The first line of pharmacotherapy fur depression. iypicaJJy a tricyclic antidcpressant (TCA), works well to relieve depression in the majority of patients, without side effects any worse than minor annoyances. Many general practitioners and other nonpsychiatric physicians can conduct such treatment effectively. Il is when the TCA does not work well that special psychiatric expertise is required.
Much commercial research has focused on developing antidepressanl medications with less likelihood of causing specific side effects. If a given antideprcssant, such as imipramine. leaves a patient intolerably oversedatcd, for example, it is easy to suggest another, less sedative drug, such as dcsipramine. Unfortunately, much less research exists to guide the psychiatrist when the patient's depression fails to improve with an otherwise adequate and welltolerated course of TCA therapy. Understandably, drug manufacturers and researchers alike would prefer robust results, with high rates of recovery, generalizable to the largest possible group of depressed patients rather than the less impressive results anticipated from a minority of depressives selected for resistance to prior pharmacotherapy.
Nevertheless, this is a treatment problem that often confronts the psych latrisi. The TCA-resistant depressive will need more, often long-term, psychiatric care than the TCA-responsivc depressive, and thereby has a prevalence in psychiatric treatment settings out of proportion to the overall incidence of rei factory depression. That incidence, while leaving drug-resistant depressives in the minority, is still substantial - at least 2O1O to 3O1V Depression being so common, this means that a great many people will suffer from treatment-resistant depression - one estimate suggests that two million Americans will face this dilemma at some point in their lives.1 Until more data accumulate from controlled studies, the psychiatrist treating such patients will often have to turn to reports of uncontrolled clinical experience in deciding what to do after several drugs have failed. There is, however, at least some research to guide us in the earlier phases of sequential pharmacotherapy for TCA-rcsistant depression.
ENSURING AN "ADEQUATE TRIAL"
One of the first questions to be asked when a patient is described as refractory to treatment should be: exactly what trcalment(s) - how much and for how long? All too often, even in the literature, the answers to these questions leave open the possibility that the patient never received an "adequate trial" of whichever antideprcssants were previously tried. There may be good reasons for this, such as when the patient has proven unable to tolerate maximal doses of a drug. Hypotension resulting as a side effect of imipramine is one example of this. However, such cases are belter referred to as "tricyclic-intolerant" rather than "tricyclic-resistant." to keep separate twro rather different problems. There is much data to guide us in getting around specific side effects.1 In the case of symptomatic hypotension, for example, a less hypotensive TCA. such as norlriptyline, can be substituted for imipramine.
A patient should not be described as TCA-resistant until that patient has failed to show reasonable improvement in depressive symptoms during a course of treatment involving at least four (preferably six) weeks on maximum daily dosage of a TCA equivalent to 300 mg of imipramine, because studies have shown that antidepressant effects continue to emerge as duration and dosage are increased up to at least these limits.2,3 In cases refractory to apparently adequate dosage and duration of TCA treatment, blood TCA levels should be measured and either shown to be within "therapeutic range" or, if not, result in dosage adjustment upwards (even above generally recommended maxima, if tolerated) or downwards (in the case of drugs with putative "therapeutic windows") to yield blood levels within the range.4
If the patient cannot tolerate these durations and doses, examining the reasons for this intolerance should lead to another, more adequate trial with a different TCA. In a few cases, such as excessive anticholtnergie effects, even with a less anticholinergic TCA like desipramine, it may be necessary to move outside the tricyelic group to a drug like trazodone, However, considering the possibility that the patient will eventually prove resistant to whichever drug is first given adequate trial, the psychiatrist should bear in mind that, limited as our knowledge is about what may work tor TCA-resistant depression, it is still more limited with respcet to depressions resistant to other, newer drugs.
