Journal Club 2015

Scabies control with ivermectin

Romani L et al. N.Engl. J. Med. 2015; 373: 2305-13.[BA1]

Background: Infestation with the mite Sarcoptes scabiei causes an extremely itchy rash, scabies, which can be complicated by impetigo due to Staphylococcus aureus or Streptococcus pyogenes. In tropical environments including Northern Australia and the Pacific, impetigo can cause pyogenic complications, but serious non-pyogenic complications of Streptococcus pyogenes infection include rheumatic fever and glomerulonephritis. Consequently, scabies control might reduce rheumatic heart disease.

Main findings: An innovative study randomised three island communities in Fiji to one of three interventions for scabies: standard care (topical permethrin to affected persons and contacts, n=803); permethrin (mass administration of permethrin, n=532); or ivermectin (mass administration of single-dose oral ivermectin, n=716). Ivermectin was taken under direct observation. Before treatment the prevalence and severity of scabies was similar. The prevalence of scabies declined in all 3 groups over the 12 month intervention period: standard care group by 49% (95%CI 37-60%), permethrin group by 62% (95%CI 49-75%) and ivermectin group by 94% (95%CI 83-100%). The prevalence of impetigo also decreased significantly in all groups: relative reduction 32% (95%CI 14-50%), 54% (95%CI 35-73%) and 67% (95%CI 52-83%) respectively. The decrease in impetigo in the ivermectin group was significantly greater than in the other two groups (p<0.05). Adverse events, mostly itch and headache, were more common in the ivermectin than the permethrin group (15.6% vs 6.8%) but were mild.

Take home message: Mass administration of ivermectin is safe and reduces both scabies and impetigo.

Submitted by: David Isaacs and Philip Britton, Children’s Hospital at Westmead, Sydney, 11 December 2015

Valganciclovir for Symptomatic Congenital Cytomegalovirus Disease

D.W.Kimberlin et al; N Engl J Med 2015; 372(10): 933-43. PMID: 25738669

Background: Congenital cytomegalovirus (CMV) infection is the leading non-genetic cause of sensorineural hearing loss, and the most common viral cause of mental retardation. Therapy with intravenous ganciclovir or oral valganciclovir for 6 weeks is now an accepted treatment option for patients with symptomatic congenital CMV disease involving the central nervous system (CNS), but there was suggestion that this benefit would wane over the first 2 years of life.

Main findings: A total of 96 neonates with symptomatic CMV disease (with or without CNS involvement) were blinded and randomized after receiving 6 weeks of valganciclovir to receive valganciclovir for a further 18 weeks versus placebo (to complete a total of 6 months of therapy). The primary end point was change in hearing in the better ear (“best-ear” hearing), from baseline to the 6-month follow-up. There was no statistically significant difference in best-ear hearing between infants who received 6 months of valganciclovir versus those who received placebo. However, infants who received 6 months of valganciclovir demonstrated improvement in several clinically significant secondary outcomes. In the assessment of total-ear hearing, participants who received 6 months of valganciclovir were 3 times more likely than those who received 6 weeks of therapy to have improved hearing or to have maintained normal hearing between baseline and 12 months, after adjustment for CNS involvement at baseline (odds ratio 3.04, 95% CI 1.26-7.35, p=0.01). Similar benefit was demonstrated at 24 months. The timing of initiation of valganciclovir within the first month of life did not correlate with audiologic outcomes at 12 months or 24 months. Of clinical significance is also the demonstration of higher Bayley-III language-composite and receptive-communication scale scores at 24 months in infants who received 6 months valganciclovir (p=0.005 and p=0.003 respectively). Importantly, the incidences of severe neutropenia and transaminitis were similar between participants in both arms from week 6 to month 7 in the study.

