Probable epidemic Mycoplasma pneumoniae disease activity in metropolitan Sydney, 2015: combining surveillance data to cross-validate signal detection

Mycoplasma pneumoniae is a leading cause of encephalitis and pneumonia in children. Active surveillance identified a cluster of children with suspected encephalitis associated with M.pneumoniae in NSW during July, 2015. An investigation that cross validated encephalitis surveillance with ED pneumonia surveillance and senitenal reference laboratory data revealed probable epidemic M.pneumoniae disease activity in Sydney during 2015.

Page last updated: 08 March 2018

Philip N Britton, Shopna K Bag, Robert Booy, Caroline HW Sharpe, Katherine B Owen, Jiaying Zhao, Melissa J Irwin, Cheryl A Jones

Abstract

Introduction

Mycoplasma pneumoniae is a leading cause of community acquired pneumonia and well-recognised cause of encephalitis in children. Sentinel hospital surveillance identified a temporal cluster of children with suspected encephalitis associated with M.pneumoniae in NSW. We aimed to determine whether this cluster was associated with epidemic M.pneumoniae activity.

Methods

A multi-faceted investigation was undertaken using existing data sources including:

  • Active clinical surveillance of suspected encephalitis (ACE study) associated with M.pneumoniae at the Children’s Hospital at Westmead (CHW).
  • Syndromic surveillance of Emergency Department presentations for pneumonia in children 0-16 years to 86 NSW hospitals.
  • Laboratory sentinel surveillance of M.pneumoniae testing performed at CHW and the Institute for Clinical Pathology and Medical Research (ICPMR), Westmead.

Results

We detected an increased number of cases of hospitalised suspected encephalitis associated with M. pneumoniae in 2015 at the Sydney Childrens Hospital Network (SCHN), with a peak in July, that were predominantly in Western Sydney. Concurrently, monthly pneumonia presentations to 86 NSW emergency departments in children aged 0-16 years increased in 2015. This increase was greater in children aged 5-16 years compared to those aged 0-4 years and in metropolitan sites compared with rural hospitals. Laboratory data from sentinel laboratories showed increased frequency of testing, and of tests returned positive for M.pneumoniae in 2015 compared to preceding years. In aggregate, this information was considered suggestive of epidemic activity of M.pneumoniae in metropolitan Sydney in 2015.

Conclusions

Active surveillance for childhood encephalitis has the potential to provide sentinel surveillance data to identify epidemic infectious disease activity. Combining multiple sources of surveillance data affords opportunities to cross-validate epidemic signals. M.pneumoniae disease activity is challenging to measure, and may be a cause of significant disease burden in Australian children during epidemic years.

Keywords: Mycoplasma pneumoniae , surveillance, encephalitis, pneumonia, children, Australia

Introduction

Mycoplasma pneumoniae infection has well-recognised neurological complications that occur most commonly in children, the most severe form being encephalitis.1, 2 M.pneumoniae most often causes respiratory tract infection and disease including atypical pneumonia, acute otitis media and coryzal illness.3, 4 The importance of M.pneumoniae as a cause of childhood pneumonia was recently emphasised in a large study of hospitalised childhood community-acquired pneumonia (CAP) from the United States. In this study, M.pneumoniae , confirmed by polymerase chain reaction assay (PCR), was the most frequent bacterial cause of childhood pneumonia (20-35%) amongst children aged 5-17 years.5 The epidemiology of M.pneumoniae disease is not well characterised, but it is considered to be a worldwide endemic infectious disease with superimposed epidemics every four to seven years.3, 4 Major epidemics were reported from several European countries, and Israel between 2010 and 2013.6-10

M.pneumoniae has been reported to be one of the most frequently identified pathogens associated with encephalitis in children.11, 12 However, the strength of its causal assocation remains controversial, because it is infrequently identified in cerebro-spinal fluid.1, 13 A further challenge is the lack of available laboratory tests with high sensitivity and specificity.14

Here we report results from a collaborative investigation undertaken following identification of an increased frequency of suspected encephalitis associated with M.pneumoniae by childhood encephalitis surveillance (the Australian Childhood Encephalitis -ACE- study). This increase in cases suggested epidemic M.pneumoniae disease activity to ACE investigators and prompted reporting to Western Sydney Local Health District (WSLHD) Public Health Unit (PHU) followed by a subsequent joint investigation with the PHU and NSW Ministry of Health (NSW MoH) Rapid Surveillance team.

Methods

We sought to determine whether the temporal cluster of M.pneumoniae encephalitis was associated with evidence of epidemic M.pneumoniae activity in NSW by aggregating existing data sources in order to determine if further public health action was required.

