Temporal lobe

Most of the reviewed studies found that ECT was followed by increased CT in the temporal lobe, including the superior, middle, and inferior temporal gyri and temporal pole (Pirnia et al., Reference Pirnia, Joshi, Leaver, Vasavada, Njau, Woods, Espinoza and Narr2016; Sartorius et al., Reference Sartorius, Demirakca, Böhringer, Clemm von Hohenberg, Aksay, Bumb, Kranaster and Ende2016; van Eijndhoven et al., Reference van Eijndhoven, Mulders, Kwekkeboom, van Oostrom, van Beek, Janzing, Schene and Tendolkar2016; Gbyl et al., Reference Gbyl, Rostrup, Raghava, Carlsen, Schmidt, Lindberg, Ashraf, Jørgensen, Larsson, Rosenberg and Videbech2019; Gryglewski et al., Reference Gryglewski, Baldinger-Melich, Seiger, Godbersen, Michenthaler, Klöbl, Spurny, Kautzky, Vanicek, Kasper, Frey and Lanzenberger2019; Xu et al., Reference Xu, Wang, Bai, Zhang, Li, Hu, Li, Zhang, Wei, Tian and Wang2019; Yrondi et al., Reference Yrondi, Nemmi, Billoux, Giron, Sporer, Taib, Salles, Pierre, Thalamas, Rigal, Danet, Pariente, Schmitt, Arbus and Péran2019; Ji et al., Reference Ji, Li, Liao, Wang, Zhang, Bai, Zhang, Xie, He, Zhu, Dukart, Baeken, Tian and Wang2022). Although most increases were observed near the placement of the electrodes, temporal changes were also observed when the electrodes were located in the frontal lobe (Xu et al., Reference Xu, Wang, Bai, Zhang, Li, Hu, Li, Zhang, Wei, Tian and Wang2019; Ji et al., Reference Ji, Li, Liao, Wang, Zhang, Bai, Zhang, Xie, He, Zhu, Dukart, Baeken, Tian and Wang2022). Despite decades of use, the precise mechanisms by which ECT exerts its therapeutic effects remain poorly understood (Fridgeirsson et al., Reference Fridgeirsson, Deng, Denys, van Waarde and van Wingen2021; Deng et al., Reference Deng, Robins, Regenold, Rohde, Dannhauer and Lisanby2023). Among the hypotheses of ECT action, three main theories have been proposed: the generalised seizure theory, the neuroendocrine–diencephalic theory, and a combined anatomical–ictal theory (Bolwig, Reference Bolwig2011). According to the generalised seizure theory, the therapeutic effects of ECT depend on the elicitation of generalised seizures. Indeed, as first demonstrated by Ottosson in 1960, generalised seizures are essential for the efficacy of ECT, and subconvulsive stimuli have weaker or no antidepressant effect (Ottosson, Reference Ottosson1960). This theory is supported by evidence showing that bilateral electrode placement, which induces a more pronounced seizure generalisation, presents a higher therapeutic efficacy compared to unilateral electrode placement (Bolwig, Reference Bolwig2011). In addition, seizure-inducing drugs, such as flurothyl and pentylendetetrazol, show similar clinical effects to ECT (Karliner and Padula, Reference Karliner and Padula1960; Cooper and Fink, Reference Cooper and Fink2014). However, recent research has challenged the notion that widespread seizure activity is the main responsible for ECT efficacy in depression, given the evidence that increasing the strength of the electric field can affect the clinical effectiveness of ECT, without affecting the characteristics of the seizure (Argyelan et al., Reference Argyelan, Oltedal, Deng, Wade, Bikson, Joanlanne, Sanghani, Bartsch, Cano, Dale, Dannlowski, Dols, Enneking, Espinoza, Kessler, Narr, Oedegaard, Oudega, Redlich, Stek, Takamiya, Emsell, Bouckaert, Sienaert, Pujol, Tendolkar, van Eijndhoven, Petrides, Malhotra and Abbott2019). Notably, current literature suggests a higher intensity of the electric field may be associated with greater neuronal plasticity after treatment (Argyelan et al., Reference Argyelan, Oltedal, Deng, Wade, Bikson, Joanlanne, Sanghani, Bartsch, Cano, Dale, Dannlowski, Dols, Enneking, Espinoza, Kessler, Narr, Oedegaard, Oudega, Redlich, Stek, Takamiya, Emsell, Bouckaert, Sienaert, Pujol, Tendolkar, van Eijndhoven, Petrides, Malhotra and Abbott2019). In this regard, animal studies have shown that electroconvulsive seizures stimulate dendritic arborisation, synaptic density, and glial proliferation (Ongür et al., Reference Ongür, Pohlman, Dow, Eisch, Edwin, Heckers, Cohen, Patel and Carlezon2007; Maynard et al., Reference Maynard, Hobbs, Rajpurohit and Martinowich2018). Furthermore, a recent quantitative MRI study that explored brain’s myelin, iron, and tissue water content at multiple time-points before, during and after ECT treatment, suggested that myelin changes rather than oedema are associated with the longitudinal effects of ECT (Gyger et al., Reference Gyger, Ramponi, Mall, Swierkocz-Lenart, Stoyanov, Lutti, von Gunten, Kherif and Draganski2021). This may be a consequence of the increased expression and release of BDNF induced by ECT (Polyakova et al., Reference Polyakova, Schroeter, Elzinga, Holiga, Schoenknecht, de Kloet and Molendijk2015), which could ultimately be the basis for the therapeutic efficacy of ECT. Taken together, this evidence seems to support the hypothesis that seizure activity synergistically with an electric field is involved in the clinical efficacy of ECT and has a profound effect on the biology of the human brain. Unfortunately, only two of the included studies reported information about seizure duration (Gryglewski et al., Reference Gryglewski, Baldinger-Melich, Seiger, Godbersen, Michenthaler, Klöbl, Spurny, Kautzky, Vanicek, Kasper, Frey and Lanzenberger2019; Van Eijndhoven et al., Reference van Eijndhoven, Mulders, Kwekkeboom, van Oostrom, van Beek, Janzing, Schene and Tendolkar2016) and no studies described the electric field strength, so we cannot draw any conclusion on the association between ECT-induced seizure activity or ECT-induced electric field, CT changes, and antidepressant effect.

