Journal Club 2013

Vaccine refusal and pertussis

Atwell JE et al. Pediatrics 2013 ; 132 : 624-30. PMID24082000

Unimmunised children are protected against exposure to some infections by high levels of immunisation of their contacts (“herd immunity”). In most industrialised countries there are population pockets where immunisation coverage is considerably lower than elsewhere. Acellular vaccines are less effective than many other vaccines and recent pertussis outbreaks in industrialised countries have been attributed largely to waning vaccine-induced immunity. A study hypothesised that clustering of unimmunised children contributed to a California pertussis outbreak. 1 People opposed to immunisation are sometimes called conscientious objectors, a term usually reserved for those opposed to war. This paper prefers the term non-medical exemptions (NMEs). The authors compared NMEs for children entering kindergarten from 2005-10 with the spatial and temporal distribution of pertussis cases in 2010. They found spatial clustering of exemptions, and children in an environment where NMEs clustered were more than twice as likely to be in a pertussis cluster (Odds ratio = 2.47, 95% CI 2.22-2.75). These data suggest children living in areas of low immunisation coverage are at increased risk of pertussis and not protected by herd immunity. Submitted by: David Isaacs, Children’s Hospital Westmead, 3 December 2013.

Vancomycin dosing in children: what is the question?

Cole TS, Riordan A. Arch Dis Child 2013;98:994-997. PMID23956256

Background: There has been much discussion and new research over recent years regarding the optimal dosing of vancomycin in paediatric patients. In 2011 the IDSA recommended paediatric dosing of 60 mg/kg/day (15 mg/kg 6 hourly). This recommendation was in regard to treating invasive MRSA infections. The dosing regimen was extrapolated from adult data in order to achieve effective concentration-dependent killing with the accepted goal of AUC/ MIC > 400. It was acknowledged that evidence for this recommendation in children was based on limited evidence. Measuring the AUC is impractical but trough serum concentrations are commonly measured: unfortunately troughs do not directly reflect the AUC. In adults, troughs of 15 -20 mg/l are recommended for serious infections. In children, this is rarely achieved. Are these levels necessary for effective clinical cure? Pharmacokinetic modeling suggests that with even higher doses these levels may not be achievable. Yet there are added toxicity risks as the safety of doses higher than 60 mg/kg/day in children is not fully known.

Main Findings: This review paper comprehensively summarises the available information to support the dosing regimen of 15 mg/ kg 6 hourly, acknowledging that serum trough levels of 15 -20 mg/l may be impractical in the paediatric population. Pharmacokinetic and modeling studies suggest that high troughs may not be necessary if the MIC of the organisms is ≤ 1 mg/l. The community MRSA infections we typically deal with in children in New Zealand (and Australia) do not generally have MICs > 1 mg/l. The studies reviewed in this paper do not comment on clinical outcome or culture negativity which in clinical practice can still be achieved in the face of lower vancomycin trough levels. Frymoyer et al have shown using simulation models that AUC/MIC > 400 is achieved in most children with the dosing regimen 15 mg/ kg 6 hourly or 20 mg /kg 8 hourly.

Take home message: There is little comment as to the usefulness of vancomycin loading doses or continuous infusions in the paediatric population, acknowledging that further studies are required.

Submitted by: Joanne McCaughan and Emma Best, Starship Children’s Hospital, 15 November 2013.

Kotloff KL et al. Burden and aetiology of diarrhoeal disease in infants and young children in developing countries (the Global Enteric Multicenter Study, GEMS): a prospective, case-control study. Lancet. 2013 Jul 20;382(9888):209-22.

http://www.ncbi.nlm.nih.gov/pubmed/23680352

Diarrhoeal disease remains one of the leading causes of morbidity and mortality in children in resource poor settings, and in the Aboriginal population of Australia. This study reports the results of an extremely large case- control study to identify the aetiology and burden of moderate to severe diarrhoeal disease in sub-Saharan Africa and south Asia.

