Effects of inner ear trauma on the risk of pneumococcal meningitis

Abstract

Objectives

To examine the risk of pneumococcal meningitis in healthy rats that received either a severe surgical trauma to the modiolus and osseous spiral lamina or the standard insertion technique for acute cochlear implantation.

Designs

Interventional animal studies.

Subjects

54 otologically normal, adult, Hooded Wistar rats.

Interventions

54 rats (18 which received a cochleostomy alone, 18 which received a cochleostomy and acute cochlear implantation using standard surgical techniques, and 18 which received a cochleostomy followed by severe inner ear trauma) were infected four weeks following surgery with S. pneumoniae via three different routes (hematogenous, middle ear and inner ear) to represent all potential routes of bacterial infection from the upper respiratory tract to the meninges in cochlear implant recipients with meningitis.

Results

Severe trauma to the osseous spiral lamina and modiolus increased the risk of pneumococcal meningitis when the bacteria were given via the middle or inner ear (Fisher’s exact test P<0.05). However, the risk of meningitis did not change when the bacteria were given via the hematogenous route. Acute electrode insertion did not alter the risk of subsequent pneumococcal meningitis for any route of infection.

Conclusion

Severe inner ear surgical trauma to the osseous spiral lamina and modiolus can increase the risk of pneumococcal meningitis. Therefore every effort should be made to ensure that cochlear implant design and insertion technique cause minimal trauma to the bony structures of the inner ear in order to reduce the risk of pneumococcal meningitis.

Introduction

Since 2002, there has been an increase in the number of reported cases of pneumococcal meningitis in patients with a cochlear implant^1,2. Many of the implant recipients had pre-existing risk factors for pneumococcal meningitis^3,4. Based on available clinical data, it is difficult to determine if the presence of a cochlear implant increases the risk of meningitis, caused by Streptococcus pneumoniae, in subjects with no pre-existing risk factors for the disease. However, animal studies have shown that a minimal threshold of S. pneumoniae is required to induce meningitis in healthy rats⁵ and that this threshold is significantly reduced in the presence of a cochlear implant⁶. Although cochlear implants may increase the risk of pneumococcal meningitis⁶, the incidence implant-related meningitis is still very low. The benefits of the cochlear implant in subjects with profound deafness far outweigh the risk of infection. Nevertheless, every effort should be made to minimise the infection risk in implant recipients.

The mechanism of cochlear implant-related pneumococcal meningitis is unclear. The surgical technique and insertion trauma to the inner ear structures are proposed to be possible mechanisms¹. Further study is required to investigate whether trauma to the osseous spiral lamina (OSL) and/or the modiolus, without the presence of an implant, can increase the risk of central nervous system (CNS) infection. Using the threshold principle developed in our previous study, we examined whether the presence of severe inner ear surgical trauma alters the risk of pneumococcal meningitis in rats. As the bacteria which cause meningitis can reach the CNS from the upper respiratory tract mucosa by different routes (hematogenous or via the inner ear) ^5–7, we examined if the threshold for infection was reduced in rats with inner ear surgical trauma for three different routes of infection: hematogenous, middle and inner ear.

Methods

Source of the animals

All the experimental animals were bred and housed in the animal house in the Department of Otolaryngology, University of Melbourne. Animals were free of endogenous pathogens, including S. pneumoniae. All procedures and animal handling were conducted in accordance with guidelines set by the Animal Research & Ethics Committee of the Royal Victorian Eye and Ear Hospital and “The Australian code of practice for the care and use of animals for scientific purposes” from the Australian National Health and Medical Research Council, 2004.

In total, 54 otologically normal, adult, Hooded Wistar rats (10 to 16 weeks old), weighing between 100 and 400 g, were randomly divided into three groups before receiving surgical intervention to the left ear 4 weeks prior to inoculation with S. pneumoniae. The first group of 18 rats (group A) had a cochleostomy performed on the left inner ear. The second group of 18 rats (group B) received a cochlear implant via a cochleostomy of the left inner ear, followed by immediate removal of the implant. The third group of 18 rats (group C) received severe surgical trauma of the modiolus and osseous spiral lamina in the left inner ear.

The rats from each group (A, B and C) were further subdivided into 3 groups of 6 to study the effect of inner ear trauma on three different routes of bacterial infection (middle ear, inner ear and intra-peritoneal (IP); Table 1). The dose of S. pneumoniae chosen for each route of inoculation was based on our previous studies which showed that bacterial inocula of this size did not induce meningitis in healthy rats with a cochleostomy, but did cause meningitis in rats with a cochlear implant ^5,6.

