22 FEBRUARY 2020 www.psychiatrictimes.com

Neuromodulation Approaches to Mood Disorders » Junao Wang, MD, driving and working. Depression re- and Alik S Widge, MD, PhD sponse rates in naturalistic studies SIGNIFICANCE FOR PRACTICING are good—30% to 60%.2 The effects appear durable, with one study show- Multiple neurostimulation are now available for treatment- ince the first antidepressant in ing 60% sustained response at 1 resistant depression, and pending technological advances will make these the 1950s, biological treat- year.3 When there is symptom re- S ments for mood disorders have treatments applicable to an even larger patient base. lapse, patients usually respond to an- largely focused on altering monoam- ◗ ECT remains the most effective treatment for depression, with response rates of other TMS series. Theta burst stimu- inergic neurotransmission, with only 70% to 80%. lation (TBS), a modified TMS that partial success. In the landmark ◗ TMS is a strong option for patients who cannot tolerate or access ECT, with takes 10 minutes or less per treat- STAR*D trial, 1 out of 3 patients did response rates up to 60% and sustained response to a year or more. ment, may be as effective as longer not achieve remission with a 4-tiered sessions, potentially allowing treat- ◗ Implantable therapies, including VNS and DBS, are rapidly gaining ground and have monoaminergic treatment algorithm. ment of many more patients with a a powerful role to play in relapse prevention. Neuromodulation therapies take an single machine. alternate approach: directly altering Although rare, seizure is the prima- the brain’s electrical activity through depression. Even with treatment resist- beneficial. A challenge, however, is ry risk of TMS. The risk is higher with electro-magnetic stimulation. ance, response rates can exceed 70% to access. An acute ECT series is psy- past seizures, brain injury, intracranial 80%. For those who respond but even- chosocially disruptive, since patients , and substances/medications Selecting an intervention tually have symptom relapse, mainte- usually are unable to work or to drive that reduce seizure threshold. TMS Neuromodulation can be divided into nance ECT (weekly to monthly) is on treatment days. find may still be usable in patients with in- invasive and non-invasive methods possible. ECT is safe and effective in ECT difficult to justify financially, as creased risk, but with careful monitor- (Table). Both are mainly indicated for bipolar, psychotic, and peripartum de- it requires an anesthesiologist, addi- ing and dose titration. For example, we depressive symptoms, generally as an pression. In cases of depression with tional support staff, and clinical routinely adjust treatment energy based adjunct to pharmacotherapy and psy- severe suicidal ideation, , or space, all of which may be limited. on changes in brain excitability (motor chotherapy and usually in patients life-threatening catatonia, ECT should Focal electrically administered threshold), which correlates with sei- with treatment resistance. Treatment strongly be considered even in the ab- seizure (FEAST) is a novel zure threshold. Regular dose adjust- resistance means lack of response to 2 sence of multiple medication trials. form of ECT that steers current away ment also reduces headache, the most to 4 medications with adequate dose While there are no absolute con- from the hippocampus, potentially common adverse effect of TMS. Com- and duration. Usually, resistance in- traindications to ECT, increased in- reducing