Predecompression and postdecompression cognitive and affective changes in Chiari malformation type I

Abstract

Objective:

The role of the cerebellum in cognitive function and psychiatric symptoms is poorly understood and particularly understudied in patients with cerebellar pathologies such as Chiari malformation type 1 (CM1). Additionally, it is unclear if interventions targeted toward the cerebellum might impact these life-altering symptoms. We sought to characterize pre- and post-operative cognitive and psychiatric function in CM1 patients as evidence for targeted cerebellar treatment for some cognitive and psychiatric conditions.

Methods:

This prospective study included surgical patients with CM1 who reported cognitive or psychiatric dysfunction. Patients completed a pre-operative assessment, and a parallel assessment 6-month following surgery. Neuropsychological evaluations included a 90-minute standardized assessment of cognitive function across multiple domains and a self-reported assessment of psychiatric symptoms.

This clinical sample consisted of 54 patients (M_age=34.17 years, M_ed=14.15 years). Any patient demonstrating pre-operative performance below 3.5 standard deviations (SD) within any cognitive domain were excluded (n=1). All patients underwent pre-operative neuropsychological assessment comprised of standard clinical tests of processing speed, attention, memory, executive function, and psychiatric symptoms.

Results:

Pre-operatively, CM1 patients performed significantly worse than a representative normative sample on measures of executive function and visuospatial memory and reported more psychiatric symptoms across all domains. On postoperative assessment, 89% of patients showed clinically significant improvements (>1 SD) in cognitive and/or psychiatric domains.

Conclusion:

We demonstrate significant, often unrecognized, impairments in cognitive function and psychiatric symptoms in a cohort of CM1 patients. Following targeted surgical posterior fossa decompression, these symptoms improved, suggesting, that at least in a subgroup of symptomatic CM1 patients, these symptoms may be treatable. This study highlights the potential role of the cerebellum in cognitive and psychiatric dysfunction.

Introduction

The role of the cerebellum in cognitive and psychiatric function is not well understood.^1,2 Given the reciprocal connections between the posterior cerebellum and the prefrontal cortex,^3,4 displaced and compressed caudal cerebellar structures might impact function.⁵ Behavior modifying cerebellar lesions in animal models further supports a wholistic view of the cerebellum’s neurological role.^{6 7} Changes in patients with focal cerebellar pathologies (e.g., lesions) result in myriad cognitive and psychiatric changes, termed the “cerebellar cognitive affective syndrome” (CCAS).⁸ CCAS encompasses cognitive impairments (e.g., executive function, visuospatial function, language), and psychiatric symptoms (e.g., depression, anxiety, and somatization disorders) from cerebellar pathology.^9,10

Chiari malformation type 1 (CM1) is a congenital or acquired structural condition where the cerebellar tonsils herniate through the foramen magnum (Figure 1). Tonsillar displacement results in cerebellar and, in some cases, cervico-medullary compression. Common CM1 symptoms include Valsalva-induced occipital headaches and upper extremity paresthesia in addition to a broad spectrum of additional symptoms, including complaints about cognitive function. ^5,11 CM1 results in a stereotypical mechanical compression localized to the foramen magnum. Studies evaluating cognitive function in CM1 demonstrate disparate findings with some commonalities emerging. Pre-operative CM1 patients have been shown to have deficits in memory^12–14 and attention.^14–17 Others investigating post-operative function showed dysfunction in executive function, verbal learning, and language,^13,18,19 and persistent visuospatial memory deficits pre- and post-posterior fossa decompression (PFD).²⁰ However, while some studies employed standard measures,^14,21 conclusions are limited by the use of cognitive screening tools with low sensitivity, single domain analysis (e.g., memory), or the utilization of non-clinically validated measuring tools. Additionally, recent work has demonstrated the presence of psychiatric symptoms (i.e., depression and anxiety) in CM1 patients, highlighting them as potential symptomatologic features.^14,16,21,22

Figure 1.

A) Sagittal MRI T1sequence demonstrating a normal midline cerebellar position relative to the foramen magnum (red line); B) Axial MRI T2 sequence demonstrating CSF (white) signal around the cervico-medullary junction; C)Sagittal MRI T1 sequence demonstrating a Chiari malformation with cerebellar descent below the foramen magnum (red line); D) Axial MRI T2 sequence demonstrating profound loss of CSF volume at the level of the foramen magnum due to cerebellar tissue crowding; E) Sagittal MRI T1 sequence demonstrating postoperative expansile duraplasty and craniectomy providing space for unobstructed CSF flow and dramatic improvement in cerebello-medullary compression; and F) CT myelogram demonstrating wide bony decompression of the foramen magnum promoting ventral and dorsal CSF continuity.

