A review of the suitability of duloxetine and venlafaxine for use in patients with depression in primary care with a focus on cardiovascular safety,suicide and mortality due to antidepressant overdose

Abstract

Depression and anxiety disorders are among the most common disorders treated by general practitioners (GPs) in the UK. Since both disorders are associated with a significantly increased risk of suicide, including with antidepressant overdose, the safety of antidepressants in overdose is of paramount importance. Numerous updates relating to antidepressant safety have been issued by regulators in the UK which may have eroded GP confidence in antidepressants. Venlafaxine, a serotonin nor adrenaline reuptake inhibitor (SNRI) had primary care prescribing restrictions placed on it in 2004 due to concerns about cardiotoxicity and mortality in overdose. Although a review of the evidence led to a reversal of the majority of restrictions in 2006, evidence suggests GPs may still be cautious in their prescribing of venlafaxine and possibly other SNRI antidepressants for patients with depression and anxiety disorders. This paper reviews the evidence pertaining to the safety of SNRI antidepressants from a perspective of cardiovascular safety and overdose. The currently available evidence suggests a marginally higher toxicity of venlafaxine in overdose compared with another SNRI duloxetine and the selective serotonin reuptake inhibitors (SSRIs), although this may be related to differential patterns of prescribing in high-risk patients. Based on this review SNRIs have a positive risk benefit profile in the treatment of depression and generalized anxiety disorder in primary care, especially as second-line agents to SSRIs.

Keywords

cardiovascular depression duloxetine mortality primary care suicide venlafaxine

Introduction

In England the adult psychiatric morbidity survey of 2007 found the weekly community prevalence of a depressive episode to be 2.3% and that for generalized anxiety disorder (GAD) 4.4% [Bebbington et al. 2009]. It is therefore not surprising that depression and anxiety are two of the most common disorders managed by general practitioners (GPs). The vast majority of depression is treated solely in primary care with only about one in four or five patients with depression referred to secondary mental health services [NICE, 2010]. In the adult psychiatric morbidity survey 24% and 13% of people with depression and GAD respectively had spoken with their GP in the last 2 weeks and 65% and 52% within the last year. This results in a considerable burden on GPs with a UK primary care survey finding that 7.2% of consecutive consultations were for probable depression [Ostler et al. 2001]. In the UK there has been a significant focus on improving treatment and services for people with depression and anxiety disorders, with the National Institute for Health and Clinical Excellence (NICE) recently updating their guidance on the management of depression [NICE, 2010] and GAD [NICE, 2011]. Much of this focus has been on increasing the availability of psychological services, but pharmacological therapies are still a core component of treatment for both conditions, particularly for more severe presentations or when psychological treatments have been ineffective.

The mainstay of pharmacological treatments for depression and GAD are antidepressants, and in the UK 95% of all prescriptions for antidepressants are issued by GPs [Henry, 1993; NICE, 2010]. In the UK, prescribing of antidepressants in primary care has been increasing steadily over the last decade, which has been explained mainly by an increase in the proportion of patients receiving long-term treatment [Moore et al. 2009]. In England during the 3 months to September 2011, over 11 million prescriptions for antidepressants were prescribed in primary care [National Health Service Business Services Authority, 2011] making them one of the most commonly prescribed classes of drugs [National Health Service Information Centre for Health and Social Care, 2010]. Selective serotonin reuptake inhibitors (SSRIs) are recommended by NICE as first-line pharmacological treatment for both depression and GAD and the dual action antidepressants serotonin and noradrenaline reuptake inhibitors (SNRIs) are among the second-line treatments [NICE, 2010; 2011]. Drug safety is one of the key considerations when prescribing in primary care and this is particularly applicable in conditions such as depression and anxiety when there is a risk of suicide by antidepressant overdose among other means and the disorders are often accompanied by other psychiatric and medical comorbidities. However, the safety of antidepressants has been called into question over the last decade. The UK Medicines and Healthcare Products Regulatory Agency (MHRA) has issued guidance on a number of safety issues, including suicidal behaviour [MHRA, 2007], discontinuation reactions [MHRA, 2004], use in pregnancy [MHRA, 2010a, 2010b], risk of fracture [MHRA, 2010c], cardiotoxicity and toxicity in overdose [MHRA, 2004, 2011] related to antidepressants. These warnings have applied to some of the most commonly prescribed antidepressants in primary care. For example, the most recent guidance regarding QT prolongation [MHRA, 2011] relates specifically to citalopram, the most commonly prescribed SSRI in the UK [National Health Service Business Services Authority, 2011], and its related compound escitalopram. These recommendations are potentially alarming for GPs and patients and may undermine the confidence in antidepressants in general. However, it is important that the safety data relating to antidepressants are fully understood in the context of treating depression when, for many patients, several different antidepressants have been prescribed before an effective treatment is found. One important group of antidepressants for which there have been safety concerns in the past is the SNRIs. There is evidence to suggest that the confidence of GPs in the safety of the SNRI venlafaxine may have been eroded following the MHRA urgent safety restriction (USR) issued in 2004 [MHRA, 2004] after which prescribing rates diminished over the next year [McAllister-Williams et al. 2006]. Venlafaxine has been licensed for the treatment of depression in the UK since 1994 and had been increasingly prescribed by GPs and specialists as a second-line treatment for depression. The USR related to potential toxicity in overdose and it restricted the initiation of venlafaxine to mental health specialists and gave new contraindications in patients with heart disease. In 2006, after a reexamination of the safety evidence and taking into account new epidemiological data, the MHRA released updated guidance on the prescribing of venlafaxine [MHRA, 2006]. This new guidance removed the recommendation for specialist initiation of venlafaxine along with some of the contraindications, and requirement for baseline electrocardiogram (ECG) monitoring, again allowing its initiation by GPs in primary care. Despite this change of guidance in 2006, prescribing of venlafaxine in primary care has since remained static [Ilyas and Moncrieff, 2012], and against a background of increasing antidepressant use, it may be conjectured that GPs are still wary of prescribing venlafaxine in primary care. In 2005 a second SNRI, duloxetine, was licensed for the treatment of depression. We speculate that many GPs may take the view that, as an SNRI, this may also carry a higher risk of cardiotoxicity or other toxicity than other antidepressants.