Implications for practice: Although the study did not demonstrate improvement in primary outcome (“best-ear” hearing) in infants who received 6 months or oral valganciclovir, it has provided the basis for 6 months of treatment by demonstrating moderately favorable effects on clinically significant long-term audiologic and neurodevelopmental outcomes. Without increased risks of severe neutropenia and transaminitis in infants who received 6 months of treatment, this study forms the basis for further prospective trials to evaluate the benefits of extended treatment in infants with congenital CMV disease.

Submitted by: Dr Jiahui Li and Dr Thoon Koh Cheng, Kandang Kerbau Women and Children’s Hospital, Singapore, 14 October 2015.

Clindamycin versus Trimethoprim-Sulfamethoxazole for Uncomplicated Skin Infections

Miller LG et al. NEJM. 2015 Mar 19;372(12):1093-1103. PMID25785967.

Background: The relative efficacy of clindamycin versus trimethoprim-sulfamethoxazole (TMP-SMZ) for treatment of skin and skin structure infections (SSSI’s) in methicillin-resistant Staphylococcus aureus (MRSA) endemic regions is not known.

Summary/Main Findings: In this multicentre, superiority DBRCT conducted in USA (2009-2011), 524 outpatients (29.6% children) with uncomplicated SSSI including abscesses >5cm (smaller for children), cellulitis or both were randomized 1:1 to receive clindamycin or TMP-SMZ for 10 days. Patients with coexisting medical conditions or immunosuppression were excluded. All abscesses were drained (44.5%). The primary outcome was clinical cure assessed 7-10 days after completion of therapy. There was no difference in cure rates between clindamycin (80.3%) and TMP-SMZ (77.7%), difference -2.6% (95% CI -10.2 to 4.9) in the intention-to-treat analysis. Per-protocol and subgroup analysis of children, adults, or patients with cellulitis, abscesses or mixed lesions were also non-significant. Cure rates were also similar at 1-month. Cultures were obtained in 296 patients (56.5%), with S. aureus most commonly isolated (41.4%), of which 31.9% were MRSA; clindamycin resistance 12.4% and TMP-SMZ resistance 0.5%. There was a trend towards lower cure rates for infections caused by clindamycin-resistant versus clindamycin-susceptible S. aureus isolates (73.3% vs. 91.7%, p=0.06), raising the possibility of spontaneous resolution in some instances. Adverse events were mild-moderate and similar between groups.

Take Home Messages: Clindamycin and TMP-SMZ have comparable efficacy and safety profiles for the treatment of uncomplicated SSSI’s (cellulitis and abscesses) in patients without coexisting medical conditions. This study builds upon existing evidence regarding utility of TMP-SMZ against S. pyogenes, and is novel in providing clinical data to support its use for cellulitis.

Submitted by: Charlie McLeod & Asha Bowen, Princess Margaret Hospital, 20 Sept 2015

Combination of Vancomycin and β-lactam therapy for methicillin-resistant Staphylococcus aureus bacteremia: A pilot, multicenter, randomized controlled trial

Davis JS et al. Clin Infect Dis. 2015 Sept 8. [Epub ahead of print]. PMID26349552.

Background: In-vitro studies demonstrate synergistic activity of vancomycin and anti-staphylococcal β-lactams for methicillin resistant Staphylococcus aureus (MRSA) bacteraemia. Prospective clinical trials are needed.

Main findings: In this pilot, open-label RCT conducted across 7 Australian hospitals (2011-2014), 60 adults with MRSA bacteraemia were randomized 1:1 within 48 hours of positive blood culture to vancomycin 1.5g IV BD (standard group) or vancomycin plus 7 days of flucloxacillin 2g IV Q6H (combination group). The primary outcome was duration of MRSA bacteraemia. There was a trend towards reduced duration of bacteraemia in the combination group (3.0 days vs. 1.94 days with a rate ratio of 0.65; 95% CI 0.41-1.02, p=0.06). Secondary endpoints showed no significant difference in ICU admissions, mortality, hepatotoxicity, nephrotoxicity or occurrence of metastatic foci.

