Propofol Has No Direct Effect on Sinoatrial Node Function or on Normal Atrioventricular and Accessory Pathway Conduction in Wolff-Parkinson-White Syndrome during Alfentanil/Midazolam Anesthesia

With Ethics Committee approval and signed patient consent, 12 patients (5 men and 7 women, aged 1839 yr) with WPW syndrome undergoing elective radiofrequency catheter ablation were studied. All patients presented with a history of paroxysmal supraventricular tachycardia or atrial fibrillation. Patients had no history of renal or hepatic disease. Antiarrhythmic therapy was discontinued for a period greater than 5 half-lives of the antiarrhythmic agent before the electrophysiologic study; no patient received amiodarone therapy.

On the day of procedure, patients received no premedication, and a blood sample was taken to determine serum electrolyte concentrations. Anesthesia was induced with 50 micro gram/kg alfentanil and 0.15 mg/kg intravenous midazolam, and tracheal intubation was facilitated with 20 mg intravenous vecuronium. Anesthesia was maintained with a continuous infusion of 2 micro gram *symbol* kg sup -1 *symbol* min sup -1 alfentanil and intermittent doses of 12 mg intravenous midazolam every 15 min, as required. Positive-pressure ventilation with air/oxygen was used to maintain normocapnia and arterial oxygen saturation > 96% (Nellcor Capnograph/Oximeter). Other monitors included a noninvasive blood pressure cuff, a three-lead electrocardiogram, and a nasopharyngeal temperature probe.

After induction, a baseline electrophysiologic study was performed using transvenous endocardial electrodes. For extrastimulus testing, the right atrium or right ventricle was paced at a standard drive cycle length of 500 or 400 ms, and the extrastimulus was coupled (steps of 20 ms from 400-300 ms and 10 ms below 300 ms) to every eighth drive beat. This protocol allowed calculation of right atrial refractory period (including right atrial effective refractory (RAERP), AV node effective refractory period (AVNERP), and accessory pathway effective refractory period (APERP)) and right ventricular effective refractory period (RVERP) at a cycle length of 400 ms. Decremental pacing of the right atrium or ventricle allowed determination of the anterograde and retrograde shortest cycle length (SCL), also known as the Wenckebach cycle length, with 1:1 conduction over the normal AV node and accessory pathway. Sinus node function was measured by right atrial pacing (overdrive suppression) to measure the sinus node recovery time and corrected sinus node recovery time, and SA conduction time was measured using the standard extrastimulus protocol. Baseline intervals measured during sinus rhythm included intraatrial conduction time and atrial-His interval. During induced reciprocating tachycardia, conduction times of each component of the reentrant circuit were measured and included cycle length, atrial-His interval, His-ventricular interval, and ventriculoatrial interval. See Table 1for definitions of parameters. A more detailed description of methodology and definitions can be found elsewhere. [8]Alfentanil/midazolam administration was then discontinued. Immediately thereafter, propofol was administered (bolus 2 mg/kg followed by a continuous infusion of 120 micro gram *symbol* kg sup -1 *symbol* min sup -1). Blood pressure measurements were recorded 1 min before and at 1-min intervals (X15 min) after the bolus of propofol, then every 5 min for the duration of the procedure. Fifteen minutes after the initiation of propofol infusion, the electrophysiologic measurements were repeated.

Comparison of arterial blood pressure and heart rate parameters measured before and at 1, 2, 3, 5, 10, and 15 min during propofol infusion was made using MANOVA. When statistical significance was achieved, the Student-Newman-Keuls test was used to determine which interaction was significant. Two-sided paired Student's t tests compared the electrophysiologic variables measured before and during propofol administration. All results are expressed as mean plus/minus SD, P < 0.05 was considered statistically significant.

The mean age and weight of the 12 patients were 26 plus/minus 7 yr and 76 plus/minus 19 kg, respectively. Serum electrolyte analysis revealed normal values in all patients (Table 2). During propofol infusion, there was a statistically significant decrease in systolic blood pressure at 3, 5, and 10 min after propofol bolus, whereas diastolic and mean arterial pressures and heart rate were similar compared to baseline measurements (Table 3). At the time of repeat measurement of the electrophysiologic parameters (15 min), all hemodynamic variables were similar to control values.

Comparison of the electrophysiologic parameters in the normal AV conduction system and accessory pathway performed before and during propofol infusion demonstrated a statistically significant prolongation of the right atrial effective refractory period (Table 4). In this study population, it was frequently not possible to measure the shortest paced cycle length permitting 1:1 conduction over the AV node (8 of 12 patients) because of concomitant conduction over the accessory pathway, because it would be impossible to determine when AV node block occurred.

Comparison of parameters of sinus node function and intraatrial condition performed before and during propofol infusion demonstrated no significant change in these parameters (Table 5).

Sustained reciprocating tachycardia was inducible in 8 of 12 patients before and during propofol infusion. Comparison of the parameters of sustained reciprocating tachycardia measured before and during propofol infusion demonstrated no significant change (Table 6). Furthermore, of the remaining four patients in whom sustained reciprocating tachycardia was not inducible, the administration of propofol did not promote inducibility of sustained reciprocating tachycardia.

