Microbiology of septic arthritis in young Auckland children.

BACKGROUND: Kingella kingae is an important cause of septic arthritis in young children, with modern laboratory methods leading to increased detection. Prevalence of this pathogen in New Zealand, where there are high rates of childhood infections due to Staphylococcus aureus and Streptococcus pyogenes, is not known. METHODS: We conducted a retrospective review of children <5 years with septic arthritis (without osteomyelitis) at a tertiary children's hospital in Auckland, over 10 years (2005-2014). Data were collected on demographics, microbiology, clinical presentation, investigations and management. RESULTS: Of the 68 cases of septic arthritis, 57 (83.8%) occurred in children aged <24 months. Among those <3 months, Streptococcus agalactiae (Group B streptococcus) was predominant (45.5% of 11 cases), followed by S. aureus (36.4%). The most common pathogen in those 3 to <12 months was Streptococcus pneumoniae (38.5% of 13 cases). In children aged 12 to <24 months, K. kingae was most common (30.3% of 33 cases). Of the 12 cases of K. kingae, 91.7% were identified from synovial fluid culture. All K. kingae isolates were susceptible to amoxicillin. CONCLUSIONS: K. kingae is the leading pathogen in septic arthritis in New Zealand children aged 12 to <24 months. Routine inoculation of synovial fluid into blood culture bottles at time of sample collection, in addition to use of polymerase chain reaction methods, should be encouraged to improve detection rates. For infants and preschool children presenting with single joint septic arthritis, empiric antibiotics should include cover for S. aureus and K. kingae.

Community-acquired invasive Staphylococcus aureus: Uncovering disparities and the burden of disease in Auckland children.

AIM: Staphylococcus aureus (SA) causes serious invasive disease in children. Large studies have measured the incidence of SA bacteraemia, but there is less information on the total burden of community-acquired invasive SA (iSA) in children. METHODS: A retrospective, cross-sectional analysis of Auckland resident children aged 0-14 years who were hospitalised with iSA between 2011 and 2015 was performed. Laboratory databases and SA-related international classification of diseases 10 discharge codes were searched to identify community-onset cases with SA isolated from a normally sterile site. Clinical records and coroner's reports were reviewed to determine clinical syndromes and exclude nosocomial infections. RESULTS: A total of 295 children with iSA were identified. The average annual incidence of iSA was 18.6 per 100 000 - for Pacific populations 44.3 per 100 000, Maori 24.3 per 100 000 and New Zealand European and other 8.8 per 100 000; 68% had bacteraemia. The incidence of iSA for Pacific infants was 10 times greater than non-Maori/non-Pacific (113.4/100 000 population vs. 11.8/100 000). Multivariate analysis found a higher risk of admission in Pacific children, males and those living in areas of high deprivation. Thirty-two patients (10.8%) were admitted to the intensive care unit; risk was higher in infants, Pacific children and those with respiratory infection (Relative Risk (RR) 12.2, 95% confidence interval (CI) 5.7-26.4) and multifocal (RR 6.9, 95% CI 3.4-13.8) and endovascular disease (RR 8.9, 95% CI 3.9-20.6). All deaths (n = 7) had respiratory infections, and four were patients <1 year of age. CONCLUSIONS: Studies investigating SA bacteraemia alone significantly underestimate the total burden of iSA disease. There are marked ethnic and socio-economic disparities in iSA disease among Auckland children. Pacific infants are at the highest risk.

Post-streptococcal glomerulonephritis: Some reduction in a disease of disparities.