cognitive adverse effects. pared with ECT, TMS works more cludes lack of response to multiple tracranial pressure is a relative con- Pilot trials have been positive, and slowly, with response usually at 4 to 6 medication classes and evi- traindication as treatment may randomized trials should start in the weeks into treatment. dence-based . That exacerbate the condition. Cardiovas- next 1 to 2 years.1 TMS is less effective than ECT, said, neuromodulation is not a pana- cular disease also increases risk of Transcranial magnetic stimulation but its safety and office-based nature cea. A careful examination of a pa- treatment as the induced seizure caus- (TMS). An office-based treatment for promote continuous, rapid innova- tients’ psychosocial structure is neces- es a surge in blood pressure and heart unipolar depression, TMS also is often tion. Controlling the context (and sary, as their mood is unlikely to show rate. Patients’ primary concern, how- used off-label for bipolar depression. It thus the brain state) under which lasting change if severe, persistent ever, is the possibility of autobio- is given daily (Monday through Fri- TMS is given may improve out- stressors are present. Similarly, co- graphical memory loss. This is usual- day) for 4 to 8 weeks with treatments comes. For instance, TMS could be morbid personality disorder may con- ly limited to memories from the usually lasting 20 to 30 minutes. delivered with cognitive training or tribute to depressive symptoms or al- treatment day, but patients (and some TMS uses a magnetic coil to in- psychotherapy that engages the cir- ter the response to neuromodulation. clinicians) still believe there is a risk duce electric currents in the prefron- cuit being stimulated, promoting of long-lasting cognitive impairment. tal cortex (PFC). The size and shape beneficial brain plasticity. Treatment Noninvasive Large studies and meta-analyses do of coils vary, and multiple parame- parameters might be tailored to a pa- neuromodulation not support these concerns. Moreover, ters can be adjusted, including fre- tient’s brain activity, ie, aspects of Electroconvulsive therapy (ECT). cognitive risk is greatly reduced with quency (eg, excitatory: 10 Hz; inhib- ongoing prefrontal rhythms. Re- This -based treatment uses an modern techniques such as ultrabrief itory” 1 Hz), treatment duration, and sponse might be accelerated by deliv- electrical current to induce a tonic- right unilateral ECT, which are just as target brain region. Some coils allow ering multiple TMS sessions within a clonic seizure under general anesthe- effective as older approaches. for deeper stimulation of cortical tis- single day.4 All of these require sub- sia. An acute series consists of 3 This highly effective treatment is sue (ie, deep TMS), but current liter- stantial further study and develop- weekly treatments for up to 4 weeks, often seen as a last resort, wasting ature does not show deep TMS to be ment but may improve outcomes. with each session requiring at least months to years of patients’ lives on superior to regular TMS when adjust- Finally, while there is minimal litera- an hour from pre-op to recovery. ineffective treatments. Discussing ing for placebo. ture for the use of TMS in pregnancy, ECT remains one of our most rapid and delivering ECT earlier in the TMS does not cause cognitive im- a small clinical trial recently showed and effective treatments for unipolar treatment algorithm would likely be pairment, so patients can continue safety and efficacy, an exciting pros-

