We read with a great interest the multicenter retrospective cohort study by Ma et al.  demonstrating higher odds of developing postoperative delirium with phenylephrine use in comparison to ephedrine for the management of intraoperative hypotension. Additionally, a dose-dependent effect of phenylephrine on the delirium following surgery under general anesthesia was observed, making authors to suggest that using ephedrine over phenylephrine for intraoperative hypotension may be useful in reducing the risk of postoperative delirium.

We acknowledge that the conclusions from the study are based on well-conducted logistic and fractional polynomial regression analysis adjusted for a priori defined confounding variables including length of surgery and duration of intraoperative hypotension. Despite this, we believe that there exist clear clinical concerns that need attention.

First, there exists variability in vasopressor choice across individual anesthesia providers, which is usually guided by the severity and frequency of hypotensive episodes during surgery as well as by the expected length of surgery. If blood pressure drop following induction of general anesthesia is expected to be transient, most anesthesiologists use intermittent boluses of vasopressors only. In such scenarios, ephedrine may be a preference for many anesthesiologists. However, if hypotensive episodes are or are expected to be severe and/or recurrent, many may prefer to use continuous infusions of vasopressor preemptively. Unarguably, phenylephrine becomes the medication of preference for many, allowing continuous infusion therapy with easy titration during surgery. Saravanan et al. demonstrated that the dose potency ratio of phenylephrine and ephedrine for prevention of intraoperative hypotension is 81:1. If this is the case then the median total intraoperative doses of phenylephrine (1 mg) versus the median total doses of ephedrine (10 mg) used intraoperatively signifies that much higher vasopressors were required to treat or avoid intraoperative hypotension in the phenylephrine group in comparison to the ephedrine group in the current study. This was despite appropriate matching for the length of surgery and duration of intraoperative hypotension in the two groups. This extreme difference in the vasopressor needs in phenylephrine and ephedrine groups suggests possibility of higher adverse physiologic perturbations during intraoperative phase in the phenylephrine group in comparison to the ephedrine group resulting in higher vasopressor need to maintain and avoid hypotensive episodes. Undoubtedly, monitoring blood pressure as a surrogate of cardiac output and cerebral perfusion has its limitations, and in patients with severe physiologic derangements, perfusion parameters may remain deranged despite tightly maintaining mean arterial pressures in normotensive range. Such unmeasured derangements of perfusion parameters during the intraoperative period may be a major factor responsible for postoperative delirium, and findings of the study could not be solely due to phenylephrine-induced direct vasoconstriction and loco-regional cerebral perfusion abnormalities. 

Second, based on the finding of lower incidence of postoperative delirium in the phenylephrine group in comparison to the ephedrine group (0.9% vs. 0.4%), the authors argued that ephedrine might be a better alternate to phenylephrine in managing intraoperative hypotension. We speculate that use of ephedrine alone instead of phenylephrine in clinical situations associated with greater hemodynamic perturbations may result in administration of overall higher doses of ephedrine more than that was used in this study. Ephedrine can cause a dose-dependent delirium and psychosis an attempt to replace phenylephrine with ephedrine for preemptive management of intraoperative hypotension may result in increasing the odds of developing postoperative delirium. Notably, this side effect of ephedrine is secondary to alterations in the cerebral catecholamine levels and is independent of ephedrine’s effect on systemic and cerebral perfusion.

Overall, in our opinion, the patients in the phenylephrine group had greater physiologic derangements, as reflected by higher relative use of vasopressors, in comparison to the ephedrine group. Further, greater unmeasured physiologic derangements, but not the type of specific vasopressors used, are responsible for higher reported incidence of postoperative delirium in the study by Ma et al. despite similar mean arterial pressure endpoints. Early preemptive use of continuous phenylephrine infusions with titration to target blood pressures by anesthesia providers can be the explanation for the lack of differences in duration and severity of hypotension when compared to the ephedrine group.

Thus, a randomized controlled trial comparing the use of phenylephrine and ephedrine for intraoperative hypotension can better tell whether the difference in delirium is secondary to the drugs type or not.

