Abstract
Objectives
(1) To describe the epidemiology of neonatal group B streptococcal (GBS) disease over five years (1997–2001) in the Netherlands, stratified for proven and probable sepsis and for very early (<12 h), late early (12 h – <7 days) and late (7–90 days) onset sepsis. (2) To evaluate the effect of the introduction in January 1999 of guidelines for prevention of early onset GBS disease based on risk factors.
Methods
Data on cases were collected in collaboration with the Dutch Paediatric Surveillance Unit and corrected for under‐reporting by the capture‐recapture technique.
Results
Total incidence of proven very early onset, late early onset and late onset GBS sepsis was 0.32, 0.11 and 0.14 per 1000 live births, respectively, and of probable very early onset, late early onset and late onset GBS sepsis was 1.10, 0.18 and 0.02 per 1000 live births, respectively. Maternal risk factors were absent in 46% of the proven early onset cases. Considerably more infants with proven GBS sepsis were boys. 64% of the infants with proven very early onset GBS sepsis were first borns compared with 47% in the general population. After the introduction of guidelines the incidence of proven early onset sepsis decreased considerably from 0.54 per 1000 live births in 1997–8 to 0.36 per 1000 live births in 1999–2001. However, there was no decrease in the incidence of meningitis and the case fatality rate in the first week of life. The incidence of late onset sepsis also remained unchanged.
Conclusion
After the introduction prevention guidelines based on risk factors there has been a limited decrease in the incidence of proven early onset GBS sepsis in the Netherlands. This study therefore recommends changing the Dutch GBS prevention guidelines.
Keywords: neonate, Group B streptococcal disease, epidemiology, prevention, guidelines
Early onset group B streptococcal (GBS) infection can be prevented by intrapartum antibiotic prophylaxis.1 In Europe, prevention strategies for this disease were developed later than in the USA.2 In 1996, the Centers for Disease Control and Prevention (Atlanta, Georgia, USA) recommended guidelines based on one of two strategies.3 The first strategy consisted of universal screening for GBS colonisation. It was calculated that this strategy would result in intrapartum chemoprophylaxis for 26.7% of all pregnant women and prevent nearly 90% of neonatal cases. The second strategy was based on the presence of risk factors, and with this strategy 18% of all pregnant women would receive intrapartum chemoprophylaxis and 69% of neonatal GBS disease would be prevented.4 In January 1999, the Dutch Paediatric Association and the Society of Obstetrics and Gynaecology introduced guidelines based on the presence of risk factors. A risk factor based strategy was chosen as the estimated total incidence of proven and probable GBS sepsis in the Netherlands is relatively low (0.9 per 1000 live births) compared with the high incidence of proven GBS sepsis only in the USA before the introduction of the guidelines (1.1–3.7 per 1000 live births).5
The Dutch guidelines recommend antibiotic prophylaxis in the case of a previous infant with GBS disease, in heavily colonised mothers and in cases of intrapartum fever (>38.0°C). To limit the numbers of women receiving antibiotic prophylaxis, the guidelines recommend that in case of labour before 37 weeks or prolonged rupture of membranes (⩾18–24 h), the woman should be first screened for GBS carriership, and antibiotic prophylaxis given if the culture is positive. The prophylactic antibiotics are given intravenously, with 5 million IU of penicillin G or 2 g amoxicillin given ideally 4 h before birth, followed by 2.5 million IU penicillin or 1 g amoxicillin, respectively, every 4 h until delivery. In case of delivery before the culture result is available, the obstetrician can give antibiotic prophylaxis based on their discretion (fig 1).6
Figure 1 Prevention strategy of early onset group B streptococcal (GBS) disease in the Netherlands. *In case of delivery before the results of culture are ready, the obstetrician should decide about antibiotic prophylaxis based on the severity of the risk factor(s).
The aim of the present study was to evaluate the incidence of GBS disease in the first three months of life during 1997–2001 in the Netherlands and to compare the incidence in the periods before and after the introduction of prevention guidelines.
Methods
The ethics committee of the University Medical Centre Nijmegen approved the study.
