Continuous spinal anaesthesia - BJA

British Journal of Anaesthesia 1998; 81: 590–597

REVIEW ARTICLE

Continuous spinal anaesthesia N. M. DENNY AND D. E. SELANDER Continuous spinal anaesthesia (CSA) is the technique of producing and maintaining spinal anaesthesia with small doses of local anaesthetic which are injected intermittently into the subarachnoid space via an indwelling catheter. The concept was described first in 1907 by Dean,11 a British surgeon, who wrote of placing a needle in the subarachnoid space and leaving it in situ so that repeated doses of local anaesthetic could be injected. In 1939, Lemmon described a malleable needle which could be left in the subarachnoid space allowing intermittent injection of local anaesthetic via a rubber tube.37 The catheter technique was described first by Edward Tuohy in 1944.71 He described CSA as a safe and versatile technique without significant problems of post-dural puncture headache (PDPH). In the early 1950s, Dripps reported a high incidence of paraesthesia and low success rates with CSA, and this led to a decline in its use.18 Fears that CSA would result in a higher incidence of PDPH and neurological complications, together with the development of the continuous epidural anaesthetic technique,10 further discouraged the use of CSA. When CSA was reintroduced in the mid 1980s, the advantages described were: excellent control of segmental spread and duration; effectiveness of small doses of local anaesthetic; and decreased risk of cardiovascular side effects.15 The incidence of PDPH also seemed low, especially in elderly patients.15 40 During the late 1980s, microcatheters were developed to make the CSA technique suitable for use in young patients without incurring an unacceptable risk of PDPH.28 However, not only was it difficult to show a decreased frequency of PDPH, but serious neurological complications were reported after the use of spinal microcatheters and high concentration, hyperbaric local anaesthetics.56 As a consequence, in 1992 the Food and Drug Administration (FDA) of the USA banned the use of spinal catheters thinner than 24-gauge,20 and the manufacturers of local anaesthetics declared that their drugs were not indicated for use with CSA.16 This reinforced the misconception that CSA was a dangerous technique. However, our experience is that CSA is a valuable and safe technique when performed correctly. The purpose of this article is to review the relevant aspects of CSA, its indications and side effects, and to comment on drugs, equipment and technique. (Br. J. Anaesth. 1998; 81: 590–597) Keywords: anaesthetic techniques, subarachnoid; anaesthetics local; equipment, catheters subarachnoid

History and experience with large-bore catheters As we know it today, CSA was first described by Edward Tuohy in 1944. He devised a technique of inserting a ureteral catheter 4–5 cm into the subarachnoid space via a 15-gauge Huber point needle and initiating spinal anaesthesia with incremental doses of local anaesthetic. He found that the dose of local anaesthetic used for single injection spinal anaesthesia could be reduced by 20–25% with the CSA technique.71 72 He recommended it for surgery below the umbilicus and reported no increase in the incidence of PDPH compared with single injection techniques and no neurological complications. His comments on the technique are as revealing and pertinent today as they were then: “A question has been raised concerning the direction the catheter will advance in the subarachnoid space after the tip of the catheter leaves the end of the guiding needle. The direction cannot be predicted positively, but if the tip of the catheter is bent slightly before introducing it into the needle, I have found that the catheter will advance cephalad (our italics) in most cases. This has been determined and verified by using leaded ureteral catheters which are opaque radiographically. If the catheter turns caudad (our italics) on entering the subarachnoid space, this fact can usually be surmised or detected shortly after the first dose of the spinal anesthetic agent has been introduced because of the segmental distribution of the resulting anaesthesia (our italics). Under these circumstances it is suggested that the remaining amount of the anesthetic solution in the syringe be diluted with an equal volume of saline to make a 1.5% solution of procaine hydrochloride. Twice as much solution (volume) should then be injected when additional doses are required”.71 In 1950, Dripps reviewed reports on single injection spinal anaesthesia (SSA) and CSA with a malleable needle and catheter techniques.18 He found an 8% (43 of 506) incidence of failed anaesthesia with CSA compared with 1.9% (37 of 1921) with SSA. Also, he found more technical difficulties with the catheter technique and a significantly higher incidence of transient paraesthesiae (33%) than with single injection techniques (13%).18 Over the next 25 yr, CSA was used little, as reflected by the paucity of references in the literature, and it is hard not to conclude that this was a direct result of Dripps’ article. NICHOLAS M. DENNY, FRCA, Department of Anaesthesia, Queen Elizabeth Hospital, King’s Lynn, Norfolk PE30 4ET. DAG E. SELANDER, MD, PHD, Sahlgren’s University Hospital, Gothenburg, Sweden and Clinical Research and Development, Astra Pain Control, S-151 85 Södertälje, Sweden.

