Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Feb 4.
Published in final edited form as: Clin Neurophysiol. 2024 Feb 16;161:52–58. doi: 10.1016/j.clinph.2024.02.010

Delays in latencies of median-nerve evoked magnetic fields in patients with succinic semialdehyde dehydrogenase deficiency

Teppei Matsubara a,b,*, Sheraz Khan a,b, Padmavathi Sundaram a,b, Steven Stufflebeam a,b, Deniz Aygun c, Melissa DiBacco c, Jean-Baptiste Roullet d, Phillip L Pearl b,c, Yoshio Okada b,e
PMCID: PMC12865411  NIHMSID: NIHMS2135674  PMID: 38447494

Abstract

Objective:

Succinic semialdehyde dehydrogenase deficiency (SSADHD) is a genetic disorder resulting in abnormal regulation of γ-aminobutyric acid, lipid metabolism, and myelin biogenesis, leading to ataxia, seizures, and cognitive impairment. Since the myelin sheath is thinner in a murine model of SSADHD compared to a wild type, we hypothesized that this also holds for human brain. We tested whether the conduction velocity in the somatosensory pathway is accordingly delayed.

Methods:

Somatosensory evoked magnetic fields (SEF) produced by transcutaneous electrical stimulation of the median nerve were measured in 13 SSADHD patients, 11 healthy and 14 disease controls with focal epilepsy. The peak latencies of the initial four components (M1, M2, M3 and M4) were measured.

Results:

The SEF waveforms and scalp topographies were comparable across the groups. The latencies were statistically significantly longer in the SSADHD group compared to the two controls. We found these latencies for the SSADHD, healthy and disease controls respectively to be: M1: (21.9 ± 0.8 ms [mean ± standard error of the mean], 20.4 ± 0.6 ms, and 21.0 ± 0.4 ms) (p < 0.05); M2: (36.1 ± 1.0 ms, 33.1 ± 0.6 ms, and 32.1 ± 1.1 ms) (p < 0.005); M3: (62.5 ± 2.4 ms, 54.7 ± 2.0 ms, and 49.9 ± 1.8 ms) (p < 0.005); M4: (86.2 ± 2.3 ms, 78.8 ± 2.8 ms, and 73.5 ± 2.9 ms) (p < 0.005).

Conclusions:

The SEF latencies are delayed in patients with SSADHD compared with healthy controls and disease controls.

Significance: This is the first study that compares conduction velocities in the somatosensory pathway in SSADHD, an inherited disorder of GABA metabolism. The longer peak latency implying slower conduction velocity supports the hypothesis that myelin sheath thickness is decreased in SSADHD.

Keywords: Myelinated axon, Conduction velocity, Somatosensory evoked field, Succinic semialdehyde dehydrogenase, deficiency (SSADHD), γ-aminobutyric acid (GABA), Cognitive impairment, Epilepsy

1. Introduction

Succinic semialdehyde dehydrogenase deficiency (SSADHD) is a rare genetic disease caused by inactivating mutations of the ALDH5A1 gene, a gene regulating the balance between excitatory and inhibitory neurotransmission. Approximately 200 published cases have been identified with a median age of diagnosis of two years, and worldwide estimates of about 1000 affected individuals (Attri et al., 2017). The clinical and metabolic consequences of SSADHD have been extensively studied by Gibson and Pearl (Pearl et al., 2021). SSADH, together with γ-aminobutyric acid (GABA) transaminase, converts GABA to succinate. The absence of SSADH leads to GABA buildup and γ-hydroxybutyrate (GHB) accumulation, and downregulation of post-synaptic GABA receptors (Lee et al., 2022).

Lipid metabolism and myelin biogenesis are also abnormal in SSADHD (Barcelo-Coblijn et al., 2007, Donarum et al., 2006, Lee et al., 2022). Myelin sheath has been found to be thinner in SSADH deficient mice compared to a wild type (Donarum et al., 2006). Multiple myelin deficiencies have also been identified in the post-mortem analysis of the brain of an SSADHD patient (Walters et al., 2021).

