Epidural and subarachnoid blockade in children

Regional anaesthetic techniques are now widely used in children undergoing surgery. Despite a published history of nearly a century these techniques were restricted to small numbers of practitioners in few centres until relatively recently. The last 10–15 years have seen these techniques become commonplace in all types of hospital where children undergo surgery. This is particularly true of subarachnoid and epidural blocks. Advantages of these techniques include dense intra-operative analgesia, reduced requirements for volatile agents and opioids and analgesia continuing into the postoperative period. Central neuraxial analgesic techniques are not restricted to specialist anaesthetists in paediatric centres but also have a place in district general hospitals and specialist surgical centres where children undergo orthopaedic, plastic, urological and thoracic surgery. In the same period there has also been a large volume of material published on many aspects of central neuraxial blocks in children. This literature, when analysed critically, can be used to guide the use of these techniques in children in order to optimise the analgesia they provide, minimise the side-effects and improve safety. Caudal epidural blockade performed with a single injection will be discussed separately from other epidural techniques performed at other levels. The technique of epidural analgesia using a single injection of local anaesthetic to the epidural space via the caudal approach combines the advantages of a simple technique with a high success rate. With appropriate selection of the patient, the procedure, drug concentration (bupivacaine 0.125%-0.25%) and maximum dose (2.5 mg.kg−1 of bupivacaine), the procedure is extremely safe, does not cause troublesome urinary retention or leg weakness and may be used in outpatients. For these reasons, single-shot caudal epidural blockade is one of the commonest local anaesthetic techniques used in paediatric anaesthesia. The sacral approach to epidural anaesthesia was first described, independently, by Sicard and Cathelin in 1901 and the first published report of the use of caudal anaesthesia in children was in 1933. The technique has a wide range of indications and several large series describe its high success rate and low incidence of complications [1–3]. Caudal epidural blockade can be used effectively in selected children to provide both intra-operative anaesthesia and postoperative analgesia for many operations below the umbilicus. These include orchidopexy, circumcision and inguinal herniotomy, lower limb and pelvic orthopaedic surgery and lower abdominal surgery in neonates and infants where low thoracic dermatomes may be blocked. It has been used as the sole technique for inguinal hernia repair in ex-premature infants at risk of postoperative apnoea. In most cases, the caudal block is combined with light general anaesthesia or sedation. Caudal blocks are popular for day case surgery, combining prolonged analgesia with few side-effects. They are also an effective alternative to spinal anaesthesia for operations below the umbilicus in awake, high-risk children such as ex-premature infants, where they produce a longer duration of action than spinal anaesthesia as well as postoperative analgesia [4]. However, the hazards associated with caudals may be more common in this group than in older children. It is usually possible to block sacral and low lumbar dermatomes consistently in any child. In children weighing up to 20 kg, inguinal dermatomes may also be blocked consistently, but in bigger children there is a significant failure rate when caudal epidural blockade is used for this purpose. Low thoracic dermatomes can only be blocked consistently by the caudal epidural route in infants. Contraindications to the technique include anatomical anomalies of the sacrum or the overlying skin, local or systemic infection or a bleeding diathesis. The neonatal sacrum is composed of five distinct semicartilaginous sacral vertebrae which gradually ossify and fuse to form the adult sacrum in the third decade of life [1]. The sacral hiatus results from the failure of fusion of the posterior arches of the fifth, sometimes the fourth and occasionally the third, sacral vertebrae. This deficiency in the neural arch of the fifth sacral vertebra is covered by a ligamentous membrane known as the sacrococcygeal membrane. The landmarks for this hiatus are the sacral cornua superiorly and the coccyx inferiorly. Posterior fusion of the sacrum becomes more complete with age resulting in reduction in the size of the hiatus which is closed by the sacrococcygeal membrane. The dimensions of the caudal space vary markedly with age and between individuals of the same age. Early work was based on the study of cadaveric specimens and skeletons [5, 6] and more recently magnetic resonance imaging has been used to study caudal anatomy in adults [7]. Of importance is the distance from the sacral hiatus to the lower extremity of the dural sac. In an infant of 2 months, the lower extremity of the dural sac may be as low as the S4 vertebral body, less than 10 mm from the sacral hiatus. Sacral anomalies are found in at least 5% of individuals [8] and can result in a reduced distance from the