The study ran over 3 years in four locations in Africa and 3 in Asia at sites selected to give a range of populations; populations varied in their GDP, HIV rates, access to water and education standards. Cases (children aged 0-59 months with moderate to severe diarrhoea) were selected from those attending health centres. Controls were children without diarrhoea.

The study recruited a total of 9439 cases and 13,129 controls. Overall, the commonest pathogens were rotavirus, Cryptosporidium, enterotoxigenic Escherichia coli producing heat-stable toxin (ST-ETEC; with or without co-expression of heat-labile enterotoxin), and Shigella. Pathogens varied from site to site, with Aeromonas frequently reported in some Asian sites and Vibrio cholerae O1 and Campylobacter jejuni also being leading causes of diarrhoeal disease in some locations. Diarrhoea was associated with a much higher risk of death (odds ratio of dying during follow-up 8·5), mainly in those aged under 2 years. Similarly, children with moderate-severe diarrhoea were much more likely to have growth failure.

This study highlights the on-going burden of diarrhoeal disease in resource poor settings. To overcome this, efforts to provide good sanitation need to continue, but vaccines may offer more realistic methods of control in the interim. Efforts to develop effective vaccines and/or to enhance the efficacy of existing vaccines against the pathogens identified are needed.

A Survival Guide on How to Keep Up-to-date With Pediatric Infectious Diseases Literature Using Internet Technology

Loher-Niederer A et al, Pediatr Infect Dis J. 2013 Jul; 32(7):786-7. PMID 23838778.

Keeping up-to-date in pediatric infectious diseases remains challenging to all clinicians and researchers as time to read new medical literature is limited. This article reviews the different online tools on offer. Most frequently cited, highly ranked journals publish electronic table of contents (eTOCs). A free of charge online subscription is required, and the TOC of the new issue will automatically when published to a personal email account. 5 minutes to peruse these abstracts is all it takes. A really simple syndication (RSS) feed delivers information about new content from an information source such as a website. RSS feeder software can be installed with simple usage instructions. Respected journals including NEJM, BMJ, The Lancet and the Cochrane Collaboration now have a twitter and/or facebook presence. Apps from open access journals can also be downloaded to mobiles. Meta-journals collect and summarize research articles and guidelines from a number of medical journals. Concise summaries, with a short commentary, are accessible online or through an e-mail alert. Saved performed searches with automated alerts can be done on MEDLINE/ PubMed, web of Knowledge or Ovid. Subscription to these automated alerts needs to be exercised with caution, and regular reevaluation is important to prevent receiving an unmanageable number of alerts. Novel tools are evolving continuously, making it important to reevaluate our techniques and stay up to date.

Trough Concentrations of Vancomycin- Adult Therapeutic Targets Are Not Appropriate for Children.

Gordon C et al. The Pediatric Infectious Disease Journal 2012. 31(12) 1269-1271. PMID 22828642.

Recent guidelines have increased the target trough serum vancomycin concentration (now 12-18mg/L using 12hrly dosing- Antibiotic Therapeutic Guidelines- ATG). This reflects a change in the principle objective of vancomycin monitoring from the traditional objective of preventing toxicity to that of ensuring efficacy. The impetus for this change has come from concern surrounding the emergence of VISA and hVISA, as well as trial evidence suggesting that clinical efficacy was best when vancomycin AUC24:MIC ration was ≥345mg.h/L. The AUC24/MIC ratio is known to be the determinant of vancomycin efficacy. There is only modest correlation between this and trough levels in adults, although trough levels are recommended for monitoring for practical reasons. There is a paucity of pharmacokinetic and pharmacodynamic studies guiding vancomycin administration in children. It is unknown how well the AUC24/MIC ratio correlates with trough levels in children. Therefore, current national recommendations for vancomycin monitoring in children are extrapolated from those for adults.

