The epidemiological interpretation of the available data was complicated by several predisposing factors for meningitis in the cochlear implant population. First, certain congenital inner ear malformations, such as the Mondini malformation, are associated with cerebrospinal fluid (CSF) fistulae and recurrent episodes of meningitis, even without cochlear implantation or other otologic surgery . Second, meningitis is the primary etiology of deafness for a significant number of adult and pediatric implant recipients, and it is well established that the risk for contracting meningitis is greater for an individual who has a prior history of this serious infection . Third, there has been a strong trend toward implanting children at an earlier age, based on data showing improved long-term functional outcomes . By the year 2000, one manufacturer had received FDA approval for use in children as young as 12 months. The incidence of meningitis is very high in this very young age group ; thus, age was considered as another possible nondevice-related factor relevant to the observed increase in meningitis reports. Finally, the positioner component for the Advanced Bionics Corporation device was introduced into the US market in 1999, and this design modification could have been a predisposing factor causing the sudden increase in meningitis cases.
Recognizing the complexity in conducting and interpreting an epidemiological study of this problem, FDA embarked on a collaborative investigation with the Centers for Disease Control and Prevention (CDC), 36 state health departments, and the local health departments of Chicago, New York City, and Washington, DC. Due to the urgent public health nature of the issue, a retrospective cohort study design was used to determine the incidence of bacterial meningitis among a study population of children receiving cochlear implants during the preceding 5.5 years who were less than 6 years old at the time of implant surgery . Using the warranty lists from the three manufacturers, a study population of 4264 children was defined. A variety of meningitis case-ascertainment methods were used for the study. Nineteen meningitis cases were identified within the study population from reports to the company, the MAUDE database, and the surveillance systems of the CDC, state and local health departments. An additional seven cases were identified by a survey that was mailed out to all families on the warranty lists, providing a total of 26 subjects for the study.
The incidence of Streptococcus pneumoniae meningitis in this population was determined to be 138.2 cases per 100 000 patient-years, which was more than 30 times the incidence reported in that age group for the general US population (data from the CDC Active Bacterial Core Surveillance Program) . Unfortunately, the study was not designed to determine the incidence of meningitis in an unimplanted cohort of deaf children. Deaf children may have a higher incidence of meningitis, even without a cochlear implant. No data on this incidence are available in the published literature.
A nested case-control study was also conducted to assess risk factors for meningitis in this cohort. Twenty-six meningitis cases were compared to a randomly selected control population of200 children from the cohort who did not develop meningitis. Information on risk factors for the two groups was obtained through parental interviews and from abstraction of medical records from cochlear implant surgeries, meningitis hospitalizations, and primary care providers.
Postimplant meningitis was strongly associated with the use of the positioner component (OR = 4.5,95% CI = 1.3-17.9) and the incidence of meningitis in patients with the positioner remained elevated for at least 24 months from the time of implantation compared to patients implanted without the positioner. Radiographic evidence of an inner ear malformation associated with a CSF leak was another strong risk factor for the development of meningitis (adjusted OR = 9.3, 95% CI = 1.2-94.5). The ability to assess the protective effects of pneumococcal vaccination in the case-control study was limited because few of the children had received the vaccine. The vaccine was not licensed in the USA until 2000, more than half-way through the study period.
Of note, this focused epidemiological study was designed, conducted, analyzed, and published in just over 1 year from FDA's initial notification of the 11 meningitis cases by Advanced Bionics. This accomplishment reflects the effective, coordinated working relationships among FDA, CDC, state and local health departments, cochlear implant manufacturers and the clinical community.
Was this article helpful?