Conceived and designed the experiments: TB UL DR. Performed the experiments: TB UL DR PW SE. Analyzed the data: MI MU DR TB. Contributed reagents/materials/analysis tools: MU MI TB UL DR. Wrote the paper: TB.
Over the past five years TB has received honoraria for speaking from Lilly, AstraZeneca, Esparma, Bristol-Myers Squibb, Sanofi-Aventis, Servier, and Lundbeck, and has accepted reimbursements of travelling expenses to congresses from AstraZeneca and Lilly. Over the past five years SE has accepted reimbursements of travelling expenses to congresses from AstraZeneca. Over the past five years MI has received grant support from the German Federal Ministry of Education and Research, and has been consultant to MSD Merck, Whitehouse Station, New Jersey, United States of America. UL has received honoraria for speaking from AstraZeneca and has accepted reimbursements of travelling expenses to congresses from AstraZeneca and Lundbeck. The authors DR, PW and MU – except for income received from their primary employer – have received no financial support or compensation from any individual or corporate entity over the past five years for research or professional service. The authors declare for all authors that there are no personal financial holdings that could be perceived as constituting a potential conflict of interest. This does not alter the authors' adherence to all the PLoS ONE policies on sharing data and materials.
Distorted activity of the hypothalamic-pituitary-adrenocortical (HPA) system is one of the most robustly documented biological abnormalities in major depression. Lithium is central to the treatment of affective disorders, but little is known about its effects on the HPA system of depressed subjects.
To assess the effects of lithium monotherapy on the HPA system of patients with major depression by means of the combined DEX/CRH test.
Thirty drug-naive outpatients with major depression (single episode or unipolar recurrent; SCID I- and II-confirmed) were treated with lithium monotherapy for four weeks. The DEX/CRH test was conducted directly before intake of the first lithium tablet and four weeks thereafter. Weekly ratings with the HDRS21 were used to determine response (≥50% symptom reduction) and remission (HDRS ≤7).
Lithium levels within the therapeutic range were achieved rapidly. Tolerability was good; no patient terminated the treatment prematurely. Response and remission rates were 50% and 33% respectively. Compared to the DEX/CRH test before the start of the treatment, a considerable and significant increase in all CRH-stimulated ACTH and cortisol parameters could be detected in the second DEX/CRH test. When analysed with particular regard to responders and non-responders, that significant increase was only present in the responders.
We were able to demonstrate that lithium leads to a significant activation of the HPA system. This is possibly connected to stimulation of hypothalamic arginine vasoporessin (AVP), to direct intracellular effects of lithium on pituitary cells and to an induction of gene expression.
drks-nue.uniklinik-freiburg.de
Depression is a frequent and severe medical problem. One of the cornerstones of treatment is antidepressant medication
A system repeatedly studied as a target for new antidepressive drugs is the hypothalamic-pituitary-adrenocortical (HPA) system
Another fundamental drug used to treat affective disorders with an alternative mode of action is lithium. Lithium induces multiple
Not established, though demonstrated effective in several older studies, is lithium monotherapy for the treatment of acute depression. At least seven studies from the 1970s and 1980s
The impact of lithium on the HPA system has repeatedly been investigated mainly in pre-clinical studies. Older cell culture and animal studies
In a previous study, we investigated a group of unipolar depressed patients (N = 30) who did not respond to a treatment with an antidepressant and who were subsequently treated with lithium augmentation
Because it couldn't be concluded from the results of this preceding study whether the increase in HPA system activity was due to a direct pharmacological effect of the lithium ion or whether it was instead an effect of the complex lithium augmentation mechanism (i. e. pre-treatment with an antidepressant, followed by a lithium/antidepressant combination), we here present the results of a study with 30 unipolar, acutely depressed, drug-naive subjects who were treated with lithium monotherapy.
The protocol for this trial and supporting CONSORT checklist are available as supporting information; see
This study was conducted at the Department of Psychiatry, Technical University of Dresden between January 2003 and February 2004. Patients who contacted the outpatient clinic of the department on their own, were transferred to the outpatient clinic by other physicians, or who followed a call for study participation in the local newspapers could be included into the study. Except for one patient, who was transferred to the inpatient ward because of the severity of the symptoms, all participants were outpatients. The study protocol was approved by the local ethical committee and the study has been carried out in accordance with the Declaration of Helsinki. After complete description of the study to the subjects, written informed consent was obtained. The laboratory ACTH and cortisol measurements and the statistical analyses were performed at the Max-Planck Institute of Psychiatry, Munich.
