ECT & CRP, quo vadis? - A retrospective study of low-grade inflammation in patients with depression undergoing electroconvulsive therapy

Full text

Original data
ECT & CRP, quo vadis? - A retrospective study of low-grade inammation in
patients with depression undergoing electroconvulsive therapy
Moritz Spangemacher
a,b,c,e,*,1
, Sebastian Karl
a,c,e,2
,
Suna Su Aksay
a,c,e,3
, Eva Kathrin Lamad´
e
a,c,d,e,4
, Jana Plemper
a,e
,
Alexander Sartorius
a,c,e,5
, Bruno Pedraz-Petrozzi
a,c,d,e,6
a
Department of Psychiatry and Psychotherapy, Central Institute of Mental Health, J5, 68159 Mannheim, Medical Faculty Mannheim - University of Heidelberg,
Mannheim, Germany
b
Department of Molecular Neuroimaging, Central Institute of Mental Health, Clinical Faculty Mannheim, University of Heidelberg, Mannheim, Germany
c
German Center for Mental Health (DZPG), partner site Mannheim, Germany
d
Research Group of Stress-related Disorders, Department of Psychiatry and Psychotherapy, Central Institute of Mental Health, Clinical Faculty Mannheim, University of
Heidelberg, Mannheim, Germany
e
Research Group Brain Stimulation, Department of Psychiatry and Psychotherapy, Central Institute of Mental Health, Clinical Faculty Mannheim, University of
Heidelberg, Mannheim, Germany
ARTICLE INFO
Keywords:
Electroconvulsive therapy
Treatment response
Depression
C-reactive protein
Low-grade inammation
ABSTRACT
Introduction: Electroconvulsive therapy (ECT) is the most effective treatment for severe depression. Depression
has been associated with low-grade inammation (LGI), as indicated by elevated C-reactive protein (CRP) levels
compared to healthy individuals. While the effect of ECT on inammation markers remains unclear, some evi-
dence suggests that higher baseline CRP levels may predict remission in ECT. Additionally, LGI could inuence
the seizure threshold and thus the required stimulation dose.
Materials and methods: This retrospective study examined the potential link between LGI and treatment outcomes
as well as stimulation doses across multiple ECT sessions, controlling for age and interparticipant variability. The
hypothesis of this study was that LGI was associated with ECT remission as well as with ECT dosing in people
with depression. Two groups were dened, depending on CRP levels, those with LGI (CRP 3–10 mg/L) and those
without (CRP <3 mg/L). Generalized linear models were used to analyze maximum ECT doses, while linear
mixed models were applied to assess changes in ECT doses over time.
Results: After 10 ECT sessions, we found no signicant difference in remission rates between patients with
(n =52) and without LGI (n =143). However, patients with baseline LGI had higher maximum ECT doses. A
linear mixed model revealed that the number of sessions and baseline LGI signicantly inuenced ECT doses,
with patients with baseline LGI needing higher doses, particularly at the seventh session. Age also was associated
with both maximum doses and dose adjustments throughout the ECT series, but the inuence of LGI on ECT dose
was independent of age, since no age differences were observed between groups.
Conclusions: Baseline inammation was not associated with remission rates, but it had a signicant association
with ECT dosing. Therefore, while CRP may not be a reliable biomarker for predicting ECT response in
depression, baseline inammation could indicate the need for a higher stimulation dose.
* Corresponding author at: Department of Psychiatry and Psychotherapy, Central Institute of Mental Health, J5, 68159 Mannheim, Medical Faculty Mannheim -
University of Heidelberg, Mannheim, Germany
E-mail addresses: moritzwilliam.spangemacher@zi-mannheim.de (M. Spangemacher), sebastian.karl@zi-mannheim.de (S. Karl), sunasu.aksay@zi-mannheim.de
(S.S. Aksay), eva.lamade@zi-mannheim.de (E.K. Lamad´
e), alexander.sartorius@zi-mannheim.de (A. Sartorius), bruno.pedraz@zi-mannheim.de (B. Pedraz-Petrozzi).
1
ORCID: 0009–0007-1268–4808
2
ORCID: 0000–0001-5406–7137
3
ORCID: 0009–0002-8112–9983
4
ORCID: 0000–0001-6687–0488
5
ORCID: 0000–0002-1243–3693
6
ORCID: 0000–0002-4119–971X
Contents lists available at ScienceDirect
Biomarkers in Neuropsychiatry
journal homepage: www.sciencedirect.com/journal/biomarkers-in-neuropsychiatry
https://doi.org/10.1016/j.bionps.2024.100115
Received 11 October 2024; Received in revised form 17 November 2024; Accepted 1 December 2024
Biomarkers in Neuropsychiatry 12 (2025) 100115
Available online 4 December 2024
2666-1446/© 2024 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (
http://creativecommons.org/licenses/by-
nc-nd/4.0/ ).

1. Introduction
Major depressive disorder is one of the most disabling and difcult-
to-treat psychiatric and medical disorders in general (McIntyre et al.,
2023). Its prevalence rises continuously and there is an urgent need to
improve treatment options (Moreno-Agostino et al., 2021). Electrocon-
vulsive therapy (ECT) continues to be the most effective treatment for
the acute management of severe major depression (Espinoza and Kell-
ner, 2022; Kirov et al., 2021). Patients with a higher symptom severity
are more likely to get treated with ECT (M.B. Jørgensen et al., 2020) and
seem to respond better to ECT (Petrides et al., 2001). For depressive
patients with psychotic symptoms and for those who do not respond
adequately to conventional antidepressants, ECT should be considered
the gold standard (Ekstrand et al., 2022; Zilles-Wegner et al., 2023).
