Abstract
Background
Successive processing, a form of working memory function detected with the Das Naglieri Cognitive Assessment System, is selectively impaired in pediatric moyamoya disease. We aimed to test whether successive processing in children with moyamoya disease was improved after bypass surgery under the control of confounding.
Methods
The present retrospective cohort study included children with moyamoya disease who underwent direct or combined bypass surgery. Neuropsychological tests including the Das Naglieri Cognitive Assessment System were administered at 2 time points, before and after surgery, approximately 1 year apart. The least squares (LS) mean standard score and LS mean difference between time points were calculated using a mixed model for repeated measures, which included 5 clinical factors along with the time point. Models including an interaction term were also generated to assess the effect of each clinical factor. Cognitive intra‐individual variability across 4 domains of the Das Naglieri Cognitive Assessment System was assessed with an analysis of variance at each time point.
Results
Of 60 patients who underwent surgery, 42 fulfilled the inclusion criteria. The median duration between assessments was 15 months. The LS mean standard scores of successive processing increased after surgery (LS mean, 95.8 versus 100.2; LS mean difference, 4.4 [95% CI, 1.5–7.3]; P = 0.004). The increase was more pronounced in those with a younger age at onset of neurological symptoms, shorter delay before surgery, preexisting infarct, posterior cerebral artery involvement, and more severe ischemic stage before surgery. Intraindividual variability, shown as the lowest score of successive processing at baseline, resolved after surgery (F = 3.56, P = 0.016 versus F = 1.21, P = 0.31). Successive processing was the domain most likely to be improved after surgery.
Conclusion
The present results suggest that successive processing is improved after bypass surgery. Larger and longer follow‐up studies are required to confirm the influencing factors and long‐term effects.
Keywords: cerebral revascularization, moyamoya disease, neurocognitive disorders, pediatrics, successive processing
Nonstandard Abbreviations and Acronyms
- CAS
Das Naglieri Cognitive Assessment System
- MMD
moyamoya disease
Clinical Perspective
What Is New?
This is the first study to demonstrate significant improvement in successive processing, a core verbal working memory function, using the Das Naglieri Cognitive Assessment System after bypass surgery in pediatric moyamoya disease.
What Are the Clinical Implications?
The cognitive improvement was more pronounced in those with a younger age at onset, a shorter delay before surgery, and other high‐risk clinical features.
The study highlights the potential benefits of timely surgical treatment in preserving and enhancing neurocognitive function in children with moyamoya disease.
Cognitive impairment is a serious issue in children with moyamoya disease (MMD), which involves the terminal portion of the internal carotid artery and causes cerebral ischemia. The impairment is often attributable to infarcts 1 , 2 , 3 ; however, it can occur even with minimal radiological infarct. 4 , 5 Working memory is the most likely to be impaired among cognitive domains 6 , 7 , 8 , 9 , 10 and is considered to be controlled by the prefrontal cortex. 5 , 10 , 11 In adult MMD, a positive effect of bypass surgery on cognitive function has been reported. 12 , 13 , 14 In pediatric MMD, however, it remains controversial whether surgery improves cognitive function. 7 , 15 , 16 , 17 Although conservatively treated children with MMD manifested extremely poor intellectual outcomes, 18 pioneering observational studies reported little intellectual improvement after surgery. 16 , 17 This controversy is partly attributable to a limited number of neuropsychological tests applicable to children.
The Das Naglieri Cognitive Assessment System (CAS), a unique 4‐domain neuropsychological test standardized for children, can successfully characterize a distinct cognitive profile of pediatric MMD 9 ; it is characterized as a selective weakness in “successive processing,” the domain reflecting verbal working memory function. 9 The CAS implements a calculation of “intraindividual differences,” which reflects cognitive variability, or strengths and weaknesses, across domains within an individual. 9 , 19 , 20 Our previous results revealed significant intraindividual differences in the domains of the CAS in children with MMD, and these differences cause underestimated difficulties in daily living. 9 Our more recent results suggest a correlation between low successive processing scores and reduced prefrontal blood flow. 21 However, no study has focused on the postoperative change in the CAS, especially in successive processing, as of this writing. Confounders have also rarely been adjusted in the comparison between pre‐ and postoperative assessments. We hypothesized that successive processing was improved after bypass surgery even with adjustment by clinical factors. The objective of our study was to compare pre‐ and postoperative standard scores detected with the CAS, especially focusing on the change in scores of successive processing. Our study might contribute to a better understanding of the role of, and indications for, bypass surgery.
Methods
Data supporting the findings of this study are available from the corresponding author upon reasonable request.
Patients and Setting
The present study was approved by the Kyoto University Graduate School and Faculty of Medicine, Ethics Committee (R1417‐1). All patients and families gave either opt‐out or written informed assent/consent in accordance with the ethical guidelines for medical and health research involving human subjects in Japan. RNF213 genotyping was conducted under a separate ethical framework (G1109). Written informed consent was obtained from the legal guardians of all pediatric participants prior to genotyping of the RNF213 p.R4810K mutation. Informed assent was also obtained from the children, based on their age and cognitive capacity. The Strengthening the Reporting of Observational Studies in Epidemiology statement was followed.
The present retrospective cohort study included children who were diagnosed with MMD according to the guideline 22 , 23 and were admitted to Kyoto University Hospital between June 2016 and December 2023. Consecutive sampling was performed. Patients were eligible if they were aged 5–17 years and underwent neuropsychological testing including the CAS before surgery, underwent bypass surgery by a single surgeon (T.F.), and completed postoperative neuropsychological testing including the CAS after surgery. Patients were excluded if they did not undergo the CAS before surgery or had special developmental backgrounds judged by the psychiatrist. 9 Patients in whom the interval between pre‐ and postoperative assessments of the CAS exceeded as long as 5 years were also excluded because factors other than surgery might affect the results.
