Autoimmunity

Low levels of autoreactive T cells and B cells are present in normal individuals. Presumably they have escaped from clonal deletion during the process of immune development and are now tolerant of their antigens. Autoimmunity develops when these cells lose tolerance and a complex process of immune reactivity in target tissues begins. One potential way tolerance can be broken is by means of molecular mimicry between self-antigens and foreign antigens—for example, viral components. Several viral and bacterial peptides share structural similarities with important proteins of myelin, and a few of them are able to activate specific T-cell clones derived from patients with MS. Another way tolerance can be broken is by CNS infection that causes tissue damage and antigen release into the peripheral circulation, where corresponding autoreactive T cells may be encountered.

Myelin basic protein (MBP) has long been considered one of the primary candidates for an autoimmune attack. T cells that respond to MBP are found in the peripheral blood in both normal persons and those with MS, possibly at higher levels in MS patients with active disease. MBP, which accounts for 30% of the protein of myelin, can be an antigen for EAE, the primary animal model of MS. Recent clinical trials with altered peptide ligands of MBP support the potential pathogenic role of MBP-reactive T cells.

Several other myelin proteins are also candidates for an autoimmune attack. Proteolipid protein accounts for 50% of CNS myelin protein and is an integral membrane protein of the myelin leaflets. In the PNS, P0 protein fulfills this role. Myelin-associated glycoprotein, myelin oligodendrocyte glycoprotein, and cyclic nucleotide phosphodiesterase are proteins that account for a few percent of myelin. Myelin oligodendrocyte glycoprotein and cyclic nucleotide phosphodiesterase are not found in peripheral nerve myelin and are therefore of interest because MS is a disease affecting only central myelin.

Although the possibility of autoimmunity as the causal mechanism for MS exists, the issue is not proven. The evidence for MS being a dysimmune condition is more compelling, with alterations in immune cell repertoire and activation state both in blood and CSF of MS patients compared to others (Conlon et al., 1999; Hafler et al., 2005).

Infection

A possible role for microbial infection in the causation of MS has been a matter of ongoing debate for decades. The epidemiology of MS (see Epidemiology in this chapter and in Chapter 39) suggests an exogenous or environmental factor. Beyond epidemiology and much speculation, little direct evidence supports the concept of a role for viral infection. Innumerable efforts to culture a virus from autopsy-derived or biopsy material have yielded no consistent result. Serological data are difficult to interpret, because titers may reflect only a nonspecific tendency toward increased immune reactivity. Specific efforts to recover a known viral genome (e.g., that of human T-cell lymphotropic virus type 1 [HTLV1]) have proven negative.

Human herpesvirus 6 (HHV6), Epstein-Barr virus (EBV), and Chlamydia pneumoniae have been the focus of interest as potential triggers for MS. Several studies of serum and CSF samples have yielded varying results. For example, one early study showed 47% of MS brains were positive for HHV6, and 80% showed elevation of antibodies in the serum. A more recent study found no HHV6 DNA in any CSF sample, and the serum antibody titers were comparable to the general population. High serum levels of EBV antibodies have also been associated with an increased risk of MS. The strongest association was found for antibodies to Epstein-Barr nuclear antigen (EBNA-2); a fourfold difference in titers was associated with increased relative risk (RR) of developing MS (Ascherio et al., 2001).

Age of Onset

Most studies agree that the mean and median age of onset in relapsing forms of MS is age 29 to 32. The peak age of onset is approximately 5 years earlier for women than for men. Primary progressive MS (PPMS) has a mean age of onset of 35 to 39 years. The onset of MS can occur as late as the seventh decade. Perhaps as many as 5% of cases of MS have their onset before age 18. Most of these cases occur in adolescents, but a small percentage begin in the first decade of life (Fig. 54.6).

Fig. 54.6 Age at onset of symptoms of multiple sclerosis in 940 patients followed at the multiple sclerosis clinic of the Montreal Neurological Institute. Mean age of onset is 30.6 years, median is 27 years, and peak incidence is 25 years.

Sex Distribution

Autoimmune diseases in general and MS in particular affect more women than men. In a summary of 30 incidence and prevalence studies, a cumulative ratio of female to male subjects was 1.77 : 1.00.

Mortality

Mortality due to MS is difficult to ascertain because of poor data collection and reporting. The U.S. Department of Health and Human Services report of deaths in the year 1992 indicates that 1900 U.S. citizens died of MS in that year, giving MS a U.S. mortality of 0.7 per 100,000. The mean age of death of all patients with MS was 58.1 years, compared with a national average of 70.5 years for all causes of death. The life expectancy of patients with MS was therefore calculated to be 82.5% of the normal lifespan. In Denmark, in an exceptionally complete survey of the country, median survival after diagnosis for men was 28 years and for women 33 years, compared with matched population death rates of 37 and 42 years, respectively. The life expectancy was 10 years shorter for MS patients than age-matched controls; this difference was becoming smaller when more recent data were studied (Bronnum-Hansen et al., 2004).

In another study, MS mortality figures were calculated for England and Wales for the years 1963-1990. Over this span of years, the death rate attributable to MS steadily and consistently declined compared with the overall death rate. Patients with MS tended to live longer, and other diseases were more likely to be the cause of death. Current estimates indicate that about half of the deaths in MS patients are directly due to the disease, slightly more than half if accidents and suicide are included as indirect causes.

