uestion: Does the patient have significant ischemic cardiac
disease?uestion: Does the patient have significant ischemic cardiac
disease?
Show this place in the flowchart.
Help:As the severity of ischemic heart disease and the risk of infarction increases, the risk of mortality from use of a tricyclic (stemming from its properties as a Type 1 antiarrhythmic drug) increases.¹ At some point, this concern overbalances the documented efficacy of a tricyclic and the morbidity associated with continued severe depression in this population.¹ However, most elderly patients do not have ischemic heart disease, even if they do have atherosclerosis.¹ Therefore, for inpatients with severe, melancholic depression, a tricyclic (or equivalent, such as venlafaxine and perhaps mirtazapine) may have superior efficacy to an SSRI. (2,3)
¹Glassman AH, Roose SP, Bigger JTJr. The safety of tricyclic antidepressants in cardiac patients: risk benefit reconsidered. JAMA May 26, 1993; 269:2673-2675. Roose SP et al. Comparison of paroxetine and nortriptyline in depressed patients with ischemic heart disease. JAMA 1998;279:287-29.
²Gram LF. Fluoxetine. NEJM Nov. 17, 1994; 331(20):1354-1361.
³Roose SP. Are SSRI's effective for melancholic depression? Currents Interview. Currents in Affective Illness 1995; 14(9):5-12.
uestion: Does the patient have significant
dementia?
Show this place in the flowchart.
Help: To further screen candidates for getting a tricyclic first line for their severe melancholic depression, we now ask if there is significant dementia. The anticholinergic side effects of any tricyclic, even desipramine, are considerable and may complicate the treatment of a demented depressed patient. Therefore, if dementia is present, the first line recommendations will be other than a tricyclic.
Pseudodementia of depression should, as best as possible, be ruled out since that generally would not offer a barrier to use of a tricylic. Differentiating features include the tendency of pseudodemented patients to complain more bitterly about their cognitive dysfunction, out of proportion to the severity of the impairment; apathetic manner; psychomotor retardation; excessive guilt. See recent review of dementias.¹ But, be alert to the fact that many patients with "pseudodementia" on long term follow up develop actual Alzheimer's Disease.
¹Goldmacher, DS, Whithouse PJ. Current Concepts: Evaluation of Dementia. N Engl J Med 1996; 335(5): 330-336.
ECT is appropriate to consider for severely depressed, melancholic patients with cardiac ischemia if there is a need for rapid and definitive response. In non-psychotic patients, efficacy is at least equal to a tricyclic and cardiac safety favors ECT in comparison with a tricyclic. However, a maintenance treatment will have to be selected. For many patients, clinicians will prefer to try to find an effective pharmacotherapy and then stay with that medication, rather than using ECT and having to guess what will be effective for maintenance. Drugs that were ineffective for the acute episode should be avoided for maintenance use.
Other drugs with apparently equivalent efficacy to tricyclics which are much safer can also be considered (venlafaxine and mirtazapine)
A choice of ECT is Recommendation #26
To continue pharmacotherapy press here.
Show this place in the flowchart.
uestion: Has the patient had an adequate trial of a
tricyclic?
Show this place in the flowchart.
Help: If cardiac and other contraindications are not present, then the first-line recommendation would be a tricyclic antidepressant. However, this remains at best an educated guess, despite the conclusions of the meta-analysis and the opinions some experts.(1-5) New studies are underway to re-examine the question of whether there is any difference in the response to an SSRI and a tricyclic. An NIMH-sponsored comparison of nortriptyline and paroxetine in elderly patients, which included separate analyses for severe and melancholic patients, found no statistically significant differences. However, severely ill melancholic patients were not randomized and the N's are small. Also there was a numerical trend in favor of nortriptyline, which was greater for both the melancholic (62% vs 43%) and severe subgroups.(6) It is also possible that paroxetine is more effective in this population than fluoxetine seemed to be in the earlier reports.
There are other options to consider for this severely and melancholically depressed patient, that would be quite reasonable as alternatives to the first-line recommendation of a tricyclic. These will be seen if you press "No" as your answer.
