RESEARCH
ARTICLE

This thesis was added
14 July 2002

The Presenting Symptoms Associated With Arachnoiditis And The Experience Of Living With Them In Everyday Life

A thesis presented in partial fulfilment of the requirements for the degree of Master of Arts in Nursing at Massey University

by Christine Hopkins - 1998

Chapter Two: What is arachnoiditis?

Introduction

This chapter reports on the results of examination of the literature and gives consideration to historical background, pathological and pathophysiological changes associated with arachnoiditis, etiological factors, epidemiology, methods of diagnosis, differential and associated diagnoses, prognosis and medical treatments. Only brief reference is made to clinical presentation in this chapter, as symptomatology is dealt with in more detail in Chapter Five.

Anatomy

Arachnoiditis is defined simply as ‘inflammation of the arachnoidea’ (Dorland, 1988). The arachnoidea, more commonly referred to as the arachnoid membrane, is a delicate, avascular membrane which envelopes the central nervous system. Composed of fibrous and elastic tissue, it is superimposed between two vascular membranes, the dura mater and the pia mater. Scanning electron microscopy reveals that the normal arachnoid membrane has a fenestrated surface embossed with parallel fibres (Guyer, Wiltse, Eskay & Guyer, 1989). The name ‘arachnoid’ refers to the similarity of the arachnoid membrane to the fine, closely woven webs spun by some spiders. The pia mater, the arachnoid membrane and the intermediate subarachnoid space are together considered as a functional unit known as the leptomeninges.

Historical perspectives

Arachnoiditis has been a recognised disease entity since the turn of the century. At that time, it was largely associated with infections such as tuberculosis, viral and bacterial meningitis, and syphilis (Long, 1992; Jackson & Isherwood, 1994). Arachnoiditis associated with infections tended to occur in the thoracic and cervical regions and had a relentlessly progressive course, leading to paraplegia and death (Bourne, 1990, Guyer et al., 1989). The introduction of antibiotics reduced the risk of arachnoiditis from infection (Long, 1992), although there has been a recent increase in cases associated with spinal tuberculosis in Western countries with high immigration rates from countries such as India (Chadwick, Cartlidge & Bates, 1989). Tuberculous spinal arachnoiditis is still an important clinical problem in India (Phadke, Kohli, Jain, Gupta, Kumar & Gujral, 1994). It is now recognised that the vast majority of cases seen today are iatrogenic cases associated with diagnostic and/or surgical intervention (Burton, 1978,1985, 1991; Guyer et al., 1989; Grahame, Clark, Watson & Polkey, 1991; Johnson, Ryan & Roche, 1991; Long, 1992; Dolan, 1993; Laitt, Jackson & Isherwood, 1996).

Since medical recognition of the condition, there has been confusion as to what should be called arachnoiditis, its relationship to myelography and spinal surgery, the frequency with which the disease occurs and with its diagnosis (Long, 1992). Termed adhesive spinal arachnoiditis in 1927 (Long, 1992), it has also been known over the years as chronic spinal meningitis, granulomatous meningitis, lumbar radiculopathy, meningeal inflammation, aseptic meningitis, chemical arachnoiditis (Arachnoiditis, 1996), meningitis serosa circumscripta spinalis (Guyer et al., 1989) and sclerosing spinal pachymeningitis (Walton, 1985). The older term ‘pachymeningitis’, associated with infectious origins such as syphilis and tuberculosis, referred to involvement of both the dura and the arachnoid. Most commonly today, it is termed chronic adhesive arachnoiditis (Long, 1992) or chronic adhesive spinal arachnoiditis or CASA (Grahame et al., 1991). Seigfried (1990) uses the term lumbo-sacral spinal fibrosis. The majority of the modern literature centres around adhesive arachnoiditis in the lumbar region and makes little mention of occurrence in higher regions of the spine.

