Abstract
Mental health and smoking redux
Lawrence and Mitrou claim that I made a “basic statistical error” in my deduction that 3.3% of adult Australians have both mental illness and are daily smokers because I used “proportions … calculated from two disjoint populations (i.e. people with no long-term mental illness, and people with a long-term mental illness” and that because these have “separate denominators” I was thus incorrect in assuming that they could be used in the one calculation [1, 2].
They are mistaken in this assertion. Like Access Economics, I sourced the data on smoking by those with mental illness from the 2004 National Health Survey, which, in questioning the same sample population and therefore using the same denominators, reported that 10.7% of adults had mental health problems and that (p.8) “Among adults who reported a mental or behavioural problem, 32% were daily smokers,” (the figure in the table was actually 31.8%) [3]. Thus, 31.8% of 10.7% of all adults means 3.4% of all adults who smoke daily and report “a long term mental or behavioural problem”.
The Australian Bureau of Statistics (ABS) 2007 National Survey of Mental Health and Well-being survey [4] – published after my original letter was accepted for publication [2] – shows that 20% (3 197 800) of 16 105 300 Australian adults reported any mental health problem in the past 12 months. Of these, 1 156 600 (36%) were current smokers. Among the 12 817 500 without mental health problems 2 410 100 (18.8%) reported current smoking. Using the definition of mental illness used in the 2007 survey, smokers with any recent mental illness thus represent 32.4% of all 3 566 800 smokers. While the 2007 ABS study apparently asked respondents to differentiate daily smoking from current smoking (i.e. daily plus less-than-daily smoking), only the latter is reported, thus further inflating the number said to be smoking.
A key difference between the 2004 National Health Survey and the 2007 Mental Health Survey was the way in which mental health problems were defined, with the near doubling in prevalence in 3 years of those defined as having mental health problems in the past 12 months (10.7% vs 20%), suggesting that the 2007 survey allowed far wider latitude in what was defined as a mental health problem. The 3 197 800 people with mental health problems include 2 303 000 cases of ‘any anxiety disorder’, 1 031 000 with post-traumatic stress disorders and 470 100 reporting ‘harmful’ alcohol consumption, as distinct from alcohol dependence. With the debate in full swing over the penchant for elements in psychiatry to medicalize sometimes ordinary human problems into treatable disorders [5], some will be comfortable with the idea that every fifth Australian adult has a ‘mental health problem’. Many others will remain circumspect.
There is a theme in much commentary about smoking and mental illness that implies that smoking cessation in this group is somehow much more difficult to achieve than that in those without mental health problems. It is often argued that specially targeted cessation support programmes are needed. There is some evidence, however, that although those with mental health problems take up smoking at higher rates than those with no mental illness, their cessation rates are broadly similar to those in the wider community [6]. Indeed, the 2007 ABS study shows that 24.2% of those with mental health problems are former smokers, while 27.6% of people with no mental health problems are ex-smokers: not wildly different figures [4]. Where the big difference lies is in the quit ratios between the groups (the percentage of ever smokers who once smoked but no longer do so). The quit ratio in those with recent mental illness is 40.1%, while it is 59.5% in those reporting no recent mental health problems. This large difference is mostly explained by the far smaller proportion of people with mental illness who have never smoked (39.7%) compared with 53.6% of those without such recent illness. There would thus seem to be a stronger case for arguing that efforts to prevent uptake in young people suffering or at risk of developing mental illness are more important than specially targeted cessation programmes. Those with mental illness may be responding to Australian cessation policies and mass reach interventions as well as the broader community.
Finally, Lawrence and Mitrou do not appear to have read the report of the Preventative Health Task Force very carefully [7]. Two of the 18 recommendations (numbers 12 and 15) explicitly address smoking and mental health, and most others concerning population-wide policies will also impact on those with mental health problems. At pages 42–43 there is an entire section addressing the issue.