This paper aims to review the data examining mortality associated with overdose of venlafaxine and duloxetine, including the data examining suicidality and cardiovascular safety. Based on this evidence, recommendations can then be made as to their suitability for use in primary care from a perspective of these major safety issues. The review will take the following structure:

(1) Mortality due to overdose of venlafaxine and duloxetine. The evidence of how these data are synthesized and analyzed using the fatal toxicity index (FTI) will be reviewed. This will be followed by a review of deaths from overdose using case series which are easy to understand and give further information on the safety of antidepressants.

(2) A brief review of cardiovascular safety of the SNRIs will be presented.

Mortality due to overdose of venlafaxine and duloxetine

Measuring overdose mortality for antidepressants is relatively straightforward in the UK, but interpreting the data is difficult due to a number of factors. The simplest way of assessing overdose mortality is to count the deaths of those who have overdosed on antidepressants [available from the Office of National Statistics (ONS) who are informed via the coroners]. The number of deaths is then divided by the number of prescriptions dispensed for each antidepressant (which in England and Wales is provided by the Department of Health) to give a FTI, which is usually quoted as the number of deaths per million prescriptions dispensed [Henry, 1989]. Several analyses using a FTI calculated by this method have been published, such as those by the ONS [Morgan et al. 2004] and Buckley and McManus [Buckley and McManus, 2002]. The results of these analyses are consistent with each other. The ONS data give a FTI of 43 deaths per million prescriptions for tricyclic antidepressants (TCAs), 4.3 for SSRIs and 17.6 for venlafaxine. Buckley and McManus found a FTI of 34.8 for TCAs, 1.6 for serotonergic drugs and 13.2 for venlafaxine [Buckley and McManus, 2002]. It was the ONS data, with a FTI for venlafaxine being four times that for the SSRIs, which was one of the concerns of the MHRA when the USR on venlafaxine was imposed in 2004. The analysis above, however, is simplistic. There is more to a FTI than the direct toxicity of the drug, and other considerations such as patient factors (e.g. severity of depression, history of self harm, other drugs involved in overdose) and even whether the antidepressant might itself increase suicidality can affect the FTI. These factors are shown in Figure 1.

Figure 1.

Patient and possible drug factors associated with suicide attempts and fatal antidepressant overdose in depression.

Additionally other factors may contribute to potential bias in this type of data. For example, the indication for the antidepressant dispensing data is not recorded and therefore it is likely some patients may have been taking the antidepressant for conditions other than depression, which may have differing inherent risks of suicide and potential for drug overdose. For example, it was found that 30% of antidepressants prescribed from a sample of 151 general practices in the UK were not prescribed for depression [Lawrenson et al. 2000]. In addition, coroners report antidepressant information voluntarily and only if they consider the antidepressant contributed to the cause of death [Morgan et al. 2004].

In order to better understand the FTI the following will be reviewed in turn:

Patient factors (e.g. severity of disease and multiple concurrent medication at overdose).

Drug factors (e.g. emergence of suicidal thoughts and inherent toxicity).

Patient factors

Most GPs will be aware that the initial antidepressant treatment of depression is with an SSRI, in line with NICE guidelines. Only patients who have a poor response, fail to reach full remission or have more severe depression may go on to receive another drug such as venlafaxine or, more recently, duloxetine. As might be anticipated, the patients who were treated with venlafaxine in the ONS data were found to have a higher burden of suicide risk factors than those prescribed SSRIs [Mines et al. 2005]. To try to compensate for the inherent risk factors using data from the UK General Practice Research Database (UKGPRD), Rubino and colleagues adjusted the hazard ratios for venlafaxine versus fluoxetine, citalopram and dothiepin (dosulepin) using proxies for severity of depression such as previous suicide attempts [Rubino et al. 2007]. Once these confounders had been adjusted for, the hazard ratios were substantially reduced [from 2.85, 95% confidence interval (CI) 1.37–5.94 to 1.63, 95% CI 0.74–3.59 for venlafaxine versus fluoxetine]. However one can only adjust for those factors that can be measured and major risk factors such as hopelessness, impulsivity, abuse, unemployment and social isolation were not measured and thus not adjusted for, meaning further confounders may still have been present in the data. Further evidence for the channelling of venlafaxine use towards patients with a higher risk of suicidal behaviour has been published using data from three primary care trusts (PCTs) in the UK [Bergen et al. 2010] and in an Australian study [Chan et al. 2010]. The MHRA has also concluded that the increased FTI is at least partially contributable to these patient factors [MHRA, 2006].

