The body of a child must be seen as a developing structure that is not anatomically and physiologically completed before the end of puberty and even in the early years of adulthood. The most evident differences pertain to the changes in size, from a length of about 50 cm and a weight of 3 to 3.5 kg at birth to more than 150 cm and 50 kg (often significantly more) at the end of the pediatric period. These considerable increases in size are neither continuous nor proportional; the relative growth of the trunk and limbs is greater than that of the head. This has marked implications in regional anesthesia because the anatomic landmarks, especially surface landmarks, are constantly changing and must be precisely identified before any block procedure is undertaken. Ultrasound technique may help considerably in locating bony structures and nerve paths (1,2). The use of surface electrolocation techniques, such as surface mapping (3) and percutaneous electrode guidance (4,5), might help in overcoming difficult identification of anatomic landmarks, especially in infants and disabled children with limb and joint malformations.

Another distinctive feature of the pediatric period is the presence of congenital malformations, genetic disorders, and the consequences of fetal/neonatal asphyxia. These conditions impair normal development, and many deformities of bone and joint, as well as neurologic structures, worsen during childhood. These patients usually require repeated orthopedic surgeries; regional block procedures are particularly useful in these indications but anatomic deformities can make the techniques particularly difficult and might expose the patient to complications not usually seen in “normal children,” especially in regard to pressure point complications at unusual locations and postoperative monitoring of neurologic functions.

Before birth, during the embryonic period, the spinal cord entirely occupies the spinal canal. During fetal life, the growth of the vertebral column exceeds that of neural structures; thus, the spinal cord and roots are encased within the spinal canal, and the lower end of the spinal cord, as well as that of the dural sac, project at progressively higher levels, L4 and S3/S4, respectively, up to the ninth month of life. By the end of the second year, adult levels are reached, namely T12–L1 and S2, respectively. As a consequence, approaches to the epidural space above L4 should be avoided in small infants to prevent any potential damage to the underlying spinal cord. Also, it should be remembered that, in infancy, the dural sac ending is close to the sacral hiatus; cephalad introduction of the block needle during caudal anesthesia must but limited to 2 to 3 mm to avoid accidental dural puncture.

Myelinization begins during the fetal period in cervical neuromeres and extends progressively both cephalad and caudad (6,7). This process is not fully completed until the end of the twelfth year of life. This lack of myelin in young patients makes the penetration of local anesthetic easier within the nerve fibers. Diluted solutions of local anesthetic can produce consistent
nerve blockade since the nerve fibers are thinner and shorter, which reduces the distance between successive nodes of Ranvier.

At birth, the skeleton is mostly cartilaginous. This has three practical implications: (a) cartilaginous structures are poorly seen on x-ray films, but ultrasonography can identify cartilage and differentiate it from bone, and visualize underlying neuraxial structures; (b) cartilage offers little resistance to sharp needles; and (c) undamaged ossification nuclei are necessary to allow normal bone growth and ossification. As a consequence, attempts to make boney contact should be avoided as often as possible in children, especially in infants, and use of sharp needles should be avoided; they may severely damage ossification nuclei and adjacent organs or vessels because they may not be halted by weak cartilaginous bone.

Development of the spine is not complete at the time of birth. It is gender-related, developing earlier in female than in male fetuses (8). The vertebral arches fuse posteriorly during the first year of life, and their junction with the vertebral body becomes ossified from the third to the sixth year (9,10). This posterior fusion remains incomplete at the level of S5 and also S4 most of the time, demarcating a V-shaped aperture, the sacral hiatus, which is covered by the sacrococcygeal membrane, a continuation of the ligamentum flavum, which is tightly adherent to different sacral ligaments. This strong membrane gives immediate access to the epidural space and can easily be penetrated by a block needle to provide caudal anesthesia.

Another peculiarity of the pediatric spine is that the sacrum is still made up of separate vertebrae, the fusion and ossification of which will not be achieved before adulthood (around the twenty-fifth year of life). Consequently, the epidural space can be approached at sacral levels in the same way as it can be at lumbar levels. On the other hand, since the sacrum is cartilaginous, it can be easily traversed by sharp needles, which can damage ossification nuclei (traumatic and infectious lesions) and pelvic/retroperitoneal organs. It should also be remembered that walking is necessary to the normal fusion of the sacral pieces. Patients confined to bed, especially those suffering from cerebral palsy, keep their intervertebral sacral discs unfused.

