| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL ARTICLE |
Department of Anesthesia Royal Hospital Muscat, Sultanate of Oman
Madan Mohan Maddali, MD Tel: +968 24499759 Fax: +968 24499759 Email: madanmaddali{at}hotmail.com, Department of Anesthesia, Royal Hospital, PB No: 1331, PC: 111, Seeb, Muscat, Sultanate of Oman.
ABSTRACT
To evaluate the role of bispectral index monitoring as an adjunct to balanced anesthesia in patients with myasthenia gravis undergoing transsternal thymectomy without the use of neuromuscular blocking agents, 10 patients were enrolled into this prospective observational study. After oral midazolam premedication, general anesthesia was induced with fentanyl, propofol, and sevoflurane. Tracheal intubation was performed without neuromuscular blocking agents. During maintenance, continuous monitoring of physiological and bispectral index parameters was used to titrate the doses of remifentanil, propofol, and sevoflurane. Sevoflurane concentration and propofol doses were adjusted to achieve bispectral index values in the high 30s to low 40 s. Propofol was discontinued when the sternum was approximated. Remifentanil infusion was stopped on subcutaneous tissue closure, and sevoflurane was switched off when nearing completion of skin closure. Tracheal extubation was performed when extubation criteria were met. On extubation, bispectral index levels were above 90. The median time from extubation to discontinuation of propofol was 28 ± 4 min, that of remifentanil was 21 ± 4 min, and it was 9 ± 5 min for sevoflurane. Bispectral index monitoring provided excellent hemodynamic control during surgery, and allowed early problem-free tracheal extubation.
Key Words: Anesthetics • Combined • Electroencephalography • Myasthenia Gravis • Thymectomy
INTRODUCTION
Myasthenia gravis (MG) is a neuromuscular disorder caused by a decrease in the number of acetylcholine receptors at neuromuscular junctions due to an antibody-mediated autoimmune attack. The thymus is abnormal in approximately 75% of patients, and thymectomy has gained widespread acceptance as the treatment for MG.1 To our knowledge, there has been no study where in bispectral index (BIS) monitoring was used to titrate the dosage of sevoflurane, propofol, and remifentanil, while avoiding neuromuscular blocking agents, for transsternal thymectomy in patients with MG.
PATIENTS AND METHODS
After hospital ethical committee approval and informed consent, 10 seropositive patients diagnosed with MG who were undergoing transsternal thymectomy were enrolled into this study conducted at our tertiary care cardiothoracic surgery center. Their demographic data, grades of MG according to the Osserman and Genkins classification of disease severity, the investigations on which the diagnosis was based, and doses and duration of various anti-MG medications were recorded (Table 1
). The preoperative complete blood count, urea and electrolyte levels, liver function tests, and coagulation studies were normal in all patients. Thyroid function tests were also normal, and no patient had an additional concomitant autoimmune disease (rheumatoid arthritis, systemic lupus erythematosus) as evidenced by the absence of rheumatoid factor, anti-dsDNA, and anti-nuclear antibodies. None of the participants exhibited any chest radiographic or electrocardiographic abnormalities. All patients were in American Society of Anesthesia class II physical status.
|
, Figure 1). In all cases, when the sternum was approximated with sternal wires, propofol infusion was discontinued. At the time of subcutaneous tissue closure, remifentanil infusion was stopped after administering a bolus of fentanyl 1 µg · kg–1 intravenously, and sevoflurane was discontinued when skin closure neared completion.