Drug factors

Drug factors can be divided into two main considerations: those involving drug-induced emergence of suicidal thoughts and behaviours, and the toxicity of individual drugs.

Emergence of suicidal thoughts

There is evidence of a small increase in suicidal behaviour in the first month after starting an antidepressant [Jick et al. 2004]. A recent review for the World Psychiatric Association has concluded that antidepressants carry a small risk of inducing suicidal ideation and behaviours in people under 25 years, although this risk reduces in those aged between 30 and 40 years [Moller et al. 2008]. There are data available on the emergence of suicidal thoughts and behaviours specific to duloxetine and venlafaxine use. Acharya and colleagues compared incidence of suicide-related events with duloxetine versus placebo in controlled trials, using Mantel–Haenszel incidence difference methods [Acharya et al. 2006]. They found no evidence of increased risks of suicidal behaviours or ideations during treatment with duloxetine compared with placebo in patients with major depressive disorder. Enstuah and colleagues found in an 8-week study that fewer patients on venlafaxine than on SSRIs developed emergent suicidal thoughts, as shown in Figure 2 [Enstuah et al. 2001]. In a recent person-level analysis of all sponsor-conducted randomized controlled trials of fluoxetine and venlafaxine, both treatments were found to decrease suicidal thoughts and behaviours compared with placebo in adult patients and older patients, although no difference was found in young patients [Gibbons et al. 2012]. This was mediated through decreases in depressive symptoms through treatment. The data therefore do not support an association of either duloxetine or venlafaxine with increased suicidal ideation or behaviours as a mechanism for increasing the frequency of poisonings with these agents. However, as suicidal thoughts and behaviours may still occur in individuals we would still recommend that patients who are started on an antidepressant should be monitored for the emergence of suicidal thoughts in line with good medical practice and the summary of product characteristics.

Figure 2.

Emergence of suicidal ideation in adults during treatment for depression with selective serotonin reuptake inhibitors (SSRIs), venlafaxine and placebo [Entsuah et al. 2001].

Inherent toxicity of an antidepressant in overdose

One way of assessing the inherent toxicity of a drug in overdose is to calculate case fatality rates. This method compares the rate of death from self-poisoning with the rate of non-fatal self-poisoning; that is, the proportion of people who die from overdose of a particular agent [Rose and Unis, 2000]. In a recent study case fatality indices were calculated for venlafaxine, TCAs and SSRIs using mortality rates and self-poisoning rates from three centres in the UK [Hawton et al. 2010]. Relative toxicity indices were calculated by standardizing the rate ratios to amitriptyline. The case fatality (mortality to self-poisoning ratio) for venlafaxine was found to be 2.5 (95% CI 2.0–3.1), for SSRIs 0.5 (95% CI 0.4–0.7) and for TCAs 13.8 (95% CI 13.0–14.7). The relative toxicity index to amitriptyline was reported as 0.29 for venlafaxine, 0.06 for SSRIs and 1.6 for all TCAs. Therefore, the case fatality rate for venlafaxine was found to be substantially lower than that for TCAs but still greater than that for SSRI antidepressants. However, the authors listed several possible limitations of these data which may have inflated the case fatality rates for venlafaxine. First, nonfatal self-poisonings not presenting to the hospitals could not be accounted for and therefore case fatality rates may be overestimated, although this probably applies equally to all drugs in the study. In addition, size of overdose was not considered and the indication for which the antidepressant was prescribed was not taken into account. This may affect case fatality rates if some drugs are used more frequently for conditions for which lower doses are often prescribed and overdoses may therefore be smaller. Venlafaxine has been previously demonstrated to be preferentially prescribed to patients with longer-term psychiatric disorders and a history of self-harm [Bergen et al. 2010; Chan et al. 2010; Mines et al. 2005] and therefore it is conceivable that this fact may increase the case fatality rate for venlafaxine in this particular study relative to SSRIs if larger overdoses are taken. However, it should be noted that there is currently no direct evidence that patients take larger venlafaxine quantities. In addition, data from the Edinburgh Poisons Unit indicate that admissions due to citalopram, mirtazapine or venlafaxine overdose involved around 15–16 times the defined daily dose for all three agents [Waring et al. 2010].

A US study has used a similar method. A hazard index was calculated for each antidepressant using US poison control data (number of major or fatal outcomes per 1000 reported antidepressant ingestions) [White et al. 2008]. There were 5510 overdoses for which venlafaxine alone was ingested but only 12 (0.22%) were fatal and cardiac conduction disturbances were seen in only 2% of all cases. For duloxetine the data were more limited, with just 36 overdoses and no fatalities. The hazard indices were 27 for venlafaxine, 0 for duloxetine, 27 for citalopram, 22 for fluvoxamine and 6 or less for the other SSRIs. Since no data were available on indication for which the drugs were prescribed and there are likely to be other non-fatal overdoses not reported to US Poison Centres, then the same limitations as in the data reported by Hawton and colleagues apply [Hawton et al. 2010].

