Australian and New Zealand Journal of Obstetrics and Gynaecology 2016; 56: 69–74
DOI: 10.1111/ajo.12378
Original Article
Early-onset neonatal group B streptococcus sepsis following national risk-based prevention guidelines Brian A. DARLOW,1 Lesley VOSS,2 Diana R. LENNON3,4 and Keith GRIMWOOD5,* 1
Department of Paediatrics, University of Otago Christchurch, Christchurch, 2Department of Paediatric Infectious Diseases, Starship Children’s Hospital, Auckland, 3University of Auckland, 4Kidz First and Starship Children’s Hospitals, Auckland, New Zealand and 5Queensland Children’s Medical Research Institute, Children’s Health Queensland, South Brisbane, Queensland, Australia
Background: Neonatal infection with group B streptococcus (GBS) is an important cause of infant mortality. Intrapartum antibiotics reduce early-onset GBS sepsis, but recommendations vary as to whether they should be offered following antenatal screening or based on risk factors alone. We aimed to determine the incidence of early-onset GBS sepsis in New Zealand five years after the publication of national risk-based GBS prevention guidelines. Materials and Methods: Prospective surveillance of early-onset GBS sepsis (defined as infection in the first 48 h of life) was undertaken between April 2009 and March 2011 through the auspices of the New Zealand Paediatric Surveillance Unit as part of a survey of infection presenting in the first week of life. Results: There were 29 cases of confirmed early-onset GBS sepsis, including one case of meningitis, giving an incidence rate of 0.23 per 1000 (95% CI 0.16–0.33) live births. Three infants (10.3%) died. In 16 cases (55%), a maternal risk factor qualifying the mother for intrapartum antibiotics was present, but only five (31%) received this intervention. A retrospective review of the major hospital laboratory databases for this period identified two additional cases. A secondary sensitivity analysis taking account of these cases provided an estimated national incidence of 0.26 (95% CI 0.18–0.37) per 1000 live births. Conclusions: Ten years after a similar survey and five years after promoting a single, risk-based prevention protocol nationally, the incidence of early-onset GBS disease in New Zealand has more than halved, but opportunities remain to further reduce the rate. Key words: newborn, prevention and control, sepsis, Streptococcus agalactiae, surveillance and monitoring.
Introduction Early-onset neonatal sepsis from Streptococcus agalactiae (group B streptococcus: GBS) is an important cause of neonatal mortality and morbidity in high-income countries.1 Intrapartum antibiotic prophylaxis (IAP) interrupts mother-to-baby transmission of GBS, reducing the risk of early-onset neonatal infection.1,2 However, different countries and jurisdictions have adopted varying recommendations as to whether IAP should be offered
Correspondence: Dr Brian A. Darlow, Cure Kids Professor of Paediatric Research, Department of Pediatrics, University of Otago Christchurch, PO Box 4345, Christchurch 8140, New Zealand. Email: [email protected] *Present address: Menzies Health Institute Queensland, Griffith University and Gold Coast University Hospital, Gold Coast Campus, Queensland, Australia. Received 31 March 2015; accepted 2 June 2015.
according to antenatal screening results for GBS carriage or by using clinical risk factors.3–5 During 1998–1999, we undertook a two-year prospective national survey of earlyonset neonatal GBS disease (defined as infection in the first 48 h of life) in New Zealand and reported the attack rate as 0.5 per 1000 live births.6 A nationally representative, multidisciplinary group met in 2004 and recommended a risk-based prevention strategy as a single national policy to reduce early-onset GBS disease in New Zealand (Table 1).7 The aim of the present study was to ascertain the incidence of early-onset neonatal GBS through prospective surveillance over a two-year period and five years after introducing the national prevention guidelines.
Materials and Methods This prospective survey was undertaken under the auspices of the New Zealand Paediatric Surveillance Unit (NZPSU).8,9 All registered paediatricians and other
© 2015 The Royal Australian and New Zealand College of Obstetricians and Gynaecologists The Australian and New Zealand Journal of Obstetrics and Gynaecology
69
B. A. Darlow et al.
Table 1 Maternal risk factors for early-onset neonatal group B streptococcus infection7 Only one of the following required for IAP* Previous group B streptococcus (GBS)-infected baby GBS bacteriuria (of any count) this pregnancy Preterm (38°C Membrane rupture >18 h *IAP, intrapartum antibiotic prophylaxis administered during labour as either intravenous penicillin or amoxycillin.