The temporal lobe plays a crucial role in the pathophysiology of depression, as it is involved in the core neuropsychological features that are commonly disrupted in MDD and bipolar depression, including emotional processing and social cognition (Takahashi et al., Reference Takahashi, Yücel, Lorenzetti, Walterfang, Kawasaki, Whittle, Suzuki, Pantelis and Allen2010; Gillissie et al., Reference Gillissie, Lui, Ceban, Miskowiak, Gok, Cao, Teopiz, Ho, Lee, Rosenblat and McIntyre2022). Interestingly, cortical abnormalities in the temporal lobe have been commonly described in MDD and BD (Schmaal et al., Reference Schmaal, Hibar, Sämann, Hall, Baune, Jahanshad, Cheung, van Erp, Bos, Ikram, Vernooij, Niessen, Tiemeier, Hofman, Wittfeld, Grabe, Janowitz, Bülow, Selonke, Völzke, Grotegerd, Dannlowski, Arolt, Opel, Heindel, Kugel, Hoehn, Czisch, Couvy-Duchesne, Rentería, Strike, Wright, Mills, de Zubicaray, McMahon, Medland, Martin, Gillespie, Goya-Maldonado, Gruber, Krämer, Hatton, Lagopoulos, Hickie, Frodl, Carballedo, Frey, van Velzen, Penninx, van Tol, van der Wee, Davey, Harrison, Mwangi, Cao, Soares, Veer, Walter, Schoepf, Zurowski, Konrad, Schramm, Normann and Schnell2017; Hibar et al., Reference Hibar, Westlye, Doan, Jahanshad, Cheung, Ching, Versace, Bilderbeck, Uhlmann, Mwangi, Krämer, Overs, Hartberg, Abé, Dima, Grotegerd, Sprooten, Bøen, Jimenez, Howells, Delvecchio, Temmingh, Starke, Almeida, Goikolea, Houenou, Beard, Rauer, Abramovic, Bonnin, Ponteduro, Keil, Rive, Yao, Yalin, Najt, Rosa, Redlich, Trost, Hagenaars, Fears, Alonso-Lana, van Erp, Nickson, Chaim-Avancini, Meier, Elvsåshagen, Haukvik, Lee, Schene, Lloyd, Young, Nugent, Dale, Pfennig, McIntosh, Lafer, Baune, Ekman, Zarate, Bearden, Henry, Simhandl, McDonald, Bourne, Stein, Wolf, Cannon, Glahn, Veltman, Pomarol-Clotet, Vieta, Canales-Rodriguez, Nery, Duran, Busatto, Roberts, Pearlson, Goodwin, Kugel, Whalley, Ruhe, Soares, Fullerton, Rybakowski, Savitz, Chaim, Fatjó-Vilas and Soeiro-de-Souza2018). Furthermore, an accumulating body of evidence shows that the neurobiology of depression involves structural and functional dysfunctions in the corticolimbic network, particularly in the temporal lobe, hippocampus, and amygdala (Zhong et al., Reference Zhong, Pu and Yao2016; Tassone et al., Reference Tassone, Demchenko, Salvo, Mahmood, Di Passa, Kuburi, Rueda and Bhat2022). Recently, Leaver et al. (Reference Leaver, Espinoza, Wade and Narr2022) hypothesised a “network mechanism” of ECT, according to which seizures originating from the middle temporal lobe improve depressive symptoms by correcting or resetting corticolimbic alterations (Leaver et al., Reference Leaver, Espinoza, Wade and Narr2022). It has been suggested that the increase in temporal CT after ECT could be due to brain damage, including reactive inflammation or oedema due to the electrical current near the site of the electrode application (Coffey et al., Reference Coffey, Weiner, Djang, Figiel, Soady, Patterson, Holt, Spritzer and Wilkinson1991); thus, an increase in temporal CT after ECT could be considered a transient epiphenomenon of treatment, rather than a mechanism underlying its efficacy (Puri et al., Reference Puri, Oatridge, Saeed, Ging, McKee, Lekh and Hajnal1998). However, although some accumulation of extracellular fluid has been observed using diffusion tensor imaging (Repple et al., Reference Repple, Meinert, Bollettini, Grotegerd, Redlich, Zaremba, Bürger, Förster, Dohm, Stahl, Opel, Hahn, Enneking, Leehr, Böhnlein, Leenings, Kaehler, Emden, Winter, Heindel, Kugel, Bauer, Arolt, Benedetti and Dannlowski2020), most studies have not detected post-ECT gliosis or brain oedema (Szabo et al., Reference Szabo, Hirsch, Krause, Ende, Henn, Sartorius and Gass2007), and specific studies, using T2-relaxometry to detect significant fluid shifts, did not support oedema as a cause of structural changes detected after ECT (Girish et al., Reference Girish, Jayakumar, Murali, Gangadhar, Janakiramaiah and Subbakrishna2001; Kunigiri et al., Reference Kunigiri, Jayakumar, Janakiramaiah and Gangadhar2007; Yrondi et al., Reference Yrondi, Péran, Sauvaget, Schmitt and Arbus2018). Furthermore, a systematic review and meta-analysis of the effects of ECT on brain volumes (Gbyl and Videbech, Reference Gbyl and Videbech2018) did not support the theory of brain damage and did not find an increase in brain damage markers (Zachrisson et al., Reference Zachrisson, Balldin, Ekman, Naesh, Rosengren, Agren and Blennow2000; Agelink et al., Reference Agelink, Andrich, Postert, Würzinger, Zeit, Klotz and Przuntek2001; Kranaster et al., Reference Kranaster, Janke, Mindt, Neumaier and Sartorius2014). These findings were confirmed by Gylb et al. (Reference Gbyl, Jørgensen and Videbech2022), who found no differences in serum S100B, a marker of brain injury, after an ECT series (Gbyl et al., Reference Gbyl, Jørgensen and Videbech2022). Crucially, cortical thickening in the temporal lobe in MDD has been previously reported in association with antidepressant drug treatment (Nemati and Abdallah, Reference Nemati and Abdallah2020), suggesting that these effects could be a common mechanism underlying the effectiveness of antidepressant treatments. Consistent with the theories of neurotrophic and neuroplasticity of depression (Duman and Li, Reference Duman and Li2012), the observed changes in CT during ECT treatment may be the result of underlying increased neural growth, synaptogenesis, and synaptic rearrangement. This is in line with an increase in neurotrophins after ECT (Yrondi et al., Reference Yrondi, Péran, Sauvaget, Schmitt and Arbus2018).

