Association between baseline serum vascular endothelial growth factor levels and response to electroconvulsive therapy
Authors
A. Minelli, E. Maffioletti, M. Bortolomasi, A. Conca, R. Zanardini, L. Rillosi, M. Abate, M. Giacopuzzi, G. Maina, M. Gennarelli, L. Bocchio-Chiavetto
First published: 20 August 2013 Abstract
Objective
Several studies have shown that vascular endothelial growth factor (VEGF) is implicated in different neuronal processes involved in major depressive disorder (MDD) and in the mechanisms of action of antidepressants. The aim of this study was to investigate whether VEGF serum levels before treatment might be associated with the antidepressant response.
Method
Two groups of patients were enrolled. One was composed of 50 MDD patients receiving an antidepressant drug treatment. Illness severity was measured before the treatment (T0) and after 12 weeks (T1). The second group was composed of 67 treatment-resistant depressed (TRD) patients undergoing electroconvulsive therapy (ECT). Illness severity was assessed before the treatment (T0) and 1 month after the end of ECT (T1). Blood samples for VEGF measurements were collected for both groups at the baseline (T0).
Results
A significant correlation was observed between baseline VEGF serum levels and the percentage reduction in depressive symptomatology after ECT (P = 0.003). In particular, VEGF levels at baseline were significantly lower in patients showing no response to ECT at follow-up (P = 0.008). No correlation between T0 VEGF concentrations and drug treatment outcome was found.
Conclusion
Our results suggest that VEGF plays a role in the mechanism of response to ECT. Significant Outcomes
A significant correlation has been shown between VEGF baseline serum levels and response to ECT. The identification of predictive clinical and biological markers could be helpful for the personalization of ECT treatment.
Limitation
The patients who undergo ECT were in pharmacological treatment, and this could represent a possible confounder.
Introduction
The vascular endothelial growth factor (VEGF) was originally described as a vascular permeability factor and an angiogenic mitogen [1]. It was later found to be expressed in the brain and subsequently implicated in neuronal processes regulating growth, differentiation, survival, neuroprotection and regeneration [2-4]. VEGF has been recently investigated as a potential biomarker for major depressive disorder (MDD), as it represents a strategic mediator of neural processes involved in MDD pathogenesis, however, the results are contradictory [5-8]. In particular, two studies [5, 7] revealed increased concentrations of VEGF in MDD patients compared with controls, whereas another work [6] reported no difference. Furthermore, in healthy subjects, VEGF was negatively associated with depressive mood trait [8].
The literature indicates that VEGF may be a common downstream target of both pharmacological and non-pharmacological antidepressant treatments. Studies on animal models have shown that the increase in hippocampal VEGF expression is required for the neurogenic and behavioural actions of different antidepressant drugs, such as SSRIs and SNRI [9, 10]. Regarding non-pharmacological treatments, VEGF has been shown to be strongly upregulated in the hippocampus by electroconvulsive seizures (ECS) in animal models [10-12] and may represent an essential mediator of the neurogenic effect of ECS in the adult brain [10]. In our previous work, we provided evidence supporting the involvement of VEGF in the antidepressant mechanism of electroconvulsive therapy (ECT), which revealed a significant increase in serum VEGF in depressed patients after ECT. Furthermore, we found a significant correlation between this increase and the reduction in symptomatology [13]. In addition, a recent study showed that sleep deprivation treatment was associated with an increase in plasma VEGF levels in depressed patients [14].