All accessory pathways were successfully identified and ablated. In addition, no tachyarrythmias occurred during general anesthesia other than those induced during the electrophysiologic study.

This study demonstrates that the addition of propofol with alfentanil/midazolam/vecuronium anesthesia has no clinically significant effect on the refractoriness or conduction of either the normal AV or the accessory pathways, no direct effect on SA node function or intraatrial conduction, and no effect on the electrophysiologic characteristics of sustained reciprocating tachycardia. These results offer no evidence for propofol to induce intraoperative bradyarrhythmias. Furthermore, propofol allows unimpeded electrophysiologic expression of the accessory pathway and is therefore an appropriate agent for use as either a general or neuroleptic anesthetic for accessory pathway ablation procedures.

Sustained reciprocating tachycardia is a useful and preferred setting to demonstrate the anatomic location and electrophysiologic characteristics of the accessory pathway in patients with WPW syndrome. After its induction, the reentrant circuit is easily identified, and detailed recordings are obtained from most of the circumference around the mitral or tricuspid annulus. After localization of the accessory pathway, high radiofrequency catheter ablation is performed. [9]It is frequently beneficial or necessary for patients who present for ablation of accessory pathways to receive sedation, especially if the procedure is long or complicated. In some cases, a general anesthetic is required. It is therefore necessary to administer anesthetic agents that do not interfere with the physiologic expression of the accessory pathway or affect the inducibility of sustained reciprocating tachycardia. This study demonstrates that propofol provided a hemodynamically stable anesthetic and did not interfere with the physiologic expression of the accessory pathway or inducibility of sustained reciprocating tachycardia. Although there was a statistically significant prolongation of the right atrial effective refractory period, the magnitude of this change was clinically unimportant. Prolongation of the right atrial effective refractory period would not be expected to interfere with reciprocating tachycardia unless the magnitude of the increase of the right atrial effective refractory period was sufficient to encroach on the conduction properties of the accessory pathway, either the accessory pathway effective refractory period or the shortest cycle length of the accessory pathway. In this study, right atrial effective refractory period was prolonged to 229 ms, at which time the shortest cycle length and effective refractory period of the accessory pathway were 320 and 309 ms, respectively.

It is well known that not all patients with a documented clinical tachycardia are inducible in the electrophysiology laboratory even in the absence of anesthetic drugs. [8]Reasons for this are uncertain, but differences in the patient's autonomic tone between the ambulatory state and the situation in the electrophysiology lab may be a major factor. It is possible that the inability to induce tachycardia in four of our patients was due to the presence of alfentanil/midazolam/vecuronium anesthesia. However, it does not change the fact that, when we repeated all measurements of conduction during the addition of a propofol anesthetic, propofol did not elicit any significant change in the measured parameters of reciprocating tachycardia (Table 6), and the ability to induce reciprocating tachycardia was not affected by propofol administration. Furthermore, in the remaining four patients who were not inducible before propofol administration, the addition of propofol anesthesia did not promote inducibility.

Clinical reports implicating propofol as causing intraoperative bradyarrhythmias are conflicting. The documented cases of propofol-associated bradyarrhythmias occurred after situations known to elicit these heart rate effects. For example, Galletly et al. studied 50 consecutive patients receiving a continuous propofol infusion and found two patients had episodes of bradycardia: one episode occurred during peritoneal manipulation, and the other occurred after anticholin-esterase administration, [4]both of which are known to cause bradyarrhythmias. [5,6]Degrood et al. presented a study of 15 patients who received propofol in which one patient developed bradycardia after succinylcholine administration, an agent also known to cause bradyarrhythmias. [3,5]Baraka presented a study of six patients, two of whom developed bradycardia, again after succinylcholine administration. [2]As a result, the remaining four patients were premedicated with atropine, and bradycardia did not occur. In the case report by Hermann and Vettermann, [10]who suggested that propofol anesthesia suppressed an acute supraventricular tachycardia in a 3-yr-old child with documented supraventricular tachycardia, the cardioversion to normal sinus rhythm occurred after intubation, a known vagotonic stimulus. [6]On the other hand, there are multiple reports involving 121 patients receiving propofol anesthesia, none of whom developed intraoperative bradyarrhthmias. [11-17]Therefore, it is not clear whether the reported episodes of intraoperative bradycardias were related to propofol or due to conditions known to elicit bradyarrhythmias. The results of this study suggest that propofol has no direct effect on the heart to induce bradyarrhythmias.