AIM: A retrospective Auckland-wide (total population approximately 1.4 million) study of hospital admissions from 2007 to 2015 was conducted to assess trends in admissions for acute post-streptococcal glomerulonephritis (APSGN) in children aged 0-14 years. METHODS: International Statistical Classification of Diseases (ICD10) discharge codes were used to identify potential cases of APSGN, and electronic clinical records and laboratory data were compared with established case definitions for definite or probable APSGN. RESULTS: A total of 430 cases of APSGN were identified (definite n = 337, probable n = 93), with a mean annual incidence of 15.2/100 000 (95% confidence interval (CI) 14.9-15.6). Incidence (0-14 years) was 17 times higher in Pacific peoples (50.2/100 000, 95% CI 48.6-51.8) and almost 7 times higher in Maori (19.6/100 000, 95% CI 18.6-20.7) than European/other populations (2.9/100 000, 95% CI 2.7-3.1). Multivariate analysis found ethnicity, deprivation, male gender, age (peak 3-8 years) and season (summer/autumn) to be associated with admission risk. Admission rates showed a significant change of -9.0% (95% CI -10.4, 7.4%) per year, with 2011 being an exception. Low C3 complement, hypertension, elevated streptococcal titres, oedema and heavy proteinuria were present in 94, 65, 67, 52 and 49% of cases, respectively. Relying on ICD10 codes without further review of clinical notes would result in an overcount of cases by 25%. CONCLUSIONS: There is severe disparity in APSGN admission rates, with a disproportionate burden of disease for Pacific and Maori children and those living in deprived circumstances. Rates trended downward from 2007 to 2015.

Respiratory virus detection during hospitalisation for lower respiratory tract infection in children under 2 years in South Auckland, New Zealand.

AIM: To describe respiratory virus detection in children under 2 years of age in a population admitted with lower respiratory infection and to assess correlation with measures of severity. METHODS: Nasopharyngeal aspirates from infants admitted with lower respiratory tract infection (n = 1645) over a 3-year time period were tested by polymerase chain reaction. We collected epidemiological and clinical data on all children. We assessed the correlation of presence of virus with length of hospital stay, intensive care admission and consolidation on chest X-ray. RESULTS: Of the children admitted 34% were Maori, 43% Pacific and 75% lived in areas in the bottom quintile for socio-economic deprivation. A virus was found in 94% of those tested including 30% with multiple viruses. Picornavirus was present in 59% including 34% as the sole virus. Respiratory syncytial virus was found in 39%. Virus co-detection was not associated with length of stay, chest X-ray changes or intensive care unit admission. CONCLUSION: In this disadvantaged predominately Maori and Pacific population, picornavirus is commonly found as a sole virus, respiratory syncytial virus is frequent but immunisation preventable influenza is infrequent. We did not find that co-detection of viruses was linked to severity.

Association study involving polymorphisms in IL-6, IL-1RA, and CTLA4 genes and rheumatic heart disease in New Zealand population of Maori and Pacific ancestry.

INTRODUCTION: Rheumatic fever (RF) incidence among New Zealand (NZ) individuals of Polynesian (Maori and Pacific) ancestry remains among the highest in the world. Polymorphisms in the IL-6, IL1RN, and CTLA4 genes have been associated with RF, and their products are modulated by new medications. Confirmation of these previous associations could help guide clinical approaches. We aimed to test IL-6, IL-1RA (IL1RN), and CTLA4 functional SNPs in 204 rheumatic heart disease (RHD) patients and 116 controls of Maori and Pacific ancestry. MATERIAL AND METHOD: Self-reported ancestry of the eight great-grandparents defined ancestry of participants. Severity of carditis was classified according to the 2012 World Heart Federation guideline for the echocardiographic diagnosis of RHD. The IL-6 promoter rs1800797, IL1RN rs447713 and CTLA4 rs3087243 SNPs were genotyped by Taqman. Correlations were assessed by logistic regression analysis adjusting for gender and ancestry. RESULTS: The IL-6 rs1800797 variant was significantly associated with RHD with carriers of the GG genotype 6.09 (CI 1.23; 30.23) times more likely to develop RHD than the carriers of the AA genotype (P=0.027). No significant associations with RHD were found for the IL1RN rs447713 and CTLA4 rs3087243 SNPs. Patients carrying the G allele (GG plus AG genotype) for the IL1RN rs447713 SNP had 2.36 times (CI 1.00; 5.56) more severe carditis than those without this allele (the AA genotype) (P=0.049). CONCLUSION: The IL-6 promoter rs1800797 (-597G/A) SNP may influence susceptibility to RHD of people of Maori and Pacific ancestry living in NZ. The IL1RN rs447713 SNP may influence the severity of carditis in this population.