Increased CT in the superior and inferior temporal gyrus was associated with an improvement in depressive symptoms only in one study (Pirnia et al., Reference Pirnia, Joshi, Leaver, Vasavada, Njau, Woods, Espinoza and Narr2016). Furthermore, Gbyl et al. (Reference Gbyl, Rostrup, Raghava, Carlsen, Schmidt, Lindberg, Ashraf, Jørgensen, Larsson, Rosenberg and Videbech2019) observed that, although CT in the temporal lobe increased after the first ECT, it returned to baseline after six months, even if the antidepressant effect persisted (Bolwig, Reference Bolwig2011).

Taken together, this evidence suggests that temporal CT is primarily affected by ECT, regardless of electrode placement. Less clear is the role of temporal CT changes in clinical improvements. Although most investigations showed that changes in CT did not correlate with clinical symptoms, this was not replicated in all included studies. Therefore, more evidence is needed to clarify the role of temporal CT in changes in depressive symptoms.

Insula

Seven investigations reported an increase in CT in the insula after ECT (Pirnia et al., Reference Pirnia, Joshi, Leaver, Vasavada, Njau, Woods, Espinoza and Narr2016; Sartorius et al., Reference Sartorius, Demirakca, Böhringer, Clemm von Hohenberg, Aksay, Bumb, Kranaster and Ende2016; van Eijndhoven et al., Reference van Eijndhoven, Mulders, Kwekkeboom, van Oostrom, van Beek, Janzing, Schene and Tendolkar2016; Gbyl et al., Reference Gbyl, Rostrup, Raghava, Carlsen, Schmidt, Lindberg, Ashraf, Jørgensen, Larsson, Rosenberg and Videbech2019; Gryglewski et al., Reference Gryglewski, Baldinger-Melich, Seiger, Godbersen, Michenthaler, Klöbl, Spurny, Kautzky, Vanicek, Kasper, Frey and Lanzenberger2019; Xu et al., Reference Xu, Wang, Bai, Zhang, Li, Hu, Li, Zhang, Wei, Tian and Wang2019; Ji et al., Reference Ji, Li, Liao, Wang, Zhang, Bai, Zhang, Xie, He, Zhu, Dukart, Baeken, Tian and Wang2022). Four of these studies presented the highest methodological scores. Previous literature has shown that the insula contributes to emotion regulation and pain perception and is also involved in the pathophysiology of MDD and bipolar depression (Nagai et al., Reference Nagai, Kishi and Kato2007; Sprengelmeyer et al., Reference Sprengelmeyer, Steele, Mwangi, Kumar, Christmas, Milders and Matthews2011; Schmaal et al., Reference Schmaal, Hibar, Sämann, Hall, Baune, Jahanshad, Cheung, van Erp, Bos, Ikram, Vernooij, Niessen, Tiemeier, Hofman, Wittfeld, Grabe, Janowitz, Bülow, Selonke, Völzke, Grotegerd, Dannlowski, Arolt, Opel, Heindel, Kugel, Hoehn, Czisch, Couvy-Duchesne, Rentería, Strike, Wright, Mills, de Zubicaray, McMahon, Medland, Martin, Gillespie, Goya-Maldonado, Gruber, Krämer, Hatton, Lagopoulos, Hickie, Frodl, Carballedo, Frey, van Velzen, Penninx, van Tol, van der Wee, Davey, Harrison, Mwangi, Cao, Soares, Veer, Walter, Schoepf, Zurowski, Konrad, Schramm, Normann and Schnell2017; Qiu et al., Reference Qiu, Liu, Zhang, Huang, Ying, Wang, Shen and Peng2020). A study showed a correlation between cortical thickening in this region and reduced severity of depression (van Eijndhoven et al., Reference van Eijndhoven, Mulders, Kwekkeboom, van Oostrom, van Beek, Janzing, Schene and Tendolkar2016). Interestingly, Bracht et al. (Reference Bracht, Walther, Breit, Mertse, Federspiel, Meyer, Soravia, Wiest and Denier2023) found an increase in CT in the right insula in both ECT and TAU responders, suggesting a potential role for CT as a marker of treatment response, which may not be specific for ECT (Bracht et al., Reference Bracht, Walther, Breit, Mertse, Federspiel, Meyer, Soravia, Wiest and Denier2023). In addition, recent evidence has shown that insular metabolic activity may represent a potential predictive biomarker of remission in MDD after treatment with antidepressants or psychotherapy (Kelley et al., Reference Kelley, Choi, Rajendra, Craighead, Rakofsky, Dunlop and Mayberg2021). Overall, this literature suggests that structural and functional insular abnormalities may play a role as biomarkers of antidepressant effects and predictive biomarkers of symptom remission, regardless of the type of treatment. In summary, the study results suggest that, although ECT is associated with an increase in insular CT, the improvement in depression symptoms after ECT does not appear to be associated with CT changes in the insula.