The proposed mechanisms of propofol-associated bradyarrhythmias are also conflicting in the literature in both animal and human studies. For example, in vitro experiments with the rabbit SA node preparation demonstrated that propofol significantly reduced atrial conduction at 33 micro gram/ml and resulted in complete block at 100 micro gram/ml. [18]In humans, the mean blood concentrations of propofol at the onset of unconsciousness after a bolus dose of 2 mg/kg is 10.5 micro gram/ml, and an adequate level of anesthesia is maintained with blood concentrations ranging from 3.4 to 4.5 micro gram/ml. [19]Therefore, the concentrations of propofol used in the rabbit in vitro model are considerably greater than what we achieve clinically. Studies in rabbits and guinea pigs receiving propofol resulting in serum concentrations consistent with clinical practice demonstrated minimal effects on SA node activity. [18,20]Yet, Colson et al., in a study of only two dogs, suggests that propofol exhibits a direct depressor effect on sinus node activity as demonstrated during in vivo measurement of RR, PR, and QRS intervals with endocardial electrodes. [21]In humans, Cullen et al. suggest propofol resets the baroreflex control, thereby allowing slower heart rates at decreased arterial pressures. [22]This hypothesis is supported by two subsequent studies that concluded propofol attenuates baroreflex sensitivity. [23,24]Our study demonstrated that a 2 mg/kg bolus of propofol followed by continuous infusion of 120 micro gram *symbol* kg sup -1 *symbol* min sup -1 had no significant direct effect on sinus node activity or intraatrial conduction. Prolongation of the sinus node recovery time, corrected sinus node recovery time, or SA conduction time would theoretically promote bradyarrhythmias; yet, propofol had no effect on these variables.

Our conclusions are supported by a recent study by Deutschmen et al., who measured heart rate variability spectra using a fast Fourier transformation technique, [25]as an indicator of parasympathetic and sympathetic tone. [26]In their study of 10 patients, propofol had no effect on heart rate, yet they demonstrated that parasympathetic tone was reduced to a lesser degree than sympathetic tone, thereby predisposing patients to develop bradyarrhythmias in response to vagotonic stimuli. [26]It is therefore unlikely that the reported intraoperative bradyarrhythmias associated with propofol on sinus node activity. Rather, these bradyarrhythmias were more likely due to surgical stimuli and/or administration of drugs known to possess cholinergic activity. Although our study demonstrated no significant change in either heart rate or arterial blood pressure, these effects may not be similar at higher propofol doses. For example, Ebert et al. demonstrated that following a larger dose of propofol (2.5 mg/kg + 200 micro gram *symbol* kg sup -1 *symbol* min sup -1) resulted in a significant impairment of baroreflex regulatory responses. [27]Therefore, the sympathetic inhibition that they demonstrated during propofol administration may lead to exacerbation of hypotension and bradyarrhythmias during its administration. We chose an induction dose of 2 mg followed by a maintenance infusion of 120 micro gram *symbol* kg sup -1 *symbol* min sup -1, which is approximately midway between the range of infusion rates (4.8 mg *symbol* kg sup -1 *symbol* h sup -1 to 12 mg/kg/h) previously reported for general anesthesia. [28]As indicated in Table 4, the "n" values for the conduction parameters are dissimilar for two reasons. First, not all accessory pathways are capable of bidirectional conduction [8]; therefore, some parameters could only be measured in one direction, in some patients. Second, conduction through the normal AV pathway may be obscured in in some patients by concomitant conduction over the accessory pathway. The limitations of our study conclusions pertain to the fact that the effects of propofol on the electrophysiologic properties of the human heart were measured in patients receiving a basal anesthetic of alfentanil/midazolam/vecuronium. For example, at the time the measurements were taken during propofol infusion, approximately 15 min had elapsed after the discontinuation of alfentanil/midazolam. Simulation of a pharmacokinetic model of alfentanil suggests that blood concentration of this agent still would be clinically significant at the time of measurement. [29]However, we submit that the effects, if any, of alfentanil/midazolam/vecuronium were minimal for two reasons. First, the baseline measurements of all conduction parameters was done during a basal alfentanil/midazolam/vecuronium anesthetic, and these measurements were repeated following the addition of propofol anesthesia. Therefore, any changes in the electrophysiologic properties demonstrated during propofol infusion had to be the result of propofol administration alone. Second, we previously measured the electrophysiologic parameters of the heart in WPW patients during alfentanil/midazolam/vecuronium anesthesia compared to the parameters measured in the same patient before anesthesia and showed conclusively that alfentanil/midazolam/vecuronium anesthesia does not affect the electrophysiologic properties of the heart. [7]On the other hand, there is a possibility that the decrease in systolic blood pressure at times 3, 5, and 10 min after the bolus infusion of propofol may have elicited an increase in reflex sympathetic activity, possibly influencing our results. However, we consider this change (100 → 92 mmHg) to be clinically insignificant, because there was no concomitant change in mean blood pressure or heart rate. Furthermore, at the time of measurement of the electrophysiologic parameters during propofol infusion, all hemodynamic parameters were similar to baseline values.

In summary, our study demonstrates that the administration of propofol in clinical doses had no effect on conduction properties of either the normal AV conduction system or the accessory pathway in patients with WPW syndrome. Furthermore, this dose of propofol (2 mg/kg + 120 micro gram *symbol* kg sup -1 *symbol* min sup -1) had no effect un SA activity, intraatrial conduction, or electrophysiologic expression of the aberrant pathway. We conclude that propofol is suitable for use in patients undergoing ablative procedures for accessory pathways who require either a neuroleptic or general anesthetic and that it has no direct effects on the heart to induce bradyarrhythmias.