(PDF) Decreased Cognitive Functioning After Electroconvulsive Therapy Is Related to Increased Hippocampal Volume: Exploring the

ORIGINAL STUDY Decreased Cognitive Functioning After Electroconvulsive Therapy Is Related to Increased Hippocampal Volume Exploring the Role of Brain Plasticity Iris van Oostrom, PhD,* Philip van Eijndhoven, MD, PhD,* Elke Butterbrod, Msc,* Maria H. van Beek, MD, PhD,* Joost Janzing, MD, PhD,* Rogier Donders, PhD,† Aart Schene, MD, PhD,* and Indira Tendolkar, MD, PhD*‡ therapeutic option for depression refractory to pharmacological Objective: Electroconvulsive therapy (ECT) is still the most effective and psychological treatment and results in relatively high re- treatment of severe and therapy-refractory major depressive disorder. Cog- sponse rates of 50% to 60%.5–7 nitive side effects are the major disadvantage of ECT. Cognitive deficits are A major concern in the clinical application of ECT is the cog- generally temporary in nature and may be mediated by the hippocampus. nitive side effects. Memory has been demonstrated to be unequivo- Recent studies have shown a temporary increase in hippocampal volume cally affected,8 with patients showing impairments in anterograde and a temporary decrease in cognitive functioning post-ECT compared amnesia for recently learned information and sometimes retrograde with pre-ECT. This study investigates whether these volumetric changes amnesia for previously learned information.9,10 Adverse changes in are related to changes in cognitive functioning after ECT. nonmemory domains have also been reported,11 including process- Methods: Nineteen medication-free patients with treatment-resistant major ing speed and executive functioning. Although reports of retrograde depressive disorder underwent a whole-brain magnetic resonance imaging amnesia extending back several years are present,9,12 the adverse ef- scan and a neuropsychological examination (including the Rey auditory ver- fects of ECT on cognitive functioning are generally transient.11,13 bal learning task, Wechsler Memory Scale Visual Reproduction, fluency, Trail Cognitive performance is significantly decreased 0 to 3 days after Making Task) within 1 week before and within 1 week after the course of ending treatment compared with baseline, but starts to improve after ECT. Electroconvulsive therapy was administered twice a week bitemporally 15 days after ending treatment, compared with baseline.11 A nota- with a brief pulse. A matched healthy control group (n = 18) received the same ble observation is that the amount of electrical current administered neuropsychological examination and at a similar interval to that of the patients. to the brain was found to be related to the clinical efficacy and to the Results: Hippocampal volumes increased significantly from pretreatment severity of cognitive impairment.7 For example, several studies found to posttreatment in patients. Mean performance on cognitive tasks declined, that bilateral stimulation is moderately more effective than right uni- or remained stable, whereas performance in controls generally improved lateral stimulation, with bilateral stimulation resulting also in more because of retesting effects. The increase in hippocampal volume was re- cognitive side effects than right unilateral stimulation.14 More re- lated to changes in cognitive performance, indicating that this increase cently, it was found that high-dose right unilateral ECT (stimulating co-occurred with a decrease in cognitive functioning. at 6–8 time seizure threshold) tends to cause about similar effects on Conclusions: Our findings tentatively suggest that the temporal increase mood and similar to less effects on cognitive functioning.15–17 in hippocampal volume after treatment, which may result from neuro- The neurobiological mechanisms of ECT-induced cognitive trophic processes and is thought to be crucial for the antidepressive effect, side effects and clinical improvement remain largely unknown to is also related to the temporary cognitive side effects of ECT. date. Based on—among others—the evidence regarding the role Key Words: ECT, cognitive functioning, brain volume, hippocampus, of brain-derived neurotrophic factor18 and neuroplasticity in the impairment pathophysiology of depression,19,20 a common hypothesis posits that neurotrophic effects mediate the clinical improvement of the (J ECT 2018;00: 00–00) treatment. The neuroplasticity hypothesis is supported by several preclinical and clinical studies. In rats, neurogenesis, synaptogenesis, ith a global point prevalence of 4.7%1 and a lifetime prev- W alence ranging from 2% to 16.2%,2,3 major depressive disor- der (MDD) is one of the most common mental disorders. Despite gliogenesis, and angiogenesis have been found to occur as a result of ECT.21–24 In humans, several studies have found ECT to induce tem- porarily increased