Take home messages: These preliminary results are encouraging, but larger trials, including in children are required to investigate the efficacy of combination therapy for MRSA bacteraemia. Recommendations for change to current clinical practice cannot be made on the basis of this data. The results of CAMERA-2 are awaited.

Submitted by: Charlie McLeod & Asha Bowen, Princess Margaret Hospital, 21 September 2015.

The microbiome of uncontacted amerindians

Clemente JC, Pehrsson EC, Blaser MJ et al. Sci Adv. 2015 Apr 3;1(3). pii: e1500183

Background: The Yanomami people are an isolated tribe in the remote Venezuelan Amazon. They are unique: there has been no documented contact with non-Yanomami in over 11,000 years; and they have not been exposed to commercial antibiotics. Hence, the Yanomami represent a unique proxy for pre-antibiotic society. The aim of this extraordinary study was to characterise the faecal, oral and skin bacterial microbiomes and resistome within this extremely isolated population. Samples were taken from the oral cavity (n = 28), forearm skin (n = 28) and faeces (n=12) from 31 of the 65 villagers at the time of the first medical expedition to the village. The age of the subjects was estimated by Yanomami health workers as being between 4-50 years. The V4 region of the 16S rRNA gene was then sequenced and compared to data from previous studies from other non-isolated populations.

Summary of Main Findings: The Yanomami microbiome demonstrated the highest skin and faecal bacterial diversity ever reported in any human group. The microbiome diversity however was more evenly shared among the Yanomami across all body sites than in other populations. The isolation and lack of transculturation is theorised to account for the high bacterial diversity and prevalence. The microbiome of the Yanomami people demonstrated colonisation with bacteria that carry functional antibiotic resistance (AR) genes despite no known exposure to commercial antibiotics with most of the identified AR genes components of a core E. coli genome. These genes confer resistance to synthetic antibiotics and have been maintained despite no apparent antibiotic selection pressure. There was no significant phenotypic resistance yet noted in the Yanomami population.

Implications for practice/opinion/conclusions: This was a captivating study which allowed a unique opportunity to peer into a geographically isolated, pre-antimicrobial human population utilising modern day sequencing technology. This population demonstrated unprecedented bacterial diversity and carried functional AR genes despite being an antibiotic naive population. The functional AR genes appear to be a feature of the human microbiome and are seemingly poised and ready-to-go following antimicrobial exposure. This finding helps explain the at-times rapid antibiotic resistance observed with the introduction of new generational antimicrobial agents observed in other populations. This study again reinforces the importance of antimicrobial stewardship in reducing the trajectory of antibiotic resistance in a population. Antimicrobial agents have since been introduced in to the Yanomami village. Follow-up community studies would therefore be useful to monitor the selective effect on this unique microbiome and track the development of phenotypic resistance - both of which would contribute to our general understanding of the evolution of antimicrobial resistance.

Submitted by: Matt O’Brien & Josh Francis, Royal Darwin Hospital, 15 September 2015.

Safety of pertussis vaccination in pregnant women in UK: observational study

Donegan K et al. BMJ. 2014;349:g4219. PMID25015137

Background: Antenatal pertussis vaccination programs have now been implemented in various forms in all Australian states and territories, as jurisdictionally funded interventions to reduce early infant morbidity and mortality. Safety data previously in pregnancy had been limited to a small number of patients. This observational cohort study compared women vaccinated in pregnancy with matched historical unvaccinated controls to determine safety.

Main Findings: This study identified pregnant vaccinated women and controls through the Clinical Practice Research Datalink (CPRD) network of 650 general practices. Adverse events were identified from clinical diagnoses with additional child health data obtained by mother-child linkage. There was no evidence of an increased risk of stillbirth within 14 days of vaccination (incidence rate ratio 0.69, 95%CI 0.23 to 1.62) or later in pregnancy (IRR 0.85, 0.44 to 1.61) compared with historical national rates. Compared with a matched historical cohort of unvaccinated pregnant women, there was no evidence that vaccination accelerated the time to delivery (hazard ratio 1.00, 0.97 to 1.02). Furthermore, there was no evidence of an increased risk of stillbirth, maternal or neonatal death, pre-eclampsia or eclampsia, haemorrhage, fetal distress, uterine rupture, placenta or vasa praevia, caesarean delivery, low birth weight, or neonatal renal failure, all serious events that can occur naturally in pregnancy.