To date, MDD treatment shows large interindividual differences in patients' clinical response: only about one-third of patients experiences full remission after the first treatment trial [15] and approximately 15% of MDD patients are classified as resistant or refractory (treatment-resistant depression, TRD) [16]. Consequently, the identification of measurable biological markers in vivo, along with non-invasive methods, could be of great value for differential diagnosis and for the optimization of patient treatments. In particular, greater relevance may come from the identification of predictive biomarkers, which are pretreatment measurements used to identify patients who are more likely to benefit from a specific therapy. Proteomic biomarkers that are predictive of treatment response in MDD remain in very early stages of development and, to date, few studies have been directed at identifying biomarkers related to different treatments and responses in depression [17]. Recent studies indicated preliminary data about a possible usefulness of the brain-derived neurotrophic factor (BDNF) as a predictive marker for antidepressant drug treatment outcome in MDD patients [18-20].
Aims of the study
The aim of our study was to investigate whether Vascular Endothelial Growth Factor serum levels might be associated with the treatment response to both antidepressant drugs and electroconvulsive therapy treatment.
Material and methods
One hundred and seventeen (DSM-IV) MDD patients were voluntarily enrolled in the study, which was approved by the local ethics committees, and written informed consent was obtained. Diagnosis of unipolar depression was confirmed using the SCID-I diagnostic structured interview. Exclusion criteria were as follows: (i) mental retardation and cognitive disorders; (ii) a lifetime history of schizophrenic, schizoaffective or bipolar disorder; (iii) personality disorders, obsessive–compulsive disorder, post-traumatic stress disorder, as primary diagnosis; (iv) comorbidity with eating disorders, substance abuse or dependency. A first group of 50 patients (Group 1: drug treated) were referred to the Psychiatric Rehabilitation Unit of IRCCS Centro S. Giovanni di Dio FBF, Brescia, to the Mood and Anxiety Disorders Unit of the University of Turin and to the Department of Psychiatry of the Central Hospital of Bolzano, Italy, to receive antidepressant treatment. Twenty-six patients were ‘drug naïve’ which had never received previous treatment with any antidepressant drug, while twenty-four were ‘drug free’, who had been previously treated with one or two antidepressants and had a washout period lasting at least 2 weeks. Thirty-seven patients were treated with SSRIs, and the other patients were treated with SNRIs or TCAs. Illness severity was assessed by the Montgomery and Asberg Depression Rating Scale (MADRS) before the start of the new antidepressant treatment (T0) and after 12 weeks of treatment (T1).
A second group of 67 subjects (Group 2: ECT treated) referred to the Psychiatric Hospital ‘Villa Santa Chiara’, Verona, Italy, and were scheduled to undergo ECT, because they had been evaluated as TRD patients. Treatment resistance to antidepressants was defined as at least the failure of the patient to respond to two or more adequate trials of two or more different classes of antidepressants and to an adequate trial of a tricyclic (TCA) drug referred to Stage III of Thase and Rush Staging Method [21]. Forty-six (69%) patients exhibited psychotic symptoms; furthermore, 31 (46%) showed comorbidity in Axis I and Axis II disorders as a secondary diagnosis. Patients were maintained on the same pharmacological treatment for at least 3 weeks before ECT treatment. Thirty-eight patients were treated with SSRIs in monotherapy or in combination with other agents; 28 patients were treated with other antidepressant drug classes, and 49 patients were treated with antipsychotic drugs. Illness severity and the outcome of ECT were assessed using the MADRS before the treatment (T0) and 1 month after the end of ECT (T1). In the month after the end of ECT, the pharmacological treatment was maintained the same, with only a possible light reduction in dosage. Any medications, in particular benzodiazepines, that reduce ECT efficacy are in general reduced or eliminated in the patients' daily treatment on the basis of the clinical judgment of the treating physicians. The mean number of ECT treatments received was 7.64 ± 2.57, and the treatment was completed on the basis of the clinical judgment of the treating physicians. ECT was performed with standard settings [22] with a bipolar brief pulse square wave and bilateral electrode placement [23], that is, two stimulus electrodes were placed over the left and right frontotemporal scalp. The ECT procedure has been described in detail elsewhere [13].
