Paracetamol for pain – Enteral or Parenteral is best?

This review is as hot off the press as they come and coincides nicely with a discussion at handover on Friday morning.

Why do we obsess and feel the patient is getting a better drug when they have it IV?
Is this something we have learnt?
Are we delivering ‘stronger’ medicine?
Does it feel more ‘hospital-ly’ giving something by a drip?
Maybe there’s something in it, or maybe its all codswallop.
We weren’t really sure, but then I stumbled across this.
I was thrilled. So much so I interrupted my colleague’s (the regional HOS) trip to see ‘The Everly pregnant brothers’ with a text and a link to the article. true FOAM. Learning anytime, anywhere. (Oh and if you’re wondering – they’re a Sheffield institution and you can sample some of their finery below….beware parental advisory!)

Clinical Question: Does Intravenous paracetamol give better analgesia than oral preparation?
Title of the paper: Intravenous Vs Oral paracetamol for acute pain in adults in the Emergency Department setting. A prospective, double blind, double dummy, randomised controlled trial.
Published: EMJ 2017

Population: Adults in moderate to severe acute pain in ED (with exclusions)
Intervention: 1 gram IV paracetamol and a dummy oral tablet
Comparitor: Placebo IV preparation and Oral paracetamol 1 gram
Outcome: Reduction in Visual analogue scale pain score at 30mins.

This was a single centre study from an Australian ED which as we know can limit the studies external validity, in that, there may be something intrinsically different about the patients in that location. Maybe they’re heroic batsmen that English bowlers haven’t got a hope in hell against. I’ve been told by people who’ve worked in Australia that folks from the Northern territory seem hardier than the rest of the population, and maybe theres something in that, akin to us nearly always triaging a farmer to resus if they present with a medical problem. I guess the point is that sometimes, just sometimes, a local population is special and doesn’t reflect yours, who are also special, but in a different way.

A power calculation was done and identified a need for 44 patients in each group to detect a clinically significant difference of 15mm in VAS scores for pain (13mm or 20mm being deemed significant depending on which studies you read – most I have read seem to suggest 13mm is a reasonable number to aim at when studying an intervention).

Patients were enrolled if they had what was deemed to be moderate/severe pain (>40mm on VAS) 5 minutes after having an IV opiate, however there was a lot of scope for clinicians to intervene and suggest they were not ‘enrollable’. Notable exclusions were those known to have or be suspected of opiate dependency (which to me is a group I’d like to know if oral or IV paracetamol had a significant effect on) and pregnant patients, again a group where one would potentially like to ‘go easy’ on the opiates with depending on complaint and how close to term mum is/was.
Another question for me is the timing of enrolment. 5 minutes after an IV opiate. Is that long enough for it to work? Maybe, depending on whats been used, and maybe depending on the dosage. but not universally. I can’t find a description in the paper as to what and how much was administered. For us the most commonly used opiate is morphine, largely due to the logistics that the nurse are happy to administer it, usually in the resus room I reach for fentanyl or ketamine. Some potential again for these patients not to represent yours, and also for the results to be influenced by what went before enrolment.

Patients and treating staff were blinded to the treatment and allocation thanks to computer generated randomisation to sealed envelopes to maintain allocation concealment to a distant (but unblinded) pharmacist providing identical looking trial drugs locally – so one could argue there is the opportunity to introduce bias here, be it a steamed envelope in pharmacy or a knowing nod from the pharmacist on delivery of medications to the dept, I was taught to think that anything may happen in the world of clinical research…..

Results: They had only 87 patients in their modified intention to treat analysis due to invalid consent in 34 patients and a a number having initial VAS pain scores <40mm. There was no significant difference (-2.6mm 95% CI -13.2-7.9mm) between the IV and oral groups. These limits are broad given the degree of difference they hoped to find, and probably reflects the small sample size. There were rescue opiates used liberally in both groups, 86% Vs 84% (IV Vs Oral), which in itself may reflect that neither group had large analgesic effect from the paracetamol (in fact 12 and 8 patients in each group had a 50% reduction in pain at 30mins). For me this paper was great in that it addressed a question I'd had the very same day it came to press. There are some flaws in the design but it is a useful piece of information in the treatment of pain in ED. It was interesting to see the high use of rescue opiates across the groups and the lack of difference between IV or Oral preparations in this group of patients, and I wonder (and hope) that this trial will be repeated on a larger scale with some of the crinkles ironed out. Thanks for reading. Keep fighting the good fight. And lets hope boxing day goes better than the last 3...... Chris Refs

Leave a Reply

Your email address will not be published. Required fields are marked *