Table 1.

Effect of inner ear trauma on the frequency of meningitis

	Frequency of meningitis in rats subjected to:
Route of inoculation and number of colony-forming units of S. pneumoniae administered	Cochleostomy only (group A)	Electrode insertion (group B)*	Severe inner ear trauma (group C)**
Intra-peritoneal 4 x 106	0/6†	0/6ns	0/6ns
Middle ear 3 x 104	0/6	2/6ns	5/6‡
Inner ear 1 x 103	0/6	0/6ns	6/6‡

^*The implant was removed immediately after insertion and the cochleostomy was sealed with fascia.

^**The OSL and modiolus were traumatized by inserting a straight micro needle (KAEN2171, Kaisers, WA, Australia) into scala tympani via cochleostomy.

^†No. of rats with meningitis/no. inoculated. Rats were monitored for 5 days after inoculation; meningitis was diagnosed clinically and confirmed by histological examination.

^ns The effect of surgical intervention on the risk of meningitis was not significantly greater than in rats which underwent cochleostomy only (P >0.05 Fisher's exact test, one-tailed)

^‡The effects of surgical intervention on the risk of meningitis was significantly greater than in rats which underwent cochleostomy only (P <0.05 Fisher's exact test, one-tailed ).

Surgical anesthesia

Rats were anesthetized with an IP injection of a mixture of 8 mg/kg xylazine and 75 mg/kg ketamine hydrochloride. A local anesthetic agent (0.1 ml of lignocaine hydrochloride with 0.0182 mg/ml of adrenalin tartrate, Troy Laboratories, NSW, Australia) was injected subcutaneously (SC) around the surgical incision. The animals were then placed on a heated pad maintained at 37°C throughout the surgery. The animals were given 0.03–0.05 mg/kg SC buprenorphine for analgesia immediately after surgery. They were assessed continually over 48 hours for signs of post-operative pain and discomfort, and buprenorphine was given 8 to 12 hourly if there were signs of post-operative pain or discomfort. Animals received 10 ml/kg of normal saline SC for fluid replacement during the post-operative recovery period.

Animal surgery: Scala tympani electrode array design and surgical technique

The dummy scala tympani electrode used in this study has been described previously⁷. In brief, 4 mm of polyimide tubing (Cole-Parmer^®, Illinois, USA) with an outer diameter of 0.10 mm was coated with a layer of silicone (Dow Corning Medical Grade Elastomer MDX4-4210, Factor II, AZ, USA) to a diameter of 0.15 mm. The dummy electrodes were cleaned with absolute alcohol in an ultrasonic cleaner then rinsed with MilliQ water and bathed in MilliQ water for 10 minutes before drying, packaging and sterilizing using H₂O₂ sterilization (STERRAD^® 100S).

Fifty-four adult, Hooded Wistar rats received a cochleostomy to the left inner ear^5,7. The bulla and round window niche were exposed and the stapedial artery was cauterized with Zencor MF1 bipolar coagulator (Zencor, Sydney, Australia) and a cochleostomy was performed just below the round window niche (RWN) and at the location of the previously cauterized stapedial artery⁸. For the control group of 18 rats, the cochleostomy was sealed immediately with fascia. The second group of 18 rats received surgical insertion of the cochlear implant as described in our previous studies^6,7. Bone dust and blood were carefully cleared away from the cochleostomy before placement of the electrode array, which was inserted 2–3 mm into the scala tympani. The electrode array was then immediately removed from the inner ear and the cochleostomy was sealed with temporalis fascia. In the final group of 18 rats, the left inner ear structure was physically traumatized by inserting a straight Micro Needle (KAEN2171, Kaisers, Perth, WA, Australia) into the scala tympani via the cochleostomy to fracture the OSL and modiolus. The cochleostomy was sealed with temporalis fascia following the fracture of inner ear bony structures. The degree of trauma to the cochleae was subsequently confirmed by histological examination.

After the surgical procedure, all animals were given two doses of prophylactic enrofloxacin 10 mg/kg SC diluted 1:1 with saline. One dose was given immediately after surgery and the second dose, 12 hours later.