Continuous spinal anaesthesia

591

In 1952, in a series of 600 poor-risk surgical patients (24% were younger than 60 yr of age) who had CSA using 16-gauge Huber point needles, Brown found a failure rate of only 2%, and a 14% incidence of transient paraesthesia on threading the catheter.2 The PDPH rate was 9.8%, and there were four cases of extraspinal neuropathies (three sixth nerve palsies and one peroneal) which recovered spontaneously. In 1972, Giuffrida and colleagues described the use of CSA for Caesarean section in 75 patients using 21-gauge Huber point needles and 24-gauge catheters.22 There were no failures, PDPH was reported in 12 patients (16%) and there were no neurological complications. In the same year, Kallos and Smith reported no failures, no complications and no PDPH in a series of 121 patients who had CSA for hip surgery (table 1).30 They used 20-gauge nylon epidural catheters which were inserted through 18-gauge Hustead or Tuohy needles. In 1981, Rao and El-Etr investigated the incidence of neurological complications arising from anticoagulant therapy after epidural and subarachnoid catheterization in 3164 and 847 patients, respectively (catheter size was not specified). Twenty patients experienced minor complications, five of which were neurological (four after epidural, one after subarachnoid blocks), and 15 low back pain (nine epidural, six subarachnoid) which were self-limiting and resolved with time.54 There was no incidence of peridural haematoma leading to spinal cord compression. There was no report on the incidence of PDPH. In 1987, Denny and colleagues, in the first prospective study of PDPH after CSA, assessed 117 patients with a mean age of 63 yr and found only one patient, a 29-yr-old man, who developed PDPH.15 They used 18-gauge Hustead Tuohy needles (which were inserted with the bevel parallel to the longitudinal fibres of the dura) with 20-gauge nylon epidural catheters inserted 2–4 cm into the subarachnoid space. Difficulty threading the catheter was experienced in five patients (4.3%) and there were six (5.1%) failures with the technique (which led to general anaesthesia), but no neurological sequelae. Denny and colleagues postulated, as an explanation for the unexpectedly low incidence of PDPH, that the spinal catheter, if left in situ long enough, might induce an inflammatory reaction around the puncture site, and “that when the catheter is removed, edema or fibrinous exudate resulting from the inflammatory reaction seals the hole in the dura, thus preventing leakage of cerebrospinal fluid (CSF)”. It was concluded that “with a PDPH incidence of less than 1% and the possibility of safely controlling the

level of anesthesia, continuous spinal anesthesia offers an excellent method for long operations, particularly in elderly or severely ill patients”.15 A similar low incidence of PDPH was reported in 1991 by Mahisekar and colleagues in a retrospective series of 226 patients after CSA with 18-gauge Tuohy needles and 20-gauge catheters, performed from 1981 to 1985. They reported no other complications.40 The hypothesis proposed by Denny and colleagues that the prolonged presence of the spinal catheter may prevent PDPH was supported by a retrospective study by Cohen and co-workers, which showed that prolonged placement of large-bore catheters in obstetric patients after accidental dural puncture appeared to reduce the incidence of PDPH.9 However, in a study of 87 patients, aged 40–90 yr, undergoing orthopaedic procedures with CSA, performed with 18-gauge Tuohy needles and 20-gauge nylon catheters, Liu and colleagues found that prolonged spinal catheterization did not reduce the incidence of PDPH (9.2%).38 In 1989, Sutter, Gamulin and Forster, in a retrospective series of 457 patients who had CSA using 20-gauge catheters, reported not only greater cardiovascular stability but a significantly lower failure rate (1.7%) than the 9% in 274 similar patients who received epidural anaesthesia. No infections or neurological complications were found, but the incidence of PDPH was not reported.69 In 1995, Van Gessel, Forster and Gamulin showed that it was feasible to use CSA in a teaching environment. They detected no incidence of PDPH after CSA with 20-gauge catheters in 100 patients aged more than 65 yr. They demonstrated that although the failure rate for their residents was 6%, the technique was 100% successful in experienced hands.75 In 1997, Horlocker and co-workers, in a retrospective series of 474 patients who had CSA with 20gauge catheters, reported a low PDPH incidence (3.4%) and a failure rate of 3.4%.27 However, one patient developed aseptic meningitis and another, a 76-yr-old, 109-kg female with a history of lumbar radiculopathy, developed a cauda equina-like syndrome (CES) after administration of 5% hyperbaric lidocaine (lignocaine) 120 mg. In summary, experience of CSA using 20-gauge catheters in elderly patients shows that the technique is safe and easy to use. The major complication was PDPH, which varied from 0 to 9.2% (tables 1, 2), but with the exception of one case of CES with tetracaine56 and one with 5% lidocaine,27 there were no major neurological complications. The overall reported failure rate was 2.5%, which is similar to single injection techniques15 27 40 46 69 70 75 (table 3).

Table 1 Incidence of post-dural puncture headache (PDPH) (retrospective studies). *All Caesarean section patients, **all obstetric patients

Microcatheters

Study

No. patients

Age (yr)

Catheter (gauge)

PDPH (%)

Kallos30 Giuffrida22 Peterson50 Mahisekar40 Horlocker27 Horlocker27

121 75* 52 226 474 127**

59pts:60 :40 70 62 63 29

20 24 20 20 20 28

0 16 0 0 3.4 33

After publication of the prospective study by Denny and colleagues in 1987,15 and the introduction of the microcatheter technique in 1990,28 there was a resurgence in popularity of CSA with both large-bore and microcatheters (28–32-gauge). The spinal microcatheter was described first by Hurley and Lambert.28 Their aim was to develop a sufficiently fine-bore catheter (32-gauge Rusch) which could be threaded through an appropriately fine spinal needle (26-gauge) into the CSF. Theoretically, com-

592

British Journal of Anaesthesia

Table 2 Incidence of post-dural puncture headache (PDPH) (prospective studies)

Study

No. patients

Age (yr, mean)

Catheter (gauge)

PDPH (%)

Denny15 Lui38 Mazze41 Van Gessel75 Hurley28 De Andres12 Standl68

117 87 100 100 58 65 100

63 70 64 965 50 33 62.5

20 20 20/24 20 32 32 28

:1 9.2 6 0 4 3 1

Table 3 Failure rate of continuous spinal anaesthesia (CSA). *one catheter broke

Table 4 Complications of continuous spinal anaesthesia (CSA) (32-gauge) catheters

Study

No. patients

Cannot thread (%)

Failure (%)

Catheter Break (%)

Hurley28 Silvanto62 Pitkanen52 Klimscha32 De Andres12

58 20 21 26 65

10 25 30 — 3

15 25 35 1 8

3.4 0 13 0 0

Table 5 Cauda equina cases. 5LH:5% lidocaine hyperbaric, 1 Tetra:1% Tetracaine hyperbaric

Study

No. patients

Catheter (gauge)

Failure (%)

Pro/ retrospective

Age (yr)

Ref. No.