sacral hiatus to the dural sac. Caudal block tends to be easier to perform in children than adults because they lack the bony irregularities, asymmetry of the sacral cornua and overlying pad of fat which develop with age and may make identification of the sacral hiatus more difficult. The sacral hiatus appears to be more cephalad than in adults and this is attributed to the less well developed gluteal muscles of children which results in the natal cleft of infants not extending as high in relation to the sacrum [9]. Under aseptic conditions a needle is advanced through the sacrococcygeal membrane until a pop or loss of resistance is felt. It is not usually necessary to perform a formal loss of resistance technique to saline or air as is necessary with lumbar and thoracic epidural techniques. The needle should be left open for 10–20 s to detect venous blood or cerebrospinal fluid in the event of a venous or dural tap. Aspiration of fine epidural veins will often be negative [1, 10] and a period of observation with the needle open to the atmosphere is essential to detect this complication. The needle should not be advanced more than 2–3 mm into the epidural space since this increases the risk of venous puncture and dural tap. Venous puncture during caudal epidural injection was seen in 6.9% of patients in a series of over 1100 cases and the incidence fell with increasing experience of the anaesthetist [10]. After a negative aspiration test the selected dose of local anaesthetic solution is injected slowly. The consequences of inadvertent intravascular injection depend not only on the plasma level of local anaesthetic reached but also on the rate of rise of this [11] and slow intravenous injection is safer than rapid injection. Intravenous injection of local anaesthetic solution may still occur after a negative aspiration test and this may be more likely in infants than in older children [10]. The dura normally ends at the level of the second sacral vertebra but variations in anatomy may cause it to be lower and in neonates the dura often extends to the third or fourth sacral vertebrae. Concerns about infection and the risk of a dural tap mean that it is essential for this block to be performed under aseptic conditions. There are a variety of needles which can be used for caudal blockade. Suggested needle gauge varies from 25G [9] in neonates to 20G [12] in older children. Finer needles cause less trauma but have been implicated in accidental intravenous or subarachnoid injection of local anaesthetic following false negative aspiration tests [13]. Specialised block needles with short 'blunt' bevels which enhance awareness of the needle penetrating the sacrococcygeal membrane have been shown to reduce the incidence of vascular penetration compared to standard long bevelled hypodermic needles [1]. Irrespective of the type of needle used is the importance of not advancing the needle into the epidural space once the sacrococcygeal membrane has been penetrated in order to reduce the likelihood of a dural puncture. As with other regional techniques practice varies concerning the administration of a test dose. Intravenous injection of an adrenaline-containing test dose may not produce the expected cardiovascular changes in children. Prior administration of atropine intravenously has been shown to improve the sensitivity of this test during halothane anaesthesia [14]. Neither 0.5 or 0.75 μg.kg−1 of adrenaline during isoflurane anaesthesia were reliable indicators of intravenous injection [15]. Isoprenaline 0.075 μg.kg−1 may be a more sensitive intravascular marker than adrenaline [16]. Given the limitations of a test dose of local anaesthetic at detecting intravascular or subarachnoid injection, the local anaesthetic solution should always be injected slowly. Failure of the technique is usually due to inability to find the hiatus and deposit the local anaesthetic solution in the correct space. The main determinant of this is the experience of the anaesthetist. Although the anatomy of the sacrum is variable it is unusual that caudal injection cannot be performed by an experienced anaesthetist. Subcutaneous placement of the caudal needle is common and occurs initially in up to 19% of cases [1]. Problems appear to be more common in children weighing less than 10 kg where difficulty in identifying the caudal hiatus has been reported in 11% of subjects [10] and children aged over 7 years. Dalens reported a 3.5% incidence of failure to identify the caudal space after up to four attempts and the majority of these cases were over 7 years of age [1]. Failure to produce an adequate block following apparently successful caudal injection has been reported in 11% [17], 2.8% [9] and 7.7% [18] of cases. The 'failed caudal' may also be due to an inappropriate indication with the use of an inadequate volume to block the required nerve roots. Caudal epidural blockade without general anaesthesia and tracheal intubation for upper abdominal surgery has been used in one series but in half of 86 cases in this series, conditions were unsatisfactory and the block required supplementation or general anaesthesia compared to 8% of the 411 patients given caudals for lower abdominal surgery. The main complication was regurgitation and aspiration of gastric contents