Procedures for infection

Four weeks after surgery, all 54 animals were inoculated with Streptococcus pneumoniae 447A which carries type 2 capsular antigen and was originally isolated from the CSF of a child with meningitis. No antibiotics were given after infection.

The meningitis animal model using S. pneumoniae 447A has been established in our previous work^5–7. The preparation of the bacterial inoculum has also been described in detail in our previous work^5–7. The bacterial inoculum for each route of infection is shown in Table 1 and was derived from our previous studies^5,6. Viable counts of the inoculum confirmed that each rat received the number of bacteria indicated in Table 1.

Intraperitoneal inoculation

Eighteen rats (6 each from groups A, B, and C) were anesthetized with a gas mixture of isoflurane and oxygen, and inoculated IP with a 1-ml suspension of 4 x 10⁶ colony-forming units (CFU) of S. pneumoniae 447A⁵ (Table 1).

Middle ear inoculation

Under general anesthesia, the left bulla of 18 rats (6 each from groups A, B and C) was surgically exposed for direct inoculation of 10 μl containing 3 x 10⁴ CFU of S. pneumoniae 447A⁵ (Table 1). To retain the micro-organisms in the bulla, the cavity was first filled with Gelfoam^® (Pharmacia & Upjohn, Michigan, USA). After the inoculation of the bacteria, the opening of the bulla was covered with temporalis fascia and the wound was sutured in two layers.

Inner ear inoculation

Under general anesthesia, the left bulla of 18 rats (6 from each group) was surgically exposed and a cochleostomy was performed with a straight Kirschner wire to access the scala tympani⁵. Two microliters of perilymph were removed and 1 μl of a bacterial suspension containing 1 x 10³ CFU of S. pneumoniae 447A was slowly inoculated into the scala tympani⁵. The cochleostomy was then covered with temporalis fascia and the wound was sutured in two layers.

Post infection monitoring

Following the inoculation, each animal was examined at least twice daily for clinical signs of meningitis for 5 days. The clinical assessment was recorded on a 12 point scored monitoring sheet⁷. Animals were euthanized if one of the following conditions was met: a score of 11 or above; a weight loss of greater than 25%; or a score of 5–10 with rectal temperature of greater than 41°C. Animals without clinical evidence of meningitis were euthanized at the end of the fifth day.

Microbiological specimen collection and tissue preparation

Once early signs of meningitis were evident, rats were deeply anesthetized with isoflurane and oxygen to allow collection of CSF, middle ear fluid and blood for microscopy and culture. The methods used to collect CSF, middle ear fluid and blood have been described previously^5–7.

The animals were then given a lethal dose of pentobarbitone sodium, 120 mg/kg of body weight, intramuscularly (IM) (Lethabarb^®, Virbac Pty. Ltd. NSW, Australia) and were transcardially perfused with 0.9% saline, then 10% neutral buffered formalin (NBF), pH 7.4 at 4°C. The brain, meninges and cochleae were harvested and placed in 10% NBF for further processing.

Fifty-four brains with meninges were harvested and stored in 10% NBF for 48 hours then embedded in paraffin. The specimens were sectioned 10 μm thick, stained with hematoxylin and eosin (H&E) or Gram’s stain and examined by light microscopy for presence of inflammation and bacteria.

Nine pairs of randomly selected cochleae were harvested from the temporal bones and fixed in 10% NBF. They were decalcified in a solution of 10% ethylene diamine tetra-acetic acid in 0.1M phosphate buffer (pH 7.4), orientated in the mid-modiolar position, and then embedded in Spurr’s resin. Two sets of twenty-one 2-μm sections were collected at 126-μm intervals throughout the cochlea. One set of 21 sections was stained with H &E and the other set was stained with Gram’s stain.

CSF, blood culture and middle ear cultures were collected as an adjunct to the histology of the CNS to determine the presence of the bacteria. The serotype of S. pneumoniae isolated from the cultures was determined using commercial typing sera (Statens Serum Institute, Denmark) to determine if the strain causing the disease was the same as that used for the initial inoculum.

Histological analysis

The brain and meninges were examined for the presence of an inflammatory cell response within the subarachnoid space and brain tissue, thickening and hyperplasia of the meningeal cells, and bacteria within the subarachnoid space and brain tissue. The cochleae were examined for the extent of trauma to the OSL and modiolus and for the presence of bacteria and inflammatory cells.