clinicians working predominantly with children are sent a card (by email or post) and asked to indicate whether or not they had seen any of the listed conditions in the previous month. Early-onset neonatal sepsis was listed as a condition under surveillance from April 2009 to March 2011. Early-onset neonatal sepsis was defined as infection in the first six days of life, where the infant had a clinical picture consistent with sepsis and a positive bacterial culture from blood, cerebrospinal fluid (CSF), or pleural fluid, together with supporting laboratory evidence of sepsis (leukopenia, leukocytosis, thrombocytopenia, or elevated serum C-reactive protein concentration). Meningitis was defined as a clinical picture consistent with meningitis, and either a positive bacterial culture in the CSF, or a positive blood culture and CSF pleocytosis (≥100 white blood cells 9 106/L).10,11 Upon notifying a case, clinicians received a two-page questionnaire. To ensure confidentiality and avoid multiple notifications, cases were identified by initials and date of birth. Information sought included maternal age; parent-ascribed ethnicity; parity; history of previous infants with GBS sepsis; risk factors for early-onset infection in this pregnancy; whether the obstetric caregiver followed an early-onset GBS prevention protocol; whether antibiotics were received; their timing, doses, and routes; the infant’s gestation; birthweight; place of birth and hospital providing primary care; age of symptom onset; and the timing of blood and CSF specimens. Laboratory test confirmation was sought, including bacterial pathogen identity. Outcome data for the infant included whether the infant survived or died and, if the latter, whether sepsis was a contributing factor. In New Zealand, pregnant women are required to choose a Lead Maternity Carer (LMC) to provide their care and who may be a midwife (either community or hospital based), obstetrician, or general practitioner.12 Most women currently choose a midwife as their LMC. In this report, we have confined data to the first 48 h of life to compare with our earlier 1998–1999 survey, undertaken using similar methodology. Data on causes of early-onset neonatal sepsis other than GBS and on sepsis presenting from days 3 to 6 of life will be reported elsewhere. 70
The major hospital laboratory databases for each of the four main centres (accounting for around 55% of the national birth cohort) were also searched retrospectively for additional cases with GBS-positive cultures during the surveillance period. Data were entered into a Microsoft excel database. Incidences were calculated with national population data from Statistics New Zealand showing there were 127 134 live births during the 24-month study period.13 The 95% confidence intervals (CI) were estimated assuming a Poisson distribution and performed using Stata 13.1 (StataCorp, College Station, TX, USA). The New Zealand Multiregional Ethics Committee approved the study as an audit/observational study not requiring individual patient consent.
Results Over 24 months of surveillance, the NZPSU response rate averaged 93%, including notification that no cases were seen. There were 91 notifications of early-onset neonatal sepsis, two questionnaires were not returned (return rate 98%), three notifications did not meet the case definition, and there were three duplicate reports, resulting in 83 valid cases, 62 (75%) of whom developed symptoms in the first 48 h of life. There were 29 reported cases of early-onset GBS disease presenting in the first 48 h, 28 presented in the first 24 h, and 20 within the first two h of life. The incidence of early-onset GBS sepsis in the first 48 h was 0.23 (95% CI 0.16–0.33) per 1000 live births, while the incidence by ethnicity is displayed in Table 2. Maternal and infant characteristics are shown in Table 3 where nearly three-quarters (72%) of infants were born at term and 13 mothers (45%) were age 24 years or younger. (Two further cases of GBS sepsis presented after the first 48 h, at 62 and 96 h of age, the latter with confirmed meningitis, and both survived.) Lumbar puncture was attempted in 22 (76%) cases and successful in 21 (72%). There was one case with Table 2 Early-onset group B streptococcus sepsis (29 cases): – incidence by ascribed maternal ethnicity
Maternal ethnicity
Cases
Births
Incidence per 1000 births
Maori Pacific Islander Non-Maori/non-PI
5 7 17*
36 595 20 600 69 939
0.14 0.34 0.24
95% Confidence intervals 0.06, 0.33 0.17, 0.70 0.15, 0.39
When the two additional cases identified by survey of main laboratory records (see text) are included, Maori cases total 7 and the incidence is 0.19 (95% CI 0.09, 0.40) per 1000 births. *Includes one case where ethnicity unknown. Births for NonMaori/non-Pacific Islander calculated from total births minus births where the primary ascribed maternal ethnicity was either Maori or Pacific Islander.