This study from the Royal Darwin Hospital is a 2-yr retrospective review of vancomycin use and monitoring in 37 paediatric patients (1 month-12 years), with at least one trough serum vancomycin concentration available, to determine how often the new target vancomycin trough concentrations were actually being achieved.Vancomycin doses and dosing frequencies varied significantly, reflecting the lack of data on pharmacokinetics and pharmacodynamics to guide use of this drug in children, despite it being the key intravenous antibiotic used for the treatment of serious MRSA infections.

Few children achieved the currently recommended target trough vancomycin concentrations within an acceptable timeframe. Younger children (<6yrs) had higher vancomycin clearances than older children. The study illustrated the common, but mistaken, practice of changing to more frequent dosing regimens in order to achieve a higher trough level. This does indeed increase the trough level, but results in the same total drug exposure and AUC24 (ie efficacy unchanged, with increased burden on patient and nursing staff). Similarly, it underlines the importance of using a different target trough level when using 6hrly dosing (the target level will be higher) from when using 12hrly dosing.

The question is raised about the appropriateness of using vancomycin trough levels for monitoring in children given that there is no data on the correlation between vancomycin trough levels in children and AUC24:MIC. Young children (<6yrs) have a higher vancomycin clearance than older children and therefore probably need different target trough levels from those in adults in order to achieve the same target AUC. It may be that the AUC24:MIC ratio targeting 400mg.h/L is a more accurate and safe method of therapeutic monitoring in children. More detailed pharmacokinetic studies are needed in children, looking at different doses, dosing frequencies, trough concentrations and AUC24.

Submitted by Nan Vasilunas, 16 August 2013

Gut microbiota in preterm infants

Berrington JE, Stewart CJ, Embleton ND, Cummings SP. Gut microbiota in preterm infants: assessment and relevance to health and disease. Arch Dis Child Fetal Neonatal Ed 2013;98(4).

In adults the gut bacterial microbiome (microbiota) influences a diverse range of health outcomes including obesity, diabetes, asthma and allergy and seemingly ‘remote’ diseases like Parkinson’s disease. In preterm infants, establishment of the gut microbiota is also of importance for key morbidities like late onset sepsis (LOS) and necrotising enterocolitis (NEC), both significant causes of mortality. Episodes of LOS are often caused by gut- derived organisms and changes in the intestinal barrier contribute to both LOS and NEC. The gut microbiota are key to developing barrier function, integrity, and mucosal and systemic immune function. They also ‘educate’ the gut-associated lymphoid tissue, allowing the establishment of a tolerant’ state between microbiota and the immune system, affecting intestinal function including tight junction structure and immune function. Patterns of initial colonisation affect host metabolic function: fat deposition, circulating leptin levels, and insulin resistance.

In the preterm gut structural and immunological immaturity contribute to inflammatory necrosis and abnormal bacterial colonisation (dysbioses). This may result in decreased microbial diversity and an increased inflammatory response exacerbated by an immature innate immune response that increases the risk of diseases like NEC or LOS. An improved understanding of the microbiota of infants cared for in neonatal intensive care, and how this is affected by current practices may allow clinicians to promote more ‘healthy’ gut microbiota patterns, and may be associated with reductions in mortality and improvements in long term outcomes. It would seem we have neglected the “internal environment” and its effect on health and disease for too long.

Submitted by Paul Goldwater, 16 August 2013

Do CAP clinical practice guidelines affect management within hospital?

Neuman MI, Hall M, Hersh AL et al Influence of hospital guidelines on management of children hospitalized with pneumonia. Pediatrics 2012. 130(5): e823-830. PMID 23090342.

Background and methods: There is marked variability in the availability and content of clinical practice guidelines (CPG) for the management of community acquired pneumonia (CAP). There is limited data regarding testing and treatment patterns, and whether the CPG affect care of patients with CAP. This multicenter retrospective cohort study obtained discharge code, diagnostic and treatment information from the pooled Kids Inpatient Dataset (http://www.hcup-us.ahrq.gov/kidoverview.jsp) from 41 tertiary pediatric hospitals with emergency departments from 1 July 2009-30 June 2011. Hospitals were also questioned regarding presence and content of CAP CPGs.