A reasonably rigorous trial of a tricylic would be 4 to 8 weeks at "therapeutic" levels: four weeks if absolutely no response, 6-8 weeks if partial response is seen in the first 4 weeks.(7) Some studies indicate occasional success with a longer trial of up to 12 weeks or an attempt to exceed the therapeutic levels if side effects permit. However, this seems unnecessary in the context of this algorithm in which some sacrifice of exhaustiveness may be acceptable in favor of moving to options that might have a better percentage chance of helping the patient recover more quickly. In the inpatient setting, there will be intense pressure to make changes sooner than 4 weeks, and it may not be possible to resist these pressures. However, these short trials can not be considered adequate.(7)
(1)Osser DN. A systematic approach to the classification and pharmacotherapy of nonpsychotic major depression and dysthymia. J Clin Psychopharmacol 1993; 13:133-144.
(2)Glassman AH, Roose SP, Bigger JTJ. The safety of tricyclic antidepressants in cardiac patients: risk-benefit reconsidered. Journal of the American Medical Association. 1993;269:2673-2675.
(3)Anderson IM, Tomenson BM. The efficacy of SSRIs in depression: a meta-analysis of studies against tricyclic antidepressants. J Psychopharmacology 1994;8:238-249.
(4)Nobler MS, Roose SP. Differential response to antidepressants in melancholic and severe depressions. Psychiatric Annals 1998;28:84-88.
(5)Gram LF. Fluoxetine. NEJM 1994;331(20):1354-1361.
(6)Mulsant BH et al. A double-blind randomized comparison of nortriptyline and paroxetine in the treatment of late-life depression: six week outcome. J Clin Psychiatry 1999;60[suppl 20]:16-20.
A tricylic antidepressant. Our preferred tricylic is nortriptyline because of its lower incidence of postural hypotension and relatively moderate anticholinergic effects. A recent meta-analysis suggests the therapeutic plasma concentration of nortriptyline is 58 to 148 ng/ml.¹ The response rate within that range was 66% vs 26% above and below that range. A plasma level should be obtained in all patients treated with a tricylic at some point in the course of treatment.² Other tricyclic level ranges according to the meta-analysis were: desipramine (threshold: 116 mg/ml) - 51% response above, and 15% response rate below: imipramine (window: 175-350 mg/ml) - 67% inside and 39% outside the window.
There are a variety of other options to consider here, as well. Venlafaxine and mirtazapine may have better efficacy than SSRI's in this situation; of the two, venlafaxine has more convincing evidence so far and as of now seems appropriate for inclusion among the first-line recommendations. Nefazodone could also be considered. However, there is still little established regarding its efficacy for severe, melancholic depression. In the one placebo-controlled study of inpatients with melancholic depression, Hamiltons dropped from 30 to a still moderately depressed level of 17 after 6 weeks on nefazodone.(3) Experts still seem somewhat unconvinced about the efficacy of nefazodone in severe depressions. Bupropion may not be the best choice if dementia is present since there is a clinical impression that it occasionally induces confusional states. Bupropion is, however, relatively safe from the cardiac standpoint, having been tested in patients with serious heart disease, although its efficacy was not impressive.(4)
Recommendation #15
Show this place in the flowchart.
¹Perry PJ et al. Tricyclic antidepressant concentrations in plasma: an estimate of their sensitivity and specificity as a predictor of response. J Clin Psychopharmacol 1994; 14:230-240.
²Preskorn SH. Therapeutic drug monitoring with tricyclic antidepressants: a response. J Clin Psychopharmacol 1994; 14:277-278.
(3)Feighner JP, Targum SD, Bennett ME, et al. A double-blind placebo-controlled trial of nefazodone in the treatment of patients hospitalized for major depression. J Clin Psychiatry 1998;59:246-253.
(4)Roose SP, Dalack GW, Glassman AH, Woodring S, Walsh BT, Giardina EGV. Cardiovascular effects of bupropion in depressed patients with heart disease. American Journal of Psychiatry. 1991;148:512-516.
uestion: Has the patient had an adequate trial of an SSRI
in combination with a tricyclic?
Show this place in the flowchart.
Help: It appears that after failure on a tricyclic, many moderately depressed outpatients will do well on an SSRI.¹ However, for the population in question here (severely depressed, melancholic) this is considered doubtful.² An augmentation is considered more likely to be helpful. Among the tricyclic augmentation choices, lithium is the most well established in the literature. SSRI augmentation has some advocares, however, and thyroid hormone augmentation has a few supporters.³
¹Thase ME et al. Double-blind crossover antidepressant study: sertraline versus imipramine. NR172 Abstract, American Psychiatric Association Annual Meeting, Miami, Florida, May 23, 1995.
