Journal Club 2012

Waning protection of acellular pertussis vaccine.

(Klein NP, Bartlett, J, Rowhani-Rahbar, A, Fireman B, Baxter R. Waning Protection after Fifth Dose of Acellular Pertussis Vaccine in Children. N Engl J Med. 2012; 367: 1012-9. PMID: 22970945)

There is no good serological correlate of protection against whooping cough, so the only truly valid measure of vaccine efficacy is protection against disease. The decision to move from whole cell to acellular pertussis vaccines in Western countries was made on the basis of equivalent short-term clinical protectiveness and reduced reactogenicity with reduction of minor but not major adverse events, but without knowledge of long-term efficacy. Large outbreaks of pertussis in Western countries have raised concerns about acellular pertussis vaccine efficacy. A US case-control study looked at time from 5th dose of acellular pertussis-containing vaccine (DTaP) and risk of pertussis. Children who were PCR-positive for pertussis were more likely to have received their fifth DTaP dose earlier than PCR-negative controls (P<0.001) or matched controls (P=0.005). After the fifth dose of DTaP, the odds of acquiring pertussis increased on average by 42% per year. The 5th dose was given age 4-6 years and the peak age of infection was 8-11 years. Children over 11 years were rarely infected which the authors postulate is because they received whole cell pertussis vaccine. Whole cell pertussis vaccines are still used in most developing countries. Will Western countries have the courage to go back to the whole cell vaccine?

Submitted by: David Isaacs, 17 December 2012.

Early primary cytomegalovirus infection in pregnancy: maternal hyperimmunoglobulin therapy improves outcomes among infants at 1 year of age.

(Visentin S, Manara R, Milanese L, Da Roit A, Forner G, Salviato E, Citton V, Magno FM, Orzan E, Morando C, Cusinato R, Mengoli C, Palu G, Ermani M, Rinaldi R, Cosmi E, Gussetti N. Clinical Infectious Diseases. 55(4):497-503, 2012 Aug. PMID: 22539662)

Background: Congenital CMV is the leading infectious cause of congenital hearing loss and neurodisability. Antiviral treatment for infants born with CNS or severe focal organ disease improves hearing outcomes, but there currently is no consensus evidence-based management for primary maternal CMV in pregnancy. Despite this, routine antenatal CMV screening is conducted in Italy and some other countries. This study was conducted between 2002 – 2009 in Padua, Italy. Women with primary maternal CMV infection diagnosed.

17 weeks gestation by seroconversion and/or positive IgM with low IgG avidity were included. From 2002 – 2007 no intervention was offered and from 2007 onwards women with CMV-infected fetuses (CMV PCR positive amniotic fluid) were offered a single dose of CMV hyper-immune globulin (HIG). Infant outcomes up to 1 year of age were evaluated.

Main findings: 92 women had positive CMV PCR on amniotic fluid. 24 /92 had terminations. 31/92 received HIG and of these, 4/31 [13%, 95% CI 1% – 25%] had abnormal antenatal scans/poor infant clinical outcomes. 16/37 [43%, 95% CI 27% – 59%] babies whose mothers did not receive HIG had poor infant outcomes. P< 0.01 [2-tailed Fisher’s Test]. 23 infants of 349 mothers with negative amniocentesis PCR went on to be diagnosed with congenital CMV on the basis of positive CMV urine in the first week of life. An additional 155 women with primary infection did not undergo amniocentesis. Of these, 37 had terminations and 5 had miscarriages. Among the 113 liveborns, 28 (25%) had a positive CMV urine test and 9 with CMV had poor clinical outcomes. The overall vertical transmission rate was 114/449 (25%).

Implications for practice: HIG appears to be associated with improved infant outcomes, however potential selection biases and small numbers mean these findings must be interpreted cautiously. The authors did not comment on whether any of the liveborn infants were treated postnatally with (val)ganciclovir, which may have modified infant hearing and developmental outcomes at 1 year. The authors did not discuss optimal dosing schedules or routes for HIG. Vertical transmission of CMV can occur despite negative amniocentesis. The majority of maternal CMV infections are asymptomatic. In Australasia and other settings where antenatal CMV screening is not routinely conducted it may be difficult to identify suitable candidates for HIG treatment as many women with CMV infected fetuses identified by ultrasound abnormalities, serology and amniocentesis elect to terminate their pregnancies.

