Procedural sedation in children: Selection of medications

INTRODUCTION — This topic will discuss the selection of medications for emergency and scheduled (elective) pediatric procedural sedation. The properties of agents commonly used for procedural sedation in children; the assessment, preparation, and proper performance of pediatric procedural sedation in children are discussed separately:

●(See "Pediatric procedural sedation: Pharmacologic agents".)

●(See "Procedural sedation in children: Preparation".)

●(See "Procedural sedation in children: Approach".)

TOPIC SCOPE — This topic provides guidance for emergency and scheduled (elective) pediatric procedural sedation performed by sedation providers from a variety of disciplines including general and pediatric emergency medicine, pediatric critical care medicine, pediatric hospital medicine, and general pediatrics.

Recommendations for procedural sedation in children undergoing gastrointestinal endoscopy, bronchoscopy, cardiac catheterization, or interventional radiology are beyond the scope of this topic.

NONPHARMACOLOGIC INTERVENTIONS — Nonpharmacologic interventions include behavioral and cognitive approaches, such as desensitization, distraction, reinforcement of coping skills, positive reinforcement, and relaxation. These techniques are complementary to pharmacologic interventions and, in some children, may prevent the need for sedation altogether. (See "Procedural sedation in children: Approach", section on 'Nonpharmacologic interventions'.)

CHOICE OF SEDATIVE AGENTS — The targeted depth of sedation and the agents used largely depend upon the anticipated degree of pain, the allowable amount of motion during the procedure, and the following patient factors [1,2] (see "Procedural sedation in children: Preparation", section on 'Pre-sedation evaluation'):

●Comorbidities (eg, asthma, upper respiratory tract infection)

●Fasting status

●Age and development level

●Ability to cooperate

●Degree of anxiety

●Any prior problems with specific medications

The dosing, administration, and properties of commonly used medications for pediatric sedation are listed in the tables and discussed in detail separately (table 1 and table 2). (See "Pediatric procedural sedation: Pharmacologic agents".)

In some facilities, the use of specific sedatives may be restricted to use by anesthesiologists or other specialists (eg, pediatric critical care or pediatric emergency medicine specialists). For example, in some settings, propofol is only approved for use by anesthesiologists or others with specialized pediatric procedural sedation training. Otherwise, bolus propofol use for procedural sedation in children requires special privileging. The clinician should be aware of local requirements.

Safety requirements — Safe use of sedative agents outside of the operating room requires careful attention to the following factors:

●Selection and preparation of patients – The patient should be evaluated and prepared for sedation in accordance with guidelines designed to maximize patient safety. This assessment, includes an American Society of Anesthesiologists (ASA) physical status classification (table 3) and helps to identify patients at higher than normal risk for sedation. (See "Procedural sedation in children: Preparation", section on 'Pre-sedation evaluation' and "Procedural sedation in children: Preparation", section on 'Preparation'.)

Those patients with ASA class IV, and V, functional needs, or airway abnormalities warrant consultation with a pediatric anesthesiologist. These patients are at increased risk for sedation-related adverse events and should be cared for by individuals who are specifically trained and experienced with high-risk pediatric procedural sedation. For elective procedures, high-risk patients should ideally receive sedation in an organized pediatric sedation service that has ancillary personnel with sedation expertise and an ongoing quality improvement program.

●Competency of the sedation provider – Health care providers who perform procedural sedation in children should have strong resuscitation and advanced pediatric life support skills, including advanced training in the assessment and management of the pediatric airway as well as specific training in pediatric procedural sedation. (See "Procedural sedation in children: Approach", section on 'Performing procedural sedation'.)

●Institutional oversight – Institutions should develop guidelines and requirements that support the safe practice of pediatric sedation, including credentialing of individual providers and measurement and assessment of the quality of sedation care. In settings where pediatric procedural sedation is routine (eg, children's hospitals), a pediatric sedation service can provide efficient and effective sedation and is a systematic support for the safe performance of sedation. (See "Procedural sedation in children: Approach", section on 'Adverse outcomes'.)

SEDATION FOR IMAGING STUDIES — Imaging can often be performed without sedation in older cooperative children and young infants (up to six months of age) who are bundled and recently fed. Furthermore, as newer technologies decrease the image capture time for computed tomography (CT), some younger infants and many uncooperative patients can be imaged without sedation [3].

However, a significant number of older infants, toddlers, and older children with intellectual disability cannot cooperate even for brief imaging tests (eg, helical CT) and warrant sedation to ensure accurate imaging without excessive radiation exposure. Thus, imaging tests that are negatively impacted by motion (eg, noninterventional CT or magnetic resonance imaging [MRI]) constitute the most common nonpainful procedures for which children undergo sedation [4].