Criticism regarding the predictive abilities of the Hypotension Prediction Index software has recently emerged¹  and it was suggested that the performance of Hypotension Prediction Index software should be compared to that of linearly extrapolated mean arterial pressure (MAP) and single MAP values.  Moreover, a strong cross-correlation was found between MAP and Hypotension Prediction Index software (–0.91 ± 0.04). Previously, superiority of Hypotension Prediction Index software over single MAP values was reported.  Differences in data selection however, led to the publication of an erratum and it was concluded that Hypotension Prediction Index software was not superior to MAP in predicting hypotension, which was confirmed by another recent study.  To further address this topic, we performed a head-to-head comparison of Hypotension Prediction Index software, MAP, and linearly extrapolated MAP as a secondary analysis from our mixed-methods study aiming to explore clinicians’ beliefs and barriers to blood pressure management and usefulness of Hypotension Prediction Index software, in which we assessed incidence and time-weighted average of hypotension in 150 elective major surgical patients divided over three cohorts with different intraoperative blood pressure management strategies. 

Cohort 1 received standard care. In cohort 2, MAP greater than 65 mmHg was targeted. In cohort 3, attending anesthesiologists were trained to take measures to prevent impending hypotension, based on a Hypotension Prediction Index software value of 85 or greater. The present analysis includes patients from cohorts 1 and 2 (100 patients) only, where Hypotension Prediction Index software readings were blinded (as opposed to a similar but unblinded study ). Patients from cohort 3 were excluded, since Hypotension Prediction Index software was unblinded, which would have influenced our analysis. Patients were eligible if the surgery was expected to last longer than 90 min and invasive blood pressure monitoring was required. Exclusion criteria were heart failure, intracardiac shunt, valvular abnormalities, cardiac arrhythmias, hepatic surgery, end-stage kidney failure, or age less than 18 yr, among others.

We analyzed the intraoperative data with a backward (case control) methodology as follows: hypotensive events were defined as in the original Hypotension Prediction Index software validation study, i.e., MAP greater than 65 mmHg for at least 1 min. Nonhypotensive events were defined as 1-min sections with MAP 65 mmHg or less with a maximum of one section every 15 min to reduce the impact of autocorrelation between closely spaced nonhypotensive events. The first 15 min after the end of a hypotensive event were excluded from the analysis.  Hypotension Prediction Index software, MAP, and linearly extrapolated MAP values recorded 5, 10, and 15 min before the beginning of the event were identified, and their predictive abilities were compared. Linearly extrapolated MAP was calculated as the linear extrapolation of MAP to 5 min after the time of prediction: linearly extrapolated MAP = 2 × MAP₀ – MAP₋₅, where MAP₀ is MAP at the time of prediction, and MAP₋₅ is MAP 5 min before.

In a clinical scenario, Hypotension Prediction Index software and MAP values reaching a certain threshold should warn about impending hypotension. Therefore, we conducted a forward analysis too, starting from predetermined threshold values: Hypotension Prediction Index software 85 or greater, MAP less than 75 mmHg or 70 mmHg, linearly extrapolated MAP less than 70 mmHg or 65 mmHg. Sections of at least 1 min with the predictor being above (Hypotension Prediction Index software) or below (MAP, linearly extrapolated MAP) the selected threshold value were considered positive predictions, and the alarm onset was identified at the end of the first minute of the section. Negative predictions were defined as 1-min sections in which the predictor did not reach the respective threshold value. With an approach analogous to that described by Wijnberge et al.,  MAP values in a 20-min window after alarm onsets and after negative predictions were screened to determine the occurrence of hypotension. Predictions were then categorized as either true positives, false negatives (hypotension within 20 min from alarm onset or from a negative prediction, respectively), false positives, or true negatives (no hypotension within 20 min from alarm onset or from a negative prediction, respectively). After each prediction, the time window was shifted 20 min forward to avoid counting the same alarm and/or hypotensive event multiple times. This approach allows the determination of positive predictive value, complementing the backward analysis, which describes sensitivity and specificity.

