A prospective, randomized, double blinded comparison of intranasal dexmedetomodine vs intranasal ketamine in combination with intravenous midazolam for procedural sedation in school aged children undergoing MRI.

BACKGROUND: For optimum magnetic resonance imaging (MRI) image quality and to ensure precise diagnosis, patients have to remain motionless. We studied the effects of intranasal dexmedetomidine and ketamine with intravenous midazolam for pre-procedural and procedural sedation in school aged children. PATIENTS AND METHODS: Children were randomly allocated to one of two groups: (Group D) received intranasal dexmedetomidine 3 μg kg(-1) and (Group K) received intranasal ketamine 7 mg kg(-1). Sedation levels 10, 20 and 30 min after drug instillation were evaluated using a Modified Ramsay sedation scale. A 4-point score was used to evaluate patients when they were separated from their parents and their response to intravenous cannulation.
RESULTS: The two groups were comparable in terms of the child's anxiety at presentation (P = 0.245). We observed that Group K achieved faster sedation at 10 min point with P < 0.05. A comparable sedation score at 20 and 30 min were noted. The two groups were comparable regarding to the child's acceptance of nasal administration (P = 0.65). The sedation failure rate was insignificantly differ between groups (13.7% vs. 20.6% for Group D and K respectively). Heart rate and systolic blood pressure showed a significant difference between the two groups starting from the point of 20 min.
CONCLUSION: Intranasal dexmedetomidine 3 μg kg(-1) or ketamine 7 mg kg(-1) can be used safely and effectively to induce a state of moderate conscious sedation and to facilitate parents' separation and IV cannulation. Addition of midazolam in a dose not sufficient alone to produce the target sedation achieved our goal of deep level of sedation suitable for MRI procedure.

RCT Entities:   Population Interventions Outcomes

INTRODUTION

Due to advances in magnetic resonance imaging (MRI) and its important role in the diagnosis of various diseases, deep sedation or anesthesia for MRI in children is increasingly requested. Scan takes about 10-30 min according to the diagnostic MRI study needed and it is quite noisy and the patient is moved into a narrow pipe with limited access. For optimum image quality and to ensure precise diagnosis, patients have to remain motionless. Goals of sedation in the pediatric groups of patients for diagnostic and therapeutic procedures can be defined as: Guard the patient's safety and welfare; decrease physical discomfort and pain; reduce anxiety, minimize psychological trauma and maximize the potential for amnesia; control behavior and/or movement to permit the safe completion of the procedure; and return the patient to a state that permit safe discharge from medical supervision, as determined by recognized criteria.[1]

Fear of physicians, injections, operation and the operation theatre in children is much more than in adults. Separation from the parents to a totally unknown world with unknown faces makes the operative experience more traumatic for young children.[2]

An a traumatic sedation can minimize the previously mentioned problems. Intranasal (IN) drug delivery is relatively easy and convenient, it also reduces first pass metabolism and has been used successfully for fentanyl, ketamine, and midazolam premedication.[345]

Dexmedetomidine (Precedex, Hospira) is a highly selective α2-adrenoceptor agonist approved by the US Food and Drug Administration (FDA) in 2008 for sedation of adults in areas outside the intensive care unit (Package Insert, Hospira, 2008) with a shorter half-life. Although dexmedetomidine is not labeled for pediatric use, its application for pediatric sedation has been described.[6]

Ketamine may be administered through intravenous, intramuscular, oral, rectal, nasal, epidural, or intrathecal routes. The use of ketamine to induce sedation and analgesia in pediatric patients has been described in various non-operating room settings, including emergency department, oncologic, dental, radiation therapy and radiology suite settings.[7]

The purpose of this investigation is to compare the efficacy of two sedation regimens in children who required sedation for an MRI procedure. The children received either intranasal dexmedetomidine or intranasal ketamine, both together with intravenous midazolam and monitored by a physician for anxiolysis, sedation, parents’ separation, acceptance of IV cannulation, respiratory and hemodynamic parameters.