Take Home Messages: This is the first large scale safety study of routine pertussis vaccination in pregnancy, utilising historical controls and primary care records linked to child health records. Despite the methodological limitations, it provides substantial reassurance that known complications of late pregnancy do not appear increased following vaccination in the 3rd trimester. It does not inform unknown unknowns, unexpected adverse events that may emerge during large program implementation and be blamed upon vaccination, but does allow patients and health care workers confidence that major pregnancy complications for women and babies are not increased following vaccination.

Submitted by: Jim Buttery, Monash Children’s Hospital,12 August 2015

Lower Newborn Bone Mineral Content Associated With Maternal Use of Tenofovir Disoproxil Fumarate During Pregnancy

Siberry GK et al. Clinical Infectious Diseases. June 2015, epub ahead of print

Background: Tenofovir disoproxil fumarate (TDF) has been increasingly prescribed as a preferred antiretroviral agent for pregnant women. Animal studies suggest impaired fetal growth and decreased bone mineralisation after maternal TDF in late pregnancy. The effects in humans have not been well studied.

Summary: Study participants were HIV-exposed, uninfected singleton infants born at ≥ 36 weeks gestation from 14 US centres. TDF-exposed (maternal TDF for ≥8 weeks third trimester) were compared with TDF-unexposed (no maternal TDF in pregnancy). DXA scans to assess bone mineral content (BMC), bone mineral density (BMD) and total mass were performed within 4-5 weeks of birth. The primary objective was to compare mean whole body BMC of TDF exposed and unexposed infants. Of 143 enrolled infants with an evaluable DXA scan, 74 were TDF-exposed and 69 TDF-unexposed. Mothers and infants were well-matched for potential confounders, except that mothers of TDF-exposed infants were more likely to have also used boosted protease inhibitors (PI). Despite this, linear regression showed boosted PI exposure had minimal impact on the TDF-effect estimate. The main finding was that the mean (SD) whole-body BMC of infants exposed to TDF was 56.0 (11.8)g compared with 63.8 (16.6)g for unexposed infants (P=.002), a difference of 12.2% between exposure groups.

Conclusions: Infants with in utero TDF exposure had a significantly lower BMC than infants without in utero TDF exposure even after adjustment for multiple potential confounders. This finding is consistent with animal studies. The magnitude of the mean BMC difference (12.2%) is larger than the effect sizes reported for adults who take TDF or certain nutritional interventions in infants. Lower BMC at birth is a well-established risk factor for bone fracture in infants, children and even into adulthood. Longitudinal evaluation of these findings is needed. Fortunately, this is possible through this ongoing study group. This study is also being replicated at multiple sites in Africa with results expected in 2016. The role of vitamin D was not investigated but other reports suggest that vitamin D supplementation in the mother or infant to counteract the bone effects of intrauterine TDF exposure may be considered.

Submitted by: Celia Cooper, Women’s and Children’s Hospital Adelaide, 19 July 2015

Bacteraemic urinary tract infection: management and outcomes in young infants

Schroeder AR, Shen MW, Biondi EA, et al. Arch Dis Child Published Online First: July 15th 2015

Background: Bacteraemia associated with UTI is now the most common source of bacteraemia in young infants. These infants receive lengthier courses of parenteral antibiotics and have increased length of hospital stay compared to those without bacteraemia.

