Secondary Logo

Journal Logo

A Comparison of the Incidence of the Oculocardiac and Oculorespiratory Reflexes During Sevoflurane or Halothane Anesthesia for Strabismus Surgery in Children

Allison, Celia E. MD*; De Lange, Jacob J. MD, PhD*; Koole, Frank D. MD; Zuurmond, Wouter W. A. MD, PhD*; Ros, Herman H. PhD*; van Schagen, Nico T. BSc*

doi: 10.1213/00000539-200002000-00012

We examined changes in the cardiorespiratory system of small children during surgical correction of strabismus with a laryngeal mask airway and spontaneous respiration with sevoflurane or halothane inhaled anesthesia. Fifty-one children, 1–7 yr old, having outpatient strabismus correction were randomized to sevoflurane (S) or halothane (H) in 66% nitrous oxide at 1.3 minimum alveolar concentration. Children breathed spontaneously through a laryngeal mask airway and were not pretreated with anticholinergics. The oculocardiac reflex (OCR), defined as a 20% decrease in heart rate (HR) from baseline, dysrhythmias, or sinoatrial arrest concomitant with ocular muscle traction occurred less frequently with sevoflurane than with halothane (S 38%, H79%, P = 0.009). The baseline HR was higher with sevoflurane (S 114 ± 13 bpm, H 101 ± 15 bpm, P = 0.002). The lowest HR occurred with halothane (S 95 ± 22 bpm, H 73 ± 19 bpm, P = 0.001). The incidence of dysrhythmias was higher in the halothane group (S 4%, H 42%, P = 0.004). Reductions in minute ventilation and PETCO2 accompanied OCRs. Airway irritability was present with halothane only (S 0, H 3). Eleven children, of whom the majority had received halothane, required measures to correct SpO2 < 95% or PETCO2 > 60 mm Hg during maintenance anesthesia (S 11%, H 32%). Sevoflurane may be a more suitable anesthetic than halothane for operations involving traction on the ocular muscles with spontaneous respiration in children because of reduced incidence of OCR, airway irritability, and ventilatory disturbances.

Implications Some children experience a sudden slowing of the heart and impaired breathing when the surgeon pulls on the eye muscles during squint operations under anesthesia. Sevoflurane, a recently developed anesthetic vapor, may reduce this problem when compared with the established vapor halothane.

Departments of *Anesthesiology and †Ophthalmology, Academic Hospital Vrije Universiteit, Amsterdam, The Netherlands

November 3, 1999.

Address correspondence and reprint requests to Celia E. Allison, MD, Department of Anesthesiology, Academic Hospital Vrije Universiteit, de Boelelaan 117, 1018 HV Amsterdam, the Netherlands. Address e-mail to

Presented in part at the Annual Meeting of the American Society of Anesthesiologists, San Diego, CA, October 18–22, 1997.

Anesthesia using the laryngeal mask airway (LMA) with spontaneous breathing has been suggested as a suitable anesthetic technique for children undergoing strabismus surgery in the outpatient setting (1). A recent survey found that 60% of replying anesthesiologists prefer to secure the airway of children 4 yr and older undergoing strabismus correction with an LMA (2). There have been no studies examining this technique in children.

During strabismus surgery, traction on the ocular muscles causes abrupt changes in the cardiovascular and respiratory systems. The oculocardiac response (OCR) is defined as a decrease in heart rate (HR) of more than 20% of the baseline value, dysrhythmias, or sinoatrial arrest (3). The oculorespiratory reflex (ORR) is the development of a reduction in tidal volume (VT) and respiratory rate (RR) (4). Because sevoflurane and halothane cause similar levels of respiratory depression in children at 1 minimum alveolar concentration (MAC) (5), and because halothane is associated with a lower HR (6) and a higher incidence of dysrhythmia (7), we hypothesized that sevoflurane might cause less a severe OCR and ORR than halothane during strabismus surgery with spontaneous respiration.

The purpose of our study was to determine the degree of change in HR, RR, VT, PETCO2, and SpO2 during strabismus surgery in children anesthetized with equipotent concentrations of sevoflurane or halothane breathing spontaneously via a LMA.

