Treating Schizophrenia Over a Lifetime
Martin L. Korn, MD


In Kraepelin's first descriptions of "dementia praecox," the progressive deteriorating course of the schizophrenic disorder was emphasized. Although Kraepelin recognized that there was a subset of patients with better outcomes, they were the exception to the rule. Until the modern psychopharmacological era, this pessimistic outlook was warranted. Since the introduction of newer "atypical" neuroleptic agents, optimism has steadily increased. The delineation of the neuropathological processes underlying deterioration in schizophrenia and the clinical guidelines for intervention in the patient with schizophrenia was the focus of a symposium on the lifetime treatment of schizophrenia held during this year's Annual Meeting of the American Psychiatric Association.

Early Intervention Strategies in Patients With Schizophrenia

In order to establish a rationale for early and prolonged interventions, it is important to review the neuropathological processes in schizophrenia. According to Diana Perkins, MD,[1] of the University of North Carolina at Chapel Hill there are several nonmodifiable aspects of schizophrenia that affect outcome. The prognosis in males, for example, is worse than in females. In addition, early age of onset as well as poor premorbid functioning are predictors of poor outcomes.

Alternatively, there are also several modifiable aspects of the disorder that may respond to clinical pharmacologic interventions. First-episode patients often present several years after the initial onset of symptoms.[2,3]The longer the duration of pretreatment symptoms, the longer the time until treatment response is achieved.[2] Therefore, decreasing the time lag between symptom onset and clinical intervention may aid in the rapid and effective stabilization of the patient. There are also data that support the commencement of prolonged drug treatment soon after the onset of the schizophrenic syndrome. By the fifth year after the first recurrence, nearly all patients will have relapsed.[4] After the second and third relapse, patients relapse even more rapidly.[4] After the third relapse, the course becomes chronic in one third of patients. Symptoms that may initially have improved with drug therapy early in the disease course may no longer respond to interventions. Other modifiable factors include frequent poor compliance with drug regimens, lack of efficacy of a particular drug, comorbid substance abuse, and psychosocial stressors.[5]

There is much evidence to suggest that schizophrenia is a neurodegenerative disorder. Increased ventricular volume, long noted to be associated with schizophrenia, has been shown to increase over time in patients with residual psychotic symptoms.[6] In an attempt to delay or reverse these changes, long-term maintenance treatment has been recommended, which may dramatically decrease the risk of recurrences.

There are several factors that must be considered before prolonged early treatment is generally recommended. Included in the decision process are the short- and long-term side effects, reluctance of the individual to take medication on a prolonged basis, and the cost of medication and clinical care. Furthermore, it is not entirely clear that prolonged treatment can actually reverse or prevent long-term neurobiological changes. Clinical studies are underway to further define this important clinical decision. It is suggested that atypical agents be used as first-line treatment due to their broader spectrum of action, greater tolerability, reduced short- and long-term side effects, and increased rates of compliance.[7,8] Lower doses are often effective early in the course, however, higher doses may be required as the illness progresses. First-episode patients may be particularly susceptible to side effects. Because the patient is often unfamiliar with the symptoms and treatment, it is particularly important to observe these individuals, as early side effects make poor compliance much more likely.

In a recently completed but unpublished international multicenter study by Lieberman and colleagues, 262 patients with schizophrenia were randomized to receive olanzapine (5-20 mg/day) or haloperidol (2-20 mg/day) over a 2-year period. The mean dose of olanzapine and haloperidol was 9.1 mg and 4.4 mg, respectively. Both medications were efficacious but there tended to be fewer dropouts on the olanzapine (68.5% vs 56.1%). Overall improvement tended to be greater with olanzapine (55% vs 46%) and neurocognitive functioning was improved with olanzapine. Average weight gain was also greater with olanzapine than with haloperidol (7.30 vs 2.64 kg).