The GABA and lipid abnormalities manifest clinically as developmental delay, hypotonia, intellectual disability, ataxia, seizures, hyperkinetic behavior, aggression, and sleep disturbances (Pearl et al., 2003a,b,2009).

In this study we addressed the possibility of abnormalities in myelination in patients with SSADHD. Based on the impairment in myelination found in the murine model of SSADHD, we hypothesized that myelin sheath may also be thinner in the brain of patients with SSADHD compared to controls. According to this hypothesis, we predicted that the central conduction velocity is slower in patients with SSADHD. We tested this prediction by measuring the latencies of the four initial components of the somatosensory evoked magnetic fields (SEFs) produced by transcutaneous electrical stimulation of the median nerve in SSADHD compared to controls. The median-nerve SEF is a standard clinical technique for measuring the conduction velocity in the somatosensory pathway (Hari et al., 2018, Matsubara et al., 2021). The SEF is well-suited for testing our hypothesis since the normative peak latencies are well established and the results can be reproduced in patients with SSADHD in our study who suffer from cognitive impairment.

2. Materials and methods

2.1. Subjects

Thirteen patients with SSADHD (age 3–29 years, mean 14.1 ± 2. 6 years, 6 males), 11 healthy controls (age 6–36 years, mean 18.2 ± 2.8 years, 4 males) and 14 disease controls diagnosed with focal epilepsy (age 7–30 years, mean 15.0 ± 2.5 years, 9 males) were included in the SEF testing study. Mean ages were not different across groups (analysis of variance [ANOVA], F(2,35) = 0.624, p = 0.5). Patients with SSADHD and some of healthy controls were the individuals enrolled in our ongoing natural history study of SSADHD who consented to undergo SEF testing. Diagnosis of SSADHD was confirmed by the presence of biallelic pathogenic variants in ALDH5A1 and by elevated urine GHB levels. Among 13 SSADHD patients, five had epilepsy, all with chronic epilepsy of combined focal and generalized type. Healthy controls were individuals with neurotypical development and no history of neurologic disorder or brain injury. As an additional control group, SEF data were obtained from patients with focal epilepsy studied at Massachusetts General Hospital Martinos Center Magnetoencephalography (MEG) Laboratory as part of presurgical evaluation. All subjects gave their written informed consent or assent from the subjects or their family before the MEG experiments (IRB protocol of Massachusetts General Hospital, #2011P000668).

2.2. Stimulus

The median nerve was transcutaneously stimulated at the wrist with an electric square pulse of 0.2 ms duration. Stimuli were presented alternatively to the left and right sides, approximately 120 times on each side. The interstimulus interval was constant at 1.5 s. The stimuli were applied using a pair of electrodes placed on the skin 3 cm apart with the cathode proximal to the anode. Stimulus intensity was just above the motor threshold of the abductor pollicis brevis muscle (approximately 4–4.5 mA). All the subjects were able to tolerate this range of intensity.

2.3. MEG recordings

MEG signals were recorded using a whole-head 306 channel sensor array (MEGIN, Helsinki, Finland) with 102 identical triple-sensor elements. Before recording, head-position-indicator coils were attached to the subject’s head. Their locations were measured with the Polhemus Fastrak digitizer (Colchester, VT, USA). Three anatomical landmarks (nasion and bilateral preauricular points) were also digitized to define head-based MEG coordinate system. At the beginning of the recording session, the subject’s head position was measured with respect to the sensor array. The continuous head position tracking was used mostly with young children to compensate for effects of the head movements on the MEG data (Taulu et al., 2004). Horizontal eye movements were recorded to monitor eye movements in some subjects. The recording was performed in a three-layer magnetically-shielded room at the MEG laboratory while subjects lay in a supine position with their head positioned inside the helmet-shaped sensor array.