Objectives

The present study characterizes pre- and post-operative cognitive and psychiatric function in a subset of CM1 patients who report typical physical symptoms of CM1 and/or cognitive and psychiatric symptoms. Specifically, we assess cognitive function and psychiatric symptoms across multiple domains and explore the impact of PFD on within-individual changes pre- to post-operatively and provide evidence for a potential role of cerebellar-mediated pathology in cognitive and psychologic dysfunction in at least a subpopulation of CM1 patients.

METHODS

Study Design: Patients in the current study presented for evaluation at the University of Pittsburgh from 2015 to 2023. All patients reporting cognitive or psychiatric symptoms were offered neuropsychological testing (NPT) pre- and postoperatively and completed a comprehensive neurological examination, including MRI. Post-operative imaging was completed 3- and 6-months postoperatively. Of the 135 patients who completed preoperative NPT, 55 returned for post-operative NPT. There were no demographic or neuropsychological differences between those patients that returned versus those that did not (Supplemental Table 1). A diagnosis of symptomatic CM1 was made based on reported symptoms and imaging. All patients parallel underwent PFD with a single surgeon (RMF). The neuropsychological battery was administered pre-operatively and patients were asked to return 6-months post-operatively to complete a assessment. Patients demonstrating pre-operative performance below 3.5 standard deviations (SD) on any measure were excluded to eliminate the impact of extreme outliers and preserve a normal distribution and verified using a modified z-score method which identified the same outlier. ²³ Thus, 54 patients were included (M_age=34.2 years; range=16-57 years, M_ed=14.2 years) (Supplemental Table 1). This prospective study was approved by the University of Pittsburgh Institutional Review Board and all patients provided written informed consent.

Study Design and Procedures

The goal of surgery in patients with symptomatic CM1 is to provide space in the posterior fossa to eliminate direct pressure on the inferior cerebellum, medulla oblongata and upper spinal cord ²⁴ and by reestablishing unobstructed CSF flow. Surgical goals were achieved by bone decompression of the foramen magnum, intradural exploration using an operating microscope with lysis of arachnoid of adhesions, tailored cautery tonsillar volumetric reduction and expansile duraplasty. Progressively more during the study period, in most patients the C1 arch was preserved with intentional shaving of the top of the arch. Complete removal of the dorsal aspect of the C1 arch was reserved for patients with tonsillar herniation rostral to the C1 arch. Prior to PFD surgery, patients completed a standardized neuropsychological assessment of cognitive function and psychiatric symptoms (see Supplementary Material). Six months post-surgery, 55 patients returned to complete a subsequent neuropsychological assessment. This included 1 patient that was excluded due to pre-operative performance that fell below 3.5 standard deviations (SD) on two measures, resulting in a sample of 54 (M_age=34.2 years; SD; 10.6; range=16-57 years).

Neuropsychological Assessment

Cognitive function was evaluated across several domains: executive function (set-shifting, inhibition, feedback integration), attention (visual scanning, simple attention, complex attention), processing speed (psychomotor speed, information processing), visuospatial memory (immediate recall and delayed recall), verbal memory (verbal learning, delayed recall), and language (phonemic fluency, semantic fluency) (see Supplementary Material). This battery was administered by a neuropsychologist (LCH) preoperatively and approximately 6-months postoperatively. All raw scores were converted to standardized z-scores, which represent performance relative to age-, education-, and sex-based normative samples. To assess potential within-individual change, scores were examined pre- and post-operatively.

Psychiatric Symptoms were assessed using a self-report questionnaire which includes 53-items assessing psychiatric symptoms across multiple domains including depression, anxiety, somatization, obsessive-compulsive, and hostility (see Supplementary Material).

Data Analysis

All data were analyzed using SPSS V28. First, to characterize cognitive function and psychiatric symptoms pre-operatively, z-score means and standard deviations (SD) are reported for all measures and represent performance relative to a representative normative sample, with 0 representing the normative sample mean. Next, one-sample t-tests were used to assess whether any measures of cognitive function or psychiatric symptoms significantly differed from the norm-based means. Finally, to assess within-individual change from pre- to post-operative assessment (i.e., post-operative score – pre-operative score), paired sample t-tests were utilized. A false discovery rate (FDR) correction was applied for all statistical comparisons. For domains with significant pre-operative deficits and significant pre- to post-operative change, analyses were conducted to characterize patterns of clinically significant improvement, defined as change > 1 SD, a commonly used metric.²⁵ Chi-square tests were used to examine whether pre-operative scores (below versus above the median) were associated with clinically significant improvements (> 1SD). The current study was not pre-registered. Demographic characteristics are described in Supplemental Table 1.