In the UK, case fatality rates can also be calculated by examining the overdose reports from the Adverse Drug Reactions Online Data Tracking (ADROIT) database at the MHRA. Case fatality rates with overdose calculated from these data are approximately the same for venlafaxine as with the other SSRIs; see Table 1 [MHRA, 2012]. As outcomes of drug overdose are not systematically collected (although once received by either a company or the MHRA they will be formally processed and reported) the information obtained from this type of analysis, although useful in helping to identify possible safety issues with medicines, has the same obvious limitations (such as under reporting) as the previous datasets and therefore cannot be used alone to base conclusions on the safety and risks of medicines.

Table 1.

Case fatality rates from overdose for selective serotonin reuptake inhibitors and venlafaxine from the Adverse Drug Reactions Online Data Tracking database [MHRA, 2012].

Drug	Citalopram	Paroxetine	Venlafaxine	Duloxetine	Sertraline	Fluoxetine
Number of reports with a fatal outcome with overdose	16	13	17	1	5	8
Total number of overdose reports	133	236	160	24	52	104
Case fatality rate for overdose	12.0	5.5	9.4	4.2	9.6	7.7

Data compiled from drug analysis prints of single active constituent overdoses and associated fatalities. Case fatality rate for overdose calculated by dividing number of fatal overdose reports by total number of overdose reports.

Thus as demonstrated above the FTI is confounded by many factors and is not a reliable way of estimating relative toxicity of antidepressants. When the confounders are adjusted for, venlafaxine appears to have similar or only slightly elevated toxicity compared with the SSRIs. The fact that the FTI is an unreliable estimate for toxicity is also demonstrated by the findings with nortriptyline, which was found to have a FTI ranging from 5.5 [Buckley and McManus, 2002] to 53.65 [Henry and Antao, 1992]. This may mean it is one of the least or one of the most toxic antidepressant agents available. The large range of FTIs reported by different authors confirms that confounders can have a large effect, as the inherent toxicity of nortriptyline would not change over time.

Case series

Observational data from clinical use regarding overdose mortality have been published as individual case reports and case series. As case series can only include reported overdoses and outcomes they cannot tell us about relative risks between antidepressants but they are still informative as to the specific risks and outcomes associated with individual antidepressants.

Venlafaxine

The largest single drug series for venlafaxine is reported by the National Poisons Information Service in London [Colbridge and Volans, 1999] who followed up 632 patients who overdosed on venlafaxine. The main finding is that there were no deaths. Two cases of serious cardiotoxicity were reported, but one was associated with concomitant chlorpromazine and the other with thioridazine. In another large case series of 235 patients admitted to the Edinburgh Royal Infirmary following venlafaxine overdose, no deaths were reported, although a dose-dependent relationship between venlafaxine ingestion, tachycardia and corrected QT interval (QTc) prolongation was found and arrhythmias were documented in three patients [Howell et al. 2007]. Prolongation of the QT interval is associated with an increased risk of the potentially lethal cardiac arrhythmia, torsades de pointes, a risk that increases with the use of QT prolonging drugs [Anderson et al. 2002]. No cases of torsades de pointes were reported in this study and to the authors’ knowledge no cases have been published in the literature, although two cases have been reported on the ADROIT database [MHRA, 2012] neither of which were fatal. The analysis by Howell and colleagues used Bazett’s formula to calculate the QTc for heart rate. This is reported to overcorrect at higher heart rates [Desai et al. 2003] and, as venlafaxine is associated with tachycardia in overdose, this may have influenced the findings in this study. Further case series of 96 cases of venlafaxine overdose [Kelly et al. 2004]and 51 cases of venlafaxine overdose [Whyte et al. 2003] also did not report any deaths on venlafaxine. The data reported by Whyte and colleagues have now been expanded, and venlafaxine overdoses for which ECGs were recorded were available for 273 patients on 369 occasions [Isbister, 2009]. No deaths were reported and venlafaxine was only found to cause minor abnormalities in the QT and QRS interval, unlikely to be associated with major arrhythmias except possibly with large doses.

There are some data associating large overdoses of venlafaxine with cardiotoxicity [Bosse et al. 2008; Hojer et al. 2008]. In a retrospective cohort study of 36 cases of venlafaxine self-poisoning and 44 randomly selected SSRI self-poisoning cases admitted to an emergency department in Australia, one death was reported in the venlafaxine group [Chan et al. 2010]. This patient had ingested 12,600 mg of venlafaxine XR with propanolol, alcohol and amphetamine. No deaths were reported in the SSRI group. No clinically relevant changes in QT interval were noted. In this study venlafaxine was found to be prescribed preferentially in patients at a higher risk of serious suicide attempt.

Duloxetine

There are far fewer published cases on duloxetine overdoses. Two individual reports of overdoses on duloxetine have been reported [Kruithof et al. 2011; Menchetti et al. 2009]. Neither case was fatal. In the first report, cardiovascular alterations with hypotension, sinus bradycardia and possibly slightly prolonged QTc interval were found, and in the second, sinus tachycardia and normal QTc interval were reported.

The case series above can be contrasted with a case series of TCA overdoses published by Serafimovski in which 68 cases of TCA overdoses were followed and resulted in 57 (83%) patients having ECG abnormalities and 8 (12%) died [Serafimovski, 1975].

