Oral melatonin as part of multimodal anxiolysis decreases emergence delirium in children whereas midazolam does not: A randomised, double-blind, placebo-controlled study : European Journal of Anaesthesiology | EJA

Secondary Logo

Journal Logo


Oral melatonin as part of multimodal anxiolysis decreases emergence delirium in children whereas midazolam does not

A randomised, double-blind, placebo-controlled study

Singla, Lily; Mathew, Preethy J.; Jain, Aditi; Yaddanapudi, Sandhya; Peters, Nitin J.

Author Information
European Journal of Anaesthesiology: November 2021 - Volume 38 - Issue 11 - p 1130-1137
doi: 10.1097/EJA.0000000000001561
  • Free



Emergence delirium whereby the patient is irritable, uncooperative and inconsolably crying, kicking or thrashing in the postoperative period is a significant problem in paediatric anaesthesia. The situation is distressing to the nursing staff as well as parents, with a reported incidence of 20 to 80%.1 Since this phenomenon is attributed to hypnotics with a rapid emergence, the principal preventive strategy is administration of sedative medications either before the procedure or intravenously during anaesthesia to smooth the emergence.1

An important risk factor for emergence delirium is pre-operative anxiety.1,2 The current approach to allay pre-op anxiety is multimodal, consisting of a combination of family-centred nonpharmacological measures employing various distraction techniques, and sedative premedication. Some drugs administered for anxiolytic premedication also prevent emergence delirium, for example midazolam, alpha-2 adrenergic agonists, melatonin and so on. Oral midazolam, widely used for pharmacological premedication,3 was found effective to prevent emergence delirium in a meta-analysis,4 but recent studies have refuted this.5,6 Midazolam has the disadvantage of persistent postoperative sedation and impairment of cognitive and psychomotor performance.7 On the contrary, melatonin has an appealing safety profile,8 and is as efficacious as midazolam for premedication.7 A recent meta-analysis on the use of melatonin to prevent emergence delirium reported a potential benefit, but the quality of evidence was deemed low due to the small amount of information available.9

It is not clear if the potential benefit of pre-operative oral drugs in preventing emergence delirium results from anxiolysis or residual sedative effects delaying emergence. There are very few studies that have evaluated emergence delirium after multimodal anxiolytic strategy with and without sedatives. Thus, we designed a randomised, blinded, placebo-controlled study to assess the incidence of emergence delirium after premedication with either 0.3 mg kg−1 of melatonin or 0.3 mg kg−1 of midazolam in the intervention arms and honey in the placebo arm. In addition, all groups received multimodal anxiolysis comprising of standardised nonpharmacological measures. Secondary objectives were to compare the incidence of pre-operative anxiety and compliance with inhalational induction in the three groups.

Materials and methods

Study design and participants

A prospective, randomised, placebo-controlled, single-centre study was undertaken wherein the patients, attending anaesthesiologists and outcome assessors were blinded to study outcomes. Institutional Ethics Committee (Intramural) of Postgraduate Institute of Medical Education and Research, Chandigarh, India (Chairperson Prof K.L. Gupta) approved the study vide IEC/2019/000591 on 19 March 2019 and the protocol was registered prospectively in the Clinical Trial Registry of India (CTRI/2019/06/019850).

Children aged 3 to 8 years with American Society of Anaesthesiologists Physical Status 1 or 2 and scheduled for elective ambulatory procedure under general anaesthesia were enrolled. Children scheduled for emergency procedures, those with neurological disease, neurodevelopmental anomalies, mental retardation or hearing impairment were excluded. Written informed consent was obtained from the parent or legal guardian of the child.

Randomisation, allocation concealment and blinding

The individuals were enrolled and randomised into three groups based on a computer-generated block randomisation process with the allocation sequence concealed in sequentially numbered, opaque, sealed envelopes. The children received either oral melatonin 0.3 mg kg−1 or oral midazolam 0.3 mg kg−1 or placebo. Oral melatonin was available as a pale brown viscous syrup (6 mg per 5 ml) similar in colour and consistency to honey. Oral midazolam was reconstituted by mixing the calculated dose of intravenous midazolam in 3 to 5 ml of plain honey. Plain honey (3 to5 ml) was used as placebo. The calculated dose was prepared and administered 45 min prior to anaesthesia induction in a 5 ml syringe by an anaesthesia resident trainee not involved in the conduct of the anaesthesia or the assessment of study outcomes. This ensured blinding of children, their families, the anaesthesia team and the investigating team to the study drug. The allocation sequence was de-coded only after completion of the study.


