British Antipsychotic Data Raise Questions

A new head-to-head comparative effectiveness study suggests there are clinically significant differences in effectiveness of the five second-generation antipsychotic (SGA) medications currently available. Yet some researchers are questioning the study's methods and its primary conclusion that olanzapine (Zyprexa) and risperidone (Risperdal), along with the first-generation drug haloperidol (Haldol), are more effective than aripiprazole (Abilify), quetiapine (Seroquel), or ziprasidone (Geodon).

The report, appearing in the December British Journal of Psychiatry, involved 327 acutely ill patients hospitalized for treatment of schizophrenia, schizoaffective disorder, or schizophreniform disorder. Patients were randomly assigned on admission to open-label treatment with one of the five SGAs or to haloperidol as an active comparator.

The research, led by Robert McCue, M.D., vice chair of psychiatry at Woodhull Medical and Mental Health Center in Brooklyn, N.Y., was funded by Woodhull, part of the New York City Health and Hospitals Corporation.

Questioning a Primary Outcome

Almost immediately after publication, researchers and clinicians began expressing reservations about the validity of the study via electronic letters submitted through the journal's Web site. At the center of the debate was the primary outcome measure used. McCue and his colleagues defined an antipsychotic drug as “effective if the patient's mental status improved sufficiently to no longer necessitate acute inpatient care.” These“ improved” patients, McCue and his coauthors added, were either discharged to the community or moved to “an alternative form of care.”

In contrast, “The antipsychotic was classified as ineffective if, in the treating psychiatrist's assessment, the patient showed no significant improvement after at least three weeks of treatment, and the drug was discontinued. If the medication was discontinued before the end of a three-week trial due to side effects or significant deterioration in the patient's mental state, it was also classified as ineffective.”

McCue and his colleagues noted that “decisions about discharge were made solely on clinical grounds and not influenced by insurance arrangements.” They also characterize their definition of effectiveness as pragmatic and one that mirrors clinical practice. “An ill patient is admitted, treated, and when sufficiently improved, is discharged. In this study, an effective antipsychotic improved a patient's psychosis enough so that he or she could be discharged. This outcome is meaningful to both clinicians and their patients.”

Yet in an electronic letter to the editor, Taiwo Ajayi, M.D., a staff psychiatrist at Kent and Medway National Health Service Trust in Kent, England, questioned the validity of the primary measure of drug effectiveness being linked to hospital discharge. “This was reportedly determined by the treating psychiatrists' assessment, but no structured scales were mentioned. This choice of outcome measure is highly subjective for several reasons: for starters, the threshold for discharge not only varies internationally and locally, but also between individual psychiatrists depending on experience and attitude towards risk.”

Ajayi added that the decision to discharge is influenced by factors other than improvement in mental status, including “social support, anticipated ease of read-mission if required, and robustness of community mental health services.” He also noted that “it is worth remembering that the improvement in mental state of a psychotic inpatient cannot be entirely attributed to effectiveness of antipsychotic medication. The associated psychosocial interventions, ward milieu, role of staff's high expressed emotion, experience, and competence all have a part to play. None of these factors were mentioned, considered as confounding factors, or controlled for in this study.”

In addition to the ability to discharge, McCue and colleagues used improvement over time in the total score on the Brief Psychiatric Rating Scale (BPRS), measured at baseline, weekly during the study, and at endpoint. Side effects were recorded along with each BPRS rating by treating psychiatrists, and side-effect data were gleaned from both spontaneous reports and clinical evaluations. A clinician who did not know which drug patients were taking assessed extrapyramidal or Parkinsonian side effects with the Simpson-Angus Scale and the Akathisia Rating Scale.

Significant Differences Found

Using their primary outcome measure, McCue and his colleagues found that the SGAs olanzapine (92 percent of inpatients discharged) and risperidone (88 percent discharged), along with the comparator FGA haloperidol (89 percent discharged), were significantly more effective, compared with aripiprazole, quetiapine, or ziprasidone (each medication group had about 64 percent of patients discharged.)

The maximum daily dosages of the antipsychotics used were 21.8 mg aripiprazole, 16 mg haloperidol, 19.1 mg olanzapine, 652.5 mg quetiapine, 5.2 mg risperidone, and 151.2 mg ziprasidone.

There was no difference between the six groups in the need for adjunct haloperidol or lorazepam (Ativan) to treat aggressive or agitated behavior. Curiously, the researchers noted, “younger patients” (defined as below the median age of 38) treated with aripiprazole needed significantly more diphenhydramine (Benedryl) as a sleep aid than did patients taking the other antipsychotics except haloperidol. McCue and his colleagues speculated that the increased need for diphenhydramine could indicate that“ aripiprazole is simply more activating in younger patients” or“ it is also possible that younger patients required diphenhydramine more often for sleep.”

Significantly more patients taking haloperidol or risperidone were prescribed benztropine (Cogentin), presumably to prevent or treat extrapyramidal effects. However, no significant differences were seen in the Simpson-Angus or Barnes rating scales between the six medication groups.

Improvement in total score on the BPRS from baseline to endpoint did not significantly differ between the six groups. However, as a group, patients taking haloperidol, olanzapine, and risperidone tended to have a greater decrease in BPRS total score (mean 15.6) compared with those in the other three groups (mean 13.8). Yet that difference was not statistically significant, and overall patients taking the three potentially superior antipsychotics, as shown by BPRS scores, had longer times to discharge than patients taking the other medications.

Clinical Implications Cited

McCue and his colleagues concluded that their findings are important because “these results were obtained with minimum bias, using a randomized design, without support from the pharmaceutical industry.”

Haloperidol, olanzapine, and risperidone “are reasonable first choices” for acute treatment of hospitalized patients “unless the patient's history suggests otherwise,” they said. The three drugs are more potent antagonists of dopamine type-2 receptors, they noted, than are aripiprazole, quetiapine, or ziprasidone, “which may account for their superior effectiveness.” In addition, olanzapine and risperidone were better tolerated than the other drugs.

Ajayi, in his letter to the editor, said, “In my opinion, though this study has a strength in its pragmatic design and inclusion of subjects with substance misuse disorder, the inappropriate choice of and flawed interpretation of outcome measures makes the conclusion unjustifiable.”

As a last note, McCue and his coauthors concluded that the number of patients who require acute treatment “is substantial, and more studies with minimal bias are greatly needed to assist clinicians in making thoughtful treatment decisions.”

“Comparative Effectiveness of Second-Generation Antipsychotics and Haloperidol in Acute Schizophrenia” is posted at<http://bjp.rcpsych.org/cgi/content/abstract/189/5/433>.▪