Otitic hydrocephalus is a rare intracranial complication of suppurative otitis media and associated mastoiditis (1). It is defined as raised intracranial pressure (ICP) secondary to the middle ear or mastoid disease without radiologic evidence of ventriculomegaly (1–4). These patients can present with papilledema (optic nerve head edema secondary to raised ICP), which can lead to permanent vision loss (3, 5). Increased ICP is thought to occur because of thrombosis of the dural venous sinuses and/or internal jugular veins, leading to a rise in cerebral venous pressure and subsequent increase in central serous fluid pressure (1). It has also been proposed that there is an inflammatory component to all cerebral venous sinus thromboses (CVST) that may affect systemic and visual outcomes (6). Previous literature suggests that 15%–22% of patients with mastoiditis develop intracranial complications despite medical treatment (7, 8).

Diagnosis and management involves a multidisciplinary team, including infectious disease, otolaryngology, neurosurgery, neuroradiology, neurology, ophthalmology, and hematology (2, 3, 9, 10). Medical treatment often includes anticoagulation, antibiotics, and acetazolamide (3, 11). Surgical interventions can include otologic surgeries such as mastoidectomy and myringotomy tubes to address the primary source of infection (7, 12). Furthermore, ICP-reducing neurosurgical procedures such as ventriculoperitoneal (VP) shunt and endovascular thrombectomy may be required to be done to prevent vision loss (13).

In CVST, two studies showed that there was significant visual morbidity in both children and adults (4, 14). Both studies found that papilledema commonly progressed after diagnosis and the visual morbidity presented as an optic atrophy in children and quantifiable visual field defects in adults. The visual outcomes following otitic hydrocephalus have not been widely reported, but one recent study reported some initial visual outcomes on a small cohort (15). In that cohort, individuals required further neurosurgery because of a progression of papilledema despite the fact that the initiation of corticosteroids had an overall positive impact on vision (15).

The present study aims to provide a detailed evaluation of interventions taken for patients presenting with otitic hydrocephalus, and their efficacy and visual outcomes.

This was a retrospective study of patients diagnosed with otitic hydrocephalus secondary to mastoiditis at a quaternary academic hospital between July 2022 and July 2024. The institutional review board approved the research protocol with a waiver of consent, given the retrospective nature of this study. The methods complied with Health Insurance Portability and Accountability Act regulations and adhered to the tenets set forth by the Declaration of Helsinki.

Pediatric patients were identified by ICD 10 codes and then a manual review was performed to confirm the diagnosis. Patients <18 years were included. Data were collected on presentation and demographics, ophthalmology examination, neurologic imaging, and medical/surgical management. Best-corrected visual acuity (BCVA), color vision, degree of papilledema as defined by Frisen grade (16), presence of optic disc atrophy, and oculomotor palsies were measured at initial and final visits by a qualified ophthalmologist. Ganglion cell layer (GCL) volume, retinal nerve fiber layer (RNFL) thickness, and retinal findings were recorded using optical coherence tomography (OCT) (Heidelberg) at initial and final visits. OCT testing is a well-established, non-invasive evaluation that can be performed in the office on awake and compliant patients. It utilizes light to create a map of the retinal tissue and can be used to identify and quantify papilledema. It can also serve as a marker of visual function and permanent visual damage.

Time to treatment was recorded as the time between initial ophthalmic review until the date of treatment initiation. The Frisen grading scale, which is a numerical scale that ranges from Grade 0 to Grade 5, was used specifically to describe optic nerve edema secondary to increased ICP or papilledema. Improvement in papilledema was defined as a decrease of at least one Frisen grade, while progression was defined as an increase of at least one Frisen grade. Time to improvement/progression of papilledema was measured from the first ophthalmic review to the date of documentation of change in papilledema. The total duration of hospital stay was measured as the number of days that the patient was admitted in hospital. If a patient was readmitted, the number of days continued from the previous admission. The other parameters measured were the opening pressure of initial lumbar puncture (LP), the peak erythrocyte sedimentation rate (ESR), white blood cell (WBC) count, C-reactive protein (CRP) level, and procalcitonin level. When available, reliable visual field (VF) testing in the form of Goldmann kinetic perimetry (Haag-Streit) and Humphrey static perimetry (Carl Zeiss Meditec) was performed. Reliable VF testing was defined as in previous pediatric studies (17). The qualitative assessment of VF testing for field defects was performed by a physician (RG) trained in neuro-ophthalmology.