Clinical Correlates of Receptor Binding Studies

The receptor-binding profile of the atypical agents are complex and are responsible for the differential therapeutic and side effect profiles. Gerald Maguire, MD,[9] of the University of California at Irvine reviewed D2-receptor-binding affinity data and used these data to predict some of the aspects of the various clinical profiles. Typical neuroleptic agents uniformly bind tightly to the receptor. This tight binding subsequently leads to a homeostatic upregulation of the dopamine receptor. Because the initial pharmacodynamic mechanism is counteracted, the therapeutic effects may also eventually be lost. Hypersensitivity of the postsynaptic D2-receptor has been posited as the mechanism responsible for disturbing side effects such as tardive dyskinesia. Among the atypical agents ziprasidone, a recently FDA-approved neuroleptic, and risperidone are the most tightly bound. There are other receptor binding properties, however, that modify the dopamine binding profile. For example, serotonergic-receptor antagonism may alter the treatment and side-effect profile.

Olanzapine demonstrates intermediate D2-binding affinities whereas quetiapine and clozapine possess loose affinities. Quetiapine may have to be used at higher doses (ie, greater than 600 mg) to compensate for this loose binding. The loose binding of quetiapine may also explain why higher doses of the medication produce no more extrapyramidal symptoms (EPS) than lower doses. Olanzapine's intermediate binding capacity explains the fewer EPS associated with this agent compared with the number seen with risperidone. This is a dose-dependent effect with risperidone however, as EPS are usually observed at doses higher than the 3-6 mg doses currently recommended.[10]

The tighter binding capacity of risperidone also accounts for the prolactin-elevating properties of this medication. Alternatively, clozapine, olanzapine, and quetiapine all demonstrate prolactin-sparing effects. Ziprasidone has not been studied enough to draw conclusions. Hyperprolactinemia may be associated with multiple side effects including menstrual disturbances, galactorrhea, sexual dysfunction, and estrogen deficiency. Prolactin elevations may be associated with depression[11] and some studies have suggested that olanzapine may be more effective in reducing depressive symptoms than either haloperidol[12] or risperidone.[10] The risk of suicide is very high in patients with schizophrenia and depressive symptoms increase this risk still further. Glazer[13] demonstrated a 2.5-fold decrease in suicide attempts in olanzapine-treated patients compared with haloperidol-treated patients.

Treatment-resistant Patients

The symptoms and course of schizophrenia worsen with successive psychotic episodes. Jeffrey Lieberman, MD,[14] of the University of North Carolina School of Medicine, Chapel Hill, suggested that either altered neurodevelopment or neurotoxic effects might explain this process. Unlike neurodegenerative disorders, such as Down's or fragile-X syndromes which manifest at birth, schizophrenia does not present in full until the second or third decades; although schizophrenics may demonstrate cognitive and motor side effects early in life. In addition, unlike the degenerative process in Alzheimer's disease, the patient with schizophrenia generally reaches a plateau phase later in life, after which, no further deterioration occurs.

Volavka and colleagues[15] studied a cohort of chronically ill schizophrenics who had been ill for more than 20 years. The medications in this study included clozapine (average dose 526.6 mg), olanzapine (average dose 30.4 mg), risperidone (average dose 11.6 mg) and haloperidol (average dose 25.7 mg). Clozapine and olanzapine produced greater improvement on the Positive and Negative Syndrome Scale (PANSS) compared with both risperidone and haloperidol, although substantial residual symptoms remained. Weight gain was greater in both clozapine- and olanzapine-treated patients; however, there was a positive association between the increased weight gain and clinical improvement. This association was not seen in either risperidone- or haloperidol-treated patients.

Treatment of Children and Adolescents

Because ethical issues as well as the general reluctance to employ pharmacologic interventions in children and adolescents, there are far fewer data concerning treatment of this population. However, Linmarie Sikich, MD,[16] of the University of North Carolina, Chapel Hill, indicated that psychotic disorders are much more prevalent among youths than previously thought. Approximately 15% to 35% of children with major depression have accompanying psychotic symptoms. Bipolar disorder presents before adulthood in 30% of patients. Forty to 60% of early-onset bipolar patients have associated psychotic symptoms. Approximately 60% of patients with schizophrenia have their first episode of psychosis during adolescence. Accurate diagnosis is often difficult in the younger individual and the diagnosis is often revised with time as the full clinical syndrome presents. For example, a substantial proportion of patients originally diagnosed as having attention deficit disorder actually have psychotic or affective syndromes.