PATIENTS AND METHODS

After approval from our New Jeddah Clinic Hospital Ethics Committee, written informed consent from the parents or legal guardian, Sixty three children of American Society of Anesthesiologists (ASA) physical status I or II, aged between 4 and 10 years, weighing between 10 and 30 kg, scheduled to undergo MRI study, were enrolled in this prospective, randomized, double-blind study. When the child was mature enough to understand and discuss the need for premedication, patient assent was also obtained. Children with known allergy or hypersensitive reaction to dexmedetomidine or ketamine, organ dysfunction, those having medical conditions that would contraindicate a child from receiving dexmedetomidine or ketamine sedation [Table 1] were excluded.

Table 1

Medical conditions that contraindicate dexmedetomidine and ketamine

A total of 58 child completed the study analysis due to dropout during study. Computer-generated assignment in a prospective fashion used for randomization and to allocate patients to one of two separate equal study groups (Group D, n = 29) and (Group K, n = 29). Children received through the nasal route, in a double-blind manner, either dexmedetomidine 3 μg kg−1 (Group D) or ketamine 7 mg kg−1 (Group K).

Intranasal dexmedetomidine was prepared from the 100 μg/mL parenteral preparation (Hospira®) in a 2-mL syringe; 0.9% saline was added to make a final volume of one mL. Nasal ketamine was prepared in one mL according to dose with saline from parenteral veterinary formulation of ketamine vial (200 mg/mL), divided between both nostril while the child was either sitting on the parent's lap or on the bed.

Before procedure by about two hours EMLA cream (mixture of local anesthetics) was applied to the site of possible venipuncture. Since the previous study of healthy adults has shown that the mean onset time for significant sedation after 1 μg kg−1 intranasal dexmedetomidine was approximately 45-60 min,[8] all children received IN medication at approximately 45 min before the planned procedure. Baseline heart rate (HR), oxygen saturation (SpO2) and systolic blood pressure (SBP), were measured before any drug administration.

Behavior was assessed before instillation of sedating drugs on a 4-point scale:

Calm, cooperative

Anxious but reassurable

Anxious and not reassurable

Crying or resisting.

The child's response to the drug was measured on a 3-point scale (1 = refusing violently, 2 = defensive action/weeping, 3 = no defensive action). An independent investigator not involved in the observation prepared all the study drugs. Observers and attending anesthesiologists were blinded to the study drug given.

Heart rate (HR), Oxygen saturation (Spo2), and Systolic blood pressure (SBP) were measured before and every 10 min after intranasal drug administration until transfer to the MRI scanner.

After 30 min of the IN dose, sedation score according to Modified Ramsay sedation scale with American Academy of pediatrics/Joint Commission/ASA Designation[9] [Table 2] and reaction to parents’ separation of the child was assessed according to separation score.[10]

Table 2

Modified Ramsay sedation scale with American academy of pediatrics/joint commission/American society of anesthesiologists designation

Separation score

Poor (crying, clinging)

Fair (crying but not clinging)

Good (whimpers, easily reassured)

Awake excellent (easy separation).

We considered 3 and 4 as satisfactory separation while 1 and 2 are unsatisfactory.

An empirical four-point scoring system devised by Gharde et al.,[10] which was used for evaluation of acceptance of the IV cannula is as follows:

Poor (terrified, crying)

Fair (fear of needle, not reassured)

Good (Slight fear of needle, easily reassured)

Excellent (unafraid, accepts IV cannula readily).

Children with scores of 3 or 4 were designated as having a satisfactory acceptance of intravenous cannula, while scores of 1 or 2 were unsatisfactory.

