A Dose-finding Study of Sugammadex for Reversal of Rocuronium in Cardiac Surgery Patients and Postoperative Monitoring for Recurrent Paralysis

This dose-finding study was approved by the Institutional Review Board of the University of Washington (Seattle, Washington). Patients were approached for the study between February 2022 and July 2022 and gave written informed consent. The study was registered at ClinicalTrials.gov (NCT05246397; Dr. Bowdle, February 18, 2022) before patient enrollment; the study was carried out as described except for minor modifications. Patients aged 18 yr or older undergoing cardiac surgery fit the inclusion criteria. Exclusion criteria were allergies or other contraindications to the use of sugammadex. The study was carried out at a single center.

Reversal or recovery from paralysis was defined as a train-of-four ratio greater than or equal to 0.9. Neuromuscular blockade was monitored using an electromyography-based twitch monitor (TwitchView, Blink DC, USA) with stimulating current set to 80 mA at a 300-ms pulse width (The default settings for the TwitchView are 65 mA at 300-ms pulse width. When testing quantitative neuromuscular blockade monitors, we utilize the maximum current output of the monitor, which is typically 60 to 80 mA.^19–21  This approach has also been taken by others.²²  We take this approach because our clinical and laboratory experience is that lower currents, including the supramaximal current, sometimes fail to produce a large enough twitch amplitude for accurate measurement of the train-of-four count or the train-of-four ratio under real world clinical conditions). The TwitchView electrode array was applied to stimulate the ulnar nerve and measure the muscle action potential at either the adductor pollicis or the first dorsal interosseous muscle. The administration of neuromuscular blocking drugs was at the discretion of the clinical anesthesia team; however, rocuronium was the nondepolarizing neuromuscular blocking drug used in all cases. Anesthesia consisted of isoflurane and opioids (including fentanyl, sufentanil, hydromorphone, and methadone); isoflurane was administered during cardiopulmonary bypass. At the end of surgery, a propofol infusion was administered for postoperative sedation. After induction of anesthesia but before administration of any neuromuscular blocking drugs, and before tucking the arm at the patient’s side, a baseline train-of-four measurement was made to verify that the monitor and electrode array were functioning properly. Near the end of the surgery, when the sternum had been reapproximated, the response to a train-of-four stimulus was determined, and sugammadex titration was performed under the supervision of the investigators unless the train-of-four ratio had spontaneously recovered to at least 0.9. A 200-mg, 2-ml vial of sugammadex was diluted in a syringe to a total of 20 ml with saline to facilitate the administration of sugammadex in 50-mg (5-ml) boluses. Five minutes after each sugammadex dose of 50 mg, the response to the train-of-four stimulus was determined. Sugammadex dosing was continued until a train-of-four ratio of at least 0.9 was achieved, before patient transport to the intensive care unit (ICU).

Electromyography-based twitch monitoring by the investigators was continued in the ICU every 15 min for the first hour after arrival and then hourly for up to 7 h or until propofol sedation was discontinued in preparation for extubation. If the train-of-four ratio was less than 0.9 during the monitoring period in the ICU, additional sugammadex doses of 50 mg were administered until the patient regained a train-of-four ratio 0.9 or greater.

The primary outcome was the response to train-of-four stimulation just before sugammadex administration and the total dose of sugammadex required to achieve a train-of-four ratio of at least 0.9. The secondary outcome was the incidence of recurrent paralysis during the postoperative monitoring period, defined as a train-of-four ratio less than 0.9 after first having achieved successful reversal to a train-of-four ratio 0.9 or greater.

Sample size was based on previous sugammadex dose-finding studies. The sample size in this study was similar to the largest previous sugammadex dose-finding studies. Descriptive statistics are presented as number (%) for categorical variables or mean ± SD for continuous variables. Outcome variables (sugammadex dose) were compared among twitch response groups using two-sample Student’s t test with the assumption of unequal variance (Satterthwaite’s degrees of freedom), ANOVA, two-sample test of proportions, Wilcoxon rank-sum test and Pearson’s chi-square test as appropriate. All P values were two-sided, and statistical significance was defined as a P value <0.05. All statistical analyses were performed with STATA 11.0 (Stata Corp LP, USA). Patient and procedure characteristics were analyzed with R 4.2.1s (R Foundation for Statistical Computing, Austria).

