Major Depressive Disorder

Depression is the second-most common cause of disability.

Background

Adult lifetime prevalence of major depressive disorder (MDD) is 13–16% and approximately 7% for twelve month prevalence in the United States. Hallmark symptoms include depressed mood (e.g. feelings of sadness or emptiness) and loss of interest or pleasure. Untreated episodes usually last six months or longer. The longer a patient has been depressed, the likelihood of spontaneous recovery of depression decreases. Approximately one-third of major depressive episodes do not resolve, and about 15% of patients with severe MDD successfully commit suicide. MDD is diagnosed through the presence of symptoms that meet criteria for a major depressive episode (MDE).

The Diagnostic and Statistical Manual of Mental Disorders 4th Edition (DSM-IV) diagnostic criteria symptoms for MDE are summarized in the table below. With the exception of suicide, the symptoms must occur throughout the day or nearly every day. Depressed mood or loss of interest or pleasure must be one of five or more symptoms that occur within the same two week period. Symptoms should cause significant distress and social impairment. MDD is considered chronic if the criteria for a MDE are continuously met for at least two years. MDD is considered recurrent if a patient has experienced two or more MDE in which criteria for MDE have been absent for at least two months between episodes. A recurrent episode is 60% likely to occur in patients who experience a single MDE, 70% likely after two MDEs, and 90% likely after three MDEs.

DSM-IV diagnostic criteria symptoms for MDE

Diagnosis must include ≥5 of the following symptoms within a two week period.

• Depressed mood*
• Markedly diminished interest or pleasure in most activities*
• Significant weight loss or weight gain, or decrease or increase in appetite
• Insomnia or hypersomnia
• Psychomotor agitation or retardation observable by others
• Fatigue or loss of energy
• Feelings of worthlessness or excessive or inappropriate guilt
• Diminished ability to think or concentrate, indecisiveness
• Recurrent thoughts of death, suicidal ideation, or suicide attempt

Table adapted from DSM-IV Criteria for Major Depressive Episode.
*Either of these must be included within the five symptoms.

For a diagnosis of MDD, symptoms must not be attributed to bereavement, other conditions (e.g. hypothyroidism), or substances (e.g. drugs of abuse). A diagnosis of MDD must not include a history of manic or hypomanic episodes. The diagnosis of mixed depressive disorder is considered to be a separate diagnosis and includes criteria for both major depressive episode and manic episode. Hypomanic episodes (Bipolar II Disorder) commonly occur before or following a MDE, and patients with Bipolar II Disorder are more likely to experience multiple MDEs. A portion of patients that experience a single MDE (10-15%) subsequently develop a manic episode (Bipolar I Disorder).

DSM-V, the most recent edition, differs from DSM-IV criteria in that it added the word “hopeless” to subjective descriptors of depressed mood, removed a statement regarding mood-incongruent delusions or hallucinations, and removed the bereavement exclusion.4 Since these definitions broaden inclusion of diagnosis, these updated definitions would not affect patients previously diagnosed with DSM-IV criteria.

There is no gold standard for the treatment of depression. Treatment of depression is dependent on patient preference and response to previous therapy. Treatment of depression includes pharmacologic therapy, behavioral therapy, somatic therapy, and a combination of pharmacologic and behavior therapy. Recommended pharmacological treatment includes selective serotonin reuptake inhibitors (SSRIs), serotonin norepinephrine reuptake inhibitors (SNRIs), tricyclic antidepressants (TCAs), monoamine oxidase inhibitors (MAOIs), and other antidepressants (e.g. bupropion, trazodone).

The use of MAOIs is generally reserved for those who fail treatment with other antidepressants because of an increased risk for food and drug interactions. The success rate for pharmacologic treatment is roughly 50–70%. Full response generally occurs within four to eight weeks. After at least four to six weeks, patients should be evaluated for partial or non-response. Patients that require dosage adjustments to their antidepressant regimen should be evaluated after an additional four to eight weeks.