MONOAMINE OXIDASE INHIBITORS
One important role for non-TCA antidcpressants lies in the treatment of depressed patients who have proven intolerani of or resistant to TCA therapy. The major type of nonTCA antidepvessanl to offer itself in the latter role is the monoaminc oxidase inhibitor (MAOI). Discovered concurrently with but independently of imipramine (the first TCA), MAOIs seem to represent a distinctly different class of drug. In contrast, drugs like trazodone or maprotilinc. while not strictly speaking tricyclics in chemical structure, could be classilied along with TCAs as "amine uptake inhibitors." though the significance of this common neuvopharmaeological effect lor clinical efficacy is still open to dispute.
Moreover, clinical experience dating back over two decades has supported the effectiveness of MAOIs in TCA-resistant depression. Studies have suggested that MAOI treatment in eases refractory to amine uptake inhibitors may work about as often, with a 60% to 75% rate of improvement, as does TCA treatment in depressions not selected for refractoriness. 6'9 Thus, MAOIs may differ sufficiently from TCAs that refractoriness to the latter has little bearing on the likelihood of response to the former.
Much interest has concerned the combining of MAOIs and TCAs for added efficacy in refractory depression. ??-" However, many reports have failed to ensure that patients treated with the combination had been previously tried and failed on an adequate course of treatment with an MAOI alone, so that eventual success on combined MAOI-TCA treatment may often reilect only the successful effect of the MAOI alone. Furthermore, the few controlled studies so far undertaken with the MAOI-TCA combination have shown no evidence for advantage,12'1^ although there remains to be done the critical study comparing the MAOI-TCA combination with single drugs in depressions resistant to adequate trials of both drugs used alone.
The only approach that really competes with a switch to an MAOl for the role of "second line" treatment in TCA-rcsistant depression is thai of adding lithium to the established TCA regimen. Much research on this approach has been published. lb-2u Although early studies left open questions about the adequacy of the antecedent TCA trial, the evidence has gradually increased that even after TCA trials ample in duration and dosage the addition of lithium will add substantially to clinical improvement in the majority of cases, even with unipolar or delusional depressions.
Lithium by itself has some antidepressant efficacy.21 and research so far has left it unclear whether the additional improvement observed with this strategem represents:
* A simpler additive antidcpressant effect of TCA plus lithium;
* An effect that could be obtained equally well with lithium alone.
Nevertheless, research has focused on the combination of lithium with a standard antidepressant, whether TCA or MAOI. Considering the relative safety and convenience of such combinations, this clinical practice is likely to continue, taking advantage of possible partial response to TCA even in treatment-resistant eases, and avoiding the risk of deterioration on withdrawal from TCA to allow several days free of medication prior to safe initiation of an MAOI.10,11
Alterations of thyroid function apparently secondary to depression and the possibility of subclinical thyroid dysfunction underlying depression make the use of thyroid hormones as an adjunct to antidepressant medication a complex issue. Several studies have indicated that TCAs combined with thyroid work better than TCAs alone, at least in females, but leave it less clear that addilion of thyroid to the TCA regimen of depressed patients refractory to wholly adequate trial of such a regimen will also work.220tl It may be that thyroid only enhances antidepressant effects of the TCA when lower than maximal doses or less than adequate duration of the TCA trial results in partial effectiveness of the primary antidepressant.27 It may also be that subtle signs of thyroid dyslunction, such as abnormalities evident only on TRH stimulation testing or screening for antithyroid antibodies, noi evaluated in earlier studies, are necessary preconditions for obtaining beneficial effects from addition of thyroid hormone to an aniidepressant.28
TRYPTOPHAN AND 5-HYDROXYTRYPTOPHAN
As precursors oiscrotonin. l.-lryptophan and 5-hydruxytryptophan (5-HTl') have been studied for the treatment of depression both alone and in combination with standard anlideprcssants. So far. results have constsently supported only the efficacy of L-lryptophan in enhancing the antidepressant effects of MAUIs: in controlled studies patients treated with MAOI plus L-tryptopban have generally done better than those treated with MAOI alone, whereas results involving 5-HTP or TCAs have been inconsistent.24,30
This sounds promising as an avenue for dealing with patients resistant to an adequate trail of an MAOI alone, but there is actually little evidence bearing directly on the effectiveness of such a strategy. As with thyroid supplementation of TCA treatment, it is possible that L-tryptophan supplementation of MAOIs represents a less relevant type of interaction - such as speeding up a response that would occur eventually in any case. Whether L-lryplophan has value specifically for refraeiory depressions remains to be seen. Moreover, the fact that L-tryptopban and 5-HTP have so far failed to show consistent antidepressant effects of their own. or even consistent potentiation of TCAs. means that their role in resistant depression must be consigned to a later stale of treatment - after TCA. MAOI, and lithium have all failed.