In recent years, a substantial body of research has demonstrated a
strong bidirectional relationship between inammation and psychiatric
disorders, specically in the context of depression (Beurel et al., 2020;
Pedraz-Petrozzi et al., 2024). Even though chronic low-grade inam-
mation processes are currently being discussed in the context of path-
ogenesis (Berk et al., 2013), it is unclear to what extent the dynamics of
these processes affect the underlying psychiatric illness and vice versa. It
has been hypothesized that patients whose depression could also be
etiologically related to chronic inammation represent only a certain
subgroup of the heterogeneous clinical picture of depression
(Musselman et al., 2001). Treatment-resistant depression in particular
can be associated with inammation (Strawbridge et al., 2019).
It has been postulated that repeated ECT treatments lead to a sus-
tained reduction in immune system activation while a single session of
ECT induces an acute, transient activation of the immune system
(Guloksuz et al., 2014; Yrondi et al., 2018). However, results investi-
gating the link between ECT and peripheral pro- and anti-inammatory
markers have been inconsistent (Mindt et al., 2020; Young et al., 2023).
Timing of the post-treatment biomarker assessments and duration of
treatment are likely to contribute to the heterogeneity of ndings
(Strawbridge et al., 2023).
Another recent study systematically investigated whether cytokine
levels, including interleukin-6 (IL-6), tumor necrosis factor-alpha (TNF-
α
) and C-reactive protein (CRP) changed after an ECT-series in older
depressed patients (Carlier et al., 2019). While there were no statisti-
cally signicant alterations in levels of inammatory markers after ECT,
CRP levels at baseline were signicantly higher in remitters than in
non-remitters, implicating that depressed patients with higher baseline
CRP benet more from ECT than other patients (Carlier et al., 2019). In
another study with 29 patients with treatment-resistant depression
treated with ECT, higher baseline CRP was a signicant predictor of
positive treatment outcome only for women (Kruse et al., 2018). CRP is
an acute phase protein that is inuenced by weight, age and sex dif-
ferences (Cho et al., 2021). Separately, meta-analyses have shown that
CRP is elevated in people with depression (Osimo et al., 2018). Since
Carlier and colleagues only included patients above 55 years of age, the
question arises as to whether the frequently elevated inammatory signs
in this cohort could have played a role. In addition, effectiveness of ECT
also appears to be closely related to the dose and seizure threshold
(Sackeim et al., 1991). These factors are strongly dependent on age as
well (Plemper et al., 2023).
Considering the mentioned aspects, we hypothesized that baseline
CRP correlates with remission in patients with major depressive disor-
der. To evaluate whether CRP has the potential as a biomarker of
response to ECT in clinical care, our retrospective study design is as
naturalistic as possible with inpatients in whom remission is based on
overall global clinical impression. Furthermore, we examined whether
low-grade inammation has an association with the maximum ECT
doses administered to patients with depression while accounting for age.
Our model aims to investigate how baseline CRP levels, and the number
of ECT sessions are associated with the dose of ECT administered while
controlling for age and accounting for variability between participants.
By including interparticipant variability as a random effect, the model
accounts for the repeated measures nature of the data and individual
differences among patients.
2. Materials and methods
2.1. Study design
This retrospective study is part of a more extensive study investi-
gating the age-dependent dose increase during an acute electroconvul-
sive therapy series (Plemper et al., 2023). As described and justied in
the introduction, we focused in this case on a subset of patients with
major depression (both in the context of unipolar and bipolar depres-
sion) from the same study. The following retrospective study (Fig. 1) was
conducted using the clinical records of the electroconvulsive therapy
department at the Central Institute of Mental Health. This retrospective
study was approved by the ethics committee of the Medical Faculty of
Mannheim, University of Heidelberg.
2.2. Denition of clinical low-grade inammation
From the clinical records, we extracted information on serum C-
reactive protein (CRP) levels at admission. The mean time between
blood sampling at admission and the rst ECT session was 0.43 ±0.73
days (Range: 0 – 2 days; maximum time between admission blood
sampling and the rst ECT session: 2 days). Following the denitions of
the American Heart Association and the Centers for Disease Control and
Prevention recommendations and previous studies (Pearson et al.,
2003), we dened clinical low-grade inammation (LGI) as a serum CRP
level between 3 and 10 mg/L. According to this denition, we classied
the sample into two groups: participants with no LGI (CRP <3 mg/L)
and participants with clinical LGI (CRP between 3 and 10 mg/L). This
classication was used throughout the entire analysis, presentation of
results, and discussion. Since CRP values >10 mg/L suggest marked
elevation and high inammation primarily caused by infections (Lelubre
et al., 2013), we excluded participants with these values at admission
from the analysis. Participants with white blood cell counts higher than
10,000 ×10
9
cells/L were excluded, as this also suggests high inam-
mation or infections.
