We propose to determine if augmentation of electroconvulsive therapy (ECT) utilized for the treatment of major depressive disorder (MDD) with daily oral creatine will lead to an accelerated response to treatment, an overall increase in response rate, and will protect against cognitive adverse effects associated with ECT. We propose to conduct a two-arm, parallel, randomized, double-blinded, placebo-controlled trial, with a treatment group receiving 20 g oral loading dose of creatine for 1 week starting the day before initiating ECT, followed by 5 g oral creatine daily for roughly five weeks, including the approximately three-week ECT treatment course and a two-week follow-up period. Response to treatment will be assessed using the Quick Inventory of Depressive Symptomatology (QIDS) at each treatment and the 17-item Hamilton Depression Rating Scale (HAM-D17) at the end of each week.

A. Prevalence and Impact of Depression

MDD has a lifetime prevalence of over 16% and is associated with significant personal and social costs, including lost work productivity, disability, diminished quality of life, increased mortality, increased rates of suicide attempts and completed suicides. The financial impact of depression in the United States is significant, with an estimated economic burden of individuals with MDD of approximately $210.5 billion dollars annually, including direct costs, suicide-related costs, and workplace costs. Although MDD is often regarded as a single disorder, it may encompass a variety of different etiologies with overlapping symptoms and signs. An additional complication of MDD is the high risk of disease recurrence; the presence of two or more chronic medical conditions, female gender, never having been married, activity limitation, and less contact with family are all significant predictors of MDD persistence.

MDD can be challenging to treat clinically, especially due to its propensity to be treatment-resistant. Treatment-resistant Depression (TRD) is currently defined as failure to achieve remission after two or more adequate pharmacologic trials. Current literature suggests that the majority of individuals with MDD do not reach and subsequently maintain a fully remitted state. Furthermore, results from the STAR-D trial indicate that the overall likelihood of failure to achieve remission is increased with increasing number of failed medication trials. At this time, the adjunctive use of atypical antipsychotics has been well-investigated, but less is known about alternative adjunctive agents.

Current clinical guidelines for the treatment of depression establish basic principles for establishing a treatment plan, preparing for potential need for long-term treatment, and assessment of remission. For moderate major depression, first line treatment includes antidepressant monotherapy and psychotherapy. For severe major depression, antidepressant therapy can be augmented with an antipsychotic or ECT. Due to the overall disease burden of unmanaged MDD, the prevalence of treatment resistant depression, high rates of primary treatment failure, the additional study of alterative adjunctive therapy is both appropriate and potentially impactful.

B. Efficacy of ECT in Treatment of MDD

ECT is considered to be a first line treatment for depression with psychotic features, but it is also often used to treat patients with treatment-resistant depression (TRD). According to a 2015 metanalysis, approximately one third of patients with MDD do not respond to ECT, with failed medication trials and longer depressive episodes being the strongest predictors of poor response. For treatment-resistant depression, the overall response rate is approximately 58%, compared to a 70% response rate in patients without TRD. When assessing the overall efficacy of ECT, both remission and response are used, although remission is more frequently used in clinical practice. In general, response has been defined as a 50% decrease in baseline depression screening scores, while remission is defined as a score <7 on the HAM-D17, or <10 with the 24-item HAM-D.

Assessments of remission in MDD after ECT treatment suggest that chronic depression, medication resistance, longer episode duration, and younger age are all statistically significant predictors of non-remission. Remission rates in ECT tend to be robust; data from the CORE trial suggested an overall remission rate of 87%, which further delineates into 95% remission rate with the presence of psychotic features and 83% without. Again, however, rates of response and remission with ECT are lower for patients with treatment-resistant depression, who may make up the majority of patients receiving ECT in clinical samples.

Although ECT is an effective treatment for MDD, up to one third of patients experience significant memory loss and other adverse cognitive effects after receiving ECT. Strategies to limit ECT's effects on memory, such as altering electrode placement (e.g., from bitemporal to bifrontal), ensuring days off between treatments, and modulating pulse width, amplitude, and frequency, may all have some benefit. Still, cognitive complaints remain one of the most significant side-effects of ECT, and concerns about these effects represent a major reason that patients who could benefit from ECT choose not to pursue it. Nootropic agents, thyroid hormone, and donepezil have all been studied to mitigate the cognitive side effects, but no consistent benefit has been shown and no current adjunctive medication is recommended.

