Introduction
Neonatal sepsis is a devastating condition associated with significant morbidity and mortality most commonly caused by GBS. The incidence for early onset GBS (EOGBS) invasive disease varies from 0.35/1000 live births in the U.S to 0.5/1000 live births in the U.K and Ireland.1,215-40% of pregnant women are colonised with GBS.3 Case fatality rate is approximately 2-3% for EOGBS disease in term infants.1 The risk of sepsis in an asymptomatic infant with risk factors for sepsis is low but clinically significant (0.5-1.0%).4 However, most infants (approx. 90%) who develop EOGBS invasive disease will be symptomatic within 24 hours after birth. Only 65% of these infants will have had at least one perinatal septic risk factor.5 In 2012 a survey of the different guidelines used in the UK showed little consensus on this issue with 125 different guidelines identified for 157 neonatal units.6 The RCOG guidelines recommend that infants with one GBS septic risk factor and no clinical signs should be carefully observed for the first 24 hours.5 In 2010 the Centre for Disease Control and prevention (CDC) changed their guidelines to improve the management of asymptomatic infants at risk for invasive EOGBS disease and thereby decrease unnecessary evaluation, antibiotic exposure and separation of mother and infant.1
Many units use broad-spectrum intravenous (IV) antibiotics such as Penicillin to cover GBS and Gentamicin to cover E-coli. The toxic effects of Gentamicin are well known including ototoxicity and nephrotoxicity. A single dose of a Gentamicin can cause permanent hearing loss in genetically predisposed individuals.12 Broad-spectrum antibiotics can also increase the risk of resistant bacteria at a later stage. Early infant antibiotic exposures have also been linked to increased risk of wheezing in children,13 and can affect the developing intestinal microbiome, increasing the risk of gastrointestinal disease,14 as well as intussusception and autoimmune diseases. However in the case of suspected neonatal invasive disease, these antibiotics are the most effective. In 2011 a retrospective audit on septic evaluations was performed in our neonatal unit. Following this, a new protocol, Figure 1, was introduced for the management of term infants with septic risk factors. This emphasised the importance of clinical examination and highlighted that for suspected maternal chorioamnionitis or in a symptomatic infant, a septic evaluation and antibiotics are always indicated. However, in infants with one septic risk factor who are asymptomatic, the new protocol recommended that four hourly clinical observations be performed by the nursing staff and that four and twelve hourly medical reviews be performed for 48 hours by the senior house officer (SHO).
The primary outcome of this study was the number of septic evaluations performed and antibiotic exposure in asymptomatic term infants with one septic risk factor following introduction of the new protocol. We hypothesised that we had performed less unnecessary blood tests and given fewer unnecessary antibiotics. Our secondary outcomes included identifying any symptomatic infants on the postnatal ward with a septic risk factor from observations and medical review, comparing these with the number of symptomatic infants identified on the postnatal ward without a septic risk factor, and analysing the economic cost implications of our new protocol.
Methods
Two independent reviewers (SG, AMD) performed a retrospective chart review of the babies born at term (= 37+0 weeks of gestation) from 1st July to 30th September 2012 in our tertiary maternity centre (Coombe Women and Infants University Hospital). We excluded infants with significant congenital defects such as congenital heart disease, spina bifida, and surgical gastrointestinal conditions which would increase their risk of sepsis. Details of birth gestation, mode of delivery, presence or absence of septic risk factors, maternal antibiotics, Apgar scores, clinical symptoms, blood tests performed, blood results and duration of any antibiotics were recorded. The clinical symptoms observed and identified as symptoms of suspected sepsis are listed in Table 1. Neither written informed consent nor research ethic committee approval was sought as this was an audit of the changed clinical practice in our centre. The data was analyzed using a PC-based statistics package (StatsDirect version 3.0.97). The data was analysed using a PC-based statistics package (StatsDirect version 3.0.97). Descriptive statistics and Fisher exact test were used for the outcomes as appropriate.