1. Data sources

The Australian Childhood Encephalitis (ACE) study:

The ACE study utilises the Paediatric Active Enhanced Disease Surveillance (PAEDS) network to undertake active, sentinel site surveillance for childhood encephalitis at five tertiary paediatric hospitals across Australia. Surveillance commenced in 2013 at the Children’s Hospital at Westmead (CHW), the NSW sentinel site and PAEDS coordinating centre. From 2014, surveillance extended to involve the national PAEDS network.15 The ACE study methodology has been previously published.16 Suspected encephalitis is defined as: a child aged 0 to 14 years AND hospitalised with acute encephalopathy AND has one or more of the following: fever, seizures, focal neurological findings, at least one abnormality of cerebrospinal fluid (CSF): age determined pleocytosis, or elevated protein ≥ 40mg/dl, or EEG/ neuroimaging findings consistent with encephalitis. An association with M.pneumoniae is defined as the presence of specific IgM antibodies in acute sera. All suspected encephalitis cases were reviewed by an expert panel of clinicians and designated as confirmed encephalitis or not encephalitis using published case definitions with higher specificity than the surveillance definition.16-18

Public Health Rapid Emergency Disease and Syndromic Surveillance (PHREDSS) system:

In 2003, NSW MoH implemented a syndromic surveillance system including administrative Emergency Department (ED) data.19 The system, now called the Public Health Rapid Emergency Disease and Syndromic Surveillance (PHREDSS) system, combines near real-time data directly from participating ED patient information management systems with cleaned, more complete Emergency Department Records for Epidemiology (EDRE) data sourced from the NSW Emergency Department Data Collection. The PHREDSS system automatically groups primary provisional ED diagnosis codes, assigned by treating clinicians, into acute illness and injury syndromes for monitoring. The diagnosis codes used include any of the Australian clinical implementations of the International Classification of Diseases (ICD) 9th revision, ICD-10th revision (ICD-10AM) or the Systematized Nomenclature of Medicine - Clinical Terminology (SNOMED-CT).20, 21 Syndromes are monitored daily to detect unusual patterns that may signify an emerging outbreak or issue in the population. The PHREDSS ‘pneumonia’ syndrome includes provisional diagnoses of viral, bacterial, atypical or unspecified pneumonia, SARS and legionnaires disease, but excludes pneumonia with influenza, which is included in the ‘influenza-like illness’ syndrome.

Sentinel laboratory surveillance of M.pneumoniae :

Laboratatory testing for M.pneumoniae is usually undertaken with serology. At the Children’s Hospital at Westmead (CHW), testing is performed for M.pneumoniae IgM using a commercial enzyme linked immunosorbent assay (ELISA:Diesse™ Chorus IgM); a positive is based on the product specified cut-off. At the Institute for Clinical Pathology and Medical Research (ICPMR), Westmead, M.pneumoniae serology is performed using complement fixation assay (CF: Virion\Serion™ reagents) and a commercial Immunofluoresence assay (IFA: Vircell™ IgM). Tests are reported positive where a CF titre of 64 or higher is measured, or IgM is identified based on the product specified cut-off. In addition at ICPMR, PCR for M.pneumoniae nucleic acid is performed as an ‘in house’ assay. CHW refers specimens to ICPMR for this test.

2. Investigation and analyses

ACE study:

ACE study surveillance continued at CHW with enhanced real-time review of cases. In addition, we contacted the Sydney Children’s Hospital, Randwick infectious diseases department to identify additional cases.

Pneumonia syndromic surveillance:

A retrospective analysis of ‘pneumonia’ ED presentations in children aged 0-16 years was conducted. Data from EDRE were used to provide greater coverage of NSW EDs and included 86 NSW facilities that participated continuously from 2010-2015 and had diagnosis complete in 75% or more records. These data represented 85% of all NSW ED presentations in 2015. ‘Pneumonia’ ED presentations in children aged 0-16 years in 2015 were compared to the mean annual count for 2010-2014. The PHREDSS system uses an automated threshold for signalling of five standard deviations (SD) above the expected count to indicate a significant increase. The expected count is the mean count for the same period over the previous five years. We compared ‘pneumonia’ presentations by age group (0-4 years and 5-16 years), and geographical location (metropolitan and rural). The number of admissions and the proportion of presentations that were admitted were analysed. In addition, we sought to sub-categorise the ‘pneumonia’ syndrome to include a subset of ICD-9, ICD-10 and SNOMED-CT diagnosis codes more specific to atypical pneumonia (CS, PB & SB agreed the included codes).

Sentinel Laboratory surveillance:

We requested records of M. pneumoniae serology and PCR nucleic acid detection tests from ICPMR and CHW diagnostic laboratories. We examined number of tests ordered and the proportion that returned positive for 2014 and 2015.

Data were collated in Microsoft (WA, USA) Excel™ v14 (2010) , and statistical testing performed using CDC Epi Info™ (GA, USA). Proportions were compared using two-tailed chi-square test with Yates correction.

