Abstract

It has been suggested that obstetric epidurals lead to chronic adhesive arachnoiditis (CAA). CAA is a nebulous disease entity with much confusion over its symptomatology. This review outlines the pathological, clinical, and radiological features of the disease. The proposed diagnostic criteria for CAA are: back pain that increases on exertion, with or without leg pain; neurological abnormality on examination; and characteristic MRI findings. Using these criteria, there is evidence to show that epidural or subarachnoid placement of some contrast media, preservatives and possibly vasoconstrictors, may lead to CAA. No evidence was found that the preservative‐free, low concentration bupivacaine with opioid mixtures or plain bupivacaine currently used in labour lead to CAA.

Br J Anaesth 2004; 92: 109–20

Accepted for publication: March 13, 2003

Chronic adhesive arachnoiditis (CAA) is an extremely rare but debilitating condition, that has recently received increased media attention. Less than 1000 cases have been reported in the last 50 yr.69 On April 15, 2001, the Sunday Express newspaper ran a double page article entitled ‘Birth Jabs Cripple Women’, outlining what they described as ‘the scandal of epidurals that have wrecked lives’. They claimed that epidurals for labour have left thousands of women disabled or paralysed; and that this ‘fact’ was one of the NHS’s most closely guarded secrets. A week later, they ran two articles: one entitled ‘Time to acknowledge this danger’ implied a reluctance of the medical profession to acknowledge the iatrogenic causes of arachnoiditis; and the other including an alleged quote, about epidurals in labour, from a former director of women and children’s health at WHO: ‘They are being told they are safe. This is a lie’. This article was supported by The Arachnoiditis Trust whose patron had written an article easily accessible on the internet (www.backtalk.nildram.co.uk/arach.htm), in which she claimed that epidural anaesthesia is implicated in the aetiology of arachnoiditis. In these days of increasing accountability and public interest in medical malpractice, issues such as these are difficult to ignore.

Although the incidence of irreversible neurological complications after epidural anaesthesia is very low,28 the Woolley and Roe case27 in 1954 serves to illustrate the catastrophic effect neural damage caused by central nerve block can have. After this report, public confidence in spinal anaesthesia disappeared for two decades. Increasingly, pregnant women are asking about the risks of arachnoiditis after an epidural. This information is not readily available. We therefore conducted a review of the current literature for any evidence of obstetric epidurals causing CAA.

Nature of arachnoiditis

Arachnoiditis was first recognized as a separate disease entity in 1909 by Victor Horsley. Since this time, it has been described by several authors using varied terminology. Titles used include: chronic spinal arachnoiditis, adhesive spinal arachnoiditis, meningitis serosa circumscripta spinalis, chronic spinal meningitis, spinal meningitides with radiculomyelopathy, lumbar adhesive arachnoiditis, spinal arachnoiditis, spinal fibrosis, and lumbosacral adhesive arachnoiditis. Arachnoiditis can be described as arachnoiditis ossificans, calcific arachnoiditis, or pachymeningitis, depending upon the extent of the radiographic or pathological findings. Indeed, there is debate as to whether this constitutes a single disease entity. It should also be noted that the radiological and pathological findings do not invariably correlate with the clinical features. Thus, there is much confusion over this disease process. More recently, CAA has been used to describe clinically significant non‐specific inflammation of the arachnoid and intrathecal neural elements. We shall use this terminology throughout this review.

Anatomy

The arachnoid consists of many layers of flat squamous cells lying one on top of another with potential space between the layers. A network containing collagen, elastic fibres, and blood vessels holds these layers together.101 Drugs pass from the epidural space, through the dura, arachnoid and pia mater to produce their effect.108 It is the arachnoid rather than the dura that is the principle barrier to drugs in the epidural space reaching the spinal nerves.11100

Intercellular pores have been demonstrated in animal and human arachnoid.101 In rabbits, these are large enough to allow the passage of erythrocytes.103 Arachnoid mater covering the ventral and dorsal roots of spinal nerves has proliferations of cells, which form villi.101 These have been classified depending upon their degree of protrusion through the dura. Types IV and V breech the dura, with Type V also protruding into the epidural space (see the two parts of Fig. 1).

Thus, the arachnoid is a dynamic structure through which substances placed in the epidural space must pass in order to have their effect. As the epidural and subarachnoid spaces are not entirely separate, substances placed in either space may have effects on this delicate structure.

Pathology

The progressive inflammation of the arachnoid that occurs with CAA was described in the 1970s by Burton.22 He described an initial stage, ‘radiculitis’, involving inflammation of the pia‐arachnoid with nerve root swelling and hyperaemia. Strands of collagen begin to form between the nerve roots and the pia‐arachnoid. ‘Arachnoiditis’ follows, characterized by collagen deposition, a decrease in nerve root swelling, and adherence of the nerve roots to each other. ‘Adhesive arachnoiditis’ is the resolution of the inflammatory process, with dense collagen deposition. This causes complete encapsulation of the nerve roots, which undergo progressive atrophy, as a result of interference with their blood supply. Microscopic studies have shown characteristic arteritis in the blood vessels of the chronically inflamed arachnoid.111 It is not yet known whether this is as a result of the arachnoiditis or the cause.

It would appear that although in the past CAA has been described mainly in the thoracic or cervical region, since the 1950s there has been a trend towards a higher incidence in the lumbar spine. In 1978, Shaw reported that 71% of cases involved the lumbosacral spine alone.102 This may reflect a change in aetiological factors.

The arachnoiditis adhesions generally occur on the dorsal segments;68 the reason for this is not fully understood. With the exception of rare cystic forms, the adhesions are arranged peripherally and have been described as looking ‘like the bark of a tree’, when viewed by myelography.68

Clinical features

CAA presents a complex clinical picture. Because of the varied symptomatology, clinical diagnosis is difficult. The precise relationship between the pathological findings and symptomatology has not been defined.

Back pain with or without leg symptoms (e.g. pain, paraesthesia, or weakness) is typical, but a wide range of neurological abnormalities have been associated with CAA.52 Physical signs are not specific, although there is generally some abnormality to be found. As yet, a typical clinical syndrome has not been identified. Reported case series show a variety of symptoms and signs (Table 1).

The following clinical features occur most frequently: back pain increased by activity; leg pain, often bilateral; hyporeflexia; decreased range of movement of the trunk; sensory abnormality; decreased straight leg raising; and urinary sphincter dysfunction. These clinical features can lead to CAA being erroneously diagnosed as spinal stenosis, spinal tumour, a lumbar disc lesion, or any other compressive lesion of the spinal cord.

Earlier case series described CAA as a progressive disease.39 Most of these patients had cervical or thoracic CAA.102 With fewer cases being secondary to infection, most CAA now occurs in the lumbar region.102 Although the course of CAA is typically irregular, more recent case series report the disease as progressive in 1.8–33% of patients and static in 50–59%.4669102

Laboratory studies are not helpful in the diagnosis of CAA. Some have reported an increase in CSF proteins with CAA,81 but this is not thought to be a reliable indicator of the disease.8194660102 Clinical neurophysiological testing (e.g. electromyelography) is also not useful in the diagnosis of CAA.194669

Radiological features

The myelographic appearance of CAA is variable and includes: a homogeneous contrast pattern without root shadows; prominent nerve roots; and subarachnoid filling defects (partial or complete block, loculation or pseudocyst formation), with narrowing and shortening of the thecal sac.2957106 MRI changes of CAA are: conglomerations of roots residing centrally in the dural sac, or adhesions tethering the nerve roots peripherally, giving rise to an ‘empty sac’ appearance, and soft tissue replacing the subarachnoid space.88 It has been reported that these changes seen on MRI have a sensitivity of 92%, a specificity of 100%, and an accuracy of 99% in the diagnosis of CAA.88

Proposed definition of CAA

From our study of the literature, we have attempted to define the common features that characterize CAA:

• Back pain, increasing with activity.

