Experimental design

Figure 1 shows the experimental design. To compare the eardrop method and intratympanic injection while avoiding potential biases as much as possible, the eardrop method was performed on one ear (eardrop ear) and intratympanic injection on the other ear (intratympanic injection ear) of the same rat (treatment rat). Ears treated with the eardrop method and intratympanic injection were randomly chosen in each rat. To validate our image analysis method, temporal bones without any treatment were harvested from one ear (control ear) randomly chosen from rats that were not treated with eardrop or intratympanic injection (control rat).

Figure 1.

Experimental design. Ears were randomly chosen for the type of treatment (eardrop or ITI). Only one ear in a control rat was randomly chosen as the control ear.

Figure 1 Long description

The diagram illustrates an experimental design involving two rats. The left side shows a 'treatment rat' with two ears labeled as 'Eardrop ear' and 'Intratympanic injection ear', indicating that ears were randomly chosen for each treatment. The right side shows a 'control rat' with one ear labeled as 'Control ear', which was randomly chosen. Arrows point from the rat's ears to the respective labels, emphasizing the random selection process for each ear in the experiment.

Animals

Sprague-Dawley rats were housed in standard polycarbonate rat cages in a temperature- and humidity-controlled room under a 14 hours:10 hours light:dark cycle, with lights on from 07:00 to 21:00 h. Animals were allowed access to water and standard chow ad libitum. Male and female adult breeder rats aged 6–10 months were assigned to this study. Thirty rats were used for the treatment group and another 30 rats for the control group. Adult breeder rats were maintained as male–female pairs, and each pair was sequentially assigned in an alternating manner to either the treatment group or the control group prior to any experimental procedures.

The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional guidelines for the care and use of laboratory animals, as approved by the Johns Hopkins University Animal Care and Use Committee (approval number: RA23M358).

Anaesthesia

Rats were placed in a medium-sized induction chamber for induction of anaesthesia with 2–3 per cent isoflurane. Animals were monitored for the depth of anaesthesia by reflexes to toe pinch and respiratory rate. Rats were maintained under anaesthesia via a nose cone with approximately 1.5–2 per cent isoflurane. Eye lubricant was placed in both eyes to prevent corneal desiccation caused by anaesthesia. Rats were placed on a low-heat-setting heating pad to maintain a normal body temperature (37°C).

Eardrop method: tympanostomy tube insertion, eardrop application and tragal pumping

For eardrop ears, tympanostomy tube insertion, eardrop application and tragal pumping were performed according to our previous publication (Figure 2a,b).Reference Yang, Zhao, Chen and Zhao⁷ Briefly, rats were placed in the lateral position with the eardrop ear upward. The ear was examined to confirm the absence of outer- and middle-ear disease using a microscope and an endoscope. Cerumen in the external auditory canal was removed if necessary. A small incision was made with a myringotomy knife on the posterior half of the pars tensa of the tympanic membrane. A tympanostomy tube (internal diameter, 1.14 mm; Medtronic, Galway, Ireland) was inserted into the tympanic membrane perforation with micro alligator ear forceps. Placement of the tympanostomy tube was confirmed using an endoscope. To detect penetration of the eardrop via the tympanostomy tube, green Davidson^® tissue marking dye (1163-1; Bradley Products, Inc., Bloomington, MN, USA) was diluted twofold with distilled water and used as the eardrop solution. A total of 100 µl of eardrop-mimicking dye solution was applied into the external auditory canal using a 22-gauge catheter connected to a 1-ml syringe under microscopic visualisation. Thereafter, the tragus was gently pushed repeatedly for 1 minute. Thirty minutes after eardrop application, the remaining solution in the external auditory canal was removed using the same-size catheter and syringe, and the tympanostomy tube was then removed with ear forceps under a microscope.

Figure 2.

A, The model section of a rat ear. B, Description of the eardrop method. C, Description of the intratympanic injection.