One hundred forty-five patients eligible for the study were approached preoperatively, and 126 were enrolled. One hundred patients (of the 126 enrolled) required reversal with sugammadex. The remaining 26 patients recovered spontaneously to a train-of-four ratio of at least 0.9 at the time of sternal reapproximation. Three patients of 100 who received sugammadex were excluded from the study: 1 patient received an incorrect dose of sugammadex during titration (200 mg instead of 50 mg), 1 patient had probable anaphylaxis to sugammadex and the titration was discontinued, and 1 patient received an additional dose of sugammadex after reversal to train-of-four ratio 0.9 or greater. Patient enrollment is further depicted in figure 1. The characteristics of the 97 patients who received sugammadex and were evaluable are shown in table 1. Of these 97 patients, there were 4 whose postreversal monitoring period was shorter than intended (2 patients required additional rocuronium after successful reversal, and 2 patients were extubated in the operating room), but these patients were not excluded from the study.

The total dose of sugammadex required to achieve a train-of-four ratio of at least 0.9 is shown in figure 2A for each patient, along with the response to the train-of-four stimulus just before sugammadex administration. The sugammadex dose on a per kilogram actual body weight basis is shown in figure 2B. A box-and-whisker plot is shown in figure 3A for the two subgroups of patients that would require either 2 mg/kg or 4 mg/kg sugammadex based on the manufacturer’s dosing recommendations; four patients with a posttetanic count of 0 were included in the 4 mg/kg subgroup although the manufacturer does not specify a dose for patients with a posttetanic count of 0. There was a significant difference in mean sugammadex dose requirement between the 2 mg/kg and 4 mg/kg recommended dose groups (P < 0.001), and this difference was maintained whether or not the four patients with a posttetanic count of 0 were included (table 2). A box-and-whisker plot is shown in figure 3B for subgroups of patients with a posttetanic count (including a posttetanic count of 0), a train-of-four count of 1, 2, or 3, a train-of-four ratio less than 0.4, or a train-of-four ratio 0.4 or greater before sugammadex administration. There was a significant difference in the mean sugammadex dose requirement between the subgroups with different train-of-four stimulus responses before sugammadex (P < 0.001; table 2).

The mean dose requirements were less than the manufacturer’s recommended dose for each subgroup; however, there was substantial variation in dose requirement between patients. While many patients required less than the recommended dose, some patients required more than the recommended dose. Of the 68 patients starting the sugammadex titration with a train-of-four count of at least 2, whose manufacturer’s recommended dose would be 2 mg/kg, 11 required more than 2 mg/kg to achieve a train-of-four ratio of at least 0.9. Of the 29 patients starting the sugammadex titration with a train-of-four count of 1 or less, whose manufacturer’s recommended dose would be 4 mg/kg, 2 required more than 4 mg/kg to achieve reversal. Although four patients had a posttetanic count of 0 and therefore fell outside of the manufacturer’s dose recommendations, none required more than 4 mg/kg for reversal; they required 0.9, 1.7, 2.6, or 3.8 mg/kg.

Ninety-five out of 97 patients (98%) did not have recurrent paralysis during the postoperative monitoring period, which averaged 2.5 h and ranged between 3 min and just greater than 7 h. One patient weighing 74 kg had a train-of-four ratio of 0.47, received 100 mg sugammadex, and achieved a train-of-four ratio greater than 0.9, but 5 min later, while still in the operating room, had a train-of-four ratio of 0.87. Two additional 50-mg doses of sugammadex were administered before achieving a train-of-four ratio greater than 0.9; the train-of-four ratio remained above 0.9 during 175 min of postoperative monitoring (fig. 4A). One patient weighing 78 kg had a posttetanic count of 0, received 250 mg sugammadex to produce a train-of-four ratio greater than 0.9, but after 45 min in the ICU had a train-of-four ratio of 0.81. An additional 50-mg dose of sugammadex produced a train-of-four ratio greater than 0.9, which was maintained during the remaining 181 min of postoperative monitoring (fig. 4B). Both patients had normal renal function.

The dose of sugammadex required to achieve a train-of-four ratio of at least 0.9 varied from 0.43 mg/kg to 5.55 mg/kg. While there was a statistically significant relationship between the twitch response before reversal and the sugammadex dose required to achieve a train-of-four ratio of at least 0.9, as shown in figure 3, the range of dose requirements between patients with the same twitch response varied greatly. Eighty-four out of 97 patients (87%) required less than the manufacturer’s recommended dose, and 13 (13%) required more than the manufacturer’s recommended dose. A subgroup analysis of these 13 subjects compared to the other 84 subjects did not reveal clinically significant differences in patient characteristics between the groups. Three patients required more than the manufacturer’s recommended dose of 4 mg/kg. Two patients required more than 4 mg/kg (5.55 and 4.26 mg/kg), despite having a train-of-four count of 1 before reversal. The third patient required more than 4 mg/kg (4.43 mg/kg) despite having a train-of-four count of 3 before reversal.

