Clonazepam in Childhood Anxiety Disorders

Clonazepam in Childhood Anxiety Disorders FLEMMING GRAAE, M.D., JUDITH MILNER, M.D., LINDA RIZZOTTO, M.A., AND RACHEL G. KLEIN, PH.D.

ABSTRACT Objective: There is evidence for the efficacy and safety of clonazepam (CZP) in adult anxiety disorders, but no formal

studies to substantiate clinical reports of similar benefit in children with anxiety disorders. Method: In this double-blind pilot study, 15 children, aged 7 to 13 years, entered a randomly assigned, double-blind crossover trial of 4 weeks of CZP (up to 2 mg/day) and 4 weeks of placebo. Results: Twelve children completed the trial. All but 1 had a diagnosis of separation anxiety disorder, and all but 2 had comorbid diagnoses. Nine children appeared to have moderate to significant clinical improvement, but statistical comparisons on several ratings failed to confirm a trend in favor of CZP. Side effects of drowsiness, irritability, and/or oppositional behavior were notable in 10 children in the CZP phase compared with 5 in the placebo phase. Conclusions: Clonazepam was believed to have clinical benefit for some children, but this was not confirmed statistically in this small sample. Problematic side effects of drowsiness and disinhibition were common and possibly were due to rapid titration. J. Am. Acad. Child Ado/esc. Psychiatry, 1994, 33,

3:372-376. Key Words: clonazepam, childhood anxiety disorders, side effects.

There is ample literature from controlled studies to support the benefit of benzodiazepines in adult anxiety disorders. In the cross-national collaborative panic study (Ballenger et al., 1988), for example, alprazolam (5 mg/day) was found superior to placebo in adults with panic disorder or agoraphobia. Controlled studies of Rickels et al. (1983), Pollack et al, (1986), and Tesar et al. (1987) found similar responses for alprazolam (2 to 3 mg/day), diazepam (30 to 40 mg/day), and clonazepam (CZP; 2 to 3 mg/day) in adults with a variety of anxiety disorders. The literature on the efficacy of benzodiazepines in child and adolescent anxiety is sparse, and the evidence mixed. Several

Accepted October 11, 1993. Drs. Graae, Milner, and Klein were at the New York State Psychiatric Institute and Columbia University, College ofPhysicians and Surgeons, New York, NY, and Ms. Rizzotto was at the Freedom from Fear Clinic, Staten Island, NY, when this work was performed. Dr. Milner is now with the Bay Psychiatric Group, Everett, WA; Ms. Rizzotto is now at St. Vincent" Medical Center, Staten Island, NY. Special thanks for crucial assistance to Kathy Amadeo, who oversaw quality controland data entry, and Mark Davies, who provided statisticalconsultation. Particular appreciation is also expressed to Mary Guardino for her support and enthusiasm. This study was supported, in part, by the Center to Study Youth, Depression, Anxiety and Suicide (NIMH center grant 30906-12), NIMH research training grant 5-T32-MHI6434, and Roche Laboratories. Reprint requests to Dr. Graae, New York State Psychiatric Institute, Unit 60, 722 West 168th Street, New York, NY 10032. 0890-8567/94/3303-0372$03.00/0© 1994 by the American Academy of Child and Adolescent Psychiatry.

372

controlled studies (Simeon and Ferguson, 1987; Simeon et al., 1992) suggest some or no benefit with alprazolam after 4 weeks in small samples of children and adolescents with overanxious disorder or avoidant disorder (N = 12 and 30, mean age = 12, average maximum dose of 1.57 mg with a range of 0.5 to 3.5 mg daily). In a case report, Biederman (1987) reported positive responses in three children with anxiety disorders and panic-like symptoms treated with CZP in doses up to 3 mg/day with few or no side effects. No studies with children have compared benzodiazepines with different half-lives. Clonazepam is a 7-nitrobenzodiazepine derivative that is lipid soluble with high affinity for benzodiazepine receptors in vitro and possibly up-regulates serotonin binding sites. It has a general distribution, is both protein-bound and free, and passively diffuses between plasma and brain with no brain sequestration. In monkeys, it is largely biotransformed to nonactive amino derivatives and a small amount of active nitro derivatives. Clonazepam has a half-life of20 to 80 hours (<40 hours in some studies), and CZP and its metabolites are excreted via urine and possibly other routes (Dreifuss and Sato, 1982; Pippenger et al., 1978). Clonazepam has been used for more than 20 years in children with medical disorders, mainly seizure disorders; it is used frequently with other antiseizure medications such as phenobarbital, in doses up to 9 mg/day

j. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 33:3, MARCH/APRIL 1994

CLONAZEPAM IN CHILDHOOD ANXIETY DISORDERS

or 0.3 mg/kg per day (Aarli, 1973; Browne, 1976; Dummermuth and Kovacs, 1972; Tyrer, 1980; Vassella et al., 1973). Side effects have been reported as transient, dose related, and infrequent, and often when used in conjunction with other antiseizure medications. Reported side effects include drowsiness, sialorrhea, irritability, and weight changes; few side effects have been reported with doses ofless than 3 mg/day. Dependence in children has not been reported. Clonazepam has the advantage of being long-acting. Because of this feature, its safety record, single daily dosing, and the indications of benefit in adult anxiety disorders, a systematic trial of its efficacy in children with anxiety disorders was undertaken.

with functioning; 2 = moderate, some interference; 3 = severe, significant interference). Ratings were done at baseline, after 4 weeks, and after 8 weeks and included the DISC 2.1 (parent and child), the Brief Psychiatric Rating Scale (BPRS; Overall and Pfefferbaum, 1982), and a Clinical Global Improvement Scale (CGI; rated on a scale of 1 to 8: 1 = completely well; 2 = much improved; 3 = improved; 4 = slightly improved; 5 = unchanged; 6 = slightly worse; 7 = worse; 8 = much worse).

METHOD

Subjects

Subjects Medically healthy children, aged 7 to 13 years, had to meet criteria for a DSM-III-R diagnosis of an anxiety disorder (American Psychiatric Association, 1987) that had persisted for at least 6 months based on "best estimate" team consensus that relied on clinical interviews by one of two psychiatrists and a Diagnostic Interview Schedule for Children (DISC 2.1, Shaffer er al., 1989). The DISC was administered in flexible fashion, with full clinical inquiry about symptomatology and functioning. Children were recruited in a university-affiliated mental health clinic located in a borough of New York City (Staten Island).

Procedure Children were recruited through local newspaper advertisement and professional and school sources. Signed consent was obtained from parents and children 12 and older, and children younger than 12 gave verbal assent. In addition to the clinical assessment, a Parent Questionnaire (Gittelman, 1985b), Teacher Rating Scale (Gittelman, 1985a), Children's Manifest Anxiety Scale (Reynolds and Richmond, 1984), and Child Global Assessment Scale (Shaffer et al., 1983) were completed. Children were to be free of medication (except for minor medical problems) and without serious medical problems and mental retardation as estimated by Peabody Picture Vocabulary Test (Dunn and Dunn, 1981). Children with a psychotic, obsessive-compulsive, or mood disorder were excluded. Children were randomly assigned to a double-blind, placebo crossover trial of matching placebo and CZP for 4 weeks each. Capsules were administered each morning before or just after breakfast in O.25-mg (or one tablet) increments every 3 days up to 1 mg, then O.25-mg increments every 2 days up to 2 mg, unless side effects or compliance issues intervened. After 4 days at a maximum 2 mg/day, medication was tapered to zero by the end of each phase.

Assessments Children were seen weekly by a child psychiatrist who provided supportive therapy, monitored clinical status and medication changes, and systematically recorded side effects on the Side Effects Rating Scale, which measures severity (l = mild, no interference

J.

Data Analysis Treatment differences for continuous scale scores on the BPRS, number of anxiety symptoms, and the CGI were subjected to analysis of covariance, controlling for baseline values. The Side Effects Rating Scale scores were averaged for each week and compared by paired t test across treatment phases.