For CRP measurement, 7.5 mL serum tubes (Serum-Gel®, Sarstedt,
North Rhine-Westphalia, Germany) were utilized. Following blood
collection, the samples were kept at 4◦C and transported to a clinical
laboratory in Mannheim for analysis. CRP levels were quantitatively
assessed on a Cobas c701 analyzer (Roche Industries, Basel,
Switzerland) using an enzymatic, particle-enhanced immunological
turbidity test. The detection range spanned from 0.6 to 350 mg/L, with a
dilution threshold of 350 mg/L; NaCl 0.9 % served as the dilution me-
dium with a factor of 2. CRP reference values were established at
<5 mg/L. Results below the detection limit of 0.6 mg/L were auto-
matically recorded as <0.6 mg/L.
2.3. Participants
A detailed description of the participants is provided in the retro-
spective study by Plemper et al. (Plemper et al., 2023). In brief, the rst
ECT series of all inpatients with no previous ECT history at the Central
Institute for Mental Health (CIMH Mannheim) between January 2010
and March 2021 was analyzed. ECT sessions more than 14 days apart
were considered maintenance ECT and excluded. Finally, of the 472
patients initially included in the database, only those with major
depressive disorder (both in the context of unipolar and bipolar
depression) were included in the analysis to create a more homogeneous
diagnostic group. All patients received some form of antidepressant
pharmacotherapy and psychotherapy. As it has been demonstrated that
this type (or “real world)” of psychotropic medication had no pro-
nounced or relevant inuence on seizure quality (Bundy et al., 2009), no
M. Spangemacher et al.
Biomarkers in Neuropsychiatry 12 (2025) 100115
2

co-medication was excluded. After applying the inclusion and exclusion
criteria of the analysis described here, 195 patients remained.
2.4. Electroconvulsive treatment
The characteristics of electroconvulsive therapy are described in the
previously mentioned study (Plemper et al., 2023). In summary, and
following the protocols of the Central Institute of Mental Health, the
class of anesthetic and its respective dose were selected in close
consultation with the anesthesiologist. For the ECT, a Thymatron IV
device (Somatics, LCC, Lake Bluff III) was used to administer ECT (either
uni- or bilaterally) with “double dose” selected (Plemper et al., 2023).
The use of high dose benzodiazepines was antagonized
dose-dependently with umazenil. Stimulation dose was typically
titrated within the rst session and in subsequent sessions; the dose was
mostly adapted based on the quality of the seizure or the clinical urgency
during the ECT series.
2.5. Clinical outcome
To describe clinical outcome we used the clinical global impression
improvement scale (CGI-I) determined by a single trained reviewer
retrospectively for another study (Plemper et al., 2023). It was based on
the documented response in the medical notes and the discharge letter.
The rater (JP) of the CGI-I was blinded for the results of the baseline
laboratory assessments. We classied a CGI-I of 2 or less as a clinical
remission to the treatment. Triggering a (hypo)manic episode or early
discontinuation of ECT was scored as 4 (=no change). The CGI is a
reliable clinical outcome measure well-suited for routine use in inpatient
settings. It provides several advantages, such as its proven effectiveness
in psychiatric research, responsiveness to changes in patient status, and
ease of administration (Berk et al., 2008).
2.6. Statistical analysis
Quantitative variables were presented in tables as either mean
(standard deviation) or median (interquartile ranges), while categorical
variables and count data were expressed as frequencies or fractions.
Depending on the normal distribution of the data, bivariate differences
for numerical data were calculated using t-tests or the U-Mann-Whitney
test, and the Chi-square test was applied to categorical data.
For the rst aim of the study, we used a generalized linear model to
determine the main effects of clinical LGI on remission after ECT ses-
sions, by correcting for age as an important covariate (Plemper et al.,
2023). In this case, a logistic model was used with a logit link function.
The GLM was specied: remission ~ 1 +LGI +age +LGI * age. In case of
ECT doses, these were analyzed using linear mixed models (LMM),
estimated using the restricted maximum likelihood (REML) method. We
examined the variables LGI (CRP between 3 and 10 mg/L and CRP <
3 mg/L), number of ECT sessions (rst, second, fourth, seventh, and tenth),
and participants’ age, along with the two-way interaction (number of ECT
sessions ×LGI). The ID variable was used as a cluster variable (intercept)
for the random effects. The results of the LMM were presented graphi-
cally with two-tailed p-values. Finally, we used a generalized linear
model (GLM) to determine the main effects of clinical LGI on maximum
ECT doses, by correcting for age as an important covariate (Plemper
et al., 2023). The GLM considered maximum ECT doses as the outcome
variable, with clinical LGI as a factor and age as a covariate, given that
previous studies have shown associations between age and ECT doses
(Plemper et al., 2023). The GLM was specied: maximum ECT doses ~
1+LGI +age +LGI * age. Signicance was determined at a two-tailed
p-value <0.05, as GLMs allow corrections for covariables. Finally, 95 %
condence intervals were bootstrapped for 5000 repetitions (95 %CI B)
and reported in the GLM and LMM results.
All statistical analyses were performed using the GAMLj toolbox of
the R-based software jamovi 2.5.0 (Love et al., 2024). It is important to
note that in GLM and LMM, we tested individual predictors for their
signicance within the model. Since each test is part of the
model-building process, they are not independent tests and should not
be treated as multiple comparisons in the traditional sense.
3. Results
3.1. Sample description and characteristics
Table 1 details the patient characteristics. No statistical differences
Fig. 1. Graphical summary of the study design (picture was elaborated using Biorender.com). Abbreviations: LGI =low-grade inammation, ECT
=electroconvulsive therapy, CRP =C-reactive protein.
