Monitoring and/or antagonizing the effects of muscle relaxant-induced neuromuscular block is a widely accepted standard in clinical practice. However, the reality is that there are international differences in this practice and, in many departments throughout the world, clinicians do not have access to equipment for measuring the degree of a neuromuscular block . Moreover, many anaesthetists do not routinely antagonize the neuromuscular block and, in some patients, 'reversal' agents may affect heart or lung function, or both [1,2]. In daily clinical practice, residual block can be excluded only by objective methods providing a quantitative measure of neuromuscular recovery. However, it is not possible to exclude clinically important residual paralysis either by clinical evaluation or tactile/visual evaluation of the response to peripheral nerve stimulation . Neuromuscular-blocking drugs are commonly given by incremental boluses or by continuous infusion during major abdominal and thoracic surgery. In the absence of monitoring, or when irrational dosage regimens are used or in absence of routine 'reversal', continuous infusions for lengthy interventions can cause overdosing and may delay recovery.
The study measured the infusion dose requirements for rocuronium and cisatracurium during propofol anaesthesia for major surgery in 30 patients. Infusions were discontinued at the beginning of surgical closure. As closure of the abdomen or thorax can often be achieved with the aid of deep anaesthesia or some mild degree of paralysis only, many anaesthetists tend to switch off the muscle-relaxant infusion at this particular moment. Spontaneous recovery of neuromuscular function in both groups was awaited and we recorded the recovery characteristics after the infusions had been discontinued. We investigated whether patients receiving a rocuronium or a cisatracurium infusion presented less residual curarization. Implications for safety from the differences observed are discussed.
Organization and recruitment
After Institutional Review Board approval and written informed consent, we studied 30 ASA I-III patients undergoing abdominal or thoracic surgery. Using data from previously published material, we performed a power analysis (two-group t-test of equal means) that revealed a sample size of 14 in each group having 80% power to detect a difference in train-of-four (TOF) ratio of 10% using a two-group t-test with a 0.05 two-sided significance level . The patients were allocated randomly by drawing lots to Group 1 (n = 15; cisatracurium) or 2 (n = 15; rocuronium). Exclusion criteria were: pregnancy; evidence of renal, hepatic, metabolic or neuromuscular disorders; and a history of medication interfering with neuromuscular transmission.
Neuromuscular transmission was monitored on the right arm by the electromyographic (EMG) response of the adductor pollicis muscle to TOF stimulation of the ulnar nerve, using surface electrodes (M-NMT™ module; Datex-Ohmeda, Helsinki, Finland). The TOF response to a supramaximal stimulus was obtained before the initial bolus of neuromuscular-blocking drug. Train-of-four was measured at 1-min intervals, using a square-wave, constant-current stimulus pulse with a width of 200 μs.
Induction and maintenance of anaesthesia
All patients were premedicated with lorazepam 2.5 mg, 1 h before induction. After 3 min of preoxygenation, anaesthesia was induced with propofol 5 mL min−1 until loss of consciousness and sufentanil 0.3 μg kg−1 intravenously (i.v.). As soon as the eyelid reflex was absent, assisted ventilation of the lungs by facemask was started with oxygen 100%. After equipotent bolus doses of cisatracurium 0.1 mg kg−1 or rocuronium 0.6 mg kg−1, endotracheal intubation was performed as soon as the first response to the TOF stimulus (T1) fell to <10%. Normocapnic ventilation was established with an ADU ventilator. Anaesthesia was maintained with oxygen 40% in air, propofol 3-6 mg kg−1 h−1 and supplemental boluses of sufentanil. No inhalation anaesthetics were used. Routine monitoring included electrocardiography (ECG), pulse oximetry, invasive and/or non-invasive arterial pressure, and central venous pressure. Temperature was monitored in the oesophagus and on the area of the skin on the right arm where neuromuscular transmission-monitoring electrodes had been applied. We used a forced air-warming system (Bair Hugger™; Augustine Medical, Inc, Eden Prairie, MN, USA) at the lower body surface to keep the oesophageal temperature between 35.5 and 37°C .
After recovery of T1/T0 to 10%, patients were given either cisatracurium or rocuronium at an infusion rate of 1.5 and 10 μg kg−1 min−1, respectively. The rate was manually adjusted to maintain T1/T0 at 10%. The concentration of the solutions used for infusion was 0.2 and 1 mg mL−1 for cisatracurium and rocuronium, respectively, so that even small changes in infusion rate led to significant changes in the volume delivered and, thus, to a quick alteration in effect . The infusion was discontinued at the beginning of surgical closure, whereas the administration of propofol was stopped when the surgeon started closing the skin.
