Transient cerebellum-related network disruption characterizes postoperative cognitive changes after dominant hemisphere subtemporal meningioma resection Authors: Ziqi Liu, Yangyang Li, Jiwei Jiang, Jun Xu, Yaou Liu, Junting Zhang, Zhen Wu, Shaowu Li, Liang Wang Published: 2026 ABSTRACT AbstractBackgroundCognitive impairment following skull base meningioma resection remains poorly understood. This prospective study aimed to investigate the cognitive function alterations and resting-state fMRI (rs-fMRI) characteristics in patients undergoing extradural subtemporal approach meningioma resection.MethodsThis study enrolled 23 right-handed primary petroclival meningioma patients from June 2024 to November 2024. Participants underwent cognitive assessments combined with rs-fMRI scans at three time points (Time1: one day before surgery; Time2: one week after surgery; and Time3: three months after surgery), utilizing comprehensive test scales including the Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), Symbol Digit Modalities Test (SDMT) and Trail Making Test (TMT-A & TMT-B). Alterations of low-frequency fluctuations (ALFF), fractional ALFF (fALFF), regional homogeneity (ReHo), and functional connectivity (FC) were calculated. Patient demographic characteristics, medical records, and neuroimaging data were systematically collected and analyzed. Independent samples t-tests, Mann–Whitney U tests, Fisher’s exact tests and paired samples t-tests were used in statistical analysis.Results13 patients were in the non-dominant side (NDS) group, and 10 patients were in the dominant side (DS) group. There were no statistically significant differences in preoperative baseline (Time1) clinical characteristics and preoperative cognitive assessment outcomes between the two groups. Patients in the DS group showed marked decline in MMSE (mean Δ = -6.4 points, corrected p = 0.009), MoCA (mean Δ = -5.1 points, corrected p = 0.047), and SDMT scores (mean Δ = -17.8 points, corrected p = 0.003), along with prolonged TMT-A completion times (mean Δ = + 32.7 s, corrected p = 0.017) at Time2. Six patients (60%) showed significant cognitive decline, while four patients (40%) maintained cognitive performance stable. Patients in the NDS group showed no significant postoperative cognitive decline. The rs-fMRI analysis revealed three characteristic alterations in cognition impaired patients: decreased ALFF in left thalamus, reduced ReHo in left middle temporal gyrus and lowered fALFF in left middle frontal gyrus (all p < 0.05). Significant weakening of functional connectivity between these regions and cerebellar networks were observed, and they show marked improvement in parallel with cognitive recovery at three months postoperatively.ConclusionsPatients undergoing dominant hemisphere extradural subtemporal approach for petroclival meningioma resection showed a propensity for transient postoperative declines in verbal memory, orientation, and executive function. These cognitive changes were associated with reduced spontaneous cortical activity and attenuated functional connectivity within cerebello-cortical and cerebello-subcortical networks. Both cognitive performance and these neural alterations demonstrated improvement at the three-month postoperative follow-up, these preliminary observations await further validation in larger independent cohorts.Supplementary InformationThe online version contains supplementary material available at 10.1007/s00701-026-06926-z.Keywords: Skull base meningioma, Extradural subtemporal approach, Postoperative cognitive changes, Resting-state fMRI, Functional connectivity Introduction Skull base meningiomas are deeply situated tumors with complex anatomical relationships to critical surrounding structures. Previous studies have identified that a subset of patients with skull base meningiomas exhibit cognitive decline during the perioperative period, manifesting as memory impairment, attentional deficits, and other neurocognitive disturbances [1, 23, 41]. These findings demonstrate that skull base meningioma patients may experience neurocognitive impairment during both perioperative and recovery phases. However, longitudinal studies tracking neurocognitive function throughout the perioperative and recovery periods remain scarce in patients with petroclival meningioma undergoing the extradural subtemporal approach, particularly regarding systematic analyses of dynamic cognitive changes and associated functional neuroimaging characteristics. Neurocognitive function, representing the core component of higher-order cerebral processing, encompasses