M. Spangemacher et al.
Biomarkers in Neuropsychiatry 12 (2025) 100115
3

were observed between the two groups (with and without LGI)
regarding age (t =-1.16, df =193, p =0.247) and the total number of
ECTs (U =3546.50, p =0.623). However, signicant differences were
observed for the maximum ECT doses between the two groups (t =-
2.38, df =193, p =0.018), showing patients with baseline LGI had
higher maximum ECT doses during treatment in comparison to patients
without LGI at baseline.
Similar proportions for the female/male ratio were observed be-
tween both groups (chi-square =0.37, df =1, p =0.541). There were no
bivariate differences in remission rates between the groups (chi-square
=0.077, df =1, p =0.782).
Concerning the ECT doses per session (uncorrected), patients with
LGI at baseline showed signicantly higher mean doses at the second
(t =-2.85, df =183, p =0.005), at the fourth (t =-2.00, df =179,
p=0.047), and the seventh session (t =- 3.35, df =142, p =0.001). No
signicant differences between groups were found for the rst (t =-
0.09, df =183, p =0.927) and the tenth session (t =-1.44, df =84,
p=0.155).
3.2. Remission status and baseline LGI in depression
Firstly, we sought to investigate if baseline LGI has an effect on
remission status in patients with depression when correcting for the
effects of age. As mentioned previously, a GLM was performed. In this
case, the model explained only 1 % of the variance (R
2
=0.01). In this
case, there were no effects of baseline LGI on the remission status (OR =
1.08, bootstrapped 95 %CI [0.567; 2.619], p =0.851) or age effects on
the remission status (OR =1.02, bootstrapped 95 %CI [0.995; 1.046],
p=0.166). No signicant interaction effects were seen between age and
baseline LGI (OR =1.00, bootstrapped 95 %CI [0.953; 1.064],
p=0.943).
3.3. Effects of baseline LGI on the ECT maximum doses in depression
We sought to investigate how baseline LGI inuences the maximum
ECT doses administered to patients with depression while accounting for
the effects of age. For this purpose, a GLM was performed. The model
explained around 11 % of the variance of the data (R
2
=0.111). Both
baseline LGI (b =13.827, bootstrapped 95 %CI [0.613; 28.288],
p=0.042, Table 2), and age (b =0.782, bootstrapped 95 %CI [0.416;
1.143], p <0.001, Table 2) had a signicant effect on the maximum ECT
doses in participants with depression. No interaction effects were seen
between age and baseline LGI concerning ECT maximum doses (b =
0.354, bootstrapped 95 %CI [-0.379; 1.098], p =0.363, Table 2). Re-
sults concerning the baseline LGI status and the maximum ECT doses are
illustrated in Fig. 2.
3.4. Baseline LGI and number of sessions on ECT dose
An LMM was used to evaluate the main effects of baseline LGI and the
number of sessions on ECT doses, adjusting for age. The model explained
72.4 % of the variance (conditional R² =0.724), while xed effects
accounted for 52.0 % of the variance (marginal R² =0.520). There was a
signicant effect for the variable "number of sessions", showing a
notable increase in ECT doses as the number of sessions increased over
time (F =262.049, p <0.001; Table 3). Additionally, age had a signif-
icant effect on ECT doses (F =46.353, p <0.001; Table 3), with higher
age correlating with higher doses (b =0.521, bootstrapped 95 % CI
[0.372; 0.672], p <0.001). Signicant main effects were also found for
baseline LGI (F =6.043, p =0.015; Table 3). Moreover, there were
signicant interaction effects between baseline LGI and the number of
sessions on ECT doses (F =3.878, p =0.004; Table 3). Specically,
signicant differences in ECT doses between patients with and without
baseline LGI were observed only at the seventh session (t =-3.843, df =
Table 1
Sample description and characteristics of both groups, mean (standard devia-
tion), with low-grade inammation at baseline (i.e., CRP between 3 and 10 mg/
L) and without low-grade inammation at baseline (i.e., CRP <3 mg/L). Ab-
breviations: ECT =electroconvulsive therapy, f =female, m =male, LGI =low-
grade inammation, mC =microcoulombs. *Median (interquartile range).
Without LGI
(n ¼143)
With LGI
(n ¼52)
Age (in years) 61.24 (17.75) 64.56 (17.39)
Sex (f/m) 84/59 28/24
Remission (no/yes) 81/30 34/14
Total number of ECTs* 9.00 (5.50) 9.50 (5.00)
Maximum ECT dose (in % of 504 mC
charge)
75.42 (40.69) 92.12 (49.80)
ECT doses (in % of 504 mC charge)  
1st session 11.88 (12.23) 12.06 (9.60)
2nd session 30.98 (15.04) 39.62 (25.40)
4th session 47.52 (26.09) 56.60 (30.36)
7th session 61.80 (29.74) 82.44 (41.16)
10th session 81.72 (42.08) 97.24 (53.12)
Table 2
Results of the generalized linear model (dependent variable: maximum ECT
doses). Factor variable was here low-grade inammation status at baseline: with
low-grade inammation (i.e., CRP between 3 and 10 mg/L) and without low-
grade inammation at baseline (i.e., CRP <3 mg/L). Age (in years) was
included in the model as confounding factor as described above. 95 % con-
dence intervals were bootstrapped for 5000 repetitions. Abbreviations: SE
=standard error of the mean, LGI =low-grade inammation.
