ABSTRACT FROM: Dougherty DD, Rezai AR, Carpenter LL, et al. A randomized sham-controlled trial of deep brain stimulation of the ventral capsule/ventral striatum for chronic treatment-resistant depression. Biol Psychiatry 2014. Published Online First.
The evidence from multiple open-label trials documents the efficacy and safety of Deep brain stimulation (DBS) in the treatment of refractory depression (TRD).1 Nonetheless, randomised sham-control trials are required to determine whether the effect is directly related to active electrical stimulation or to the non-stimulation effects.
This was a multicentre randomised sham-controlled trial of DBS at the ventral capsule/ventral striatum (VC/VS). Thirty patients with chronic, non-psychotic, unipolar TRD were recruited from five centres that used comparable surgical targeting, in the USA. Patients received active or sham stimulation (1:1) in a double-blinded fashion for 16 weeks, followed by a 24-month open-label phase of active stimulation. During the blinded phase, high-frequency bilateral stimulation with two different fixed pulse width (90 and 210 µs) was used and medication regimens were kept stable (not listed in the paper). Bilateral stimulation was used to reduce the side effects and protect the blinding. The primary outcome measure was response (50% on Montgomery-Asberg Depression Rating Scale after 16 weeks).
No difference was found in efficacy between the active arm (3 responders of the 15 participants) and sham-control group (2 of the 14) at the end of the blinded phase.
The response rate of the active stimulation group (20% at 4 months) was much lower than that of another open trial (53.3% and 46.7% at 3 and 6 months, respectively).2
Only the active stimulation group experienced adverse events suggestive of dopaminergic overstimulation, such as hypomania (2 participants), mania (1 participant) and disinhibition (2 participants). Furthermore, cognitive inflexibility to response inhibition was significantly associated with the active group (results of this analysis were not reported). One suicide occurred in a non-responder who was taken off stimulation.
Response rate in the open-label phase at 12 and 24 months was only 20% and 23%, respectively—similar to the response rate of active stimulation (20%) during the sham-control phase. The lack of additional clinical benefit during the open phase, despite prolonged stimulation, changes in stimulation parameters and medications contradicts the previous reports of incremental increase in response rates of DBS over time.3
The stimulation parameter settings during the blinded phase were fixed (pulse width 90 or 210 µs), which might have limited the optimisation of stimulation, as some patients may improve with longer pulse width (210–450 µs) stimulation.4
There was no report on planned algorithm used to standardise the stimulation adjustments, including contact changes during the open phase across the centres.
Lack of standardised protocol in stimulation adjustments/contact changes may be partly responsible for poor results in the open phase of this multicentre study compared with a single-centre open-label study involving the same target.2
The duration of the blinded phase may not be long enough. The active-sham stimulation differences in improvement may be greater with longer blinded phase (6–12 months) as the initial clinical effects of the microlesion (due to electrode insertion) and placebo (expectation) may not be sustained long term.
Future trials should consider adaptive designs with a longer duration of the blinded phase. In adaptive design, initial or interim results of the study can be used to modify the study design without compromising the validity and integrity of the trial. This will allow modifications in optimal targeting and stimulation based on initial clinical responses during the controlled blinded phase. Additionally, having a crossover design within the blinded phase is statistically powerful for a small sample if order and carryover effects are balanced.
Given the heterogeneity in patients with TRD, the selection of should be based on clinical and imaging biomarkers such as anhedonia and deficits in reward circuits matching the functionality of VC/VS.
More research is needed to optimise strategies for DBS including closed loop stimulation. Contrary to hand-tuned stimulation adjustments, the closed loop system involves automated adjustments based on neural oscillations.
No. Given the disappointing results of both the blinded and open phase, clinical or off-label use of VC/VS stimulation is not recommended for TRD at present. Although the failure of this trial is a major setback to the future of DBS trials, there is growing consensus that DBS is beneficial at least for a subgroup of patients with TRD. Additionally, most of the responders to active stimulation seem to relapse during planned discontinuation or battery depletion suggesting a cause–effect relationship between active stimulation and outcome.2 ,3
Competing interests: None declared.
Provenance and peer review: Commissioned; internally peer reviewed.
Cook IA, Espinoza R, Leuchter AF, et al. Neuromodulation for depression: invasive and noninvasive (deep brain stimulation, transcranial magnetic stimulation, trigeminal nerve stimulation). Neurosurg Clin N Am 2014; 25:103–16. Malone DA, Dougherty DD, Rezai AR, et al. Deep brain stimulation of the ventral capsule/ventral striatum for treatment-resistant depression. Biol Psychiatry 2009; 65:267–75. Holtzheimer PE, Kelley ME, Gross RE, et al. Subcallosal cingulate deep brain stimulation for treatment-resistant unipolar and bipolar depression. Arch Gen Psychiatry 2012; 69:150–8. Ramasubbu R, Anderson S, Haffenden A, et al. Double-blind optimization of subcallosal cingulate deep brain stimulation for TRD: a pilot study. J Psychiatry Neurosci 2013; 38:325–32. 
