EEG Characteristics of Clinically Defined On-Drug-Responders and Non-Responders - A Comparison Clomipramine vs. Maprotiline

Abstract

lntroduetion80th Wirz-Justieeet al. (1976) and Fähndrieh (1983) found that those patients showing a day-I response to sleep deprivation proved to be responders to subsequent c10mipramine treatment (Clo), while those showing a day-2 response to sleep deprivation were shown to respond to maprotiline (Map). Amin (1978), Wirz-Justieeet al., (1979), and Höehli et al. (1986) could not replicate this finding. Thus, the assumption of "distinct biochemical subtypes" of depression remained controversial. Our study is an attempt to elucidate this problem by means of a vigilance-oriented analysis of the waking EEG which takes into account the time-dependent increase of subvigilant epochs. Material and MethodsAnalyzed were the resting EEGs (leads: F3-CZ, F4-CZ, Ol-CZ, 02-eZ) of 45 endogenously-<iepressed inpatients -Hami/ton Depression Scorẽ 20, meanage 51.1 ± 10.5 years, range: 25-72years, 35 9 and 10 dwho were recorded before medication (day 1,9.00 a. m.), two hours following the firstdrug infusion (day 3, 11.00 a. m.), and on the 21st day of the medication period (day 23, 11.00 a. m.). The patients were random-Iyseleeted fortreatmentwitheither Map(n = 22)orClo(n = 23). 80th drugs were given in a daily dosage of 100 mg per infusion. As a criterion of c1inical response, we agreed on a decrease of at least 50 %in the Hami/ton Depression Score on day 23 compared to day I. There were 19 re· sponders (R, mean age: 48 ± I I.3 years, range: 25-64 years) including 7 Map-R and 12 Clo-R who could be contrasted with 26 non-responders (NR, mean age: 52.2 ± 10.1 years, range: 29-72 years) including 15 Map-NR and 11 Clo-NR. As an EEG variable of drug response we chose the frequency of 2-second epochs which do not represent stage A (nonA epochs). The method ofassessing nonA epochs which makes use of4 leads has been described in detail elsewhere ( U/rich and Frick, 1986; Ulrich et al., 1987). Results a) Before medication (day I), the groups which were later deterrnined R (n = 19) and NR (n = 26) did not differ with regard to the total amount or the time-course of nonA epochs. A main effect noted was the time-dependent increase of nonA epochs (F4,172 = 4.37, P < 0.01 Greenhouse-Geisser correction) as would be expected in healthy subjects ( UJrieh et al., 1987). b) Prior to medication (day I), the subgroups of Map-R, Map-NR, Clo-R, and Clo-NR difTered with regard to the depend-r8ble 1 Frequencies of nonA epochs determined for consecutive 2-second epochs of the resting EEG: group mean values and standard deviations recording time (in minutes)

Early Changes in Prefrontal Activity Characterize Clinical Responders to Antidepressants

“…In this study, absolute and relative power did not show a statistically significant increase in the presence of medication, and we did not find any differences in absolute or relative between responders and nonresponders. The inconsistency between our findings and the prior reports of Ulrich (1988) or Knott et al (1996) may reflect differences in the medications used (maprotoline and clomipramine in Ulrich's report; imipramine in Knott's study), since those tricyclic antidepressants have patterns of activity that involve more neurotransmitter systems known to have non-specific effects on EEG activity (e.g., acetylcholine). Since absolute and relative power measures individually failed to detect the early predictive changes that were seen with cordance in our dataset, there may be an advantage in combining power measures via the cordance algorithm.…”

“…These regional measures were calculated for absolute power, relative power, and cordance values. In order to avoid spurious findings from comparing a large number of potential variables, analyses were limited to the theta band, since it has previously been shown to be sensitive to changes in activity in depression (Ulrich et al 1988;Leuchter et al 1997;Cook et al 1998a;Cook and Leuchter 2001). …”

Neuronal generators of posterior EEG alpha reflect individual differences in prioritizing personal spirituality

“…Using a larger sample and stronger methods (dense recording montage, current source density [CSD] transformation, frequency principal components analysis [fPCA]), Tenke et al (2011) identified prominent posterior alpha as predictive of a good antidepressant treatment response to a SSRI or dual therapy with a serotnergic plus nonserotonergic antidepressant. Similar differences have also been found between responders and nonresponders to a nonserotonergic antidepressant (Ulrich et al, 1988; Knott et al, 1996; Tenke et al, 2013). However, it is not yet known to what extent individuals who have positive outcomes following other treatments, or who have spontaneous remissions, might also show differences in posterior alpha.…”

Review on EEG and ERP predictive biomarkers for major depressive disorder

“…In addition, the R also exhibited a decrease in the absolute alpha power at the baseline [94]. Further studies [95,96] reported early changes in the alpha band that were shown in R only. A similar study [97] reported that less alpha current source density was associated with non-responders (NR) compared with both R and controls.…”