Parazzini M, Sibella F, Paglialonga A, Ravazzani P. Assessment of the exposure to WLAN frequencies of a head model with a cochlear implant. Bioelectromagnetics. Aug 3, 2010. Ahead of print.
Wireless Local Area Network (WLAN) technology is a form of short-range communication between an access point and many personal devices/computers. WLANs use radiofrequency (RF) frequencies of 2.4, 5.2 and 5.8 GHz which could be of health concern because of the exposure duration is often long. Currently, only a few studies looked at the exposure to WLAN, and assessed exclusively exposure assessments. Very few studies have investigated the interactions between new EMF-emitting technologies and cochlear implants (CI). CI are auditory prostheses which can bring back partial hearing in profoundly deaf adults and infants.
The objective of the research was to estimate internal RF distribution in an anatomical model of a head, with a cochlea that includes an electrode array of a CI, exposed to WLAN frequencies of 2.4, 5.2 and 5.8 GHz in far-field.
The human model and the electrode array model were previously developed and used in the exposure assessment study. The exposure source was modelled as a uniform plane wave at frequencies of 2.4, 5.2 and 5.8 GHz on the side of the head where the electrode array was inserted. The power density of the plane wave was adjusted at the ICNIRP limit for general public exposure at these frequencies (10W/m2). The simulation time was 2 hours to 7 hours and 20 minutes, depending on the characteristics of the geometrical and electromagnetic model used for the computational analysis. For each WLAN frequency, simulations were conducted with and without the electrode array, with a reference using the array. The data was analyzed by evaluating the RF distribution inside the cochlea and near the cochlear implant to compare the amplitude distribution of the fields for the simulation (with and without the implant at each WLNA frequency).
Results show that CI increases both the mean (range of ratio with Implant/without CI: 1.31–2) of the E amplitude distribution near the electrode array for all frequencies and polarizations. The change of mean (range of ratio with CI/without CI: 0.93–1.13) of the E amplitude distribution was less evident at the cochlear surface far away from the electrode array as compared to the electrode itself. Data show that the CI had a tendency to increase the mean value of the point SAR up to a factor of about 4 (5800 MHz, vertical polarization). Data also show small insignificant differences in maximum SAR10 mg values in the cochlea between simulations with and without the cochlear implant. For all scenarios, the maximum SAR10 g values were well below the ICNIRP local basic restrictions for general public exposure (max SAR10 g = 2 W/kg).
EMF distribution in a person wearing an implant is different because of dielectric properties of the implants in comparison to human tissues. These possible EMF interactions have to be evaluated as a primary and new safety issue. One hypothesis is that the area possibly affected by EMF exposure is localized close to the stimulating interface between the electrode array and the auditory fibres within the cochlea. The results of this study confirm this as it showed larger variations in the E, and marginal in the H, amplitude distributions on the electrode array surface. The estimation of maximum SAR10 g values in the whole head indicate that the CI complies with ICNIRP RF safety limits for the general public, even if simulation exposure conditions were higher than a true WLAN exposures. It must be noted that compliance with the ICNIRP limits does not necessarily eliminate any functional influence of the RF fields on the stimulating currents produced by the implants.
The conclusion is that insertion of a CI causes only moderate localized differences in the E, H and point SAR distribution when evaluated close to the electrode array in the cochlea. The effects were found to be minimal at the high frequencies (5.2 and 5.8 GHz) compare with the lower frequency (2.4 GHz).