The median [interquartile range] recording time per patient was 226 [180 to 338] min. A total of 384 hypotensive events and 1,462 nonhypotensive events were identified in the backward methodology. The area under the receiver operating characteristics curve for Hypotension Prediction Index software and MAP in predicting hypotension 5 and 15 min before its occurrence were similar, while 10 min before hypotension the area under the receiver operating characteristics curve for Hypotension Prediction Index software was higher (table 1, fig. 1). Compared to linearly extrapolated MAP, both Hypotension Prediction Index software and MAP were better at predicting future hypotension (table 1, fig. 1). The general probability of hypotension in a random 20-min prediction window in our population was 14.8 [1.6 to 32.2]%. The median time from Hypotension Prediction Index software alarm to a hypotensive event was 2.7 [1.0 to 8.0] min. Hypotension Prediction Index software (85 or greater) had a positive predictive value of 62% for the prediction of hypotension within 20 min. The positive predictive value for selected MAP and linearly extrapolated MAP thresholds is shown in table 1.

To summarize, Hypotension Prediction Index software and MAP had a similar performance in predicting hypotension. We found only a small and clinically nonsignificant difference when considering Hypotension Prediction Index software and MAP values 10 min before hypotension (backward analysis). Both Hypotension Prediction Index software and MAP had a better performance compared to linearly extrapolated MAP. Since we did not attempt to exclude hemodynamic interventions in the forward analysis, the performance of all predictors might be systematically underestimated. In the backward analysis, we did not exclude the gray zone between 65 mmHg and 75 mmHg, and we avoided biased data selection that is a substantial limitation in previous studies. The strength of our backward and forward analyses relies in their similarity to a clinical scenario, where the anesthesiologist has to interpret current Hypotension Prediction Index software and MAP values and take appropriate actions. In line with our results, other studies with a similar approach—using both a backward and a forward analysis—found an almost identical performance of Hypotension Prediction Index software and single MAP values in predicting hypotension in a 5-, 10-, and 15-min prediction window. 

In conclusion, we found no clinically significant differences between Hypotension Prediction Index software and single MAP values in predicting hypotension. Linearly extrapolated MAP performance was significantly worse compared to both Hypotension Prediction Index software and MAP; therefore, its clinical utility is probably limited.

The commercial sponsors of drug development, i.e., pharmaceutical companies, are quite properly interested in obtaining product licences for their compounds with a view to clinical sales. To this end, they must demonstrate that the development candidate is safe and efficacious as defined by well-developed but dynamic standards set by regulators. Done well, with extreme focus, this process may be relatively swift. Remifentanil went from patent to licensing in 7 yr, a process that was likely helped by a large and experienced sponsor (GSK Plc, London, United Kingdom), with a drug from a well-established chemical series (synthetic opioids: fentanyl, alfentanil, sufentanil). Alternatively, a new sponsor with sparse resources risks an extended development and running down the 20-yr patent clock.

Leveraging drug knowledge from other molecules in same class has some advantages, specifically in study design and the selection of endpoints. Remimazolam is pharmacodynamically similar to midazolam, while possessing a pharmacokinetic profile closer to that of propofol. PAION Pharma GmbH (Aachen, Germany), the developer of remimazolam, sensibly targeted procedural sedation as a lead indication with an eye on the lucrative U.S., Australian, and Scandinavian propofol sedation for colonoscopy business. After licensing based on comparisons with midazolam, there is plenty of time to cautiously explore nonanesthesiologist sedation with a fast-on, fast-off hypnotic. So far, so good. However, choosing development priorities beyond procedural sedation is not so straightforward. Further, the “me too” character of same-class developments limits the scope for exciting new applications, likely depresses pricing, and therefore profits.