   95 % condence
intervals
 
Estimate SE Lower Upper Z P-value
LGI 13.827 6.793 0.613 28.288 2.036 0.042
Age 0.782 0.194 0.416 1.143 4.023 <0.001
LGI * Age 0.354 0.389 −0.379 1.098 0.910 0.363
Fig. 2. Graphical representation of the differences between patients with
baseline LGI (i.e., CRP between 3 and 10 mg/L) and without baseline LGI (i.e.,
CRP <3 mg/L) concerning maximal ECT doses. * p ≤0.05. LGI =low-grade
inammation.
Table 3
Results of the linear mixed model (dependent variable: ECT doses). *Group
variable was here LGI status at baseline: with LGI (i.e., CRP between 3 and
10 mg/L) and without LGI at baseline (i.e., CRP <3 mg/L). Age (in years) was
included in the model as confounding factor as described above. LGI =low-
grade inammation.
F P-value
Group-by-session interaction 3.88 0.004
Number of session main effect 262.05 <0.001
Group main effect* 6.04 0.015
Age main effect 46.35 <0.001
M. Spangemacher et al.
Biomarkers in Neuropsychiatry 12 (2025) 100115
4

589.797, p
HOLM
=0.001). Post hoc results of this interaction are
graphically illustrated in Fig. 3.
4. Discussion
Our main results indicated that, although there were signicant ef-
fects of LGI on the applied ECT doses, no signicant differences in
remission following ECT were found between participants with baseline
LGI and those without it. Regarding the association between baseline
LGI and ECT doses, we identied signicant relationships, revealing that
patients with depression and baseline LGI required higher maximum
ECT doses compared to those without baseline LGI. Furthermore, in the
analysis of bivariate uncorrected differences, the baseline LGI group
exhibited signicantly higher ECT doses in the second, fourth, and
seventh ECT sessions. However, the LMM revealed a signicant inter-
action effect between baseline LGI and the number of sessions, after
correcting for age as a covariate, indicating that the baseline LGI group
received higher ECT doses than those without LGI only during the sev-
enth session.
First, our retrospective study found no signicant associations be-
tween remission rates and baseline LGI, which contrasts with previous
reports (Carlier et al., 2019; Kruse et al., 2018). However, a larger
prospective study, which included a broader cohort of patients with
depression, was also unable to replicate the ndings of the previous
studies and is thus in line with the results of our study (Ryan and
McLoughlin, 2022).
Concerning ECT doses, this study is, to the best of our knowledge, the
rst retrospective study to analyze baseline LGI and maximum ECT
doses in depression. Our study revealed that patients with depression
and LGI required higher maximum stimulation doses compared to those
without LGI. Over the last few decades, evidence linking LGI to mood
disorders has been growing. Studies have reported, that high CRP levels
are signicantly correlated with symptom severity, increased use of
antidepressants, poor outcomes with conventional treatment, and, in
some cases, treatment resistance (Orsolini et al., 2022). Previous studies
have demonstrated that the effects of ECT are dose-dependent, meaning
that higher doses are more likely to result in treatment response
(Sackeim et al., 1991), and a higher charge is associated with better
treatment efcacy (Thirthalli et al., 2023). For instance, in depression,
treatment with high-dose ECT leads to a greater reduction in depressive
symptoms, according to a meta-analysis (The UK ECT Review Group,
2003). Since LGI is associated with symptom severity and a poor
response to conventional treatment, and ECT effects are
dose-dependent, it is plausible to suggest that individuals with depres-
sion and baseline LGI received higher maximum ECT doses compared to
those without baseline LGI. This could be because LGI is related to dis-
ease severity, and higher doses are generally associated with greater
efcacy, especially in more severe forms of the disease.
Additionally, we initially found no bivariate differences in sex and
age between the group of patients with baseline LGI and those without
it. We considered age an important covariate, as it has been established
and described with this sample in a previous study (Plemper et al.,
2023). Concerning the age of participants, this variable showed signif-
icant associations in the statistical models with maximum ECT doses and
doses per session. However, since there were no signicant age differ-
ences between patients with LGI and without LGI, the effect of LGI on
maximum ECT doses and ECT doses over the ECT sessions was inde-
pendent of the variable age. This suggests that there were no mediation
effects between age, LGI, ECT dose and remission after ECT.
It is important to acknowledge several limitations of our study.
First, the retrospective nature of the study prevents the systematic
assessment of several confounding factors. Potential variables related to
inammation, stimulation dosage or clinical outcome that could not be
thoroughly assessed due to missing documentation in the electronic
records were: Body mass index (BMI) (severely ill psychiatric patients
are not routinely weighed and measured); psychiatric and physical
comorbidities that could have an inuence on clinical outcome and low-
grade inammation; and concomitant medication. This lack of infor-
mation on possible confounding factors prevents a direct causal associ-
ation between baseline CRP and, in our case, stimulation dose. At the
same time, our retrospective study used data from routine clinical
practice rather than controlled research environments, so it reects
everyday clinical outcomes. In this sense, it may more accurately predict
in clinical practice whether baseline CRP could serve as a biomarker for
therapeutic response or the required level of stimulation dose.