multiple complex domains including memory, attention, language, executive function, and spatial perception. The past decade has witnessed remarkable advancements in neuropsychological assessment methodologies. Standardized instruments such as the Mini-Mental State Examination (MMSE) [11], Montreal Cognitive Assessment (MoCA) [24], Symbol Digit Modalities Test (SDMT) [8], and Trail Making Test (Parts A and B) [6] have become indispensable tools in contemporary clinical neurocognitive assessment practice. The progression of neuroimaging technologies, particularly the growing utilization of resting-state functional magnetic resonance imaging (rs-fMRI), has revolutionized our capacity to delineate the intricate associations between brain functional networks and cognitive processes [21, 26]. The rs-fMRI enables direct observation of abnormal activation/inhibition patterns and functional connectivity alterations across brain regions during various cognitive states. These developments have substantially deepened our comprehension of cognitive mechanisms while creating novel opportunities for both anatomical localization of cognitive deficits and prognostic evaluation of cognitive recovery in clinical settings [3, 35]. Although previous clinical studies have reported the phenomenon of cognitive risks among meningioma patients, the underlying neural mechanisms and their dynamic patterns over time remain unclear. To address this gap, we conducted a prospective, longitudinal study combining detailed neuropsychological assessments with multimodal rs-fMRI to comprehensively examine the dynamic alterations of cognitive function and rs-fMRI metrics during perioperative and recovery phases in petroclival meningioma patients treated via the extradural subtemporal approach. Methods Inclusion/exclusion criteria The inclusion criteria comprised of the following. Consecutive collection of patients with primary petroclival meningioma who scheduled for extradural subtemporal approach surgical treatment in the Beijing Tiantan Hospital Neurosurgical Department. Consecutive collection of patients with primary petroclival meningioma who scheduled for extradural subtemporal approach surgical treatment in the Beijing Tiantan Hospital Neurosurgical Department. Age between 18 and 65 years, right-handed, with the participant or their family members providing signed informed consent and voluntary participation. Age between 18 and 65 years, right-handed, with the participant or their family members providing signed informed consent and voluntary participation. The exclusion criteria included the following, Significant visual impairment affecting cognitive assessment compliance. Significant visual impairment affecting cognitive assessment compliance. Participants aged < 18 years or > 65 years, or women who are pregnant or breastfeeding. Participants aged < 18 years or > 65 years, or women who are pregnant or breastfeeding. Preoperative use of medications such as vasoactive drugs and psychotropic agents that may interfere with cognitive assessments. Preoperative use of medications such as vasoactive drugs and psychotropic agents that may interfere with cognitive assessments. Patients with a KPS score below 70 were excluded from cognitive assessments, as significant physical impairment or systemic illness at this functional level may confound reliable neuropsychological testing and reflect non-neurological causes of poor performance. Patients with a KPS score below 70 were excluded from cognitive assessments, as significant physical impairment or systemic illness at this functional level may confound reliable neuropsychological testing and reflect non-neurological causes of poor performance. Participants with postoperative MRI-confirmed temporal lobe or cerebellar contusions and/or hematomas. Participants with postoperative MRI-confirmed temporal lobe or cerebellar contusions and/or hematomas. This prospective study initially enrolled 30 patients with petroclival meningioma. After excluding seven patients who did not meet the inclusion criteria, a total of 23 patients were eligible for analysis (Fig. 1). Flow diagram of patient enrollment and follow-up Demographic and clinical information collection Detailed epidemiological characteristics, medical records, and neuroimaging data of all enrolled patients were collected, including: patients’ age, gender, educational attainment (quantified in years of formal education), handedness (categorized as right, left, or ambidextrous), tumor location, tumor maximum diameter (measured in millimeters on contrast-enhanced T1-weighted MRI