Denny15 Van Gessel75 Sutter69 Mahisekar40 Horlocker27 Petros51 Standl68 Chan7 Horlocker27

117 100 457 226 474 90 100 55 127

20 20 20 20 20 28* 28 27 28

5.1 0 1.7 4 3.2 4.4 0 16 3.9

Prospective Prospective Retrospective Retrospective Retrospective Prospective Prospective Prospective Retrospective

67 60 68 45 56 67 76

58 58 56 56 56 56 27

pared with large-bore catheters, this would enable CSA to be performed in younger patients with a reduced risk of PDPH. Their initial study with the 32gauge microcatheter in 58 patients in 1990 showed a 20% incidence of technical complications, which included failure to thread the catheter, inability to inject the local anaesthetic and inadequate anaesthesia, resulting in an incidence of failed spinal block of 15% and a 3.4% incidence of broken catheters (table 4).28 Similar experiences were also reported by others (table 4).12 52 62 The 32-gauge spinal microcatheter was difficult to handle, CSF could not be aspirated and it had a very high internal resistance, making injection of local anaesthetic very slow. A 28-gauge catheter which could be passed through a 22-gauge spinal needle was then developed (Kendall). This catheter proved easier to use and did not have as many technical complications. Recent studies with 28-gauge catheters have shown a technical complication rate similar to that of large-bore catheters, including the incidence of PDPH.27 51 68 The authors’ own experience with this catheter revealed a higher incidence of slow onset and inadequate blocks compared with large-bore catheters. However, it was not long before cases of neurological complications, in the form of cauda equina syndrome, were described after CSA with microcatheters.56 Further reports of problems with the microcatheter technique58 led the FDA in 1992 to ban the use of spinal catheters smaller than 24-gauge in the USA.20 In all, approximately 12 cases of cauda equina syndrome after CSA with microcatheters have been reported.

Sex

Surg. position

M M M F M M F

Lithotomy ? Lithotomy Supine Supine Supine Supine

Drug

Initial dose (mg)

Total dose (mg)

Cathete (gauge)

5LH 5LH 5LH 5LH 5LH 1 Tetra 5LH

150 150 150 100 110 20 30

290 220 175 300 190 37 120

28 28 28 28 28 20 20

sensory disturbance in the saddle area, weakness of the legs, bowel and/or bladder dysfunction.29 In the first six cases of cauda equina syndrome reported, 28-gauge catheters and 5% hyperbaric lidocaine were used, except in one patient where a 20-gauge catheter and 1% hyperbaric tetracaine had been used.56 58 All patients showed insufficient spread of anaesthesia which became adequate only after excessive doses of local anaesthetic had been injected. After operation it became evident that these patients had developed a serious neurological complication which was diagnosed as cauda equina syndrome. As an explanation, mechanical trauma to nerve roots was considered unlikely. Instead, it was postulated that this was a neurotoxic injury caused by pooling of highly concentrated local anaesthetic around the cauda equina.56 The six first cases are summarized in table 5. STUDIES WITH SPINAL MODELS

Studies using spinal column models have shown that insufficient spread of a hyperbaric local anaesthetic can occur if the catheter is positioned caudally and especially when the local anaesthetic is injected slowly.17 35 55 57 Under such circumstances, the hyperbaric local anaesthetic could accumulate in the caudal end of the dural sac, without or with only little mixing with CSF, thus allowing exposure of the thinly protected nerve roots to an abnormally high concentration of the local anaesthetic. It was postulated that if this exposure time was long enough, the nerve roots could suffer from a toxic injury. CLINICAL EVIDENCE

Cauda equina syndrome and local anaesthetic toxicity Cauda equina syndrome results from toxic damage to the nerve roots of the cauda equina, and is characterized by the following symptoms: low back pain,

The suggested explanation is supported in clinical practice by the quoted remark of Tuohy (p. 1)71 and by two reports of insufficient dermatomal spread of spinal anaesthesia caused by caudally positioned catheters which resulted in “failed CSA”.6 43 The first

Continuous spinal anaesthesia case was described by Morch, Rosenberg and Truant in 1957 in a remarkable study of intrathecal pharmacokinetics in patients during CSA with 5% hyperbaric lidocaine.43 Large-bore catheters were inserted 17–20 cm into the CSF (aiming to place the tip at T11–12) both for injection of the local anaesthetic and to enable CSF samples to be withdrawn for analysis of lidocaine. In one patient, only restricted anaesthesia was detectable, despite injection of lidocaine 200 mg (in 2–3 separate doses). They found that the CSF concentration of lidocaine was approximately 5–7 times greater than in patients who had a successful block. Radiographically it was demonstrated that the catheter had gone caudally, and had curled up on itself with its end in the dural cul-de-sac at the level of the second sacral vertebra. Morch, Rosenberg and Truant stated that “as the lidocaine injected in this patient would probably form a pool in the dural sac it is easy to see why the concentration of lidocaine in the spinal fluid samples was so high”.43 The second report is of two cases of failed CSA with 28-gauge spinal catheters.6 In both patients a very restricted block (sacral) was demonstrated after an initial intrathecal injection of 1.5 ml (75 mg) of 5% hyperbaric lidocaine. Radiographic evidence showed that in both patients the catheters were placed caudally with contrast dye pooling sacrally. In none of these cases were any neurological complications reported. MIXING OF LOCAL ANAESTHETIC IN CSF: EFFECT OF SPEED OF INJECTION

Normally, injection of local anaesthetic via a spinal needle or a large-bore (spinal) catheter causes turbulence at its tip, enabling the local anaesthetic to mix and dilute with the CSF to become evenly distributed. This process varies with the rate of injection (ml s91), so that in principle, the faster the local anaesthetic is injected the more thoroughly it mixes with the CSF. However, because of its very high internal resistance it is difficult to inject local anaesthetic fast enough through a microcatheter to ensure adequate mixing with CSF. With a caudally positioned catheter and the use of hyperbaric, highly concentrated local anaesthetics, such as 5% lidocaine, there is a risk that the poorly diluted local anaesthetic may remain in the caudal part of the dural sac for long enough to cause toxic lesions to the nerve roots.36 56 58 LOCAL ANAESTHETIC TOXICITY