• Leg pain, which may be bilateral.

• Some neurological abnormality on examination, most commonly hyporeflexia.

• MRI changes consistent with CAA (myelography changes were accepted for earlier studies).

Suspected aetiologies

Since it was first described in 1909, various factors have been implicated in the aetiology of lumbosacral adhesive arachnoiditis. In the 19th century, infections such as syphilis, gonorrhoea, and tuberculosis were the most prevalent causes, whereas in the 1940s blood in the CSF (after subarachnoid haemorrhage (SAH) or surgery) became the most important cause. More recently, the following have been implicated in the aetiology of CAA: contrast media, epidural steroids, trauma, blood, preservatives, contaminants, vasoconstrictors, and local anaesthetics. Unfortunately, because of the rarity of this disease, there are no randomized controlled trials available. Case reports of CAA and studies involving these suspected aetiologies were reviewed. The proposed definition of CAA was applied to each study to assess its relevance based on our criteria. In older studies, where MRI was unavailable, a diagnosis of CAA on myelography findings as outlined was allowed.

Contrast media

Since the 1970s, contrast myelography has been a common cause of arachnoiditis. Ethyliophendylate (‘Myodil’ in UK, ‘Pantopaque’ in USA) is an ionized fatty acid compound that is extremely radio‐opaque. The incidence of adhesive arachnoiditis after its use is dose‐dependant and has been quoted as 1%.102 It has a prolonged excretion time, sometimes up to 1 yr. It has therefore been suggested that the contrast media should be removed from the CSF immediately after imaging.61 However, there is no evidence that this reduces the incidence of arachnoiditis. A study in dogs has shown an increased inflammatory reaction with ethyliodophendylate plus blood and it is now advised to abandon the procedure if a bloody tap occurs.57 The use of oil‐based iodine agents has largely been abandoned,56 but water‐based agents are also capable of producing arachnoiditis, thought to be related to their tonicity.105 Metrizamide is thought to be the safest; its clearance from the CSF has a half‐life of 4 h, and there have been no reported cases of arachnoiditis after its use in humans (although arachnoiditis can be induced in monkeys using very high concentrations).61

Epidural steroids

Injection of corticosteroid preparations into the epidural space in an attempt to relieve back pain is a common procedure. Corticosteroids used include: hydrocortisone acetate, methylprednisolone acetate (MPA), methylprednisolone succinate, and triamcinolone. MPA (e.g. Depomedrone) is the most commonly used. There have been reports of CAA after intrathecal injection of MPA,12137591 leading to calls to abandon its use.1013 Some doubt the validity of these claims.112

MPA is suspended in polyethylene glycol (a non‐ionic detergent), and myristyl gamma chloride (a long chain fatty acid) to reduce its aqueous solubility. It is thought that polyethylene glycol is the trigger for CAA.101275 Other steroid agents do not contain polyethylene glycol but do contain bacteriostatic agents, for example benzyl alcohol or phenol, which are considered more noxious.13

Some authors feel that MPA should not be used in the epidural space, because of the potential for transfer to the intrathecal space, possibly leading to CAA.75 However, animal experiments have not shown significant inflammatory changes in the meninges after epidural MPA or triamcinolone.42531 Dilution of steroid with saline or local anaesthetic before injection into the epidural space lowers the concentration of polyethylene glycol.10 This may be why there are no reports of CAA after uncomplicated injection of epidural corticosteroids. Indeed, Abram and O’Connor found no cases of CAA after administration of epidural corticosteroids in their large review.3

Blood

It has been suggested that blood in the CSF can lead to an inflammatory reaction. Cases of CAA have been reported after SAH, the so called ‘aseptic haemogenic meningitis’ occurring several days post‐SAH.48107 Nelson reported a case in which post‐mortem studies showed an increase in inflammation of the pia‐arachnoid leading to fibrosis after SAH.76 It has been suggested that the breakdown products of haemoglobin form free radicals, which can cause damage to nerves.5882 Indeed, placement of the breakdown products of blood into the CSF of dogs causes more meningeal inflammation than does fresh blood.58 Other researchers have found that the deliberate placement of autologous blood in the epidural space produced no more tissue reaction than a normal lumbar puncture and so does not result in chemical meningitis.33 Obviously, this is relevant to those patients requiring an epidural blood patch for treatment of postdural puncture headache. Furthermore, bleeding into the epidural space can occur on insertion of an epidural needle or catheter, a ‘bloody tap’. Incidences for this have been quoted up to 18%.72 Evidence that minor bleeding is not uncommon with the insertion of epidural catheters has been found at epiduroscopy.15

A case has been reported in which an epidural blood patch was alleged to have caused CAA.5 Several attempts were made to locate the epidural space in a 34‐yr‐old woman in labour. After an epidural catheter had been inserted, the injection of 10 ml of bupivacaine 0.25% resulted in sensory analgesia to T2, suggestive of subdural placement of the catheter. After delivery, 19 ml of autologous blood were injected down the catheter as prophylaxis against postdural puncture headache. Five days later, the patient complained of backache, a burning sensation in both feet, and photophobia. An MRI scan showed a subdural haematoma, atypical clumping of the nerve roots, and also an extradural collection of blood. The patient was treated with anti‐inflammatory drugs and phenytoin, but did not improve significantly. Although the patient had good clinical and MRI evidence of arachnoiditis, it was probably a result of the subdural, rather than epidural, blood patch. The use of the catheter to place the blood patch must be questioned, as it was already doubtful that the tip of the catheter was in the epidural space.

Abouliesh2 followed up 118 epidural blood patches over a 2‐yr period. He found 19 cases of residual backache, three cases of limited back movement, and two cases with occasional radicular pain down both legs, but no cases of CAA. However, the study contains little detail of how the patients were followed up. Although it would appear that some were examined, there is no mention of further investigations including myelography or MRI.

Trauma

It is well known that CAA occurs after spinal surgery, particularly if it is either extradural,1781or repeated. It has been implicated for many years as a factor in Failed Back Surgery Syndrome.

It has been suggested that the epidural catheter may cause an inflammatory reaction in the epidural space, particularly if left in the epidural space long‐term.6265 In rats, a fibrous sheath has been shown to form around an epidural catheter after it had been left in situ for several days.36 Moderate inflammatory changes can be seen at post‐mortem in some patients who have had continuous epidural catheters in situ, with thickening along the indentations of the dura where there had been contact with the catheter.114 It should be noted that epidural catheters are not commonly left in situ for long periods of time in obstetric practice.

It was recognized in the 1960s that traumatic lumbar puncture led to an increase in CSF proteins, which did not occur in uncomplicated cases.971 This was postulated to be evidence of meningeal irritation; an inflamed meningeal barrier would allow more protein to cross into the CSF. Transient paraesthesiae occur during 24–44% of epidural catheter placements indicating possible trauma.87 There is no evidence, however, to suggest that this meningeal irritation progresses to CAA. It has been suggested that the incidence of prolonged neurological abnormalities may be increased if paraesthesiae are elicited during insertion of an epidural needle or catheter.114

Reynolds documented seven cases, six obstetric and one surgical, in which neurological damage followed spinal or combined spinal epidural (CSE) anaesthesia.83 All patients experienced pain during the insertion of the spinal needle, which was believed by the operator to be at the L2/L3 interspace. MRI showed irrefutable evidence of spinal cord damage. However, there was no CAA demonstrable on MRI.