Back to Top | Article Outline


The study protocol was approved by our institutional review board. Children between the ages of 1 and 7 yr, ASA physical status I or II, scheduled to undergo surgical correction of strabismus were eligible to enter the study. Children and their parents were seen 1 wk before the planned operation in the anesthesia preassessment clinic. In our population, very few children chose IV induction despite the availability of local anesthetic skin creams. Those requesting inhaled induction were asked for consent for participation in the trial.

Children were allowed food and milk products on the evening before surgery until midnight. On the day of surgery, the children were advised to drink sweet drinks until 2 h before daycare admission. Surgery took place within 1 h of admission. Children were not premedicated. Parents accompanied their children into the operating room and were present during the induction of anesthesia. Before the induction a Clip-Tip SpO2 probe was attached to a finger or toe (Oh- meda, Louisville, CO). Anesthesia was induced through an unscented face mask with oxygen in 66% nitrous oxide (N2O) via the Bain system with fresh gas flows sufficient to eliminate rebreathing. Inspiratory and expiratory gas concentrations and volumes were measured with a pediatric Datex Capnomac Ultima by using the Datex D-lite sensor (Datex, Helsinki, Finland). Children were randomly allocated to the study groups at the time of the induction of anesthesia by having a nurse pick a colored marble out of an opaque bag with equal numbers of red and blue marbles signifying sevoflurane (Group S) and halothane (Group H), respectively. The overpressure technique was used for induction. After four breaths of oxygen and N2O, the vaporizers were set at the maximal concentration deliverable, 8% for sevoflurane or 4% for halothane.

After loss of consciousness, a three-lead electrocardiogram and noninvasive blood pressure monitoring were applied. Clinical readiness for LMA insertion was jaw relaxation. The presence of an IV line was not a prerequisite for this procedure. An appropriately sized LMA was placed and checked for absence of obstruction. The polyethylene gas-sampling catheter (Lectocath, Vygon, France, internal diameter 1.0 mm) was advanced via a latex membrane on the Luer lock port at the elbow connector to within the distal end of the LMA. Gas sampling speed was 250 mL/min. If hypoventilation was present at this time, ventilation was briefly assisted manually. Vaporizers were then adjusted to an approximate MAC multiple of 1.3 for either anesthetic, 2.5% in the sevoflurane group and 0.5% in the halothane group. In calculating equipotent concentrations the following data were used. The MAC reducing effect of 60% N2O on 1 MAC sevoflurane is 24% (8), and on that of 1 MAC halothane is 60% (9).

Sevoflurane 2.5% is 1 MAC [MAC is 2.5% in this age group (8)]; 66% N2O adds approximately 0.24 MAC when combined with sevoflurane; sevoflurane contributes 1 MAC and N2O 0.24 MAC = 1.24 total MAC. 0.5% halothane is 0.55 MAC, [1 MAC is 0.91% in this age group (10)]; 66% N2O adds 0.66 MAC; halothane contributes 0.55 MAC and N2O contributes 0.66 = 1.21 total MAC.

Ringer’s lactate was administered via an IV cannula. Paracetamol 40 mg/kg was given, but no opiates or prophylactic anticholinergics.

After the operative field was prepared, HR (average of 3 s) was noted, and this was taken as baseline. The OCR was considered present if the HR decreased by 20% from this value or if dysrhythmias or sinoatrial arrest occurred during traction of the ocular muscles. If the HR did not increase after release of muscle tension, atropine 0.02 mg/kg was administered. If SpO2 decreased to values <95% fraction of inspired oxygen (FIO2) was increased to 50%. If there was no improvement in the SpO2, ventilation was assisted manually. Similarly, ventilation was assisted in children whose PETCO2 increased above 60 mm Hg. Children with persistent dysrhythmias were to be changed to isoflurane as the maintenance anesthetic, and those with persistent airway irritability were to be changed to IV propofol.

After surgery was completed, the child was given 100% oxygen, and the LMA was removed under deep anesthesia with the vaporizer on maintenance concentrations. The child was then turned on his or her side and transported to the recovery room.

Binary data were presented as frequencies, and continuous variables were presented as mean and standard deviations. Comparisons between Groups S and H were made by using Fisher’s exact test, unpaired t-tests, or Mann-Whitney U-tests for nonnormally distributed variables as appropriate.