Patients of both genders, aged 18 to 75 years, with a major depressive episode, single episode or recurrent (DSM IV criteria) were included into the study. Diagnoses were confirmed by the Structured Clinical Interview for DSM IV (SCID I; German version)
The exclusion criteria were (SCID I): a history of a manic or hypomanic episode (bipolar disorder), current alcohol abuse or dependency, organic brain disease, schizophrenia or schizoaffective disorder. Further exclusion criteria were: current intake of psychotropic drugs (see above); acute or chronic somatic conditions that might contraindicate lithium or that might influence the regulation of mineralo- or glucocorticoids; pregnancy or lactation. Personality disorders were not an exclusion criterion but were systematically recorded by conducting a SCID II interview with each subject.
Patients started lithium treatment on the evening of the first combined DEX/CRH test (baseline). Lithium was administered at a daily dose of two times 12.2 mEq lithium carbonate (2×450 mg). Standardized 12-hour serum lithium levels were measured at day 3, 7, 14, 21, and 28 of lithium treatment. The lithium dose was adapted individually to achieve a serum level within the established therapeutic range
The HDRS21, the Clinical Global Impressions Scale (CGI)
The combined DEX/CRH test was performed in accordance with the established research procedure
The plasma cortisol and ACTH concentrations in the first specimen (baseline), collected at 3 p.m., reflect the suppressive effects of the dexamethasone administered the day before, whereas the other five plasma cortisol and ACTH concentrations reflect the additional effects of the CRH stimulation. Peak levels, CRH response values (Delta) and area under the curve (AUC) values of the cortisol and ACTH in the six consecutive blood samples were used as indicators of the response to the combined DEX/CRH test. The Delta values were obtained by subtracting the 3 p.m. hormone values (before CRH injection) from the individual peak value. The AUC values were defined as the natural logarithms of the trapezoidal integration of the six cortisol and ACTH concentrations. Missing values (about 1%) were estimated by the individual values of the previous and subsequent sample.
All probes were analysed together in one assay at the end of the entire study. In order to determine the plasma cortisol concentration, a commercially available radioimmunoassay (RIA) kit (ICN Biomedicals, Carson, CA) was used. The detection limit was 0.3 ng/ml plasma; intra- and interassay coefficients of variation for 20 and 40 ng/ml were <7%. For plasma ACTH measurements, an immunoradiometric assay without extraction (Nichols Institute, San Juan Capistrano, CA) was used providing a detection limit of 4.0 pg/ml. The intra- and interassay coefficients of variation at 20 pg/ml plasma were <8%.
The study was designed according to the above mentioned previous results from our group
Changes between the initial clinical or endocrine variables at baseline and at follow-up were evaluated with Wilcoxon's matched pairs test. Differences between responders and non-responders were assessed with Fisher's exact test and the Mann-Whitney U test. Differences were regarded as significant when p<0.05. All analyses were performed with SPSS, Version 17.0. The sample size estimation was performed with G*Power 3.1
Forty patients were screened for participation and 30 were able to be included in the study (
Age (years) | 45.97±10.93 |
Gender (male:female) | 15∶15 |
Age at onset of mood disorder (years) | 36.83±10.24 |
Duration of index episode (months) | 17.03±16.26 |
Number of previous depressive episodes | 1.77±1.78 |
diagnosis: single episode (DSM-IV: 296.2) : recurrent episodes (DSM-IV: 296.3) | 9∶21 |
Lithium serum levels and lithium doses are presented in
lithium carbonate doses | Lithium serum level (mEq/L) | |||
mean ± SD | Range | mean ± SD | range | |
day 3, mEq/d | 24.4±0 | – | 0.46±0.11 | 0.26–0.69 |
mg/d | 900±0 | – | ||
day 7, mEq/d | 26.84±4.96 | 18.3–36.6 | 0.56±0.14 | 0.26–0.88 |
mg/d | 990±183 | 675–1350 | ||
day 14, mEq/d | 33.55±6.94 | 18.3–48.8 | 0.71±0.19 | 0.45–1.31 |
mg/d | 1238±256 | 675–1800 | ||
day 21, mEq/d | 35.99±8.68 | 18.3–48.8 | 0.81±0.18 |
0.51–1.38 |
mg/d | 1327±320 | 675–1800 | ||
day 28, mEq/d | 35.99±9.21 | 18.3–48.8 | 0.82±0.19 | 0.45–1.39 |
mg/d | 1327±340 | 675–1800 |
*the data from two patients are missing.
Lithium tolerability was good. Side effects were recorded weekly. The majority of patients actually showed signs of probably side-effects of lithium (70% of the patients at day 28), however such side-effects were generally mild. No patient terminated the lithium treatment prematurely. The side-effects observed were those that are typically expected under lithium treatment (in order of appearance: polyuria, ataxia, increased thirst, diarrhoea, and tremor).