Schedule of Neuropsychological Assessments
The schedule of neuropsychological assessments and surgical treatment is shown in Figure 1. The CAS and the Wechsler Intelligence Scale for Children Fourth Edition were routinely administered before and after surgery at intervals of 1 year. According to the manual, the interval was set to at least 6 months to minimize practice effects by repeating tests. 20 Administration of the tests in the acute phase of stroke was avoided. Assessments at baseline were performed during the first admission, and postoperative assessments were performed during the follow‐up admission or at the outpatient clinic. Alternate forms were not used. The same occupational therapist performed both pre‐ and postoperative tests for each patient.
Figure 1.
Flow chart showing the schedule of neuropsychological assessment, imaging, and surgical treatment. ACA indicates anterior cerebral artery; CAS, Das Naglieri Cognitive Assessment System; DSA, digital subtraction angiography; MCA, middle cerebral artery; MRI, magnetic resonance imaging; OA, occipital artery; PCA, posterior cerebral artery; SPECT, single‐photon emission tomography; STA, superficial temporal artery; and WISC, Wechsler Intelligence Scale for Children.
The details of the CAS have been described elsewhere. 9 In brief, the CAS is applicable for children aged 5–17 years and comprises full scale and 4 domains: planning, attention, simultaneous, and successive processing. The full‐scale score is a standard score derived from an equally weighted composite of the 4 domain scores and serves as an index of an individual's overall cognitive functioning. 20 The task of successive processing includes word recall, sentence repetition, speech rate, and sentence question, all of which require verbal working memory. 9 The standard scores of each domain are defined as a mean of 100 with an SD of 15, and higher standard scores reflect better cognitive performance. The manual of the CAS defines the calculation of “intraindividual differences” as mentioned previously; they are calculated by subtracting the mean of the 4 standard scores from each standard score. 19 , 20
Surgical Procedure
Bypass surgery was indicated for children with ischemic or hemorrhagic symptoms according to the guideline. 24 We have adopted conventional superficial temporal artery‐middle cerebral artery (MCA) anastomosis with or without encephalomyosynangiosis as a first‐line treatment. 1 , 25 For patients with bilateral internal carotid artery involvement, staged bypass surgeries for the bilateral MCA territories were performed at intervals of at least 1 month. Patients underwent follow‐up imaging examinations including conventional angiography approximately 3 months after surgery and then magnetic resonance imaging annually. During follow‐up, additional revascularization of the anterior cerebral artery (ACA) or posterior cerebral artery (PCA) territory was considered if indicated. The indication criteria and procedures of revascularization of the ACA and PCA territories have been described in detail elsewhere. 26 In brief, additional revascularization of the ACA territory was considered if patients exhibited both remaining transient ischemic attacks predominantly in the lower extremities and reduced cerebral blood flow (CBF) in the corresponding ACA territories. Superficial temporal artery‐ACA direct anastomosis combined with encephalo‐pericranio‐synangiosis was performed as the revascularization procedure. Similarly, additional revascularization of the PCA territory was considered if patients exhibited both transient visual symptoms and reduced CBF in the corresponding PCA territories. Occipital artery‐PCA direct anastomosis combined with encephalo‐pericranio‐synangiosis was performed as the revascularization procedure.
Outcome
The primary outcome variable was the standard score of successive processing, regarded as a continuous variable. The standard scores of the other domains (full scale, planning, attention, and simultaneous processing) were also evaluated.
According to the manual of the CAS, the change of the standard scores was also classified into 1 of the 3 categories: “markedly improved,” “unchanged,” and “markedly deteriorated.” The manual describes the predicted score range at the second session by each score at the first session, which is approximately ±11 points. The change was defined as “markedly improved” and “markedly deteriorated” when the second score increased and decreased beyond the predicted range, respectively. The change was defined as “unchanged” when the second score was within the predicted range.
Clinical Factors Affecting Postoperative Effect
Clinical factors affecting outcome, in addition to bypass surgery, included the following variables: age at onset of neurological symptoms, sex, initial manifestations, radiological evidence of infarct on magnetic resonance imaging, Suzuki's angiographic stage, 27 preoperative hemodynamic stage in single‐photon emission tomography (SPECT) graded as either stage 2 or nonstage 2, 28 laterality of disease involvement, presence or absence of PCA involvement, 1 and additional revascularization of the ACA or PCA territory. According to the previous study, 1 age at onset, delay before surgery, preexisting radiological evidence of infarct on magnetic resonance imaging, and PCA involvement were selected as potential confounders and incorporated into the statistical model. Preoperative SPECT stage was also incorporated into the model according to our recent results suggesting an association between successive processing and SPECT stage. 21
Genotyping
Genomic DNA was extracted from peripheral blood or buccal smear samples using the QIAamp DNA Blood Mini Kit (Qiagen). Genotyping of the RNF213 p.R4810K mutation was performed using TaqMan SNP Genotyping Assays (Applied Biosystems), as previously described. 29
Radiographic Analysis
This analysis was performed in response to peer reviews. Resting‐state CBF was acquired before and 3 months after bypass surgery with SPECT with N‐isopropyl‐[123I]‐p‐iodoamphetamine. Regions of interest were automatically set according to the vascular territories using NEUROSTAT software (University of Utah). Regional CBF (rCBF) in the MCA territory was calculated as the count ratio to that of the ipsilateral cerebellar hemisphere (CBF ratio). Baseline and postoperative rCBF were compared in a by‐person analysis. For patients undergoing bilateral surgery, data from the hemisphere with the greater improvement in CBF ratio (greater difference before and after surgery) were used for the analysis.