Back to Top | Article Outline


Demographic data are given in Table 1. Fifty-one children were studied, and there were no statistically significant differences. The number of muscles operated on was similar, and the average duration of surgery was 35 min. Figure 1 shows an example of HR, RR, VT, and PETCO2 in a study patient anesthetized with halothane.

Table 1

Table 1

Figure 1

Figure 1

The incidence of the OCR is presented in Table 2. Baseline HR was higher in the sevoflurane group (P = 0.002). The OCR occurred less often in the sevoflurane group (S 38%, H 79%;P = 0.009). Halothane was found to have the lower mean lowest HR during OCR. There was no difference in the baseline systolic blood pressure (S 102 ± 12 mm Hg, H 98 ± 12 mm Hg). Dysrhythmias were seen after traction on the ocular muscles and were more common in the halothane group (S 4%, H 42%;P = 0.004). One child in the sevoflurane group experienced nodal rhythm. In the halothane group, five children had nodal rhythm, one had irregular sinus rhythm, one had a transient sinus arrest, that is, a missing P wave with a delayed QRS by 0.4 s, and three had ventricular extrasystoles. One of the children in the halothane group had severe ventricular extrasystoles on traction of the ocular muscles and was switched to isoflurane for the maintenance anesthetic.

Table 2

Table 2

The ORR and respiratory parameters are summarized in Table 3. The baseline RR was higher in the sevoflurane group (S 38.2 ± 6, H 31.7 ± 8.2;P = 0.005). The baseline VT was similar in both groups, as was the baseline PETCO2. Significant changes in VT accompanied the OCR in both groups (P < 0.01). There were no significant differences in VT and PETCO2 changes between groups. During the OCR, the RR remained higher in the sevoflurane group than in the halothane group (P = 0.016). Overall there was a significant decline in the RR from baseline values (P = 0.03).

Table 3

Table 3

Three children had evidence of hypoventilation after LMA insertion (S 2, H 1) on high end-tidal vapor concentrations, and ventilation was briefly assisted until spontaneous respiration returned. Another 11 children, most of whom had received halothane (S 3, H 8), required an increase in FIO2 and/or assisted ventilation to correct hypoventilation during maintenance anesthesia. These children included three, all receiving halothane, who had irritable airways with laryngospasm and coughing, four children who had SpO2 values < 95% (S 2, H 2), and four who had PETCO2 > 60 mm Hg (S 1, H 3). One child receiving halothane had a persistent irritable airway and could not be settled, even after deepening the level of anesthesia, and was switched to IV propofol for maintenance. This child was excluded from the analysis of the OCR and ORR, but included in the results on respiratory complications.

Back to Top | Article Outline


In spontaneously breathing children undergoing strabismus correction, the OCR occurred significantly less often with 1.3 MAC sevoflurane in N2O than with halothane. Baseline HR was higher in children receiving sevoflurane. These findings have also been observed in similarly aged children during manual ventilation with 1 and 2 MAC sevoflurane and halothane in 100% oxygen (6). In that study, the authors postulated that the higher HR with sevoflurane was possibly caused by the differing effects of sevoflurane and halothane on the baroreflex. The vagal nerve is involved not only in the baroreflex but also in the OCR. Because we found that the OCR occurred less often in the sevoflurane group, a comparatively greater depression of the vagal activity by sevoflurane could theoretically lead to a less pronounced bradycardia on stimulation of the OCR.

We also observed considerably fewer dysrhythmias in children receiving sevoflurane than in those receiving halothane. This has also been observed in other studies in children (7) and may be attributed to the lack of effect of sevoflurane on myocardial conduction as compared with halothane (11).

