Cognitive Function After Electroconvulsive Therapy For Depression: Relationship To Clinical ResponseⅡ

Results 

Seventy-nine patients were randomized with the 37 patients who were assessed 4 months after the end of ECT (18 remitted, 19 non-remitted at final assessment) included in this analysis, together with 56 healthy controls. Clinical and cognitive assessments over time in the intention-to-treat patient population and patient flow through the study are given in the main study report (Anderson et al., 2017b) with timing and reasons for dropping out summarised in online Supplementary Material. Patients who dropped out were broadly similar to those completing the study in their baseline characteristics and cognitive function (online Supplementary Table S1). 

Click to cistanche tubulosa dosage for memory improvement

There was a wide separation in depression scores at the 4-month follow-up (Fig. 1a) with the highest MADRS score in the remitted group 9, and the lowest in the non-remitted group 16. Remitted patients still scored higher than the healthy controls (3.8, 95% CI 2.2–5.4 v. 0.8, 0.5–1.3, p = 0.045), while the non-remitted patients remained moderately to severely depressed (MADRS 27.2, 23.0–31.5, p < 0.001 v. healthy controls and remitted patients). The two patient groups did not differ significantly on demographic or illness-related measures at baseline (Table 1). Apart from one patient in the non-remitted group, all were on an antidepressant which was combined with an antipsychotic drug in just over half of the patients. 

Most patients were taking a selective serotonin reuptake inhibitor or serotonin and noradrenaline reuptake inhibitor; combination antidepressant treatment (usually with mirtazapine) occurred in five patients in the remitted group and four in the non-remitted group. Table 2 shows treatment-related variables. The only significant difference was that more patients in the remitted group had remitted by the end of ECT treatment than in the non-remitted group. Non-remitted patients received non-significantly more ECT treatments and hence had a slightly longer gap between baseline and end of ECT assessment. There was only a modest change in medication which did not differ between the groups.

Baseline comparisons of mood and cognition 

At baseline, the patient groups did not differ significantly from each other or healthy controls in age, sex distribution, IQ, or education (Table 1), although patients in the non-remitted group were a little younger with a higher proportion of men than the other groups. Patients had higher depression scores than healthy controls and had slightly lower MMSE scores, with impairment on all neuropsychological tests and poorer subjective memory. The two patient groups did not differ in depression severity or cognitive measures and about two-thirds expected a negative effect of ECT on memory (see Table 1).

Change in cognitive measures over time 

Subjective memory assessment 

GSE-My current memory scores (Fig. 1b) showed no significant effect of time (F3,105 = 1.363, p = 0.26) but a significant effect of group (F1,35 = 10.003, p = 0.003) and an interaction between group and time (F3,105 = 3.692, p = 0.02). Contrasts showed no significant effect of ECT (time contrast between baseline and end of ECT, p = 0.65) but a significant group × time interaction between baseline and 4-month follow-up ( p = 0.001) and a trend between baseline and 1-month follow-up ( p = 0.063). Non-remitted patients did not significantly change over time, whereas remitted patients showed improvement after ECT, with values at 1- and 4-month follow-ups similar to healthy controls. 

In the remitted group, 2/18 rated their memory worse at the 4-month follow-up than baseline compared with 10/19 in the non-remitted group ( p = 0.02). Patients’ self-evaluation of the effect of ECT on the GSE-My (Fig. 1c) showed a significant effect of time (F3,105 = 10.868, p < 0.001) and a trend to a group effect (F1,35 = 2.928, p = 0.096) but no interaction between group and time (F3,105 = 1.300, p = 0.28). Overall patients reported a negative effect of ECT on memory which plateaued between 1- and 4-month follow-up in non remitters but returned towards no effect in remitters. Time contrasts showed this negative evaluation was significant at all time points compared to the baseline for patients taken together ( p ⩽ 0.001). 

There were no significant group × time contrasts in keeping with the overall result ANOVA (baseline compared with a 4-month follow-up, p = 0.057) although the final value in remitters did not significantly differ from 0 (i.e. no effect of ECT, p = 0.3). Of the 25 patients who expected a negative effect of ECT on memory before ECT, 9/13 (69%) non-remitters and 4/12 (33%) remitters reported a negative effect at follow-up; for those who expected no effect or a positive effect, the respective figures for a negative effect at follow-up were 3/6 (50%) and 2/6 (33%). The differences in proportions were not significant ( p = 0.3).

Objective anterograde memory and retrograde biographical memory 

Delayed verbal memory measured with the HVLT-R-DR (Fig. 2a) showed a significant effect of time (F3,105 = 3.425, p = 0.02) and a trend for a group × time interaction (F3,105 = 2.490, p = 0.065). Time contrasts showed that the slight decline in the number of words recalled between baseline and end of ECT was not significant ( p = 0.21), but was followed by a significant improvement between the end of ECT and 4-month follow-up ( p = 0.02). Remitters improved significantly compared to non-remitters between baseline and 4-month follow-up (time × group contrast p = 0.04) with a trend at 1-month follow-up ( p = 0.07) and did not differ significantly from the healthy controls at both follow-up assessments ( p > 0.5). Non-remitters remained impaired compared to healthy controls at the 4-month follow-up ( p < 0.01).

