Skip to main content
NCBI home page
Neuro-Oncology Advances logoLink to Neuro-Oncology Advances
. 2026 Jun 8;8(1):vdag151. doi: 10.1093/noajnl/vdag151

Neuro-ophthalmological disturbances in pediatric posterior fossa tumors: Prevalence, risk factors, and outcomes

Maria Aliotti Lippolis 1,2,3, Sabine Koinig 4, Christian Dorfer 5, Luca Conci 6, Martin Bertich 7, Christoph Stapf 8, Lisa Mayr 9,10,11, Natalia Stepien 12,13,14, Cora Hedrich 15,16,17, Alicia-Christina Baumgartner 18,19,20, Hannah Schned 21,22,23, Clara Köller, Abed Ahmadi 24,25,26, Oliver Eichmüller 27,28,29, Christine Haberler 30,31,32, Irene Slavc 33,34,35, Thomas Czech 36, Karl Rössler 37, Gregor Kasprian 38, Karl Kircher 39, Andreas Reitner 40, Andreas Peyrl 41,42,43, Johannes Gojo 44,45,46, Amedeo A Azizi 47,48,49,✉, Berthold Pemp 50
PMCID: PMC13635747  PMID: 42835364

Abstract

Background

Neuro-ophthalmological disturbances present frequently in pediatric posterior fossa (PF) tumors, contributing significantly to morbidity. This study evaluates their prevalence, risk factors, and outcomes in affected children.

Methods

This retrospective study analyzed 119 children with PF tumors treated over a decade at a tertiary care center, including 94 presenting at initial diagnosis and 25 at relapse. Clinical and ophthalmologic evaluations were performed preoperatively and 3 months, 1 year, and 2 years postoperatively. Prevalence of visual disturbances, central and/or peripheral oculomotor dysfunctions, optic nerve changes, and corrective ophthalmological interventions, were evaluated over time. Associations between visual/oculomotor outcomes and tumor location, histology, hydrocephalus, cerebellar mutism syndrome (CMS), and intraventricular chemotherapy were analyzed using χ2 tests.

Results

At diagnosis, n = 21/94 presented due to visual symptoms, while thorough preoperative examination revealed central and peripheral oculomotor disturbances in n = 12/27 and n = 6/27, respectively. Their prevalence increased postoperatively but improved over time in one-third of patients, with earlier resolution in peripheral disturbances. Vermis/brainstem location and CMS were associated with higher rates of postoperative oculomotor disorders. Medulloblastomas exhibited an elevated risk of central disturbances. Prism correction was initially required in n = 16/94 of cases, decreasing to n = 4/64 after 2 years; strabismus surgery was performed in n = 4/64. Papilledema was common (n = 10/33) at diagnosis, and n = 8/56 developed optic atrophy during follow-up. Among patients presenting at relapse, neuro-ophthalmological disorders were frequent (central: n = 24/25; peripheral: n = 15/25), while visual acuity remained satisfactory.

Conclusions

Neuro-ophthalmological disturbances are common in pediatric PF tumors and evolve throughout treatment. Regular ophthalmologic examinations, including detailed orthoptic assessments, are essential for timely intervention.

Keywords: cerebellar mutism syndrome, comitant esotropia, medulloblastoma, oculomotor disorder, pediatric brain tumors

Key Points.

  • Ophthalmological symptoms are frequent in posterior fossa (PF) tumors, and rates increase postsurgery.

  • Risk factors: brainstem/vermis location, PF syndrome, medulloblastoma.

  • Symptoms resolve in over a third of affected patients during follow-up.

Importance of the Study

This study represents a comprehensive investigation of the prevalence and progression of neuro-ophthalmological disturbances in pediatric patients with posterior fossa tumors. By analyzing a population-based cohort of 119 pediatric patients consecutively presenting at our center, retrospectively identified over a 10-year inclusion period, the study provides valuable insights into the relationship between tumor characteristics and visual outcomes. A significant percentage of children presented with visual disturbances, emphasizing the importance of early detection. High-risk features, including location in the vermis and brainstem or posterior fossa syndrome, are identified, advocating the need for focused monitoring strategies. The study also illustrates symptom evolution following surgery: while the prevalence of disturbances initially increases, symptoms may resolve over time. Given the known impact of oculomotor disorders on quality of life and the potential benefits of corrective procedures, early intervention in these patients may offer similar advantages. Thus, regular neuro-ophthalmological and orthoptic assessments are recommended to facilitate timely interventions and improve outcomes in this vulnerable population.

Central nervous system (CNS) malignancies represent the most common solid tumors in childhood, frequently localized in the posterior fossa (PF). Among those, medulloblastoma (MB), low-grade glioma (LGG), and ependymoma (EPN) are the main tumor entities.1,2 Visual impairment is a frequent initial symptom, along with headaches, ataxia, nausea, and vomiting.3,4 As ocular abnormalities may represent the first manifestation of a PF lesion in up to 10% of cases, neuro-ophthalmological examinations are important for a timely diagnosis, especially as visual symptoms were correlated with a prolonged prediagnostic interval in children with brain tumors.5-8

Although survival rates have improved over time, many survivors experience severe disabilities,9 some of which may be prevented by timely diagnosis and treatment.

Many factors including surgical management, elevated intracranial pressure (ICP), or tumor location itself may affect visual or oculomotor function by damaging neuroanatomical structures involved in the optic or oculomotor pathways.10-13 The nature of this damage varies greatly. Tumors can cause direct damage to the afferent optic pathway, the oculomotor system, and/or eloquent areas of the brain. In some patients, cranial nerves (CNs) are affected centrally (ie by impairment of their nuclei) due to brainstem compression and/or involvement, or peripherally along their intracranial course, resulting in different symptoms of oculomotor disturbance.

Increased ICP caused by intracranial tumors or tumor-induced obstructive hydrocephalus can lead to acute papilledema and potentially irreversible optic nerve damage.12-14

Besides tumor-related factors, visual impairment and ocular motility disorders can also have an iatrogenic origin. Resection of PF tumors can cause various postoperative deficits, including cerebellar mutism syndrome (CMS).15 Cerebellar mutism syndrome comprises motor, cognitive, and affective disturbances, such as severe ataxia, hypotonia, CN deficits, emotional lability, cerebellar mutism, and significant neurocognitive impairment.15 Symptoms appear days to weeks after surgery and vary in duration and severity.

In addition to surgery, oncological drug treatment can sometimes lead to CN neuropathy and visual impairment.16-20 While irradiation of the anterior optic pathway can cause optic neuropathy, in pediatric patients with PF tumors, radiotherapy does not appear to be correlated with poor visual outcomes.21-24

Methods

This study is a single-center, retrospective analysis with a mainly descriptive, exploratory nature. Approval was obtained from the ethical committee of the Medical University of Vienna, and informed consent was waived (EK Nr. 1918/2021).

Patient Cohort

All patients aged 0-19 years, diagnosed with a PF tumor at the Department of Pediatrics and Adolescent Medicine at the Medical University of Vienna between January 2012 and August 2021, were extracted from our database. While during this period a change of the WHO classifications of tumors of the CNS occurred, the overall grouping of LGG, MB, and EPN applied for the present analysis has not seen any changes. Diagnosis of novel entities only introduced in the 2016 classification (such as embryonal tumors with multilayered rosette [ETMR]) did, even retrospectively, not occur in our cohort previous to its introduction. Diffuse midline glioma and specifically diffuse intrinsic pontine glioma (DIPG) were per se excluded from the analysis, as they harbor a distinct other mechanism of impacting oculomotor and CN functions compared to cerebellar and fourth ventricle tumors (ie by infiltrating the nuclear and supranuclear oculomotor areas of the brainstem).25-29 For this analysis, we therefore defined PF tumors as tumors localized either in the cerebellum or the fourth ventricle. Patients operated at another center but undergoing subsequent treatment and regular follow-up at our institution (including clinical and ophthalmological examinations) were also included if sufficient data were available. Additionally, a cohort of children referred to our center at the time of relapse was evaluated descriptively as a separate group but not included in the main comparative analysis.

Medical records were reviewed for clinical and ophthalmological parameters; patients with insufficient data were excluded (Figure S1). All children routinely received ophthalmologic and orthoptic evaluations at similar timepoints during their treatment and follow-up.

Patients’ characteristics and neuro-ophthalmological findings at diagnosis and during follow-up were collected by chart review.

Tumor location was defined by an experienced pediatric neurosurgeon (C.D.) at our center, either by direct intraoperative observation or by analyzing the MRI scans and surgical reports of patients operated at an external center.30

Ophthalmological Examinations

Age-appropriate determination of best-corrected visual acuity (VA) and detailed testing of central and peripheral oculomotor functions were performed by an orthoptist (S.K.) specialized in neuro-orthoptic examination, who also assessed visual field (VF) in many cases. Fundoscopy and optical coherence tomography (OCT) were performed by neuro-ophthalmologists at the same clinical visits.