Main findings: This was a retrospective multicenter study across 20 US hospitals, re-examining cases of UTI with bacteraemia in infants <3 months of age from databases prepared for other purposes, during different but overlapping time periods from 1998-2013. The sickest infants (i.e. initially admitted to ICU) and those with major co-morbidities were excluded. The median age of 251 included infants was 35 days (IQR 17-58). The most common organism was E. coli, in 89.6% of infants. Renal ultrasound was eventually obtained in 95.6% and VCUG in 83.7% of infants. Duration of parenteral antibiotics varied substantially, with a mean+/-SD duration of 7.4 +/- 4 days. The most popular choices of duration (in order) were 7, 10, 14, 3, and 5 days. Younger age, non-E. coli organism, and positive repeat blood culture were the strongest patient-level predictors of longer duration, however variability in duration was explained mostly by institution rather than recorded patient factors. Six infants (2.4%, 95% CI 0.8-5.1%) had relapsed UTI with the same organism. There was no difference in duration of parenteral antibiotics between infants with and without relapsed UTI. All infants with relapsed UTI had VCUG abnormalities (100% vs 32% for those without relapse who had VCUG (p=0.001). There were no cases of relapsed bacteraemic UTI. The authors highlight a number of important study limitations, most inherent in the heterogeneous retrospective design, and anticipate the difficulties likely to be encountered in coordinating a randomised trial in this area.

Take home messages: This is the largest paediatric study of antibiotic therapy for bacteraemic UTI. While hardly definitive, its findings are generally supportive of including uncomplicated bacteraemic UTI on the growing list of indications where early intravenous-to-oral switch may be safely applied with expected benefits in terms of patient safety and reduced costs to patients, families, and the public.

Submitted by: Josh Osowicki, BC Children’s Hospital, Vancouver, 15 July 2015

No Survival Benefit With Empirical Vancomycin Therapy for Coagulase-negative Staphylococcal Bloodstream Infections in Infants

Ericson JE et al The Pediatric Infectious Disease Journal • Volume 34, Number 4, April 2015 PMID25760564

Background: Coagulase-negative Staphylococcus (CoNS) is the most common organism associated with late-onset neonatal sepsis (LOS) in hospitalised infants. This retrospective study compared mortality in infants with CoNS Bloodstream infections (BSI) who received empiric versus delayed vancomycin therapy.

Main Findings: Between 1997 to 2012, 4364 Infants (< 120 days) with CoNS BSI were identified from 348 NICUs in the Pediatrix Database (USA). CoNS BSI definition required at least 2 positive blood cultures. Gestational age (27 weeks (IQR 25,29) in both groups) and VLBW (91% vs 89%) were similar between groups. Use of empirical vancomycin between units varied considerably (8-100%) with no significant secular trends over time. Overall, 65% of infants with CoNS BSI were treated with empiric vancomycin. There was no difference between unadjusted 30-day mortality between the empiric (6%) and delayed vancomycin (4%) treated groups (OR 1.14 (95% CI 0.84 – 1.56). The median duration of bacteraemia was 1 day longer for infants with delayed vancomycin therapy (4 days (IQR 2,6) vs 3 days (IQR 2,5). No details were provided about cause of mortality, attainment of therapeutic vancomycin levels, adverse drug reactions, presence of a CVC or further episodes of sepsis. The proportion of babies receiving vancomycin in each unit suggests that a substantial number of units do not have, or do not often follow, an empiric antibiotic policy.

Take Home Messages: Crude mortality was similar in (mostly VLBW) infants receiving empiric and delayed vancomycin therapy. Several small, mostly retrospective studies have shown no difference between delayed and empiric vancomycin policies. This study, whilst impressively large, omits important clinical details and provides no information on cause of mortality. In addition, as a retrospective study, it inevitably suffers from the bias of potential unmeasured factors that influenced whether babies received empiric versus delayed vancomycin. Therefore this ‘big data’ approach cannot definitively answer the question of the efficacy and safety of a delayed vancomycin policy for neonatal sepsis.