Ideally, the chosen agent or agents should have a quick onset of action that permits successful and safe completion of the imaging study, maintains airway reflexes, and has limited impact on breathing and hemodynamic stability [5]. The regimen should also permit rapid recovery with few side effects, such as nausea or agitation. Because imaging studies are not painful, analgesia is not necessary.

Anesthesia for children undergoing MRI or CT is discussed separately. (See "Anesthesia for magnetic resonance imaging and computed tomography procedures".)

Computed tomography

Approach — Successful imaging with helical CT is less sensitive to patient movement than MRI and, given the rapid speed of imaging, can frequently be done without sedation or requires only brief sedation (approximately 5 to 10 minutes). As an example, in one case series of 104 patients undergoing helical CT, only 9 percent received sedation [6].

When sedation is required, the clinician has a choice of several different agents as discussed below.

The intravenous (IV) route is preferred where maximal efficiency and throughput is desirable, such as patients undergoing urgent imaging for diagnostic purposes, because the onset of action is shorter and more predictable, titration is easier, and (depending upon the agent), recovery is quicker (table 1) [5]. (See 'Intravenous medications' below.)

Other routes of administration are feasible for children without IV access, especially in settings where CT imaging is elective and longer sedation duration and recovery times are acceptable (table 2). However, if inadequate sedation or serious adverse events occur, lack of IV access may impede further management [7]. Thus, we prefer to provide IV sedation whenever possible. (See 'No intravenous access' below.)

Intravenous medications — We recommend that healthy infants and children (American Society of Anesthesiologists [ASA] class I or II) (table 3) who have vascular access and are undergoing sedation for CT receive IV propofol, dexmedetomidine, ketamine, or etomidate rather than IV short-acting barbiturates (eg, pentobarbital or methohexital) or midazolam (table 1).

Evidence identifying the best sedative agent for children undergoing CT is limited. Based upon observational studies, IV propofol, dexmedetomidine or etomidate is associated with effective sedation in over 99 percent of children undergoing CT compared with 97 to 98 percent of children receiving short-acting barbiturates or midazolam (table 1) [5,8-12]. Although lower efficacy was described for IV etomidate (76 percent of 17 patients) in one small trial, more than 99 percent of 446 children receiving etomidate for imaging successfully completed the procedure in a subsequent observational study [8,9,13]. Based upon a review of prospectively collected registry data from over 22,000 children, IV ketamine is also frequently used for imaging; radiologic procedures accounted for almost 25 percent of reported use in the study [13]. Although not specifically reported for radiologic procedures, 99.8 percent of all procedures were successfully completed which suggests that IV ketamine has similar efficacy for CT when compared with propofol, etomidate, or dexmedetomidine. However, when used as a single agent, ketamine may be associated with random movement in some patients and may not be ideal.

IV ketamine, etomidate and propofol have a more rapid onset of action than dexmedetomidine (1 to 2 minutes versus 8 to 16 minutes, respectively) and etomidate has a shorter duration of action than dexmedetomidine (approximately 15 minutes compared with 30 minutes, respectively) [8-12]. Duration of action with ketamine and propofol varies by dose and mode of administration (bolus or continuous infusion) and compares favorably with etomidate.

Evidence is lacking to suggest that major adverse events (eg, laryngospasm, aspiration, apnea requiring airway intervention, or need for resuscitation) varies by the medication used to sedate for CT. Adverse events that are not necessarily life-threatening have been described in 2 percent of infants and children receiving etomidate or propofol sedation for imaging and 1 percent receiving dexmedetomidine. By contrast, approximately 5 percent of patients receiving IV short-acting barbiturates (eg, pentobarbital or methohexital) and up to 7 percent receiving IV midazolam experience adverse events, primarily respiratory depression with transient oxygen desaturation [8,10,14]. Less commonly apnea has been described, especially in patients receiving short-acting barbiturates.

Other adverse reactions that are specific to the agent are described in the table (table 1) and discussed separately. (See "Pediatric procedural sedation: Pharmacologic agents", section on 'Sedative-hypnotic agents'.)

No intravenous access — We prefer administration of IV medications, whenever possible, for uncooperative children undergoing CT. (See 'Intravenous medications' above.)