Summary of main findings: 13 of 41 hospitals reported having a CPG. CPGs showed marked variability in recommended diagnostic testing (including routine FBC in 6/13, blood culture in 6/13 and CXR in 8/13) In the 19700 eligible separations from 38957 children admitted with a diagnosis of CAP, the presence of a CPG did not affect the outcomes of, diagnostic testing, length of stay, readmission with 14 days or hospital costs. CPG presence did increase the chance of narrow spectrum penicillins being used as first line therapy. CPG with absent recommendations regarding macrolide use in children >5 years were associated with lower macrolide use.

Implications for practice: CPGs for hospitalized CAP patients can influence antimicrobial choice, but appear to have little impact upon other measured aspects of clinician behavior. CPG for CAP vary widely in their recommendations, especially for diagnostic tests.


Submitted by Jim Buttery, 31 July 2013.

First-Day Step-Down to Oral Outpatient Treatment Versus Continued Standard Treatment in Children With Cancer and Low-Risk Fever in Neutropenia. A Randomized Controlled Trial Within the Multicenter SPOG 2003 FN Study.

(Brack E, Bodmer N, Simon A, MD, Leibundgut, Kuhne T, Niggli FK, Ammann RA. Blood Cancer 2012;59:423–430. PMID 22271702)

Background: This is a well designed and clearly reported non-inferiority randomized control trial in “low risk” children with febrile neutropenia to establish that the efficacy and safety of step down to oral antibiotics (amoxicillin and ciprofloxacin) and discharge home within 8-22 hours of initial presentation is non-inferior to standard inpatient management. Children were recruited if they presented with a temperature >38.5 once or 38 twice and with neutrophils < 0.5. They underwent FBC, blood culture and physical examination with extra investigations as appropriate. Between 8 and 22 hours they were reviewed by a Paediatric Oncologist to determine their status as a low risk patient based on stringent exclusion criteria. At this point they were randomized to be discharged home on oral antibiotics with daily clinical review, an FBC every second day and repeat blood cultures if febrile or standard intravenous treatment as an inpatient. Post discharge antibiotics were stopped if fever resolved for 48 hours and cultures were negative. Efficacy was defined as no change to treatment protocol and no adverse events including a serious medical complication, development of a positive blood culture, diagnosis of a viral infection or development of radiologically confirmed pneumonia. Safety was defined as the absence of a serious medical complication including death, admission to PICU and complications attributable to infection as judged by the treating physician. Outcomes of efficacy and safety were monitored until 7 days post discharge.

Main Findings: In both intention to treat and per protocol analyses statistical significance, that is that OPD oral antibiotics were non-inferior to standard inpatient IV antibiotics within a 3.5% margin of error, could not be achieved due to insufficient sample size. This is despite the fact that the only serious medical complication during the trial (fatal adenovirus infection) occurred in the standard care group. Analysis of efficacy outcomes however did demonstrate less than 10% non-inferiority in the OPD group.
Take home message: This trial is notable for its use of a non-inferiority design to try and provide methodological sound evidence and effectively answer this important clinical question. The authors rationalized the use of this design as previous superiority trials had only succeeded in showing no or marginal difference between groups. In choosing a non-inferiority methodology for this trial the authors are hypothesizing that outpatient oral antibiotic treatment may be an alternative to standard treatment, not superior to it. Power calculations and statistical analysis have to be based on predefined margins of inferiority resulting in much bigger sample sizes with this trial design. The major limitation of the trial was the failure of recruitment resulting in early termination and an inability to reach significance for the major outcome of safety. The difficulties encountered by the group in recruiting patients when stringent safety criteria were applied underpins the difficulty in conclusively answering this question in clinical trials and the limited generalizability of the intervention. The most recently published paediatric febrile neutropenia guidelines from NICE recommend consideration of OPD oral therapy in “low risk” patients after 48 hours only.