Central oculomotor disturbances included spontaneous nystagmus, gaze-evoked nystagmus, anomalies of smooth pursuit and saccadic eye movement, anomalies of visual fixation suppression of the vestibular ocular reflex, saccadic intrusions and oscillations, ocular tilt reaction (including skew deviation, ocular torsion, abnormal subjective visual vertical, and/or head tilt), and horizontal or vertical gaze palsy. Peripheral disturbances included CN palsies of the third, fourth, sixth, and/or seventh CN and comitant esotropia.

Visual acuity results were evaluated for the whole primary cohort, by comparing patients with or without hydrocephalus and symptoms of elevated ICP at diagnosis. Visual acuity was classified as normal or abnormal for patient age. An amelioration of at least 20% in VA in relation to the previous measurement was considered an improvement, whereas a deterioration of at least 20% as worsening.31-34 Patients showing diverging responses in both eyes are reported accordingly.

Visual field testing was performed using machine perimetry, including Goldmann, Dicon, or Humphrey perimeters. Visual field deficiencies included central scotomas, hemianopsia, and others. If pathological findings in VF testing generated findings that were not identical in both eyes, these were considered “divergent”. The optic nerve head was assessed by fundoscopy and, if feasible, by OCT measuring the average peripapillary retinal nerve fiber layer (pRNFL) thickness in both eyes. Pathological findings were classified as atrophy or edema.

Statistical Analysis

Data were collected at 4 timepoints: before surgery, 3 months after surgery (including outliers from early post­operatively and up to 6 months), 1 year postsurgery (±6 months), and 2 years postsurgery (±6 months).

The prevalence of central and peripheral oculomotor disturbances was examined in subgroups according to location, histology, application of intraventricular therapy, presence of symptoms of elevated ICP at diagnosis, and presence of CMS.

Data were transferred to Excel, and statistical analysis was performed using the software SPSS (Version 29.0.2.0, released 2023; IBM Corp., Armonk, NY). Descriptive statistics were performed on all data sets. χ2 test was performed to elucidate the relationship between tumor location, tumor histology, MB subgroups, and visual/oculomotor outcome at different timepoints. Additionally, 95% CIs and adjusted odds ratios were calculated. Statistical significance was considered from a P value of <.05. The time of symptom resolution during follow-up was measured. Kaplan-Meier Analysis of symptom resolution was performed with R (R Core Team, version 4.3.2, released 2023).

In the relapse cohort, only a descriptive analysis was performed due to the great inhomogeneity. Furthermore, a multivariate analysis was performed with exploratory aims. As expected, the full multivariable logistic regression model did not reach convergence, and estimates were unstable due to sparse data and quasi-complete separation. Therefore, parameter estimates for selected variables are not reliable. Results are presented for transparency but should be interpreted with caution (Table S4).

Results

Patients

This study identified 158 consecutive patients with a PF tumor (excluding DIPG) seen at our center (Figure S1). After excluding patients with insufficient data availability, atypical location (eg cerebellopontine angle, n = 3/152), or tumor histology (eg sarcoma, neuroendocrine carcinoma), 119 patients were eligible for the analysis. Twenty-five patients presented at our center only at relapse or disease recurrence and following treatment in other centers and were solely included in the descriptive part of this study as a separate cohort. The remaining 94 patients formed the main cohort of newly diagnosed patients included in the comparative analysis.

From the longitudinal cohort, 17 patients (18.1% [n = 17/94]) relapsed/progressed during follow-up.

The median oncologic follow-up after initial diagnosis (main cohort) was 3.5 (1.3-6.1) years, the median ophthalmologic follow-up was 1.4 (0.4-2.6) years.

Clinical Characteristics

Patient characteristics, including tumor histology and location, oncological treatment, the most common clinical signs leading to initial diagnosis, permanent surgical resolution of hydrocephalus, the and presence of CMS, are summarized in Figure 1 and Table S1.

Figure 1.

This figure shows that localisation of the tumours was almost equally distributed between cerebellar hemispheres, vermis and brainstem. Histology was most commonly LGG (49%), followed by medulloblastoma and Ependymoma. Symptoms at diagnosis is listed as well as oncological treatment strategies. Patients characteristics at relapse are given.

Open in a new tab

Population characteristics. (A) Histology of patients presenting at primary diagnosis; (B) tumor localization of patients presenting at primary diagnosis; (C) surgical treatment of hydrocephalus in patients presenting at primary diagnosis; (D) further characteristics of patients presenting at primary diagnosis; (E) histology of patients presenting at relapse; and (F) further characteristics of patients presenting at relapse. Further details can be found in Table S1. 1Other location = diffuse/not classifiable;  2Other CN symptoms = torticollis, hearing impairment, neck pain, weakness of the shoulder;  3The symptoms for elevated intracranial pressure considered for this study are cephalea, vomiting, and nausea. 4Other symptoms in this analysis are reduced appetite, weight loss, weakness, change of character, regression of development, and dizziness. 5Other histologies of patients presenting at relapse were: atypical teratoid rhabdoid tumor, embryonal tumor with multilayered rosettes. CN, cranial nerve; EPN, ependymoma; FU, follow-up; ICP, intracranial pressure; IQR, interquartile range; LGG, low-grade glioma; MB, medulloblastoma; SHH, Sonic hedgehog; WNT, wingless; VP, ventriculo-peritoneal.

In the main cohort, the age ranged from 0.3 to 19 years, with a median age at diagnosis of 6.6 years (IQR: 3.8-9.7). The female to male ratio was 36:58. The most prevalent tumor histology was LGG (48.9% [n = 46/94]) and MB (41.5% [n = 39/94]), while EPN (9.6% [n = 9/94]) was the least common. Most tumors originated in the vermis (38.3% [n = 36/94]), followed by the cerebellar hemispheres (33% [n = 31/94]) and the brainstem (27.7% [n = 26/94]). One MB was initially metastasized and diffusely infiltrating the cerebellum, and its anatomical origin was therefore nonclassifiable (classified as “other”). One in 5 children was diagnosed with metastatic disease (20.2% [n = 19/94]). Almost half of the patients received nonsurgical oncological treatment, including chemotherapy, radiotherapy, or a combination of both (48.9% [n = 46/94], 41.5% [n = 39/94], 39.4% [n = 37/94], respectively). Of all patients, 13 (13.8% [n = 13/94]) received additional intraventricular chemotherapy.

The most common clinical signs leading to initial diagnosis were symptoms of elevated ICP (including headache, nausea, or vomiting) (84% [n = 79/94]), ataxia (62.8% [n = 59/94]), visual disturbances (22.3% [n = 21/94]), or other CN disturbances, including hearing impairment, neck pain, and torticollis/accessory nerve palsy (eg weakness of the shoulder) (14.9% [n = 14/94]). Among patients who presented with visual disturbances at diagnosis, the most common findings were diplopia, blurred vision, anisocoria, and strabismus. Fifty-nine children (62.8% [n = 59/94]) presented with other symptoms such as reduced appetite, weight loss, weakness, change of character, regression of development, or dizziness at diagnosis.

Of the 79 patients with elevated ICP or hydrocephalus, 18 (22.8% [n = 18/79]) required surgical intervention (third ventriculostomy: 13.9% [n = 11/18]; ventriculo-peritoneal shunt: 8.9% [n = 7/18]). Seven children experienced CMS (7.4 [n = 7/94]) after tumor surgery.

Ophthalmological Assessment

Initial presentation

Table 1 outlines the occurrence of visual symptoms at diagnosis, distinguishing between patients with and without signs of elevated ICP at diagnosis. At initial presentation, 22.3% (n = 21/94) of patients and families sought medical advice due to visual disturbances, the most common being double vision and strabismus (10.6% [n = 10/94] and 7.4% [n = 7/94], respectively). Other symptoms included blurred vision, anisocoria (3.2% [n = 3/94] each), reduction in VA, and photophobia (1.1% [n = 1/94] each). However, the professional examination at diagnosis revealed a far higher rate of ophthalmological symptoms (see Table 2).

Table 1.