Imaging consisted of computed tomography (CT), brain magnetic resonance (MR) imaging, and MR venography (MRV). CT tests included high-resolution thin section images of the temporal bones with intravenous contrast. MR images were obtained on a Siemens 3T system with multiplanar sequences of the brain and high-resolution sequences of the temporal bones without and with intravenous gadolinium. All patients underwent MR venography. Imaging was reviewed by an experienced pediatric neuroradiologist.

Patients were split into three distinct groups based on the treatment received (Figure 1):

Group 1—Patients who had further ICP controlling neurosurgery and no corticosteroids (n = 2).

An analysis was carried out using IBM SPSS Statistics, Version 29.0.0.0 (IBM Corp., Armonk, NY, USA). The data were not normally distributed therefore we used non-parametric tools for the statistical analysis. The Kruskal–Wallis test was used to compare the three patient groups. The Wilcoxon rank-sum test was used to compare two groups (Group 2 vs. Group 3). The exact method was applied because of the small sample size. The alpha level (type I error) was set at 0.05.

A total of 14 patients were identified with otitic hydrocephalus. The mean age at presentation was 6 (range 2–9) years. Full demographic and treatment information is available in Table 1. Otalgia was the most common presenting complaint (86%), followed by fever (79%), headache (79%), diplopia/sixth cranial nerve palsy (71%), lethargy (64%). The average duration of symptoms before presentation was 11 days (range 2–29 days).

At presentation, patients' BCVA in the worse seeing eye ranged from 20/20 to 20/40, except one who was only able to fix and follow. Optic disc edema at presentation ranged from Frisen grade 1 to grade 4 (Figure 2), with eight (57%) patients presenting with grade 1 or 2 papilledema (Table 2). Six patients (43%) had cranial nerve VI palsy. A total of 11 patients underwent an OCT examination on presentation (3 were not developmentally able), and the results revealed an average RNFL thickness of 262 µm and an average GCL volume of 1.1 mm³) in the worse eye. Five patients had abnormal retinal findings such as subretinal fluid (SRF), intraretinal fluid (IRF), retinal exudates, inner segment and outer segment junction (IS/OS) disruption, and paracentral acute middle maculopathy (PAMM) (Figure 3).

CT and MR examinations were reviewed for 13 of 14 patients; the reports were reviewed for 1 patient imaged at an outside institution, for whom images were not available. The CT of the temporal bones was obtained in 13 of 14 patients, and MR with MRV was obtained in all patients, with 12 MR images obtained on a 3T system at our institution. CVST was present in all patients and involved a part or most of one transverse sinus (TVS) (n = 13) extending to the contralateral TVS in one of these patients, one sigmoid sinus (SS) (n = 14), and one internal jugular vein (IJV) (n = 12). CVST involved the dominant TVS and SS in seven patients, with the remaining seven patients having codominant (n = 5), nearly symmetric sinuses (n = 2). Papilledema was present in all 13 patients for whom MR was available for review as evidenced by a protrusion of the optic papillae on T2-weighted axial images. Ipsilateral perisinus epidural abscesses were present in 8 patients and bilateral abscess in 1. Mastoid and middle ear space fluid signal was present ipsilateral to CVST in all patients and was bilateral in 11 patients. All 13 patients imaged with CT had evidence of coalescent mastoiditis with a mastoid air cell and middle ear space opacification and demineralization or erosion of the mastoid cortex.

All patients were started on broad-spectrum intravenous antibiotics (IV) within one day of admission. All were anticoagulated. Acetazolamide was started after the first ophthalmic examination, at 10–20 mg/kg/day, and in some patients, it was increased up to 30 mg/kg/day. Corticosteroids were started for 12 patients after controlling infection, with the average time to corticosteroids of 10 days (range 0–30 days). The patients were started on either 20–30 mg/kg IV methylprednisolone, 0.50–1 mg/kg dexamethasone, or 0.75–2 mg/kg oral prednisolone.

Thirteen patients were treated with initial otologic surgery with the aim of primary source control. These surgeries included canal wall up mastoidectomy, tympanostomy tube insertion, and drainage of postauricular abscess.

LP was performed for five patients as part of their diagnostic workup or for evaluation of refractory symptoms and to guide further management. The average opening pressure was 38 cm H₂O (range 27–55). Most of the time, LP was performed for refractory cases (n = 4).

