMV.2,24 This differs from recent data estimating significantly more cases of congenital CMV (75%) result from non-primary CMV infection.3 However, this discrepancy most likely results from the method of case ascertainment used in the APSU programme, which involves recognition by a paediatrician of congenital CMV in young infants, likely biasing cases to those most severely affected.1,4 In addition, Wang et al.3 identify that although primary maternal infection has a much higher rate of transplacental transmission than non-primary infection, the majority of pregnancies occurring in CMV seropositive women and that pregnancy represents a relatively short period of time in which to acquire primary infection. We observed no variation in rates of CMV seropositivity in children of different SES (Fig. 1A); however, there was an expected trend that CMV seropositivity was higher in older (>8 years) than younger children (Fig. 1A). The data suggest that at a young age (