Recovery of neuromuscular function
When surgery had been accomplished, the TOF ratio was assessed, and once it had recovered spontaneously to 0.9, the awakening patient's neuromuscular function was assessed clinically. The patient's response to the clinical tests was recorded. Neostigmine (50 μg kg−1) was administered only when a patient started to wake with TOF <0.9. When the TOF ratio had recovered to 0.9, patients were extubated when they were breathing spontaneously, fully awake and able to follow commands. All clinical tests were completed within 1 min and repeated within 5 min of the patient's arrival in the postanaesthesia care unit (PCU). In contrast to most studies published on this matter, we extubated our patients when the TOF had recovered to 0.9. The evidence that residual effects of neuromuscular-blocking drugs may persist until a TOF ratio of 0.9 is reached, enforced this strategy . Moreover, Eriksson and colleagues suggested an effect of vecuronium on carotid body hypoxic chemosensitivity at a TOF ratio of 0.7, thus impairing the hypoxic ventilatory response . In another study by Eriksson and colleagues, TOF < 0.9 was accompanied by a reduced ability to protect the airway . Return of the TOF ratio to 0.9 at the thumb is thus probably necessary to counteract any residual block in the pharynx and the facial muscles.
The results of monitoring the neuromuscular transmission and of all other vital parameters were displayed on an AS/3® monitor (Datex-Ohmeda) and stored in a spreadsheet on a PC (Compaq®; Houston, TX, USA). Infusion rate and duration were recorded throughout the procedure for each patient. The mean infusion rates in both groups were computed. We derived the patient characteristics and calculated the body surface area in both groups. The number of patients with a TOF ratio of ≥0.9 at the end of surgery was recorded, as well as the mean TOF ratio in each group at that time. Also, the number of patients needing neostigmine and the time-points, as well as the mean TOF ratio at which 'reversal' was attempted, were derived. The time intervals between discontinuing the infusion and the end of surgery and between the end of surgery and a TOF ratio of 0.9 were recorded. Additionally, the time interval between the end of surgery and extubation, as well as between the end of surgery and arrival at the PCU, were computed. We looked at unplanned admissions to the intensive care unit (ICU), as well as other adverse postoperative events. Finally, we investigated the following potential confounding factors: the infusion duration and the period between discontinuing the neuromuscular-blocking drug infusion and the end of surgery.
Statistical analysis (GraphPad®; InStat, San Diego, CA, USA) was performed to compare both groups and to investigate significant differences. A Kolmogorov and Smirnov normality test and an unpaired t-test were performed to compare body surface area. The calculated T1/T0 was compared with the preset value (T1/T0 = 10%) by means of a Kolmogorov and Smirnov test to assess normality and a one-sample t-test. We compared the duration of surgery as well as the infusion duration using a non-parametric U-test. The recovery time intervals, as well as the TOF ratio at the end of surgery and at the time of reversal, were compared between the cisatracurium and rocuronium groups using a non-parametric U-test. A Fisher's exact test compared the results of the clinical neuromuscular function tests between cisatracurium and rocuronium groups at two different time-points. Results are presented as mean ± SD. Significance was set at P < 0.05.
In the cisatracurium group, one patient suffered from bronchospasm following intubation, and another developed atrial fibrillation requiring intraoperative electrical conversion. In the rocuronium group, surgery in one patient was complicated by significant blood loss (<2000 mL). Another patient in the rocuronium group lost 2000 mL blood and required a small dose of norepinephrine infusion intraoperatively (<200 ng kg−1 min−1). Eight minutes after surgery, this patient had a neuromuscular transmission count of 1 and T1% = 9. It was decided to continue sedation and transfer the patient to the intensive care unit for postoperative ventilation of the lungs. Because of the infiltration seen on the chest radiograph and fever, this patient was not extubated until 3 days later and was excluded from the recovery analysis. Anaesthesia was uneventful in all other patients.
The duration of surgery in the cisatracurium and rocuronium groups was 242 ± 83 and 212 ± 137 min, respectively (P = 0.10). Table 1 shows the physical characteristics of the patients in the cisatracurium and rocuronium groups. The body surface area in the cisatracurium and rocuronium groups was 1.74 ± 0.18 and 1.82 ± 0.24 m2, respectively (P = 0.30). In the cisatracurium group, T1/T0 during the infusion was 10.3 ± 3.5%; this mean was not significantly different from the preset value (T1/T0 = 10%) (P = 0.76). The infusion rate of cisatracurium required to keep T1/T0 at 10% was 1.0 ± 0.3 μg kg−1 min−1. In the rocuronium group, the T1/T0 during the infusion was 10.0 ± 6.3%; this was not significantly different from the preset value (T1/T0 = 10%) (P = 0.99). The infusion rate of rocuronium required to keep T1/T0 at 10% was 5.6 ± 2.6 μg kg−1 min−1. The infusion duration was 174 ± 67 min for cisatracurium and 159 ± 142 min for rocuronium (P = 0.15).
