Relapse risk is greater for rapid drug withdrawal

May 28, 2010 In patients with major depressive disorder, panic disorder, or bipolar disorder, the risk for illness recurrence is far greater following abrupt or rapid, vs gradual, discontinuation of clinically effective antidepressant treatment, according to a study published online May 17 in the American Journal of Psychiatry.

"Our findings of greater clinical risk following rapid versus gradual discontinuation of antidepressants is one of several similar findings that we have reported over the years," Ross J. Baldessarini, MD, from Harvard Medical School, Boston, and McLean Hospital, Belmont, Massachusetts, told Medscape Psychiatry.

"We found this phenomenon first with discontinuing lithium in bipolar disorder patients and later with antipsychotic drugs in schizophrenia patients. Similar risks also are well known in rapidly stopping sedative-anxiolytics and anticonvulsants (both of which can lead to epileptic-like seizures)," he added.

According to Dr. Baldessarini, it was not easy to obtain data on effects of discontinuing the use of antidepressants rapidly vs gradually because single studies involving both conditions had virtually been unknown. "To my knowledge, our report is the first to involve such a comparison with antidepressants of various types," he said.

The study involved 398 patients with a Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition) diagnosis of recurrent major depressive disorder (n = 224), panic disorder (n = 75), bipolar II disorder (n = 62), or bipolar I disorder (n = 37).

Roughly two-thirds were women (65.6%), a little more than half were married (54.8%), and most were employed (95.6%). The mean age of study subjects was 42 years, and the mean length of antidepressant therapy was 8.5 months. The use of antidepressants was discontinued rapidly during 1 to 7 days in 188 subjects (largely at the patient's request) and gradually during 14 days or more in 210 subjects.

Patients who discontinued antidepressant use rapidly were similar to those who discontinued gradually in most demographic and clinical characteristics. All patients include in the analysis were "clinically well" at the time of discontinuing antidepressant treatment and had no clinical evidence of even mild depression or anxiety, the study authors note.

According to the investigators, during an average of 2.8 years of follow-up, the latency to first new illness with rapid discontinuation was 0.4 times that with gradual discontinuation (3.62 months [SD, 8.78] compared with 8.42 months [SD, 22.2]; P < .0001).

"Notably," they report, the 3.62-month latency period after rapid discontinuation was one-fourth the estimated average previous interepisode interval in the same patients (14.9 months [SD, 21.0]; P < .0001) and about one-fifth of the mean previous interval following gradual discontinuation (16.6 months [SD, 30.5]; P < .0001).

In patients with bipolar I disorder and panic disorder, differences in illness latency between rapid and gradual antidepressant discontinuation were larger (3.1- to 4.6-fold) than among those with bipolar II disorder or recurrent major depressive disorder (2.4- to 2.5-fold). The reasons for these differences require further study, the investigators say.

The effect was similar across antidepressant classes, although the pace of discontinuation had less effect with drugs of prolonged half-life, they note.

"The general point for clinical practice," Dr. Baldessarini said, "is that it appears that most psychotropic drugs, when discontinued abruptly or rapidly, can lead to early and severe exacerbations of the illnesses being treated.

"It is my impression that this concept has been widely accepted and that clinical practice has been modified appropriately in many cases to include gradual dose tapering and slow discontinuation when feasible clinically," he added.

Optimal dose-tapering times and protocols still need to be worked out, Dr. Baldessarini noted. However, in general, he said, it is wise to taper off most psychotropics during at least several weeks.

"Nevertheless, modern psychiatry includes a great deal of trial and error in attempting to find effective treatments for individual patients, and this process can tend to encourage rapid changes in treatment, which can be risky."

The findings also have important research implications, Dr. Baldessarini said. "Many modern treatment trials involve discontinuing medications at some point, often when patients are barely recovering from an acute episode of illness," he noted. On the basis of the current results and prior research, abruptly or rapidly discontinuing psychotropic medications whether by clinical decision on the part of patients or clinicians or by experimental design "can present substantial clinical and ethical problems," the investigators note in their report.

The study's strengths include its large sample size with a range of disorders for which antidepressants are prescribed, with consistent assessments, treatment, and follow-up and the inclusion of different antidepressant drug types. The lack of randomized assignment to precisely scheduled dose tapering options is a major limitation.

Other limitations include uncontrolled, and often patient-determined, drug therapy discontinuation pace, as well as the possibility that some patients reported inaccurate drug therapy discontinuation data or became ill after loss to follow-up.

Dr. Baldessarini and colleagues say their findings "underscore the importance of warning patients that abrupt discontinuation of antidepressant treatment can lead not only to early adverse physiological [withdrawal] responses but also, over several months, to a return of the illness being treated."

The study was supported in part by the National Institutes of Health. Dr. Baldessarini and several coauthors report being consultants or receiving research support from several pharmaceutical companies that make antidepressant medications.

Am J Psychiatry. Published online May 17, 2010.
[CLOSE WINDOW]
Authors and Disclosures
Journalist
Megan Brooks

The measure of achievement lies not in how high the mountain,
but in how hard the climb.

The measure of success lies only in how high one feels he must
climb to get there.

Relapse risk is greater for rapid drug withdrawal

Thread

Post a new follow-up

Notify the administrators

Start a new thread