Hyperbaric 5% lidocaine has been used uneventfully for more than 40 yr as a spinal anaesthetic, although mainly via a single injection technique, and it was not until recently that doubts were raised about the safety of this drug for spinal anaesthesia.14 25 36 59 64 After reports of the described severe neurological complications and in vitro toxicity tests on frog nerves, some investigators have declared 5% hyperbaric lidocaine neurotoxic, and recommend anaesthetists not to use it for spinal anaesthesia.14 16 36 However, the reported complications with 5% lidocaine were associated exclusively with its use in microcatheters. Lambert and Strichartz have shown in vitro with bullfrog sciatic nerves that both 5% lido-

593 caine and 0.5% tetracaine were neurotoxic, whereas both 0.75% bupivacaine and 7.5% dextrose were not.36 What seems to have happened in patients who developed cauda equina syndrome is an incompatibility between a “normally” safe drug and a new injection device. Thus when combining the old reliable 5% hyperbaric lidocaine with microcatheters, a new risk of neurological complications was created. Theoretically, the use of very thin spinal needles, through which aspiration of CSF is difficult, could cause a similar situation to that described for microcatheters. Consequently, the FDA has recommended that 5% hyperbaric lidocaine should be diluted with an equal volume of CSF before injection.21 “Transient radicular irritation” (TRI) syndrome, sometimes seen after spinal anaesthesia, is frequently confused with the neurotoxic cauda equina syndrome.14 TRI is a transient condition characterized by symmetric low back pain which usually resolves within 72 h. It normally responds to simple analgesics, and the described symptoms indicate that it is not a neurotoxic lesion.14 25 59 TRI is neither concentration nor dose dependant in contrast with the cauda equina syndrome.26 53

Factors affecting intrathecal distribution of local anaesthetic An even spread of local anaesthetic through the CSF giving an adequate block may not always occur with CSA6 7 15 40 43 71 and newcomers to the use of CSA can be surprised by the occasional failure or restricted block. As mentioned previously, one reason for a caudally restricted block is a sacral position of the catheter. CATHETER POSITION

Most anaesthetists intuitively aim to place the catheter in a cephalad direction to facilitate cranial spread of local anaesthetic. Its importance was described first by Tuohy71 and this has been confirmed by two recent studies.66 67 An important factor determining the direction of the catheter seems to be the length it is inserted into the subarachnoid space. Standl and Beck, in a study of 68 patients, found that microcatheters (28-gauge) inserted more than 4 cm had a 30% incidence of being placed caudally compared with 15% for those inserted less than 4 cm.65 Other studies with 28-gauge catheters by the same group reported an incidence of caudal catheter placement of 16–35% (table 6).65 67 The limited information available indicates a lower risk of caudal placement of 20-gauge catheters.74 In another study, Standl and colleagues found that the risk of caudal placement of 28-gauge catheters was reduced by using a directional spinal 22-gauge Sprotte needle.67 PATIENT POSITION

The position of the patient while the catheter is inserted is unlikely to influence the position of the catheter. Studies using 28-gauge microspinal catheters by Standl and colleagues have shown contrasting results. In one study they found that patients

594

British Journal of Anaesthesia

Table 6 Incidence of caudally placed catheters Study 74

Van Gessel Standl65 Standl67

PAIN MANAGEMENT

Catheter (gauge)

No. patients

Incidence (%)

Postoperative analgesia

20 28 28

28 68 21

7 16 35

There are a few published studies on the use of spinal catheters for postoperative analgesia,1 4 48 49 68 and only one comparative study with epidural analgesia or other techniques has been published.47 Postoperative analgesic regimens have been described for bupivacaine, morphine and diamorphine infusions. While experience with morphine and diamorphine has been successful,4 47 49 that with bupivacaine solutions has been inconsistent, in terms of both motor block and hypotension.48 68 There was also a high failure rate with microcatheters.47 49 68 A recent study in patients after hip and knee arthroplasty showed that infusion of a combination of bupivacaine 1 mg h91 and morphine 8 ␮g h91 provided similar postoperative analgesia as bupivacaine alone infused at a rate of 2 mg h91, but with less motor block.1 There was a high incidence of nausea and vomiting in the morphine group.

who had catheters inserted in the sitting position had only a 6% incidence of caudally directed catheters compared with 25% if placed with the patient in the right lateral position.65 However, in the other study they found a 35% incidence of caudal catheters despite insertion in the sitting position.67 VOLUME AND DOSE OF LOCAL ANAESTHETIC

Provided isobaric solutions are used,76 the spread of analgesia appears to be related mainly to the volume of local anaesthetic, while the total dose generally determines duration of block.3 34 60 78 The height of the block is best controlled by titration of intermittent injections of small volumes of local anaesthetic through the catheter and assessment of block level after each injection.2 8 15 32 34 60 67 69 BARICITY

When equal volumes and doses of local anaesthetic are injected with patients supine, segmental spread of block is greater and onset is faster with hyperbaric than either isobaric or hypobaric solutions,3 76 provided the spinal catheter is placed correctly. It should be remembered that plain bupivacaine solutions are slightly hypobaric at body temperature.39 CONCENTRATION

The concentration of local anaesthetic itself has little effect on the spread of spinal block, the spread primarily being a feature of speed of injection, volume and baricity.3 34 61 77 Provided equal volumes are used, block duration is shorter with less concentrated local anaesthetics. Except for the smaller volume, no advantage seems to be gained by use of local anaesthetics of a higher concentration than used normally, for example in epidural anaesthesia.5 77 As mentioned previously, highly concentrated local anaesthetics may cause neurotoxicity.