RESULTS

Pre-operative Assessment

Table 1 presents means and SD for pre-operative assessments across all cognitive and psychiatric domains, as well as results from one-sample t-tests. A False Discovery Rate correction was implemented to control multiple comparisons.

Table 1.

pre/Post-operative Cognitive Function and Psychiatric Symptoms in Surgical Patients with CM1

	Norm-Based Comparisons
(N=54)	Pre-operative				Post-operative				Pre- to Post-operative Change
	Mean	SD	t	p	Mean	SD	t	p	Mean ∆	SD	t	p
Estimated IQ	0.21	0.75	2.08	0.28	0.33	0.74	1.2	.09	0.12	0.29	−2.39	.26
Cognitive Domains
Executive Function
Set Shifting	0.19	1.14	1.20	0.28	0.47	0.99	3.40	0.004	−0.27	0.89	−2.22	0.063
Inhibition	0.23	0.69	2.44	0.03	0.44	0.86	3.78	0.001	−0.21	0.70	−2.21	0.063
Feedback Integration	−0.42	0.85	−3.65	0.002	−0.29	0.79	−2.7	0.02	−0.13	0.71	−1.37	0.235
Attention
Visual Scanning	0.24	1.06	1.68	0.14	0.28	1.03	1.99	0.086	0.04	0.86	−.31	0.796
Simple Attention	−0.22	0.85	−1.87	0.11	−0.21	0.85	−1.79	0.122	0.01	0.79	−.08	0.936
Complex Attention	0.04	0.92	0.34	0.74	0.10	0.94	.77	0.525	0.06	0.97	−4.2	0.752
Processing Speed
Psychomotor Speed	0.46	0.98	3.43	0.003	0.67	0.81	6.03	<.0001	0.21	−1.67	−0.39	0.184
Information Processing	−0.17	0.94	−1.35	.23	−0.07	0.97	−.52	0.67	0.10	−.75	0.42	0.551
Visuospatial Memory
Immediate Recall	−0.53	1.28	−3.04	.007	0.49	0.92	3.87	0.001	1.02	1.37	−5.44	<.001
Delayed Recall	−0.51	1.17	−3.18	.005	0.43	0.96	3.27	0.005	0.94	1.32	−5.22	<.001
Verbal Memory
Verbal Learning	0.15	1.17	.93	0.39	0.51	1.29	2.88	0.013	0.36	1.10	−2.38	0.052
Short Delay	−0.26	1.16	−1.64	0.14	0.01	1.33	.05	0.96	0.27	1.28	−1.54	0.208
Long Delay	−0.29	1.23	−1.71	0.65	−0.06	1.23	−.33	0.78	0.23	1.22	−1.39	0.235
Language
Phonemic Fluency	0.07	1.04	.50	0.65	0.15	1.09	1.01	0.40	0.08	0.80	−7.3	0.551
Semantic Fluency	0.43	0.99	3.19	0.05	0.63	1.18	3.95	0.001	0.20	0.99	−1.52	0.208
Psychiatric Domains
Depression	0.66	1.02	4.76	<.0001	0.16	1.04	1.15	0.34	−0.50	0.89	4.14	0.0004
Anxiety	0.87	1.08	5.94	<.0001	0.31	1.15	1.99	0.09	−0.56	0.95	4.34	0.0003
Somatization	1.62	0.96	12.43	<.0001	0.81	1.19	5.01	<.0001	−0.81	1.11	5.37	<.0001
Obsessive Compulsive	1.81	1.06	12.55	<.0001	1.03	1.19	6.39	<.0001	−0.77	1.16	4.89	<.0001
Hostility	0.69	0.90	5.66	<.0001	0.21	0.99	1.5	0.19	−0.49	0.86	4.17	0.0004

Cognitive Function

Patients exhibited lower performance relative to the normative sample on an executive function measure of feedback integration, information processing and visuospatial memory (immediate recall and delayed recall). Patients in the current sample performed better than the normative sample on a processing speed measure of simple psychomotor speed and on a language measure of semantic verbal fluency.