C. ECT Augmentation

In order to maximize the efficacy of ECT in the treatment of MDD, several studies have analyzed the benefit of augmentation with a variety of antidepressants, anesthetic agents, and nutritional supplements. The most thoroughly studied has been adjunctive ketamine administration with ECT, although findings have been inconsistent. A metanalysis of RCTs investigating adjunctive ketamine and ECT in 2019 did not find that ketamine improves the efficacy of ECT when compared to other anesthetic agents, although it was suggested that ketamine could lead to improvement of depressive symptoms in the acute phases of ECT when used in combination. There was no improvement in depressive symptoms with ketamine augmentation by the end of the ECT series. The effect of ketamine on the neurocognitive side effects associated with ECT remains unclear, but no clear benefit has been shown. In addition, the long-term efficacy and safety of ketamine use in ECT is unknown, particularly in the setting of maintenance treatment.

Many secondary agents have been studied in conjunction with ECT therapy, including caffeine sodium benzoate (CSB), hyperventilation, and methylxanthines. However, these agents have primarily been studied for their potential to lower seizure threshold or increase overall seizure duration during ECT, without clear effects on overall ECT efficacy or symptom improvement apart from effects on seizure characteristics. Less is understood about the effects of nutritional supplementation, such as folate, thyroid hormone, tryptophan, or S-adenosylmethionine (SAM-e). A case study from 2015 demonstrated improvement in response to ECT after folate supplementation.

There is some research that has demonstrated associations between response/remission in ECT and serum levels of various vitamins and essential nutrients. A 1994 study examining the association between serum 5-methyltetrahydrofolate (5-MeTHF) levels and ECT response did not demonstrate any significant association between ECT response and serum 5-MeTHF levels, although low serum 5-MeTHF levels were positively correlated with depression symptom severity. A pilot study has further analyzed serum levels of vitamin B12, folate, S100B, homocysteine, and procalcitonin in patients undergoing ECT, and found that decreased vitamin B12 and folate levels in conjunction with elevated homocysteine and S100B levels lead to increased sensitivity to ECT, as evidenced by increased remission rates. Thyroid hormone has been studied in the setting of ECT therapy, both as an adjunctive treatment for depressive symptoms and as an agent to reduce neurocognitive deficits associated with ECT and has been demonstrated to reduce ECT associated amnesia and promote a decrease in depressive symptoms. Still, thyroid hormone has not been widely used in clinical practice, possibly because of concerns about adverse effects.

D. Creatine and Depression

There is a growing body of literature surrounding the use of standardized, pharmaceutical-grade nutrients (nutraceuticals) as augmentation therapy in the setting of treatment resistant depression. A 2016 metanalysis concluded that adjunctive use of SAM-e, l-methylfolate, omega-3, and vitamin D with antidepressant therapy leads to a reduction in depressive symptoms, while isolated studies showed a similar effect with the use of creatine, folinic acid, and an amino acid combination.

Creatine is a naturally-occurring organic acid that is known to play a role in brain energy homeostasis and is hypothesized to be involved in the pathophysiology of depression via altered energy metabolism. Oral creatine supplementation has been shown to increase cerebral phosphocreatine levels, which is hypothesized to shift cerebral creatine kinase activity, leading to increased ATP production. Early literature suggests that creatine may have an antidepressant effect when used as adjunctive therapy in MDD, due to its role in altering brain bioenergetics. Creatine has been shown to lead to an earlier treatment response in patients treated with escitalopram, with positive response to therapy as early as 2 weeks after beginning treatment. To date, no studies have investigated the use of creatine as an adjunctive therapy to ECT.

Inadequate tissue bioenergetic functioning is thought to be related to disease pathology that affects predominately organs that are comparatively highly metabolically active, like the brain, liver, heart, and skeletal muscle. Because creatine supplementation has the potential to increase bioenergetic stores, it may produce an antidepressant response by enabling synaptogenesis, increasing connectivity between frontal cortical regions and the amygdala, or enhancing frontal cortical functioning. ECT is thought to produce an antidepressant effect largely by promoting synaptogenesis in frontal cortical regions via alterations in the activity of NMDA and AMPA receptors, leading to upregulations in BDNF. Accordingly, creatine has the potential to augment the efficacy of ECT by increasing bioenergetic stores available for synaptogenesis. There is also some evidence to suggest that creatine has activity at NMDA receptors, functioning as a neuromodulator that is released in response to electrical stimulation. Its activity at NMDA receptors is potentially another explanation for its antidepressant effect.

