Imipramine Treatment for Chronic Depression

Imipramine Treatment for Chronic Depression James H. Kocsis, MD; Allen J. Frances, MD; Carlyle Voss, MD; J. John Mann, MD; Barbara J. Mason, PhD; J...
Author: Tracey Hudson
2 downloads 0 Views 805KB Size
Imipramine Treatment for Chronic Depression James H.

Kocsis, MD;

Allen J.

Frances, MD; Carlyle Voss, MD; J. John Mann, MD; Barbara J. Mason, PhD; John Sweeney,

in the treatment of chronic depressystematic study. We used a two\x=req-\ week, single-blind placebo washout followed by a six-week, double-blind comparison of imipramine hydrochloride and placebo in a sample of 76 outpatients with DSM-III dysthymic disorder entered into a trial at two centers. Subjects were preponderantly female, had insidious onset at an early age, and had depressions of moderate severity; 96% also met the DSM-III criteria for major depressive disorder at the time of presentation. Sixty percent had a history of persistent depressive symptoms sufficient to meet criteria for major depression for longer than two years. Markedly favorable responses occurred in 45% of imipramine-treated (n 29) and 12% of placebo-treated (n 25) patients and, respectively, 59% and 13% of those who completed the study. Imipramine produced significant advantage in measures of depressive symptoms, global severity of illness, and self-rated social and vocational function. Recovered patients experienced remission from both long-standing symptoms and deficits as well as more recently exacerbated aspects of their syndrome. Patients with pure dysthymic disorder of a mild, subsyndromal type were uncommon in these clinical settings. However, anti\x=req-\ depressant medication was effective for many moderately severe chronic depressions, which had previously been untreated or undertreated, presumably related to misdiagnosis. (Arch Gen Psychiatry 1988;45:253-257) \s=b\

Antidepressant drugs

sions have received little



clinical conceptualization of chronic, "dysthymic," The"lifelong" depressive historically


states has fluctuated


view of such conditions as "affective" or bio¬ logically determined vs "characterologic" or psychological constructs.1 In recent years the pendulum has swung toward the classification of chronic depressions among the affective disorders, as reflected by the diagnostic category "dysthymic disorder" in the affective disorders section of the DSM-III.2 This shift has been accompanied by a new interest in determining the role of pharmacologie treat¬ ment, and several uncontrolled trials have suggested efficacy for antidepressant medications in some chronic forms of depression. Ward et al3 treated 15 patients, who met Research Diag¬ nostic Criteria (RDC) for major depressive disorder, with doxepin (up to 300 mg) on an open basis for four weeks. All subjects had been depressed for at least one year, although 11 had durations longer than two years and three longer a

Accepted for publication Oct 16, 1987. From the Departments of Psychiatry, New York Hospital\p=m-\CornellMedi-

cal Center (Drs Kocsis, Frances, Mann, Mason, and Sweeney), and Maine Medical Center, Portland (Dr Voss). Reprint requests to New York Hospital\p=m-\CornellMedical Center, 525 E 68th St, New York, NY 10021 (Dr Kocsis).


than ten years. Eighty percent of the sample experienced marked or moderate improvement on the Clinical Global Impression Scale during this study. A double-blind compar¬ ison of amitriptyline, perphenazine, and a combination of the two for treatment of outpatients with RDC major depressive disorder reported by Rounsaville et al4 included a subsample of 23 patients who also met RDC for intermit¬ tent depressive disorder. Improvement was significant and equivalent in the groups having major plus intermittent depression vs major depression alone. Àkiskal et al6 treated 50 "characterologic depressives," described as outpatients with chronic, mild depression of insidious onset, on an open basis using a variety of anti¬ depressant medications. After at least six months of treat¬ were categorized as responders or nonre¬ sponders by "global clinical criteria." Forty percent were reported to be responsive. Paykel et al6 randomly assigned 131 outpatients with mild-moderate depression to six weeks of double-blind treatment with phenelzine (up to 75 mg), amitriptyline (up to 187.5 mg), or placebo. A subset of 56 subjects (43%) met criteria of Klein and Davis7 for "neurotic dysphoria," described as "a pattern of long-standing or recurrent depression, without clear differentiation of epi¬ sodes from personality and with evidence suggesting de¬ pressive personality." No significant differences occurred for outcome measures in this subgroup between those patients taking amitriptyline and those taking placebo, both of whom improved considerably. Phenelzine-treated patients experienced significantly worse outcome than the

ment, patients

other two groups on most measures. Harrison et al8 re¬ cently presented results of a study in which 12 patients who met DSM-III criteria for dysthymic disorder and who had experienced favorable responses to phenelzine were ran¬ domly assigned to a continuation trial of phenelzine or placebo for six months. All seven patients randomized to receive placebo and only one of the five phenelzine-treated patients experienced relapse, a significant difference. These five studies vary considerably in diagnostic crite¬ ria, treatment strategies, degree of placebo control, and methods of outcome evaluation. Overall, however, the re¬ sults do suggest that some chronically depressed patients have been helped by treatment with antidepressant medica¬ tions. The current study sought to investigate this issue further by (1) selection of a sample of depressed patients based on the presence of chronicity of symptoms, (2) provision of an adequate and standardized trial of a tricyclic antidepressant (TCA), (3) employment of a placebo-con¬ trolled, double-blind design, and (4) rigorous definition of response to require relief from chronic or "dysthymic" symptoms of depression. We hypothesized that a TCA, imipramine hydrochloride, would be more likely to produce recovery than placebo.

