Effects of magnesium sulfate and lidocaine in the treatment of ventricular arrhythmias in experimental amitriptyline poisoning in the rat

OBJECTIVES: Amitriptyline poisoning is associated with ventricular arrhythmias. Standard treatment is sodium bicarbonate but further intervention may be necessary. The present study compared the actions of lidocaine and magnesium sulfate on ventricular tachycardia induced by amitriptyline. DESIGN: Nonrandomized, controlled, intervention trial. SETTING: University laboratory. SUBJECTS: Thirty male Wistar rats anesthetized with pentobarbital and mechanically ventilated. INTERVENTIONS: After pretreatment with norepinephrine, the animals were subjected to a continuous infusion of amitriptyline. After the appearance of ventricular tachycardia, they were treated with magnesium sulfate (45 mg/kg + 15 mg/kg/min) or lidocaine (1 mg/kg + 0.5 mg/kg/min) or glucose infusion as a control. MEASUREMENTS AND MAIN RESULTS: In the group treated with magnesium sulfate, electrocardiogram tracings demonstrated that nine of ten animals converted from ventricular tachycardia to sinus rhythm compared with one of ten in both the lidocaine- and glucose-treated groups (p < .001). The animals treated with magnesium sulfate also had a significantly longer total time in sinus rhythm (10.0 +/- 1.6 mins) than those rats treated with lidocaine (1.7 +/- 1.5 mins) or glucose (1.5 +/- 1.5 mins). Magnesium sulfate significantly decreased blood pressure and heart rate, but no severe hemodynamic side effects were observed. CONCLUSIONS: Magnesium sulfate is effective in converting ventricular tachycardia in hyperadrenergic amitriptyline poisoning. In contrast, lidocaine had no effect on arrhythmias.     

Relationship between age and tricyclic antidepressant plasma levels

Older depressed patients treated with imipramine or amitriptyline developed higher steady-state plasma levels of imipramine, desipramine, and amitriptyline. In imipramine-treated patients this finding was associated with a decreased rate of drug elimination from plasma. These findings provide at least a partial explanation for the increased susceptibility of the older patient to tricyclic antidepressant side effects and also provide a pharmacological rationale for use of lower dosages in this age group.     

Electrocardiographic findings in patients undergoing amitriptyline treatment

Fifteen depressed patients were treated with amitriptyline in a dosage ranging from 75 mg to 200 mg/day for a minimum of three weeks, two of which were at a fixed dosage. Plasma samples were drawn 12 to 16 hours after the bedtime dose of medication for determination of amitriptyline and its metabolite nortriptyline. Electrocardiograms were taken prior to treatment and after three weeks of drug treatment. The only abnormality noted prior to treatment was bradycardia in one patient. After treatment, two patients exhibited nonspecific T-wave abnormalities. Otherwise, the electrocardiograms were unremarkable. A significant increase in heart rate (p less than 0.001) was noted. The mean increase in rate was sixteen beats per minute. Those patients having a rate change greater than sixteen beats per minute had significantly higher amitriptyline levels (p less than 0.05), and total tricyclic antidepressant levels (p less than 0.05), than those patients having a mean rate increase less than sixteen. While tricyclic antidepressants can produce multiple cardiac effects, the risk of cardiac morbidity and mortality is relatively low in patients undergoing tricyclic antidepressant treatment with moderate dosages.     

Chondrosarcoma of the thoracic wall

In this paper, three placebo-controlled trials of moclobemide, and four three-way comparison trials of moclobemide, placebo and one of the standard tricyclic antidepressants, imipramine, clomipramine, or amitriptyline, are reviewed. The results of the placebo-controlled trials indicate that about twice as many of the patients receiving moclobemide showed a marked improvement during a four-week treatment period, as compared with patients receiving placebo. Of the three-way trials, two studies indicated that the efficacy of moclobemide was significantly greater than that of placebo and similar to that of other antidepressants (imipramine or amitriptyline). Tolerance of moclobemide was good and better than that of the tricyclic antidepressants. In the other three-way comparison trials, the differences in efficacy between the placebo, moclobemide, and the tricyclic antidepressant were not significant.     

Management of gynecologic oncology patients with a preoperative deep vein thrombosis

Study of the effect of a tricyclic antidepressant amitriptyline on an isolated papillary muscle of the rat left ventricle showed a decrease of the contractility due to depression of the potential of action and the slow calcium channels, as well as decreased rate and prolongation of contraction caused by impairment of calcium reabsorption from the sarcoplasma. Injection of dobutrex after amitriptyline has a positive inotropic effect by completely removing the disorders in the time parameters of contraction.     