When an IV cannot be placed, medication options and route of administration include (table 2):

●Oral or intranasal (IN) midazolam

●Intramuscular (IM) ketamine

●IN dexmedetomidine

Midazolam by the oral or IN route achieves successful minimal sedation in 50 to 87 percent of patients undergoing CT, with higher efficacy when delivered IN [15-17]. Onset of sedation is more rapid when midazolam is administered IN (generally, less than 10 minutes) [18], while oral midazolam takes approximately 30 minutes [19]. IM ketamine may be preferred when timely imaging is necessary. Dissociative sedation with IM ketamine is associated with more adverse events and serious adverse events than minimal sedation with midazolam or dexmedetomidine. Furthermore, random motion may occur during ketamine sedation which can impair successful completion of the study. (See "Pediatric procedural sedation: Pharmacologic agents", section on 'Ketamine'.)

Limited evidence from two randomized trials suggests that IN dexmedetomidine produces adequate sedation in 60 to 74 percent of children undergoing CT. The onset of sedation (approximately 30 minutes) and time to discharge (approximately 40 minutes) appears similar to oral midazolam [20,21]. Further study is needed to determine the optimal dosing for intranasal dexmedetomidine.

Regardless of agent and route chosen, children should undergo monitoring of oxygen saturation and heart rate according to recommendations because the risk of adverse events is significant regardless of the route of administration. (See "Procedural sedation in children: Approach", section on 'Monitoring'.)

Despite their relatively high efficacy when compared with oral or intranasal midazolam or intranasal dexmedetomidine, we do not endorse the use of oral or rectal short-acting barbiturates (eg, pentobarbital or methohexital) because of their long duration of sedation relative to the short imaging time, high rates of adverse events during sedation, and potential for delayed adverse events, including prolonged sleepiness and ataxia [15-17,22-27].

Chloral hydrate is no longer recommended for sedation in children and is not available in many countries, including the United States. Some countries have removed chloral hydrate from national health formularies because of potential carcinogenicity although the risk of cancer from a single dose is inconclusive [28,29].

Magnetic resonance imaging — MRI often necessitates sedation for up to one hour. Furthermore, machine noise and lack of patient access pose additional challenges to achieving safe and effective sedation. We suggest that healthy infants and children (ASA class I or II) (table 3) undergoing MRI receive sedation using propofol, dexmedetomidine, midazolam and dexmedetomidine, or dexmedetomidine bolus followed by propofol sedation (table 1). Because it has a shorter time to recovery and discharge [30], experienced providers often administer propofol given as a bolus of 2 to 4 mg/kg followed by an infusion at 150 to 200 mcg/kg/minute. The use of dexmedetomidine reduces the dose of propofol during sedation. (See "Pediatric procedural sedation: Pharmacologic agents", section on 'Propofol'.)

Propofol may also be preferred in children with increases in pulmonary artery pressure, decreases in cardiac output, children with AV node conduction delay, or those receiving digoxin, beta blockers or other medications that slow AV node conduction.

For some providers, dexmedetomidine may be preferred to propofol for sedation of children with sleep apnea. As an example, in an observational study of 82 children undergoing MRI sleep studies, dexmedetomidine provided an acceptable level of anesthesia with fewer patients requiring an artificial airway compared with propofol (13 versus 30 percent) [31].

Sedation with continuous infusion of propofol or dexmedetomidine permits successful completion of MRI in approximately 97 to 99 percent of children when administered by experienced practitioners [12,30,32-34]. Time to sedation is longer with dexmedetomidine than with propofol (10 versus <1 minutes, respectively) [12,30,32,33]. Recovery is typically complete 20 to 30 minutes after stopping a propofol infusion and 25 to 40 minutes after ending a dexmedetomidine infusion [12,30,32,33,35,36].

However, to achieve this level of success with dexmedetomidine alone sometimes requires bolus dosing that exceeds current manufacturer recommendations (table 1) [33]. Alternatively, co-administration of midazolam can permit effective sedation at lower doses of dexmedetomidine. (See "Pediatric procedural sedation: Pharmacologic agents", section on 'Dexmedetomidine'.)

Either propofol or dexmedetomidine can be used safely for MRI sedation, propofol may cause more need for airway intervention or IV fluid administration. Dexmedetomidine can be associated with marked bradycardia. (See "Pediatric procedural sedation: Pharmacologic agents", section on 'Propofol' and "Pediatric procedural sedation: Pharmacologic agents", section on 'Dexmedetomidine'.)

Evidence from observational studies supports the combination of moderate-dose dexmedetomidine 1 mcg/kg with low-dose propofol infusion for MRI sedation. The combination appears effective and with fewer adverse effects than propofol used alone. However, a large randomized trial is needed to confirm these findings [37,38].