PSY0220_019-022-26_widge-mouch-constantino-knoll.indd 22 1/28/20 4:22 PM FEBRUARY 2020 PSYCHIATRIC TIMES 23 Neuropsychiatry www.psychiatrictimes.com

pect for an understudied population.5 However, study design had a short sult in dysphagia or vocal cord paral- ed in clinical trials, all with similar comparison period. ysis. The risk of these complications effects, which suggests that they may Invasive neuromodulation FDA approval was based on data is less than 1% with a well-trained all be nodes in a common “mood reg- (VNS). An from a one-year open-label follow-up surgeon. The stimulation frequently ulation” network.8 DBS has only been implanted device that repeatedly stim- that showed increasing response switches on and off, and many pa- tested in patients who no longer re- ulates the left cervical vagus nerve, rates with time. A 5-year open-label tients note throat sensations or voice spond to any other therapy. VNS was FDA-approved in 2005 for study of 795 patients replicated this changes, both of which remit over Initially, DBS patients report “dis- adjunctive treatment of severe unipo- result, with significantly higher VNS time. appearance of the void,” reduced anx- lar and bipolar depression. VNS lever- response rates compared with treat- The primary limitations of VNS iety, increased connectedness, and ages the vagus nerve’s afferent projec- ment as usual (67.6% vs 40.9%).7 are logistical. Because the effects euphoria, subjective changes that tions to brainstem nuclei. Moreover, it This study has brightened VNS’ take months to be noticeable, patients usually fade within minutes to hours. may augment norepinephrine and prospects—Medicare will soon cov- may prefer a more rapid-acting treat- Continued stimulation over weeks to dopamine signaling throughout the er VNS as part of a large observation- ment. Most payors still do not cover months leads to more gradual im- brain through those projections, in al study, and commercial insurers are VNS, limiting patients’ access. provement. In early open label stud- turn enhancing the effects of mono- slowly following. A large multisite, randomized, ies, response rates were 40% to 66% aminergic medication. The implant VNS offers long-term benefit double-blind controlled study is in severely treatment-refractory cas- procedure takes approximately 2 even for treatment-resistant depres- about to start, which should clarify es.9,10 The evidence from randomized, hours and can be completed under lo- sion. A subanalysis in the 5-year the efficacy of VNS. Non-invasive well-controlled studies is less clear. cal or general . Patients re- study showed significantly higher VNS devices may greatly expand its Two US industry-funded trials, where turn for office-based stimulator pro- response rate to VNS compared with reach while reducing costs. A trans- all patients received implants but only gramming by a trained clinician. treatment as usual (59.6% vs 34.1%). cutaneous stimulation device (tVNS) some of the implants were activated, Programming is typically repeated The response rate was even higher in was FDA-cleared in 2018 for treat- failed to show a separation between biweekly to monthly over the first 6 patients who previously responded to ment of migraines and cluster head- DBS and sham stimulation.11,12 Euro- months of treatment. VNS is usually ECT (71.3% vs 56.9%). In addition, aches. Similar systems may be very pean trials that used a different de- reserved for patients who have not re- patients who remitted remained in re- useful for mood disorders. sign, however, showed a separation.13 sponded to medications or to noninva- mission longer with VNS compared Deep Brain Stimulation (DBS). An DBS may require long durations to sive forms of neuromodulation. to treatment as usual (40 months ver- experimental treatment for unipolar demonstrate its effect. When it is ef- Although FDA-approved, VNS is sus 19 months), although this did not and bipolar depression, DBS involves fective, the benefits appear to last—a rarely covered by commercial insur- reach statistical significance. MRI-assisted electrode implantation recent study reported over 50% re- ance because of its limited evidence As with any surgical procedure, directly into brain regions associated sponse rate at 8 years post-DBS.14 base. The published large rand- VNS carries a risk for pain and infec- with mood. The procedure takes 3 to 6 Risks include intracranial bleed- omized, controlled trial failed to tion. The vagus nerve innervates hours and is done under a combina- ing, stroke, seizure, and , but show separation between active and muscles involved in swallowing and tion of local and general anesthesia. all are very rare (less than 1%). Ad- sham stimulation after 10 weeks.6 speech, and surgical damage can re- Multiple brain targets have been test- verse psychological effects are more

Table. Choosing a neuromodulation intervention Noninvasive Intervention Pros Cons Transcranial Magnetic Stimulation (TMS) • Pairs well with behavioral activation because • Time-intensive, requires 4-8 weeks of daily FDA-approved office-based treatments it provides daily structure for patient treatment using electromagnetic conduction to • Patients can continue daily routine (work, • Small risk for seizures stimulate areas of the brain, typically the driving) dorsolateral prefrontal cortex • Minimal adverse effects; good option in severe depression without psychosis or acute suicidality if patients worry about effects of ECT

Electroconvulsive Therapy (ECT) • Proven safe and effective in unipolar, bipolar, • Acute cognitive effects limits ability to work FDA-approved treatment given under psychotic, and peripartum depression and drive general anesthesia using a short burst of • Rapid improvement over the span of 2-4 • Risk for memory impairment electrical stimulation to induce a weeks generalized tonic clonic seizure Invasive Intervention Pros Cons Vagus Nerve Stimulation (VNS) • May be effective in a highly treatment- • Surgical risks as well as risks for dysphagia FDA-approved treatment using a resistant depression population and dysphonia surgically implanted stimulator to deliver • Good option in patients who respond to ECT • May require long duration of treatments repetitive stimulation to the left cervical but quickly relapse (also effective in ECT before response is seen vagus nerve nonresponse) • Not yet covered by most insurance

Deep Brain Stimulation (DBS) • Some evidence for benefit in severe • May require long duration of treatment Experimental treatment with growing body treatment-resistant depression before response is seen of evidence using a surgically implanted • Durable effect for responders • Expensive, invasive, and only available at stimulator in brain regions thought to be very specialized centers part of a mood-regulation network

Other emerging neuromodulation interventions under study: transcranial direct current stimulation (tDCS), transcranial alternating current stimulation (tACS), (MST)