Treatment-resistant depression (TRD) occurs in roughly 15% of those diagnosed with MDD. Although no definition of treatment-resistant depression exists, there is a consensus that an individual must fail at least two adequate trials of different classes of antidepressants. A sufficient trial includes adequate dosage, duration, and compliance. A score of at least 16 on the 17-item Hamilton Depression Rating Scale may be considered sufficient to confirm TRD.

Electroconvulsive therapy (ECT) improves depression for 70–90% of patients and should be considered for patients with severe depression who have failed prior antidepressant therapies. ECT treatments are usually two to three times per week and an acute course usually requires six to twelve treatments. Less frequent administrations result in delayed onset of action and the risk of relapse is greater with abrupt discontinuation of therapy. Perceivably, frequency of hospitalization and duration may represent a treatment burden to patients.

Patients must attend sessions frequently to derive benefit from ECT. As well, patients may not experience symptom improvement until four to eight weeks on conventional antidepressants alone. Considering the delayed time of onset with current therapies, a medication that exhibits rapid onset of action with a single dose seems attractive. Multiple case reports have demonstrated rapid improvement in subjective symptoms in patients with MDD and TRD treated with ketamine (Ketalar) alone or in combination with ECT. For example, one patient treated with a ketamine continuous infusion daily for five days reported improved mood after 24 hours from start of infusion.

Within the same case series, another patient treated for five days improved and subsequently was treated again at 2.5 months and 8.5 months from first infusion. Both patients held response for 12 months after their last infusion while also being maintained on an antidepressant regimen.7 Other, open-label studies have also shown benefit with ketamine in the treatment of MDD.19-30Treatment scales used in studies evaluating ketamine’s efficacy for treating depression include: Hamilton Rating Scale for Depression (HDRS), Montgomery-Åsberg Depression Rating Scale (MADRS), Beck Depression Inventory (BDI), Clinical Global Impression (CGI), Inventory of Depressive Symptomatology (IDS or IDS30) or Quick Inventory of Depressive Symptomatology (QIDS or QIDS16), Clinician Administered Dissociative States Scale (CADSS), Visual Analog Scale (VAS), Brief Psychiatric Rating Scale positive symptom subscale (BPRS+), and Young Mania Rating Scale (YMRS). The gold standard rating scales for depression are HDRS, BDI, and IDS or QIDS. HDRS17, a clinician rated scale consisting of 17 items, is the most utilized scale in clinical trials examining depression.

However, HDRS scales may include different versions such as the 21 and 24 item HDRS. In trials with ketamine, HDRS or MADRS were used to assess primary outcomes of decreased depression severity. Other scales were used to monitor for potential adverse effects unrelated to depressive symptoms.

CADSS, VAS, BPRS+, and YMRS were used as safety measures to monitor for potential negative psychological effects of administering a dissociative anesthetic. The CADSS consists of 27 items and each item is rated 0 to 4. With this scale, patients are asked to rank questions relating to specific alterations in perceptions (e.g. objects moving in slow motion, differences in color perception). A high score represents increased intensity or frequency of dissociative symptoms.32 VAS is used to quantify a single subjective experience on a numeric continuum. The range used in the scale can be arbitrary and, because of the subjective nature, ratings can vary widely between patients.

Due to this limitation, comparisons of scores between patients are not reliable.33The BPRS+ ranks severity of positive symptoms with 1 being not present to 7 being extremely severe. 34 Lastly, the YMRS contains 11 items and each item is ranked 0 to 4. Increased scores correlate to increased severity of manic symptoms such as irritability, mood, frequency of speech, and hallucinations.35 Often, studies that examined the use of ketamine for MDD used more than one scale to examine the safety profile of ketamine. This is also true for the measurement of depressive symptoms. Studies often used other scales as secondary outcomes to measure reduction in depressive symptoms.