Haisa47 reported a case in Tokyo of CAA after obstetric epidural anaesthesia. At the time of epidural insertion, the patient felt a sudden sharp pain radiating down the left leg, and continued to complain of pain in her left leg and buttock. After a presumed diagnosis of disc herniation, the symptoms were treated by repeated epidurals each containing local anaesthetic and steroid. She also underwent myelography before an MRI, which was diagnostic of CAA. Unfortunately, this case report did not reveal details as to which drug was placed in the epidural and subarachnoid spaces during the many injections the patient received. It is therefore difficult to determine the exact aetiology of the CAA, but it should be noted that there were contributing factors including the traumatic epidural insertion, use of repeated epidural steroids, and myelography.

Detergents and contaminants

The Woolley and Roe case of 1954 famously advanced contamination with detergents as a cause of neurological abnormalities after spinal anaesthesia.27 Indeed, at laminectomy, Cecil Roe had thickening and cyst formation of his arachnoid mater, suggestive of CAA. Several cases have been described of neurological abnormalities,113 aseptic meningitis,4396 and CAA,79 after subarachnoid blocks. The authors all postulated that these were because of contamination with detergents used to clean instruments. However, none put forward good evidence to support their claims.

Aseptic meningitis has been reported after CSE for analgesia during labour, thought to be a result of contamination by the chlorhexidine spirit used to clean the patient’s back.49 Experimentally, intrathecal detergents can cause a pronounced cellular proliferation of the arachnoid in monkeys, dependant upon the detergent used and its concentration.32111 However, concentrations used were far in excess of those that could contaminate spinal anaesthetic equipment under normal clinical conditions.

The needle‐through‐needle technique is probably the commonest method of establishing CSE.26 Some are concerned that friction between the needles produces metallic fragments that are then introduced into the subarachnoid space, causing an inflammatory reaction.37 This has been postulated as a cause of aseptic meningitis.38 However, it is debatable whether metallic fragments are indeed produced.235051 Medical grade stainless steel needles do not cause inflammation in nickel‐sensitive patients.40

Vasoconstrictors

Boiardi and colleagues reported four cases of CAA occurring after non‐obstetric epidural anaesthesia, using epidural bupivacaine with epinephrine.16 The authors suggested that a subarachnoid hyperaemic reaction had occurred secondary to drug placement in the epidural space. No detergents were used, but it is not clear whether preservatives were present in the solutions used. Chliapparini and colleagues reported 16 cases of serious neurological deficit after lumbar epidural anaesthesia over a 7‐yr period, including nine cases of CAA developing 1 month to 8 yr later.24 Only one case involved an obstetric epidural. All used bupivacaine, with six also using epidural epinephrine (concentration unstated). CAA was diagnosed using MRI in seven cases and myelography in the remaining two. The obstetric case had epidural bupivacaine with epinephrine for analgesia during labour. Ten months later, she developed spastic paraparesis and decreased sensation. There was a delay of 10 yr before MRI findings diagnosed CAA. The authors allude to the fact that preservatives in the epidural solution may have been the causative factor, although which preservatives were used was not stated.

Preservatives

Sghirlanzoni and colleagues99 reported six patients from Italy, diagnosed with CAA using myelography, up to 3 yr after receiving non‐obstetric epidural anaesthesia between 1983 and 1988. All their cases had received epidural local anaesthetics from multiple dose vials containing the preservatives, methyl and propyl paraben. Sklar and colleagues106 reported seven cases of CAA diagnosed using MRI, which were referred to their Miami hospital after epidural analgesia in labour, 2 months to 12 yr previously. All these women originated from South America and it is thought that they may all have received lidocaine 2% with the preservatives metabisulfite and methylparaben. Both these preservatives were banned in the USA at the time of this study, as they were known to have toxic effects. Indeed, Gissen’s study in rabbits demonstrated the neurotoxicity of sulfite‐containing preservatives, although there was no comment on the reaction of the meninges.42

Local anaesthetics

Gemma and colleagues41 reported a case of CAA diagnosed by MRI after a non‐obstetric epidural of bupivacaine 0.5% without epinephrine. An epidural catheter was not used. Little information was given on this case. No information was reported concerning the use of detergents or preservatives. The patient suffered from lumbar pain with paraesthesiae in his big toes, more prominent on the left side.

Arachnoid reactions to local anaesthetics are thought to be a function of their histotoxic properties.44 Topically applied local anaesthetics cause altered perineural permeability and oedema of nerve fibres.74 Myers74 suggested that ester local anaesthetics (e.g. chloroprocaine and tetracaine) are more neurotoxic, producing significant oedema of perineural tissues at clinically relevant concentrations. Topical application of increasing concentrations of bupivacaine to rabbit sciatic nerves in vitro showed increased adherence of nerves to each other, suggestive of a dose‐dependant inflammation of perineural tissues.98 Although hyperglycaemia worsens neurological recovery after an ischaemic event, the addition of glucose 7.5% does not appear to increase neurotoxicity.92 There is no proof that ‘allergic sensitivity’ of the meninges to local anaesthetic occurs.44 Cases of CAA who were tested for allergy to local anaesthetics have proved negative.6364

CAA and cauda equina syndrome have been widely reported after spinal anaesthesia using a continuous micro‐catheter technique.63668693 Drasner reported a case of cauda equina syndrome after continuous epidural analgesia in a 52‐yr‐old man.34 However, there was doubt as to whether the tip of the epidural catheter was extradural throughout the case. It has been shown by the use of glass spine models that injection of lidocaine 5% with glucose 7.5% has a non‐uniform distribution in the CSF, because of the slow flow rate as it leaves the tip of the microcatheter.6685 It is thought that this maldistribution unmasks the neurotoxic potential of local anaesthetics, by facilitating areas of high concentration within the CSF.86 This is potentiated by the use of hyperbaric solutions.86 Animal studies have shown that neurotoxicity can occur at clinically used concentrations of lidocaine and bupivacaine with a continuous infusion down a spinal microcatheter.3567

Opioids

Opioids are widely used in the epidural and intrathecal space for pain relief in the perioperative, obstetric, and chronic pain settings. Fentanyl, alfentanil, diamorphine, and morphine are those most commonly used in the UK. Unfortunately there are no data from controlled trials investigating the long‐term side effects of neuraxial opioids. Histopathological studies after long‐term administration of intrathecal morphine in monkeys,1 and epidural administration of morphine and bupivacaine in humans,104109 showed no evidence of arachnoiditis. A review of the literature by a panel of experts in chronic pain relief noted that the intrathecal administration of morphine and fentanyl at clinically effective concentrations appeared to be safe.9

Epidural solutions commonly used in UK

In 1992, Holdcroft53 found that 69% of respondents at the annual meeting of the Obstetric Anaesthetists Association used epinephrine in epidurals, either as a test dose or to establish block. Twenty per cent of those using epinephrine whilst establishing analgesia, and 76% of those using epinephrine as part of their test dose, chose a premixed solution of bupivacaine with epinephrine (which contains the preservatives sodium metabisulfite and hydrochloric acid). In total, 17 (19.5%) respondents used a premixed solution of local anaesthetic containing preservatives.