hippocampal volumes in the weeks after ECT25–30 the growing knowledge of its etiology and advances in treatment and hippocampal gray matter volumes,29,31 with mixed results re- options, MDD is reported to be treatment resistant in approximately garding hippocampal connectivity.29,30 The extent of hippocam- one-third of patients.4 Electroconvulsive therapy (ECT) involves pal volume increase was associated with the extent of clinical the elicitation of a general seizure through unilateral or bilateral improvement, further supporting the neuroplasticity hypothesis.28 administration of an electrical pulse. It is a recognized effective Initial evidence suggests that the neurotrophic effect of ECT resulting in the enlargement of the hippocampus is temporary, be- From the *Department of Psychiatry, Donders Institute for Brain, Cognition and cause hippocampal volumes return to pre-ECT levels 6 months af- Behaviour, and †Department for Health Evidence, Radboud University Medical ter ending treatment.25,32,33 Besides its often impressive clinical Center, Nijmegen, the Netherlands; and ‡Faculty of Medicine and LVR Clinic improvement, ECT thus results in temporarily increased hippo- for Psychiatry and Psychotherapy, University of Duisburg-Essen, Essen, Germany. Received for publication July 25, 2017; accepted December 19, 2017. campal volumes and—at the same time—temporarily decreased Reprints: Iris van Oostrom, PhD, Department of Psychiatry, Radboud cognitive functioning. This raises questions concerning the role University Medical Centre, PO Box 9101, 6500 HB Nijmegen, the of neurogenesis in the development of cognitive problems after Netherlands (e‐mail:

). ECT. Cognitive dysfunction may not only result from hippocam- The authors have no conflicts of interests. The study was funded by the Department of Psychiatry. pus atrophy or damage, as extensively described in the literature, Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved. but may also arise as a result of prompt, externally induced neu- DOI: 10.1097/YCT.0000000000000483 rogenesis. Memory encoding and retrieval could be impaired Journal of ECT • Volume 00, Number 00, Month 2018 www.ectjournal.com 1 Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. van Oostrom et al Journal of ECT • Volume 00, Number 00, Month 2018 because of remodeling of hippocampal circuits when newborn of ECT. Bitemporal ECT was administered twice per week at the granule cells integrate into established hippocampal circuits.34 temporal window using a brief pulse, constant current device, of To the best of our knowledge, only 2 studies have reported on which the maximum stimulus output was 1008 mC (200%; the relationship between the cognitive side effects and ECT- Thymatron System IV, Somatics, IL). Anesthesia and muscle re- induced volumetric or connectivity changes. Nordanskog et al25 laxation were induced by intravenous administration of etomidate reported an increase in hippocampal volume and a decrease in ver- and succinylcholine, respectively. Seizure threshold was deter- bal memory 1 week posttreatment compared with pretreatment, mined using stimulus titration. Stimulus application was set at but no significant relationship between the 2 factors, in a small 1.5 times the original seizure threshold. Treatment was discontinued sample of 12 patients. Increased left hippocampal volumes were if a plateau was reached in symptom improvement in the last 4 related to processing speed after ECT. In addition, Abbott et al29 sessions or if patients reached complete remission. Patients who found no significant relationship between change in hippocampal showed no or partial response underwent a minimum of 10 ade- volume and connectivity with neuropsychological performance in quate ECT sessions before treatment was discontinued. Response a group of 15 patients. This study examined cognitive performance to ECT was defined as a reduction of at least 50% of depressive at a relatively long interval of 11 ± 8 days after last ECT session, symptoms as measured with the Hamilton Rating Scale for De- when the acute neurotrophic effects may have already passed. pression (HDRS) from pre- to post-ECT. Remission was defined The aim of this study is to investigate the relationship be- by a HDRS score of less than 7 points after ECT treatment.39 The tween increased hippocampal volumes and changes in cognitive study has been approved by the local medical ethical committee. functioning after ECT. In view of the critical role of the hippo- campus in episodic and spatial memory and its involvement in Materials executive tasks,20,35–37 verbal and visual memory will be studied as well as verbal fluencies, attention, and processing speed. These Cognitive Functioning cognitive functions—also generally attenuated in depressed Processing speed, executive functioning, and verbal and vi- patients—will be studied within 