Downloaded From: on 01/15/2017


Subjects Subjects were recruited among the adults applying for outpa¬ tient treatment for depression at the Payne Whitney Clinic, New York, and the Maine Medical Center, Portland. Initial screening was conducted by research psychiatrists using a checklist of

DSM-III symptoms of major depressive disorder and dysthymic disorder. Subjects were included if they (1) fulfilled DSM-III criteria for dysthymic disorder, ie, depressed or dysphoric mood

for at least two continuous years plus at least three associated symptoms; (2) had a Global Assessment Scale (GAS)9 score of 70 or less; (3) had a score on the 24-item version of the Hamilton

Depression Scale (HAM-D)10 of greater than 13; and (4) had given

signed informed consent. Patients were excluded if they had a history of bipolar disorder, ie, mania or hypomania, or "secondary depression" as indicated by a history of psychosis, alcohol or substance abuse, or severe or chronic medical illness. Also excluded were patients having a contraindication to imipramine or an apparently adequate trial of antidepressant medication within the past six months. The pres¬ ence of Axis I and Axis II disorders other than those already stated was systematically assessed but was not used to exclude patients from the study. Treatment Treatment was with imipramine hydrochloride (Tofranil), 50 mg, identical placebo tablets. All subjects initially received one placebo pill twice a day for two weeks on a single-blind basis. At the end of this period diagnostic and behavioral measures were re¬ peated and patients who still met inclusion criteria were ran¬ domized to a six-week, double-blind treatment with imipramine hydrochloride or placebo according to the following daily equiva¬ lent dose schedule: day 1 and 2, 100 mg; day 3 and 4, 150 mg; day 5 through 7, 200 mg; day 7 through 9, 250 mg; and day 10 through 42, 300 mg. Vital signs and side effects were checked at weeks 1 and 2, and patients intolerant to increased doses were stabilized at lower doses. An effort was made to hold doses constant after week 2. Concurrent psychotropic medications or psychotherapy were not allowed except for occasional nighttime sedation with a ben¬

tion of chronic dysthymic symptoms, from a more modest level of improvement that might be viewed as a phase in the natural history of chronic depression with a fluctuating course. Criteria by which patients were included in the responder group were (1) a total score of 6 or less on the 24-item HAM-D, (2) a ten-point or greater symptoms improvement on the GAS, and (3) absence of sufficient as derived to meet the DSM-III criteria for dysthymic disorder, from ratings on the dysthymic symptom checklist. Clinical ratings and diagnostic evaluations were performed by psychiatrists assigned to the affective disorder clinics of the two centers. All raters were trained, and joint rating sessions were held to establish and assess interrater reliability. Intraclass cor¬

relation coefficients were .93 for the HAM-D total score and .89 for the GAS. Nine rating sessions involving nine patients resulted in perfect interrater agreement for the presence of DSM-III dysthymic disorder (two patients positive, seven negative). Data


Clinical and demographic variables, diagnostic subtypes, and treatment response were first compared for the samples from New York and Maine. With few exceptions (see "Results" section) these variables did not show significant differences; therefore, further sample. Outcome analyses were undertaken using the combined for differences in for the two treatments was analyzed using 2 analysis of covariance categorical response and repeated-measures and t tests for differences in scale scores.



Clinical Measures Baseline diagnostic evaluations included a clinical interview by a trained research psychiatrist, which was conducted using a check¬ list of DSM-III symptoms for affective disorders. Durations of

each symptom of major depressive and dysthymic disorder were included in this assessment. Other evaluations included an HAMD, GAS, DSM-III Axis IV rating, and Atypical Depression Diagnosis Scale. The criteria for atypical depression by Liebowitz et alu include mood reactivity, reverse vegetative symptoms (hypersomnia, hyperphagia), anergy, and rejection sensitivity. The intake interview included systematic inquiries about prior psychotherapy and antidepressant medication. Patients were rated as having had prior psychotherapy if they had been seen by a therapist at least weekly for six months or longer, and as having had adequate TCA treatment if they had received at least three weeks of 150 mg of imipramine hydrochloride or its equivalent. Repeated measures were done every two weeks and included a checklist oí DSM-III dysthymic symptoms, the HAM-D, GAS, and a side-effect checklist. The self-rated version of the Social Adjust¬ ment Scale (SAS-SR)·2 was administered before and again at the end of the double-blind period. On this form we asked subjects to rate their social functioning over the past two weeks on five-point scales for items measuring work, family, and leisure functioning. The total score represents an averaging of scores on eight subscales. The SAS-SR was completed by a consecutive subset of subjects (13 taking imipramine, 14 placebo) who entered after the form was introduced during the course of the study. Based on our clinical experience and prior literature on chronic depression, we believe such patients usually presented for treat¬ ment at a peak of symptom severity in a waxing and waning clinical course. This phenomenon has been referred to as "double-depres¬ sion,"13*14 when episodes of major depressive disorder are superim¬