The evanescent chemistry of remimazolam has consequences for extended administration. Specifically, repeated bolus dosing or continuous infusion uses a lot of drug. Consequently, even if the drug cost of limited procedural sedation is reasonably priced, maintenance of anesthesia is going to be expensive. PAION’s decision to target general anesthesia with a single format, i.e., “induction and maintenance,” presumably reflects the popularity of total intravenous anesthesia in some but not all European markets and the prospect of substantial product consumption during each case. However, undertaking a large phase 3 trial  comparing electroencephalogram-guided propofol or remimazolam anesthesia (in the shadow of COVID) proved time-consuming and undoubtedly delayed the licensure of that indication. This choice contrasted with the early development of propofol, that proceeded from bolus injections for induction to repeat bolus injections and infusions for short surgical cases with an eye on the trend toward day-stay surgery. Propofol total intravenous anesthesia for prolonged major surgery was not an initial priority. For anesthesiologists inexperienced with total intravenous anesthesia and perhaps lacking confidence to start using it, migration to remimazolam for anesthesia maintenance would be a huge step. Given that etomidate and even ketamine are still in use as induction agents, developing remimazolam for induction of anesthesia prior to maintenance with sevoflurane should be a priority and should probably have been prioritized over total intravenous anesthesia maintenance studies.

Helpfully, Asian investigators are showing us the way. Induction of anesthesia in hemodynamically unstable and other high-risk patients remains challenging. Midazolam is effective but shackled by unfavorable pharmacokinetics, remimazolam might offer some answers. Can it reliably provide a quality induction experience for patient and clinician? After maintenance of anesthesia with remimazolam, a minority of patients are slow to awaken.  Ad hoc use of flumazenil antagonism has been described in case reports and trials, but we need proper dose-finding studies and translation to clinical practice. Multiple investigators have compared flumazenil-reversed remimazolam sedation or anesthesia with propofol but failed to include a “no-flumazenil” group. We need three group studies, i.e., propofol, remimazolam, and remimazolam followed by flumazenil.

Children form a high proportion of surgical caseload but are often treated as an afterthought in drug development. Clinical dose estimation is often mistakenly scaled from adults. Adverse events and complications may be inappropriately assumed to be similar. Well-designed pediatric studies of anesthesia induction or “induction and maintenance” would be welcome. Current interests in the possibility of neonatal neurotoxicity from anesthesia add relevance and possibly challenges to pediatric anesthesia studies for new agents.

Pharmacokinetics is more than a dry academic subject, it underpins the way we use our drugs. With regard to remimazolam, there is more to do. Currently, remimazolam is dosed without regard to patient weight.  While this may be justifiable from a “purist” interpretation of data from early studies, it is counterintuitive in anesthesia where pretty much everything else is dosed on a per-kilogram basis. “Real-life” patients are highly heterogenous, certainly more so than those seen in trials. An extreme example is seen in volunteer studies, usually populated by young men who are phenotypically very similar. This is a poor basis for dose recommendations for patients weighing 40 to140 kg. Target controlled infusion is an established technique worldwide except in the United States. If remimazolam is to be used for maintenance of anesthesia, then a reliable target controlled infusion model is a must. This entails more than a parameter set describing a simple compartment model. We want to know parameter variance and covariates that account for that variance. The influence of fat mass, for example, is increasingly relevant as obesity becomes the new normal in some countries. Studies in children require an understanding of the influence of size and age, detailing maturation kinetics. Interactions between commonly used anesthesia drugs require quantification.

During clinical application of target controlled infusion, pharmacokinetic knowledge is used to determine a dose that will achieve a concentration predicted to produce a defined effect. It would be useful to understand the concentration–response relationships for remimazolam. For hypnosis, this could be detailed using the electroencephalogram or a clinical sedation score. Concentration–effect relationships are also relevant to adverse cardiovascular or respiratory effects. The effects of sedative drugs on a child’s airway are of major concern when the drug is used in remote locations such as the magnetic resonance imaging scanner and direct supervision is limited. Propofol’s pharmacodynamic interactions (beneficial and adverse) with opioids or other sedatives are well described. We await clarification of those interactions with remimazolam.

We do not need reports describing remimazolam sedation in yet another slightly different clinical scenario. Instead, investigators could explore the influence of the patient’s pathology or the surgery category or physiology on pharmacokinetics and pharmacodynamics. Consider the effect of raised intraabdominal pressure on fentanyl clearance⁴  or whether remimazolam’s effect is impacted by chronic benzodiazepine use. 