Anxious children were given a second dose of nasal dexmedetomidine 1 μg.kg−1 (Group D) or ketamine 2 mg kg−1 (Group K) and reassessed after 10 min. then shifted inside MRI scanner. All children included in the study in both groups will receive 0.2 mg.kg−1 IV midazolam just before the start of MRI study. In case of failed sedation or movements that need rescue sedation, administration of a sedative drug that is different from the study sedative drugs (dexmedetomidine or ketamine) to achieve deeper sedation was started by propofol IV (0.5 mg kg−1 bolus and 2 mg kg−1 h−1 by infusion).

Sedation was evaluated upon entrance to MRI scanner on a Modified Ramsay sedation scale with American Academy of pediatrics/Joint Commission/ASA Designation [Table 2][9].

A single observer, who was blinded to the treatment given, was present throughout the MRI and recovery periods.

During the procedure, respiratory rate, hemoglobin oxygen saturation (using a magnetic field compatible pulse oximeter) and CO2 tension (via the nasal cannula), noninvasive blood pressure and HR measurements were recorded at 5-min intervals. During the procedure, the degree of sedation assessed at 10-min intervals by means of recording the responses to gentle touch and shaking of patients shoulder. Recovery of all patients in the post anesthesia recovery room in the radiology department was monitored. During the recovery period, the observer recorded SpO2, noninvasive blood pressure and HR measurements every 15 min.

A HR under 50 beats per minute or a 20% decrease from the baseline was considered an indication of bradycardia, whereas a HR of over 110 or an increase in the baseline level of more than 20% was considered as a tachycardia. Mean arterial pressure levels that were lower than 60 mmHg or 20% less than the baseline were regarded as hypotension and a mean arterial pressure value of over 150 mmHg or a 20% increase from the baseline was regarded as hypertension. Respiratory adverse events was defined as wheezing, laryngeal spasm, apnea, desaturation (decrease in SpO2 less than 92% for more than 10 s) with the need for maneuvers to improve patency of the airway, such as a shoulder roll, oropharyngeal, nasopharyngeal airway, laryngeal mask airway, or endotracheal intubation.

The discharge criteria were based on Modified Aldrete score ≥ 8 (each parameter from scale of 0-2 for activity, respiration, circulation, consciousness, color).

Sedation onset time (the time in minutes from the administration of the primary sedative drug to the documented time of satisfactory sedation), procedure duration (time in minutes from the start of the procedure to the termination of the MRI procedure), recovery times (time in minutes from the end of the procedure up to the time that the patient was discharged from the recovery area). Adverse events including failed sedation, need for additional boluses or rescue sedation and psychologic disturbances during the procedure, in the recovery room and at home (according to 24h and 7-day follow-up calls) were recorded.

Failed sedation was defined as inadequate sedation after the administration of the 2nd dose of the primary study sedative drug and need for rescue sedation, resulting in an inability to perform the MRI studies.

The primary end-point of the study was adequate pre procedural (moderate) and procedural (deep) sedation to complete the MRI study. Secondary end-points included acceptance of IV cannulation, parents’ separation, systolic BP (SBP), HR, respiratory rates changes and the incidence of complications. Sedation onset times, procedure durations, recovery times and number of failed sedation were also recorded.

Statistical analyses

Statistical processing of data was performed using IBM® SPSS® Statistics 20. We chose the mean sedation score at the time of transfer to the MRI to calculate the required sample size for this study. The required sample size was calculated to be 28 patients per group with α = 0.05 and a power of 80% to detect a difference of at least 30% in the mean sedation score. The data are presented as means ± standard deviation for continuous variables and as counts and percentages for nominal data. The independent t-test was used to compare means. The demographic data were analyzed by independent t- test and the Chi-squared test. Mann-Whitney test was used for evaluating the sedation level, child acceptance of the intranasal drug, ease of IV cannulation scores. The Fisher's exact test was used to compare nominal data or percentages. Changes in HR and arterial blood pressure were analyzed with paired t-tests. A P < 0.05 was considered to be statistically significant.