Although pentobarbital by the oral or IV route was used in the past to sedate children undergoing MRI, the potential for adverse events may be increased and recovery time tends to be longer with a high rate of delirium during emergence when compared with propofol or dexmedetomidine infusion (table 1) [23,39-42].

Proper administration, dosing, adverse effects, contraindications, and precautions for propofol and dexmedetomidine are discussed in greater detail separately. (See "Pediatric procedural sedation: Pharmacologic agents", section on 'Sedative-hypnotic agents'.)

SEDATION FOR OTHER NONPAINFUL PROCEDURES — In some children, physical examination (eg, genital examination to document sexual assault or routine physical examination in children with intellectual disability) or other nonpainful procedures (eg, echocardiography or electroencephalogram) can cause anxiety and lack of cooperation with the medical provider. In many situations, nonpharmacologic interventions can permit completion of the examination or test. (See "Procedural sedation in children: Approach", section on 'Nonpharmacologic interventions'.)

When nonpharmacologic interventions are not sufficient and mild sedation is necessary for nonpainful procedures, we suggest that healthy children (American Society of Anesthesiologists class I or II) (table 3) receive sedation with oral, sublingual (SL), or intranasal (IN) midazolam or IN dexmedetomidine rather than short-acting barbiturates (table 2). Inhaled nitrous oxide is also a reasonable option, although use of a nasal or face mask can be challenging in selected patients. In addition, a combination of IN ketamine and dexmedetomidine or "ketodex" (1 mg/kg ketamine and 2 mcg/kg dexmedetomidine) shows promise for sedation, at least for nonpainful procedures. For example, in two large, retrospective observational studies, this combination had a success rate of 93 to 96 percent for sedation of children undergoing nonpainful procedures [43,44]. However, further study comparing this combination with midazolam or dexmedetomidine alone, as described above, is warranted before use is routine.

Intravenous (IV) sedation as described for CT is suggested for patients who fail midazolam or dexmedetomidine sedation by the routes described above. (See 'Computed tomography' above.)

Multiple trials describe acceptable efficacy for pediatric sedation with IN dexmedetomidine for pediatric sedation that appears equivalent or superior to oral midazolam, depending upon the procedure [45-49]. In a systematic review of 12 trials that compared IN dexmedetomidine with other sedatives for nonpainful procedures, dexmedetomidine provided adequate sedation, according to validated sedation instruments, for 86 percent of 1021 children compared with 77 percent of 277 children receiving other types of sedation (eg, IN or oral midazolam or oral ketamine) and was similar to IN midazolam in two trials [49]. Adverse effects after IN dexmedetomidine administered in doses ranging from 1 to 4 mcg/kg included bradycardia (2.2 percent), hypotension (1.2 percent), oxygen desaturation (0.5 percent), and vomiting (0.4 percent); none required advanced life support.

Midazolam has both anxiolytic and amnestic properties. After oral or SL administration, it has an onset of action of 5 to 10 minutes with recovery occurring in approximately 60 minutes [50]. Onset of action with IN midazolam administration is typically more rapid than oral administration but duration of sedation is shorter (20 to 30 minutes) [51]. Due to the low pH and the benzyl alcohol preservative, IN midazolam can be irritating when administered. Pretreatment with lidocaine spray (10 mg per puff) one minute prior to IN midazolam administration decreases nasal mucosal irritation. An atomizer is preferred for the IN delivery of midazolam because of more reliable nasal mucosa absorption, better comfort, and reduction of sneezing and cough when compared with direct instillation [51]. Although midazolam may be administered rectally, this route is less reliable.

Midazolam has good efficacy for nonpainful procedures that permit motion [50,51] and has a shorter duration of action than oral or rectal pentobarbital (<1 hour versus 1 to 4 hours, respectively). Unlike barbiturates, midazolam is not associated with prolonged symptoms of ataxia, sleepiness, or irritability (table 2). (See 'No intravenous access' above.)

Flumazenil is an effective reversal agent for the few patients who develop significant respiratory depression or apnea after sedation with midazolam. Flumazenil should not be used in patients with seizure disorders or those who receive benzodiazepines on a chronic basis because of the risk of precipitating seizures or withdrawal symptoms, respectively. The use of flumazenil to reverse adverse effects of benzodiazepines, including dosing and re-dosing recommendations is discussed in detail separately. (See "Benzodiazepine poisoning", section on 'Role of antidote (flumazenil)'.)

