Introduction
Young children may not understand the necessity of invasive procedures due to a lack of mental maturity. Also, the natural fear of injection and separation from parents can exacerbate this situation [1, 2, 3, 4]. Different levels of sedation can be applied to children before invasive procedures. The European Society for pediatric gastroenterology, hepatology, and nutrition (ESPGHAN) recommends general anesthesia (GA) or deep sedation for pediatric gastrointestinal (GI) endoscopy [1]. However, some medical centers may not have access to anesthesiologists, and some parents may not have consent due to the possible complications and the long period until discharge. 
On the other hand, there are a limited number of safe sedative medications for children [5, 6, 7, 8]. The selective drugs should have a rapid onset, provide adequate sedation, have few side effects, and not threaten respiration and hemodynamics [9, 10]. Midazolam is a sedative, hypnotic, amnestic, anxiolytic, and anticonvulsant agent [9, 11, 12]. It is a safe medication for short procedures in the pediatric group due to its 6–15 min distribution half-life and an elimination half-life of 1.5–2 h [10]. It can be administered through several routes, including oral, intravenous, muscular, buccal, nasal, and rectal [3, 13]. Blister taste and nasal irritation are the most common side effects of intranasal midazolam [13, 14]. Its use may be limited because of side effects, including mild respiratory depression leading to apnea [10]. Moreover, oral ketamine is a safe rapid-onset drug with sedative, analgesic, and amnestic effects [9, 15, 16, 17]. It is recommended to use ketamine cautiously due to its side effects, such as delirium, aspiration, excitation, stridor, laryngospasm, and post-sedation agitation [2, 17, 18].
To the best of our knowledge, the efficacy and safety of oral ketamine and nasal midazolam have not been compared through the time of pediatric upper GI endoscopy. Both drugs are prescribed without any invasive injections and without inducing additional anxiety. Because of no agreement on the route and drug of choice to provide adequate and safe sedation in pediatric invasive procedures, we conducted a study to compare the utility, safety, and sedation effects of intranasal midazolam and oral ketamine during upper gastrointestinal endoscopy in children.
Materials and Methods
This randomized double-blinded clinical trial was performed on 100 children who were candidates for upper gastrointestinal endoscopy and whose parents did not consent to GA. They were referred to 17-Shahrivar Hospital, Rasht city, Iran, from January 2014 to January 2015. 
The exclusion criteria were as follows: a history of allergy to benzodiazepines and ketamine and the appearance of endoscopy-related complications such as gastrointestinal bleeding or excessive procedure prolongation. After obtaining written informed consent, the eligible children were randomly assigned to two groups, including those who received oral ketamine (+placebo) and intranasal midazolam (+placebo). The patients were fasting for solid or non-clear liquids for 8 hours preoperatively. In the first group, 4 mg/kg oral ketamine (Rotex Medica Company, Germany) in combination with 0.5 mL/kg USP suspension (the United States of pharmacopoeia) was prescribed. Also, normal saline was dropped into the nasal cavity as the placebo. In another group, midazolam (0.1 mg/kg, intranasal) (Daroopakhsh Company, made in Iran) was administered in each nostril and slowly divided equally, drop by drop. Cooperated children over 2 years old were asked to stick out their tongues to abstain from swallowing the drug inside the nose until complete drug absorption, and the USP suspension was also prescribed as the placebo. The USP suspension was provided as a mixture of sucrose and water at 85%. Prescribing these drugs was done in parents’ arms to minimize patients’ stress 30 minutes before endoscopy.
Although the onset time of the sedation effect of intranasal midazolam is 10-15 min, and the duration of its effect is 60 minutes, oral ketamine effects appear between 20-30 min after prescription. Endoscopy was performed 30 min after drug and placebo administration for each patient. The fact that the onset of oral ketamine sedative effect is later than intranasal midazolam justifies the time of endoscopy in our study. For this purpose, if the patient is receiving ketamine, the onset of its sedative effect has occurred. The sedative effects of intranasal midazolam continue for up to 60 min, so half an hour after the administration was the logical time for endoscopy.
