Abstract #15: Efficacy and Safety of Long-Acting Injectable Paliperidone Palmitate Relative to Long-Acting Injectable Haloperidol, Bromperidol, and Fluphenazine Decanoate for Long-Term Treatment in Patients with Schizophrenia Using Number Needed to Treat and Number Needed to Harm

Abstract

Background Atypical antipsychotic long-acting injectables (LAIs) such as paliperidone palmitate (PP) are believed to provide a favorable safety profile compared with typical LAIs. Because no trials have compared PP directly with typical LAIs, we assessed the benefit–risk of PP relative to typical LAIs using published placebo-controlled trials. Methods A computerized search in MEDLINE was conducted. Number needed to treat (NNT), number needed to harm (NNH), and the likelihood of being helped or harmed (LHH) were calculated from published placebo-controlled studies of PP and the following decanoate LAIs: haloperidol (HD), bromperidol (BD), and fluphenazine (FD). NNTs were calculated for prevention of psychotic relapse. NNHs were calculated for extrapyramidal symptoms: akathisia, tremor, and tardive dyskinesia reported as an adverse event, anticholinergic use, and positive Abnormal Involuntary Movement Scale (AIMS) total score. Results NNTs for relapse were similar for all LAIs: 2–5. NNHs varied considerably: akathisia 205 (PP) vs. 10 (BD); tremor 69 (PP) vs. −5 (BD); anticholinergic use 30 (PP) vs. 5 (FD); AIMS positive score −33 (PP) vs. 13 (FD). For tardive dyskinesia the NNH was infinity for PP vs. 7 for FD. LHH for preventing relapse vs. anticholinergic use was 15 for PP and 2.5 for FD. Conclusions Effect size as measured by NNT for relapse was similar across different LAIs. NNHs for extrapyramidal symptoms and related events were lower for typical LAIs, compared with PP. Although the benefits in prevention of relapse are similar among LAIs, the risks of extrapyramidal symptoms vary considerably. PP has a favorable LHH compared with typical LAIs.