Indications and clinical use SURGICAL

The main advantages of CSA over epidural anaesthesia and SSA are its easier technique and the possibility of providing an adequate level and duration of anaesthesia with small intermittent doses of local anaesthetic, which also minimizes the risks of cardiovascular and respiratory disturbances. These qualities are of special value for lower abdominal and lower limb surgery in elderly and high-risk patients; they constitute the primary indication for CSA.2 8 15 19 23 32 33 40 60 69 73 When used in trauma patients, CSA appears to be associated with fewer technical problems and faster onset of anaesthesia compared with combined spinal–epidural anaesthesia.79

Chronic pain relief For long-term analgesia in cancer patients, continuous spinal analgesia with dilute mixtures of bupivacaine and morphine, or bupivacaine and buprenorphine, were effective for several months, with few side effects and no signs of neurotoxic reactions.31 63

Adverse effects/side effects CARDIOVASCULAR

Clinical studies have shown that haemodynamic stability is greater with CSA than with SSA or epidural anaesthesia. This is because of the possibility of titrating the level of anaesthesia with small incremental doses of local anaesthetic until adequate block is achieved.2 8 15 19 32 34 60 67 69 From this aspect, isobaric solutions were more predictable to use than either hypo- or hyperbaric solutions.3 76 POST-DURAL PUNCTURE HEADACHE (PDPH)

The incidence of PDPH after CSA remains controversial, with widely differing reported incidences. Some prospective studies have shown a very low incidence, while others found incidences of 6–9% (table 2).12 15 28 38 41 68 75 In retrospective studies, a generally low incidence of PDPH after CSA using large-bore catheters was found in older patients (table 1).22 27 30 40 50 The few published studies of the incidence of PDPH after CSA with microcatheters in older patients have indicated no difference compared with what could be expected with large-bore spinal catheters.12 28 62 68 The incidence of PDPH after CSA is probably commensurate with the age of the patient, size of the needle used and dural puncture technique—the latter was demonstrated by Mihic, who detected a much lower incidence of PDPH if the dura was punctured with the needle bevel plane parallel to the axis of the spine as opposed to being at right angles.42 In older patients, that is more than 60 yr, the incidence of PDPH is generally low after CSA with 20–24-gauge catheters (tables 1, 2). De Andres and colleagues found no difference in the incidence of PDPH after CSA with a 20-gauge catheter compared with SSA using 24-gauge Sprotte needles.13

Continuous spinal anaesthesia In younger patients, experience with large-bore catheters is limited22 and no large-scale prospective studies on the incidence of PDPH with microcatheters have been reported in this age group. However, a recent retrospective study showed a surprisingly high incidence of PDPH (33%) after use of 28-gauge microcatheters in obstetric patients for both vaginal and Caesarean delivery.27

Technical complications

595 attached to the needle, with the tip of the wire protruding from the injection end of the catheter (“Spinocath” B Braun).45 A modified Tuohy needle is placed in the epidural space, the “Spinocath” is then pushed through the dura into the CSF and the needle is pulled out leaving the catheter in the CSF. The theoretical advantage of this system is that the hole in the dura is the same size as the catheter, thus minimizing CSF leak. However, there is little published information on this technique and prospective studies are needed to evaluate its use, particularly in younger patients.

ANAESTHETIC

The average failure rate of CSA with 20-gauge catheters is 2.5% (table 3).7 15 27 40 51 68 69 75 This is comparable with the 4% found in a prospective study of 200 patients,46 and with 3.1% in another prospective study of 1891 patients who all underwent SSA.70 It is similar to failure rates of 0, 3.9 and 4.4% with 28gauge catheters found in three studies27 51 65 but in a fourth study, a 16% failure rate with 28-gauge catheters was reported (table 3).7 In contrast, the overall failure rate for 32-gauge microspinal catheters was 17%12 28 32 52 62 (table 4). Occasionally the catheter can be difficult to thread. This incidence varies from 4.5 to 20%15 51 67 75 76 and is usually a result of the needle tip being pushed too far into the subarachnoid space and lying too close to the opposite side of the spinal canal.44 Often this can be overcome by withdrawing the needle slightly within the subarachnoid space and readvancing the catheter. The incidence of transient paraesthesia on insertion of the catheter into the subarachnoid space can be as high as 24%.75 While there is a potential risk of infection when the dura is punctured, especially with long-term use, there are no reports of infective meningitis or sepsis after CSA. POSTOPERATIVE PAIN MANAGEMENT

There are limited data on the technical complications of the use of spinal catheters for postoperative pain control. The problems encountered range from disconnection of the catheter from the filter, obstruction of the catheter as a result of kinking and migration of the catheter tip out of the CSF.4 48 49 68 The 19% failure rate with 28-gauge catheters in one study made the authors change to 22-gauge catheters which reduced the failure rate to only 4%.49

Conclusion Continuous spinal anaesthesia is a well established technique and has clear advantages over epidural and single shot spinal anaesthesia, especially in elderly or high-risk patients. CSA offers possibilities of reduced interference with the cardiovascular and respiratory systems by allowing the anaesthetist to titrate small doses of local anaesthetic to reach the required level of spinal anaesthesia, and the duration can be extended to match the needs of prolonged surgery. As with all techniques, proper training is vital for the optimal use of CSA. The use of 20-gauge catheters for CSA can be recommended in older patients (i.e. more than 60 yr) in whom the incidence of PDPH is low. Most experience of large-bore catheters has been gained with 20gauge catheters and 18-gauge needles, and there is little published data on the use of 22- or 24-gauge catheters. Therefore, it is not possible to say if there is any benefit from the use of these thinner catheters. Experience to date indicates that there is no advantage in the use of microcatheters. There are little data on the use of CSA in younger patients, with particular reference to the incidence of PDPH; at present it cannot be recommended for routine use in younger patients. In obstetric patients, CSA, in common with single shot spinal anaesthesia, seems to carry an increased risk of PDPH, which reduces its advantages in these patients. Bupivacaine (0.25–0.5%) appears to be a safe local anaesthetic for use with CSA. The use of short-acting, high concentration, hyperbaric local anaesthetics for CSA is not recommended. In summary, correctly used, CSA is an effective, reliable and safe technique.