We found that at 1.3 MAC sevoflurane or halothane the RR was higher with sevoflurane. We also observed higher minute volumes with sevoflurane, but the PETCO2 was similar to that of halothane. One possible explanation for this apparent discrepancy is a lower cardiac output with halothane when compared with an equivalent MAC of sevoflurane (12). Data from the literature on sevoflurane are inconclusive regarding whether the RR increases or decreases with increasing MAC. Doi et al. (13) and Komatsu et al. (14) showed a dose-dependent tachypnea. In contrast, Yamakage et al. (15) found a progressive decrease in RR, although in that study, halothane interestingly produced no change in RR. Johannesson et al. (16) found that the RR and PETCO2 were similar with both anesthetics during surgery. These findings may be explained by their definition of equipotency, which was based on clinical signs, and the fact that the authors used larger halothane concentrations than in our study. Brown et al. (5), in a group of younger children aged 6–24 months, also found no difference in VT or PETCO2 between sevoflurane and halothane in 60% N2O, but observed a higher RR with halothane. As higher halothane MAC values were used (S 1.4 MAC, H 1.8 MAC), and halothane is known to cause a dose-related tachypnea (17), this might explain the difference in RR observed by the authors.

Three children in our study, all anesthetized with halothane, had irritable airways and required assisted ventilation. Attempts to settle the children by deepening the level of anesthesia only resulted in apnea, suggesting that these respiratory problems were not caused by inadequate levels of anesthesia.

In our study, during positive OCRs, a decrease in VT and PETCO2 occurred, and this was similar in both groups. The observed decrease in minute ventilation during the OCR was not accompanied by an increase in PETCO2 but by a transient decrease. This may have been caused by the decrease in cardiac output during the OCR bradycardia.

A proportion of children had evidence of hypoventilation with spontaneous respiration through a LMA. Three had problems after LMA insertion, presumably because of high vapor concentrations, and this may be considered acceptable as a temporary consequence of this technique. However, 22% of the total number of children, of whom most had received halothane, required an increase in FIO2 and/or assisted ventilation to maintain the criteria of adequate ventilation during maintenance.

The crucial issue in this study is whether the compared alveolar vapor concentrations were equipotent. Various authors have commented on this problem (5).

The principle of overpressure induction consists of the creation of the steepest possible gradient between inspired and alveolar vapor concentrations to hasten the achievement of a specified endpoint. The endpoint chosen determines the MAC reached and will be similar to the accepted MAC for that endpoint. The inspired vapor must then be immediately reduced to prevent overdosing. If this is not done, this technique may create potentially dangerous large tissue vapor concentrations, similar to the circumstances seen with intubation under deep volatile anesthesia.

The endpoint in this study was readiness for LMA insertion which was between two and three minutes after beginning the induction for both anesthetics. Therefore, both were equipotent at the time that the vaporizer was switched off and the mask removed for LMA insertion.

We chose to use the overpressure technique with sevoflurane and halothane as it had been shown to be associated with minor complications (18). Interestingly, in that study, the authors were of the opinion that the lack of respiratory problems was the result of the use of preoperative oral atropine. We found that there were more respiratory complications with halothane than with sevoflurane, and this has also been observed in two other studies in children—one with incremental vapor induction (19) and one on ease of LMA insertion (20).

As halothane has a slower uptake and washout than sevoflurane, the time to the achievement of maintenance alveolar concentrations may have differed and have led to discrepancies. An average of 10 minutes passed from the time that the vaporizer concentration was reduced to the time surgery began. During this time, the remaining monitors were applied and the operative field was prepared. End tidal and inspired vapor concentrations had then reached equilibrium. The OCR and ORR were determined only after surgery started.

In determining equipotent maintenance concentrations, the term “MAC multiple” has been used, although it is not known what distortions arise from this concept. Moreover, the use of N2O creates further difficulties in calculating equipotency. When N2O and halothane are used together in children, it has been stated that MAC multiples of halothane and N2O can be simply added when the total MAC is 1 (9). From the only two existing studies in children on this subject (8,21), the evidence suggests that MAC multiples of sevoflurane and N2O are less than purely additive, and the hypothesis is that they act at a common site of action or mechanism. Furthermore, the amount of the MAC-reducing effect of N2O on sevoflurane is different for MAC skin incision, as it is for MAC tracheal intubation. The MAC multiples we used were 1.21 for the sevoflurane group and 1.24 for the halothane group. Considering the many points where inaccuracies can enter in the concept of MAC multiples, we feel that the comparison of the two anesthetics in our study was acceptable.