Frontal lobe

The frontal lobe, in particular the OFC, the superior, middle, and inferior frontal gyrus, have been consistently implicated in the pathophysiology of unipolar and bipolar depression (Drevets, Reference Drevets2007; Altshuler et al., Reference Altshuler, Bookheimer, Townsend, Proenza, Sabb, Mintz and Cohen2008; Fitzgerald et al., Reference Fitzgerald, Laird, Maller and Daskalakis2008). Three of the ten studies included in the present review showed cortical thickening in these areas (Gbyl et al., Reference Gbyl, Rostrup, Raghava, Carlsen, Schmidt, Lindberg, Ashraf, Jørgensen, Larsson, Rosenberg and Videbech2019; Xu et al., Reference Xu, Wang, Bai, Zhang, Li, Hu, Li, Zhang, Wei, Tian and Wang2019; Bracht et al., Reference Bracht, Walther, Breit, Mertse, Federspiel, Meyer, Soravia, Wiest and Denier2023) and, interestingly, only in the study by Xu et al. (Reference Xu, Wang, Bai, Zhang, Li, Hu, Li, Zhang, Wei, Tian and Wang2019) the electrodes were localised in the frontal lobe. Remarkably, Gbyl et al. (Reference Gbyl, Rostrup, Raghava, Carlsen, Schmidt, Lindberg, Ashraf, Jørgensen, Larsson, Rosenberg and Videbech2019) reported a strong association between cortical thickening of the right lateral OFC and the antidepressant effect of ECT (Gbyl et al., Reference Gbyl, Rostrup, Raghava, Carlsen, Schmidt, Lindberg, Ashraf, Jørgensen, Larsson, Rosenberg and Videbech2019). In depression, a reduced volume along with altered activity in OFC has previously been reported (Drevets, Reference Drevets2007). A recent study showed that direct stimulation of this region produced an acute improvement in depressive symptoms in epileptic subjects with moderate to severe depression (Rao et al., Reference Rao, Sellers, Wallace, Lee, Bijanzadeh, Sani, Yang, Shanechi, Dawes and Chang2018). Noticeably, OFC CT increased immediately after the ECT series but returned to baseline levels at six months of follow-up, despite the continued antidepressant effect, suggesting that the persistence of the clinical efficacy of ECT was not related to orbitofrontal CT. Interestingly, a recent GEMRIC consortium study identified a multivariate discriminative pattern of volume alterations in the cortical midline, striatal, and lateral prefrontal areas that differentiated responders from non-responders, suggesting that structural changes in prefrontal areas might play a role in the clinical response to ECT (Mulders et al., Reference Mulders, Llera, Beckmann, Vandenbulcke, Stek, Sienaert, Redlich, Petrides, Oudega, Oltedal, Oedegaard, Narr, Magnusson, Kessler, Jorgensen, Espinoza, Enneking, Emsell, Dols, Dannlowski, Bolwig, Bartsch, Argyelan, Anand, Abbott, van Eijndhoven and Tendolkar2020).