Serum VEGF determination
Venous blood samples were collected between 8:00 and 9:00 a.m. after an overnight fast in anticoagulant-free tubes. Samples were kept at room temperature for 1 h followed by 1 h at 4°C before serum separation by centrifugation (1620 rcf for 15 min at 4°C). Serum samples were stored at −80°C until the assay was performed. VEGF levels were measured in duplicate using an enzyme-linked immunosorbent assay (ELISA) method (Human VEGF Quantikine kit, R&D Systems, MN, USA) following the manufacturer's instructions. VEGF content was expressed as pg of protein/ml of serum, with a detection limit of 9 pg/ml.
Blood samples for VEGF measurements were collected at T0 for all the patients.
Statistical analysis
Demographic and clinical characteristics in our patient samples were described either in quantitative terms of mean ± standard deviation (SD) or as percentage proportions. Analysis of variance (anova) was used for computing possible differences, whereas the chi-square (χ2) test was used to evaluate categorical variables, along with clinical and biological changes. Pearson's coefficient was used to evaluate bivariate correlations. The identification of the VEGF cut-off score was made using receiver operating characteristic (ROC) curve analysis. The area under the curve (AUC) is a measure of the overall discriminative power. A value of 0.5 for the AUC represents the absence of discriminative power, whereas a value of 1.0 indicates perfect discrimination. The curve shows different probabilities (different decision cut-points of the logistic model) of a patient to achieve or not achieve response to ECT.
Moreover, we have computed the positive and negative predictive values (PPV and NPV), which respectively indicate the probability of achieving or not achieving the response given an above/under threshold VEGF level.
All analyses were conducted using SPSS Version 17.0 statistical software (SPSS Inc. Chicago, IL). Results
Considering the whole group, no significant correlation was detected between VEGF levels at the baseline (T0) and gender (P = 0.10), age (P = 0.13), BMI – body mass index (P = 0.41), smoking (P = 0.23) and T0
MADRS scores (P = 0.42). The group treated with antidepressant drugs differed from the one treated with ECT in the categories of age, education, BMI and illness severity at the baseline (for all features at the baseline of both groups see Table 1).
Baseline characteristics Drug treated(N = 50) ECT treated(N = 67) P-value
Age (years), mean (SD) 41.0 (13.7) 54.7 (15.0) <0.001
Gender (F), n (%) 35 (70.0) 47 (70.1) 0.99
Education (years), mean (SD) 11.7 (2.9) 8.4 (3.9) <0.001
BMI, mean (SD) 24.0 (3.2) 26.8 (5.0) <0.001 % Smokers 47.5 52.5 0.36 MADRS at T0, mean (SD) 24.9 (5.0) 33.7 (6.6) <0.001 VEGF at T0, mean (SD) pg/ml 360.27 (210.41) 377.72 (206.44) 0.65 % Responders 72.0 68.7 0.70 % psychotic symptoms 0% 69% <0.001 % of recurrent MDD 29% 93% <0.001
Table 1. Demographic, clinical and biological features at the baseline both for the depressed patients treated by antidepressant drugs and for the treatment-resistant depressed patients undergoing electroconvulsive therapy
Regarding Group 1 (drug treated), the antidepressant drug treatment reduced depression symptomatology as measured with MADRS (T0 = 24.94 ± 5.02; T1 = 10.00 ± 8.82; F1,49 = 146.50; P < 0.0001), and 72.0% of the patients were considered responders (patients were defined as responders if percentage MADRS reduction at T1 was >50%).
No significant correlation was found between baseline VEGF serum levels and the percentage MADRS score reduction (ΔMADRS%) (P = 0.25) and between baseline VEGF level and response (P = 0.24), demonstrating a lack of predictive power of VEGF levels at T0 for the drug treatment.
Regarding Group 2 (ECT treated), the ECT treatment reduced depression symptomatology as measured with MADRS (T0 = 33.69 ± 6.60; T1 = 12.72 ± 11.55; F1,66 = 197.50; P < 0.0001), and 68.7% of the patients were considered responders.