BDI, CGI, and IDS or QIDS have been used as supplemental scales to assess particular features of depression or as adjuncts to severity classification. CGI measures consist of three distinct dimensions: severity (CGI-S), improvement (CGI-I), and efficacy. CGI-S measures severity on a 7-point scale as 1 being normal to 7 being extremely ill. CGI-I measures improvement from baseline on a 7-point scale with 1 being very much worse to 7 being very much improved. Efficacy is measured on a different 4-point scale.

An estimated comparison of scores for HDRS, MADRS, BDI, IDS, and QIDS is listed in Table 2. These comparisons are estimations on equivalency of scores between scales. Severity of depression of individual scales may differ from what is represented below. For example, HDRS17 score interpretation differs from the table in that scores can be interpreted as the following: 0−6 indicates no depression, 7−17 indicates mild depression, 18−24 indicates moderate depression, and scores over 24 indicate severe depression.31 For purpose of analyzing different scales, the following chart will be used to qualitatively describe data from ketamine trials in an effort to compare results.

Estimated comparison of total scores used to evaluate depression

Scale	Severity of Depression
	None	Mild	Moderate	Severe−Very Severe
Hamilton Depression Rating Scale, 17-item (HDRS17)	0−7	8−13	14−19	20−25 (severe) 26-52 (very severe)
Hamilton Depression Rating Scale, 21-item (HDRS21)	0−8	9−15	16−22	23−28 (severe) 29-64 (very severe)
Hamilton Depression Rating Scale, 24-item (HDRS24)	0−9	10−18	19−26	27−34 (severe) 35−75 (very severe)
Montgomery-Åsberg Depression Rating Scale (MADRS)	0−6	7−19	20−34	35−60
Beck Depression Inventory (BDI)	0−9	10−18	19−29	30−63
Inventory of Depressive Symptomatology (IDS30)	0−11	12−23	24−36	37−46 (severe) 47−84 (very severe)
Quick Inventory of Depressive Symptomatology (QIDS16)	0−5	6−10	11−15	16−20 (severe) 21−27 (very severe)

The effects of antidepressants may be related to changes in synaptic connections in the brain. These synaptic changes result in reduced NMDA receptor function and up-regulation of AMPA (α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid) receptor sub-unit phosphorylation. Ketamine, an N-methyl-D-aspartate (NMDA) receptor antagonist and phencyclidine derivative, is thought to increase presynaptic release of glutamate and increase glutamatergic effects on AMPA receptors over NMDA receptors due to selective blockade. Glutamate may play an important role in the treatment of MDD since patients with MDD have higher levels of plasma glutamate and antidepressant treatment is associated with lower plasma glutamate levels. Rapid effects seen in case studies using ketamine may be related to rapid induction of synaptic proteins and changes in synaptic function. These changes may also explain the relative longevity of therapeutic effects that persist after administration of ketamine.

Currently, ketamine is indicated for induction of anesthesia in adjunct to general anesthetic agents, and as a single anesthetic for procedures that do not require skeletal muscle relaxation. Contraindications include hypersensitivity and predisposition to serious injury with significant elevation of blood pressure. Surgical anesthesia is usually achieved within 30 seconds with a 2 mg/kg intravenous (IV) dose of ketamine. Without further administration, the effects of anesthesia generally last five to ten minutes. Compared to IV administration, intramuscular (IM) doses exhibit delayed onset of action and longer duration. Since patients respond differently to ketamine, doses should be titrated to response. Generally, the initial IV dose required for surgical anesthesia ranges from 1─4.5 mg/kg. To maintain sedation, one-half to the full initial dose is used. Most studies that examined the use of ketamine in the treatment of MDD utilized an IV dose of 0.5 mg/kg which is below the dosage range used for initiating surgical anesthesia.

At this point in time, there are no pharmacologic treatments indicated for the rapid reversal of depressive symptoms. Current treatment modalities for depression do not resolve symptoms of depression in a timely manner.

Methodology

A systematic search using PubMed was conducted using a combination of the following MeSH terms: “ketamine” and “depression,” “depressive disorder,” or “depressive disorder, treatment-resistant.” Search results were limited to randomized, double-blind clinical trials, humans, and English. Results were excluded if the focus was not on treating major depressive disorder.