In this double-blind study, the endoscopist or the person who recorded the data and the parents or patients were unaware of the medications prescribed. A single experienced nurse, unaware of the groups assigned, administered both nasal and oral solutions. Only one nurse in the endoscopy ward was involved in the grouping and type of prescription drugs in patients. 
Variables including sex, age, height, sedation score, fear levels, cooperation and behavior of children at the time of separation from parents, arterial oxygen saturation by digital pulse oximetry, and vital signs were recorded before, during, and after the procedure. A child’s blood pressure status is evaluated based on age and sex, and the standard height percentile [19]. Furthermore, the time to being completely conscious after the procedure and complications, including nausea and vomiting, seizure, laryngospasm, stridor, and unconsciousness (confusion, delirium, etc.), were recorded. The severity of impatience at the time of separation from parents was scored as follows: 1 (no fear, good cooperation, or asleep), 2 (slight fear or cry, relaxed by ensuring), 3 (moderate fear or crying, no relaxed by ensuring), or 4 (crying needed to another person to keep the child). The fear scale was also labeled as none, mild, moderate, or severe. Ramsay scale [20] was used to assess sedation level as 1 (patient awake and anxious, agitated, or restless), 2 (patient awake and cooperative, oriented and tranquil, 3 (patient asleep, responsive to commands), 4 (patient asleep, with brisk response to stimuli (light and noise), 5 (patient asleep, with response only to pain), or 6 (patient with no response to any stimuli [light, noise, or pain]). The recovery time and recall of unpleasant experiences in >6 years old children, 1 hour after the procedure, were recorded and compared (1: recall in detail, 2: relative recall, 3: amnesia). Due to the possibility of midazolam-induced respiratory depression or postoperative seizure due to ketamine, we prepared flumazenil and diazepam vials before the procedure.
Nasal oxygenation was also considered when oxygen desaturation occurred. In cases of severe vomiting, ondansetron was prescribed. Also, labetalol was available for probable hypertension crises. In those with hypotension, normal saline was administered. Resuscitation equipment was ready as routine. 
For statistical analysis, the results were presented as Mean±SD for quantitative variables and were summarized by absolute frequencies and percentages for categorical variables. The normality of data was analyzed using the Kolmogorov-Smirnoff test. Categorical variables were compared using the Chi-square or Fisher exact test. Quantitative variables were also compared with the t-test, analyses of variance (ANOVA), the Mann-Whitney U test, or the Kruskal Wallis test. The change in study parameters after the procedure was examined using the paired t-test or Wilcoxon test. The SPSS software, version 22.0 for windows (SPSS Inc., Chicago, IL) was used for the statistical analysis. P<0.05 were considered statistically significant. 
Results
In this study, 103 children were assessed for eligibility, and after excluding 3 children, 100 patients in the two groups of ketamine and midazolam were compared (Figure 1). Most patients in both groups were girls, indicating 26(52%) in the ketamine and 28(56%) in the midazolam group. There was no significant difference regarding the demographic characteristics such as sex (P=0.841) and mean age between the two groups receiving ketamine or midazolam (7.52±2.69 years vs 7.54±3.12 years, respectively, P=0.880). Comparing hemodynamic parameters before, during, and after the procedure, data showed statistically significant lower arterial oxygen saturation during (98.14±2.41% vs 96.65±4.05%, respectively, P=0.023) and after the procedure (98.02±2.98% vs 96.86±3.48%, respectively, P=0.01) in children who received intranasal midazolam, rather than who received oral ketamine. Also, higher systolic blood pressure (based on child age and height percentile) was seen in children medicated with ketamine (111.53±12.38, 101.86±12.53, P=0.012) (Table 1).