Submitted by: Rachel Webb, 12 November, 2012

Household versus Individual Approaches to Eradication of Community-Acquired Staphylococcus aureus in Children: A randomised trial.

(Fritz SA, Hogan PG, Hayek G, Eisenstein KA, Rodrigues M, Epplin EK, Garbutt J, Fraser VJ. Clinical Infectious Diseases 2012;54(6):743–51; DOI: 10.1093/cid/cir919) PMID 22198793

Background: Infectious diseases physicians are frequently asked to manage recurrent skin and soft tissue infections (SSTI). A number of different regimens exist including topical and systemic antiseptics and antimicrobials. Treatment of household family members and hygiene measures are also frequently recommended, but there is little evidence to support these recommendations. Fritz et al performed an open label, randomised control trial in children with community onset S. aureus SSTI with confirmed nasal S. aureus colonisation. Following resolution of their acute SSTI, decolonisation therapy included i) 5 days of daily 4% chlorhexidine and twice daily 2% nasal mupirocin ointment and ii) standardized hygiene advice. Participants were randomised to either individual or household decolonisation.

Main Findings: One hundred and eighty three cases were randomised. The adherence to the protocol was high in both groups (index group: 82%; household group: 80%). No differences in S. aureus colonisation were observed at 1, 6 and 12 months (1 month - index group: 50%, household group: 51%, NS). Despite this, the rate of SSTI recurrence was significantly less in children from undergoing household decolonisation at 3, 6 and 12 months (12 months: self-reported recurrence of SSTI - index group: 72%, household group: 52%, p = 0.02; physician documented recurrent SSTI – index group; 55%, household group: 36%)

Take-Home Message: This paper provides evidences that household decolonisation strategies following community-acquired S. aureus infection can further reduce the burden of SSTI in children. However, it also demonstrates a high recurrence rate with both treatment strategies in a highly motivated and adherent population.

Submitted by: Chris Blyth, 15 October 2012

Rapid molecular diagnosis of pulmonary tuberculosis in children using nasopharyngeal specimens

(Zar HJ et al, Clin Infect Dis. 2012 Jul 2. [Epub ahead of print], PMID: 22752518)

Background: Diagnosis of pulmonary tuberculosis (PTB) in children remains challenging. Culture remains the gold standard but results may take weeks. For rapid diagnosis, the traditional role of smear microscopy has been diminished with the development of molecular techniques such as Xpert. Due to difficulties in expectorating sputum in young children, specimen collection in children where PTB is suspected has been by gastric lavage (GL) and more recently by induced sputum (IS). Nasopharyngeal aspirate (NPA) is a potentially attractive alternative, more accessible than IS and less invasive than GL.

Main Findings: Of 674 children hospitalised consecutively with suspected PTB in Cape Town, South Africa between 2009 and 2011, 535 children had one IS and NPA; 396 had two paired specimens. A positive smear, Xpert or culture occurred in 30 (5.6%), 81 (15.1%) and 87 (16.3%) respectively. Xpert detected all smear positive cases and 40/57 (70.2%) of smear negative cases. Xpert was significantly more sensitive than smear (80.5% vs. 34.5%). Culture yield from IS was significantly greater than from NPA (84/87 96.1% vs. 61/87 70.1%, p<0.001). Incremental culture yield and sensitivity of Xpert for both IS and NPA improved with a second paired specimen set, an effect that was greatest when specimens were taken on sequential days rather than on the same day. Sensitivity of two Xperts on IS was 71.4% and for two NPAs was 65.1%. Xpert specificity was 99.1% on IS and 98.2% on NPA. Sensitivity for determination of Rifampicin susceptibility for Xpert was 83.3% and specificity 99.1%. As would be expected, results from Xpert were faster than culture (0 vs. 15 days).

