Efficacy and tolerability of selective serotonin reuptake inhibitors compared with tricyclic antidepressants in depression treated in primary care: systematic review and meta-analysis
BMJ 2003; 326 doi: https://doi.org/10.1136/bmj.326.7397.1014 (Published 10 May 2003) Cite this as: BMJ 2003;326:1014
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
MacGillivray et al (1) published a meta-analysis on antidepressant
treatments for primary care depressed patients. They compared the efficacy
and tolerability of selective serotoninergic reuptake inhibitors (SSRI) to
those of tricyclics. Studying primary care patients was taken as the
criteria for selecting studies.
We question the validity of this criteria.
First, it is not sure that the patients treated in a primary care setting
are similar between the UK, Greece and South Africa, for example, in terms
of diagnosis of depression, depression severity, comorbidities or general
health care. One could expect a large heterogeneity in the population of
patients defined by primary care depressed patients.
Second, we wonder to
which extent primary care depressed patients differ from the secondary
care population of patients. We would like to know according to what
clinical parameters differences should be observed and, according to the
hypothesized differences, what differences in efficacy and safety should
be expected between the two antidepressant classes. When an
antidepressant, whose efficacy has been demonstrated in clinical trials,
is used in a patient with a depression, why should one expect that the
patient will respond to tricyclics or SSRI in a comparatively different
manner simply because the clinical setting is different? Do patients
become psychiatric patients only when in secondary care?
Did the authors
select studies where patients received a prescription for an
antidepressant while not being psychiatric patients?
A difference that might exist between primary and secondary care patients
might be a lesser comorbidity of other psychiatric disorders but a
greater comorbidity of physical disorders. This might impact on the
efficacy of antidepressants as well as on their tolerability. One could
also expect that secondary care patients would be managed in a more
careful way by psychiatrists aware of psychotropic drug administration
than by GPs who are less familiar with these treatments. It is therefore
surprising to read the conclusion that the drop-out rate could be lower in
primary care patients compared to secondary care patients.
The results of the meta-analysis by Gillivray et al (1) are in line with
what is known about antidepressant treatments (2) but the patients
included might be quite heterogenous.
1.MacGillivray S, Arroll B, Hatcher S, Ogston S, Reid I, Sullivan F,
Williams B, Crombie I. Efficacy and tolerability of selective serotonin
reuptake inhibitors compared with tricyclic antidepressants in depression
treated in primary care: systematic review and meta-analysis. BMJ
2003;323:1014
2.Anderson IM. Selective serotonin reuptake inhibitors versus
tricyclic antidepressants: a
meta-analysis of efficacy and tolerability. J Affect Disord 2000;58:19-36
Competing interests:
None declared
Competing interests: No competing interests
The valuable article by MacGillivray and his colleagues serves to
underline the general inadequacy of therapy with antidepressant drugs.
Frequently they are ineffective, and relatively often they are employed to
commit suicide, and in this setting they are quite lethal (1,2). Studies
of their efficacy are frequently small, methodologically flawed, and all
too often are sponsored by the manufacturers of the drug under study. As
the article points out, the majority of patients with depression are
treated in a primary care setting, but many of the drug studies are
conducted in a secondary care setting and their results - even when
reliable - cannot safely be extrapolated to the patient population that is
encountered in primary care.
The authors found no significant difference in effectiveness when
tricyclic antidepressants were compared with selective serotonin reuptake
inhibitors, although there was a significant advantage for the latter in
terms of greater tolerability and lower withdrawal rates. This might imply
that the serotonin reuptake inhibitors are actually inferior, as the lower
withdrawal rate was not accompanied by greater efficacy.
Perhaps the best conclusion is that none of the currently available
drugs is of sovereign value. I recall prescribing amitriptyline - then a
relatively new drug - as a house officer in 1961. Since then we have seen
a multitude of new antidepressants introduced - each one reported to be
superior to all preceding drugs - but after more than forty years,
amitriptyline remains on the market and many of the newer drugs are no
longer prescribed. Amitriptyline is not a very good drug, but its
persistence on the market surely indicates that there has been no dramatic
advance in antidepressant drug therapy.
1.Hawton K, Simkin S, Deeks J. Co-proxamol and suicide: a study of
national mortality statistics and local non-fatal self poisonings. BMJ
2003;326:1006-8.
2. Hall DW, Mant A, Mitchell PB, Rendle VA, Hickie IB, McManus P.
Association between antidepressant prescribing and suicide in Australia,
1991-2000: trend analysis. BMJ 2003;326:1008-11.