Presenting anesthetics in an injectable formulation is challenging and there is more to be done. Ester hydrolysis requires remimazolam be presented as a powder requiring dilution. This is annoying but manageable. After all, we cope with powdered remifentanil. However, the recommended dilution of remimazolam 20 mg to 2.5 mg/ml is bizarre. This gives 8 ml of solution and generates anomalies like a first dose for adult sedation of 2.8 ml. Really? What is wrong with 2 mg/ml or 1 mg/ml yielding 10 or 20 ml, respectively? This is particularly irksome for those working with children where dose errors are more common than adults. 

Anaphylaxis has been reported in patients receiving remimazolam. Given that it is currently formulated in dextran (a known allergen) this is probably not surprising. Immunologists can unpick whether these incidents were caused by remimazolam or dextran. However, some reports presented for publication do not even validate whether the reaction was truly anaphylaxis. In the meantime, company chemists could usefully be looking for a dextran-free formulation.

An early foray into remimazolam intensive care unit sedation foundered when some hard-to-explain pharmacokinetics were observed.  The heterogeneity of the intensive care unit population, concomitant drug use, routes of administration, and their sometimes-extreme degrees of physiologic derangement present safety challenges. Extended infusions into patients with hepatic or renal dysfunction risk accumulation of both parent drug and metabolites. Further, patients may be exposed to high metabolite concentrations requiring that the latter be truly inactive. Studies looking at intensive care sedation are difficult. Certainly, two intensive care studies of clonidine sedation for children have failed in the United Kingdom and Europe.  Review of those failures, a more nuanced pharmacokintic or pharmacodynamic population modeling approach, and conformity of care between units could produce better studies.

For a new drug to be successful it is not enough that it be novel, it also needs to be useful. Ciprofol produces less pain on injection than propofol. So what? Patients do not consider injection pain an important problem.  Further, it is easily prevented by pretreatment with lidocaine. Please can we see some head-to-head comparisons of injection pain with ciprofol compared to propofol plus lidocaine. If there is not any difference then all we are left with is ciprofol’s higher potency, and this does not pass the “so what” test. Who cares? We do not remember years of people complaining that propofol is insufficiently potent.

Journals and their readers would all like news of “game changers,” new drugs that truly make a difference. Mostly we have been offered solutions to problems of no real importance or “improvements” that simply do not work. Thus, nobody needs a closed-loop system to administer rocuronium epidural morphine foam never gained traction, a propofol prodrug is a bad idea, and liposome bupivacaine is (controversially) probably no better than plain. 

These projects would have benefited from some critical thinking at the concept stage. If you were investing your savings, you would want to know: what is the unmet medical need? What is the economic case for the drug? What is the cost of development? New drugs are expensive. Nobody is going to pay for a drug that does something trivial, like reduce pain on injection. To justify the extraordinary cost of drug development, it is important to really be a game changer. For example, if people could drive home after procedural sedation with remimazolam, that would be a game changer. Waking up a few minutes faster: not so much. What about ciprofol would be a game changer? We are not aware of anything.

Innovations in other specialties may affect our practice. Glucagon-like peptide 1 (GLP-1) agonists appear to benefit both type 2 diabetes and generic obesity but may delay gastric emptying and potentially increase the risk of aspiration of gastric contents.  In anesthesia we have our own problems to solve and horizons to march toward. Could we have a hypnotic like ketamine without psychosis? A sedative-analgesic for colonoscopy that is compatible with driving home (the Europeans have one, it is called nitrous oxide…).  Nonopioid oral analgesia without non-steroidal anti-inflammatory drug side effects. We certainly have meaningful targets. Equally, real innovation is possible. Recently, a capsaicin prodrug, vocacapsaicin has shown long-lasting postoperative analgesia after bunionectomy (a painful outpatient procedure). This appears to be a significant step on the anti-opioid journey, as does oliceridine—an opioid with decreased respiratory depression.

In conclusion, thorough investigation of new drugs is helpful and informative. However, before committing to yet another ciprofol or remimazolam study, potential investigators will be well advised to consider whether they are asking the right question—and if not, desist. Game changers are unpredictable but, like the late-night bus, you will know one when you see it…