Table 2 shows the recovery time intervals. In the cisatracurium and rocuronium groups, four (27%) and one (7%) patients, respectively, had a TOF ratio ≥0.9 at the end of surgery. The TOF ratio in each group at that time was 51 ± 32% for cisatracurium and 47 ± 31% for rocuronium (P = 0.78). Six (40%) patients in the cisatracurium group and seven (47%) in the rocuronium group required neostigmine; for cisatracurium, neostigmine was administered 10 ± 5.5 min after the end of surgery and for rocuronium, it was given 11.4 ± 7.2 min after the end of surgery. These time-points were comparable between both groups (P = 0.72). The TOF ratio at the time of reversal was 63 ± 7% for cisatracurium and 40 ± 19% for rocuronium (P = 0.01). Patients requiring neostigmine recalled no undesirable experiences while awakening.
Table 3 shows the results of the neuromuscular tests in the operating room and at arrival in the PCU. There was some variation in correspondence between the different neuromuscular tests and a TOF ratio of 0.9 in the operating room, as well as in the PCU. However, at both time-points there were no significant differences in the results of the clinical neuromuscular function tests between cisatracurium and rocuronium groups.
Baillard and colleagues reported the incidence of postoperative residual curarization in the PCU after a vecuronium bolus-induced neuromuscular block without the use of a nerve stimulator and without reversal . The results were alarming: 30-40% of patients had TOF < 0.7. Viby-Mogensen also formulated some evidence-based guidelines . Subsequently, we sought to investigate the incidence of postoperative residual curarization after infusions of the newer muscle relaxants, cisatracurium and rocuronium, during lengthy interventions (3-4 h), without attempting systematic 'reversal'. It is routine clinical practice for many anaesthetists to administer a continuous infusion of a muscle relaxant during protracted abdominal or thoracic surgery. Most often, anaesthetists discontinue the infusion at the start of surgical closure, as this moment is a recognizable event during the procedure and the method is close to common clinical practice. Throughout the world there are differences in the practice of antagonizing a neuromuscular block; in many countries, systematic antagonism of the effects of the block is not attempted routinely and especially in abdominal surgery, many anaesthetists do not use cholinergic drugs in order to avoid stress on the intestinal anastomoses . Moreover, anti-cholinesterases may induce serious adverse effects . Although there was a rational intraoperative dosage regimen for the muscle-relaxant infusions, we still found dramatic results: at the end of surgery, only one patient in the rocuronium group and four in the cisatracurium group recovered spontaneously to a TOF ratio of 0.9. This TOF level is apparently needed for safety [6-8]. In the other patients in both groups, we found important time intervals between the end of surgery and recovery to a TOF ratio of 0.9. Achieving spontaneous recovery of neuromuscular function took so long that six patients in the cisatracurium group and seven in the rocuronium group required pharmacological 'reversal', meaning that they began to wake while their TOF ratio was <0.9. As it only takes seconds before the onset of the effect of neostigmine, there was probably no risk that these patients would become conscious without full recovery of neuromuscular function [2,11]. Moreover, none of our patients recalled undesirable experiences while awakening. However, if partial curarization had not been recognized by monitoring or if pharmacological 'reversal' had not been attempted, there would have been a risk of evolution to full awakening with partial paralysis in these patients, and to a much more serious extent for rocuronium than for cisatracurium. However, these numbers of post-operative residual curarization were the result of rational dosage regimens of these muscle relaxants. We administered cisatracurium at 1 μg kg−1 min−1 and rocuronium at 6 μg kg−1 min−1, regimens that are in accordance with those reported in other investigations [11, 12]. If overdosing had occurred, the recovery results would be probably even worse.
The possible confounding factors in this study were not important: there was no difference in duration of infusion/surgery between the cisatracurium and the rocuronium groups. Moreover, there was no difference in the time interval between discontinuing the infusion and the end of surgery. We found some variable correspondence between the different clinical and neuromuscular tests, and a TOF ratio of 0.9 in the operating room as well as in the PCU. It is well known that clinical signs that require patient cooperation during recovery from anaesthesia offer partial degrees of reliability. Moreover, head lifting causes pain in patients with an abdominal incision and is notoriously useless. Beemer and Rozental described an inherent incidence of incomplete antagonism when only clinical methods were used . In the present study, we could only confirm the lack of a relationship between the results of the clinical neuromuscular tests and a TOF ratio of 0.9.