Antimicrobial stewardship using pharmacy data for the nurse-led school-based clinics in Counties Manukau District Health Board for management of group A streptococcal pharyngitis and skin infection.

AIM: To evaluate antimicrobial usage in the school-based clinics against operating guidelines. METHOD: Antimicrobial prescribing data (2014) from 10/18 participating pharmacies serving 14,153/23,588 primary school children of the programme were accessible. Prescriptions from 5/10 pharmacies were available for identifying type, amount, and indication of the medicine. One pharmacy serving a defined population (n=3,513) with single healthcare provider delivering the school programme was selected for detailed evaluation and identifying individuals receiving multiple treatments. RESULTS: Data from 10 pharmacies (n=7,889 prescriptions) showed 91.2% of prescriptions were for group A streptococcal-positive throat swab, 8.8% for skin infections. More detail from 5/10 pharmacies showed only 2% of group A streptococcal pharyngitis treatments (107/4,672) were not first-line (56 cephalexin and 51 rifampin prescriptions). Fusidic acid (159/452, 35.18%) or cephalexin (169/452, 37.39%) were most commonly used for skin infection. Analysis in the defined population showed <4% (151/4,325) of assessed skin conditions received antimicrobials, and only 6 individuals received more than one course of oral antimicrobial over the year. CONCLUSION: Antimicrobial administration demonstrates high compliance with the protocol. There was very limited use of second-line antimicrobials for recurrent pharyngitis. Most skin infections did not require antimicrobial treatment. Repeated antimicrobials for individuals were rare.

Early-onset neonatal group B streptococcus sepsis following national risk-based prevention guidelines.

BACKGROUND: Neonatal infection with group B streptococcus (GBS) is an important cause of infant mortality. Intrapartum antibiotics reduce early-onset GBS sepsis, but recommendations vary as to whether they should be offered following antenatal screening or based on risk factors alone. We aimed to determine the incidence of early-onset GBS sepsis in New Zealand five years after the publication of national risk-based GBS prevention guidelines. MATERIALS AND METHODS: Prospective surveillance of early-onset GBS sepsis (defined as infection in the first 48 h of life) was undertaken between April 2009 and March 2011 through the auspices of the New Zealand Paediatric Surveillance Unit as part of a survey of infection presenting in the first week of life. RESULTS: There were 29 cases of confirmed early-onset GBS sepsis, including one case of meningitis, giving an incidence rate of 0.23 per 1000 (95% CI 0.16-0.33) live births. Three infants (10.3%) died. In 16 cases (55%), a maternal risk factor qualifying the mother for intrapartum antibiotics was present, but only five (31%) received this intervention. A retrospective review of the major hospital laboratory databases for this period identified two additional cases. A secondary sensitivity analysis taking account of these cases provided an estimated national incidence of 0.26 (95% CI 0.18-0.37) per 1000 live births. CONCLUSIONS: Ten years after a similar survey and five years after promoting a single, risk-based prevention protocol nationally, the incidence of early-onset GBS disease in New Zealand has more than halved, but opportunities remain to further reduce the rate.

Registered nurse assessment and treatment of skin sepsis in New Zealand schools: the development of protocols.

BACKGROUND: Skin infection is the commonest medical cause of hospitalisation in school children. Disadvantaged children, usually Maori or Pacific, have high rates of preventable diseases. AIM: To improve access to early treatment for skin infections using nurse-led school clinics in South Auckland, including provision of antibiotics under delegated standing orders. METHOD: Evidence-based protocols for the recognition and treatment of skin sepsis were developed following a literature search. A training package was developed for health professionals involved and outcome data were collected from a pilot study in which the protocols were trialled. RESULTS: An algorithm for diagnosis of skin infections was adapted from Steer et al (Bull World Health Organ. 2009;87:173-9). Fusidic acid ointment was recommended as first-line treatment for localised impetigo. Twice daily oral cephalexin was recommended for extensive impetigo and cellulitis, for palatability and simplicity of dosing. Fifty-six episodes of skin infection received treatment under standing orders in the first 15 weeks of the pilot study. CONCLUSION: Robust evidence to determine optimal choice, dosage and duration of antibiotic therapy for skin sepsis in children is lacking. The algorithms described are consistent with available evidence and provide a pragmatic approach for use in registered nurse (RN)-led school clinics.

