A Comparison between the Effects of Propofol and Sevoflurane in Pediatric Strabismus Surgery on the Quality and Depth of Anesthesia

Abstract

Background:

Strabismus surgery may be associated with several undesirable complications as increased incidence of the oculocardiac reflex (OCR), hemodynamic changes, emergency agitation (EA), postoperative pain, nausea, and vomiting. Previous studies suggested that deeper anesthesia monitored by bispectral index (BIS) protects against OCR. This study aims to evaluate the effect of the type of anesthesia on the quality of anesthesia in pediatric patients.

Patients and Methods:

One hundred American Society of Anesthesiologists physical status classes I and II pediatric patients, aged between 3 and 6 years old of both genders, who were subjected to strabismus surgery under general anesthesia were enrolled in this study. Patients were randomized into two equal groups (each = 50); in the first group, anesthesia was induced and maintained with sevoflurane (Group S), and in the second group, anesthesia was induced and maintained with propofol (Group P). Hemodynamics and BIS were monitored, and OCR and the need for atropine were recorded. Furthermore, EA using the Cravero scale was recorded.

Results:

The propofol group showed a higher incidence of OCR while the sevoflurane group had a higher incidence of postoperative agitation, pain, nausea, and vomiting, without statistically significant differences regarding hemodynamics.

Conclusion:

Although sevoflurane anesthesia may be superior to propofol in ameliorating OCR, it has been associated with an increased incidence of postoperative complications.

INTRODUCTION

Strabismus surgery is one of the most common pediatric surgeries. However, it may be associated with several undesirable intraoperative and postoperative complications.[1] Increased incidence of oculocardiac reflex (OCR) is one of the most important complications during strabismus surgery, and it is usually induced by manipulation of the eye or extraocular muscles (EOMs); therefore, anesthesiologists and ophthalmic surgeons make attempts to protect patients from OCR. Other side effects include hemodynamic changes induced by OCR, postoperative nausea and vomiting (PONV), emergency agitation (EA), and postoperative pain. These effects are mainly related to anesthesia.[2]

The choice of the proper anesthetic agent is the main interest for anesthesiologists trying to protect their patients from these unwanted side effects. Sevoflurane is an inhalational anesthetic agent frequently used in pediatric surgery. It has the advantages of minimal airway irritation, rapid induction, and recovery from anesthesia due to low blood-gas partition coefficient. Unfortunately, EA following general anesthesia in children is commonly reported after using sevoflurane.[3] Propofol is an intravenous (i.v.) agent, causes sedation, hypnosis, and amnesia, and has anti-emetic properties. It is characterized by rapid onset with smooth recovery and dose-dependent effect. It acts by inhibiting N-methyl-d-aspartate receptors in the hippocampal region.[4]

The OCR is more frequent in pediatric patients and specifically in those subjected to surgery on the medial rectus muscle; however, deep levels of anesthesia decrease its incidence in pediatric patients.[5]

Several studies have found that bispectral index (BIS) may be a good tool for measuring the depth of anesthesia and have validated its utility for this use. It provides earlier awakening and better recovery profiles at the end of anesthesia.[6]

The exact definition of OCR, its incidence, and the risk factors for its development is not yet clearly described. Moreover, few studies have focused on OCR during strabismus surgery in relation to the quality of anesthesia, not only the depth, especially in terms of the anesthetic agent used.[7,8]

This study aimed to compare the quality of anesthesia by propofol versus sevoflurane in pediatric patients undergoing strabismus surgery. Points of assessment included depth of anesthesia as measured by BIS, incidence of OCR, and other postoperative complications associated with these two anesthetic agents.

PATIENTS AND METHODS

This prospective randomized study was designed to include 100 pediatric patients with the American Society of Anesthesiologists physical status classes I and II, aged between 3 and 6 years old of both genders, who were scheduled for elective strabismus surgery under general anesthesia. The study was conducted in Mansoura Ophthalmology Center in the interval between November 2019 and August 2020. Approval of the Institutional Board Review with number R/20.01.732 and clinical trial registry (NCT04485117) was obtained. Furthermore, written informed consent was obtained from parents of all children who participated in the study after ensuring confidentiality.