Delayed ossification of the iliac crest results in lowering of the crystal line (Tuffier line), which is used to locate the L4–L5 interspace to perform a lumbar epidural. In neonates and infants, this line passes one interspace lower; that is, at the level of the L5–S1 interspace (Fig. 27-1).

At birth, the spine has a single regular flexure. With the acquisition of head support, the cervical lordosis develops progressively by the sixth month, while the lumbar lordosis appears with the acquisition of stable sitting position between 9 and 12 months of life. This has implications for the performance of an epidural block depending on the spinal level. At birth, the orientation of the epidural needle is the same regardless of the intervertebral space. With the development of the spinal flexures, the orientation of the epidural needle must be adapted accordingly. At age 1, the spine has acquired its definitive shape, and the orientation of epidural needles should be the same as that suitable in adult patients at the same intervertebral space. However, these flexures can be easily emended by forced flexion almost throughout childhood due to the long-lasting persistence of spine flexibility, a major benefit of the pediatric period (in addition to the absence of osteophytes).

Ability to walk is necessary to acquire normal flexures of the spine. Patients with cerebral palsy usually develop severe deformities of the spine, mainly scoliosis, which often necessitate surgical treatment and make epidural anesthesia particularly challenging, occasionally impossible.

An important feature of the pediatric period is the loose attachment of fascias and sheaths to nerve trunks and muscles. This allows extended spread of local anesthetic, resulting in high-quality nerve blockade whatever the technique used. This extended diffusion can occasionally result in unwanted or even undesirable distant nerve blocks. This also applies to the sheaths surrounding the spinal roots. The younger the patient, the greater the distal spread of the local anesthetic injected epidurally (caudal or lumbar epidural approach) along the spinal roots. Due to this extended spread, in addition to the incomplete myelination of nerve fibers, the quality of blockade is improved in relevant dermatomes and allows the use of lower concentration of local anesthetics, thus reducing the danger of systemic toxicity. However, leakage along spinal roots has also negative consequences: comparatively larger volumes of local anesthetics are necessary in infants to achieve the same cephalad level as in adults. This may be an incitation to administering potentially toxic amounts of local anesthetics or to attempting hazardous techniques such as distant insertion of epidural catheters (especially via the caudal route) to achieve satisfactory blockade.

Another factor affecting the spread of epidural solution is the consistency of the epidural fat. Up to 6 to 8 years of age, this fat is very fluid. After that age, it becomes more densely packed, with an increased proportion of elastic and collagenous fibers that can significantly impede the distribution of the local anesthetic and the progression of epidural catheters. Caudal anesthesia becomes less reliable after age 6, and the insertion of a distant epidural catheter becomes very hazardous and often impossible.

Not only is the longitudinal spread of local anesthetic increased in children, the diffusion process across nerve sheaths is increased too. In young patients, the structure of the endoneurium is loose, which makes it easily traversed in both directions (“fast in, fast out”). As the child grows, the
endoneurium becomes enriched with connective fibers, thus less permeable. The latency and also the duration of nerve blockade increase with age. The most significant characteristics of the pediatric patient that can influence the selection or performance of a regional block procedure are summarized in Table 27-1.

Although the pharmacodynamics of local anesthetics is similar in children and adults, pharmacokinetic properties differ significantly, especially in neonates and infants (12).

The main pharmacokinetic difference between infants and older patients in regard to opioids (mainly morphine) pertains to the elimination half-life, which is consistently increased in infants. After lumbar epidural injection, peak plasma concentration of morphine is reached within 10 minutes. However, following therapeutic epidural or spinal doses, this plasma concentration is far too low to play a significant role in the development of analgesia, which depends only on the drug crossing the dura mater (43,44). The elimination half-life from cerebrospinal fluid (CSF) is similar to that of plasma and follows the same monoexponential elimination curve. As peak CSF concentration of morphine is very high shortly after an epidural injection, it takes 12 to 24 hours before the concentration falls below the minimal effective concentration (10 ng/mL). Therefore, adverse effects, including respiratory depression, can be observed more than 12 hours after an epidural injection, thus requiring close monitoring of vital parameters 24 hours after the last injection. Maximum recommended doses are 30 μg/kg for epidural morphine and 10 μg/kg for intrathecal morphine (Table 27-4). Patients given neuraxial morphine are not eligible for outpatient surgery (see also Chapter 40).