A significant correlation was found between baseline VEGF levels and ΔMADRS% (T0 VEGF 377.72 ± 206.44 pg/ml; ΔMADRS 61.91% ± 32.83%; r = 0.35; P = 0.003). The t-test indicated that VEGF serum levels at baseline were significantly lower in patients that showed no response to ECT treatment at follow-up (responders T0 VEGF 422.05 ± 200.25 pg/ml; non-responders T0 VEGF 280.63 ± 189.73 pg/ml; P = 0.008; Fig. 1).
Figure 1. Difference of the VEGF serum levels at baseline between patients that did or did not respond to ECT treatment by the time of follow-up. Error bars show mean ± 1.0 SD.
As we found a significant association between response and the presence of psychiatric comorbidity (P = 0.02), a logistic regression was used to adjust the result, and the effect of VEGF serum levels on the ECT outcome remained highly significant (P = 0.02). No significant association was found between presence of psychosis and ECT response (P = 0.74).
Receiver operating characteristic (ROC) curve for predicting response showed an AUC = 0.72. In our model, the best fitting decision rule (referred to as ‘prediction’) was a VEGF serum concentration of 191.36 pg/ml at the baseline, revealing a sensitivity of 93.5%, a specificity of 42.9%, a PPV of 78.2% and a NPV of 75.0%. Discussion
Our results have not shown any significant correlation between baseline VEGF serum levels and response to antidepressants, confirming our previous study regarding a lack of involvement of VEGF in antidepressant drug treatment in depressed patients [6]. In literature, only a recent study has evidenced an association trend between higher baseline VEGF plasma levels and poorer response to antidepressants [24], even though the sample size was extremely limited and consisted of patients with both MDD and bipolar disorder diagnosis.
In parallel, we have highlighted a significant correlation between baseline VEGF levels and response to ECT. This suggests that the VEGF content before treatment may provide a neurotrophic support in TRD patients undergoing ECT and therefore could represent one of the factors influencing the treatment response. These data go in the same direction of our previous study [13] showing a significant increase in the VEGF serum content following ECT and support a role of VEGF in the efficacy of this non-pharmacological antidepressant therapy.
Neurogenic and neurotrophic hypothesis of depression assumes that development of this disease is, at least partially, related to the reduced neurogenesis and/or depletion of neurotrophic factors. A low concentration of trophic factors, such as BDNF and/or VEGF, can contribute to the progression of depression [5, 25, 26]. A neurotrophic/neuroprotective role for VEGF has been widely demonstrated. Studies of cultured neurons subjected to hypoxia showed that endogenous VEGF exerts a neuroprotective effect, and in parallel, exogenous VEGF administered to rat brains can reduce ischaemic infarct and hypoxic neuronal death [27]. Moreover, the increase in hippocampal VEGF expression is required for the therapeutic action of antidepressants [9, 10], and hippocampal neurogenesis seems to be essential for the effectiveness of these drugs because its suppression has been shown to prevent their effects in behavioural animal models of depression [28].
We hypothesize that an anomalous VEGF regulation in TRD patients, and a consequent decrease in neurotrophic support, could represent one of the reasons for a lack of response to ECT.
Vascular Endothelial Growth Factor (VEGF) is also directly involved in the mechanism of action of ECT, as shown by studies in animal models, where it has been described to be strongly regulated in the hippocampus by electroconvulsive seizures (ECS) [11, 12, 29] VEGF signalling is necessary and sufficient to produce ECS induction of quiescent neural progenitor cell proliferation [29]. It is interesting to note that ECS can increase both neuronal and endothelial cell expression of VEGF in the brain [9].
