Response to Desipramine Treatment in Adolescent Depression: A Fixed-Dose, Placebo-Controlled Trial

Response to Desipramine Treatment in Adolescent Depression: A Fixed-Dose, Placebo-Controlled Trial STAN KUTCHER, M.D., CAROLYN BOULOS, M.D., BRIDGETTE WARD, R.N., PETER MARTON, PH.D., JOVAN SIMEON, M.D., H. BRUCE FERGUSON, PH.D., JOHN SZALAI, PH.D., MARKO KATIC, B.A., NASREEN ROBERTS, M.D., CHANTAL DUBOIS, B.A., AND KENTON REED, PH.D.

ABSTRACT Objective: To determine the efficacy and tolerability of the tricyclic antidepressant desipramine (DMI) in the treatment

of DSM-III-R-diagnosed major depressive disorder in adolescents. Method: Sixty adolescents (42 female, 18 male; aged 15 to 19 years) diagnosed with major depressive disorder using clinical interview and Schedule for Affective Disorders and Schizophrenia for School-Age Children were randomized to receive either DMI (200 mg daily in divided doses) or placebo for six consecutive weeks following a 1-week placebo period. Treatment outcome was determined using the Hamilton Depression Rating Scale and the Beck Depression Inventory. Tolerability was determined using a symptom side effects scale. In addition, a variety of laboratory and cardiovascular monitoring was performed. Results: No significant differences in treatment outcome between DMI- and placebo-treated groups were determined. Neither DMI, nor its metabolite 2-hydroxy-DMI, nor their ratio, was positively correlated to treatment outcome. The DMI group endorsed more side effects but there were no significant between-group differences in any laboratory, electrocardiographic, or other cardiovascular parameters apart from heart rate, which was increased in the DMI-treated group (p

=

.03). Conclusions: Given the findings of this study and our review of previously published reports of tricyclic antidepressant treatment in this population, the routine use of short-term (6 weeks) DMI in the treatment of adolescent depression is not supported by the data on hand. Further investigations into what constitutes optimal psychopharmacological treatment of adolescent depression are warranted. J. Am. Acad. Child Ado/esc. Psychiatry, 1994, 33, 5:686-694. Key Words: adolescent, depression, desipramine, treatment.

Adolescent depression is a serious psychiatric disorder with significant morbidity and associated high rate of mortality, especially through suicide (Brent et al., 1988; Kutcher and Marton, 1989; Robins and Alessi, 1985;

Accepted January 18, 1994. From the Sunnybrook Health Science Centre, North York, Ontario, Canada; University of Toronto, Ontario; University of Ottawa, Ontario; Institute of Mental Health Research; Clarke Imtitute ofPsychiatry, Limetrees, York; Royal Ottawa Hospital,' and Queen Street Mental Health Centre. This study was supported in part by a grant from Nordic Merrell Dow Research and by an Ontario Mental Health Foundation Senior Scientist Award to the principal investigator (S.K). Data base development for this project was supportedin part by agrantfrom the Canadian Psychiatric Research Foundation. Our thanks to Betty Rychlewski, Jill Green, David Gardner, and Patrick Abtan for secretarial and logistical support. Reprint requests to Dr. Kutcher, Division ofAdolescent Psychiatry, Room FG60, Sunnybrook Health Science Centre, 2015 Bayview Ave., North York, Ontario M4N 3M5. 0890-8567/94/3305-0686$03.0010©1994 by the American Academy of Child and Adolescent Psychiatry.

686

Ryan and Puig-Antich, 1986). Although major depressive disorder (MDD) may have prepubertal onset, the teenage years are associated with an increased prevalence of the disorder, which reaches up to 8% of the teenage population by approximately age 19 (Angold, 1988; Fleming and Offord, 1990; Kashani et al., 1987b). Although a variety of treatment modalities have been directed toward ameliorating depressive symptoms in this population, no clear evidence for the efficacy of any specific treatment has been clearly established (Kutcher and Marton, 1990). Tricyclic antidepressant (TeA) treatment is well established as an effective treatment for depression in adult populations (Baldessarini, 1989; Morris and Beck, 1974; Nelson et al., 1988; Quitkin et al., 1984). However, adult findings cannot be directly extrapolated to adolescent populations (Kutcher and Marton, 1990) and thus, tricyclic antidepressants need to be empirically evaluated in adolescents as well.

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 33:5, JUNE 1994

DESIPRAMINE IN ADOLESCENT DEPRESSION

To date, published studies of tricyclics in depressed adolescents are few and have often consisted of open clinical trials. Ryan et al. (1986) treated 34 depressed adolescent outpatients with imipramine (1M I) with dosage titrated to a maximum of 5 mg/kg per day. Mean plasma level of IMI and desipramine (OMI) of 284 ng/mL was achieved. Forty-four percent were considered to show a positive response. Similarly, Strober et al. (1990) reported on 35 depressed adolescent inpatients treated with IMI (mean plasma level of IMI and OMI of 237 ng/mL). Thirty-eight percent of teenagers with nondelusional MOD were deemed responders. Kutcher and Marton (1989) reported a 60% response of depressed adolescent outpatients to OMI at doses ranging from 150 to 250 mg daily. Thus, data from open studies suggest a range of responses to TCAs of approximately 40% to 60%, in adolescents with MOD. In controlled studies, somewhat similar but generally poorer response rates have been reported. Kramer and Feiguine (1981) described a negative outcome of a placebo study using a fixed dose of amitriptyline of 200 mg/day in 10 depressed adolescents. Geller (1990) conducted an 8-week, double-blind, placebo-controlled trial of nortriptyline after a 2-week placebo washout using a fixed plasma-level design (80 ± 20 ng/mL). Only 1 of 31 adolescents showed a positive response. Koplewicz and Klein (1990) presented preliminary data from a double-blind placebo-controlled study of OMI (dose, 150 to 300 mg/day) which showed a response rate of approximately 60% in OMI-treated subjects compared with about a 40% placebo response. Finally, in a 6-week, double-blind, placebo-controlled trial of 30 depressed teenagers, Boulos et al. (1991) reported a response rate of 50% in OMI-treated against 33% of placebo-treated depressed adolescent outpatients. Because of differences in populations studied, outcome measures, lengths of treatment, and the relatively small numbers of probands studied, it is difficult to draw conclusions from these studies (Connors, 1992; Strober, 1992). We report here on the final sample of