The following patients were excluded from the study: children with any developmental delays, mental or neurological disorders, hyperactive airway diseases, hemostasis disorders, previous hypersensitivity to the used drugs, any ocular pathology other than strabismus, in addition to parental refusal of consent.

Sample size calculation

A priori G-power analysis (program version 3, Erdfelder E, Lang AG 2007, Universitat Dusseldorf) was done to estimate study sample size. Assuming α1 error = 0.05 and β2 error = 0.02 (power = 80%), 47 patients in each group were enough to detect a difference of 30% in the occurrence of OCR by BIS between the two groups. A dropout of 5% of cases was expected. Therefore, 50 patients were needed in each group to find out this difference.

All patients were subjected to preoperative assessment (history taking, clinical evaluation, and routine laboratory investigations). The day before the surgery, the study protocol was discussed with parents of the children who participated in the study, and they were instructed to keep them fasting for 8 h before the surgery. No premedication drug was used.

Patients were randomly allocated into two equal groups (each n = 50): sevoflurane group (Group S) and propofol group (Group P) using a permuted block randomization method to set a table of random numbers. Closed opaque envelopes containing group allocation were opened only after getting the written informed consent.

On arrival to operating area, peripheral i.v. cannula was inserted. Basic monitoring was applied (electrocardiogram, noninvasive blood pressure, and pulse oximeter and capnography) using Datex-Ohmeda Cardiocap (Helsinki, Finland) monitoring system, in addition to the BIS sensor electrodes which were applied to the patient forehead. Anesthesia was induced using face mask with 8% sevoflurane in 40% O₂ in Group S and by giving i.v. propofol (2 mg.kg⁻¹) in Group P. Laryngeal mask airway (LMA) was used to maintain airway after adequate jaw relaxation, and capnography was connected to it. Then, sevoflurane concentration was decreased to 2%–3% in 40% O₂ throughout the operation for maintenance of anesthesia in the sevoflurane group (Group S). Anesthesia was maintained in the propofol group (Group P) with propofol infusion (10–15 mg.kg⁻¹.h⁻¹) as titrated by the anesthesiologist. Analgesia for all children was achieved by i.v. acetaminophen (10 mg.kg⁻¹). Ventilation was maintained to keep end-tidal CO₂ (ETCO₂) within the range of 35 and 40 mmHg.

At the end of the operation, the used anesthetic drugs were stopped and LMA was removed while the patient was still deeply anesthetized. Face mask with 100% O₂ was applied with careful observation for any upper airway obstruction, laryngospasm, or breath holding. Then, children were transferred to the postanesthesia care unit (PACU) with their parents.

Collected data

All patients were monitored for hemodynamic variables (heart rate [HR] and mean arterial blood pressure [MAP]). These variables were recorded at 5-min intervals during the entire procedure.

BIS values were recorded using Covidien monitor (Medtronic UK) at induction of anesthesia and every 5 min till the end of the surgery. The occurrence of OCR and the need for atropine were recorded. The level of BIS accompanied by occurrence of OCR during surgery was recorded. During the procedure, if there was dysrhythmia or rapid decrease in HR by ≥25% from the baseline, it was considered as an OCR and was managed by asking the surgeon to release traction on muscles, but if this was ineffective or HR decreased to <60 beats/min, i.v. atropine (0.01 mg.kg⁻¹) was administered.

EA was evaluated using the Cravero scale [Table 1] to be recorded every 5 min from awakening for 30 min. If the score was ≥4 for more than 5 min, the child was considered agitated and i.v. propofol (1 mg.kg⁻¹) was used as rescue medication.[9] Emergence time was measured from the discontinuation of anesthesia to the first response to verbal command.

Table 1.