Statistical analysis

The effects of inserting a cochlear implant and surgically induced, severe intra-cochlear trauma in rats on the risk of developing pneumococcal meningitis for three different routes of bacterial inoculation were evaluated statistically using Fisher’s exact test (one- tailed). A P value of <0.05 was considered to be significant.

Results

In total, 13 of 54 rats developed meningitis (Table 1). They appeared tired, lethargic, unresponsive to sound and light, had a hunched body posture, poor grooming, weight loss and a rectal temperature above 38°C. When these signs developed, the histology of the brain consistently showed evidence of meningitis with inflammatory cells and Gram-positive diplococci within the subarachnoid space. The correlation between the clinical signs of meningitis and histopathological evidence of meningitis was established in our previous work^5–7.

Two of the 18 rats in the cohort which underwent insertion and immediate removal of the scala tympani electrode array (group B) acquired meningitis between110 and 112 hours after middle ear inoculation. Of the cohort of 18 rats subjected to severe left intra-cochlear trauma (group C), 11 (5 inoculated via the middle ear and 6 inoculated directly into the inner ear) developed meningitis 64 to 120 hours after inoculation.

Compared to the control cohort (rats with cochleostomy only), rats subjected to implantation and immediate withdrawal of the scala tympani electrode array from the inner ear exhibited no significant increase in the incidence of meningitis for any of the three routes of inoculation (P>0.05, Fisher's exact test, one-tailed). However, the attack rate of meningitis was significantly raised in rats with a severe intra-cochlear trauma when bacteria were given via the middle ear (P=0.008) or inner ear (P=0.001). Severe intra-cochlear trauma did not increase the risk of pneumococcal meningitis when the bacteria were administered IP (P >0.05).

Microbiology

Culture results of the CSF, blood and middle ear samples are summarized in Table 2. Examination of the pneumococci isolated from these samples showed them to be the same serotype as that used for the initial inoculum.

Table 2.

Results of microbiological and pathological examination of rats with cochlear implantation and severe intra-cochlear trauma after inoculation with S. pneumoniae via three different routes.

Routes of inoculation and procedure	Blood culture	CSF culture	Middle ear swab culture	Meningitis confirmed histologically
Intraperitoneal:
Cochleostomy only (group A)	2*	0	0	0
Electrode insertion (group B)	2	0	0	0
Severe inner ear trauma (group C)	3	1	0	0
Middle ear :
Cochleostomy only (group A)	2	0	6	0
Electrode insertion (group B)	5	3	5	2
Severe inner ear trauma (group C)	6	4	6	5
Inner ear :
Cochleostomy only (group A)	1	0	3	0
Electrode insertion (group B)	0	0	0	0
Severe inner ear trauma (group C)	5	5	6	6

^*No. of rats (of 6 inoculated with S. pneumoniae) with S. pneumoniae cultured at necropsy. Serotyping of the isolated bacteria showed them to be the same as that used for inoculation.

Cochlear histology

There was minimal damage to the inner ear structure of the left cochlea of rats where the intervention was either a cochleostomy or short-term implantation of the electrode array (Figure 1). In two cochleae of rats which underwent electrode insertion only, absence of the organ of Corti in the basal turn was observed but this did not increase the risk of cochlear or CNS infection. By contrast, extensive fractures of the OSL and modiolus were observed in rats subjected to the procedure aiming to traumatize the inner ear architecture (Figure 2).

Figure 1.

Lower power H & E photomicrographs illustrating representative cochlear histology in acute implanted (immediate withdraw of the electrode array following insertion) non-meningitic animals. Left (a) and right (b) cochleae 120 hours following IP inoculation of 4 x 10⁶ CFU S. pneumoniae. No inflammatory cells and bacteria were evident within the cochleae. The section of the cochlea at the level of cochleostomy (co) for acute electrode insertion and fascial plug (fas) is illustrated in (a). Left (c) and right (d) cochleae 120 hours after middle ear inoculation of 3 x 10⁴ CFU S. pneumoniae. Similarly, no inflammatory cells and bacteria were evident within the cochleae. Left (e) and right (f) cochleae 120 hours following direct inner ear inoculation of 1 x 10³ CFU S. pneumoniae into the left cochlea. Inflammatory cell response (ir) was evident in the basal turn of the ipsilateral cochlea. There were no inflammatory cells or bacteria in more apical region of the cochlea and not in the contralateral cochlea. There were no evidence of trauma to the OSL and the modiolus from acute insertion of a scala tympani electrode array. However, an absence of organ of Corti was observed in the basal turn in the ipsilateral acutely implanted cochleae (a and e). The loss of organ of Corti did not increase the risk of meningitis in these two rats. Scale bar: (a–f) 200 μm.

