For 2 mellow guys, our 2003 editorial about target-controlled drug infusions (TCI) of anesthetic drugs in Anesthesiology was pretty harsh. The subtitle was barbed: Surfing USA Not!1 We made the snarky observation that “exactly 0 of the estimated 13 million propofol anesthetics administered worldwide with TCI since the introduction of the Diprifusor™ (AstraZeneca, Gothenburg, Sweden) in Europe, Asia, the South Pacific, South America, and Africa have been performed in North America.”1 We placed the blame for the fact that TCI was only unavailable in the United States directly on the Food and Drug Administration (FDA), who spent 8 years reviewing a TCI application from Graseby Medical (Graseby Medical Ltd., Upper Pemberton, Kennington, UK): “Over the course of the 8-yr review, the FDA has demonstrated a poor understanding of the underlying scientific basis of TCI. Specifically, the FDA has not recognized that TCI devices can neither increase nor decrease underlying pharmacokinetic variability. As a result, the FDA has expressed unfounded concerns that the TCI mode of administration may lead to a greater frequency of adverse events.” Yikes! We said that?
We have matured somewhat since 2003. We now have a better understanding of the statutory limitations that define exactly what the FDA can and cannot do. We are more aware of the difficult regulatory path for TCI 20 years ago. We also recognize that no sponsor has attempted to bring TCI to the United States since 2003. The FDA obviously cannot approve devices for which an application has not been submitted.
Twenty years ago, the primary obstacle preventing commercialization of TCI in the United States was indeed a regulatory barrier (influential factions within the FDA were categorically opposed to computers making therapeutic decisions about drug doses). Nowadays, financial considerations may constitute the main obstacle. The business model for the Diprifusor depended largely on the sale of proprietary propofol (i.e., Diprivan™—the commercial linkage of the 2 products is obvious from the names! [Fresenius Kabi, Lake Zurich, IL]), not on the sale of the pump. With propofol now at generic price points, that business model has evaporated. That pump companies have not submitted a TCI application to the FDA since propofol went generic suggests that perhaps they do not see an attractive business proposition. Of course part of the calculation on the attractiveness of the TCI business proposition relates to the estimated cost of meeting the regulatory requirements, so it is complicated.
In any case, we still do not have TCI in the United States. However, commercialization of TCI has continued elsewhere. Three articles in the current issue of Anesthesia & Analgesia, solicited by the Editor-in-Chief (SLS) and coordinated by Michel Struys and his colleagues, document the current status of TCI devices.
The first article reviews the history of TCI,2 starting with the initial prototypes by Schüttler et al.3 in Bonn, Germany. The article offers a simple explanation of the pharmacokinetic principles of TCI, based on moving chocolate among different boxes. The authors explain how the early research systems eventually led to the Diprifusor, the first commercially available TCI system. The Diprifusor was approved in Europe in 1996. The Diprifusor was limited to administering proprietary propofol (Diprivan). A decade later, it gave way to second-generation “Open TCI” devices that could give many drugs. The Open TCI systems were far more flexible, but presented clinicians with potentially confusing selections of pharmacokinetic models and implementations.
The second article reviews the safety of TCI.4 The authors use the medical literature, a survey of TCI manufacturers, and data available on government regulatory sites to account for all the reported adverse events associated with TCI. The only risk that they could identify was a TCI-mediated syringe swap, specifically exchanging remifentanil and propofol syringes in a 2-pump device. This is not unique to TCI systems, as one of us (SLS) made exactly the same mistake with a conventional multichannel pump a month ago. The lack of even a single adverse event specifically related to TCI drug delivery in an estimated 20,000,000 uses suggests by application of the “Rule of Three” that the upper 95% confidence interval of the risk of a serious event is <1 in 7 million.5,a To be clear, these figures regarding TCI-associated complications are estimates. Because adverse events are well-known to be underreported, we must view these assumptions about the rate of adverse events with prudence. Nevertheless, even in the worst case scenario, the adverse event rate appears vanishingly low.
The third article reviews the current status of TCI worldwide.6 The authors review 14 TCI systems developed by academic laboratories and 12 commercially available TCI devices. As of 2015, TCI is available in 90 countries. More than 60,000 TCI devices have been sold. TCI has been used millions of times and continues to be used daily worldwide. TCI is described in over 500 peer-reviewed articles. As the authors note, it is now a mature technology. Indeed, the first commercial device, the Diprifusor, has already reached the end of its product life cycle!
So, how do we get this safe and mature technology approved in the United States? Paul Dryden, a regulatory consultant with extensive experience in anesthesia device approval, explains the regulatory options available to the FDA.7 If the device is fundamentally identical to an existing device, it can be approved through a 510(k) process. This is not expensive and simply requires verification, in a laboratory, that the device matches the “predicate” device. There is no obvious predicate device for TCI, so that does not seem like an option.
