“Medicine is a science of uncertainty and an art of probability.”
Sir William Osler (July 12, 1849–December 29, 1919)
In this issue of Anesthesia & Analgesia, Bronsert et al1 report on patient outcomes (30-day morbidity and mortality and long-term survival) after the use of intermediate-duration nondepolarizing neuromuscular blocking drugs (NMBDs). We wish to comment on the effect of pharmacologic reversal on respiratory complications and relate the findings to previous studies.
First, we consider the impact of using neostigmine (or avoiding its use) for reversal of neuromuscular blockade on the risk of postoperative respiratory complications. Bronsert et al1 reported that patients who did not receive neostigmine had more respiratory complications than patients who did (propensity modeling odds ratio [OR] 1.75 [1.23–2.50] and multivariate logistic regression 1.71 [1.24–2.37]; P < .001 and P < .0001, respectively). This finding is similar to that reported by Bulka et al.2 Of the patients who received an intraoperative NMBD, those who did not receive pharmacologic reversal were 2.26 times more likely to develop postoperative pneumonia than the patients who received neostigmine reversal (95% bootstrapped confidence interval [CI], 1.65–3.03).2 One of the conclusions from the reports by Bronsert et al and Bulka et al might be that pharmacologic reversal of intermediate NMBDs is needed to reduce postoperative respiratory complications.
On the contrary, it also has been suggested that neostigmine reversal may actually increase the risk of developing postoperative respiratory complications.3–6 In addition, it seems alarming that the use of intermediate NMBDs is associated with a 36% increase in the incidence of desaturation (defined as oxygen saturation of <90% with a decrease in oxygen saturation after extubation of >3%), a 40% increase in risk of reintubation requiring intensive care unit admission, and a 91-fold increased risk of death in patients who required tracheal intubation and intensive care unit admission.3 Not surprisingly, clinicians are left wondering how to handle this apparent conundrum.7–10
The main message of the study by Bronsert et al appears to be, “always reverse.” However, neostigmine, when used without the guidance of a device (a quantitative monitor that measures and displays the train-of-four [TOF] ratio in real time or at least a conventional peripheral nerve stimulator [PNS]), may result in suboptimal antagonism and lead to adverse clinical outcomes. Subjective evaluation of TOF stimulation with the use of a PNS requires the clinician to determine the number of twitches (TOF count) or the strength of the first response in the train and compare it with the fourth evoked response by tactile or visual (ie, subjective) means. The safe use of NMBDs requires both quantitative monitoring (to determine optimal timing and dosing of reversal agents) and pharmacologic reversal.8,11,12 Grosse-Sundrup et al3 found that patients who received both neostigmine and intraoperative PNS and whose tracheas were extubated in the operating room had fewer episodes of hypoxia (oxygen saturation <90% and oxygen saturation <93%) than patients who received neostigmine without a PNS (P = .003 and P < .001, respectively). Similarly, McLean et al6 found that administration of 60 μg/kg or less of neostigmine at a TOF count of ≥2 depth of block was associated with a decrease in the risk for postoperative pulmonary complications (OR, 0.79; 95% CI, 0.69–0.92; P = .002). However, just as written so elegantly by Meretoja and Olkkola8 in their 2012 commentary letter, the clinical study data do not simply and solely support the conclusion that intermediate-duration NMBDs need pharmacologic reversal: “Administration of neuromuscular blocking agents remains misty if neuromuscular transmission is not objectively monitored.” Consequently, the use of intraoperative monitoring impacts patient outcome, and the use of pharmacologic reversal agents that is not guided by quantitative monitoring probably is insufficient to ensure patient safety.