Like lithium , the psychostimulants - principally methylphenidate. dextroamphetamine. and pcrnoline - represent in themselves a long-available and plausible, if not universally accepted, amidepressant pharmaeotherapy different in various seemingly important respects from more standard antidepressants. They do raise serious concerns about risks of dependence and abuse, but such concerns should not bar their judicious use in cases proven resistant to less risky alternatives. What is lacking at this point is much evidence that they do work where more standard agents have failed. There is some evidence thai methylphenidale may help by raising blood levels ol TCAs with which il is combined,31 but the same effect might be more easily obtained by simply increasing the dose of TCA.
An incomplete chapter of psychopbarmacological research concerns the seemingly paradoxical reactions manifested by some depressed patients on TCAs to which reserpine is added, sometimes dramatically altering the course of a previously resistant depression. However, though lhe lew published studies are consistent about this, this work needs further replication to become convincing enough lor general clinical use.32-34
One group of investigators have conducted studies (35) suggesting that high-dose oral coniugateci estrogeni* in women and mesterolone in men can benefit depressions that have responded poorly to conventional treatments." However, the documentation of treatment resistance in these cases is poor, the antidepressant effects reponed are modest, and the agents pose sufficient risk of side effects to deter their clinical trial by most psychiatrists even in severely refraclory cases, at least until further supporting data are forthcoming.
Certain newer drugs with known or suspected antidepressanl efficacy have received isolated reports of effectiveness in refractory eases. Trazodone is an example of a nonTCA. non-MAOl anlideprcssant which warrants further study as an alternative lo MAOIs as a second line of treatment for TCA-resislanl depression."^ Carbamazepine. though structurally a lricyelic and still in need of studies confirming its antidepressant efficacy, has proven sufficiently useful as a mood-stabilizing agent in lithium-resistant cases of bipolar disorder to warrant further trial in TCA-resistam depression as well.37
NEUROLEPTICS AND DELUSIONAL DEPRESSION
Most of the preceding discussion applies specifically to nonpsychotic depression; there is a sizable body of data indicating that psychotic or delusional depression represents a different case with respect to issues of treatment and prognosis.38'40 Given antidepressant medication alone, delusional depressives appear lo do rather poorly. Given neuroleptics along with antidepressants. or given electroconvulsive therapy (ECT). they appear to do quite well, so that both these measures might be considered relatively early, perhaps even intially. in the treatment of major depression with psychotic features. Yet, both neuroleptics and ECT raise special concerns about risks of l o n g - 1 a s t i n g adverse effects - albeit perhaps unjustified in the case of ECT - that their use in nonpsychotic depression will probably occur much later in the sequence of alternatives for refractory cases.
A TREATMENT PLAN FOR RESISTANT DEPRESSIONS
With so little data available from controlled research bearing directly on the issue of refractory depression, there is insufficient basis for a dogmatic and restrictive approach to treatment choices. Some authors have published flow charts to guide successive treatment decisions, but even this seems premature.41-43
What does seem clear is that the clinician should first assure that the apparently treatment-resistant depressive really has had a trial of a standard, tricyclic anudepressant lhai is wholly adequale in terms of dosage, duration, and, ii available, blood levels as well.
Tricyclic intolerance should not be confused with tricyclic resistance; where intolerance ol side effects hinders adequate treatment, generally another standard drug can be found that will prove tolerable enough for an adequate antidepressant trial.
