PRODEO: PROfylactic Haloperidol in patients defined as high risk for DElirium with delirium risk mOdel

Polypharmacy is defined as the chronic use of different medications (>5) per
day. In the Netherlands, 30% to 45% of the people over 65 years use more than
5 medicines per day. 20% of the 70 year olds even use more than nine different
medicines a day. Polypharmacy increases the risk of side effects, drug-drug
interactions and has a negative influence on the compliance. It can even
contribute to under-treatment because physicians can be reluctant to prescribe
new medications to patients who already use five or more medicines. Besides
this, polypharmacy is also a risk factor for developing a delirium1,2.
A delirium is a rapidly altered mental state which is characterized by a
fluctuating consciousness, attentional deficit, psychotic features and altered
behavior. A delirium has a higher prevalence among the elderly people and is a
common postoperative complication within geriatric patients with an incidence
varying from 9 to 87%3. A delirium is associated with higher costs, an
increased morbidity and mortality. Costs range from $ 16,303 to $ 64,421 per
patients. Mortality figures rates ranging from 25% to 33%5.

In 2013, the hospital pharmacy at the Zuyderland Medical Centre developed the
DElirium MOdel (DEMO) to determine the change on development of a delirium in
patients aged 60 or older. Patients who were enrolled were not yet diagnosed
with delirium. The study used electronically available data such as age,
polypharmacy, use of anti -dementia, anti-depressants, anti-parkinsonian,
anti-diabetics, analgesics and / or sleeping medication. The DEMO was developed
retrospectively . A logistic regression analysis was used and showed a AUROC
of 0.770 (95% CI (0.736 to 0.804) with a sensitivity of 78.2% and a specificity
of 63.7%. In the study of Gonzalvo et al., the DEMO-model was validated
prospectively.9 Through an observational study newly hospitalized patients were
screened for five days. The average stay was 4.5 days. In total, 383 patients
were analyzed.
According to the original protocol, the primary endpoint was a delirium that
occurred within 24 after admission but there was also an analysis for the
following days. The table below shows the sensitivity and specificity, with the
search terms "delirium" and "delirious" in the medical files (in Dutch:
*delier*, *delirant* of *delirium*).

Table 1:
Sensitivity and specificity analyses at day 1, 3 and 5.

Analyze day Sensitivity Specificity
1 88 73
3 91 75
5 92 76

Research question
Will prophylactic administration of haloperidol to patients with a high risk of
developing a delirium according to DEMO, reduce the incidence of a delirium?

Hypotheses
Prophylactic administration of haloperidol in post-operative patients with an
increased risk of a delirium according to DEMO gives a reduction of 30% on the
incidence of a delirium, compared with the control group.

The PRODEO study is designed as a randomized, controlled, double blind single
center study with two parallel groups and a primary endpoint of delirium
incidence within 5 days after surgery. Randomization will be performed as block
randomization with a 1:1 allocation.
Haloperidol (1mg) or placebo will be administered at set times. Administration
of medication will take place every day 8(am) and 10(pm). If the patient
returns postoperatively to the ward before 16:00(pm), the subject will be given
an extra dose of haloperidol(1mg) or placebo. In this way, a maximum of 2mg
haloperidol will be administered daily to patients in the intervention group.
On the third postoperative day 07.00(am), blood samples will be obtained in
order to determine the circulating concentration of haloperidol. Blood samples
of all subjects who develop a delirium will be assessed for the according
haloperidol concentration. A random sample of blood will be tested of subjects*
negative for delirium in order to compare circulating concentrations of
haloperidol.

Sample size calculation
The sample size was calculated based on the primary study outcome. Delirium
rates for subjects after DEMO-analysis are estimated to occur at a rate of
0.20. (C. Mestres Gonzalvo). The trials by Wang et al. and Fukata et al. show
ambiguous results of incidence of delirium after administration of prophylactic
haloperidol in the general population. After risk-analysis based on the
DEMO-model we expect a reduction of delirium incidence of more than 30%.
Therefore, we estimate a delirium rate of 0.14 in the intervention group
treated with oral prophylactic low-dose Haloperidol. To detect a reduction of
30% with a significance level of a = 0.05 and power of 1-b = 0.80 with equal
allocation to two arms at least 614 subjects are required in each arm of the
trial. To allow for 10% dropout, 683 subjects will be recruited per arm, i.e.
1366 in total.

First of all the risk to develop a delirium will be calculated. If there is an
increased risk, the patient will be randomised.
Haloperidol (1mg) or placebo will be administered at set times. Administration
of medication will take place every day 8(am) and 10(pm). If the patient
returns postoperatively to the ward before 16:00(pm), the subject will be given
an extra dose of haloperidol(1mg) or placebo. In this way, a maximum of 2mg
haloperidol will be administered daily to patients in the intervention group.
On the third postoperative day 07.00(am), blood samples will be obtained in
order to determine the circulating concentration of haloperidol. Blood samples
of all subjects who develop a delirium will be assessed for the according
haloperidol concentration. A random sample of blood will be tested of subjects*
negative for delirium in order to compare circulating concentrations of
haloperidol.

If a patient experiences a delirium, it is decribed as a very stressful
experience for the patient which can lead to additional morbidity and even
increased mortality. The preventive use of low-dose haloperidol has also some
risks, but this appears to be low as the haloperidol is now only used in
patients who are at high risk of delirium.