Competing interests:
None declared
Competing interests: No competing interests
The paper by MacGillivray et al (BMJ 2003; 326: 1014) potentially
makes a huge contribution to the understanding of the relative utility of
serotonin selective reuptake inhibitors (SSRIs) and older antidepressants
in primary care. Sadly, the usefulness of this meta-analysis is limited by
the original data, as is so often the case. These results are very
difficult to interpret, not least because the comparitors used in the key
studies are pharmacologically heterogeneous. Two studies, including one of
the key three crucial studies providing data in an unambiguous format,
used mianserin as a comparitor. Mianserin is a tetracyclic, not a
tricyclic as the paper’s title might imply. It is not a potent reuptake
inhibitor and is a potent and thus sedative antihistamine. A third study
used lofepramine as a comparitor, more ‘modern’ tricyclic with a much more
benign anticholinergic side effect profile than its older cousins, with an
adverse event rate similar to the SSRIs (1).
Two of the three studies providing unambiguous data for the efficacy
analysis were of paroxetine against amitriptyline. Paroxetine may be
considered an ‘atypical’ SSRI, in that is has some adrenergic reuptake
inhibition and anticholinergic properties, so that pharmacologically is
has a profile lying between the more selective SSRIs and the classic
tricyclics, as does lofepramine (2).
As an underlying principle of meta-analysis is homogeneity of the
material, it is no surprise that a clear answer does not emerge from this
work, given the pharmacological heterogeneity.
References:
1. Katona CL, Abou-Saleh MT, Harrison DA, Nairac BA, Edwards DR, Lock
T, Burns RA, Robertson MM. Placebo-controlled trial of lithium
augmentation of fluoxetine and lofepramine. Br J Psychiatry. 1995
Jan;166(1):80-6.
2. Hale A S. The treatment of depression: the reuptake inhibitors.
In: Honig A, van Praag HM. Depression: Neurobiological, Psychopathological
and Therapeutic Advances. 1997. John Wiley & Sons Chichester UK.
Competing interests:
Past investigator and speaker for companies producing all the drugs concerned
Competing interests: No competing interests
Dear Sir/Madam
We all agree that SSRI`S have greater tolerability and reach therapeutic
potential sooner but evidence does slightly favour Tricyclics in terms of
efficacy.Tricyclics can be very useful for resistant and severe depression
in inpatient setting because of wider recepter profile.If the risk of
suicide is minimal and there are no contraindications tricyclics can be
tried as a first change antidepressant in place of trying other SSRI,all
of which have the same receptor profile.
Thanks
Dr Singh
Competing interests:
None declared
Competing interests: No competing interests
Dear Sir,
MacGillivray etal(BMJ 2003)did show in their review article
similar effectiveness for the SSRIs and TCAs in primary care depressive
illness, also they did show that SSRIs are better tolerated.
But from clinical experience, although SSRIs are better tolerated than
TCAs, they tend to be of similar effectiveness for mild and moderate
depressive illness. For severe depressive illness with or without psychotic
features, TCAs tend to be better.
With my best regards,
A.K.Al-Sheikhli,MB;ChB,MRCPsych DPM.
Competing interests:
None declared
Competing interests: No competing interests
The conclusion of MacGillivray et al. (1) of comparable efficacy
between serotonin reuptake inhibitors and tricyclic antidepressants (TCA),
may result in increase usage of TCA. This raises again the question of TCA
safety, with particular reference to desipramine. Higher fatality rate
from desipramine compared with that of other TCA has been documented based
on national data in the UK (2,3) and the US (4,5). Additionally,
unexpected and unexplained death has been reported in several children
treated with therapeutic doses of desipramine.
We have suggested an explanation for this excessive mortality rate.
Desipramine and nortriptyline are desmethyl metabolites of their parent
compounds; imipramine and amitriptyline, respectively. Desipramine and
nortriptyline share similar pharmacokinetic profiles, and differ from
their parent compounds by having higher distribution volumes,
erythrocyte/plasma ratio, and higher fraction of unbound drug (5). Thus,
desipramine and nortriptyline would have higher tissue concentrations than
imipramine and amitriptyline, for a given plasma level. Consequently, the
recommended therapeutic plasma levels for nortriptyline is 30-150 ng/mL,
compared with 150-300 ng/mL of amitriptyline. The recommended daily dose
of nortriptyline (30-150 mg) and the maximum pill strength (75 mg) are
half of that of amitriptyline. Indeed, with these dosage adjustments, the
mortality index for nortriptyline and amitriptyline are similar. By
contrast, similar adjustments have not been done for desipramine. The
therapeutic desipramine plasma level listed in the US Pharmaceutical Drug
Index (USPDI) is 125-300 ng/mL, the daily dose recommended by the USPDI
and the PDR are identical with that of imipramine (50-300 mg), as is the
maximum pill strength (150 mg). Of note, therapeutic plasma levels of 40-
160 ng/mL were listed for desipramine in the Goodman and Gilman’s textbook
of Pharmacology since 1990. To obtain such plasma levels, desipramine dose
should be reduced to half.