© 2015 The Royal Australian and New Zealand College of Obstetricians and Gynaecologists
Early-onset neonatal GBS sepsis in New Zealand
Table 3 Maternal and infant characteristics for 29 cases of earlyonset group B streptococcus sepsis Maternal age (years)
DOI: 10.1111/ajo.12378
Original Article
Early-onset neonatal group B streptococcus sepsis following national risk-based prevention guidelines Brian A. DARLOW,1 Lesley VOSS,2 Diana R. LENNON3,4 and Keith GRIMWOOD5,* 1
Department of Paediatrics, University of Otago Christchurch, Christchurch, 2Department of Paediatric Infectious Diseases, Starship Children’s Hospital, Auckland, 3University of Auckland, 4Kidz First and Starship Children’s Hospitals, Auckland, New Zealand and 5Queensland Children’s Medical Research Institute, Children’s Health Queensland, South Brisbane, Queensland, Australia
Background: Neonatal infection with group B streptococcus (GBS) is an important cause of infant mortality. Intrapartum antibiotics reduce early-onset GBS sepsis, but recommendations vary as to whether they should be offered following antenatal screening or based on risk factors alone. We aimed to determine the incidence of early-onset GBS sepsis in New Zealand five years after the publication of national risk-based GBS prevention guidelines. Materials and Methods: Prospective surveillance of early-onset GBS sepsis (defined as infection in the first 48 h of life) was undertaken between April 2009 and March 2011 through the auspices of the New Zealand Paediatric Surveillance Unit as part of a survey of infection presenting in the first week of life. Results: There were 29 cases of confirmed early-onset GBS sepsis, including one case of meningitis, giving an incidence rate of 0.23 per 1000 (95% CI 0.16–0.33) live births. Three infants (10.3%) died. In 16 cases (55%), a maternal risk factor qualifying the mother for intrapartum antibiotics was present, but only five (31%) received this intervention. A retrospective review of the major hospital laboratory databases for this period identified two additional cases. A secondary sensitivity analysis taking account of these cases provided an estimated national incidence of 0.26 (95% CI 0.18–0.37) per 1000 live births. Conclusions: Ten years after a similar survey and five years after promoting a single, risk-based prevention protocol nationally, the incidence of early-onset GBS disease in New Zealand has more than halved, but opportunities remain to further reduce the rate. Key words: newborn, prevention and control, sepsis, Streptococcus agalactiae, surveillance and monitoring.
Introduction Early-onset neonatal sepsis from Streptococcus agalactiae (group B streptococcus: GBS) is an important cause of neonatal mortality and morbidity in high-income countries.1 Intrapartum antibiotic prophylaxis (IAP) interrupts mother-to-baby transmission of GBS, reducing the risk of early-onset neonatal infection.1,2 However, different countries and jurisdictions have adopted varying recommendations as to whether IAP should be offered
Correspondence: Dr Brian A. Darlow, Cure Kids Professor of Paediatric Research, Department of Pediatrics, University of Otago Christchurch, PO Box 4345, Christchurch 8140, New Zealand. Email: [email protected] *Present address: Menzies Health Institute Queensland, Griffith University and Gold Coast University Hospital, Gold Coast Campus, Queensland, Australia. Received 31 March 2015; accepted 2 June 2015.
according to antenatal screening results for GBS carriage or by using clinical risk factors.3–5 During 1998–1999, we undertook a two-year prospective national survey of earlyonset neonatal GBS disease (defined as infection in the first 48 h of life) in New Zealand and reported the attack rate as 0.5 per 1000 live births.6 A nationally representative, multidisciplinary group met in 2004 and recommended a risk-based prevention strategy as a single national policy to reduce early-onset GBS disease in New Zealand (Table 1).7 The aim of the present study was to ascertain the incidence of early-onset neonatal GBS through prospective surveillance over a two-year period and five years after introducing the national prevention guidelines.
Materials and Methods This prospective survey was undertaken under the auspices of the New Zealand Paediatric Surveillance Unit (NZPSU).8,9 All registered paediatricians and other
© 2015 The Royal Australian and New Zealand College of Obstetricians and Gynaecologists The Australian and New Zealand Journal of Obstetrics and Gynaecology
69
B. A. Darlow et al.
Table 1 Maternal risk factors for early-onset neonatal group B streptococcus infection7 Only one of the following required for IAP* Previous group B streptococcus (GBS)-infected baby GBS bacteriuria (of any count) this pregnancy Preterm (38°C Membrane rupture >18 h *IAP, intrapartum antibiotic prophylaxis administered during labour as either intravenous penicillin or amoxycillin.