Second, regarding the remission rates, while we utilized the CGI for
assessing disease severity—a valuable tool for identifying clinically
signicant effects—more subtle changes may have been better detected
through semi-structured interviews, such as the Montgomery–Åsberg
Depression Rating Scale used by Carlier and colleagues (Carlier et al.,
2019). It further limits our conclusions regarding antidepressant therapy
outcomes, as the CGI only reects the overall clinical picture and does
not allow for statements specic to individual diagnoses. Third,
although CRP is a well-documented clinical marker for inammation,
future studies should include additional inammatory parameters,
whether proteins or cellular markers. Finally, it would be ideal to have
information on previous psychiatric treatments to better control for in-
ammatory parameters.
In a future observational study, data from real clinical practice
should be assessed more thoroughly with a focus on baseline inam-
matory parameters and baseline patient characteristics. The course of
the clinical symptoms should also be recorded using a diagnosis-specic
psychometric questionnaire (e.g. MADRS for depression). Furthermore,
as elaborated by van Buel and his colleagues, the activation of the im-
mune system might be needed to mobilize neurotrophin expression (van
Buel et al., 2015). In this context, it would be interesting to evaluate the
inuence of stimulation dose on the initial increase of inammatory
markers after ECT treatment. Future studies should also explore the
inuence of inammation on factors such as seizure threshold and
seizure duration (Gillving et al., 2024). Additionally, the interactions
with inammatory processes should be examined in longitudinal
studies, such as in maintenance ECT (A. Jørgensen et al., 2024).
5. Conclusion
The relationship between LGI and the treatment of major depressive
disorder with ECT remains complex. While elevated baseline CRP levels
were linked to changes in dosage in our study, remission rates did not
Fig. 3. ECT doses (in % of 504 mC charge) and number of sessions in patients
with baseline low-grade inammation and without low-grade inammation –
results of the linear mixed models. * ** p ≤0.001. Abbreviations: S =session,
ECT =electroconvulsive therapy. LGI =low-grade inammation.
M. Spangemacher et al.
Biomarkers in Neuropsychiatry 12 (2025) 100115
5

differ between both groups. Thus, CRP cannot be used as a biomarker for
ECT response in depression, but as a possible indicator of the treatment
course, e.g., dose increase due to LGI.
CRediT authorship contribution statement
Bruno Pedraz-Petrozzi: Writing – original draft, Visualization, Su-
pervision, Methodology, Investigation, Formal analysis, Data curation,
Conceptualization. Alexander Sartorius: Writing – review & editing,
Supervision, Formal analysis, Data curation, Conceptualization. Jana
Plemper: Methodology, Investigation, Data curation. Eva Kathrin
Lamad´
e: Writing – review & editing, Supervision, Conceptualization.
Suna Su Aksay: Writing – review & editing, Supervision. Sebastian
Karl: Writing – review & editing, Methodology, Formal analysis, Data
curation, Conceptualization. Moritz Spangemacher: Writing – original
draft, Methodology, Investigation, Formal analysis, Data curation,
Conceptualization.
Declaration of Competing Interest
All authors declared no conicts of interest.
Data Availability Statement
The datasets generated and analyzed during the current case study
are not publicly available, but are available from the corresponding
author on reasonable request.
References
Berk, M., Ng, F., Dodd, S., Callaly, T., Campbell, S., Bernardo, M., Trauer, T., 2008. The
validity of the CGI severity and improvement scales as measures of clinical
effectiveness suitable for routine clinical use. J. Eval. Clin. Pract. 14 (6), 979–983.
https://doi.org/10.1111/j.1365-2753.2007.00921.x.
Berk, M., Williams, L.J., Jacka, F.N., O’Neil, A., Pasco, J.A., Moylan, S., Allen, N.B.,
Stuart, A.L., Hayley, A.C., Byrne, M.L., Maes, M., 2013. So depression is an
inammatory disease, but where does the inammation come from? BMC Med. 11,
200. https://doi.org/10.1186/1741-7015-11-200.
Beurel, E., Toups, M., Nemeroff, C.B., 2020. The bidirectional relationship of depression
and inammation: double trouble. Neuron 107 (2), 234–256. https://doi.org/
10.1016/j.neuron.2020.06.002.
Bundy, B.D., Hewer, W., Andres, F.-J., Gass, P., Sartorius, A., 2009. Inuence of
anesthetic drugs and concurrent psychiatric medication on seizure adequacy during
electroconvulsive therapy. J. Clin. Psychiatry 71 (6), 9063. https://doi.org/
10.4088/JCP.08m04971gre.
Carlier, A., Berkhof, J.G., Rozing, M., Bouckaert, F., Sienaert, P., Eikelenboom, P.,
Veerhuis, R., Vandenbulcke, M., Berkhof, J., Stek, M.L., Rhebergen, D., Dols, A.,
Exel, E. van, 2019. Inammation and remission in older patients with depression
treated with electroconvulsive therapy; ndings from the MODECT study☆.
J. Affect. Disord. 256, 509–516. https://doi.org/10.1016/j.jad.2019.06.040.
Cho, S.H., Lim, J.-E., Lee, J., Lee, J.S., Jeong, H.-G., Lee, M.-S., Ko, Y.-H., Han, C.,
Ham, B.-J., Han, K.-M., 2021. Association between high-sensitivity C-reactive
protein levels and depression: Moderation by age, sex, obesity, and aerobic physical
activity. J. Affect. Disord. 291, 375–383. https://doi.org/10.1016/j.
jad.2021.05.040.