We also assessed Matsushima angiographic grade of revascularization area 30 with conventional angiography performed 3 to6 months after surgery. Revascularization area was classified into 1 of 3 grades: Grade A (more than two thirds of the MCA territory), Grade B (two thirds to one third of the MCA territory), and Grade C (slight or none). 30
Statistical Analysis
The sample size was determined by the number of cases treated during the study period, as this study was designed as an exploratory, observational study. Summary statistics were constructed using frequencies and proportions for categorical data. For continuous variables, the mean and SD were obtained when a distribution of normality was observed, and the median and interquartile range when a distribution of normality was not observed. A mixed model for repeated measures was used for outcomes collected at 2 time points. We incorporated into the model the time point (baseline, postoperation), age at onset (<6, ≥6 years old), delay before surgery (≥3, <3 years), preexisting radiological evidence of infarct (yes, no), PCA involvement (yes in either hemisphere, no), and preoperative SPECT stage (stage 2 in either hemisphere, nonstage 2) as fixed effects. The least squares (LS) mean at each timepoint and LS mean difference between time points were calculated with the corresponding 2‐sided 95% CIs. In addition, to evaluate the effect of clinical factors that might influence postoperative change, a model incorporating an interaction term between time point and each clinical factor was also analyzed for each clinical factor, along with time point and all 5 clinical factors. The LS mean difference between time points for each clinical factor was then calculated. A one‐factor repeated measures analysis of variance was used to compare the standard scores across the 4 measures of the CAS within subject at each time point because the standard scores from an individual are related. 9 Regarding the comparison of the intraindividual differences across the 4 measures of the CAS, a one‐way analysis of variance was used. To compare baseline and postoperative CBF ratios in each patient, paired t‐tests were used. For patients who underwent bilateral surgery, data from the hemisphere with the better CBF ratio (greater difference before and after surgery) were used in the analysis. All statistical analyses were performed with JMP Pro software (version 17, SAS Institute Inc.).
Results
Patients
The flow diagram for patient inclusion is summarized in Figure 2. A total of 70 patients aged between 5 and 17 years underwent baseline neuropsychological testing including the CAS. Of these, 60 patients underwent bypass surgery by a single surgeon. Five patients were excluded for having a special developmental background judged by the psychiatrist, and 13 patients were excluded because they did not complete postoperative psychological testing for various reasons, as listed in Figure 2. Consequently, the remaining 42 patients were analyzed.
Figure 2.
Study profile. CAS indicates Das Naglieri Cognitive Assessment System.
All bypass procedures were successfully completed. Perioperative complications were observed in 2 patients (4.8%). One patient exhibited a new radiological infarction remotely located from the anastomosis site after superficial temporal artery‐ACA anastomosis despite good patency of bypass, and another developed an epidural hematoma outside the craniotomy; however, both patients remained asymptomatic during follow‐up.
The clinical factors at baseline are summarized in Table 1. The median time interval between the baseline and postoperative assessment of the CAS was 15 months (interquartile range, 12.75–20.25). Genotyping of the RNF213 p.R4810K mutation was performed in 16 patients: 1 patient (6.3%) was homozygous for the mutation, 12 (75.0%) were heterozygous, and 3 (18.8%) had wild‐type alleles.
Table 1.
Baseline Clinical Factors
| Variable | |
|---|---|
| No. of patients | 42 |
| Median age at onset (IQR) | 7 (6–10 |
| < 6 y old, n (%) | 19 (45.2) |
| ≥ 6 y old, n (%) | 23 (54.7) |
| Female, n (%) | 25 (59.5) |
| Initial manifestation, n (%) | |
| Transient ischemic attack | 35 (83.3) |
| Others | |
| Ischemic stroke | 1 (2.4) |
| Hemorrhagic stroke | 2 (4.7) |
| Involuntary movement | 1 (2.4) |
| Headache | 3 (7.1) |
| Radiological evidence of infarct (%) | 20 (47.7) |
| Median Suzuki stage (IQR) | |
| Higher stage | 3 (2.75–4) |
| Right hemisphere | 3 (2–4) |
| Left hemisphere | 2 (1–3) |
| SPECT stage 2, n (%) | |
| Either hemisphere | 23 (54.8) |
| Right hemisphere | 17 (40.5) |
| Left hemisphere | 16 (38.1) |
| Bilateral involvement (%) | 30 (71.4) |
| Right unilateral | 8 (19.0) |
| Left unilateral | 5 (11.9) |
| Presence of PCA involvement (%) | 8 (19.0) |
| Additional revascularization | |
| ACA territory | 6 (14.3) |
| PCA territory | 1 (2.4) |
| Median interval btw baseline and postop CAS, month, (IQR) | 15 (12.75–20.25) |
ACA indicates anterior cerebral artery; CAS, Das Naglieri Cognitive Assessment System; IQR, interquartile range; PCA, posterior cerebral artery; and SPECT, single‐photon emission tomography.
