Invasive meningococcal disease (IMD) has emerged as the most common infectious cause of death in childhood in developed countries.1 Outcome data on survival following IMD (survived or died) are only available for 10% (25/241) of laboratory-confirmed IMD cases notified in Australia during 2011,2 revealing a knowledge gap about deaths and sequelae resulting from IMD in Australian children. The aim of this study was to describe the clinical burden of IMD in South Australian children including clinical features, outcomes and sequelae and determine risk factors for sequelae following IMD.
The medical records of all children admitted to the Women’s and Children’s Hospital (WCH) with a diagnosis of IMD and aged <18 years, between May 2000 and April 2011, were reviewed. To identify cases, hospital separation data were accessed to identify any admissions coded with ICD10–A39.0 to A39.9. This was cross checked with laboratory notifications. Both laboratory-confirmed and clinically suspected (probable) cases were included in the final analysis. Medical records were reviewed to collect information about clinical disease, management, complications and sequelae. Sequelae were defined as any complications related to IMD that were not resolved at hospital discharge or occurred after discharge.
The presence or absence of sequelae was the primary outcome measure for the statistical analysis. Variables tested for association with the outcome measures included socio-demographic and clinical indicators. Univariate associations were reported as prevalence estimates and as ORs with 95% confidence intervals (CI). A multivariate logistic regression model was developed to identify significant and independent predictors of developing sequelae post IMD infection. Modeling commenced with a fully saturated model, which included the above-mentioned covariates with a P = 0.20 or lower on a univariate analysis of association with sequelae development; nonsignificant variables were then removed from the model using methods proposed by Hosmer and Lemeshow.3 All effects were assessed at a 5% alpha level of significance. The analyses were completed using SAS 9.3 (SAS Institute Inc., Cary, NC). The study was approved by the Women’s and Children’s Health Network Human Research Ethics Committee.
A total of 132 IMD cases were identified. Of these, 23 cases were excluded due to the following reasons: IMD contacts only (n = 2), error coding (n = 2), noninvasive meningococcal infection (n = 5), duplicate coding (n = 5); adults (n =2) and readmissions (n = 7). Of 109 eligible cases, 102 (93.6%) were laboratory confirmed.
There were more females (54.1%; n = 59) than males (45.9%; n = 50) with an age range of 22 days to 17 years of age (mean: 3.9 years; median: 2 years) and the majority were Caucasian. One-third of children (29.4%) were infants aged <1 year with half of the infants (51.6%) developing sequelae.
The most common presenting symptom was rash (76.2%, n = 83) which was described as petechial, purpuric or nonblanching. Almost all children (92.7%, n = 101), had pyrexia on presentation or a history of pyrexia. Other symptoms included headache (26.6%, n = 29), drowsiness (26.6%, n = 29), neck stiffness (27.5%, n = 30), joint pain (4.6%, n = 5), photophobia (9.2%, n = 10) and focal cerebral deficit (1.8%, n = 2). Meningococcal septicemia and meningitis occurred in 27.5% (n = 30) of cases, meningitis in 23.9% (n = 26) and septicemia in 48.6% (n = 53) of cases. Many cases were caused by serogroup B (70.6%, 77/109), with 9.2% (10/109) caused by serogroup C and for 20.2% (22/109) the serogroups were unknown (n = 19), W135 (n = 2) and Y (n = 1). Transferred children were more likely to require high dependency unit (HDU) and intensive care unit (ICU) management (72.2%, 39/54) compared with those who were not transferred (41.8%, 23/55; P = 0.002).
Almost one-third (27.5%, n = 30) of patients had been previously vaccinated with a MenC vaccine. Two children with serogroup C disease had previously been vaccinated with a MenC vaccine 2 or 3 years before the IMD infection. Both recovered with a short hospital stay (4 days for both) with no ICU/HDU admission and no sequelae.
Sequelae occurred in 37.6% (n = 41) of 109 IMD patients. A higher proportion of children with serogroup B disease developed sequelae [41.3% (31/75)] than those with serogroup C disease [22.2% (2/9)], although this was not statistically significant (P = 0.280). For 2 cases, outcome was unknown, as children were transferred back to a rural hospital to complete their treatment and were lost to follow up. Inpatient death occurred in 2 patients, both with meningococcal septicemia. The most commonly observed sequelae were skin necrosis/scarring (10.1%, n = 11), neurological problems (10.1%, n = 11) and bone/joint conditions (8.2%, n = 9). Minor sequelae such as chronic lethargy (n = 5) or headaches (n = 5) were reported in 4.5% of IMD children. Children with sequelae were more likely to be managed in ICU/HDU (70.7%, 29/41) than those without sequelae (45.3%, 29/64; P = 0.012).
Prematurity (n = 8) was 1 of the most frequent past medical histories identified in children with IMD. For children aged <1 year, sequelae occurred in all infants with a history of prematurity (100%, 4/4) with admission ages ranging from 4 to 10 months, compared with full-term infants (44.4%, 12/27; χ2 1df = 4.306, P = 0.038).
Hypotension, defined as systolic blood pressure below the fifth percentile for age, occurred in 29 patients (26.6%). Children with hypotension recorded during admission were 2.3 times more likely to develop sequelae than children who were normotensive (P = 0.073).