In light of the above, are we prepared to accept the conclusion by Bronsert et al1 that short- and long-term morbidity in patients who did not receive NMBDs is the same as that of patients who received NMBDs plus neostigmine reversal in the absence of a quantitative monitoring or a PNS? The authors’ Table 4 third row suggests that the odds of adverse respiratory outcomes in patients who received NMBDs of intermediate duration followed by reversal with neostigmine were similar to those of patients who did not receive a neuromuscular blocker (P = .99; OR 1.36; 95% CI, 0.69–2.70).1 This proposition might imply that clinicians can properly manage intraoperative neuromuscular blockade without a monitoring device; we already know this assumption to be false.13 In fact, we know that the lowest TOF ratio (fade) that can be discerned by subjective means (with PNS), even by experienced clinicians, is approximately 0.40.14 We also know that detection of TOF count, not just TOF fade, is inaccurate when clinicians use subjective means (ie, PNS); when TOF count assessed subjectively was compared with the TOF count measured with acceleromyography, an agreement was present in just over 50% of the patients.15 In those cases in which there was no agreement, the clinicians overestimated the TOF count 96% of the time.15 Even more troubling is that, despite this knowledge, more than half of clinicians still incorrectly estimate the incidence of significant residual block to be <1%.16
What do these results imply then? In our opinion, it is unclear how best to apply these findings to clinical practice. In part, this is because the reported CI for respiratory complications is wide (CI, 0.69–2.70), even though the sample size is 1944 per group. For this secondary finding, the authors’ sample was limited, and we would recommend that future sample sizes be larger. More troubling, though, is that the propensity matching presented in Table 4 did not control for the surgical procedure or for the method of postoperative analgesia (eg, the use of opioids). In fact, the surgical procedure is not included at all in the propensity score model. The issue, therefore, is not that there is an imbalance between the groups with regard to the type of surgical procedure but rather that the procedure itself is not even included in the analysis.
The last set of rows in Supplementary Digital Content Table 4 shows that Bronsert et al1 achieved a very good balance between groups for the surgical specialty. However, general surgery cases without neuromuscular blockade are typically not the same procedures as general surgery cases that require neuromuscular blockade. The same holds true for orthopedic surgery cases. Consistent with this limitation is the issue of opioid use (ie, another respiratory depressant) because opioid requirements differ among procedures. Epstein et al17 studied the timing of events in the postanesthesia care unit that would require the anesthesiologists’ and/or nurse anesthetists’ time. Episodes of hypoxemia occurred throughout the time when patients were in the postanesthesia care unit, mostly after >30 minutes from admission (ie, not immediately upon arrival, as might be expected to occur as a result of residual neuromuscular block). Patients receiving neuromuscular blocking agents (but not opioids) accounted for less than 5% of the episodes of hypoxemia in the postanesthesia care unit.
Consequently, we recommend that if future investigators seek to compare patients receiving intermediate NMBDs plus neostigmine reversal with patients who receive no neuromuscular blockade, they need to sufficiently control for the surgical procedure to account for the intraoperative and postoperative analgesia, including the use of opioids. However, this is a challenging comparison to make because, in general, the choice is moot for the surgical procedure (ie, neuromuscular blocker use is integrally related to the procedure). For example, McLean et al6 stated that, “in the cases with appropriate neostigmine reversal, total neuromuscular blockade dose given during surgery no longer predicted the risk of postoperative respiratory complications (composite respiratory outcome, highest versus lowest quintile of NMBD dose: OR, 0.98; 95% CIs, 0.63 to 1.52; P = .94).” They also were unable to control for the surgical procedure. Their article included analysis for the one most common procedure, laparoscopic cholecystectomy, but then this comparison was not made. We presume the comparison was omitted because laparoscopic cholecystectomies are nearly always performed with some degree of neuromuscular blockade. For instance, urologic procedures (eg, cystoscopies) may represent a suitable operative model in which muscle relaxation is not always indicated, but where one could compare patients undergoing surgery with or without muscle relaxation.
The study by Bronsert et al also suggests that minor degrees of neuromuscular block may not be associated with increased morbidity. There may be little evidence that TOF ratios of 0.8 to <0.90 are associated with increased morbidity, but this certainly does not mean that this assumption is true.18 Nor should we be tempted to infer that the study1 may have employed more frequent objective monitoring or that the dose and timing of neostigmine reversal was improved, leading to better outcome given the high incidence of disinclination to use even a PNS among clinicians.16
Although a number of observational studies have assessed postoperative pulmonary risk after the use of an NMBD, none has fully addressed the variables that impact residual block.1–6 We do not know when intraoperative neuromuscular monitoring devices were used, and we do not know the timing of administration of neostigmine in relation to the last dose of neuromuscular blocking drug.
Based on the above, the authors wish to commend all the investigators who continue to attempt to determine the relationship between surgical procedures, the use of NMBDs, pharmacologic reversal, the use of perioperative objective monitoring techniques, and postoperative outcome. We also call for appropriate study designs that will answer the few (but important) remaining questions. Until then, we will continue to make inferences under conditions of uncertainty. To paraphrase Sir William Osler, we need to reduce the uncertainty and increase the probability of good patient care.
Name: Mohamed Naguib, MB BCh, MSc, FCARCSI, MD.
Contribution: This author helped write the manuscript.
Conflicts of Interest: None.
Name: Franklin Dexter, MD, PhD.
Contribution: This author helped write the manuscript.