which resulted in two cardiac arrests and one death. Unsupplemented caudal epidural blockade is an inappropriate anaesthetic technique for upper abdominal surgery [9]. This series also reported unsatisfactory results with the use of caudal anaesthesia for meningomyelocoele repair and laparoscopic procedures. In older children, there may be unexplained failures of the technique possibly due to the development of fibrous septa within the epidural space which are well described in adult patients and can result in unilateral or limited blocks despite appropriate volumes being injected [1]. Because of its long duration of action, bupivacaine is the local anaesthetic agent used most commonly for caudal epidural blockade. Several formulae have been described to determine dosages and volumes required to produce blockade to various levels. Multiple regression analysis on data from three studies demonstrated the existence of a linear relationship with a correlation coefficient of 0.88 between the dose requirement expressed as millilitres per dermatome to be blocked and age [12]. In most cases, however, the volume of local anaesthetic to be injected is calculated using body weight which is readily available for every child. In children up to 7 years of age the dose requirement per dermatome correlates with body weight (correlation coefficient 0.93) and a dose of 0.056 ml.kg−1 for each dermatome to be blocked has been recommended [19]. The extent of spread varies markedly between patients especially over 7 years of age possibly as a result of the changing consistency of epidural fat with age. Armitage recommends 0.5 ml.kg−1 for a lumbosacral block as for circumcision, 1.0 ml.kg−1 for a thoracolumbar block as for inguinal hernia repair and 1.25 ml.kg−1 for a midthoracic block as for an orchidopexy [20]. Dalens reported excessive spread of local anaesthetic in 30% of patients receiving 1.25 ml.kg−1 and found 0.75–1.0 ml.kg−1 to be appropriate for a wide variety of operations [1]. The use of 0.75 ml.kg−1 for both hernia repair and orchidopexy resulted in 69% of children being pain free 4 h postoperatively [18]. The use of 0.5 ml.kg−1 or 0.75 ml.kg−1 of bupivacaine 0.5% for circumcision resulted in successful blocks in 96% and 100% of patients, respectively [21, 22]. This area is plagued by different criteria for the performance of a successful block with few studies involving rigorous objective postoperative assessments with standard stimuli. A simple working rule is for children less than 6 months of age 1 ml.kg−1 of bupivacaine 0.125% will block low thoracic dermatomes. In children above this age 1 ml.kg−1 of bupivacaine 0.25% will consistently block inguinal dermatomes in children less than 20 kg while above this weight the technique becomes inconsistent at blocking inguinal dermatomes. At all ages, sacral dermatomes may be blocked reliably by 0.3–0.5 ml.kg−1 of solution. With these smaller volumes a concentration of 0.5% bupivacaine will provide a longer duration of block than 0.25%. Pharmacokinetic data from several studies show that single epidural doses of 2–2.5 mg.kg−1 of bupivacaine are associated with low plasma levels of bupivacaine (Table 1). The incidence of motor block as shown by urinary retention and leg weakness is very variable after caudal epidural blockade. There is an impression that urinary retention is less common than in the past partly because of more liberal pre-operative fasting regimens which allow clear fluid intake until 2 h pre-operatively. Varying criteria for the definition of this problem make comparisons difficult. If this potential problem is explained to the parents and child (if appropriate) as a likely consequence of the technique, its occurrence will not cause anxiety and it need not be a reason to delay discharge from hospital. Walking independently is a common discharge criterion for suitable children in day surgery units. Moderate motor block causing a degree of weakness of the legs is common and may delay discharge. Extensive motor block in the legs may also frighten and distress some older children and is a relative contraindication to the use of large volumes of local anaesthetic solution in children aged over about 5 years [10]. This is seen in 30% of subjects after bupivacaine 0.5% given to block lumbar dermatomes [28] but the incidence is lower when smaller volumes of bupivacaine 0.5% are used to block sacral dermatomes only. The incidence is also much lower when more dilute solutions of bupivacaine are used. There was a 4% incidence of motor block lasting for 4 h after 1 ml.kg−1 of bupivacaine 0.2% [17], a 4.6% incidence of inability to walk at 4 h postoperatively following caudal block with 1 ml.kg−1 of bupivacaine 0.25% [29] and an 8% incidence of being unable to stand unaided at 4 h following caudal block with 0.5 ml.kg−1 of bupivacaine 0.25% [21]. More importantly, perhaps, than motor weakness is the occurrence of proprioceptive blockade in a child who is unaware of any weakness and who appears to have normal power when tested in bed but who may be unable to walk unaided. Because of this, all children should be closely supervised when first walking after caudal epidural blockade. Leg weakness is much more of a real problem after caudal epidural blockade than urinary retention and