The sampling rate was 2000 or 2035 Hz for all SSADHD and healthy controls; for disease controls, the sampling rate was 1000 (before 2017), 2000 (after 2022), or 2035 Hz (between 2017 to 2022) depending on the study date. The data were downsampled or up-sampled to 2000 Hz.

2.4. Data analysis

2.4.1. Preprocessing

MEG data were preprocessed using Maxfilter (Elekta Oy, Finland). In addition to a spatiotemporal signal space separation method applied to reduce external artifact signals (Sekihara, 2018), a default head position was used along with the maxmove option of Maxfilter. This allowed for direct within-channel comparison of signals across patients and groups without concern of differences in head position within the MEG helmet during data collection. The MEG signals were bandpass filtered at 2–130 Hz. The MEG data were averaged off-line from −50 ms before and 150 ms after stimulus onset. Trials exceeding 2000 fT/cm in amplitude on a gradiometer and 3000 fT on a magnetometer were excluded before averaging. Artifacts such as eye movements and epileptic spikes were carefully excluded by visual inspection. The average numbers of excluded trials were 13.3 (standard deviation, SD 7.3) for left stimulation and 12.8 (SD 8.7) for right stimulation in the SSADHD group, 11.6 (12.4) for left and 9.6 (SD 9.7) for right in healthy controls, and 10.9 (8.8) for left and 9.5 (8.3) for right in disease controls. There were no significant differences in the number of excluded trials observed across groups.

2.4.2. SEF latency

Four components were identified in the sensor space analysis: M1, M2, M3 and M4. These components corresponded to N20m, P30m, N40m and P60m, respectively (Kakigi, 1994). The first cortical SEF component, N20m response (M1), is an established marker of primary somatosensory function. Each component was identified contralaterally in each subject separately based on the topographical map and its peak referencing global field power of each subject. Due to the lack of magnetic resonance imaging (MRI) acquisition from all of the patients, we assessed the latency difference that is measurable in the sensor space analysis.

2.4.3. Partial correlation

Pearson partial correlation analyses were performed between the latencies of four components and a clinical severity scoring (CSS) system. Our group has validated a CSS system according to the clinical characteristics of SSADHD in the domains reflecting the principal manifestations (Tokatly Latzer et al., 2023): cognition, communication, motor skills, psychiatric presentation, and epilepsy. The scores of each domain range from 1 to 5, and lower scores indicate greater impairment. The total CSS is compounded from the additive scores of these five domains (25). Pearson partial correlation analyses were performed between the latencies of the components of the SSADHD group and (1) CSS total score, and (2) CSS of each domain, compensating for effects of age.

2.5. Statistical analysis

First, three-way repeated measures analysis of covariance (ANCOVA) was used to examine differences in the latency of each component across hemispheres, and across groups after removing effects of age. Age is a confounding variable that may affect the conduction time in the peripheral as well as central nervous systems. The side (left vs right hemisphere) and the group (SSADHD vs healthy control vs disease control) were the between-subjects factors, whereas component (M1–M4) was the within-subjects factor. After we found no effects of side or interaction between side and the rest of the factors, we used a two-way repeated measures ANCOVA to examine differences in the latency of each component across groups after averaging the responses of left and right hemispheres. Post-hoc analysis was conducted using t-test. False discovery rate (FDR) correction was employed for multiple comparisons. In all analyses, the significance level was set at 0.05, 0.001, or 0.005.

3. Results

The latency of each component was determined for each subject. Fig. 1 shows the butterfly plots of SEF measured by the magnetometers in the MEG system for left median nerve in representative subjects from a 27-year-old SSADHD (top), 25-year-old healthy control (middle), and 20-year-old disease control (bottom). Since the peak of each component was not always clear, we determined the latency based on the butterfly plots of all sensors, the topography of the isofield map, and global field power. The butterfly plots showed a peak in many of the sensors for each component M1–M4 at the individual subject level. The dipole isofield maps indicated that the fields at those latencies were produced by a relatively simple single equivalent current generator. Fig. 2A shows the grand-averaged butterfly plots of SEFs for left median nerve stimulation in SSADHD (top), healthy controls (middle), and disease controls (bottom). The waveforms are comparable across the three groups. Fig. 2B shows the averaged global field power waveforms across stimuli in the three groups. Again, the waveforms are quite comparable across the three groups. The four components selected for this analysis are indicated by arrows. M1 has a recognizable sharp peak, as is also the case for M2. M3 is the initial inflection in the slower component. M4 is the peak of the slow component. The combination of these three types of information enabled us to determine the peak latency for M1–M4.