RESULTS

Totally 63 children undergoing MRI scheduled to receive either intranasal dexmedetomidine or ketamine. Fifty eight children successfully completed the imaging studies and were included in the analysis. Two patients were excluded from the study due to very anxious state and inability to administer nasal medications. Another three patients were excluded due to difficult venous access with multiple trials.

The patient's characteristics, type and duration of the MRI studies showed no statistical significant difference between both groups [P > 0.05; Tables 3]. Four patients, one in the Group D and three in the Group K, received a second dose of dexmedetomidine and ketamine respectively to achieve a preprocedural sedation on Modified Ramsay sedation score of B i.e., moderate sedation level [Table 4].

Table 3

Patient demographics of the two groups

Table 4

Onset of sedation, second dose, rescue propofol, sedation failure rate and recovery time

The child's behavior, before the start of the procedure was assessed, on a four-point scale. Most of the children were anxious and afraid of the procedure. Only 12 (20.06%) children were calm and cooperative at presentation. Five (8.62%) children were crying. Anxious and not reassurable 25 (43.1%) children and anxious but reassurable 16 (27.58%). The two groups were comparable in terms of the child's anxiety at presentation (P = 0.245).

The majority of children accepted the nasal medications with minimal discomfort. Although 54% of the children showed no defensive action, 32.3% wept when the drug was instilled into the nostrils. Only 13.7% of the children fought violently against the drug administration. The two groups were comparable regarding to the child's acceptance of nasal administration (P = 0.65).

All the patients successfully completed the MRI studies. The sedation failure rate was insignificantly differ in the D group (13.7%), in comparison with the K group (20.6%) [Table 4].

The onset of sedation was significantly shorter in the K group in comparison with the D group (14.65 ± 4.9 min vs. 22.4 ± 5.6 min [Table 4] (P < 0.001). Three patients (10.3%) of Group K required a second dose of nasal ketamine while one patient (3.4%) of Group D required a second dose (P = 0.3). Furthermore, there were no statistically significant values regarding sedation failure rate and recovery time (P > 0.05). Patients of Group K required more dose of rescue propofol than Group D but did not reach significant limit (P = 0.21) [Table 4].

We observed that Group K achieved faster sedation at 10 min point with P < 0.05 [Figure 1]. A comparable sedation score at 20 and 30 min were noted with better insignificant sedation score of Group D than Group K [Table 5].

Figure 1

Sedation score.(*p < 0.05)

Table 5

Sedation score, separation score at 30 min of drug adminstration and sedation score at cannulation time

Table 5 shows the sedation and separation scores at 30 min and cannulation acceptance. Sedation score showed no statistically significant difference between Group D and Group K children (4.89 ± 0.85 vs. 4.58 ± 0.9 respectively, P > 0.05). Separation score also showed no statistically significant difference between Group D and Group K children (3.86 ± 0.35 vs. 3.75 ± 0.43 respectively, P > 0.05). Both groups showed a satisfactory acceptance to IV cannulation with P > 0.05.

Reaction to parents’ separation score also showed no statistically significant difference between Group D and Group K children (3.86 ± 0.35 and 3.75 ± 0.43 respectively, P > 0.05). The intravenous cannulation was insignificantly accepted between Group D than Group K children (sedation score was 3.65 ± 0.9 and 3.72 ± 0.78 respectively, P > 0.05).

Overall, we did not observe any clinically significant effects of any of the studied drugs on SpO2 below 97% and respiratory rate during the observation period after administering study drugs. There was a significant difference in heart rate and systolic blood pressure between the two groups starting from the point of 20 min. In Group D, the HR and SBP were significantly decreased at 30 min value when compared with basal readings (P < 0.05) [Figure 2].