It is worth noting that blood pressure in ketamine recipients never exceeded stage 1 of hypertension.
As Table 2 shows, 75% and 84% of children who received ketamine and midazolam had respectively excellent cooperation during the procedure, and no significant difference was noted (P=0.262).


The lack of fear was observed in 74% and 82% of patients in the two groups, respectively (P=0.413). Comparing the level of sedation according to the Ramsay score showed no significant difference in children who received ketamine vs midazolam (P=0.490).
There was no significant relationship between the groups regarding cooperation at the time of separation from parents and age (P=0.262). However, the highest level of cooperation was seen in children aged 10 to 14 years compared to younger children (respectively, 88.0% vs 68.8%, P<0.0001) (Table 3).


As shown in Table 4, comparing the sedation score between the groups showed no inter-group difference between the three age subgroups (P=0.445).


The sedation score was significantly better in the group aged 10 to 14 years compared with the lower ages (P=0.010). The sedation score status was not different in the two groups regarding sex (P=0.490).
Side effects, including hypertension, dizziness, and loosening of extremities, were shown only in the ketamine group but not in the midazolam group (P=0.012, P=0.006, and P=0.027, respectively). Some complications, including nausea, nystagmus, mild vomiting, diplopia, blurred vision, and headache, were more common in the ketamine group than in the midazolam group, with no statistically significant difference (Table 5).


Laryngospasm, apnea, and seizure did not occur in any group.
In this study, drug-induced hypertension was mild and resolved without any treatment after the procedure. Decreased O2 saturation was quickly corrected with nasal oxygenation. Sublingual ondansetron was administered in cases of persistent nausea/vomiting that continued after the procedure. There were no cases of drug-induced hypotension that required the administration of normal saline. 
On the other hand, 44.1% of children aged >6 years receiving ketamine remembered the procedure in detail, while remembering the procedure was revealed in 58.8% of those who received midazolam without any difference (P=0.114). There was no significant difference in terms of the mean length of recovery time between the two groups (P=0.474), but in a few patients in the ketamine group, recovery time was prolonged up to 90 minutes. 
Discussion
In the current study, we compared the safety and effectiveness of the two sedative drugs: oral ketamine and nasal midazolam. Regarding efficacy and hemodynamic stability, a slight decrease in arterial oxygen saturation was revealed in the midazolam group but not in the ketamine group. 
 Midazolam is one of the most commonly used agent in children with hypnotic, sedative, amnesic, anticonvulsant, and anxiolytic effects before diagnostic procedures, despite concerns about respiratory depression [22, 23]. Consistent with the present study, Lane et al. found no severe respiratory depression by intranasal midazolam [24]. Also, Miqdady et al. found a mild decrease in oxygen saturation in children who received midazolam [25]. 
We did not find any significant decrease in O2 saturation in children who received ketamine. As noted, ketamine can maintain airway reflexes during sedation, which may induce minimal side effects on the respiratory system [26]. Although it can also be accompanied by an increased risk of laryngospasm [2, 27, 28, 29], this complication did not occur during the current study.
Ketamine is a dissociative agent with a rapid onset of action that induces profound sedation, analgesia, and amnesia, with a short duration of action (15-30 min) which is adequate for routine diagnostic endoscopy, allowing fast recovery [26]. It induces functional dissociation between the limbic and the cortical systems. Impaired sensory recognition of painful stimuli impacts this cataleptic state, resulting in memory-inducing, a condition known as “dissociative anesthesia” [27]. In the present study, no significant difference was seen in remembering the detail of the procedure between the two groups, despite the expected difference between groups in terms of recalling the procedure. Viana et al. assessed the occurrence of amnesia after a dental procedure which was slightly higher in the oral midazolam group than in intranasal midazolam and a combination of oral ketamine /midazolam [30]. 
Overall, no significant difference was found in other parameters, including the severity of impatience at the time of separation from parents, sedation score, level of fear, and the presence of restlessness and agitation during the procedure between the two groups. Consistent with our study, Khoshrang et al. revealed no significant difference between the sedation score of children who received intranasal Midazolam vs intranasal Ketamine [3]. Akçay et al. showed better sedation scores in children who received the combination of intra-nasal ketamine and midazolam than in children who received these drugs alone [31]. Rubinstein et al. found that the level of sedation during the procedure in children treated with ketamine was not significantly different from those treated with midazolam which was consistent with the current study. However, failure to achieve adequate sedation was more common in the ketamine group [15]. Recovery time in the previous research was longer in the intranasal ketamine group than in children who received intranasal midazolam, contrary to our study [3]. However, the method of ketamine administration was different in the two studies. 
Regarding complications, our results showed that dizziness, nystagmus, diplopia, and loosening extremities were more common in children who received ketamine, consistent with some other studies [17, 26]. However, the difference in the results is justifiable, according to the different doses and administration methods.
Conclusion
Either oral ketamine or intranasal midazolam before endoscopy is a safe and non-invasive method that induces sedation, and regardless of the slight differences in blood pressure and oxygen saturation, they had no superiority to each other in terms of sedation score, fear, or impatience at the time of separation from parents. In conclusion, ketamine might be preferred to midazolam regarding respiratory and O2 saturation stability, while midazolam can be preferred concerning drug side effects and maintaining blood pressure. 