Impact of pneumococcal vaccine on hospital admission with lower respiratory infection in children resident in South Auckland, New Zealand.

AIM: To assess the change in admission rates for all Lower Respiratory Infection (LRI) including pneumonia for children resident in Counties Manukau District Health Board (CMDHB) with the introduction of the Pneumococcal Conjugate Vaccine 7 valent (PCV7) in June 2008. METHOD: National Minimum dataset ICD10 coded LRI admissions to any NZ hospital August 2001-July 2011 for children <2 year resident in CMDHB were analysed using Poisson regression, omitting 1 August 2008 to 31 July 2009, the first-year post vaccine introduction. RESULTS: Pneumonia but not bronchiolitis admissions have been declining since 2001. Pneumonia admissions decreased significantly after PCV7 introduction (incidence risk ratio (IRR) (95% CI) 1.51 (1.08-1.77), additional to the gradual decline since 2001. There was significant decline for Pacific children post PCV7 introduction IRR 1.70(1.39, 2.07) but not for Maori children, IRR 1.05 (0.78-1.40). Maori and Pacific children are at increased risk of admission with LRI compared to European children (relative risk (RR) (95%CI) 4.6 (4.3-5.0) and 5.0(3.7-5.3) respectively) as are those living in Decile 9, 10 compared with those from other deciles, RR 1.43 (1.36-1.50). CONCLUSION: The introduction of PCV7 is associated with reduced admissions for pneumonia in young children yet there has been less impact for Maori in CMDHB.

Prospective population-based study on the burden of disease from post-streptococcal glomerulonephritis of hospitalised children in New Zealand: epidemiology, clinical features and complications.

AIM: A nationwide 24-month study was conducted (2007-2009), via the New Zealand Paediatric Surveillance Unit to define epidemiology and clinical features of acute poststreptococcal glomerulonephritis (APSGN) in children hospitalised with the illness. METHODS: Paediatricians (n = 215) were requested to report new hospitalised cases fulfilling a case definition of definite (haematuria with low C3 and high streptococcal titres or biopsy proven APSGN) or probable (haematuria with low C3 or high streptococcal titres). RESULTS: A total of 176 cases were identified (definite: n = 138, probable: n = 38) with 63% residing in the Auckland metropolitan region. Sixty-seven percent were in the most deprived quintile. Annual incidence (0-14 years) was 9.7/100,000 (Pacific 45.5, Maori 15.7, European/other 2.6 and Asian 2.1/100,000). Annual incidence was highest in the South Auckland Metropolitan region (31/100,000), Central Auckland 14.9, West/North Auckland metropolitan region 5.9 and for the remainder of New Zealand 5.5/100,000. Age-specific incidence was highest in age 5-9 years (15.1/100,000). Reduced serum complement C3, gross haematuria, hypertension, impairment of renal function and heavy proteinuria were present in 93%, 87%, 72%, 67% and 44% of patients, respectively. Severe hypertension was closely associated with either symptoms of an acute encephalopathy or congestive heart failure. CONCLUSIONS: New Zealand children carry a significant disease burden of hospitalised APSGN with socio-economically deprived; Pacific and Maori children are being over-represented. Significant short-term complications were observed in hospitalised children with APSGN. Persistently very low rates in European/other suggest a preventable disease.

Incidence of acute rheumatic fever in New Zealand children and youth.