Furthermore, another intriguing hypothesis regarding the role played by VEGF in antidepressant mechanisms can be formulated considering that it has been shown to downregulate the activity of multiple drug resistance (MDR) P-glycoprotein (a drug efflux transporter) at the brain–blood barrier (BBB) [30], resulting in an increased brain content of its substrates, which include antidepressant drugs [31]. As ECT may increase VEGF levels [10-12] and was mainly administered to subjects receiving antidepressant medication, it could augment the cerebral concentration of these drugs. In TRD patients, a low VEGF level could contribute to a low cerebral concentration of antidepressant drugs, which could not be sufficient to produce an adequate therapeutic response. However, it remains unclear whether serum concentrations reflect VEGF expression in the central nervous system; however, at this regard, a study reported a parallel alteration of VEGF serum and CSF (cerebrospinal fluid) levels in subjects with neurotuberculosis [32].
Our significant effect in TRD represents an important finding since approximately one-third of real-world MDD patients show an insufficient response to various drug treatments, and approximately 15% of MDD patients have significant depression despite aggressive pharmacologic and psychotherapeutic approaches, and for this reason, they are classified as TRD [16, 33]. In this scenario, despite controversial issues, ECT remains one of the most eligible therapies among TRD patients or intolerance to antidepressant medications or when a rapid and definitive response is required (e.g. because of psychosis or a risk of suicide) [23, 34]. Moreover, to date, meta-analyses have shown that ECT is more efficacious than simulated ECT and also more efficacious than antidepressant drugs [34]. With the aim to attain a reasonable balance between effectiveness and safety, the identification of predictive markers, both biological than clinical, associated with the response to the ECT could allow the clinicians to classify patients as good or poor candidates for ECT treatment and/or plan other options. Indeed, it might be possible to arrange in advance a plan of care that provides the integration with other non-pharmacological therapies, such as psychotherapy procedures [35, 36]. The available evidence suggests that the combined use of ECT and psychotherapy may confer additional, positive functional outcomes [36]. Moreover, the identification of TRD patients who could obtain a real benefit from ECT could be relevant for avoiding the immediate exclusion of ECT as a potential treatment. In fact, in many countries, the choice of this treatment is still largely governed by the principle of ‘ultima ratio’ [37], where patients are treated with ECT only after complex and controversial combinations of drugs or time is consuming for non-evidence-based psychotherapy, resulting in a high suffering of patients and their families.
The applicability of the VEGF serum measurement as a biomarker is limited as the ROC curve analyses indicated a high sensitivity but a low specificity and further studies in larger and independent samples are necessary for the confirmation of these first findings. On the other hand, the association of serum VEGF with ECT response remained significant after the correction for clinical predictors of poorer outcome as psychiatric comorbidity demonstrating the robustness of the results. At this regard, replications in larger samples may be helpful also to evaluate the added value of VEGF dosage in relation to other clinical predictors of response and other putative molecular markers.
A limitation of our study is the heterogeneity of antidepressant drugs administered to TRD patients undergoing ECT, which makes it difficult to define a possible augmentation effect of ECT and/or VEGF for specific drug classes, as the sample is too small for this subanalysis.
In conclusion, these results suggest that VEGF might play a role in the mechanism of ECT, and the dosage of serum VEGF might be helpful, together with other clinical and biological markers, for the prediction of response for the improvement of the remission rate and lower healthcare costs.
Acknowledgements
This research was supported by grants from the Italian Ministry of Health and Regione Lombardia (ID: 17387 SAL-13). We thank Federica Centin and Barbara Bertasi for laboratory support.
Declaration of interest
Minelli, E. Maffioletti, M. Bortolomasi, R. Zanardini, L. Rillosi, M. Abate, M. Giacopuzzi, G. Maina, M. Gennarelli and L. Bocchio-Chiavetto declare that they have no conflict of interest. A. Conca has served as a consultant for Lilly, BMS, Pfizer and on the speakers' bureau of Lilly, BMS, Astra Zeneca, Lundbeck, Italfarma and Janssen. He declares that he has no conflict of interest with this publication.