Burnstein’s 1996–7 survey of epidural analgesia in the UK revealed that the most commonly used epidural test dose was 3 ml bupivacaine 0.25%, with only 9.5% regularly using lidocaine 2% as a test dose.21 Only five (3%) regularly added epinephrine to the test dose in this survey, in contrast to Holdcroft’s findings. No comment was made as to whether premixed solutions containing epinephrine were used. Ten millilitre bupivacaine 0.25% was most often used to initiate analgesia. A trend towards the increasing use of low concentrations of bupivacaine was shown with 89.1% using bupivacaine 0.125% plain or less for maintenance of analgesia. Opioids were added to the epidural solutions in 88.1% of units, with fentanyl being the most common, but alfentanil and diamorphine were also used. A CSE for labour analgesia was regularly used in 24% of units, the Queen Charlottes’ regime being the most common (1 ml bupivacaine 0.25% plain with 15–25 µg fentanyl intrathecally, and bupivacaine 0.1% with fentanyl 2 µg ml–1 as an infusion or bolus doses).

Evidence for back pain after epidural analgesic

There have been many studies performed to discover the association between low back pain and epidurals for labour.101856899094 As back pain is a common symptom of CAA, these studies might be expected to expose any cases of CAA occurring after obstetric epidurals. The data obtained is inconsistent, with the retrospective studies showing an association between epidurals and postpartum backache, and the prospective studies refuting this. This illustrates more about the potential for bias in retrospective studies than the link between epidurals and postpartum backache. Few researchers have examined their patients, and no study involved MRI scanning, making it difficult to determine the prevalence of CAA. However, Russell examined those who complained of new onset backache in his retrospective study,90 and found that most back pain was mild and not suggestive of serious pathology, which would seem to exclude CAA. A recent randomized study of long‐term outcome after epidural analgesia during labour found no evidence of a causal link between epidural analgesia during labour and low back pain.56 Patients in this study were examined (although not MRI scanned), and none diagnosed with CAA. Back pain is common in the population as a whole and during pregnancy. Swedish studies have found between 49 and 67% of pregnant women suffer from back pain; most improved within 6 months of delivery, but 7% had serious backache 18 months after delivery.1078 Aetiological factors for back pain during pregnancy include mechanical factors, sacroiliac dysfunction, hormonal influence on joint laxity, and local factors, with a small fraction suffering a herniated disc.6 The current evidence suggests no causative link between epidural analgesia and back pain.56 As back pain is a cardinal feature of CAA, it would appear unlikely that epidural analgesia in labour is a major cause of CAA.

Evidence for neurological deficit after epidural analgesic

The true incidence of neurological sequelae attributable to epidural anaesthesia is difficult to quantify.2084 Studies of neurological complications after epidural anaesthesia are not designed to specifically discover CAA, particularly as MRI scanning is often not undertaken. However, a large survey would be expected to uncover cases of CAA if it were prevalent after epidural anaesthesia.

A review of adverse drug reactions in Sweden over 30 yr included 21 reports of neurological abnormalities secondary to epidural analgesia.80 Since 1965, it has been compulsory to report all suspected new or adverse drug reactions to the Swedish Adverse Drug Reactions Advisory Committee. All reports from this large database concerning peripheral nerve injury associated with epidural or subarachnoid administration were reviewed. No cases of CAA were reported from this large database. There was only one case of an obstetric epidural leading to neurological deficit. This patient had a plain bupivacaine epidural for labour analgesia. She developed low back pain radiating to both legs with perineal dysaesthesia and incontinence, 27 months after delivery. Complicating factors included blood in the epidural catheter on first insertion and a forceps delivery. The incidence of peripheral neurological deficit is increased by use of the lithotomy position,110 and instrumental delivery.4577 Indeed, Murray found that 85% of postpartum obstetric paralyses were associated with instrumental deliveries.73

The Patient Injury Act in Finland provides a ‘no fault’ scheme for all patients. Claims are therefore made against the Patient Insurance Association (PIA) rather than the party implicated. A review spanning 5 yr uncovered 38 reports of neurological damage after central nerve block among the 23 500 claims for compensation filed with the PIA.7 It was estimated that 55 000 spinals and 170 000 epidurals were performed during this time. This large study revealed no documented cases of CAA. Two obstetric cases were reported in this review. After epidural analgesia for labour, one patient suffered an L5 lesion whilst the other had an unspecified permanent neurological deficit. There are no details as to the clinical findings or investigations, making it difficult to determine the cause or type of neurological deficit.

Scott and Hibbard published the results of an extensive retrospective questionnaire sent to all obstetric units in the UK, requesting data of any serious adverse events during and after extradural block in the previous 5 yr.95 Information was received from 203 units covering 516 000 deliveries, thus representing 78% of all births reported to the Royal College of Obstetricians and Gynaecologists during that time. Scott and Hibbard estimated that 506 000 epidurals for labour were performed during the 5 yr. There were 38 cases of neuropathy, all of which were a result of the damage of a single nerve or nerve root. They were of limited duration (up to 3 months), except for one case of permanent peripheral nerve damage. No cases of CAA were reported.

A multidisciplinary prospective audit in North‐West Thames had 35 notifications of neurological deficits.54 During the 1‐yr study period, notification of any postpartum neurological deficit was requested from obstetricians, anaesthetists, neurologists, rheumatologists, urologists, orthopaedic surgeons, GPs, and health visitors. Of 48 066 deliveries, 13 007 patients had an epidural and 629 had a spinal. An independent neurologist reviewed the notes of the 35 women with neurological deficits. Seven women had no neurological problems (joint problems), eight had no clinically identifiable lesion, one had multiple sclerosis, and 19 had neurological problems associated with pregnancy and delivery (an incidence of one in 2530). Seven of these patients had deficits persisting at 1 yr. No anaesthetic technique could be identified as a contributory factor. However, none of the women were examined or given an MRI scan, and no diagnosis was given to the neurological deficits.

Conclusions

CAA is an extremely rare condition with a wide variety of presentations. Criteria for diagnosis are: back pain that increases with exertion, with or without leg pain, which may be bilateral; some neurological abnormality on examination, most often hyporeflexia; and characteristic MRI findings. Its quoted incidence varies depending upon the criteria used for diagnosis and the prevalence of the most common aetiology at that time. Often it presents many years after the suspected causative event. This highlights the need for meticulous records of epidural interventions, including documentation of pre‐existing neurological abnormality, complications during epidural insertion, details of drugs used, and any post‐epidural neurological deficit.

It is now generally accepted that contrast media can lead to CAA, ethyliodophentylate in particular.5761102105 A link has been suggested between the use of epidural steroids and CAA.75 It would seem that meningeal irritation can be caused by blood or its breakdown products in the CSF.485876106 The epidural catheter itself may lead to inflammation of the meninges.114 There is no evidence to suggest that this meningeal inflammation causes long‐term problems, including CAA. It is possible that bupivacaine with epinephrine may cause CAA, although the evidence is not conclusive.1624 There is fairly good evidence to link the use of preservatives with CAA.99106 From Burnstein’s survey, it would seem that local anaesthetics containing preservatives or epinephrine are not used regularly in anaesthetic practice in the UK.21

Prospective studies show that epidurals do not cause chronic backache.567081 Studies of backache or neurological complications after central nerve block do not show a link between epidural anaesthesia and CAA. However, all are lacking in detail, and do not cover a sufficient time period to be sure that they would detect CAA. Reviews of insurance claims,7 adverse drug reactions,80 and reports of adverse events after epidurals,95 rely on a connection being made between the symptoms of CAA and the epidural. We cannot be certain that this connection would have been made, as CAA presents in a wide variety of ways and often many years after the aetiological event.

There are a few cases in the literature of CAA after epidural anaesthesia (Table 2). However, only one case has been reported after an uncomplicated (non‐obstetric) epidural using bupivacaine without preservatives, detergents, or vasoconstrictors.41 This report contains insufficient detail to be used as scientific evidence of a link between epidural bupivacaine and CAA.

