Results

Case Numbers

We reviewed data from 331 cases, excluding 12 cases (6 in nonresidents, 5 that were noninvasive disease, and 1 that lacked sufficient data). The remaining 319 cases of laboratory-confirmed IMD occurred in 318 children. One child had 2 unrelated episodes of IMD that occurred in 2006 and 2017. There were no documented relapses after treatment in the cohort. The average annual incidence of IMD across the study period was 5.9/100,000 population. Incidence rates declined from the tail end of the epidemic in 2004 to a nadir in 2014, then increased to a second peak in 2019 (Figure 1). Overall, we found a trend toward reduced incidence over the study period (Poisson coefficient −0.07 [95% CI −0.14 to −0.01; p<0.01]; rate ratio 0.92 [95% CI 0.90–0.95]). Cases were more common in winter (135/319, 42.3%), followed by spring (87/319, 27.3%), autumn (52/319, 16.3%), and summer (45/319, 14.1%) (p<0.0001).

Enlarge

Figure 1. Timeline of 319 cases of confirmed invasive meningococcal disease in children <15 year of age, by serogroup, reported by year, Auckland, New Zealand, 2004–2020. Numbers along data line indicate exact rates for all cases by year. Men, Neisseria meningitidis serogroup.

Demographic Factors

Median age at time of diagnosis was 18 months (interquartile range [IQR] 7–60 months). The highest average incidence rates were among infants <1 year of age (31.5/100,000 population/year), followed by those 1–4 years of age (8.2/100,000 population/year), 5–9 years of age (2.6/100,000 population/year), and 10–14 years of age (2.0/100,000 population/year) (Table 1). Average incidence rates by ethnic group were highest in Pacific peoples (13.4/100,000 population/year), followed by Māori (11.6/100,000 population/year) and those who were neither Māori or Pacific (1.9/100,000 population/year) (Table 1). Based on census data and compared with non-Māori and non-Pacific groups, the unadjusted relative risk of IMD was 5.9 (95% CI 4.4–8.1) for Māori (p<0.0001) and 6.9 (95% CI 5.1–9.3) for Pacific peoples (p<0.0001). Most children (189/317, 59.6%) lived in NZDep quintile 5 (most deprived 20%) areas. The unadjusted relative risk of IMD for children living in NZDep quintile 5 areas compared with quintile 1 areas was 17.2 (95% CI 9.7–33.2; p<0.0001).

Microbiology and Laboratory Features

Of the 319 cases, we confirmed a microbiological diagnosis by both culture and PCR for 81 (25.4%), on culture alone for 114 (35.7%), and on PCR alone for 124 (38.9%). We compared N. meningitidis culture and PCR from blood and from CSF (Appendix Table 1). Blood culture was negative for 56 (78.9%) of 71 cases in children who received antibiotics before hospital admission (odds ratio 5.1 [95% CI 2.7–9.5]) compared with no prehospital antibiotics (p<0.0001). Of those 56 cases, N. meningitidis blood PCR was positive in all 50 cases tested. CSF analysis was performed in 138 (67.6%) of the 204 cases classified as meningitis (Appendix Table 2). CSF leukocytosis for age was present in 130 (97.7%) of 133 cases of meningitis where a CSF leukocyte count was performed. The serogroup was identified for 301 (94.4%) of the 319 cases: 245 (81.4%) were MenB, 26 (8.6%) MenW, 19 (6.3%) MenC, and 11 (3.7%) serogroup Y. Beginning in 2017, there was an increase in disease caused by MenW, which accounted for 8 (29.6%) of the 27 cases serogrouped in 2019; MenB was the remaining predominant serogroup (Figure 2). The epidemic B:P1.7–2,4 strain accounted for 135 cases (44.9%), waning over time, from 42 cases (14.0/100,000 population) in 2004 to 3 cases (0.90/100,000 population) in 2020 (Poisson coefficient −0.20 [95% CI −0.28 to −0.11]; p<0.01; rate ratio 0.82 [95% CI 0.78–0.85]). The proportion of isolates with reduced penicillin susceptibility increased during the study period. A MIC of >0.06 mg/L was identified in 28 (22.6%) of 124 isolates in 2004–2012 and 47 (66.2%) of 71 isolates in 2013–2020 (p<0.0001). Reduced penicillin susceptibility was identified in 20 (76.9%) of 26 MenW isolates compared with 55 (32.5%) of 169 non-MenW isolates (p = 0.012). All isolates were susceptible to ceftriaxone, ciprofloxacin, and rifampin.

