Literature search results

A total of 290 studies were identified initially after screening various databases and searching additional sources. Subsequently, 199 duplicate records were removed, and 125 records were excluded by a thorough review of titles and abstracts. In these 125 excluded items, 10 were focused on adult patients, 1 was conducted on animals, 12 were conference abstracts, comments notes or letters, 50 were protocols or ongoing trials, 6 were reviews, and 46 were studies with irrelevant topics. Consequently, 64 items were further excluded following full-text review, and 3 of them were not relevant to the outcomes of the study, 4 of them did not focus on ketamine, 56 of them focused on ketamine administered not via nebulization, and 1 of them was not a randomized controlled trial. Eventually, 10 studies were selected for subsequent analysis [14, 21,22,23,24,25,26,27,28,29]. The PRISMA flowchart (Fig. 1) provides details on the identification of the literature.

Fig. 1

PRISMA flow diagram showing literature search results

Basic characteristics of enrolled studies

The involved studies were published from 2015 to 2022, with a total of 727 eligible pediatric patients ranging in age from 1 to 12 years. Among the included studies, six examined the effects of nebulized ketamine versus nebulized dexmedetomidine, while three focused on the effects of nebulized ketamine versus nebulized midazolam. Furthermore, three studies examined the effects of nebulized ketamine versus combination of nebulized ketamine and dexmedetomidine. In addition, one study reported findings on effects of nebulized ketamine versus oral ketamine, and one study examined the effects of nebulized ketamine versus intravenous ketamine. An overview of the main characteristics of the enrolled studies was presented in Table 1 including the following information: first author, publication year, range of age, American Society of Anesthesiologists status, type of surgery/procedure, drug dosage, sample size, scale used for sedation measurement and outcomes.

Table 1 The general characteristics of the enrolled studiesRisk of bias assessment

Cochrane Collaboration’s risk of bias tool was employed to appraise the validity and quality of the RCTs by us. In all 10 enrolled studies, 8 studies (80.00%) delineated an appropriate method of random sequence generation, 7 studies (70.00%) reported adequate allocation concealment, 8 studies (80.00%) showed a low risk in blinding of participants and personnel domain, and all studies described the blinding procedure of outcome assessment. The detailed information about risk of bias assessment is presented in Fig. 2.

Fig. 2

Risk of bias summary of included the trails: evaluation of bias risk items for each included study. Green circle, low risk of bias; red circle, high risk of bias; yellow circle, unclear risk of bias

Primary outcomesNumber of patients with satisfactory sedation levels

Five studies compared nebulized ketamine to nebulized dexmedetomidine described the number of patients with satisfactory sedation levels [14, 23, 26,27,28]. Owing to existence of statistical heterogeneity, the random-effects model was chosen in present analysis. And the results indicated that no significant differences were observed between nebulized ketamine group and nebulized dexmedetomidine group (54.79% vs. 60.69%, RR = 0.88, with 95%CI [0.61, 1.27], P = 0.49, I2 = 71%; Fig. 3; Table 2). The sensitivity analysis indicated that the heterogeneity (I2 = 71%) derived from the Geetha K et al. study [27]. And heterogeneity was resolved (I2 = 0%) by omitting this study, the more reliable results indicated that the summary estimate was changed (46.36% vs. 63.64%, RR = 0.77, 95% CI [0.63, 0.94], P = 0.009).

Fig. 3

Forest plot: Number of children with satisfactory sedation. No significant differences were observed between nebulized ketamine group and nebulized dexmedetomidine group (RR = 0.88, with 95%CI [0.61, 1.27], P = 0.49); Nebulization of dexmedetomidine plus ketamine can provide better sedative effect than nebulized ketamine alone (RR = 0.50, with 95%CI [0.27, 0.92], P = 0.03)

Three studies compared nebulized ketamine to nebulized ketamine plus dexmedetomidine reported the number of patients with satisfactory sedation levels [14, 26, 29]. Existence of statistical heterogeneity prompted us to applied random-effects model. The results indicated that nebulization of dexmedetomidine plus ketamine can provide better sedative effect than nebulized ketamine alone (33.82% vs. 68.11%, RR = 0.50, with 95%CI [0.27, 0.92], P = 0.03, I2 = 58%; Fig. 3; Table 2). The sensitivity analysis showed that the heterogeneity (I2 = 58%) was attributed to the Dharamkhele SA et al. [29] study. Following excluding this study, the heterogeneity was resolved (I2 = 0%), and the summary estimate was unchanged essentially (46.67% vs. 75.56%, RR = 0.63, 95% CI [0.44, 0.89], P = 0.009).

According to the GRADE summary of findings table, the quality of evidence pertaining to these outcomes was low. It was attributed to both inconsistency (I2 > 50%) and imprecision (lack of events number) (Table S1).