In conclusion, children undergoing outpatient strabismus surgery with spontaneous breathing must be intensively monitored for bradycardia and hypoventilation. We found that sevoflurane was associated with a lower incidence of the OCR, airway irritability, and fewer ventilatory interventions compared with halothane. Sevoflurane may be the better choice of inhaled anesthetic for this procedure.

Back to Top | Article Outline


1. Barry Smith G, Hamilton RC, Carr CA. Ophthalmic anaesthesia. 2nd ed. London: Arnold, 1996.
2. Dell R, Williams B. Anaesthesia for strabismus surgery: a regional survey. Br J Anaesth 1999; 82:761–3.
3. Loewinger J, Friedmann-Neiger I, Cohen M, Levi E. Effects of atracurium and pancuronium on the oculocardiac reflex in children. Anesth Analg 1991; 73:25–8.
4. Blanc VF, Jacob JL, Milot J, Cyrenne L. The oculorespiratory reflex revisited. Can J Anaesth 1988; 35:468–72.
5. Brown K, Aun C, Stocks J, et al. A comparison of the respiratory effects of sevoflurane and halothane in infants and young children. Anesthesiology 1998; 89:86–92.
6. Kawana S, Wachi J, Nakayama M, Namiki A. Comparison of haemodynamic changes induced by sevoflurane and halothane in paediatric patients. Can J Anaesth 1995; 42:603–7.
7. Paris ST, Cafferkey M, Tarling M, et al. Comparison of sevoflurane and halothane for outpatient dental anaesthesia in children. Br J Anaesth 1997; 79:280–4.
8. Lerman J, Sikich N, Kleinman S, Yentis S. The pharmacology of sevoflurane in infants and children. Anesthesiology 1994; 80:814–24.
9. Murray DJ, Mehta MP, Forbes RB, Dull DL. Additive contribution of nitrous oxide to halothane MAC in infants and children. Anesth Analg 1990; 71:120–4.
10. Gregory GA, Eger EI II, Munson ES. The relationship between age and halothane requirement in man. Anesthesiology 1969; 30:488–91.
11. Nakaigawa Y, Akazawa S, Shimizu R, et al. Comparison of the effects of halothane, isoflurane, and sevoflurane on atrioventricular conduction times in pentobarbital-anesthetized dogs. Anesth Analg 1995; 81:249–53.
12. Holzman RS, van der Velde ME, Kaus SJ, et al. Sevoflurane depresses myocardial contractility less than halothane during induction of anesthesia in children. Anesthesiology 1996; 85:1260–7.
13. Doi M, Ikeda K. Respiratory effects of sevoflurane. Anesth Analg 1987; 66:241–4.
14. Komatsu H, Chujo K, Morita J, et al. Spontaneous breathing with the use of a laryngeal mask airway in children: comparison of sevoflurane and isoflurane. Paediatr Anaesth 1997; 7:111–5.
15. Yamakage M, Tamiya K. Horikawa D, et al. Effects of halothane and sevoflurane on the paediatric respiratory pattern. Paediatr Anaesth 1994; 4:53–6.
16. Johannesson GP, Floren M, Lindahl SGE. Sevoflurane for ENT-surgery in children. Acta Anaesthesiol Scand 1995; 39:546–50.
17. Murat I, Chaussain M, Hamza J, Saint-Maurice C. The respiratory effects of isoflurane, enflurane and halothane in spontaneously breathing children. Anaesthesia 1987; 42:711–8.
18. Sigston PE, Jenkins AMC, Jackson EA, et al. Rapid inhalational induction in children: 8% sevoflurane compared with 5% halothane. Br J Anaesth 1997; 78:362–5.
19. Walker SM, Haugen RD, Richards A. A comparison of sevoflurane with halothane for paediatric day case surgery. Anaesth Intensive Care 1997; 25:643–9.
20. Ignagaki Y, Wat LI, Joshi MD, et al. Sevoflurane facilitates insertion of the laryngeal mask airway in children: a comparison with halothane [abstract]. Anesth Analg 1997; 84:S434.
21. Swan H, Crawford MW, Ling Pua H, et al. Additive contribution of nitrous oxide to sevoflurane minimum alveolar concentration for tracheal intubation in children. Anesthesiology 1999; 91:667–71.
© 2000 International Anesthesia Research Society