We have suggested limiting the maximum pill strength of desipramine
to 75 mg, reducing its therapeutic plasma levels and to consider
downscaling of the recommended dose (5).
Yet, eight years elapsed since our suggestions were made, and dosage
recommendations, therapeutic plasma levels and maximal pill strength of
desipramine remained unchanged.
Compilation of the Annual Reports from the Toxic Exposure
Surveillance System of the American Association of Poison Control Centers
for the years 1983-2001, show high mortality index for desipramine which
exceeds that of amitriptyline, imipramine, and nortriptyline by 2.28, 2.67
and 2.35, respectively (Table).
We urge for prompt revision of desipramine dosage regime to improve
its safety.
Yona Amitai, M.D.
Ministry of Health, Israel
Jerusalem, Israel
Henry Frischer, MD. Ph.D.
Rush-Presbyterian-St. Luke’s
Medical Center Chicago, IL
References:
1. MacGillivary S, Arroll B, Hatcher S, Ogston S, Reid I, Sullivan F et
al. Efficacy and
tolerability of selective serotonin reuptake inhibitors compared with
tricyclic
antidepressants in depression treated in primary care: systematic
review and meta-analysis.
BMJ;2003:326:1014-7.
2. Cassidy S, Henry J. Fatal toxicity of antidepressant drugs in overdose.
BMJ 1987:295: 1021-4.
3. Buckley NA, McManus PR. Can the fatal toxicity of antidepressant drugs
be predicted with pharmacological and toxicological data? Drug
Safety 1998; 18; 369-81.
4. Kapur S, Mieczkowski T, Mann J. Antidepressant medications and the
relative risk of
suicide attempt and suicide. JAMA 1992: 268: 3441-5.
5. Amitai Y, Frischer H. The toxicity and dose of desipramine
hydrochloride. JAMA 1994:
272: 1719-20.
Table: Mortality Index of Four Tricyclic Antidepressant Medications During the Years 1983-2001 (the period 1995-2001 is since our report in 1994) Mortality Index: No. of Death / No. of Exposures (%)*_ 1983-8 1989-94 1995-2001 Total for 1983-2001 Amitriptyline 121/17,302 256/38,347 387/56,100 764/111,749 (0.70) (0.67) (0.69) (0.68) Imipramine 68/9,9596 110/19,961 85/15,678 263/45,235 (0.71) (0.55) 0.54) (0.58) Nortriptyline 28/2,904 110/15,173 58/11,485 196/29,562 (0.96) (0.72) (0.50) (0.66) Desipramine 83/4,113 143/9,869 53/4,050 279/18,032 (2.02)* 1.44)** (1.31) *** (1.55)**** Chi2=(df=3)*70;p<_.00179p.001 xmlns:_183="urn:x-prefix:_183" xmlns:data="urn:x-prefix:data" _34p.001="_34p.001" _183:p.001="_183:p.001" _="_" source="source" of="of" data:_="data:_" the="the" annual="annual" reports="reports" american="american" association="association" poison="poison" control="control" centers="centers" toxic="toxic" exposure="exposure" surveillance="surveillance" system="system" for="for" years="years" _1983="_1983" _-2001="_-2001" pre="pre"/>
Competing interests:
None declared
Competing interests: Table: Mortality Index of Four Tricyclic Antidepressant Medications Duringthe Years 1983-2001 (the period 1995-2001 is since our report in 1994)Mortality Index: No. of Death / No. of Exposures (%)*_1983-8 1989-94 1995-2001 Total for 1983-2001 Amitriptyline 121/17,302 256/38,347 387/56,100 764/111,749 (0.70) (0.67) (0.69) (0.68)Imipramine 68/9,9596 110/19,961 85/15,678 263/45,235(0.71) (0.55) 0.54) (0.58)Nortriptyline 28/2,904 110/15,173 58/11,485 196/29,562(0.96) (0.72) (0.50) (0.66)Desipramine 83/4,113 143/9,869 53/4,050 279/18,032(2.02)* 1.44)** (1.31) *** (1.55)****Chi2=(df=3)*70;p
ECONOMIC APPROPRIATENESS OF THE EXPENDITURE FOR SELECTIVE SEROTONIN REUPTAKE INHIBITORS IN DEPRESSIVE DISORDER: ANALYSIS OF NATIONAL PRESCRIPTION DATA IN ITALY
The systematic review by MacGillivray et al. (1) discusses the
efficacy and tolerability of selective serotonin reuptake inhibitors
(SSRIs) compared with tricyclic antidepressants. While this review is
almost exclusively focused on clinical effectiveness, the high cost of
these agents raises the need to evaluate also their cost-effectiveness
ratio.