In our evaluation of the impact of corticosteroids on clinical outcomes, we found that patients who were treated with corticosteroids for otitic hydrocephalus had an overall reduction in hospital stay. When comparing our three groups, we found no significant difference between the presenting symptoms and examination findings between them, suggesting that it could be difficult to predict which patients would require further surgeries despite intravenous corticosteroids. However, involvement of the dominant TVS, SS and IJV was seen more frequently in Group 3 patients.

A previous study by Chen et al., noted the potential benefit of using systemic corticosteroids in patients who were poorly responsive to medical management (15). They hypothesized that the worsening of papilledema was due to inflammation occurring locally. In patients with acute pseudotumor cerebri, high-dose corticosteroids were found to be effective in managing acute vision loss (18). In our study, 66% of those who were treated with corticosteroids did not require further neurosurgery.

Antibiotics and otologic surgical intervention are part of established management for intracranial hypertension secondary to mastoiditis (3, 10, 12, 19) While these work well to control the primary disease process, increased ICP and threat to vision remain a serious sequelae. The main morbidity resulting from them is headache and potential vision loss, and while the visual acuity outcomes of all 14 patients were good, there was evidence of optic nerve atrophy in 35.7% (5/14) of patients. All patients in our series received anticoagulation, and despite this, half the patients required neurosurgical intervention as there was refractory intracranial hypertension despite treatment of the infection and thrombus (20). Our group previously made similar observations and hypothesized that a secondary inflammatory process, following the initial infection, may cause dysregulation of the cerebrospinal fluid cycle; this has been noted in other studies of CVST without infection as well (6, 15). This phenomenon has been described to occur in 21.5%–29% of patients with CVST from all causes (4, 14).

In the cohort of patients who had corticosteroids only, 33% went on to require further neurosurgical intervention for vision-threatening papilledema. Cerebrospinal fluid shunting (ventriculostomy catheter, lumboperitoneal shunt, or lumbar drain) is a traditional option for treatment for persistently increased ICP. Previous studies have shown that these interventions can be effective in reducing ICP (13). As fewer patients in our study required surgery following systemic corticosteroids, we suggest that including steroids as part of treatment may reduce the need for neurosurgical intervention.

We performed a subanalysis, only looking at patients who had corticosteroids vs. those who went on to require further neurosurgery. We evaluated whether there were any differences between VA and papilledema grade on presentation. We found that there was no significance between the two groups; however, VA was trending toward significance (p = 0.10). This could be investigated further with a larger cohort of patients. Patients receiving corticosteroids had reduced hospital length of stay in between the three groups of patients (p = 0.05). This may indicate that there is a role for corticosteroids in patients with mastoiditis complicated by papilledema and CVST as they can help reduce psychological and physical burden on children and their families by reducing their length of stay. They may also serve as an indirect surrogate marker of overall systemic benefit in these patients.

Within the subcohort of patients who had retinal abnormalities on presentation (n = 5), all patients went on to develop optic disc atrophy after resolution of their papilledema. Fortunately, in these patients, final VA returned to normal despite OCT findings. Previous literature suggests that there can be a discrepancy between VA and optic disc atrophy; however, there may be a higher risk of future visual limitations (21). Furthermore, given the developmental age of the patients in our cohort, it was sometimes difficult to obtain reliable visual field assessments that may have demonstrated visual field loss. This emphasizes the importance of closer follow-up and more aggressive intervention. Curley et al. found that patients with retinal manifestations in idiopathic interstitial hypertension, such as retinal fluid, were more likely to have worse visual function and outcomes, indicating that macular findings at presentation may be a factor for poor prognosis (22). Therefore, in addition to monitoring patients who demonstrate signs of optic disc atrophy, a close evaluation of patients with retinal changes at presentation is warranted. OCT retinal imaging in patients with a similar clinical presentation may represent a useful tool for risk stratification. Prospective studies can help determine whether retinal changes on OCT can predict disease progression.

In conclusion, our series shows that there could be a potential role for systemic corticosteroids in reducing the hospital duration of patients with mastoiditis complicated by CVST and increased ICP. When patients are started on corticosteroids, it can be difficult to predict which patients may require further neurosurgery to control ICP. Therefore, it is important to monitor patients closely for improvement in the condition of papilledema.

Patients who presented with retinal changes on OCT may be at a higher risk of developing optic disc atrophy, which, in turn, may impact visual function such as the visual field. Thus, such patients require closer monitoring and more aggressive therapy.