Fig. 1.

Fig. 1.

Open in a new tab

Butterfly plots of the SEFs produced by left median nerve stimulation in representative subjects from a 27-year-old patient with succinic semialdehyde dehydrogenase deficiency (SSADHD) (top), 25-year-old healthy control (middle), and 20-year-old disease control (lower), together with the topographical maps of identifiable components (M1–M4). The waveforms of magnetometer are plotted.

Fig. 2.

Fig. 2.

Open in a new tab

The grand-averaged butterfly plots of the evoked responses to left median nerve stimulation in patients with succinic semialdehyde dehydrogenase deficiency (SSADHD) (top), healthy controls (middle) and disease controls (lower) (A). The waveforms of magnetometer are plotted. Global field power averaging left and right median nerve stimuli in three groups (B). In each group, four components corresponding to M1–M4 are identified (arrows).

Fig. 3 shows SEF topographies produced by left median nerve stimulation for the four components along the columns and for the three groups along the rows. The topography of M1 is similar across the three groups with comparable positions of the field maximum and minimum. Topographies are also comparable across the three groups for M2, M3, and M4. The similar topographies indicate that the generators are comparable across the three groups for each component. Thus, it was possible to compare the latency of each component across the three groups.

Fig. 3.

Fig. 3.

Open in a new tab

The topographical map corresponding to each component (M1–M4) in response to left median nerve stimulation. Each component shows the consistent single-dipole pattern in the contralateral hemisphere.

The three-way repeated measures ANCOVA showed a significant main effect of group [F(2, 32) = 7.4, p < 0.005] and component [F(3, 96) = 306.7, p < 0.001]. However, there were no significant main effects of side [F(1, 96) = 0.1, p = 0.8], or interaction between side and group [F(2, 96) = 1.4, p < 0.3], side and component [F(3, 96) = 0.4, p = 0.8], and side, component and group [F(6, 96) = 1.4, p = 0.2]. These results indicated that there were no differences in latency across hemispheres. Therefore, latencies of each component were averaged across two hemispheres in the subsequent analyses.

Fig. 4 shows the histogram of averaged latencies across groups, separately for the four components. Note that the vertical scale is different for the four components. The average latency of M1 was 21.9 ± 0.8 ms (mean ± SEM), 20.4 ± 0.6 ms, and 21.0 ± 0.4 ms for SSADHD, healthy control and disease control, respectively. The average latency of M2 was 36.1 ± 1.0 ms, 33.1 ± 0.6 ms, and 32.1 ± 1.1 ms. The average latency of M3 was 62.5 ± 2.4 ms, 54.7 ± 2.0 ms, and 49.9 ± 1.8 ms. The average latency of M4 was 86.2 ± 2.3 ms, 78.8 ± 2.8 ms, and 73.5 ± 2.9 ms.

Fig. 4.

Fig. 4.

Open in a new tab

Comparison of somatosensory evoked field latency between groups. Averaged latency of peak of each component compared between succinic semialdehyde dehydrogenase deficiency (SSADHD) and combined controls shows the significant delay in each component in SSADHD. Error bar indicates standard error of the mean.