Magnesium intake and depression
We commend Jacka et al. for their attempt to link dietary intake to psychiatric disorders of high prevalence [1]. As they correctly point out, psychiatry is in need of population-based primary prevention strategies. The difficulty it seems is the complex interactions and pathways by which magnesium and other nutrients may work. Several authors believe that magnesium may be used therapeutically for premenstrual syndrome [2, 3]. Others believe it may assist in a broader group of disorders such as asthma, cardiovascular disease, diabetes, migraine and even osteoporosis [4].
In addition to the pro-inflammatory hypothesis discussed in the paper, magnesium is required for the conversion of thiamine (vitamin B1) to its active form, is an essential co-factor in the Krebs cycle and plays a role in calcium homeostasis. Magnesium may be depleted by medications (such as loop diuretics and corticosteroids) and by the excess use of alcohol, caffeine or soft drinks (which is common in the psychiatric population) [3]. The multitude of potential mechanisms of actions and interactions highlights the complexity of research in this area. Furthermore, serum magnesium is a poor marker of magnesium status. Like calcium, only approximately 1% is found in extracellular fluid. This would explain inconsistencies in research results attempting to evaluate the relationship between serum deficiency and major depression [5]. When considering similar lines of research involving folate and omega-3 fatty acids, one cannot underestimate the role of good nutrition and healthy lifestyle. Perhaps future research should prospectively assess mental health outcomes associated with educational programmes emphasizing such holistic approaches.
Tranylcypromine dependence and withdrawal
Monoamine oxidase (MAO) inhibitors are used in the management of psychiatric as well as some medical conditions. Although MAO inhibitors are generally presumed to be free of addictive potential, there have been at least 15 cases of abuse reported, the majority of these involving tranylcypromine [13–15], and some describing withdrawal symptoms [4]. Tranylcypromine has a similar chemical structure to amphetamine and has a similar serotonin-releasing potential to amphetamines. Tranylcypromine abuse may present with euphoria and hyperactivity; confusion, anxiety, paranoia and insomnia may suggest withdrawal.
A 70-year-old single Caucasian man was referred for admission following increasing agitation, manic presentation and psychotic symptoms. His history suggested that while being managed for a medical concern, he had ceased tranylcypromine, which he had been abusing in doses up to 300 mg for approximately 20 years. Mental state examination on admission indicated expansive mood and rapid dysarthric speech that was not pressured. Investigations indicated thrombocytopenia, and computed tomography showed diffuse calcification of basal ganglia.
Surprisingly, his diet included significant amounts of tyramine-rich food including sausages, salami, processed meat and beer; history did not suggest any episodes of hypertensive crisis. Collateral history suggested that he had been taking excessive amounts of tranylcypromine and used to ‘prescription shop’, by using different names. Further investigation ascertained that there was a significant overlap in his attendance at various clinics, and he was probably not seen by any practitioner for 2 years. He was also known to abuse benzodiazepines, but his symptoms were not consistent with benzodiazepine withdrawal because there were no perceptual disturbances or sensory distortion. History, supporting information and monitoring ruled out any suspicion of psychotic disorder, factitious disorder or Munchausen's syndrome.
Considering the chemical similarity of tranylcypromine to amphetamines, he was placed on an amphetamine withdrawal scale, and given a sliding scale regimen of oxazepam. His mental state stabilized, his thrombocytopenia normalized, but his rapid dysarthric speech went unexplained. He was discharged to local services following advice regarding potential adverse effects of continued abuse.
It is likely that the patient initially exploited the tranylcypromine to seek a euphoric state, because he had been taking the tranylcypromine and other drugs of abuse potential for a number of years. His knowledge of the hypertensive crisis phenomena and continued consumption of tyramine-rich foods may suggest that he was getting a high from them, concurrently developing a tolerance to amphetamines.
His rapid dysarthric speech continues to remain unexplained. Familial idiopathic basal ganglia calcification presents with paranoid delusions, rapid but non-pressured speech and abnormal deposits of calcium in the basal ganglia and cerebral cortex. The syndrome, however, has also been described with other common features that were not evident in the present patient [5].