We collaborated with the Dutch Paediatric Surveillance Unit (DPSU) for data collection. This unit coordinates an “active” case reporting structure to facilitate the investigation of uncommon paediatric diseases. Every month, the DPSU sends a card containing a variable menu of 10 conditions to all practising paediatricians in the Netherlands. During 1997–2001, neonatal GBS disease was included in the DPSU surveillance. Respondents were requested to report both proven and probable cases of GBS disease during the first three months of life, diagnosed in the preceding month. The DPSU forwarded the names of the reporting paediatricians to one of the authors (MTS). She then sent the paediatricians a questionnaire to collect clinical, microbiological and biochemical data on both mothers and infants.
As there was a possibility that the data reported to the DPSU might have been incomplete, the results were corrected for under‐reporting. For this, we used the capture‐recapture technique to estimate the number of cases missed by the DPSU surveillance. The technique uses numbers of cases reported to two independent sources. The corrected numbers of proven and probable cases can be calculated from the numbers reported by each source and the overlap between them.7 The second independent source that we used was the response to a call for patients, published in a national parents' magazine, in which parents were asked to report to one of us (GdJ) if they had a child with GBS disease during the first three months of life in the years 1997–2001. We then checked whether the individual cases reported by parents had been reported to the DPSU by paediatricians. If not, relevant information was requested from the paediatrician involved. Finally, each case was classified as proven or as probable GBS sepsis or it was discarded, similar to the cases reported to the DPSU. We also estimated the level of completeness of reporting cases of GBS meningitis to the DPSU by the capture‐recapture technique. Data from the microbiology laboratory at the Academic Medical Centre in Amsterdam, the national reference laboratory for bacterial meningitis, were used as the second independent source for data collection on cases of GBS meningitis.
To evaluate the relevance of the time between birth and the onset of the first symptoms of neonatal GBS infection we stratified the patients into three groups:
very early onset (<12 h);
late early onset (12 h – <7 days);
late onset (7–90 days).
We analysed the data on neonates with proven and with probable GBS sepsis separately. The diagnosis of proven GBS sepsis was based on a positive GBS culture of blood and/or cerebrospinal fluid (CSF) along with presence of clinical features of septicaemia and abnormal results of laboratory investigations (leucocytosis or leucopenia, granulocytopenia, shift to the left, thrombocytopenia, elevated C‐reactive protein). The diagnosis of probable GBS sepsis was based on the presence of the same clinical and laboratory features as for proven GBS sepsis and isolation of GBS from various sites, but without positive blood or CSF culture.8
Four obstetric risk factors of early onset GBS disease were documented: delivery before 37 weeks, ruptured membranes for ⩾18 h, intrapartum fever (>38.0°C) and previous delivery of a child with GBS disease.9,10 The risk factor heavy maternal GBS colonisation (which may present as GBS urinary tract infection or GBS bacteriuria during the current pregnancy) was excluded from the analysis because in the Netherlands urine cultures are not routinely done.
We also collected data on sex, gestational age, parity, intrapartum antibiotic treatment, multiple pregnancies, mode of delivery, meconium staining of amniotic fluid, place of delivery, exact age when first symptoms occurred, meningitis and fatal outcome. The effect of introduction of the guidelines on the incidence of GBS disease was expressed in terms of risk ratio (RR) and 95% CI. p Values <0.05 were considered to be significant.
Results
Incidence of GBS disease
Over the whole study period (1997–2001) the corrected incidence of proven very early onset, late early onset and late onset GBS sepsis was 0.32, 0.11 and 0.14 per 1000 live births, respectively. The corrected incidence of probable very early onset, late early onset and late onset GBS sepsis was 1.10, 0.18 and 0.02 per 1000 live births, respectively.