Results

The use of ketamine for MDD has been evaluated in case studies and clinical trials. A small number of trials were double-blind, randomized clinical studies. Some of these studies utilized other medications in concert with ketamine, such as propofol (Diprivan), fentanyl (Sublimaze), and thiopental (Pentothal). Thiopental and propofol are used in the induction of anesthesia. Thiopental is a barbiturate anesthetic and propofol is thought to augment the inhibitory effects of γ-aminobutyric acid (GABA) on GABAA receptors.41,42 Parenteral fentanyl is an anilidopiperidine opioid indicated as an adjunct to general anesthesia.43 A single study used midazolam (Versed), a benzodiazepine, as an active control in order to mimic the sedative properties of ketamine. Other studies used normal saline (NS) as placebo.

Randomized, Double-Blind Clinical Studies

Randomized, double-blind studies that evaluated the use of ketamine in the treatment of depression utilized small study sizes and often used normal saline (NS) as placebo infusion. Eight studies specifically evaluated the use of ketamine in treating major depressive disorder. Of these eight studies, three allowed patients with history of bipolar depression (BD) to enroll. Half of these studies used other medications as comparators or an active placebo. Studies excluded patients with a recent history of substance abuse. Primary outcomes were reduction in depressive symptoms as measured by either HDRS or MADRS. Five studies utilized HDRS and the other studies utilized MADRS for the primary outcome.

Berman, et al. (2000)

In a small, placebo-controlled, crossover study with nine patients, ketamine significantly reduced HDRS and BDI scores. Patients received normal saline and 0.5 mg/kg ketamine infusions separated by at least one week. HDRS25­ and BDI were the primary outcomes measured. The order of treatment was randomized. HDRS scores after treatment were compared to baseline scores to determine significance. Baseline was measured prior to treatments. Baseline BDI scores were 29.5±8.2 for ketamine and 23.0±8.2 for placebo. BDI scores post-infusion were 16.8±10.5 for ketamine and 25.2±6.0 for placebo. Roughly translated, post-infusion scores for ketamine ranged from no symptoms of depression to moderate symptoms, while placebo post-infusion scores ranged from moderate to severe symptoms of depression. This comparison was made from Table 2 and may be potentially inaccurate, but it does give a rough estimation.44

Treatment with ketamine was not found to significantly alter the effects or results of placebo treatment in terms of HDRS change from baseline. In other words, patients treated with NS as the second treatment did not experience significant change from second baseline HDRS, measured before NS treatment, to post-NS infusion HDRS. Only one patient showed improvement with NS. Three patients remained similar to baseline HDRS. Four patients demonstrated >50% reduction in HDRS scores with ketamine treatment and opposed to one patient on NS. Ketamine significantly decreased individual HDRS items of depressed mood, suicidal ideation, helplessness, and worthlessness. Ketamine also significantly increased VAS and BPRS+ scores. VAS scores returned to baseline by 110 minutes post infusion. The positive effects of ketamine subsided one to two weeks after infusion. Subsidence, or relapse of scores, was considered to be an HDRS score within 5 points of baseline or as decided by a clinician. One patient maintained response for two weeks and had a HDRS score of 41 at baseline, 7 at day three, and 15 at two weeks. Of nine patients enrolled, three patients did not complete both treatment arms.44

Three patients dropped from the study. Two patients dropped to initiate antidepressant treatment. Of these two, one patient completed only the NS infusion and the other completed ketamine infusion only. One patient only completed the ketamine infusion and was excluded from analysis. This patient did, however, experience a 17 point reduction in HDRS score from baseline (16 v. 33, significance not reported). This study included patients with recurrent MDD, but one of the patients had BD. This inclusion may or may not have been a limitation in this study. The difficulty of ascertaining the effect of this patient inclusion involved an overall limitation in the lack of patient specific characteristics, especially in terms of baseline HDRS items. The results did not differentiate the effects of ketamine on the one patient with bipolar depression. Also, VAS and BPRS+ scores were not well defined.44 Another study that used the HDRS scale also examined MDD patients, but used a different dose of ketamine.