SEDATION FOR PAINFUL PROCEDURES — Clinicians frequently employ procedural sedation for infants and children undergoing a variety of painful procedures including fracture reduction, laceration repair, bone marrow aspiration, central line placement, and lumbar puncture. For these procedures, chosen agents or combinations of agents must safely provide sedation and analgesia.

Approach — Patient factors (eg, last oral intake, urgency of the procedure, prior sedation experience, and comorbidities [eg, asthma, upper respiratory infection]) are key considerations for sedation strategies in children undergoing painful procedures and are discussed in detail separately. (See "Procedural sedation in children: Preparation", section on 'Pre-sedation evaluation'.)

In healthy infants and children, anticipated pain during the procedure is also an important determinant of the depth and type of sedation:

●Minimally painful procedures (eg, peripheral intravenous [IV] cannula insertion or laceration repair in regions of the body where occasional movement does not interfere with the procedure) – Minimal sedation, often best achieved with midazolam or inhaled nitrous oxide. (See 'Minimally painful procedures' below.)

●Moderate to severely painful procedures (eg, fracture reduction) or laceration repair where movement will interfere significantly with performance of the procedure) – Deep sedation, typically reached with ketamine alone, propofol alone (brief procedures), ketamine combined with propofol, or propofol combined with fentanyl. The provider should be aware that any of these regimens can produce general anesthesia depending upon the initial dose and frequency of re-dosing. (See 'Moderately or severely painful procedures' below.)

Effective local or regional anesthesia can often lower the amount of sedative agent needed to provide adequate sedation and increase the safety of the procedure.

The need for supplementary analgesia varies by the agents used for sedation:

●Ketamine has both sedative and analgesic properties and can thus be used alone to provide sedation for moderate to severely painful procedures. However, for wound repair, local anesthetic (eg, topical lidocaine-epinephrine-tetracaine [LET] gel or infiltration of a local anesthetic) or a regional nerve block is also typically used, especially if the repair is estimated to take longer than 10 to 15 minutes. (See 'Moderately or severely painful procedures' below.)

●Dexmedetomidine and nitrous oxide have limited analgesic properties that may be inadequate and warrant additional analgesic medications for moderately or severely painful procedures.

●Propofol, midazolam, and etomidate do not have analgesic properties and should be combined with other analgesic agents. Although propofol is used alone for brief, painful procedures by some practitioners, the doses necessary to achieve adequate analgesia and control unwanted motion during the procedure may approach those used for general anesthesia. Propofol alone may be appropriate when local or regional anesthesia provides complete pain control. (See 'Moderately or severely painful procedures' below.)

Although options vary, typical approaches to analgesia by type of procedure include:

●Wound repair

•Topical anesthetic (eg, lidocaine-epinephrine-tetracaine) (see "Clinical use of topical anesthetics in children")

•Local infiltration of anesthetic (see "Subcutaneous infiltration of local anesthetics" and "Subcutaneous infiltration of local anesthetics", section on 'Procedure')

•Regional anesthesia (eg, lacerations of the face, fingertip, distal toe, plantar foot, or penis) as described separately:

-Facial lacerations (see "Assessment and management of facial lacerations", section on 'Facial nerve blocks')

-Fingertip and distal toe (see "Digital nerve block", section on 'Digital block procedures')

-Plantar foot (see "Lower extremity nerve blocks: Techniques", section on 'Ankle block')

-Penis (see "Management of zipper entrapment injuries", section on 'Dorsal penile block')

●Closed reduction of fractures

•Opioid analgesia (eg, fentanyl or morphine; IV opioids have been associated with an increased risk of oxygen desaturation, vomiting, or need for positive-pressure ventilation, especially when administered less than 30 minutes prior to sedation) (see 'Moderately or severely painful procedures' below)

•Hematoma block (distal radial fractures) (see "Distal radius fractures in adults", section on 'Performance of hematoma block')

•Bier block (fractures of the hand or forearm) (see "Overview of anesthesia", section on 'Intravenous regional anesthesia')

•Regional anesthesia (eg, pain control for femoral fractures) (see "Lower extremity nerve blocks: Techniques")

●Lumbar puncture or bone marrow aspiration – Topical anesthetic followed by local infiltration of anesthetic (see "Clinical use of topical anesthetics in children" and "Subcutaneous infiltration of local anesthetics")

Minimally painful procedures — In many children undergoing IV cannula insertion or laceration repair, local anesthetics can be delivered topically or by direct infiltration to diminish or abolish the pain without the need for sedation, especially when age-appropriate nonpharmacologic interventions are used. (See 'Nonpharmacologic interventions' above and "Clinical use of topical anesthetics in children", section on 'Lidocaine-prilocaine' and "Subcutaneous infiltration of local anesthetics".)