may prevent the discharge of an older child who is too big to carry. The optimal dose of bupivacaine for caudal epidural blockade in children may be 0.125% which has been shown to produce a block with an equal duration of analgesic effect and the same requirements for supplementary analgesia as 0.25% but with a much lower incidence of leg weakness at 1 h postoperatively [30]. The incidence of vomiting associated with caudal epidural blockade ranges from 0% [22], 4.3% [31], 10% [32] to 30% [21, 28, 33]. Whether this is a consequence of caudal blockade itself or is a background incidence associated with general anaesthesia in children is difficult to say. Dalens noted that postoperative restlessness is a problem in 3% of patients [1]. There was a strong association with residual motor block. Venous puncture has an incidence of 1.6–10.6% [1, 10] and is reduced if short bevelled rather than hypodermic needles are used. If this occurs, the procedure should be restarted from fresh with an unstained needle to prevent the occurrence of uncertainty about a subsequent venous tap if traces of blood are seen in the needle hub. There are no reports of extradural haematoma after caudal epidural blockade in children. It is possible to place the tip of the needle subperiostally in the marrow cavity of the sacrum which has been estimated to have an incidence of 0.2% [9]. Injection of local anaesthetic under these circumstances is equivalent to an intravenous injection but this should be detected by the marked resistance to injection. The incidence of dural tap has been described as 1/500 [9], 1/750 [1], 1/1100 [10] and 1/1154 [2] in various series. Negative aspiration for cerebrospinal fluid is possible despite a dural puncture and is another reason why the needle should be left open to atmosphere after entering the space before injection of local anaesthetic when the flow of cerebrospinal fluid is usually vigorous and unmistakable. The caudal block is usually abandoned if this occurs. A total spinal block is a rare but serious complication of caudal epidural blockade caused by inadvertent subarachnoid injection of a significant volume of local anaesthetic. The risk is increased due to the variable distance between the sacrococcygeal membrane and the lower extremity of the dural sac. Desparmet described a case in a 3.5-kg ex-premature infant who developed bradycardia, apnoea, severe hypotension and fixed dilated pupils following caudal injection of 0.7 ml.kg−1 of local anaesthetic [13]. Afsan reported a case in an 18-month-old child who, following 1 ml.kg−1 of local anaesthetic, exhibited similar signs of a total spinal [34]. Both cases followed negative aspiration tests. The overall incidence appears to be ≈1 in 80 000 [3]. Rectal penetration with a caudal needle has been described and is estimated to have an incidence of 1 in 80 000 cases [3]. When performed safely by experienced anaesthetists, caudal epidural blockade in children can provide excellent intra-operative and postoperative analgesia with a very low incidence of serious complications. A large North American survey encompassing an estimated 158 229 caudal epidural blocks calculated an estimated incidence of 1 in 10 000 for major complications and 1 in 40 000 for catastrophic complications [3]. Various additives to local anaesthetic solutions have been used in attempts to prolong the duration of caudal analgesia provided by a single injection of local anaesthetic. The duration of caudal blockade is usually defined as the time to the first requirement for supplementary analgesia. When plain bupivacaine 0.25% is used the median duration of block ranges from 4 to 8 h [22, 31, 35, 36]. When plain bupivacaine 0.5% is used the mean duration of analgesia is about 10 h [33]. Vasoconstrictors have long been used to prolong the duration of caudal epidural blockade. They decrease the rate of vascular absorption of local anaesthetic and allow a greater mass of local anaesthetic molecules to reach the nerve membranes and increase the density and duration of the block produced. The commonest is adrenaline in a concentration of 5 μg.ml−1 (1/200 000). The effects of adrenaline depend on the site of injection and on the local anaesthetic used. Although it reliably prolongs the duration of blockade when used for infiltration anaesthesia and peripheral nerve blocks with all local anaesthetic agents, when used for epidural injection the effects of adrenaline tend to be less with local such as bupivacaine than with more such as The high of bupivacaine it to be in epidural fat and and the relatively of adrenaline tend to have effect on its duration of The effect of adrenaline on the duration of action of epidural bupivacaine also on the concentration of bupivacaine and effect is seen when 0.5% or are used compared with 0.125% or 0.25%. The of 000 adrenaline to bupivacaine 0.25% for caudal epidural blockade has been found both to prolong the duration of the block and to have no effect [18]. any effects of adrenaline on the duration of single-shot caudal epidural blockade in children are relatively Several studies have provided of the analgesia which is in children after the administration of caudal epidural The median duration of analgesia after mg.kg−1 of for lower body surgery was h compared with 