According to the criteria mentioned above, fifteen of the 30 patients (8 women, 7 men) responded within 4 weeks after initiation of lithium treatment (in mean after 2.33±1.18 weeks). Ten of these 15 responders even showed remission. There was no significant difference between responders and non-responders with regard to age, gender, baseline severity of depression, lithium doses or lithium serum level (data not shown). The HDRS21, BDI and CGIseverity scores (mean ± SD) on day 0 were 21.7±4.70, 24.0±8.1, and 5.7±0.7 respectively, and declined to 11.1±5.1, 15.0±9.1, and 4.6±1.2 on day 28 (p<0.001 for all three changes; Wilcoxon's matched pairs test).
At day 0 (before start of lithium treatment), the ACTH peak was 25.37±27.19 pg/ml and the cortisol peak was 83.77±81.49 ng/ml (for baseline, AUC, and delta values, and cortisol/ACTH ratios cf.
pre-treatment (day 0) (N = 30) | after 4 weeks of lithium (day 28) (N = 29) | pWilcoxon's matched pairs test | ||
baseline (before CRH injection) | 16.40±7.60 | 16.25±6.47 | 0.991 | |
ACTH peak | 25.37±27.19 | 28.48±24.65 | 0.048 | |
ACTH AUC | 1829.3±1488.9 | 1978.59±1129.47 | 0.033 | |
ACTH delta | 8.97±26.20 | 12.23±19.74 | 0.008 | |
baseline (before CRH injection) | 17.28±13.40 | 28.34±35.99 | 0.002 | |
Cortisol peak (ng/ml) | 83.77±81.49 | 104.84±78.95 | 0.012 | |
Cortisol AUC | 4933.6±4640.2 | 5927.76±4323.2 | 0.013 | |
Cortisol delta | 66.48±73.51 | 76.5±52.46 | 0.045 |
With regard to response or non-response to the lithium treatment, differences could be detected between responders and non-responders. Before the start of the treatment (day 0), patients who responded in the following four weeks to the lithium medication displayed numerically lower values on all ACTH and cortisol parameters than subsequent non-responders. However, only the difference for baseline ACTH (before CRH injection) was significant (14.16±1.85 vs. 18.64±10.27 pg/ml, p = 0.049, Mann-Whitney U test, cf.
Pre-treatment (day 0) | after 4 weeks of lithium (day 28) | day 0 vs. day 28 | |||||||
Responders(n = 15) | Non-Responders(n = 15) | p(Responders vs. Non-Responders. Mann-Whitney U Test) | Responders(n = 14) | Non-Responders(n = 15) | p(Responders vs. Non-Responders. Mann-Whitney U Test) | Respondersp(Wilcoxon's matched pairs test) | Non-Respondersp(Wilcoxon's matched pairs test) | ||
baseline (before CRH injection) | 14.16±1.85 | 18.64±10.27 | 0.049 | 17.61±8.74 | 14.97±3.02 | 0.81 | 0.221 | 0.233 | |
ACTH peak | 19.56±3.62 | 31.18±38.02 | 0.95 | 35.32±34.34 | 22.10±5.41 | 0.57 | 0.008 | 0.733 | |
ACTH AUC | 1497.9±186.3 | 2160.8±2078.9 | 0.66 | 2308.4±1558.1 | 1670.8±294.7 | 0.66 | 0.013 | 0.609 | |
ACTH delta | 5.41±4.32 | 12.54±37.10 | 0.08 | 17.71±27.20 | 7.13±5.84 | 0.16 | 0.035 | 0.084 | |
baseline (before CRH injection) | 15.08±9.68 | 19.49±16.365 | 0.69 | 28.22±28.53 | 28.44±42.83 | 0.76 | 0.004 | 0.125 | |
Cortisol peak (ng/ml) | 73.39±43.05 | 94.14±108.03 | 0.55 | 114.95±67.25 | 95.39±89.82 | 0.22 | 0.026 | 0.112 | |
Cortisol AUC | 4231.1±2585.8 | 5394.8±6046.2 | 0.63 | 6564.2±3827.5 | 5333.8±4795.2 | 0.26 | 0.026 | 0.173 | |
Cortisol delta | 58.31±44.01 | 74.65±95.46 | 0.74 | 86.73±51.51 | 66.95±53.27 | 0.27 | 0.041 | 0.334 |
After four weeks of lithium treatment (day 28) a substantial rise in the ACTH and cortisol parameters (significant for all parameters except baseline ACTH) was observed only in responders to the lithium treatment, while the non-responders showed a (non-significant) decrease for most of the parameters of the DEX/CRH test (cf.