This was a public health investigation conducted using provisions in the NSW Public Health Act , 2010 therefore ethical approval was not required. De-identified ED data were released for the purposes of this investigation by the Executive Director, Centre for Epidemiology and Evidence using provisions in the NSW Health Administration Regulation , 2015. The ACE study was approved by the Sydney Children’s Hospitals Network human research ethics committee.

Results

Suspected encephalitis surveillance

Between January and October 2015 (10 months), M.pneumoniae was associated with 29% (13/45) of suspected encephalitis cases identified by the ACE study, including five in July alone (Table 1). Between May 2013 and December 2014 (20 months), the ACE study identified four cases of suspected encephalitis associated with M.pneumoniae at CHW, of 79 total cases (5%; two tailed chi square p-value <0.001 comparing 2015 to 2013/14). Of the 13 cases in 2015, median age was 10.1 years (range 3-11.7); ten were male. Seven cases resided in outer western Sydney. All cases presented with an altered level of consciousness. Other symptoms/signs included: fever (8/13), headache (9/13), focal neurological signs (7/13), seizure(s) (6/13), cerebrospinal fluid pleocytosis (7/11), and abnormal neuroimaging (6/13). Four of the thirteen children were admitted to intensive care. The median length of stay in hospital was 7 days (inter-quartile range 5.5, 15.5). Following review of these patients by the expert panel, two of the thirteen were deemed not to have encephalitis but other neurological syndromes; one with cerebral venous sinus thrombosis, and the other a non-specific seizure episode.

Table 1: Cases of suspected encephalitis* associated with Mycoplasma pneumoniae infection.#^

This table lists the 15 cases of M.pneumoniae suspected encephalitis, showing their sex, date of admission, age, and geographic location of residence (by statistical subdivision) and highlights time (July), place (Western Sydney), and person (age 8-11 years) clustering of cases.
Sex DOA Age (yrs) Geographic Statistical Sub-division (SSD)
Cases identified by PAEDS surveillance at the Children’s Hospital at Westmead
M 29/12/2014 5.7 Sydney-Inner Western SSD
M 15/03/2015 11.7 Sydney-Blacktown SSD
M 17/04/2015 11.5 Sydney -St George/Sutherland SSD
M 22/04/2015 10.8 Sydney-Central Western SSD
F 4/07/2015 5.7 Sydney-Central Western SSD
F 10/07/2015 3.0 Sydney-Blacktown SSD
M 16/07/2015 10.8 Sydney-Blacktown SSD
M 25/07/2015 10.1 Sydney-Blacktown SSD
F 25/07/2015 10.7 Sydney-Blacktown SSD
M 6/08/2015 8.2 Sydney-Blacktown SSD
M 23/09/2015 10.7 Central West-Lachlan SSD
M 30/09/2015 8.5 Sydney-Central Northern SSD
M 25/10/15 8.4 Sydney-Blacktown SSD
Additional cases identified by infectious diseases at Sydney Children’s Hospital, Randwick
M 03/02/2015 9.9 Sydney-St George-Sutherland SSD
M 22/07/2015 10.9 Sydney-Northern Beaches SSD
*Suspected encephalitis: Age 0 to 14 years AND Hospitalised with acute encephalopathy (defined by altered level of consciousness, lethargy and/or personality change lasting ≥24 hours) AND Has one or more of the following: fever, seizures, focal neurological findings, at least one abnormality of cerebrospinal fluid (CSF; age determined pleocytosis, or elevated protein ≥ 40mg/dl), or EEG/ neuroimaging findings consistent with infection-related encephalitis.
#Defined as the presence of positive M.pneumoniae specific IgM antibodies on acute sera.
^In three cases an additional possible infectious cause was identified (one enterovirus PCR positive on stool; one respiratory syncytial virus (RSV) PCR positive on naso-pharyngeal specimen; one RSV and human meta pneumovirus (hMPV) PCR positive on naso-pharyngeal specimen).
Abbreviations: DOA=date of admission; PAEDS=Paediatric Active Enhanced Disease Surveillance network; SSD=statistical sub-division

 

SCH-Randwick reported two cases of M.pneumoniae associated suspected encephalitis hospitalised in 2015; one in February and one in July (Table 1). This was considered to be in keeping with expected case numbers (A Bartlett, personal correspondence).

Pneumonia syndromic surveillance

In 2015, among children aged 0-16 years there were 5,337 ED presentations to 86 NSW hospitals, which was significantly greater (> 5 SD) than the mean number of annual presentations (3,824) during the previous five years (2010-2014). Monthly counts of “pneumonia” presentations were increased from March to September 2015, peaking in August with 663 “pneumonia” presentations (Figure 1).