The two-way repeated measures ANCOVA showed a significant main effect of group [F(2, 34) = 10.2, p < 0.001] and component [F(2.0, 69.3) = 285.65, p < 0.001 (Greenhouse-Geisser corrected)]. There was a statistically significant interaction between group and component [F(4.1, 69.3) = 4.5, p < 0.005 (Greenhouse-Geisser corrected)]. Pairwise comparisons across groups with FDR correction showed a significant difference between SSADHD and healthy controls (p < 0.05), and SSADHD and disease controls (p < 0.001), but there was no significant difference between healthy controls and disease controls (p = 0.12). Furthermore, there were no significant differences between healthy controls and disease controls for each component [M1, p = 0.7; M2, p = 1.0; M3, p = 0.4; M4, p = 0.3]. Thus, we combined data from the healthy and disease controls to compare with the SSADHD group. For the combined controls, the average latency was M1 − 20.7 ± 0.3 ms, M2 − 32.5 ± 0.7 ms, M3 − 52.0 ± 1.4 ms, and M4 − 75.8 ± 2.0 ms. Pairwise comparisons across components with FDR correction showed a significant difference for all four components suggesting that SSADHD showed a longer latency for each component compared to the combined controls: M1, p < 0.01; M2, p < 0.001; M3, p < 0.001; and M4, p < 0.001.

We next analyzed whether the longer latency seen in the SSADHD group is due to the pathophysiology of SSADHD as a disease process per se, independently of other confounding factors. One possible confounding factor is epilepsy. We found no latency differences between the healthy controls and epilepsy disease controls (see above). However, we carried out a further analysis. As mentioned in the demographic information for the SSADHD patients, 5/13 SSADHD patients had chronic epilepsy. We tested whether epilepsy on its own affects the SEF latency by combining data from all subjects with and without epilepsy across the three groups. We divided all the subjects into epilepsy (n = 19) and non-epilepsy groups (n = 19). A two-way repeated measures ANCOVA showed that there was a significant main effect of component [F(3, 105) = 249.2, p < 0.001]. However, there was no main effect of group [F(1, 35) = 1.1, p = 0.3] or interaction between component and group [F(3, 105) = 1.1, p = 0.3]. These results further support our conclusion that epilepsy is not a contributing factor affecting the latency alteration seen in SSADHD.

We then asked whether the pathophysiology of SSADHD explains the increased latency independently of epilepsy. We divided all non-epilepsy subjects and patients into two groups (SSADHD, n = 8; healthy controls, n = 11). A two-way repeated measures ANCOVA was conducted to examine differences in the latency of each component across groups. The results showed that there was a significant main effect of component as expected [F(3, 48) = 143.4, p < 0.001]. In addition, there was a marginal significance in the main effect of group [F(1, 16) = 4.0, p = 0.06], while there was no significant interaction between component and group [F(3, 16) = 0.5, p = 0.7]. These results suggest the pathophysiology is an independent factor for the latency alteration seen in SSADHD. The marginal significance is most likely due to the small sample size.

We also tested whether the increase in SEF latency correlates with the severity of symptoms within the SSADHD patients using the partial correlation analysis (see section 2.4.3). Marginal significance was seen only between M2 latency and the psychiatric component of the CSS: M1 (r = −0.11, p = 0.74), M2 (r = −0.55, p = 0.06), M3 (r = −0.14, p = 0.65), M4 (r = −0.10, p = 0.76). The lack of significance may be due to small sample size.

4. Discussion

We demonstrated for the first time that the peak latencies of the initial components of the median-nerve evoked SEF are delayed in patients with SSADHD compared to age-matched healthy controls and disease controls. There was no statistically significant difference in latency of the SEF components between the healthy controls and disease (patients with epilepsy) controls. Some of the patients with SSADHD did exhibit epilepsy. Thus, the disease group consisting of patients with focal epilepsy provided a useful control.

It is generally accepted that epilepsy is not related to prolongation of SEF latency (Ishida et al., 2020). Consistent with this notion, the mean latencies of our epilepsy group were not different from those of healthy controls. Extrapolating from this result, we believe it is reasonable to assume that the latency delays in the SEF of the SSADHD patients, some of whom had epilepsy, were not caused by epilepsy per se.