Results
The results of the primary audit showed that 444 (19%) out of 2375 term infants born in the Coombe Hospital in the three month audit period in 2011 had septic evaluations which included an FBC, a serum CRP and a blood culture performed and received IV antibiotics. This was similar to International data (approx. 15%). Of these 444 septic evaluations, 242 (54.5%) were asymptomatic infants with one septic risk factor such as prolonged rupture of membranes (PROM) or maternal GBS positive high vaginal swab (HVS). The other 202 (45.5%) were performed on symptomatic infants with suspected invasive disease or those with multiple septic risk factors. Of the total septic evaluations performed, 2.5% (11) had positive blood cultures, 1.2% (5) of which were GBS positive and only one of whom had a septic risk factor. All of these GBS positive infants were clinically symptomatic. They concluded that the 4 (80%) of GBS positive infants had no known septic risk factors and that the 0% (0 of 242) asymptomatic infants who had septic evaluations and antibiotics for one septic risk factor had a positive blood culture or raised CRP. The annual incidence of EOGBS in our hospital ranged from 0.45/1000 live births to 0.68/1000 live births between 2009 and 2011. Following the introduction of the new protocol, Figure 1, a re-audit was performed. 1855 charts of infants who were eligible for inclusion were reviewed from the time period.
We found that 344 infants had septic risk factors and of these infants, 83 (4.5%) had septic evaluations and received IV antibiotics immediately after delivery. See Figure 2. Of the initial septic evaluations, two had a raised CRP while none had a positive blood culture. Of these infants, five were clinically unwell and received antibiotics for >48 hours despite their negative blood cultures. 261 infants with one septic risk factor were allocated as per the new protocol to four hourly observations for 48 hours if the mother had received adequate intrapartum antibiotics (IAP) (i.e >2 doses or 1 dose > 4 hours before delivery) or four hourly observations and a four and twelve hourly medical review for 48 hours if the mother had received inadequate IAP. 241 infants remained asymptomatic. 20 infants were identified as symptomatic on the postnatal ward. These infants all had septic evaluations performed on identification and received IV antibiotics.
None of these infants were blood culture positive however 6 had raised CRPs and received >48 hours IV antibiotics. See Table 2. 1511 infants in this time period had no septic risk factors. 18 infants were identified as having clinical symptoms of sepsis either at birth or during routine observations/ checks on the postnatal ward. None had a blood culture positive sepsis, however one had a blood culture positive suspected as contamination. Two other infants with raised CRPs had >48 hours IV antibiotics. We calculated that at least eleven thousand five hundred euro was saved by the unit in this three month audit period due to less on call blood tests being performed and analysed, which equates to annual savings of nearly fifty thousand euro.
Discussion
As previously mentioned, the RCOG and the CDC have recently changed their guidelines encouraging a more clinical and less invasive approach to infants with one septic risk factor. We confirmed our hypothesis that we are performing less septic evaluations and giving fewer courses of antibiotics. We reduced the numbers of septic evaluations performed and antibiotics given from 19% to 6.5% (p<0.001), of the total number of term babies born in those three months. In 2011, 242 (10%) of all term infants had septic evaluations and received antibiotics immediately after birth for one septic risk factor despite being asymptomatic, while in 2012, 241(13%) asymptomatic infants with one septic risk factor were safely managed conservatively. We also showed that four hourly clinical observations and four and twelve hourly medical reviews of infants with one septic risk factor successfully identified symptomatic infants. There was no difference between the numbers of infants with and without septic risk factors that were identified as symptomatic on the postnatal ward. No infants became significantly unwell and there were no known morbidities and no mortalities. We cannot calculate a sensitivity or specificity of our screening tool as we do not have any blood culture confirmed sepsis during this period. It could be argued that we could not definitively state that no unwell infants had been missed, however as all infants get a postnatal check before discharge, a symptomatic infant would have been identified. In the US and in many European countries, all mothers are screened for GBS antenatally, with a positivity rate of 15-40%.3
The policy is that these mothers should be covered with IAP, usually in the form of a Penicillin. However, it has been found that even with adequate IAP, EOGBS infection in term infants in these countries is still a significant burden.1,8,9 In Ireland and in the UK, screening is not performed routinely. IAP are recommended in certain situations in our hospital such as intrapartum pyrexia >38°C, previous baby with GBS disease or delivery at < 37 weeks with PROM >18 hours. However, the evidence for this is inconsistent.9 Some centres perform 18 hour CRP measurement as an indication of sepsis in infants. This has a 64% sensitivity and 56% specificity with a negative predictive value (NPV)of 93% and a positive predictive value (PPV) of 14%.10 However, many diagnostic tests for neonatal sepsis have a poor PPV. The use of procalcitonin can also be used in the evaluation of sepsis.11
We did not evaluate late onset GBS (LOGBS) invasive disease, which has an incidence of 0.24 per 1000 live births for term infants.15 Most mothers in these cases are GBS carriers and IAP have been associated with delayed presentation of symptoms and milder late onset sepsis. However, neither septic evaluations nor early antibiotics have been shown to prevent infants from developing LOGBS invasive disease. Our study shows that we were able to perform less septic evaluations and give fewer antibiotics to asymptomatic term infants with 1 septic risk factor without compromising the safety of infants who became symptomatic or unwell. We calculated that nearly fifty thousand euro per year were saved with this new protocol. We interviewed the SHOs who performed the frequent medical reviews. Subjectively they advised that the collective time performing the medical reviews was less than the collective time and stress it would take to perform septic evaluations on these infants. We suggest that this new protocol, Figure 1, of observational and medical review could be introduced in other hospitals who are performing septic evaluations and giving antibiotics unnecessarily to well term infants with one septic risk factor. We emphasise the importance of clinical exam in all cases whether or not septic risk factors are known.