sequences), presence and severity of peritumoral edema, hydrocephalus status, preoperative (Time1) KPS and baseline cognitive assessment outcomes, surgical approach, postoperative (Time2) KPS and cognitive assessment outcomes, and recovery phase KPS and cognitive assessment outcomes three months after surgery (Time3). The data are summarized in Table 1. Handedness was assessed using the Edinburgh Handedness Inventory. All patients in this study were right-handed. Accordingly, the left hemisphere was considered the dominant hemisphere for language and cognitive functions [16, 19]. Table 1Demographic and clinical characteristicsDS GroupNDS Groupp Valuesn = 23n = 10n = 13Age (year)0.813Mean49.5 ± 10.148.6 ± 5.7Gender (n)0.104Male04Female109Educational attainment (year)0.295Mean13.1 ± 4.311.5 ± 3.1Tumor maximum diameter (mm)0.739Mean35.5 ± 16.339.2 ± 10.5Peritumoral edema (n)0.604Present13Absent910History of hydrocephalus (n)0.179Present15Absent98Time1 KPS0.380Median (Q1, Q3)90.0 (90.0, 90.0)90.0 (90.0, 90.0)Time2 KPS0.380Median (Q1, Q3)80.0 (80.0, 80.0)80.0 (80.0, 80.0)Time3 KPS0.088Median (Q1, Q3)85.0 (80.0, 90.0)90.0 (90.0, 90.0)"n" represents the sample number; p Values are from independent samples t‑tests, Mann–Whitney U tests or Fisher’s exact tests Demographic and clinical characteristics "n" represents the sample number; p Values are from independent samples t‑tests, Mann–Whitney U tests or Fisher’s exact tests The cohort comprised two surgically defined groups: 10 patients (43.5%) undergoing dominant hemisphere (left-sided) extradural subtemporal approach and 13 patients (56.5%) receiving non-dominant hemisphere (right-sided) procedure. There were 10 females in the DS group, four males and nine females in the NDS group. The DS group demonstrated a mean age of 49.5 ± 10.1 years, formal education of 13.1 ± 4.3 years, and tumor maximum diameter of 35.5 ± 16.3 mm, while the NDS group showed a mean age of 48.6 ± 5.7 years, formal education of 11.5 ± 3.1 years, and tumor maximum diameter of 39.2 ± 10.5 mm. No significant intergroup differences in baseline characteristics were observed, including mean age, gender distribution, educational attainment, tumor maximum diameter, peritumoral edema severity, hydrocephalus status, preoperative Karnofsky Performance Status (KPS), or postoperative KPS scores at one-week and three-month follow-ups (Table 1). Cognitive assessment and rs-fMRI scan protocol All participants were scheduled for comprehensive standardized cognitive assessments plus rs-fMRI scans at three time points (Time1: one day before surgery; Time2: one week after surgery; and Time3: three months after surgery), utilizing comprehensive test scales including the MMSE for global cognitive screening, MoCA for detecting subtle cognitive impairment, SDMT for assessing information processing speed, and TMT-A & TMT-B for evaluating executive function and attention control. Cognitive assessments were administered by two trained doctors (Z.L. and J.J.). The same examiners conducted all assessments across the three time points, and all cognitive assessments were conducted through in-person evaluations. Patients underwent rs-fMRI scanning while awake, in a quiet environment with eyes closed. Imaging was performed on a 3.0 T MRI system (uMR 880, UIH, China) at Beijing Tiantan Hospital, Capital Medical University (Beijing, China). T1-weighted images were obtained using an inversion recovery gradient echo sequence (field-of-view = 240 mm, 196 slices, 0.5 mm isotropic voxels). Whole-brain functional echo-planar images (EPIs) were acquired using a 25-channel coil (TR = 2000 ms, TE = 30 ms, flip angle = 90 voxels). The acquired DICOM-format raw data were subsequently exported for analysis. Additionally, two blinded investigators independently reviewed all surgical documentation and postoperative structural MRI scans to identify and exclude cases with evidence of procedure-related organic brain injury. Surgical procedure All surgeries were performed by one of three senior neurosurgeons (Z.W., J.Z., or L.W.). The extradural subtemporal approach was performed with the patient in the supine position, head rotated contralaterally. A frontotemporal craniotomy was followed by extradural dissection of the middle fossa floor to expose the petroclival region. Intraoperative neurophysiological monitoring (including somatosensory and motor evoked potentials) was routinely used. Postoperative MRI was obtained within 72 h to confirm the extent of resection and assess for complications. Statistical analysis and rs-fMRI processing Continuous variables were compared between groups using