After ECT: The longitudinal outcome in patients with severe late-life depression treated with electroconvulsive therapy

Background Patients with severe LLD suffer from a life-threatening disorder, with higher suicide rates and comorbid cognitive impairment relative to younger patients with depression. In patients suffering from severe LLD, including psychotic LLD, ECT is recommended when patients are pharmacotherapy resistant or when a delay of ECT is irresponsible due to life-threatening symptoms. ECT is the most effective treatment option for LLD, with overall remission rates between 60-80 %. Despite being safe and highly effective for the most severely depressed patients, ECT may lead to mild and transient cognitive side effects. In addition the high relapse rates following a successful acute treatment represent a major clinical problem. There is a paucity of knowledge on longitudinal outcomes after ECT in patients with severe LLD. Thesis aim The main objective of this thesis is to expand the knowledge on longitudinal outcomes after ECT in patients with severe LLD, regarding cognitive functioning, relapse of depressive symptoms and mortality. The patient cohorts used in the analyses were part of the MODECT study (N = 110) [and the Valerius cohort (N = 76), both prospective clinical cohort studies. In this thesis we examined the association between baseline clinical characteristics and the risk of relapse within six months after ECT (chapter 2). We focused on psychotic LLD and addressed whether patients with psychotic LLD show better ECT remission because of a shorter index episode duration (chapter 3). We looked into cognitive impairment during and after ECT, and we examined transient cognitive impairment and its association with structural brain characteristics (chapter 4). Also, we explored the association between BrainAge, depression severity, and ECT outcome (chapter 5). We ended with demonstrating our findings of clinical outcome five years after ECT. Relapse, cognitive outcome, and survival were studied in relation to clinical characteristics (chapter 6). Lastly, all findings are summarized and evaluated in light of existing literature. In addition, methodological considerations are discussed, implications for clinical practice will be put forward, and ideas for future research will be delineated. Concluding remarks The main message that can be drawn from this thesis is that ECT in severe LLD is highly effective, both on the short and long-term. Especially patients with psychotic LLD show high remission rates, and have a lower risk of relapsing within the first six months after ECT. These remission rates could not be explained by a shorter index episode duration. Five years after ECT their risk of relapse does not differ from those without psychotic symptoms. We demonstrated that ECT induces transient cognitive impairment. This is independent of severity of WMH, GCA, and MTA. Although patients with severe WMH show worse cognitive functioning before, during, and after ECT, they do not show a less favorable trajectory as opposed to patients with less severe WMH. After follow-up of five years, roughly one-third of remitted patients showed cognitive impairment, which was associated with worse cognitive functioning post-ECT. A survival rate of 72% five years after ECT was found. Besides age, a more recurrent course of depression influenced long-term survival. In an exploratory study, we observed an older biological than chronological brain (BrainAge gap) for LLD patients. This BrainAge gap was not associated with clinical characteristics indexing chronicity and current severity, or ECT outcome. Future studies should focus on replicating these findings in a large database of patients treated with ECT, to evaluate alternatives to measure cognitive functioning before, during and after ECT, to focus on prevention of cognitive impairment and relapse, and lastly to study whether limiting the time before ECT, would enhance effectiveness of ECT.