In a mouse model study of hyperhomocystinemia, a condition that leads to impaired creatine kinase activity, creatine supplementation was shown to have neuroprotective effects, with suggestion of memory improvement. Other mouse model studies have shown that creatine can generate improvement of spatial memory, when compared to a traditional diet, as well as improve learning and mitochondrial function. Several human studies have also demonstrated that creatine supplementation is associated with multiple cognitive improvements, including effects on attention, mood, working and long-term memory, and mental fatigue. These data suggest that creatine is a potential approach for addressing the cognitive side effects associated with ECT therapy.

E. Creatine Safety and Toxicity

Retrospective and prospective studies in humans have found no evidence for long-term or short-term significant side effects from creatine supplementation taken at recommended doses. Most controlled studies of creatine report an absence of side effects or report no differences in the incidence of side effects between creatine and placebo. Mihic and colleagues have demonstrated that creatine loading increases fat-free mass, but does not affect blood pressure or plasma creatinine in adult men and women.

Reports in the popular media of links between creatine use and muscle strains, muscle cramps, heat intolerance, and other side effects are not supported by the medical literature. Studies conducted in athletes and military personnel indicate a substantial safety level of both short- and long-term creatine supplementation in healthy adults. Concerns about high-dose creatine's association with renal toxicity are based exclusively on two published case reports; in one of the cases the patient had a documented pre-existing kidney condition. Literature reviews and expert consensus panels have concluded there is no evidence supporting an association between creatine and renal disease.

Concern has been raised regarding creatine's potential for adverse effects on the kidneys and renal system, in part because creatine supplementation can increase urinary creatine and creatinine excretion. In response to the concerns regarding creatine and renal toxicity, Poortman's conducted studies of the effect of creatine supplementation on renal function, showing that short-term supplementation does not alter glomerular filtration rate, and that chronic supplementation of up to five years' duration did not impair renal function in healthy athletes. Other researchers conducted a retrospective study of participants who had been taking oral creatine from 0.8 to 4 years, at an average dose of 9.7 grams per day. Data was collected on 65 health-related variables. These included a complete blood count, 27 serum chemistries, and anthropometric data including vital signs and % body fat. On all 65 variables, group means fell within the normal clinical range. The authors concluded that that long-term creatine supplementation does not result in adverse health effects.

Evidence to date suggests that even aged, debilitated, medically fragile patients are able to tolerate creatine supplementation. Bender and colleagues studied elderly patients with Parkinson's Disease who had received either placebo or four grams/day of creatine for two years. They found no differences between the creatine and placebo groups in laboratory markers of renal dysfunction. Interestingly, the participants who received creatine performed better on the depression subscale of the Unified Parkinson Disease Rating Scale.

No strong evidence exists linking creatine supplementation and gastrointestinal discomfort. These reports remain anecdotal, as there are no documented reports of creatine over placebo resulting in stomach concerns.

1. Overview

Study Timeline

Week Procedures

Visit 1 Eligibility screening: MINI, HAM-D17, MOCA, QIDS, Labs, CGI, BSS, ATRQ

Visit 2 QIDS

31P-MRS Scan 1

Start creatine 20g per day or placebo

Start ECT (treatments 1 up to 3) as per routine

Week 2 (Subjects may begin to complete ECT, moving to follow-up phase) Creatine 5g per day or placebo

Continuing ECT (treatments 3 or 4 through 6 or 7) as per routine

QIDS before each ECT

HAM-D17 , CGI, BSS, MOCA at end of week

Week 3 (Most subjects will complete ECT, moving to follow-up phase) Creatine 5g per day or placebo

Continuing ECT (treatments 3 or 4 through 6 or 7) as per routine

QIDS before each ECT

HAM-D17 , CGI, BSS, MOCA at end of week

31P-MRS Scan 2 (if ECT complete)

Week 4 (Almost all subjects will complete ECT, moving to follow-up phase) Creatine 5g per day or placebo