Psychiatric Symptoms

In the sample, patients were significantly more impaired than the normative sample across all domains including depression, anxiety, somatization, obsessive-compulsive, and hostility.

Post-operative Assessment

Analyses were conducted to characterize patterns of clinically significant change following surgery, defined as change >1 SD. Most patients in the sample (48/54, 89%) showed clinically significant improvements in either cognitive or psychiatric domains. Approximately one-third of patients (17/54) demonstrated improvements in both domains by the 6-month follow-up assessment. Within the sample, 55% (30/54) of patients demonstrated clinically significant improvement in visuospatial memory (immediate recall, delayed recall). Similarly, 65% (35/54) of patients demonstrated clinically significant improvement in at least one psychiatric domain.

Table 1 presents means and SD for post-operative assessments across all cognitive and psychiatric domains. Paired-samples t-tests were conducted to determine the impact of surgery on cognitive function and psychiatric symptoms. A false discovery rate correction was implemented to control for multiple comparisons.

Cognitive Function

Patients demonstrated improvements on visuospatial memory measures of immediate recall (Figure 2, 2A; upper) and delayed recall (Figure 2, 2B; lower).

Figure 2.

Pre- and post-operative patient performance on a visuospatial memory task. Each pair of blue (pre-operative) and red (post-operative) dots on the X-ordinate represent scores from the same patient, with lower z-scores indicating poorer performance and higher z-scores indicating better performance. Panel insets show the paired-sample t-test demonstrating significant within-individual improvements in visual memory immediate delay (2A; upper), and delayed recall (2B; lower) by 6-month post-operative follow-up.

Psychiatric Symptoms

Patients evidenced significant improvements across psychiatric domains of anxiety (Figure 3, 3B), somatization (Figure 3, 3C), obsessive-compulsive (Figure 3D), and hostility symptoms (Figure 3, 3E). Pre-operatively, patients reported significant distress in each of these domains when compared to the normative sample (Table 1); however, post-operatively (Table 1), depression, anxiety, and hostility symptoms no longer differed from the normative sample. Most patients suffering multiple psychiatric symptoms improved in multiple domains (Supplemental Table 3). Post-operative somatization and obsessive-compulsive symptoms also improved, although they remained clinically elevated relative to the normative sample.

Figure 3.

Pre- and post-operative symptom levels across all psychiatric domains: depression (A), anxiety (B), somatization (C), obsessive-compulsive (D), and hostility (E). Each pair of blue (pre-operative) and red (post-operative) dots on the X-ordinate represent scores from the same patient, with higher z-scores indicating more symptoms and lower z-scores indicating fewer symptoms. Panel insets show results from the paired-sample t-test and demonstrate significant within-individual improvements across all psychiatric domains by 6-month post-operative follow-up.

Magnitude of Effect

Chi-square tests were conducted to assess associations between pre-operative function and the magnitude of clinically significant improvements. Patients who performed the worst preoperatively evidenced the most clinically meaningful improvement (>1SD). Among patients whose visuospatial memory performance was below the mean preoperatively, 78% (21/26 patients; χ=18.9, p<.001) demonstrated clinically meaningful improvement on immediate recall and 77% (20/26 patients; χ=14.5, p<.001) on delayed recall. Within neuropsychiatric symptoms, two domains emerged showing significant association between preoperative distress and degree of change. First, those with higher reported symptoms of depression evidenced the greatest reduction of associated distress postoperatively with 82% (9/11; χ=7.0, p=.008) evidencing clinically meaningful change. Likewise, 79% (15/19) patients with the greatest degree of distress over cognitive complaints showed the greatest reduction of symptom burden (χ=6.5, p=.021).

DISCUSSION

We demonstrate that a patient population suffering from an isolated cerebellar pathology differ from a representative normative sample across domains of cognitive and psychiatric function. In this cohort, 89% (48/54 patients) of CM1 patients with cognitive or psychiatric symptoms who underwent PFD evidenced clinically significant improvement (> 1 SD) in cognitive function and/or reductions in psychiatric symptoms. Specifically, 56% (30/54) of patients evidenced improved cognitive function, 64% (35/64) evidenced significant reduction in psychiatric symptoms, and 31% (17/54) evidenced improved cognitive function and reduced psychiatric symptoms.