1 week after ending treatment sual memory were assessed as part of a larger neuropsychological in a homogenous sample of treatment-resistant, medication- battery. The Trail Making Test (TMT)40 part A (seconds to com- free unipolar depression patients who underwent ECT.27 We ex- plete the test) was used to assess processing speed and part B (sec- pect the change in hippocampal volumes to be related to the onds to complete the test) to assess divided attention. Verbal changes in neuropsychological functioning. fluency was measured by registering the number of words produced in 1 minute. Both letter and semantic fluency (animals, professions) MATERIALS AND METHODS were assessed. Verbal episodic memory was assessed using the Rey Auditory Verbal Learning Test41 direct recall (5 trials, total number of words reproduced), and recognition (total number of word cor- Sample rectly recognized after 15 minutes). Visual episodic memory was Patients with a clinical diagnosis of MDD established through assessed using the Wechsler Memory Scale III42 Visual Reproduc- use of the Structured Clinical Interview for DSM-IV 38 and eligible tion I (direct recall) and Visual Reproduction II (delayed recall after for ECT according to the Dutch guidelines for ECT (2006) were re- ±20 minutes). The National Adult Reading Test43 was used to esti- cruited between 2009 and 2011 at our department. The patient mate the level of intellectual functioning. group consisted of a homogeneous sample of therapy refractory pa- tients who had not responded tot a stepwise treatment on serotonine Hippocampal Volumes reuptake inhibitors, serotonine-noradrenaline reuptake inhibitors, tricyclic antidepressants, and lithium or antiepileptics. A control A whole-brain magnetic resonance imaging scan was acquired using a 1.5-T Siemens Sonata whole-body scanner (Siemens, group of individuals with no history of or current psychiatric illness was recruited at the Donders Institute for Brain, Cognition and Erlangen, Germany). A 3-dimensional T1-weighted magnetization- Behaviour. Controls were matched with regard to sex, age, and ed- prepared rapid acquisition gradient-echo sequence with the following ucational level. Exclusion criteria were current or a history of neu- acquisition parameters; inversion time, 850 milliseconds; repeti- rological disorder, magnetic resonance imaging contraindications tion time, 2250 milliseconds; echo time, 3.68 milliseconds; flip (eg, claustrophobia or pregnancy), bipolar depression, comorbid angle, 15 degrees; field of view, 256 256 176 mm; and voxel schizophrenia or substance abuse, and ECT within 1 year before size, 1.0 1.0 1.0 mm. The longitudinal pipeline of FreeSurfer, the current round. We used a healthy control group instead of a version 5, was used for the volumetric analysis of the hippocam- control group with treatment-resistant patients not receiving ECT, pus. Volume is expressed in millimeters cube. because we did not want to withhold depressed patients from a treat- ment with a large chance of improvement. Moreover, a comparison Severity of Depressive Symptoms with patients with complex psychopharmacotherapy instead of Depressive symptoms were measured using the HDRS ECT may reflect other confounds. Somatic comorbidity (chronic (Hamilton, 1960) within 1 week before and within 1 week after lung disease, cardiac disease, peripheral atherosclerotic disease, the treatment. diabetes mellitus, malignant neoplasms, osteoarthritis, or rheuma- toid arthritis) and objectified cardiovascular risk factors (diabetes Statistical Analyses mellitus, hypertension, elevated cholesterol) were recorded. Data were analyzed using SPSS (IBM SPSS statistics version 22). t Tests were used to analyze differences between patients and Procedure controls in cognitive functioning. To investigate whether the This study is part of a larger study described earlier by change in cognitive functioning before treatment and after treat- Tendolkar et al.27 Neuropsychological assessment and neuroim- ment differed between patients and controls, repeated-measures aging took place 1 week before the start of ECT and within 1 week analysis of variance was used with 2 levels (measurement pre- after cessation. The controls received the same neuropsychologi- ECT and post-ECT) and 2 groups (patients and controls). To avoid cal assessments, at a similar interval. Pharmacological treatment type I errors, we performed a Bonferroni correction per neuropsy- of depression was discontinued at least 1 week before the start chological test. Differences between hippocampal volumes before 2 www.ectjournal.com © 2018 Wolters Kluwer Health, Inc. All rights reserved. Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Journal of ECT • Volume 00, Number 00, Month 2018 Cognition and Hippocampal Volume After ECT treatment and after treatment were analyzed using a paired- Changes in Cognitive Functioning From samples t test. In the subsequent analyses, we used as volumetric Pretreatment to Posttreatment Compared variable the percentage change in volume of the hippocampus With Controls structure that emerged between pretreatment and posttreatment. Results are displayed in Table 2 and Figure 1. Before ECT, Changes in cognitive functioning variables were calculated as dif- patients and controls did not differ on the neuropsychological ference scores between the pretreatment and posttreatment mea- measures, except for letter fluencies. Patients produced less words surements. By calculating variables this way, we adjusted for than did controls (t [df ] = −3.0 [31], P = 0.006); the difference re- individual differences such as age, educational level, sex, and total mained significant after Bonferroni correction. After ECT, patients brain volume, because patients were their own controls. Examina- performed less well than did controls on 4 neuropsychological mea- tion of distribution of scores using a Kolmogorov-Smirnov test sures, although some results were only borderline significant. After showed no violation of the assumption of normality in any of Bonferroni correction, one neuropsychological measure remained the variables used. Preliminary analyses with regard to linearity significant (letter fluencies, P < 0.006). There was a significant in- and homoscedasticity were conducted and no violations were teraction effect between the time of measurement and group on found. No standardized residuals of more than 3.3 or less than the semantic fluency (F1,22 = 10.2; P = 0.004) and Rey Auditory −3.3 were found, pointing out that there were no significant out- Verbal Learning Test direct recall task (F1,29 = 7.8; P = 0.009), liers. Pearson correlations were used to analyze the relationship with semantic fluency remaining significant after Bonferroni between changes in cognitive functioning and changes in hippo- correction (P < 0.008). campal volumes and depression severity from pretreatment to posttreatment. The method of Deming regression was used to dis- play the relation between changes in cognitive functioning and Changes in Hippocampal Volumes From hippocampal volumes to take measurement errors into account. Pretreatment to Posttreatment Hippocampal volumes increased significantly from pretreat- ment to posttreatment in patients (see Table 3). Pretreatment, pa- RESULTS tients' hippocampal volumes did not differ significantly from those of the controls. General and Clinical Characteristics General and clinical characteristics of the participants are Relationship Between Increase in Hippocampal displayed in Table 1. Patients did not differ from controls with Volume and Change in Cognitive Functioning regard to age, sex, educational level, and estimated IQ, demon- Correlations between the percentage change in hippocampal strating that matching was successful. A significant decrease in volume and change in cognitive functioning variables are shown depressive symptoms was found in the patients from pretreatment in Table 4 and Figure 2. Overall, the increase in hippocampal vol- (mean [SD],22.1 [5.9]) to posttreatment (mean [SD], 14.6 [7.9]; ume was negatively correlated with the performance on the neuro- t = 4.6, (df 18); P = <0.001). psychological tests, indicating that the increase in hippocampal volume co-occurred with a decrease in cognitive functioning TABLE 1. General and Clinical Characteristics of Participants (TMT-A r = 0.54, P < 0.05; TMT-B r = 0.52, P < 0.05; verbal memory recognition r = −0.5, P < 0.05). Patients Controls Correlations between change in depression severity and the (n = 19) (n = 18) P changes in cognitive performance were close to zero (see Table 4). Age, mean (SD), y 49.7 (8.1) 52.1 (8.6) 0.40 Sex, % male 37 39 0.90 DISCUSSION Education The findings of this exploratory, prospective longitudinal Lower vocational or secondary 11% 0% 0.24 study show that the increase in hippocampal volume, that is Intermediate 53% 44% known to be temporary,25,32 co-occurs with a decrease in cognitive Higher vocational or secondary 37% 56% functioning shortly after electroconvulsive stimulation, also Estimation of IQ, NART IQ (SD) 101.0 (15.3) 109.2 (14.1) 0.10 known to be temporary.11 As was found in similar studies,25–30 our patients demonstrated a significant increase in hippocampal MDD volume from pretreatment to posttreatment. Our patients also re- Recurrent 77.8% vealed a decrease in cognitive performance, comparable to that With psychotic symptoms 31.6% found in a meta-analysis of cognitive functioning 0 to 3 days after Mean (SD) age at onset, y 39.3 (10.2) ECT.11 Although preliminary and still tentative in this small sam- Mean (SD) duration of 31.7 (57.9) ple, our findings suggest for the first time that the temporal in- current episode, mo crease in hippocampal volume may be related to the cognitive