It is interesting to note that Medicare Australia, which has taken over the services of the Health Insurance Commission, no longer provides the Doctor Shopping Program. This has now been replaced by the Prescription Shopping Program, which provides information regarding Pharmaceutical Benefits Scheme items supplied to patients, but does not actually restrict collection from chemists.
Serotonin syndrome induced by duloxetine
Serotonin syndrome is an adverse reaction due to the overstimulation of serotonin receptors within the central nervous system. It is characterized by mental state changes, autonomic instability and neuromuscular abnormalities [1]. All selective serotonin reuptake inhibitors (SSRIs) and venlafaxine have been implicated in serotonin syndrome [1]. The following case describes serotonin syndrome primarily caused by duloxetine, a serotonin-norepinephrine reuptake inhibitor.
A 70-year-old woman with a 35 year history of major depressive disorder and generalized anxiety disorder was admitted to hospital from a residential facility, with an acute relapse, characterized by depressed mood, suicidal thoughts, severe anxiety and withdrawal from her usual activities. She had been well for 1 year on sertraline 200 mg day−1, sodium valproate 400 mg day−1 and ziprasidone 40 mg day−1. She has had several trials of medications with varying degrees of success, including multiple tricyclic antidepressants, monoamine-oxidase inhibitors, SSRIs, venlafaxine and mirtazapine. There had been augmentation trials with therapeutic doses of lithium carbonate, sodium valproate, numerous typical and atypical antipsychotics, and several courses of electroconvulsive therapy. There had been no known history of serotonin syndrome. Sertraline and ziprasidone were ceased during the first week of admission followed by an 8 week trial of escitalopram, reaching a dose of 30 mg day−1, without benefit.
Sodium valproate was changed to lithium carbonate (500 mg day−1, serum level of 0.2 mmol L−1). Her escitalopram was decreased and ceased over 3 days and duloxetine 60 mg day−1 commenced 24 h after her last dose of escitalopram (10 mg). Within 48 h the patient developed marked tremors, agitation, hypertonia, hyperreflexia and diaphoresis. Her temperature (35.8°C), heart rate (88 b.p.m.), blood pressure (160/61), respiratory rate (40 min−1) and oxygen saturation (93% on air) were not clinically significant. Neurological examination indicated increased tone and brisk reflexes on both upper and lower limbs, with normal power and intact sensation. Inducible clonus was difficult to elicit due to her tremors. There was no clear evidence of focal neurology or myoclonus. She had a mildly raised white cell count (13.0×109/L; reference range = 3.5−11.0 109/L), which was inconclusive for infection. Urinalysis and chest X-ray were unremarkable. Her duloxetine was ceased with rapid resolution of her symptoms over 24 h. Duloxetine 60 mg day−1 was reintroduced with a re-emergence of the aforementioned symptoms after one dose. A diagnosis of serotonin syndrome was made and the medication ceased indefinitely. The patient consented to publication.
The signs of tremors, severe agitation, hypertonia, hyperreflexia and diaphoresis fulfilled both the Hunter and Sternbach criteria for serotonin syndrome [2, 3]. There have been recent reports of serotonin syndrome with duloxetine following a rapid switch from moclobemide [4] and with other serotonin-enhancing medications [5, 6]. A Medline and PsychInfo search to December 2008 found no reports of serotonin syndrome with duloxetine alone.
To date this is the first published case of serotonin syndrome with either duloxetine alone or in combination with lithium carbonate. In our opinion the latter is unlikely given that the patient had previously been on high doses of serotonergic antidepressants with lithium carbonate. Furthermore, her symptoms ceased when duloxetine was stopped and they re-emerged when there was an attempt to restart duloxetine. The case highlights that serotonin syndrome occurs with newer serotonergic antidepressants, such as duloxetine, at standard doses, even in patients who have previously tolerated high doses of serotonergic antidepressants at high doses.