posed on chronic dysthymia. Therefore, we developed a stringent definition of response to treatment, to distinguish through résolu-

RESULTS Patient Characteristics

Table 1 gives the demographic, clinical, and diagnostic measures for groups completing treatment with imipramine and placebo as well as noncompleters. Differences between the two treatment groups were not statistically significant. Noncompleters were characterized by older age at onset, shorter duration of illness, and lower average HAM-D total scores. Failure to complete the singleblind placebo period or enter the double-blind stage occurred for the following reasons: (1) improvement such that severity criteria for randomization were no longer met (n ll); (2) lack of cooperation or failure to return for appointments (n = 6); (3) both improvement and lack of cooperation (n = 2); and (4) condition worsened requiring active treatment (n 3). Dropouts subsequent to randomization occurred in seven imipramine- and one placebotreated patient. Two imipramine-treated patients were dropped because of severe medication reactions (urinary retention, gener¬ alized skin rash). Five imipramine-treated patients became noncompliant for unknown reasons before receiving four weeks of treatment. One placebo-treated patient worsened and required active medication. Dropouts occurred in the imipramine-treated group at a significantly greater rate than in the placebo-treated group ( 2 4.3, Pplacebo imipramine, =.05. =




-¡•Imipramine vs noncompleters,



Liebowitz et alu for atypical depression. Twenty-eight (37%) were given an additional DSM-III Axis II diagnosis by the treating psychiatrist. The most frequent types of personality disorders were atypical-mixed (n 11) and dependent (n 10). Severity of depressive illness was measured by the 24-item version of the HAM-D and by the GAS. Average depressive severity was in the moderate range as rated on the HAM-D and was normally distributed in all groups. The GAS score was rated in the mild to moderate range (51 to 70) in 65 cases (85% of the sample), whereas the remaining cases were rated as "serious" (n 8) or "major" (n 3) in overall severity. Inquiry about prior treatment revealed that 54 (71%) of the patients had received prior psycho¬ therapy, whereas only six (8%) had ever received an adequate trial of antidepressant medication, and 25 (33%) had ever received any antidepressant drug treatment. Forty-seven patients (62%) were seen in New York and 29 (38%) in Maine. Sex distribution, duration of illness, prior psychiatric treatment, proportion of patients with atypical depression or Axis =


2.—Categorical Outcome in Completers by Treatment and Center




II diagnoses, and SAS-SR and HAM-D total scores at entry were not significantly different for the samples studied at the two centers. Maine patients were significantly younger (34 ±11 years and 42 ±10 years for Maine and New York, respectively; .01) and had less severe GAS scores (58 ±7 and 53 ±8, respectively;


Imipramine Hydrochloride (n 22) Nonre-

Center New York*

Mainef Total, No. (%)t



Results of Double-blind Treatment

Forty-six patients completed double-blind treatment and could be classified as responders or nonresponders. Fifty-nine percent of 22 imipramine-treated and 13% of 24 placebo-treated completers were categorized as responders (Table 2). Rates of response were quite similar at the two treating centers. Average ( ± SD) daily equivalent peak doses were 198 ±59 mg/d for imipramine and 247 ±42 mg/d for placebo. Seventeen (78%) of 22 imipraminetreated and 18 (75%) of 24 placebo-treated patients were actually maintained on a constant dose from week 2 to week 6. Average plasma drug concentration at steady state (imipramine and desipramine) was 276 ± 116 ng/mL. Because the dropout rate in the group randomized to receive imipramine (7/29) exceeded the rate in the placebo-treated group (1/25), the 2 analysis was repeated classifying all dropouts as nonresponders. Response rates to


(n 24) =



spender 13 8










Fisher's exact test, P= .02. fFisher's exact test, P= .05. *X2=11, =.001.

Table 3.—Week 6 HAM-D Total Scores

Among Completers*

Imipramine Hydrochloride





Score 0-6




13-18 19-24

25-30 *HAM-D indicates Hamilton

Depression Scale.

imipramine (13/29, 45%) vs placebo (3/25, 12%) remained signifi¬ cantly different ( 2 6.9, P

Suggest Documents