Cravero scale

Behavior	Score
Obtunded with no response to stimulation	1
Asleep but responsive to movement or stimulation	2
Awake and responsive	3
Crying (for 3 min)	4
Thrashing behavior that requires restraint	5

Postoperative pain was assessed by Face, Legs, Activity, Cry, and Consolability (FLACC) Pain Score in a scope of 0–10 (0 = no pain and 10 = worst pain).[10] It was assessed upon arrival to PACU, 1, 2, 4, 6, 12, and 24 h after surgery. Acetaminophen i.v. in a dose (10 mg.kg⁻¹⁾ was given as a rescue analgesia if the FLACC score is ≥4 or upon the patient request. The time of the first request for analgesia and the total analgesic requirements during the first 24 h postoperatively were recorded.

The incidence of PONV was recorded during the first 24 h postoperatively. It was treated with i.v. ondansetron (0.1 mg.kg⁻¹). The total dose of anti-emetic drug and the number of patients who required anti-emetic therapy were also recorded.

Statistical methods

Collected data were coded, processed, and analyzed using SPSS program (version 22) IBM company, Mansoura, Dakahlia, Egypt for Windows. Kolmogorov–Smirnov test was used for testing the normality of numerical data distribution. Unpaired Student's t-test was applied for testing continuous data of normal distribution which were presented as mean ± standard deviation. Mann–Whitney U-test was used for testing data of nonnormal distribution which were presented as median (range). Chi-square test was used for testing categorical data which were presented as number (percentage). When P ≤ 0.05, it was considered statistically significant.

RESULTS

One hundred and twenty-eight patients were recruited and only one hundred patients were enrolled in this study, as shown in the flowchart [Figure 1]. Patients in the two groups were comparable with respect to age, weight, gender, and durations of surgery and anesthesia [Table 2].

Figure 1.

Flowchart of the study

Table 2.

Patient demographics

	Group S (n=50)	Group P (n=50)	P
Age (years)	4.42±1.12	4.60±1.06	0.41
Weight (kg)	13.32±3.32	14.04±3.01	0.27
Sex, n (%)
Male	25 (50)	25 (50)	0.42
Female	29 (58)	21 (42)
Duration of anesthesia (min)	74.18±13.71	76.62±13.59	0.94
Duration of surgery (min)	62.90±15.15	62.60±14.68	0.50

Data are presented as mean±SD or n (%). SD=Standard deviation

In the propofol group (Group P), there was a significant increase in the incidence of OCR compared to the sevoflurane group (Group S). Comparing both the groups, not only the number of patients experiencing OCR was significantly higher in Group P than Group S (20 [40%] versus 36 [72%]), but also the number of episodes (mean 1, range 0–2 versus mean 2, range 0–4). Moreover, the number of patients requiring atropine was significantly higher in Group P than Group S (10 patients vs. 21 patients, respectively). The mean BIS reading of OCR incidence was 57.93 ± 2.66 in Group P compared to 62.53 ± 2.84 in Group S with a significant P value (P 0.001) [Table 3].

Table 3.

Incidence of oculocardiac reflex, number of episodes, number of patients required atropine, and mean bispectral index of oculocardiac reflex incidence

Variables	Group S (n=50), n (%)	Group P (n=50), n (%)	P
No of patients with OCR	20 (40)	36 (72)	0.001*
Number of episodes of OCR	1 (0-2)	2 (0-4)	0.001*
Number of patients required atropine	10 (20)	21 (42)	0.01*
Mean BIS of OCR incidence	57.93±2.66	62.53±2.84	0.001*

Data are expressed as mean±SD, n (%), or median (range). *P≤0.05 was considered statistically significant. OCR=Oculocardiac reflex, BIS=Bispectral index, SD=Standard deviation

In the PACU, there was a statistically significant increase in the number of patients with high EA scores (scores 4 and 5) than in Group S (P = 0.001). Emergence times were significantly longer in Group S (9.6 ± 1.17) as compared to Group P 7.08±0.0.8 (P < 0.05) [Table 4].

Table 4.