Continuing ECT (treatments 3 or 4 through 6 or 7) as per routine

QIDS before each ECT

HAM-D17 , CGI, BSS, MOCA at end of week

31P-MRS Scan 2 (if ECT completed this week)

Follow-up phase: 2 weeks (to start after completion of ECT or after week 4, whichever comes first) Creatine 5g per day or placebo (for two weeks)

HAM-D17, CGI, BSS, MOCA two weeks after completion of ECT series

31P-MRS Scan 2 (if not yet completed)

All procedures performed by study personnel are research-related, but will be performed in addition to routine care (see Table 1). None of the study activities will be considered standard of care. There will be no cost to study subjects for their participation. Participants will be compensated for their time and travel. Study visits will be supervised by a board-certified/board-eligible psychiatrist or psychiatry resident and will be conducted either by a board-certified/board-eligible psychiatrist, psychiatry resident, or an at least baccalaureate degree level research assistant with training in the specific measures used. Laboratory and other study interpretation will be conducted by a board-certified/board-eligible psychiatrist.

To determine if an individual is eligible for study participation, a screening visit will be conducted. Initially, a HAM-D17 will be administered to determine if the patient exhibits depressive symptoms that are sufficiently severe for inclusion in the study. Next, the Mini International Neuropsychiatric Interview (MINI) will be administered to confirm a diagnosis of a current major depressive episode. Study subjects will receive a baseline basic metabolic panel (BMP) to assess for renal insufficiency, and vitals (these may have been obtained in the course of ongoing clinical care). The QIDS, Antidepressant Treatment Response Questionnaire (ATRQ), Clinical Global Impression (CGI), Beck Suicide Scale (BSS), and a MoCA will be administered. Each participant will be assessed for history of ECT therapy, current medications, and any other medical history.

Once entered into the study, depressed subjects will be randomized to receive either creatine or placebo using a random-digit method that is based on computer-generated numbers. Block randomization created by Investigational Drug Services (IDS) will be used to ensure equal treatment allocation within each block. 50% of the trial's clinical subjects will be randomized to placebo and the other 50% to active treatment. The study will be conducted as a double-blind trial, with neither participants nor research staff aware of participant assignment. Except in cases of medical emergency, the double-blind will not be "broken" until recruitment is closed and the final participant has completed 6 weeks of treatment and 4 weeks of follow-up. The blind will be broken following the culmination of the study or at the request of a medical professional dealing with a medical emergency in a case in which it would help a study participant.

2. Electroconvulsive Therapy

Participants will be recruited from a population of patients who have already been referred for ECT and who have had initial clinical assessments to determine whether ECT is indicated. As this study is an add-on to standard clinical care only, routine ECT procedures will be followed, though the ECT service (attending psychiatrist, anesthesiologist) will be notified of subject participation. In general, subjects will, after being medically cleared for ECT (psychiatric exam, physical exam, EKG, laboratory studies if indicated), received bifrontal ECT every other day for between 6 and up to roughly 14 treatments, with a total duration of treatment lasting between two and four weeks. Anesthesia for ECT is provided by a board-certified anesthesiologist and comprises the use of methohexital, midazolam, etomidate, or ketamine, as indicated, and at the discretion of the anesthesiologist. Participants may receive interventions designed to augment the likelihood of a seizure being achieved, such as a caffeine infusion, hyperventilation, or other techniques, at the discretion of the treating ECT psychiatrist and anesthesiologist. After ECT, participants are monitored for recurrent seizure/status epilepticus and for vital sign abnormalities for roughly thirty minutes, then released to home (if outpatient) with 24 hour per day supervision by adult family members, or else escorted back to the inpatient unit, where they again receive 24 hour per day supervision by unit staff. Prior to each treatment, subjects complete a Quick Inventory of Depressive Symptoms (QIDS) to assess their overall burden of depressive symptoms. The exact duration of treatment/number of treatments is determined by the treating ECT psychiatrist, based on clinical response. The study will record all pertinent variables related to ECT as noted in the clinical record, including number of treatments, QIDS score, number of seizures per treatment session, seizure duration, adverse effects, augmentation strategies, anesthesia type, and vitals.