Take-Home Messages: This study of rapid diagnostic techniques for children with suspected PTB in a high prevalence setting confirms the superiority of Xpert over smear microscopy, reinforces the value of repeated specimens taken on sequential days, and establishes the place of NPA alongside IS and GL for specimen collection. The findings suggest that, where possible, paediatricians in Australia and New Zealand should consider requesting Xpert in preference to traditional smear microscopy. While sensitivity of NPA for rapid diagnosis of PTB by Xpert is lower than for IS, NPA is a familiar technique for paediatric ward staff in our setting, requires no extra training, is especially attractive at centers with limited access to IS at times such as after hours or weekends when specially trained staff are unavailable, has lower risk of nosocomial transmission and in instances where GL yields are of consistently poor quality. As with other TB diagnostics, thorough assessment of pre-test probability for children in our relatively low-prevalence setting will be important in making best use of Xpert, whichever specimen collection technique is employed.

Submitted by: Josh Osowicki, Asha Bowen, 4 September 2012

Long-term effects of clearing Helicobacter pylori on growth in school-age children

(Mera R.M. et al Paediatr Infect Dis J 2012;31:263-266 PMID: 22315005)

Background: An apparent negative impact of H. pylori on growth velocity has been noted in previous cohort and cross-sectional studies, however it is not known whether attained height and weight in children will be impacted over the long term. Robertino and colleagues investigated 295 school-aged children in 2 cohorts in two rural communities in Colombia and followed them for an average of 3.7 years. The children were tested for H. pylori using the C-urea breath test every 6 months and those who tested positive in the intervention cohort were treated with amoxicillin, metronidazole, lansoprazole and bismuth using DOT for 14 days. Anthropometric measurements were taken of all children every 3 months.

Main findings: H. pylori negative and positive children had similar average heights at the commencement of the study and by the last visit H. pylori negative children had an average height of 130.5cm, which was a full centimeter taller than the average height of H. pylori positive children. The differences in height were progressively larger with time, as the children aged. Similarly H. pylori negative and positive children had nearly identical average weights at the commencement of the study and by the last visit H. pylori negative children had an average weight of 28.6kg, nearly a kilogram more than H. pylori positive children. A multivariate mixed model adjusted for age, gender, father’s education, number of siblings and presence of helminthes/protozoa in stool found no significant differences in height or weight at baseline by H. pylori status.

Take home message: This study suggests that school-age children’s growth benefits from being treated for H. pylori infection, not only in the short term but in the long term as well. H. pylori infection may produce partial loss of the gastric acid barrier among children, and this may result in chronic diarrhoea, malnutrition due to impaired nutrient absorption or direct nutrient losses. 78% of the children involved in this study tested positive for H. pylori at the first breath test, data from Australia suggests that less than 10% of children are positive for H. pylori. The applicability of this data to Australian children in general may be limited, however it is certainly comparable to high-risk populations including refugees and Indigenous Australians. Growth in children is complex and there may be some confounders unaccounted for. This study did however account for obvious variables such as diet, co-infection and socioeconomic status. The rate of growth differences may be underestimated as reoccurrence/recrudescence of the infection made the initial 73% clearance rate decline after one-year. Nonetheless this study showed a significant and durable effect of clearing H. pylori infection on increased height and weight.

Submitted by: Anita Campbell, 1 August 2012

Efficacy of immunoglobulin plus prednisolone for prevention of coronary artery abnormalities in severe Kawasaki disease (RAISE study): a randomised, open-label, blinded-endpoints trial.

(Kobayashi et al, Lancet 2012, PMID: 22405251 andSon & Newburger, Lancet 2012, PMID: 22405252)

BACKGROUND: In contrast to other inflammatory vasculitides, the benefit of steroids as primary adjunctive therapy in Kawasaki disease (KD) is controversial. A randomised controlled trial (RCT) of primary adjunctive steroids in unselected KD patients from the US did not find any benefit on coronary artery (CA) outcomes (PMID: 17301297), although a post hoc analysis suggested a significant beneficial effect in a subgroup who subsequently required further IVIG. The recently published Japanese RCT (The RAISE study; PMID: 22405251) investigated whether primary adjunctive steroids were of benefit in selected KD patients, assessed as being at high risk of IVIG non-response.