In assessing the effects of a given drug on a specific patient, the clinician should bear in mind the possibility of partial antidepressanl efficacy which, while at first seeming far short of "satisfactory," may later appear belter than other agents tried and well worth returning to in such a case. In other words, treatment resistance is not always an "all or nothing" phenomenon: use of standard depression rating scales such as the Hamilton scale in the course of clinical treatment may aid in measuring the extent of partial treatment response for comparison of different treatments with each individual patient.
When adequate trial of a TCA has failed, it is certainly worth trying an MAOI; indeed, this seems to be the major indication for use of MAOIs.44 Once considered nearly obsolete, these drugs have enjoyed a renaissance ol clinical use. and we may appreciate this increasingly as the need for second-line treatments becomes more apparent. As with TCAs, it is important to assure adequacy of dosage and duration belorc deciding that a given depressed patient is also refractory to trial of an MAOI.
Because it can simply be added to the ongoing TCA regimen that may have yielded partial even though insufficient benefits, lithium may represent an even stronger candidate than MAOIs for the role of secondline treatment in TCA-rcsistant depression.
Thyroid, like lithium, has appeal as an alternative because it can be used as an adjunct to an ongoing TCA regimen; however, the data for efficacy of this intervention in refractory cases seem less convincing than that for lithium or MAOIs.
When all these measures have tailed, the clinician will face an array of alternatives that have at most only suggestive evidence to support their value in refractory depression: combining MAOI and TCA, combining L-tryptophan with an MAOI. adding meihyiphenidate ov rescvpine to a TCA. or trying one of the newer drugs with amidepressant effects that have become available in recent years or may become available shortly - trazodone, maprolline. alprazolam. S-adenosylmethionine. fluoxetinc, fluvoxamine, sertraline. A comprehensive list would be much longer than this, and undoubtedly will continue to grow.
QUESTIONS IN NEED OF ANSWERS
Adequate trial of all known antidepressanls and combinations with all putative potentiating agents for a persistantlv refractory depression could easily lake years, during which the patient's hoplessness may deepen from the experience of interminable treatment failures. The chronic depression adversely impacts the patient's life in many ways - through divorce, social isolation, lack of income, loss of social and occupational status, alcohol and drug abuse. Of course, it is equally possible thai the hopes raised by having a series of different drugs to try may help support the patient through a íong period of chronic depression until spontaneous remission supervenes.
It is not enough, however, to have an array of treatment alternatives that might work, to be chosen more or less randomly for use in treatment on a trial-and-error basis. We need to choose what is to be tried next on the basis of data indicating what is most likely to work. So far, we are largely lacking in such data, which requires at least systematic follow-up of patients initially selected for treatment-resistance according to stringent criteria, as they proceed through successive treatment trials administered on a clinical basis.
While treatment approaches that seem most promising have been reviewed, these are not necessarily the usual approaches tried first in clinical situations. For example, many psychiatrists, speculating about "serotonergic" versus "noradrenergic" effects ol dubious relevance, may proceed from lailure on an adequate trial of amitriptyline. to a subsequent trial of desipramine. There is virtually no evidence to indicate lhc likelihood of success with such a measure. Such a clinically common, but poorly documented stratagem needs ?? be compared in controlled fashion with some of the approaches reviewed and recommended here. New antidepressant drugs will also deserve controlled comparison with established treatments for TCA-resistant depression. because much of their usefulness may lie in such cases.
Refractory depression is an area where not onlv controlled research. but also careful clinical reports, can significantly increase our knowledge. Where a patient has stable symptomatology, often present on a chronic basis, without response to some plausible prior treatments, the clinician is in a belter than usual position to document efficacy from an instance of therapeutic success. This efficacy is less likely to be attributable to placebo effects of spontaneous remission (though this may occur even in the most chronic cases) than is improvement in a case of acute, previously untreated depression. Many of the treatments discussed here have their basis for consideration largely in such uncontrolled clinical reports, a situation likely to alter only slowly as research turns increasingly to this difficult but important issue.