Figure 2

Heart rate, systolic blood pressure, SpO2 and respiratory rate changes. There was a significant difference in heart rate and systolic blood pressure between the two groups starting from the point of 20 min. In Group D, the heart rate and SBP were significantly decreased at 30 min value when compared with basal readings (P < 0.05)

The incidence of side-effects was similar among the both groups (P = 0.213). Two patients in Group D had bradycardia (HR < 20% of baseline) that did not require treatment with atropine. Five patients in Group K and two patients in Group D had an episode of nausea and vomiting that was treated with injected ondansetron (0.08 mg.kg−1) [Figure 2].

DISCUSSION

Many sedative analgesic agents and routes of drug administration for facilitation of painful procedures have been studied, with varying degrees of patient acceptance, efficacy and safety.[11]

The present study was performed to evaluate the efficacy and safety of intranasal dexmedetomidine and ketamine as sedatives in school age pediatric groups of patients.

The current study did not include a placebo group as these drugs when compared with placebo the effectiveness was found to be superior to placebo in previous studies.[12131415]

Usually the Observer's Assessment of Alertness and Sedation (OAAS) scale or the Ramsey score are used to describe sedation depth clinically.[1] For children the American Academy of Pediatrics defined four sedation steps: Anxiolysis, conscious sedation, deep sedation and anesthesia.[9]

The primary aim of our study was to compare the effectiveness of both intranasal dexmeditomidine and ketamine to achieve adequate pre procedural (moderate) and procedural (deep) sedation in combination of both drugs with IV medazolam to complete the MRI study. Secondary end-points included status at IV cannulation and separation from the parents, SBP, HR, respiratory rates changes and the incidence of complications. Sedation onset times, procedure durations, recovery times and number of failed sedation also recorded.

Concerning children and MRI, because of noise and tube narrowness, deep sedation is the required depth for examination in most cases. Stopping an MRI scan is expensive and ineffective, thus the failure rate has to be minimized.

Intranasal delivery offers unique advantages that may allow more efficient use of resources, more rapid patient care and higher patient and provider satisfaction.[16]

Minimizing drug volume while maximizing drug concentration, adequate dosing, use of both nostrils to double the absorptive mucosal surface and use of atomized particles to enhance nasal medication absorption. Concentrated medications in a small volume (0.2-0.3 mL per nostril) are ideal, whereas volumes in excess of 1 mL per nostril are not reliably absorbed as a result of the mucosal surface saturation and runoff from the nasal cavity.[1718]

The highly vascularized nasal mucosa and the olfactory tissue are in direct contact with the central nervous system that allow nasally delivered drugs to be rapidly transported into the bloodstream and brain, with onsets of action approaching that of IV therapy. First-pass metabolism via the liver is also avoided, resulting in high bioavailability of many drugs. Administration of IN medication is also relatively painless, inexpensive and easy to deliver with a minimum of training[19]

Children presenting to the MRI suite are frequently anxious. In our study, only 20.06% of the children were calm at presentation to the MRI. Securing an IV access in these children, is a difficult procedure.

Dexmedetomidine is most often administered as a continuous infusion in the intensive care unit, but now-a-days is increasingly being studied as an alternative to standard agents for IN administration.

In the present study, a relatively higher dose of both drugs were used than that used in the previously performed studies.

In children undergoing MRI, the use of IN dexmedetomidine (3 μg kg−1) or IN ketamine (7 mg kg−1) had achieved a moderate level of sedation (mean of 4.89 ± 0.85 and 4.58 ± 0.9 respectively) on Modified Ramsay sedation scale with American Academy of pediatrics/Joint Commission/ASA Designation scale suitable for pre procedural period.

In response to this level of sedation children in the two groups separated more easily from their parents (mean of 3.86 ± 0.35 and 3.75 ± 0.43 respectively).

School aged children was chosen due the possible pharmacokinetic and pharmacodynamic differences between young and older children. Vilo et al.[20] evidenced that younger children require larger bolus doses of dexmedetomidine to achieve satisfactory sedation due to pharmacokinetic factors.

