Secondary Logo

Journal Logo

When Duzitol Does Not Do It All

The Two Sides of Drug Synergy

Krejcie, Tom C., MD; Avram, Michael J., PhD

doi: 10.1213/ANE.0b013e3182251509
Editorials: Editorials

From the Department of Anesthesiology, Northwestern University Feinberg School of Medicine, Chicago, Illinois.

Funding: None.

The authors declare no conflicts of interest.

Reprints will not be available from the authors.

Address correspondence to Tom C. Krejcie, MD, Department of Anesthesiology Northwestern University Feinberg School of Medicine, Ward 13–195, 303 East Chicago Avenue, Chicago, IL 60611-3072. Address e-mail to

Accepted May 6, 2011

The range of effects that are part of a clinically adequate state of general anesthesia are hypnosis (loss of appropriate response to verbal command), amnesia (lack of recall of intraoperative events), analgesia or pain relief along with blunting of the adrenergic response to noxious stimuli, and suppression of movement in response to noxious stimuli. These effects are produced by distinct pharmacologic actions even when they are produced by one drug.1 In their 1992 editorial, Shafer and Stanski provided the pharmacokinetic profile of a fictitious IV-administered anesthetic, Duzitol (a combination amnestic, hypnotic, analgesic, and muscle relaxant).2 Unfortunately, such an IV anesthetic has yet to be introduced into clinical practice. The IV-administered anesthetic agents that are available are drugs that have an affinity for specific receptors and, unless given in very high doses, produce very specific effects. Therefore, to produce a clinically adequate state of general anesthesia with the drugs in the contemporary anesthetic armamentarium, drugs from at least 2 different classes must be used in combination. When drugs are used in combination, their effect can be infra-additive (i.e., antagonistic), additive, or supra-additive (i.e., synergistic). Antagonistic interactions generally involve deliberate reversal of the effects on one drug by another (e.g., flumazenil reversal of the effects of a benzodiazepine).3 Additive interactions are typically observed when drugs exert their effect through interaction with the same receptors or receptor subtypes (e.g., minimum alveolar concentration [MAC] additivity of inhalation anesthetics), whereas synergistic interactions are observed when drugs produce their effects by different or slightly different mechanisms.3,4

One method to graphically present drug interactions is by using an isobologram.5 The 2 axes of the isobologram ideally represent the effect-site concentrations of the 2 drugs of interest (Fig. 1). The line on the graph represents the drug concentration pairs that elicit the same response. Infra-additive interactions are represented by a line that curves away from the graph origin (zero–zero concentration point); adding a second drug at a low concentration requires a higher concentration of the first drug to elicit the same response. If the drug interaction is additive, the x- and y-axis intercepts are connected by a negative-sloping straight line. Finally, with a synergistic relationship, the interaction line is curved toward the graph origin, which is equivalent to a left shift of the first drug's concentration–response curve. Synergy, then, is the addition of a low concentration of a second drug to a greatly reduced concentration of the first drug to attain the same response. In their analysis of the literature on anesthetic drug interactions, Hendrickx et al. concluded that opioids interact synergistically with not only volatile anesthetics but also IV-administered hypnotics.4 Unfortunately, drug interactions can be a double-edged sword; synergy is often observed for both the desirable aspects of anesthesia or sedation and undesirable side effects, including respiratory and cardiovascular depression; little is known or has been studied about these latter interactions.3,4

Figure 1

Figure 1

Volatile anesthetics can produce a clinically adequate state of general anesthesia with a single drug. When the volatile anesthetic is classified as an immobilizer, the MAC of that drug can be determined as the minimum brain-equilibrated alveolar concentration that prevents half of those tested from moving in response to a noxious stimulus (i.e., the median concentration). Although opioids can significantly reduce the MAC of a volatile anesthetic, there is no dose of an opiate alone that will prevent movement; there is always a low concentration of the volatile anesthetic required to achieve MAC (synergy by definition).4 In the same way, high concentrations of a sedative/hypnotic can be used to provide anesthesia as long as ventilation of the subject is assured. However, the recovery from such an anesthetic can be quite prolonged. High concentrations of any drug may also increase the incidence of unwanted side effects. In an effort to minimize side effects and hasten anesthetic recovery, a balanced anesthetic is preferred. A balanced anesthetic takes advantage of the primary effect of each drug by matching each component of the anesthetic to the effect of a single drug. To this end, the sedative/hypnotic assures lack of recall, muscle relaxants prevent movement, and the opioids blunt the adrenergic response and reflex movement to noxious stimuli while providing pain relief on emergence. Moderate to deep sedation as a part of monitored anesthesia care (the other MAC) has different goals and almost never includes a volatile anesthetic. The patient is expected to maintain adequate spontaneous ventilation and oxygenation; assisted ventilation or the insertion of an airway device becomes a general anesthetic by definition. It is particularly important for the patient to maintain spontaneous ventilation or respond to prompts to breathe when the patient is undergoing upper gastrointestinal endoscopy and the anesthesia caregiver must “share” the airway. Typically during sedation, the nociceptive input is at least partially blocked by local infiltration or topical application of a local anesthetic. When the noxious stimuli are not completely blocked, an analgesic or higher doses of sedative/hypnotics are required. By making use of the interaction of sedatives and analgesics, the dose of each can be lowered; hence, a more rapid recovery can be realized. If the 2 drugs being used have a synergistic effect, large dose reductions of each can be accomplished. However, synergy of adverse effects, including respiratory depression, can also occur.