References Needle design It is well known that the risk of PDPH after SSA is lower if pencil-point spinal needles are used rather than cutting edged ones of similar gauge.24 In spite of this, surprisingly little attention has been paid to the use of pencil-point needles with CSA. The other potential advantage with the use of this type of needle is that the side hole would enable a more accurate placement of the catheter, as was found in a recent study by Standl and colleagues.67 Further studies need to be performed to evaluate the use of pencil-point needles for CSA with reference to the incidence of PDPH, particularly in younger patients. Recently, a new 22-gauge spinal catheter over a 27gauge needle has been developed with a thin wire

1. Bachman M, Laakso E, Niemi L, Rosenberg PH, Pitkanen M. Intrathecal infusion of bupivacaine with or without morphine for postoperative analgesia after hip and knee arthroplasty. British Journal of Anaesthesia 1997; 78: 666–670. 2. Brown S. Fractional segmental spinal anesthesia in poor risk surgical patients. Anesthesiology 1952; 13: 416–428. 3. Brown DT, Wildsmith JAW, Covino BG, Scott DB. Effect of baricity on spinal anaesthesia with amethocaine. British Journal of Anaesthesia 1980; 52: 589–596. 4. Burchett KR, Denny NM. Initial experience of continuous subarachnoid diamorphine infusion for postoperative pain relief. Regional Anesthesia 1991; 16: 253–256. 5. Chambers WA, Littlewood DG, Edstrom HH, Scott DB. Spinal anaesthesia with hyperbaric bupivacaine: effects of concentration and volume administration. British Journal of Anaesthesia 1982; 54: 75–80. 6. Chan VWS, Smyth RJ. Radiographic examination of catheter position in restricted sacral block after continuous spinal anesthesia. Anesthesia and Analgesia 1992; 75: 449–452.

596 7. Chan VWS, Chung F, Gomez M, Seyone C, Baylon G. Anesthetic and hemodynamic effects of single versus incremental titration of hyperbaric spinal lidocaine through microcatheter. Anesthesia and Analgesia 1994; 79: 117–123. 8. Collard CD, Eappen S, Lynch EP, Concepcion M. Continuous spinal anesthesia with invasive hemodynamic monitoring for surgical repair of the hip in two patients with severe aortic stenosis. Anesthesia and Analgesia 1995; 81: 195–198. 9. Cohen S, Amar D, Pantuck EJ, Singer N, Divon M. Decreased incidence of headache after accidental dural puncture in caesarean delivery patients receiving continuous intrathecal analgesia. Acta Anaesthesiologica Scandinavica 1994; 38: 716–718. 10. Curbello MM. Continuous peridural segmental anesthesia by means of a urethral catheter. Current Research in Anesthesiology 1949; 28: 13–16. 11. Dean HP. Discussion on the relative value of inhalation and injection methods of inducing anaesthesia. British Medical Journal 1907; 2: 869–877. 12. De Andres J, Bellver J, Bolinches R. Comparison of continuous spinal anaesthesia using a 32G catheter with anaesthesia using a single dose 24G atraumatic needle in young patients. British Journal of Anaesthesia 1994; 73: 747–750. 13. De Andres JA, Febre E, Bellver J, Bolinches R. Continuous spinal anaesthesia versus single dosing. A comparative study. European Journal of Anaesthesiology 1995; 12: 135–140. 14. de Jong RH. Last round for a “heavyweight”? Anesthesia and Analgesia 1994; 78: 3–4. 15. Denny N, Masters R, Pearson D, Read J, Sihota M, Selander DE. Postdural puncture headache after continuous spinal anesthesia. Anesthesia and Analgesia 1987; 66: 791–794. 16. Denny NM. Continuous spinal anaesthesia and cauda equina syndrome. Anaesthesia 1995; 50: 474. 17. Drasner K. Models for local anesthetic toxicity from continuous spinal anesthesia. Regional Anesthesia 1993; 18: 434–438. 18. Dripps RD. A comparison of the malleable needle and catheter techniques for continuous spinal anesthesia. New York Journal of Medicine 1950; 50: 1595–1599. 19. Favarel-Garrigues JF, Sztark F, Petietjean ME, Thicoipe M, Lassie P, Dabad P. Hemodynamic effects of spinal anesthesia in the elderly: single dose versus titration through a catheter. Anesthesia and Analgesia 1996; 82: 312–316. 20. FDA. Safety alert. Cauda equina syndrome associated with use of small-bore catheters in continuous spinal anesthesia. May 29, 1992. 21. F-D-C Reports—“Pink Sheet”. Astra 5% hyperbaric xylocaine labeling for spinal anesthesia should recommend 1-to-1 drug dilution to reduce neurotoxicity risk. FDA CMTE, concludes June 13, 1994. 22. Giuffrida JG, Bizzarri DV, Masi R, Bondoc R. Continuous procaine spinal anesthesia for cesarian section. Anesthesia and Analgesia 1972; 51: 117–124. 23. Grace D, Orr DA. Continuous spinal anaesthesia in acute respiratory failure. Anaesthesia 1993; 48: 226–228. 24. Halpern S, Preston R. Postdural puncture headache and spinal needle design. Anesthesiology 1994; 81: 1376–1388. 25. Hampl KF, Schneider MC, Ummenhofer W, Drewe J. Transient neurologic symptoms after spinal anesthesia. Anesthesia and Analgesia 1995; 81: 1148–1533. 26. Hampl KF, Schneider MC, Pargger H, Gut J, Drewe J, Drasner K. A similar incidence of transient neurological symptoms after spinal anesthesia with 2% and 5% lidocaine. Anesthesia and Analgesia 1996; 83: 1051–1054. 27. Horlocker TT, McGregor DG, Matsushige DK, Chantigian RC, Schroeder DR, Besse JA. Neurological complications of 603 continuous spinal anesthetics using macrocatheter and microcatheter techniques. Anesthesia and Analgesia 1997; 84: 1063–1070. 28. Hurley RJ, Lambert DH. Continuous spinal anesthesia with a microcatheter technique. Preliminary experience. Anesthesia and Analgesia 1990; 70: 97–102. 29. Jaradeh S. Cauda equina syndrome: A neurologist’s perspective. Regional Anesthesia 1993; 18: 473–480. 30. Kallos T, Smith TC. Continuous spinal anesthesia with hypobaric tetracaine for hip surgery in lateral decubitus. Anesthesia and Analgesia 1972; 51: 76–62. 31. Karlsson P-Å, Nordborg C, Sjöberg M, Nitescu P, Appelgren L, Curelaru I. Neuropathological findings after long-term intrathecal infusion of buprenorphine and bupivacaine for pain treatment in cancer patients. European Journal of Pain 1994; 15,4: 69–76.