Submitted by Bridget Freyne, 19 June 2013.

Central Venous Catheters and Catheter locks in Children with cancer: A Prospective randomized control Trial of Taurolidine Vs Heparin.

(Handrup MM, Møller JK, Schrøder H. Pediatric Blood Cancer 2013 Feb 15. Epub ahead of print. PMID 23417891)

Background: Central venous lines (CVL) play a major role in the management of cancer in all age groups. CVL’s easily form biofilms, creating an excellent environment for bacterial growth. Eradication of CVC related blood stream infection continues to be one of the goals in management of cancer patients. The taurine amino-acid derivative, taurolidine, has broad antibacterial activity against both gram positive and negative organisms, possibly facilitated by bacterial cell wall disruption thus neutralizing endotoxin and exotoxin, and antiadherence properties. Previous small trials have shown decreased blood stream infections when taurolidine is used as a CVL lock (Taurolock) .

Method: This single centre open label RCT compared Taurolock versus standard heparin lock (250 IE heparin in 0.9% saline) in CVL’s with otherwise identical standard of insertion and care. Study period was from April 2008 till August 2012. Out Of 141 CVL’s eligible, 130 CVL’s from 113 patients participated in the trial after exclusion, with equal numbers in both arms. Most of the participants receiving Taurolock experienced a transient bad oral taste as a side effect and hence one patient withdrew from the trial. End points of the trial were: catheter related blood stream infection (CRBSI), time to first CRBSI, removal of CVL, death of the patient with CVL, transfer of the patient to other specialty where adherence to the trial protocol was in doubt or reaching end of the trial period. The inherent observer bias within the open label design was offset by the laboratory end points.

Interesting findings: Rate of total blood stream infection in Taurolocked CVL was 1.2 per 1,000 CVL days while it was 2.5 per 1,000 CVL days in heparin locked CVL`s. Incidence rate ratio (IRR)= 0.49 and P value = 0.004. The rate of CRBSI in Taurolock group was 0.4 per 1,000 CVL days compared to 1.4 per 1,000 CVL days. CRBSI with coagulase negative staphylococci decreased by 66% in Taurolock group. Exit site infections however were no different in both groups (0.3 Taurolock v 0.1 standard). CVL infection was indentified as the major reason for premature line removal in standard Heparin lock when compared to CVL locked with Taurolock. No line with Taurolock was removed due to occlusion.

Take home messages:
1. Taurolock is independently protective against CVL related BSI in children with cancer. 2. Taurolock was well tolerated and did not appear to compromise CVL patency in this small series – however larger numbers may be required to ensure this.
3. Taurolock can be considered as CVL locks for children with long term CVL’s which could reduce the associated infection related complications
4. Taurolock is however not a substitute for exemplary CVL care, but rather an adjunct.

Submitted by: Dr Arun Bojarajan (Paediatric ID registrar) and Dr Julia Clark, 16 May 2013.


Treatment of pyelonephritis caused by extended-spectrum beta-lactamase-producing enterobacteriaceae in children

(Katsuta T, Shoji K, Watanabe Y, Saitoh A.Infect Dis J. 2013 Apr;32(4):417-9)

Background and methods: Gram negative resistance is on the rise, and paediatric ID teams are increasingly being asked to assist with the management of children infected with ESBL producing Enterobacteriaceae. Therapeutic options are limited, and carbapenems are usually indicated for serious infections. Principles of antimicrobial stewardship would favour use of narrower spectrum antibiotics if appropriate, in order to reserve carbapenems where possible. Carbapenem therapy can also be problematic for the treatment of children with pyelonephritis who may be well enough to be managed at home.

This study retrospectively examined 54 children treated for ESBL positive pyelonephritis (defined as fever, pyuria and positive culture on catheter or clean catch samples), treated in a large tertiary children’s hospital in Tokyo between April 2009 and December 2010. Antibiotic choices were at the discretion of individual physicians, without any control over these decisions. No control group was included in the study. Treatment outcomes were resolution of fever, infection relapse within 4 weeks of completion of treatment, and cure (defined as absence of symptoms and no recurrence).

