Guest Commentary by Rajnish Mago, MD
GME Research Review is a monthly newsletter where internationally recognized experts select, summarize, and provide a clinical commentary on the latest published research in psychiatry. Each summary has been derived from the relevant article’s abstract and the clinical commentary has been provided by our expert.
- Bipolar disorder: What happens in real life to patients discharged from the hospital?
- A systematic review of the data on rapid cycling in bipolar disorder
- For alcohol detoxification, is symptom-triggered lorazepam better than fixed tapering dose?
- The two key clinical trials of suvorexant (Belsomra®)
Simhandl et al. A prospective 4-year naturalistic follow-up of treatment and outcome of 300 bipolar I and II patients. J Clin Psychiatry. 2014;75(3):254-262. PubMed PMID: 24717379
Why is this paper important?
- There is a concern among clinicians that randomized, controlled clinical trials of medications for bipolar disorder study a highly selected group of patients. On the other hand, patients seen in the real world are much more complicated and have many comorbidities
- Thus, we need more studies and more data about treatment and prognosis of real-world patients with bipolar disorder
- The study aimed to assess the subsequent course of illness in patients discharged from the hospital after admission for bipolar disorder
- 300 patients with bipolar I (n = 158) or II (n = 142) disorder who were consecutively admitted to a psychiatric community hospital were included
- After discharge from the hospital, patients were followed up naturalistically for 4 years. (“Naturalistically” means that the patients were treated by their own clinicians in the usual manner. The researchers did not modify the treatment but only collected observations.)
- Medications were chosen by the patients’ own treating psychiatrists on the basis of their clinical judgment
- Prescribed medications included lithium, carbamazepine, valproate, lamotrigine, antidepressants, and atypical antipsychotics
- 68% of the patients relapsed within 4 years
- The number of days to the next affective episode varied widely; the mean was 208 days
- In both bipolar I and bipolar II disorders, the new episodes of illness correlated with the original (index) episode. For example, for patients who were discharged after an episode of mania, the next episode tended to be that of mania as well
- Patients relapsed statistically significantly sooner when prophylactic medication was either changed by the psychiatrist or stopped by the patient
- These data confirm a high risk of relapse in the real world
- The study’s data provides a warning that we need to be very cautious when reducing or changing medication. Even though the changes may not be the cause, it was found that changes made by clinicians are associated with higher relapse rates
- Of course, these were patients who needed hospitalization and were sicker than others who did not need hospitalization
- Nevertheless, the important message of this study, in my opinion, is for us to realize that things are not going well and that we need to do a lot more in trying to reduce relapse rates in bipolar disorder
- We need to distinguish between “relapse” into the same episode and “recurrence” or a new episode
- Other studies, reviewed in the article, have all similarly found high rates of recurrence.
- One of the most important issues is that we need to make specific and energetic efforts to improve adherence to the medications. Details of how to actually do that are beyond the purview of my discussion here. Briefly, some of the things we need to do are: discuss a relapse prevention and early detection strategy repeatedly, involve family members in the prevention plan, use mood charting, actively treat adverse effects, include psychotherapy to reduce the effect of stress, provide substance abuse treatment, use higher doses when needed, and use combinations of mood stabilizers where appropriate
- When making changes to the medications, we should become alert to the possibility that the patient may become ill. Making changes more slowly, cross tapering medications over many months if possible, increasing frequency of visits, and involving the family in monitoring the patient are some of the measures that should be considered
- We need more data about comparisons between different monotherapies and different drug combinations in order to choose the best options for reducing relapse and recurrence
Carvalho et al.
J Clin Psychiatry. 2014;75(6):e578-86
Why is this paper important?
- Rapid cycling in bipolar disorder is common and has an important bearing on the treatment and prognosis of the disorder
- Therefore, the existing data on this issue needs to be combined and summarized to make it easier for clinicians to become better informed about it.