To discover if obstetric epidurals lead to CAA, an enormous prospective study would need to be undertaken, over many years. It would need to look at any neurological abnormality in detail, including full clinical examination and MRI scan. Such a study, unsurprisingly, has not yet been done. Current evidence does not support the claim that epidural analgesia in obstetrics using preservative‐free, low concentration bupivacaine with opioids or plain bupivacaine, if performed in the standard way with disposable equipment, causes CAA.

Fig 1 Drawings of the spinal cord, dorsal and ventral roots, dorsal root ganglion, and common nerve trunk. The different types of arachnoid villi are illustrated (refs 1112131415). Reproduced with permission from Lippincott Williams & Wilkins. See reference 101 for more details. Key: 1, spinal‐cord dura mater; 2, spinal‐root dura mater; 3, perineurium and epineurium of peripheral nerve; 4, dural collar; 5, spinal‐cord arachnoid mater; 6, spinal‐root arachnoid mater; 7, spinal‐root pia and arachnoid coming together, with obliteration of the subarachnoid space; note the arachnoid proliferation at this point (circled); 8, perineural epithelium, a continuation of pia arachnoid membrane on to the peripheral nerve; 9, spinal‐cord pia mater; 10, spinal‐root pia mater; 11, arachnoid proliferations (Type I) protruding into spinal‐root subdural space; 12, arachnoid villi (Type II) partially penetrating the spinal‐root dura; 13, arachnoid villi (Type III) completely penetrating the spinal‐root dura and then exposing itself to the epidural space; 14, arachnoid villi (Type IV) protruding out of the spinal‐root dura into the epidural space; 15, arachnoid villi (Type V) protruding into a vein in the epidural space after emerging out of the spinal‐root dura; 16, epidural vein; 17, intervertebral foramina; 18, dorsal root ganglion; 19, dorsal spinal root; 20, substantia gelatinosa; 21, ventral spinal root; 22, spinal‐cord subdural space; 23, spinal‐cord subpial space; 24, peripheral nerve subperineural space (a continuation of the root subpial space (25)); 25, spinal‐root subpial space; 26, spinal‐root subdural space; 27, inter‐root or lateral epidural space (between dorsal and ventral spinal roots), CSF, Cerebrospinal fluid in the spinal cord and spinal‐root subarachnoid spaces

Fig 1 Drawings of the spinal cord, dorsal and ventral roots, dorsal root ganglion, and common nerve trunk. The different types of arachnoid villi are illustrated (refs 1112131415). Reproduced with permission from Lippincott Williams & Wilkins. See reference 101 for more details. Key: 1, spinal‐cord dura mater; 2, spinal‐root dura mater; 3, perineurium and epineurium of peripheral nerve; 4, dural collar; 5, spinal‐cord arachnoid mater; 6, spinal‐root arachnoid mater; 7, spinal‐root pia and arachnoid coming together, with obliteration of the subarachnoid space; note the arachnoid proliferation at this point (circled); 8, perineural epithelium, a continuation of pia arachnoid membrane on to the peripheral nerve; 9, spinal‐cord pia mater; 10, spinal‐root pia mater; 11, arachnoid proliferations (Type I) protruding into spinal‐root subdural space; 12, arachnoid villi (Type II) partially penetrating the spinal‐root dura; 13, arachnoid villi (Type III) completely penetrating the spinal‐root dura and then exposing itself to the epidural space; 14, arachnoid villi (Type IV) protruding out of the spinal‐root dura into the epidural space; 15, arachnoid villi (Type V) protruding into a vein in the epidural space after emerging out of the spinal‐root dura; 16, epidural vein; 17, intervertebral foramina; 18, dorsal root ganglion; 19, dorsal spinal root; 20, substantia gelatinosa; 21, ventral spinal root; 22, spinal‐cord subdural space; 23, spinal‐cord subpial space; 24, peripheral nerve subperineural space (a continuation of the root subpial space (25)); 25, spinal‐root subpial space; 26, spinal‐root subdural space; 27, inter‐root or lateral epidural space (between dorsal and ventral spinal roots), CSF, Cerebrospinal fluid in the spinal cord and spinal‐root subarachnoid spaces

Table 1

Summary of clinical findings in CAA. SLR=straight leg raise

Author Date published Number of patients Symptoms Frequency Basis of diagnosis 
    (%)  
Lombardi68 1961 41 Pain and parasthesiae 63 Myelography 
   Sphincter abnormality 63  
   Hypoaesthesia 95  
   Motor abnormality 98  
De La Porte30 1973 38 Back pain 50 Myelography 
   Leg pain 74  
   Sphincter disturbance 29  
   Abnormal reflexes 66  
Jorgensen60 1975 72 Back pain 93 Myelography 
   Leg pain 43  
   Sphincter disturbance  
Benner8 1978 68 Back pain 84 Myelography 
   Leg pain 91  
   Motor deficits 72  
   Sensory deficits 82  
   Abnormal reflexes 88  
   Urinary incontinence 25  
   Bowel incontinence 16  
Burton22 1978 100 Back pain +/– leg pain 100 Myelography 
   Decreased SLR Frequent Direct surgical observation 
   Decreased trunk movement Often  
Quiles81 1978 38 Back pain 76 Myelography 
   Leg symptoms 63  
   Decreased SLR 42  
   Decreased trunk movement 24  
   Abnormal reflexes 79  
   Motor weakness 37  
   Sensory abnormality 21  
   Sphincter abnormality 26  
Shaw102 1978 80 Back pain +/– leg pain 95  
   Bilateral sciatica 97  
   Decreased SLR 49  
   Motor weakness 23  
   Progressive symptoms 25  
   Static symptoms 50  
Guyer46 1989 50 Back pain 96 Myelography 
   Leg pain 98 Direct surgical observation 
   Decreased trunk movement 87  
   Motor weakness 66  
   Sensory abnormality 74  
   Decreased SLR 53  
   Abnormal reflexes 70  
   Sphincter abnormality 23  
   Progressive symptoms 33  
   Static symptoms 59  
   Able to walk unaided 72  
Long69 1992 321 Back p ain 94 Myelography 
   Leg pain 81  
   Neurogenic claudication 92  
   Decreased trunk movement 91  
   Motor weakness 74  
   Sensory abnormality 81  
   Abnormal reflexes 96  
   Decreased SLR 61  
   Sphincter abnormality 14  
   Progressive symptoms 18  
   Able to walk unaided 84  
Author Date published Number of patients Symptoms Frequency Basis of diagnosis 
    (%)  
Lombardi68 1961 41 Pain and parasthesiae 63 Myelography 
   Sphincter abnormality 63  
   Hypoaesthesia 95  
   Motor abnormality 98  
De La Porte30 1973 38 Back pain 50 Myelography 
   Leg pain 74  
   Sphincter disturbance 29  
   Abnormal reflexes 66  
Jorgensen60 1975 72 Back pain 93 Myelography 
   Leg pain 43  
   Sphincter disturbance  
Benner8 1978 68 Back pain 84 Myelography 
   Leg pain 91  
   Motor deficits 72  
   Sensory deficits 82  
   Abnormal reflexes 88  
   Urinary incontinence 25  
   Bowel incontinence 16  
Burton22 1978 100 Back pain +/– leg pain 100 Myelography 
   Decreased SLR Frequent Direct surgical observation 
   Decreased trunk movement Often  
Quiles81 1978 38 Back pain 76 Myelography 
   Leg symptoms 63  
   Decreased SLR 42  
   Decreased trunk movement 24  
   Abnormal reflexes 79  
   Motor weakness 37  
   Sensory abnormality 21  
   Sphincter abnormality 26  
Shaw102 1978 80 Back pain +/– leg pain 95  
   Bilateral sciatica 97  
   Decreased SLR 49  
   Motor weakness 23  
   Progressive symptoms 25  
   Static symptoms 50  
Guyer46 1989 50 Back pain 96 Myelography 
   Leg pain 98 Direct surgical observation 
   Decreased trunk movement 87  
   Motor weakness 66  
   Sensory abnormality 74  
   Decreased SLR 53  
   Abnormal reflexes 70  
   Sphincter abnormality 23  
   Progressive symptoms 33  
   Static symptoms 59  
   Able to walk unaided 72  
Long69 1992 321 Back p ain 94 Myelography 
   Leg pain 81  
   Neurogenic claudication 92  
   Decreased trunk movement 91  
   Motor weakness 74  
   Sensory abnormality 81  
   Abnormal reflexes 96  
   Decreased SLR 61  
   Sphincter abnormality 14  
   Progressive symptoms 18  
   Able to walk unaided 84  
Table 2