Figure 1: Monthly counts of pneumonia Emergency Department presentations in children for 2015 (black solid lines), compared with each of the mean of the previous 5 years (coloured dashed lines lines), persons aged 0-16 years, to 86 NSW hospitals in NSW, by age group

Figure 1: Monthly counts of pneumonia Emergency Department presentations in children for for 2015 (solid lines), compared with the mean of the previous 5 years (dashed lines lines), to 86 NSW hospitals, by age group

Text version of the Figure (Text 1 KB)

In 2015 the increase in “pneumonia” ED presentations, compared to the mean number of annual presentations of the previous 5 years, was greater in children aged 5-16 years than in children aged 0-4 years (Figure 1) and in metropolitan hospitals compared to the rural hospitals (Figure 2).

Figure 2: Monthly counts of pneumonia Emergency Department presentations in children for for 2015 (solid lines), compared with the mean of the previous 5 years (dashed lines lines), to 86 NSW hospitals, by location

Figure 2: Monthly counts of pneumonia Emergency Department presentations in children for for 2015 (solid lines), compared with the mean of the previous 5 years (dashed lines lines), to 86 NSW hospitals, by location

*Abbreviation: Metro = Metropolitan
Source (Figures 1-2): Emergency Department Records for Epidemiology (EDRE), held by the NSW Ministry of Health Secure Analyics for Population Health Research and Intelligence (SAPHaRI).

Text version of the Figure (Text 1 KB)

Of the “pneumonia” presentations in 2015, 3,032 were admitted from ED to a ward (not critical care), which was significantly greater than the mean number of annual admissions (2,483) in the previous 5 years. However, the proportion of presentations that were admitted from ED in 2015 (57%) was lower than the mean proportion admitted between 2010 and 2014 (64%).

Of the pneumonia ED presentations in children aged 0-16 years in 2015, 81% had non-specific diagnosis codes assigned (Table 2).

Table 2: Frequency of top 5 diagnosis codes included in the PHREDSS “pneumonia” syndrome in 2015.
This table shows the top fve diagnosis codes contained in the ED “pneumonia” syndrome surveillance and highlights the lack of specificity of the majority (>80%) of clinican assigned diagnoses in this surveillance syndrome.
Diagnosis Code Coding System* Description Frequency Percentage
233604007 SNOMED-CT Pneumonia (disorder) 2421 45%
J18.9 ICD-10AM Diagnosis: Pneumonia, unspecified (Ed.1-Ed.9) 1135 21%
385093006 SNOMED-CT Community acquired pneumonia (disorder) 540 10%
233606009 SNOMED-CT Atypical pneumonia (disorder) 273 5%
53084003 SNOMED-CT Bacterial pneumonia (disorder) 218 4%
*Abbreviations: SNOMED-CT = the Systematized Nomenclature of Medicine - Clinical Terminology; ICD-10AM = the Australian clinical implementations of the International Classification of Diseases ICD-10th revision.

 

Laboratory sentinel surveillance

Analysis of diagnostic laboratory data obtained from CHW showed both an increased number of M.pneumoniae tests ordered and an increase in positive results in 2015 compared to 2014 (Figure 3a). From March to September, there was a significantly higher proportion of positive tests in 2015 compared to 2014 (35% vs 9%; two tailed chi square p-value <0.001). The proportion was similarly higher comparing full years (34% vs 12%; p-value <0.001).

Figure 3a: Mycoplasma pneumoniae IgM testing at the Children’s Hospital at Westmead (CHW) 2014 – 2015.

Figure 3a: Mycoplasma pneumoniae IgM testing at the Children’s Hospital at Westmead (CHW) 2014 – 2015.

Text version of the Figure (Text 1 KB)

Data from ICPMR showed that a greater number of M.pneumoniae serology tests were ordered in 2015 compared to 2014, and there was a peak in positive serology results (CFT or IgM) in August (Figure 3b). The proportion of positive tests (CFT and IgM) was not significantly higher in 2015 compared to 2014 for the whole year (10% vs 7%; p-value <0.2), but showed statistical significance when stratified to a 5-14 years age group (11% vs 3%; p-value <0.001) and approached significance for March to September (5% vs 2%; p-value 0.07). Additionally, a significantly higher proportion of M.pneumoniae PCR tests returned positive in 2015 relative to the preceding two years (Table 3).

Figure 3b: Mycoplasma pneumoniae serology testing performed at the Institute for Clinical Pathology and Medical Research (ICPMR) at Westmead 2014 – 2015.

Figure 3b: Mycoplasma pneumoniae serology testing performed at the Institute for Clinical Pathology and Medical Research (ICPMR) at Westmead 2014 – 2015.

