“Their observations […] indicate that clinicians who do not see the necessity for quantitative or objective neuromuscular monitoring when reversing residual block with sugammadex need to rethink their position.”

In the current issue of Anesthesiology, Bowdle et al. report the results of an observational study examining the dose of sugammadex required to achieve adequate reversal of a rocuronium neuromuscular block in a group of elective cardiac surgical patients. At the end of surgery, the authors gave 97 patients small (50 mg), repeated (every 5 min) incremental doses of sugammadex until a train-of-four ratio of 0.90 or greater was achieved and then continued to monitor the train-of-four ratio for up to 7 h postoperatively. This work contains a number of findings that deserve attention from our entire profession. Most importantly, it contradicts the widespread belief that simply giving the recommended 2- or 4-mg/kg dose of sugammadex, based solely on an observed twitch count (or post-tetanic count) ensures adequate reversal. While we have no objections to these recommended (and Food and Drug Administration–approved) doses, the Bowdle et al.  study strongly argues that the recommendations should be combined with careful monitoring. Their observations are in accord with the recent American Society of Anesthesiologists (ASA) guidelines regarding objective monitoring of neuromuscular blockade and indicate that clinicians who do not see the necessity for quantitative or objective neuromuscular monitoring when reversing residual block with sugammadex need to rethink their position.

Until the late 1990s, a train-of-four ratio of 0.70 at the adductor pollicis muscle was considered to represent clinically acceptable recovery from nondepolarizing neuromuscular block. This target changed in the late 1990s, when it was shown that pharyngeal function (and hence the risk for aspiration) was impaired at train-of-four ratios less than 0.90 as measured by mechanomyography. In a study using electromyographic monitoring (EMG), Kopman et al.  evaluated various “bedside tests” of neuromuscular recovery in healthy young volunteers at train-of-four values of 0.70 and above. Grip strength measured with a dynamometer was reduced to 57 ± 11% of control at a train-of-four ratio of 0.70 and had only returned to a value of 82 ± 10% of control at a train-of-four ratio of 0.90. In addition, at a train-of-four ratio of 0.70, subjects had trouble sitting without assistance, displayed variable degrees of difficulty in sipping water from a straw, and apposing their incisor teeth, and all experienced visual difficulties. None felt remotely “street ready.” Therefore, it is clear that even “minor” degrees of residual paralysis have meaningful consequences. Since that time, the accepted target for reversal has been a ratio of 0.90 or greater, as used by Bowdle et al. and as recommended in the ASA Guidelines. Note, however, that this target is dependent on the method of assessment. For example, uncalibrated acceleromyography, meaning that a prerelaxant baseline train-of-four ratio was not established, may overestimate recovery compared to the EMG or mechanomyography; displayed acceleromyography train-of-four ratios of 0.90 may only represent EMG values of 0.75 to 0.80.  For this reason, some experts recommend that an uncalibrated acceleromyography ratio of 1.0 is a more appropriate target. 

The first observation of Bowdle et al.  was that the mean dose of sugammadex required in patients with a starting train-of-four count of 2 or greater was 1.24 ± 0.83 mg/kg (less than the recommended dose of 2 mg/kg) and was only 2.3 ± 1.18 mg/kg when sugammadex was given at a train-of-four count of less than 2 (less than the recommended 4 mg/kg dose for a post-tetanic count of 1 to 2 with no response to train-of-four stimulation). This finding is not, however, surprising for several reasons. First, nearly all studies supporting the recommended sugammadex doses used time to recovery (not adequacy of recovery) as their primary endpoint.  Nearly all showed that somewhat lower doses (e.g., 1 mg/kg vs. 2 mg/kg) achieved full recovery, but it took a minute or so longer. Since Bowdle et al.  were not interested in time, their findings are consistent with published work. In addition, their slow titration protocol meant that there was probably a degree of spontaneous recovery occurring during the observational period, something that again might reduce the required sugammadex dose.

Their second observation was that some degree of “reparalysis” occurred in 2 of their 97 patients. One reached a train-of-four ratio of greater than 0.90 and then dropped back to 0.87 within a few minutes. The other was found to have a train-of-four ratio of 0.81 roughly 45 min after achieving at train-of-four ratio of greater than 0.90. Both responded to an added dose of sugammadex. Reparalysis after sugammadex is described in the literature but the picture is inconsistent and the explanation is unclear. The mechanism is thought to be related to the dosing of sugammadex relative to the degree of neuromuscular blockade (and hence the amount of circulating relaxant) but it has been described with recommended doses and rapid reversal, and numerous studies with less than recommended doses have shown no problems (as was observed in the great majority of the patients in this study). Until we better understand this phenomenon, we believe that the best recommendation is careful quantitative monitoring and verification that the train-of-four ratio reaches a value of 0.90 and persists at this level for at least a few monitor cycles or continues to increase.

