Haloperidol

Placebo-controlled studies

There has been a randomized, placebo-controlled comparison of haloperidol (mean dose 1.8 mg/day), trazodone (200 mg/day), and behavior management techniques in 149 patients with Alzheimer’s disease [ ]. Although 34% of the subjects improved relative to baseline, there were no significant differences in outcomes among the four arms; there were significantly fewer cases of bradykinesia and parkinsonian gait in those given behavioral therapy. These results suggest that other treatments for agitation in dementia need to be considered and evaluated; likewise, they are consistent with the results of a meta-analysis [ ] and a clinical trial [ ].

Patients with acute migraine were randomized to two groups and received either haloperidol 5 mg intravenously in 500 ml of isotonic saline or saline alone [ ]. There was significant pain relief in 80% of the patients treated with haloperidol (n = 20), whereas only three patients (15%) responded to placebo (n = 20). The most common adverse reactions to haloperidol were sedation and akathisia, the latter being more troublesome and occurring in 53% of the patients; 16% of the patients considered the adverse reactions intolerable and would not like their migraine attacks to be treated with haloperidol again.

Cardiovascular

Intravenous haloperidol is often prescribed to treat agitation, and torsade de pointes has on occasions occurred [ , ]. In a cross-sectional cohort study QT _c intervals were measured before the intravenous administration of haloperidol plus flunitrazepam, and continuous electrocardiographic monitoring was performed for at least 8 hours after (n = 34) [ ]; patients who received only flunitrazepam served as controls. The mean QT _c interval after 8 hours in those who were given haloperidol was longer than in those who were given flunitrazepam alone; four patients given haloperidol had a QT _c interval of more than 500 ms after 8 hours. However, none developed ventricular tachydysrhythmias.

In a case–control study, haloperidol-induced QT _c prolongation was associated with torsade de pointes [ ]. The odds ratio of developing torsade de pointes in a patient with QT _c prolongation to over 550 ms compared with those with QT _c intervals shorter than 550 ms was 33 (95% CI = 6, 195). The sample consisted of all critically ill adult patients in medical, cardiac, and surgical intensive care units at a tertiary hospital who received intravenous haloperidol and had no metabolic, pharmacological, or neurological risk factors known to cause torsade de pointes, or if the dysrhythmia developed more than 24 hours after intravenous haloperidol. Of 223 patients who fulfilled the inclusion criteria, eight developed torsade de pointes. A group of 41 patients, randomly selected from the 215 without torsade de pointes, served as controls. The length of hospital stay after the development of haloperidol-associated torsade de pointes was significantly longer than that after the maximum dose of intravenous haloperidol in the control group. The overall incidence of torsade de pointes was 3.6% and 11% in patients who received intravenous haloperidol 35 mg or more over 24 hours.

Several cases of torsade de pointes have been reported with intravenous haloperidol used with low-dose oral haloperidol [ ].

The effects of haloperidol dose and plasma concentration and CYP2D6 activity on the QT _c interval have been studied in 27 Caucasian patients taking oral haloperidol (aged 23–77 years, dosages 1.5–30 mg/day) [ ]. Three patients had a QT _c interval longer than 456 ms, which can be considered as the cut-off value for a risk of cardiac dysrhythmias. There was no correlation between QT _c interval and haloperidol dosage or plasma concentrations or CYP2D6 activity.

Asystolic cardiac arrest has been reported after intravenous haloperidol [ ].

In one case, the use of carbamazepine and haloperidol led to prolongation of the QT _c interval and cardiac complications [ ].

• A 75-year-old man developed ventricular fibrillation and cardiac arrest after intravenous haloperidol [ ]. His past history included coronary bypass surgery and coronary angioplasty. As he continued to have severe chest pain, emergency angioplasty was performed. On day 3 he received haloperidol by infusion 2 mg/hour, with 2 mg increments every 10 minutes (up to 20 mg in 6 hours) as needed for relief of agitation. Before haloperidol, his QT _c interval was normal; after haloperidol it increased to 570 ms. The next day he developed ventricular fibrillation. Subsequent electrocardiograms showed prolonged QT _c intervals of 579 and 615 ms, and haloperidol was withdrawn; the QT _c returned to normal.

• A 76-year-old man developed torsade de pointes while taking tiapride 300 mg/day; the QT _c interval 1 day after starting treatment was 600 ms; the dysrhythmia resolved when tiapride was withdrawn [ ].

• A 39-year-old man died suddenly 1 hour after taking a single oral dose of haloperidol 5 mg [ ]. He had myasthenia, alcoholic hepatitis, and electrolyte abnormalities due to inadequate nutritional state. His electrocardiogram showed prolongation of the QT _c interval (460 ms). Autopsy showed a cardiomyopathy but no explanation for sudden death.

However, malignant dysrhythmias can occur without changes in the QT interval.

• A 64-year-old woman underwent coronary artery bypass surgery and was given intravenous haloperidol for agitation and to avoid postoperative delirium; she developed torsade de pointes [ ].

• Asystolic cardiac arrest occurred in a 49-year-old woman after she had received haloperidol 10 mg intramuscularly for 2 days; no previous QT _c prolongation had been observed [ ].

Respiratory

A single case of fatal pulmonary edema was reported in 1982 in association with haloperidol [ ].

Nervous system

Three cases of radial nerve palsy were reported in demented elderly patients confined to wheelchairs who were treated with haloperidol. The combination of extrapyramidal and sedative adverse reactions, added to wheelchair confinement, may have resulted in pressure on the upper arm with subsequent neuropathy [ ].

Endocrine

The prevalence of hyperprolactinemia in patients with chronic schizophrenia taking long-term haloperidol has been studied in 60 patients in Korea (28 women; illness mean duration, 15.5 years) [ ]. There was hyperprolactinemia, defined as a serum prolactin concentration over 20 ng/ml in men and 24 ng/ml in women, in 40; the prevalence of hyperprolactinemia in women (93%) was significantly higher than in men (47%). There was also a significant correlation between haloperidol dose and serum prolactin concentration in women, but not in men.

The relation of prolactin concentrations and certain adverse events has been explored in large randomized, double-blind studies. In 813 women and 1912 men, haloperidol produced dose-related increases in plasma prolactin concentrations in men and women, but they were not correlated with adverse events such as amenorrhea, galactorrhea, or reduced libido in women or with erectile dysfunction, ejaculatory dysfunction, gynecomastia, or reduced libido in men [ ].

No further rise in plasma prolactin concentration was observed with dosages of haloperidol over 100 mg/day, which was explained as being related to saturation of the pituitary dopamine receptors by a modest amount of haloperidol [ ].

The time-course of the prolactin increase has been examined in 17 subjects whose prolactin concentrations rose during the first 6–9 days of treatment with haloperidol [ ]. The increase was followed by a plateau that persisted, with minor fluctuations, throughout the 18 days of observation. Patients whose prolactin concentrations increased above 77 ng/ml (n = 2) had hypothyroidism, and it is known that TRH (thyrotropin) stimulates the release of prolactin [ ].

The effects of haloperidol and quetiapine on serum prolactin concentrations have been compared in 35 patients with schizophrenia during a drug-free period for at least 2 weeks in a randomized study [ ]. There was no significant difference in prolactin concentration between the groups at the start of the study; control prolactin concentrations were significantly lower with quetiapine than with haloperidol. Two patients taking haloperidol had galactorrhea related to hyperprolactinemia.

Metabolism

Glucose metabolism has been studied in 10 patients taking haloperidol; none had impaired glucose tolerance and only one had a glycemic peak delay [ ].

Liver

Occasional reports of cholestatic jaundice with haloperidol have been published [ , ].