AIM: To estimate acute rheumatic fever (ARF) incidence rates for New Zealand children and youth by ethnicity, socioeconomic deprivation and region. METHODS: National hospital admissions with a principal diagnosis of ARF (ICD9_AM 390-392; ICD10-AM I00-I02) were obtained from routine statistics and stratified by age, ethnicity, socioeconomic deprivation index (NZDep2006) and District Health Board (DHB). RESULTS: The mean incidence rate for ARF in 2000-2009 peaked at 9 to 12 years of age. Incidence rates for children 5 to 14 years of age for Maori were 40.2 (95% confidence interval 36.8, 43.8), Pacific 81.2 (73.4, 89.6), non-Maori/Pacific 2.1 (1.6, 2.6) and all children 17.2 (16.1, 18.3) per 100 000. Maori and Pacific incidence rates increased by 79% and 73% in 1993-2009, while non-Maori/Pacific rates declined by 71%. Overall rates increased by 59%. In 2000-2009, Maori and Pacific children comprised 30% of children 5-14 years of age but accounted for 95% of new cases. Almost 90% of index cases of ARF were in the highest five deciles of socioeconomic deprivation and 70% were in the most deprived quintile. A child living in the most deprived decile has about one in 150 risk of being admitted to the hospital for ARF by 15 years of age. Ten DHBs containing 76% of the population 5 to 14 years of age accounted for 94% of index cases of ARF. CONCLUSIONS: ARF with its attendant rheumatic heart disease is an increasing public health issue for disadvantaged North Island communities with high concentrations of Maori and/or Pacific families.

Optimising echocardiographic screening for rheumatic heart disease in New Zealand: not all valve disease is rheumatic.

AIMS: Echocardiography detects a greater prevalence of rheumatic heart disease than heart auscultation. Echocardiographic screening for rheumatic heart disease combined with secondary prophylaxis may potentially prevent severe rheumatic heart disease in high-risk populations. We aimed to determine the prevalence of rheumatic heart disease in children from an urban New Zealand population at high risk for acute rheumatic fever. METHODS AND RESULTS: To optimise accurate diagnosis of rheumatic heart disease, we utilised a two-step model. Portable echocardiography was conducted on 1142 predominantly Maori and Pacific children aged 10-13 years. Children with an abnormal screening echocardiogram underwent clinical assessment by a paediatric cardiologist together with hospital-based echocardiography. Rheumatic heart disease was then classified as definite, probable, or possible. Portable echocardiography identified changes suggestive of rheumatic heart disease in 95 (8.3%) of 1142 children, which reduced to 59 (5.2%) after cardiology assessment. The prevalence of definite and probable rheumatic heart disease was 26.0 of 1000, with 95% confidence intervals ranging from 12.6 to 39.4. Portable echocardiography overdiagnosed rheumatic heart disease with physiological valve regurgitation diagnosed in 28 children. A total of 30 children (2.6%) had non-rheumatic cardiac abnormalities, 11 of whom had minor congenital mitral valve anomalies. CONCLUSIONS: We found high rates of undetected rheumatic heart disease in this high-risk population. Rheumatic heart disease screening has resource implications with cardiology evaluation required for accurate diagnosis. Echocardiographic screening for rheumatic heart disease may overdiagnose rheumatic heart disease unless congenital mitral valve anomalies and physiological regurgitation are excluded.

The primary prevention of rheumatic fever.

Rheumatic fever (RF), caused by untreated group A streptococcal (GAS) pharyngitis, is a major cause of morbidity and mortality throughout much of the less developed world and disadvantaged populations (Indigenous and other) in the developed world. Through systematic literature searches, our group has identified potential risk factors for RF and possible interventions for its prevention. The causes can be divided into biological factors, socio-economic, and lifestyle factors and health-care systems and services. Currently, the most promising medical areas look to be improving access to health care and introducing community and school-based sore throat interventions (which aim to diagnose and treat GAS pharyngitis). We could find no convincing support for skin sepsis causing RF. Overall evidence suggests that measures that aim to alleviate poverty and crowding may also reduce the incidence of RF. In comparatively rich countries such as New Zealand and Australia, urgent measures based on available evidence should be undertaken to reduce the very striking health disparity seen with RF and its sequela, rheumatic heart disease in our at-risk populations.