All parents were instructed about the pre-operative fasting protocol as per the institutional policy of 6 h fasting for solids and 2 h for clear liquids before the procedure. On the day of procedure, a standardised bundle of at least three simple nonpharmacological measures commonly used in our set-up was employed: familiarisation with the anaesthesia breathing circuit and play therapy to blow up the reservoir bag like a balloon; using a flavour of the child's choice to smear on the mask during induction; and parental presence at induction. The parent or guardian who would accompany the child for induction was counselled regarding their role during mask induction of anaesthesia. In addition, the attending anaesthesiologist was free to choose other age-appropriate nonpharmacological interventions such as a pacifier, video cartoons and so on as required to minimise a child's anxiety.

The assigned drug for premedication was administered 45 min before the procedure. During mask induction, parental presence in the operating room was ensured in all groups. A pulse oximeter probe was placed on the child's finger following which 70% nitrous oxide in oxygen was delivered through the flavoured facemask while distracting the child with the balloon (reservoir bag) being blown. After 1 to 2 min, sevoflurane was introduced in a concentration of 0.5 to 1% and increased every three to four breaths to a maximum of 6%. After induction, an intravenous cannula was inserted and fentanyl 1 μg kg−1 was administered. Anaesthesia was maintained with sevoflurane in a mixture of oxygen and nitrous oxide. An airway, either supraglottic, or endotracheal tube, was placed by the attending anaesthesiologist as required, or anaesthesia was continued through the facemask. The need for skeletal muscle relaxation was decided by the attending anaesthesiologist depending on the procedure. After placement of the chosen airway device, caudal or regional block was given as appropriate. Adequate analgesia was ensured during the procedure with additional boluses of fentanyl if required. At the end of the procedure, the airway device was removed after the return of consciousness and muscle power. Emergence time was defined as the interval from discontinuation of sevoflurane to spontaneous eye opening. Postoperatively, a rescue fentanyl bolus was administered by the attending anaesthesiologist to the child if the patient complained of pain on verbal query after waking up.

Data collection and outcome variables

The baseline data were obtained for every patient in the pre-operative ward. Temperament, as reported by the primary caregiver, was used to assess the child's temperament.10 The pre-operative anxiety behaviour was assessed at two time points, in the ward before administration of the study drug and in the operating room before mask induction, using the modified Yale Preoperative Anxiety scale (mYPAS).11 Two minutes of the child's interactions at both these times were recorded by video: these were assessed later and a mYPAS assigned for each. The alertness level of the child during transfer into the operating room, sleeping, drowsy, awake or crying was recorded. The compliance to mask induction was assessed using the Induction Compliance checklist (ICC).12

The primary outcome of the study was the incidence of emergence delirium reported during the first 30 min in the postanaesthesia care unit (PACU). Emergence delirium was assessed using the four-point Watcha scale (1 = calm, quiet; 2 = crying, but can be consoled; 3 = crying, cannot be consoled; and 4 = agitated and thrashing around) soon after emergence from anaesthesia in the operating room and later in PACU at 5,10,15, 20 and 30 min after arrival.13 A score of at least 3 on the Watcha scale was defined as emergence delirium. Sedation was also assessed in PACU every 30 min for 2 h using the Children's Hospital of Wisconsin Sedation Scale.14 All outcomes were assessed by a blinded observer (LS or AJ) not involved in the care of the patient.

Statistical analysis

Sample size calculation

The incidence of emergence delirium in an ambulatory setting without oral sedative premedication in a similar age group was reported to be 32% while using sevoflurane anesthesia.15 We deemed a 25% decrease in this incidence to be clinically significant. To detect this effect size with an alpha error of 0.05 and power of 85%, a sample size of 41 children per group was calculated. Adjusting for a possible dropout of up to 10%, a total of 135 children were randomised into three groups. We formulated the null hypothesis that, as part of a multimodal pre-operative anxiety treatment before sevoflurane anaesthesia, the incidence of emergence delirium would be similar in the midazolam and melatonin groups when compared with the placebo group.

