Biofilm Formation In Cochlear Implants With Cochlear Drug Delivery Channels In An

1
Biofilm Formation In Cochlear Implants With
Cochlear Drug Delivery Channels In An In Vitro
Model Trey A. Johnson, BS1, Kimberly A. Loeffler,
BS1, Robert A. Burne, PhD2, Claude N. Jolly,
PhD3, Patrick J. Antonelli, MD1 From the
Departments of Otolaryngology1and Oral Biology2,
University of Florida, Gainesville, Florida and
Med-El, Innsbruck, Austria.3 Supported in part
by a grant from Med-El Medical Electronics.
Abstract
Results
All samples had biofilms resistant to 24-hour
treatment with oxacillin, 100 ?g/ml (MIC 1
?g/ml) (Figure 4). Examination of external
surfaces of the septae demonstrated immature
biofilm growth (Figure 5). The condition of the
model septae was highly significant (p lt 0.0001).
Quantitative bacterial counts were significantly
higher in models with septae penetrated by a 19g
coring needle than in models with a 30g
non-coring needle remaining in the septum (p
0.0003), 30g non-coring needle removed (p lt
0.0001), and models with non-penetrated septae (p
lt 0.0001). Bacterial counts were not
significantly higher in models with a 30g
non-coring needle remaining in the septum than in
models penetrated by a 30g non-coring needle with
needle removal (p 0.7016) or models with
non-penetrated septae (p0.7963) (Figure 4).
Bacteria could be seen in the needle tract of the
septae penetrated with the 19g coring needle and
septae where a 30g non-coring needle was left in
place during bacterial challenge. No bacteria
were seen in the needle tract of septae
penetrated with a 30g non-coring needle, with
removal of the needle prior to bacterial
challenge. The internal surface of septae and
inner channels revealed relatively robust biofilm
formation only on septae penetrated with a 19g
coring needle (Figures 6 and 7). Septae
penetrated with 30g non-coring needles showed
sparse bacteria and no biofilm growth.
Background Cochlear drug delivery may improve
preservation of residual hearing with cochlear
implant (CI) placement in candidates for
concomitant acoustic and electrical stimulation.
Cochlear therapy may require a drug delivery (DD)
channel. CI DD may promote sequestration of
bacterial contaminants, biofilm formation, and
suppurative complications. The aim of this study
was to evaluate the impact of DD port
considerations on CI biofilm formation.
Methods Sterilized silastic models were
constructed to represent CIs with a DD channel,
with in intact port, a widely opened port, a
non-coring needle penetrating the port, and a
non-coring needle removed from the port. CIs
were suspended in culture media, inoculated with
a biofilm forming strain of Staphylococcus
aureus, incubated for 96 hours, then exposed to
oxacillin for 24 hours to kill non-biofilm
bacteria. Biofilms were dispersed with
ultrasonification and quantitative bacterial
counts were performed. Scanning electron
microscopy was performed to evaluate bacterial
colony architecture. Results Bacterials counts
were higher in CIs with widely open ports than
CIs with ports containing a non-coring needle,
CIs with ports that had been previously
penetrated by a non-coring needle and CIs with
intact ports (p 0.0003). Conclusions The CI
DD channel port may significantly impact CI
biofilm formation. CI biofilm formation may be
minimized by delivering drugs with fine,
non-coring needles and limiting the duration of
port penetration. Clinical Significance CI
design and treatment considerations may impact
the risk of suppurative complications with CIs.

Introduction
Cochlear implants (CIs) may be exposed to
bacteria either during surgery or as a result of
subsequent otitis media. Infections involving
CIs are uncommon with reported rates of 1 4.
These infections may be refractory to treatment
with systemic antibiotics, often necessitating CI
removal imparting significant morbidity and
expense. CI infections refractory to systemic
antibiotic therapy have been found to be due to
formation of biofilms. Biofilms are complex
microbial ecosystems that may develop after
bacterial or candidal binding to a surface and
secretion of an exopolymeric matrix, which
protects microbes from host immune responses and
antimicrobial therapy. Given the risk of
bacterial contamination cochlear implants should
be designed to minimize the risk of bacterial
biofilm formation. Cochlear drug delivery
(DD) is being investigated as an adjunct to CI
surgery. Intracochlear DD of neurotrophins may
promote the outgrowth of axons to more precisely
interface with the electrodes. Cochlear DD of
agents that block hair cell apoptosis may improve
hearing preservation results with hybrid
electroacoustic CIs. Such agents may be
delivered into the cochlea through a channel that
extends down the electrode array in a single
bolus or a continuous microinfusion pump (Figure
1). Incorporation of a DD channel into a CI
could create a nidus for sequestration of
microorganisms, and biofilm formation. The
purpose of this study was to examine the impact
of different DD channel port considerations on
the rate of biofilm formation in an in vitro
model.
1 2
3 4
Figure 2. (Left) Experimental conditions of
cochlear implant models. (1) non-penetrated port,
(2) port penetrated once using a 30 gauge
non-coring needle mimicking a single bolus drug
infusion, (3) port penetrated with a 30
non-coring gauge needle which remained in the
septum mimicking continuous drug infusion, (4)
port penetrated by a 19 gauge coring needle.
Figure 3. (Right) Labeled diagram of
experimental samples including placement of glass
bead to allow for free floating suspension.

Figure 4. Quantitative bacterial counts,
measured in total colony forming units, for each
experimental condition sample set

Figure 5. (Left) Scanning electron microscopy of
external surfaces of the experimental septae.
Figure 6. (Below left) Scanning electron
microscopy of inner surfaces of the experimental
silicone septae. Figure 7. (Below right)
Scanning electron microscopy of the inner channel
surfaces
Materials and Methods
Silastic CI models were constructed from silastic
tubing to represent CIs with a DD channel and a
penetrable, self-sealing experimental septum or
port to seal the channel. The non-experimental
end of the sample was sealed using silicone glue.
The DD channel port was subjected to one of four
conditions (1) left intact, (2) penetrated
transiently with a 30 gauge non-coring needle,
(3) penetrated with a 30 gauge non-coring needle
left in the septum, (4) or transiently penetrated
with a 19 gauge large bore, coring needle
(Figures 2 and 3). A glass bead was attached to
the non-septum end of each sample before
sterilization with gamma irradiation. CI models
were suspended in 120 ml specimen containers,
bathed in culture media (Tryptic soy broth)
inoculated with a biofilm forming strain of
Staphylococcus aureus, and incubated for 96 hours
(Figure 3). All samples used for quantitative
study were then exposed to oxacillin, at 100
times the minimal inhibitory concentration, for
24 hours to kill planktonic (ie, non-biofilm)
bacteria. Quantitative microbiology was performed
on nine samples for each condition. Biofilms were
dispersed using ultrasonication and quantitative
bacterial counts were performed using standard
microbiological techniques. Scanning electron
microscopy was performed on two samples for each
condition to evaluate biofilm architecture.
(1) a non-penetrated port, (2) a port penetrated
with a 30 gauge needle mimicking a single bolus
drug infusion, (3) a port penetrated with a 30
gauge needle which remained in the septum
mimicking continuous drug infusion, and (4) a
port penetrated by a 19 gauge needle
Conclusions
CI DD channels may increase the risk of bacterial
biofilm development unless protected with an
impermeable septum in an in vitro model.
Further, in vivo study would be necessary to
establish the relative risk of biofilm formation
with different CI DD port conditions.
Figure 1. Cochlear implant with Drug Delivery
Channel shown with and without microinfusion
needle in place.