When nonpharmacologic interventions and local anesthetics are not sufficient and minimal sedation is necessary for minimally painful procedures (eg, peripheral IV line placement, urethral catheterization, small laceration repair, or nasogastric tube placement), we suggest that healthy children (American Society of Anesthesiologists [ASA] class I or II) (table 3) receive sedation with inhaled nitrous oxide (N₂O); oral, sublingual, or intranasal (IN) midazolam; or IN dexmedetomidine rather than short-acting barbiturates (table 2). When N₂O is available and tolerated by the patient, it is preferred to midazolam as summarized below. IN dexmedetomidine has been used for minimal sedation but requires more study before best practice guidelines for administration are determined.

When compared with N₂O, IN midazolam, or IN dexmedetomidine, oral or rectal barbiturates are associated with higher rates of adverse effects, especially respiratory depression, (thiopental or methohexital) or prolonged duration of sedation (pentobarbital) (table 2).

●Nitrous oxide – Because it has a shorter recovery time and fewer overall adverse effects, nitrous oxide (N₂O) may be preferred to midazolam in cooperative children for the following reasons [52-55]:

•The onset of action is short (<1 minute).

•Recovery is typically faster (<20 minutes versus 30 to 60 minutes, respectively).

•For selected procedures, N₂O may also have greater efficacy. For example, in a trial, of over 250 children undergoing facial laceration repair, N₂O or N₂O combined with midazolam decreased distress to a greater extent than midazolam alone or lidocaine with comforting [52]. In another trial of 90 children in whom IV access was expected to be difficult, use of N₂O was associated with a significantly increased success rate for IV placement when compared with oral midazolam [53]. Procedure time was also significantly shorter for patients who received 50 percent N₂O when compared with oral midazolam.

•Adverse events following N₂O use are typically self-limited with vomiting being most common (approximately 6 to 7 percent of patients, primarily those with longer duration of inhalation or who also received opioids) [52,54-56]. Serious adverse effects (eg, postprocedure desaturation, apnea, stridor, aspiration, laryngospasm, or airway obstruction) are rare (eg, 0.2 percent in two observational studies and one case report of aspiration and laryngospasm) [55-57]). Thus, N₂O use requires the same level of preparation and vigilance as other sedative agents. (See "Procedural sedation in children: Preparation" and "Procedural sedation in children: Approach", section on 'Performing procedural sedation'.)

The method for administering N₂O for procedural sedation and additional precautions are discussed in greater detail separately. (See "Pediatric procedural sedation: Pharmacologic agents", section on 'Nitrous oxide'.)

●Midazolam – Based upon one large observational study, up to one-third of patients, especially children younger than 4 years of age, may not tolerate N₂O delivery [58]. Such patients are good candidates for oral or IN midazolam [54]. Midazolam is also a reasonable alternative when N₂O is not available. If the IN route of administration is chosen, pretreatment with lidocaine spray (10 mg per puff) one minute prior to administration and use of an atomizer is suggested [51].

●Dexmedetomidine – Preliminary evidence suggests that IN dexmedetomidine may be equivalent to midazolam for minimally painful procedures. As an example, in a trial of 40 young children (mean age three years) undergoing laceration repair, intranasal (IN) dexmedetomidine (2 mcg/kg) was similar to IN midazolam (0.4 mg/kg) for anxiolysis during repair but was associated with less anxiety during positioning with significantly more children showing no anxiety at this time point (70 versus 11 percent) [48]. Although evidence for safety and efficacy for this specific use of IN dexmedetomidine in children is limited, results from small trials where IN dexmedetomidine was used for other purposes (eg, nonpainful procedures, dental procedures, or preoperative sedation) support safety and efficacy that is potentially equivalent to midazolam [49,59-64]. Larger studies are needed to determine optimal dosing and to identify more infrequent side effects.

Moderately or severely painful procedures — For healthy infants and children (ASA class I or II) (table 3) who are undergoing moderately or severely painful procedures of short duration (eg, fracture reduction, bone marrow aspiration), we suggest procedural sedation with IV ketamine, ketamine combined with propofol, or fentanyl combined with propofol rather than opioids combined with benzodiazepines (eg, midazolam) or etomidate. When ketamine is used, a prophylactic dose of ondansetron may reduce postprocedure vomiting. For selected procedures, effective regional anesthesia can reduce the required medication dose or eliminate the need for moderate or deep sedation altogether. (See "Pediatric procedural sedation: Pharmacologic agents", section on 'Ketamine'.)