However, it should not be ruled out too quickly. There are FDA-approved devices that accurately administer IV anesthetics based on drug choice and patient covariates. Every infusion pump that infuses an anesthetic drug at a rate proportional to body weight infers such a claim. There is an implied pharmacokinetic model: distribution volumes and compartment clearances are proportional to weight. The claim for TCI is not very different, other than that the pharmacokinetic model accounts for drug accumulation in tissue as well. In fact, TCI is the only way to implement fully the detailed information contained in modern pharmacokinetic models. For example, a TCI pump can administer an infusion profile in which the pharmacokinetic model relates central clearance to body weight and a distribution volume to age. Standard infusion pumps are not capable of this level of sophistication.
In addition, GE’s Navigator and Drager’s Pilot are FDA-approved products to guide the administration of IV anesthetics. These devices constitute “passive TCI,” in that they perform calculations of drug concentration based on the same pharmacokinetic models used in TCI. They are also a predicate device for the TCI pumps. Even though there is not a definitive predicate device, infusion pumps that assume weight-based pharmacokinetics, along with the Navigator and Pilot systems that incorporate full pharmacokinetic/pharmacodynamic models, may collectively represent a predicate device, facilitating 510(k) approval of TCI.
The second path is called PMA, for PreMarket Approval. This pathway assumes that the device has high risk (class 3) and thus requires extensive clinical testing in humans. That can take many years and cost tens of millions of dollars. This seems hugely wasteful for a technology that has been used routinely for 2 decades worldwide, with zero evidence of safety issues associated with the pharmacokinetic dosing algorithm. Fortunately, Mr. Dryden has identified a third route, called “de novo.” This is a pathway for devices that have no predicate, but are low to moderate risk, and for which there is a reasonable assurance of safety and efficacy. The articles in this issue of Anesthesia & Analgesia2,4,6 provide a comprehensive review of safety and efficacy. The articles also benefit from the imprimatur of peer review and were not funded by the medical device industry. It is our hope that the comprehensive reviews in this issue of Anesthesia & Analgesia will help facilitate a de novo determination for TCI devices submitted to the FDA for consideration.
Finally, Schnider et al.4 offer unique insight into TCI device product labeling. No standard infusion pump manufacturer claims that a pump produces a certain drug concentration. The pump makers merely claim that the device infuses the drug within standard limits of accuracy. A TCI device can make exactly the same claim. For a particular set of patient covariates (e.g., age, body weight, among others), the device will accurately administer a given infusion profile. The labeling nuance for TCI is that the infusion profile was chosen to permit the patient to have reasonably steady drug concentrations over time. However, the device need not claim to achieve a particular concentration. The actual concentration depends on the patient’s pharmacokinetics, just as it does with all forms of drug delivery.
The claim for a TCI device is prosaic and perhaps dull. For a given patient, a TCI device delivers a given profile of drug within conventional limits of infusion accuracy. Otherwise, it is just administering an approved drug, for an approved indication, at doses entirely consistent with the package insert.
It is simple. It is elegant. It is safe. It is useful. It is fun. Given the comprehensive reviews in this issue of Anesthesia & Analgesia,2,4,6 we hope to be surfing the concentration versus response curve for IV anesthetic drugs soon, as we proposed in our editorial 12 years ago.1
Name: Steven L. Shafer, MD.
Contribution: This author helped write the manuscript.
Attestation: Steven L. Shafer approved the final version of the manuscript.
Conflicts of Interest: Steven L. Shafer is the author of the STANPUMP program. STANPUMP has been made freely available for 20 years, and Dr. Shafer does not collect fees or royalties from use of the program.
Name: Talmage Egan, MD.
Contribution: This author helped write the manuscript.
Attestation: Talmage Egan approved the final version of the manuscript.
Conflicts of Interest: Talmage Egan is the co-developer of the Navigator system introduced by General Electric that uses pharmacokinetic algorithms to assist with titration of IV drugs during anesthesia.
Steven L. Shafer is the Editor-in-Chief of Anesthesia & Analgesia. This manuscript was handled by Dr. James G. Bovill, Guest Editor-in-Chief, and Steven L. Shafer was not involved in any way with the editorial process or decision.
a Available at: https://en.wikipedia.org/wiki/Rule_of_three_%28statistics%29. Accessed September 10, 2015.
1. Egan TD, Shafer SL. Target-controlled infusions for intravenous anesthetics: surfing USA not! Anesthesiology. 2003;99:1039–41
2. Struys MMRF, De Smet T, Glen JB, Vereecke HEM, Absalom AR, Schnider TW. The history of target-controlled infusion. Anesth Analg. 2016;122:56–69
3. Schüttler J, Schwilden H, Stoekel H. Pharmacokinetics as applied to total intravenous anaesthesia. Practical implications. Anaesthesia. 1983;38(Suppl):53–6
4. Schnider TW, Minto CF, Struys MMRF, Absalom AR. The safety of target-controlled infusions. Anesth Analg. 2016;122:79–85
5. Hanley JA, Lippman-Hand A. If nothing goes wrong, is everything all right? Interpreting zero numerators. JAMA. 1983;249:1743–5
6. Absalom AR, Glen JB, Zwart GJC, Schnider TW, Struys MMRF. Target-controlled infusion: a mature technology. Anesth Analg. 2016;122:70–8
7. Dryden PE. Target-controlled infusions: paths to approval. Anesth Analg. 2016;122:86–9