Figure 2 Long description

A shows a diagram of a rat ear, highlighting the external auditory canal, tympanic membrane, middle-ear cavity and Eustachian tube. B illustrates the eardrop method in three steps: tympanostomy tube insertion, eardrop application using a syringe and catheter and tragal pumping with a finger. C depicts the intratympanic injection process, showing a syringe with a spinal needle administering dye solution into the ear.

Intratympanic injection

Figure 2c shows a diagram of this procedure. Rats were placed in the lateral position with the intratympanic injection ear upward. Using an endoscope, the integrity of the external auditory canal and tympanic membranes was examined. Cerumen in the external auditory canal was removed if necessary. To perform intratympanic injection precisely and consistently, we used an instrument with an endoscope and a 27-gauge spinal needle (Figure 3). To visualise the tip of the needle during the procedure and insert it accurately into the posterior half of the pars tensa of the tympanic membrane, the tip of the needle was positioned 5 mm ahead of the tip of the endoscope. With needle insertion into the tympanic membrane, 100 µl of dye solution was slowly injected into the middle-ear cavity of the intratympanic injection ears. The overall average volume of the middle-ear gas space in Sprague-Dawley rats has been reported to be 42.0 µl.Reference Kania, Herman, Tran Ba Huy and Ar⁸ Therefore, an injection volume of 100 µl was determined to be sufficient for middle-ear delivery, consistent with volumes used in prior intratympanic injection studies in Sprague-Dawley rats.Reference Park, Kim, Han, Park and Park⁹ Thirty minutes after injection, the solution in the external auditory canal was removed using the same method as in the eardrop-ear procedure.

Figure 3.

Our instrument for intratympanic injection. A, The instrument was made with a 1-mL syringe, a 27-gauge spinal needle, an endoscope, and an adapter for a camera. B, The tip of the instrument.

Figure 3 Long description

The image A shows an instrument for intratympanic injection held by a gloved hand. It consists of a one milliliter syringe attached to a 27-gauge spinal needle, an endoscope and a camera adapter. The image B shows a close-up of the instrument's tip, highlighting the needle and endoscope alignment.

Histological analysis

Rats were euthanised immediately after completion of all procedures described above and were under general anaesthesia until euthanasia. After euthanasia, both temporal bones were harvested for histological analysis. Histological and image analyses were performed according to our previous publication.Reference Yang, Zhao, Chen and Zhao⁷ Briefly, after fixation in formaldehyde solution and decalcification, the tissue was processed for embedding. Multiple thin sections were then cut horizontally and stained with haematoxylin and eosin.

To compare solution delivery between the two methods, the area stained by dye in the middle ear on histological sections was analysed using Image J software. Figure 4 shows the representative images of a temporal bone section analysed using Image J software by a scorer blinded to the experimental groups. First, the temporal bone section was cropped to include only the middle-ear cavity (Figure 4a,b). Next, the green-stained area and unstained area were selected and measured using the colour threshold function of Image J software. In this process, the selected area for measurement is shown in red. Figure 4c shows the green-stained area highlighted in red. Residual pale green areas were considered unstained (below the threshold on Image J software) because the green staining is weak. Figure 4d shows the unstained area highlighted in red. The stained area ratio (per cent) was calculated as (stained area)/(stained area + unstained area).

Figure 4.

Representative temporal bone section images during the image analysis. A, An original image. B, The cropped image of the middle ear cavity. C, The green-stained area. D, The white unstained area.

Figure 4 Long description

The image A shows a temporal bone section with labeled anatomical features: external auditory canal, tympanic membrane, middle-ear cavity and Eustachian tube. The image B shows the cropped section focusing on the middle-ear cavity. The image C highlights the green-stained area in red, indicating the stained region. The image D highlights the unstained area in red, showing the region below the staining threshold.

Statistical analyses

Descriptive statistics were presented as numbers (per cent) and means ± standard deviations for categorical variables and continuous variables, respectively. The stained area ratio was compared between the eardrop and intratympanic injection ears within the same rat using a paired t-test. Baseline characteristics between the treatment and control cohorts (age and body weight) were compared using independent samples (Student’s) t-tests. Categorical variables (sex and side of the ear) were compared between cohorts using chi-square tests. A p value less than 0.05 was considered statistically significant. All analyses were performed using SPSS Statistics version 28 (IBM, Armonk, NY, USA).

