Inpatient Delirium

  • Melatonin and Melatonin Agonists
  • Evidence-based Tests
  • Antipsychotics in Delirium


  • Ultra-Brief Delirium Screen:  1.  What day of the week is it?  2.  Months of the year backward.  (Sens 93%, Spec 64%) JHM 2015;10:645-650.
  • However, though the Ultra-brief screen is a good SCREEN, it’s not a great DIAGNOSTIC test to differentiate between dementia vs delirium!

A patient who is elderly and sick enough to be in the hospital has a 25% chance of becoming delirious!

MODIFIABLE RISK FACTORS:  (NEJM 2999:340:669ff.) – notice that 1/3 of these are sensory deprivation issues.

  • Visual impairment
  • Hearing impairment
  • Cognitive impairment
  • Immobility
  • Dehydration
  • Sleep deprivation

Sensory deprivation experiments done in the 1960s by the CIA, Dr Hebb at McGill University

– in less than 24 hours, “subjects progressively lose touch with reality, focus inwardly, and produce hallucinations, delusions…”

BED REST ORDERS ARE THE DEVIL – patients aren’t getting up because WE ORDER IT!

J Am Geriat Soc 2009 – elderly patients spend 95+% of their time IN BED even when the order is “ad lib” movement!

RESTRAINTS – only appropriate for behavior that is a risk to life or to necessary medical care (tubes, lines).  Restraints are associated with 4x increased risk of delirium.

– Try a distraction / activity vest – it’s not a restraint, it’s paper and has a bunch of distracting shit on it!

SLEEP DEPRIVATION –> irritability, impaired attention (sound like delirium?).  The hospital environment is designed to impair sleep!  The 0500 blood draw.  The 0200 vital signs.  Noise from machines.  Noise from staff.  Light from the hallways.  Pain.  Skin care.

  • Stop the vital signs in the wee hours.



– 1 small trial of ramalteon (melatonin agonist) – dramatically lower risk of delirium (3% v 32%, p = 0.003). (JAMA Psychiatry 2014: 71(4).)

– Meta-analysis:  n = 669 elderly patients.  Significant reduction in delirium for elderly patients on medical wards, RR 0.25.  No difference for surgical patients.


  • 3D-CAM: 3 minute diagnostic exam for delirium.
  • Ann Intern Med 2014; 161(8): 554.
  • If not able to do months of year backward, choose a semi-difficult test that requires concentration and attention.
  • Use that same test day after day to track changes.
    • Chem 7, CBC
    • UA, CXR, ABG – UA driven by clinical cues, not routine!  ABG:  hypercarbic respiratory failure looks IDENTICAL to hypoactive delirium!
    • Troponin, EKG – in a patient with dementia, cardiac ischemia may not present classically – may just be delirium!
  • Low yield tests for hospital-onset delirium:  EEG (use only if sxs of seizures), LP (use only if fever with nuchal signs), CT or MRI (yield was only 2.7% – 6 of 220, 4 bleeds, 2 masses/infarcts – half were on anticoagulation).
    • JHM 2014; 9:497-501


  • Review the med list and remove or minimize all the problem medications (BDZ, TCAs, opioids, antihistamines, muscle relaxants, antiemetics, partial anticholinergics (digoxin, prednisolone, furosemide, theophylline)).
    • Arch Int Med 2009;169:1952ff
    • Arch Int Med 2003;163:2716ff
  • Treat withdrawal (EtOH or BDZ)
  • Correct metabolic disturbances (electrolytes, glucose, hydration)
  • Reduce level of invasion (urinary catheters and lines)
  • Assess for and treat infection (new UTI or pneumonia)
  • Improve environment and mobility (stop the bed rest, hearing aids, glasses, better lighting, etc.)
  • Adequately treat pain – uncontrolled pain is a more potent trigger of delirium than opioids!  Scheduled pain meds better than as needed.


What doesn’t work for delirium?  Cholinesterase inhibitors (donepezil, rivastigmine).  Benzodiazepines (except for EtOH withdrawal).

What kind of works for delirium?  Haloperidol and other antipsychotics.  The data is not strong, by the way – RCTs have not shown clear benefit.

  • American Geriatrics Society 2014 – DO NOT PRESCRIBE ANTIPSYCHOTICS for the postop delirious patient unless they are severely agitated and threatening to themselves and others.

If necessary:

  • use haloperidol 0.25-0.5 mg (NOT 10mg for Zeus’s sake!) PO/IM/IV.
  • monitor EKG when using haloperidol (JHM 2010;5:E8-E15) 
    • IV haloperidol is often used unsafely!  16% had QTc > 500 BEFORE the first dose was given!  Only 42% got safe prescribing in one study.  In another study, 9% of elderly got antipsychotics; 1/4 had a prolonged QTc after treatment!


Antipsychotics are often inappropriately continued on discharge!  (Like PPIs – these are very often inappropriately continued on discharge.)

  • Lancet Neurology 2009;8:151ff – stopping an antipsychotic on discharge did not increase bad outcomes, and actually lowered mortality in hospitalized elderly.
  • Stop them on discharge.
  • If you absolutely cannot stop them on discharge, WRITE A STOP DATE so someone (else) has to write to re-start it.

The conclusion about antipsychotics in delirium:  IS THE CURE WORSE THAN THE DISEASE?  They are not clearly effective, but they are clearly harmful…

– From Rocky Mountain Hospital Medicine, 2016, Cumbler


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s