In this prospective, randomized, double-blind, controlled trial, sixty-six patients [American Society of Anesthesiologists (ASA) physical status I to II, aged 18 to 60 years] were recruited who were scheduled for surgery (estimated surgical time greater than 90 min, with similar pain stimulus and no need for a neuromuscular block during the surgical procedure). Exclusion criteria were patients with diseases or on medications known to interfere with neuromuscular transmission, hepatic or renal dysfunction, electrolyte abnormalities, allergy to drugs used in the study, body mass index <18 or >29 kg.m-2, expected to have difficulties during mask ventilation or intubation, pregnant or breastfeeding.
Using sealed opaque envelopes, numbered sequentially, sixty-four patients were randomly allocated to four parallel treatment groups (figure 1). The seal of the envelope was broken before the induction of general anesthesia by trained study personnel not involved in the data collection. Throughout the perioperative period, care providers, patients, and research team members were blinded to group assignment. The L group received lidocaine 3 mg.kg-1 as an IV bolus before the induction of anesthesia and 3 mg.kg-1.h-1 IV continuous. infusion during the operation period; the M group received magnesium sulfate 40 mg.kg-1 as an IV bolus before the induction of anesthesia and 20 mg.kg-1h-1 IV continuous infusion during the operation period. The ML group received equal doses of magnesium sulfate combined with lidocaine at the same conditions during the operation period. The control group received an equivalent volume of isotonic solution.
Patients were monitored using electrocardiography, noninvasive blood pressure measurements, pulse oximetry, capnography (Draeger Medical Systems, Telford, Pennsylvania, USA). Total Intravenous (IV) Anesthesia was standardized for all patients and performed without the use of benzodiazepines, using a propofol target dose (plasma targeting, Injectomat TIVA Agilia, Brazil) of 4 µg·mL-1 and a remifentanil infusion of 0.5 mg·kg-1·min-1. After induction, the propofol infusion target was decreased to 2.5 µg·mL-1, and infusion of remifentanil was adjusted to 0.1-0.3 µg·kg-1·min-1 as needed. If systolic arterial pressure (SAP) or HR increased or decreased by > 30% of baseline for > 60 sec, remifentanil infusion was respectively increased/decreased at 0.05 µg kg-1 min-1 until achieving the goal value within the range. If necessary, ephedrine´bolus (2.5 to 5 mg) was allowed. Hemodynamic parameters were considered stable when BP and HR were within 20% of baseline values.
After induction of anesthesia and loss of consciousness, the neuromuscular function was assessed by monitoring the adductor pollicis muscle via acceleromyography with a TOF-Watch SX device (Organon Ireland Ltd., a subsidiary of Merck & Co., Inc., Swords, Co., Dublin, Ireland) according to the neuromuscular research consensus. 28
The monitoring system was positioned on the side opposite to the blood pressure cuff and IV line. Pediatric surface electrodes (Red Dot®, 3M Health Care, Neuss, Germany) were placed on cleaned skin over the ulnar nerve on the volar side of the wrist. The transducer's position was secured by placing the thumb in a hand adapter and fixed a temperature sensor at the distal end of the forearm. TOF tracing was stabilized by administering 1 min of TOF repetitive stimulation, followed by 5 s of 50-Hz tetanus stimulation, and then another period of repetitive TOF stimulation for 3-4 min. CAL 2 mode determined the supramaximal threshold and to calibrate the transducer of the accelerometer. After calibrating the device and stabilization, the mean of three TOF's values was recorded in each patient and used as a reference. Then, the TOF recovery value was considered the measured equivalent of 90% of this predefined value. The same procedure was performed for the T1 measurements. Then, bolus doses of solutions were administered to assigned groups over 5 min. Subsequently, a total of 0.15 mg.kg–1 cisatracurium over 5 seconds (time point zero) was administered, which was followed by tracheal intubation when the TOF ratio reached zero. Patients were monitored until they achieved spontaneous recovery from NMB (TOF ratio=0.9). Values of T1, T2, T3, T4, and TOF ratio, as well as skin temperature, were recorded. No additional cisatracurium injections were permitted. After measuring onset time, stimulation mode was changed to TOF (2 Hz, stimulus duration of 200 µs, square wave, 15 s intervals). Finally, adequate normalization of the TOF recovery results according to the baseline values was provided to detect residual paralysis reliably.
All neuromuscular monitoring data were transferred in real-time and stored on a laptop using the TOF-Watch SX monitor computer program (version 2.5.INT; Organon Ltd., Dublin, Ireland).
The following variables were measured:
- Speed of onset – time in seconds required to reduce T1 response to 5% of initial contraction force.
- Clinical duration (Dur25%) – elapsed time in minutes for T1 response to recover 25% of its initial value.
- Recovery index – elapsed time in minutes between the recovery of the T1 response from 25% of its initial value (Dur25%) to 75% of its initial value (Dur75%).
- Pharmacological duration – elapsed time in minutes for the T1 response to recover 95% of its initial value (Dur95%).
- Spontaneous recovery – time in minutes to the recovery of T4/T1 to 90% of its initial value.
Body temperature, respiratory (end-tidal CO2) and hemodynamic parameters (systolic, diastolic, mean blood pressure and heart rate) were recorded and annotated at various times: M1- when the patient arrived in the operating room; M2- immediately before induction of anesthesia; M3- before the infusion of the tested solutions (saline, magnesium sulfate or magnesium sulfate associated with lidocaine); M4- five minutes after M3 (end of the infusion loading dose of test solutions); M5 immediately before intubation; M6- one minute after tracheal intubation and M7 every fifteen minutes until the end of the study. Heating elements were used to maintain their skin and central temperatures above 32 and 36 °C, respectively. All unexpected events that occurred during the study were recorded as adverse effects.
The Shapiro-Wilk test was used to assess normality. The clinical and demographic characteristics are expressed as the means ± SD or medians (IQR [range]) and were compared by analysis of variance, the Kruskal-Wallis test, or the chi-square test, where appropriate. Given that studies have suggested that the area under the curve (AUC) can provide a more accurate analysis of the hemodynamic data 29, 30, this approach was used to compare the hemodynamic responses among the study groups. The pharmacodynamic variables (i.e., speed of onset, clinical duration, recovery rate, and total duration) were represented as box-and-whisker plots showing the range, quartiles, and median. The AUCs of the changes in mean arterial pressure (MAP) and HR were expressed as mean ± SD (normally distributed data). The pharmacodynamic variables were compared among the groups via the Kruskal-Wallis test, followed by Dunn’s multiple comparison test. The AUCs of the changes in mean arterial pressure (MAP) and HR were compared among the groups by the one-way ANOVA followed by Tukey multiple comparison test. Both multiple comparison tests were used to control the type I error at 5%. The percentage of patients who achieved a TOF ratio of 90% without reaching 95% recovery of the first twitch (T1) response was compared among the groups by the chi-square test. A p-value <0.05 was considered statistically significant for all outcome variables.
For the sample size calculation, we considered a previous study showing that magnesium sulfate prolongs rocuronium-induced NMB.16 Having chosen a significance level of 5% and a power of 80%, we used the spontaneous recovery means of that previous study (73.2 ± 22 min with MgSO4 and 57.8 ± 14.2 min with saline) to calculate the number of participants required to detect a similar effect.31 The calculation revealed that N=14 patients were needed per group. As we included N=15-16 per group, we had 85% power to detect the same differences as planned a priori.