Since the dose required for an individual patient could not be predicted with certainty based on the train-of-four twitch response immediately before reversal, we conclude that quantitative twitch monitoring is essential to evaluate the effectiveness of reversal with sugammadex. Even if sugammadex had been administered at 4 mg/kg to all patients, some patients would not have attained a train-of-four ratio of at least 0.9.

Previous studies of reversal of rocuronium with sugammadex have utilized experimental designs with substantial differences from this study, making direct comparisons problematic (see table in Supplemental Digital Content 1, https://links.lww.com/ALN/D134). In previous studies, sugammadex was typically given at several fixed doses, such as 0.5, 1, 2, or 4 mg/kg, and the time required to reach a train-of-four ratio of 0.9 was measured. In some of these studies, sugammadex was administered when a train-of-four count of 2 was reached.^{1,6–9,11,12}  In some other studies, sugammadex was administered when neuromuscular blockade was profound, such as a posttetanic count of 1 or 2.^2–5,10  In other studies, small doses of sugammadex, 1 mg/kg or less, were administered when the train-of-four ratio reached 0.2 or 0.5.^13,14  Most previous studies utililzed acceleromyography, without normalization. While a dose of sugammadex of 2 mg/kg for a train-of-four count of 2 or 4 mg/kg for a posttetanic count of 1 or 2 usually resulted in a train-of-four ratio of 0.9 in less than 5 min in these studies, there were outliers. For example, Jones et al. reported that when the posttetanic count was 1 or 2, the time required to reach a train-of-four ratio of 0.9 after administering sugammadex 4 mg/kg ranged from 0.9 to 16.1 min; in addition, one patient did not reach a train-of-four ratio of 0.9 “during the observation period.”⁵  When the posttetanic count was 1 or 2 and a dose of sugammadex less than 4 mg/kg was administered, the time required to reach a train-of-four ratio was highly variable. For example, Duvaldestin et al. found that in patients receiving rocuronium with a posttetanic count of 1 or 2, a dose of sugammadex at 1 mg/kg achieved a train-of-four ratio of at least 0.9 in a median time of fewer than 10 min but required as much as nearly 120 min.³  Groudine et al. found that in patients receiving rocuronium with a posttetanic count of 1 or 2, a dose of sugammadex at 1 mg/kg achieved a train-of-four ratio of at least 0.9 in a mean time of 19 min. However, the range was 5 to 33 min.⁴  Sugammadex, 2 mg/kg, achieved a train-of-four ratio of at least 0.9 in a median time of 5 min, but the range was 2 to 15 min.

Most previous studies of sugammadex dose-response for reversal of neuromuscular blockade have utilized acceleromyography and have reported raw train-of-four ratio data without normalization. Using acceleromyography without normalization will tend to produce a nonnormalized train-of-four ratio of 0.9 with either smaller doses of sugammadex or shorter time intervals, as many patients have a baseline train-of-four ratio of greater than 1 with acceleromyography.^25,26  A train-of-four ratio of 0.9 with acceleromyography may not represent adequate reversal in patients with a baseline train-of-four ratio greater than 1. As this study utilized electromyographic twitch monitoring, which is not complicated by baseline train-of-four “overshoot,” the doses of sugammadex and the time required to achieve a train-of-four ratio of at least 0.9 may be greater than found with acceleromyography.

There were two patients with mild recurrent paralysis, one with a posttetanic count of 0 before reversal, which is outside of the manufacturer’s dosing recommendations. The other patient with recurrent paralysis had a train-of-four ratio of 0.47 before reversal and received 100 mg sugammadex. The train-of-four ratios at the time of recurrent paralysis were 0.81 and 0.87, respectively. While these train-of-four ratios do not represent severe recurrent paralysis, they were nevertheless less than 0.9. There have been numerous case reports of imputed recurrent paralysis after sugammadex, as well as some data from clinical trials (see table in Supplemental Digital Content 2, https://links.lww.com/ALN/D135).^{2,3,7,15–17}  The circumstances from these reports are highly variable and, in some cases, are obviously related to either failure to achieve reversal to a train-of-four ratio of at least 0.9 or to the lack of quantitative twitch monitoring. Determination of the response to a train-of-four stimulus after surgery can be complicated by patient discomfort when using the same stimulating current commonly used intraoperatively (approximately 45 to 80 mA).^27,28  As the stimulating current is reduced to avoid patient discomfort, ulnar nerve depolarization may be incomplete, resulting in a diminished twitch response compared to that which would occur at a higher current.²⁷  In this study, we performed quantitative twitch monitoring in sedated, intubated patients at 80 mA, the same current utilized for intraoperative monitoring, to ensure an adequate twitch response. Taken as a whole, the literature concerning recurrent paralysis after reversal with sugammadex suggests that achieving a train-of-four ratio of at least 0.9 does not, by itself, guarantee that recurrent paralysis will not occur.

