There are several reasons why multimodal concepts are used in anesthesia. One of the most important reasons is to reduce the severity of postoperative pain.

Whereas sometimes the combination of paracetamol and NSAIDs or the combination of general anesthesia with peripheral nerve blocks is named multimodal analgesia, most use the term of multimodal analgesia for adding drugs like gabapentinoids, dexamethasone, ketamine and/or magnesium.

If starting with enhanced multimodal concepts one of the first questions is: which of the above named substances should I use in which dosing and in which combination?

The answer is not too easy and it is important to understand why the answer is not too easy.

Let’s compare it with the multimodal concepts against PONV using the three substances dexamethasone, ondansetron and droperidol. a single bolus of a widely accepted dose for all three drugs is applied. In contrast, for multimodal concepts for postoperative analgesia, there is much more discussion about single bolus versus continuous application, which dose, when to start and how long to continue.

For each of the four substances gabapentinoids, dexamethasone, ketamine and magnesium this could be solved with dose findings studies. But several problems arise:

1. the PROSPECT group claims: ‘Existing general guidelines for perioperative pain management do not consider procedure-specific differences in analgesic efficacy or applicability of a given analgesic technique. For the clinician, an evidence-based, procedure-specific guideline for perioperative pain management is therefore desirable.’

Interestingly, no one claims that for PONV also a procedure-specific approach should be chosen. Dose finding studies for each of the four substances for every surgical procedure are a very challenging target.

1. the clinical endpoints for PONV studies are well defined. Vomiting, yes or no, is an objective binary measure. Nausea is more subjective but still binary, yes or no.

Postoperative pain as a clinical endpoint is more difficult. Effectively, there are two endpoints: subjective patient-rated pain scores and consumption of analgetic drugs (e.g. postoperative opioids). But these two measures are not independent of each other. Theoretically, the amount of postoperative analgetic drugs can be fixed and then the patient-rated pain scores can be measured. But this approach is not common. In statistical terms pain scores are an ordinal variable. They can be ordered but the spacing between the values may not be the same across the levels of the variables. For example, a pain score of 4 is not twice as painful as a pain score of 2, statistical calculations of a mean are not meaningful and care has to be taken for choosing the right statistical measure which commonly needs bigger numbers, i.e. more patients in each group, for reaching statistical significance compared to statistical measures used for interval variables. Alternatively, the patients could be given easy and (nearly) unlimited access to postoperative opioids with a PCA pump, assuming that then all patients would titrate themselves to even acceptable pain scores. This is the more common approach. Interestingly, this is also statistically challenging. It is known that the postoperative consumption of analgetic drugs like opioids has a skewed distribution. But most statistical test like t-test or ANOVA assume that the distribution of the sample means are normally distributed. Additionally, choosing this study design often leads to a conclusion like: ‘Whereas substance Y reduced postoperative opioid consumption by X percent, no difference in pain scores was seen. This questions the clinical meaning of substance a for reducing postoperative pain.’

If these are the problems for a dose finding study of one single substance used in multimodal concepts for one single procedure it can be easily extrapolated to studying all possible combinations of multimodal components in all possible specific surgical procedures.

One additional problem is what I call the ‘ground fog problem’. For most patients the most important item is whether the post-operative pain is acceptable or not. Assuming that a pain score of 4 or lower is acceptable, it is very difficult for patients with acceptable pain to choose a reliable reproducible pain score between 0 and 4 having in mind all other possible uncomfortable sensations associated with the standard postoperative situation. That creates the ‘ground fog situation’: if the control group already has in average an acceptable pain score, it is very difficult to demonstrate a statistically significant improvement with which analgetic (multimodal or not) drug ever.

Comparing different dosing regimens of a multimodal component or comparing different combinations of multimodal components especially for procedures with not the highest postoperative pain scores are prone to the ‘ground fog problem’ (as are comparisons with control groups in procedures with relatively low postoperative pain scores).

Therefore, if in the literature not much difference in effectivity between various dosing regimens could be found, it may make sense to start with the dosing scheme that is most safe and easy to apply. If in the literature the combination of different multimodal components (the more the better) are better than a single multimodal component, it may make sense to start with all four components together.

A possible regimen could be:

150 mg pregabaline p.o. as premedication e.g. one hour before surgery (50% dose reduction in elderly or frail patients)

+ 0,1 mg/kg dexamethasone iv

+ 2000 mg magnesium iv over 10–15 minute (or added in the drip chamber)

+ 10 mg esketamine bolus followed by 10 mg/h esketamine until 30 min before end of surgery (50% dose reduction in elderly or frail patients).

Another important advantage of multimodal anesthesia is the improved hemodynamic stability. Koc et al demonstrated in 2007 in Anesthesia & Analgesia in their article ‘The preoperative use of gabapentin, dexamethasone, and their combination in varicocele surgery: a randomized controlled trial’ that 800 mg gabapentin p.o. and 8 mg dexamethasone iv substantially diminished the increase of heart rate and blood pressure after intubation compared with the control group, while the combination of both gabapentin and dexamethason eliminated the increase of heart rate and blood pressure after intubation completely.

These results proved to be reproducible.

Forget and Cata concluded in 2017 in Best Practice & Research Clinical Anaesthesiology in their article ‘Stable anesthesia with alternative to opioids: are ketamine and magnesium helpful in stabilizing hemodynamics during surgery? a systematic review and meta-analyses of randomized controlled trials’: ‘In conclusion, these meta-analyses of nine trials confirm that ketamine and magnesium, differently but consistently, reduce hemodynamic variability during surgery and may be seen as complementary not only for pain control but also to provide stable anesthesia. This study supports the use of those drugs to control the sympathetic response to surgery in the context of opioid-free anesthesia.’

In our own experience this hemodynamic stability can even further be improved by combining propofol (e.g. 3 mg/kg/h) with a volatile anesthetic (e.g. sevoflurane 0,5 MAC).

Kawano et al. in their 2016 article in Revista Brasileira De Anesthesiologia termed this method of anesthesia ‘combined intravenous-volatile anesthesia (CIVA)’ and showed an effective decrease of the incidence of postoperative nausea in laparoscopic gynecological surgery.

Whereas the improvement in the quality of postoperative analgesia may not always be directly visible for the anesthesiologist working on the OR, the improved hemodynamic stability translates into a direct clinical advantage for the anesthesiologist using multimodal concepts.