Their third observation involves patient-to-patient variability. Just as there is wide variability in patient sensitivity to nondepolarizing blockers, it appears that the same is true for the efficacy of sugammadex. From the authors’ Table 2, the average dose of sugammadex necessary to return the train-of-four ratio to a value of 0.90 from starting train-of-four-counts of 1 to 3 was 2.01 mg/kg with a SD of 1.22. This predicts that 16% of subjects (the area under the Gaussian curve one SD to the right of the mean value) will require a dose of at least 3.23 mg/kg to achieve satisfactory recovery at this level of block. Another 16% will require 0.79 mg/kg or less. Perhaps more to the point, the average dose of sugammadex necessary to return the train-of-four ratio to a value of 0.90 at train-of-four-counts of 2 or greater was 1.24 mg/kg with a SD of 0.83. This suggests that 16% of subjects will require a dose more than 2.0 mg/kg (the recommended dose at this level of block) to achieve satisfactory recovery. While perhaps a coincidence, this is almost exactly number of individuals in the current protocol who required doses above the usual dosing guidelines, which brings us to their fourth observation.

The authors’ fourth observation is by far the most important and most concerning. “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 [or 16%] required more than 2 mg/kg to achieve a train-of-four ratio of at least 0.9.” In addition, 2 of 29 patients with a train-of-four count of 1 or less required more than 4 mg/kg. This is obviously a potentially serious patient safety issue. Reversal based solely on twitch counts (or, worse, not based on any twitch assessment at all) does not guarantee full reversal. We believe that this constitutes some of the strongest evidence to date for the value of routine quantitative neuromuscular blockade monitoring—and clearly demonstrates the fallacy of believing that sugammadex obviates the need for such monitoring. Residual paralysis after sugammadex has been described before. Kotake et al.  working in an unmonitored operating room environment, showed a 46% incidence of an uncalibrated acceleromyography less than 1.0. Della Rocca et al. while showing a 100% incidence of full reversal (calibrated acceleromyography) 20 min after sugammadex, also noted that 27% of patients reversed from a deep block were not fully recovered 5 min after dosing (when most patients are probably extubated), and 7.1% were still not completely reversed after 10 min (which could be well after arrival in the postanesthesia care unit). Batistaki et al.  working (like Kotake et al. ) in a largely unmonitored environment and with uncalibrated acceleromyography, had an incidence of train-of-four ratio less than 0.90 of 9.5%, meaning a much higher incidence was likely if a train-of-four ratio target of 1.0 had been used. In 2017, before the introduction of quantitative monitoring to the University of Minnesota, a residual paralysis incidence of 24% postsugammadex (based on a uncalibrated acceleromyography ratio of 1.0 or less) was found in the postanesthesia care unit (personal communication).

Readers should remember that subjective or qualitative evaluation of the train-of-four (visual, tactile) cannot exclude a substantial degree of residual paralysis (train-of-four ratios greater than 0.40). At the current time, the only method for verifying full reversal—after any nondepolarizing agent or any reversal agent regardless of dose—is quantitative monitoring.

A final comment: We have repeatedly made reference to a train-of-four ratio of 0.90 with EMG or calibrated acceleromyography as “full reversal” (1.0 or greater by uncalibrated acceleromyography), but this is really incorrect. Considerable muscle receptor occupancy still exists at this level of block.  The following EMG study is instructive.  When a small dose of mivacurium (25 µg/kg) was initially administered to 12 subjects, the average twitch depression was about 12%. A larger dose resulting in 100% T1 depression was then administered. When the train-of-four ratio had recovered to 0.95, the same small dose of mivacurium resulted in 79% T1 depression, more than a 50% reduction in the calculated ED50 of the drug. Thus, at an EMG train-of-four ratio of 0.90, while a subject may have few symptoms of residual block, his/her neuromuscular reserve is still markedly diminished. This may be of little consequence in young, healthy patients, but in our sickest and oldest patients having major surgery, we should probably consider a target value closer to 1.0 with any form of quantitative monitoring.