Short-acting lipid-soluble opioids such as fentanyl (1–2 μg/kg) and sufentanil (0.5 μg/kg) can be administered epidurally instead of morphine. They improve the quality of intraoperative analgesia, especially when coadministered with diluted local anesthetic solution. However, they do not prolong significantly the duration of postoperative pain relief unless a continuous infusion or repeat injections are made. Part, if not all, of their action is related to their plasma concentration, which can lead to early and acute respiratory depression (sudden apnea). This condition is very different from the progressive and delayed respiratory depression, preceded by generalized pruritus, sedation, and progressive bradypnea, which is observed following excessive doses of epidural morphine.

Another limitation must be considered before prescribing opiates in pediatrics. Experimental administration of a few doses—even a single dose—of opiates can induce tolerance in newly born animals (45,46,47); this tolerance similarly develops in human neonates and occurs more rapidly in preterm than in term neonates (48,49). Gender plays a consistent role, both in animals (50) and in human (51), with males developing a more intense tolerance than females. The underlying mechanisms of this early tolerance are essentially unknown but should encourage the practitioner to limit exposure of neonates to opiates as often as possible, using alternate medications (continuous local anesthetic infusion and/or other additives).

If based on medical indications alone, regional procedures would be extensively used for surgery on the extremities and lower part of the abdomen in children, as it is in adults. This is common practice in a few countries, especially in Northern Europe, where blocks are culturally well established and accepted both by adults and children. However, in the vast majority of countries, children usually fear needle sticks, and regional block procedures have to be performed in anesthetized children, a practice well-established and accepted in this age group (68,69).

Analgesia is currently the main indication for regional anesthesia in children, especially in elective surgery involving the extremities, abdomen and, to a lesser extent, the thorax, neck, and head. Intra- and postoperative pain is easily and efficiently prevented by single-injection or continuous techniques depending on the expected duration and intensity of postoperative pain (70,71,72). All pediatric patients can benefit from these techniques including neonates, preterm infants, and critically ill children (73,74,75). In some institutions, regional techniques are used in children undergoing cardiac surgery (76,77,78), but this application remains controversial. Outpatient surgery currently represents a major and cost-effective indication of regional blocks, mainly peripheral nerve blocks, with a very low morbidity provided that appropriate precautions are taken.

In recent years, catheter techniques allowing intermittent, on-demand, or continuous infusion of local anesthetics have become increasingly popular. Originally restricted to neuraxial techniques (28,79,80), catheter placement along a peripheral nerve path or a fascial plane is becoming routine practice in many institutions to provide long-lasting pain relief both in hospitalized and nonhospitalized patients (81,82,83). Regional blockade is commonly performed to provide analgesia for cancer pain and chronic pain relief. These techniques are detailed in Chapter 47 and are not considered in this chapter.

Occasionally, a regional block procedure may be performed in children for whom analgesia is not the first objective. Examples of these mainly nonanalgesic indications include vasodilation in patients presenting with ischemic disorders (84,85,86) and maintenance of a motor block throughout the recovery period in the patient with an unstable fracture, fragile nerve suture, or tendon repair.

Absolute contraindications to neuraxial blockade include (87,88): infection at the puncture site, septicemia and meningitis; bleeding disorders, either constitutional or secondary to anticoagulant therapy; true allergy to local anesthetic, a very rare event (even with amino esters); uncorrected hypovolemia; and major vertebral anomalies (not spina bifida occulta—for many but not all authors, these include tethered cord syndromes and ventriculoperitoneal shunts).

Patients with mild seizure disorders should not be denied a regional anesthesia if they do not suffer any additional disorder prone to increase the morbidity of the procedure. Preexisting central nervous disorders and degenerative axonal diseases have long been considered as contraindications, at least relatively, even though few data support the hypothesis that a regional block worsens their course. A recent pediatric study involving 139 patients showed that preexisting neurologic disorders were not associated with bad neurologic outcome (the reverse was observed) after central block procedures (89).

Peripheral nerve blocks have few contraindications. Although infection close to the insertion site and allergy to local anesthetics are absolute contraindications, bleeding disorders may not contraindicate the performance of most peripheral blocks. Only procedures at risk of damaging large arteries (such as a supraclavicular block) or requiring deep needle insertion in closed muscular compartments (such as psoas compartment blocks or sciatic blocks at the thigh)
should be considered with caution.