Ethical Considerations
Compliance with ethical guidelines
All study procedures followed the ethical guidelines of the Declaration of Helsinki 2013. The study protocol was approved by the Ethics Committee of the Vice-Chancellor of Research at Guilan University of Medical Sciences (Code: 1930309905, Date: 2014/9/6) and registered at IRCT (CODE: IRCT 2014111419936N1) (Date: 2014/12/12).

Funding
This study was financially supported by the Vice Chancellor of Research at Guilan University of Medical Sciences.

Authors contributions
Conceptualization and investigation: Shohreh Maleknejad, Seyyedeh Azade Hoseini Nouri, Afshin Safaei-Asl, and Farnoush Farrzi; Methodology: Abtin Heidarzadeh and Afagh Hasanzadeh Rad; Data analysis: Abtin Heidarzadeh; Original draft: Shohreh Maleknejad and Seyyedeh Azade Hoseini Nouri; Writing, review, and editing: Seyyedeh Azade Hoseini Nouri, Afagh Hasanzadeh Rad, Shohreh Maleknejad, and Afshin Safaei; Supervision: Shohreh Maleknejad.

Conflict of interest
The authors declared no competing interests.

Acknowledgements
We appreciate our colleagues for cooperating and participating in this study, especially the staff of the Pediatric Diseases Research Center at 17-Shahrivar Hospital. We are also grateful for the kind cooperation of the parents and participants throughout the study. 


References

1. Kedia S, Sharma R, Makharia G, Ahuja V, Desai D, Kandasamy D, et al. Indian guidelines on imaging of the small intestine in crohn's disease: A joint Indian society of gastroenterology and indian radiology and imaging association consensus statement. Indian J Radiol Imaging. 2019; 29(2):111-32 [DOI:10.4103/ijri.IJRI_153_18] [PMID] [PMCID]
2. Kerkhofs L, Allegaert K, Toelen J, Vanhonsebrouck K. Pediatric procedural sedation and analgesia (PROSA) in the Leuven University Hospitals: An audit on efficacy and safety. Children (Basel). 2022; 9(6):776. [DOI:10.3390/children9060776] [PMID] [PMCID]
3. Khoshrang H, Alavi CE, Rimaz S, Mirmansouri A, Farzi F, Biazar G, et al. Efficacy of intranasal ketamine and midazolam for pediatric sedation: A double-blind, randomized clinical trial. Caspian J Intern Med. 2021; 12(4):539-43. [DOI:10.22088/cjim.12.4.539] [PMID] [PMCID]
4. Ramburuth M, Rodseth RN, Reddy P. Ketamine and midazolam as procedural sedation in children: A structured literature review and narrative synthesis. South African J Anaesth Analg. 2020; 26(1):10-6. [DOI:10.36303/SAJAA.2020.26.1.2302]
5. Shetabi H, Shafa A, Zare M. Effect of ketamine-sufentanil and ketamine-midazolam to induce sedation and analgesia in pediatric with lumbar puncture or bone marrow aspiration. Arch Anesth Crit Care. 2018; 4(3):497-500. [Link]