5 h for 1 ml.kg−1 of bupivacaine 0.25% 000 adrenaline and for mg.kg−1 of intravenous After surgery, caudal mg.kg−1 a mean duration of analgesia of 20 h compared with 6 h for 0.5 ml.kg−1 of bupivacaine 0.25% After surgery, mg.kg−1 combined with 0.75 ml.kg−1 of bupivacaine 0.25% the need for postoperative analgesia while over of receiving bupivacaine required analgesia postoperatively of blood for plasma after the administration of caudal epidural show levels much less than required for analgesia after systemic administration and that the effect of epidural opioids on analgesia is due to a local action at spinal level as to an effect after systemic absorption Although has been used most commonly for caudal epidural use in children, the effects with local are also seen when other opioids are used include urinary and vomiting as well as The most of these is This is a risk when epidural opioids are used in adults and is particularly because its may be for several after particularly if is the used. The reported series of children given caudal epidural reported an incidence of cases of in cases Of these cases, were aged less than months of age and opioids in to caudal epidural The sensitivity of children up to the age of or 4 months to opioids is well known and these results are not It is likely that a group of children aged less than months given significant doses of opioids by any route will have a significant incidence of is an This of are widely used in and anaesthesia as and The analgesic action of when is due to of which the of in the of the spinal A site of action may also be Several studies in children have and the effect of when used to caudal epidural blockade with a local anaesthetic solution. of patients undergoing general surgical and urological under general anaesthesia with caudal epidural blockade of bupivacaine 0.25%. to this 1 μg.kg−1 or adrenaline 5 The and duration of postoperative analgesia as the time to first analgesic requirement using an objective pain was longer with than with plain bupivacaine or bupivacaine adrenaline The of children no analgesia was with than with plain bupivacaine or bupivacaine adrenaline In a of bupivacaine 0.25% 1 ml.kg−1 with or without 2 μg.kg−1 for lower limb orthopaedic surgery in children, there was a increased duration of postoperative analgesia in the group given compared with receiving plain bupivacaine and requirements for postoperative were reduced at and h in the group The of 2 μg.kg−1 to adrenaline 5 μg.ml−1 to bupivacaine 0.25% 1 ml.kg−1 for caudal use was also seen in a study of undergoing orchidopexy The mean duration of analgesia was h in the group receiving compared with h in the group receiving This effect is less than that seen in other studies using of bupivacaine and in children for caudal epidural use and may be explained by the that patients were not in the studies and analgesic was a effect with epidural The of 2 μg.kg−1 to 7 mg.kg−1 for caudal epidural use in children aged years undergoing surgery increased the duration of effective analgesia by an objective pain of from to studies have found no between receiving caudal and in the occurrence of significant or changes postoperative longer following caudal bupivacaine 1 ml.kg−1 with 2 μg.kg−1 than plain bupivacaine These closely to duration of analgesia h and and this effect was attributed by the to analgesia. The of adrenaline 5 μg.ml−1 or 2 μg.kg−1 to caudal bupivacaine in no between the in the incidence of motor urinary retention or postoperative its anaesthetic and analgesic effects by to a of by the where it as an These are found the central the lumbar spinal in the of pain has shown that as well as analgesia after systemic analgesic at a spinal level in This with the effects of has in its epidural and subarachnoid administration in patients to provide postoperative analgesia. of caudal epidural in children have demonstrated its In a study of caudal analgesia with 1 ml.kg−1 of bupivacaine 0.5 mg.kg−1 or both given after of general anaesthesia for inguinal herniotomy, postoperative analgesia with caudal was similar in to 1 ml.kg−1 of bupivacaine 0.25% or a of the The duration of action of the of and bupivacaine was particularly with only of subjects any analgesia in the first h postoperatively This compared with and in the and bupivacaine These results have been by a study which that 0.5 mg.kg−1 provided a longer duration of postoperative analgesia after orchidopexy duration than 2 μg.kg−1 or adrenaline 5 μg.ml−1 when to 1 ml.kg−1 of bupivacaine 0.25% for In this study of there were no between in the incidence of urinary motor block or postoperative sedation. major have been reported after the use of epidural studies have demonstrated the of without The use of epidural analgesia in children was described in as an alternative to general anaesthesia in high-risk neonates and children The technique not become widely used in children because of in it and about potential complications when it is performed in It has become more popular in years because of increased to the requirements for effective analgesia in children, that the postoperative may be by dense block and the of of appropriate The of small gauge epidural needles and and fine epidural and which through have epidural analgesia in the of children