References
1. Leung DW, Cachianes G, Kuang WJ, Goeddel DV, Ferrara N. Vascular endothelial growth factor is a secreted angiogenic mitogen. Science 1989;246:1306–1309.
2. Greenberg DA, Jin K, Experiencing VEGF. Nat Genet 2004;36:792–793.
3. Schanzer A, Wachs FP, Wilhelm D et al. Direct stimulation of adult neural stem cells in vitro and neurogenesis in vivo by vascular endothelial growth factor. Brain Pathol 2004;14:237–248.
4. Storkebaum E, Lambrechts D, Carmeliet P. VEGF: once regarded as a specific angiogenic factor, now implicated in neuroprotection. BioEssays 2004;26:943–954.
5. Kahl KG, Bens S, Ziegler K et al. Angiogenic factors in patients with current major depressive disorder comorbid with borderline personality disorder. Psychoneuroendocrinology 2009;34:353–357.
6. Ventriglia M, Zanardini R, Pedrini L et al. VEGF serum levels in depressed patients during SSRI antidepressant treatment. Prog Neuropsychopharmacol Biol Psychiatry 2009;33:146–149.
7. Lee BH, Kim YK. Increased plasma VEGF levels in major depressive or manic episodes in patients with mood disorders. J Affect Disord 2012;136:181–184.
8. Katsuura S, Kamezaki Y, Yamagishi N et al. Circulating vascular endothelial growth factor is independently and negatively associated with trait anxiety and depressive mood in healthy Japanese university students. Int J Psychophysiol 2011;81:38–43.
9. Greene J, Banasr M, Lee B, Warner-Schmidt J, Duman RS. Vascular endothelial growth factor signaling is required for the behavioral actions of antidepressant treatment: pharmacological and cellular characterization. Neuropsychopharmacology 2009;34:2459–2468.
10. Fournier NM, Duman RS. Role of vascular endothelial growth factor in adult hippocampal neurogenesis: implications for the pathophysiology and treatment of depression. Behav Brain Res 2012;227:440–449.
11. Altar CA, Laeng P, Jurata LW et al. Electroconvulsive seizures regulate gene expression of distinct neurotrophic signaling pathways. J Neurosci 2004;24:2667–2677.
12. Newton SS, Collier EF, Hunsberger J et al. Gene profile of electroconvulsive seizures: induction of neurotrophic and angiogenic factors. J Neurosci 2003;23:10841–10851.
13. Minelli A, Zanardini R, Abate M, Bortolomasi M, Gennarelli M, Bocchio-Chiavetto L. Vascular Endothelial Growth Factor (VEGF) serum concentration during electroconvulsive therapy (ECT) in treatment resistant depressed patients. Prog Neuropsychopharmacol Biol Psychiatry 2011;35:1322– 1325.
14. Ibrahim L, Duncan W, Luckenbaugh DA, Yuan P, Machado-Vieira R, Zarate CA Jr. Rapid antidepressant changes with sleep deprivation in major depressive disorder are associated with changes in vascular endothelial growth factor (VEGF): a pilot study. Brain Res Bull 2011;86:129–133.
15. Gaynes BN, Rush AJ, Trivedi MH, Wisniewski SR, Spencer D, Fava M. The STAR*D study: treating depression in the real world. Cleve Clin J Med 2008;75:57–66.
16. Berlim MT, Turecki G. Definition, assessment, and staging of treatment-resistant refractory major depression: a review of current concepts and methods. Can J Psychiatry 2007;52:46–54.
17. Schmidt HD, Shelton RC, Duman RS. Functional biomarkers of depression: diagnosis, treatment, and pathophysiology. Neuropsychopharmacology 2011;36:2375–2394.
18. Dreimuller N, Schlicht KF, Wagner S et al. Early reactions of brain-derived neurotrophic factor in plasma (pBDNF) and outcome to acute antidepressant treatment in patients with Major Depression. Neuropharmacology 2012;62:264–269.