Postoperative Change in Outcome
The baseline and postoperative standard scores in each domain of the CAS are summarized in Table 2. The LS mean standard scores of the full scale increased after surgery (LS mean, 99.6 versus 102.2; LS mean difference, 2.5 [95% CI, 0.6–4.5]; P = 0.011). The LS mean standard scores of successive processing also increased after surgery (LS mean, 95.8 versus 100.2; LS mean difference, 4.4 [95% CI, 1.5–7.3]; P = 0.004). No difference between baseline and postoperative LS mean standard scores was observed in the remaining domains: planning, attention, and simultaneous processing.
Table 2.
Baseline and Postoperative Least Squares Mean Standard Scores in Each Domain of the Das Naglieri Cognitive Assessment System
| Domains |
Baseline LS mean * |
Postop LS mean * |
Difference (postop − baseline) | ||
|---|---|---|---|---|---|
| LS mean * | 95% CI | P value | |||
| Full scale | 99.6 | 102.2 | 2.5 | 0.6 to 4.5 | 0.011 |
| Planning | 99.6 | 97.8 | −1.8 | −5.5 to 1.9 | 0.33 |
| Attention | 103.2 | 105.1 | 1.8 | −2.4 to 6.1 | 0.39 |
| Simultaneous | 100.2 | 102.6 | 2.4 | −0.7 to 5.5 | 0.13 |
| Successive | 95.8 | 100.2 | 4.4 | 1.5 to 7.3 | 0.004 |
LS indicates least squares; PCA, posterior cerebral artery; and SPECT, single‐photon emission tomography.
MMRM (mixed model for repeated measures). Model including time point (baseline, postop), age at onset (<6, ≥6 years old), delay before surgery (≥3, <3 years), radiological evidence of infarct (yes, no), PCA involvement (yes, no), and SPECT stage (stage 2, nonstage 2) as fixed effects.
Effect of Clinical Factors
The effect of each clinical factor on the difference between baseline and postoperative successive processing scores is summarized in Table 3. Postoperative scores were larger than baseline when the age at onset was less than 6 years (LS mean difference, 7.7 [95% CI, 3.6–11.8]; P < 0.001), delay before surgery was less than 3 years (LS mean difference, 4.7 [95% CI, 1.5–7.9]; P = 0.005), preoperative infarction was present (LS mean difference, 6.1 [95% CI, 1.9–10.3]; P = 0.006), and either hemisphere was SPECT Stage 2 (LS mean difference, 5.7 [95% CI, 1.8–9.7]; P = 0.005). The postoperative scores increased for both the presence (LS mean difference, 7.0 [95% CI, 0.3–13.7]; P = 0.041) and absence (LS mean difference, 3.8 [95% CI, 0.5–7.0]; P = 0.023) of PCA involvement, but the change in scores was larger for the presence of PCA involvement.
Table 3.
Effect of Each Confounder on Difference Between Baseline and Postoperative Successive Processing Scores
| Variable used for interaction | Difference (postop − baseline) of successive processing | ||
|---|---|---|---|
| LS mean * | 95% CI | P value | |
| Age at onset | |||
| <6 y | 7.7 | 3.6 to 11.8 | <0.001 |
| ≥6 y | 1.6 | −2.1 to 5.3 | 0.39 |
| Delay before surgery | |||
| ≥3 y | 2.7 | −4.5 to 9.9 | 0.45 |
| <3 y | 4.7 | 1.5 to 7.9 | 0.005 |
| Preexisting infarct | |||
| Yes | 6.1 | 1.9 to 10.3 | 0.006 |
| No | 2.8 | −1.2 to 6.8 | 0.16 |
| PCA involvement | |||
| Yes † | 7.0 | 0.3 to 13.7 | 0.041 |
| No | 3.8 | 0.5 to 7.0 | 0.023 |
| SPECT stage | |||
| Stage 2 † | 5.7 | 1.8 to 9.7 | 0.005 |
| Stage non‐2 | 2.7 | −1.6 to 7.0 | 0.20 |
LS indicates least squares; PCA, posterior cerebral artery; and SPECT, single‐photon emission tomography.
LS means for the interaction term using repeated measures mixed effects model (MMRM) with fixed effects for time (pre‐ and postoperative), age of onset, delay before surgery, infarct, PCA, SPECT, and variable used for the interaction and time (pre‐ and postoperative).
In either hemisphere.
Change in Cognitive Profile
We subsequently evaluated statistical variances of the standard scores and intraindividual differences across the 4 domains of the CAS by each time point (Table 4 and Figure 3) because cognitive intraindividual variability can cause potential difficulties in daily living. At baseline, the standard scores differed across the 4 domains of the CAS (F = 3.56, P = 0.016 in one‐factor repeated measures analysis of variance), and that of successive processing was the lowest (98.7±14.8); however, these differences resolved after surgery (F = 1.21, P = 0.31). Similarly, the intraindividual differences differed across the 4 domains at baseline (F = 4.74, P = 0.003 in one‐way analysis of variance), and that of successive processing was the lowest (−4.8±11.7); however, these differences resolved after surgery (F = 1.61, P = 0.19).
Table 4.
Standard Score and Intraindividual Difference of Each Domain of the Das Naglieri Cognitive Assessment System
| Time point | Planning | Attention | Simultaneous | Successive | F | P value * | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| Mean | SD | Mean | SD | Mean | SD | Mean | SD | |||
| Baseline | 106.5 | 14.4 | 106.3 | 15.4 | 102.2 | 15.4 | 98.7 | 14.8 | 3.56 | 0.016 |
| Postop | 104.7 | 12.7 | 108.2 | 16.5 | 104.6 | 16.1 | 103.1 | 14.0 | 1.21 | 0.31 |
| IID mean | SD | IID mean | SD | IID mean | SD | IID mean | SD | F | P † | |
|---|---|---|---|---|---|---|---|---|---|---|
| Baseline | 3.1 | 9.0 | 2.9 | 12.2 | −1.2 | 11.5 | −4.8 | 11.7 | 4.74 | 0.003 |
| Postop | −0.4 | 9.9 | 3.0 | 11.9 | −0.5 | 10.2 | −2.1 | 11.9 | 1.61 | 0.19 |
ANOVA indicates analysis of variance; and IID, intraindividual difference.