Antiseizure medications (ASMs) may modify SEF due to their effect on the synaptic neurotransmission and the modification of cortical excitability. However, recent studies reported no significant differences (Borah and Matheshwari, 1985, Carenini et al., 1988, Sendrowski et al., 2010) or prolonged latency (Mervaala et al., 1987) in somatosensory evoked potential during treatment with ASMs. Vigabatrin used to control seizures by elevating GABA may potentially affect the SEF latency. However, none of the SSADHD and epilepsy patients received this ASM.

The delays seen in all four components support our prediction that the conduction velocity of the myelinated axons in the central nervous system of patients with SSADHD should be relatively slow as a consequence of the hypothesized thin myelin sheath in the myelinated axons in the central nervous system of patients with SSADHD similar to the murine model (Barcelo-Coblijn et al., 2007, Donarum et al., 2006). We know of no published results showing thinner myelin sheaths in the clinical condition. An earlier post-mortem study of one patient identified myelin deficiency by showing reduced volume of phospholipids, one of the most abundant classes of lipids in the myelin sheath (Walters et al., 2021). However, this is an indirect support for our hypothesis.

The peak latency of the cortical response is the sum of the conduction latency in the initial peripheral pathway (i.e., median nerve) and that in the central pathway. Clinical studies have not demonstrated peripheral neuropathy in SSADHD (Fattahi et al., 2020). Thus, our results most likely reflect delay due to poor myelination in the central pathway. In a follow up study, we plan to determine the central conduction velocity using transcranial magnetic stimulation (TMS). We also plan to directly test our hypothesis by measuring the caliber of myelinated axons using the g-ratio MRI, which measures the ratio between the caliber of myelinated axons and the total caliber of the axon plus its myelin sheath (Yu et al., 2019). We expect to provide further insights into abnormal myelination using TMS and g-ratio imaging.

Slower central conduction velocity due to thinner myelination implies that speed of communication among neurons, especially long-range connections, may be impaired in patients with SSADHD. Consequently, this slower communication might impair cognitive functioning. The conduction velocity may thus be useful for understanding the pathophysiological mechanisms underlying cognitive impairment in this disease. Consistent with this notion, we found that the SEF latencies and the psychiatric component of the CSS were marginally correlated. We plan to carry out this analysis with a larger number of subjects in the ongoing study.

Although the results of our study mainly imply slower conduction velocity due to poor axonal myelination, the latency may be affected in part by abnormal GABA regulation. In our analysis, we did not find a correlation between SEF latencies and GABA concentration determined by magnetic resonance spectroscopy (MRS) for any of the components M1–M4. Egawa et al. (Egawa et al., 2008), however, found that M1 of the SEF has a longer latency in patients with Angelman syndrome (AS), associated with decreased gene expression of GABA-A receptors. The authors proposed that functional impairment of the GABAergic system reduces synchronous feed-forward and feedback inhibition (Amitai et al., 2002, Gibson et al., 1999, Swadlow, 2003). There may be a similar mechanism in SSADHD patients delaying the SEF latencies, especially of later components M2–M4, since GABA concentration is higher and postsynaptic GABA-A receptors are downregulated in this disease. Further investigation is required to understand the pathological mechanism of SEF latency prolongation in relation to GABAergic impairment.

5. Conclusions

SEF latencies are delayed in patients with succinic semialdehyde dehydrogenase deficiency (SSADHD) compared with healthy controls and disease controls. The delay in latency supports our hypothesis that the myelin sheath thickness is reduced in the brain of patients with SSADHD, as observed previously in a murine model of SSADHD.

HIGHLIGHTS.

  • Conduction velocities of somatosensory pathway were experimentally determined in patients with succinic semialdehyde dehydrogenase deficiency (SSADHD), an inherited disorder of GABA metabolism.

  • Somatosensory latencies were delayed in SSADHD patients compared with healthy and disease controls.

  • The findings support the hypotheses that myelin sheath thickness is reduced in SSADHD.