Figure 2.

Lower power H & E photomicrographs illustrating the ipsilateral (a–c) cochleae of 3 rats that received a severe inner ear trauma and 120 hours following IP inoculation of 4 x 10⁶ CFU S. pneumoniae. These animals did not exhibit any clinical and histological (CNS) evidence of meningitis. The scalae of cochleae were devoid of gross infection or inflammation. Extensive fractures (arrow heads) to the OSL and modiolus were evident in the photomicrographs. There were no osteogenesis and very minimal fibrosis were observed within the cochleae as a result of a severe inner ear trauma. (c) a section of ipsilateral cochlea taken at the level of basal turn at the level of cochleostomy. The location of cochleostomy (co) to allow surgically inflicted fractures to the OSL and modiolus and the fascial plug (fas) can be seen in (c). Scale bar: (a–c) 200 μm.

The pattern and the distribution of bacteria and inflammatory cells within the cochleae of rats with clinical and histological evidence of meningitis were consistent with those in our previous studies^5,7. In rats with meningitis following middle or inner ear inoculation, the cochlea ipsilateral to the inoculation contained bacteria and inflammatory cells within the scala tympani, scala vestibuli, Rosenthal’s canal, the canaliculi perforantes (the pores within the wall of the OSL, adjacent to the scala tympani), and the peri-neural and peri-vascular spaces within the modiolus and internal acoustic meatus (IAM) (Figure 3). Far fewer bacteria and inflammatory cells were observed within the ipsilateral scala media. The contralateral cochlea also exhibited bacteria and inflammatory cells within the IAM, modiolus, scala tympani and scala vestibuli. However, less severe labyrinthitis was observed in these cochleae compared to the ipsilateral side (Figure 3). In the scala tympani of the contralateral ear, bacteria and inflammatory cells were prominent in the basal turn, close to the OSL; there were also more inflammatory cells in the scala tympani than in the scala vestibuli.

Figure 3.

Lower power H&E photographs illustrating a severely traumatized cochlea [fractures of OSL and modiolus (arrow heads)] (a) and contralateral control (b) cochleae of a rat 72 hours following middle ear inoculation of 3 x 10⁴ CFU S. pneumoniae. This animal exhibited clinical and histological (CNS) evidence of meningitis. In this example there is severe labyrinthitis throughout the ipsilateral cochlea (a) while the contralateral cochlea exhibits evidence of infection predominately localized to the scala tympani (arrows)(b). Higher power micrograph of H & E stain of the fractured modiolus (c) and OSL (d) of the ipsilateral cochlea showing extensive inflammatory cells and bacterial infiltration within the fibrous tissue between the bony fragments. Arrow heads: fracture sites. bn: bone; ic: inflammatory cells; fib : fibrous tissue. * histology process artifact. Scale bar: (a) & (b) 200 μm ; (c) & (d) 100 μm.

In the presence of severe trauma to the OSL and modiolus, numerous bacteria were present within the neural elements at the fracture sites. When the basic architecture of modiolus and OSL were severely damaged, the perilymphatic spaces of the scala tympani and vestibuli were in direct contact with the neural elements within the modiolus and IAM. In these cochleae, there was no intervening fibrosis or osteogenesis between the bony fragments. Gram-positive cocci were observed within the perilymphatic space of the scalae and within the peri-neural and peri-vascular spaces of the modiolus and IAM. In a small number of traumatized cochleae, extensive fibrosis and osteogenesis were seen within the scalae. Despite this, numerous bacteria and inflammatory cells were seen within the fibrous tissue between the new and fractured bones as well as in the modiolus and IAM (Figures 4 and 5).

Figure 4.