SPSS Version 22.0 was used to analyse the data. Continuous data are expressed as mean ± SD or median [IQR] based on the normality of distribution as checked by the Shapiro--Wilk test. Frequency and percentage were calculated for categorical data. The baseline data were compared using one-way analysis of variance.

The incidence of emergence delirium was analysed using the Chi-squares test and calculation of risk ratios. The secondary outcomes, anxiety behaviour recorded on mYPAS scale, and ICC were analysed using the Kruskal--Wallis test. Children with mYPAS at least 30 were identified as high pre-operative anxiety. Similarly, children with ICC = 0 were deemed to have perfect mask induction. The proportion of children who had high anxiety and perfect induction were analysed using the Chi-squared test. The Bonferroni correction was applied when multiple group testing was undertaken.


One hundred and thirty-five children were enrolled and randomised between July 2019 and January 2020. The data from 132 children were analysed (Fig. 1). The baseline characteristics of the children in the three groups were comparable, including their age, weight, sex, type of surgical procedure and temperament (Table 1). The various aspects of intra-operative management were similar in the three groups, duration of anaesthesia, intra-operative fentanyl usage, use of airway devices, muscle relaxants and regional anaesthesia as well the emergence time (Table 1).

Fig. 1:
CONSORT flow chart
Table 1 - Pre-operative and intra-operative characteristics
Melatonin (n = 45) Midazolam (n = 43) Placebo (n = 44) P
Age (years) 5.3 ± 1.5 5.1 ± 1.7 5.2 ± 1.9 0.73a
Weight (kg) 18.0 ± 6.2 17.0 ± 5.4 18.4 ± 6.9 0.54a
Sex (male: female) n (%) 31: 14 (69: 31) 30: 13 (70: 30) 32: 12 (73: 27) 0.91b
ASA Physical status- I 45 43 44 NA
Type of surgery 0.38c
 Urogenital 15 (33) 14 (33) 9 (20)
 Ophthalmology 28 (62) 24 (56) 29 (66)
 Superficial 2 (4) 5 (12) 6 (14)
Child's temperament
 Sociability 4.1 ± 0.8 4.0 ± 0.8 3.9 ± 0.9 0.80a
 Emotionality 4.0 ± 0.7 4.1 ± 0.7 4.0 ± 0.8 0.95a
 Energy 4.1 ± 0.8 4.2 ± 0.7 4.1 ± 0.7 0.71a
 Distractibility 4.1 ± 0.8 4.2 ± 0.6 3.9 ± 1.0 0.35a
 Rhythmicity 4.0 ± 1 4.2 ± 0.7 4.0 ± 1.0 0.45a
Duration of anaesthesia (min) 48.1 ± 44.1 45.4 ± 30.9 43.6 ± 26.5 0.83a
Intra-operative fentanyl use (μg kg−1) 1.0 ± 0.2 1.0 ± 0.1 1.0 ± 0.2 0.94a
Airway device 0.66c
 Supraglottic Device 27 (60) 28 (65) 25 (57) --
 Face Mask 14 (32) 10 (23) 12 (27) --
 Endotracheal tube 3 (7) 5 (12) 7 (16) --
Minimum Alveolar Concentration 1.2 ± 0.1 1.2 ± 0.0 1.1 ± 0.1 0.22a
Use of muscle relaxant 3 (7) 5 (12) 9 (21) 0.14c
Regional anaesthesia technique 0.37c
 Caudal block 10 (22) 14 (33) 8 (18) --
 Penile block 2 (4) 0 (0) 1 (2) --
Emergence time (min) 5.3 ± 1.8 6.2 ± 2.4 6.3 ± 2.7 0.11a
ASA, American Society of Anaesthesiologists. Emergence time was defined as time interval from discontinuation of sevoflurane to spontaneous eye opening. Statistical tests:
aAnalysis of variance.
bPearson Chi-square test.
cFisher's exact test.All numbers are stated in mean ± SD if not stated otherwise (number (% of patients)).

