By Tom Heaton
Any ideas? After a longer than expected gap, welcome back to the second part of our quick look through the recent guidance from the Society for Ambulatory Anesthesia on the management of post-operative nausea and vomiting (PONV). In the first post we had a quick look at some of the risk factors for PONV, particularly looking at the weighting of different factors. As a follow up, this post looks at the next components of the guidelines where the authors look at the evidence of different interventions for reducing the incidence of PONV.
Reducing Risk
They start by outlining strategies to reduce the baseline risk of PONV. They summarise these as:
Other than the recommendations to avoid volatile anaesthetics and minimise intraoperative opioids (which are strength A2), these all have a grade A1 level of recommendation. The absolute impact of these interventions is a little bit harder to unpick, though it is probably clear from the risk ratios described in the first post how avoiding the risk factors might work, with many of the recommendations being fairly obvious.
They do outline the impact of some of the strategies that they describe though. Of note, the incidence of PONV is around 9 times less in patient receiving regional rather than general anaesthetic. In addition, for those that do require general anaesthesia, the use of total intravenous anaesthesia (TIVA) reduced the incidence of PONV by 25% in one study, with an NNT of 5 in another study. Not bad.
Their description of the impact of nitrous oxide on PONV is limited to just a couple of lines, noting some increased risk but that this was probably limited in low risk patients. This is something I hope to have a look at in the next post though (which I have found pretty interesting) so if you’re not using nitrous due to its emetic effect, keep your eyes peeled..
- Avoiding general anaesthesia through regional techniques
- Use of propofol infusion for general anaesthesia
- Avoidance of nitrous oxide
- Avoidance of volatile anaesthetics
- Minimisation of perioperative opioids
- Adequate hydration
Other than the recommendations to avoid volatile anaesthetics and minimise intraoperative opioids (which are strength A2), these all have a grade A1 level of recommendation. The absolute impact of these interventions is a little bit harder to unpick, though it is probably clear from the risk ratios described in the first post how avoiding the risk factors might work, with many of the recommendations being fairly obvious.
They do outline the impact of some of the strategies that they describe though. Of note, the incidence of PONV is around 9 times less in patient receiving regional rather than general anaesthetic. In addition, for those that do require general anaesthesia, the use of total intravenous anaesthesia (TIVA) reduced the incidence of PONV by 25% in one study, with an NNT of 5 in another study. Not bad.
Their description of the impact of nitrous oxide on PONV is limited to just a couple of lines, noting some increased risk but that this was probably limited in low risk patients. This is something I hope to have a look at in the next post though (which I have found pretty interesting) so if you’re not using nitrous due to its emetic effect, keep your eyes peeled..
Antiemetic Prophylaxis
After this brief run through of how best to reduce the baseline risk, the authors then turn their attention to the use of prophylactic antiemetic agents. Obviously the list of different anti-emetic drugs is lengthy, and indeed with this being an American group some of the drugs are quite different from those commonly in use in the UK (I could see no mention of cyclizine for example). They do have a more detailed look at many of the antiemetic classes, including some values for number needed to treat (NNT) and number needed to harm (NNH) which I found useful, but I won’t spend time going into these here (it’s well worth a look at these though).
I’ll make one exception to this by having a look at ondansetron which they described as the ‘gold-standard’ antiemetic by which the others are compared (and fortunately the one I see used the most). They quote a NNT of 6 for preventing vomiting and NNT of 7 for preventing nausea and that it is best given at the end of surgery. With regards to side effects, the NNH from a single dose is quoted as 36 for headache, 31 for liver enzyme derangement, and 23 for constipation. The risk of QT prolongation at doses less than 16mg is thought to be not significant, with a handful of case reports described where arrhythmias may have been attributed to its use.
