Cryoneurolysis: Major Problems Rarely Discussed (or “Don’t Be Fooled—The Ice Ball Lies!”)

“[T]he three temperature ‘zones’—too warm, too cold, and just right—are both unpredictable and indistinguishable using current imaging techniques.”Image: J. P. Rathmell.Cryoneurolysis has been cleared by the U.S. Food and Drug Administration (Silver Spring, Maryland) to manage acute and chronic pain for a half century. This technique involves the treatment of a nerve with exceptionally cold temperatures, with the goal of inducing Wallerian degeneration and resulting in a predictable yet reversable inhibition of nerve conduction that lasts weeks or months. Modern cryoprobes function by passing a gas at high pressure through a small annulus, which results in a drop in pressure and therefore temperature due to the Joule-Thomson effect (fig. 1).1 An ice ball forms around the end of the probe, which induces neuronal injury within the affected area.2 Wallerian degeneration resulting in reliable, prolonged effects requires the tissue temperature to reach colder than –20° to –60°C (the precise temperature remains undetermined).3 Temperatures warmer than this threshold result in a neuropraxia with unpredictable effects lasting a few minutes to a few weeks.3 However, temperatures colder than –100° to –140°C can result in permanent nerve injury (again, the precise temperature remains undetermined, but most practitioners use –100°C to retain a margin of safety).3 Therefore, the temperature of the target tissue is of critical importance: treatment success is dependent upon ensuring the entire diameter of the target nerve is encompassed in the “therapeutic zone” between approximately –60° to –100°C (fig. 2). And here lies the colossal—yet rarely discussed—problem with cryoneurolysis: the three temperature “zones”—too warm, too cold, and just right—are both unpredictable and indistinguishable using current imaging techniques.Fig. 1.: A modern cryoneurolysis probe produces extremely cold temperatures at its tip due to the Joule-Thomson effect resulting from gas flowing from a high- to low-pressure chamber (used with permission from Brian M. Ilfeld, M.D., M.S.).Fig. 2.: The three possible thermal “zones” within a cryoneurolysis ice ball and the potential results for a target nerve within each range. The precise warmest temperature at which the desired Wallerian degeneration occurs remains undefined, but lies between –20° and –60°C. Similarly, the coldest temperature before permanent nerve injury occurs remains undefined, but lies between –100° and –140°C. Therefore, the optimal temperature range target is often defined as –60° to –100°C, which reliably induces Wallerian degeneration and prolonged analgesic effects while avoiding permanent ablation. If a nerve is partially or completely in the “too cold” zone of the ice ball, permanent nerve injury is possible (A). If a nerve is partially or completely in the “too warm” zone of the ice ball—which is not identifiable by current bedside imaging techniques—a neuropraxia may occur, the results of which are unpredictable, lasting a few minutes to a few weeks (C). Ideally, the entire cross-section of the target nerve should lie within the therapeutic zone (B).Consequently, many practitioners desiring reversible effects (e.g., postoperative analgesia) prefer using carbon dioxide or nitrous oxide since these two gasses become solid below –78°C and –88°C, respectively, thus avoiding any chance of colder temperatures and resulting nerve injury. In contrast, irreversible nerve ablation may be desired when treating chronic pain conditions, in which case a different gas such as argon (solid below –189°C) is required. Unfortunately, many factors other than the type of gas dramatically influence tissue temperature, including cryoprobe gauge, shape, and design; tissue baseline temperature, vascularity, cardiac output, and the concurrent use of a sympathectomy-inducing local anesthetic peripheral nerve block; local injection of a vasoconstrictor such as epinephrine; as well as the duration of gas application, number of applications (“cycles”), time between cycles, and previously treated nerves in the general anatomic area.4 The result is that it is virtually impossible to accurately predict tissue temperature at various distances from the cryoprobe at any specific timepoint5 and therefore difficult to assess whether the entire diameter of the target nerve was encompassed in the cryoneurolysis “therapeutic zone” (fig. 2). A trial by Nygaard et al. in the current issue of Anesthesiology illustrates this problem.6 In the study, 87 participants with chronic knee pain due to osteoarthritis were randomized to either a sham procedure or ultrasound-guided percutaneous cryoneurolysis of the anterior femoral cutaneous nerve and infrapatellar branch of the saphenous nerve.6 No statistically significant differences between treatments were identified for the primary (pain after 2 weeks) and majority of secondary outcomes at 0.5, 6, and 12 months, leading the authors to conclude that, “when viewed across all assessed timepoints, the results indicate that cryoneurolysis has no meaningful, robust benefit over sham for chronic knee pain.”6 In contrast, a very similar randomized, sham-controlled investigation identified multiple benefits within the first 1 to 3 months after treatment,7 raising the possibility that the freezing protocol used by Nygaard et al. was inadequate. This is not a criticism of Nygaard et al. because they themselves noted, and we concur, that “there is no consensus on the optimal protocol, i.e., freeze duration, no. of cycles, thaw duration.”6 The authors also explain that “to ensure patient safety, the present study used a conservative protocol…”6 In short, it remains unknown if “cryoneurolysis has no meaningful” benefit in treating chronic knee pain, or if the study protocol of Nygaard et al. failed to induce cryoneurolysis due to inadequate therapeutic temperatures. As noted previously, use of carbon dioxide and nitrous oxide guarantee avoiding temperatures colder than –100°C. Since they cannot induce a level of cold resulting in permanent nerve injury, it would appear that they are the perfect cryoneurolysis gasses. Unfortunately, ice is an exceptional insulator. The