60 adolescent outpatients with MOD treated in a 6-week, double-blind, placebo-controlled trial of OMI to clarify the potential efficacy of OMI in the treatment of depressed teenagers. This report extends the sample described by Boulos et al. (1991) and constitutes the largest randomized sample of OMI treatment in depressed adolescents published to date.

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY. 33:5. JUNE 1994

METHOD Adolescents (male and female, aged 15 to 19 years) with a presumptive diagnosis of depression (N = 150) were referred to one of two university outpatient clinics-Sunnybrook Health Science Centre, University of Toronto; or the Royal Ottawa Hospital, University of Ottawa. Informed consent to enter the study was obtained from the patient plus a parent or legal guardian. After informed consent was obtained, all subjects were assessed using a clinical interview and the Schedule for Affective Disorders and Schizophrenia for School-Age Children (K-SADS) (Chambers et al., 1985) by an adolescent psychiatrist trained in its use. For each proband, a parent (usually the mother) was also interviewed using the K-SADS. Discrepancies between informants were resolved by reviewing the K-SADS material with both informants to arrive at subject and parent agreement on the scoring of individual K-SADS items. In the event that agreement between subject and parent could not be reached, the investigators accepted the subject's score for those K-SADS items measuring internal states and the parent's score for those K-SADS items measuring observable behaviors. After all the above information was compiled, diagnoses were assigned consensually by two investigators using DSM-JII-R (American Psychiatric Association, 1987) criteria. Medically fit patients with a DSM-JII-R diagnosis of nonpsychotic MDD, who were medication-naive and did not have a concurrent or lifetime diagnosis of anorexia nervosa, bulimia, manic depression, schizophrenia, schizoaffective disorder, alcohol abuse. any substance abuse disorder, or any organic psychiatric disorder were included in the study. Furthermore, each proband was required to meet a Hamilton Depression Rating Scale (HDRS) score of 17 or greater and a Beck Depression Inventory (BDI) score of at least 16. Each proband completed a I-week, single-blind, placebo washout during which each was required to maintain an HDRS score of ~17 and a BDI score of ~16. Each subject underwent a complete medical assessment including laboratory investigations, electrocardiogram, and the DMI side effects scale (SES). The subjects were then assessed using the depression section of the K-SADS to confirm that they continued to meet diagnostic criteria for MDD. After 1 week on single-blind placebo, patients who continued to meet diagnostic and severiry criteria (HDRS, BDI) were randomized to DMI or placebo. After randomization, treatment was begun using a b.i.d. dosage schedule with either DMI or identical placebo. Treatment lasted for 6 consecutive weeks after randomization. Patients were assessed weekly on a variety of measures including the HDRS, BDl, SES, and symptom checklist-58 (SCL-58). Electrocardiograms were done at every visit. An evaluable subject was defined as one who completed the entire 6 weeks of the postrandomization treatment protocol (Fig. 1). Active drug consisted of a fixed dose of 200 mg of DMI daily, given in divided doses at 8:00 A.M. and 8:00 P.M .. DMI treatment was started at 100 mg given at 8:00 P.M., followed by an addition of 50 mg at 8:00 A.M. on successive days to reach a total of 200 mg daily. At study completion, blood was drawn in the morning, 12 to 14 hours after the last DMI dose, for repeat of laboratory investigations and DMI plus 2-hydroxydesipramine (2-0H-DMI) evaluations. Compliance was monitored using weekly pill counts and study completion DMI plus 2-0H-DMI serum levels.

Analytical Procedures For all continuous measures, the DMI and placebo groups were compared using two-way analyses of variance with group (DMI,

687

KUTCHER ET AL.

N=14

N=10

N=18

refused study

placebo responder

dropouts

N=150

N=70

N=60

entry evaluation

one week placebo run in

6 weeks fixed dose DMljplacebo

completers N=42

N=66

did not meet criteria Fig. 1 Flow diagram: desipramine/placebo study. DM!

=

desipramine.

placebo) and time (baseline, end of treatment) factors. Some analyses included more levels in the time factor whereas others involved an extra berween-subjects factor such as the atypical or endogenous classifications. t Tests were used to compare "well" and "ill" subgroups of the DMI subjects at the end of treatment on some of the cardiovascular and laboratory measures. For those measures that were dichotomized into "well" and "ill" or "improved" and "not", the DM! and placebo groups were compared using the X2 or Fisher's Exact Test.

RESULTS

Of the 150 patients (I12 female, 38 male) initially assessed, 66 did not meet intake criteria. Fourteen met criteria but refused to enter the study. Thus, 70 probands entered the single-blind placebo washout phase. Of this number, 10 were deemed placebo responders after 1 week (HDRS