Different effects of nifedipine and amlodipine on circulating catecholamine levels in essential hypertensive patients

OBJECTIVE: To compare the acute and chronic effects of nifedipine retard (NPA), nifedipine gastrointestinal therapeutic system (NGITS) and amlodipine at trough and peak plasma concentrations of drug on blood pressure and heart rate, and on plasma norepinephrine and epinephrine levels in patients with mild-to-moderate hypertension (diastolic blood pressure 95-115 mmHg). DESIGN AND METHODS: After 3-4 weeks' placebo treatment, patients of both sexes were randomly allocated to be administered 10 or 20 mg NPA twice a day, 30 or 60 mg NGITS once a day, and 5 or 10 mg amlodipine once a day for 6 weeks. Initially, for the first 2 weeks, the lowest dose of each drug was used, but higher doses were administered after 2 weeks if sitting diastolic blood pressure was > 90 mmHg. Patients were evaluated after administration of the first dose and after 6 weeks' therapy in a hospital setting. Blood samples were taken for high-performance liquid chromatography measurement of catecholamine and drug levels at various intervals for a period covering trough to peak drug level ranges. RESULTS: Administration of all three drugs reduced clinic blood pressure to the same level after 6 weeks' therapy, but heart rate was increased slightly only with amlodipine (P < 0.05). Administration of NPA reduced blood pressure more abruptly whereas administrations of NGITS and amlodipine induced smoother falls after acute and chronic treatments: a significant increase in heart rate was observed with amlodipine after chronic treatment. Both acute and chronic treatments with NPA (n = 19) increased norepinephrine levels (P < 0.01) transiently (2-4 h). In contrast, administration of NGITS (n = 22) did not increase norepinephrine levels and even induced a slight but significant decrease in norepinephrine levels 5-6 h after chronic treatments. Although administration of amlodipine (n = 22) did not increase norepinephrine levels transiently either after acute or after chronic administration, it did induce a sustained rise in basal norepinephrine levels by more than 50% after chronic therapy (P < 0.01). Plasma epinephrine levels were not increased by any of the treatments and even a slight decrease was observed 4 h after administration of a dose following chronic treatments with NGITS and amlodipine (P < 0.05). CONCLUSIONS: The transient increase in norepinephrine levels observed with NPA and the sustained increases in norepinephrine levels observed after chronic treatment with amlodipine suggest that sympathetic activation occurs with those two drugs. The lack of increase in norepinephrine levels after administration of NGITS suggests that this formulation does not activate the sympathetic system. The lowering of epinephrine levels after administrations of NGITS and amlodipine suggests that inhibition of release of epinephrine by the adrenal medulla occurs with longer-acting dihydropyridine formulations.     

Fluoxetine

Fluoxetine was developed as an antidepressant drug. It is more effective than placebo, but a dose-effect relation has not been established. Fluoxetine is almost as effective as tricyclic antidepressant drugs, but the available studies do not allow accurate comparisons. Fluoxetine may be less effective than tricyclic antidepressant drugs for the treatment of inpatients with severe melancholic depression, and it should not be the first choice of a drug for them. Fluoxetine may be most appropriate for patients with moderate depression who can be treated as outpatients. If there is little improvement after treatment for four to six weeks, an alternative treatment should be offered. Fluoxetine does not have the anticholinergic, hypotensive, and sedative effects of tricyclic antidepressant drugs and has no particular cardiovascular effects; overdoses do not cause serious toxic effects. Nausea, anorexia, insomnia, and nervousness--the most common side effects--may be controlled with a careful adjustment to the dose. Clinically important drug interactions may occur with monoamine oxidase inhibitors, tricyclic antidepressant drugs, and other drugs. The published data on the antidepressant effect of fluoxetine do not fully explain its popularity. One may speculate that fluoxetine has psychobiologic effects not strictly related to the biology of depression and that it acts primarily as a mood- or affect-modulating agent.     

