Abstract
Chief Complaint and Presenting Problem
History of Present Illness
W. was 14 years old when she first presented to the clinic with a past history significant for cerebral palsy, seizure disorder, and mild mental retardation (full scale IQ of 64). First signs of serious psychiatric illness started at the age of 13 years; while W. was described as a generally anxious youngster, her anxiety increased, and she began to show mild signs of depression after being reportedly bullied at school. She responded initially to treatment with psychotherapy and sertraline, and returned to baseline within a few months.
At the time of presentation to the clinic, W. was reportedly experiencing auditory and visual hallucinations of birds and animals and was having grandiose delusions of having the power to hypnotize birds and people. She exhibited signs of anxiety and depression. She was first treated for a presumed affective psychosis, and ultimately was diagnosed with schizoaffective disorder when her psychotic symptoms persisted in the absence of mood symptoms.
W. was started on clozapine at the age of 16 years and 4 months, after having had an inadequate response to previous adequate trials of multiple antipsychotics and mood stabilizers. Early in the course of her treatment, W. received a combination treatment of olanzapine to a maximum dose of 12.5 mg in combination with valproic acid with blood levels targeted to the high end of 80-120 μg/mL. This treatment was discontinued in part due to rapid weight gain. Other antipsychotic treatments were as follows in chronologic order and all were combined with carbamazepine treatment with levels adjusted to the high end of 4-12 μg/mL: quetiapine, one month to maximum dose of 250 mg; haloperidol, 10 months to maximum dose of 5 mg (dose limited by symptomatic prolactinemia); risperidone, 7 months to maximum dose of 6 mg; paliperidone, 4 months to maximum dose of 6 mg; and ziprasidone, 6 weeks to maximum dose of 120 mg. Following the ziprasidone trial, a combination treatment of lithium with levels adjusted to the high end of 0.8-1.1 mEq/L, carbamazepine with levels adjusted to the high end of 4-12 μg/mL, and paliperidone dosed at 6 mg daily did not lead to significant improvement in psychotic symptoms.
W.'s BMI at the age of 16 years and 4 months in the week before initiation of the clozapine trial was 28.9 kg/m2 (95th percentile). At the time of initiation of clozapine, W.'s symptoms included active auditory and visual hallucinations, disorganized speech and behavior, depressed mood and social withdrawal, and lack of speech. Her illness was newly complicated by catatonic features and refusal of food and water. These symptoms had resulted in weight loss, and her BMI decreased to 27.6 kg/m2 over the first six months of clozapine treatment.
By 6 months of clozapine treatment, W.'s psychiatric symptoms had improved substantially. She was evaluated as being “overall much improved,” and her ability to communicate her needs improved. However, at this point in time W.'s weight began to increase. Over the next 10 months, W.'s weight increased by a total of 22.5 kg (BMI = 37.7 kg/m2 [>97th percentile]). During this period of weight gain, consultation was ongoing with a multidisciplinary “Healthy Weight” team to address nutritional and exercise needs. Consultation originated at W.'s initial period of psychosis and treatment and spanned over a three to four year period. The Healthy Weight team offered W. and her mother nutritional counseling, sample menus, estimated calorie targets, and exercise goals. This intervention had little effect on the upward trajectory of her weight gain, and W.'s mother noted the difficulties in managing weight gain in a youngster who was “always hungry.” Nutrition and exercise were also consistently discussed and monitored at W.'s weekly to biweekly psychiatric appointments. Topiramate and metformin were both considered as pharmacologic interventions. However, W. had been prescribed topiramate for migraine headaches while taking risperidone, and weight control was not obtained. After reviewing available studies to date of the role of metformin in controlling antipsychotic-induced weight gain, a consultation was obtained through the endocrine department, and the decision was made to add metformin to W.'s medication regimen.
Past Psychiatric History
W. was diagnosed with attention deficit hyperactivity disorder (ADHD) at age 4 years and was treated with methylphenidate for several years. W. was described as an anxious youngster and was engaged in therapy for school refusal at the age of eight, and again at the age of 13 years. Her diagnosis is recalled as being generalized anxiety disorder. She did not have symptoms of depression at the age of 8 years, but she did at the age of 13 years, and a contributory factor was school bullying.