Primary outcome

The overall incidence of emergence delirium in PACU was significantly less in the melatonin group (27%) than in the midazolam (56%) or placebo groups (50%) (Table 2) Compared with placebo, the absolute risk of emergence delirium decreased significantly when premedicated with melatonin [absolute risk ratio = 23.3 (95% CI 3.7 to 42.9), P = 0.03], whereas it did not differ when premedicated with midazolam [absolute risk ratio = -5.8 (95% CI -26.8 to 15.1), P = 0.59 (Table 3)]. Melatonin significantly decreased the risk of emergence delirium compared with midazolam [absolute risk ratio = 29.2 (95% CI 9.5 to 48.8)].

Table 2 - Incidence of emergence delirium and fentanyl usage
Melatonin (n = 45) Midazolam (n = 43) Placebo (n = 44) P
Number of patients with ED in PACU 12 (27) 24 (56) 22 (50) 0.01b ,
ED at different time points
 In OR after emergence from anaesthesia 13 (29) 14 (33) 15 (31) 0.86b
 In PACU at 5 min 11 (24) 17 (40) 19 (43) 0.15b
 In PACU at 10 min 4 (9) 14 (33) 14 (32) 0.01b ,
 In PACU at 15 min 1 (2) 16 (37) 11 (25) <0.001c ,
 In PACU at 20 min 0 (0) 11 (26) 4 (9) <0.001c ,
 In PACU at 30 min 0 (0) 6 (14) 2 (5) 0.009c ,
Fentanyl use after emergence
 Number of children 4 (9) 7 (16) 7 (16) 0.52c
 Dose of fentanyl (μg kg−1) 0.3 (0.3 to 0.7) 0.3 (0.3 to 0.6) 0.5 (0.3 to 0.7) 0.59d
 Time interval to first fentanyl bolus (min) 3.2 ± 13.6 2.9 ± 7.3 4.6 ± 14.9 0.79a
Postoperative sedation scale in PACU
 At 30 min 4.8 ± 0.7 5.0 ± 1.0 5.1 ± 1.0 0.22a
 At 60 min 4.9 ± 0.3 5.0 ± 0.6 5.0 ± 0.3 0.62a
 At 90 min 4.9 ± 0.2 4.9 ± 0.4 5.0 ± 0.2 0.15a
 At 120 min 5.0 ± 0.3 4.9 ± 0.4 5.1 ± 0.2 0.56a
Data are mean ± SD, median [IQR], n (%).ED, emergence delirium; PACU, postanaesthesia care unit; OR, operating room. Statistical tests:
aOne-way analysis of variance.
bPearson Chi-squared test.
cFisher's exact test.
dKruskal--Wallis test.
P < 0.05.All numbers are stated in mean ± SD if not stated otherwise (number (% of patients)).

Table 3 - The absolute risk, reduction relative risk and relative risk reduction of emergence delirium
Melatonin vs. Placebo (Control) Midazolam vs. Placebo (Control) Melatonin vs. Midazolam (Control)
Absolute risk reduction (95% CI) 23.3 (3.7 to 42.9) −5.8 (−26.8 to 15.1) 29.2 (9.5 to 48.8)
Relative risk (Test/Control) (95% CI) 0.53 (0.30 to 0.94)P = 0.03 1.12 (0.75 to 1.66)P = 0.59 0.48 (0.27 to 0.83)P = 0.008
Relative risk reduction 0.47 (0.06 to 0.70) (risk decreases with melatonin) −0.12 (0.25 to −0.66) (risk increases with midazolam) 0.52 (0.17 to 0.73) (risk decreases with melatonin)
Numbers needed to prevent (95% CI) 4.29 (2.33 (benefit) to 26.79 (benefit)) 17.2 (3.73 (harm) to 6.61 (benefit) 3.43 (2.05 (benefit) to 10.50 (benefit))

The incidence of emergence delirium was lowest in the melatonin group at all time points and was significantly lower at 10, 15, 20 and 30 min in PACU, with P values of 0.01, less than 0.001, less than 0.001 and 0.009, respectively. The postoperative sedation scores were similar across the groups until 2 h after anaesthesia (P > 0.05) (Table 2).