Over the two‐year period before the introduction of guidelines (1997–8), 642 cases of GBS disease were reported to the DPSU. Of these, 66 reports were excluded from the study because they were double reports. Another 198 reports were excluded because they were not about GBS and did not meet the inclusion criteria. Among the latter, 124 reports were of GBS colonisation without clinical features and two were of positive GBS blood culture with no clinical features of septicaemia (bacteriaemia). The remaining 378 reports were included in the study, of which 184 had proven GBS sepsis and 194 had probable GBS sepsis (table 1). In the three‐year period after the introduction of the guidelines (1999–2001), 948 cases of GBS disease were reported to the DPSU. Of these, 72 were excluded from the study because they were double reports and 312 because they were incorrect and did not meet the inclusion criteria. Among the latter, 164 cases had GBS colonisation without clinical features. The remaining 564 cases were included in the study, of which 246 had proven GBS sepsis and 318 had probable GBS sepsis (table 1).
Table 1 Corrected incidence of proven and probable neonatal GBS sepsis before (1997–8) and after (1999–2001) the introduction of guidelines for prevention of early onset GBS disease in the Netherlands.
| DPSU reported cases | Corrected incidence* per 1000 live births (95% CI) | Risk ratio (min–max)† | |||
|---|---|---|---|---|---|
| 1997/98 | 1999/01 | 1997/98 | 1999/01 | ||
| Live births | 391 851 | 608 665 | |||
| Proven GBS sepsis | |||||
| Early onset | 157 | 196 | 0.54 (0.42 to 0.67) | 0.36 (0.32 to 0.41) | 0.67 (0.48 to 0.98) |
| Very early | 128 | 140 | 0.38 (0.30 to 0.46) | 0.28 (0.22 to 0.34) | 0.74 (0.48 to 1.13) |
| Late early | 29 | 56 | 0.13 (0.07 to 0.19) | 0.09 ‡ | 0.69 (0.47 to 1.30) |
| Late onset | 27 | 50 | 0.14 (0.07 to 0.27) | 0.14 (0.06 to 0.23) | 1.0 (0.22 to 32.9) |
| Probable GBS sepsis | |||||
| Early onset | 190 | 313 | 1.3 (0.61 to 2.1) | 1.4 (0.9 to 1.9) | 1.08 (0.43 to 3.11) |
| Very early | 159 | 257 | 0.99 (0.44 to 1.53) | 1.1 (0.65 to 1.52) | 1.11 (0.42 to 3.45) |
| Late early | 31 | 56 | 0.4 (0.0 to 1.1) | 0.3 (0.06 to 0.53) | 0.75 (0.05 to 53.0) |
| Late onset | 4 | 5 | 0.01 | 0.02 (0.0 to 0.04) | 2.0 (0.0 to 4.0) |
DPSU, Dutch Paediatric Surveillance Unit; GBS, group B streptococci.
*Corrected for under‐reporting to the DPSU by using the capture‐recapture technique; this may lead to discrepancies in the data.
†Minimum–maximum calculated with lowest and highest estimates of incidence.
‡No under‐reporting for this category.
In 1997–8, parents reported 27 cases of proven and 22 cases of probable GBS sepsis. Of these, 19 with proven and 8 with probable GBS sepsis were also reported to the DPSU. Therefore, the DPSU captured only 70% of proven and 36% of probable GBS sepsis. In 1999–2001, parents reported 33 cases of proven and 51 cases of probable GBS sepsis. Of these, 27 with proven and 19 with probable GBS sepsis were also reported to the DPSU. Therefore, the DPSU captured only 82% of proven and 37% of probable GBS sepsis.
After correction for under‐reporting, the decrease in the incidence of very early onset GBS sepsis from 0.38 per 1000 live births in 1997–8 to 0.28 per 1000 live births in 1999–2001 was not statistically significant (p>0.05; table 1). Total proven early onset GBS sepsis, on corrected for under‐reporting, decreased significantly from 0.54 per 1000 live births to 0.36 per 1000 live births (p<0.05; table 1). There was no decrease in the corrected incidence of total probable early onset GBS sepsis (1.3 per 1000 live births and 1.4 per 1000 live births, respectively; table 1). In the small category of late onset GBS sepsis, there was no decrease in the corrected incidence of proven cases.