W. was psychiatrically hospitalized twice in close succession at the age of 16 years during the more severe phase of her illness. Development of catatonia for 3 weeks precipitated her second hospitalization, which lasted for one month. She was treated with clozapine and high-dose lorazepam during this hospitalization. Electro-convulsive therapy was being considered but ultimately not necessary.
Developmental History
Mother was 27 years old when she became pregnant; the pregnancy was complicated by the use of propylthiouracil for hyperthyroidism. W. was born at term at 6 lbs and 6 oz, although later it was questioned whether she was premature. She was cyanotic at birth. W. was treated with oxygen and went home with her parents 3 days after birth.
In the first 2 weeks of the postnatal period, W. developed a high fever, turned blue, and had seizures and apnea at home. She was admitted to the hospital and given antibiotics. She went home again and returned after another episode of apnea. An electroencephalogram (EEG) was performed at that time, and a seizure disorder was diagnosed.
W.'s developmental milestones were delayed. She sat without support at the age of one year, walked at 2 years, and spoke single words at 2 years of age. She was diagnosed with mild cerebral palsy around the age of one year with left-sided weakness. She was diagnosed with mild mental retardation during grade school with a full scale IQ of 64 and a score at the first percentile on the Vineland Adaptive Behavior Scale.
Educational History
W. attended a private parochial school until fourth grade. She was transferred to a public school when the private school staff felt they could no longer meet her needs. W. has consistently had an individualized education plan for learning that also included services for physical, occupational, and speech therapy. She started at a specialized private school for children and adolescents with developmental disabilities at the age of 14 years and continues there today.
Social History
W. lives with her biological mother and maternal grandmother. Her mother and father separated when W. was 15. W.'s illness was stressful for the family and preceded her parents' separation. W. enjoys summer camps, school, and her cat. She is friendly with and enjoys her peers.
Family History
W. has a family history of depression and drug and alcohol problems. A maternal uncle has a diagnosis of epilepsy. A maternal great aunt has a history of depression and similar behaviors as W., such as talking with birds, and has been hospitalized many times. A paternal aunt had a history of schizophrenia.
Medical History
W. has a history of seizures since infancy and was initially treated with carbamazepine. She remained on carbamazepine, valproic acid, or oxcarbazepine relatively consistently until the start of her clozapine treatment, with 2 month-long interludes when she was briefly lost to follow-up. An ambulatory EEG performed early in her psychotic illness indicated low probability of ongoing seizures.
Current medical problems include asthma, for which she is treated with Advair, obesity, and migraine headaches. She has been treated for the migraine with topiramate, which was started during the ninth month of her metformin treatment. The upward trajectory of her weight gain continued while taking topiramate. She had also been treated with topiramate for migraine during her more acute psychotic illness in the past and before initiation of clozapine, but it was terminated due to lack of efficacy.
W. was prescribed an oral contraceptive pill about 4 months after the start of clozapine treatment due to fleeting psychotic symptoms that emerged around the time of her menstrual cycle. The oral contraceptive pill appears to have helped with these psychotic symptoms, and her menses are regular.
Medication History
The following medication trials (in chronologic order) were undertaken with 2 periods of brief loss to follow-up: olanzapine to maximum dose of 12.5 mg daily with valproic acid for 4 weeks, quetiapine to maximum dose of 250 mg daily with valproic acid for 4 weeks, oxcarbazepine to maximum dose of 750 mg daily for 4 weeks, haloperidol to maximum dose of 5 mg daily with carbamazepine for 10 months, risperidone to maximum dose of 6 mg daily with carbamazepine for 7 months, ziprasidone to maximum dose of 60 mg daily with carbamazepine for one month, paliperidone to a maximum dose of 6 mg daily with carbamazepine and lithium for 4 months, and ziprasidone titrated to 120 mg daily with carbamazepine for 6 weeks.
At the time of metformin initiation, W.'s psychotropic medications were clozapine 250 mg twice daily (BID) and lorazepam 2 mg four times daily (QID).
Mental Status Examination
At the initial time of presentation to our clinic W. appeared as an appropriately groomed adolescent. She was cooperative and not aggressive. During the examination she appeared to stare and talked to herself without cessation. She occasionally paced around the room. Her affect was generally constricted and sad. W. described grandiose delusions as well as paranoid delusions such as her birds “cheating” her. She described auditory hallucinations. Her insight and judgment were limited secondary to her internal preoccupation.