- Rapid-cycling bipolar disorder refers to the presence of at least four mood episodes in a twelve month period
- The purpose of this study was to synthesize data regarding prevalence, clinical correlates, and familial/genetic aspects related to rapid cycling in bipolar disorder
- The authors searched the MEDLINE database through September, 2013 for articles regarding rapid cycling in bipolar disorder
- In addition, they also looked at the reference lists of previous review articles on the topic
- Only articles in English were included
- The final review included 119 articles
- Two investigators independently reviewed the articles for eligibility (Good. This reduces bias of any one individual). Final decisions regarding eligibility were made by consensus
- Rapid cycling affects a significant proportion of bipolar patients
- Lifetime prevalence of rapid cycling ranges from 26% to 43%
- One year prevalence of rapid cycling ranges from 5% to 33%
- Rapid cycling is related to: a) a longer course of illness, b) an earlier age at onset, c) more drug and alcohol abuse, and d) increased suicidality
- Overall, rapid cycling is associated with a worse outcome of the illness
- Etiology of rapid cycling is unclear
- Rapid cycling clearly seems to be associated with female gender and with substance use but the reasons for these associations are not clear. (Don’t assume that an association means that one is causing the other.)
- On the other hand, contrary to popular belief, the association of rapid cycling with hypothyroidism and with use of antidepressants is not well proven. The studies are contradictory and there are plausible alternative explanations for the apparent associations.
- Most studies show that rapid cycling does not run in families.
- The long-term prognosis of rapid cycling varies greatly
- Rapid cycling is often a transient phenomenon rather than a stable pattern that characterizes the individual patient
- Rapid cycling is a common but under-recognized condition in bipolar disorder
- It constitutes a worsening of the primary disorder. There is no good evidence that rapid cycling represents a discrete subtype of bipolar disorder
- The authors think that early recognition of this pattern can lead to better treatment strategies and improvement of the long-term course.
- This paper emphasizes that we need to pay more attention to the problem of rapid cycling in patients with bipolar disorder because it is common and is associated with significant adverse outcomes
- About one third of patients with bipolar disorder will develop rapid cycling at some time in their lives. About half of those, i.e., about one sixth of patients, will have rapid cycling during a given year
- In assessing patients with bipolar disorder, we should always watch for the number of episodes in the most recent 12 months
- A new episode is separated from a previous episode by two months of significant improvement (not meeting criteria for that episode). However, it is important to note that switch to an episode of the opposite polarity constitutes a new episode even when there is little or no interval of wellness.
- What should clinicians do? Strongly consider tapering off any antidepressant in a patient with rapid cycling. I recently treated a woman who was switching back and forth about 24 times a year while on venlafaxine. Tapering off the venlafaxine had a dramatic effect on reducing the switching, though, of course, the illness still had to be treated with mood stabilizers
- Secondly, check for hypothyroidism and for substance abuse
- Thirdly, keep in mind that lithium may be less efficacious for patients with rapid cycling and a switch to another mood stabilizer should be considered
Sachdeva et al.
Alcohol Alcohol. 2014;49(3):287-291.
Why is this paper important?
- Detoxification from alcohol is a very common clinical problem both in psychiatric and nonpsychiatric settings
- We need to know which regimens are safe and simple to implement, yet can reduce hospital stay
- There is some prior data suggesting the utility of symptom-triggered regimens of lorazepam for alcohol detoxification
- This was described as a randomized, double blind controlled trial (However, it was not a truly blinded trial because no placebo pills were used. Thus, the patients knew how much medication they were getting, just not why they were getting a certain dose at that time.)
- Sixty-three consecutive consenting male patients admitted with diagnosis of uncomplicated alcohol withdrawal were included
- Patients were randomized into two groups: symptom-triggered (n = 33) and fixed tapering dose regimen (n = 30)
- Alcohol withdrawal symptoms were rated on Clinical Institute Withdrawal Assessment – Alcohol- revised scale (CIWA-Ar)
- The fixed tapering dose started with 2 mg/day, 4 mg/day, and 8 mg/day for patients with mild, moderate, or severe withdrawal, respectively, on the CIWA-Ar
- For the symptom-triggered group, CIWA-Ar assessment was done every few hours (frequency determined by the severity of withdrawal). When the score was less than 8, no lorazepam was given. Each time the score was higher, 2 mg of lorazepam was given orally.