One‐factor repeated measures ANOVA.
One‐way ANOVA.
Figure 3.
Graph showing the change in each domain of the Das Naglieri Cognitive Assessment System. A, Standard score. B, Intraindividual difference. Error bar indicates SD.
The category of the change in each domain, defined according to the manual, is shown in Figure 4. Regarding successive processing, 9 (21.4%) of 42 patients were classified as “markedly improved,” 32 (76.2%) as “unchanged,” and 1 (2.4%) as “markedly deteriorated.” Successive processing was the most likely to be improved and the least likely to deteriorate among domains (Figure 4).
Figure 4.
Graph showing the category of the change in each domain of the Das Naglieri Cognitive Assessment System.
Radiographic Result
Postoperative changes in rCBF within the MCA territory were analyzed in 40 patients. CBF data were unavailable in 2 patients who underwent baseline SPECT outside of our hospital. CBF ratio increased after surgery as compared with baseline (1.10±0.13 versus 1.21±0.10, P<0.001, Table 5).
Table 5.
Baseline and Postoperative Cerebral Blood Flow Ratio
| Number | Baseline | Postop | P value * | |||
|---|---|---|---|---|---|---|
| Mean | SD | Mean | SD | |||
| Patient † | 40 | 1.10 | 0.13 | 1.21 | 0.10 | <0.001 |
| Hemisphere | 69 | 1.13 | 0.14 | 1.20 | 0.10 | – |
Paired t‐test.
Data were unavailable in 2 patients who underwent baseline single‐photon emission tomography outside of our hospital. For patients undergoing bilateral surgeries, data from the hemisphere with the greater improvement in cerebral blood flow ratio (greater difference before and after surgery) were used for the analysis.
Angiographic revascularization area was evaluated in 62 hemispheres undergoing surgery. Postoperative angiography was not completed in 4 patients. According to Matsushima angiographic grade, 29 hemispheres (46.8%) were classified as Grade A (more than two thirds of the MCA territory), and 33 (53.2%) as Grade B (two thirds to one third of the MCA territory). None were classified as Grade C (slight or none).
Effect on Cerebral Blood Flow Improvement and RNF213 Genotype on Successive Processing
To explore the relationship between postoperative changes in successive processing scores and the degree of CBF improvement and genotype, scatter plots were generated comparing pre‐ and postoperative changes in rCBF within the MCA territory and successive processing scores (Figure S1). No clear correlation was observed between the degree of CBF change and improvement in successive processing (r = 0.148). With regard to the RNF213 genotype, patients with the heterozygous mutation tended to cluster in the higher range of successive processing improvement (Figure S1A). In contrast, when stratified by Matsushima angiographic grade of the revascularization area, no clear trend was observed between changes in rCBF and changes in successive processing scores among patients with Grades A and B (Figure S1B).
Discussion
Building on prior studies of postoperative neurocognitive improvement, our study offers three key findings. First, the standard score of successive processing increased after surgery under the control of confounding. Second, the improvement in successive processing score was affected by several clinical factors, including age at onset, delay before surgery, preexisting infarct, PCA involvement, and ischemic stage before surgery. Third, cognitive intraindividual variability, shown as the lowest score of successive processing among domains at baseline, resolved after surgery. The present study is perhaps the first to reveal the cognitive improvement detected with the CAS after bypass surgery for pediatric MMD.
Our results are in line with those of previous studies suggesting the beneficial effect of bypass surgery on cognitive function in MMD. Kimura et al analyzed the pre‐ and postoperative neuropsychological tests in 52 adult patients and found an increase of scores in verbal and performance intelligence quotients in the Wechsler Adult Intelligence Scale‐Revised, memory quotient of the Wechsler Memory Scale, and Rey copy and recall tests. 12 Two other studies on adult patients also suggested improvements in verbal and performance intelligence quotients in Wechsler Adult Intelligence Scale in the group in which CBF was ameliorated after bypass surgery. 13 , 14 Regarding pediatric MMD, Lee et al analyzed the pre‐ and postoperative cognitive profiles of 65 patients using Wechsler Intelligence Scale for Children‐R. They found that performance intelligence quotient, especially the Coding subtest, increased after surgery. Coding tasks measure processing speed and reflect various cognitive functions, including working memory. Our results add novel information to these foundational studies because successive processing, a cognitive function intrinsically impaired in pediatric MMD, 9 , 21 was improved after surgery.