The target patient population was chosen upon findings of Yuen et al.,[21] who randomized 116 children between 1 and 8 years of age to receive an IN dexmedetomidine dose of either 1 μg kg−1 or 2 μg kg−1 as a pre-induction sedative. Satisfactory sedation at the time of induction was observed in 53% of the children in the lower dose group versus 66% in the higher dose group (P = 0.049). The difference between the two doses was more obvious in the children between 5 and 8 years of age, where the 2 μg kg−1 dose resulted in significantly more patients achieving satisfactory sedation, than in the 1-4 year olds who responded the same to both doses. Our results nearly matched with this results in regard to percentage of adequate moderate sedation (B score = 69%) at 30 min [Table 5].

Yuen et al.[8] in a randomized, crossover evaluation of healthy adult volunteers, demonstrated that intransal 1 and 1.5 μg kg−1 dexmedetomidine produces sedation in 45-60 min and peaks in 90-105 min. In addition, they observed only a modest reduction of HR and arterial blood pressure (BP). In a follow-up study these authors found the onset of action of IN dexmedetomidine to be 25 min and duration of action to be 85 min.[22]

Many authors[2223] investigated the pharmacokinetics of IN ketamine. They documented that nasal ketamine peaks and sedation begins at 15-20 min. Furthermore[22] they found that 9 mg kg−1 of nasal ketamine resulted in plasma levels comparable to the IV induction dose used to provide deep anesthesia.

Three other studies[242526] tested various doses of intranasal ketamine ranging from 3-9 mg kg−1. They revealed statistically significant differences in levels of sedation between small and larger doses. Children who received a 9 mg kg−1 dose, were considered to have adequate sedation in contrast to those received 1-3 mg kg−1 had unsatisfactory sedation.

Intranasal ketamine at 6 mg kg−1 dose was evaluated by Pandey et al.,[27] for sedation in uncooperative children undergoing dental procedures and documented that this dose resulted in a goal of adequate sedation in 91% and successful procedural efficacy in 85% of cases. These results come in agreement with our results that revealed mean sedation onset of 22.4 ± 5.6 min and 14.65 ± 4.9 min after intranasal dexmedetomidine and ketamine respectively.

Furthermore, these results explained why the percentage of sedation failure was higher in Group K and in turn need for rescue propofol.

Gyanesh et al.,[28] compared intranasal dexmedetomidine (1 μg kg−1) and ketamine (5 μg kg−1) for procedural sedation in children undergoing an MRI. There were no significant differences in the children's response to administration of the drug and IV cannulation between dexmedetomidine and ketamine. There were no significant differences in adverse effects among the groups and no serious events. These results agreed with our findings revealing that intranasal administration of dexmeditomidine (3 μg kg−1) or ketamine (7 mg kg−1) to be very effective in providing adequate conditions for the placement of the IV cannula. Most of the children accepted the IN instillation of the drugs with minimum discomfort. Both groups showed insignificant differences regarding sedation scale at 30 min point, parents’ separation and acceptance of IV cannulation (P > 0.05).

There was a significant difference in HR and SBP between the two groups starting from the point of 20 min. In Group D, the HR and SBP were significantly decreased at 30 min value when compared with basal readings (P < 0.05). These findings required no treatment and we attributed it to relatively large dose of dexmedetomidine.

One limitation of this study was that we did not measure parents’ and radiologist satisfaction. Another limitation may lie with the method of IN drug delivery. We instilled the study medications with the use of tuberculin syringes. Sprayed or atomized IN medication delivery is a recent technique. Use of these methods of drug delivery may lead to more bioavailability and higher satisfaction with the use of IN medications in children.

CONCLUSION

IN administration is pain free and more acceptable to children. IN dexmedetomidine 3 μg kg−1 or ketamine 7 mg kg−1 can be used safely and effectively to induce a state of moderate conscious sedation and to facilitate parents’ separation and IV cannulation. Addition of midazolam in a dose not sufficient alone to produce the target sedation achieved our goal of deep level of sedation suitable for MRI procedure.