Secondary outcomes

Pre-operative anxiety

The incidence of high anxiety (melatonin 24%, midazolam 30%, placebo 20%; P = 0.57) as well as mYPAS scores before premedication (median [IQR], melatonin 23.3 [23.3 to 25.0], midazolam 23.3 [23.3 to 31.7] and placebo 23.3 [23.3 to 23.3]; P = 0.53) were similar in the three groups (Table 4). During transfer to the operating room for the procedure, significantly more children in the melatonin (25%) and midazolam (33%) groups were drowsy or sleeping than in the placebo group (0%) (P < 0.001). Intergroup analysis revealed that midazolam and melatonin had similar sedation, which was significantly different from placebo group (melatonin vs. placebo, P = 0.003; midazolam vs. placebo, P = <0.001; melatonin vs. midazolam, P = 1.0). On arrival in the operating room, there was no difference in the mYPAS scores (P = 0.23) or proportion of highly anxious children (melatonin, 42%; midazolam, 42%; placebo, 23%; P = 0.09) between the three groups. The direction of change in anxiety, that is the number of children in whom anxiety remained the same, increased or decreased after premedication was also similar in the three groups (P = 0.13). Overall, the number of children with high anxiety increased from the baseline to the operating room in all groups (melatonin 24 to 42%, P = 0.04; midazolam 30 to 42%, P = 0.23; placebo 20 to 23%, P = 1.0)

Table 4 - Anxiety characteristics before and during induction
Variable Melatonin (n = 45) Midazolam (n = 43) Placebo (n = 44) P
Before premedication
 Anxiety in pre-operative ward (mYPAS score) 23.3 [23.3 to 25.0] 23.3 [23.3 to 31.7] 23.3 [23.3 to 23.3] 0.53d
 Number of highly anxious children (mYPAS ≥30) 11 (24) 13 (30) 9 (20) 0.57b
Alertness level of child during transfer to OR <0.001c ,
 Crying 2 (5) 1 (2) 6 (14)
 Awake 32 (73) 28 (65) 38 (86)
 Drowsy 9 (21) 14 (33) 0 (0)
 Sleeping 2 (5) 0 (0) 0 (0)
After arrival inside OR
 Child anxiety in OR (mYPAS score) 23.3 [23.3 to 49.2] 23.3 [23.3 to 40.0] 23.3 [23.3 to 27.9] 0.23d
 Number of highly anxious children (mYPAS ≥30) 19 (42) 18 (42) 10 (23) 0.09b
Direction of change in mYPAS after premedication 0.13b
 No change 26 (58) 25 (58) 32 (73)
 Increase 15 (33) 9 (21) 6 (14)
 Decrease 4 (9) 9 (21) 6 (14)
Induction Compliance Checklist 0 [0 to 2.5] 0 [0 to 1.0] 0 [0 to 1.0] 0.76d
Number of children with perfect induction (ICC score = 0) 26 (58) 28 (65) 28 (64) 0.75b
Data are median [IQR], and n (%).mYPAS, modified Yale Preoperative Anxiety Scale; ICC, Induction Compliance Checklist; OR, operating room. Statistical tests:
aOne-way analysis of variance.
bPearson Chi-square test.
cFisher's exact test.
dKruskal--Wallis test.
P < 0.05.

Mask acceptance

The acceptance of the facemask for inhalation induction as measured by the ICC was similar across the groups, P = 0.76 (Table 4). The majority of children had perfect induction (ICC = 0): melatonin 58%, midazolam 65%, placebo 64%, P = 0.77).


The current study found that oral melatonin premedication as part of multimodal anxiolysis significantly reduced the risk of emergence delirium by 23% in children undergoing ambulatory procedures under sevoflurane anaesthesia when compared to placebo, and by 29% when compared to midazolam. Oral midazolam premedication did not decrease the risk of emergence delirium.

Children who undergo ambulatory day-care procedures are reported to have a high incidence of pre-operative anxiety and a high incidence of emergence delirium.16,17 The potential risk factors known to contribute to emergence delirium are the volatile agent used, type of surgery and age. Sevoflurane predisposes to a high incidence of emergence delirium.18,19 A significant proportion of the children in all three groups had ophthalmic procedures, which also predispose to a high incidence of emergence delirium. These factors mentioned above may have contributed to the higher than expected incidence of emergence delirium in the current study, but these confounding factors applied to all three study groups.