On the other hand, our results are partly inconsistent with a historical work by Matsushima et al, in which full‐scale and performance intelligence quotient in Wechsler Intelligence Scale for Children was deteriorated during follow‐up after indirect bypass surgery. 16 Digit span, which measures working memory, was the most likely to be deteriorated among subtests. 16 Several possible reasons for this inconsistency are considered, including differences in surgical procedures and length of follow‐up period. Another possible reason is that the tasks of successive processing of the CAS might sensitively detect cognitive improvement after surgery, considering that the CAS is a sensitive method for detecting cognitive characteristics in pediatric MMD. 9
The improvement of successive processing shown in the present study is reasonably explained by the probable association between working memory function and CBF especially in the prefrontal cortex. Karashima et al revealed that the working memory index in Wechsler Intelligence Scale for Children was associated with blood flow in the anterior area of the MCA territory. 11 In adult MMD, neuronal loss in the medial frontal lobe, which is detected with 123I‐iomazenil SPECT, was related to cognitive dysfunction. 5 Kuroda et al showed that the type of bypass surgery, which widely covered the frontal lobe, was a factor associated with favorable intellectual outcomes. 31 Our previous study also revealed that selective intraindividual weakness in successive processing was associated with reduced blood flow in the prefrontal cortex. 21 All these findings lead to the hypothesis that the amelioration of blood flow in the prefrontal cortex results in the recovery of successive processing.
To support this hypothesis, our overall findings may suggest a potential causal relationship between improved CBF and cognitive outcomes, as postoperative increases in both rCBF (Table 5) and successive processing scores (Table 3) were confirmed. Notably, the latter was demonstrated using the mixed model for repeated measures, which included time point (baseline versus postoperative) as a fixed effect. However, the supplemental analysis revealed no clear correlation between the degree of changes in rCBF within the MCA territory and in successive processing (Figure S1). This discrepancy may be attributable to several factors, including the limited sample size and the preliminary nature of the CBF analysis. The present study primarily focused on cognitive improvement. The relationship between CBF and cognitive outcomes should be further investigated in larger cohorts using more detailed analyses of CBF parameters in more specific brain regions, such as the prefrontal cortex.
We observed that the postoperative increase of successive processing scores was more pronounced in patients who underwent surgery within 3 years after symptom onset than it was in those with a longer delay (Table 3). Viewed from a clinical perspective, this may suggest that early surgical intervention is recommended for children with MMD in terms of cognitive recovery. This speculation is supported by the study by Imaizumi et al, in which longer duration after onset without surgery was associated with unfavorable intellectual outcomes. 32 Larger studies are required to identify factors affecting the improvement of successive processing.
Our results suggest that onset at a younger age (<6 years old) accrues more pronounced neurocognitive improvement than that at an older age (≥6 years). The present study focused on neurocognitive change at a relatively short interval (median, 15 months) because it might minimize contamination of effects other than surgery. On the other hand, several long‐term follow‐up studies revealed that onset at a younger age was associated with poorer outcomes. 25 , 33 Long‐term follow‐up of our cohort is required to elucidate the outcomes of patients, especially those with a younger onset. In addition, our exploratory findings suggest a potential association between the RNF213 genotype and postoperative improvement in successive processing. However, genotype data were available for only a limited number of patients in this study, and further accumulation of RNF213 genotype data is warranted.
Our study has several limitations. First, the study was a retrospective one and did not have a control group not undergoing bypass surgery. Nowadays, however, it is very difficult to acquire neurocognitive follow‐up data of pediatric patients treated conservatively because most symptomatic patients with pediatric MMD undergo surgery. Randomization would no longer be applicable from an ethical perspective. Our conclusions might be justified considering the extremely poor intellectual outcomes in conservatively treated children found by an earlier study. 18 Second, the sample size of the present study was not sufficient to conduct a logistic regression analysis identifying factors associated with the improvement of successive processing as discussed previously. Third, we observed postoperative cognitive improvement at the median interval of 15 months; however, it remains unclear whether the improvement lasts for a long time after surgery. As shown in Table 4 and Figure 3, the score of successive processing remained the lowest among domains even after surgery. Long‐term careful follow‐up is required, and adequate educational support should be considered for those with lasting weakness of successive processing.
Conclusions
The results support our hypothesis that successive processing, a working memory function intrinsically impaired in pediatric MMD, is improved after bypass surgery, at least during a short follow‐up period. The benefit of surgery on neurocognitive function, however, should be carefully interpreted. Larger studies are required to identify factors affecting the improvement. Long‐term follow‐up is also required to confirm whether the improvement lasts for a long time after surgery.
Sources of Funding
This work was supported by JSPS KAKENHI Grant Number JP24K23529.
Disclosures
Takeshi Funaki received a consigned research fund from Nihon Medi‐Physics Co., Ltd.
Supporting information
Figure S1: Scatter plots showing the relationship between postoperative changes in successive processing scores and potential influencing factors: change in cerebral blood flow (CBF) ratio (A and B), RNF213 genotype (A), and Matsushima angiographic grade of the revascularization area (B). Data from the hemisphere with the greater improvement in CBF ratio or the better Matsushima grade were used. AA indicates homozygous mutation; GA, heterozygous mutation; and GG, wild type.
Acknowledgments
None.