MAIN FINDINGS: High-risk KD patients, identified using the authors’ own severity score, were randomised to receive standard therapy (IVIG and aspirin) +/- a prolonged course of steroids. The primary outcome was CA abnormalities and analysis was by number needed to treat (NNT). Of 2014 KD patients assessed from 74 hospitals, 248 (12%) were included. The incidence of CA abnormalities was 4/121 (3%) in the steroid group and 28/121 (23%) in the standard treatment group (p<0.0001) during the study period; NNT to prevent CA abnormalities was five. Four weeks after onset, the incidence of CA abnormalities was 4/120 (3%) in the steroid treatment group and 15/120 (13%) in the standard treatment group (p=0.014); NNT was ten.

TAKE-HOME MESSAGES: These are impressive results but caution is required before these findings are generalised to non-Japanese populations. Importantly, risk scores developed for the Japanese, including that used in the RAISE study, have unacceptably low sensitivity in non-Japanese populations (PMID: 21168857). Other important differences include the timing of the start of treatment in the RAISE study, which was at a median of day 4 of illness. This is earlier than is usual (or often operationally possible) in Australia and other countries. The steroid regimen in the RAISE study included an initial minimum 5 days of intravenous treatment, which potentially would increase inpatient costs. Overall the RAISE and other studies indicate that steroids may have a role in high-risk KD patients. The challenge is to identify this small sub-group in non-Japanese populations, so that trials of steroids and other adjunctive therapies can be performed.

Submitted by: David Burgner (, Katherine Chen and Nigel Curtis, 20 July 2012

Is malodorous urine associated with urinary tract infection in children?

(Gauthier M. et al. Pediatrics 2012 May;129(5):885-90. [Epub 2012 Apr 2])

Background and methods: Although data are contradictory, malodorous urine has been reported in between 2 to 20% of children with urinary tract infection (UTI). In this two-year prospective study in a tertiary care Canadian ED, Gauthier and colleagues investigated whether parental reporting of malodorous urine was associated with confirmed UTI in children aged 1 to 36 months that presented with fever without focus. Results from a standardised questionnaire from parents of 331 children were included in the analysis.

Summary of main findings: Of 331 children (median age 12 month), 51(15%) had a UTI. The majority of urine specimens (88%) were obtained via catheter. Malodorous urine was reported in 57% of children with UTI and 32% of children without UTI. Children with malodorous urine were more likely to have UTI (OR 2.83 (1.54-5.2)) compared to children without malodorous urine, even after adjustment for sex and the presence of VUR. However the sensitivity and specificity of malodorous urine for UTI was only 57% (42-70) and 68% (62-74) respectively.

Implications for practice: Malodourous urine is more likely in children with confirmed UTI. However, 40% of children with a confirmed UTI do not have malodorous urine and up to one third of children with parental reported malodorous urine do not have a UTI, limiting the ability to rule out or rule in a UTI based on this sign alone. Despite the study limitations, the findings suggest that parental reporting of malodorous urine, albeit not perfect, may be a useful clinical sign to aid clinical decision-making in routine clinical practice.

Submitted by: Tom Connell, 13 June 2012

Acyclovir suppression and neurodevelopment outcome after neonatal HSV

Kimberlin DW et al. NEJM 2011 Oct 6;365:1284

This study reports a potential neurodevelopment benefit for 6 months of AVC suppressive treatment in infants following CNS or SEM disease in the neonatal period. Although numbers were relatively small, infants in the ACV achieved higher Bayley scores compared to infants in the placebo group (88.2 vs. 68.1, p=0.04). Benefit was greatest in infants with previous CNS disease. There was a trend towards neutropenia in the ACV arm but no infant developed any significant infections. Longterm follow-up results of the infants in recruited to trial are eagerly awaited.

Submitted by: Tom Connell 15 Feb 2012