Submitted by: Jeremy Carr and Nigel Curtis, Royal Children’s Hospital Melbourne, 24 June 2015

Three-weekly doses of azithromycin for Indigenous infants hospitalized with bronchiolitis: a multicentre, randomized, placebo-controlled trial

McCallum et al. Frontiers in Pediatrics. April Volume 3, 2015 doi: 10.3389/fped.2015.00032

Background: Bronchiolitis is a major health burden in infants globally, notably amongst both indigenous populations of Australia and New Zealand. Increasingly severe sequelae such as bronchiectasis are thought to be in part due to recurrent early respiratory infections such as bronchiolitis.

Summary: This multicentre DBRCT involved Australian Aboriginal, Torres Strait Islander, Maori, and/or Pacific Island infants from Darwin, Townsville and Auckland investigating whether once weekly azithromycin for 3 weeks versus placebo shortened length of stay or oxygen supplementation during a bronchiolitis episode. Secondary measures evaluated symptoms at day 21 and re-hospitalisations within 6 months plus nasopharyngeal bacterial load. 219 infants were randomised to azithromycin (106) or placebo (113) and no difference was found in median LOS (54 hrs both groups, difference 0 hrs 95% CI -6 to +8%) oxygen duration, symptoms at day 21 or rehospitalisation in 6 months. Changes in nasopharyngeal flora and carriage density were seen in the group receiving antibiotics.

Conclusions: No evidence was found to support azithromycin in acute bronchiolitis in high risk but otherwise well indigenous infants nor that azithromycin changes recurrent or persistent respiratory symptoms in the first 6 months. The same research group have previously shown effect of prolonged weekly azithromycin courses in a more select group of young children with established early chronic lung disease in lessening exacerbations (Valery et al. Lancet Respir Med 2013). Other antibiotics were used in 60% of the entire bronchiolitis cohort recruited, showing that treating viral respiratory infections with antibiotics is commonplace in these high risk infants; yet persistent symptoms and respiratory re-hospitalisation (25%) (both independent risk factors for subsequent development of chronic suppurative lung disease) remain high.

On a slightly related note another recent paper confirms what we know about use of azithromycin in the neonatal period and increased risk of pyloric stenosis. The risk is highest in the first 14 days (aOR 8.2, 95% CI 2.6 to 26) but persists to day 42 of life (aOR 2.9, 95% CI 1.2 to 7.2) and did appear to be less than erythromycin (13 fold increased risk in the first 14 days of life and 4 fold from days 15-42). Eberly et al Pediatrics 2015

Submitted by: Emma Best, Starship Hospital, Auckland (currently in lovely Darwin) on 5 May 2015

Efficacy of combined antiparasitic therapy with praziquantel and albendazole for neurocysticercosis: a double-blind randomised controlled trial

Garcia HH et al Lancet Infectious Diseases 2014;14:687-95. PMID24999157

Background: Neurocysticercosis caused by Taenia solium is widely considered the most common cause of acquired epilepsy worldwide. Treatment with either praziquantel or albendazole has suboptimal efficacy. No previous studies have assessed their combined efficacy for parenchymal cerebral disease, and seizure control remains problematic.

Summary: In this double-blind placebo-controlled phase-3 trial, 124 adult patients (16 – 65 years) were randomised to receive either combined (albendazole + praziquantel, n=41) or single antiparasitic therapy (standard albendazole, n=43 and high dose albendazole, n=40) for 10 days. All patients received steroids. The primary outcome was complete cyst resolution on 6-month MRI. In the combined treatment arm, 64% had complete resolution of cysts compared to 37% in standard albendazole, RR 1.75 95% CI 1.1 – 2.79. The study demonstrates that albendazole plus praziquantel is significantly more efficacious in clearing parasites than single therapy without increased side-effects, and more importantly, complete cyst resolution appears associated with fewer seizure relapses.