|
|
RESULTS
All patients were seropositive for MG with plasma acetylcholine receptor antibody levels above the normal range. The Tensilon (edrophonium) test was positive in all cases following intravenous administration of 5 mg of the drug. Repetitive nerve conduction studies of the facial muscles showed significant decremental responses in all patients, but none demonstrated significantly decreased responses on stimulation of the ulnar innervated muscles. The duration of induction of anesthesia up to the time of tracheal intubation was 6 ± 1 min. All patients were mechanically ventilated with a pressure-control mode and fixed positive end-expiratory pressure of 4 cm H2O. Peak airway pressures were not elevated at any point in any patient. On tactile assessment, TOF at the time of tracheal intubation showed a minimal fade of T4 compared to T1. The double-burst stimulus also exhibited some degree of fade. This phenomenon was noticed throughout the duration of sevoflurane administration. Once sevoflurane was discontinued, the fade phenomenon disappeared. At the time of tracheal extubation, all twitches were of the same intensity. None of the patients showed any movement during the course of surgery. Sevoflurane concentration, propofol and remifentanil doses were adjusted to achieve a BIS value below 40 during surgery. The duration of surgery was 35 ± 24 min. In all cases, there was a slight increase in systolic blood pressure in response to tracheal intubation and sternotomy. After deepening the plane of anesthesia with increases in sevoflurane concentration and propofol dose, further increases in systolic blood pressure were managed by increasing the remifentanil infusion rate until a steady state was achieved. When the stressful stimuli were overcome, the infusion rate of remifentanil was reduced, followed by adjustment of the propofol infusion rate and sevoflurane concentration (Table 2
).
The BIS values started rising once propofol was discontinued and were >90 at the time of return of spontaneous breathing and consciousness. All patients met the criteria for tracheal extubation: awake, neurologically intact, and with adequate ventilatory frequency to maintain arterial blood gas measurements of SaO2 
95%, PaCO2 35–45 mm Hg, and pH >7.35. All could raise their head from the bed and sustain a handgrip for 5 sec. The median time interval from discontinuation of propofol to tracheal extubation was 28 ± 4 min (25th–75th percentile: 23–30 min), that of remifentanil was 21 ± 4 min (25th–75th percentile: 16–23 min), and for sevoflurane it was 9 ± 5 min (25th–75th percentile: 5–14 min). At the time of tracheal extubation, BIS levels were high (low 90 s). Neuromuscular blocking drugs were not administered, and no reversal agents were given prior to tracheal extubation. Subsequently, all patients were comfortable, easily aroused (sedation score of 2, based on the modified Ramsay scale), and recovery was uneventful. Postoperative analgesia was provided with fentanyl infusion 1–2 µg · kg–1 · h–1. No patient had any recall or awareness during surgery. According to the unit’s protocol, all were given pyridostigmine, the initial dose being administered through a nasogastric tube on reaching the postoperative cardiac care unit; this was gradually reduced to nil over a month in all patients
DISCUSSION
The incidence of MG is 1 in 10,000–20,000 adults.2 This muscular disorder is generalized in 85% of patients, and confined to the extraocular muscles in 15%. Extremity musculature is symmetrically affected, with weakness of proximal muscle groups preferentially involved. Women are more commonly affected than men, with a ratio of 3 : 2. Approximately 10% of patients with MG have an associated autoimmune disorder. These include hyper- or hypothyroidism, rheumatoid arthritis, or systemic lupus erythematosus.3 Patients with MG pose significant management problems for the anesthesiologist. The major concerns are an unpredictable response to neuromuscular blocking agents and increased susceptibility to postoperative respiratory failure that may necessitate mechanical ventilation of variable duration. Various non-muscle relaxant techniques have been suggested for these patients, taking into consideration their inherent muscle weakness and dependency on preoperative anticholinesterase drugs. The hypothesis of this study was that by titration of anesthetic drugs based on BIS values, neuromuscular blocking agents could be totally avoided, and an anesthetic regimen could be developed to meet the procedural requirements. BIS has not been used widely for monitoring the effects of neuromuscular blocking agents nor as an indicator for tracheal intubation and extubation. Therefore, we had to correlate the results of accepted clinical indicators for tracheal intubation and extubation, such as TOF, double-burst stimulation, clinical and blood gas analysis data, with the observed BIS values to enable us to concluded that BIS values can be used as surrogate markers for tracheal intubation and extubation as well as for providing an immobile patient during surgery.