Acknowledgments

Funding provided by Japan Society for the Promotion of Science Overseas Research Fellowship (Matsubara), NIH grant 5R01HD091142 (PI: Gibson), NIH grant Brain Initiative grant 1R01NS112183 (PI: Sundaram), NIH grant 1R01NS104585 (MNE-Python grant, PI: Hämäläinen) and NIH grant 5U01EB023820 (MPI: Hämäläinen, Okada and Mosher).

Abbreviations:

ANCOVA

analysis of covariance

ANOVA

analysis of variance

AS

Angelman syndrome

ASM

antiseizure medication

CSS

clinical severity scoring

FDR

false discovery rate

GABA

γ-aminobutyric acid

GHB

γ-hydroxybutyrate

MEG

magnetoencephalography

MRI

magnetic resonance imaging

MRS

magnetic resonance spectroscopy

SEF

somatosensory evoked magnetic field

SD

standard deviation

SEM

standard error of the mean

SSADHD

succinic semialdehyde dehydrogenase deficiency

TMS

transcranial magnetic stimulation

Footnotes

Conflict of interest statement

Declarations of conflicts of interest: none to declare.

References

  1. Amitai Y, Gibson JR, Beierlein M, Patrick SL, Ho AM, Connors BW, et al. The spatial dimensions of electrically coupled networks of interneurons in the neocortex. J Neurosci 2002;22:4142–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Attri SV, Singhi P, Wiwattanadittakul N, Goswami JN, Sankhyan N, Salomons GS, et al. Incidence and geographic distribution of Succinic Semialdehyde Dehydrogenase (SSADH) deficiency. JIMD Rep 2017;34:111–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Barcelo-Coblijn G, Murphy EJ, Mills K, Winchester B, Jakobs C, Snead OC 3rd, et al. Lipid abnormalities in succinate semialdehyde dehydrogenase (Aldh5a1−/−) deficient mouse brain provide additional evidence for myelin alterations. Biochim Biophys Acta 2007;1772:556–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Borah NC, Matheshwari MC. Effect of antiepileptic drugs on short-latency somatosensory evoked potentials. Acta Neurol Scand 1985;71:331–3. [DOI] [PubMed] [Google Scholar]
  5. Carenini L, Bottacchi E, Camerlingo M, D’Alessandro G, Mamoli A. Carbamazepine does not affect short-latency somatosensory evoked potentials: a longitudinal study in newly diagnosed epilepsy. Epilepsia 1988;29:145–8. [DOI] [PubMed] [Google Scholar]
  6. Donarum EA, Stephan DA, Larkin K, Murphy EJ, Gupta M, Senephansiri H, et al. Expression profiling reveals multiple myelin alterations in murine succinate semialdehyde dehydrogenase deficiency. J Inherit Metab Dis 2006;29:143–56. [DOI] [PubMed] [Google Scholar]
  7. Egawa K, Asahina N, Shiraishi H, Kamada K, Takeuchi F, Nakane S, et al. Aberrant somatosensory-evoked responses imply GABAergic dysfunction in Angelman syndrome. Neuroimage 2008;39:593–9. [DOI] [PubMed] [Google Scholar]
  8. Fattahi M, Bushehri A, Alavi A, Asghariazar V, Nozari A, Ghasemi Firouzabadi S, et al. Bi-allelic Mutations in ALDH5A1 is associated with succinic semialdehyde dehydrogenase deficiency and severe intellectual disability. Gene 2020. 144918. [DOI] [PubMed] [Google Scholar]
  9. Gibson JR, Beierlein M, Connors BW. Two networks of electrically coupled inhibitory neurons in neocortex. Nature 1999;402:75–9. [DOI] [PubMed] [Google Scholar]
  10. Hari R, Baillet S, Barnes G, Burgess R, Forss N, Gross J, et al. IFCN-endorsed practical guidelines for clinical magnetoencephalography (MEG). Clin Neurophysiol 2018;129:1720–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Ishida M, Jin K, Kakisaka Y, Kanno A, Kawashima R, Nakasato N. Awake state-specific suppression of primary somatosensory evoked response correlated with duration of temporal lobe epilepsy. Sci Rep 2020;10:15895. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Kakigi R Somatosensory evoked magnetic fields following median nerve stimulation. Neurosci Res 1994;20:165–74. [DOI] [PubMed] [Google Scholar]