In a small study of 21 treatment-resistant children, clozapine was more effective than haloperidol for both positive and negative symptoms.[17] Although weight gain was similar in the 2 groups, this similarity may be due to the short 6-week duration of the study. In another study, haloperidol (mean dose, 5.3 mg), olanzapine (mean dose, 4.1 mg), and risperidone (mean dose, 12.5 mg) were compared.[16] A greater proportion of the olanzapine-treated patients were classified as responders compared with patients in the other treatment groups. There were also fewer dropouts in the olanzapine-treated group, however, this effect tended to diminish over the 20-week course of the study. As with studies in adults, weight gain tended to be greater with olanzapine while prolactin elevation was greater with risperidone.

Treatment of children and adolescents is more difficult due the complex social and family environments that affect outcome. Furthermore, younger individuals suffer from increased side effects from the medications. Overall, Dr. Sikich recommended a multimodal approach to treatment, which includes a substantial psychoeducational component. The lowest effective dose of medication should be used to minimize side effects and increase compliance. Side effects such as sexual dysfunction or weight gain may be even more problematic in younger individuals than in adults.

Cognitive Functioning

Since the introduction of the atypical agents, the focus in the treatment of schizophrenia has shifted from the more overt positive symptoms to negative symptoms such as social withdrawal and cognitive impairment. Negative symptoms are often associated with greater overall functional impairment and poor adaptation in the community. Richard Keefe, PhD,[18] from Duke University, Durham, North Carolina, reviewed the subject of cognitive impairment.

Multiple areas of cognitive functioning are often impaired in patients with schizophrenia. These areas include verbal memory, executive functioning, attention, motor speed, visuospatial capacity, and verbal fluency.[19] Deficits in these functions are strongly associated with poor outcome. Harvey and colleagues[20] compared 3 groups of geriatric patients with schizophrenia, including 31 lifelong ambulatory patients hospitalized with acute exacerbations, 37 chronic patients maintained in a nursing home for at least 2 years, and 97 patients hospitalized at a long-term state psychiatric facility. Negative rather than positive symptoms appeared to differentiate the groups. Nursing home and state hospital patients scored 3 to 4 standard deviations (SD) below ambulatory patients with schizophrenia on a composite score of cognitive functioning

In first-episode patients, reduced frontal and temporal lobe volumes have been correlated with decreased improvement in negative symptoms, decline in neurocognitive performance, and higher typical antipsychotic doses.[21] Typical neuroleptics often fail to improve and may worsen attention and motor performance. They also may result in a decreased capacity to learn from previous experience or "practice effects." Enhancement of practice effects is mediated through frontal and striatal areas of the brain. Dopamine antagonism in these regions by typical agents may be responsible for the decrement in functioning.[22,23] Cognitive functioning was compared in 262 atypical- vs typical-antipsychotic-treated (olanzapine vs haloperidol, respectively) patients in the unpublished study by Lieberman and colleagues cited above. A composite score of neurocognitive tests was significantly higher in olanzapine-treated patients. The continuous performance test has been strongly associated with social problem solving, skill acquisition, and community functioning.[24]The patients taking atypical antipsychotic performed significantly better on this particular test.


Schizophrenia is a neurodegenerative disorder that often results in a progressive downward course. Prior to the introduction of typical antipsychotic agents, patients with schizophrenia were seen as "doomed from the womb." The atypical agents have extended our therapeutic reach, as we are now able to effectively treat a broader array of patients in a more comprehensive fashion. As new treatments are introduced, however, important clinical and research questions are beginning to emerge. For example, the proper time and extent of treatment intervention is being actively studied. In particular, the effect of early intervention in reversing the neuropathological and neurocognitive deficits is of great importance. Furthermore, efforts at understanding which specific pharmacologic treatments are most effective for particular patients and symptoms is central to further enhancing treatment. As these issues continue to be resolved, our optimistic outlook toward this disorder can only be enhanced.