PSY0220_019-022-26_widge-mouch-constantino-knoll.indd 23 1/28/20 4:22 PM 24 PSYCHIATRIC TIMES www.psychiatrictimes.com &

common—up to 50% of patients grows and patients continue to de- may experience worsened mood, mand better options, this barrier will Substance anxiety, or hypomania while their fall, and neuromodulation will oc- clinicians search for optimal cupy a larger role in the psychiatric stimulation settings. Further- toolbox. more, patients often see DBS as a Disorders last resort. If response is delayed Dr Wang is Resident, and or difficult to achieve, they can Dr Widge is Assistant Professor, FEBRUARY 2020 www.psychiatrictimes.com lose hope. This is life-threaten- Department of Psychiatry and ing, as evidenced by a suicide Behavioral Sciences, University of rate close to 10% in DBS clinical Minnesota, Minneapolis, MN. Dr MALPRACTICE GRAND ROUNDS trials. Wang reports no conflicts of inter- The largest challenge of DBS is est concerning the subject matter the wide range of stimulation pa- of this article; Dr Widge reports rameters that can be applied, with patent applications in the area of A Rock and no clear guidance on how to search deep brain stimulation that are for the best settings. not licensed to any entity. research seeks to better define the target mood circuit, and to use ad- a Hard Place REFERENCES vanced electric field modeling to 1. Sahlem GL, Short EB, Kerns S, et al. Expand- optimally activate the desired neu- ed safety and efficacy data for a new method of ral pathways. Closed loop ap- performing electroconvulsive therapy. J ECT. 2016;32:197-203. » Eduardo Constantino, MD proaches seek to sense brain activ- 2. Taylor SF, Bhati MT, Dubin MJ, et al. A natural- cessfully detoxified and treated. His BD ity in real time and allow the device istic, multi-site study of repetitive transcranial symptoms were well controlled on his to self-adjust until a desired activity magnetic stimulation therapy for depression. J have worked in university health cen- psychiatric medication regimen, and he pattern is achieved. Either strategy Affect Disord. 2017 15;208:284-290. ters for many years and treated psychi- presented to our after discharge 3. Dunner DL, Aaronson ST, Sackeim HA, et al. A may lead to higher and more dura- atric disorders in adult patients, often for his initial evaluation with no evident multisite, naturalistic, observational study of I ble clinical response rates. transcranial magnetic stimulation for patients with concomitant substance use disor- mood symptoms. with pharmacoresistant major depressive disor- ders (SUDs). Despite a background in At that time, Lenny was also man- Emerging der: durability of benefit over a 1-year follow-up addiction psychiatry, I have found that dated to a substance intensive outpa- period. J Clin Psychiatry. 2014;75:1394-1401. neuromodulation 4. Duprat R, Desmyter S, Rudi DR, et al. Accel- the current opioid crisis has led to diffi- tient program (IOP) where he was sub- While ECT, TMS, VNS, and DBS erated intermittent theta burst stimulation cult ethical and legal scenarios that are ject to random urine toxicology. Lenny represent some of our best-studied treatment in medication-resistant major de- different from past epidemics. The risk refused our offer for naltrexone (wheth- tools for mood disorders, the land- pression: a fast road to remission? J Affect Dis- of sudden death from overdose with opi- er PO or by long-acting injection) citing ord. 2016;200:6-14. scape of neuromodulation is rap- 5. Kim DR, Wang E, McGeehan B, et al. Rand- oids (intentional or not) has added a new his relative stability. However, three idly changing, and several emerg- omized controlled trial of transcranial magnetic level of complexity and risk to my day- months later, he had another serious ing interventions are actively stimulation in pregnant women with major de- to-day work. Particularly troubling is overdose of IV heroin use that resulted being studied. These encompass a pressive disorder. Brain Stim. 2019;12:96-102. that when I successfully help patients in a brief ICU admission. The admitting 6. Rush AJ, Marangell LB, Sackeim HA, et al. variety of techniques including Vagus nerve stimulation for treatment-resistant stay off opioids for an extended period, hospital contacted us for aftercare but direct electrical stimulation at depression: a randomized, controlled acute I am effectively increasing their risk of did not contact the patient’s substance sub-threshold energy levels with phase trial. Biol Psychiatry. 