Physical and Laboratory Examination at Initial Presentation
Physical examination was remarkable for obesity and acanthosis nigricans. Laboratory studies obtained at initiation of metformin treatment included fasting triglycerides 118 mg/dl (40–136), fasting HDL 67 mg/dl (36–76), fasting LDL 91 mg/dl (<110), fasting plasma glucose of 97 mg/dl, and fasting plasma insulin level of 12.8 μU/mL. Insulin levels and plasma glucose at 2-hr during an oral glucose tolerance test (OGTT) were 71.4 μU/mL and 120 mg/dl, respectively. Percent HgbA1C was normal at 5.4. Biochemical markers did not indicate insulin resistance. Homeostasis model assessment of insulin resistance (HOMA-IR) was 3.07 (a level >3.16 indicative of insulin resistance) (Keskin et al. 2005).
When W. was age 17 years and 8 months, metformin was initiated starting at 250 mg BID and was titrated in two weeks to 500 mg BID.
Course of Treatment on Metformin
Labs at the 6-month follow-up included fasting triglycerides 169 mg/dl (38–135), fasting HDL 54 mg/dl (36–84), fasting LDL 110 mg/dl (59–137), and a fasting glucose of 78 mg/dl (65–105). Physical examination was remarkable for diminished acanthosis nigricans. Psychiatric symptoms remain significantly improved; W. attends school in a special education program that she enjoys, is socially interactive with her family and treatment team, is competent in her activities of daily living, and her mood is generally happy. W. has fleeting auditory hallucinations and low mood mostly around the time of her menstrual cycle.
Her psychiatric medications remain generally consistent at 250 mg of clozapine BID and 2 mg of lorazepam QID. In the 9-month follow-up since initiation of metformin treatment, there has been no further weight gain. W. is currently 2.3 kg less than her peak weight of 83.6 kg (Fig. 1).
Body mass index (BMI) changes during treatment.
Brief Formulation
In summary, W. is an 18-year-old young woman with cerebral palsy, mild mental retardation, history of seizure disorder, and a family history of affective and psychotic illness. The pregnancy was complicated by maternal hyperthyroidism and treatment with propylthiuracil. W. also experienced some degree of perinatal distress, with early apnea and seizures. W. was diagnosed with ADHD at the age of 4 years, and also had some anxiety symptoms prior to the development of her psychosis at the age of 14 years.
Biologically, this young woman had a diathesis for major mental illness based on family history; perinatal medical problems also rendered it more likely that she would experience developmental and psychiatric problems later in life.
W. was ultimately diagnosed with schizoaffective disorder during adolescence, the working diagnosis at the present time. W. was reported to be treatment-resistant, with numerous medication failures. She experienced her first major improvement with clozapine, but potential response was limited due to uncontrolled weight gain. Behavioral measures were ineffective in managing diet and exercise. The addition of metformin stabilized her excessive weight gain and has allowed her to continue the course of clozapine.
Multi-Axial Diagnoses
Discussion
This case clearly illustrates the challenges of treatment of early onset psychotic illness in youth, and the need to balance efficacy and safety in young patients. The use of antipsychotic agents in children and adolescents can be associated with significant adverse effects, and prominent among these side effects is weight gain (Correll 2008; Correll et al. 2009). In the case of some affected adolescents with psychotic disorders, clozapine has been shown to have superior efficacy over other antipsychotic agents (Gogtay and Rapoport 2008). However, clozapine may have more liability for weight gain than other antipsychotic agents (Allison and Casey 2001). One prospective study showed that adults taking clozapine had significant weight gain for at least 4 years after initiation of treatment, and 2 other studies observed significant weight gain for at least 1 year, which was the length of the study (Haddad and Sharma 2007; Henderson et al. 2000; Bustillo et al. 1996; Briffa and Meehan 1998). The time course of weight gain with clozapine treatment in adults is in contrast to what might be an earlier plateau to weight gain in adults with some other atypical antipsychotic agents (Haddad and Sharma 2007).