Kudoh, et al. (2002)

A randomized, double-blind, clinical study examined the antidepressant effects of ketamine with patients that underwent orthopedic surgery. The subjects in this study included patients with MDD and patients without depression. The subjects were randomized into three groups. HDRS21 was the primary outcome measurement. Two groups, patients with MDD and patients without MDD, were given ketamine 1 mg/kg, propofol 1.5 mg/kg, and fentanyl 2 mcg/kg to induce anesthesia. The third group also included patients with MDD and was given the same doses of propofol and fentanyl, but was not given ketamine.

Baseline HDRS values were taken two days prior to surgery and were 12.7±5.4 for Group A and 12.3±6.0 for Group B. HDRS was significantly decreased one day after surgery in the group of patients with MDD treated with ketamine, Group A, as opposed to the group of patients with MDD not treated with ketamine, Group B (9.9±4.1 v. 14.4±3.8, respectively; p 0.05). Using Table 2, scores roughly translate to no symptoms to mild symptoms for Group A, and mild to moderate symptoms of depression for Group B. Group A, compared to Group B, also experienced significant decreases in depressed mood, suicidal tendencies, somatic anxiety, and hypochondriasis (p 0.001). This study did not use VAS, BPRS+, or CADSS as safety measures. Hypotension (systolic pressure 70mmHg) was experienced by two patients in Group A and one patient in Group B during induction of anesthesia. The next study was conducted similarly to the first study mentioned, but also used HDRS21 instead of HDRS25.45

Zarate, et al. (2006)

In a crossover study, eighteen patients with MDD were selected to receive 0.5 mg/kg ketamine and NS separated by one week. Treatment was infused over 40 minutes. The order of treatment given was randomized. HDRS21 was the primary outcome measure and BDI was the secondary outcome measure. Response to treatment was considered to be a ≥50% reduction in HDRS score. According to this criterion, twelve patients met response one day after ketamine infusion. Of these twelve, six maintained response for at least one week and two maintained response for at least two weeks. Remission was considered to be a HDRS score of ≤7 (absence of depressive symptoms). Remission criterion was met by five subjects one day after infusion. Placebo did not attribute to remission status. Ketamine was associated with significant improvement versus placebo from 110 minutes to 7 days post-infusion with both primary and secondary measures. BPRS+ and YMRS were significantly elevated after ketamine infusion. For both measurements, scores peaked at 40 minutes but did not persist beyond 110 minutes. The most common adverse reactions for ketamine were perceptual disturbances, confusion, elevations in blood pressure, euphoria, dizziness, and increased libido. The frequency of adverse events was not stated.46

Three studies were conducted to assess the effects of ketamine with ECT. Of these three studies, only one study found significant reduction in depressive symptomatology with the use of ketamine as an adjunct treatment. The one study that found benefit used the HDRS scale while the other two studies used either HDRS or MADRS. The studies used different interventions. For example, one study used thiopental in combination with ketamine.

Abdallah, et al. (2012)

A study was conducted with eighteen subjects that were diagnosed with either MDD or BD. Exclusion criteria included substance dependence, presence of a serious medical condition, lifetime diagnosis of primary psychotic disorder, mental retardation, dementia, or mood disorder due to a general medical condition. Patients were allowed to take other psychiatric medications (e.g. antidepressants, mood stabilizers, antipsychotics) during the study. Patients had severe to very severe depression at baseline. Significantly more patients in the placebo group had BD (N=4 v. N=1, p 0.04). All subjects underwent ECT three times a week for two weeks. Before the initial ECT session, subjects were given 0.5 mg/kg ketamine and 3.5 mg/kg thiopental IV push, or only 3.5 mg/kg thiopental IV push. No major adverse effects were noted. No significant differences in HDRS25 reduction were noted between groups after first or sixth treatment.47 Another study found similar results.