References
- 1. Funaki T, Takahashi JC, Takagi Y, Yoshida K, Araki Y, Kikuchi T, Kataoka H, Iihara K, Miyamoto S. Impact of posterior cerebral artery involvement on long‐term clinical and social outcome of pediatric moyamoya disease. J Neurosurg Pediatr. 2013;12:626‐632. 10.3171/2013.9.PEDS13111. [DOI] [PubMed] [Google Scholar]
- 2. Karasawa J, Kikuchi H, Furuse S, Kawamura J, Sakaki T. Treatment of moyamoya disease with STA‐MCA anastomosis. J Neurosurg. 1978;49:679‐688. 10.3171/jns.1978.49.5.0679. [DOI] [PubMed] [Google Scholar]
- 3. Kim SK, Seol HJ, Cho BK, Hwang YS, Lee DS, Wang KC. Moyamoya disease among young patients: its aggressive clinical course and the role of active surgical treatment. Neurosurgery. 2004;54:840‐844; discussion 844–846. [DOI] [PubMed] [Google Scholar]
- 4. Karzmark P, Zeifert PD, Bell‐Stephens TE, Steinberg GK, Dorfman LJ. Neurocognitive impairment in adults with moyamoya disease without stroke. Neurosurgery. 2012;70:634‐638. 10.1227/NEU.0b013e3182320d1a. [DOI] [PubMed] [Google Scholar]
- 5. Kikuchi T, Takagi Y, Nakagawara J, Ueno T, Ubukata S, Houkin K, Araki Y, Takahashi JC, Nakase H, Murai T, et al. Neuronal loss in the bilateral medial frontal lobe revealed by (123)I‐iomazenil single‐photon emission computed tomography in patients with moyamoya disease: the first report from Cognitive Dysfunction Survey of Japanese Patients with Moyamoya Disease (COSMO‐Japan Study). Neurol Med Chir (Tokyo). 2023;63:334‐342. 10.2176/jns-nmc.2023-0041. [DOI] [Google Scholar]
- 6. Kronenburg A, Deckers PT, van den Berg E, van Schooneveld MM, Vonken EJ, van der Zwan A, van Berckel BNM, Yaqub M, Otte W, Klijn CJM, et al. The profile of cognitive impairment and hemodynamic compromise in moyamoya: a single‐center prospective cohort study. J Neurosurg. 2023;138:173‐184. 10.3171/2022.3.Jns212844. [DOI] [PubMed] [Google Scholar]
- 7. Kronenburg A, van den Berg E, van Schooneveld MM, Braun KPJ, Calviere L, van der Zwan A, Klijn CJM. Cognitive functions in children and adults with moyamoya vasculopathy: a systematic review and meta‐analysis. J Stroke. 2018;20:332‐341. 10.5853/jos.2018.01550. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Hsu YH, Kuo MF, Hua MS, Yang CC. Selective neuropsychological impairments and related clinical factors in children with moyamoya disease of the transient ischemic attack type. Childs Nerv Syst. 2014;30:441‐447. 10.1007/s00381-013-2271-9. [DOI] [PubMed] [Google Scholar]
- 9. Kusano Y, Funaki T, Ueda K, Nishida N, Tanaka K, Miyamoto S, Matsuda S. Characterizing the neurocognitive profiles of children with moyamoya disease using the Das Naglieri cognitive assessment system. Sci Rep. 2022;12:3638. 10.1038/s41598-022-07699-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Kazumata K, Tokairin K, Sugiyama T, Ito M, Uchino H, Osanai T, Kawabori M, Nakayama N, Houkin K. Association of cognitive function with cerebral blood flow in children with moyamoya disease. J Neurosurg Pediatr. 2019:1‐7. 10.3171/2019.7.PEDS19312. [DOI] [Google Scholar]
- 11. Karashima S, Nakamizo A, Arimura K, Yoshimoto K. Intellectual function and memory in children with moyamoya disease: relationship between Wechsler Intelligence Scale and Benton Visual Retention Test scores and regional cerebral blood flow. J Neurosurg Pediatr. 2024;33:301‐306. 10.3171/2023.11.Peds23317. [DOI] [PubMed] [Google Scholar]
- 12. Kimura K, Kubo Y, Dobashi K, Katakura Y, Chida K, Kobayashi M, Yoshida K, Fujiwara S, Terasaki K, Kawamura T, et al. Angiographic, cerebral hemodynamic, and cognitive outcomes of indirect revascularization surgery alone for adult patients with misery perfusion due to ischemic moyamoya disease. Neurosurgery. 2022;90:676‐683. 10.1227/neu.0000000000001907. [DOI] [PubMed] [Google Scholar]
- 13. Hara S, Kudo T, Hayashi S, Inaji M, Tanaka Y, Maehara T, Ishii K, Nariai T. Improvement in cognitive decline after indirect bypass surgery in adult moyamoya disease: implication of (15)O‐gas positron emission tomography. Ann Nucl Med. 2020;34:467‐475. 10.1007/s12149-020-01473-8. [DOI] [PubMed] [Google Scholar]
- 14. Uchida S, Kubo Y, Oomori D, Yabuki M, Kitakami K, Fujiwara S, Yoshida K, Kobayashi M, Terasaki K, Ogasawara K. Long‐term cognitive changes after revascularization surgery in adult patients with ischemic moyamoya disease. Cerebrovasc Dis Extra. 2021;11:145‐154. 10.1159/000521028. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Lee JY, Phi JH, Wang KC, Cho BK, Shin MS, Kim SK. Neurocognitive profiles of children with moyamoya disease before and after surgical intervention. Cerebrovasc Dis (Basel, Switzerland). 2011;31:230‐237. 10.1159/000321901 [DOI] [Google Scholar]
- 16. Matsushima Y, Aoyagi M, Nariai T, Takada Y, Hirakawa K. Long‐term intelligence outcome of post‐encephalo‐duro‐arterio‐synangiosis childhood moyamoya patients. Clin Neurol Neurosurg. 1997;99:S147‐S150. [DOI] [PubMed] [Google Scholar]