Incidence of GBS meningitis
In 1997–8 and 1999–2001, 29 and 57 patients, respectively, with proven GBS meningitis were reported to the reference microbiology laboratory. As 16 and 36 of these were also reported to the DPSU, the DPSU captured only 55% of GBS meningitis cases in 1997–8 and 63% in 1999–2001. Therefore, the corrected incidence of proven GBS meningitis in these two periods was 0.14 (95% CI 0.11 to 0.17) per 1000 live births and 0.17 (95% CI 0.15 to 0.19) per 1000 live births, respectively, and in the total five‐year period it was 0.16 (95% CI 0.14 to 0.17) per 1000 live births. The proportion of patients with GBS meningitis was clearly related to the time of onset of GBS disease and increased from 6% if the onset was <12 h to 53% if the onset was ⩾7 days after birth (table 2).
Table 2 Incidence of reported proven GBS meningitis in the Netherlands: 1997–2001.
| Reported to DPSU 1997–2001 | No. | Lumbar puncture performed n (%) | Positive lumbar puncture n (%) | Proved meningitis (%) |
|---|---|---|---|---|
| Early onset | 353 | 132 (37) | 55 (42) | 16 |
| Very early | 268 | 79 (29) | 17 (22) | 6 |
| Late early | 85 | 53 (62) | 38 (72) | 45 |
| Late onset | 77 | 47 (61)* | 41 (87) | 53 |
DPSU, Dutch Paediatric Surveillance Unit, GBS, group B streptococci.
*Cochran‐Armitage test for trend with time of onset, p<0.0001.
Case fatality rate
In the whole five‐year period, the case fatality rate (CFR) of proven early and late onset GBS sepsis was 8% and 5%, respectively. The CFR of proven early onset GBS sepsis was 7% versus 9% in the two periods and the mortality was similar (0.028 per 1000 live births). The CFR of proven late onset GBS sepsis was 7% in the first period and 4% in the second period (table 3). The CFR of probable early and late onset GBS sepsis was 2% and 0%, respectively.
Table 3 Case fatality rates of proved cases of group B streptococcal (GBS) sepsis (<3 months) in the past three decades.
| GBS sepsis | ||
|---|---|---|
| Early onset | Late onset | |
| Remington and Klein 11* | ||
| 1982 | 55 | 23 |
| 1995 | 10–15 | 2–6 |
| 2001 | 5–10 | 2–6 |
| The Netherlands (all hospitals)† | ||
| 1997–1998 | 11/157 (7) | 2/27 (7) |
| 1999–2001 | 17/196 (9) | 2/50 (4) |
| 1997–2001 | 28/353 (8) | 4/77 (5) |
| The Netherlands (NICUs)† | ||
| 1977–198112 | 13/21 (62) | 4/5 (80) |
| 1985–199313 | 18/78 (23) | 2/4 (50) |
| 1994–199614 | 21/66 (32) | 3/7 (43) |
| 1997–2001 (this study‡) | 14/111 (13) | 2/19 (10) |
*Values are %.
†Values are n/N (%).
‡Case fatality rate in the same eight neonatal intensive care units (NICUs) in the Netherlands as in 1994–6.
Timing of onset of GBS disease
In 1997–8, in 70% (n = 128/184) of all patients with proven GBS sepsis reported to the DPSU and in 1999–2001, in 57% (n = 140) of all patients with proven GBS sepsis reported to the DPSU (n = 246), first symptoms were observed within 12 h after birth. In 85% (157/184) and 80% (196/246) of the cases, respectively, the first symptoms occurred within 7 days after birth.
Risk factors and characteristics of infants and mothers in proven and probable GBS sepsis
In patients with proven very early onset (<12 h) GBS disease, the presence of the risk factor preterm delivery was significantly lower in 1999–2001 (40/140; 29%) compared with 1997–8 (54/128; 42%) (p<0.05). The prevalence of the risk factors prolonged rupture of membranes, intrapartum fever and a previous child with GBS disease did not change during this time. Table 4 shows the patient characteristics over the period 1997–2001 because there were no differences between 1997–8 and 1999–2001.