Higher levels of pre-operative anxiety are associated with higher levels of emergence delirium postoperatively.2 A child's temperament and prehospital anxiety are risk factors for emergence delirium, which are difficult to control in a study setting. The current study design incorporated a multimodal anxiolytic strategy to maximise the chances of reducing pre-operative anxiety and thus emergence delirium. We found that the level of pre-operative anxiety was low with a median score of 23.3 in all the groups. As the pre-operative anxiety levels were similar, the difference in the effect on emergence delirium of pre-operative melatonin and midazolam indicates a beneficial pharmacological effect of melatonin on factors other than anxiolysis in preventing emergence delirium. It is known that neural remodelling of GABA receptors with a consequent decrease in binding sites is a possible mechanism of chronic posttraumatic delirium.20 And there is recent evidence that an action on GABA receptors may be involved in emergence delirium.21 Melatonin enhances GABAergic transmission and increases in GABA concentrations in the centra nervous system, thus facilitating neural inhibition,22 which may explain the efficacy of melatonin in decreasing emergence delirium.

Earlier studies of oral melatonin for preventing emergence delirium used varying doses, ranging from 0.05 to 0.5 mg kg−1. Because of the different doses, previous studies on the effect of melatonin in preventing emergence delirium have reported variable results, with Samarkandi et al.7 showing a lower incidence, Ozcengiz et al.15 reporting a similar incidence and Kain et al.23 recording a higher incidence of emergence delirium after melatonin administration in comparison with midazolam. However, a multimodal approach to decrease pre-operative anxiety was not employed in these studies. A meta-regression analysis found that the effect of melatonin on emergence delirium was significantly dose-dependent with possible efficacy at a dose at least 0.2 mg kg−1.9 The dose of 0.3 mg kg−1 chosen in the current study strikes a balance between efficacy and avoidance of postoperative sedation. Midazolam was administered at a dose of 0.3 mg kg−1 as it is the dose routinely used in our centre and within the range of doses recommended for anxiolysis.24

Nonpharmacological interventions are known to reduce pre-operative anxiety as effectively as sedative premedication.25,26 On analysing the changes in anxiety between the pre-operative period and the operating room, 60 to 70% of children did not have any change, neither increase nor decrease with sedative premedication or placebo. The bundle of simple nonpharmacological interventions instituted before the procedure in all three groups strengthens the study design by adding pragmatism, as many of these measures are integral in today's practice of paediatric anaesthesia. The high proportion of perfect mask induction in all groups (60 to 65%) and excellent quality of induction in the remaining children emphasise the benefits of multimodal anxiolytic interventions for mask induction of anaesthesia in children. It also leads to the inference that sedative premedication may not offer additional peri-induction advantage when nonpharmacologic measures are chosen wisely.

Although more children in the melatonin and midazolam groups scored higher on mYPAS scale in the operating room than in pre-operative room when compared to the placebo group, this did not reach statistical significance nor was it clinically significant. This may be explained by two possibilities: as the sedative and anxiolytic effects of melatonin and midazolam are dose-dependent,7,23,27 the moderate doses used in the study may not have produced maximal effect. However, the residual postop sedation after higher doses is undesirable as it prolongs PACU stay in paediatric ambulatory procedures.28,29 Second, the drowsiness induced by premedication decreased the tendency of the child to actively interact at induction and thus increased their mYPAS scores. This argument is corroborated by the finding that 25 to 32% of children in melatonin and midazolam groups were sleeping in contrast to placebo group where none were sleeping.

We used the Watcha scale to assess emergence delirium although Pediatric Anaesthesia Emergence Delirium (PAED) score is another established tool. A detailed evaluation of emergence delirium scales had found Watcha scale to possess higher overall sensitivity and specificity over PAED score.30 Watcha scale is a practical and simple tool, whereas PAED score is relatively cumbersome to use in a busy clinical setting.31 The alternative ED-1 score comprising the first three components of PAED score may be specific to the definition of delirium in children, but the nonspecific symptoms of delirium excluded from ED-1 score are equally challenging to the healthcare personnel in the PACU. Hence, it is important that an intervention to prevent ED targets both specific and nonspecific components of delirium. On the basis of these considerations, Watcha scale maybe a better metric of outcome to study interventions to prevent ED.