Somatic comorbidities side effects of ECT. Chronic somatic comorbidity 21.1% The increase in hippocampal volume after ECT may result Cardiovascular risk factors 15.8% from increased adult neurogenesis24,28 that is assumed to also en- No. ECT sessions gender the antidepressive effect.44 In rats, adult neurogenesis from Mean (SD) 17.7 (7.3) new excitatory granule cells in the dentate gyrus goes through sev- Range 7–37 eral developmental stages that take several weeks.44 The differ- ence in properties of young neurons compared with those that Effect of ECT have already matured can offer some explanation for the acute ex- Responders 47.4% periences of cognitive dysfunction after ECT. The depolarization Remitters 26.3% threshold of new neurons is lower than that of those who have NART indicates National Adult Reading Test. matured, which means that long-term potentiation can be elic- ited through relatively weak stimulation. Moreover, neurons are © 2018 Wolters Kluwer Health, Inc. All rights reserved. www.ectjournal.com 3 Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. van Oostrom et al Journal of ECT • Volume 00, Number 00, Month 2018 TABLE 2. Performance on Neuropsychological Tasks Pre-ECT and Post-ECT in Patients and Controls Pre-ECT Post-ECT Patients, Controls, Patients, Controls, Mean (SD) Mean (SD) T (df ) P Mean (SD) Mean (SD) T (df ) P Psychomotor speed and attention TMT-A, s* 37.8 (11.6) 32.1 (9.3) 1.6 (35) 0.11 43.2 (18.5) 32.4 (11.2) 1.8 (29) 0.079 TMT-B, s* 77.3 (31.8) 75.8 (29.9) 0.1 (35) 0.89 76.0 (41.1) 69.9 (25.3) 0.46 (29) 0.65 Fluencies Letter, n words 22.0 (5.9) 29.1 (7.1) −3.1 (31) 0.004 17.4 (6.0) 29.6 (8.9) −4.5 (28) 0.000 Semantic, n words 37.3 (8.5) 32.3 (14.3) 1.1 (24) 0.28 28.8 (8.4) 35.4 (15.8) −1.5 (28) 0.15 Verbal memory Direct recall, n words 40.1 (11.6) 41.9 (7.3) −0.6 (35) 0.57 37.5 (12.0) 44.3 (8.2) −1.7 (29) 0.096 Recognition, n words 26.8 (4.3) 28.2 (1.6) −1.3 (35) 0.21 25.7 (4.2) 28.5 (1.9) −2.2 (28) 0.040 Visual memory Direct recall, VR I 76.9 (11.0) 86.7 (10.7) −1.9 (23) 0.070 81.5 (16.5) 83.6 (16.9) −0.3 (26) 0.80 Delayed recall, VR II 45.6 (25.8) 57.7 (11.0) −1.1 (23) 0.23 44.6 (28.6) 63.0 (31.4) −1.3 (26) 0.23 *The lower the score, the better the performance. VR indicates Visual Reproduction. FIGURE 1. Performance on neuropsychological tasks pre and post-ECT in patients and controls. In TMT, a lower score indicates a better performance; in other tasks, a higher score indicates a better performance. 4 www.ectjournal.com © 2018 Wolters Kluwer Health, Inc. All rights reserved. Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Journal of ECT • Volume 00, Number 00, Month 2018 Cognition and Hippocampal Volume After ECT TABLE 3. Changes in Hippocampal Volume Pre-ECT and Post-ECT in Patients Patients Controls Hippocampal Volume Pre-ECT, Mean (SD) Post-ECT, Mean (SD) T (df ) P Pre-ECT, Mean (SD) P Right 3719.4 (417.0) 3955.6 (398.0) −7.8 (18) 0.000 3762.0 (332.8) 0.80 Left 3606.6 (399.3) 3825.7 (426.7) −6.0 (18) 0.000 3639.63 (341.5) 0.84 Total 7326.0 (788.9) 7781.3 (791.7) −7.8 (18) 0.000 7401.6 (665.1) 0.81 tonically activated up to 1 week after birth. This discrepancy in ex- We found that changes in hippocampal volume were related citability might influence functionality within the entire hippo- to performance on tests that assess both memory and executive campal circuitry.45 Because adult neurogenesis takes place in the functioning. This study focused on the hippocampus only, but it dentate gyrus, this circuit could be particularly sensitive to remod- is generally acknowledged that the neuroplastic effects occur eling because of these inadequate “behaviors” of newborn neurons beyond that, for example, into the temporal lobe. Moreover, and also glial cells.46 Learning, memory, and cognitive perfor- the hippocampus is known to be involved in executive tasks mance can therefore first become temporarily impaired and subse- as well.20,35–37 Neuropsychological tests do not measure purely quently improved when new cells have been integrated. This may one cognitive function; they generally address one cognitive explain the improvement in cognitive performance after 15 days function more than others. For example, the trail making task post-ECT that can even outperform pre-ECT levels.11 and verbal fluency task not only assess executive functioning This is the first study reporting a significant relationship be- but also imply processing speed, semantic memory, and lan- tween structural and cognitive changes. The 2 other studies that guage.47,48 The relationship between hippocampal volume and have examined the relationship between structural and cognitive memory and executive functioning we found in this study can changes after ECT treatment25,29 did not support our findings. therefore be estimated as meaningful. One study may have been underpowered with only 12 patients25, Some strengths