2005;58:347-354. dying if they relapse. This increased risk IOP or his probation officer. The hospital transcranial direct current stimula- 7. Aaronson ST, Sears P, Ruvuna F, et al. A 5-year is due both to their loss of tolerance to staff assumed that the patient’s proba- observational study of patients with treat- tion officer was aware of this new viola- tion (tDCS) or transcranial alter- ment-resistant depression treated with vagus narcotic medications and to recent nating current stimulation (tACS) nerve stimulation or treatment as usual: com- changes in opioid preparations that in- tion of his probation. as well as other forms of seizure parison of response, remission, and suicidality. clude the potential mix of high potency When Lenny presented to our clinic therapy such as magnetic seizure Am J Psychiatry. 2017;174:640-648. agents (eg, fentanyl and carfentanil). one week after this overdose, he was 8. Widge AS, Malone DAJ, Dougherty DD. Closing calm and psychiatrically stable, report- therapy (MST). Further evidence the loop on deep brain stimulation for treatment-re- is needed to support their clinical sistant depression. Front Neurosci. 2018;12:175. ing full sobriety since he had left the hos- use at this time. 9. Mayberg HS, Lozano AM, Voon V, et al. Deep CASE VIGNETTE pital. He refused our recommendation to brain stimulation for treatment-resistant de- start monthly long-acting injectable nal- pression. Neuron. 2005;45:651-660. “Lenny” was a divorced unemployed trexone, but he agreed to take PO nal- Accessing 10. Malone DA, Dougherty DD, Rezai AR, et al. neuromodulation Deep brain stimulation of the ventral capsule/ salesman in his 30s with bipolar disorder trexone daily. He also told us that he was While neuromodulation services ventral striatum for treatment-resistant depres- (BD). He presented to our outpatient de- attending his court-mandated substance are limited in community practice, sion. Biol Psychiatry. 2009;65:267-275. partment after completing a 14-day IOP for use three times per week. most academic institutions now em- 11. Holtzheimer PE, Husain MM, Lisanby SH, et court-mandated inpatient rehabilitation However, he rescinded his consent al. Subcallosal cingulate deep brain stimulation ploy at least one neuromodulation for treatment-resistant depression: a multisite, program. He had a long history of daily for us to speak to that program and also specialist. Referral to such special- randomised, sham-controlled trial. Lancet Psy- alcohol and IV heroin use, and intermittent asked us not to speak to his probation ists is often the first step in helping chiatry. 2017;4:839-849. use of intranasal cocaine and oral (PO) officer about his recent heroin relapse patients access these next-genera- 12. Dougherty DD, Rezai AR, Carpenter LL, et al. amphetamines. He had undergone mul- and overdose. His reasons for this A randomized sham-controlled trial of deep tion treatments. A critical aspect of brain stimulation of the ventral capsule/ventral tiple inpatient and outpatient treatments change were simple and clearly stated: referral is a detailed treatment histo- striatum for chronic treatment-resistant de- but no extended periods of sobriety. he needed only five more months to ry, especially medications tried and pression. Biol Psychiatry. 2015;78:240248. Before his recent admission, Lenny “graduate” from his outpatient sub- reasons for nonresponse. That histo- 13. Bergfeld IO, Mantione M, Hoogendorn et al. had been hospitalized medically for re- stance program and, shortly thereafter, Deep brain stimulation of the ventral anterior ry often influences treatment selec- limb of the internal capsule for treatment-re- spiratory depression after an inadver- his probation period would be over. He tion and is critical for obtaining pay- sistant depression: a randomized clinical trial. tent heroin overdose that required four told us that if we notified either party or approval. Lack of reimbursement JAMA Psychiatry. 2016;73:456-464. doses of naloxone. This episode was (who were both unaware of his most remains the single largest barrier to 14. Crowell AL, Riva-Posse P, Holtzheimer PE, et al. considered a violation of his probation, recent overdose), his probation would Long-term outcomes of subcallosal cingulate offering patients these circuit-based deep brain stimulation for treatment-resistant de- and he was mandated to the inpatient be extended and he would likely face treatments. As the evidence base pression. Am J Psychiatry. 2019;176:949-952. ❒ substance program where he was suc- jail time.

PSY0220_019-022-26_widge-mouch-constantino-knoll.indd 24 1/28/20 4:22 PM