Lower power H&E photographs illustrating a severely traumatized cochlea (fractures of OSL and modiolus) (a) and contralateral control (b) cochlea of a rat 64 hours following inner ear inoculation of 1 x 10³ CFU S. pneumoniae. This animal exhibited clinical and histological (CNS) evidence of meningitis. In this example there is extensive osteogenesis and fibrosis of the traumatized cochlea especially at the upper cochlear turns. A severe labyrinthitis was also observed throughout the ipsilateral cochlea (a) while the contralateral cochlea exhibits a less severe labyrinthitis with the infection predominately localized to the scala tympani (arrows)(b). Higher power micrograph of H & E stain of the ossified scalae (c,d) of the ipsilateral cochlea illustrates extensive inflammatory cells and bacterial infiltration within the fibrous tissue between the newly formed bony fragments. ic: inflammatory cells; os: osteogenesis of new bone fragments. The approximate location of the higher power micrographs (c,d) are illustrated in (a). Scale bar: (a) & (b) 200 μm; (c) & (d) 100 μm.

Figure 5.

Lower power H&E photographs illustrating another example of a rat with a severe traumatized ipsilateral cochlea (fractures of OSL and modiolus) (a) 60 hours following inner ear inoculation of 1 x 10³ CFU S. pneumoniae. This animal also exhibited clinical and histological (CNS) evidence of meningitis. In this example there is also an extensive osteogenesis and fibrosis of the traumatized cochlea especially at the upper cochlear turns. A severe labyrinthitis was also observed throughout the ipsilateral cochlea (a). Higher power micrograph of H & E stain of the ossified and fibrosed scalae (b,c) of the ipsilateral cochlea illustrates extensive inflammatory cells and bacterial infiltration within the fibrous tissue between the newly formed bony fragments. ic: inflammatory cells; fib: fibrous tissue; os: osteogenesis of new bone fragments. The approximate location of the higher power micrographs (b,c) are illustrated in (a). Scale bar: (a) 200 μm ; (c) & (d) 100 μm.

In rats without meningitis, the histological appearance of the cochleae was normal in animals inoculated IP. Small numbers of inflammatory cells and scant serofibrinous exudate were seen in the basal turn of the cochleae of rats inoculated via the middle or inner ear.

Macroscopic examination of the ipsilateral middle ear mucosa of the round window niche revealed evidence of middle ear inflammation in rats inoculated directly into the middle ear. The contralateral control bullae showed no evidence of middle ear inflammation. There were also no inflammatory changes in the middle ear mucosa of rats inoculated IP directly into the inner ear.

Discussion

This study has demonstrated that the presence of severe inner ear trauma increases the risk of pneumococcal meningitis when bacteria were inoculated directly into the middle or the inner ear of the ipsilateral side. However, severe inner ear trauma to the OSL and modiolus did not increase the risk of pneumococcal meningitis when the bacteria were given IP.

We have previously demonstrated that a minimum threshold of bacteria is required to induced meningitis in healthy rats and that this threshold differs according to the route of bacterial inoculation⁵. Furthermore, the presence of a cochlear implant significantly reduces the threshold of bacteria required to induce meningitis for three different routes of inoculation routes (IP [hematogenous, or directly into the middle or inner ear)⁶. In this study, we showed that severe inner ear trauma as a result of surgery also reduced the threshold of bacteria required to cause meningitis via direct routes of infection, but only for direct routes of infection from the middle or inner ear. The likely explanation is that severe trauma to the OSL and modiolus creates a direct communication route between the inner ear and the subarachnoid space but does not alter the pathway for the bacteria to reach the meninges via the hematogenous route. This and our previous studies^5,7 have shown that bacteria can reach the CNS via peri-neural and peri-vascular spaces within the modiolus and IAM. In normal cochleae, the bacteria were found to traverse the canaliculi perforantes to Rosenthal's canal and from there to the CNS via the peri-neural and peri-vascular spaces within the modiolus and IAM⁵. In the present study, there were more bacteria localized in neural structures at the fracture site compared to the cochleae with minimal or no trauma to the OSL.

On the other hand, inner ear trauma can lead to fibrosis and osteogenesis in the long term^9–13 and this may obliterate the direct opening to the CNS created by the initial surgery. Under these circumstances, the threshold of infection may be similar to that in control animals to the level of control animals with no inner ear trauma. Although fibrosis and osteogenesis were observed in a number of traumatized cochleae in this study, bacteria and inflammatory cells were found within the fibrous tissue between the bony fragments and in the neural elements within the modiolus and IAM. This is consistent with previous studies of human temporal bone where inflammatory cells were observed to infiltrate the mature fibrous tract of the fractured temporal bones following meningitis^14–16, and the observation that patients with fibrous union following a base of skull fracture still have a higher risk of pneumococcal meningitis in the long term¹⁶.