Text version of the Figure (Text 1 KB)

This table shows the number of M.pneumoniae PCR tests performed, and proportion that were positive at ICPMR from 2011 to 2015 and shows the significantly higher proportion of tests that were positive in 2015 compared to 2014 and 2013 (two tailed Chi square testing).
  2015 2014 2013 2012 2011
Total samples tested 617 501 409 401 211
Samples positive (%) 40 (6.5) 9 (1.8) 8 (2.0) 15 (3.7) 12 (5.7)
Two tailed Chi square p-value* <0.001 <0.01 0.08 0.7
*compared with 2015. p-value <0.05 considered as statistically significant

 

Discussion

In 2015, active clinical surveillance for childhood encephalitis showed an increase in suspected encephalitis cases in NSW associated with M.pneumoniae . The cluster peaked in July in one geographic location, the Blacktown area. A similar increase in encephalitis cases was not observed, albeit passively, in Eastern Sydney (at SCHN-Randwick). In 2015, “pneumonia” ED presentations significantly increased in NSW in the age group in which M.pneumoniae is most common5 (i.e. children aged 5-16 years) and in metropolitan locations. Laboratory surveillance showed increased frequency of testing, and tests returned positive for M.pneumoniae in 2015 compared with previous years at two sentinel referral laboratories. Together, this information was suggestive of epidemic M.pneumoniae activity in metropolitan Sydney in 2015. As this information was gathered, it was forwarded on to the WSLHD PHU and the Communicable Diseases Branch, Health Potection NSW. Clinicians (neurology, infectious diseases) at CHW and SCH-Randwick were made aware by email of the increased identification of M.pneumoniae associated encephalitis cases and to consider early testing and treatment. No further public health actions were undertaken.

We reported this cluster of M.pneumoniae associated encephalitis in Western Sydney and probable epidemic M.pneumoniae activity to demonstrate that active surveillance for childhood encephalitis has the potential to provide sentinel surveillance data to identify epidemic infectious disease activity. We have shown that combining surveillance data affords opportunities to cross-validate epidemic signals, and facilitate hypothesis testing. We have also shown that M.pneumoniae is likely an underappreciated cause of hospitalisation and acute morbidity in children in epidemic years.

There are scarce published data describing the epidemiology of M.pneumoniae infection and disease from Australia. A recent molecular epidemiology study showed the circulation of diverse genotypes of M.pneumoniae in Sydney with a low prevalence of genetic markers of antibiotic resistance.22 Furthermore, a clinical study from the Children’s Hospital at Westmead showed M.pneumoniae infection to be most often associated with respiratory tract infection and children aged 5-9 years.23 In this five year study, half of the cases occurred in a single ‘epidemic’ year (2001).23 We could not identify any other contemporary published studies.

Epidemics of M.pneumoniae infection can result in a significant burden of disease. In 2015 there was a considerable increase in ‘pneumonia’ ED presentations amongst school aged children and an increase in admissions. A similar pattern was observed during an epidemic of M.pneumoniae in Scotland in 2010-11 where infection resulted in a high hospitalisation rate (59%) of cases with acute respiratory illness.8 Furthermore, M.pneumoniae encephalitis results in one third to a half of cases suffering neurological morbidity (e.g. motor or cognitive dysfunction or seizures).24-26 Given the significant increase in M.pneumoniae associated encephalitis that we observed in 2015, we are concerned about potential long-term morbidity arising from this epidemic year. We are undertaking a follow-up study that includes many of the cases described here that will further clarify neuro-psycological outcomes.

Monitoring M.pneumoniae disease activity is challenging for several reasons. Firstly, M.pneumoniae causes a variety of clinical syndromes, most commonly respiratory tract infections. Respiratory tract infections, including pneumonia, are among the most common reasons for children to present to primary care practitioners and emergency departments and are caused by a variety of pathogens.5, 27 Furthermore, a Cochrane systematic review showed that M.pneumoniae cannot be reliably distinguished from other causes of pneumonia by clinicians on clinical grounds alone.28 Despite these limitations, given the likely high proportion of pneumonia in school aged children caused by M.pneumoniae, we suggest that pneumonia when restricted to this age group is a potentially useful proxy for M.pneumoniae disease activity. The limitations of the laboratory diagnosis of M.pneumoniae infection preclude its use as a sole mechanism for disease surveillance. Serological diagnosis is hampered by cross-reactivity of IgM assays, limited sensitivity and specificity of single elevated titres of total antibody, and infrequent convalescent sampling required to demonstrate a fourfold rise in titres.14, 29, 30 PCR is also limited by high rates of asymptomatic carriage in children.4 As a result, cross validation of syndromic surveillance with laboratory surveillance, as performed in this outbreak investigation, is important in monitoring M.pneumoniae disease activity. This may still provide conflicting results, as shown in this investigation. In Norway, investigators have studied the possible monitoring of macrolide prescriptions at a population level, cross referenced with laboratory reports.31 Further work is required to identify the best combination of tools for ongoing M.pneumoniae surveillance.