Conclusions: This study shows an advantage of combination albendazole and praziquantel over monotherapy for cerebral neurocysticercosis. Cyst destruction appeared associated with fewer seizure relapses and thus better control. Note that patients with intraventricular or subarachnoid neurocysticercosis were not assessed as treatment differs.

Take Home Message: Let's use combination therapy for cerebral neurocysticercosis.

Submitted by: Alberto Pinzon, Lady Cilento Hospital, 22 April 2015

Maternal titres after adequate syphilotherapy during pregnancy

Rac MW et al. Clin Inf Dis 2015 Mar 1 60(5). PMID25414264

Background: Effective therapy for syphilis diagnosed during pregnancy is essential to reduce the risk of infection in the infant. However, judging the effectiveness of therapy in pregnancy is challenging.

Summary: This study looked retrospectively at maternal non-treponemal titres in a cohort of women with syphilis diagnosed during pregnancy at a single institution over a 30-year period. All had been treated adequately according to CDC protocols. They included women who were diagnosed after 18 weeks gestation and had results available from the time of diagnosis and at birth. Only 63 of 166 women (38%) achieved a 4-fold decline in titres by delivery, however none had treatment failure (4-fold increase in titres). There was no difference in the proportion of infants treated for syphilis between those who did and those who did not achieve a 4-fold decline in titres at birth. Unfortunately the clinical findings from the infants who were treated for congenital syphilis are not well described in the paper. Women were less likely to have achieved a 4-fold decline if they had latent syphilis or syphilis of unknown duration, or were treated later in pregnancy.

Conclusions: In the context of documented adequate treatment for syphilis during pregnancy a failure to achieve a 4-fold decline in titres is more a reflection of the timing of treatment than of treatment failure. Its not realistic to expect women to have all had an adequate serological response to treatment of syphilis at delivery and failure to do so doesn’t necessarily mean a higher risk of congenital syphilis. P.S. I don’t think syphilotherapy is a word.

Submitted by: Tony Walls, Christchurch Hospital NZ, 12 March 2015

Factors Associated with Intrapartum Transmission of Group B Streptococcus

Berardi A et al. Factors Associated with Intrapartum Transmission of Group B Streptococcus. Pediatric Infectious Disease Journal 2014;33:1211–1215, PMID25037035

Background: Intrapartum antibiotic prophylaxis (IAP) for pregnant women is effective in reducing risk of early onset Group B Streptococcus infection (EOGBS) in neonates. Although current guidelines recommend administering antibiotics at least 4 hours prior to delivery, the minimum duration of IAP required to prevent EOGBS is uncertain. Understanding factors that influence neonatal colonization might help explain factors associated with prophylaxis failure.

Main findings: In this prospective cohort study from Italy, 502 term neonates from 499 mothers eligible for IAP based on prenatal screening were included and 458 neonates were given IAP. Risk of neonatal Group B Streptococcus colonization (in throat and rectal swabs at 24 - 48h of life) was associated with a lack of IAP exposure (P<0.01), intra- partum fever ≥37.5°C (P<0.01) and African ethnicity (P < 0.01). In the 458 IAP-exposed neonates, the neonatal colonization rate was low (3.7%) and did not vary significantly for neonates given IAP from <1h – 12h before delivery (score test for trend of odds, P = 0.13). There were no cases of GBS sepsis, although 5 IAP-exposed infants and 1 without IAP received postpartum antibiotics.

Take home message: Neonatal colonization was associated with heavy maternal colonization, lack of any IAP exposure, maternal fever and African ethnicity. IAP with ampicillin/ penicillin has rapid onset of action and appears to be effective in reducing NC with Group B Streptococcus, even if given <4h prior to delivery. Note: Breastfeeding among non-IAP and IAP infants, which may have been a factor associated with colonization, was not investigated.

Submitted by: Gilkrist Ewe, Brendan McMullan and Pamela Palasanthiran, Sydney Children’s Hospital NSW, 20 January 2015