the independent samples t‑tests (for normally distributed data) or the Mann–Whitney U tests (for non‑normally distributed data), with normality assessed by the Shapiro–Wilk tests. Normally distributed continuous data are expressed as mean ± standard deviation, while non-normally distributed data are presented as median (first quartile, third quartile). Categorical variables were analyzed using the Fisher’s exact tests. Bonferroni correction was applied where applicable. Statistical analyses were performed using IBM SPSS Statistics 26.0 (Armonk, NY, USA), with a two-tailed p-value < 0.05 considered statistically significant. For rs-fMRI data processing, we utilized the RESTplus toolbox implemented in MATLAB [17]. The analysis focused on four key metrics [9]: alterations of low-frequency fluctuations (ALFF) [39], fractional ALFF (fALFF) [40], regional homogeneity (ReHo) [4], and functional connectivity (FC) [10]. Paired tests were performed to identify significant changes in these rs-fMRI metrics across perioperative and recovery period, with a significance threshold set at p < 0.05, the false discovery rate (FDR) method was applied at q < 0.05. Statistically significant results were visualized by overlaying them onto a standardized, tumor-removed T1-weighted Colin 27 brain template [15]. Functional connectivity patterns were further examined and displayed using BrainNet Viewer (http://www.nitrc.org/projects/bnv/). Results All participants maintained KPS scores > 70 throughout the study period. Pathological examination confirmed World Health Organization Grade 1 meningiomas in all 23 cases, with three-month follow-up data available for six of DS group and nine of NDS group. Loss to follow-up occurred in eight patients who underwent postoperative monitoring at local hospitals due to geographical constraints and were unable to return to our center. Consequently, cognitive assessment and fMRI data at Time3 were unavailable for these eight patients (Fig. 1). A comparison of baseline characteristics—including age, tumor size, preoperative KPS scores, and cognitive scores—between the completed follow-up and lost-to-follow-up groups demonstrated no statistically significant differences (Supplementary Table 1). Cognitive assessment outcomes analysis Preoperative neurocognitive assessments revealed no significant differences between the two groups (Supplementary Table 2). However, at the one-week postoperative evaluation, patients in the DS group demonstrated significantly lower scores on both MMSE and MoCA compared to patients in the NDS group (Table 2). This cognitive disparity resolved at the three-month follow-up, with no significant intergroup differences observed in subsequent testing (Supplementary Table 3). Table 2Comparison of one-week postoperative cognitive functionn = 23DS GroupNDS Groupp Valuesn = 10n = 13Time2 MMSE22.7 ± 6.127.9 ± 1.20.026*Time2 MoCA20.2 ± 7.125.6 ± 3.10.044*Time2 SDMT29.2 ± 13.331.0 ± 12.30.741Time2 TMT-A (sec)75.9 ± 37.455.2 ± 16.80.089Time2 TMT-B (sec)138.2 ± 51.3133.9 ± 37.30.816p Values are from independent samples t‑tests, * represents statistically significant differences Comparison of one-week postoperative cognitive function p Values are from independent samples t‑tests, * represents statistically significant differences In the DS group, pairwise comparisons demonstrated significant cognitive changes. Compared to preoperative baselines, patients showed marked decline in MMSE (mean Δ = −6.4 points, corrected p = 0.009), MoCA (mean Δ = −5.1 points, corrected p = 0.047), and SDMT scores (mean Δ = −17.8 points, corrected p = 0.003), along with prolonged TMT-A completion times (mean Δ = + 32.7 s, corrected p = 0.017) at Time2. Six patients (60%) showed significant cognitive decline, while four patients (40%) maintained cognitive performance stable. At three-month follow-up, all cognitive metrics except MMSE had returned to preoperative levels (all p > 0.05). Significant recovery was observed in TMT-A & TMT-B at Time3 versus Time2 (corrected p = 0.037, 0.027 respectively), with other tests showing nonsignificant improvement (Fig. 2). Temporal variations of cognitive assessment results in the DS and NDS groups. A-C Longitudinal MMSE scores in the DS and NDS groups. D-F Longitudinal MoCA scores in the DS and NDS groups. G-I Longitudinal SDMT scores in the DS and NDS groups. J-L Longitudinal TMT-A time in the DS and NDS groups. M–O Longitudinal TMT-B time in the DS and NDS groups (*corrected p < 0.05, ** corrected p < 0.01, *** corrected p < 0.001) In the NDS group, postoperative (Time2) deficits were observed only in SDMT (mean Δ = −11.4 points, corrected p = 0.035) and TMT-B performance (mean Δ = + 23.0 s, corrected p = 0.046). Notably, the cognitive assessments at Time3 demonstrated recovery trends toward preoperative baseline levels, although these improvements did not reach statistical significance (Fig. 2). Comparative analysis showed significant improvement in all cognitive measures except MoCA between Time2 and Time3 evaluations (all corrected p < 0.05, Fig. 2). Rs-fMRI metrics analysis We calculated the changes in ALFF, fALFF, and ReHo values for three sub-groups: Group 1 (six patients with significant postoperative cognitive decline from the DS group), Group 2 (four patients without significant cognitive decline from the DS group), and Group 3 (13 patients from the NDS group). The results showed that Group 1 patients exhibited widespread reduction in spontaneous neural metabolic activity across multiple cortical regions at Time2, with the most significant decreases observed in: left thalamus, posterior portion of left middle temporal gyrus and left middle frontal gyrus (Fig. 3 A, D, G). In contrast, Group 2 (Fig. 3 B, E, H) and 3 (Fig. 3 C, F, I) showed no significant reductions in these three metrics. Meanwhile, comparative analysis revealed no significant reductions in ALFF, fALFF, or ReHo values at Time3 relative to preoperative baselines across all three sub-groups (Supplementary Fig. 1). Based on these findings, we identified three regions of interest (ROIs): left thalamus (coordinates: −20, −33, 3), posterior left middle temporal gyrus (−52, −27, −8) and left middle frontal gyrus (−41, 26, 21) (all with 6 mm radius). Perioperative alterations in rs-fMRI metrics. A-C Whole-brain ALFF changes pre- versus post-operation in (A) Group 1, (B) Group 2, and (C) Group 3. D-F Corresponding fALFF changes in (D) Group 1, (E) Group 2, and (F) Group 3. G-I ReHo alterations in (G) Group 1, (H) Group 2, and (I) Group 3. Blue clusters indicate regions with significant decreases (p < 0.05, FDR-corrected), while red clusters denote areas of significant increase (p < 0.05, FDR-corrected) Analysis of functional connectivity changes between these ROIs and whole-brain regions revealed that Group 1 showed significantly decreased connectivity strength between all three ROIs and cerebellar regions at Time2 compared to preoperative baseline (Fig. 4). No similar decreases in functional connectivity strength were observed in either Group 2 or Group 3. Alterations in whole-brain functional connectivity strength of three ROIs in Group 1 after surgery. A ROI1 (left thalamus) connectivity changes; B ROI2 (posterior left middle temporal gyrus) connectivity changes; C ROI3 (left middle frontal gyrus) connectivity changes. Blue clusters represent regions with significantly decreased functional connectivity (p < 0.05, FDR-corrected) In patients with available three-month follow-up data, we performed whole-brain FC analysis using the predefined ROIs. The results demonstrated significant recovery of FC strength in Group 1 patients, with previously observed postoperative decreases (at Time2) showing substantial improvement at the three-month scans (Fig. 5). Notably, ROI 2 exhibited significantly enhanced FC with cerebellar hemispheres and vermis compared to preoperative baseline levels. Longitudinal changes in whole-brain functional connectivity strength of three ROIs at three months post-operation compared to preoperative baseline in Group 1. A ROI1 (left thalamus) connectivity changes; B ROI2 (posterior left middle temporal gyrus) connectivity changes; C ROI3 (left middle frontal gyrus) connectivity changes. Red clusters represent regions with significantly increased functional connectivity (p < 0.05, FDR-corrected) Discussion Previous studies have been limited in comprehensively evaluating the dynamic changes in cognitive function and neural network reorganization in skull base meningioma patients during perioperative and recovery periods. This exploratory study aims to enrich the research in this field by systematically integrating standardized neuropsychological assessments with rs-fMRI to track the longitudinal recovery of both cognitive function and network integrity up to three months postoperatively. While literature has highlighted the inherent limitations of brain-behavior correlation studies in small samples [12], the present study is explicitly designed as an exploratory investigation aimed at generating hypotheses regarding the neural correlates of postoperative cognitive changes, rather than establishing definitive causal relationships. These exploratory findings, while