Ocular symptoms prompting initial medical consultation

Ocular symptoms All patients Symptoms of elevated ICP at diagnosis
 P OR
No Yes
n = 94 n = 15 n = 79
Ocular symptoms at diagnosis 21/94 (22.3%) 6/15 (40%) 15/79 (19%) .073 0.352 (0.108-1.14)
 Double vision 10/94 (10.6%) 1/15 (6.7%) 9/79 (11.4%) .586 1.8 (0.211-15.363)
 Strabismus 7/94 (7.4%) 3/15 (20%) 4/79 (5.1%) .043 0.213 (0.042-1.074)
 Blurred vision 3/94 (3.2%) 0/15 (0%) 3/79 (3.8%) .443 NAa
 Anisocoria 3/94 (3.2%) 2/15 (13.3%) 1/79 (1.3%) .015 0.083 (0.007-0.986)
 Decrease in visual acuity 1/94 (1.1%) 1/15 (6.7%) 0/79 (0%) .021 NAa
 Photophobia 1/94 (1.1%) 0/15 (0%) 1/79 (1.3%) .661 NAa
Open in a new tab

Subgrouping by patients presenting with and without concomitant symptoms of elevated ICP at diagnosis.

a

Not applicable due to absence of cases in one of the groups.

Abbreviations: ICP, intracranial pressure; NA, not applicable. Significant results are highlighted in bold.

Table 2.

Description of central and peripheral oculomotor symptoms

Preoperative Three months One year Two years Patients presenting at relapse
n = 94 n = 94 n = 83a n = 64a n = 25
Central symptoms 12/27 (44.4%) 45/75 (60%) 33/59 (55.9%) 15/35 (42.9%) 24/25 (96%)
Spontaneous nystagmus 1/27 (3.7%) 12/74 (16.2%) 6/57 (10.5%) 2/35 (5.7%) 8/25 (32%)
Gaze-evoked nystagmus 7/21 (33.3%) 26/75 (34.7%) 17/55 (30.9%) 8/33 (24.2%) 20/25 (80%)
Anomalies of smooth pursuit 9/15 (60%) 41/62 (66.1%) 30/46 (65.2%) 10/25 (40%) 23/25 (92%)
Anomalies of saccadic movements 4/14 (28.6%) 30/62 (48.4%) 23/46 (50%) 9/25 (36%) 20/25 (80%)
Anomalies of visual fixation 2/5 (40%) 15/30 (50%) 13/22 (59.1%) 0/0 (0%) 16/25 (64%)
Intrusions and oscillations 0/19 (0%) 4/72 (5.6%) 4/55 (7.3%) 1/34 (2.9%)
Ocular tilt reaction 2/21 (9.5%) 18/75 (24%) 6/56 (10.7%) 5/34 (14.7%) 9/25 (36%)
Skew deviation 1/21 (4.8%) 9/75 (12%) 4/56 (7.1%) 1/34 (2.9%)
Ocular torsion 1/21 (4.8%) 14/74 (18.9%) 5/55 (9.1%) 4/34 (11.8%)
Subjective visual vertical 1/20 (5%) 12/73 (16.4%) 5/56 (8.9%) 2/34 (5.9%)
Head tilt 0/21 (0%) 4/75 (5.3%) 1/56 (1.8%) 2/34 (5.9%)
Gaze palsy 0/24 (0%) 3/54 (5.6%) 1/56 (1.8%) 1/32 (3.1%) 2/25 (8%)
Horizontal gaze palsy
 Rightward 0/24 (0%) 1/54 (1.9%) 1/56 (1.8%) 1/32 (3.1%)
 Leftward 0/24 (0%) 2/54 (3.7%) 0/56 (0%) 0/32 (0%)
Vertical gaze palsy 0/24 (0%) 0/54 (0%) 0/56 (0%) 0/0 (0%)
Peripheral symptoms 6/27 (22.2%) 32/75 (42.7%) 17/56 (30.4%) 10/34 (29.4%) 15/25 (60%)
CN palsies 1/27 (3.7%) 11/75 (14.7%) 10/56 (17.9%) 7/34 (20.6%) 7/25 (32%)
CN III 0/27 (0%) 1/75 (1.3%) 1/56 (1.8%) 0/34 (0%) 0/25 (0%)
CN IV 0/27 (0%) 0/75 (0%) 1/56 (1.8%) 0/34 (0%) 1/25 (4%)
CN VI 0/27 (0%) 8/75 (10.7%) 7/56 (12.5%) 5/34 (14.7%) 5/25 (20%)
CN VII 1/27 (3.7%) 4/75 (5.3%) 4/56 (7.1%) 2/34 (5.9%) 1/25 (4%)
Comitant esotropia 5/22 (22.7%) 29/75 (38.7%) 15/56 (26.8%) 9/34 (26.5%) 13/52 (52%)
Diplopia corrections
Prism foils 0/94 (0%) 15/94 (16%) 9/83 (10.8%) 4/64 (6.3%) 5/25 (20%)
Strabismus surgery 0/94 (0%) 0/94 (0%) 1/83 (1.2%) 4/64 (6.3%)b

  • Strabismus surgery

  • (months postoperative; median, IQR)

 25.1 (18.7-27)
Open in a new tab

Following timepoints were evaluated: preoperatively, 3, 12, and 24 months after surgery (± 6 months). The total number of patients included for each visit is reported in the column title. As at a specific timepoint, not every examination was performed in every patient, the number of patients may vary for each variable.

a

At the time of database closure, 11/94 (11.7%) and 30/94 (31.9%) patients at 1 year and 2 years postoperatively were out of follow-up, respectively.

b

Two additional patients required strabismus surgery after the study was completed.

Abbreviations: CN, cranial nerves.

Central and peripheral oculomotor symptoms

The occurrence of central and peripheral symptoms assessed by neuro-orthoptic examination as well as diplopia corrections is described in Table 2. The prevalence of central as well as peripheral disturbances was higher in the first postoperative follow-up compared to preoperatively and then decreased over time (central/peripheral disturbances: 44.4% [n = 12/27]/22.2% [n = 6/27] preoperatively; 60% [n = 45/75]/42.7% [n = 32/75] early postoperatively; 55.9% [n = 33/59]/30.4% [n = 17/56] at 1 year; and 42.9% [n = 15/35]/29.4% [n = 10/34] 2 years postoperatively).

At the first follow-up after surgery, 16% (n = 15/94) of examined patients needed diplopia correction with prism foils, and this number later decreased to 10.8% (n = 9/83) and 6.3% (n = 4/64) after 1 and 2 years. Strabismus surgery was required in 4 children (median 25.1 months after tumor surgery, IQR: 18.7-27). Two additional patients of the cohort underwent strabismus surgery after data collection was completed.

Evaluation of potential risk factors

The subgroup analysis of central and peripheral oculomotor symptoms is presented in Table 3. Regarding tumor location, preoperative central disturbances had the highest prevalence in patients with tumors originating from the vermis (66.7% [n = 4/6]) or the pons (55.6% [n = 5/9]) compared to the cerebellar hemispheres (27.3% [n = 3/11]). At the first postoperative follow-up, the prevalence of central symptoms was 66.7% (n = 20/30) and 81% (n = 17/21) for tumors with origin in the vermis and brainstem, respectively, and 34.8% (n = 8/23) for hemispheric cerebellar neoplasms (P = .007), while at 1 year after resection, the prevalence was 61.5% (n = 16/26), 81.3% (n = 13/16), and 25% (n = 4/16), respectively (P = .007).

Table 3.