Propofol was found to be a suitable drug that allows tracheal intubation, fine-tuning of anesthetic depth, with a good recovery profile in MG patients undergoing surgery.4 A totally intravenous anesthetic technique with propofol and remifentanil, without recourse to neuromuscular blocking agents, for transsternal thymectomy for MG has also been reported.5 However, delayed postoperative arousal following remifentanil-based anesthesia in a patient with MG undergoing transsternal thymectomy has been described.6 Thus there was a need for an adequate anesthetic technique that would avoid the use of neuromuscular blocking agents. By titrating the dose of anesthetic drugs with BIS monitoring, early recovery is possible. BIS monitoring is known to facilitate optimization of drug delivery in anesthetized patients. Hence we used BIS monitoring to titrate the doses of propofol, sevoflurane, and remifen-tanil to achieve stable hemodynamics with early return of spontaneous breathing and problem-free tracheal extubation. Awake, non-premedicated patients have BIS values 93. Loss of recall occurs at BIS values of 75 to 80. Loss of consciousness correlates with BIS values of 68–75. BIS values of 40–60 have been recommended for anesthetic maintenance during general anesthesia. As BIS values fall from the mid 30 s to zero, EEG burst suppression increases to cortical silence.7 BIS values in the high 30 s to low 40 s were aimed at in this study, which correspond to deep sedation and near burst suppression.7
One should keep in mind that although a BIS value around 40 indicates that the patient is deeply unconscious, it does not necessarily signify that the patient will not respond by movement or hemodynamic changes to noxious stimulation such as surgical incision. Many factors including specific anesthetic agents, use of analgesics, and degree of surgical stimulation, all contribute to the potential for the unconscious patient with a low BIS value to respond. Therefore, we aimed for BIS values <40 to avoid any movement, especially during sternotomy and sternal retraction, which are very potent stimuli. We found that the patients maintained excellent hemodynamic parameters with no delay in extubation despite achieving such low BIS values. There were no signs of increased autonomic response as reflected in blood sugar levels, sweating, or sialorrhea at any time intraoperatively, which would indicate inadequate depth of anesthesia. By potentiation of the drug’s individual sedative and cardiovascular properties, we could prevent stress responses during surgical stimulation, except for changes in heart rate and systolic blood pressure. Measurement of stress hormone levels would have ensured to what extent stress responses were prevented, but that was not undertaken in this series. Hypotension was not encountered postoperatively, probably due to optimization of the dosage of the anesthetic drugs aided by their short duration of action, along with adequate hydration.
The proposed dose of propofol for producing deep hypnosis is 2.5 µg · mL–1 at the effect-site concentration, corresponding to an infusion rate of 6 mg · kg–1 · h–1, but we found that a dose of 2.2 ± 0.4 mg · kg–1 ·h–1 was adequate throughout the intraoperative period.8 The hypnotic effect of propofol is enhanced by µ-agonist opioids such as fentanyl or remifentanil, which might not be reflected in BIS values, but could result in reduced propofol requirements.9 Hence the synergistic effects of the other drugs administered might explain the low dose requirements of propofol in this series.
Remifentanil is an esterase-metabolized opioid with a short half-life.10 When remifentanil is used intraoperatively, the transition to an effective alternative analgesic regimen is a key consideration in postoperative pain management. In this study, remifentanil was discontinued 15 min before completion of surgery and replaced with a bolus dose and subsequent infusion of fentanyl. The terminal half-life of remifentanil is 15 min, and the time taken to achieve the peak effect after a bolus of fentanyl is only 5 min. It would be helpful to remember that, remifentanil could pose a problem in MG patients. Those patients maintained on pyridostigmine up until the time of surgery might be extremely sensitive to drugs metabolized by nonspecific esterases, such as remifentanil.6 Pyridostigmine might inhibit not only acetyl and butyryl cholinesterases but also non-specific esterases that metabolize drugs like remifentanil, resulting in delayed arousal on some occasions. It is uncertain whether this hypothesis also holds good for the total dose of remifentanil administered. We did not resort to a bolus dose of remifentanil and used the lowest possible rate of infusion. If esterase inhibition is a problem with pyridostigmine, it is logical that only the optimal dose of remifentanil should be used to avoid delayed arousal. This would strengthen the case for BIS monitoring as it would ensure just this. In this study, the remifentanil infusion rate was increased only if BIS values were persistently elevated, even after increases in sevoflurane and propofol. The remifentanil infusion rate was decreased first on achieving the desired BIS value.