and limitations of our study need to be men- and the other study examined cognitive functioning at a relatively tioned. This is the first study—to our knowledge—to report on the long interval of 11 ± 8 days after last ECT session29; by that time, potential relationship between hippocampal and cognitive the temporary decrease in cognitive performance may have been changes after ECT using a large and balanced battery of neuropsy- disappeared. Changes in cognitive performance after ECT were chological tasks. This study included a relatively homogenous overall unrelated to clinical improvement in this sample. This may sample and eliminated some major possible confounders of previ- suggest that cognitive performance shortly after ECT is more deter- ous research, such as electrode placement differences, medication mined by current effects on the brain than by changes in depressive effects, disorder, and age differences. Other strengths are the re- symptoms, which often occur somewhat later after commencement stricted time window, assessing patients within 1 week before of antidepressant treatment. and after cessation of treatment, and the successful matching of the control group. A limitation is the sample size (n = 19), although larger than similar studies including cognitive testing. Probably be- cause of low power, changes over time were not or only borderline TABLE 4. Correlations Between Change in Cognitive significant. We included healthy controls instead of treatment- Functioning and Change in Hippocampal Volume From resistant patients not treated with ECT as a control group, mainly Pre-ECT to Post-ECT; Correlations Between Change in Cognitive for ethical reasons. It can therefore not be ruled out that some of Functioning and Change in Depression Severity (HDRS) from the changes in cognitive functioning were related to the progression Pre-ECT to Post-ECT of the disorder itself. Bonferroni corrections for multiple testing re- sulted in less significant changes in cognitive functioning after Change in Change in ECT, further questioning the validity of our findings of the decline Hippocampal Depression in cognitive functioning after ECT. However, the modest decline Change in Performance Volume Severity that we found in cognitive functioning in patients shortly after Processing speed and attention ECT has been found in many studies and is known to disappear after approximately 14 days, making it unlikely that changes in TMT-A, s* 0.54† −0.08 cognitive functioning can be attributed to the progression of the TMT-B, s* 0.52† 0.08 disorder or to multiple testing problems.11 Despite our efforts to Fluencies include a homogeneous sample, several other confounding factors Letter, n words −0.35 0.05 may have affected our findings. No information on the number of Semantic, n words −0.33 0.13 patients with white-matter hyperintensities, which can have an im- Verbal memory pact on cognitive functioning during ECT, was available.49 Not Direct recall, n correct words −0.40 0.12 too many white-matter hyperintensities would, however, be ex- Recognition, n correct words −0.50† −0.12 pected in our sample, because the mean age was 49.7 years. Also, Visual memory the poor response and remission rates in the studied patients may Direct recall, VR I −0.10 −0.22 have resulted in attention50 and sleep deficits and elevated anxiety levels, negatively influencing performance on neuropsychological Delayed recall, VR II −0.39 −0.01 tasks. The sample was furthermore not homogeneous with regard *A lower score (less seconds) indicates a better performance. to the number of ECT sessions, including one outlier receiving 37 †Correlation is significant at the 0.05 level (2-tailed). sessions. We recommend using larger samples in future research, VR indicates Visual Reproduction. so that all these confounders can be adjusted for in the analyses. Taking all these limitations into account, our findings can only © 2018 Wolters Kluwer Health, Inc. All rights reserved. www.ectjournal.com 5 Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. van Oostrom et al Journal of ECT • Volume 00, Number 00, Month 2018 FIGURE 2. Relation between changes in cognitive functioning and hippocampal volumes using Deming regression analysis. be considered as preliminary and tentative until they are corrobo- REFERENCES rated in other studies. Our results support the call for more exten- sive research with regard to the role of neurobiological factors that 1. Ferrari AJ, Somerville AJ, Baxter AJ, et al. Global variation in the are influenced by ECT in unwanted treatment side effects. Linking prevalence and incidence of major depressive disorder: a systematic review structural changes to interhippocampal and hippocampal-cortical of the epidemiological literature. Psychol Med. 2013;43:471–481. connectivity51 in the context of cognitive impairment would add 2. Moussavi S, Chatterji S, Verdes E, et al. Depression, chronic diseases, and to the more traditional structure-function view that was used as ba- decrements in health: results from the World Health Surveys. Lancet. sis for this study. 