- The main outcome measures were the total amount and duration of lorazepam treatment and the incidence of adverse events or complications
- These were quite significantly alcohol dependent patients as indicated by moderate or severe withdrawal on the CIWA-Ar in the vast majority of patients in both groups
- The mean lorazepam dose that the symptom-triggered group received was significantly lower than in the fixed tapering dose group (9.5 mg versus 19.9 mg)
- The symptom-triggered group also received lorazepam for a significantly shorter duration of time (mean of 47.8 hours versus 146 hours).
- These advantages of symptom-triggered treatment applied to both patients with moderate withdrawal and with severe withdrawal
- There were no significant differences between both the groups in terms of the incidence of complications like seizures or delirium tremens. One patient in the fixed tapering dose had a seizure and none in the symptom-triggered group. One patient in each group developed delirium.
- Symptom-triggered lorazepam treatment for alcohol withdrawal had advantages over use of a fixed tapering dose and was just as safe
- As the authors noted, a symptom-triggered regimen can avoid both under- and over-medicating the patient since the benzodiazepine dose given is adjusted based on systematically assessed severity of withdrawal symptoms
- The symptom-triggered approach, of course, requires trained staff and more frequent assessment
- A combination of a lower dose fixed taper AND symptom-triggered additional doses could be another option.
Herring et al.
Biol Psychiatry. 2014
Why is this paper important?
- Suvorexant (Belsomra®) is a hypnotic that was approved by the FDA in 2014
- The two clinical trials described in this paper are the most important (“pivotal”) phase III clinical trials and are the ones that led to the FDA approval
- The orexin (also known as hypocretin) system was discovered in the late 1990s
- Loss of orexin neurons leads to narcolepsy
- The orexin system is involved in, among other functions, the regulation of wakefulness and the transition between sleep and wakefulness, with increased activity during the day and decreased activity at night
- Two randomized, double-blind, placebo-controlled, 3-month trials in patients with primary insomnia
- Patients were either nonelderly (18 to 64 years) or elderly (> 65 years)
- 1021 and 1019 patients were randomized in the two trials
- Suvorexant was given in either high dose or low dose. In nonelderly patients, these were 40 mg and 20 mg, respectively. In elderly patients, these were 30 mg and 15 mg, respectively
- At the end of the trials, there was a one-week randomized, double-blind discontinuation period to assess whether suvorexant is associated with withdrawal symptoms or with rebound insomnia
- Efficacy was assessed in two ways: patient report and polysomnography (in a subset of patients)
- Patients reported a) time to sleep onset, and b) total sleep time. This was done after one week, one month, and three months of treatment
- Polysomnography assessed a) latency to onset of persistent sleep, and b) wake after persistent sleep onset. This was done after one night, one month, and three months of treatment
- Suvorexant higher dose (40 or 30 mg) was statistically significantly better than placebo on all the subjective and polysomnographic end points, at all three assessment points, in both trials (except only one assessment at one time point)
- Suvorexant lower dose (20 or 15 mg) was also statistically significantly better than placebo on most but not all measures
- There was no significant rebound insomnia or withdrawal symptoms when suvorexant was discontinued at the end of the study
- Suvorexant improved sleep onset and maintenance over 3 months of nightly treatment
- The Abstract states the following about adverse effects: “Both doses of suvorexant were generally well tolerated, with < 5% of patients discontinuing due to adverse events over 3 months. The results did not suggest the emergence of marked rebound or withdrawal signs or symptoms when suvorexant was discontinued,” and “Suvorexant… was generally safe and well tolerated.” However, let us look deeper
- An important potential adverse event for any hypnotic is somnolence. It occurred in 5 to 8% of patients on the lower dose, 10 to 11% of patients on the higher dose, and in 3% of patients on placebo. (These are only the percentages of patients who spontaneously reported the adverse event.)