2. Drug Dosing

Participants who have been assigned to the creatine arm of the trial will receive a 20g loading dose daily for 1 week starting as soon as possible before ECT begins and after completion of the 31P-MRS; this will be administered in 4 divided doses of 5g each. Participants will then receive 5 g creatine daily throughout the course of ECT therapy (~3 weeks), with continuation for an additional 2 weeks after the completion of the acute series of ECT, again at 5g daily (thus, up to 5 weeks total supplementation with creating 5g per day, depending on the length of ECT) Placebo recipients will receive an inert, relatively tasteless powder matched to creatine (e.g., glucose). Creatine doses are based on doses that have previously been shown to be safe and efficacious.

3. Measures

We plan to use the following for determining participant baseline and data collection:

• Hamilton Depression Rating Scale (HAM-D17) (at baseline, the end of week 1, the end of week 2, the end of week 3, and two weeks after completion of the ECT series)

• Quick Inventory of Depressive Symptomatology (QIDS) (at baseline and prior to each ECT session)

• Antidepressant Treatment Response Questionnaire (ATRQ) (at baseline)

• Mini International Neuropsychiatric Interview (MINI) (at baseline)

• Clinical Global Impressions Scale (CGI) illness improvement subscale (CGI-I) (at baseline, week 1, week 2, week 3, and two weeks after completion of the ECT series)

• Beck Suicide Scale (BSS) at baseline, week 1, week 2, week 3, and two weeks after completion of the ECT series)

• Montreal Cognitive Assessment (MoCA) at baseline, week 1, week 2, week 3, and two weeks after completion of the ECT series)

  4. Imaging

  1. Magnetic Resonance Imaging (Siemens 3T MRI system)

     MRI scans will be conducted twice: after the baseline visit and prior to initiating ECT, and following completion of the ECT series (i.e., after ~3 weeks, and during the 2 week post-ECT follow-up period). The 3.0 Tesla Siemens Prisma whole-body clinical scanner (Siemens Medical Solutions, Erlangen, Germany) located within the University Neuropsychiatric Institute (UNI) will be used to acquire this data. Participants will first undergo a routine anatomic MRI protocol, which includes MRI images acquired in the axial and coronal planes. Specifically, the anatomic scan protocol consists of a T1 weighted structural scan (MP2RAGE), and double-echo T2 weighted scan, and a Fluid Attenuated Inversion Recovery scan (FLAIR). The purposes of the MR anatomic screening session include screening subjects for gross structural abnormalities and acquiring images for use in brain cortical thickness measurements. Anatomic MRI examinations will be performed with Siemens 64 channel head coil. After localization, anatomical imaging will be obtained using a T1-weighted, sagittal oriented 3D-Magnetization Prepared Rapid Gradient Echo (MPRAGE) sequence (TR/TE/TI 5000/2.93/700 ms, matrix 256x256, FOV 256x256 mm, flip angle 4 degree, slice thickness 1.0 mm, slab 176 mm, bandwidth 240 Hz/pixel). Axial proton-density and T2 weighted images will be acquired to screen for brain structural abnormalities using 2D Double echo T2 weighted turbo spin echo (TSE) sequence (TR 7110 ms, TE 28/84 ms, FOV 240x210, slice thickness 3 mm, flip 150°, bandwidth 179 Hz/pixel). FLAIR sequence (TR/TE/TI 8000/90/2500 ms, slice thickness 5 mm, FOV 240x168, voxel size 0.8x0.6x5.0 mm, bandwidth 200 Hz/pixel, turbo factor 13) will be used to detect juxtacortical-cortical lesions. All anatomic MRI images will be read by a board-certified, CAQ neuroradiologist to screen for structural abnormalities.

  2. Measurement of In-Vivo Brain Chemistry Using Phosphorus-31 Magnetic Resonance Spectroscopy (31P-MRS)

     Phosphorus spectroscopy data will be acquired on the same Siemens 3T system. We aim to keep the duration of each MRSI examination at or under 25 minutes. A 3D-MRSI sequence with elliptically weighted phase-encoding will be used to collect 31P-MRSI data to minimize T2 signal decay. Acquisition parameters will be: data matrix size 16x16x8; TR 2000 ms; tip-angle 90 degree for hard RF pulse; Rx bandwidth ±1 kHz; complex-points 1024; readout duration 256 ms; pre-acquisition delay 0.3ms; FOV 240x240 mm2 ; 16 NEX.