When comparing recovery characteristics between cisatracurium and rocuronium, we found that there was no significant difference in TOF at the end of surgery. However, the difference in time interval between the end of surgery and a TOF ratio of 0.9 was obvious but not statistically significant. Of importance to us was the significant difference in TOF in those patients who received neostigmine: when patients began to awaken, those in the cisatracurium group had a TOF ratio of nearly 0.7, whereas those in the rocuronium group had a TOF ratio of only 0.4, a value that is associated with a much more dangerous degree of inadequate 'reversal' of neuromuscular function. At time points thereafter, there was no longer a difference in effect between cisatracurium and rocuronium: the time intervals between the end of surgery and tracheal extubation and between the end of surgery and leaving patients safely in the PCU. One patient in the rocuronium group had to be transferred to the ICU due to an unforeseen event requiring prolonged mechanical ventilation of the lungs; a possible explanation may have been the extremely long surgical procedure (544 min). Although there were intraand postoperative complications in this patient, postoperative residual curarization may have contributed to the multifactorial causes of this expensive ICU transfer.
In summary, we recorded the recovery characteristics of cisatracurium and rocuronium infusions and looked at the safety implications for postoperative residual curarization in the event that the infusions were discontinued at the beginning of surgical closure and no routine 'reversal' was attempted. We found alarming numbers of residually curarized patients at the end of surgery. The time needed for spontaneous recovery was longer for rocuronium compared with cisatracurium. In both groups, some patients were given neostigmine as an escape. At that moment, the TOF ratio in the cisatracurium group was in a safer range than in the rocuronium group. In the rocuronium group, there was one unforeseen admission to the ICU. Postoperative residual curarization can perhaps be avoided by discontinuing the neuromuscular-blocking drug infusion earlier than at the beginning of surgical closure or by systematically antagonizing the neuromuscular block at the end of surgery. When 'reversal' is not attempted or not indicated, the choice of cisatracurium seems to be safer than rocuronium, as spontaneous recovery is longer with rocuronium. However, both drugs are unsuitable for continuous relaxation in abdominal/thoracic surgery when relaxation is mandatory until the abdomen/thorax is started to be closed and the patient has to be extubated at the end of surgery. In order to administer rational dosage regimens and to recognize possible postoperative residual curarization, the only objective and reliable guide is to use the neuromuscular transmission monitor [14,15]. Moreover, neuromuscular transmission monitoring must provide a quantitative TOF, as the present study confirms the relative uselessness of clinical recovery tests and previous reports have shown a discordance between TOF and visual/tactile fade detection .
1. Osmer C, Vogele C, Zickmann B, Hempelmann G. Comparative use of muscle relaxants and their reversal in three European countries: a survey in France, Germany and Great Britain. Eur J Anaesthesiol
2. Bevan DR, Donati F, Kopman AF. Reversal of neuromuscular blockade. Anesthesiology
3. Viby-Mogensen J, Jensen NH, Engbaek J, et al.
Tactile and visual evaluation of the response to train-of-four nerve stimulation. Anesthesiology
4. Cammu G, Coddens J, Hendrickx J, Deloof T. Dose requirements of infusions of cisatracurium or rocuronium during hypothermic CABG. Br J Anaesth
5. Heier T, Caldwell JE, Sessler DI, Kitts JB, Miller RD. The relationship between adductor pollicis twitch tension and core, skin, and muscle temperature during nitrous oxide-isoflurane anesthesia in humans. Anesthesiology
6. Kopman AF, Yee PS, Neuman GG. Relationship of the train-of-four fade ratio to clinical signs and symptoms of residual paralysis in awake volunteers. Anesthesiology
7. Eriksson LI, Sato M, Severinghaus JW. Effect of a vecuronium-induced partial neuromuscular block on hypoxic ventilatory response. Anesthesiology
8. Eriksson LI, Sundman E, Olsson R, et al.
Functional assessment of the pharynx at rest and during swallowing in partially paralyzed humans: simultaneous videomanometry and mechanomyography of awake human volunteers. Anesthesiology
9. Baillard C, Gehan G, Reboul-Marty J, et al.
Residual curarization in the recovery room after vecuronium. Br J Anaesth
10. Viby-Mogensen J. Postoperative residual curarization and evidence-based anaesthesia. Br J Anaesth
11. McCoy EP, Mirakhur RK, Maddineni VR, Loan PB, Connolly F. Administration of rocuronium by continuous infusion and its reversibility with anticholinesterases. Anaesthesia
12. Belmont MR, Lien CA, Quessy S, et al.
The clinical neuromuscular pharmacology of 51W89 in patients receiving nitrous oxide/opioid/barbiturate anesthesia. Anesthesiology
13. Beemer GH, Rozental P. Postoperative neuromuscular function. Anaesth Intens Care
14. Ballantyne JC, Chang Y. The impact of choice of muscle relaxant on postoperative recovery time: a retrospective study. Anesth Analg
15. Shorten GD. Postoperative residual curarisation: incidence, aetiology and associated morbidity. Anaesth Intens Care