Cardiovascular safety

Cardiovascular safety in relation to overdose has been alluded to above, but here we will briefly review the preclinical, clinical and postmarketing cardiotoxicity data on venlafaxine and duloxetine.

Preclinical data

Of interest is whether duloxetine or venlafaxine have activity at sodium or potassium ion channels, which are the main cause of arrhythmias. Two studies in animals have shown that venlafaxine can inhibit cardiac ion channels [Fossa et al. 2007; Khalifa et al. 1999], but the concentrations of venlafaxine associated with inhibition were much greater than those seen in humans taking therapeutic doses so these are difficult to interpret. Preclinical data have demonstrated that duloxetine has no adverse effect on human cardiac sodium and potassium channels [Detke et al. 2005]. As there is now a wealth of clinical data for both duloxetine and venlafaxine, there is little point in dwelling on preclinical data which are of more use when a drug is under development.

Clinical trial data

A large review of the duloxetine clinical trial database which included 8504 patients on duloxetine has been published [Wernicke et al. 2007]. The review concluded that the use of duloxetine did not appear to be associated with significant cardiovascular risk in patients with conditions for which the drug has been approved or studied. In particular, there was nothing of concern regarding QTc interval, and this is reflected in the duloxetine summary of product characteristics (SPC) (available from www.emc.medicines.org.uk) which states ‘The heart rate-corrected QT interval in duloxetine-treated patients did not differ from that seen in placebo-treated patients’.

In a review of the venlafaxine clinical trial database by Rudolph and colleagues, of 2897 patients who took venlafaxine, there were no serious arrhythmias or significant increase in QTc interval [Rudolph and Derivan, 1996].

Postmarketing data

A large nested case control study has also been performed to assess whether venlafaxine is associated with an increased risk of sudden cardiac death or near death compared with other antidepressants [Martinez et al. 2010]. This study using the UKGPRD followed 207,384 new users of venlafaxine and other antidepressants with a diagnosis of depression or anxiety for an average of 3.3 years. There were 568 cases of sudden cardiac death or near death, which were matched to 14,812 controls. The adjusted odds ratio (OR) of sudden cardiac death or near death associated with venlafaxine use was 0.66 (95% CI 0.38–1.14) relative to fluoxetine use, whereas compared with citalopram it was 0.89 (95% CI 0.50–1.60) and with dosulepin 0.83 (95% CI 0.46–1.52). In addition, no evidence was found that venlafaxine use was associated with a higher risk of out-of-hospital haemodynamically significant acute ventricular tachyarrhythmia compared with the other antidepressants. It was therefore concluded that venlafaxine was not associated with an excess risk of cardiac death or near death compared with fluoxetine, citalopram or dosulepin in patients with depression or anxiety.

A nationwide study performed in Denmark examining the association of antidepressant use and out-of-hospital cardiac arrest (OHCA) has recently been published [Weeke et al. 2012]. All patients in Denmark who experienced an OHCA between 2001 and 2007 were identified (19,110 in total) and associations between specific antidepressants and OHCA examined with conditional logistic regression in case–time–control models. A total of 2913 patients were receiving antidepressants at the time of the OHCA. TCAs (OR 1.69, 95% CI 1.14–2.50) and SSRIs (OR 1.21, 95% CI 1.00–1.47) were both associated with comparable increases in risk of OHCA. No association was found for SNRIs/noradrenergic and specific serotonergic antidepressants (NaSSAs) (OR 1.06, 95%CI 0.81–1.39). Citalopram (OR 1.29, 95%CI 1.02–1.63) and nortriptyline (OR 5.14, 95% CI 2.17–12.2) had the strongest associations. Venlafaxine had the lowest OR of 0.68 (95% CI 0.38–1.22) from 177 identified cases of OHCA and hence no evidence was found that venlafaxine increased the risk of OHCA. A review of 37 patients with depression taking high therapeutic doses of venlafaxine (mean dose 346.15 mg/day) did not reveal any clinically significant change in QTc intervals [Mbaya et al. 2007]. However, there is a case report of significant QTc prolongation associated with venlafaxine 150 mg/day in an older lady with depression who had a QTc interval of 582 ms which reduced to 430 ms several days after discontinuing venlafaxine [Letsas et al. 2006].

Other cardiovascular-related adverse events

Another area of potential concern for SNRIs is well recognized and relates to the potential to increase pulse and blood pressure because of inhibition of reuptake of noradrenaline. These are covered under the SPCs for duloxetine and venlafaxine (both available from http://www.emc.medicines.org.uk). The SPC for duloxetine gives a warning that blood pressure monitoring is recommended in patients with known hypertension or other cardiac disease. It also states that duloxetine should be used with caution in patients whose conditions could be compromised by an increase in heart rate or blood pressure. The SPC for venlafaxine (Efexor XL, Pfizer Ireland Pharmaceuticals, County Kildare, Republic of Ireland) is slightly different in that it recommends all patients should be screened for hypertension prior to initiation and all patients should have their blood pressure monitored. Similar to duloxetine, there is a warning for patients who might be compromised by an increase in heart rate.