The current study has a few limitations. First, we did not record postoperative pain as an outcome using standardised assessment tools and this may have confounded the assessment of emergence delirium due to pain. However, we took precautions to decrease the impact of pain on the occurrence of emergence delirium as the attending anaesthesiologist, who was blind to group allocation, addressed pain management and administered rescue fentanyl as and when required after emergence and in PACU. The surrogate indicators of pain, the time interval to first fentanyl bolus and the total dose of fentanyl administered, were similar in all the three study groups. Second, logistically, we were limited as we could afford only one blinded assessor from the investigating team to collect data from a given randomised patient and were concerned about the introduction of bias in observations if the same investigator assessed pain and emergence delirium of the same patient. This was resolved by assigning pain assessment and management to the attending anaesthesiologist who was blinded to group allocation. Third, there were two independent observers (LS and AJ) involved during the study period to assess emergence delirium. Although these independent observers were blinded to group allocation, we have not assessed inter-observer agreement of this assessment. Fourth, although we recorded postoperative sedation, time to discharge was not recorded as other surgical and logistic factors also influence this parameter. Also, the effect of melatonin on behavioural changes days or weeks following surgery would have been interesting to study.

In conclusion, we found that melatonin as part of multimodal anxiolysis significantly reduced the risk of emergence delirium after sevoflurane anaesthesia in children who underwent ambulatory procedures. In contrast, oral midazolam did not decrease the risk of emergence delirium.

Acknowledgements relating to this article

Assistance with the article: none.

Financial support and sponsorship: this work was supported by the Department of Anaesthesia and Intensive Care, Post Graduate Institute of Medical Education and Research, Chandigarh, India.

Conflict of interest: none.

Presentation: none.