Milnacipran. A review of its use in depression

Milnacipran is a cyclopropane derivative which acts by inhibiting noradrenaline (norepinephrine) and serotonin (5-hydroxytryptamine; 5-HT) reuptake at presynaptic sites; no postsynaptic receptor activity has been demonstrated. It is most commonly administered at a dosage of 50 mg twice daily for the treatment of major depressive disorder. Improvement usually occurs within 2 weeks of treatment initiation, but some patients do respond sooner. Most studies which evaluated milnacipran were of short (4 to 8 weeks) duration and results were not published in full with rigorous peer review. Nonetheless, the drug is significantly more effective than placebo for the treatment of in- or outpatients with moderate to severe major depressive disorder. Limited data suggest that it may prevent relapse and be effective for long term use, although this requires confirmation. Milnacipran 200 mg/day is generally not significantly different from amitriptyline 150 mg/day in terms of onset and efficacy. However, when doses are titrated (not a requirement for milnacipran), milnacipran 50 or 100 mg/day has a slower onset than the tricyclic antidepressant. At a dosage of 100 mg/day for 4 to 12 weeks, milnacipran generally has similar efficacy to imipramine and clomipramine 150 mg/day, although milnacipran 50 to 150 mg/day had a faster onset of activity than imipramine 50 to 150 mg/day in Japanese patients. In a 6-month trial, milnacipran was less effective than clomipramine. Milnacipran 50 or 100 mg twice daily was as effective as fluoxetine 20 mg once daily or fluvoxamine 100 mg twice daily in 4- to 12-week studies. At a dosage of 50 then 100 mg daily it was also as effective as mianserin 30 then 60 mg daily in a 4-week study. However, when administered once daily (in the evening), milnacipran 100 mg/day was not as effective as fluoxetine 20 mg/day after 6 weeks. The drug is generally well tolerated, producing no more adverse events (including anticholinergic events) than placebo, selective serotonin reuptake inhibitors or mianserin and fewer adverse events than tricyclic antidepressants in clinical trials. However, dysuria has been reported in 7% of male patients receiving milnacipran. CONCLUSIONS: Data from predominantly short term trials suggest that milnacipran generally has similar efficacy to tricyclic antidepressants and SSRIs. Although further published data are required to confirm its efficacy, good tolerability profile and pharmacokinetic profile which suggests a low potential for drug interactions, milnacipran should be considered a promising agent for the treatment of patients with major depressive disorder.     

Effects of oral clonidine premedication on plasma glucose and lipid homeostasis associated with exogenous glucose infusion in children

BACKGROUND: Oral clonidine may influence plasma glucose and lipid homeostasis by modulating endocrinologic responses to surgical stress. The effect of oral clonidine premedication on plasma glucose and lipid homeostasis associated with exogenous glucose infusion were investigated in children undergoing minor surgery. METHODS: Otherwise healthy children (n, 120; aged 3-13 yr) were assigned randomly to six groups according to the glucose concentration of the intravenous solution (0%, 2%, or 5%, at a rate of 6 ml kg(-1) x h(-1)) and the preoperative medications (4 microg/kg clonidine or placebo given 100 min before anesthesia) they were to receive. The plasma concentrations of glucose, nonesterified fatty acid, ketone bodies, epinephrine, norepinephrine, and cortisol were determined. RESULTS: Infusion of 5% glucose caused hyperglycemia (mean glucose concentration >200 mg/dl) in six children receiving placebo and two receiving clonidine. Although the mean plasma glucose concentration increased in three placebo groups, it was unchanged and the plasma concentrations of total ketone bodies and nonesterified fatty acid were increased in children receiving clonidine and glucose-free solution. The plasma epinephrine, norepinephrine, and cortisol levels in children receiving placebo increased in response to surgery. Clonidine attenuated the increase in catecholamines and cortisol. CONCLUSIONS: Oral clonidine premedication attenuated the hyperglycemic response, probably by inhibiting the surgical stress-induced release of catecholamines and cortisol. Infusion of 2% of glucose maintained plasma glucose concentrations within physiologic ranges in children receiving clonidine.     