Loo, et al. (2012)

A study that examined the use of ketamine as an adjunctive treatment with ECT found no significant difference in MADRS reduction between intervention and control groups. Subjects included patients with MDD and nine patients with BD. Patients with a diagnosis of schizophrenia, schizoaffective disorder, rapid cycling bipolar disorder, current psychotic symptoms, history of ECT within the last three months, history of alcohol or substance abuse within the last 6 months, history of significant neurological disease, or history of ketamine hypersensitivity were excluded. Patients were allowed to take other psychiatric medications during the study. ECT was administered three times a week for up to six treatments. Each group was given 3─5 mg/kg thiopental IV for induction of anesthesia, with the exception of three patients who received 1─2 mg/kg propofol due to drug shortage. The intervention group was given a 0.5 mg/kg ketamine IV push and the control group received NS IV push. Patients within the study were diagnosed with either MDD or BD. Even after patients with BD were excluded from analysis, no significant differences in MADRS reduction for the whole duration of the study were noted. The ketamine group showed significant response during the first week, but this difference was not significant after subjects with BD were excluded from analysis. Regarding adverse effects seen in the ketamine treatment group, one patient with BD became hypomanic and one patient developed rapid cycling manic symptoms. Other adverse effects were not stated.48

Wang, et al. (2012)

The study that showed response with ketamine and ECT included only patients diagnosed with MDD. HDRS25 was the primary outcome measured. In this study, patients were given ECT followed by either 0.8 mg/kg ketamine, 1.5 mg/kg propofol, or 0.8 mg/kg ketamine and 1.5 mg/kg propofol. HDRS scores were measured one day before ECT and 1, 2, 3, and 7 days post-ECT. HDRS scores were significantly lower at days 1, 2, and 3 in the groups that received ketamine as opposed to the group that received only propofol. No significant reduction in HDRS was noted between ketamine groups. After seven days post-ECT, no significant difference was noted between any of the groups. Adverse events in the ketamine group that occurred significantly more often than the propofol group were hypertension during ECT (67% v. 25%) and sense of fear with hallucinations upon awakening from anesthesia (25% v. 0%). Vascular pain or angialgia occurred significantly more frequently in the propofol group versus the ketamine group (42% v. 0%).49 ECT was only given once in this study but is usually given two to three times a week.

Murrough, et al. (2013)

The previous studies compared ketamine to NS or used ketamine in addition to propofol, propofol and fentanyl, or thiopental. The next study is unique because midazolam was used as an active control. The study included only patients with TRD. Subjects were randomized in a 2:1 ratio to receive either 0.5 mg/kg ketamine or 0.045 mg/kg midazolam. Treatments were infused over 40 minutes. Of 73 patients, 47 received ketamine and 25 received midazolam. The 2:1 ratio was determined to provide 80% power to the ketamine group and 96% power to the midazolam group to detect a reduction in MADRS scores at 24 hours. Ketamine was expected to decrease MADRS scores in 60% of subjects while midazolam was expected to cause a decrease in MADRS scores in 15% of subjects. These estimations were based on the results of previous studies that used NS as placebo. Midazolam was expected to show the same response of placebo.50