Conclusion

There is now a substantial amount of clinical information on both duloxetine and venlafaxine taken at therapeutic doses and in overdose. FTI data are confounded by a number of variables, in particular differential antidepressant prescribing in patients with different baseline risks of taking antidepressant overdose. This and other confounders must be considered when interpreting FTI data as without adjustment the data prove unreliable. It must be remembered, however, that many confounders are unknown or unrecorded and therefore cannot be adjusted for. Case fatality indices suggest a slightly higher toxicity of venlafaxine than that of duloxetine or SSRIs, although venlafaxine users may have taken larger overdoses. Further studies would be required to investigate this possibility. In addition, a large nested case–control study and clinical data do not currently suggest that venlafaxine is significantly more toxic than first-line SSRIs or that it is associated with increased rates of attempted suicide, successful or otherwise. Duloxetine data are more limited but demonstrate no significant safety issues relating to suicidality or mortality from overdose compared with SSRIs. Considering as few as one-third of patients treated with SSRIs achieve full remission from their symptoms [Trivedi et al. 2006] there is a clinical need for alternative antidepressants. Switching to another antidepressant class after SSRI treatment failure may be a more effective strategy than within class switching [Papakostas et al. 2008]and data specifically support the safety and efficacy of switching SSRI treatment failures to SNRIs, such as duloxetine [Perahia et al. 2008, 2009; Romera et al. 2012]. Based on this review of the safety data from the SNRIs duloxetine and venlafaxine and the evidence of their efficacy in depression and GAD [Allgulander et al. 2008; Smith et al. 2002; Thase et al. 2007], both drugs appear to have a positive benefit risk profile. This evidence suggests both duloxetine and venlafaxine are appropriate for use in primary care, especially as a second-line option following an SSRI.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial or not for profit sectors.

Conflict of interest statement

ALS and AB are employees of Eli Lilly and Company who manufacture and market duloxetine, an SNRI antidepressant.

References

Acharya

Rosen

Polzer

D’Souza

Perahia

Cavazzoni

. (2006) Duloxetine: meta-analyses of suicidal behaviors and ideation in clinical trials for major depressive disorder. J Clin Psychopharmacol 26: 587–594.

Allgulander

Nutt

Detke

Erickson

Spann

Walker

. (2008) A non-inferiority comparison of duloxetine and venlafaxine in the treatment of adult patients with generalized anxiety disorder. J Psychopharmacol 22: 417–425.

Anderson

Al-Khatib

Roden

Califf

(2002) Cardiac repolarization: current knowledge, critical gaps, and new approaches to drug development and patient management. Am Heart J 144: 769–781.

Bebbington

Brugha

Coid

Crawford

Deverill

D’Souza

. (2009) Adult Psychiatric Morbidity in England, 2007. Results of a household survey. London: The NHS Information Centre for Health and Social Care.

Bergen

Murphy

Cooper

Kapur

Stalker

Waters

. (2010) A comparative study of non-fatal self-poisoning with antidepressants relative to prescribing in three centres in England. J Affect Disord 123: 95–101.

Bosse

Spiller

Collins

(2008) A fatal case of venlafaxine overdose. J Med Toxicol 4: 18-20.

Buckley

McManus

(2002) Fatal toxicity of serotoninergic and other antidepressant drugs: analysis of United Kingdom mortality data. BMJ 325: 1332–1333.

Chan

Gunja

Ryan

(2010) A comparison of venlafaxine and SSRIs in deliberate self-poisoning. J Med Toxicol 6: 116–121.

Colbridge

Volans

(1999) Venlafaxine in oversdose – experience of the national poisons information service (London centre). J Toxicol Clin Toxicol 37: 383.

10.

Desai

Desta

Malik

Flockhart

(2003) Variability of heart rate correction methods for the QT interval. Br J Clin Pharmacol 55: 511–517.

11.

Detke

Iyengar

Henck

(2005) Cardiovascular effects of duloxetine: preclinical and clinical findings. In: Proceedings of the American Collage of Neuropsychopharmacology 44th Annual Meeting, Waikoloa, HI, 11–15 December 2005.

12.

Entsuah

(2001) Venlafaxine/venlafaxine XR for suicidal ideation World J Biol Psychiatry 2: 331S.

13.

Fossa

Gorczyca

Wisialowski

Yasgar

Wang

Crimin

. (2007) Electrical alternans and hemodynamics in the anesthetized guinea pig can discriminate the cardiac safety of antidepressants. J Pharmacol Toxicol Methods 55: 78–85.

14.

Gibbons

Brown

Hur

Davis

Mann

(2012) Suicidal thoughts and behavior with antidepressant treatment: reanalysis of the randomized placebo-controlled studies of fluoxetine and venlafaxine. Arch Gen Psychiatry 69: 580–587.

15.

Hawton

Bergen

Simkin

Cooper

Waters

Gunnell

. (2010) Toxicity of antidepressants: rates of suicide relative to prescribing and non-fatal overdose. Br J Psychiatry 196: 354–358.

16.

Henry

(1989) A fatal toxicity index for antidepressant poisoning. Acta Psychiatr Scand Suppl 354: 37–45.

17.

Henry

(1993) Debits and credits in the management of depression. Br J Psychiatry Suppl (20): 33–39.

18.

Henry

Antao

(1992) Suicide and fatal antidepressant poisoning. Eur J Med 1: 343–348.

19.