The Public Health response to infectious diseases outbreaks can be categorised into three core activities (i) monitoring and surveillance, (ii) communication and (iii) control of transmission. In this outbreak investigation, the PHU and Rapid Surveillance team undertook additional monitoring and surveillance of available data (PHREDSS) to identify the pattern of M. pneumoniae disease. Communication of the the evolving findings to children’s hospital clinicians was facilitated through SCHN in August, 2015, to alert them to the possible epidemic period. Such alerting of clinicians supports earlier diagnosis and treatment, potentially reducing complications of severe illness, or exclusion of a diagnosis in critically ill patients. No specific community or healthcare transmission control measures were implemented in this situation. Here, our main goal was to cross validate surveillance mechanisms which proved complex and resulted in a delay in confidently identifying the epidemic. Community outbreaks remain largely undetected as M.pneumoniae surveillance is not routinely conducted.32 As a result, this outbreak was a novel consideration for the PHU. These factors and the absence of clear nosocomial or institutional transmission resulted in the level of public health action being minimal. However, we note that M. pneumoniae outbreaks within healthcare facilities have been reported, with high attack rates and significant morbidity despite control measures.14 The optimal public health response to community outbreaks of M. pneumoniae infection is unclear.32 Given the likely community transmission, during future epidemics, additional public health actions for consideration include active communication with primary care practitioners to optimise testing and antimicrobial use, and with schools and child care facilities to emphasise preventive health behaviors. The possible need for direct communication with the public is emphasised by evidence that households may be central to amplifying transmission.32

We acknowledge a number of limitations with this analysis. Firstly, the use of a single positive IgM in the case definition for M.pneumoniae encephalitis lacks specificity; however this definition was unchanged across the surveillance period so should not affect trends. Secondly, we do not have an explanation for why the encephalitis cases were clustered in Western Sydney, despite evidence of an increase in pneumonia cases across metropolitan NSW. It is possible that cases at Sydney Children’s Hospital, Randwick may have been missed given that case ascertainment at this site was based on clinician identification retrospectively rather than on established active surveillance. Furthermore, possible shared exposures were not explored directly with cases. Thirdly, ED provisional diagnoses are allocated at the end of the ED episode of care by a clinician (nurse or doctor) from a drop down list and there may be differences in ED provisional diagnosis coding practices across time and facility. Although we attempted to develop a more specific “atypical pneumonia” syndrome sub-category, the non-specific nature of ED diagnosis coding precluded this being useful. Fourthly, the assumption that up to a third of community acquired pneumonia in children aged 5-19 years is caused by M.pneumoniae is unproven in this study.5 Finally, given the variability in the absolute number of laboratory tests performed at CHW and ICPMR for M.pneumoniae it is likely that testing is selective, and we do not know what additional factors apart from disease activity may have influenced this across the time period from which data were generated.

In conclusion, active, sentinel site surveillance for childhood encephalitis has the capacity to identify epidemic infectious diseases in children. Combining, and cross validating syndromic surveillance signals requires a collaborative approach and the combination of clinical, public health, and biostatistical expertise. We suggest that M.pneumoniae disease burden in Australia is currently inadequately measured, but that it may be considerable and should be the focus of future research.

Acknowledgements

We would like to thank the PAEDS nurses and investigators at CHW: A/Prof Kristine Macartney, Dr Nicholas Wood, Prof Peter McIntyre and Prof Elizabeth Elliot.

We would like to thank Archana Karlekar and Briony Hazleton from CHW; Linda Hueston and Glenys Connor from ICPMR; and Dr Adam Bartlett (SCH-Randwick) all of whom supplied data to this investigation.

We would like to thank the other ACE study investigators at CHW: Prof Russell Dale, Prof Alison Kesson.

The authors would like to acknowledge the Centre for Epidemiology and Evidence, NSW Ministry of Health for providing the EDRE data, the NSW Emergency Departments that contribute data to the PHREDSS system and the NSW Public Health and Biostatistics Training Programs.

Author details

Dr Philip Britton, Staff Specialist Infectious Diseases Physician, The Children’s Hospital at Westmead and Senior Lecturer, Discipline of Child and Adolescent Health, The University of Sydney.1,2,3

Dr Shopna Kumari Bag, Manager-Communicable Diseases, Western Sydney Public Health Unit, Western Sydney Local Health District.4

Professor Robert Booy, Professor and NHMRC Practitioner Fellow, Discipline of Child and Adolescent Health, The University of Sydney.1,2,6

Dr Caroline Sharpe, NSW Public Health Training Program, Centre for Epidemiology and Evidence, NSW Ministry of Health.5

Dr Katherine Bridget Owen, NSW Biostatistics Training Program, Centre for Epidemiology and Evidence, NSW Ministry of Health.5

Dr Jiaying Zhao, Senior Surveillance Officer, Rapid Surveillance, Population Health Intelligence Systems, Centre for Epidemiology and Evidence, NSW Ministry of Health.5

Ms Melissa Jane Irwin, Principal Analyst, Rapid Surveillance, Population Health Intelligence Systems, Centre for Epidemiology and Evidence, NSW Ministry of Health.5