Other brain areas

ECT was associated with cortical thickening in several other areas, including the ACC, the parahippocampal gyrus, the entorhinal cortex, the fusiform gyrus, the superior and inferior parietal gyrus, the pre- and postcentral gyrus, and the lateral occipital gyrus. Although all these areas have been consistently implicated in the pathogenesis of unipolar and bipolar depression and appear to be influenced by antidepressant treatments (Bartlett et al., Reference Bartlett, DeLorenzo, Sharma, Yang, Zhang, Petkova, Weissman, McGrath, Fava, Ogden, Kurian, Malchow, Cooper, Trombello, McInnis, Adams, Oquendo, Pizzagalli, Trivedi and Parsey2018; Klimes-Dougan et al., Reference Klimes-Dougan, Westlund Schreiner, Thai, Gunlicks-Stoessel, Reigstad and Cullen2018; Yi et al., Reference Yi, Wang, Yang, Huang, Liu, Chen, Zhang, Zhao, Li, Fang and Liu2022), only increased CT in the right lateral occipital gyrus, an area involved in the perception of the face and emotion (Nagy et al., Reference Nagy, Greenlee and Kovács2012), was associated with clinical improvement (Xu et al., Reference Xu, Wang, Bai, Zhang, Li, Hu, Li, Zhang, Wei, Tian and Wang2019). Future studies with larger sample sizes are needed to clarify the effect of ECT on CT in these areas and the potential association with clinical improvements.