Table 4 Prevalence of risk factors among and characteristics of mothers and infants with proved and probable group B streptococcal (GBS) sepsis (1997–2001).
| Very early onset GBS sepsis | Late early onset GBS sepsis | Late onset GBS sepsis | |||||
|---|---|---|---|---|---|---|---|
| Proven n (%) | Probable n (%) | Proven n (%) | Probable n (%) | Proven n (%) | Probable n (%) | Dutch population* | |
| Number | 268 | 415 | 85 | 88 | 77 | 9 | |
| Preterm delivery (<37 weeks) | 94 (35) | 124 (30) | 17 (20) | 10 (11) | 30 (39) | 3 (38) | 7.6 |
| Prolonged rupture of membranes (⩾18 h) | 91 (34) | 19 (5) | 14 (16) | 4 (4.6) | 9 (12) | 0 (0) | 10.8 |
| Intrapartum fever (>38°C)† | 52 (19) | 18 (4) | 0 (0) | 3 (3) | 2 (3) | 1 (11) | 5.7 |
| Previous child with GBS disease | 2 (1) | 9 (2) | 1 (0) | 1 (1) | 2 (3) | 0 (0) | <0.17‡ |
| No risk factor | 109 (41) | 131 (32) | 55 (65) | 41 (47) | 37 (48) | 4 (44) | >80 |
| Male sex | 158 (59) | 247 (60) | 41 (48) | 47 (53) | 45 (58) | 6 (67) | 51.1% |
| First born | 171 (64) | 243 (59) | 39 (46) | 52 (59) | 28 (36) | 2 (22) | 47.0% |
| Twin | 19 (7) | 21 (51) | 2 (2) | 7 (8) | 10 (13) | 0 (0) | 3.5% |
| Caesarean section | 55 (20) | 104 (25) | 12 (14) | 20 (22.7) | 13 (17) | 1 (11) | 10.5% |
| Meconium‐stained amniotic fluid | 42 (16) | 90 (22) | 8 (9) | 16 (18.2) | 2 (3) | 0 (0) | 7.6% |
| Home delivery | 20 (7) | 24 (6) | 21 (25) | 6 (7) | 15 (19) | 2 (22) | ± 30% |
| Hospital delivery | 228 (85) | 361 (87) | 56 (66) | 74 (84) | 56 (73) | 7 (78) | ± 70% |
*Best estimates from several sources (eg Statistics Netherlands, Prismant, Institute for Health Care Management).
†The risk factor intrapartum fever was often not reported in the medical record (60%).
‡The probability of having a previous child with early onset GBS is calculated by multiplying the rate of attack of 3 cases per 1000 live births by 55.2%, the percentage of multiparous infants.
Risk factors
The presence of the risk factor preterm delivery was two to five times higher in case of proven and probable GBS sepsis than in the Dutch population. In proven very early onset GBS sepsis only, the presence of prolonged rupture of membranes and intrapartum fever was about three times higher (table 4). In 46.7% of all infants with proven early onset GBS sepsis and in 34.1% of all infants with probable early onset GBS sepsis none of the four risk factors was present.
Sex
Of the 430 infants with proven GBS sepsis, 57% were boys and of the 512 infants with probable GBS sepsis, 300 (59%) were boys. In the general population, this is 51.1% (p<0.05). There was no sex differences between the GBS early and late meningitis groups nor between the early and late fatal cases.
Birth order
Of the 268 infants with proven very early onset GBS sepsis 171 (64%) were first borns and of the 415 infants with probable very early onset GBS sepsis 243 (59%) were first borns, compared with 47% in the general population (p<0.0001).
Other characteristics
None of the mothers of infants with proven late early onset sepsis received antibiotics during delivery. The proven very early GBS sepsis group had two times more cases of twins, caesarean section and meconium stained amniotic fluid than expected in the Dutch population. Most of the infants with proven and probable very early onset sepsis (85% and 87%, respectively) were born in hospital (table 4).
Discussion
What is already known on this topic
After implementation of guidelines in the USA, the incidence of early onset GBS disease markedly dropped from 1.7 per 1000 live births in 1993 to 0.6 per 1000 live births in 1998.
In 1999, only 4 of 29 European countries (Denmark, Norway, Spain and the Netherlands) had nationwide guidelines for prevention of neonatal GBS infection.