requiring validation in larger independent cohorts, suggest a potential role for cerebellar-cortical networks in postoperative cognitive recovery, generating testable hypotheses for future mechanistic studies. Findings of standardized cognitive assessment Standardized cognitive instruments, including the MMSE, MoCA, SDMT, and TMT, were selected for their established utility in assessing memory, attention, processing speed, and executive function [6, 8, 11, 24]. Our study employed the culturally adapted Beijing version of the MoCA [37] to enhance diagnostic accuracy across diverse populations. Moreover, all patients in this study had a KPS score ≥ 70 at the time of cognitive assessment, thereby avoiding the confounding effects of poor physical status on cognitive performance. Our multimodal assessment revealed that patients undergoing dominant-hemisphere extradural subtemporal resection exhibited significantly greater postoperative declines in memory, orientation, and executive function compared with their non-dominant-hemisphere counterparts, with substantial recovery by three months. This transient pattern aligns with previous studies documenting reversible cognitive changes following tumor resection [5, 23, 41], while the finding that patients undergoing surgery in the dominant hemisphere are more vulnerable to postoperative cognitive decline may be attributed to the lateralized linguistic specialization of the dominant hemisphere [14], as well as intensive neural interconnections within the temporo-occipital network [22]. Existing literature has demonstrated the correlations between brain networks and cognitive function as well as neural neuroplasticity [27, 31], yet few studies have reported their perioperative dynamic alterations. In this study, the parallel recovery of cognitive performance and cerebello-cortical functional connectivity suggests that the observed postoperative cognitive trajectory may be attributable to a reversible network perturbation rather than permanent structural injury, although this hypothesis requires further validation in larger cohorts. Notably, while higher scores on the MMSE, MoCA, and SDMT indicate better cognitive performance, the TMT outcomes reflect completion time, with longer times indicating poorer performance in this study [18]. This inverse relationship accounts for the opposing trends observed between the TMT and other scales in Fig. 2. By integrating these cognitive scales, this combined approach addresses the limitations of single-test methodologies while remaining feasible for routine clinical use, with minimal equipment requirements and low patient burden. Biological significance of rs-fMRI metrics Rs-fMRI measures blood oxygen level-dependent (BOLD) signals to investigate spontaneous neural activity and functional network organization in the human brain [25]. The most commonly used rs-fMRI metrics: ALFF, fALFF, ReHo, and FC provide complementary perspectives on brain functional architecture. This study demonstrated consistent alterations across all these metrics in patients with postoperative cognitive decline, showing widespread and significant reductions in spontaneous neural activity, while patients without measurable cognitive deterioration maintained stable neural activity levels across these metrics. These findings suggest that postoperative cognitive impairment is closely associated with decreased cortical metabolic activity, reflecting the dominant hemisphere extradural subtemporal approach may disrupt multiple neural circuits and functional networks, resulting in a transient hypometabolic state of cortex and subcortical regions. FC analysis extends this perspective by characterizing inter-regional temporal correlations, revealing functional cooperation between anatomically distinct brain areas. The cerebellum and cognitive function Traditional neuroanatomical teaching has primarily associated the cerebellum with motor coordination and control. However, accumulating evidence from contemporary neuroscience research has substantially expanded this view, demonstrating the cerebellum's critical involvement in diverse cognitive domains including language processing, memory function, attentional regulation, emotional control, and executive functioning [7, 20, 28, 29, 33, 34]. Neuroimaging studies have revealed robust functional connections between the cerebellum and cortical language centers [7, 38]. Furthermore, the cerebellum maintains extensive connections with limbic structures and prefrontal cortical regions, forming a neural substrate for its involvement in emotional regulation and social