RESULTS

Of 89 seemingly appropriate referrals, 57 were determined not to have anxiety disorders, 6 had prominent attention deficit disorder (exclusionary if prominent) in addition to an anxiety disorder, 5 refused treatment, 3 had subclinical dysfunction, 3 had anxiety disorders but were responsive to psychotherapy before screening, and 3 had subclinical anxiety states. Fifteen children were entered into the study. Diagnoses were as follows: separation anxiety disorder, 14; overanxious disorder, 6; avoidant disorder, 2; social phobia, 5; simple phobia, 5; oppositional disorder, 3; conduct disorder, 1; and attention-deficit hyperactivity disorder, 3. All children except one had a diagnosis of separation anxiety disorder, and all but two had two or more diagnoses (Table 1). Four had four diagnoses, two of them dropouts. All children were white; eight were male and seven female, aged 7 to 13 (mean = 9.8, SD = 2.1). Drug Regimen

In phase I, eight children received CZP and seven received placebo. Three boys dropped out during the active phase, two because of serious disinhibition with marked irritability, tantrums, and aggressivity (one with attempted self-injury by tying a rope around his neck) and one because of noncompliance. Of the 12 completers (Table 1), 3 were assessed as having no overall improvement in anxiety symptoms or level of functioning while taking CZP, 5 had moderate improvement, and 4 had marked improvement.

AM. ACAD. CHILD ADOLESC. PSYCHIATRY. 33:3. MARCH/APRIL 1994

373

GRAAE ET AL.

TABLE 1 Characteristics of Subjects and Response to Clonazepam Patient

Age (yr)

Sex

Diagnosis

Completers 1

7

F

SAD

2

10

M

3

7

M

4

10

F

5

8

F

6

7

F

7

8

F

8

10

F

9

9

F

10

12

M

SAD ODD AD SAD ADHD SoPh CD SAD SiPh SAD SiPh OAD SAD SAD OAD ADHD ODD SAD SiPh SoPh SAD Enuresis OAD SAD AD OAD

CZP Max Dose (mg)

Clinical Response to CZP/ Side Effects

5

1.5

9 7 5 3 5 5 6 10 10 6 3 6 7 8 7 5 8 5 4 8 5 4 5

0.5

No clinical improvement/ Irr,Opp Some clinical improvement/Oro

Age Onset (yr)

1.75

No clinical improvement/ No side effects

2.0

No clinical improvement/ Irr,Opp Good clinical improvement/Irr, Opp Good clinical improvernent/Irr, Oro Good clinical improvement/Irr, Oro

1.25 1.25 2.0

2.0

1.25 2.0

11

? 5

2.0

11

11

M

12

13

M

OAD SAD SoPh

3 4 4

1.5

Dropouts" 13

13

M

SAD SoPh

3 13

0.75

14

11

M

11

M

4 5 5 5 3 5 5 6

0.75

15

SAD SoPh SiPh ADHD SAD OAD ODD SiPh

1.25

Some clinical improvernenr/Irr, Opp, Oro Some clinical improvernent/Irr, Opp Some clinical irnprovernent/Irr, Opp, Oro Good clinical improvement/No side effects Some clinical improvement/Oro

Unclear whether decreased anxiety/Extreme disinhibition: aggressive and dangerous behavior including self-injury Reported less anxiety/ Marked disinhibition: tantrums, Opp, Irr Slight improvement reported/Oro (dropped for nonstudy reasons)

Note: CZP = clonazepam; SAD = separation anxiety disorder; ODD = oppositional disorder; AD = avoidant disorder; ADHD = attention-deficit hyperactivity disorder; SoPh = social phobia; SiPh = simple phobia; CD = conduct disorder; OAD = overanxious disorder; lrr = irritablellabile; Opp = oppositional; Oro = drowsiness. aAil dropped in drug phase.

