In this study, we showed the effects of LEV as first ASM in a cohort of people with a new diagnosis of TLE through resting-state EEG microstate analysis. Our main findings can be summarized as follows: (i) LEV treatment induced a reduction of microstates B and D’s mean duration and (ii) microstates B and D’s occurrences per second and; (iii) LEV treatment increased the directional predominance of microstate A to C and microstate B to D.
4.1 Microstate Metrics
Pharmaco-EEG studies in epilepsy have usually focused on assessing frequency modifications induced by old-generation ASMs, either visually or through quantitative analysis (Sannita et al. 1989; Wu and Xiao 1996, 1997; Höller et al. 2018). Only recently, previous works from our group showed that new-generation ASM therapy can induce a “normalization” of the EEG power spectrum and connectivity features in people with different types of epilepsy (Pellegrino et al. 2018; Lanzone et al. 2021; Ricci et al. 2021) and that such modifications are also predictive of good clinical outcome in TLE (Croce et al. 2021). Yet, to our best knowledge, microstate EEG analysis to evaluate the effects of ASMs in people with epilepsy has never been attempted before.
Microstate analysis is able to quantify long-range functional balances of networks, and such information has already been shown to provide prognostic information in patients after acute ischemic stroke (Zappasodi et al. 2017), probably because focal ischemic lesions directly impair the whole brain’s multi-scale systemic activity (Zappasodi et al. 2014).
Here, we found a reduction in microstate duration and occurrence in patients with TLE after three months of LEV therapy (Fig. 4). This is consistent with the literature since there is evidence that the metrics derived from microstate dynamics share rather similar neurophysiological implications (Zappasodi et al. 2017). The mean duration, named in several works as average lifespan, is considered to reveal the stability of the underlying neural networks, whereas the frequency of occurrence may reflect the propensity of a particular microstate and its neural generator to become activated.
Hence, a drop in microstate metrics can be considered as disentanglement and instability of the neural network generating the microstate topography, whereas an increase may be a sign of dysfunctional hyperactivity (Zappasodi et al. 2017). This is interesting since epilepsy has been increasingly recognized as a disorder of cortical networks (Englot et al. 2015; Assenza et al. 2020), and several studies have proposed that the EEG of people with epilepsy is characterized by abnormally synchronized networks and enhanced hyperconnectivity of the epileptogenic focus (Mormann et al. 2000; Iandolo et al. 2021; Ricci et al. 2021). Taken together, our results and previous works suggest a scenario in which LEV is able to disrupt a probably hyperfunctioning and abnormally active epileptic network. This notion is supported by the reduction in microstate metrics and the general good clinical response to LEV therapy in our cohort (11.1% with a < 50% reduction in seizure frequency; see Table 1).
4.2 Microstate Templates and Directional Predominance
There is evidence that EEG microstates may be different in people with epilepsy as opposed to healthy controls. Indeed, a recent study showed that people with TLE presented alterations of microstate C parameters and that such features could differentiate epilepsy from healthy controls with an accuracy of 76.1% (Rajagopalan et al. 2018). Yet, there is a lack of knowledge about the modulation of EEG microstates induced by ASMs in people with epilepsy.
The EEG microstates display recurrent topographic distributions of the ongoing scalp potential fields and are proposed to represent the neurophysiological equivalent of the typical fMRI resting-state networks (Britz et al. 2010). In particular, microstate A has been linked to phonological and auditive processes, microstate B to visual activation, microstate C to the insula-cingulate salience network, and microstate D to the central executive resting-state network (Britz et al. 2010).
Our results showed a reduction in microstate B and D metrics and an increase in the directional predominance of microstate A to C and microstate B to D in newly diagnosed TLE patients after the initiation of LEV therapy. Template B is mainly associated with visual area activity. Lehman and colleagues found that a single administration of the ASM Piracetam in healthy subjects was able to cause clockwise rotation of the topography of fronto-occipital microstates, with increasing rotation as a function of dose (Lehmann et al. 1993). We may speculate that in analogy with Piracetam, LEV may induce a perturbation in the occipital visual networks, which may represent a specific effect of ASMs activity, regardless of the mechanism of action.