British Journal of Anaesthesia 32. Klimscha W, Weinstabl C, Ilias W, Mayer N, Kashanipour A, Scheider B, Hammerle A. Continuous spinal anesthesia with a microcatheter and low-dose bupivacaine decreases the hemodynamic effects of centro-neuroaxis blocks in elderly patients. Anesthesia and Analgesia 1993; 77: 275–280. 33. Kopacz D. Continuous spinal anaesthesia for abdominal surgery in a patient receiving amiodorone. Canadian Journal of Anaesthesia 1991; 38: 341–344. 34. Labaille L, Benhamou D, Westermann J. Hemodynamic effects of continuous spinal anesthesia: A comparative study between low and high doses of bupivacaine. Regional Anesthesia 1992; 17: 193–196. 35. Lambert DH, Hurley RJ. Cauda equina syndrome and continuous spinal anesthesia. Anesthesia and Analgesia 1991; 72: 817–819. 36. Lambert DH, Strichartz GR. Irreversible conduction block in isolated nerve by high concentrations of local anesthetics. Anesthesiology 1994; 80: 1082–1093. 37. Lemmon WT. A method for continuous spinal anesthesia. Annals of Surgery 1940; 111: 144. 38. Liu N, Montefiore A, Kermarec N, Rauss A, Bonnet F. Prolonged placement of spinal catheters does not prevent postdural puncture headache. Regional Anesthesia 1993; 18: 110–113. 39. Logan MR, McClure JH, Wildsmith JA. Plain bupivacaine: an unpredictable spinal anesthetic agent. British Journal of Anaesthesia 1986; 58: 292–296. 40. Mahisekar UL, Winnie AP, Vasiroddy AO, Masters RW. Continuous spinal anesthesia and postdural puncture headache. A retrospective study. Regional Anesthesia 1991; 16: 107–111. 41. Mazze RI, Fujinaga M. Postdural puncture headache after continuous spinal anesthesia with 18G and 20G needles. Regional Anesthesia 1993; 18: 447–451. 42. Mihic DN. Postspinal headache and relationship of needle bevel to longitudinal dural fibres. Regional Anesthesia 1985; 10: 76–81. 43. Morch ET, Rosenberg MK, Truant AT. Lidocaine for spinal anesthesia. A study of the concentration in the spinal fluid. Acta Anaesthesiologica Scandinavica 1957; 1: 105–115. 44. Mollmann M, Holst D, Lubbesmeyer H, Lawin P. Continuous spinal anesthesia: Mechanical and technical problems of catheter placement. Regional Anesthesia 1993; 18: 469–472. 45. Mollmann M, Van Steenberge A, Sell A, Pitkanen M, Holst D, Van Dongen A, Berg S. Spinocath, a new approach to continuous spinal anaesthesia—Preliminary results of a multicentre trial. International Monitor on Regional Anaesthesia 1996; 8: 74. 46. Munhall RJ, Sukhani R, Winnie AP. Incidence and etiology of failed spinal anesthetics in a university hospital: A prospective study. Anesthesia and Analgesia 1988; 67: 843–848. 47. Niemi L, Pitkanen M, Tuominen M, Rosenberg PH. Technical problems and side effects associated with continuous intrathecal or extradural post-operative analgesia in patients undergoing hip arthroplasty. European Journal of Anaesthesiology 1994; 11: 469–474. 48. Niemi L, Pitkanen M, Dunkel P Laakso E, Rosenberg PH. Evaluation of the usefulness of intrathecal bupivacaine infusion for analgesia after hip and knee arthroplasty. British Journal of Anaesthesia 1996; 77: 544–545. 49. Niemi L, Pitkanen MT, Tuominen MK, Rosenberg PH. Comparison of intrathecal fentanyl infusion with intrathecal morphine infusion or bolus for postoperative pain relief after hip arthroplasty. Anesthesia and Analgesia 1993; 77: 126–130. 50. Peterson DO, Borup LJ, Chestnut JS. Continuous spinal anesthesia, case review and discussion. Regional Anesthesia 1983; 8: 109–113. 51. Petros AJ, Barnard M, Smith D, Ronzoni G. Continuous spinal anesthesia: Dose requirements and characteristics of the block. Regional Anesthesia 1993; 18: 52–54. 52. Pitkanen M, Tuominem M, Rosenberg P, Wahlstrom T. Technical and light microscopic comparison of four small catheters used for continuous spinal anesthesia. Regional Anesthesia 1992; 17: 288–291. 53. Pollock JE, Neal JM, Stephenson CA, Wiley CE. Prospective study of transient radicular irritation in patients undergoing spinal anesthesia. Anesthesiology 1996; 84: 1361–1367. 54. Rao TLK, El-Etr AE. Anticoagulation following placement of extradural and subarachnoid catheters: an evaluation of neurologic sequelae. Anesthesiology 1981; 55: 618–620. 55. Rigler ML, Drasner K. Distribution of catheter-injected local anesthetic in a model of the subarachnoid space. Anesthesiology 1991; 75: 684–692.