Previous animal studies have also demonstrated that bacteria and inflammatory cells can infiltrate the well-formed peri-implant fibrous seal and fibro-muscular wall of stapedial artery in the event of a severe pneumococcal infection^6,7. Even when there is significant fibrous tissue repair within the inner ear, it is still unclear whether the fibrous tissue can resist pneumococcal infection. As no bacteria were observed within the newly formed bony tissues in this study, ossification (without intervening fibrosis) may resist direct spread of bacteria from the middle ear to the CNS. It is interesting to note that complete ossification of the cochlea may follow meningitis and it is plausible that this may protect against meningitis via the direct otologic route.

In traumatized cochleae where there was minimal fibrosis and osteogenesis, it is possible that the damage to the inner ear structure was too extensive for complete healing to occur, or that a period of 4 weeks was insufficient for the rats used for this study to mount sufficient fibrous and bony tissue re-growth to resist direct spread of infection from the inner ear to the CNS.

This study has demonstrated that insertion implantation of a scala tympani electrode array per se using the standard insertion technique in rats does not alter the threshold for pneumococcal meningitis by any of the three different routes of infection examined. Histological examination of cochleae indicated that damage to the membranous labyrinth, OSL and modiolus in these animals was minimal, and insufficient to alter the threshold required for CNS infection. The absence of the organ of Corti in the basal turns was observed in two cochleae and this may have been due to insertion of the electrode array. Nevertheless, loss of the organ of Corti in these animals did not increase their risk of developing meningitis.

Our study also provides novel insights into the possible mechanisms underlying the development of pneumococcal meningitis in patients with different cochlear implant designs. There is a greater incidence of meningitis in patients implanted with electrode arrays containing a positioner¹. In a recent clinical study, children receiving an implant with a positioner had 4.5 times the risk of developing meningitis compared with those who received other types of implant ⁴. It has been postulated that a two-part electrode system may increase the likelihood of trauma to the osseous spiral lamina and/or modiolus¹⁷, thus allowing bacteria direct access to the subarachnoid space once they have entered the inner ear. Our data support the hypothesis that an extensive fracture to the OSL and modiolus as a result of inner ear surgery may allow S. pneumoniae easier access to the peri-neural space and peri-vascular space within Rosenthal’s canal and the modiolus. In a recent study, severe insertion trauma, including fracture of the OSL was observed when an implant device with a positioner was fully inserted into the cochlea¹⁸. This injury appeared to occur primarily because the device was too large to permit full insertion into the scala tympani in 70% of temporal bones examined¹⁸.

It is important to point out that this study compared the effect of severe surgical trauma within the inner ear with that of standard insertion of a single component electrode array on the risk of pneumococcal meningitis. The effect of a moderate degree of trauma to the inner ear structures on the risk of pneumococcal meningitis was not examined because of the difficulty involved in creating inner ear trauma of varying severity consistently in our animal model. However, the principle that severe trauma can lead to a higher incidence of meningitis is clearly demonstrated for the direct route of infection, i.e., from the middle ear to the CNS via the inner ear.

Apart from risk of infection, inner ear trauma has further implications for the survival of neural structures within the inner ear. Human temporal bone studies¹⁹ and animal studies^20–23 have revealed the loss of spiral ganglion cells in regions of intra-cochlear damage following cochlear implantation. Although there is no clear correlation between speech discrimination performance and spiral ganglion cell counts obtained post mortem ²⁴, the consequences of the loss of spiral ganglion neurons loss through cochlear implant trauma warrants further research. Preventing loss of spiral ganglion cells may be especially important in young children, in whom lifetime use of the cochlear implant is anticipated. As the indications for cochlear implantation expand to include individuals with some residual hearing, it is also important to reduce inner ear trauma to prevent damage to the residual hair cells. For these reasons, it is important to minimize the extent of insertion trauma within the cochlea to reduce the risk of meningitis and improve survival of hair cells and spiral ganglion neurons.

Conclusion

Severe trauma to the OSL and modiolus increased the risk of pneumococcal meningitis in healthy rats when S. pneumoniae was introduced directly into the middle or inner ear four weeks after the trauma. However, the same degree of surgical trauma to the bony structure of the inner ear did not alter the risk of pneumococcal meningitis when the bacteria were given via the hematogenous route. The standard insertion technique of a cochlear implant in healthy rats produced minimal trauma to the membranous OSL and modiolus and did not alter the risk of pneumococcal meningitis.