clinicians working predominantly with children are sent a card (by email or post) and asked to indicate whether or not they had seen any of the listed conditions in the previous month. Early-onset neonatal sepsis was listed as a condition under surveillance from April 2009 to March 2011. Early-onset neonatal sepsis was defined as infection in the first six days of life, where the infant had a clinical picture consistent with sepsis and a positive bacterial culture from blood, cerebrospinal fluid (CSF), or pleural fluid, together with supporting laboratory evidence of sepsis (leukopenia, leukocytosis, thrombocytopenia, or elevated serum C-reactive protein concentration). Meningitis was defined as a clinical picture consistent with meningitis, and either a positive bacterial culture in the CSF, or a positive blood culture and CSF pleocytosis (≥100 white blood cells 9 106/L).10,11 Upon notifying a case, clinicians received a two-page questionnaire. To ensure confidentiality and avoid multiple notifications, cases were identified by initials and date of birth. Information sought included maternal age; parent-ascribed ethnicity; parity; history of previous infants with GBS sepsis; risk factors for early-onset infection in this pregnancy; whether the obstetric caregiver followed an early-onset GBS prevention protocol; whether antibiotics were received; their timing, doses, and routes; the infant’s gestation; birthweight; place of birth and hospital providing primary care; age of symptom onset; and the timing of blood and CSF specimens. Laboratory test confirmation was sought, including bacterial pathogen identity. Outcome data for the infant included whether the infant survived or died and, if the latter, whether sepsis was a contributing factor. In New Zealand, pregnant women are required to choose a Lead Maternity Carer (LMC) to provide their care and who may be a midwife (either community or hospital based), obstetrician, or general practitioner.12 Most women currently choose a midwife as their LMC. In this report, we have confined data to the first 48 h of life to compare with our earlier 1998–1999 survey, undertaken using similar methodology. Data on causes of early-onset neonatal sepsis other than GBS and on sepsis presenting from days 3 to 6 of life will be reported elsewhere. 70
The major hospital laboratory databases for each of the four main centres (accounting for around 55% of the national birth cohort) were also searched retrospectively for additional cases with GBS-positive cultures during the surveillance period. Data were entered into a Microsoft excel database. Incidences were calculated with national population data from Statistics New Zealand showing there were 127 134 live births during the 24-month study period.13 The 95% confidence intervals (CI) were estimated assuming a Poisson distribution and performed using Stata 13.1 (StataCorp, College Station, TX, USA). The New Zealand Multiregional Ethics Committee approved the study as an audit/observational study not requiring individual patient consent.
Results Over 24 months of surveillance, the NZPSU response rate averaged 93%, including notification that no cases were seen. There were 91 notifications of early-onset neonatal sepsis, two questionnaires were not returned (return rate 98%), three notifications did not meet the case definition, and there were three duplicate reports, resulting in 83 valid cases, 62 (75%) of whom developed symptoms in the first 48 h of life. There were 29 reported cases of early-onset GBS disease presenting in the first 48 h, 28 presented in the first 24 h, and 20 within the first two h of life. The incidence of early-onset GBS sepsis in the first 48 h was 0.23 (95% CI 0.16–0.33) per 1000 live births, while the incidence by ethnicity is displayed in Table 2. Maternal and infant characteristics are shown in Table 3 where nearly three-quarters (72%) of infants were born at term and 13 mothers (45%) were age 24 years or younger. (Two further cases of GBS sepsis presented after the first 48 h, at 62 and 96 h of age, the latter with confirmed meningitis, and both survived.) Lumbar puncture was attempted in 22 (76%) cases and successful in 21 (72%). There was one case with Table 2 Early-onset group B streptococcus sepsis (29 cases): – incidence by ascribed maternal ethnicity
Maternal ethnicity
Cases
Births
Incidence per 1000 births
Maori Pacific Islander Non-Maori/non-PI
5 7 17*
36 595 20 600 69 939
0.14 0.34 0.24
95% Confidence intervals 0.06, 0.33 0.17, 0.70 0.15, 0.39
When the two additional cases identified by survey of main laboratory records (see text) are included, Maori cases total 7 and the incidence is 0.19 (95% CI 0.09, 0.40) per 1000 births. *Includes one case where ethnicity unknown. Births for NonMaori/non-Pacific Islander calculated from total births minus births where the primary ascribed maternal ethnicity was either Maori or Pacific Islander.
© 2015 The Royal Australian and New Zealand College of Obstetricians and Gynaecologists
Early-onset neonatal GBS sepsis in New Zealand
Table 3 Maternal and infant characteristics for 29 cases of earlyonset group B streptococcus sepsis Maternal age (years)