2007;370:851–858. In conclusion, we found first evidence of a relationship be- 3. Richards D. Prevalence and clinical course of depression: a review. tween structural increases and cognitive outcomes directly after Clin Psychol Rev. 2011;31:1117–1125. ECT, suggesting that there may be a link between the cognitive impairment shown in patients during and shortly after treatment 4. Ruhe HG, van Rooijen G, Spijker J, et al. Staging methods for treatment and structural changes that occur within the brain. If this holds resistant depression. A systematic review. J Affect Disord. 2012;137:35–45. true, the plasticity of the brain that is hypothesized to be crucial 5. Baldinger P, Lotan A, Frey R, et al. Neurotransmitters and electroconvulsive for the therapeutic effect of ECT may also be responsible for its therapy. J ECT. 2014;30:116–121. most important side effects. Further studies are needed to confirm 6. Pagnin D, de Queiroz V, Pini S, et al. Efficacy of ECT in depression: a or disconfirm our findings and more specifically to study the changes meta-analytic review. J ECT. 2004;20:13–20. in cognitive functioning and brain volumes at several time win- dows (1–3 days, 14 days, 3 months after ECT), in view of the pos- 7. UK ECT Review Group. Efficacy and safety of electroconvulsive therapy sible diverse biological effects of ECT that work together and the in depressive disorders: a systematic review and meta-analysis. Lancet. different developmental stages of neurogenesis after ECT. For 2003;361:799–808. clinical practice, we recommend to inform patients about these 8. Ingram A, Saling MM, Schweitzer I. Cognitive side effects of brief pulse cognitive side effects and possible ways of dealing with it. electroconvulsive therapy: a review. J ECT. 2008;24:3–9. 9. Sackeim HA, Prudic J, Fuller R, et al. The cognitive effects of electroconvulsive therapy in community settings. Neuropsychopharmacology. 2007;32:244–254. ACKNOWLEDGMENTS 10. McClintock SM, Choi J, Deng ZD, et al. Multifactorial determinants of The authors thank all patients for their participation in the neurocognitive effects of electroconvulsive therapy. J ECT. 2014;30: the study. 165–176. 6 www.ectjournal.com © 2018 Wolters Kluwer Health, Inc. All rights reserved. Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Journal of ECT • Volume 00, Number 00, Month 2018 Cognition and Hippocampal Volume After ECT 11. Semkovska M, McLoughlin DM. Objective cognitive performance 31. Dukart J, Regen F, Kherif F, et al. Electroconvulsive therapy-induced brain associated with electroconvulsive therapy for depression: a systematic plasticity determines therapeutic outcome in mood disorders. Proc Natl review and meta-analysis. Biol Psychiatry. 2010;68:568–577. Acad Sci U S A. 2014;111:1156–1161. 12. Donahue AB. Electroconvulsive therapy and memory loss: a personal 32. Bouckaert F, Dols A, Emsell L, et al. Relationship between hippocampal journey. J ECT. 2000;16:133–143. volume, serum BDNF, and depression severity following electroconvulsive 13. Bodnar A, Krzywotulski M, Lewandowska A, et al. Electroconvulsive therapy in late-life depression. Neuropsychopharmacology. 2016;41: therapy and cognitive functions in treatment-resistant depression. World J 2741–2748. Biol Psychiatry. 2016;17:159–164. 33. Oltedal L, Bartsch H, Sorhaug OJ, et al. The Global ECT-MRI Research 14. Semkovska M, Keane D, Babalola O, et al. Unilateral brief-pulse Collaboration (GEMRIC): establishing a multi-site investigation of the electroconvulsive therapy and cognition: effects of electrode placement, neural mechanisms underlying response to electroconvulsive therapy. stimulus dosage and time. J Psychiatr Res. 2011;45:770–780. Neuroimage Clin. 2017;14:422–432. 15. Semkovska M, Landau S, Dunne R, et al. Bitemporal Versus High-Dose 34. Frankland PW, Kohler S, Josselyn SA. Hippocampal neurogenesis and Unilateral Twice-Weekly Electroconvulsive Therapy for Depression forgetting. Trends Neurosci. 2013;36:497–503. (EFFECT-Dep): a pragmatic, randomized, non-inferiority trial. Am J 35. Keller SS, Baker G, Downes JJ, et al. Quantitative MRI of the prefrontal Psychiatry. 2016;173:408–417. cortex and executive function in patients with temporal lobe epilepsy. 16. Kellner CH, Knapp R, Husain MM, et al. Bifrontal, bitemporal and right Epilepsy Behav. 2009;15:186–195. unilateral electrode placement in ECT: randomised trial. Br J Psychiatry. 36. Frodl T, Schaub A, Banac S, et al. Reduced hippocampal volume 2010;196:226–234. correlates with executive dysfunctioning in major depression. J Psychiatry 17. McCall WV, Dunn A, Rosenquist PB, et al. Markedly suprathreshold right Neurosci. 