HISTORY: Six years of psychosis
Ms. G, age 37, has had paranoid schizophrenia f or 6 years, resulting in numerous hospitalizations and continuous outpatient follow-up. Her family is supportive and supervises her when she’s not hospitalized.
Though fluent in English, Ms. G—a Polish immigrant—speaks primarily in her native tongue during psychotic episodes and becomes increasingly paranoid toward neighbors. As her condition degenerates, she hears her late father’s voice criticizing her. Because of marked social withdrawal and isolation, she cannot maintain basic interpersonal skills or live independently. Her psychosis, apathy, avolition, withdrawal, and lack of focus have persisted despite trials of numerous antipsychotics, including olanzapine, 25 mg nightly for 1 month, and quetiapine, 300 mg bid for 3 weeks.
What are the drug therapy options for this patient?
The authors’ observations
“Treatment-refractory” schizophrenia has numerous definitions. One that is widely accepted but cumbersome—used in the multicenter clozapine trial1 —requires a 5-year absence of periods of good functioning in patients taking an antipsychotic at dosages equivalent to chlorpromazine, 1,000 mg/d. In that time, the patient must have received two or more antipsychotic classes for at least 6 weeks each without achieving significant relief. The Brief Psychiatric Rating Scale (BPRS) score must be at least 45, with item scores of moderate severity for two or more of the following:
- disorganization
- suspiciousness
- hallucinatory behavior
- unusual thought content.
The Clinical Global Impression (CGI) Scale score must be at least 4 (moderately ill). Also, a 6-week trial of haloperidol, with a mean dosage of 60 mg/d:
- must fail to decrease the BPRS score by 20% or to below 35
- or must fail to decrease the CGI severity score to 3 (mildly ill).1
In 1990, an international study group defined treatment-refractory schizophrenia as “the presence of ongoing psychotic symptoms with substantial functional disability and/or behavioral deviances that persist in well-diagnosed persons with schizophrenia despite reasonable and customary pharmacological and psychosocial treatment that has been provided for an adequate period.”2 This definition is far more useful to clinical practice and also considers psychosocial function. Seven levels of treatment response and resistance were suggested, based on presence of positive and negative symptoms, personal and social functioning, and CGI scores.2
Meltzer3 proposed that any person not returning to his or her highest premorbid level of functioning with a tolerable antipsychotic be considered refractory and thus a possible candidate for clozapine therapy.
Ms. G’s illness meets the definition of treatment-refractory schizophrenia. Her CGI score at baseline was 5—severely ill—and several medication trials at sufficient dosages failed to control her positive or negative symptoms. Upon psychotic decompensation, she required prolonged hospitalization and could no longer live independently or work. At this point, she is a possible candidate for clozapine therapy.
TREATMENT: Starting clozapine
Ms. G was started on clozapine, 25 mg at night, titrated to 300 mg at bedtime.
Two weeks later, her paranoia and auditory hallucinations diminished, her interpersonal relationships improved, she was less withdrawn, her thoughts became more organized, and her range of affect expanded. She functioned at her highest level since her initial presentation based on clinical observation and family reports. Her CGI Global Improvement score at this point was 2 (much improved).
Ms. G. continued to take clozapine, 300 mg/d, for 2 years while undergoing weekly blood tests for white blood cell counts (WBC) with differentials. She did not require hospitalization for schizophrenia during this time, and her WBC count averaged between 4,000 and 4,500/mm3, well within the normal range of 3,500 to 12,000/mm3.
Then one day—after maintaining a relatively stable WBC for several weeks—a blood test revealed a WBC of 2,700/mm3. Ms. G exhibited no objective signs of immunosuppression, such as fever or infection. Still, the psychiatrist immediately discontinued clozapine.
Was the treating psychiatrist justified in immediately stopping clozapine after one low WBC reading?
The authors’ observations
Leukopenia, defined as a WBC <3,000/mm3, and agranulocytosis, defined as an absolute neutrophil count <500/mm3, are well-documented adverse reactions to clozapine. Early data on clozapine-associated agranulocytosis cases prior to 1989 suggest that up to 32% were fatal,4 but relatively few cases have occurred since the Clozaril National Registry was instituted in 1977.4,5 Between 1977 and 1997, 585 clozapine-associated agranulocytosis cases were reported in the United States; 19 of these were fatal, suggesting a mortality rate of 3.2% and attesting to the effectiveness of FDA-mandated WBC testing. During this period, 150,409 patients received clozapine.4
The agranulocytosis risk does not appear to be dose-related but declines substantially after the 10th week. Three out of 1,000 patients who take clozapine for 1 year are likely to develop agranulocytosis at the 3- to 6-month mark.4 Although the incidence continues to drop after month 6, it never reaches zero.4-6