Ekstrand, J., Fattah, C., Persson, M., Cheng, T., Nordanskog, P., Åkeson, J.,
Tingstr¨
om, A., Lindstr¨
om, M.B., Nordenskj¨
old, A., Movahed Rad, P., 2022. Racemic
ketamine as an alternative to electroconvulsive therapy for unipolar depression: a
randomized, Open-Label, Non-Inferiority Trial (KetECT). Int. J.
Neuropsychopharmacol. 25 (5), 339–349. https://doi.org/10.1093/ijnp/pyab088.
Espinoza, R.T., Kellner, C.H., 2022. Electroconvulsive therapy. N. Engl. J. Med. 386 (7),
667–672. https://doi.org/10.1056/NEJMra2034954.
Gillving, C., Ekman, C.J., Hammar, Å., Land´
en, M., Lundberg, J., Movahed Rad, P.,
Nordanskog, P., von Knorring, L., Nordenskj¨
old, A., 2024. Seizure duration and
electroconvulsive therapy in major depressive disorder. JAMA Netw. Open 7 (7),
e2422738. https://doi.org/10.1001/jamanetworkopen.2024.22738.
Guloksuz, S., Rutten, B.P.F., Arts, B., van Os, J., Kenis, G., 2014. The immune system and
electroconvulsive therapy for depression. J. ECT 30 (2), 132. https://doi.org/
10.1097/YCT.0000000000000127.
Jørgensen, A., Gronemann, F.H., Rozing, M.P., Jørgensen, M.B., Osler, M., 2024. Clinical
outcomes of continuation and maintenance electroconvulsive therapy. JAMA
Psychiatry. https://doi.org/10.1001/jamapsychiatry.2024.2360.
Jørgensen, M.B., Rozing, M.P., Kellner, C.H., Osler, M., 2020. Electroconvulsive therapy,
depression severity and mortality: Data from the Danish National Patient Registry.
J. Psychopharmacol. 34 (3), 273–279. https://doi.org/10.1177/
0269881119895518.
Kirov, G., Jauhar, S., Sienaert, P., Kellner, C.H., McLoughlin, D.M., 2021.
Electroconvulsive therapy for depression: 80 years of progress. Br. J. Psychiatry 219
(5), 594–597. https://doi.org/10.1192/bjp.2021.37.
Kruse, J.L., Congdon, E., Olmstead, R., Njau, S., Breen, E.C., Narr, K.L., Espinoza, R.,
Irwin, M.R., 2018. Inammation and Improvement of Depression Following
Electroconvulsive Therapy in Treatment-Resistant Depression. J. Clin. Psychiatry 79
(2), 9042. https://doi.org/10.4088/JCP.17m11597.
Lelubre, C., Anselin, S., Zouaoui Boudjeltia, K., Biston, P., Piagnerelli, M., 2013.
Interpretation of C-reactive protein concentrations in critically Ill patients. BioMed.
Res. Int. 2013 (1), 124021. https://doi.org/10.1155/2013/124021.
2024. Love, J., Dropmann, D., Selker, R., Gallucci, M., Jentschke, S., Balci, S., Seol, H., &
Agosti, M., 2024, The jamovi project (Version 2.5) [Computer software]. https://
www.jamovi.org/.
McIntyre, R.S., Alsuwaidan, M., Baune, B.T., Berk, M., Demyttenaere, K., Goldberg, J.F.,
Gorwood, P., Ho, R., Kasper, S., Kennedy, S.H., Ly-Uson, J., Mansur, R.B., McAllister-
Williams, R.H., Murrough, J.W., Nemeroff, C.B., Nierenberg, A.A., Rosenblat, J.D.,
Sanacora, G., Schatzberg, A.F., Maj, M., 2023. Treatment-resistant depression:
Denition, prevalence, detection, management, and investigational interventions.
World Psychiatry.: Off. J. World Psychiatr. Assoc. (WPA 22 (3), 394–412. https://
doi.org/10.1002/wps.21120.
Mindt, S., Neumaier, M., Hoyer, C., Sartorius, A., Kranaster, L., 2020. Cytokine-mediated
cellular immune activation in electroconvulsive therapy: A CSF study in patients
with treatment-resistant depression. World J. Biol. Psychiatry: Off. J. World Fed. Soc.
Biol. Psychiatry 21 (2), 139–147. https://doi.org/10.1080/
15622975.2019.1618494.
Moreno-Agostino, D., Wu, Y.-T., Daskalopoulou, C., Hasan, M.T., Huisman, M., Prina, M.,
2021. Global trends in the prevalence and incidence of depression:a systematic
review and meta-analysis. J. Affect. Disord. 281, 235–243. https://doi.org/10.1016/
j.jad.2020.12.035.
Musselman, D.L., Miller, A.H., Porter, M.R., Manatunga, A., Gao, F., Penna, S., Pearce, B.
D., Landry, J., Glover, S., McDaniel, J.S., Nemeroff, C.B., 2001. Higher than normal
plasma interleukin-6 concentrations in cancer patients with depression: preliminary
ndings. Am. J. Psychiatry 158 (8), 1252–1257. https://doi.org/10.1176/appi.
ajp.158.8.1252.