Moving on, the general recommendations they make for prophylactic antiemetic use can be summarised as such (a more user friendly flow chart is presented in the paper itself):
I’ll make one exception to this by having a look at ondansetron which they described as the ‘gold-standard’ antiemetic by which the others are compared (and fortunately the one I see used the most). They quote a NNT of 6 for preventing vomiting and NNT of 7 for preventing nausea and that it is best given at the end of surgery. With regards to side effects, the NNH from a single dose is quoted as 36 for headache, 31 for liver enzyme derangement, and 23 for constipation. The risk of QT prolongation at doses less than 16mg is thought to be not significant, with a handful of case reports described where arrhythmias may have been attributed to its use.
Moving on, the general recommendations they make for prophylactic antiemetic use can be summarised as such (a more user friendly flow chart is presented in the paper itself):
- At the moment blanket prophylaxis isn’t recommended (though may become more so with cheaper generics becoming available).
- Each patient’s risk should be calculated based on the different patient, anaesthetic and surgical factors.
- Baseline risk factor reduction should be used where possible and the impact of both PONV and antiemetic side effects on the patient should be considered.
- If the PONV risk if low, a wait and see approach is advocated
- If the risk is medium, single or dual therapy prophylaxis should be used
- If the risk is high, more than 2 interventions should be employed
Anything Else
As well as going through the role of prophylaxis they look at an approach to take for responding to PONV when it has occurred. The main point they make for this is the recommendation that treatment is with a different class to any antiemetic that has already been used, rather than more of the same (I’m specifically thinking of ondansetron here). That being said, the 5-HT3 antagonists are the main class that have been studied for this role and this is at a lower dose than when used for prophylaxis. Ondansetron of just 1mg IV has a NNT of between 4 and 5 for treating established PONV. Also not bad, though it is more than likely that it will have been used already for prophylaxis.
Finally, a point I found particularly interesting was the evidence behind propofol's antiemetic potential. I had previously thought that some of known antiemetic effect may have been attributable to the avoidance of volatile (e.g. in TIVA). However, sub-hypnotic doses have been demonstrated to be as effective as ondansetron for prevention and treatment of PONV. They describe the option of a small dose (20mg) for treating PONV in an environment with close monitoring when PONV is very problematic, the main downside being the short duration (reducing the conscious level of someone who is vomiting though…?).
Finally, a point I found particularly interesting was the evidence behind propofol's antiemetic potential. I had previously thought that some of known antiemetic effect may have been attributable to the avoidance of volatile (e.g. in TIVA). However, sub-hypnotic doses have been demonstrated to be as effective as ondansetron for prevention and treatment of PONV. They describe the option of a small dose (20mg) for treating PONV in an environment with close monitoring when PONV is very problematic, the main downside being the short duration (reducing the conscious level of someone who is vomiting though…?).
Final Thoughts
Well that’s about it. The document is a pretty extensive one and there is a huge amount of other stuff that I’ve not mentioned to try and prevent this post getting even longer. Whilst there were still areas that I wish they had been able to provide more evidence for, I’ve come away from reading these guidelines with a greater understanding of how best to assess risk of, prevent and treat PONV with a best-evidence based approach. My personal feeling (based on limited illness experience) is that nausea is worse than mild to moderate pain. It is recognised as a big fear that many patients have about having an anaesthetic and so is something I want to try and get right for my patients. I hope this can help provide me with some structure to achieve this.
Thanks for reading and, as always, please let me know your thoughts on the topic (or any others). We shall hopefully have some more interesting posts heading this way shortly. And if anyone was still wondering, the picture is of ondansetron's molecular structure (couldn't find any other good ones is my excuse). Until next time...
Thanks for reading and, as always, please let me know your thoughts on the topic (or any others). We shall hopefully have some more interesting posts heading this way shortly. And if anyone was still wondering, the picture is of ondansetron's molecular structure (couldn't find any other good ones is my excuse). Until next time...
Useful links
Consensus Guidelines for the Management of PONV - Society for Ambulatory Anesthesia
Medscape - Prevention of PONV
APAGBI - Guidelines on preventing PONV in children (2009)
Consensus Guidelines for the Management of PONV - Society for Ambulatory Anesthesia
Medscape - Prevention of PONV
APAGBI - Guidelines on preventing PONV in children (2009)
RSS Feed