Comparations on central and peripheral oculomotor symptoms

Preoperative Three months One year Two years
n = 94 n = 94 n = 83a n = 64a
Central oculomotor symptoms
Histology MB 4/7 (57.1%) 28/34 (82.4%) 23/28 (82.1%) 10/17 (58.8%)
LGG 7/18 (38.9%) 14/35 (40%) 8/26 (30.8%) 3/14 (21.4%)
EPN 1/2 (50%) 3/6 (50%) 2/5 (40%) 2/4 (50%)
Total 15/27 (44.4%) 45/75 (60%) 33/59 (55.9%) 15/35 (42.9%)
P (Pearson Chi square) .702 .001 .001 .107
Tumor location Brainstem 5/9 (55.6%) 17/21 (81%) 13/16 (81.3%) 5/9 (55.6%)
Vermis 4/6 (66.7%) 20/30 (66.7%) 16/26 (61.5%) 8/17 (47.1%)
Cerebellar hemispheres 3/11 (27.3%) 8/23 (34.8%) 4/16 (25%) 2/9 (22.2%)
Other 0/1 (0%) 0/1 (0%) 0/1 (0%) 0/1 (0%)
Total 12/27 (44.4%) 45/75 (60%) 33/59 (55.9%) 15/35 (42.9%)
P (Pearson Chi square) .288 .007 .007 .320
Symptoms of elevated ICP at diagnosis Yes 8/18 (44.4%) 42/66 (63.6%) 29/49 (59.2%) 11/30 (36.7%)
P (Pearson Chi square) 1.000 .082 .265 .070
Intraventricular therapy Yes 0/2 (0%) 7/11 (63.6%) 3/5 (60%) 1/3 (33.3%)
P (Pearson Chi square) 0.189 0.790 0.848 0.727
Posterior fossa syndrome Yes 0/1 (0%) 7/7 (100%) 7/7 (100%) 2/4 (50%)
P (Pearson Chi square) .362 .023 .012 .759
Peripheral oculomotor symptoms
Histology Medulloblastoma 1/9 (11.1%) 15/34 (44.1%) 8/27 (29.6%) 6/16 (37.5%)
LGG 4/16 (25%) 14/35 (40%) 7/24 (29.2%) 3/14 (21.4%)
Ependymoma 1/2 (50%) 3/6 (50%) 2/5 (40%) 1/4 (25%)
Total 6/27 (22.2%) 32/75 (42.7%) 17/56 (30.4%) 10/34 (29.4%)
P (Pearson Chi square) .448 .877 .886 .615
Tumor location Brainstem 2/9 (22.2%) 10/21 (47.6%) 9/16 (56.3%) 4/9 (44.4%)
Vermis 1/7 (14.3%) 13/30 (43.3%) 7/24 (29.2%) 5/16 (31.3%)
Cerebellar hemispheres 3/10 (30%) 9/23 (39.1%) 1/15 (6.7%) 1/9 (11.1%)
Other 0/1 (0%) 0/1 (0%) 0/1 (0%) 0/1 (0%)
Total 6/27 (22.2%) 32/75 (42.7%) 17/56 (30.4%) 10/34 (29.4%)
P (Pearson Chi square) .828 .782 .023 .293
Symptoms of elevated ICP at diagnosis Yes 5/20 (25%) 28/66 (42.4%) 14/46 (30.4%) 8/29 (27.6%)
P (Pearson Chi square) .557 .908 .978 .574
Intraventricular therapy Yes 0/2 (0%) 4/11 (36.4%) 1/5 (20%) 1/3 (33.3%)
P (Pearson Chi square) .432 .647 .598 .876
Posterior fossa syndrome Yes 0/1 (0%) 4/7 (57.1%) 3/7 (42.9%) 2/3 (66.7%)
P (Pearson Chi square) .586 .416 .442 .138
Open in a new tab

Following timepoints were evaluated: preoperatively, 3, 12, and 24 months after surgery (± 6 months). The total number of patients included for each visit is reported in the column title. As at a specific timepoint, not every examination was performed in every patient, the number of patients may vary for each variable.

a

At the time of database closure 11/94 (11.7%) and 30/94 (31.9%) patients, respectively, at 1 year and 2 years postoperatively were out of follow-up.

Abbreviations: EPN, ependymoma; ICP, intracranial pressure; LGG, low-grade glioma; MB, medulloblastoma. Significant results are highlighted in bold.

Similarly, a lower prevalence of peripheral oculomotor disturbances was seen 1 year after surgery in tumors originating from the cerebellar hemispheres (6.7% [n = 1/15]) compared to those originating from the vermis or the brainstem (29.2% [n = 7/24] and 56.3% [n = 9/16] [P = .023]).

In relation to the histological entity, patients with MB showed a significantly elevated prevalence of central oculomotor symptoms, both at the early postoperative and 1-year postoperative follow-up (82.4% [n = 28/34] and 82.1% [n = 23/28] [P = .001]). No significant correlation with histology could be identified for peripheral disturbances.

The presence of elevated ICP was not associated with a higher prevalence of central or peripheral disturbances at any timepoint. However, a trend toward higher rates of central disturbances was noted at the 2-year postoperative follow-up (P = .070). Patients without elevated ICP at diagnosis showed a trend toward higher incidence of visual disturbances at diagnosis (P = .073) and presented with a significantly higher rate of strabismus (P = .043), decreased VA (P = .021), and anisocoria (P = .015; OR: 0.083 [0.007-0.986]) (Table 1).

Patients experiencing CMS showed a higher prevalence of central oculomotor disturbances (early postoperative: P = .023, 1-year postoperative: P = .012). Intraventricular chemotherapy was not associated with an increase of ophthalmological symptoms during follow-up.

Evaluation over time

The rate of complete resolution of central and peripheral disturbances over time is presented in Figure 2. In the case of central disorders, low but ongoing resolution was identified (6.7% [n = 3/45] at 400 days, 11.1% [n = 5/45] at 800 days, and 13.3% [n = 6/45] at 1000 days postsurgery). Regarding peripheral symptoms, an early tendency toward resolution was observed in the immediate postoperative period, 18.8% [n = 6/32] had recovered after 200 days, and 31.3% [n = 10/32] at 400 days. The remaining patients showed no complete resolution of symptoms at the last follow-up.

Figure 2.

The two graphs show that a potential resolution of peripheral symptoms occurs earlier and at a higher rate than that of central symptoms.

Open in a new tab

Kaplan-Meier analysis depicting the complete resolution of central (A) and peripheral (B) neuro-ophthalmological symptoms over time. It includes symptomatic patients that presented at initial diagnosis.

VA assessment

Visual acuity data are given in Table S2 and Figure S2. Both the preoperative and the 1- and 2-year assessments showed VA normal for age in most patients, with no differences between patients with or without elevated ICP at diagnosis. Counterintuitively, patients with elevated ICP had a lower prevalence of VA impairment at the first postoperative assessment (OR: 0.14 [0.02-0.85], P = .018) and experienced a higher rate of improvement at 3 months, although this trend did not reach statistical significance (P = .087).

Evaluations of the optic disk and VF

The status of the optic disk assessed by OCT and fundoscopy is described in Table S3 and Figure S3. Preoperatively, OCT and fundoscopy were performed in 11 and 30 patients, respectively. Pathological findings, including papilledema (n = 10), optic atrophy (n = 1), and divergent findings (n = 1), were identified in at least 1 eye of 40% (n = 12/30) of patients who underwent fundoscopy. Patients with initial symptoms of increased ICP exhibited papilledema on fundoscopy both preoperatively and at the first postoperative assessment (P = .022 and P = .024, respectively).

The prevalence of papilledema decreased during follow-up (preoperative: 33.3% [n = 10/30], postoperative: 32% [n = 24/75] at 3 months, 1.8% [n = 1/56] at 1 year), and no patient exhibited papilledema at the 2-year follow-up. Conversely, the number of patients exhibiting optic nerve atrophy increased gradually (preoperative: 3.3% [n = 1/30], postoperative: 5.3% [n = 4/75] at 3 months, 14.3% [n = 8/56] at 1 year, and 14.3% [n = 5/35] at 2 years). Twenty-three patients with papilledema recovered over time without developing optic atrophy, while 1 experienced chronic damage leading to atrophy. Among patients with optic atrophy, 4 had no previous fundoscopic pathology, 3 were consistently atrophic, and 1 exhibited divergent findings (1 eye showing papilledema, the other atrophy). In patients examined with OCT, median pRNFL thickness was 120 µm (IQR: 97-185) at diagnosis and decreased to a median thickness of 95 µm (IQR: 84-104), with values within the normal range in most patients after 2 years.

Preoperative VF testing was performed in 11 patients. Of these, 3 children (27.3% [n = 3/11]) showed VF deficits in at least 1 eye, including central scotoma, hemianopsia, and diverse other changes (relative VF loss, constricted VF, arcuate scotoma, and enlarged blind spot). Postoperative deficits were present in a small percentage of the cohort and remained constant during follow-up. Despite the young age of the cohort, machine-based VF testing could be performed in many patients at each timepoint (9.6% [n = 9/94], 30.9% [n = 29/94], 31.9% [n = 30/83], and 18.1% [n = 17/64], respectively).

Patients at Relapse

Twenty-five patients (29.8% [n = 25/119]) treated in our center presented only at relapse, and patient characteristics including age, tumor histology, oncological, and ophthalmological FU are described in Figure 1 and Table S1.

The age at diagnosis ranged from 1.4 to 15 years, and median age at diagnosis was 6.2 years (IQR: 4.5-7.6). The female-to-male ratio was 9:16. The histological diagnoses included MB (68% [n = 17/25]), EPN (20% [n = 5/25]), atypical teratoid rhabdoid tumors (ATRTs), and ETMRs (in total 12% [n = 3/25]). After presentation at our center, the median oncologic FU was 2.3 years (IQR: 1-4.7), and the median ophthalmologic FU was 2.0 years (IQR: 0.5-3.0).

In contrast to the high prevalence of oculomotor disturbances, VA was not or only mildly affected (according to ICD-10 classification), with a median VA of 1.0 decimal in both eyes (IQR OD: 0.5-1.0; OS: 0.6-1.0). Optic nerve evaluation showed papilledema in 4% (n = 1/25) and optic atrophy in 20% (n = 5/25) of children (Table S3).