Orsolini, L., Pompili, S., Tempia Valenta, S., Salvi, V., Volpe, U., 2022. C-reactive protein
as a biomarker for major depressive disorder? Int. J. Mol. Sci. 23 (3), 1616. https://
doi.org/10.3390/ijms23031616.
Osimo, E.F., Cardinal, R.N., Jones, P.B., Khandaker, G.M., 2018. Prevalence and
correlates of low-grade systemic inammation in adult psychiatric inpatients: An
electronic health record-based study. Psychoneuroendocrinology 91, 226–234.
https://doi.org/10.1016/j.psyneuen.2018.02.031.
Pearson, T.A., Mensah, G.A., Alexander, R.W., Anderson, J.L., Cannon, R.O., Criqui, M.,
Fadl, Y.Y., Fortmann, S.P., Hong, Y., Myers, G.L., Rifai, N., Smith, S.C., Taubert, K.,
Tracy, R.P., Vinicor, F., 2003. Markers of inammation and cardiovascular disease.
Circulation 107 (3), 499–511. https://doi.org/10.1161/01.
CIR.0000052939.59093.45.
Pedraz-Petrozzi, B., Spangemacher, M., Deicher, A., Drews, L., Defert, J., Silva-
Colmenero, A.Y., Wein, P., Riedinger, E., Gründer, G., Gilles, M., Sartorius, A.,
Reinwald, J.R., 2024. Baseline monocyte count predicts symptom improvement
during intravenous ketamine therapy in treatment-resistant depression: A single-arm
open-label observational study. Front. Psychiatry 15. https://doi.org/10.3389/
fpsyt.2024.1415505.
Petrides, G., Fink, M., Husain, M.M., Knapp, R.G., Rush, A.J., Mueller, M., Rummans, T.
A., O’Connor, K.M., Rasmussen, K.G.J., Bernstein, H.J., Biggs, M., Bailine, S.H.,
Kellner, C.H., 2001. ECT remission rates in psychotic versus nonpsychotic depressed
patients: a report from CORE. J. ECT 17 (4), 244.
Plemper, J., Sartorius, A., Karl, S., 2023. Age-dependent dose increase during an acute
electroconvulsive therapy series. J. ECT 39 (3), 193–196. https://doi.org/10.1097/
YCT.0000000000000896.
Ryan, K.M., McLoughlin, D.M., 2022. Peripheral blood inammatory markers in
depression: Response to electroconvulsive therapy and relationship with cognitive
performance. Psychiatry Res. 315, 114725. https://doi.org/10.1016/j.
psychres.2022.114725.
Sackeim, H.A., Devanand, D.P., Prudic, J., 1991. Stimulus intensity, seizure threshold,
and seizure duration: Impact on the efcacy and safety of electroconvulsive therapy.
Psychiatr. Clin. North Am. 14 (4), 803–843.
Strawbridge, R., Hodsoll, J., Powell, T.R., Hotopf, M., Hatch, S.L., Breen, G., Cleare, A.J.,
2019. Inammatory proles of severe treatment-resistant depression. J. Affect.
Disord. 246, 42–51. https://doi.org/10.1016/j.jad.2018.12.037.
Strawbridge, R., Izurieta, E., Day, E., Tee, H., Young, K., Tong, C.C., Young, A.H.,
Cleare, A.J., 2023. Peripheral inammatory effects of different interventions for
treatment-resistant depression: A systematic review. Neurosci. Appl. 2, 101014.
https://doi.org/10.1016/j.nsa.2022.101014.
The UK ECT Review Group, 2003. Efcacy and safety of electroconvulsive therapy in
depressive disorders: A systematic review and meta-analysis. Lancet 361 (9360),
799–808. https://doi.org/10.1016/S0140-6736(03)12705-5.
Thirthalli, J., Sinha, P., Sreeraj, V.S., 2023. Clinical practice guidelines for the use of
electroconvulsive therapy. Indian J. Psychiatry 65 (2), 258. https://doi.org/
10.4103/indianjpsychiatry.indianjpsychiatry_491_22.
van Buel, E.M., Patas, K., Peters, M., Bosker, F.J., Eisel, U.L.M., Klein, H.C., 2015.
Immune and neurotrophin stimulation by electroconvulsive therapy: Is some
inammation needed after all? Transl. Psychiatry 5 (7), e609. https://doi.org/
10.1038/tp.2015.100.
Young, J.R., Evans, M.K., Hwang, J., Kritzer, M.D., Kellner, C.H., Weiner, R.D., 2023.
Electroconvulsive therapy changes immunological markers in patients with major
M. Spangemacher et al.
Biomarkers in Neuropsychiatry 12 (2025) 100115
6

depressive disorder: a scoping review. J. ECT. https://doi.org/10.1097/
YCT.0000000000001021.
Yrondi, A., Sporer, M., P´
eran, P., Schmitt, L., Arbus, C., Sauvaget, A., 2018.
Electroconvulsive therapy, depression, the immune system and inammation: A
systematic review. Brain Stimul. 11 (1), 29–51. https://doi.org/10.1016/j.
brs.2017.10.013.
Zilles-Wegner, D., Kellner, C.H., Sartorius, A., 2023. Electroconvulsive therapy and its
new competitors: ECT remains the gold standard. J. ECT. https://doi.org/10.1097/
YCT.0000000000001034.
M. Spangemacher et al.
Biomarkers in Neuropsychiatry 12 (2025) 100115
7