19. Rojas PS, Fritsch R, Rojas RA, Jara P, Fiedler JL. Serum brain-derived neurotrophic factor and glucocorticoid receptor levels in lymphocytes as markers of antidepressant response in major depressive patients: a pilot study. Psychiatry Res 2011;189:239–245.
20. Wolkowitz OM, Wolf J, Shelly W et al. Serum BDNF levels before treatment predict SSRI response in depression. Prog Neuropsychopharmacol Biol Psychiatry 2011;35:1623–1630.
21. Thase ME, Rush AJ. When at first you don't succeed: sequential strategies for antidepressant nonresponders. J Clin Psychiatry 1997;58(Suppl 13):23–29.
22. Abrams R, Swartz CM, Vedak C. Antidepressant effects of right versus left unilateral ECT and the lateralization theory of ECT action. Am J Psychiatry 1989;146:1190–1192.
23. Lisanby SH. Electroconvulsive therapy for depression. N Engl J Med 2007;357:1939–1945.
24. Halmai Z, Dome P, Dobos J et al. Peripheral vascular endothelial growth factor level is associated with antidepressant treatment response: results of a preliminary study. J Affect Disord 2013;144:269–273. 25. Warner-Schmidt JL, Duman RS. VEGF is an essential mediator of the neurogenic and behavioral actions
of antidepressants. Proc Natl Acad Sci USA 2007;104:4647–4652.
26. Bergstrom A, Jayatissa MN, Mork A, Wiborg O. Stress sensitivity and resilience in the chronic mild stress rat model of depression; an in situ hybridization study. Brain Res 2008;1196:41–52.
27. Sun FY, Guo X. Molecular and cellular mechanisms of neuroprotection by vascular endothelial growth factor. J Neurosci Res 2005;79:180–184.
28. Yan HC, Cao X, Gao TM, Zhu XH. Promoting adult hippocampal neurogenesis: a novel strategy for antidepressant drug screening. Curr Med Chem 2011;18:4359–4367.
29. Segi-Nishida E, Warner-Schmidt JL, Duman RS. Electroconvulsive seizure and VEGF increase the proliferation of neural stem-like cells in rat hippocampus. Proc Natl Acad Sci USA 2008;105:11352– 11357.
30. Hawkins BT, Sykes DB, Miller DS. Rapid, reversible modulation of blood-brain barrier P-glycoprotein transport activity by vascular endothelial growth factor. J Neurosci 2010;30:1417–1425.
31. O'brien FE, Dinan TG, Griffin BT, Cryan JF. Interactions between antidepressants and P-glycoprotein at the blood-brain barrier: clinical significance of in vitro and in vivo findings. Br J Pharmacol 2012;165:289–312.
32. Husain N, Awasthi S, Haris M, Gupta RK, Husain M. Vascular endothelial growth factor as a marker of disease activity in neurotuberculosis. J Infect 2008;56:114–119.
33. Schosser A, Serretti A, Souery D et al. European Group for the Study of Resistant Depression (GSRD)– where have we gone so far: review of clinical and genetic findings. Eur Neuropsychopharmacol 2012;22:453–468.
34. Ottosson JO, Odeberg H. Evidence-based electroconvulsive therapy. Acta Psychiatr Scand 2012;125:177–184.
35. Fenton L, Fasula M, Ostroff R, Sanacora G. Can cognitive behavioral therapy reduce relapse rates of depression after ECT? a preliminary study. J ECT 2006;22:196–198.
36. McClintock SM, Brandon AR, Husain MM, Jarrett RB. A systematic review of the combined use of electroconvulsive therapy and psychotherapy for depression. J ECT 2011;27:236–243.
37. Gazdag G, Takacs R, Ungvari GS, Sienaert P. The practice of consenting to electroconvulsive therapy in the European Union. J ECT 2012;28:4–6.