Compartment syndromes are (very) infrequent but can result in devastating complications. Certain trauma lesions and surgeries are more often concerned, such as displaced humeral supracondylar fractures or tibial shaft and forearm fractures treated by intramedullary nail fixation.

The extremities are divided into distinct muscular compartments by inextensible fascias, within which run nerves and vessels. The intracompartmental pressure must remain low to allow effective venous return. Any expanding lesion or disorder including muscle necrosis, edema, hematoma, injection of a large volume of fluid, or external compression by a cast (90) or a wound dressing inevitably increases the intracompartmental pressure, which may interrupt blood flow. Depending on the local lesions and state of consciousness, this interruption produces “tourniquet” pain, hypoesthesia, or a combination of both, typically summarized by the “5 P’s” eponym: pain, pallor, paresthesia, paralysis, and pulselessness. This severe condition must be recognized early, in order to perform extended fasciotomies before irreversible lesions have occurred.

Since pain is one of the cardinal symptoms, any technique aimed at attenuating pain, especially regional blockade, is often claimed to be contraindicated because of the possibility of hiding this presenting symptom and thus delaying the rescuing surgery. However, avoidance of analgesic strategies is not acceptable medically or ethically (91). Intense pain is often associated with bone fractures. Although fractures are very frequent in the pediatric population, compartment syndromes are very rare. Therefore, children should not be deliberately allowed to experience pain because of fear of compartment syndrome, which could be easily and precociously diagnosed using appropriate and easily available monitoring.

Adequate pain management, including continuous epidural analgesia (92), does not preclude early diagnosis but, on the contrary, favors it. In effect, the most sensitive clinical indicator of compartment syndrome is the increase in pain medication requirements, both in dosage and frequency, and this indicator precedes other clinical symptoms of compartment syndrome by an average of 7.3 hours (93). Therefore, adequate pain management, either by the IV route or by using regional procedures with low concentrations of local anesthetics (0.1%–0.15% bupivacaine, levobupivacaine, or ropivacaine) and avoiding motor blockade may considerably facilitate early diagnosis when significant changes happen: (a) because pain is no longer relieved and the child asks for additional analgesia; (b) less often, because the child becomes oversedated with the usual “normal” dose of parenteral narcotics; or (c) because the intensity and distribution of anesthesia is changing (progressive loss of any sensory perception and motor function).

The development of excruciating pain in a child deprived of analgesic medications should no longer be considered the “presenting” symptom of a compartment syndrome. Rather, patients at risk should not have a closed plaster cast and should be monitored by serial clinical evaluation of the distal perfusion and tissue oxygenation of the limb and by noninvasive monitoring of the intracompartmental pressure using, for example, near infrared spectroscopy (94), noninvasive “hardness” measuring devices (95), or ultrasound technique (96), even though these noninvasive techniques are not fully reliable. If the risk is considered high (e.g., displaced humeral fractures, intramedullary nail fixation of tibia or radius, obtunded patients), the intracompartmental pressure close to the fracture site should be invasively monitored because the technique is easy and inexpensive, requiring only a venous cannula, an IV line, and a pressure gauge (as for central venous pressure measurement) (93,97,98,99).

Among the pediatric population, regional anesthesia is mostly used for analgesia, rather than anesthesia (as in adults). Therefore, it must offer a risk–benefit ratio superior to alternative analgesic approaches. In making a selection, the following points must be considered:

It is recommended that an alternative strategy be selected and explained to patients and family, in case of failure of the primary technique.

As most block procedures are performed on anesthetized children, a reliable technique of nerve localization independent of the patient’s cooperation is necessary (nerve stimulation, ultrasound guidance, or both); the overall morbidity of the two combined techniques, regional and general, should be lower than that of any other technique of anesthesia care, meaning usually that general anesthesia should be light (no muscle relaxants and no/minimal narcotics).

Basically, these techniques are the same as in adults; namely, loss-of-resistance technique for epidural anesthesia and compartment blocks, CSF aspiration for spinal anesthesia, and nerve stimulation for peripheral nerves and plexuses.