Main findings: 32/54 (59%) were treated empirically with an agent to which the organism was not susceptible. There was no significant difference in the length of fever between these and the ones who were initially treated with susceptible antibiotics (2.2 vs 2.0 days respectively; p=0.62). 30/32 were switched to effective antibiotics, after an average of 4.4 days. Definitive antibiotics were chosen based on susceptibility testing. 21 were given intravenous cefmetazole (2nd gen cephalosporin), and a handful were given pip/taz (n=3), gent (n=3), cipro (n=2) or mero (n=2). Oral agents were used in 28/54 (52%): cotrimoxazole (n=16), fosfomycin (n=7) and quinolones (n=5). Mean treatment duration was 13.7 days (range 7-24 days). All patients were cured of infection and no relapses were documented within 4 weeks of completion of therapy (although it is unclear how hard they looked).

Take-home messages: Non-carbapenem antibiotics, such as cephamycins (and possibly beta-lactam/beta-lactamase combinations) may have a role in the treatment of non-bacteraemic ESBL pyelonephritis, although the retrospective, observational nature of this study does not really provide robust evidence for such an approach. Prospective, randomised studies comparing carbapenem head-to-head against alternatives would be useful. Whilst the use of an alternative intravenous agent would have the advantage of ‘sparing’ carbapenems, finding a safe and effective oral option is a more pressing problem for individual children with pyelonephritis. Resistance to cotrimoxazole and quinolones is increasing, and fosfomycin use is limited by availability and a lack of paediatric data. This study describes treatment success for a small cohort of patients treated with fosfomycin, but further research into its use in paediatric pyelonephritis would be very valuable.


Submitted by Josh Francis, 9 April 2013


Two studies describing the herd effects of paediatric vaccines (both submitted by Tony Walls 15 March 2013)

Impact of the 7-valent pneumococcal conjugate vaccine on invasive pneumococcal disease in infants younger than 90 days in England and Wales

(Ladhani et al. Clin Infect Dis. 2013 Mar;56(5):633-40)

Streptococcus pneumonia is an uncommon but well-recognised cause of invasive bacterial infection in young infants. The UK has had active surveillance for invasive pneumococcal disease (IPD) since 1998 through the Health Protection Agency. This study looked at trends in IPD in infants <90 days of age before and after the introduction of the 7-valent pneumococcal conjugate vaccine (PCV7) in September 2006. Prior to PCV7 introduction the incidence of IPD in infants <90 days of age was 13/100,000 live births (95%CI 12.0-14.0) with serotypes contained in PCV7 accounting for 44% of isolates. The highest rates of disease occurred in infants in the first week of life, and there was an overall case fatality ratio of 7%.

Following the introduction of PCV7 there was an 83% reduction in IPD due to vaccine serotypes. Most of these infants would have been too young to receive PCV vaccine and the authors suggest that this herd effect is due to a reduction in household transmission from vaccinated siblings. However the overall reduction of IPD in this group was small with an increase in non-vaccine serotypes causing disease.

Indirect protection of adults from rotavirus by pediatric rotavirus vaccination

(Anderson et al, Clin Infect Dis. 2013 Mar;56(6):755-60)

Routine use of rotavirus vaccine has been shown to provide indirect protection of unvaccinated children in both Australia and the USA. However it is not clear if this effect extends to adults. This study tested stored stool samples from adults that had been initially processed for routine bacterial stool cultures. This was used as a marker for significant diarrhoeal illness in this population. Inpatients and outpatients were included and the study compared results from 2006-7 (pre-vaccination era) and 2008-10 (post vaccination). They found a 48.4% decline in rotavirus detection from stools with an absolute reduction of 2.11% (p=0.0007). There was no associated decline in rotavirus-associated hospitalisations.