We found that the incidence (corrected for 77% response) of proven neonatal GBS sepsis during 1997–2001 was 0.56 per 1000 live births in the Netherlands. For probable GBS sepsis the incidence (corrected for 37% response) was 1.38 per 1000 live births. We collected data on infants with probable disease because most of them perhaps had GBS sepsis. Early onset disease is usually defined as the occurrence of first symptoms in the first seven days after birth. We considered it appropriate to subcategorise early onset GBS sepsis in “very early onset” (<12 hours) and “late early onset” (12 h – <7 days) because: (1) with this approach, we found that in 76% (268/353) of proven early onset cases, first symptoms occurred within 12 h; (2) the risk factor prolonged rupture of membranes occurred more frequently in the very early group; (3) the risk factor intrapartum fever was only present in the very early group (52 v 0 infants); and (4) mothers of infants with proven very early onset GBS sepsis were more often primigravid (64% v 46%). The difference in birth order may partly be due to the longer duration of the first delivery.15 Furthermore, more infants with proven GBS sepsis were boys. Two other Dutch studies also described this sex difference.12,14 In our study this male preponderance was restricted to the very early onset and late onset groups. Following the introduction of the guidelines, after correction for under‐reporting, there was a significant decrease in the total incidence of proven early onset GBS sepsis, from 0.54 per 1000 live births (1997–8) to 0.36 per 1000 live births (1999–2001) (p<0.05; table 1). This can be partly explained by the increase in use of intrapartum antibiotic prophylaxis in the Netherlands from 1.0% of all deliveries in 1997 to 5.9% in 2000 after the introduction of the Dutch guidelines.16 The incidence of late onset GBS sepsis remained unchanged, probably because of nosocomial acquisition of the infection.17,18
What this study adds
This study provides nationwide epidemiological data on neonatal GBS disease.
It also presents an evaluation of the effect of nationwide guidelines for prevention of neonatal GBS infection in the Netherlands.
The overall effect of the Dutch guidelines is disappointing because of the limited decrease in the incidence of proven early onset GBS sepsis, and no decrease in mortality, incidence of meningitis and probable sepsis. In the USA, the incidence of proven early onset GBS sepsis was higher, varying between 1.1 per 1000 live births and 3.7 per 1000 live births before the introduction of the prevention guidelines. In one study the incidence dropped from 1.7 per 1000 live births in 1993 to 0.6 per 1000 live births in 199819 after the introduction of prevention guidelines in 1996 based on screening or on the presence of risk factors.3 In 2002 the US guidelines recommended a strategy based only on screening20 and the incidence of early onset GBS disease dropped further to 0.3 per 1000 live births in 2004.21 We carried out a study in 1999 and found that the incidence of neonatal GBS disease in Europe varied from 0.5 per 1000 live births to 2 per 1000 live births.2 A recent national study in the UK (2000–1) found the incidence of early onset GBS disease was 0.48 per 1000 live births but antibiotic prophylaxis was rarely used at the time.22,23 The Dutch guidelines are based on recognition of risk factors. To limit use of antibiotic prophylaxis during delivery, in case of labour before 37 weeks and/or prolonged rupture of membranes women are first screened and then given antibiotic prophylaxis if a positive culture is obtained. In case of delivery before the culture result is available, the obstetrician decides whether prophylaxis should be given.5 Therefore, the Dutch guidelines can be expected to have limited effectiveness because in the setting of preterm labour and/or prolonged rupture of membranes cases can be missed as a result of delay in culture results. Furthermore, in about 40–50% of cases of early onset GBS sepsis no maternal risk factors are present24 and it has been shown in the USA that the screening strategy is more effective than the risk factor based strategy.25
As meningitis occurs more often in late onset GBS disease and chemoprophylaxis is effective in preventing early onset GBS disease, our study found that the incidence of meningitis did not decrease over the study period. Moreover, there is under‐reporting of meningitis because in unstable (preterm) newborns, lumbar puncture is often omitted.