Pathology

Arachnoiditis involves a non-specific, inflammatory process which gives rise to an exudate involving the leptomeninges and nerve roots. The avascularity of the arachnoid is thought to prevent the eradication of the fibrous bands that proliferate during the second stage of the inflammatory process. Fibroblast proliferation then gives rise to fibrosis and hyalinization of the arachnoid membrane, which becomes thickened and adherent to both pia and dura mater. (Laitt, Jackson & Isherwood, 1996). Changes range from clumping of nerve roots and blunting of nerve root pouches to multi-segmental transverse obliteration of the subarachnoid space. Atrophic, devascularised nerve roots become enmeshed in scar tissue (ibid).

I n advanced cases, a surgeon will encounter a nonpulsatile mass of matted nerve roots and a partial or complete block of spinal fluid flow. (Dolan, 1993) In the majority of patients studied by Burton (1978), loculated cysts containing spinal fluid and/or Pantopaque were present. When large, these cysts can compress local nervous tissue. Small areas of calcification commonly occur within the collagenous tissue (Long, 1993 ).

Pathophysiology

It is only relatively recently that techniques for studying pathophysiological mechanisms have been available. Walton (1985) notes that spinal cord compression, such as that linked to excessive scar formation in the spinal canal, affects the cord in several ways, indicating that a number of neuropathophysiological mechanisms may be involved. Overt pressure may interfere with conduction in spinal roots and in the cord itself. Pressure on the ascending longitudinal veins may lead to oedema in the cord below the site of compression. Compression of longitudinal and radicular spinal arteries may lead to ischaemia of the segments of the cord they supply. These vascular disturbances possibly create a local oedema and/or degeneration of ganglion cells and white matter. Areas of softening may develop (myelomalacia). Obstruction of the subarachnoid space can interrupt the flow of cerebro-spinal fluid (CSF). CSF can become trapped in small cavities (loculation) with changes in its composition.

The role of CSF in cord nutrition and the correlation of the resultant malnutrition with the presence of symptoms was not discussed in any of the literature examined. Pathophysiology associated with interference in CSF flow in arachnoiditis was noted as significant in the development of syringomyelia subsequent to arachnoiditis (Klekamp, Batzdorf, Samii & Bothe, 1997), in which impaired CSF flow is believed to initiate a series of events that ultimately lead to an increase in the extracellular water content of the spinal cord. This fluid is visible within a cavity in the spinal cord itself. Usually thoracic, it is associated with progressive symptoms and can be relieved by surgical procedures. As such, this late pathophysiological development in some people with arachnoiditis requires recognition.

Walton (1985) saw the association between cord compression from the scarring and the experiencing of symptoms as probably less important in itself than the ischaemia related to interference with the blood supply of the cord. Guyer et al. (1989) cited studies drawing an analogy with cancer in which the condition is painless unless a nerve is entrapped against a rigid structure or interferes with vital organ function. Most recently, Burton (1997) attributed the pain of arachnoiditis to a nutritional, vascular nerve impairment of variable severity induced by the encapsulating collagenous scar tissue creating neural tissue anoxia that similar to that associated with causalgia.

Research into central pain syndromes investigated the importance of the thalamus, the significance of lesion location and changes in membrane receptor function (Casey, 1991), and deficiencies in Aß-mediated inhibition of denervation sensitivity, and the role of substance P (Price, 1991) in pain that appears to have no direct correlation to, or immediate relationship with, an injury. No literature was found that specifically considered central pain syndromes in arachnoiditis.

Aetiology

Aetiology is a medical term relating to the study or theory of factors that cause disease and the method of their introduction to the host; the cause(s) or origin of a disease or disorder (Dorland, 1988). In a minority of cases, the development of arachnoiditis can be idiopathic, with an increased likelihood of misdiagnosis (Vloeberghs, Herregodts, Stadnik, Goossens & D’Haens, 1992). Roeder, Bazan and Jinkins (1995) report the development of arachnoiditis after the rupture of a spinal dermoid cyst. The most clinically apparent pathological forerunners of arachnoiditis have been a progressive, inflammatory reaction of the pia-arachnoid to infection, such as pyogenic syphilis and tuberculous leptomeningitis. Arachnoiditis may also be an associated complication of spinal stenosis (Jackson & Isherwood, 1994) or may arise from an exuberant reaction to meningeal injury as part of the normal reparative process (Smith & Blaser, 1991).