The current results join burgeoning work demonstrating the underestimated role of the cerebellum in cognition^{12–14,18,20,26} and affect.^22,27–29 Cognitive function across multiple domains (executive function, visuospatial memory) differed relative to normative samples. This is broadly consistent with other studies^{13–16,18,21,30} showing differences in executive function,^13,16,18,21 and visuospatial memory.^20,31 Additionally, elevated psychiatric symptoms were observed pre-operatively, including depression, anxiety, somatization, obsessive-compulsive, and hostility, broadly consistent with other work.^18,22,27 Our observation of diminished visuospatial memory is consistent with the role of the cerebellum in visuospatial organization/memory^20,32,33 and prior work showing the role of the cerebellum in implicit learning and visuospatial processing and attention.^34–36 While practice effects and regression (or progression) to the mean are ever-present factors in interpreting change, the stability of all other cognitive measures along suggests that the improved visuospatial memory is not primarily driven by these factors; furthermore, lower scores may be less prone to drifting toward the norm.²⁵

Critically, the patients in this cohort exhibit significant psychiatric symptoms, with higher levels of depression, anxiety, somatization, obsessive compulsive, and hostility relative to normative samples. Several studies have also shown affective dysfunction in CM1 patients ^{13,18,22,27,30,31} with anxiety and depression commonly reported. Changes in affect are well described as part of the manifestation of injury and pathology in the vermis and perivermian regions of the cerebellum,^10,37,38 playing a critical role in affect regulation via direct connections to the ventral tegmental and limbic areas.^10,38,39 Accordingly, the lack of broad recognition of psychiatric distress in such patients may result in misinterpretation or misattribution of their presentation as psychosomatic by medical providers.

While some studies have compared pre- and post-operative cognitive function in CM1,^{12,13,20–22,27,40} only one other study has used a within patient design.²⁰ Unique to our study, patients who performed the most poorly preoperatively demonstrated improved visuospatial memory. The other study employing a similar within group design, did not show improvement of visuospatial deficits, nor were changes seen across other domains. ²⁰ Although the potential impact of a placebo effect cannot be ruled out, a key surgical difference may provide additional insight. Bone only decompression provides some decompression with purportedly less morbidity but has an inferior efficacy rate due to a less thorough decompression.⁴¹ Seaman et al²⁰ provides fortuitus insight as most patients (13/19) in that study underwent a bone-only decompression and while all experienced resolution of their headaches, no improvement in cognitive or psychiatric function was observed in any patient, suggesting that headache resolution in it of itself did not result in cognitive or psychiatric improvement. In the Seaman et al study, the visuospatial memory test results are not reported between the different (intra- versus extradural) surgical groups and the much smaller sample may also account for their report stability in cognitive functions. By contrast, in the current study 30/54 patients evidenced significantly improved visuospatial memory, all of whom underwent an intradural procedure with direct cerebellar decompression and thorough confirmation of CSF pathways at the fourth ventricle. During each surgery, extensive dissection of the foramen magnum to the edge of the paracondylar protuberances is performed, with all medial bone being removed. A wide duraplasty is performed with additional decompression provided by lysing all arachnoid tonsillar adhesions, fourth ventricle obstructive membranes were resected when present, and when appropriate, gentle cauterization of the pia to promote rostral contraction to achieve focal foramen magnum decompression. The difference between bone-only and intradural decompression may be related to a greater degree of decompression,⁴² thereby lowering the likelihood that the current results are purely the result of a placebo effect. Interestingly, in the current sample, patients who performed most poorly pre-operatively, generally demonstrated the greatest improvements postoperatively. The predictive utility of cognitive factors on surgical outcome and recovery remains unclear, though continued research may provide important prognostic information.

Studies have demonstrated behavior changes, especially aggression, with cerebellar lesioning in animal models. ^6,7 Others have identified cerebellar dysfunction as a contributor to social inflexibility and dysfunction in a range of conditions.⁴³ Higher levels of psychiatric symptoms are commonly reported in patients with cerebellar pathology. For example, emotional lability is classically associated with iatrogenic injury to the cerebellar vermis.⁴⁴ In CM1 patients,^14,16,21,27 a small number of studies have demonstrated post-PFD changes in elevated psychiatric symptoms.^22,27 In this cohort, all psychiatric domains (depression, anxiety, somatization, obsessive-compulsive, and hostility) showed significant improvement after PFD. Following PFD, depression, anxiety, and hostility normalized, and while somatization and obsessive-compulsive symptoms improved, they did not normalize.