Biological significance of CT findings

Cortical thickening has been previously reported in patients with depression after antidepressant medications (Bartlett et al., Reference Bartlett, DeLorenzo, Sharma, Yang, Zhang, Petkova, Weissman, McGrath, Fava, Ogden, Kurian, Malchow, Cooper, Trombello, McInnis, Adams, Oquendo, Pizzagalli, Trivedi and Parsey2018; Koenig et al., Reference Koenig, Westlund Schreiner, Klimes-Dougan, Ubani, Mueller, Lim, Kaess and Cullen2018; Krause-Sorio et al., Reference Krause-Sorio, Kilpatrick, Siddarth, Ercoli, Laird, Aguilar-Faustino, Milillo, Narr and Lavretsky2020) and neurostimulation treatments (Phillips et al., Reference Phillips, Batten, Tremblay, Aldosary and Blier2015; Dalhuisen et al., Reference Dalhuisen, Ackermans, Martens, Mulders, Bartholomeus, de Bruijn, Spijker, van Eijndhoven and Tendolkar2021; Wu et al., Reference Wu, Zhang, Sun, Lei, Liu, Wang, Li, Yang and Zhang2022), suggesting that cortical changes may not be specifically associated with ECT, but rather represent a generally shared mechanism of effectiveness of various antidepressant treatments. Surprisingly, however, both pretreatment CT and the early increase in CT seem to represent predictive biomarkers of the clinical response to antidepressants (Phillips et al., Reference Phillips, Batten, Tremblay, Aldosary and Blier2015), suggesting that various processes can underlie its clinical effectiveness. Neuroplasticity, including the processes of neurogenesis, synaptogenesis, dendrogenesis, gliogenesis, and angiogenesis, has been hypothesised to be the biological mechanism for the effects of ECT on CT (Wennström et al., Reference Wennström, Hellsten and Tingström2004; Hellsten et al., Reference Hellsten, West, Arvidsson, Ekstrand, Jansson, Wennström and Tingström2005; Jaggar et al., Reference Jaggar, Ghosh, Janakiraman, Chatterjee, Maheshwari, Dewan, Hare, Deb, Figueiredo, Duman and Vaidya2023; Zhang et al., Reference Zhang, Tang, Wei, Xu, Hu, Cui, Zeng, Ye, Xie, Tang, Liu, Chen, Li, Liu and Wang2023). In particular, animal studies have reported that rats subjected to electroconvulsive stimulation show increased genesis of hippocampal neurons with the potential for long-term survival (Altar et al., Reference Altar, Whitehead, Chen, Wörtwein and Madsen2003; Madsen et al., Reference Madsen, Yeh, Valentine and Duman2005; Olesen et al., Reference Olesen, Wörtwein, Folke and Pakkenberg2017). In humans, ECT is associated with a rapid and widespread increase in GM volume in the post-treatment phase (Ousdal et al., Reference Ousdal, Argyelan, Narr, Abbott, Wade, Vandenbulcke, Urretavizcaya, Tendolkar, Takamiya, Stek, Soriano-Mas, Redlich, Paulson, Oudega, Opel, Nordanskog, Kishimoto, Kampe, Jorgensen, Hanson, Hamilton, Espinoza, Emsell, Eijndhoven, Dols, Dannlowski, Cardoner, Bouckaert, Anand, Bartsch, Kessler, Oedegaard, Dale, Oltedal, Erchinger, Haavik, Sørhaug, Jørgensen, Bolwig, Magnusson, Cano, Pujol, Menchón, Petrides and Sienaert2020), as well as an increased BDNF and vascular endothelial growth factor (VEGF) (Vanicek et al., Reference Vanicek, Kranz, Vyssoki, Komorowski, Fugger, Höflich, Micskei, Milovic, Lanzenberger, Eckert, Kasper and Frey2019; Sorri et al., Reference Sorri, Järventausta, Kampman, Lehtimäki, Björkqvist, Tuohimaa, Hämäläinen, Moilanen and Leinonen2021). However, the