Discussion

This study is one of the largest and most in-depth analyses of neuro-ophthalmological symptoms in pediatric patients with PF tumors ever conducted, focusing not only on prevalence but also on potential risk factors and outcomes. Although retrospective in nature, it was possible to retrieve high-quality longitudinal data for most patients. In addition to a well-characterized cohort of newly diagnosed patients, it was for the first time possible to include descriptive data on a substantial cohort of patients at relapse, who exhibited a different and more severe pattern of symptoms.

We analyzed the most common neuro-ophthalmological symptoms and their connection with tumor-related factors, such as the presence of elevated ICP, tumor location, histology, and treatment-related factors, including surgery, intraventricular chemotherapy, and CMS.

The issue of inter-observer variability may be a potential concern in ophthalmological examinations conducted on children. Indeed, the experience, patience, and training of the examiner may well determine the likelihood of successful completion and the assessment of valid and usable results when dealing with young children who are often already anxious and stressed. It is important to note that, in this cohort, besides the preoperative evaluation, which is often conducted in an emergency setting, all subsequent follow-up examinations have been conducted by ophthalmologists and orthoptists experienced with children, either in the neuro-ophthalmology clinic or the strabismus clinic of the Department of Ophthalmology, mostly by the same healthcare providers on different occasions. The highly trained and skilled personnel that is routinely working with children with brain tumors should help to minimize the degree of inter-observer variability in the results obtained.

A preoperative assessment was possible only in a limited number of our cohort, since many children presented with life-threatening symptoms and underwent immediate surgery for tumor resection or resolution of obstructive hydrocephalus. In patients seen preoperatively, the ophthalmological examination at diagnosis revealed more visual symptoms than initially identified by patients and families. Whereas this may well be due to thorough examination, it might also be biased as patients with apparent visual disturbances might have been sent for a preoperative exam more often.35 More complete preoperative data could help to discriminate the influence of tumor location and infiltration versus the effect of surgery. It is therefore recommended that a comprehensive neuro-ophthalmological examination should be incorporated into the diagnostic checklist during the preoperative phase, in addition to the standard laboratory, clinical, and neurological examinations as well as neuroradiological imaging procedures, unless the clinical condition of the patient prohibits the assessment.

Ophthalmological symptoms can arise from local tumor effects, with tumor location being critical in the pathogenesis of visual disturbances. Tumor infiltration in the midbrain, pons, and brainstem can directly cause CN neuropathies with ophthalmoplegia, gaze limitation, or ocular tilt reaction. Cerebellopontine pathway disruption affects visual fixation, causing nystagmus, skew deviation, or impaired saccadic movements.5,6,11,36-39 Consequently, in the present study, patients with cerebellar hemispheres and vermis tumors showed significantly fewer central and peripheral symptoms compared to patients with tumors arising from the brainstem (Table 3). As this finding impacts the first and second postoperative assessments, but not the preoperative examination or the 2-year follow-up, this emphasizes the role of surgery in the origin of neuro-ophthalmological disturbances. On the other hand, it also depicts that neuro-ophthalmological deficits may improve over time. In fact, our findings reveal a notable resolution of both peripheral and central oculomotor disorders, 1- and 2-years postoperatively (Figure 2).

As oncological surgery remains the mainstay of brain tumor treatment, and residual tumor is directly associated with postoperative progression,40-42 only surgeons experienced in pediatric neurosurgery should perform such procedures. Development of novel, risk-adapted surgical techniques and approaches is therefore warranted that might help to reduce ophthalmological sequelae in the future.

In historical cohorts, duration and intensity of preoperative hydrocephalus seemed to have a direct impact in terms of visual outcomes, potentially resulting in papilledema with risk of irreversible impairment of visual function due to axonal damage to the optic nerve.3,4,7,12-14 Despite limited available data, preoperative tests in our cohort indicated a significantly higher prevalence of papilledema in patients with symptoms of elevated ICP at diagnosis, both preoperatively and 3 months after surgery, compared to those without. No case of papilledema was observed after 2 years (Table S3 and Figure S3), reinforcing existing knowledge of papilledema improvement after resolution of hydrocephalus. Yet, this study found no correlation between hydrocephalus and poor VA or its worsening over time. Most patients had normal VA, with constant improvement until the last follow-up. This result differs from the findings of some previous studies investigating pediatric patients with PF tumors.3,43 Surprisingly, our findings even indicate that patients initially presenting without elevated ICP exhibited a markedly increased incidence of visual disturbances, strabismus, and reduced VA upon diagnosis (Table 1).

Our data suggest that elevated ICP is only causing visual symptoms at diagnosis in a quarter of patients. Moreover, such symptoms tend to resolve rapidly following surgical intervention. Nonetheless, in cases where direct damage to the optic pathway or CNs is the underlying cause of visual symptoms, these may be less likely to resolve over time. In our cohort, we observed unexpected high rates of direct damage to the optic system, for example by metastasis, which explains a skewing of the analysis. The small cohort of patients without hydrocephalus in the present study (only 16% [n = 15/94]) may have hampered these findings, especially as one of these patients initially presented with a metastatic lesion on the optic nerve and consecutive loss of VA. In total, metastatic optic pathway lesions were described in 4 patients, also including 2 cases during follow-up with occipital metastases and 1 case with leptomeningeal carcinomatosis (Table S3 and Figure S3). Furthermore, no statistically significant correlation was found between the presence of hydrocephalus at diagnosis and oculomotor disturbances at any time during follow-up. Yet, in children with initial hydrocephalus, a trend for more frequent central symptoms was found 2 years after surgery (Table 3).

We also evaluated the influence of histology on ophthalmological outcomes, as previous studies postulated a correlation between tumor pathology and inferior visual outcome in terms of VA and strabismus.44,45 Children with PF LGGs have been found to experience fewer long-term oculomotor dysfunctions, including nystagmus and strabismus, compared to those with MB or EPN.3 It has previously been hypothesized that the aggressive growth pattern observed in high-grade malignancies, characterized by compression and invasion of the brainstem and involvement of CN structures, could result in a poorer ophthalmological outcome, especially as these tumors necessitate intensive surgical treatment.3 Our cohort consisted mostly of LGG and MB, with fewer EPN, therefore representing well the numbers expected in such a population.1,2,43 Coherently with historical findings, patients with MB in our cohort had a significantly higher rate of central oculomotor disturbances, both at 3 months and 1 year of follow-up (Table 3). This may be caused by the frequent origin in the brainstem on the floor of the fourth ventricle or infiltration thereof, prone to primary tumor-related or secondary surgical damage. The further need for intensive oncological treatments might also have a negative impact on recovery and cannot be seen separately, as the visual outcome is influenced by both tumor and therapy-related factors.

The surgical treatment itself comes with a risk of neurological impairment, as seen with CMS. This study identifies a greater prevalence of central oculomotor disturbances among patients with CMS at both 3 months and 1 year follow-ups, confirming the influence of surgery on oculomotor outcomes (Table 3).

While ophthalmological sequelae are common, information regarding potential resolution of neuro-ophthalmological symptoms is crucial in postoperative counseling of families. Peripheral disorders often resolved during the first postsurgical year, whereas central disorders showed improvement only to a lesser extent but ameliorated even more than 2 years postoperatively (Figure 2). It may be speculated that the resolution of oculomotor disorders observed at a late stage, potentially only after completing intensive therapy, may be attributed to the role of anticancer treatment. This finding also corroborates the existing knowledge that a variable degree of improvement of CMS and its associated central disorders may be anticipated over time.46 Indeed, trial designs currently being developed are focusing on evaluating the role of surgery in preserving neurological functions by aiming at randomizing the standard of care of gross total to a less radical resection.

Regarding medical oncological interventions such as chemotherapy using neurotoxic agents, it is hard to separate their effects from other measures required during treatment. Intraventricular chemotherapy, analyzed separately due to its potential for CN neuropathy, showed no significant impact neither on central nor peripheral oculomotor disturbances (Table 3).47-50

Strabismus Sequelae and Treatment

Acute comitant esotropia is a frequent finding in pediatric neuro-oncology.4,6,51-53 Prism corrections and strabismus surgery are often necessary to restore visual integrity. Yet, oculomotor fusion might not be restorable even with strabismus surgery, especially if the eye misalignment has existed for a longer period.12 Orthoptic treatment with prism foils was often indicated immediately after tumor surgery, and its prevalence decreased during follow-up. In our cohort, we could observe a gradual reduction of both central and peripheral oculomotor disturbances in the years following tumor resection. Nevertheless, orthoptic treatment was often necessary due to permanent long-term sequelae. Strabismus surgery had to be performed in 4 children after a median of 2 years following tumor surgery. As each case may evolve differently over time, individualized treatment plans are needed to determine the best timing and approach for strabismus intervention. In our series, the timing of surgery was likely influenced by the intensive multimodal oncological treatment that hindered earlier interventions. Patients requiring strabismus surgery in the present cohort were affected by either MB or LGG (3:1), demonstrating that this was not linked to histological grade. Furthermore, 2 additional patients required strabismus surgery following data cut off for the present analysis. One patient with LGG underwent surgery 12 months after tumor resection, and a second patient with multiple MB recurrences, who had already received a strabismus surgery years earlier, developed a new trochlear nerve palsy and a worsening of the preexisting abducens palsy, so that a second strabismus surgery is currently being planned 31 months following resection of the last recurrent lesion. Oculomotor disorders are known to significantly impair quality of life, and interventions such as strabismus surgery have demonstrated clear improvements.54-56 Early corrective intervention in this high-risk pediatric population (such as the use of prisms and/or strabismus surgery) may therefore lead to visual benefit and improvements in quality of life.