Dosing and administration of specific IV sedation drugs used for moderate, dissociative, or deep sedation are provided in the table (table 1) and discussed separately. (See "Pediatric procedural sedation: Pharmacologic agents".)

Based upon observational studies and randomized trials, similar sedation efficacy for moderate to severely painful procedures, such as fracture reduction or bone marrow aspiration, is achieved with ketamine alone, ketamine combined with propofol, or fentanyl combined with propofol [65] compared with regimens that combine opioids (typically fentanyl) with midazolam [66-68] or etomidate [69,70]. All of the above IV regimens have a rapid onset of effect ranging from 30 seconds for etomidate and propofol to 1 to 2 minutes for ketamine alone or midazolam with fentanyl (table 1). Recovery from sedation using ketamine alone or ketamine with propofol ranges from 15 to 30 minutes which is longer than for etomidate or propofol (5 to 15 minutes) but shorter than for midazolam with fentanyl (15 to 60 minutes). Providers using propofol should have specific training in the use of propofol and the provision of deep sedation or anesthesia [71].

The IV regimens vary in terms of the frequency of important adverse events (see "Pediatric procedural sedation: Pharmacologic agents"):

●With recommended dosing, ketamine sedation tends to preserve airway reflexes and is associated with fewer airway or respiratory adverse events that require intervention than fentanyl combined with propofol, midazolam, or etomidate. On the other hand, vomiting and agitation during emergence are more frequent following ketamine sedation than with the other regimens although avoidance of high dosing (≥2.5 mg/kg initial dose or total dose ≥5 mg/kg) may reduce the frequency of these adverse events [72] and prophylactic ondansetron significantly reduces vomiting [73]. Hypotension, which occurs frequently during sedation with propofol, is not described following ketamine sedation. (See "Pediatric procedural sedation: Pharmacologic agents", section on 'Ketamine'.)

●The combination of propofol and ketamine for procedural sedation appears to provide effective sedation and less vomiting than reported for ketamine alone and less hypotension than described with propofol alone. However, adverse respiratory events including laryngospasm can still occur [74]. Suggested dosing for this regimen is provided separately. (See "Pediatric procedural sedation: Pharmacologic agents", section on 'Ketamine'.)

●Propofol combined with fentanyl is preferred by some sedation practitioners because it is equally efficacious to ketamine or ketamine combined with propofol and may produce a shorter and smoother recovery [75]. (See "Pediatric procedural sedation: Pharmacologic agents", section on 'Propofol'.)

●Nitrous oxide alone for these procedures is associated with high levels of responsiveness to pain (up to 40 percent of patients) and need for restraint (up to one-third of patients) and is not recommended [58].

Appropriate analgesia is an important adjunct to sedation that is frequently necessary for children undergoing moderate to severely painful procedures. However, administration of IV opioids prior to sedation is associated with increased frequency of oxygen desaturation, vomiting, and need for positive-pressure ventilation during sedation although the absolute effect on vomiting or the need for positive-pressure ventilation is small. The provider should anticipate and be prepared to manage these adverse events and, when possible, administer alternative, nonopioid analgesics, especially within 30 minutes prior to sedation. For example, in a prospective, multicenter, observational study of almost 6300 children undergoing sedation for painful procedures in the emergency department, opioid administration prior to the procedure versus no opioid analgesia was associated with an increased risk of oxygen desaturation (9 versus 4 percent), vomiting (7 versus 5 percent), and need for positive-pressure ventilation (1.5 versus 0.9 percent) [76]. These risks were greatest when opioid analgesia was administered closer to the time of sedation (eg, less than 30 minutes). Compared with morphine, fentanyl was associated with decreased odds of oxygen desaturation (odds ratio [OR] 0.5, 95% CI 0.3-0.8) and vomiting (OR 0.5, 95% CI 0.34-0.95), but this finding requires confirmation in randomized trials.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Procedural sedation in children".)

SUMMARY AND RECOMMENDATIONS

●Topic scope – This topic provides guidance for emergency and scheduled (elective) pediatric procedural sedation performed by sedation providers from a variety of disciplines including general and pediatric emergency medicine, pediatric critical care medicine, pediatric hospital medicine, and general pediatrics. (See 'Topic scope' above.)

●Nonpharmacologic interventions – Techniques such as distraction, desensitization, reinforcement of coping skills, positive reinforcement, and relaxation are complementary to pharmacologic interventions and, in some children, may prevent the need for sedation altogether. (See "Procedural sedation in children: Approach", section on 'Nonpharmacologic interventions'.)