Hojer

Hulting

Salmonson

(2008) Fatal cardiotoxicity induced by venlafaxine overdosage. Clin Toxicol (Phila) 46: 336–337.

20.

Howell

Wilson

Waring

(2007) Cardiovascular toxicity due to venlafaxine poisoning in adults: a review of 235 consecutive cases. Br J Clin Pharmacol 64: 192–197.

21.

Ilyas

Moncrieff

(2012) Trends in prescriptions and costs of drugs for mental disorders in England, 1998–2010. Br J Psychiatry 200: 393–398.

22.

Isbister

(2009) Electrocardiogram changes and arrhythmias in venlafaxine overdose. Br J Clin Pharmacol 67: 572–576.

23.

Jick

Kaye

Jick

(2004) Antidepressants and the risk of suicidal behaviors. JAMA 292: 338–343.

24.

Kelly

Dhaun

Laing

Strachan

Good

Bateman

(2004) Comparative toxicity of citalopram and the newer antidepressants after overdose. J Toxicol Clin Toxicol 42: 67–71.

25.

Khalifa

Daleau

Turgeon

(1999) Mechanism of sodium channel block by venlafaxine in guinea pig ventricular myocytes. J Pharmacol Exp Ther 291: 280–284.

26.

Kruithof

Bruins

van Roon

(2011) Coma after overdose with duloxetine. Ann Pharmacother 45: e5.

27.

Lawrenson

Tyrer

Newson

Farmer

(2000) The treatment of depression in UK general practice: selective serotonin reuptake inhibitors and tricyclic antidepressants compared. J Affect Disord 59: 149–157.

28.

Letsas

Korantzopoulos

Pappas

Evangelou

Efremidis

Kardaras

(2006) QT interval prolongation associated with venlafaxine administration. Int J Cardiol 109: 116–117.

29.

Martinez

Assimes

Mines

Dell’aniello

Suissa

(2010) Use of venlafaxine compared with other antidepressants and the risk of sudden cardiac death or near death: a nested case-control study. BMJ 340: c249.

30.

Mbaya

Alam

Ashim

Bennett

(2007) Cardiovascular effects of high dose venlafaxine XL in patients with major depressive disorder. Hum Psychopharmacol 22: 129–133.

31.

McAllister-Williams

Foran

Forrest

Ingram

McMahon

Taylor

. (2006) Prescribing of venlafaxine and dosulepin in primary care. J Psychopharmacol 20: 868.

32.

MHRA (2004) Medicines and Healthcare Products Regulatory Agency. Available at: http://www.mhra.gov.uk/NewsCentre/Pressreleases/CON002060 (accessed 16 December 2012).

33.

MHRA (2006) Updated prescribing advice for venlafaxine (Efexor/Efexor XL). Medicines and Healthcare Products Regulatory Agency. Available at: http://www.mhra.gov.uk/home/groups/pl-p/documents/websiteresources/con2023842.pdf (accessed 16 December 2012).

34.

MHRA (2007) Antidepressants: suicidal behaviour. Medicines and Healthcare Products Regulatory Agency. Available at: http://www.mhra.gov.uk/Safetyinformation/DrugSafetyUpdate/CON079102 (accessed 16 December 2012).

35.

MHRA (2010a) Fluoxetine: possible small risk of congenital cardiac defects. Medicines and Healthcare Products Regulatory Agency. Available at: http://www.mhra.gov.uk/Safetyinformation/DrugSafetyUpdate/CON091089 (accessed 16 December 2012).

36.

MHRA (2010b) S.S.R.I.s and S.N.R.I.s: risk of persistent pulmonary hypertension in the newborn. Medicines and Healthcare Products Regulatory Agency. Available at: http://www.mhra.gov.uk/Safetyinformation/DrugSafetyUpdate/CON085130 (accessed 16 December 2012).

37.

MHRA (2010c) Antidepressants: risk of fractures. Medicines and Healthcare Products Regulatory Agency. Available at: http://www.mhra.gov.uk/Safetyinformation/DrugSafetyUpdate/CON085136 (accessed 16 December 2012).

38.

MHRA (2011) Citalopram and escitalopram: QT interval prolongation – new maximum daily dose restrictions (including in elderly patients), contraindications, and warnings. Medicines and Healthcare Products Regulatory Agency. Available at: http://www.mhra.gov.uk/Safetyinformation/DrugSafetyUpdate/CON137769 (accessed 16 December 2012).

39.

MHRA (2012) Drug Analysis Prints (DAPS). Medicines and Healthcare Products Regulatory Agency. Available at: http://www.mhra.gov.uk/Safetyinformation/Howwemonitorthesafetyofproducts/Medicines/TheYellowCardScheme/YellowCarddata/Druganalysisprints/index.htm?secLevelIndexChar=Dq%20-%20Du#retainDisplay (accessed 16 December 2012).

40.

Menchetti

Gozzi

Saracino

Mercolini

Petio

Raggi

(2009) Non-fatal overdose of duloxetine in combination with other antidepressants and benzodiazepines. World J Biol Psychiatry 10: 385–389.

41.

Mines

Hill

Novelli

(2005) Prevalence of risk factors for suicide in patients prescribed venlafaxine, fluoxetine, and citalopram. Pharmacoepidemiol Drug Saf 14: 367–372.

42.