  13. Lee HHC, McGinty GE, Pearl PL, Rotenberg A. Understanding the molecular mechanisms of Succinic Semialdehyde Dehydrogenase Deficiency (SSADHD): towards the development of SSADH-targeted medicine. Int J Mol Sci 2022:23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Matsubara T, Ahlfors SP, Mima T, Hagiwara K, Shigeto H, Tobimatsu S, et al. Bilateral representation of sensorimotor responses in benign adult familial myoclonus epilepsy: an MEG study. Front Neurol 2021;12 759866. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Mervaala E, Keranen T, Tiihonen P, Riekkinen P. The effects of carbamazepine and sodium valproate on SEPs and BAEPs. Electroencephalogr Clin Neurophysiol 1987;68:475–8. [DOI] [PubMed] [Google Scholar]
  16. Pearl PL, DiBacco ML, Papadelis C, Opladen T, Hanson E, Roullet J-B, et al. Succinic semialdehyde dehydrogenase deficiency: review of the natural history study. J Child Neurol 2021;36:1153–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Pearl PL, Gibson KM, Acosta MT, Vezina LG, Theodore WH, Rogawski MA, et al. Clinical spectrum of succinic semialdehyde dehydrogenase deficiency. Neurology 2003a;60:1413–7. [DOI] [PubMed] [Google Scholar]
  18. Pearl PL, Gibson KM, Cortez MA, Wu Y, Carter Snead O 3rd, Knerr I, et al. Succinic semialdehyde dehydrogenase deficiency: lessons from mice and men. J Inherit Metab Dis 2009;32:343–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Pearl PL, Novotny EJ, Acosta MT, Jakobs C, Gibson KM. Succinic semialdehyde dehydrogenase deficiency in children and adults. Ann Neurol 2003b;54(Suppl 6):S73–80. [DOI] [PubMed] [Google Scholar]
  20. Sekihara K Signal space separation method for a biomagnetic sensor array arranged on a flat plane for magnetocardiographic applications: a computer simulation study. J Healthc Eng 2018;2018:7689589. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Sendrowski K, Sobaniec W, Bockowski L, Kulak W, Smigielska-Kuzia J. Somatosensory evoked potentials in epileptic children treated with carbamazepine or valproate in monotherapy - a preliminary study. Adv Med Sci 2010;55:212–5. [DOI] [PubMed] [Google Scholar]
  22. Swadlow HA. Fast-spike interneurons and feedforward inhibition in awake sensory neocortex. Cereb Cortex 2003;13:25–32. [DOI] [PubMed] [Google Scholar]
  23. Taulu S, Simola J, Kajola M. MEG recordings of DC fields using the signal space separation method (SSS). Neurol Clin Neurophysiol 2004;2004:35. [PubMed] [Google Scholar]
  24. Tokatly Latzer I, Roullet J-B, Gibson KM, Pearl PL. Establishment and validation of a clinical severity scoring system for succinic semialdehyde dehydrogenase deficiency. J Inherit Metab Dis 2023;46:992–1003. [DOI] [PubMed] [Google Scholar]
  25. Walters DC, Lawrence R, Kirby T, Ahrendsen JT, Anderson MP, Roullet J-B, et al. Postmortem analyses in a patient with succinic Semialdehyde Dehydrogenase Deficiency (SSADHD): II. Histological, lipid, and gene expression outcomes in regional brain tissue. J Child Neurol 2021;36:1177–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Yu F, Fan Q, Tian Q, Ngamsombat C, Machado N, Bireley JD, et al. Imaging G-ratio in multiple sclerosis using high-gradient diffusion MRI and macromolecular tissue volume. AJNR Am J Neuroradiol 2019;40:1871–7. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES