Abstract
Discontinuation symptoms following antidepressant cessation have been reported with increasing frequency in recent years. Tricyclic antidepressants have been associated with withdrawal symptoms, believed to be caused by an excessive rebound of cholinergic activity [1–4]. There are numerous reports of withdrawal symptoms following discontinuation of selective serotonin re-uptake inhibitors (SSRIs). These have occurred more frequently with the shorter half-life SSRIs such as paroxetine and fluvoxamine than with those with longer half-life metabolites such as sertraline and fluoxetine [5–6]. Venlafaxine, a serotonin and noradrenaline re-uptake inhibitor (SNRI), has also been reported to produce withdrawal symptoms upon discontinuation [7–8].
Nefazodone is a new antidepressant marketed for the treatment of depression. It is a phenylpiperazine derivative and is structurally different from tricyclic antidepressants, monoamine oxidase inhibitors and SSRIs. Nefazodone has a distinctive pharmacological profile and exhibits potent antagonism of post-synaptic serotonin type 2 (5-HT2) receptor sites and inhibits the presynaptic re-uptake of serotonin, and to a lesser extent, noradrenaline. The efficacy of nefazodone as an antidepressant has been postulated to be a consequence of both serotonin re-uptake blockade and 5-HT2 antagonism, which may result in the facilitation of 5-HT1A neurotransmission. Blockade of noradrenaline re-uptake may also play a role. The mean elimination half-life of nefazodone is 2–4 h. Nefazodone has three active metabolites (α-hydroxy-nefazodone, desethyl-hydroxynefazodone and m-chlorophenylpiperazine) with half-lives ranging from 1 to 18 h.
We are not aware of any reports of reactions following discontinuation of nefazodone in published literature. There were no reports of adverse experiences in a controlled placebo substitution study where patients were treated with nefazodone for 16 weeks and then abruptly randomised to either placebo or nefazodone. The manufacturer does not consider it necessary to taper nefazodone prior to discontinuation (product information, Bristol-Myers Squibb, Melbourne, Australia). We would like to report a patient who experienced withdrawal symptoms following gradual cessation of nefazodone.
Case report
Ms LW, a 41-year-old woman presented with a 2-month history of depressed mood, poor sleep, poor concentration, irritability and suicidal ideation. She was started on moclobemide, which was gradually increased to 300 mg twice daily and she remained on this for 3 months. However, she had not improved completely and requested a change in medication. Moclobemide was uneventfully reduced and stopped, and after 1 week she commenced on nefazodone 50 mg twice daily. She experienced initial weight gain, headaches and poor concentration but persisted on the same dosage. When reviewed after another week, she was beginning to become depressed again but wished to continue with the medication. When reviewed at the end of the third week, she was worse with a further reduction in her mood, and an increase in her anger and irritability. It was then decided to reduce and stop nefazodone. Although the manufacturer had not recommended tapering of nefazodone upon cessation, she was asked to take 50 mg once daily for 3 days and then cease.
Five days later (2 days after she had received no medication), LW began to feel non-specifically restless and tired. On the third day without nefazodone, she felt worse and went to bed. She reported feeling markedly nauseous on the following day with a severe headache, feeling light headed with generalised aches and pains. On the fifth day she was vomiting, had loose bowel motions, and was sweating, trembling and felt that her heart was coming up into her throat. She slept poorly, experienced nightmares, and contacted her treating psychiatrist for advice. A presumptive diagnosis of nefazodone withdrawal was made. LW had disposed of all her remaining nefazodone and was hesitant about recommencing and withdrawing more slowly. She was seen by her general practitioner to exclude other causes and was prescribed 7.5 mg oxazepam for sleep. The above symptoms had reduced 24 h later, but she became frightened by what she described as the worst feeling throughout. Although she found the experience difficult to describe, she reported that her head felt itchy inside, had racing thoughts, felt anxious and hyped-up to the point that she wanted to hit her head on the wall in order to make it stop. She continued with oxazepam 7.5 mg twice daily and improved over the next 2 days. She eventually felt well again 10 days after the initial discontinuation. No further medication was prescribed.
On specific inquiry, LW recalled similar experiences when she had reduced and stopped other anti-depressants. She reported that a 2-week reduction in sertraline had been associated with nausea, sweating, loose bowel motions, dizziness, feeling hot, cold, and tremulous but that this had subsided without recourse to a health practitioner. Following the gradual discontinuation of a 4-month course of fluoxetine 20 mg, she recalled a 1–2-day period when she felt nauseous and sweaty. She was seen by her general practitioner who prescribed metoclopramide 10 mg. LW also reported feeling strung out, and hyped-up after venlafaxine 37.5 mg b.d. was discontinued after a 2-month course.
Discussion
Zajecka et al. [6] have defined an antidepressant discontinuation syndrome as the onset of a cluster of somatic and psychic symptoms following the discontinuation of an antidepressant drug, not attributable to other causes.
The emergence of new symptoms, a close temporal relationship with nefazodone cessation and gradual resolution over the following 10 days suggests that this was a withdrawal reaction [9]. Withdrawal symptoms have been reported following abrupt discontinuation, gradual dose reductions and even after missing a single dose [6–7]. Our patient developed symptoms even though she was on a low dose of nefazodone and this was tapered over 3 days. Other authors have also reported discontinuation symptoms during slow tapering as well as with low doses tapering [10–13].
Common withdrawal symptoms seen with SSRIs (paroxetine, fluvoxamine, sertraline and fluoxetine) include influenza-like symptoms, dizziness, lethargy, paraesthesia, nausea, vivid dreams, irritability and lowered mood [5]. Anxiety, anorexia, agitation, tremors, insomnia, headache, sweating, confusion and memory problems are also known to occur. The common somatic syndromes have been grouped into five clusters: (i) disequilibrium features, (ii) gastrointestinal features, (iii) flu-like symptoms, (iv) sensory disturbances, and (v) sleep disturbances. Core psychological symptoms reported include anxiety, agitation, crying spells and irritability [9]. Withdrawal symptoms reported with venlafaxine are generally similar to those with short half-life SSRIs, with headaches often a common and prominent feature [7],[8],[14].
The symptoms experienced by our patient were nausea, diarrhoea, ataxia, insomnia, marked agitation, headache and flu-like symptoms. These are similar to the clusters of somatic symptoms following SSRI and SNRI withdrawal and this suggests that a common mechanism may be involved in withdrawal reactions with these agents. Our patient had also experienced withdrawal reactions to sertraline, venlafaxine and possibly fluoxetine, indicating a probable increased sensitivity to antidepressant withdrawal effects.
Antidepressants that block the re-uptake of serotonin (such as SSRIs) are thought to result in the down regulation (or desensitisation) of postsynaptic serotonin receptors [6]. While nefazodone inhibits serotonin re-uptake, it possesses the added property of blocking postsynaptic 5-HT2 receptors. Nefazodone's antidepressant potential may derive from its 5-HT2 antagonism and 5-HT re-uptake blockade, which together may ultimately result in the facilitation of 5-HT1A mediated neurotransmission [15]. Long-term use of nefazodone results in a down regulation of 5-HT2 receptor binding sites. When discontinued, there is a restoration or even acute enhancement of serotonin re-uptake leading to a deficiency of serotonin in the synapse. The re-adaptation of these neurotransmitter systems should theoretically occur over 2–3 weeks, which correlates with improvement of withdrawal symptoms [6].
Withdrawal symptoms can occur in several common clinical situations: after routine discontinuation of treatment, after switching antidepressant treatments or in a patient who is non-compliant. Awareness of antidepressant-related withdrawal effects are important as these symptoms can be mistaken for a physical illness or an exacerbation of the original psychiatric illness. New symptoms which emerge during a switch from one antidepressant to another may be the result of side effects of the new medication, drug interactions or withdrawal effects from the discontinued antidepressant [6]. Prompt recognition of these effects is essential to institute appropriate management, including reintroduction of the drug and a more gradual tapering schedule. A recent survey reported that only 72% of psychiatrists and 30% of general practitioners were aware that patients could experience antidepressant discontinuation events [16]. Because of the possible withdrawal effects, we suggest that discontinuation of nefazodone be done gradually, by tapering doses over at least a week. Longer periods may be needed where high doses are involved and it has been suggested that tapering of the shorter half-life SSRIs may have to continue for up to several weeks [9].
Individuals may have specific genetic or psychological differences that place them at risk for discontinuation symptoms [3]. Patients who have a history of antidepressant noncompliance, who have experienced discontinuation symptoms in the past, or who have treatment-emergent anxiety are at highest risk for experiencing discontinuation phenomena [17] and will, therefore, need more gradual reductions and closer monitoring. It is suggested that patients who have previously received antidepressants be routinely asked as to whether they had experienced any difficulties at the time of change over or discontinuation. Only during the withdrawal phenomenon did it become apparent that our patient was, and presumably continues to be, at an increased risk of such withdrawal-related events.