Immunogenicity and tolerability in infants of a New Zealand epidemic strain meningococcal B outer membrane vesicle vaccine.

BACKGROUND: An outer membrane vesicle meningococcal vaccine (MeNZB), was developed for the New Zealand epidemic strain of Neisseria meningitidis B:4:P1.7-2,4. METHODS: A phase II, randomized, observer blind, controlled study evaluating the safety, reactogenicity, and immunogenicity of MeNZB administered with routine New Zealand immunizations at 6 weeks, 3 months, and 5 months of age (n = 375). Group 1 (n = 250) received 25 mug MeNZB and routine immunizations with a fourth MeNZB dose given at 10 months (n = 51). Group 2 (n = 125) received routine immunizations only. Sero-response was a > or =4-fold rise in vaccine strain serum bactericidal antibody titer compared with baseline or a titer of at least 1:8 for baselines <1:4. Reactogenicity was monitored for 7 days after vaccination. RESULTS: Sero-response in Group 1 was achieved in 53% (95% Confidence interval [CI]: 46-59, n = 239) and 69% (95% CI: 54-80, n = 45) with geometric mean antibody titers of 9 (95% CI: 7-10) and 22 (95% CI: 12-39) after the third and fourth doses, respectively. No negative interference by MeNZB on routine immunizations was detected. There were no serious adverse events judged to be vaccine related. CONCLUSIONS: In this group of New Zealand infants, 4 MeNZB doses were required to demonstrate titers comparable with those achieved after 3 doses in older children. MeNZB was safe when used concomitantly with routine New Zealand immunizations to 5 months of age.

Acute rheumatic fever in the Waikato District Health Board region of New Zealand: 1998-2004.

AIM: To outline the epidemiology and clinical pathway of acute rheumatic fever (ARF) cases in the Waikato District Health Board (DHB) region of New Zealand. METHOD: An audit was carried out of the clinical notes of all recognised ARF cases from 1998 to 2004 (inclusive) residing within the Waikato DHB region at diagnosis. Cases were identified by using the hospital admissions coding system and the EpiSurv notification system. The case definition used was the New Zealand criteria for ARF diagnosis, which includes echocardiographic evidence of carditis as a major criteria. RESULTS: A total of 77 ARF cases were found, 8 of which were recurrences. An annual rate of 3.0 per 100,000 initial cases or 3.3 per 100,000 population (initial and recurrent cases) was documented. Over 80% of the total initial ARF cases in the Waikato DHB were in the 5-14 year age group. The overall annual incidence in this age group was 12.9 per 100,000 (age specific incidence for Maori aged 5-14 years 39.6 per 100,000 and for NZ European aged 5-14 years 2 per 100,000). The majority of cases found were Maori (83%), and residing in low socioeconomic status (80% living at the time of diagnosis within the most deprived three deciles according to NZDep01). CONCLUSION: The presence of 77 cases in the Waikato DHB region from 1998-2004 compares unfavourably with other regions, and implies a significant burden from this disease. ARF is a preventable chronic disease with potential life-long sequelae. If the rate of ARF in Maori was reduced to that of non-Maori non-Pacific, then the burden of this disease to New Zealand communities and to the health sector would be virtually eliminated and inequalities improved.

Rapid sterilisation of cerebrospinal fluid in meningococcal meningitis: Implications for treatment duration.

AIM: The aim of this study was to determine the time and total cumulative dose of parenteral antibiotic, required to sterilize the cerebrospinal fluid (CSF) of children presenting with meningococcal meningitis (MM). METHODS: The study was a retrospective audit of children aged 0-14 years who presented between January 1995 and December 1999 with MM. All cases had a delayed lumbar puncture (LP) at least 1 h after commencing antibiotic therapy and demonstrated at least one of the following: (i) a positive CSF culture of Neisseria meningitidis (n = 6); (ii) Gram negative diplococci on Gram stain (n = 16) or (iii) a positive CSF plasma clearance rate test for N. meningitidis (n = 26). RESULTS: Forty-eight children were identified with a mean age of 4.4 years. The cumulative dose of antibiotic prior to LP, ranged from 22 to 440 mg/kg body weight. All cases (n = 24) who received a cumulative dose of at least 150 mg/kg of antibiotic, prior to LP, had a sterile CSF. No CSF taken more than 5 h after commencing antibiotics grew N. meningitidis. CONCLUSIONS: Children in this study with MM had rapid sterilisation of the CSF in less than 6 h. This would support recent recommendations to reduce the duration of antibiotic therapy to 4 days. There is however, lack of long-term data on sequelae with 4 days of treatment.

