Carbamazepine for Acute Mania
From MindLinc Wiki
Clinical Question: What is the efficacy and safety of monotherapy with extended-release carbamazepine in patients with bipolar I disorder experiencing manic or mixed episodes?
Background Info: Carbamazepine has been long used in the treatment of bipolar disorder, however its efficacy was proven only recently in a well conducted study. Due to the limitations in response and tolerability of the available agents to treat bipolar disorder, the availability of additional options is likely to improve outcomes.
Question Type: Therapy
Validity Criteria for Appropriate Question Type:
Follow-up: 144 patients (60.3%) completed the study. All but 4 patients who entered the trial were properly accounted for and attributed at its conclusion.
Randomization: Randomized. Intention to treat: from the 239 patients, 235 were included in the intention to treat sample.
Similar groups: There were no important differences between groups.
Blinding: Double blinded.
Equal treatment: There was equal treatment aside from the intervention. Non intervention treatment: Lorazepam prn was the only additional medication allowed.
Study Design Type: Randomized Controlled Trial
Relevant Criteria for Appropriate Design:
Allocation: Randomized (no comments on how randomization was achieved).
Blinding: Double blind (no comments on how blinding was achieved).
Follow up period: 21 days, following a 5 day single-blind placebo lead in period.
Setting: 25 study sites (19 US and 6 in India). There is no comment on what kind of study sites these were - university, community, public, or private.
Patients/Population: n= 239. >18 y/o. DSM-IV criteria for bipolar disorder most current episode manic or mixed, with a prior manic or mixed episode, with YMRS score of > 20. Information was not given on how these patients were recruited. Intervention/ exposure: Extended Release Carbamazepine (beaded capsules) up to 1600 mg daily (started at 200 mg and titrated by 200 mg daily).
Outcomes: Primary outcome was change in YMRS total score. Secondary outcomes included: responder rate (defined as decrease> 50% in YMRS score), change in CGI and HAM-D scores and time to outpatient status. Patient Follow Up: 144 completed the study. 95 terminated prematurely: 4 lost to follow up, 17 adverse effects, 35 lack of efficacy, 13 protocol violations, 4 other.
Main Results: Primary outcome: Treatment group with statistically significant decreases in YMRS scores at day 7, 14 and 21 compared to placebo (Last observation carried forward analysis in ITT sample p <.0001 at all points) Figure 1. Secondary outcomes: ERC-CBZ patients had significantly higher response rates on all measured days (p<0.0001), with by my calculations a relative risk of 2.1 at the end point. CGI-S and CGI-I scales scores were significantly improved in the treatment group (p<.0001), as were HAM-D scores (p<.01). 48.3% in ERC-CBZ group vs. 34.8% in the placebo group were discharged from the hospital during double-blind treatment period (P>.05). Cohen’s effect size was 0.85 (large effect). Significantly more subjects in the placebo group discontinued because of lack of efficacy (23.1% vs. 6.6% p <.001). ERC-CBZ group had significantly higher adverse events (91.8% vs. 56.4%).
Conclusions: Overall, this was a well designed study. The validity criteria for appropriate question type were met, also the sample was big enough. However, the period of follow up was short, the participants seemed to be less sick than the usual manic population we handle, and because the study provides little information on study sites and recruitment it could be difficult to extrapolate results to our own patients. Monotherapy treatment with extended release carbamazepine was statistically significantly more efficacious than placebo in the treatment of manic or mixed episode in the first 21 days of treatment in this RCT. Although the treatment group had a significantly higher incidence of adverse events, this does not fully account for the drop outs in the treatment group, from which the authors conclude that the medication is safe and well tolerated.
Two Teaching Points: 1. Intention to treat analysis, which is analysis of outcomes based on the treatment arm to which patients are randomized, preserves the value of randomization. Analyses restricted to patients who adhered to assigned treatment can provide misleading estimate of impact. 2. The intention to treat principle is violated by excluding patients who did not begin the treatment to which they were allocated.
Two Questions for Discussion: 1. Why is carbamazepine less used if efficacy has been proven? 2. At what point would you consider carbamazepine monotherapy in your manic (or mixed episode) patients?
Presented by Carolina Aponte Urdaneta on 9/18/06