The CFR of GBS disease has decreased dramatically in the past 25 years26 and in proven neonatal GBS sepsis it was less than 10% in the present study. In a more selected (neonatal intensive care unit) population, the CFR of proven early onset GBS sepsis has decreased in the past 25 years from about 62% to about 13% (table 3). The decline in CFR of GBS sepsis may be attributed to improvements in neonatal care, antibiotic prophylaxis and change of the virulence of group B streptococci.27 The CFR and the mortality of proven early onset GBS sepsis did not decrease further after the introduction of the guidelines, partly because the administration of antibiotics is primarily for prevention and not for treatment of the infant with early onset GBS disease. Of the 77 infants with proven late onset GBS disease, 31 were born before 37 weeks. This confirms that preterm delivery is the most important perinatal risk factor in late onset disease.28
Most of the patients with very early onset GBS sepsis and very early onset probable GBS sepsis were born in hospital (85% and 87%, respectively) in contrast with about 70% of all deliveries in the Netherlands. This may be explained by the fact that in the Netherlands, pregnant women with risk factors are usually referred to hospital. Fifty‐six infants with proven GBS sepsis were born at home, and only in 9 (16%) of them was one of the four risk factors present.
Thus, due to the limited effectiveness of the Dutch risk factor based guidelines for prevention of neonatal GBS disease, we recommend that these guidelines should be changed.
Acknowledgements
We thank all the paediatricians who contributed to the data collection, and the Dutch Paediatric Surveillance Unit and the Reference Laboratory for Bacterial Meningitis at the Academic Medical Centre Amsterdam for invaluable help in collecting data.
Abbreviations
CFR - case fatality rate
DPSU - Dutch Paediatric Surveillance Unit
GBS - group B streptococci
Footnotes
This study was funded by Zorg Onderzoek Nederland (project no. 2200004) and by Stichting Group B Streptococcen, the Netherlands.
Competing interests: None.
References
- 1.Boyer K M, Gotoff S P. Prevention of early‐onset neonatal group B streptococcal disease with selective intrapartum chemoprophylaxis. N Engl J Med 19863141665–1669. [DOI] [PubMed] [Google Scholar]
- 2.Trijbels‐Smeulders M A J M, Kollée L A A, Adriaanse A H.et al Neonatal group B streptococcal (GBS) infection: incidence and strategies for prevention in Europe. Pediatr Infect Dis 200423172–173. [DOI] [PubMed] [Google Scholar]
- 3.Centers for Disease Control and Prevention Prevention of perinatal group B streptococcal disease: a public health perspective. MMWR Recomm Rep 1996451–24. [PubMed] [Google Scholar]
- 4.Rouse D J, Goldenberg R L, Cliver S P. Strategies for the prevention of early‐onset neonatal group B streptococcal sepsis: a decision analysis. Obstet Gynecol 199483483–494. [DOI] [PubMed] [Google Scholar]
- 5.Schuchat A, Wenger J D. Epidemiology of group B streptococcal disease. Risk factors, prevention strategies and vaccine development. Epidemiol Rev 199416374–402. [DOI] [PubMed] [Google Scholar]
- 6.Trijbels‐Smeulders M A J M, Adriaanse A H, Gerards L J.et al Prevention strategy for neonatal early‐onset group‐B‐streptococcal (GBS) disease in the Netherlands. Rev Med Microb 20031435–39. [Google Scholar]
- 7.Wittes J T, Colton T, Sidel V W. Capture‐recapture methods for assessing the completeness of case ascertainment when using multiple information sources. J Chronic Dis 19742725–36. [DOI] [PubMed] [Google Scholar]
- 8.Gerdes J S. Clinicopathologic approach to the diagnosis of neonatal sepsis. Clin Perinatol 199118361–381. [PubMed] [Google Scholar]