Emergence agitation scale and emergence time

EA score	Group S (n=50), n (%)	Group P (n=50), n (%)	P
Incidence of score (4)	28 (56)	12 (24)	0.001*
Incidence of score (5)	10 (20)	5 (10)	0.001*
Emergence time	9.6±1.17	7.08±0.0.8	0.001*

Data are expressed as mean±SD or n (%.). *P≤0.05 was considered statistically significant. SD=Standard deviation, EA=Emergency agitation

Considering the FLACC score, it was significantly lower in Group P in the first 4 postoperative hours. The time to first analgesic request was significantly longer in Group P than in Group S (71.28 ± 21.73 vs. 95.81 ± 37.47, respectively). Moreover, patients in Group P experienced less total analgesic requirements as compared to Group S (500 ± 139.14 vs. 406.50 ± 141.79) (P < 0.05) [Table 5].

Table 5.

Face, Legs, Activity, Cry, and Consolability Scale and time to first analgesic request and total analgesic requirements

FLACC score	Group S (n=50)	Group P (n=50)	P
After surgery	5 (2-9)	3 (0-9)	0.001*
1	5 (1-9)	2 (0-6)	0.001*
2	5 (2-9)	4 (0-8)	0.01*
4	4 (1-7)	3 (0-6)	0.03*
6	1 (0-4)	1 (0-3)	0.44
12	2 (0-4)	1 (0-3)	0.06
24	1 (0-2)	1 (0-2)	0.68
Time to first analgesic request (min)	71.28±21.73	95.81±37.47	0.003*
Total analgesic requirements (mg)	500±139.14	406.50±141.79	0.003*

Data are expressed as mean±SD, n (%), or median (range). *P≤0.05 was considered statistically significant. FLACC=Face, Legs, Activity, Cry, and Consolability, SD=Standard deviation

Regarding PONV, there were a statistically significant less number of patients requiring anti-emetics and less total anti-emetic requirements in Group P as compared to Group S (P < 0.05) [Table 6].

Table 6.

Incidence of postoperative nausea and vomiting and number of patients who required anti-emetics and total anti-emetic requirements

PONV	Group S (n=50), n (%)	Group P (n=50), n (%)	P
Incidence of PONV	34 (68)	14 (28)	0.001*
Number of patients who required anti-emetics	21 (42)	6 (12)	0.001*
Total anti-emetic requirements (mg)	1.29±32	0.9±0.45	0.01*

Data are expressed as mean±SD, or n (%.). *P≤0.05 was considered statistically significant. PONV=Postoperative nausea and vomiting, SD=Standard deviation

No statistically significant differences were detected between both the groups as regards HR and MAP (P > 0.05) [Figures 2 and 3].

Figure 2.

Intraoperative heart rate mean values (beat/min)

Figure 3.

Intraoperative mean arterial blood pressure mean values (mmHg)

DISCUSSION

The main finding of this study is that sevoflurane was more successful than propofol in minimizing the incidence of OCR. In the current study population of pediatric patients undergoing strabismus surgery, the depth of anesthesia was observed to be a major factor in the incidence of OCR, as BIS values <50 were linked with decreased incidence of OCR.

The OCR is more common in children than adults as they are characterized by increased vagal tone which made them more liable to the OCR during strabismus surgery. In the present study, patients in the propofol group developed more evident OCR. This may be referred to the fact that propofol has the ability to increase the incidence of bradycardia by a central sympatholytic effect and vagal stimulation.[11]

Sevoflurane inhibits the vagal activity leading to less bradycardia which may help in lowering the incidence of OCR.[12] Nevertheless, this could be in contrast with the studies demonstrating that sevoflurane has little or no impact on cardiac parasympathetic tone.[13]

This protective effect of sevoflurane may be also explained by its effect on the cortical and subcortical regions in the brain. Increased anesthetic concentration can suppress the nociceptive and autonomic reflexes.[14]

The current study is consistent with several studies that reported a close association between OCR and depth of anesthesia monitored by BIS. One study involved three groups of children to achieve three BIS levels by varying sevoflurane concentration. The incidence of OCR was 11% for the deep anesthesia group (40–49), 32% for the mid-anesthesia group (50–59), and 71% for the light anesthesia group (60–69).[7] Consequently, the other two studies investigated the relation between BIS and OCR, concluding that less OCR was observed in the study group with lower BIS values.[8,15]

