Thread by @WellingtonICU

Our weekly #tweetorial returns. This week’s paper is from last month’s NEJM - Haloperidol & Ziprasidone for Treatment of Delirium in Critical Illness
nejm.org/doi/full/10.10…
Prepared by @Taffydoc1

ICU delirium remains a huge challenge, affecting 50-75% of our ventilated patients. The implications on patients and healthcare include a higher mortality, longer hospital LOS and periods on mechanical ventilation and long-term effects on cognitive function.

Despite over 40yrs of treatment by clinicians with antipsychotics, there remains little evidence for their use.
Haloperidol and Ziprasidone are dopamine D2 antagonists and are classed as typical and atypical antipsychotics respectively.

This DB PC RCT in 16 US ICUs from Dec 2011 to Aug 2017 enrolled patients with acute respiratory failure or shock (invasive ventilation (MV), NIV, vasopressors or mechanical cardiac support) with hypo- or hyperactive delirium.

They were randomised to receive 12 hrly boluses of haloperidol (max. 20mg/day), ziprasidone (max. 40mg/day) or placebo for the 14-day intervention period.

Primary end-point no. days alive free of delirium or coma in 14d intervention period. Secondary endpoints included 30d and 90d survival, freedom from MV, time to ICU and hospital discharge. Safety endpoints included extrapyramidal symptoms, excessive sedation and prolonged QTc.

The authors hypothesised that typical and atypical antipsychotics would result in shorter periods of delirium and coma than placebo and would improve other outcomes.

Delirium was assessed using the CAM-ICU tool for 5 days, until death or ICU discharge.
Of 20,914 screened, 1183 were eligible and consented. 566 (48%) developed delirium, 89% hypoactive. Baseline characteristics were similar between groups.

The median number of days alive without delirium or coma was 8.5, 7.9 and 8.7 in the placebo, haloperidol and ziprasidone groups respectively (P=0.26 for overall effect across groups).

There was no significant effect on the primary endpoint (OR 0.88 [95% CI 0.64-1.04, and 1.04 [CI 0.73-1.48] for haloperidol and ziprasidone Vs placebo respectively.

There were no significant between-group differences for secondary endpoints or extrapyramidal symptoms although prolongation of the QTc was commoner in the ziprasidone group, & 2 patients in the haloperidol group developed Torsades de Pointes.

Limitations include: potential selection bias (large number (74%) refused consent) & may be underpowered to detect 2-day difference between groups as lower-than-predicted incidence of delirium (48%)

Other potential limitations include use of antipsychotics to treat hypoactive delirium (not standard practice) & generalisability to ICUs outside the US given variations in practice.

These results are in keeping with those from two smaller placebo-controlled trials (MIND, 2010 – haloperidol Vs ziprasidone Vs placebo and HOPE-ICU, 2013 – haloperidol Vs placebo).
thelancet.com/journals/lanre…
journals.lww.com/ccmjournal/ful…

Commentary: it seems to be a challenging task to assess effectiveness of antipsychotic medications on ICU delirium given multitude of mechanisms leading to its development.

These include the use of GABA agonists, pre-morbid cognitive function and lifestyle choices, and even dare I say, the effect of the lunar calendar!

There remains little evidence to favour the use of these drugs, and I often wonder whether we are truly treating our patients, or merely our workloads.

This #tweetorial was prepared by @wellingtonICU SR, Dr.Nia Davies @Taffydoc1

@threadreaderapp unroll