1. Price LH, Charney DS. Hemnger GR: Variability of response to lithium augmentation in refractory depression. Am J Psychiatry 1986; 143:1387-1392.
2. Quitkin KM, Rabkin IG. Ross D. et al: Duration of antidepressant drug treatment. Arch UfM Psvchiatry 1984; 41:238-245.
3. Simpson GM, Lee |H. Cuculic Z. et al: Two dosages of imipramine in hospitalized endogenous and neurotic depressives. Arch GfH Psychiatry 1976: 33:1093-1102.
4. Simpson G. Pi E, White K; Plasma drug levels and clinical response to antidcpressants. J Clin Psychiatry 1983; 44:91-93.
5. Schicle BC: The Parnate-spedie patient. Mimi Medi 96 5; 48:355-357.
6. Himmelhoch JM, Detrc T. Kupfer Dl. et al: Treatment of previously intractable depressions with iranylcypromine and lithium. J Nerv Men't DAs 1972; 155:216-220.
7. Georgotas A. Friedman E. McCarthy M, et al: Resistant geriatric depressions and therapeutic response to monoaminc oxidase inhibitors. Biol Psvchiatry 1983; 18:195-205.
8. Price LH. Charncy DS, Hcningcr GR: Efficacy of lithium -tranykypromine treatment in rclractory depression. Am I Psvchiatry 1985; 142:619-023.
9. Nolen WA. Van De Putte II, DiJ ken WA, et al: L-5HTP in depression resistant to re-uptake inhibitors: An open comparative study with tranvicyromine. Br J Psychiatry 1985: 147:16-22.
10. White K, Simpson G: Combined MAOliricyclic antidepressant treatment: A revaluation. J Clin Psychopharmacol 1481; 1:264-282.
11. White K. Simpson G: The combined use of MAOIs and tricyclics. iClin Psvchiatry 1984i 45:67-69.
12. Davidson J, McLeod M. Law- Yone B. et al: A comparison of electroconvulsive therapy and combined phcnclzineamitriptyline in refractory depression. Arch Gen Psychiatry 1978; 35:639-642.
13 Young IPR. Lader MH. Hughes WC: Controlled trial of trimipramine. monoamine oxidase inhibitors, and combined phenelzine-amitriptyline in refractory depression. Arch Gen Psychiatry 1979. 35:639-642.
14. White K. Pistole JL. Boyd IL: Combined MAOI-TCA vs. single antidepressant treatment: A pilot study. Am J Psychiatry 1980: 137:1422-1425.
15. Kazan: J. White K. O'Leary J. et al: The safety and efficacy of combined amitriptyline and iranylcypromine amidepressant treatment: A controlled trial. Arch Gen Psychiatry 1983; 40:657-661.
16. De Montigny C. Grunberg F. Mayor A. et al: Lithium induces rapid relief of depression in tricyclic antidepressanl drug non-responders. Br J Psychiatry 1981 : 138:252-256.
17. De Montigny C, Cournoyer G. Morisette R, el al: Lithium carbonate addition in tricyclic ant idepressant-resistant unipolar depression. Arch Gen Psychiatry 1983; 40:1327-1334.
18. Heninger GR. Charney DS. Sternberg DE: Lithium carbonate augmentation of antidepressanl treatment. .Arch Gen Psychiatry 1983; 40:1335-1342.
19. Nelson IC. M azure CM: Lithium augmentation in psychotic depression refractory to combined drug treatment. Am J Psychiatry 1986; 143:363-366.
20. Price LH. Charncy DS. Hcninger GR: Variability of response to lithium augmentation in refractory depression. Am J Psychiatry 1986; 143M387-1392.
21. Ramsey TA. Mendels I: Lithium ion as an antidepressant. in Enna SJ. Malick JB. Richelson E (eds): Antidepressants: Neurochemical, Bechavioral, and Clinical Perspectives. New York. Raven Press. 1981.
22. Karle BV: Thyroid hormone and tricyclic antideprcssants in resistant depressions. Am J Psychiatry 1981; 138:252-256.
23. Coppen A. Whybrow PC. Noguera R. et al: The comparative antidepressant value of L-tryptophan and imipramine with and without attempted potentiation by liothyronine. Arch Gen Psychiatry 1972; 26:234-241.
24. Goodwin FK. Prange AJ. Post RM. et al: Potentiation of antidepressant effects by L-triiodothyronine in tricyclic non res ponders. Am J Psychiatry 1982: 139:34-38.
25. Ogura C, Okuma T. Uchida Y. et al: Combined thyroid (triodothyronine)-tricyclic antidepressanl treatment in depressive states. Fulid Psychiaitr Neurol Jpn 1974: 28:179-186.
26. Prange AJ Jr. Wilson IC. Breese GR. et al: Hormona! alteration of imipramine response: A review, in Sachar EJ (ed): Hormones, Behavi r. and Psychopathology. New York, Raven Press, 1976.
27. Feighner IP. King LJ, Schtickit MA, et al: Hormonal potentiation of imipramine and ECT in primary epression. Am J Psychiatry- 1972: 128:"50-58.
28. Targum SD. Green berg RD. Harmon RL. et al: The TRH test and thyroid hormone in retract o ry depression. Am i Psychiatry 1984; 141:463.
29. D'Elia G, Hanson L. Raolma H: L-tryptophan and 5-hydroxytryptophan in the treatment of depression. A review. Acta Psychiatr Scand 1978; 57:239-252.
30. Van Praag HM: Management of depression with scrotonin precursors, Biol Psychiatry 1981: 16:291-310.
31. Wharton RN. Perel JM. Daytun PG. et al: A potential clinical use for methylphenidate with tricyclic antidepressants. Am J Psychiatry 1971; 127:55-61.
32. Poldinger W: Combined administration of desipramine and reserpine or tetrahenzinc in depressive patients. Psychophartnacolugia 1963; 4:308-310.
33. Haskovec L, Rybanek K: The action of reserpine in imipra mine-resistant depressive patients. Psvchupharmacologia 1967; 11:18-30.
34. Hopkinson G. Kenny F: Treatment with reserpine of patients resistant to tricyclic antidcpressanls. Psychiair ? a; 1975; 8:109-114,
35. Vogel W. Klaiber EL. Oro verm a n DM: Roles ol the gonadal steroid hormones in psychiatric depression in men and women. Prog Nettropsychopharmacol Biol Psychiatry 1978: 2:487-503.
36. Cole. 10, Schatzberg AF. Sniffin C. et al: Trazodonc in treatment-resistant depression: An open study. J Clin Psvchophanncol 1981; l(Sup'pl):49-54.
37. Post RM. Uhde TW. Roy-Hyrne PP, et al: Antidcpressant effects of carbamazepinc. Am J Psychiatry 1986: 143:29-34.
38. Davidson JRT. McLeod MN. Kurland AA. et al: Antidepressant drug therapy in psychotic depression, Br J Psychiatry 1977: 131:493-496.
39. Charney DS. Nelson JC: Delusional and nondelusional unipolar depression: Further evidence for distinct subtypes. Am J Psychiatry 1981: 138:328-333".
40. Spiker DG. Weiss JC, Dealy RS, et al: The pharmacological treatment of delusional depression. Am J Psychiatry 1985: 142:430-436.
41. Shaw DM. Hewland R: The management of resistant depression. Br J Psychiatry 1973: 123:489-492.
42. Shaw DM: The practical management of affective disorders. Br J Psychiatry 1977: 130:432-451.
43. Gemer RH : Systematic treatment approach to depression and treatment resistant depression. Psychiatric Annuls 1983; 13:37-49.
44. Simpson G, Whiff K: Clinical uses of monoamine oxidase inhibitors. Hosp Community Psychiatry I 982 : 3 3: 615-616.