CAA cases reported after epidural anaesthesia

Author Date published Number of cases Suspected aetiology Diagnostic criteria Confounding factors Obstetric? 
Aldrete5 1997 Blood MRI Subdural injection myelography Yes 
Haisa47 1995 Trauma MRI Nil Yes 
Paddison79 1954 Detergents PM Pneumoencephalogram No 
Boiardi16 1983 Epinephrine MRI Nil No 
Chiapparinni24 2000 Epinephrine Myelography Nil Yes 
Sghirlanzoni99 1989 Preservatives MRI Nil No 
Sklar106 1988 Preservatives MRI Laminectomy(1 patient) Yes 
Gemma41 1994 Bupivacaine MRI Nil No 
Author Date published Number of cases Suspected aetiology Diagnostic criteria Confounding factors Obstetric? 
Aldrete5 1997 Blood MRI Subdural injection myelography Yes 
Haisa47 1995 Trauma MRI Nil Yes 
Paddison79 1954 Detergents PM Pneumoencephalogram No 
Boiardi16 1983 Epinephrine MRI Nil No 
Chiapparinni24 2000 Epinephrine Myelography Nil Yes 
Sghirlanzoni99 1989 Preservatives MRI Nil No 
Sklar106 1988 Preservatives MRI Laminectomy(1 patient) Yes 
Gemma41 1994 Bupivacaine MRI Nil No 

References

1
Abouleish E, Barmada MA, Nemoto EM, Tung A, Winter P. Acute and chronic effects of intrathecal morphine in monkeys.
Br J Anaesth
 
1981
;
53
:
1027
–31
2
Abouleish E, De la Vega S, Bledinger I, Tio T‐O. Long‐term follow‐up of epidural blood patch.
Anesth Analg
 
1975
;
54
:
459
–63
3
Abram SE, O’Connor TC. Complications associated with epidural steroid injections.
Reg Anesth
 
1996
;
21
:
1198
–205
4
Abram S, Marasala M, Yaksh T. Analgesic and neurotoxic effects of intrathecal corticosteroids in rats.
Anesthesiology
 
1994
;
81
:
149
–62
5
Aldrete JA, Brown T. Intrathecal hematoma and arachnoiditis after prophylactic blood patch through a catheter.
Anesth Analg
 
1997
;
84
:
228
–36
6
Alexander J, McCormick P. Pregnancy and discogenic disease of the spine.
Neurosurg Clin N Am
 
1993
;
4
:
153
–9
7
Arommaa U, Landensuu M, Cozanitis D. Severe complications associated with epidural and spinal anaesthesias in Finland 1987–1993. A study based on patient insurance claims.
Acta Anaesthesiol Scand
 
1997
;
41
:
445
–52
8
Benner B, Ehni G. Spinal arachnoiditis. The postoperative variety in particular.
Spine
 
1978
;
3
:
40
–4
9
Bennett G, Serafini M, Burchiel K, et al. Evidence‐based review of the literature on intrathecal delivery of pain medication.
J Pain Sympt Manag
 
2000
;
20
:
S12
–36
10
Berg G, Hammar M, Moller‐Nielsen J, Linden U, Thorblad J. Low back pain during pregnancy.
Obstet Gynecol
 
1988
;
71
:
71
–5
11
Bernards C, Hill F. Morphine and alfentanil permeability through the spinal dura, arachnoid, and pia mater of dogs and monkeys.
Anesthesiology
 
1990
;
73
:
1214
–9
12
Bernat JL. Intraspinal steroid therapy.
Neurology
 
1981
;
31
:
168
–70
13
Bernat JL, Sadowsky CH, Vincent FM, Nordgren RE, Margolis G. Sclerosing spinal pachymeningitis: a complication of intrathecal administration of depo‐medrol for multiple sclerosis.
J Neurol Neurosurg Psychiatry
 
1976
;
39
:
1124
–8
14
Benzon HT. Epidural steroid injections for low back pain and lumbosacral radiculopathy.
Pain
 
1986
;
24
:
277
–95
15
Blomberg R, Olsson S. The lumbar epidural space in patients examined with epiduroscopy.
Anesth Analg
 
1989
;
68
:
157
–60
16
Boiardi A, Sghirlanzoni A, La Mantia L, Bussone G, Lombardi B, Girotti F. Diffuse arachnoiditis following epidural analgesia.
J Neurol
 
1983
;
230
:
253
–7
17
Bourne IHJ. Lumbo‐sacral adhesive arachnoiditis: a review.
J R Soc Med
 
1990
;
83
:
262
–5
18
Breen TJ, Ransil BJ, Groves PA, Oriol NE. Factors associated with back pain after childbirth.
Anesthesiology
 
1994
;
81
:
29
–34
19
Brodsky AE. Cauda equina arachnoiditis. A correlative clinical and roentgenologic study.
Spine
 
1978
;
3
:
51
–9
20
Bromage PR. Neurological complications of subarachnoid and epidural anaesthesia.
Acta Anaesthesiol Scand
 
1997
;
41
:
439
–44
21
Burnstein R, Buckland R, Pickett JA. A survey of epidural analgesia for labour in the United Kingdom.
Anaesthesia
 
1999
;
54
:
634
–40
22
Burton C. Lumbosacral arachnoiditis.
Spine
 
1978
;
3
:
24
–30
23
Carrie LES. Metallic fragments and the combined spinal‐extradural technique.
Br J Anaesth
 
1992
;
69
:
662
–3
24
Chiapparini L, Sghirlanzoni L, Paryson D, Savoiardo M. Imaging and outcome in severe complications of lumbar epidural anaesthesia: a report of 16 cases.
Neuroradiology
 
2000
;
42
:
564
–71
25
Cicala RS, Turner R, Moran E, Henley R, Wong R, Evans J. Methylprednisolone acetate does not cause inflammatory changes in the epidural space.
Anesthesiology
 
1990
;
72
:
556
–8
26
Cook TM. Combined spinal‐epidural techniques.
Anaesthesia
 
2000
;
55
:
42
–64
27
Cope RW. The Woolley and Roe case.
Anaesthesia
 
1954
;
9
:
249
–70
28
Dahlagren N, Tornebrant K. Neurological complications after anaesthesia. A follow‐up of 18 000 spinal and epidural anaesthetics performed over 3 years.
Acta Anesthesiol Scand
 
1995
;
39
:
872
–80
29
Delamarter RB, Ross JS, Masaryk TJ, et al. Diagnosis of lumbar arachnoiditis by magnetic resonance imaging.
Spine
 
1990
;
15
:
304
–10
30
De La Porte C, Seigfried J. Lumbosacral spinal fibrosis (spinal arachnoiditis). Its diagnosis and treatment by spinal cord stimulation.
Spine
 