²Roose SP et al. Comparative efficacy of selective serotonin reuptake inhibitors and tricyclics in the treatment of melancholia. Am J Psychiatry 1994; 151(12):1735-1739.
³Nelson JC. Treatment of antidepressant nonresponders: augmentation or switch? J Clin Psychiatry 1998;59[suppl 15]:35-41.
Addition of an SSRI affords the opportunity to achieve the broader spectrum of pharmacodynamic activity associated with combining a serotonin stimulating and a noradrenergic agent, without having to introduce the side effects of lithium. The main problem with introduction of an SSRI is the need to attend to the consequences of inhibition of the Cytochrome P450 enzyme system.¹ It may be that the better results reported on the combination, in people who failed on both drugs alone, is due to higher blood levels of tricyclic in combination than when previously used as monotherapy.² However, in this algorithm, the clinician will have optimized blood levels as part of completing an adequate trial of the tricyclic. That may make the likelihood of response lower than in the above study, in which 7 or 13 patients responded to the combination.
Mirtazapine³ and venlafaxine (4) are reasonable alternatives, and also may well be better than an SSRI in this population.
If adding an SSRI, always be aware if the patient is on other drugs metabolized by Cytochrome enzyme IID6 (especially when using fluoxetine or paroxetine, and with lesser concern if the patient is going to receive sertraline or citalopram, since the latter two may raise desipramine levels by up to 50%) or by IIIA3,4 (especially with fluoxetine and sertraline and perhaps with lesser concern if the patient is going to receive paroxetine, and there is no problem if adding citalopram). Be prepared to start with a lower SSRI dose or make adjustments to the doses of the drugs that are substrates for the affected enzymes).
Check tricyclic levels and start with a low dose of the SSRI. Fluoxetine and paroxetine seem to raise nortriptyline and desipramine levels more than sertraline or citalopram do, but there will be individual variation. Hydroxylation of these secondary amines involves IID6 and the result is a buildup of the parent compound and lower levels of the hydroxy-metabolites. Since the hydroxy-metabolites are thought to be more cardiotoxic, it has been speculated that there may be a cardio-protective effect of this combination, so long as the parent compound levels are kept in the therapeutic range.
Recommendation #16
Show this place in the flowchart.
¹Harvey AT and Preskorn SH (reply by Popli A et al.) Interactions of SSRI's with tricyclic antidepressants. Arch Gen Psychiatry 1995; 52:783-785.
²Levitt AJ, Joffe RT, Kamil R, McIntyre R. Do depressed subjects who have failed both fluoxetine and TCAs respond to the combination? J Clin Psychiatry 1999;60:613-616.
³Wheatley DP et al. Mirtazapine: efficacy and tolerability in comparison with fluoxetine in patients with moderate to severe major depressive disorder. J Clin Psychiatry 1998;59:306-312.
(4)Rudolph RL, Feiger AD. A double-blind, randomized, placebo-controlled trial of once-daily venlafaxine XR and fluoxetine for the treatment of depression. J Affect Disord 1999;56:171-181.
Help:As the severity of ischemic heart disease and the risk of infarction increases, the risk of mortality from use of a tricyclic (stemming from its properties as a Type 1 antiarrhythmic drug) increases.¹ At some point, this concern overbalances the documented efficacy of a tricyclic and the morbidity associated with continued severe depression in this population. (1) However, most elderly patients do not have ischemic heart disease, even if they do have atherosclerosis.¹ Therefore, for most inpatients with severe, melancholic depression, a tricyclic is preferred as it is considered to have superior efficacy to an SSRI. (2,3)
¹Glassman AH, Roose SP, Bigger JTJr. The safety of tricyclic antidepressants in cardiac patients: risk benefit reconsidered. JAMA May 26, 1993; 269:2673-2675,
²Gram LF. Fluoxetine. NEJM Nov. 17, 1994; 331(20):1354-1361.
³Roose SP. Are SSRI's effective for melancholic depression? Currents Interview. Currents in Affective Illness 1995; 14(9):5-12.