Conflicts of Interest: Several years ago, the University of Iowa Division of Management Consulting performed research for Merck & Co, Inc. Dr Dexter receives no funds personally other than his salary and allowable expense reimbursements from the University of Iowa and has tenure with no incentive program. He and his family have no financial holdings in any company related to his work other than indirectly through mutual funds for retirement. Income from the Division’s consulting work is used to fund Division research.
Name: Sorin J. Brull, MD, FCARCSI (Hon).
Contribution: This author helped write the manuscript.
Conflicts of Interest: Dr Brull is a shareholder and board of directors member, Senzime AB; funded research, Merck, Inc (Dr Brull receives no funds personally); is a member, board of scientific advisors, ClearLine MD.
This manuscript was handled by: Ken B. Johnson, MD.
1. Bronsert MR, Henderson WG, Monk TG, et al. Intermediate-acting nondepolarizing neuromuscular blocking agents and risk of postoperative 30-day morbidity and mortality, and long-term survival. Anesth Analg2017;124:1476–1483.
2. Bulka CM, Terekhov MA, Martin BJ, et al. Nondepolarizing neuromuscular blocking agents, reversal, and risk of postoperative pneumonia. Anesthesiology2016;125:647–655.
3. Grosse-Sundrup M, Henneman JP, Sandberg WS, et al. Intermediate acting non-depolarizing neuromuscular blocking agents and risk of postoperative respiratory complications: prospective propensity score matched cohort study. BMJ2012;345:e6329.
4. Meyer MJ, Bateman BT, Kurth T, et al. Neostigmine reversal doesn’t improve postoperative respiratory safety. BMJ2013;346:f1460.
5. Sasaki N, Meyer MJ, Malviya SA, et al. Effects of neostigmine reversal of nondepolarizing neuromuscular blocking agents on postoperative respiratory outcomes: a prospective study. Anesthesiology2014;121:959–968.
6. McLean DJ, Diaz-Gil D, Farhan HN, et al. Dose-dependent association between intermediate-acting neuromuscular-blocking agents and postoperative respiratory complications. Anesthesiology2015;122:1201–1213.
7. Nielsen JR. Re: Neostigmine reversal doesn’t improve postoperative respiratory safety [letter]. Available at: http://www.bmj.com/content/346/bmj.f1460/rapid-responses
. BMJ. 2013. Accessed November 30, 2016.
8. Meretoja OA, Olkkola KT. Re: Intermediate acting non-depolarizing neuromuscular blocking agents and risk of postoperative respiratory complications: prospective propensity score matched cohort study [letter]. Available at: http://www.bmj.com/content/345/bmj.e6329/rr/621252
. BMJ. 2012. Accessed November 30, 2016.
9. Kopman AF, Naguib M. Neostigmine: you can’t have it both ways. Anesthesiology2015;123:231–233.
10. Murphy GS, Kopman AF. “To reverse or not to reverse?”: the answer is clear! Anesthesiology2016;125:611–614.
11. Brull SJ, Kopman AF. Current status of neuromuscular reversal and monitoring: challenges and opportunities. Anesthesiology2017;126:173–190.
12. Naguib M, Johnson KB. Innovative disruption in the world of neuromuscular blockade: what is the “State of the Art?” Anesthesiology. 2017;126:12–15.
13. Naguib M, Kopman AF, Ensor JE. Neuromuscular monitoring and postoperative residual curarisation: a meta-analysis. Br J Anaesth. 2007;98:302–316.
14. Viby-Mogensen J, Jensen NH, Engbaek J, et al. Tactile and visual evaluation of the response to train-of-four nerve stimulation. Anesthesiology1985;63:440–443.
15. Bhananker SM, Treggiari MM, Sellers BA, et al. Comparison of train-of-four count by anesthesia providers versus TOF-Watch® SX: a prospective cohort study. Can J Anaesth2015;62:1089–1096.
16. Naguib M, Kopman AF, Lien CA, et al. A survey of current management of neuromuscular block in the United States and Europe. Anesth Analg2010;111:110–119.
17. Epstein RH, Dexter F, Lopez MG, et al. Anesthesiologist staffing considerations consequent to the temporal distribution of hypoxemic episodes in the postanesthesia care unit. Anesth Analg2014;119:1322–1333.
18. Piccioni F, Mariani L, Bogno L, et al. An acceleromyographic train-of-four ratio of 1.0 reliably excludes respiratory muscle weakness after major abdominal surgery: a randomized double-blind study. Can J Anaesth2014;61:641–649.