Since our institution is a reference center for relapsed pediatric high-grade tumors, a considerable proportion of PF tumor patients were initially referred at relapse, including MB, EPN, ATRT, and ETMR. Despite the numerous ocular motility and optic nerve disorders observed at presentation at our center, VA was not or only mildly affected (see Table 3 and Table S2).

While, to our knowledge, this is the most extensive study of its kind, possible limitations have to be considered. The retrospective nature may have impaired the analysis. Especially, the paucity of preoperative data is hampering in-depth statistical assessments. Yet, the availability of preoperative evaluations is unlikely to improve in future studies, as many patients require immediate surgery for tumor resection or drainage of a hydrocephalus. Furthermore, aggressive tumors such as MB and EPN may relapse during follow-up and require further treatment. Thus, in some patients of the longitudinal cohort neuro-ophthalmological findings may have been caused by disease progression and relapse treatment rather than by the primary tumor or first-line therapy.

Conclusion

Neuro-ophthalmological findings are common in children with PF tumors and can be caused by both tumor and therapy-related factors at any time during the course of the disease. Strategies to reduce the risk of treatment-related factors are therefore warranted. Regular preventive neuro-ophthalmological examinations, including detailed orthoptic assessments, are essential to ensure timely corrective treatment and an improvement of visual function. While many symptoms may resolve over time, some patients show long-term sequelae, which should be recognized and treated, aiming at optimizing quality of life in childhood brain tumor survivors.

Supplementary Material

vdag151_Supplementary_Data
vdag151_supplementary_data.zip (2.3MB, zip)

Contributor Information

Maria Aliotti Lippolis, Division of Neonatology, Pediatric Intensive Care and Neuropediatrics, Department of Pediatrics and Adolescent Medicine, Medical University of Vienna, Vienna, Austria; Comprehensive Center for Pediatrics, Medical University of Vienna, Vienna, Austria; Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria.

Sabine Koinig, Department of Ophthalmology and Optometry, Medical University of Vienna, Vienna, Austria.

Christian Dorfer, Department of Neurosurgery, Medical University of Vienna, Vienna, Austria.

Luca Conci, Department of Cardiac Surgery, Medical University of Vienna, Vienna, Austria.

Martin Bertich, Department of Ophthalmology and Optometry, Medical University of Vienna, Vienna, Austria.

Christoph Stapf, Department of Ophthalmology and Optometry, Medical University of Vienna, Vienna, Austria.

Lisa Mayr, Division of Neonatology, Pediatric Intensive Care and Neuropediatrics, Department of Pediatrics and Adolescent Medicine, Medical University of Vienna, Vienna, Austria; Comprehensive Center for Pediatrics, Medical University of Vienna, Vienna, Austria; Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria.

Natalia Stepien, Division of Neonatology, Pediatric Intensive Care and Neuropediatrics, Department of Pediatrics and Adolescent Medicine, Medical University of Vienna, Vienna, Austria; Comprehensive Center for Pediatrics, Medical University of Vienna, Vienna, Austria; Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria.

Cora Hedrich, Division of Neonatology, Pediatric Intensive Care and Neuropediatrics, Department of Pediatrics and Adolescent Medicine, Medical University of Vienna, Vienna, Austria; Comprehensive Center for Pediatrics, Medical University of Vienna, Vienna, Austria; Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria.

Alicia-Christina Baumgartner, Division of Neonatology, Pediatric Intensive Care and Neuropediatrics, Department of Pediatrics and Adolescent Medicine, Medical University of Vienna, Vienna, Austria; Comprehensive Center for Pediatrics, Medical University of Vienna, Vienna, Austria; Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria.

Hannah Schned, Division of Neonatology, Pediatric Intensive Care and Neuropediatrics, Department of Pediatrics and Adolescent Medicine, Medical University of Vienna, Vienna, Austria; Comprehensive Center for Pediatrics, Medical University of Vienna, Vienna, Austria; Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria.

Abed Ahmadi, Division of Neonatology, Pediatric Intensive Care and Neuropediatrics, Department of Pediatrics and Adolescent Medicine, Medical University of Vienna, Vienna, Austria; Comprehensive Center for Pediatrics, Medical University of Vienna, Vienna, Austria; Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria.

Oliver Eichmüller, Division of Neonatology, Pediatric Intensive Care and Neuropediatrics, Department of Pediatrics and Adolescent Medicine, Medical University of Vienna, Vienna, Austria; Comprehensive Center for Pediatrics, Medical University of Vienna, Vienna, Austria; Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria.

Christine Haberler, Division of Neuropathology and Neurochemistry, Department of Neurology, Medical University of Vienna, Vienna, Austria; Comprehensive Center for Pediatrics, Medical University of Vienna, Vienna, Austria; Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria.

Irene Slavc, Division of Neonatology, Pediatric Intensive Care and Neuropediatrics, Department of Pediatrics and Adolescent Medicine, Medical University of Vienna, Vienna, Austria; Comprehensive Center for Pediatrics, Medical University of Vienna, Vienna, Austria; Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria.

Thomas Czech, Department of Neurosurgery, Medical University of Vienna, Vienna, Austria.

Karl Rössler, Department of Neurosurgery, Medical University of Vienna, Vienna, Austria.

Gregor Kasprian, Division of Neuroradiology and Musculoskeletal Radiology, Department of Biomedical Imaging and Image-Guided Therapy, Medical University of Vienna, Vienna, Austria.

Karl Kircher, Department of Ophthalmology and Optometry, Medical University of Vienna, Vienna, Austria.

Andreas Reitner, Department of Ophthalmology and Optometry, Medical University of Vienna, Vienna, Austria.

Andreas Peyrl, Division of Neonatology, Pediatric Intensive Care and Neuropediatrics, Department of Pediatrics and Adolescent Medicine, Medical University of Vienna, Vienna, Austria; Comprehensive Center for Pediatrics, Medical University of Vienna, Vienna, Austria; Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria.

Johannes Gojo, Division of Neonatology, Pediatric Intensive Care and Neuropediatrics, Department of Pediatrics and Adolescent Medicine, Medical University of Vienna, Vienna, Austria; Comprehensive Center for Pediatrics, Medical University of Vienna, Vienna, Austria; Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria.

Amedeo A Azizi, Division of Neonatology, Pediatric Intensive Care and Neuropediatrics, Department of Pediatrics and Adolescent Medicine, Medical University of Vienna, Vienna, Austria; Comprehensive Center for Pediatrics, Medical University of Vienna, Vienna, Austria; Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria.

Berthold Pemp, Department of Ophthalmology and Optometry, Medical University of Vienna, Vienna, Austria.

Supplementary Material

Supplementary material is available online at Neuro-Oncology Advances (https://academic.oup.com/noa).

Author Contributions

Maria Aliotti Lippolis (Conceptualization, Data collection, Data curation, Statistical analysis, Investigation, Methodology, Writing—original draft, Project administration, Writing—review and editing), Sabine Koinig (Conceptualization, Data collection, Data curation, Methodology, Project administration, Writing—review and editing), Christian Dorfer (Conceptualization, Data collection, Data curation, Writing—review and editing), Luca Conci (Conceptualization, Data curation, Statistical analysis, Writing—review and editing), Karl Kircher (Conceptualization, Writing—review and editing), Andreas Reitner (Conceptualization, Writing—review and editing), Martin Bertich (Conceptualization, Writing—review and editing), Christoph Stapf (Conceptualization, Writing—review and editing), Andreas Peyrl (Conceptualization, Writing—review and editing), Johannes Gojo (Conceptualization, Writing—review and editing), Lisa Mayr (Conceptualization, Writing—review and editing), Irene Slavc (Conceptualization, Writing—review and editing), Thomas Czech (Conceptualization, Writing—review and editing), Karl Rössler (Conceptualization, Writing—review and editing), Gregor Kasprian (Conceptualization, Writing—review and editing), Christine Haberler (Conceptualization, Writing—review and editing), Natalia Stepien (Conceptualization, Writing—review and editing), Cora Hedrich (Conceptualization, Writing—review and editing), Alicia-Christina Baumgartner (Conceptualization, Writing—review and editing), Hannah Schned (Conceptualization, Writing—review and editing), Clara Köller (Conceptualization, Writing—review and editing), Abed Ahmadi (Conceptualization, Writing—review and editing), Oliver Eichmüller (Conceptualization, Writing—review and editing), Amedeo A. Azizi (Conceptualization, Data curation, Supervision, Formal analysis, Investigation, Methodology, Writing—original draft, Project administration, Writing—review and editing, Correspondence), and Berthold Pemp (Conceptualization, Data curation, Supervision, Formal analysis, Investigation, Methodology, Writing—original draft, Project administration, Writing—review and editing)

Conflict of Interest Statement

The authors have no competing interests to declare that are relevant to the content of this article.