This is, to our knowledge, the first systematic study of lithium monotherapy in treating acute unipolar depression in more than 20 years. Lithium serum levels can be considered as adequate and well-tailored, reaching the therapeutic range after just one week, with a low degree of variance. The response (50%) and remission (33%) rates have to be called moderate, but match the efficacy typically seen in clinical studies on antidepressive compounds. In a recent analysis including 182 RCTs of antidepressants, the pooled response rate was 53.8%
After 4 weeks of lithium treatment the HPA system showed a clear activation compared to the first DEX/CRH test directly before the start of lithium (p<0.05 for all ACTH and cortisol parameters after CRH stimulation). This is in line with previous results from pre-clinical studies and our study on lithium augmentation. Previous studies in cell cultures and in animals
In an earlier study with the DEX/CRH test, our group found that lithium augmentation had a clear HPA-system-activating effect in unipolar depressed patients who did not respond to preceding monotherapy with an antidepressant
However, with the results of the presented study it becomes likely that it is a direct effect of lithium that activates the HPA system in such way that it becomes more responsive to CRH and ACTH and probably to external (stress) stimuli as well. A probable explanation of this finding is related to the osmotic effect of lithium salts. It is well known that lithium treatment affects renal tubular function and can lead to polyuria
The hypothetical functional connection between elevated AVP and hyper-responsiveness of the HPA axis is further supported by animal studies. Male Wistar rats selectively bred for high innate anxiety are characterized by elevated ACTH and corticosterone responses to an adaptation of the combined DEX/CRH test, and treatment with an antagonist against the vasopressin receptor 1 normalized the HPA axis hyperactivity in these animals
In addition, the activating effect of lithium on the HPA system might be mediated by direct intracellular effects of lithium on pituitary cells
Interestingly, when viewed separately with regard to responders and non-responders in the present study, the significant rise in the post-CRH ACTH and cortisol parameters under lithium treatment is only present in those who responded to the therapy. Most antidepressants, as tricyclic antidepressants
In addition, the influence of psychotropic drugs on the HPA system seems to be time-dependent: the inhibiting effect of some antidepressants on the ACTH and cortisol release seems to need a number of weeks of treatment, whereas the acute effects tend to result in stimulation of the hormone release, at least in healthy volunteers
The study presented here has methodological strengths and shortcomings. One key strong point is the fact that the study participants were all drug-naive. Although the initial inclusion criteria required a drug free period of only two weeks, we were able to include only subjects without pharmacological pre-treatment for the depressive index episode. This excludes the influence of any preceding pharmacotherapy on the clinical or endocrinological results. The drop-out rate was extremely low (no patients dropped out, and only one patient declined the second DEX/CRH test). Diagnoses and depressive symptoms were monitored and validated using established standardised clinical interviews and rating scales. With regard to the neuroendocrinological evaluation, however complex, the best-established and validated test – the combined DEX/CRH test – was used. Lithium therapy can be considered as optimal provided that lithium serum levels are very frequently monitored and that lithium levels are rapidly achieved within the therapeutic range (0.5 to 1.0 mEq/L) despite a good tolerability.
As the most important limitation, one could mention the lack of a control group. However, in keeping with most studies with the DEX/CRH test, our study concentrated on comparing endocrinological parameters before the start of the treatment (right before the intake of the first lithium tablet) and the parameters four weeks after the start of treatment. Since no other systematic changes took place except the commencement of the lithium therapy, it is rather unlikely that the changes found in the HPA system might be due to factors other than the lithium therapy and the possible therapeutic effects induced by it. However, it cannot be excluded that other aspects such as the passage of time (4 weeks) or placebo effects might have had an impact on the endocrinological results. The absence of a control group means that the clinical efficacy of the lithium monotherapy found in our study cannot reliably be interpreted. However, this was not the objective of the study.
In conclusion, this is the first study on lithium monotherapy for acute unipolar depression in more than 20 years, and the first prospective study of lithium monotherapy to ever use the DEX/CRH test. Given the results of the study presented here, of our preceding study with the DEX/CRH test on lithium augmentation, of previous studies of the HPA status in depressed or healthy subjects and the results of the pre-clinical studies in animals and cell cultures, it can now be regarded as having been established that lithium has a stimulating effect on the human HPA system. This is presumably mainly due to an activation of central AVP.
The pronounced effects found in our study confirm the fundamental significance of the HPA system to the pathophysiology and the treatment of depressive disorders. Developing fundamentally new treatment approaches continues to be a challenge, given the fact that some approaches directly targeting the HPA system have failed in recent years
(DOC)
(DOC)