A number of situations related to the iatrogenic factors in arachnoiditis development are recognised. Burton (1978) related the development of arachnoiditis to a combination of etiologic factors initiating and maintaining a progressive inflammatory reaction, usually in relation to a herniated disc, amplified by the presence of blood and an irritative foreign body substance (such as Myodil/ Pantopaque) in the subarachnoid space. While Guyer et al. (1989) felt repeated trauma at the site of a protruded disc could lead to localised arachnoiditis in the absence surgery or myelography, Jackson and Isherwood (1994) found no evidence to substantiate an aetiological linkage between disc prolapse per se and arachnoiditis. Arachnoiditis has been shown to occur after extradural lumbar disc surgery independent of myelographic contrast media use (Fitt & Stevens, 1995).

The precise role of myelographic contrast agents continues to be debated. Known commercially as Myodil in Commonwealth countries and Pantopaque in the USA, iophendylate (see Appendix One ) has been suspected of toxicity since the 1940’s. Since that time there has been an increasing recognition of the relationship between iophendylate and lumbo-sacral arachnoiditis (Burton, 1978). Burton (1991) continues to maintain that iophendylate is the single, most significant, toxic agent in the development of lumbo-sacral arachnoiditis. Others maintained that there is no evidence that a single case of chronic arachnoiditis can be attributed solely to Pantopaque (Peterson, 1976). In Britain, evidence is also felt to be lacking that Myodil alone leads to arachnoiditis (Jackson & Isherwood, 1994). Research there indicated that adhesive arachnoiditis tended to centre on the level of surgery, suggesting synergy between Myodil and surgery (Laitt, Jackson & Isherwood, 1996) and concluded that adhesive arachnoiditis is a rare complication of Myodil myelography alone but that exposure to Myodil dramatically increases the incidence of adhesive arachnoiditis when other risk factors are present (ibid).

Disagreement also continues over the role of water soluble contrasts, some authors declaring that, without any doubt, water soluble contrast media for myelography may produce arachnoiditis (Jorgensen, Hansen, Steenskov & Oveson, 1975) and others believing the side effects are only significant in the immediate period post-myelogram. Long (1992) found no evidence to suggest that the incidence of arachnoiditis has reduced since the introduction of water soluble agents, particularly the early ones, which were all as reactive, or more so, than Pantopaque. He sounds a note of caution for those making any predictions that newer agents, such as metrizamide, can definitively be declared safer. Even so, Long (ibid) believes the paucity of reported cases in the literature in the face of millions of myelograms done throughout the world suggests that the actual incidence of symptomatic arachnoiditis related to myelograms alone is less than 1%.

The reasons why arachnoiditis develops in some individuals, and not in others, remain unclear. The role of autoimmune factors has yet to be established but may be important (Burton, 1994; Guyer et al., 1989). There is the possibility that this process only occurs in those who have a fibrinolytic defect, affecting fibrotic processes which are self-limiting in most cases, but which potentiate in others. Research in England is reported as focusing on the nature of the abnormal scar tissue and factors leading to its development, as well as on abnormal amounts of substance P production in those with arachnoiditis (Jayson, Information sheet). The Fact Sheet (Arachnoiditis - questions, 1994) produced by the Back Pain Association of America and the Arachnoiditis Trust of Great Britain leaflet (Arachnoiditis Trust Leaflet, undated) declare that arachnoiditis is an autoimmune disease but offer no supporting references for this assertion.