Funding

Partial financial support for this study was received from the “Verein unser_kind” (A-1150 Vienna, Austria) (J.G.) and the “Forschungsgesellschaft für Cerebrale Tumore” (A-1180 Vienna, Austria) (M.A.L.).

Ethics Approval Statement

Ethics approval was obtained from the ethics committee of the Medical University of Vienna, and informed consent was waived (EK Nr. 1918/2021).

Data Availability

Deidentified data may be provided from the corresponding author upon written reasonable request by a qualified researcher and upon approval by the data-clearing unit of the Medical University Vienna.

References

  • 1. Brandão LA, Young Poussaint T.  Posterior fossa tumors. Neuroimaging Clin N Am. 2017;27:1-37. 10.1016/J.NIC.2016.08.001 [DOI] [PubMed] [Google Scholar]
  • 2. Udaka YT, Packer RJ.  Pediatric brain tumors. Neurol Clin. 2018;36:533-556. 10.1016/J.NCL.2018.04.009 [DOI] [PubMed] [Google Scholar]
  • 3. Peeler CE, Edmond JC, Hollander J, et al.  Visual and ocular motor outcomes in children with posterior fossa tumors. J AAPOS. 2017;21:375-379. 10.1016/J.JAAPOS.2017.05.032 [DOI] [PubMed] [Google Scholar]
  • 4. Lee JM, Kim SH, Lee JI, Ryou JY, Kim SY.  Acute comitant esotropia in a child with a cerebellar tumor. Korean J Ophthalmol. 2009;23:228-231. 10.3341/KJO.2009.23.3.228 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Girardi F, Di Carlo V, Stiller C, et al. ; CONCORD Working Group. Global survival trends for brain tumors, by histology: analysis of individual records for 67,776 children diagnosed in 61 countries during 2000-2014 (CONCORD-3). Neuro Oncol. 2023;25:593-606. 10.1093/NEUONC/NOAC232 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Simon JW, Waldman JB, Couture KC.  Cerebellar astrocytoma manifesting as isolated, comitant esotropia in childhood. Am J Ophthalmol. 1996;121:584-586. 10.1016/S0002-9394(14)75442-X [DOI] [PubMed] [Google Scholar]
  • 7. Nuijts MA, Stegeman I, Porro GL, et al.  Ophthalmological evaluation in children presenting with a primary brain tumor. J Neuroophthalmol. 2022;42:e99-e108. 10.1097/WNO.0000000000001421 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Azizi AA, Heßler K, Leiss U, et al.  From symptom to diagnosis—the prediagnostic symptomatic interval of pediatric central nervous system tumors in Austria. Pediatr Neurol. 2017;76:27-36. 10.1016/j.pediatrneurol.2017.08.006 [DOI] [PubMed] [Google Scholar]
  • 9. Goldenberg-Cohen N, Ehrenberg M, Toledano H, et al.  Preoperative visual loss is the main cause of irreversible poor vision in children with a brain tumor. Front Neurol. 2011;2:62. 10.3389/FNEUR.2011.00062 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Gadgil N, Edmond J, Stormes K, Lam S, Shah V.  Visual complications of pediatric posterior fossa tumors: analysis of outcomes. Pediatr Neurol. 2019;92:48-54. 10.1016/J.PEDIATRNEUROL.2018.09.016 [DOI] [PubMed] [Google Scholar]
  • 11. Kotecha M, Gotecha S, Chugh A, Punia P.  Neuroophthalmic manifestations of intracranial tumours in children. Case Rep Ophthalmol Med. 2021;2021:7793382-7793388. 10.1155/2021/7793382 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Shalev B, Repka MX.  Restoration of fusion in children with intracranial tumors and incomitant strabismus. Ophthalmology. 2000;107:1880-1883. 10.1016/S0161-6420(00)00345-6 [DOI] [PubMed] [Google Scholar]
  • 13. Xie JS, Donaldson L, Margolin E.  Papilledema: a review of etiology, pathophysiology, diagnosis, and management. Surv Ophthalmol. 2022;67:1135-1159. 10.1016/J.SURVOPHTHAL.2021.11.007 [DOI] [PubMed] [Google Scholar]
  • 14. Wirtschafter JD, Rizzo FJ, Smiley BC.  Optic nerve axoplasm and papilledema. Surv Ophthalmol. 1975;20:157-189. 10.1016/0039-6257(75)90001-6 [DOI] [PubMed] [Google Scholar]
  • 15. Gadgil N, Hansen D, Barry J, Chang R, Lam S.  Posterior fossa syndrome in children following tumor resection: knowledge update. Surg Neurol Int. 2016;7:S179-S183. 10.4103/2152-7806.178572 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Zhang S.  Chemotherapy-induced peripheral neuropathy and rehabilitation: a review. Semin Oncol. 2021;48:193-207. 10.1053/J.SEMINONCOL.2021.09.004 [DOI] [PubMed] [Google Scholar]
  • 17. Dasgupta S, Adilieje C, Bhattacharya A, Smith B, Sheikh M. U H.  Capecitabine and sixth cranial nerve palsy. J Cancer Res Ther. 2010;6:80-81. 10.4103/0973-1482.63555 [DOI] [PubMed] [Google Scholar]
  • 18. Bakbak B, Gedik S, Koktekir BE, et al.  Assessment of ocular neurotoxicity in patients treated with systemic cancer chemotherapeutics. Cutan Ocul Toxicol. 2014;33:7-10. 10.3109/15569527.2013.787087 [DOI] [PubMed] [Google Scholar]
  • 19. Kissoon T, Gururangan S, Sladky J.  Acute neurotoxicity following vincristine due to charcot–marie–tooth disease in a young child with medulloblastoma. Neurooncol Pract. 2019;6:179-184. 10.1093/nop/npz002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Frustaci S, Barzan L, Comoretto R, Tumolo S, Lo Re G, Monfardini S.  Local neurotoxicity after intra-arterial cisplatin in head and neck cancer. Cancer Treat Rep. 1987;71:257-259. [PubMed] [Google Scholar]
  • 21. Hamilton SN, Nichol A, Truong P, et al.  Visual outcomes and local control after fractionated stereotactic radiotherapy for optic nerve sheath meningioma. Ophthalmic Plast Reconstr Surg. 2018;34:217-221. 10.1097/IOP.0000000000000914 [DOI] [PubMed] [Google Scholar]
  • 22. Kortmann RD, Timmermann B, Taylor RE, et al.  Current and future strategies in radiotherapy of childhood low-grade glioma of the brain. Part II: treatment-related late toxicity. Strahlenther Onkol. 2003;179:585-597. 10.1007/S00066-003-8104-0 [DOI] [PubMed] [Google Scholar]
  • 23. Tishler RB, Loeffler JS, Lunsford LD, et al.  Tolerance of cranial nerves of the cavernous sinus to radiosurgery. Int J Radiat Oncol Biol Phys. 1993;27:215-221. 10.1016/0360-3016(93)90230-S [DOI] [PubMed] [Google Scholar]
  • 24. Hasegawa T, Shintai K, Kato T, Iizuka H.  Stereotactic radiosurgery as the initial treatment for patients with nonfunctioning pituitary adenomas. World Neurosurg. 2015;83:1173-1179. 10.1016/J.WNEU.2015.01.054 [DOI] [PubMed] [Google Scholar]
  • 25. Jang SW, Song SW, Kim YH, et al.  Clinical features and prognosis of diffuse midline glioma: a series of 24 cases. Brain Tumor Res Treat. 2022;10:255-264. 10.14791/btrt.2022.0035 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Hennika T, Becher OJ.  Diffuse intrinsic pontine glioma: time for cautious optimism. J Child Neurol. 2016;31:1377-1385. 10.1177/0883073815601495 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Thabet F, Sawahreh M, Thaher D, Al Maadid F.  Teaching video neuroImage: oculomotor apraxia as the only presentation of diffuse intrinsic pontine glioma. Neurology. 2023;101:E854-E855. 10.1212/WNL.0000000000207376 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Bayramova E, Manklow L, Filyridou M, Pilat A.  Ophthalmic presentation of diffuse intrinsic pontine glioma in children (case series and literature review). Neuroophthalmology. 2025;49:200-205. 10.1080/01658107.2024.2402726 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Donaldson SS, Laningham F, Fisher PG.  Advances toward an understanding of brainstem gliomas. J Clin Oncol. 2006;24:1266-1272. 10.1200/JCO.2005.04.6599 [DOI] [PubMed] [Google Scholar]