  3. Spectral Analysis of 31P-MRS Data

Spectroscopy will be analyzed using Liner Combination of Model Spectra (LCModel), which analyzes an in vivo spectrum as a linear combination of model in vitro spectra from individual metabolite solutions. This model is fully automatic and user independent. A nearly model-free constrained regularization method is used for convolution and baseline. For quantification, absolute metabolite concentrations (institutional units) will be estimated using the unsuppressed water signal as an internal concentration reference. Also, total creatine levels will be used as a denominator for calculating the relative concentration for the comparison with previous reports. The standard Siemens libraries of model metabolite spectra provided with LCModel will be used in the basis set. The metabolites from the basis set will include alanine, aspartate, creatine, gamma-amino butyric acid, glucose, glutamine, glutamate, glycerophosphocholine, glutathione, myo-inositol, scyllo-inositol, lactate, N-acetylaspartate, N-acetylaspartylglutamate, phosphocholine, phosphocreatine, phosphoethanolamine, and taurine. For the reliability of detection, the Cramer-Rao lower bounds (CRLB) will be determined: the acceptable upper limit of estimated standard deviations will be set at 20%.

Post processing of 31P-MRS data will be conducted using jMRUI software (jMRUI v. 4.0, European Community) with the AMARES algorithm (Advanced Method for Accurate, Robust and Efficient Spectral fitting of MRS data with use of prior knowledge). Before fitting the FID (Free-induction-decay) data, a Hamming filter will be applied to reduce signal contamination from neighboring voxels, with apodization of 10 Hz line broadening. Fourier transformation, frequency shifts correction, and zero-order/first order phase correction as well as baseline correction will be applied. The structural image-processing tool FSL (FMRIB Software Library, Release 4.1, University of Oxford) will be used to account for gray matter, white matter, and cerebrospinal fluid (CSF), in order to correct the partial volume effects on metabolite concentrations. The MRS grid will be positioned over the images in an identical fashion between baseline and treatment scans for each participant. The peak area for each 31P-MRS metabolite will be calculated as a percentage of the total phosphorus signal.

Inclusion Criteria:

Participants with a diagnosis of Major Depressive Disorder with moderate to severe symptoms will be randomly assigned in a 1:1 ratio to either ECT with creatine augmentation or ECT with placebo augmentation or 6 weeks. Individuals must meet criteria for major depressive episode for a duration of at least 2 months in order to participate. Participants must be greater than 18 years of age and considered to be a good candidate for ECT based on clinical assessment. Individuals who are pregnant or breast-feeding or who possess any other contraindication to ECT will not be invited to participate. Participants can receive ECT treatments in either inpatient or outpatient setting at the University Neuropsychiatric Institute.

Exclusion Criteria:

Individuals who meet diagnostic criteria for other psychiatric conditions apart from major depressive disorder (including Bipolar I, Bipolar II, or personality disorders) will not be invited to participate. Individuals with substance use disorders will be excluded because substance use disorders typically confound the diagnosis of depression and can contribute to treatment resistance. Individuals will not be considered for study participation if they have renal disease because to date it cannot definitively be stated if short and long-term creatine usage is or is not harmful to the kidneys. Appropriate renal function will be determined based on normal creatinine clearance, determined by routine laboratory work (basic metabolic panel).

Participants who are already undergoing electroconvulsive therapy (ECT) or transcranial magnetic stimulation (TMS) for the treatment of depression, or who have completed a course of ECT within a month of the baseline visit, will not be invited to participate given the possibility of confounding treatment effects as well as increased seizure risk. Individuals currently undergoing psychotherapy remain eligible to participate.

Participants who have implanted ferromagnetic hardware, implanted electronic devices, or retained ferromagnetic materials from surgery or injuries will not be invited to participate as these represent contraindications to MRI. Likewise, individuals who are unable to tolerate confinement in the MRI scanner will not be invited to participate.

Individuals who lack capacity to consent to treatment or to participate in the study will be excluded. Patients who are hospitalized involuntarily will also be excluded. Individuals demonstrating active psychosis or any other clinical characteristic making them inappropriate candidates for treatment will be excluded. Patients with pre-existing neurologic condition, any major neurocognitive disorder, or known traumatic brain injury will not be invited to participate.

During six weeks of study participation.

Standard definitions used.

Randomized, double-blind, placebo-controlled. Block randomization protocol.