Basal and stimulated sympathetic responses after epinephrine: no evidence of augmented responses

Delayed facilitation of norepinephrine release through the action of epinephrine (NE) at presynaptic beta-adrenoceptors has been postulated to account for the delayed hemodynamic effects of epinephrine and to be a mechanism causally related to the development of hypertension. To determine whether a short-term increase in epinephrine concentrations resulted in subsequent facilitation of sympathetic responses, 9 healthy subjects (age, 21+/-0.9 years) were studied at rest and during physiological stress on 2 occasions when they received an infusion of either saline or epinephrine (20 ng/kg per minute) in random order. Heart rate, blood pressure, forearm blood flow, epinephrine concentrations, and NE spillover were measured at rest, during mental stress (Stroop test), and during a cold pressor test. Measurements were performed before, during the 1-hour infusion of epinephrine or placebo, and 1 hour after the infusion. A radioisotope dilution method was used to measure NE spillover. Hemodynamic measurements and NE spillover were increased during the infusion of epinephrine, but 1 hour after discontinuation of epinephrine there was no significant augmentation of hemodynamic or sympathetic responses. NE spillover 1 hour after saline or epinephrine infusion was similar (0.85+/-0.2 versus 0. 87+/-0.2 microg/min; P=0.92). In addition, there was no delayed facilitation of stress-induced hemodynamic or NE responses after epinephrine. These findings do not support the hypothesis that epinephrine results in delayed facilitation of NE release.     

Effects of viloxazine, an antidepressant agent, on biogenic amine uptake mechanisms and related activities

The effects of viloxazine, a clinically effective antidepressant, on noradrenaline (NA) and 5-hydroxytryptamine (5-HT) uptake and various related pharmacological activities were determined and compared to those of the tricyclic antidepressants desimipramine, imipramine, and amitriptyline. Viloxazine inhibitied [3H]NA uptake in the mouse and rat heart, being maximally about one half as potent as imipramine with a similar onset, but shorter duration of action than imipramine. The drug did not inhibit [3H]NA uptake in rat medulla or hypothalamus in contrast to desimipramine and imipramine, but it did alter [3H]NA metabolites in a similar manner. Viloxazine, like desimipramine, was a weak blocker of mouse brain 5-HT uptake, but differed from desimipramine as it poteniated 5-HT-mediated functions in the mouse and rat, as did imipramine and amitriptyline, the latter drugs being relatively potent blockers of 5-HT uptake. Viloxazine potentiated the L-DOPA behavioural syndrome in the mouse, antagonized reserpine-induced ptosis and hypothermia in the mouse, and inhibited gastric acid secretion in the rat, but was less potent than the tricyclic antidepressants. No appreciable in vivo inhibition of monoamine oxidase (EC 1.4.3.4.) activity in the mouse was exhibited. Like imipramine, the drug potentiated the ocular effects of L-adrenaline in the rabbit. It was similar to imipramine in potency in potentiating the apomorphine-induced gnawing in the mouse. The drug antagonized oxotremorine-induced hypothermia in the mouse but differed from the tricyclic antidepressants in not exhibiting the anticholinergic effects of blocking the tremors, salivation and lacrimation. Thus, viloxazine exhibits activities related to the biogenic amines both similar to and different from the tricyclics desimipramine, imipramine, and amitriptyline. These actions appear to be of relevance with respect to the antidepressant action of this drug.     

Dosing of antidepressants--the unknown art

Data from a therapeutic drug monitoring service, in total 2,393 observations in 1,606 patients, were used to analyze factors associated with the prescribed daily doses of the tricyclic antidepressants amitriptyline and nortriptyline. The achieved concentrations in plasma were evaluated in relation to suggested therapeutic ranges. The doses of both drugs were greatly reduced with increasing age, despite the fact that age is of minor importance for their kinetics. Interactions with concomitantly given drugs were not handled by dose adjustments of the antidepressant. Therapeutic drug monitoring increased the proportion of concentrations within the therapeutic range for patients on amitriptyline, but not for those on nor-triptyline. The large interindividual kinetic variability for most antidepressants requires individualized dosing, but this individualization is performed on incorrect grounds.     

The influence of reserpine on the responsiveness to norepinephrine and electrolyte contents of rabbit atria

This study was designed to investigate the relationships between reserpine-induced supersensitivity and electrolyte levels in isolated rabbit atria. Atria from reserpine-treated (1, 3 or 4 mg/kg 24 hr before) and untreated rabbits were placed in an isolated organ bath for determination of dose-response relationships for norepinephrine, or were used for electrolyte assay. Reserpine-treated atria had greater resting and maximum rates of beat than did untreated atria. However, the ED50 concentrations of norepinephrine producing a rate response were similar. Atria pretreated with 1 or 3 mg/kg reserpine were supersensitive to the inotropic effect of norepinephrine. 4 mg/kg reserpine did not induce inotropic supersensitivity. Atrial sodium contents were significantly increased by 3 mg/kg reserpine. Potassium contents were increased by 1 and 4 mg/kg reserpine. Calcium contents were significantly decreased by 4 mg/kg reserpine, while magnesium contents were increased by 1 and 4 mg/kg. Reserpine induced supersensitivity to the inotropic effects of norepinephrine but did not alter the sensitivity to the chronotropic effect of the drug. This supersensitivity may be linked to alterations in the tissue's homeostatic mechanisms for calcium.     