6. Gomes HS, Miranda AR, Viana KA, Batista AC, Costa PS, Daher A, et al. Intranasal sedation using ketamine and midazolam for pediatric dental treatment (NASO): Study protocol for a randomized controlled trial. Trials. 2017; 18(1):172. [DOI:10.1186/s13063-017-1919-2] [PMID] [PMCID]
7. Pickthorn S, Terpstra E, Conde TP, Kolba J, Ali M. Use of midazolam for minimal sedation in pediatric outpatients: A survey analysis of patient experience and parent/guardian satisfaction. South Dakota Medicine. 2019; 72(9):414-7. [PMID]
8. Bollinger LL, Yao LP. Pediatric sedatives and the food and drug admini, stration (FDA): Challenges, limitations, and drugs in development. In: Mason KP, editor. Pediatric sedation outside of the operating room. A multispecialty international collaboration. Cham: Springer; 2021. [DOI:10.1007/978-3-030-58406-1_32]
9. Heydari F, Shabani H, Majidinejad S, Nasr-esfahani M. Efficacy and safety of intranasal midazolam versus chloral hydrate as sedation for quality computed tomography imaging in children. J Pediatr Res. 2019; 6(4):286-91. [Link]
10. Jain SA, Rathi N, Thosar N, Baliga S. Midazolam use in pediatric dentistry: A review. J Dent Anesth Pain Med. 2020; 20(1):1-8. [DOI:10.17245/jdapm.2020.20.2.105] [PMID] [PMCID]
11. Cheng X, Chen Z, Zhang L, Xu P, Qin F, Jiao X, et al. Efficacy and safety of midazolam oral solution for sedative hypnosis and anti-anxiety in children: A systematic review and meta-analysis. Front Pharmacol. 2020; 11:225. [DOI:10.3389/fphar.2020.00225] [PMID] [PMCID]
12. 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(11):1094-106. [DOI:10.1111/pan.13747] [PMID] [PMCID]
13. Fantacci C, Fabrizio GC, Ferrara P, Franceschi F, Chiaretti A. Intranasal drug administration for procedural sedation in children admitted to pediatric emergency room. Eur Rev Med Pharmacol Sci. 2018; 22(1):217-22. [DOI:10.26355/eurrev_201801_14120] [PMID]
14. Peerbhay F, Elsheikhomer AM. Intranasal midazolam sedation in a pediatric emergency dental clinic. Anesth Prog. 2016; 63(3):122-30. [DOI:10.2344/15-00016.1] [PMID] [PMCID]
15. Rubinstein O, Barkan S, Breitbart R, Berkovitch S, Toledano M, Weiser G, et al. Efficacy of oral ketamine compared to midazolam for sedation of children undergoing laceration repair: A double-blind, randomized, controlled trial. Medicine. 2016; 95(26):e3984 [DOI:10.1097/MD.0000000000003984] [PMID] [PMCID]
16. Motamed F, Aminpour Y, Hashemian H, Soltani AE, Najafi M, Farahmand F. Midazolam-ketamine combination for moderate sedation in upper GI endoscopy. J Pediatr Gastroenterol Nutr. 2012; 54(3):422-6. [DOI:10.1097/MPG.0b013e3182323c75] [PMID]
17. Akbulut UE, Cakir M. Efficacy and safety of low dose ketamine and midazolam combination for diagnostic upper gastrointestinal endoscopy in children. Pediatr Gastroenterol Hepatol Nutr. 2015; 18(3):160-7. [DOI:10.5223/pghn.2015.18.3.160] [PMID] [PMCID]