Reduction in MADRS at 24 hours post-infusion was the primary outcome. Baseline MADRS scores were 32.6±6.1 and 31.1±5.6 for ketamine and midazolam groups, respectively. The ketamine group demonstrated lower MADRS scores than the midazolam group (7.95 mean reduction; 95% CI, 3.20─12.71). Mean MADRS scores 24 hours post infusion were 14.77 (95% CI, 11.73─17.80) and 22.72 (95% CI, 18.85─26.59) for ketamine and midazolam groups, respectively. Response, defined as ≥50% decrease in MADRS scores, was a secondary outcome. In the ketamine group, 64% met response and 28% met response in the midazolam group (Odds Ratio, OR 2.18; [95% CI, 1.21─4.66; p≤0.001]). Other secondary outcomes included QIDS, CGI-I, and CGI-S. QIDS was lower in the ketamine group by 3.4 points (95% CI, 0.78─6.01; p≤0.02). The ketamine group was also more likely to show improvement on the CGI-I measure (OR 2.31; [95% CI, 1.25─4.14; p≤0.004]). The odds that a subject would be rated minimally ill or not ill at all on the CGI-S measure were greater for the ketamine group (OR 4.08; [95% CI, 1.76─13.51; p≤0.001]). The most common side effects associated with ketamine were dizziness, blurred vision, headache, nausea or vomiting, dry mouth, poor coordination, poor concentration, and restlessness. Dissociative symptoms occurred in 17% of ketamine subjects but resolved by two hours post-infusion. No severe psychotic symptoms were reported.50

Sos, et al. (2013)

Instead of a single dose of ketamine administered at one time or as a single infusion, a study examined the effects of ketamine given as two infusions with different rates. In a crossover study with 30 subjects with either recurrent MDD or single MDE, ketamine was given as a 0.27 mg/kg infusion over 10 minutes followed by another 0.27 mg/kg infusion over 20 minutes. The ketamine dosage used in the study was calculated based on the pharmacokinetics of ketamine in an effort to provide stable levels, and to provide a dose that was close to those used in previous trials (i.e. 0.5 mg/kg). Unlike the previously mentioned study that compared ketamine to midazolam, the placebo treatment was NS.

Patients included had a MADRS score of at least 20. Exclusion criteria included suicide risk, psychiatric comorbidity, serious unstable medical illness or neurological disorder (e.g. epilepsy), lifetime history of psychotic symptoms or disorder in first or second degree relatives, and ECT therapy within three months of the study. Patients were on stable doses of antidepressants for at least three weeks prior and throughout the study. Response, defined as ≥50% MADRS reduction, was significantly greater in the ketamine group versus placebo at days 1, 4, and 7 post-infusion. Five patients maintained response through post-infusion days 1 through 7 and up to eleven patients achieved response during the study. Change in BPRS total score was not significant. Ketamine was well tolerated and side effects noted were dissociation and perceptual disturbances, confusion, mild increases in blood pressure, emotional blunting, and euphoria. Emotional blunting, euphoria, or dissociation did not persist 60-minutes post-infusion and other symptoms resolved usually within 30 minutes post-infusion.51

Summary of Double-Blind, Randomized Studies

Table 3 summarizes the randomized, double-blind studies that examined the use of ketamine and shows patient characteristics in terms of diagnosis of MDD and BD. The other therapy heading refers to the other medications used with ketamine in the studies. The scale heading refers to the primary outcome measured. Studies were considered to show reduction if ketamine significantly differed in terms of primary outcome.

Study	Population	IV Ketamine Dose	Other Therapy	Scale	Reduction
Berman, et al. (2000)	N=9 MDD, BD	0.5 mg/kg infusion over 40 min.	None.	HDRS25	Yes
Kudoh, et al. (2002)	N=95 MDD, non-MDD	1 mg/kg (rate not given)	1.5 mg/kg propofol, 2 mcg/kg fentanyl	HDRS21	Yes
Zarate, et al. (2006)	N=18 MDD	0.5 mg/kg infusion over 40 min.	None.	HDRS21	Yes
Abdallah, et al. (2012)	N=18 MDD, BD	0.5 mg/kg IV push	ECT 3.5 mg/kg thiopental	HDRS25	No
Loo, et al. (2012)	N=51 MDD, BD	0.5 mg/kg bolus	ECT 3─5 mg/kg thiopental 1─2 mg/kg propofol	MADRS	No
Wang, et al. (2012)	N=48 MDD	0.8 mg/kg (length of infusion not given)	ECT 1.5 mg/kg propofol*	HDRS17	Yes
Murrough, et al. (2013)	N=73 TRD	0.5 mg/kg infusion over 40 min.	None.	MADRS	Yes
Sos, et al. (2013)	N=30 MDD	0.54 mg/kg infused over 30 min. (0.27 mg/kg infused twice)	None.	MADRS	Yes

*Two groups were given ketamine: one group was given only ketamine and another group was given ketamine and propofol.