- 17. Matsushima Y, Aoyagi M, Koumo Y, Takasato Y, Yamaguchi T, Masaoka H, Suzuki R, Ohno K. Effects of encephalo‐duro‐arterio‐synangiosis on childhood moyamoya patients–swift disappearance of ischemic attacks and maintenance of mental capacity. Neurol Med Chir (Tokyo). 1991;31:708‐714. [DOI] [PubMed] [Google Scholar]
- 18. Kurokawa T, Tomita S, Ueda K, Narazaki O, Hanai T, Hasuo K, Matsushima T, Kitamura K. Prognosis of occlusive disease of the circle of Willis (moyamoya disease) in children. Pediatr Neurol. 1985;1:274‐277. [DOI] [PubMed] [Google Scholar]
- 19. Naglieri JA, Das JP. Cognitive Assessment System Interpretive Handbook. Nihon Bunka Kagakusha; 2007. [Google Scholar]
- 20. Naglieri JA. Essentials of CAS assessment. CAS Interpretation. John Wiley & Sons Inc; 1999. [Google Scholar]
- 21. Kusano Y, Funaki T, Ueda K, Ueno T, Tanaka K, Nishida N, Tabata A, Fushimi Y, Mitsumoto K, Kikuchi T, et al. Lower prefrontal blood flow associated with intra‐individual weakness in successive processing: a neurocognitive study of pediatric moyamoya disease. J Neurosurg Pediatr. 2025;35: 581–590. 10.3171/2024.11.PEDS24495. [DOI] [PubMed] [Google Scholar]
- 22. Kuroda S, Fujimura M, Takahashi J, Kataoka H, Ogasawara K, Iwama T, Tominaga T, Miyamoto S. Diagnostic criteria for moyamoya disease – 2021 revised version. Neurol Med Chir (Tokyo). 2022;62:307‐312. 10.2176/jns-nmc.2022-0072. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Research Committee on the Pathology and Treatment of Spontaneous Occlusion of the Circle of Willis, Health Labour Sciences Research Grant for Research on Measures for Intractable Diseases . Guidelines for diagnosis and treatment of moyamoya disease (spontaneous occlusion of the circle of Willis). Neurol Med Chir (Tokyo). 2012;52:245‐266. [DOI] [PubMed] [Google Scholar]
- 24. Fujimura M, Tominaga T, Kuroda S, Takahashi JC, Endo H, Ogasawara K, Miyamoto S. 2021 Japanese Guidelines for the Management of Moyamoya Disease: guidelines from the Research Committee on Moyamoya Disease and Japan Stroke Society. Neurol Med Chir (Tokyo). 2022;62:165‐170. 10.2176/jns-nmc.2021-0382. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Karasawa J, Touho H, Ohnishi H, Miyamoto S, Kikuchi H. Long‐term follow‐up study after extracranial‐intracranial bypass surgery for anterior circulation ischemia in childhood moyamoya disease. J Neurosurg. 1992;77:84‐89. 10.3171/jns.1992.77.1.0084. [DOI] [PubMed] [Google Scholar]
- 26. Funaki T, Takahashi JC, Miyamoto S. Direct anterior or posterior cerebral artery territory bypass in pediatric moyamoya disease: perioperative and Mid‐term results of a case series. Nerv Syst Child [Jpn]. 2019;44:16‐22. [Google Scholar]
- 27. Suzuki J, Kodama N. Moyamoya disease–a review. Stroke. 1983;14:104‐109. [DOI] [PubMed] [Google Scholar]
- 28. Takahashi JC, Funaki T, Houkin K, Kuroda S, Fujimura M, Tomata Y, Miyamoto S. Impact of cortical hemodynamic failure on both subsequent hemorrhagic stroke and effect of bypass surgery in hemorrhagic moyamoya disease: a supplementary analysis of the Japan Adult Moyamoya Trial. J Neurosurg. 2021;134:940‐945. 10.3171/2020.1.Jns192392. [DOI] [PubMed] [Google Scholar]
- 29. Liu W, Morito D, Takashima S, Mineharu Y, Kobayashi H, Hitomi T, Hashikata H, Matsuura N, Yamazaki S, Toyoda A, et al. Identification of RNF213 as a susceptibility gene for moyamoya disease and its possible role in vascular development. PLoS ONE. 2011;6:e22542. 10.1371/journal.pone.0022542. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Matsushima T, Inoue T, Suzuki SO, Fujii K, Fukui M, Hasuo K. Surgical treatment of moyamoya disease in pediatric patients–comparison between the results of indirect and direct revascularization procedures. Neurosurgery. 1992;31:401‐405. [DOI] [PubMed] [Google Scholar]
- 31. Kuroda S, Houkin K, Ishikawa T, Nakayama N, Ikeda J, Ishii N, Kamiyama H, Iwasaki Y. Determinants of intellectual outcome after surgical revascularization in pediatric moyamoya disease: a multivariate analysis. Childs Nerv Syst. 2004;20:302‐308. 10.1007/s00381-004-0924-4. [DOI] [PubMed] [Google Scholar]
- 32. Imaizumi C, Imaizumi T, Osawa M, Fukuyama Y, Takeshita M. Serial intelligence test scores in pediatric moyamoya disease. Neuropediatrics. 1999;30:294‐299. 10.1055/s-2007-973508. [DOI] [PubMed] [Google Scholar]
- 33. Imaizumi T, Hayashi K, Saito K, Osawa M, Fukuyama Y. Long‐term outcomes of pediatric moyamoya disease monitored to adulthood. Pediatr Neurol. 1998;18:321‐325. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1: Scatter plots showing the relationship between postoperative changes in successive processing scores and potential influencing factors: change in cerebral blood flow (CBF) ratio (A and B), RNF213 genotype (A), and Matsushima angiographic grade of the revascularization area (B). Data from the hemisphere with the greater improvement in CBF ratio or the better Matsushima grade were used. AA indicates homozygous mutation; GA, heterozygous mutation; and GG, wild type.