Template D is mainly associated with the dorsal attention network (Britz et al. 2010). Several studies have demonstrated a selective impairment in dorsal attention networks in patients with TLE using resting-state fMRI (Zhang et al. 2009; Zhou et al. 2020). In particular, Zhou et al., reported an aberrant dorsal attention network homogeneity in patients with right TLE (Zhou et al. 2020). The reduction in microstate D observed in our cohort metrics may represent a neurophysiological biomarker of a reduction in the aberrant homogeneity of the dorsal attention network after the initiation of LEV. Yet, modifications in neuropsychological tests could not be verified in our cohort, given the retrospective nature of the study design. Changes in microstate metrics are reported to be influenced by certain behavioural states (Cantero et al. 1999) and impaired in patients with cognitive decline (Nishida et al. 2013). However, it should be mentioned that LEV is not associated with relevant cognitive side effects, as opposed to other ASMs (Gomer et al. 2007). Moreover, side-effects in our cohort of patients were rare and generally mild (18.5% of patients, see Table 1) and are unlikely to have influenced microstate parameters in our cohort. Crucially, microstate directional predominance unbalances with opposite alterations of B vs. D and A vs. C microstates as a modulation effect of LEV therapy underlines the relevance of network activity balances for brain functionality.
It is interesting to note that alterations in the dorsolateral prefrontal cortex (DLPFC), which is a key element of the salience network (microstate C) and central executive networks (microstate D), have been already described in TLE using quantitative MRI (Keller et al. 2009) and functional MRI (Qin et al. 2020). Indeed, Qin and colleagues found that TLE patients displayed impaired executive function, intrinsic alertness, and phasic alertness and orientation over time, and such cognitive decline was coupled with alterations DLPFC activity. They also described a slight improvement in cognitive functions over three years, suggesting that ASM therapy, better control of seizures, and shorter disease duration in TLE may have contributed to such modification in cognitive performance (Qin et al. 2020).
In this scenario, we may speculate that LEV may have induced a perturbation in the focal epileptic network, with reduced influence on the activity of resting-state global cortical networks. Definitely, our data support the notion that the focal alteration in the temporal lobe experienced by patients with TLE drives a global modulation of metrics and dynamics of microstates, with an increased representation of some states compared to others and higher transition probability which is triggered by the initiation of first ASM with LEV.
4.3 Limitations and Future Directions
Our study has some limitations which should be reported. The first is the non-randomized, retrospective nature of the study design since our analysis was not directly designed for a clinical application and the results found on the group level in retrospection are not applicable in a clinical setting at single-subject level. However, our study showed that EEG microstate metrics extracted with low-density scalp EEG could unveil significant effects of ASM therapy and, once confirmed in larger cohorts, might be considered for possible future clinical applications with the aim of improving medical management and offering new potential biomarkers in people with epilepsy.
All TLE patients in our cohort were drug naïve to other ASMs, and our results showed the modulation in microstate metrics induced by LEV therapy. However, the specificity of the effect of LEV on EEG microstates could not be completely pointed out at this time, given the lack of patients on other ASMs. Future studies evaluating different ASMs with multiple mechanisms of action may further clarify this point. We did not find an association between clinical variables (i.e., seizure-freedom after LEV, structural aetiology of TLE) and modulation in microstate metrics in our cohort. However, an attempt to find an association between different clinical variables (i.e., age, epilepsy duration, aetiology, location of the epileptic focus, type of ASM) and microstate metrics is beyond the scope of this work.
Finally, it is important to emphasize that our analysis was carried out using conventional low-density scalp EEG, which is low-cost and widely available in most neurophysiology units. Therefore, our approach can be applied even in centers that lack more advanced neurophysiological methods (i.e., magnetoencephalography or high-density EEG).
4.4 Conclusions
This study shows that starting LEV treatment in newly diagnosed TLE patients induces a global modulation of resting-state EEG microstates metrics, which suggests an involvement of the focal epileptic network in the more global and widespread resting-state cortical networks. Microstate modifications induced by ASMs offer new insights into the identification of the neurophysiological effects of ASMs in the epileptic brain. Microstate’s analysis has the potential to identify a common neurophysiological “fingerprint” of ASMs therapeutic efficacy that probably involves an increased neuronal inhibition of the epileptic network, irrespective of the primary pharmacodynamic target.