Hemodynamic effects of epinephrine, bicarbonate and calcium in the early postnatal period in a lamb model of single-ventricle physiology created in utero

OBJECTIVES: A reproducible fetal animal model of single-ventricle physiology was created to examine the effects of pharmacologic agents commonly used in the perinatal and perioperative intensive care management of patients with a single ventricle. BACKGROUND: Single-ventricle physiology is characterized by parallel pulmonary and systemic circulations, with effective blood flow to each determined by the relative resistances in the pulmonary and systemic vascular beds. Perinatal and perioperative management of these patients is largely based on empiric observations and differs considerably between institutions and is further complicated by the transitional physiology of the newborn. The lack of animal models of single-ventricle physiology has hindered the understanding of this problem. METHODS: A 10-mm, Damus-Kaye-Stansel-type aortopulmonary anastomosis was created in 10 fetal sheep at 140 +/- 1.2 days of gestation. The main pulmonary artery was ligated distally, and pulmonary blood flow (Qp) was provided through a 5-mm aortopulmonary shunt. Eight lambs were delivered at term and placed on cardiopulmonary bypass (30 min) 48 to 72 h after birth. Pharmacologic interventions (0.1 microgram/kg body weight per min of epinephrine, 2 mEq/kg of sodium bicarbonate and 10 mg/kg of calcium chloride) were performed before and after bypass, and hemodynamic responses were observed. The response to the epinephrine bolus was determined only in the postbypass study. RESULTS: Both before and after bypass, epinephrine infusion and calcium and bicarbonate administration increased Qp and systemic blood flow (Qs) (total cardiac output) but produced only small changes in the Qp/Qs ratio (-0.5% to -7.3% change). With the epinephrine bolus, Qp increased enormously, and the Qp/Qs ratio increased by 584% (p < 0.001). CONCLUSIONS: In neonatal lambs with single-ventricle physiology created in utero, epinephrine infusion and calcium and bicarbonate administration increased total cardiac output without significantly compromising the Qp/Qs ratio. However, epinephrine bolus seems to be hemodynamically detrimental in circumstances of single-ventricle physiology and should be used with caution and probably in relatively lower doses in the resuscitation of patients with single-ventricle physiology. Further investigation of the dose-dependent effects and the effects of prolonged administration of common pharmacologic agents will enable better management of patients with single-ventricle physiology.     

Hemoperfusion with coated activated charcoal for treating acute poisoning by remedies, plant protectants, and fungi. I. Remedies (author's transl)

Hemoperfusion with coated activated charcoal is a novel procedure for treating acute poisoning. It enables the elimination of both, water-soluble and liposoluble toxins. Hemoperfusion with coated activated charcoal has proved to be superior to hemodialysis in the treatment of barbiturate or bromocarbamide poisoning both under experimental conditions as well as in the ward. Analogous statements may be made for the therapy of glutethimide poisoning. Methaqualone, on the other hand, could not be eliminated sufficiently well in animal trials. Intoxications by "mild" analgetics, such as paracetamol and acetylsalicylic acid, may be treated successfully with hemoperfusion. Treatment of acetylsalicylic acid poisoning is equally effective with hemoperfusion as with hemodialysis. Prospects for the success of hemoperfusion in treating intoxication from tricyclic antidepressants and neuroliptics are slight. It is simply the danger of antidepressant poisoning that justifies using this method of treatment in the first few hours after ingestion in order to reduced the flow of the psychopharmaceutical substance into the tissue.     