In this issue of Anesthesia & Analgesia, LaPierre et al. have looked at the combined effect of a sedative/hypnotic and an opioid in blunting the responses to esophageal instrumentation, which is likened to upper gastrointestinal endoscopy.6 The aim of their study was to “explore selected effects of combinations of remifentanil–propofol effect-site concentrations that lead to a loss of response to esophageal instrumentation (blunt-end bougie insertion; positive outcome), along with a loss of responsiveness, and/or the onset of intolerable ventilatory depression (negative outcomes).” A secondary aim was to create response surface models (3-dimensional [D] isobolographic interactions) for each of the 3 observed effects. The authors hypothesized that they could determine selected effect-site concentrations of the combination of remifentanil and propofol that would permit esophageal instrumentation in a high percentage of these volunteers while avoiding loss of responsiveness or intolerable ventilatory depression. They also hypothesized that the interaction of propofol and remifentanil would be synergistic.

When administering an anesthetic, we strive to attain our goal in more than half or, preferably, >99% of our patients. Indeed, in the article by LaPierre et al. the authors include isobolograms for the 5% and 95% responses in addition to that for 50% (their Fig. 2, A and C, and Fig. 3).6 But what if you are interested in other response percentages? The Utah anesthesia research group frequently and effectively uses response surfaces to illustrate their interactive data. A response surface is a 3-D rendition of the more common (2-D) isobologram, which illustrates the response of a group of subjects to various concentration combinations of 2 drugs.7 A response surface is a 3-axis plot of the isobolographic response with the third (vertical) axis representing the percentage of responders (or nonresponders) as a continuous function (LaPierre et al.'s Fig. 2, B and D). The surface is particularly useful in illustrating the transition zone separating responders from nonresponders. Individual isobolograms can be illustrated on the surface as a subset of the continuous relationship.

Figure 2

Figure 2

Figure 3

Figure 3

Comparing the results of LaPierre et al.6 to those of Lallo et al.8 reveals a very interesting similarity. In the Lallo et al. study the authors compared the predicted effect-site concentrations of propofol to those of remifentanil when each drug was administered alone, as the only “sedative” for nasotracheal intubation. They found that the propofol group lacked recall and would breathe spontaneously, but were more likely to display an uncooperative response to noxious stimulation. The remifentanil group could typically recall events, were cooperative, and would appropriately respond to reminders to breathe. Similar results can be gleaned from Figure 1A in the study by LaPierre et al.6 If the preferential endpoint is loss of response to esophageal instrumentation and no intolerable ventilatory depression, there are 2 groups of concentration pairs that could be acceptable. One group, along the propofol axis, is the combination of effect-site concentrations of propofol of 1.5 to 2.7 μg · mL−1 and remifentanil up to 0.8 ng · mL−1. The other group, along the remifentanil axis, has effect-site concentrations of remifentanil of 3 to 4 ng · mL−1 and propofol up to 0.6 μg · mL−1. The authors state that there appears to be a region of low remifentanil (0 to 1.5 ng · mL−1) and high propofol (4 to 6 μg · mL−1) concentrations in which there is a high probability (>80% to 95%) of loss of response to esophageal instrumentation and a moderate probability (40% to 70%) of intolerable ventilatory depression. There is no “sweet spot” (LaPierre et al.'s Fig. 3) for which there is greater than a 95% probability of loss of response to esophageal instrumentation and less than a 5% probability of intolerable ventilatory depression.6 However, the authors did find that 2 to 3 μg · mL−1 of propofol and 0.8 ng · mL−1 of remifentanil blocked the response to esophageal instrumentation and avoided intolerable ventilatory depression in a majority of volunteers.