●Choice of sedative agents – The targeted depth and the agents used for pediatric procedural sedation depend on (see 'Choice of sedative agents' above):

•The specific procedure, especially the anticipated degree of pain and length of the procedure

•Allowable patient movement during the procedure

•Patient factors, especially age/developmental level, degree of anxiety, ability to cooperate, fasting status, airway assessment, and ASA classification (table 3)

All drug combinations used to provide targeted deep sedation for painful procedures (eg, propofol alone, propofol and dexmedetomidine, propofol and fentanyl) can result in general anesthesia depending upon the initial dose(s) and frequency of redosing. Providers must always be prepared to provide rescue interventions (eg, bag-mask ventilation for apnea) for these patients.

●Safety – The safe performance of pediatric sedation requires proper selection and evaluation of patients, appropriate preparation, and the performance of sedation by providers with appropriate training and institutional support in accordance with guidelines designed to maximize patient safety. (See 'Safety requirements' above.)

●Analgesia – Local or regional anesthesia and/ or opioid analgesia are important adjuncts for patient comfort and can decrease the depth or total dose of sedative needed to successfully complete the procedures. However, IV administration of opioids prior to sedation is associated with increased frequency of oxygen desaturation, vomiting, and need for positive-pressure ventilation during sedation and the provider should be prepared to manage these adverse events. (See 'Moderately or severely painful procedures' above.)

●Medications – The properties of the specific agents and dosing for pediatric sedation are listed in the tables (table 1 and table 2) and discussed in detail separately. (See "Pediatric procedural sedation: Pharmacologic agents".)

●Imaging studies – Imaging can often be performed without sedation in older cooperative children and young infants (up to six months of age) who are bundled and recently fed. When necessary, the selection of medications for these procedures are determined by the duration of the test and whether vascular access is present (see 'Sedation for imaging studies' above):

•Noninterventional CT – For healthy infants and children (American Society of Anesthesiologists [ASA] class I or II (table 3)) who have vascular access and are undergoing sedation for CT, we recommend IV bolus doses of propofol, dexmedetomidine, ketamine, or etomidate rather than IV short-acting barbiturates (eg, pentobarbital or methohexital) or midazolam (Grade 1B). When used as a single agent, ketamine may be associated with random movement in some patients and may not be ideal. (See 'Intravenous medications' above.)

We prefer administration of IV medications, whenever possible, for uncooperative children undergoing CT. When an IV cannot be placed, medication options and route of administration include intramuscular (IM) ketamine, oral or intranasal (IN) midazolam, and IN dexmedetomidine. However, successful completion of the procedure is lower than for IV agents. (See 'No intravenous access' above.)

•MRI – For healthy infants and children (ASA class I or II) undergoing MRI, we suggest continuous IV infusion of propofol, dexmedetomidine, midazolam and dexmedetomidine, or dexmedetomidine bolus followed by propofol sedation (Grade 2C). (See 'Magnetic resonance imaging' above.)

●Other nonpainful diagnostic procedures – For healthy children undergoing minimal sedation for other nonpainful (eg, echocardiography or electroencephalogram), we suggest IN midazolam or IN dexmedetomidine rather than short-acting barbiturates (Grade 2C). While oral or sublingual midazolam may also be efficacious, longer onset of action and prolonged duration of action will limit its use in many settings. Nitrous oxide (N₂O) is also a reasonable option when mask delivery is tolerated. (See 'Sedation for other nonpainful procedures' above.)

●Minimally painful procedures – When nonpharmacologic interventions and topical anesthetics are not sufficient and mild sedation is

necessary for minimally painful procedures (eg, IV cannula placement), we suggest that healthy children (ASA class I or II) receive inhaled nitrous oxide (N₂O); oral, sublingual, or IN midazolam; or IN dexmedetomidine (Grade 2C). IV sedation as described for CT is suggested for patients who fail sedation with these agents. (See 'Minimally painful procedures' above and "Pediatric procedural sedation: Pharmacologic agents", section on 'Nitrous oxide'.)

●Moderate or severely painful procedures – For healthy infants and children (ASA class I or II) who are undergoing moderately or severely painful procedures of short duration (eg, fracture reduction or bone marrow aspiration), we suggest IV ketamine, ketamine and propofol, or propofol and fentanyl rather than opioids combined with benzodiazepines (eg, midazolam) or etomidate (table 1) (Grade 2C). When ketamine is used, a prophylactic dose of ondansetron may reduce postprocedure vomiting. (See 'Moderately or severely painful procedures' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Deborah C Hsu, MD, MEd, who contributed to earlier versions of this topic review.