Discussion

All trials used criteria from DSM-IV for MDD. The changes in DSM-V likely do not affect the interpretation of these study results.

Limitations of these studies include differing scales, and comparisons between studies are difficult because of the variations between scales. Another limitation of these studies includes different populations. Some studies allowed BD patients to enroll while others did not. The effect of patients with BD in the studies was not clear. One trial demonstrated significant difference in favor of ketamine when patients with BD were included in analysis but did not show significance when this subject population was excluded. Small population sizes could be considered another limitation. Future studies should examine larger populations and examine specific facets of depressive disorder (i.e. chronic MDD versus recurrent MDD). Since studies have either excluded patients with BD, or included only small numbers of patients with BD, ketamine should not be considered for this population until further studies have been evaluated. Ketamine should also not be given to patients that would be at risk for hypertensive crises considering that ketamine was shown to increase blood pressure in a few studies.

In open label trials, inclusion and exclusion criteria differed only slightly. With the exception of three studies, subjects were excluded if diagnosed with BD. Considering bipolar I depression is undetected in 35–45% of patients, excluding patients due to previous history of manic or hypomanic episodes is appropriate to distinguish between patients who truly have MDD. Trials generally excluded subjects with a history of substance abuse or dependence within the last three to six months. Only a few randomized, double-blind trials included a wash-out period for antidepressants. Further studies should examine the safety of ketamine in conjunction with traditional antidepressant therapy.

Studies used either HDRS or MADRS to assess primary outcomes relating to reduction of depressive symptoms. While HDRS is the most commonly used scale for clinical trials studying depression, it is criticized for the failure to include all symptoms of MDD and uneven weight of particular items, such as insomnia. Due to the uneven weight attributed to sleep integrity, HDRS is believed to be better at detecting superiority for some antidepressants, such as tricyclic antidepressants (TCA), due to hypnotic effect. MADRS has been shown to be as effective as HDRS in detecting changes between antidepressants and placebo.

Ketamine seems to be a promising adjunctive therapy for use in MDD. The onset of response is rapid. Significant improvement can occur in less than 24 hours; however, duration of effect seems to wane after seven days. The recommended dose of ketamine for MDD treatment should be an IV infusion of 0.5 mg/kg over 40 minutes. This dosage has been studied in previous open-label trials and three randomized, placebo-controlled trials. Treatment with ketamine should be considered for patients with TRD. Until trials directly compare ketamine against ECT, ketamine should not replace ECT. Studies that have used ketamine in conjunction with ECT have not shown clear benefit when ECT was given frequently. Regarding this, ketamine may not be a useful adjunct to therapy when ECT is administered frequently or up to three times a week. Ketamine may be useful when ECT is given only once a week. Further studies with larger sample populations should be conducted to assess if ketamine is able to decrease the number of ECT treatments needed to elicit response or reduction of depressive symptoms.

Further studies should compare the antidepressant effects of ketamine between patients with MDD and patients with BD. Also, further studies should be conducted to determine if the positive effects of ketamine can be stabilized with current antidepressant therapies. As well, the safety of ketamine in conjunction with other antidepressant therapy should be addressed. If found to have an acceptable safety profile with other antidepressants, ketamine should be studied as a possible adjunct to decrease the latency of onset associated with traditional antidepressants.

Conclusion

Ketamine infusion has been shown to effectively reduce symptoms of major depressive disorder in randomized, double-blind, placebo-controlled trials. Further studies should seek to prolong the durability of ketamine’s efficacy or examine the use of ketamine in addition to other antidepressant therapy to determine if ketamine is effective in decreasing the lag-time associated with traditional antidepressants.