Animal profiles

Table 1 shows the animal profiles of the eardrop, intratympanic injection and control ears. In the treatment group, 15 males and 15 females were included for the eardrop and intratympanic injection ears. The mean age was 266.5 ± 32.6 days, and the mean body weight was 452.1 ± 89.6 g. The eardrop method was applied to the left ear in 15 rats and to the right ear in the remaining 15 rats. An additional 15 males and 15 females were included in the control group for the control ears. The mean age was 262.4 ± 37.2 days and the mean weight was 453.6 ± 93.7 g. For the control ears, left ears were used in 15 rats and right ears in the remaining 15 rats. There were no significant differences in age (p = 0.649), body weight (p = 0.949) or sex (p = 1.000) between the treatment and control groups.

Table 1.

Profiles of rats in the treatment and control groups

Table 1 Long description

The table compares characteristics of rats in treatment and control groups, each consisting of 30 rats. Both groups have similar mean ages, with treatment group at 266.5 days and control group at 262.4 days, showing no significant difference (p=0.910). Sex distribution is identical, with 50% males in each group (p=1.000). Body weight averages are also similar, with treatment group at 452.1g and control group at 453.6g (p=0.949). The side of the ear treated is evenly distributed, with 50% right ears in both groups (p=1.000). All p values indicate no significant differences between the groups, suggesting comparable profiles.

Treatment group rats were used for eardrop ears and intratympanic injection ears. Control group rats were used for control ears.

* Student’s t-test;

^† chi-square test

Histological analysis

Figure 5 shows representative images of the eardrop, intratympanic injection and control ears. Images of the eardrop and intratympanic injection ears, obtained from the same rat, showed that large regions of the middle-ear cavities were stained (Figure 5d,e). In contrast, no substantial green signal was observed in the control ear, confirming the absence of false-positive signal detection (Figure 5f). Image analysis and data processing revealed that the stained area ratios in the eardrop and intratympanic injection ears were 37.4 ± 19.3 and 36.8 ± 14.8 per cent, respectively. A paired t-test was used to compare the stained area ratio between the eardrop and intratympanic injection ears, as both were derived from the same rat under similar conditions. Figure 6a shows that no significant difference was detected in the stained area ratio of the middle-ear cavity between the eardrop and intratympanic injection ears (p = 0.897; 95 per cent confidence interval, −7.81–8.88), suggesting that the solution delivery efficacy of the eardrop method is comparable to that of intratympanic injection. Figure 6b shows that the stained area ratio in the unstained control ear was 0.15 ± 0.09 per cent, validating the visual inspection.

Figure 5.

Representative images of temporal bone sections in eardrop, ITI, and control ears. A, B, C. Original images. D, E, F. The green-stained areas. G, H, I. The white unstained areas.

Figure 5 Long description

The image consists of three sets of temporal bone sections labeled as eardrop ear, intratympanic injection ear and control ear. Each set includes three images: original images, stained areas highlighted in red and unstained areas highlighted in red. The original images show the anatomical structure of the ears. The stained areas highlighted in red indicate regions where staining occurred, with the eardrop and intratympanic injection ears showing substantial staining compared to the control ear. The unstained areas highlighted in red show regions without staining, with the control ear having more unstained areas compared to the other two. Each section is clearly labeled to differentiate between the types of ears and the staining conditions.

Figure 6.

The mean stained area ratio (large black dots) with standard deviation error bars for each ear. A, Eardrop and ITI ears. The paired ears of individual rats are connected. B, Control ear.

Figure 6 Long description

The image contains two graphs labeled A and B. Graph A shows the stained area ratio in percentage for eardrop ears and intratympanic injection ears, each with a sample size of 30. The data points are connected and the mean values are indicated with large black dots. The p-value is 0.897, suggesting no significant difference between the two methods. Graph B displays the stained area ratio for control ears, also with a sample size of 30, showing minimal staining with clustered data points near the bottom of the graph.