Clinical application of the predicted effective concentrations derived in these studies is particularly problematic for anesthesia practitioners in the United States who do not, as yet, have access to target-controlled infusions other than in a research setting. We are more accustomed to thinking in terms of infusion rates than in terms of plasma or effect-site concentrations. Therefore, to provide a more familiar reference, we have used the Schnider (propofol)9,10 and Minto (remifentanil)11 models to simulate a bolus-infusion regimen that will attain effect-site concentrations near the optimum concentration pairs as determined by LaPierre et al.6 For propofol, 0.8 mg · kg−1 (56 mg/70 kg) infused over 1 minute followed by a 83 μg · kg−1 · min−1 infusion will attain and maintain an effect-site concentration within 10% of 2 μg · mL−1 for the first 60 minutes of the infusion (Fig. 2). For remifentanil, 0.15 μg · kg−1 infused over 1 minute (10 μg/70 kg) followed by a 0.03 μg · kg−1 · min−1 infusion will attain and maintain an effect-site concentration of about 0.9 ng · mL−1 (Fig. 3). A larger-bolus dose would obviously obtain this concentration sooner with the risk of higher initial concentrations. These simulations were performed using SAAM II software and the aforementioned model parameters (SAAM Institute, Seattle, WA).

We can conclude that the adverse synergy of adding remifentanil to propofol, resulting in the negative endpoints of loss of responsiveness, intolerable respiratory depression, or both, outweighed the positive synergy gained by adding remifentanil. It would seem that the use of either propofol alone (the patient will breath spontaneously despite loss of response to verbal stimuli) or remifentanil alone (the patient is able to respond to prompts to breathe), as demonstrated by Lallo et al.,8 may be the preferred method of sedation.

The message must be heard loud and clear by all proceduralists. If they are performing a procedure on a mild to moderately sedated patient, or even with deep sedation (with the patient maintaining a patent airway and spontaneous ventilation), a significant percentage of those patients will respond, in an uncooperative manner, to a noxious stimulus. If opiate is added to alleviate this unwelcome response, the patient will possibly stop breathing and require ventilatory support, precluding continuation of an upper endoscopy or airway intervention.

Back to Top | Article Outline


Name: Tom C. Krejcie, MD.

Contribution: This author helped write the manuscript.

Attestation: Tom C. Krejcie approved the final manuscript.

Name: Michael J. Avram, PhD.

Contribution: This author helped write the manuscript.

Attestation: Michael J. Avram approved the final manuscript.

Back to Top | Article Outline


1. Stanski DR, Shafer SL. Quantifying anesthetic drug interaction: implications for drug dosing. Anesthesiology 1995;83:1–5
2. Shafer SL, Stanski DR. Improving the clinical utility of anesthetic drug pharmacokinetics. Anesthesiology 1992;76:327–30
3. Rosow CE. Anesthetic drug interaction: an overview. J Clin Anesth 1997;9:27S–32S
4. Hendrickx JFA, Eger EI II, Sonner JM, Shafer SL. Is synergy the rule? A review of anesthetic interactions producing hypnosis and immobility. Anesth Analg 2008;107:494–506
5. Tallarida RJ. An overview of drug combination analysis with isobolograms. J Pharmacol Exp Ther 2006;319:1–7
6. LaPierre CD, Johnson KB, Randall BR, White JL, Egan TD. An exploration of remifentanil–propofol combinations that lead to a loss of response to esophageal instrumentation, a loss of responsiveness, and/or onset of intolerable ventilatory depression. Anesth Analg 2011;113:490–9
7. Short TG, Ho TY, Minot CF, Schnider TW, Shafer SL. Efficient trial design for eliciting a pharmacokinetic–pharmacodynamic model based response surface describing the interaction between two intravenous anesthetic drugs. Anesthesiology 2002; 96:400–8
8. Lallo A, Billard V, Bourgain JL. A comparison of propofol and remifentanil target-controlled infusions to facilitate fiberoptic nasotracheal intubation. Anesth Analg 2009;108:852–7
9. Schnider TW, Minto CF, Gambus PL, Andresen C, Goodale DB, Shafer SL. The influence of method of administration and covariates on the pharmacokinetics of propofol in adult volunteers. Anesthesiology 1998;88:1170–82
10. Schnider TW, Minto CF, Shafer SL, Gambus PL, Andresen C, Goodale DB, Youngs EJ. The influence of age on propofol pharmacodynamics. Anesthesiology 1999;90:1502–16
11. Minto CF, Schnider TW, Egan TD, Youngs E, Lemmens HJM, Gambus PL, Billard V, Hoke JF, Moore KHP, Hermann DJ, Muir KT, Mandema JW, Shafer SL. Influence of age and gender on the pharmacokinetics and pharmacodynamics of remifentanil. Anesthesiology 1997;86:10–23
© 2011 International Anesthesia Research Society