Lone acute severe headaches are a common and sometimes complex presentation to all Emergency Departments. There are obviously life threatening causes of a headache that we rule out with judicious history taking and examination. The NICE guidelines have been talked about on St Emlyn’s here (1).
Headaches are often debilitating and cause significant pain and suffering to our patients. Treating them promptly and effectively should be bread and butter to emergency physicians.
I suspect most of us have developed our own cocktail for the treatment of migraines from the use of high dose aspirin, oral triptans to IV anti-emetics. In the UK our NICE guidelines advise combinations of oral triptans with an NSAID or paracetamol, or mono-therapy with an oral triptan, aspirin (900mg), NSAID or paracetamol.
This was a single centre, double-blind, randomised control trial conducted in a teaching hospital in the United States. In this trial the investigators wanted to determine if low-dose haloperidol was an effective treatment for acute benign headaches. As always with any JC please take a look at the original article (2).
Who was included?
Patients aged 13-55 presenting to the ED with a headache or migraine were enrolled in the study. Exclusion criteria was fairly pragmatic and included signs and symptoms for other causes of headache. For example: sudden or rapid onset, abnormal neurological examination, fever, trauma, severe hypertension, or clinical concern leading to the ordering of a CT Scan. It is also important to note that patients with a prolonged QT (>450ms) were also excluded.
This trial includes any benign headache and does not differentiate between migraines and other causes of benign headaches.
What did they do?
Once enrolled, patients were randomised to receive 2.5mg IV haloperidol (diluted with 2.5ml of NaCl 0.9%) or to the control group who received a placebo of 5ml NaCl 0.9% flush. Both groups were administered the 5ml IV injection over 1-2 minutes. This administration method meant that the patient, nurses and treating physicians were all blinded.
All patients were placed on cardiac monitoring and had VAS pain scores and vital signs recorded 0, 30, 60, 90 minutes and 25 hours post discharge.
Further analgesia in the form of IV ketorolac (dose adjusted for age and weight) was given to patients at 60 minutes if they had not had at least a 50% reduction in pain scoring,
In other words, this trial compared giving low dose haloperidol to doing nothing about a headache. An alternative would have been to compare the study treatment against standard treatment; mainly because doing nothing (giving a placebo) is probably not the most effective treatment for pain. More about this later.
What did they find?
A total of 118 patients were recruited, with 58 patients in the treatment group and 60 in the placebo. There were no significant differences in baseline characteristics for each group. In total, 78% of patients in this trial were female, only 9 paediatric patients were recruited. Just over half of the patents (59.3%) reported a past medical history of migraines.
The majority of these patients (107) scored their headache as severe on presentation (VAS >7). However, it is worth mentioning that patients in the placebo group started with a mean VAS score of 7.54 compared to 8.5 to 9 in the treatment group.
Unsurprisingly, 47 of the patients in the placebo group (78.3%) required analgesia at 60 minutes. In the treatment group 18 patients (31%) were given further analgesia at their 60 minute assessment.
Overall, it was found that Haloperidol gave a statistically significant reduction in pain relief (reduction in VAS >50%) in comparison to the control group, p < 0.001.
Side effects of treatments were also recorded at each assessment. At 30 minutes Haloperidol had significantly (p= 0.024) more side effects reported. The most common were anxiety and restlessness (total of 10 patients), and the majority of which resolved with antihistamine treatment. Two patients required Lorazepam, and a further two patients required treatment after the 24 hour follow up.
There were no cardiovascular complications reported with no significant difference in QT between the two groups.
What does this all mean?
On first glance this trial ticks many boxes for good methodology. Bias and confounders have been reduced by study design, it is a randomised control study, with double-blinding. Giving a 5ml clear injection to both groups of patients is a sensible way of reducing bias. Its pragmatic approach with convenience sampling means that patient selection could be applied to the majority of emergency departments.
However, the trial compares giving a drug to giving nothing, for what was in most cases severe pain. So at most it has shown that giving Haloperidol is better than giving 5mls of salty water to a patient with a severe headache which is benign in origin. You may recall Rick and Simon expressing similar concerns about other analgesic trials, notably a trial of Methoxyflurane vs. placebo(3) in musculoskeletal pain. Their editorial argued(4) that it is likely unethical to test analgesics against placebo and that we must test them against usual care.
When presented with a patient in severe pain, we would all give analgesia, and not wait 60 minutes to do so. For the patients in the placebo group that did experience a reduction in their pain, it is not clear if these were the same patients with lower baseline VAS scores. It may have been that placing a patient in a dark, quiet room was also effective for treating headache.
It is reassuring that there were no cardiovascular complications with the administration of Haloperidol in this trial, though it still was not without side effects.
This trial is unlikely to change our clinical practice in Virchester for the management of severe benign headache. It would be interesting to see a further RCT comparing Haloperidol to standard treatment.
- NICE on headaches https://www.stemlynsblog.org/nice-on-headaches/
- Treatment of Headache in the Emergency Department: Haloperidol in the Acute Setting (THE-HA Study): A Randomized Clinical Trial. https://pubmed.ncbi.nlm.nih.gov/32402480/
- STOP!: a randomised, double-blind, placebo-controlled study of the efficacy and safety of methoxyflurane for the treatment of acute pain https://pubmed.ncbi.nlm.nih.gov/24743584/
- Methoxyflurane is a better painkiller than placebo: but do we want to know more? https://pubmed.ncbi.nlm.nih.gov/24743586/