Delayed reproduction of a complex figure assessed with the MCGCFT (Fig. 2b) changed significantly with time (F3,105 = 13.082, p < 0.001) with no differential effect by group (group × time F3,105 = 0.205, p = 0.88). Time contrasts showed a significant decline in score after ECT compared with baseline ( p = 0.005) and a subsequent improvement so that at 4-month follow-up, it was significantly better than baseline, end of ECT, and 1-month follow-up ( p ⩽ 0.007) and did not differ significantly from healthy controls for either group ( p ⩾ 0.12). Retrograde autobiographical memory consistency measured by the AMI-SF (Fig. 2c) changed significantly with time (F3,105 = 46.693, p < 0.001) with a nadir at the end of ECT and no differential effect by group (group × time F3,105 = 0.482, p = 0.69). Time contrasts showed consistency was significantly lower than 100% at all subsequent time points ( p < 0.001) as expected, but significantly increased from the end of ECT to 4-month follow-up ( p = 0.005) reaching 79% and 84% in non-remitters and remitters, respectively.

Verbal fluency and working memory 

COWAT letter fluency (Fig. 3a) showed no effect of time (F3,105 = 1.706, p = 0.18), group (F1,35 = 0.031, p = 0.86) or group × time interaction (F3,105 = 1.312, p = 0.28). At the 4-month follow-up, both groups produced fewer words than healthy controls, significantly for non-remitters ( p = 0.02) but not remitters ( p = 0.19). COWAT category fluency (Fig. 3b) changed significantly over time (F3,105 = 3.699, p = 0.014) with a significant group × time interaction (F3,105 = 3.874, p = 0.011). Remitters, but not nonremitters, showed an increase in the number of words produced from the end of ECT to the final assessment. 

Time contrasts showed a trend decrease overall between baseline and end of ECT ( p = 0.075) and a significant increase between the end of ECT and 4-month follow-up ( p = 0.002). Remitters, compared with nonremitters, had a significant increase from both baseline and end of ECT to 4-month follow-up ( p ⩽ 0.008). At the final assessment, however, both remitters and non-remitters still performed less well than healthy controls ( p ⩽ 0.02). Digit span backward (Fig. 3c) showed a trend effect of time (F3,105 = 2.153, p = 0.098), but no effect of group (F1,35 = 0.189, p = 0.67) or group × time interaction (F3,105 = 0.872, p = 0.46). Time contrasts showed no significant change between baseline and end of ECT ( p = 0.14) but a greater digit span at 4-month follow-up compared with baseline ( p = 0.049) for all patients considered together. At the 4-month follow-up, both remitters and non-remitters were still impaired compared with healthy controls ( p ⩽ 0.05).

Correlations 

At the 4-month follow-up, the GSE-My global memory correlated negatively with the MADRS score (ρ = −0.56, p < 0.001) and positively with the COWAT category fluency (ρ = 0.48, p = 0.003) and the AMI-SF percentage consistency (ρ = 0.34, p = 0.04).

how does cistanche protect neurons? why

Cistanche contains several bioactive compounds that may have neuroprotective effects. One of these compounds is echinacoside, which has been shown to have antioxidant and anti-inflammatory effects. These properties may help protect neurons from oxidative stress and inflammation, two factors that can damage and kill neurons.

Another potential mechanism by which cistanche may protect neurons is through its ability to modulate certain signaling pathways in the brain. For example, cistanche has been found to increase levels of a protein called brain-derived neurotrophic factor (BDNF), which is important for the growth and survival of neurons.

Finally, cistanche may also have neuroprotective effects through its ability to improve blood flow to the brain. By improving blood flow, cistanche may help deliver oxygen and nutrients to neurons, which can also help prevent damage.

Overall, while more research is needed to fully understand how cistanche protects neurons, its antioxidant and anti-inflammatory effects, ability to modulate signaling pathways, and improve blood flow all suggest that it may have potential as a neuroprotective agent.

Ian M. Anderson1 , R. Hamish McAllister-Williams2 , Darragh Downey3, Rebecca Elliott1 and Colleen Loo4 

1 Neuroscience and Psychiatry Unit, University of Manchester, Manchester Academic Health Science Centre, Manchester, UK; 

2 Newcastle University, Cumbria, Northumberland Tyne and Wear NHS Foundation Trust, Newcastle upon Tyne, UK; 

3 Faculty of Biology, Medicine and Health, University of Manchester, Manchester, UK and 

4 University of New South Wales, Black Dog Institute & St George Hospital, Sydney, Australia