Professor Cheryl Jones, Stevenson Chair of Paediatrics, University of Melbourne and Infectious Diseases Consultant, Royal Children’s Hospital, Melbourne.2,7

Affiliations

  1. Discipline of Child and Adolescent Health, Sydney Medical School, University of Sydney;
  2. Marie Bashir Institute of Infectious Diseases and Biosecurity Institute (MBI), University of Sydney;
  3. Department of Infectious Diseases and Microbiology, Children’s Hospital at Westmead;
  4. Western Sydney Public Health Unit, Western Sydney Local Health District;
  5. Rapid Surveillance, Centre for Epidemiology and Evidence, Population and Public Health Division, NSW Ministry of Health;
  6. National Centre for Immunisation Research and Surveillance;
  7. Department of Paediatrics, University of Melbourne.

Address for correspondence

Dr. Philip Britton

c/o Discipline of Child and Adolescent Health

The Children’s Hospital at Westmead

Locked Bay 4001, Westmead NSW2145

Phone: 0612 9845 3274

E-mail: [email protected]

Funding Source: The ACE study is funded by grants from the Australian Department of Health (Surveillance branch), and the National Health and Medical Research Council (NHMRC) Centre for Research Excellence (CRE) in Critical Infections (1001021); both to CJ and RB; and NHMRC CRE in emerging infectious diseases (1102962) to CJ.
PB is supported by a NHMRC post-graduate scholarship (1074547), the Royal Australasian College of Physicians (RACP) NHMRC Award for Excellence, and Norah Therese Hayes-Ratcliffe/Sydney Medical School Dean’s paediatric infectious diseases fellowship.
Financial Disclosure: The authors have no financial relationships relevant to this article to disclose.
Conflict of Interest: The authors have no conflicts of interest to disclose.
We could not identify any other contemporary published studies.