On the other hand, the finding of a prospective cohort study on OCR from 2009 to 2013, in which anesthetic depth was evaluated using either BIS or Narcotrend monitors, reported that confirmation of direct correlation between brain wave monitors and OCR when using multiple anesthetic agents is difficult. They found also that no level of BIS can predict the absence of OCR as seen when profound OCR occurs despite deep BIS readings. This conflict with other studies can be attributed to the fact that there are multiple factors that can influence the incidence of OCR. These factors include the type of general anesthetic agent, the type of surgery including tension on EOMs, and the type of the EOM.[16] Furthermore, the power of traction on EOM is a great determinant of OCR. Hypercarbia and hypoxemia are added factors that can cause OCR.[17] Hence, the depth of anesthesia is not the only factor affecting the incidence of OCR.

Another finding of the current study was a lower incidence of EA with a shorter emergence time after propofol as compared to sevoflurane anesthesia. It may be explained by the epileptiform EEG changes and temporary neurological dysfunction associated with sevoflurane anesthesia.[18,19] Propofol has been documented to be more efficacious than sevoflurane in reducing EA.[20,21]

This was in agreement with a study that found greater scores of EA in children anesthetized with sevoflurane for MRI imaging.[22] An incidence of 42% for EA after sevoflurane anesthesia and 5% after propofol anesthesia has been reported in preschool-aged children.[23] Nevertheless, extubation and recovery times were similar between the sevoflurane and propofol groups, but EA was more common with sevoflurane anesthesia.[24]

Multiple studies have revealed that propofol infusion for anesthesia maintenance has unpredictable effects on the incidence of EA; some studies demonstrated no significant effect on EA in comparison with volatile anesthesia,[25,26] while others have reported an increased EA with propofol.[27] The variations in the incidences of EA in different studies may be referred to the type of surgery, the anesthetic technique, operation time, age group, difference in the assessment scales, and use of poorly validated EA assessment tools.[25,26,27]

Additional differences were observed between the present two-study arms; a higher FLACC score was found in the sevoflurane group as compared to the propofol group, with a greater effect on the intensity of postoperative pain for 4 h. Furthermore, propofol prolonged the time for the first request analgesic request with decreased total analgesic requirements. This may be attributed to the analgesic effect of propofol that reduced EA through facilitation of a smoother emergence. The behavior categories of the EA score overlap with those of pain scales.[28]

In accordance with the current results, multiple studies stated that propofol anesthesia in different surgeries decreased postoperative pain and the need to rescue analgesia.[25,29] Its use has decreased postoperative pain as measured by FLACC scales in children aged between 2 and 6 years who underwent strabismus surgery. Otherwise, higher analgesic consumption in PACU was noted in the sevoflurane group.[29] As well, they found a significant positive correlation between FLACC and EA scores, with lower incidences of EA and postoperative pain in the propofol group.[25]

In the present study, the incidence of PONV and the need for administration of anti-emetics were higher in the sevoflurane group. In a similar study, propofol-based anesthesia should still be preferred to inhalational agents in ophthalmic surgery due to high patient satisfaction and less frequent incidence of PONV.[30]

Indeed, PONV is not only influenced by the regimen of general anesthesia, but rather by many other factors, in particular the type of operation. Pediatric strabismus surgery increases both OCR and PONV, with a great association between them.[31] In agreement with results of the present study, it is documented that the patients who received large concentrations of sevoflurane, resulting in BIS <50 had an increased risk of PONV.[32]

CONCLUSION

BIS monitoring can be used as a valuable tool to evaluate pharmacodynamic effects of propofol and sevoflurane anesthesia facilitating their titration to improve overall anesthetic outcomes. Furthermore, titration of anesthetic agents guided by BIS may be advantageous to optimize the depth of anesthesia in order to increase safety by influencing OCR during pediatric strabismus surgery. Although sevoflurane anesthesia may be superior to propofol in ameliorating OCR, it has been associated with an increased incidence of postoperative agitation, pain, nausea, and vomiting.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.