promptness of these volumetric changes suggests that neurogenesis is unlikely to be the only mediating factor for the volumetric effects of ECT (Ousdal et al., Reference Ousdal, Brancati, Kessler, Erchinger, Dale, Abbott and Oltedal2022). Taking into account this evidence, Ousdal et al. (Reference Ousdal, Brancati, Kessler, Erchinger, Dale, Abbott and Oltedal2022) conceptualised a time-dependent model of the neural effects of ECT that includes three consecutive phases: disruption as the immediate effect, increased neuroplasticity as the short-term effect, and rewiring as the long-term effect (Ousdal et al., Reference Ousdal, Brancati, Kessler, Erchinger, Dale, Abbott and Oltedal2022). Most of the studies included in our review examined the short-term mechanisms of ECT. The only investigation that evaluated the long-term effects of ECT reported that CT returned to baseline levels after 6 months (Gbyl et al., Reference Gbyl, Rostrup, Raghava, Carlsen, Schmidt, Lindberg, Ashraf, Jørgensen, Larsson, Rosenberg and Videbech2019). Although these results seem to indicate that CT is more sensitive in detecting immediate changes associated with brain atrophy (Lemaitre et al., Reference Lemaitre, Goldman, Sambataro, Verchinski, Meyer-Lindenberg, Weinberger and Mattay2012) compared to GM volume morphometry that can better measure neuroplasticity and rewiring processes after electrical stimulation (Argyelan et al., Reference Argyelan, Oltedal, Deng, Wade, Bikson, Joanlanne, Sanghani, Bartsch, Cano, Dale, Dannlowski, Dols, Enneking, Espinoza, Kessler, Narr, Oedegaard, Oudega, Redlich, Stek, Takamiya, Emsell, Bouckaert, Sienaert, Pujol, Tendolkar, van Eijndhoven, Petrides, Malhotra and Abbott2019), this study only included 14 patients, so the results should be interpreted with caution. Importantly, the findings of broad cortical thickening after ECT are in line with volumetric studies (Ousdal et al., Reference Ousdal, Brancati, Kessler, Erchinger, Dale, Abbott and Oltedal2022). Interestingly, two of the included studies explored GM volume in addition to CT (Sartorius et al., Reference Sartorius, Demirakca, Böhringer, Clemm von Hohenberg, Aksay, Bumb, Kranaster and Ende2016; Gryglewski et al., Reference Gryglewski, Baldinger-Melich, Seiger, Godbersen, Michenthaler, Klöbl, Spurny, Kautzky, Vanicek, Kasper, Frey and Lanzenberger2019). Although Gryglewski et al. (Reference Gryglewski, Baldinger-Melich, Seiger, Godbersen, Michenthaler, Klöbl, Spurny, Kautzky, Vanicek, Kasper, Frey and Lanzenberger2019) focused only on hippocampal and amygdala volume (Gryglewski et al., Reference Gryglewski, Baldinger-Melich, Seiger, Godbersen, Michenthaler, Klöbl, Spurny, Kautzky, Vanicek, Kasper, Frey and Lanzenberger2019), Sartorius et al. (Reference Sartorius, Demirakca, Böhringer, Clemm von Hohenberg, Aksay, Bumb, Kranaster and Ende2016) examined whole-brain GM volume and CT changes and showed a pattern of similar abnormalities involving the temporal lobe and the insula (Sartorius et al., Reference Sartorius, Demirakca, Böhringer, Clemm von Hohenberg, Aksay, Bumb, Kranaster and Ende2016), suggesting that CT and GM volume may provide complementary information that could help to understand the mechanisms underlying the differential changes in the brain across the cortical regions after ECT.