The results of Abdel-Ghaffar HS et al. [23] study indicated that no significant differences were observed between group midazolam and group ketamine (22/30 (73.33%) vs. 25/30 (83.33%); Table 2). According to Abdel-Ghaffar HS et al. [21] study, children in nebulized ketamine group showed more satisfactory sedation levels compared with children in the intravenous ketamine group (5/25 (20.00%) vs. 0/25 (0.00%); Table 2) and the control group (5/25 (20.00%) vs. 0/25 (0.00%); Table 2). However, the results of Kamel AAF et al. [22] study described that number of patients with satisfactory sedation levels was highly statistically significant difference in oral ketamine group than in nebulized ketamine group (9/31 (29.03%) vs. 31/31 (100.00%); Table 2).

Number of children with satisfactory separation from parents

Four studies compared nebulized ketamine to nebulized dexmedetomidine reported number of children with satisfactory separation from parents [14, 23, 26, 28]. The value of I2 (I2 = 59%) indicated that the statistical heterogeneity was existed, then we chose the random-effects model for analysis. Compared to nebulized dexmedetomidine, nebulized ketamine provided no obvious advantage in satisfactory separation from parents (57.27% vs. 73.64%, RR = 0.81, with 95%CI [0.61, 1.08], P = 0.15, I2 = 59%; Fig. 4; Table 2). After excluding the source of heterogeneity (Mohammad Hazem I et al. [26]), the heterogeneity was resolved (I2 = 39%) and the summary estimate was unchanged (68.24% vs. 80.00%, RR = 0.87, 95% CI [0.70, 1.08], P = 0.21).

Fig. 4

Forest plot: Number of children with satisfactory separation from parents. No significant differences were observed between nebulized ketamine group vs. nebulized dexmedetomidine group (RR = 0.81, with 95%CI [0.61, 1.08], P = 0.15), and nebulized ketamine group vs. nebulized ketamine plus dexmedetomidine group (RR = 0.92, with 95%CI [0.74, 1.14], P = 0.42)

Three studies compared nebulized ketamine to nebulized ketamine plus dexmedetomidine described the number of children with satisfactory separation from parents [14, 26, 29]. On account of existed statistical heterogeneity, the random-effects model was applied in present analysis. Analysis from the three studies found that nebulized ketamine plus dexmedetomidine has no statistical difference in number of children with satisfactory separation from parents compared to nebulized ketamine alone (64.71% vs. 73.91%, RR = 0.92, with 95%CI [0.74, 1.14], P = 0.42, I2 = 57%; Fig. 4; Table 2). Sensitivity analysis indicated that the heterogeneity (I2 = 57%) was attributable to the Dharamkhele SA et al. [29] study. Heterogeneity was resolved (I2 = 0%) by removing the study, and the summary estimate was unchanged (48.89% vs. 62.22%, RR = 0.84, with 95%CI [0.69, 1.03], P = 0.09, I2 = 0%).

The GRADE summary of findings table indicated that quality of evidence for present outcomes low. Inconsistency (I2 > 50%) and imprecision (limited number of events) were main factors (Table S1).

Abdel-Ghaffar HS et al. [23] found that no significant differences were observed between midazolam group and ketamine group(21/30 (70.00%) vs. 28/30 (93.33%); Table 2) in number of children with satisfactory separation from parents. Kamel AAF et al. [22] found that number of patients with satisfactory sedation levels was highly statistically significant difference in oral ketamine group than in nebulized ketamine group (8/31 (25.81) vs. 31/31 (100.00%); Table 2). Abdel-Ghaffar HS et al. [21] study indicated that patients in nebulized ketamine groups showed higher sedation scores compared with patients in the intravenous ketamine group (0.5 mg/kg) and the control group (P = 0.041), and there was no significant difference between nebulized ketamine group 1 (1 mg/kg) and nebulized ketamine group 2 (2 mg/kg) (P = 0.763).

Number of children with satisfactory mask acceptance

Four studies compared nebulized ketamine to nebulized dexmedetomidine described number of children with satisfactory mask acceptance [14, 23, 26, 28]. We applied random-effects model in analysis as the existed statistical heterogeneity (I2 = 50%). Analysis from the four studies found that no significant differences were observed between Nebulized Ketamine Group and Nebulized Dexmedetomidine Group (37.27% vs. 52.73%, RR = 0.71, with 95%CI [0.45, 1.10], P = 0.13, I2 = 50%; Fig. 5; Table 2). Sensitivity analysis showed that the heterogeneity (I2 = 50%) was attributable to the Mohammad Hazem I et al. [26] study. After omitting this study, the heterogeneity was resolved (I2 = 20%) and the summary estimate was unchanged (42.35% vs. 52.94%, RR = 0.84, with 95%CI [0.59, 1.19], P = 0.32, I2 = 20%).