- A subset of patients did the Digit Symbol Substitution Test immediately after waking up. Patients on suvorexant did not differ from those on placebo
- Since the orexin system may be involved in weight regulation as well, it should be noted that mean changes from baseline in weight and proportions of patients with > 7% increase or decrease in weight were similar between suvorexant and placebo
- Other hypnotics have been associated with complex sleep-related behavior, so it is important to ask whether suvorexant is associated with such behaviors. Of over 2000 patients in these two trials, such behavior occurred in only one patient (parasomnia and sleepwalking) on suvorexant and none on placebo. This incidence may be similar to that in patients in the general population who are taking other hypnotics.
- Given that loss of orexin neurons is associated with narcolepsy, we are very interested in knowing whether any of the symptoms found in narcolepsy may be adverse effects of suvorexant
- Hypnagogic or hypnopompic hallucinations occurred in three patients on suvorexant and none on placebo
- Sleep paralysis occurred in four patients on suvorexant and none on placebo
- The paper does not give the number of patients who reported symptoms suggestive of cataplexy. It only states that none of the patient-reported events that could have been cataplexy were considered to be cataplexy by an expert panel. Admittedly, deciding whether what a patient reported was really an episode of cataplexy is tricky.
- The prescribing information approved by the FDA has warnings about somnolence, driving, complex sleep-related behaviors, increased depression, increased suicidal ideation, sleep paralysis, hypnagogic/hypnopompic hallucinations, and cataplexy-like symptoms
- My view is that we should neither overestimate such risks nor ignore them. It is hard to know about rare events until a drug is used in even larger numbers of patients
- It is very important to note that due to concerns about potential adverse effects with hypnotics, the FDA insisted on approving suvorexant for use at doses are much less than those evaluated in the studies: a usual dose of 10 mg with a maximum dose of 20 mg. Therefore, we may expect both efficacy and adverse effects to be lower than what was found in the clinical trials
Michelson et al.
Lancet Neurol. 2014;13(5):461-71
Why is this paper important?
- Many mental health patients need to take hypnotics for longer periods. Therefore, it is important to know about how patients tolerate the medication over a longer period
- Suvorexant is an orexin receptor antagonist shown to be efficacious for insomnia over three months
- This study aimed to assess its clinical profile during and after one year of treatment
- This was a randomized, placebo-controlled, clinical trial
- Patients: age 18 years or older, diagnosis of primary insomnia by DSM-IV-TR criteria
- 522 patients were randomly assigned to receive suvorexant, 259 to receive placebo
- Suvorexant was given every night, not as needed
- The dose of suvorexant given was 40 mg for patients younger than 65 years, 30 mg for patients aged 65 years or older
- After one year, there was a two-month randomized discontinuation phase in which patients on suvorexant either continued suvorexant or were abruptly switched to placebo
- Efficacy was assessed based on patient report. Polysomnography was not used (since the main aim of the study was to assess safety/ tolerability and not efficacy)
- Two thirds of the patients stayed in the study for the entire year (this is a pretty good retention rate for a one year study). The percentage was similar for patients on suvorexant or on placebo
- The most common adverse event was somnolence: in 13% of patients on suvorexant and 3% on placebo
- Assessed after one month of treatment, suvorexant was statistically significantly more efficacious than placebo
- Patient-reported total sleep time improved by an average of 38.7 minutes in patients on suvorexant vs. 16.0 minutes in patients on placebo
- Patient-reported time to sleep onset decreased by an average of 18.0 minutes in patients on suvorexant vs. 8.4 minutes in patients on placebo
- The authors concluded that “suvorexant was generally safe and well tolerated over 1 year of nightly treatment” and efficacious based on subjective measures
- The Abstract does not tell us what the adverse effects were. We have to go into the paper to find out more about adverse effects. The table below summarizes the commonest adverse events that occurred more often than on placebo:
- We are, of course, very interested in knowing about adverse events that may be rare but are nevertheless important. The table below summarizes those:
- So, it appears that in addition to residual somnolence and persistent excessive daytime sleepiness, we should watch for rare occurrence of suicidal ideation or sleep-related symptoms (including sleep paralysis)
- Very, very important: patients with narcolepsy were excluded in all suvorexant studies. Before giving suvorexant to a patient, we should screen for symptoms of narcolepsy including sleep paralysis, hypnagogic/hypnopompic hallucinations, and cataplexy