Moller

Baldwin

Goodwin

Kasper

Okasha

Stein

. (2008) Do SSRIs or antidepressants in general increase suicidality? WPA Section on Pharmacopsychiatry: consensus statement. Eur Arch Psychiatry Clin Neurosci 258(Suppl. 3): 3–23.

43.

Moore

Yuen

Dunn

Mullee

Maskell

Kendrick

(2009) Explaining the rise in antidepressant prescribing: a descriptive study using the general practice research database. BMJ 339: b3999.

44.

Morgan

Griffiths

Baker

Majeed

(2004) Fatal toxicity of antidepressants in England and Wales, 1993–2002. Health Stat Q (23): 18–24.

45.

National Health Service Business Services Authority (2011) Available at: http://www.nhsbsa.nhs.uk/PrescriptionServices/Documents/PPDPrescribingAnalysisCharts/CNS_Dec_2011_National.pdf (accessed 16 December 2012).

46.

National Health Service Information Centre for Health and Social Care (2011). Available at: http://www.ic.nhs.uk/pubs/quarterlypcajanmarch11 (accessed 16 December 2012).

47.

NICE (2010) The Treatment and Management of Depression in Adults. NICE Clinical Guideline 90. London: National Institute for Health and Clinical Excellence.

48.

NICE (2011) Generalised Anxiety Disorder and Panic Disorder (With or Without Agrophobia) in Adults: Management in Primary, Secondary and Community Care (Partial Update). National Clinical Guideline 113. London: National Institue for Health and Clinical Excellence.

49.

Ostler

Thompson

Kinmonth

Peveler

Stevens

(2001) Influence of socio-economic deprivation on the prevalence and outcome of depression in primary care: the Hampshire Depression Project. Br J Psychiatry 178: 12–17.

50.

Papakostas

Fava

Thase

(2008) Treatment of SSRI-resistant depression: a meta-analysis comparing within- versus across-class switches. Biol Psychiatry 63: 699–704.

51.

Perahia

Quail

Desaiah

Corruble

Fava

(2008) Switching to duloxetine from selective serotonin reuptake inhibitor antidepressants: a multicenter trial comparing 2 switching techniques. J Clin Psychiatry 69: 95–105.

52.

Perahia

Quail

Desaiah

Montejo

Schatzberg

(2009) Switching to duloxetine in selective serotonin reuptake inhibitor non- and partial-responders: effects on painful physical symptoms of depression. J Psychiatr Res 43: 512–518.

53.

Romera

Perez

Manchon

Schacht

Papen

Neuhauser

. (2012) Early switch strategy in patients with major depressive disorder: a double-blind, randomized study. J Clin Psychopharmacol 32: 479–486.

54.

Rose

Unis

(2000) A mortality index for postmarketing surveillance of new medications. Am J Emerg Med 18: 176-179.

55.

Rubino

Roskell

Tennis

Mines

Weich

Andrews

(2007) Risk of suicide during treatment with venlafaxine, citalopram, fluoxetine, and dothiepin: retrospective cohort study. BMJ 334: 242.

56.

Rudolph

Derivan

(1996) The safety and tolerability of venlafaxine hydrochloride: analysis of the clinical trials database. J Clin Psychopharmacol 16(3 Suppl. 2): 54S-59S; discussion 59S–61S.

57.

Serafimovski

Thorball

Asmussen

Lunding

(1975) Tricyclic antidepressive poisoning with special reference to cardiac complications. Acta Anaesthesiol Scand Suppl 57: 55–63.

58.

Smith

Dempster

Glanville

Freemantle

Anderson

(2002) Efficacy and tolerability of venlafaxine compared with selective serotonin reuptake inhibitors and other antidepressants: a meta-analysis. Br J Psychiatry 180: 396–404.

59.

Thase

Pritchett

Ossanna

Swindle

Detke

(2007) Efficacy of duloxetine and selective serotonin reuptake inhibitors: comparisons as assessed by remission rates in patients with major depressive disorder. J Clin Psychopharmacol 27: 672–676.

60.

Trivedi

Rush

Wisniewski

Nierenberg

Warden

Ritz

. (2006) Evaluation of outcomes with citalopram for depression using measurement-based care in STAR*D: implications for clinical practice. Am J Psychiatry 163: 28–40.

61.

Waring

Graham

Gray

Wilson

Howell

Bateman

(2010) Evaluation of a QT nomogram for risk assessment after antidepressant overdose. Br J Clin Pharmacol 70: 881–885.

62.

Weeke

Jensen

Folke

Gislason

Olesen

Andersson

. (2012) Antidepressant use and risk of out-of-hospital cardiac arrest: a nationwide case-time-control study. Clin Pharmacol Ther 92: 72–79.

63.

Wernicke

Lledo

Raskin

Kajdasz

Wang

(2007) An evaluation of the cardiovascular safety profile of duloxetine: findings from 42 placebo-controlled studies. Drug Saf 30: 437–455.

64.

White

Litovitz

Clancy

(2008) Suicidal antidepressant overdoses: a comparative analysis by antidepressant type. J Med Toxicol 4: 238–250.

65.

Whyte

Dawson

Buckley

(2003) Relative toxicity of venlafaxine and selective serotonin reuptake inhibitors in overdose compared to tricyclic antidepressants. QJM 96: 369–374.