Ultrasound guidance is opening a new era, one that is particularly exciting in pediatric patients, especially in infants, whose nerves are very superficial and easily identified. However, the size of the probe can be difficult to handle and its application on the skin can change the anatomic landmarks and displace the block needle. The sterility of the procedure can occasionally be difficult to maintain throughout the procedure. Currently, block needles can only be reliably seen and their tip clearly identified when the major axis of the probe is in line with the shaft of the needle. If the probe is perpendicular to the shaft, only a cross-section of the needle can be seen (not always) and no information is obtained regarding the position of the needle tip. Two “tricks” can be helpful (personal data): (a) inject a small amount of 5% dextrose while the ultrasound machine is in Doppler mode—the stream produced at the tip of the needle will give a conical colored signal (dextrose will not compromise
further neurostimulation); (b) inject the local anesthetic after dispersing microbubbles in the solution.

Selection of the most appropriate local anesthetic depends on the surgical indication, the expected duration of intense postoperative pain, and the postoperative location of the patient. Commonly administered local anesthetics and their usual doses are displayed in Table 27-5. Short-acting local anesthetics (lidocaine and mepivacaine) are preferred for outpatient surgery. Longer-lasting agents do not preclude hospital discharge, provided motor functions have been restored at least in part and on the condition that the familial environment is competent (i.e., caring and able to make sound judgments). Levobupivacaine and ropivacaine are preferred for continuous infusion or repeat/on-demand injections after a catheter has been inserted along a peripheral nerve or neuraxially to provide long-lasting pain relief. Additives may consistently improve the quality and duration of blockade. Nonopioid additives (clonidine, ketamine) are particularly useful because they have few (and minor) adverse effects, thus not precluding early hospital discharge, and they prolong postoperative pain relief without requiring catheter placement.

For decades, long-lasting postoperative pain relief depended exclusively on continuous epidurals and patient hospitalization. Recently, the development of peripheral catheter techniques has simplified the management of postoperative pain, offering a better risk–benefit ratio than continuous neuraxial blockade, and even allowing hospital discharge and management at home in selected patients. For these techniques, continuous infusion (2–5 mL/h) or on-demand injections (2–5 mL) of low-concentration levobupivacaine or ropivacaine (0.1%–0.2%) represent the best and safest choice.

The required equipment to perform regional block in children is basically the same as in adults (see Chapter 8), and this holds particularly true for peripheral nerve blockade (single-injection as well as catheter techniques).

For neuraxial blocks, some notable differences must be considered. For obvious reasons, shorter and thinner needles must be used in infants. For epidural anesthesia, the Tuohy bevel is virtually the only bevel used in children. The recommended medium for the loss-of-resistance technique has elicited many controversies (100). Basically, in children over 6 to 7 years of age, use of normal saline offers the same advantages as in adults. Conversely, in infants and, particularly in neonates, air is the most dependable technique of identification of the epidural space. For spinal blocks, the bevel/tip of the needle does not have the same importance as in adults; the incidence of postdural puncture headache remains (very) low and is not influenced by the design of the needle tip (101,102). Most critical is the distance from the tip of the needle to the distal orifice of the lumen; this distance must be as short as possible to avoid leakage outside the dura mater if the needle is not advanced enough past the dura mater, and to avoid potential damage to the cauda equina if the needle is introduced too deeply. Consequently, pencil-point needles do not improve the overall quality of the procedure and are even suspected of decreasing the success rate by favoring subdural spread of the local anesthetic. The most commonly used devices for regional block procedures in the pediatric patient are listed in Table 27-6.

The technique of injection is a critical factor of safety. Meticulous technique must be systematically followed:

Regional anesthesia is an anesthetic procedure. As such, it must be performed by anesthesiologists in an operating room environment with the usual monitoring recommended for any general anesthesia—monitoring of electrocardiogram (ECG), blood pressure, temperature, respiratory rate, oxygen saturation by pulse oximetry and, in patients (especially infants) under assisted ventilation, evaluation of tidal volume and end-tidal CO₂. An IV line must be established prior to any injection of local anesthetic (110), and vitals parameters are to be reported on the anesthesia chart (see Chapter 8).

The quality and distribution of blockade must be evaluated prior to surgery, whatever the patient’s age. This evaluation is not easy even in conscious children in the absence of a motor block. It requires that the anesthesiologist has gained the confidence of the child, prevents the child from seeing what he is doing (especially handling of needles), and makes comparative tests in nonblocked areas. Gentle skin pinching is the most dependable technique of sensory testing, especially in lightly anesthetized patients. Electrical stimulation using a nerve stimulator proved to be suitable in healthy volunteers (111,112) but little data are available in children.