Correspondence: S Glackin
Our Lady’s Children’s Hospital, Crumlin, Dublin 12
References
1. Verani JR, McGee L, Schrag SJ. Prevention of perinatal group B streptococcal disease. Morbidity and Mortality Weekly Report (MMWR), Revised Guidelines from CDC, Recommendations and Reports. 2010;59:1-32
2. NHS U. The UK NSC policy on Group B Streptococcus screening in pregnancy. nhs.uk: NHS; 2012
3. Stoll BJ, Hansen NI, Sánchez PJ, Faix RG, Poindexter BB, Van Meurs KP, Bizarro M, Goldberg R, Ivan F, Hale E. Early onset neonatal sepsis: the burden of group B Streptococcal and E. coli disease continues. Pediatrics. 2011;127:817-26
4. Polin RA, Papile L-A, Baley JE, Bhutani VK, Carlo WA, Cummings J, Kumar P, Tan R, Wang K, Watterberg K. Management of neonates with suspected or proven early-onset bacterial sepsis. Pediatrics. 2012;129:1006-15.
5. RCOG. The Prevention of Early-onset Neonatal Group B Streptococcal Disease. 2012.
6. Behjati S, Prentice P, Rennie J. Management of Group B streptococcal sepsis risk in well, term newborns. Acta Paediatrica. 2012;101:128-31
7. Escobar GJ, Li D-k, Armstrong MA, Gardner MN, Folck BF, Verdi JE, Xiong B, Bergen R. Neonatal sepsis workups in infants= 2000 grams at birth: a population-based study. Pediatrics. 2000;106:256-263
8. Schrag SJ, Verani JR. Intrapartum antibiotic prophylaxis for the prevention of perinatal group B streptococcal disease: Experience in the United States and implications for a potential group B streptococcal vaccine. Vaccine. 2013;31:D20-D6.
9. Ohlsson A, Shah VS. Intrapartum antibiotics for known maternal Group B streptococcal colonization. Cochrane Database Syst Rev. 2013;1.
10. Lacaze-Masmonteil T, Rosychuk RJ, Robinson JL. Value of a single C-reactive protein measurement at 18 h of age. Archives of Disease in Childhood-Fetal and Neonatal Edition. 2014;99:F76-F9.
11. Stocker M, Hop WC, van Rossum AM. Neonatal Procalcitonin Intervention Study (NeoPInS): Effect of Procalcitonin-guided decision making on Duration of antibiotic Therapy in suspected neonatal early-onset Sepsis: A multi-centre randomized superiority and non-inferiority Intervention Study. BMC pediatrics. 2010;10:89.
12. Bitner-Glindzicz M, Pembrey M, Duncan A, Heron J, Ring SM, Hall A, Rahman S. Prevalence of mitochondrial 1555A? G mutation in European children. New England Journal of Medicine. 2009;360:640-2.
13. Alm B, Erdes L, Möllborg P, Pettersson R, Norvenius SG, Åberg N, Wennergren, G. Neonatal antibiotic treatment is a risk factor for early wheezing. Pediatrics. 2008;121:697-702.
14. Torrazza RM, Neu J. The developing intestinal microbiome and its relationship to health and disease in the neonate. Journal of Perinatology. 2011;31:S29-S34.
15. Berardi A, Rossi C, Lugli L, Creti R, Reggiani MLB, Lanari M, Memo L, Pedna MF, Venturelli C, Perrone E. Group B streptococcus late-onset disease: 2003–2010. Pediatrics. 2013;131:e361-e8.