One common challenge to surgical studies is the rarity of a sham control group, and the same is true of the current study. Insight on the magnitude of placebo effects on psychiatric symptoms can be taken from the deep brain stimulation literature.^45,46 For example, Holtzheimer et al.⁴⁶ found that at 6 months, 20% of sham patients reported a 40% or greater improvement in depression symptoms, corresponding to approximately a 1SD improvement among cohort participants. This is lower than the 65% of patients with greater than 1SD in at least one psychiatric domain and 58% with the same magnitude of improvement in at least one cognitive domain within our cohort. Furthermore, the fact that all patients in the Seaman study²⁰ experienced headache resolution and no improvement in cognitive or psychologic symptoms provide further evidence that the cognitive and psychologic improvement in our study is unlikely to be a sole function of headache reduction/resolution.

The precise mechanism of CM1-induced cognitive and psychiatric alterations and PFD related improvement is unclear. The differential impacts of episodic elevated intracranial pressure (ICP) and direct mechanical compression of the cerebellum on neural function cannot be delineated based on the current results, nor can the postoperative improvements be attributed to the alleviation of one over the other. Persistent or transient ICP elevation in CM1 may impact cognition and psychiatric symptoms. Future work investigating the interplay between neuroanatomical and cognitive aspects of CM1 will offer valuable prognostic indicators for PFD outcome. While open questions remain, we demonstrate that cognitive and psychiatric dysfunction, often underappreciated components of the condition, affect patients with cerebellar-specific pathologies such as CM1. Critically, this dysfunction is not fixed and may be reversible in some cerebellar conditions. This work provides a foundation for more in-depth studies to delineate the proportion of patients with cognitive and psychiatric symptoms attributable to cerebellar pathologies to help delineate a specific syndrome in this condition and likelihood of improvement with PFD or other targeted cerebellar therapies.

LIMITATIONS

This study is limited by the lack of a control group. Adding healthy and nonsurgical control groups will help delineate precise aspects of CM1, separating nonspecific aspects of headache pain, psychiatric symptoms, and surgical intervention. Also, a specific headache pain scale was not used for all patients and including this is critical for future work given the prevalence of this symptom and its valence (along with other physical symptoms) as a part of outcome. Pain specifically is important to consider as a factor in the improved outcomes after PFD where chronic pain does exert influence on cognitive performance across multiple cognitive domains such as processing speed, attention, and memory.²³ Pain/symptom reduction is likely a factor in the current results, though the specific nature of the cognitive deficits and changes in the current study and the absence of more broad cognitive dysfunction reflects a more CM1-specific finding..⁴⁷ Pain/symptom reduction is likely a factor in the current results, though the specific nature of the cognitive deficits and changes in the current study and the absence of more broad cognitive dysfunction reflects a more CM1-specific finding. When noted in the chart, pre- and post-op headaches, NIH pain scales, presence of dysphagia, and upper extremity paresthesias were compared and all statistically improved (Supplemental Table 2). Importantly, not all patients returned for postoperative neuropsychological evaluation, potentially biasing the magnitude of the effect. However, it is clear from the data that at least a proportion of patients with cognitive or psychiatric symptoms improve following PFD and that our available headache data is equivalent between those that returned for post-operative neuropsychological assessment and those who did not (Supplemental Table 2). In the same vein, this cohort consists of patients who are self-reporting cognitive dysfunction and therefore may not be reflective of all patients with CM1 and may also present a confirmation bias given their reported symptomatology. That said, perceived cognitive dysfunction does not necessarily equate to objective deficits as reported cognitive problems do not necessarily correlate with measured function in neurological populations.^48,49 We acknowledge that at present, we do not endorse posterior fossa decompression in CM1 patients with neuropsychiatric complaints but without any of the other well established CM1 symptoms. This is an important question which requires additional study. This is an important question which requires additional study. Additionally, while methodological advantages exist from a single surgeon/institution study, generalizability is a relative limitation. Finally, the current study is unable to comment specifically about direct physiological changes from surgery (e.g., tractography) on cognition, a promising avenue of future study.

CONCLUSION

We demonstrate significant, often unrecognized, impairments in cognitive function and psychiatric symptoms in a cohort of patients with focal cerebellar pathology. Following targeted surgical intervention relieving cerebellar compression and thorough reestablishment of CSF flow, these symptoms can be alleviated. This study highlights the role of the cerebellum in cognitive and psychiatric dysfunction and suggests that cerebellar pathology should be considered as a potential neuroanatomical site for similar dysfunction from other etiologies.