- 9.Bramer S, Wijk van F H, Mol B W.et al Risk factors for neonatal early‐onset GBS‐related disease. A case‐control study. J Perinat Med 199725469–475. [DOI] [PubMed] [Google Scholar]
- 10.Schuchat A, Deaver‐Robinson K, Plikaytis B D.et al Multistate case‐control study of maternal risk factors for neonatal group B streptococcal disease. The Active Surveillance Study Group. Pediatr Infect Dis J 199413623–629. [DOI] [PubMed] [Google Scholar]
- 11.Remington J S, Klein J O.Infectious diseases of the fetus and newborn infant, 5th edn. Philadelphia: WB Saunders Company 2001
- 12.Gerards L J. Group B streptococci in the perinatal period. Thesis, Utrecht University, the Netherlands 1985
- 13.Adriaanse A H, Lagendijk I, Muytjens H L.et al Neonatal early onset group B streptococcal infection; a nine‐year retrospective study in a tertiary care hospital. J Perinat Med 199624531–538. [DOI] [PubMed] [Google Scholar]
- 14.Sprij A J, de Jonge G A. Group B streptococcal infection in eight tertiary care hospitals 1994−96, a retrospective study. Tijdschr Kindergeneeskd 199967224–228. [Google Scholar]
- 15.The Netherlands Perinatal Registry Obstetrics in the Netherlands, trends 1995−1999. The Netherlands Perinatal Registry 2005
- 16.Trijbels‐Smeulders M. Group B streptococcal disease: effect of the Dutch guidelines for prevention. Thesis, Nijmegen University, the Netherlands 2006
- 17.Anthony B F, Okada D M, Hobel L J. Epidemiology of the group B streptococci: maternal and nosocomial sources for infant acquisition. J Pediatr 197995431–436. [DOI] [PubMed] [Google Scholar]
- 18.Boyer K M, Vogel L C, Gotoff S P.et al Nosocomial transmission of bacteriophage type 7/11/12 group B streptococci in a special care nursery. Am J Dis Child 1980134964–966. [DOI] [PubMed] [Google Scholar]
- 19.Schrag S J, Zywicki S, Farley M M.et al Group B streptococcal disease in the era of intrapartum antibiotic prophylaxis. N Engl J Med 200034215–20. [DOI] [PubMed] [Google Scholar]
- 20.Centers for Disease Control and Prevention Prevention of perinatal group B streptococcal disease: revised guidelines from CDC. MMWR Recomm Rep 2002511–24. [PubMed] [Google Scholar]
- 21.Gibbs R S, Schrag S, Schuchat A. Perinatal infections due to group B streptococci. Obstet Gynecol 20041041062–1076. [DOI] [PubMed] [Google Scholar]
- 22.Heath P T, Balfour G, Weisner A M.et al Group B streptococcal disease in UK and Irish infants younger than 90 days. Lancet 2004363292–294. [DOI] [PubMed] [Google Scholar]
- 23.Kenyon S, Brocklehurst P, Blackburn A.et al Antenatal screening and intrapartum management of group B streptococcus in the UK. BJOG 2004111226–230. [DOI] [PubMed] [Google Scholar]
- 24.Lopez Sastre J B, Fernandez Colomer B, Coto Cotallo G D.et al Trends in the epidemiology of neonatal sepsis of vertical transmission in the era of group B streptococcal prevention. Acta Paediatr 200594451–457. [DOI] [PubMed] [Google Scholar]
- 25.Schrag S J, Zell E R, Lynfield R.et al A population‐based comparison of strategies to prevent early‐onset group B streptococcal disease in neonates. N Engl J Med 2002347233–239. [DOI] [PubMed] [Google Scholar]
- 26.Yagupsky P, Menegus M A, Powell K R. The changing spectrum of group B streptococcal disease in infants: an eleven‐year experience in a tertiary care hospital. Pediatr Infect Dis J 199110801–808. [DOI] [PubMed] [Google Scholar]
- 27.Fleming K E, Bohnsack J F, Palacios G C.et al Equivalence of high‐virulence clonotypes of serotype III group B Streptococcus agalactiae (GBS). J Med Microbiol 200453505–508. [DOI] [PubMed] [Google Scholar]
- 28.Lin F C, Weisman L E, Troendle J.et al Prematurity is the major risk factor for late onset Group B Streptococcus disease. J Infect Dis 200318826. [DOI] [PubMed] [Google Scholar]