Novel environment induced inhibition of corticosterone secretion: physiological evidence for a suprachiasmatic nucleus mediated neuronal hypothalamo-adrenal cortex pathway

The effects of two cognition enhancers on avoidance impairment induced by the tricyclic antidepressant amitriptyline were assessed during shuttle-box avoidance acquisition and in previously trained mice of the DBA/2 strain. The nootropic agent piracetam (50, 100 or 200 mg/kg, i.p.) had slight or no effect in mice receiving amitriptyline (5 or 10 mg/kg, i.p.). Conversely, the acetylcholinesterase inhibitor tacrine (0.5, 1, 2 or 3 mg/kg, i.p.) prevented the avoidance impairment induced by 5 mg/kg amitriptyline on shuttle-box avoidance acquisition as well as on a previously learned avoidance response. The avoidance disrupting action produced by 10 mg/kg of the antidepressant drug was not affected by the anticholinesterase drug. The preventing action of tacrine seems specifically related to the avoidance impairment induced by amitriptyline, since the acetylcholinesterase inhibitor did not reduce, but enhanced the avoidance impairing action of the neuroleptic chlorpromazine. Taken together, the results indicate that amitriptyline-induced avoidance impairment, and the related preventing action of tacrine, may be ascribed to drug effects on the performance of the avoidance response, rather than to interferences with learning processes.     

Paroxetine. A review of its pharmacology, therapeutic use in depression and therapeutic potential in diabetic neuropathy

Paroxetine is a selective serotonin reuptake inhibitor effective in a once-daily administration regimen in the treatment of depression. In elderly patients (aged > or = 60 years) with major depression, short term (6 weeks) treatment with paroxetine produces clinical improvements significantly superior to those seen with placebo and similar to those with tricyclic antidepressant agents, mianserin and fluoxetine. There is evidence that paroxetine has positive effects on co-existing anxiety and does not precipitate agitation. Paroxetine has also shown potential in the symptomatic treatment of diabetic neuropathy; however, further clinical experience is needed to confirm this preliminary result. Short term paroxetine therapy is associated with fewer anticholinergic and CNS adverse effects, but generally more gastrointestinal disturbances, than tricyclic antidepressants and mianserin. Unlike the tricyclic agents, paroxetine does not significantly affect cardiovascular function or impair psychomotor performance. This tolerability profile should be particularly beneficial in elderly patients, who are generally more susceptible than younger patients to the anticholinergic and CNS adverse events associated with tricyclic antidepressant drugs, and in whom there is a higher prevalence of pre-existing cardiovascular disease. It also suggests an important potential advantage over tricyclic antidepressants in the setting of overdosage. Thus, primarily because of its better tolerability profile and potentially lower toxicity in overdosage and in patients with cardiovascular disease, paroxetine appears to be a more attractive option than tricyclic antidepressants for the treatment of depression in late life. Future research should attempt to define more fully the efficacy of paroxetine as long term prophylactic therapy for recurrent depression and to assess how its overall therapeutic profile compares with other selective serotonin reuptake inhibitors in the elderly.     

Treatment of depression in bipolar disorder: new directions for research

The objective of the study was to review the clinical literature on the acute, somatic treatment of the depressed phase of bipolar disorder. We reviewed all available published studies of "standard" somatic treatments (lithium, antidepressant and anticonvulsant agents, and electroconvulsive therapy [ECT]) reporting three or more depressed bipolar patients who were not psychotic, rapid cycling, or previously treatment refractory. We also reviewed all studies of "nonstandard" pharmacologic treatments involving even a single case of a depressed bipolar patient. Data sources included the MEDLINE database and relevant references from articles obtained in this search and in major reviews. Five of seven studies comparing ECT with antidepressant agents find ECT more efficacious. Eight of nine controlled comparisons find lithium superior to placebo in depressed bipolar patients. Three controlled comparisons of lithium to tricyclic antidepressants suggest that lithium is equivalent to tricyclic drugs in such patients. Three double-blind, controlled studies indicate that carbamazepine is more effective than placebo. Limited data on other antidepressant classes suggest that monoamine oxidase inhibitors, bupropion, and serotonergic agents may offer some advantages over tricyclic antidepressants in this population. Some "nonstandard" treatments also show some potential in bipolar patients. The possibility of switching into a manic episode is an important consideration with many of the agents studied, although little remains known about spontaneous versus treatment-associated mood shifts. In contrast to the extensive literature on the acute treatment of the manic phase of bipolar disorder and on the prophylaxis of manic and depressive episodes, there are few studies of treatment of the depressed phase of bipolar disorder, and their results generally are limited or inconclusive. Lithium generated a revolution in psychiatric treatment, but the treatment of the depressed phase of bipolar disorder remains a relatively neglected corner of the field. Several study designs may help to augment knowledge in the treatment of bipolar depression.