374

j. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 33:3, MARCH/APRIL 1994

CL ON AZEPAM IN CHILDH OO D AN XIETY DISORDERS

TABLE 2 BPRS and CGI Ratings: Clonazepam and Placebo at W eek 4

Drug

Placebo

Mean

SD

Mean

SD

p

14.5

8.7 0.9

15.3

7.9

4.0

1.1

NS NS

BPRS' CGI

3.6

Note: BPRS = Brief Psychiatric Rating Scale; CGI = Clin ical Global Improvement Scale; NS 'Analysis of covariance was used to adjust for baseline values.

At the end of the study, six children (50%) no longer met criteria for an anxiety disorder. Two of three children with a comorbid disruptive disorder retained only the disruptive disorder. No order effect was found. Comparisons on the BPRS and CGI at end-phases, when medication or placebo had been tapered off (Table 2), and for CGI at week 3 of each phase (Table 3), when medication or placebo was at maximum, did not reveal significant improvement relative to baseline for placebo or CZP. The number of anxiety symptoms derived from the DISC and the Children's Manifest Anxiety Scale did not show significant treatment differences with regard to severity or frequency.

= not

significant.

in children and adolescents. The study sample is not large, but in 12 cases, a crossover design should detect a clearly beneficial effect, especially since 11 children had the same diagnosis of separation anxiety disorder. Because only 5 children had an overanxious disorder, the disorder is insufficiently represented for detecting a specific drug effect on this condition. A dose of 2 mg was selected because it has been reported effective in adults. However, it is conceivable that children require higher levels for clinical effect. Measurement of treatment efficacy is problematic. One study (Klein er aI., 1992) found anxiety scales in children to be insensitive to clinical improvement. More to the point, probably, is the relatively short exposure to treatment in the present study. It is plausible that longer therapy would yield better effect. However, we found the use of CZP problematic in some children . Although the level of side effects was not significantly higher on CZP than placebo, these comparisons exclude the 3 children who had to be removed from the trial. Furthermore, side effects reached clearly troublesome proportions in several children. The presence of disinhibition, oppositionalism, and irritability was notable . They occurred in children who otherwise clinically improved and in those who did not. It is possible that the rate of medication induction was too steep and contributed to this phenomenon and that much more gradual dose increments might avoid these troublesome side effects (Biederman, personal communication). Nevertheless, some children

Side Effects

Drowsiness, irritability or lability, and oppositional behavior were the most frequent CZP side effects. In the 12 children who remained in the study, the mean rate of side effects did not differ significantly between CZP and placebo at any study point (weeks 2, 3, and 4). Side effects were reported in 10 children (83%) receiving CZP and in 7 children (58%) receiving placebo. Five experienced drowsiness, and 8 experienced irritability and concurrent oppositional behavior. DISCUSSION

This study does not support the efficacy of CZP in doses up to 2 mg/day in treatment of anxiety disorders

TABLE 3 CGI Ratings: Clonazepam and Placebo at Week 3

Drug CGI

Placebo

Mean

SD

Mean

SD

4.08

1.2

3.75

1.0

Note : CGI: rated 1 (much improved ) to 4 (no change) NS = not significant .

to

p 1.08

. NS

7 (much worse). CGI = Clinical Glob al Improvement Scale;

j . AM. ACAD. CHILD AD OLESC. PSYCHIATRY, 33:3, MAR CH/APRIL 1994

375

GRAAE ET AL.

experienced side effects, including drowsiness, at relatively low doses (as low as 0.5 to 0.75 mg/day). We believe there is sufficient evidence to warrant a larger study of the efficacy of benzodiazepines in the treatment of childhood anxiety disorders and that additional study features might be incorporated such as slow gradual titration and longer treatment, as well as a comparison with non medication treatment such as behavioral therapy. Quantifying clinical improvement remains problematic, however, and evidence suggests a lack of descriptive validity for scales of childhood anxiety. Klein et al. (1992) have suggested that the best approach, currently, may be to rate specific symptoms for each anxiety disorder in terms of severity and impairment. In summary, the clinical impression that some children improved while taking CZP could not be statistically confirmed in this small sample as a whole, while there was more evidence for clinically significant behavioral disinhibition, possibly due to induction rate.