Fig. 5

Forest plot: Number of children with satisfactory mask acceptance. No significant differences were observed between nebulized ketamine group and nebulized dexmedetomidine group (RR = 0.71, with 95%CI [0.45, 1.10], P = 0.13); Nebulized ketamine plus dexmedetomidine was associated with more satisfactory mask acceptance in pediatric patients compared to nebulized ketamine alone (RR = 0.69, with 95%CI [0.56, 0.86], P = 0.001)

The GRADE summary of findings table showed that quality of evidence for this outcome was low. Inconsistency (I2 > 50%) and imprecision (lack of events number) were considered as main reasons (Table S1).

Three studies compared nebulized ketamine to nebulized ketamine plus dexmedetomidine described the number of children with satisfactory mask acceptance [14, 26, 29]. Given that no statistical heterogeneity (I2 = 0%) was detected, the fixed-effects model was used for analysis. The results indicated that using of ketamine plus dexmedetomidine via nebulization was associated with more satisfactory mask acceptance in pediatric patients compared to nebulized ketamine alone (45.59% vs. 71.01%, RR = 0.69, with 95%CI [0.56, 0.86], P = 0.001, I2 = 0%; Fig. 5; Table 2).

According to GRADE summary of findings table, quality of evidence for present outcome was moderate. The imprecision (lack of events number) was considered as the main reason (Table S1).

In addition, the results of Abdel-Ghaffar HS et al. study [23] indicated that no significant differences were observed between midazolam group and ketamine group (20/30 (66.67%) vs. 17/30 (56.67%); Table 2) in number of children with satisfactory mask acceptance.

Secondary outcomes

Results of secondary outcomes including onset of sedation, recovery time, various adverse effects and hemodynamic status were summarized in Table 2.

Onset of sedation and recovery time

Geetha K et al. [27] found that the time to onset of sedation was significantly less in nebulized dexmedetomidine group compared to nebulized ketamine group (19.73 ± 8.43 min vs. 26.00 ± 7.33 min, P = 0.002). However, analysis of two studies found that no significant differences were observed between nebulized ketamine group and nebulized dexmedetomidine group in recovery time (MD = -2.96, with 95% CI [-8.69, 2.77], P = 0.31, I2 = 98%; Table 2).

Various adverse effects

The results involving various adverse effects indicated that no significant differences were found between nebulized ketamine group and nebulized dexmedetomidine group in incidence of vomiting (7.06% vs. 3.53%, RR = 1.86, with 95%CI [0.53, 6.55], P = 0.34,I2 = 0%; Fig. 6; Table 2), and nebulized ketamine was associated with higher incidence of emergence agitation (18.18% vs. 3.33%, RR = 4.98, with 95%CI [1.88, 13.16], P = 0.001, I2 = 0%; Table 2) compared to nebulized dexmedetomidine. And no significant differences were observed between nebulized ketamine group vs. nebulized midazolam group (13.33% vs. 1.67%, RR = 5.67, with 95%CI [1.03, 31.20], P = 0.05, I2 = 8%; Fig. 6) and nebulized ketamine group vs. nebulized ketamine plus dexmedetomidine group (4.44% vs. 6.67%, RR = 0.71, with 95%CI [0.15, 3.48], P = 0.68, I2 = 31%; Fig. 6) in the incidence of vomiting. In addition, for the occurrence of other adverse effects (e.g., hypotension, bradycardia, abnormal movement, nystagmus), the existing evidence was still lacking, and it was difficult to judge whether nebulized ketamine brings benefits compared with other sedative approaches.

Fig. 6

Forest plot: Incidence of Vomiting. No significant differences were observed between nebulized ketamine group vs. nebulized midazolam group (RR = 5.67, with 95%CI [1.03, 31.20], P = 0.05), nebulized ketamine group vs. nebulized dexmedetomidine group (RR = 1.86, with 95%CI [0.53, 6.55], P = 0.34), and nebulized ketamine group vs. nebulized ketamine plus dexmedetomidine group (RR = 0.71, with 95%CI [0.15, 3.48], P = 0.68)

Hemodynamic parameters

The results of general hemodynamic parameters indicated that nebulized ketamine provided more steady value of MAP (MD = 3.35, with 95% CI [0.61, 6.09], P = 0.02, I2 = 3%; Table 2) after administration compared to nebulized midazolam. And according to Shereef KM et al. study [25], the hemodynamic parameters (HR and MAP) showed statistically significant decrease throughout the perioperative period in nebulized dexmedetomidine group when compared with nebulized ketamine group.