References

  1. Christie LJ, Honarmand S, Talkington DF, Gavali SS, Preas C, Pan C-Y, et al. Pediatric encephalitis: what is the role of Mycoplasma pneumoniae? Pediatrics . 2007;120(2):305-13.
  2. Bitnun A, Ford-Jones E, Blaser S, Richardson S. Mycoplasma pneumoniae ecephalitis. Seminars in pediatric infectious diseases. 2003;14(2):96-107. Epub 2003/07/26.
  3. Atkinson TP, Waites KB. Mycoplasma pneumoniae Infections in Childhood. Pediatr Infect Dis J. 2014;33(1):92-4.
  4. Meyer Sauteur PM, van Rossum AM, Vink C. Mycoplasma pneumoniae in children: carriage, pathogenesis, and antibiotic resistance. Curr Opin Infect Dis. 2014;27(3):220-7.
  5. Jain S, Williams DJ, Arnold SR, Ampofo K, Bramley AM, Reed C, et al. Community-acquired pneumonia requiring hospitalization among U.S. children. N Engl J Med. 2015;372(9):835-45.
  6. Uldum SA, Bangsborg JM, Gahrn-Hansen B, Ljung R, Molvadgaard M, Fons Petersen R, et al. Epidemic of Mycoplasma pneumoniae infection in Denmark, 2010 and 2011. Euro Surveill. 2012;17(5):02.
  7. Blystad H, Anestad G, Vestrheim DF, Madsen S, Ronning K. Increased incidence of Mycoplasma pneumoniae infection in Norway 2011. Euro Surveill. 2012;17(5):02.
  8. Ferguson GD, Gadsby NJ, Henderson SS, Hardie A, Kalima P, Morris AC, et al. Clinical outcomes and macrolide resistance in Mycoplasma pneumoniae infection in Scotland, UK. Journal of medical microbiology . 2013;62(Pt 12):1876-82.
  9. Linde A, Ternhag A, Torner A, Claesson B. Antibiotic prescriptions and laboratory-confirmed cases of Mycoplasma pneumoniae during the epidemic in Sweden in 2011. Euro Surveill. 2012;17(6).
  10. Nir-Paz R, Abutbul A, Moses AE, Block C, Hidalgo-Grass C. Ongoing epidemic of Mycoplasma pneumoniae infection in Jerusalem, Israel, 2010 to 2012. Euro Surveill. 2012;17(8).
  11. Glaser CA, Honarmand S, Anderson LJ, Schnurr DP, Forghani B, Cossen CK, et al. Beyond viruses: clinical profiles and etiologies associated with encephalitis. Clin Infect Dis. 2006;43(12):1565-77.
  12. Pillai SC, Hacohen Y, Tantsis E, Prelog K, Merheb V, Kesson A, et al. Infectious and autoantibody-associated encephalitis: clinical features and long-term outcome. Pediatrics. 2015;135(4):e974-84.
  13. Domenech C, Leveque N, Lina B, Najioullah F, Floret D. Role of Mycoplasma pneumoniae in pediatric encephalitis. Eur J Clin Microbiol Infect Dis. 2009;28(1):91-4.
  14. Waites KB, Talkington DF. Mycoplasma pneumoniae and its role as a human pathogen. Clinical microbiology reviews. 2004;17(4):697-728, table of contents.
  15. Zurynski Y, McIntyre P, Booy R, Elliott EJ, Group PI. Paediatric active enhanced disease surveillance: a new surveillance system for Australia. J Paediatr Child Health. 2013;49(7):588-94.
  16. Britton PN, Dale RC, Elliott E, Festa M, Macartney K, Booy R, et al. Pilot surveillance for childhood encephalitis in Australia using the Paediatric Active Enhanced Disease Surveillance (PAEDS) network. Epidemiol Infect. 2016:1-11.
  17. Sejvar JJ, Kohl KS, Bilynsky R, Blumberg D, Cvetkovich T, Galama J, et al. Encephalitis, myelitis, and acute disseminated encephalomyelitis (ADEM): case definitions and guidelines for collection, analysis, and presentation of immunization safety data. Vaccine. 2007;25(31):5771-92. Epub 2007/06/16.
  18. Venkatesan A, Tunkel AR, Bloch KC, Lauring AS, Sejvar J, Bitnun A, et al. Case Definitions, Diagnostic Algorithms, and Priorities in Encephalitis: Consensus Statement of the International Encephalitis Consortium. Clin Infect Dis. 2013. Epub 2013/07/19.
  19. Muscatello DJ, Churches T, Kaldor J, Zheng W, Chiu C, Correll P, et al. An automated, broad-based, near real-time public health surveillance system using presentations to hospital Emergency Departments in New South Wales, Australia. BMC public health . 2005;5:141.
  20. Classifications, International Classification of Diseases (ICD). In: Organization WH , editor. 2016.
  21. Organisation IHTSD. SNOMED CT: The global language of healthcare. 2016.
  22. Xue G, Wang Q, Yan C, Jeoffreys N, Wang L, Li S, et al. Molecular characterizations of PCR-positive Mycoplasma pneumoniae specimens collected from Australia and China. J Clin Microbiol. 2014;52(5):1478-82.
  23. Othman N, Isaacs D, Kesson A. Mycoplasma pneumoniae infections in Australian children. J Paediatr Child Health. 2005;41(12):671-6.
  24. Al-Zaidy SA, MacGregor D, Mahant S, Richardson SE, Bitnun A. Neurological Complications of PCR-Proven M. pneumoniae Infections in Children: Prodromal Illness Duration May Reflect Pathogenetic Mechanism. Clin Infect Dis. 2015;61(7):1092-8.
  25. Kammer J, Ziesing S, Davila LA, Bultmann E, Illsinger S, Das AM, et al. Neurological Manifestations of Mycoplasma pneumoniae Infection in Hospitalized Children and Their Long-Term Follow-Up. Neuropediatrics. 2016.
  26. Daxboeck F, Blacky A, Seidl R, Krause R, Assadian O. Diagnosis, treatment, and prognosis of Mycoplasma pneumoniae childhood encephalitis: systematic review of 58 cases. J Child Neurol. 2004;19(11):865-71.
  27. Acworth J, Babl F, Borland M, Ngo P, Krieser D, Schutz J, et al. Patterns of presentation to the Australian and New Zealand Paediatric Emergency Research Network. Emergency medicine Australasia : EMA. 2009;21(1):59-66.
  28. Wang K, Gill P, Perera R, Thomson A, Mant D, Harnden A. Clinical symptoms and signs for the diagnosis of Mycoplasma pneumoniae in children and adolescents with community-acquired pneumonia. Cochrane Database of Systematic Reviews. 2012;10:CD009175.
  29. Thurman KA, Walter ND, Schwartz SB, Mitchell SL, Dillon MT, Baughman AL, et al. Comparison of laboratory diagnostic procedures for detection of Mycoplasma pneumoniae in community outbreaks. Clin Infect Dis. 2009;48(9):1244-9.
  30. Waites KB. What’s new in diagnostic testing and treatment approaches for Mycoplasma pneumoniae infections in children? Advances in Experimental Medicine & Biology. 2011;719:47-57.
  31. Blix HS, Vestrheim DF, Hjellvik V, Skaare D, Christensen A, Steinbakk M. Antibiotic prescriptions and cycles of Mycoplasma pneumoniae infections in Norway: can a nationwide prescription register be used for surveillance? Epidemiol Infect. 2015;143(9):1884-92.
  32. Walter ND, Grant GB, Bandy U, Alexander NE, Winchell JM, Jordan HT, et al. Community outbreak of Mycoplasma pneumoniae infection: school-based cluster of neurologic disease associated with household transmission of respiratory illness. J Infect Dis. 2008;198(9):1365-74.