It is hard to know if these findings reflect a true reduction in adult rotavirus disease in the community. This was a single centre lab-based study, and a high proportion of patients who tested positive for rotavirus were immunocompromised. In addition the pre-vaccine era was quite short and included just one rotavirus season. However, it makes sense that rotavirus vaccination may reduce disease in adults who care for young children. The authors comment that even a small absolute reduction in adult rotavirus disease could lead to enormous cost savings in the USA. The potential effects on adults were never factored in to cost-effectiveness calculations made prior to rotavirus vaccine introduction.



Antibiotic exposure and IBD development among children: a population-based cohort study

(Kronman MP, Zaoutis TE, Haynes K, Feng R, Coffin SE. Pediatrics 2012;130:2011-3886.)

Background: This is a well designed, retrospective, population based cohort study that investigated the association between exposure to antianaerobic antibiotics in childhood (£ 17 years) with development of inflammatory bowel disease (IBD). Electronic records of > 9 million patients (5.7% of all UK (out)patient visits) from a network of 464 UK ambulatory practices (The Health Improvement Network, THIN, 1994-2009) were captured. Patients were classified as “ever” (64%) or “never” having had antibiotic exposures to antianaerobic antibiotics (penicillin, amoxicillin, ampicillin, and penicillin/ß-lactamase inhibitor combinations, tetracyclines, clindamycin, metronidazole, cefoxitin, carbapenems, and oral vancomycin), by link to written prescriptions.* Primary outcome was IBD development (by validated definiton). Secondary outcome was by IBD according to classes of antibiotics. Patients were excluded if they recieved antibiotics during the “latency period” (study-defined time betwen onset of IBD symptoms and formal diagnosis) to address misclassification bias. Multivariate analysis included known co-mordities for IBD e.g. family history, CGD, primary sclerosing cholangitis. *Note (PP): the study assumes a “prescription” equates to actual consumption of antibitoics

Main findings: A total of 1, 072,426 children were included in analysis; 748 (0.07%) developed IBD. IBD incidence rates were 0.83 and 1.52 per 10,000 person-years in unexposed and exposed children, for an 84% relative risk increase. In multivariate analysis: the highest association was with co-morbidities. However, there was an independent, significant association with age of exposure to antibiotics, with a cumulative dose-response effect seen. Exposure before 1 year of age was associated with a 5.5 fold increased IBD risk [95% confidence interval, 1.66 - 18.28), decreasing to 2.62 [95% confidence interval, 2 - 18.28] at 5 years and to 1.57 [95% confidence interval, 1.66 - 18.28] by 15 years. Each anti-anaerobic course increased the risk for IBD by 6%. The effect was greatest for penicillins, broad spectrum penicillins, cephalosporins, metronidazole and flourquinolones. Antibiotics were prescribed for GIT reasons in only 1.6% and in 38.9%** no specific infectious diagnosis was found. [** Note: PP: implying that in nearly 40% of cases, antibiotics may have not been needed - commensurate with current literature where ~50% antibiotics use is inappropriate]. Results for patients by Crohn’s disease and ulcerative colitis were similar.

The results mean that for every 14 300 prescriptions for antibiotics with anaerobic cover, 1 child will develop IBD (UK) or an additional 1700 cases of IBD in the USA yearly.

Take home message: As infectious diseases physicians, we champion judicious antibiotic use. It is noteworthy when we can cite evidence that early antibiotic use is associated with a long term adverse outcome.This paper provides further fodder against the (repeated), inappropriate use of antibiotics, particularly in the young child. So, the message is not new: “If one uses antibiotics, think first if it is really needed, especially in the young child where the increase in IBD risk is 5 x above baseline, and higher with repeated courses”

Of note, the basis of this predisposition to IBD may lie with the diruption of normal healthy flora in early life, a subject that is the focus of many recent publications.

Submitted by: Pam Palasanthiran, 15 February, 2013