1. Nair S, Wolf A. Emergence delirium after paediatric anaesthesia: new strategies in avoidance and treatment. BJA Educ 2018; 18:30–33.
2. Kain ZN, Caldwell-Andrews AA, Maranets I, et al. Preoperative anxiety and emergence delirium and postoperative maladaptive behaviors. Anesth Analg 2004; 99:1648–1654.
3. Manso MA, Guittet C, Vandenhende F, Granier LA. Efficacy of oral midazolam for minimal and moderate sedation in pediatric patients: a systematic review. Paediatr Anaesth 2019; 29:1094–1096.
4. Zhang C, Li J, Zhao D, Wang Y. Prophylactic midazolam and clonidine for emergence agitation in children after emergence from sevoflurane anaesthesia: a meta-analysis. Clin Ther 2013; 35:1622–1631.
5. Yao Y, Sun Y, Lin J, et al. Intranasal dexmedetomidine versus oral midazolam premedication to prevent emergence delirium in children undergoing strabismus surgery. Eur J Anaesthesiol 2020; 37:1143–1149.
6. Breschan C, Platzer M, Jost R, et al. Midazolam does not reduce emergence delirium after sevoflurane anaesthesia in children. Paediatr Anaesth 2007; 17:347–352.
7. Samarkandi A, Naguib M, Riad W, et al. Melatonin vs. midazolam premedication in children: a double-blind, placebo-controlled study. Eur J Anaesthesiol 2005; 22:189–196.
8. Andersen LPH, Rosenberg J, Gögenur I. Perioperative melatonin: not ready for prime time. Br J Anaesth 2014; 112:7–8.
9. Mihara T, Nakamura N, Ka K, et al. Effects of melatonin premedication to prevent emergence agitation after general anaesthesia in children: a systematic review and meta-analysis with trial sequential analysis. Eur J Anaesthesiol 2015; 32:862–871.
10. Malhotra S, Malhotra A. Psychological adjustment of physically sick children: relationship with temperament. Indian Pediatr 1990; 27:577–584.
11. Kain ZN, Mayes LC, Cicchetti DV, et al. The Yale Preoperative Anxiety Scale: how does it compare with a ‘gold standard’? Anesth Analg 1997; 85:783–788.
12. Kain ZN, Mayes LC, Wang SM, et al. Parental presence during induction of anesthesia versus sedative premedication: which intervention is more effective? Anesthesiology 1998; 89:1147–1156.
13. Watcha MF, Ramirez-Ruiz M, White PF, et al. Perioperative effects of oral ketorolac and acetaminophen in children undergoing bilateral myringotomy. Can J Anaesth 1992; 39:649–654.
14. Soetenga D, Frank J, Pellino TA. Assessment of the validity and reliability of the University of Wisconsin Children's Hospital Pain Scale for Preverbal and Nonverbal Children. Pediatr Nurs 1999; 25:670–676.
15. Özcengiz D, Gunes Y, Ozmete O. Oral melatonin, dexmedetomidine, and midazolam for prevention of postoperative agitation in children. J Anesth 2011; 25:184–188.
16. Kain ZN, Mayes LC, O’Connor TZ, Cicchetti DV. Preoperative anxiety in children. Predictors and outcomes. Arch Pediatr Adolesc Med 1996; 150:1238–1245.
17. Ahmed MI, Farrell MA, Parrish KA, Karla A. Preoperative anxiety in children risk factors and nonpharmacological management. Middle East J Anaesthesiol 2011; 21:153–164.
18. Dahmani S, Delivet H, Hilly J. Emergence delirium in children: an update. Curr Opin Anaesthesiol 2014; 27:309–315.
19. Driscoll JN, Bender BM, Archilla CA, et al. Comparing incidence of emergence delirium between sevoflurane and desflurane in children following routine otolaryngology procedures. Minerva Anestesiol 2017; 83:383–391.
20. Geuze E, van Berckel B, Lammertsma A. Reduced GABAA benzodiazepine receptor binding in veterans with posttraumatic stress disorder. Mol Psychiatry 2008; 13:74–83.
21. Eom W, Lee JM, Park J, et al. The effects of midazolam and sevoflurane on the GABAA receptors with alternatively spliced variants of the (2 subunit. Korean J Anesthesiol 2011; 60:109–118.
22. Kurdi MS, Patel T. The role of melatonin in anaesthesia and critical care. Indian J Anaesth 2013; 57:137–144.
23. Kain ZN, MacLaren JE, Herrmann L, et al. Preoperative melatonin and its effects on induction and emergence in children undergoing anesthesia and surgery. Anesthesiology 2009; 111:44–49.
24. Heikal S, Stuart G. Anxiolytic premedication for children. BJA Educ 2020; 20:220–225.
25. Kain ZN, MacLaren J, McClain BC, et al. Effects of age and emotionality on the effectiveness of midazolam administered preoperatively to children. Anesthesiology 2007; 107:545–552.
26. Stewart B, Cazzell M, Pearcy T. Single-blinded randomized controlled study on use of interactive distraction versus oral midazolam to reduce pediatric preoperative anxiety, emergence delirium, and postanesthesia length of stay. J Perianesth Nurs 2019; 34:567–575.
27. Kurdi MS, Muthukalai SP. A comparison of the effect of two doses of oral melatonin with oral midazolam and placebo on preoperative anxiety, cognition and psychomotor function in children: a randomised double-blind study. Indian J Anaesth 2016; 60:744–750.
28. Viitanen H, Annila P, Viitanen M, Tarkkila P. Premedication with midazolam delays recovery after ambulatory sevoflurane anesthesia in children. Anesth Analg 1999; 89:75–79.
29. Faghihian R, Eshghi A, Faghihian H, Kaviani N. Comparison of oral melatonin and midazolam as premedication in children undergoing general anesthesia for dental treatment. Anesth Pain Med 2018; 8:e64236.
30. Bajwa SA, Costi D, Cyna AM. A comparison of emergence delirium scales following general anesthesia in children. Paediatr Anaesth 2010; 20:704–711.
31. Reduque LL, Verghese ST. Paediatric emergence delirium. Cont Educ Anaesth Crit Care Pain 2013; 13:39–41.
Copyright © 2021 European Society of Anaesthesiology and Intensive Care. Unauthorized reproduction of this article is prohibited.