18. Orhurhu VJ, Vashisht R, Claus LE, Cohen SP. Ketamine toxicity. Treasure Island: Stat Pearls Publishing; 2022. [Link]
19. Dalili S, Mohammadi H, Rezvany SM, Dadashi A, Novin MH, Gholaminejad H, et al. The relationship between blood pressure, anthropometric indices and metabolic profile in adolescents: A cross sectional study. Indian J Pediatr. 2015; 82(5):445-9. [DOI:10.1007/s12098-014-1573-6] [PMID]
20. Rasheed AM, Amirah MF, Abdallah M, Parameaswari PJ, Issa M, Alharthy A. Ramsay sedation scale and Richmond agitation sedation scale: A cross-sectional study. Dimens Crit Care Nurs. 2019; 38(2):90-5. [DOI:10.1097/DCC.0000000000000346] [PMID]
21. Poonai N, Canton K, Ali S, Hendrikx S, Shah A, Miller M, et al. Intranasal ketamine for procedural sedation and analgesia in children: A systematic review. Plos One. 2017; 12(3):e0173253. [DOI:10.1371/journal.pone.0173253] [PMID] [PMCID]
22. Uludağ Ö, Doğukan M, Kaya R, Tutak A, Dumlupınar E. Comparison of the effects of midazolam-ketamine or midazolam-propofol combinations on hemodynamic stability, patient comfort, and post-anesthesia recovery in children undergoing sedation for magnetic resonance imaging procedures. Ain-Shams J Anesth. 2020; 12(1):1-7. [DOI:10.1186/s42077-019-0037-7]
23. Conway A, Chang K, Mafeld S, Sutherland J. Midazolam for sedation before procedures in adults and children: A systematic review update. Syst Rev. 2021; 10(1):69. [DOI:10.1186/s13643-021-01617-5] [PMID] [PMCID]
24. Lane KJ, Nelson TM, Thikkurissy S, Scott JM. Assessing temperament as a predictor of oral sedation success using the children’s behavior questionnaire short form. Pediatr Dent. 2015; 37(5):429-35. [PMID]
25. Miqdady MI, Hayajneh WA, Abdelhadi R, Gilger MA. Ketamine and midazolam sedation for pediatric gastrointestinal endoscopy in the Arab world. World J Gastroenterol. 2011; 17(31):3630-5. [DOI:10.3748/wjg.v17.i31.3630] [PMID] [PMCID]
26. Majidinejad S, Ebrahimi M, Azizkhani R, Mansouri M, Ahmadpour M, Esmailian M, et al. The effects of different doses of submucosal vs. intravenous ketamine for conscious-sedation in children candidates for diagnostic-therapeutic procedures in emergency department. Front Emerg Med. 2021; 5(1):e5. [DOI:10.22114/ajem.v0i0.315]
27. Motamed H, Maleki Verki M, Mozafari J, Forouzan A. Ketamine associated vomiting in children requiring sedation: A prospective randomized open trial study. Int J Pediatrics. 2018; 6(12):8611-8. [DOI:10.22038/ijp.2018.33524.2960]
28. Cosgrove P, Krauss BS, Cravero JP, Fleegler EW. Predictors of laryngospasm during 276,832 episodes of pediatric procedural sedation. Ann Emerg Med. 2022; 80(6):485-96. [DOI:10.1016/j.annemergmed.2022.05.002] [PMID]
29. Sedasyon KM, ve Porofole AP, Sağlar GD. Ketofol provides better upper airway size than ketamine and propofol in pediatric population receiving sedation for magnetic resonance imaging. Turkiye Klinikleri J Pediatr. 2020; 29(2):57-64. [DOI:10.5336/pediatr.2019-71846]
30. Viana KA, Moterane MM, Green SM, Mason KP, Costa LR. Amnesia after midazolam and ketamine sedation in children: A secondary analysis of a randomized controlled trial. J Clin Med. 2021; 10(22):5430. [DOI:10.3390/jcm10225430] [PMID] [PMCID]
31. Akçay ME, Kılıç ET, Akdemir MS. The comparison of the efficacy and safety of midazolam, ketamine, and midazolam combined with ketamine administered nasally for premedication in children. Anesth Essays Res. 2018; 12(2):489-94. [DOI:10.4103/aer.AER_80_18] [PMID] [PMCID]