cognition [2, 3, 13]. Surgical interventions may transiently disrupt these cerebellar-cerebral networks, potentially leading to postoperative cognitive impairments such as attentional deficits and working memory decline [32, 36]. The rs-fMRI studies have provided novel insights into the organization of cerebellar-cerebral networks, demonstrating the cerebellum's functional integration with multiple cortical regions including prefrontal, parietal, and temporal areas [7]. Our rs-fMRI analyses revealed significantly reduced FC strength between the identified ROIs and cerebellar in patients exhibiting postoperative cognitive decline, and the changes disappeared in parallel with cognitive recovery. We hypothesize that the extradural subtemporal approach petroclival meningioma resection may transiently compromise cerebellar network function. The subsequent recovery of both cognitive performance and connectivity strength during follow-up underscores the cerebellum's dynamic role in perioperative cognitive changes and highlights the remarkable plasticity of cerebellar-cerebral networks [30]. Furthermore, the observed temporal correlation between FC restoration and cognitive improvement in this study supports the potential utility of rs-fMRI in monitoring postoperative recovery and guiding rehabilitation strategies. Limitations Due to the limited sample size, the findings of this study regarding the relationship between neural network alterations and postoperative cognitive changes should be interpreted as preliminary and exploratory. Future large sample studies integrating fMRI, diffusion tensor imaging (DTI), and early postoperative perfusion imaging are essential to causally link the surgical act to the observed functional network dynamics. Conclusions Patients undergoing dominant hemisphere extradural subtemporal approach for petroclival meningioma resection showed a propensity for transient postoperative declines in verbal memory, orientation, and executive function. These cognitive changes were associated with reduced spontaneous cortical activity and attenuated functional connectivity within cerebello-cortical and cerebello-subcortical networks. Both cognitive performance and these neural alterations demonstrated improvement at the three-month postoperative follow-up, these preliminary observations await further validation in larger independent cohorts. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material (DOCX 2.15 MB) Abbreviations Amplitude of Low-Frequency Fluctuations Blood Oxygen Level-Dependent Dominant Side Diffusion Tensor Imaging Echo-planar Images Functional Connectivity Fractional Amplitude of Low-Frequency Fluctuations Functional Magnetic Resonance Imaging Interquartile Range Karnofsky Performance Status Mild Cognitive Impairment Mini-Mental State Examination Montreal Cognitive Assessment Non-dominant Side Regional Homogeneity Resting-state functional Magnetic Resonance Imaging Symbol Digit Modalities Test Trail Making Test Repetition Time Echo Time Author Contribution Z.L. and Y.Li contributed equally to this work. Z.L., J.J., and J.X. conceived and designed the study. Y.Liu, J.Z., and Z.W. collected and curated the clinical and imaging data. Z.L. and Y.Li performed the statistical analysis and neuroimaging processing. S.L. and L.W. supervised the project. All authors participated in data interpretation, reviewed the manuscript drafts, and approved the final version of the manuscript. Funding This study was supported by funding from the Beijing Municipal Science and Technology Commission (Grant No. 20220484162). Data availability The data that support the findings of this study are available from the corresponding author upon reasonable request. Declarations Ethical approval All procedures performed in this study involving human participants were in accordance with the ethical standards of the Beijing Tiantan Hospital Research Ethics Committee and with the Helsinki declaration. Informed consent Informed consent was obtained from all individual participants included in the study. Consent to publish Written informed consent for publication of their clinical details and clinical images was obtained from the patients. Competing interest The authors declare no competing interests. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Ziqi Liu and Yangyang Li contributed equally to this work. Contributor Information Shaowu Li, Email: lys5@sina.com. Liang Wang, Email: saintage7@126.com. References Associated Data Supplementary Materials Supplementary Material (DOCX 2.15 MB) Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request.