Rotavirus hospitalisation in New Zealand children under 3 years of age.

OBJECTIVE: To describe the epidemiology of severe rotavirus gastroenteritis and to estimate the hospitalisation rates of this illness in New Zealand children under 3 years of age. METHODS: Children under 3 years of age with acute diarrhoea admitted to 1 of 8 study hospitals between 1 May 1998 and 30 April 2000 were surveyed. Their socio-demographic, treatment and length-of-stay data were recorded and stool samples tested by a rotavirus-specific enzyme-linked immunoassay. National hospital discharge data for infectious diarrhoea (International Classification of Diseases, ninth revision, 003-009) were reviewed, allowing population-based estimates for rotavirus-related hospitalisation in New Zealand. RESULTS: Of 2019 enrolled children, 1138 (56.4%) provided stools for testing, and of these 485 (42.6%) tested rotavirus positive. Rotavirus detection varied significantly by age (26.8% for 0 to 5 months, 42.5% for 6 to 11 months and 52.1% for children aged 12 to 35 months; P < 0.001), and by season (51.2% in winter/spring vs. 24.5% in summer/autumn; P < 0.001). While those infected with rotavirus were more likely to be dehydrated (50.6% vs. 37.4%; P < 0.001), their median hospital stay was similar (1.0 vs. 2.0 days; P = 0.09) to other children with acute gastroenteritis. The estimated national hospitalisation rate for rotavirus diarrhoea in children under 3 years, standardised for age and season, was 634 (95% CI 597, 672) per 100,000. In New Zealand, rotaviruses result in 1 in 52 children being hospitalised by 3 years of age. CONCLUSIONS: Rotavirus diarrhoea is an important, potentially vaccine-preventable cause of hospitalisation in New Zealand children, especially during winter and spring seasons.

Late antenatal carriage of group B Streptococcus by New Zealand women.

OBJECTIVES: To determine in New Zealand women the prevalence of group B Streptococcus (GBS) carriage late in pregnancy and to identify GBS colonisation risk factors, antibiotic susceptibility and serotype distribution. DESIGN: Prospective, observational study. SETTING: Community and hospital antenatal clinics in Wellington and Auckland during 1998-1999. SAMPLE: Convenience sample of 240 women between 35-37 weeks gestation. METHODS: Sociodemographic data, obstetric details and anogenital swabs were collected from each subject. Swabs were inoculated into selective media. GBS isolates underwent serotyping and antibiotic susceptibility testing. RESULTS: Two hundred and forty women (9% Maori, 11% Pacific) aged 15-41 years were recruited. Fifty-two (22%; 95% CI 17, 27) were colonised by GBS. Carriage was independently associated with younger age (59% < or = 30 years; adjusted OR 3.25; 95% CI 1.53, 6.95) and least social deprivation (57% NZ Dep 96 score +/- 3; adjusted OR 1.22; 95% CI 1.06,1.39). All GBS isolates were penicillin-susceptible, but resistance to clindamycin (15%) and erythromycin (7.5%) was detected and associated with serotype V strains. Predominant serotypes were: III (29%), Ia (21%), Ib (20%) and V (20%). CONCLUSIONS: Approximately 20% of New Zealand women carry GBS late in pregnancy, with young age a major risk factor. Increased risk in the socially advantaged, development of resistance to erythromycin and clindamycin, and emergence of new GBS serotypes are findings with important implications for prevention strategies requiring further confirmation.