Continuous spinal anaesthesia 56. Rigler ML, Drasner K, Krejcie TC, Yelish SJ, Scholnick FT, Defontes J, Bohner D. Cauda equina syndrome after continuous spinal anesthesia. Anesthesia and Analgesia 1991; 72: 275–281. 57. Ross BK, Coda B, Heath CH. Local anesthetic distribution in a spinal model: A possible mechanism of neurologic injury after continuous spinal anesthesia. Regional Anesthesia 1992; 17: 69–77. 58. Schell RM, Brauer FS, Cole DJ, Applegate RL. Persistent sacral nerve root defects after continuous spinal anaesthesia. Canadian Journal of Anaesthesia 1991; 38: 908–911. 59. Schneider M, Ettlin T, Kaufman M, Schumacher P, Urwyler A, Hampl K, Hochsetter A. Transient neurologic toxicity after hyperbaric subarachnoid anesthesia with 5% lidocaine. Anesthesia and Analgesia 1993; 76: 1154–1157. 60. Schnider W, Mueller-Duysing S, Johr M, Gerber H. Incremental dosing versus single-dose spinal anesthesia and hemodynamic stability. Anesthesia and Analgesia 1993; 77: 1174–1178. 61. Sheskey MC, Rocco AG, Bizzarri-Schmid M, Francis DM, Edstrom H, Covino BG. A dose response study of bupivacaine for spinal anesthesia. Anesthesia and Analgesia 1983; 62: 931–935. 62. Silvanto M, Pitkanen M, Tuominen M, Rosenberg PH. Technical problems associated with the use of 32G and 22G spinal catheters. Acta Anaesthesiologica Scandinavica 1992; 36: 295–299. 63. Sjöberg M, Karlsson P-A, Nordborg C, Wallgren A, Nitescu P, Appelgren L, Linder L-E, Curelaru I. Neuropathologic findings after long-term intrathecal infusion of morphine and bupivacaine for pain treatment in cancer patients. Anesthesiology 1992; 76: 173–186. 64. Sjostrom S, Blass J. Severe pain in both legs after spinal anaesthesia with hyperbaric 5% lignocaine solution. Anaesthesia 1994; 49: 700–702. 65. Standl T, Beck H. Radiological examination of the intrathecal position of microcatheters in continuous spinal anaesthesia. British Journal of Anaesthesia 1993; 71: 803–806. 66. Standl T, Beck H. Influence of the subarachnoid position of micro-catheters on the onset of analgesia and dose of plain bupivacaine 0.5% in continuous spinal anesthesia. Regional Anesthesia 1994; 19: 231–236.

597 67. Standl T, Eckert S, Rundshagen I, Schulte am Esch J. A directional needle improves effectiveness and reduces complications of microcatheter continuous spinal anaesthesia. Canadian Journal of Anaesthesia 1995; 42: 701–705. 68. Standl T, Eckert S, Schulte am Esch J. Microcatheter continuous spinal anaesthesia in the post-operative period: a prospective study of its effectiveness and complications. European Journal of Anaesthesiology 1995; 12: 273–279. 69. Sutter P-A, Gamulin Z, Forster A. Comparison of continuous spinal and continuous extradural anaesthesia for lower limb surgery in elderly patients. A retrospective study. Anesthesia 1989; 44: 47–50. 70. Tarkkila PJ. Incidence and causes of failed spinal anesthetics in a university hospital: A prospective study. Regional Anesthesia 1991; 16: 48–52. 71. Tuohy EB. Continuous spinal anesthesia, its usefulness and technique involved. Anesthesiology 1944; 5: 142–148. 72. Tuohy EB. Continuous spinal anesthesia: A new method of utilizing ureteral catheter. Surgery Clinics of North America 1945; 25: 834–840. 73. Underwood RJ. Continuous spinal anesthesia in physical status group IV patients. Anesthesia and Analgesia 1968; 47: 18–26. 74. Van Gessel EF, Forster A, Gamulin Z. Continuous spinal anesthesia: Where do spinal catheters go? Anesthesia and Analgesia 1993; 76: 1004–1007. 75. Van Gessel E, Forster A, Gamulin Z. A prospective study of feasibility of continuous spinal anesthesia in a University hospital. Anesthesia and Analgesia 1995; 80: 880–885. 76. Van Gessel EF, Forster A, Schweizer A, Gamulin Z. Comparison of hypobaric, hyperbaric and isobaric solutions of bupivacaine during continuous spinal anesthesia. Anesthesia and Analgesia 1991; 72: 779–784. 77. Van Zundert AAJ, Grouls RJE, Korsten HHM, Lambert DH. Spinal anesthesia volume or concentration—What matters? Regional Anesthesia 1996; 21: 112–118. 78. Wildsmith JAW, McClure JH, Brown DT, Scott DB. Effects of posture on the spread of isobaric and hyperbaric amethocaine. British Journal of Anaesthesia 1981; 53: 273–278. 79. Wilhelm S, Standl T, Burmeister M, Kessler G, Schulte am Esch J. Comparison of continuous spinal with combined spinal-extradural anesthesia using plain 0.5% bupivacaine in trauma patients. Anesthesia and Analgesia 1997; 85: 69–74.