In Italy, SSRIs are widely prescribed to patients with depressive
disorder. To interpret the data of yearly national expenditure (EXPEND)
for a drug class, the method recently proposed by Messori et al. (2-10)
assigns an index of “economic appropriateness” to the treatment(s) under
examination. This analysis is based on the comparison between the health
gain theoretically expected from EXPEND according to current cost
effectiveness benchmarks (expected health gain, EHG) and the amount of
health that is estimated to be gained in the "real" patients (real health
gain, RHG). The final index is calculated as RHG/EHG.
Considering the Italian expenditure for SSRIs in 2002 (372 million
Euros) along with the average cost of a DDD (12), the number of DDDs
administered over the year under examination can be calculated to be
323,478,269. If one designs this economic analysis as a comparison between
SSRIs and tricyclic antidepressants, the increase in cost related to the
use of SSRIs is 356,958,261 Euros in 2002 (assumptions: cost of 1 DDD =
1.15 Euros for SSRIs and 0.0465 Euros for tricyclic antidepressants). What
is the "economic appropriateness" of this expenditure? The value of EHG is
35,696 quality-adjusted life years (QALYs) or 7,139 QALYs using the
benchmark of either 10,000 or 50,000 Euros per QALY gained, respectively.
To calculate the value of RHG, since the average duration of the treatment
with these agents is around 2 years per patient (13), the number of
patients who have been started on these drugs in 2002 is calculated as
443,121. The lifetime gain per patient is 0.25 QALYs (13), and so 443,121
patients are thought to gain a total of 110,780 QALYs. Comparing EHG
(either 35,696 or 7,139 QALYs depending on which benchmark is chosen) with
RHG (110,780 QALYs) yields a ratio RHG/EHG of 3.10 or 15.5, respectively.
In summary, our analysis produces a pharmacoeconomic profile of SSRIs
(RHG much greater than EHG) more favourable than the findings previously
published in the literature (13). There are some important approximations
in our analysis: firstly the uncertainties about the average duration of
treatment and secondly the methodological quality of the study by Revicki
et al. (13) (which is flawed by the large number of unproven assumptions).
Both of these approximations might have contributed to overestimating the
value of RHG.
REFERENCES
1. MacGillivray S, Arroll B, Hatcher S, Ogston S, Reid I, Sullivan F,
Williams B, Crombie I. Efficacy and tolerability of selective serotonin
reuptake inhibitors compared with tricyclic antidepressants in depression
treated in primary care: systematicreview and meta-analysis.
2. Messori A et al. Economic appropriateness of the expenditure for
alendronate: cost-effectiveness analysis of national prescription data in
Italy.
http://bmj.com/cgi/eletters/327/7406/89#40333
3. Santarlasci B, Trippoli S, Messori A. Economic appropriateness of
the expenditure for beta-interferon in multiple sclerosis: analysis of
national prescription data in Italy.
http://bmj.bmjjournals.com/cgi/eletters/326/7388/522#46835
4. Cecchi M, Pelagotti F, Santarlasci B, Trippoli S, Brutti C,
Messori A. Economic appropriateness of the expenditure for infliximab in
rheumatoid arthritis : analysis of national prescription data in Italy.
http://bmj.com/cgi/eletters/324/7333/312#44561
5. Santarlasci B, Brutti C, Messori A. Economic appropriateness of
the expenditure for aromatase inhibitors: analysis of national
prescription data in Italy.
http://bmj.com/cgi/eletters/327/7420/885#47864
6. Messori A et al. Economic appropriateness of the expenditure for
coxibs: cost-effectiveness analysis of national prescription data in
Italy. http://bmj.com/cgi/eletters/327/7420/933-b#39700
7. Santarlasci B. Economic appropriateness of national drug
expenditures: improved estimation of cost-effectiveness ranking for
coxibs. http://bmj.com/cgi/eletters/327/7420/933-b#42273
8. Messori A et al. Spending on statins.
http://bmj.com/cgi/eletters/327/7420/933-b#38400
9. Messori A et al. Questionable cost-effectiveness of statins for
primary prevention of cardiovascular events.
http://bmj.com/cgi/eletters/326/7404/1407#34612
10. Vacca F et al. Cost effectiveness of statins in men vs women.
http://bmj.com/cgi/eletters/327/7420/933-b#39209
11. Messori A et al. Absolute and relative rankings in cost-
effectiveness comparisons. http://bmj.com/cgi/eletters/327/7426/1237#42094
12. Autori Vari. L’uso dei farmaci in Italia. Rapporto nazionale
2002. Osservatorio Nazionale sull’impiego dei farmaci.
13. Revicki DA, Brown RE, Keller MB, Gonzales J, Culpepper L, Hales
RE. Cost-effectiveness of newer antidepressants compared with tricyclic
antidepressants in managed care settings. J Clin Psychiatry. 1997
Feb;58(2):47-58.
Competing interests:
None declared
Competing interests: No competing interests