There are also reports of significant weight gain in children and adolescents treated with clozapine. A double-blind randomized study comparing olanzapine and clozapine in childhood-onset schizophrenia over an 8 week period showed weight gain of 3.8 kg in the clozapine group, although this did not differ significantly from the olanzapine group (Shaw et al. 2006). Another non-randomized study comparing clozapine, risperidone and olanzapine in children and adolescents with multiple diagnoses over a 45-week period showed significant weight gain in all groups with an average weight gain of 9.5 kg in the clozapine group (Fleischhaker et al. 2008). Finally, an older study showed a 0.9 kg weight gain over 6 weeks in children and adolescents prescribed clozapine but this did not differ significantly in comparison to those prescribed haloperidol in the treatment of childhood-onset schizophrenia. However, this study was limited by the length of the trial (Kumra et al. 1996).
Some previous work has indicated that adjunctive use of metformin can be associated with minimization of continued weight gain and some improvement in lipid and triglyceride profiles in adults taking antipsychotic agents, although results have been mixed (Miller 2009).
The data on the use of metformin as an agent to attenuate antipsychotic-induced weight gain in children and adolescents is sparse, and no trials have lasted longer than 16 weeks. In a 16-week double-blind, randomized-controlled study of 39 adolescent patients ages 10–17 years, patients who were prescribed an atypical antipsychotic (olanzapine, risperidone, or quetiapine) in conjunction with metformin showed a significant weight stabilization relative to those in the placebo group who continued to gain weight (Klein et al. 2006). Conversely, in a 12-week double-blind, placebo-controlled, randomized study of children and adolescents less than 20 years of age, patients who were prescribed up to 6 mg of risperidone in conjunction with metformin had no significant change in weight compared to the placebo group (Arman et al. 2008). In a recent 12-week open-label study of 11 adolescent patients ages 10–18 years who were taking atypical antipsychotics, (among them was one patient prescribed clozapine), those receiving metformin in addition to antipsychotics showed no significant reduction in weight. However, patients receiving metformin did show a cessation in weight gain over a 12-week interval (Shin et al. 2009).
To date, the problem of weight control during antipsychotic treatment remains a significant limitation in the management of patients requiring antipsychotic treatment. We share this case where stabilization of weight in an adolescent on clozapine and metformin has been maintained for 9 months. In the case of this adolescent, previous adequate trials of other antipsychotic agents did not adequately control psychotic symptoms. This adolescent had a robust response to clozapine, but treatment was associated with progressive weight gain despite maximal nutritional and exercise counseling interventions. Ongoing weight gain would have limited the possibility of long-term clozapine treatment.
The use of metformin has shown mixed results in adults and thus far in children and adolescents. In studies to date, it is unclear whether there may be differences in the effect of metformin on antipsychotic-induced weight gain associated with specific antipsychotic agents. To our knowledge, no randomized trials have been conducted in a pediatric population treated with clozapine or olanzapine, the two agents associated with the greatest degree of weight gain. We are aware of only one other published instance where a youth on clozapine was treated with metformin for 12 weeks (Shin et al. 2009). We are aware of no follow-up studies that have been longer than 16 weeks in either adults or adolescents.
While this youth did not meet American Diabetes Association criteria for impaired glucose tolerance (The Expert Committee on the Diagnosis and Classification of Diabetes Mellitus 2003), she had fasting glucose at the high range of normal and some physical manifestations of insulin resistance (acanthosis nigrans), although HOMA-IR did not indicate insulin resistance. The use of metformin does not seem to have impacted her lipid profile or other parameters, although her acanthosis nigrans has improved.
Given the fact that children and adolescents may be especially vulnerable to weight gain with antipsychotic agents (Correll 2008; Safer 2004) and are inclined to spend more years exposed to these medications by virtue of an early onset of illness, active research into ways to limit antipsychotic weight gain is warranted.
Disclosures
Drs. Weaver and De León have no conflicts of interest or financial ties to disclose. Dr. Borgmann-Winter has received financial support from Pfizer and Eli Lilly. Dr. Coffey has received research support from Eli Lilly, NIMH, NINDS, Tourette Syndrome Association, Bristol-Myers Squibb, and Boehringer Ingelheim.
Footnotes
Acknowledgment
We would like to acknowledge and thank Stephanie Samar, M.A. for her review and assistance with preparation of the manuscript.