Recently, concerns about the role of epidural anaesthesia and analgesia in relation to the subsequent development of arachnoiditis (Arachnoiditis: What is, 1996) have been voiced. Reasons for the development of arachnoiditis after epidural administrations are unclear, with possibilities including unintentional injection of drugs into the CSF, use of large volumes of anaesthetic causing temporary compression ischaemia, the action of vasoconstrictors such as epinephrine, and the effect of contaminants such as parabens (Rosenberg & Stacey, 1996). Epidural Depo-Medrol, while it can be effective for the treatment of pain continuing despite back surgery (Shipton, 1989), is implicated in the development of arachnoiditis. There is the suggestion that the polyethylene glycol Depo Medrol contains is the likely irritant responsible (Nelson, 1993; Johnson et al., 1991). The makers state in their product information data sheet that Depo Medrol is not recommended for intrathecal use.

Epidemiology

The actual incidence and prevalence of arachnoiditis is not known. Burton (1994) reported that very little incidence and prevalence data has been accumulated. If studied per se, arachnoiditis would probably be common, since it occurs as a normal response to traumas and infections in the subarachnoid space (ibid), making arachnoid adhesions a very common entity that typically are of no clinical significance (Burton, 1997). Opinion on the incidence of symptomatic arachnoiditis varies. Danish authors in the seventies described chronic arachnoiditis in the spinal canal as a rather well known condition (Jorgensen et al., 1975). Burton (1991) considered arachnoiditis to constitute 11% of failed back surgery syndrome patients since 1981. Using extrapolations from United States statistics of spinal surgeries performed per annum, this indicates 1,659,090 cases of lumbo-sacral arachnoiditis worldwide (Burton, 1997). However, variably long latency periods and confusion with other conditions, particularly concurrently present spinal stenosis, confuse the issue and lead to delays in diagnosis (Long, 1992). Improved surgical and radiological techniques should have decreased the potential incidence (Burton, 1997) but this view requires balancing against the speed with which countries adopt new, safer technology and surgical techniques, the increasingly frequent use of epidural anaesthesia and the ongoing use of intrathecal and epidural steroids and other medications. In areas without access to advanced imaging techniques, the most important measure in reducing the incidence of arachnoiditis development is the avoidance of routine myelography prior to surgery (Dolan, 1993). Dolan (ibid) maintains that recognition of the clinical picture associated with classical lumbar disc protrusion and the findings of physical examinations should be sufficient to make a diagnosis indicating the need for any surgery.

Diagnosis

The wide range of opinions as to which clinical signs and symptoms may properly be considered to be those of arachnoiditis, and the possibility that arachnoiditis has been used as a scrapheap diagnosis for failed back surgery (Burton, 1978, 1985, 1991; Shipton, 1989), has meant that a diagnosis of arachnoiditis, if not confirmed by direct surgical observation, has often only been considered valid when radiological evidence was present. It is ironic, therefore, that myelography has been both the test by which the diagnosis must be confirmed and the procedure implicated in its development.

Radiological evidence consists of radiculographic/myelographic findings of obliteration of nerve root sleeves, and/or thickening of nerve-roots, and/or irregular distribution, fixation and/or loculation of contrast medium (Grahame et al., 1991). Myeloscopy has been recognised as having a valuable, but limited, role, particularly in avoiding unnecessary surgery when CT myelogram is not diagnostically conclusive (Peek, Thomas & Wiltse, 1993). MRI is now the most commonly used diagnostic tool (Laitt, Jackson & Isherwood, 1996). The noninvasive evaluation it offers is considered to excel at both diagnosing arachnoiditis and minimising the need for invasive procedures, which might potentiate its development (Gundry & Fritts, 1997). Advances in MRI permit recognition of mild or even minimal cases (Fitt & Stevens, 1995) at a level of near perfect agreement between myelographic and MRI diagnosis. A lack of correlation between the severity of radiological change and clinical status (Jorgensen et al., 1975) has led to the belief that the essential assessment in diagnosing arachnoiditis must be clinical (Johnson et al. 1991), supported only if necessary by radiological evidence.