REFERENCES Aarli J (1973), Effect of clonazepam on epileptic seizures, Acta Neurol

Scand SuppI53:11-17 American Psychiatric Association (1987), Diagnostic and Statistical Manual of Mental Disorders, 3rd edition, revised. Washington, DC: American Psychiatric Association Ballenger JC, Burrows GD, DuPont RL et al. (1988), Alprazolam in panic disorder and agoraphobia: results from a multicenter trial. Arch Gen

Psychiatry 45:413-422 Biederman J (1987), Clonazepam in the treatment of prepubertal children with panic-like symptoms. J Clin Psychiatry 48:38-41 Browne T (1976), Clonazepam, a review of a new anti-convulsant drug.

Arch NeuroI33:326-332

376

Dreifuss E, Sato S (1982), Benzodiazepines. In: Antiepileptic Drugs, Woodbury DM, Penry JK, Schmidt RP et al., eds. New York: Raven Press, pp 737-752 Dumerrnuth G, Kovacs E (1972), The effect of clonazepam in the syndrome of infantile spasms with hypsarryrhmia and in petit mal variant or Lennox syndrome. Acta Neurol Scand SuppI53:91-96 Dunn LM, Dunn LM (1981), Peabody Picture Vocabulary Test-Revised. Circle Pines, MN: American Guidance Service Gittelman R (l985a), Teacher rating scale (modified Conners): anxiery and mood items added. Psychopharmacol Bull 21:945-948 Gittelman R (1985b), Parent's questionnaire (modified Conners): anxiery and mood items added. Psychopharmacol Bull 21:939-943 Klein R, Koplewicz H, Kanner A (1992), Imipramine treatment of children with separation anxiety disorder. J Am Acad Child Adolesc Psychiatry 31:21-28 Overall J, Pfefferbaum B (1982), Brief reports and reviews: the Brief Psychiatric Rating Scale for Children. Psychopharmacol Bull 18:10-16 Pippenger C, Penry J, Kutt H (1978), Antiepileptic Drugs: Quantitative Analysis and Interpretation. New York: Raven Press Pollack M, Tesar G, Rosenbaum J, Speir S (1986), Clonazepam in the treatment of panic disorder and agoraphobia: a one year follow-up. J

Clin Psychopharmacol 6:302-304 Reynolds CR, Richmond BO (1984), Revised Children's Manifest Anxiety Scale. Los Angeles, CA: Western Psychological Services Rickels K, Case G, Downing RW (1983), Long-term diazepam therapy and clinical outcome. JAMA 250:767-771 Shaffer D, Fisher P, Piacentini J, Schwab-Stone M, Wicks J (1989), Diagnostic Interview Schedule fOr Children (DISC 2.1). New York: Division of Child and Adolescent Psychiatry, New York State Psychiatric Institute Shaffer D, Gould M, Brasic Jet al. (1983), A children's global assessment scale (CGAS). Arch Gen Psychiatry 40:1228-1231 Simeon J, Ferguson H (1987), A1prazolam effects in children with anxiety disorders. Can J Psychiatry 32:570-574 Simeon JG, Ferguson HB, Knott V et aI. (1992), Clinical, cognitive, and neurophysiological effects of alprazolam in children and adolescents with overanxious and avoidant disorders. J Am Acad Child Adolesc Psychiatry 31:29-33 Tesar G, RosenbaumJ, Pollack M, Herrnan ], Sachs G (1987), Clonazepam versus alprazolam in the treatment of panic disorder: interim analysis of data from a prospective, double-blind, placebo-controlled trial. J Clin Psychiatry 48(suppl):16-21 Tyrer J, ed (1980), The Treatment of Epilepsy. Philadelphia: Lippincott Vassella F, Pavlincova H, Schneider H, Rudin H, Karbowski K (1973), Treatment of infantile spasms and Lennox-Gastaut syndrome with clonazepam. Epilepsia 14:165-175

]. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 33:3, MARCH/APRIL 1994