Enlarge

Figure 2. Timeline of 319 cases of confirmed invasive meningococcal disease in children <15 years of age by age group (A) and prioritized ethnicity (B), reported by year, Auckland, New Zealand, 2004–2020.

Clinical Features

The median duration of illness before care was sought was 1 day (IQR 1–3 days). Bacteremia was present in 251 cases (78.7%), meningitis in 204 (63.9%), and septic arthritis in 10 (3.1%) (Table 1). Concomitant bacteremia and meningitis occurred in 141 (44.2%) cases. No cases of chronic meningococcemia were recorded. Sepsis occurred in 172 (80.8%) of the 213 cases with complete systemic inflammatory response syndrome data.

Treatment

Of the 319 cases, 317 (99%) were treated in a hospital; 2 children (0.6%) died before arrival. The median duration of hospitalization was 5 nights (IQR 3–7 nights). Prehospital parenteral antibiotics were administered in 52 (16.3%) cases. In the hospital, empiric antibiotics included a third-generation cephalosporin in 294 (92.7%) cases. There were 303 children who completed a full targeted treatment course of antibiotics, which included a third-generation cephalosporin in 230 cases (75.9%), benzylpenicillin in 67 cases (22.1%), amoxicillin in 8 cases (2.6%), and another antibiotic in 4 cases (1.3%). The median duration of antibiotic treatment was 5 days (IQR 5–7 days). Dexamethasone was administered in 47 (23%) of the 204 meningitis cases. Of the 100 (31.3%) children who were admitted to an ICU (median duration of stay 1 night [IQR 1–3 nights]), 55 received ≥1 life-saving measure: 46 received invasive ventilation, 44 received inotropic/vasopressor support, and 7 received renal replacement therapy. Plastic surgical procedures were performed in 12 (3.8%) cases, orthopedic procedures in 12 (3.8%), and neurosurgical procedures in 6 (1.9%).

Outcomes

Thirteen children died, resulting in a CFR of 4.1% (Table 2). The average death rate over the study period was 0.24/100,000 population/year. Of the 13 children who died (Appendix Table 3), 12 (92.3%) were Māori or Pacific peoples, 11 (84.6%) were living in NZDep quintile 5 areas, and 9 (69.2%) were infants <1 year of age. Ten deaths (76.9%) occurred in the community or within the first 48 hours of hospitalization. Of the 306 survivors, outcome data were complete for 258 cases (84.3%): cure without sequelae occurred in 197 (76.4%) and cure with sequelae 61 (23.6%). We classified outcome as unknown in 48 cases; all had meningitis with no available audiometry results (none had other sequelae identified on follow-up). Documented audiologic assessment occurred in 143 (72.6%) of 197 cases after meningitis. Of the 142 cases with audiometry results available, 32 (22.5%) had sensorineural impairment. Māori children with IMD had 2.5 (95% CI 1.1–6.4) times the odds for death or sequelae compared with non-Māori, non-Pacific peoples (p = 0.0366) (Table 3). Pacific peoples with IMD had 2.9 (95% CI 1.3–7.2) times the odds for death or sequelae compared with non-Māori, non-Pacific peoples (p = 0.0128). Results of univariate comparisons of age, sex, NZDep quintile, season, MeNZB vaccination status, sepsis criteria, serogroup, reduced penicillin susceptibility, and prehospital parenteral antibiotics were not significant.

MeNZB Vaccination

Of the 163 children with complete vaccination records who were eligible for MeNZB, 114 (69.9%) had received ≥1 dose and 64 (39.3%) were fully vaccinated at time of hospital admission. For the 97 eligible children with vaccine subtype IMD, 55 (56.7%) had received ≥1 dose and 31 (32%) were fully vaccinated at time of hospital admission. The mean number of days between the date of last MeNZB vaccine and IMD onset increased with the number of doses received (p<0.00027) (Appendix Table 4).