1983
;
8
:
593
–603
31
Delany TJ, Rowlingson JC, Carron H, Butler A. Epidural steroid effects on nerves and meninges.
Anesth Analg
 
1980
;
59
:
610
–4
32
Denson JS. Effects of detergents intrathecally.
Anesthesiology
 
1957
;
18
:
143
–4
33
Digiovanni AJ, Galbert MW, Wahle WM. Epidural injection of autologous blood for postlumbar‐puncture headache II. Additional clinical experiences and laboratory investigation.
Anesth Analg
 
1972
;
51
:
226
–32
34
Drasner K, Rigler M, Sessler D, Stoller M. Cauda equina syndrome following intended epidural anesthesia.
Anesthesiology
 
1992
;
77
:
582
–5
35
Drasner K, Sakura S, Chan VWS, Bollen AW, Ciriales R. Persistent sacral sensory deficit induced by intrathecal local anesthetic infusion in the rat.
Anesthesiology
 
1994
;
80
:
847
–52
36
Durant P, Yaksh T. Epidural injections of bupivacaine, morphine, fentanyl, lofentanil, and DADL in chronically implanted rats: a pharmacologic and pathologic study.
Anesthesiology
 
1986
;
64
:
43
–53
37
Eldor J, Brodsky V. Danger of metallic particles in the spinal‐epidural spaces using the needle‐through‐needle approach.
Acta Anaesthesiol Scand
 
1991
;
35
:
461
–3
38
Eldor J, Guedj P. Aseptic meningitis due to metallic particles in the needle‐through‐needle technique.
Reg Anesth
 
1995
;
20
:
360
39
Elkington JS. Arachnoiditis.
Modern Trends Neurol
 
1951
;
1
:
149
–61
40
Fisher A. The nickel controversy at home and abroad.
Cutis
 
1993
;
52
:
134
–6
41
Gemma M, Bricchi M, Grisoli M, Visintini S, Pareyson D, Sghirlanzoli A. Neurologic symptoms after epidural anaesthesia. Report of three cases.
Acta Anaesthesiol Scand
 
1994
;
38
:
742
–3
42
Gissen AJ, Datta S, Lambert D. The Chloroprocaine controversy II. Is chloroprocaine neurotoxic?
Reg Anesth
 
1984
;
9
:
135
–45
43
Goldman W, Sanford J. An ‘epidemic’ of chemical meningitis.
Am J Med
 
1960
;
29
:
94
–101
44
Greene N. Neurological sequelae of spinal anesthesia.
Anesthesiology
 
1961
;
22
:
682
–98
45
Grove LH. Backache, headache, and bladder dysfunction after delivery.
Br J Anaesth
 
1973
;
45
:
1147
46
Guyer DW, Wiltse LL, Eskay ML, Guyer BH. The long‐range prognosis of arachnoiditis.
Spine
 
1989
;
14
:
1332
–41
47
Haisa T, Todo T, Kondo T. Lumbar adhesive arachnoiditis following attempted epidural anesthesia.
Neurol Med Chir (Tokyo)
 
1995
;
35
:
107
–9
48
Hammes E.M. Reaction of the meninges to blood.
Arch Neurol Psych
 
1944
;
52
:
505
–14
49
Harding S, Collis RE, Morgan BM. Meningitis after combined spinal‐extradural anaesthesia in obstetrics.
Br J Anaesth
 
1994
;
73
:
545
–7
50
Hargreaves J. Metal particle generation caused by the combined spinal‐extradural technique.
Br J Anaesth
 
1993
;
70
:
706
51
Herman N, Molin J, Knape KG. No additional metal particle formation using the needle‐through‐needle combined epidural/spinal technique.
Acta Anaesthesiol Scand
 
1996
;
40
:
227
–31
52
Hoffman GS. Spinal arachnoiditis. What is the clinical spectrum?
Spine
 
1983
;
8
:
538
–40
53
Holdcroft A. The use of adrenaline in obstetric analgesia.
Anaesthesia
 
1992
;
47
:
987
–90
54
Holdcroft A, Gibberd FB, Hargrove RL, Hawkins DF, Dellaportas CI. Neurological complications associated with pregnancy.
Br J Anaesth
 
1995
;
75
:
522
–6
55
Holloway J, Seed P, Reynolds F, O’Sullivan G. Paraesthesiae and nerve root damage following combined spinal‐epidural and spinal anaesthesia.
Int J Obstet Anesth
 
1999
;
9
:
201
–2
56
Howell CJ, Dean T, Lucking L, et al. Randomised study of long term outcome after epidural versus non‐epidural analgesia during labour.
Br Med J
 
2002
;
325
:
357
–9
57
Howland WJ, Curry JL. Pantopaque arachnoiditis: experimental study of blood as a potentiating agent and corticosteroids as an ameliorating agent.
Acta Radiol (Diagn)
 
1966
;
5
:
1032
–41
58
Jackson IJ. Aseptic hemogenic meningitis.
Arch Neurol Psych
 
1949
;
62
:
572
–89
59
Johnson C, Sze G. Benign lumbar arachnoiditis: MR imaging with gadopentetate dimeglumine.
Am J Neuro Radiol
 
1990
;
155
:
873
–9
60
Jorgensen J, Hansen V, Steenskov V, Ovesen N. A clinical and radiological study of chronic lower spinal arachnoiditis.
Neuroradiology
 
1975
;
9
:
139
–44
61
Junck L, Marshall WH. Neurotoxicity of radiological contrast agents.
Ann Neurol
 
1983
;
13
:
469
–84
62
Kane RE. Neurologic deficits following epidural or spinal anesthesia.
Anesth Analg
 
1981
;
60
:
150
–61
63
Kennedy F, Effron AS, Perry G. The grave spinal cord paralyses caused by spinal anesthesia.
Surg Gyn Obst
 
1950
;
91
:
385
–97
64
Kennedy F, Somberg HM, Goldberg BR. Arachnoiditis and paralysis following spinal anesthesia.
JAMA
 
1945
;
129
:
664
–7
65
Kytta J, Rosenburg PH, Wahlstrom T. Long‐term epidural bupivacaine in pigs.
Acta Anaesthesiol Scand
 
1985
;
29
(Suppl 80):
114
66
Lambert D, Hurley R. Cauda equina syndrome and continuous spinal anesthesia.
Anesth Analg
 
1991
;
72
:
817
–9
67
Lambert LA, Lambert DH, Strichartz GR. Irreversible conduction block in isolated nerve by high concentrations of local anesthetics.
Anesthesiology
 
1994
;
80
:
1082
–93
68
Lombardi G, Passerini A, Migliavacca F. Spinal arachnoiditis.
Br J Radiol
 
35
:
314
–20
69
Long DM. Chronic adhesive spinal arachnoiditis: pathogenesis, prognosis and treatment.
Neurosurgery Q
 
1992
;
2
:
296
–319
70
Macarthur A, Macarthur C, Weeks S. Epidural anaesthesia and low back pain after delivery: a prospective cohort study.
BMJ
 
1995
;
11
:
1336
–9
71
Marx GF, Saifer A, Orkin LR. Cerebrospinal fluid cells and proteins following spinal anesthesia.
Anesthesiology
 
1963
;
24
:
305
–12
72
McNeill MJ, Thorburn F. Cannulation of the epidural space. A comparison of 18‐ and 16‐gauge needles.
Anaesthesia
 
1988
;
43
:
154
–5
73
Murray RR. Maternal obstetrical paralysis.
Am J Obstet Gynecol
 
1964
;
88
:
399
–403
74
Myers R, Kalichman MW, Reisner LS, Powell HC. Neurotoxicity of local anesthetics: altered perineural permeability, edema, and nerve fiber injury.
Anesthesiology
 