Differential and/or associated diagnoses

There are a number of conditions various authors believe need to be differentially considered. These include cauda equina claudication, prolapsed intervertebral disc, failed back surgery, prolonged postoperative pain, recurrent prolapsed intervertebral disc (Shipton, 1989), disc herniation (Johnson et al., 1991), spinal stenosis (Long, 1992), spinal cord atrophy (Donaldson & Gibson, 1982), spinal intradural arachnoid cysts (Kriss & Kriss, 1997) and multiple sclerosis (Arachnoiditis, 1996). Other concomitants of arachnoiditis requiring differential diagnosis include syringomyelia, (Guyer et al., 1989; Brammah & Jayson, 1994) and calcific or arachnoiditis ossificans (Kitagawa, Kanamori, Tatezaki, Itoh, & Tsuji, 1990; Toribatake, Baba, Maezawa, Umeda & Tomita, 1995). Differential diagnosis is considered necessary because arachnoiditis may be masking other more treatable pathologies (Grahame et al., 1991).

Prognosis

Prior to the development of antibiotics, arachnoiditis was largely associated with infectious processes and occurred most frequently in the thoracic and cervical regions. Such cases tended to be inexorably progressive, resulting in paraplegia and, possibly, death (Guyer et al., 1989). Most cases now occur in the lumbo-sacral region and studies on this form of arachnoiditis have tended to be short term with little followup. A longer study (Guyer et al., 1989) followed fifty patients for ten years, and concluded that the natural course of the disease was mildly progressive in only a few, and that increased neurological deficits were more frequently linked to surgical intervention than to arachnoiditis itself. For the majority of people, the symptoms, disabling and painful though they may be, would wax and wane over the years without clear evidence of progression. People may function well in a sedentary lifestyle but few would return to full-time work. Alcohol and drugs could become a problem for some. Authors who mention prognosis tend do so in terms of incurability (Shipton, 1989; Guyer et al., 1989). The pain is mentioned as disabling and intractable (Guyer et al., 1989) or ineradicable (Shipton, 1989). Johnson et al., (1991) recognise the temptation to assign these patients to an untreatable category, with few practitioners feeling able to look after them. The poor prognosis may be a reflection of the lack of consensus over appropriate treatment for arachnoiditis, which mirrors a divergence of opinions over the management of neurogenic pain (Davies, Crombie & Macrae, 1993).

Treatment for arachnoiditis

A number of possible treatments have been researched and attempted. Pharmocological approaches include d-Penicillamine (Grahame et al., 1991) and epidural administration of bupivacaine and methylprednisolone, clonidine, buprenorphine, morphine, and intrathecal baclofen, combined with psychological support, evaluation and psychotherapy (Shipton, 1989).

The operative approach remains controversial. Dooley, McBroom, Taguchi and MacNab (1988) found that those with arachnoiditis posed a special problem, with a poor response to surgical intervention. Carroll and Wiesel (1992) noted that no surgical technique has been proven successful in either eliminating the scar tissue or significantly reducing the pain. Dolan (1993) considered that when all other approaches had been exhausted, any surgical treatment that offered a chance of improvement, even if for a variable time, might be justified, at the same time emphasising the expertise needed for such surgery if the situation was not to be worsened.

Tasker, DeCarvalho and Dolan (1992) noted that steady pain, the most common type suffered by people with cord central pain, the type present in arachnoiditis, responds poorly to destructive surgery and better to chronic stimulation of the cord or brain, which induces paraesthesia in the area of pain. However, this technique is recognised as only working in a relatively small proportion of patients (Martin & Yuan, 1996). Overall, the results of many surgical treatments have been discouraging, and are now reserved for specific circumstances, such as acute cauda equina syndrome and syringomyelia.

Clinical presentation

Previous literature reports that clinical presentation ranges from nonspecific back pain to definite signs of radiculopathy or myelopathy (Rosenberg & Stacey, 1996). Radiologically evident arachnoiditis can be asymptomatic (Johnson et al., 1991). Arachnoiditis occurs most commonly in the lumbar spine (Guyer et al., 1989) but can also occur in the cervical and thoracic regions, either from primary traumas in those areas or ascending from lumbar origins. Clinical presentation varies accordingly. The symptoms detailed in clinical presentation are the aspect of medical writings most pertinent to this study. They have, accordingly, been considered separately in Chapter Five.