2006;31:316–323. unilateral ECT versus minimally suprathreshold bilateral ECT: 37. Leirer VM, Wienbruch C, Paul-Jordanov I, et al. Hippocampal activity antidepressant and memory effects. J ECT. 2002;18:126–129. during the transverse patterning task declines with cognitive competence 18. Hashimoto K, Shimizu E, Iyo M. Critical role of brain-derived neurotrophic but not with age. BMC Neurosci. 2010;11:113. factor in mood disorders. Brain Res Brain Res Rev. 2004;45:104–114. 38. First MB, Spitzer RL, Gibbon M. Structured Clinical Interview for 19. Malberg JE, Schechter LE. Increasing hippocampal neurogenesis: a DSM-IV-TR Axis I Disorders. New York: Biometrics Research; 2002. novel mechanism for antidepressant drugs. Curr Pharm Des. 2005;11: 39. Heijnen WT, Birkenhager TK, Wierdsma AI, et al. Antidepressant 145–155. pharmacotherapy failure and response to subsequent electroconvulsive 20. Sahay A, Hen R. Adult hippocampal neurogenesis in depression. therapy: a meta-analysis. J Clin Psychopharmacol. 2010;30:616–619. Nat Neurosci. 2007;10:1110–1115. 40. Reitan RM. The relation of the trail making test to organic brain damage. 21. Madsen TM, Treschow A, Bengzon J, et al. Increased neurogenesis in a J Consult Psychol. 1955;19:393–394. model of electroconvulsive therapy. Biol Psychiatry. 2000;47: 41. Brand N, Jolles J. Learning and retrieval rate of words presented auditorily 1043–1049. and visually. J Gen Psychol. 1985;112:201–210. 22. Warner-Schmidt JL, Madsen TM, Duman RS. Electroconvulsive seizure 42. Wechsler D. Wechsler Memory Scale. 3rd ed. San Antonio: The restores neurogenesis and hippocampus-dependent fear memory after Psychological Corporation; 1997. disruption by irradiation. Eur J Neurosci. 2008;27:1485–1493. 43. Nelson HE. The National Adult Reading Test. NFER-Nelson: 23. Bouckaert F, Sienaert P, Obbels J, et al. ECT: its brain enabling effects: a Windsor; 1982. review of electroconvulsive therapy-induced structural brain plasticity. J ECT. 2014;30:143–151. 44. Kempermann G, Song H, Gage FH. Neurogenesis in the Adult Hippocampus. Cold Spring Harb Perspect Biol. 2015;7:a018812. 24. Olesen MV, Wortwein G, Folke J, et al. Electroconvulsive stimulation results in long-term survival of newly generated hippocampal neurons in 45. Doetsch F, Hen R. Young and excitable: the function of new neurons in rats. Hippocampus. 2017;27:52–60. the adult mammalian brain. Curr Opin Neurobiol. 2005;15:121–128. 25. Nordanskog P, Larsson MR, Larsson EM, et al. Hippocampal volume in 46. Akers KG, Martinez-Canabal A, Restivo L, et al. Hippocampal relation to clinical and cognitive outcome after electroconvulsive therapy in neurogenesis regulates forgetting during adulthood and infancy. Science. depression. Acta Psychiatr Scand. 2014;129:303–311. 2014;344:598–602. 26. Nordanskog P, Dahlstrand U, Larsson MR, et al. Increase in hippocampal 47. Bryan J, Luszcz MA, Crawford JR. Verbal knowledge and speed of volume after electroconvulsive therapy in patients with depression: a information processing as mediators of age differences in verbal fluency volumetric magnetic resonance imaging study. J ECT. 2010;26:62–67. performance among older adults. Psychol Aging. 1997;12:473–478. 27. Tendolkar I, van Beek M, van Oostrom I, et al. Electroconvulsive therapy 48. Greenaway MC, Smith GE, Tangalos EG, et al. Mayo older americans increases hippocampal and amygdala volume in therapy refractory normative studies: factor analysis of an expanded neuropsychological depression: a longitudinal pilot study. Psychiatry Res. 2013;214: battery. Clin Neuropsychol. 2009;23:7–20. 197–203. 49. Oudega ML, van Exel E, Wattjes MP, et al. White matter hyperintensities 28. Joshi SH, Espinoza RT, Pirnia T, et al. Structural plasticity of the and cognitive impairment during electroconvulsive therapy in severely hippocampus and amygdala induced by electroconvulsive therapy in major depressed elderly patients. Am J Geriatr Psychiatry. 2014;22:157–166. depression. Biol Psychiatry. 2016;79:282–292. 50. Culpepper L. Impact of untreated major depressive disorder on cognition 29. Abbott CC, Jones T, Lemke NT, et al. Hippocampal structural and and daily function. J Clin Psychiatry. 2015;76:e901. functional changes associated with electroconvulsive therapy response. 51. Kong XM, Xu SX, Sun Y, et al. Electroconvulsive therapy changes the Transl Psychiatry. 2014;4:e483. regional resting state function measured by regional homogeneity (ReHo) 30. Jorgensen A, Magnusson P, Hanson LG, et al. Regional brain volumes, and amplitude of low frequency fluctuations (ALFF) in elderly major diffusivity, and metabolite changes after electroconvulsive therapy for depressive disorder patients: an exploratory study. Psychiatry Res. 2017; severe depression. Acta Psychiatr Scand. 2016;133:154–164. 264:13–21. © 2018 Wolters Kluwer Health, Inc. All rights reserved. www.ectjournal.com 7 Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.