  • 30. Ciobanu-Caraus O, Czech T, Peyrl A, et al.  The site of origin of medulloblastoma: surgical observations correlated to molecular groups. Cancers (Basel). 2023;15:4877. 10.3390/CANCERS15194877 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Azizi AA, Walker DA, Liu JF, et al. ; SIOPE NF1 OPG Nottingham, UK, Workshop 2014. NF1 optic pathway glioma: analyzing risk factors for visual outcome and indications to treat. Neuro Oncol. 2021;23:100-111. 10.1093/neuonc/noaa153 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. de Blank PMK, Fisher MJ, Liu GT, et al.  Optic pathway gliomas in neurofibromatosis type 1: an update: surveillance, treatment indications, and biomarkers of vision. J Neuroophthalmol. 2017;37:S23-S32. 10.1097/WNO.0000000000000550 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Fisher MJ, Avery RA, Allen JC, et al. ; REiNS International Collaboration. Functional outcome measures for NF1-associated optic pathway glioma clinical trials. Neurology. 2013;81:S15-S24. 10.1212/01.wnl.0000435745.95155.b8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Fisher MJ, Loguidice M, Gutmann DH, et al.  Visual outcomes in children with neurofibromatosis type 1-associated optic pathway glioma following chemotherapy: a multicenter retrospective analysis. Neuro Oncol. 2012;14:790-797. 10.1093/neuonc/nos076 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Nuijts MA, Stegeman I, van Seeters T, et al.  Ophthalmological findings in youths with a newly diagnosed brain tumor. JAMA Ophthalmol. 2022;140:982-993. 10.1001/jamaophthalmol.2022.3628 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Gudrunardottir T, Morgan AT, Lux AL, et al. ; Iceland Delphi Group. Consensus paper on post-operative pediatric cerebellar mutism syndrome: the Iceland delphi results. Childs Nerv Syst. 2016;32:1195-1203. 10.1007/S00381-016-3093-3 [DOI] [PubMed] [Google Scholar]
  • 37. Helmchen C, Rambold H, Kempermann U, Büttner-Ennever JA, Büttner U.  Localizing value of torsional nystagmus in small midbrain lesions. Neurology. 2002;59:1956-1964. 10.1212/01.WNL.0000038387.90128.8D [DOI] [PubMed] [Google Scholar]
  • 38. Girardi F, Allemani C, Coleman MP.  Worldwide trends in survival from common childhood brain tumors: a systematic review. J Glob Oncol. 2019;5:1-25. 10.1200/JGO.19.00140 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Raimondi AJ, Tomita T.  Hydrocephalus and infratentorial tumors. Incidence, clinical picture, and treatment. J Neurosurg. 1981;55:174-182. 10.3171/JNS.1981.55.2.0174 [DOI] [PubMed] [Google Scholar]
  • 40. Grin K, Azizi A, Haberler C, et al.  Cerebellar pilocytic astrocytoma: predictors of recurrence based on MRI morphology—a single-centre experience. Childs Nerv Syst. 2024;41:73. 10.1007/s00381-024-06733-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Ramaswamy V, Hielscher T, Mack SC, et al.  Therapeutic impact of cytoreductive surgery and irradiation of posterior fossa ependymoma in the molecular era: a retrospective multicohort analysis. J Clin Oncol. 2016;34:2468-2477. 10.1200/JCO.2015.65.7825 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Haratek K, Bubeníková A, Entenmann CJ, et al.  Predictors of postoperative complications and functional outcomes in pediatric patients with surgically treated fourth ventricle tumors. Acta Neurochir (Wien). 2023;165:4279-4292. 10.1007/s00701-023-05729-w [DOI] [PubMed] [Google Scholar]
  • 43. Wilne S, Collier J, Kennedy C, Koller K, Grundy R, Walker D.  Presentation of childhood CNS tumours: a systematic review and meta-analysis. Lancet Oncol. 2007;8:685-695. 10.1016/S1470-2045(07)70207-3 [DOI] [PubMed] [Google Scholar]
  • 44. Sønderkaer S, Schmiegelow M, Carstensen H, Nielsen LB, Müller J, Schmiegelow K.  Long-term neurological outcome of childhood brain tumors treated by surgery only. J Clin Oncol. 2003;21:1347-1351. 10.1200/JCO.2003.08.009 [DOI] [PubMed] [Google Scholar]
  • 45. Armstrong GT, Conklin HM, Huang S, et al.  Survival and long-term health and cognitive outcomes after low-grade glioma. Neuro Oncol. 2011;13:223-234. 10.1093/NEUONC/NOQ178 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Khan RB, Patay Z, Klimo P, et al.  Clinical features, neurologic recovery, and risk factors of postoperative posterior fossa syndrome and delayed recovery: a prospective study. Neuro Oncol. 2021;23:1586-1596. 10.1093/NEUONC/NOAB030 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47. Küker W, Bader P, Herrlinger U, Heckl S, Nägele T.  Transient encephalopathy after intrathekal methotrexate chemotherapy: diffusion-weighted MRI. J Neurooncol. 2005;73:47-49. 10.1007/S11060-004-2929-Y [DOI] [PubMed] [Google Scholar]
  • 48. Bay A, Oner AF, Etlik O, Yilmaz C, Caksen H.  Myelopathy due to intrathecal chemotherapy: report of six cases. J Pediatr Hematol Oncol. 2005;27:270-272. 10.1097/01.MPH.0000162527.85024.E9 [DOI] [PubMed] [Google Scholar]
  • 49. Jabbour E, O’Brien S, Kantarjian H, et al.  Neurologic complications associated with intrathecal liposomal cytarabine given prophylactically in combination with high-dose methotrexate and cytarabine to patients with acute lymphocytic leukemia. Blood. 2007;109:3214-3218. 10.1182/BLOOD-2006-08-043646 [DOI] [PubMed] [Google Scholar]
  • 50. Hilgendorf I, Wolff D, Junghanss C, et al.  Neurological complications after intrathecal liposomal cytarabine application in patients after allogeneic haematopoietic stem cell transplantation. Ann Hematol. 2008;87:1009-1012. 10.1007/S00277-008-0546-0 [DOI] [PubMed] [Google Scholar]
  • 51. Williams AS, Hoyt CS.  Acute comitant esotropia in children with brain tumors. Arch Ophthalmol. 1989;107:376-378. 10.1001/ARCHOPHT.1989.01070010386029 [DOI] [PubMed] [Google Scholar]
  • 52. Musazadeh M, Hartmann K, Simon F.  Late onset esotropia as first symptom of a cerebellar tumor. Strabismus. 2004;12:119-123. 10.1080/09273970490517557 [DOI] [PubMed] [Google Scholar]
  • 53. Dikici K, Cicik E, Akman C, Kendiroglu G, Tolun H.  Cerebellar astrocytoma presenting with acute esotropia in a 5 year-old girl. Case report. Int Ophthalmol. 1999;23:167-170. 10.1023/A:1010676913684 [DOI] [PubMed] [Google Scholar]
  • 54. Fieß A, Elflein HM, Urschitz MS, et al.  Prevalence of strabismus and its impact on vision-related quality of life. Ophthalmology. 2020;127:1113-1122. 10.1016/j.ophtha.2020.02.026 [DOI] [PubMed] [Google Scholar]
  • 55. Lee YH, Repka MX, Borlik MF, et al.  Association of strabismus with mood disorders, schizophrenia, and anxiety disorders among children. JAMA Ophthalmol. 2022;140:373-381. 10.1001/jamaophthalmol.2022.0137 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56. Holmes JM, Hercinovic A, Melia BM, et al. ; Pediatric Eye Disease Investigator Group. Improvement in health-related quality of life following strabismus surgery for children with intermittent exotropia. J AAPOS. 2021;25:82.e1-82-e7. 10.1016/j.jaapos.2020.11.021 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

vdag151_Supplementary_Data
vdag151_supplementary_data.zip (2.3MB, zip)

Data Availability Statement

Deidentified data may be provided from the corresponding author upon written reasonable request by a qualified researcher and upon approval by the data-clearing unit of the Medical University Vienna.

Articles from Neuro-Oncology Advances are provided here courtesy of Oxford University Press

RESOURCES