Limitations and conclusions

The findings of this review should be interpreted in light of the limitations of primary studies. First, the small sample size of the included studies could have skewed the results and decreased the power of the studies, leading to a type II error. Thus, the abnormalities described in this review might be limited to some brain areas that presented changes with large effect sizes, neglecting changes with smaller effect sizes. This could, at least partially, explain why the mega-analysis by Ousdal et al. (Reference Ousdal, Argyelan, Narr, Abbott, Wade, Vandenbulcke, Urretavizcaya, Tendolkar, Takamiya, Stek, Soriano-Mas, Redlich, Paulson, Oudega, Opel, Nordanskog, Kishimoto, Kampe, Jorgensen, Hanson, Hamilton, Espinoza, Emsell, Eijndhoven, Dols, Dannlowski, Cardoner, Bouckaert, Anand, Bartsch, Kessler, Oedegaard, Dale, Oltedal, Erchinger, Haavik, Sørhaug, Jørgensen, Bolwig, Magnusson, Cano, Pujol, Menchón, Petrides and Sienaert2020) showed more widespread ECT-related brain changes (Ousdal et al., Reference Ousdal, Argyelan, Narr, Abbott, Wade, Vandenbulcke, Urretavizcaya, Tendolkar, Takamiya, Stek, Soriano-Mas, Redlich, Paulson, Oudega, Opel, Nordanskog, Kishimoto, Kampe, Jorgensen, Hanson, Hamilton, Espinoza, Emsell, Eijndhoven, Dols, Dannlowski, Cardoner, Bouckaert, Anand, Bartsch, Kessler, Oedegaard, Dale, Oltedal, Erchinger, Haavik, Sørhaug, Jørgensen, Bolwig, Magnusson, Cano, Pujol, Menchón, Petrides and Sienaert2020). Furthermore, only six studies had a control group, and not all studies reported baseline differences in CT between patients and HC. Second, the different clinical phenotypes of the participants (e.g., comorbidities, illness phase, duration of the illness) may have contributed to the heterogeneity of the results. In addition, two studies included individuals with bipolar depression, which may have affected their results. Third, the studies did not report effect size but only significance levels. The significance of a result depends on both effect and sample size, so that with a small effect and few individuals, a low number of regions can be detected. Fourth, the heterogeneity in the positioning of the electrodes during the administration of the ECT may have also influenced CT results. Similarly, the timing of the second MRI scan was not clearly defined in all studies, leading to the speculation that some studies conducted the second MRI immediately after ECT and others completed it some days after ECT, which could have affected the CT findings. Lastly, in this literature, the sex-related difference in CT changes after ECT was not investigated, and this could be due to the lack of evidence of sex differences in CT in depression.

In conclusion, available studies lend support to the hypothesis that ECT in depression is associated with cortical thickening of the temporal lobe, insula, and, to a lesser extent, frontal lobe, and several other areas of the brain involved in the pathophysiology of depression. These findings are consistent with evidence of increased CT after pharmacological and neuromodulation treatments, suggesting that cortical changes may not be specific to ECT, but rather reflect mechanisms of therapeutic response. Future studies with higher methodological quality, longer follow-up periods, and multiple methods of evaluating morphological changes are warranted to increase knowledge of the effects of ECT on CT in MDD. Such studies may also clarify the potential role of CT and other cortical measures, including surface area, gyrification, and cortical complexity, as biomarkers of the clinical response in depression.