Under constant propofol target-controlled infusion, it was found that µ-agonist opioids attenuate increases in BIS in response to tracheal intubation and endotracheal suction.9 This means that BIS values under a constant anesthetic regimen not only indicate a certain level of hypnosis but also reflect the degree of opioid-induced inhibition of the response to noxious stimuli. If BIS values suddenly increase in response to a noxious stimulus, this could be the result of a cortical arousal reaction reflecting a deficit in the analgesic component of anesthesia that would necessitate an increase in the analgesic dosage.9 This was the strategy we also adopted in our study. One should keep in mind that remifentanil might not have a direct effect on BIS in analgesic doses, but by ensuring good analgesia, rapid fluctuation in BIS values secondary to a noxious stimulus can be prevented.9 However, remifentanil at higher doses can directly influence BIS monitoring by attenuation of painful stimuli.11
The fade phenomenon on TOF and double-burst stimulation, which was noticed with the peripheral nerve stimulator, could be attributed to the administration of sevoflurane, which is consistent with the suggestion that sevoflurane has an inhibitory effect on neuromuscular transmission in both MG and control patients, probably due to its depressant action at pre-synaptic sites.12 We monitored the end-tidal concentration of sevoflurane and not the minimal alveolar concentration (MAC) values of the anesthetic gases. The MAC value for a volatile anesthetic provides the probability of patient movement, and is very useful in estimating the correct dose. One particular advantage of BIS monitoring is the ability to measure anesthetic effect in an individual patient throughout an operation. Because our anesthetic gas monitors display end-tidal gas concentrations and not MAC values, we combined the end-tidal measure of anesthetic dosing with the BIS value as a measure of anesthetic effect. We assumed that this would give a more comprehensive picture of the dose-response relationship. Nitrous oxide has a weak cortical action because its effect is mainly through activation of the descending inhibitory noradrenergic pathway in the brainstem and spinal cord. This effect is completely undetectable by the algorithm that computes BIS.9 However, adding nitrous oxide reduces the MAC of sevoflurane especially when using a remifentanil infusion.13
Limitations for BIS monitoring exist. High electromyographic activity and electric device interference could create subtle artifact signal pollution without necessarily being displayed as artifacts. This could be misinterpreted by the BIS algorithm as EEG activity, and assigned a spuriously increased BIS value. However, the new BIS XP system is designed to detect and filter interference from muscle artifacts, and BIS XP technology is resistant to electrocautery devices. Numerous clinical conditions have a direct effect on EEG cerebral function that could influence the BIS values. Even with the advanced BIS XP system, paradoxical BIS changes resulting in high BIS values due to diverse cortical effects were observed with anesthetics such as halothane. On the other hand, hypoglycemia could result in low BIS values due to elevated 
and 
waves and decreased 
waves. Cardiac arrest could also result in lower BIS values due to decreased cerebral perfusion.9
In a solitary case of transsternal thymectomy for MG, in which a total intravenous technique with propofol and remifentanil was used, a time of 9 min from cessation of anesthesia to extubation was reported by Lorimer and Hall.5 Earlier, Della Rocca and colleagues14 observed slightly longer times to extubation with either sevoflurane or propofol and nitrous oxide plus opioids, without neuromuscular blocking agents, in patients undergoing transsternal thymectomy for MG; the extubation time from discontinuation of propofol was 15 min (range, 10–18 min) vs. 12 min (range, 7–14 min) for the sevoflurane-based technique. In our study, the median time interval from discontinuation of sevoflurane to tracheal extubation was 9 ± 5 min which compares favorably with the earlier reported studies.
This was a pilot study performed at our institute as part of an ongoing project involving a larger patient population. The aim of this submission was to describe a technique with comparable, reproducible, and consistent results that would add to the armamentarium of the anesthesiologist. We recommend the use of BIS monitoring to titrate the doses of short-acting anesthetic drugs for intraoperative management of thymectomy.
ACKNOWLEDGMENTS
We are thankful for the support of Dr. John McDonnell and Dr. Geraldine O’Leary, Senior Consultants, Department of Anaesthesia, Our Lady’s Hospital, Navan, Ireland, in the preparation of the manuscript.
REFERENCES
Asian Cardiovasc Thorac Ann 2009;
17:389-394
© 2009 by SAGE Publications
DOI: 10.1177/0218492309338120
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
