Dr Matt Hart

@propofoldream @MichiIntrona We have one going through ethics at the moment. Main limitation for generalisability will be it's only designed to investigate this in adolescents, so a similar one in adult land would be most helpful...

@MichiIntrona So we desperately need a clinical study designed to find: (1) Time to LOC (2) Time to onset of NMB (3) Range/variance of (1) and (2) For pump vs manual bolus For a clinically useful range of pts

We are, of course, fully aware that this is a technical paper, and that clinical data will eventually be needed(like LOC, BIS, so on). However, we deliberately chose TTPE and Ce/Cp as endpoints because they allow a fair, model-based comparison between the two techniques. In parallel, we are already working on additional analyses that are more explicitly clinical, which will be presented separately. The infusion rate is certainly open to discussion. To be honest, a 3-second bolus appears quite extreme, and the literature provides no clear guidance on what should be considered an appropriate or realistic administration speed. Our intent was simply to explore a range of conditions, including such an upper-limit scenario. What our results clearly show is that the difference in the rise of Ce between the two techniques is far smaller than one might intuitively expect. This, in itself, is an important point that addresses a common misconception. What is certain, however, is that boluses delivered at such high speeds mainly serve to rapidly saturate the central compartment, producing a marked haemodynamic impact, with all the associated consequences. @propofoldream @DocMattHart

@MichiIntrona (Apologies for the broken reply, the character limitation is at times frustrating!)

@MichiIntrona Meaning the difference in TTPE of 20-30 for 2-2.5mg/kg boluses might be seen as unacceptable locally. Were a patient to aspirate prior to intubation, you would likely be criticised by a large number of colleagues and medicolegal system here if you used a "slower" technique.

@MichiIntrona In e.g. AUS/NZ/UK there is emphasis on rapid administration of induction agents. 3s might seem "extreme" from your perspective, but wouldn't be called that here. A quick straw poll of my department here suggests 15s is "slow"! Highlighting again cultural variation.

@MichiIntrona I actually think it's a good study. I think what is highlighted in this discussion though is how the lack of evidence for individual components of the RSI leads to a large variation in acceptable practice in different societal / geographic contexts. This affects practice.

@NicholasChrimes onlinelibrary.wiley.com/share/author/Q… The anecdotal speed difference is why we worked with our local pump manufacturer to add an RSI mode to our pumps, which allows clinicians to give a manual bolus, and the pump will then account for that in its PK modelling. Best of both worlds.

A few things: Manual bolus is over 15secs. I reckon I’d typically deliver the initial bolus over ~3 secs. Time to peak effect is not what’s relevant, it’s time to sleep. Clinically this is not the case. TCI inductions (even for non-RSI) are much slower (too slow) inductions.

@NicholasChrimes Using time to peak effect (~200s) also makes the relative difference percentage they quote smaller. They say a 2.5mg/kg bolus is only 14% slower, based on 229s TTPE vs 200s TTPE for a bolus. 29s difference in time to LOC may be more problematic.

@NicholasChrimes @Anaes_Journal The Orihara study retrospectively analysed 5 years worth of anaesthetic records, so underestimating less likely. Other factors may be at play though, e.g. environmental exposure or sensitisation from high use. Either way I use it like roc: only if I have a good reason 😂.

@DocMattHart I’m not sure what to make of the sugammadex thing. It seems very uncertain, skewed by one Japanese study. This from France just out in @Anaes_Journal suggests incidence is low. …-publications.onlinelibrary.wiley.com/doi/epdf/10.11…

In Australia & NZ you should only be using rocuronium if there’s a benefit to the patient from its rapid onset. That doesn’t necessarily require it to be RSI, could just be challenging FMV, short safe apnoea time, etc. If you just want to reverse w sugammadex, use vecuronium.

@NicholasChrimes @Anaes_Journal I suppose it's one of the problems of trying to make inferences from rare outcomes. Pharmacovigilance data like that from France risks underestimating, but aligns with similar studies from Korea and UK.

@sherlockedverne @Anaes_Journal We used @simtiva_app for our modelling. A really excellent tool!

@Anaes_Journal The simulated pumps such as iTIVA and simTIVA have this feature.... Can be easily incorporated in tci pumps .... A good start!

Total intravenous anaesthesia (TIVA) for rapid sequence induction (RSI) Worried your TIVA infusion is too slow for RSI? Why not work with your local manufacturer to develop an ‘RSI-TCI’ mode for your target-controlled infusion pumps? #anaesthesia #MedTwitter #pharmacology doi.org/10.1111/anae.7…

@mick_kerr @Anaes_Journal We were motivated to develop this precisely because as you say some people still think letting a pump deliver the initial bolus over a minute or more is fine.

@Anaes_Journal Set the induction duration to longer, bolus by hand, press go on the pump, done. People are too worried about the tci being inaccurate from the hand bolus that they forget it always was and end up doing stupid things like letting a pump "R"SI a full stomach at 1200ml/hr(20ml/min)

@mick_kerr @Anaes_Journal That's what this new mode does. You deliver the bolus by hand, and because you've told the pump what bolus amount was given it accounts for it in the TCI model. You get the "best of both worlds"; delivering a manual bolus but preserving the model.

@Andyrooz22 @Anaes_Journal That's what this mode lets clinicians do. You deliver a manual bolus that you feel is appropriate for the patient, as a rapid manual push, and the pump accounts for this because you've told it what manual bolus you gave.

@Anaes_Journal I just bolus what I think is needed for the clinical situation.

@DrGeorgeHarvey @Anaes_Journal The orange line represents the simulation of what this new mode does (a clinician delivered bolus that the pump is aware of and incorporates in its modelling) vs what pumps do at present (20ml/min infusion speed).

@Anaes_Journal Still looks like at 60 seconds concentration is much lower

@foi_requests @Anaes_Journal The orange line represents the simulation of what this new mode does (a clinician delivered bolus that the pump is aware of and incorporates in its modelling) vs what pumps do at present (20ml/min infusion speed).

@DrNickB_ObAnaes @Anaes_Journal It's available on all Arcomed pumps now I believe. Other manufacturers could likely also implement similar functions if they felt so inclined.

@DocMattHart @Anaes_Journal Would love to have access to that function

"... a ‘RSI–TCI’ mode with a pre-programmed manual bolus offers a pragmatic solution for those who wish to preserve an intact pharmacokinetic model for accurate propofol delivery, while allowing the rapid induction clinicians expect from a manual bolus." #anaesthesia #MedTwitter doi.org/10.1111/anae.7…

@DrNickB_ObAnaes @Anaes_Journal The slight difference here is that the anaesthetist themself delivers the bolus manually (not limited by pump infusion rates), but the TCI pump is aware of this bolus and incorporates it into the PK-PD modelling.

@Anaes_Journal This was possible with the older BBraun TCI pumps - you would set the target hit pause, hit bolus program how many mls bolus to give. The pump would deliver bolus & then calculate the value & start TCI accordingly The new Braun touch screen versions do not allow this!! Please fix

@johncampbell01 @Anaes_Journal The idea being "best of both worlds". Clinicians get to deliver their RSI bolus manually, but the TCI model remains "intact".

@johncampbell01 @Anaes_Journal The blue line simulates what TCI pumps currently do (bolus delivered at 20ml/min), whereas the orange line simulates this new mode; a clinician delivered rapid manual bolus which the pump knows about and incorporates into its modelling.