Haloperidol versus placebo for delirium prevention in acutely hospitalised older at risk patients: a multi-centre double-blind randomised controlled clinical trial

• Published in: Age and ageing Vol. 47; no. 1; pp. 48 - 55
• Main Authors: Schrijver, Edmée J M, de Vries, Oscar J, van de Ven, Peter M, Bet, Pierre M, Kamper, Ad M, Diepeveen, Sabine H A, van Marum, Rob J, van Strien, Astrid M, Anten, Sander, Lagaay, Anne M, Boelaarts, Leo, Bloemers, Frank W, Kramer, Mark H H, Nanayakkara, Prabath W B
• Format: Journal Article
• Language: English
• Published: England Oxford University Press 01.01.2018; Oxford Publishing Limited (England)
• Subjects: Aged patients; Analysis; Clinical outcomes; Clinical research; Clinical trials; Delirium; Dosage; Dosage and administration; Double-blind studies; Efficacy; Elderly patients; Emergency services; Haloperidol; Health aspects; Hospitals; Missing data; Patient admissions; Patients; Pharmacovigilance; Prevention; Severity; Side effects; Statistical power; Treatment outcome; haloperidol; older people; aged 80 and over; delirium; prevention
• ISSN: 0002-0729; 1468-2834
• DOI: 10.1093/ageing/afx124

Abstract Background because the few randomised placebo-controlled trials investigating the potential role for prophylactic haloperidol in delirium prevention have focused on specific surgical populations, we investigated its efficacy and safety in acutely hospitalised older patients. Methods this multi-centre, double-blind, stratified, block randomised, placebo-controlled trial was conducted at six Dutch hospitals. Patients age ≥70 years, acutely admitted through the emergency department for general medicine or surgical specialties and at risk for delirium were randomised (n = 245) to haloperidol or placebo 1 mg orally twice-daily (maximum of 14 doses) on top of standard nonpharmacological prevention strategies. The primary outcome was delirium incidence. Other endpoints included delirium severity and duration, drug safety and clinical outcomes. Results intention-to-treat analysis included 242 participants (calculated sample size n = 390, statistical power of current sample 59%) allocated to haloperidol (n = 118) or placebo (n = 124). In the haloperidol and placebo group, delirium incidence was 19.5 versus 14.5% (OR 1.43, 95% CI 0.72 to 2.78); median (IQR) delirium duration 4 (2, 5) versus 3 (1, 6) days (P = 0.366); maximum DRS-R-98 score 16 (9.8, 19.5) versus 10 (5.5, 22.5) (P = 0.549; 53.7% missing data); hospital LOS 7 (4, 10.3) versus 7 (5, 11.8) days (P = 0.343); 3-month mortality 9.9 versus 12.5% (OR 0.77, 95% CI 0.34 to 1.75), respectively. No treatment-limiting side effects were noted. Conclusions prophylactic low-dose oral haloperidol did not reduce delirium incidence in acutely hospitalised older patients. Therefore, prophylactic use of haloperidol in this population is not recommended.