1986
;
64
:
29
–35
75
Nelson DA. Dangers from methylprednisolone acetate therapy by intraspinal injection.
Arch Neurol
 
1988
;
45
:
804
–6
76
Nelson J. Intramedullary cavitation resulting from adhesive spinal arachnoiditis.
Arch Neurol Psychiatry
 
1943
;
50
:
1
–7
77
Newman B. Postnatal paraparesis following epidural analgesia and forceps delivery.
Anaesthesia
 
1983
;
38
:
350
–1
78
Ostgaard HC, Andersson GBJ. Postpartum low‐back pain.
Spine
 
1992
;
17
:
53
–4
79
Paddison R, Alpers B. Role of intrathecal detergents in pathogenesis of adhesive arachnoiditis.
AMA Arch Neurol Psych
 
1954
;
71
:
87
–100
80
Pleym H, Spigset O. Peripheral neurologic deficits in relation to subarachnoid or epidural administration of local anaesthetics for surgery.
Acta Anaesthesiol Scand
 
1997
;
41
:
453
–60
81
Quiles M, Marchisello P, Tsairis P. Lumbar adhesive arachnoiditis—etiologic and pathologic aspects.
Spine
 
1978
;
3
:
45
–50
82
Renk H. Neurological complications of central nerve blocks.
Acta Anaesthesiol Scand
 
1995
;
39
:
859
–68
83
Reynolds F. Damage to the conus medullaris following spinal anaesthesia.
Anaesthesia
 
2001
;
56
:
238
–47
84
Reynolds F. Maternal sequelae of childbirth.
Br J Anaesth
 
1995
;
75
:
515
–6
85
Rigler M, Drasner K. Distribution of catheter‐injected local anesthetic in a model of the subarachnoid space.
Anesthesiology
 
1991
;
75
:
684
–92
86
Rigler M, Drasner K, Krejcie T, et al. Cauda equina syndrome after continuous spinal anesthesia.
Anesth Analg
 
1991
;
72
:
275
–81
87
Roblin SH, Hew E, Olgilvie G. A comparison of two types of epidural catheters.
Can J Anaesth
 
1987
;
34
:
459
–61
88
Ross JS, Masaryk TJ, Modic MT, et al. MR imaging of lumbar arachnoididtis.
Am J Roentgenol
 
1987
;
149
:
1025
–32
89
Russell R, Dundas R, Reynolds F. Long term backache after childbirth: prospective search for causative factors.
BMJ
 
1996
;
312
:
1384
–8
90
Russell R, Groves P, Taub N, O’Dowd J, Reynolds F. Assessing long‐term backache after childbirth.
BMJ
 
1993
;
306
:
1299
–303
91
Ryan MD, Thomas KFT. Management of lumbar nerve root pain by intrathecal and epidural injection of depot methylprednisolone acetate.
Med J Austr
 
1981
;
2
:
532
–4
92
Sakura S, Chan V, Ciricales R, Drasner K. The addition of 7.5% glucose does not alter the neurotoxicity of 5% lidocaine administered intrathecally in the rat.
Anesthesiology
 
1995
;
82
:
236
–40
93
Schell RM, Brauer FS, Cole DJ, Applegate RL. Persistent sacral nerve root deficits after continuous spinal anaesthesia.
Can J Anaesth
 
1991
;
38
:
908
–11
94
Schneider MC. Pleading not guilty for long‐term maternal morbidity following dural puncture.
Int J Obs Anaesth
 
2001
;
10
:
1
–3
95
Scott D, Hibbard B. Serious non‐fatal complications associated with extradural block in obstetric practise.
Br J Anaesth
 
1990
;
64
:
537
–41
96
Seinge TD. Aseptic meningitis following spinal anaesthesia.
Anaesthesia
 
1970
;
25
:
402
–7
97
Sekel R. Epidural Depo‐Medrol revisited.
Med J Austr
 
1981
;
2
:
532
–4
98
Selander D, Brattsand R, Lundborg G, Nordborg C, Olsson Y. Local anaesthetics: Importance of mode of application, concentration and adrenaline for the appearance of nerve lesions.
Acta Anaesthesiol Scand
 
1979
;
23
:
127
–36
99
Sghirlanzoni A, Marazzi R, Pareyson D, Olivieri A, Bracchi M. Epidural anaesthesia and spinal arachnoiditis.
Anaesthesia
 
1989
;
44
:
317
–21
100
Shantha TR. Spinal nerve root is one of the preferred routes for transfer of drugs to the nerve roots and spinal cord from the epidural space.
Anesthesiology
 
1992
;
77
:
216
–8
101
Shantha TR, Evans JA. The relationship of epidural anesthesia to neural membranes and arachnoid villi.
Anesthesiology
 
1972
;
37
:
543
–57
102
Shaw M, Russell JA, Grossart KW. The changing pattern of spinal arachnoiditis.
J Neurol Neurosurg Psych
 
1978
;
41
:
97
–107
103
Simmonds WJ. Observation of labelled erythrocytes from the subarachnoid space in rabbits.
Aus J Exp Biol Med Sci
 
1953
;
312
:
77
–83
104
Sjoberg M, Karlsson P, Nordborg C, et al. Neuropathologic findings after long‐term intrathecal infusion of morphine and bupivacaine for pain treatment in cancer patients.
Anesthesiology
 
1992
;
76
:
173
–86
105
Skalpe IO. Adhesive arachnoiditis following lumbar radiography with water‐soluble contrast agents.
Radiology
 
1976
;
21
:
647
–51
106
Sklar EML, Quencer RM, Green BA, Montalvo BM, Post MJD. Complications of epidural anesthesia: MR appearance of abnormalities.
Radiology
 
1991
;
181
:
549
–54
107
Tjandra JJ, Varma TRK, Weeks RDW. Spinal arachnoiditis following subarachnoid haemorrhage.
Aust New Z J Surg
 
1989
;
59
:
84
–7
108
Usubiaga JE, Wikinski J, Wikinski R, Usubiaga LE, Pontremoli M. Transfer of local anesthetics to the subarachnoid space and mechanisms of epidural block.
Anesthesiology
 
1964
;
25
:
752
–9
109
Wagemans MF, Van Der Valk P, Spoelder EM, Zuurmond WWA, De Lange JJ. Neurohistopathological findings after continuous intrathecal administration of morphine or a morphine/bupivacaine mixture in cancer pain patients.
Acta Anaesth Scand
 
1997
;
41
:
1033
–8
110
Warner M, Martin J, Schroeder D, Offord K, Chute C. Lower extremity motor neuropathy associated with surgery performed on patients in a lithotomy position.
Anesthesiology
 
1994
;
81
:
6
–12
111
Weston‐Hurst E. Adhesive arachnoiditis and vascular blockage caused by detergents and other chemical irritants: an experimental study.
J Path Bact
 
1955
;
70
:
167
–78
112
Wilkinson HA. Intrathecal Depo‐Medrol: a literature review.
Clin J Pain
 
1992
;
8
:
49
–56
113
Winkelman NW. Neurologic symptoms following accidental intraspinal detergent injection.
Neurology
 
1952
;
2
:
284
–91
114
Wulf H, Striepling E. Post‐mortem findings after epidural anaesthesia.
Anaesthesia
 
1990
;
45
:
357
–61

Author notes

1Shackelton Department of Anaesthesia, Southampton General Hospital, Tremona Road, Shirley, Southampton SO14 6YD, UK. 2Department of Anaesthesia, Poole Hospital NHS Trust, Poole, UK. 3Department of Anaesthesia, North Hampshire Hospital, UK

Comments

0 Comments