Interestingly, a well-conducted study of reversal of vecuronium by sugammadex at the reappearance of the fourth twitch of the train-of-four found recurrent paralysis in 9 of 27 (33%) patients. These instances of recurrent paralysis occurred during 60 min of quantitative twitch monitoring in the recovery room using the same stimulating current used in the operating room (Laszlo Asztalos, M.D., Ph.D., Department of Anesthesiology and Intensive Care, University of Debrecen, Debrecen Hungary; written communication, July 16, 2022). Recurrent paralysis was observed in patients receiving sugammadex doses at 0.5 (3 of 9 patients), 1.0 (4 of 8 patients), or 2.0 (2 of 10 patients) mg/kg.¹⁵  Vecuronium is not as avidly bound by sugammadex as rocuronium.²⁹  Further research may be needed to characterize the conditions in which recurrent paralysis occurs after sugammadex administration for reversal of vecuronium, which may differ from reversal of rocuronium.

This study has several limitations. We deliberately studied cardiac surgery patients because of the opportunity to measure train-of-four twitch response for several hours after surgery in sedated, intubated patients. This enabled the use of an 80-mA stimulation current throughout the study, which is not feasible in awake patients due to discomfort. In addition, the protocol for fast-track extubation of cardiac surgery patients utilized by our medical center recommends the reversal of neuromuscular blockade at the conclusion of surgery. The patients underwent cardiopulmonary bypass with a variable extent of deliberate hypothermia, although the patients were rewarmed at the time of sugammadex administration (table 1). Whether this influences the reversal of rocuronium with sugammadex and whether the results would be the same in noncardiac surgery patients is uncertain. However, we are unaware of any specific reasons the results would not apply to noncardiac surgery patients.

Because we titrated sugammadex in 50 mg increments every 5 min, administration of larger doses of sugammadex required more time than would be the case if the entire dose were administered as a single bolus. For example, administration of sugammadex 400 mg would have required more than 40 min. Throughout the sugammadex administration, rocuronium clearance is ongoing, which would likely reduce the total dose of sugammadex required. Therefore, for the larger doses of sugammadex given in this study, the amount required for reversal could be an underestimate of the dose that would have been required had the entire dose been administered as a single bolus.

Additionally, all of the patients in this study received rocuronium. Because rocuronium is bound more avidly to sugammadex than vecuronium, the results and conclusions of this study may not apply to vecuronium. The time interval between sugammadex doses, 5 min, may have influenced the result. An argument could be made that in some cases, the maximum effect of sugammadex may not be seen within 5 min. However, clinical trial data suggest that sugammadex reversal to a train-of-four ratio of at least 0.9 frequently occurs within 5 min. Also, from a practical perspective, it seemed unlikely that anesthesia providers in a real-world setting would be willing to wait longer than 5 min before giving additional sugammadex if reversal were not achieved. Finally, the choice of an incremental sugammadex dose of 50 mg may have influenced the result. This dose was chosen from a practical perspective, given that a vial of sugammadex contains 200 mg in 2 ml; dilution of this vial to a total of 20 ml with saline makes the administration of 50-mg increments reasonably accurate.

In conclusion, sugammadex was titrated in 50-mg increments every 5 min until a train-of-four ratio of at least 0.9 was reached at the conclusion of cardiac surgery, and quantitative twitch monitoring was continued postoperatively in the ICU while the patients remained sedated and intubated. The sugammadex dose requirement for reversal was generally less than the manufacturer’s recommended doses. However, there was a wide range of dose requirements for individual patients, and some patients required more than the recommended dose. Therefore, quantitative twitch monitoring is required to determine whether a train-of-four ratio of at least 0.9 has been reached, even at a sugammadex dose of 4 mg/kg. There were two cases of mild recurrent paralysis, suggesting that recurrent paralysis after sugammadex requires additional study.

Dr. Michaelsen’s research is supported by the Foundation of Anesthesia Education and Research (Schaumburg, Illinois) and the Washington Research Foundation (Seattle, Washington). Dr. Bowdle’s research is supported in part by the Laura Cheney Professorship in Anesthesia Patient Safety (Seattle, Washington). Support was otherwise provided from departmental sources.

The authors declare no competing interests.

Supplemental Digital Content 1—Sugammadex dose-finding clinical trials, https://links.lww.com/ALN/D134

Supplemental Digital Content 2—Case reports and clinical trial recurrent paralysis, https://links.lww.com/ALN/D135