result is a steep temperature gradient between the probe shaft and even just a few millimeters away: even if the cryoprobe itself reaches the therapeutic cold treatment zone, the temperature beyond the first few millimeters of the ice surrounding the probe will frequently be inadequate to induce Wallerian degeneration. As demonstrated in figure 3A, while the shaft of the probe reaches below –87°C (green line), tissue just 5 mm from the probe does not reach therapeutic cold (blue line), even after 10 min of treatment, and this laboratory model does not mimic continuous warm blood flow as occurs in vivo. Due to the insulation effect of successive layers of ice, nerves greater than 1 to 2 mm in diameter may require a gas such as argon capable of far colder temperatures to induce a larger therapeutic zone. As seen in figure 3B, the very cold probe temperature of –137°C resulted in tissue 5 mm from the shaft reaching a therapeutic temperature—and thus adequate treatment of larger nerves. Unfortunately, any nerve fibers immediately adjacent to the probe could be exposed to low enough temperatures to risk permanent injury (fig. 3B also illustrates the extraordinary insulation property of ice with the 76-degree difference in the 2 mm beyond the cryoprobe shaft). Practitioners are therefore between a rock and a cold place when treating all but the smallest nerves.Fig. 3.: Minimum probe temperatures versus time for an argon-based cryoneurolysis system at 2,000 and 2,450 PSI in a laboratory model. At the lower argon pressure (A), note that while the probe shaft reaches a therapeutic –87°C (green line), tissue just 5 mm from the probe does not reach therapeutic cold (blue line) due to the insulation property of ice, even after 10 min of treatment. This laboratory model does not mimic continuous warm blood flow as occurs in vivo, so ice ball size and temperature in patients would be even smaller and lower, respectively. Nerves greater than 1 to 2 mm in diameter may require a far colder probe temperature to induce a larger therapeutic zone. A higher pressure of argon (B) results in tissue 5 mm from the probe reaching a therapeutic temperature—and thus being able to adequately treat larger nerves—but any nerve fibers immediately adjacent to the probe could be exposed to cold enough temperatures to risk permanent injury. This figure also illustrates the extraordinary insulation property of ice with the 76-degree difference just 2 mm beyond the cryoprobe shaft (–137°C vs. –61°C). Varian Medical Systems (Palo Alto, California) has been a research partner in compiling the data of this figure, which illustrates the steep temperature gradient that occurs with all cryoablation systems.One potential partial solution is to integrate a thermocouple into the cryoprobe’s wall, providing the minimum temperature to which an adjacent nerve could be exposed. Unfortunately, nearly all cryomachines with percutaneous probes designed for reversible cryoneurolysis of peripheral nerves—as opposed to permanent tissue ablation—provide practitioners with the temperature of returning gas, which is always lower than the actual probe temperature (unpublished data, Ilfeld et al.). While an integrated thermocouple would provide a measure of safety by allowing practitioners to avoid potentially hazardous low temperatures, it still would not deliver the information needed to assure that the entire nerve diameter reaches a therapeutic level. Larger nerves—the definition of which remains undefined—may require a separate thermocouple inserted on the opposite side of the nerve from the cryoprobe. Perhaps the largest nerves might require applying cryoneurolysis to multiple points around the nerve circumference to reach all integrated fibers.8,9 Partial cryoneurolysis of a nerve not only fails to provide analgesia but also can induce a painful neuropraxia and therefore must be avoided.10 Due to the numerous factors influencing tissue temperature making tissue temperature unpredictable, it is improbable that an “optimal” cryoneurolysis protocol will ever be identified for all equipment and clinical situations.4,5 Like so many things in medicine, treatment will most likely need to be tailored for each individual; in the case of cryoneurolysis, this will probably require one or more thermocouples either integrated into the probe or inserted separately. Until improved equipment becomes available, clinical practice can be based on previously published successful cryoneurolysis protocols, such as the use of nitrous oxide after knee arthroplasty, mastectomy, and traumatic rib fractures.7–9 Protocols are being investigated for larger nerves as well (NCT06071715). Additional research is required to determine the minimal temperature before risking permanent neural injury (somewhere between –100° and –140°C) as well as the maximum temperature that will still reliably result in prolonged analgesia (somewhere between –20° and –60°C). Determining the minimum freeze time is also vital as procedure duration is perhaps the greatest limiting factor of cryoneurolysis.11 Even the optimal application timing relative to surgery remains unclear.12 Finally, the nerve length requiring adequate treatment should be studied, given that laboratory data suggest lesions less than 4 mm are inadequate in cats.13 In the meantime, it will be difficult to interpret negative/neutral studies such as that by Nygaard et al., given our current inability to distinguish or predict therapeutic temperature zones within an ice ball. Acknowledgments Elan F. Ilfeld, B.S., M.S. (Northeastern University, Boston, Massachusetts), created the three-dimensional renderings of figures 1 to 3. Competing Interests Dr. Ilfeld’s institution has received funding and/or product from Epimed International (Dallas, Texas), SPR Therapeutics (Cleveland, Ohio), Infutronix (Natick, Massachusetts), Masimo (Irvine, California), Varian Medical Systems (Palo Alto, California), and Avanos Medical (Irvine, California). Dr. Prologo is a cofounder of Focused Cryo, Inc. (Atlanta, Georgia), and has received funding for research and/or consultation in the past from Boston Scientific (Boston, Massachusetts), Varian Medical Systems (Palo Alto, California), and Merit Medical Devices (South Jordan, Utah).