Diagnosis type, temperature ≥ 39°C at admission and parenteral antibiotics given before admission were independent predictors of development of sequelae (Table 1). Children diagnosed with meningitis and septicemia were more likely to develop sequelae than those diagnosed with septicemia (OR: 15.48; P < 0.001) or meningitis alone (OR: 7.83; P = 0.002). Children with a high temperature at presentation had a 4.5 times higher risk of developing sequelae than those whose body temperature was < 39°C (P = 0.012). Unexpectedly, compared with children who received their first parenteral antibiotics in the hospital setting, parenteral antibiotics administered at the family physician clinic before admission was a risk factor (OR: 11.97; P = 0.007) for development of sequelae. The children who were treated with parenteral antibiotics at the family physician clinic, presented with severe IMD symptoms, such as rash (80.0%, 8/10), fever (100.0%, 10/10), seizure (20.0%, 2/10), irritability (30.0%, 3/10) or confusion (20.0%, 2/10), and most (90.0%, 9/10) required ICU and/or HDU management during hospitalization.
Many cases in our study were associated with serogroup B, with a high overall rate of sequelae, consistent with prior studies in the United States and England.4,5 However, a lower sequelae rate of 9–19% has been observed in the literature.6 Because the WCH is a tertiary pediatric institution, severe cases are more likely to be transferred or referred to the WCH which may lead to bias and a higher sequelae rate than those previously reported in other studies. The risk of developing sequelae is often the highest in the very young compared with older children and adults7 and our study supported this finding. In addition, chronic lethargy and headaches were reported by children in this study and high levels of mental fatigue were noted in Norwegian and United Kingdom studies.8 Sequelae occurred more frequently following serogroup B than serogroup C disease, which is different to the findings in other studies9 but the difference was not significant possibly due to the small sample size. This finding may differ from studies in other countries due to difference in circulating meningococcal subtypes. Transfer from another hospital for treatment was associated with an increased risk of requirement for ICU and HDU management. These findings were also observed in a study in Western Australia.6 It may indicate that patients transferred from peripheral hospitals have more severe disease requiring intensive treatment in a tertiary pediatric hospital or alternatively to a less rapid diagnosis and onset of treatment. Early antibiotic treatment, namely parenteral antibiotic administration before admission, was identified as an unexpected risk factor of developing sequelae. This result should be interpreted with caution due to the small number of children who received antibiotic treatment before admission (n = 10). It is arguable that the severity of disease may account for poor outcomes, as these children presented to the family physician with severe IMD symptoms. A systematic review showing the conflicting results of the effects of early antibiotic treatment in studies in Denmark, United Kingdom and New Zealand, suggests confounding factors and the proportions of cases treated could explain the heterogeneity in findings between studies.10 No studies have previously identified that a body temperature higher than or equal to 39°C was a strong predictor of development of sequelae. This finding may be helpful for triage nurses and clinicians to prioritize patient assessment, management and follow up.
Two cases of serogroup C infection occurred in previously vaccinated children. Vaccine failures may be due to waning immunity, host factors or problems in storage or administration of the vaccine. Several countries (United Kingdom, United States) have implemented an adolescent MenC/MenACWY booster program to avoid the potential for vaccine failure following reduction in antibody titres after 3–5 years.
Our study has the limitations of a retrospective audit study including likely imperfect hospital records and incomplete data due to loss to follow up. In addition, the psychological impact on a child’s life is difficult to evaluate through hospital note review. Larger national studies in Australia are required to more accurately describe the burden of sequelae from IMD.
Information Previously Presented
Portions of the study data shown in this article were previously presented in abstract format at the 13th National Immunisation Conference 2012 on Public Health Association Australia held on June 19–21, 2012 in Darwin, Australia.
The authors wish to acknowledge Novartis Vaccines and Diagnostics Pty Ltd for partially funding this study with a nonrestricted grant. Associate Professor Helen Marshall acknowledges support from the National Health and Medical Research Council of Australia: Career Development Fellowship (1016272).
1. Hart CA, Thomson AP. Meningococcal disease and its management in children. BMJ. 2006;333:685–690
2. Lahra MM, Enriquez RP. Annual report of the Australian Meningococcal Surveillance Programme, 2011. Commun Dis Intell Q Rep. 2012;36:E251–E262
3. Hosmer DW, Lemeshow S Applied Logistic Regression. 20002nd ed New York Wiley
4. Davis KL, Misurski D, Miller J, et al. Cost impact of complications in meningococcal disease: evidence from a United States managed care population. Hum Vaccin. 2011;7:458–465
5. Viner RM, Booy R, Johnson H, et al. Outcomes of invasive meningococcal serogroup B disease in children and adolescents (MOSAIC): a case-control study. Lancet Neurol. 2012;11:774–783
6. Olesch CA, Knight GJ. Invasive meningococcal infection in Western Australia. J Paediatr Child Health. 1999;35:42–48
7. Jones DCartwright K. Epidemiology of meningococcal disease in Europe and the USA. Meningococcal disease. 1995 Chichester John Wiley & Sons:147–57
8. Borg J, Christie D, Coen PG, et al. Outcomes of meningococcal disease in adolescence: prospective, matched-cohort study. Pediatrics. 2009;123:e502–e509
9. Pace D, Pollard AJ. Meningococcal disease: clinical presentation and sequelae
. Vaccine. 2012;30(suppl 2):B3–B9
10. Hahné SJ, Charlett A, Purcell B, et al. Effectiveness of antibiotics given before admission in reducing mortality from meningococcal disease: systematic review. BMJ. 2006;332:1299–1303