Geographical and Racial Distribution

More than 250 prevalence surveys have been carried out, serving as the basis for the delineation of geographical risk for MS depicted in Fig. 54.7. High-frequency areas of the world, with current prevalence of 60 per 100,000 or more, include all of Europe (including Russia), southern Canada, the northern United States, New Zealand, and the southeastern portion of Australia. In many of these areas, the prevalence is more than 100 per 100,000, with the highest reported rate of 300 per 100,000 occurring in the Orkney Islands. In the United States, the prevalence when measured during the 1990s was approximately 350,000. Several studies suggest that the prevalence is increasing beyond what might occur because of enhanced recognition and better-appreciated diagnostic techniques.

Fig. 54.7 Worldwide distribution of multiple sclerosis as of 1980. High-frequency areas (>30 per 100,000 population) are orange, medium-frequency areas (5 to 25 per 100,000) are yellow, and low-frequency areas (<5 per 100,000) are green. Open areas are regions without data. South American frequencies are tentative. The low frequency in South Africa is for native South Africans, with medium frequency for persons of European stock native to South Africa.

(Reprinted with permission from Kurtzke, J.F., 1980. The geographical distribution of multiple sclerosis—an update with special reference to Europe and the Mediterranean region. Acta Neurol Scand 62, 65-80.

One possible conclusion is that MS is a location-related illness with a latitude gradient. However, notable exceptions then have to be explained. Japan, situated at the same latitude as areas of high prevalence in Europe, is a low-risk area. Second-generation Japanese in the United States retain their parents’ low risk of MS. The white population of South Africa, with medium prevalence of MS, is surrounded by a black population in whom the disease is very uncommon. Native North Americans, especially of pure Amerindian background, have a very low prevalence but are surrounded by a white population with a medium or high risk for MS.

It seems plausible that race is a determinant of MS risk, with populations of white extraction, especially from Northern Europe, being most susceptible. People of Asian, African, or Amerindian origin have the lowest risk, whereas other groups are variably intermediate.

Migration data have often been used to support the view that an environmental agent is involved in the pathogenesis of MS. The data indicate that persons migrating from an area of high risk to an area of low risk after the age of puberty carry their former high risk with them. With migration during childhood, the risk seems to be that of the new area to which the person has migrated.

Genetics

MS is a genetically complex disease. Many genetic and nongenetic factors of moderate effect interact to influence disease probability. Compared to the general population, a higher frequency of familial occurrence of MS suggests a strong but non-Mendelian inheritance of susceptibility. Twin studies established the importance of genetic factors: the concordance rate for a clinical diagnosis of MS in monozygotic twins is about 30%, whereas in dizygotic twins it is 2% to 5%. (Ebers et al., 1995; Willer et al., 2003). The risk is highest for siblings: 3% to 5%, or 30 to 50 times the background risk for this same population. Adoptive relatives, when raised from infancy with the patients with MS, are no more likely to develop MS than the general population. This indicates that the familial aggregation of MS is more likely to be related to shared genetic variations rather than to a shared family environment. In some studies, unaffected family members may have been found to have abnormalities on MRI, implying the risk may be even higher.

MS is associated with both MHC and human leukocyte antigen (HLA) class I A3 and B7 antigens. The class II polymorphisms, Dw2 and DR2, also show strong association, and specifically the HLA DRB1*1501 allele (Oksenberg et al., 2004). Additional HLA alleles that carry protective as well as detrimental effects with regard to MS susceptibility have been identified. HLA-A*02, for example, has a protective effect relative to MS susceptibility. In recent years, additional MS susceptibility loci outside of the MHC have been described. Specifically, the loci coding for IL-7 receptor and IL-2 receptor are strongly linked with MS susceptibility (Zuvich et al., 2010). IL-7 and IL-2 receptor signaling is critical for the differentiation of CD4⁻ CD8⁻ thymocytes and has a role in survival of CD4⁺ CD8⁺ cells after positive selection. This may be important not only in MS predisposition but also in disease course and outcome.

Recently an aggregate measure of MS risk that combines weighted odds ratios from 16 genetic susceptibility loci (the weighted genetic risk score [wGRS]) was created and used to predict a diagnosis of MS in three independent cohorts (De Jager et al., 2009). The results revealed that the wGRS can modestly but robustly predict MS risk. Individuals with MS have a greater “load” of susceptibility factors than healthy people, although the 16 loci assessed so far are insufficient to differentiate MS cases from healthy controls. While this tool is still in early stages of development, it is quite likely a useful application of the wGRS may be to predict individuals who are at higher risk for developing the disease, such as first-degree relatives of MS patients. Identifying individuals during the clinically silent phase of the disease could be extremely helpful in guiding the selection of those who would benefit most from early imaging screening, and eventually from early treatment.

Cognitive Impairment

Cognitive involvement in MS was documented as early as 1877 by Charcot. He observed that patients with “multilocular sclerosis” are slow to form conceptions, have “marked enfeeblement of the memory, and blunting of intellectual and emotional faculties.” However, the subsequent era of research in MS focused primarily on physical disability. There was a paucity of rigorous studies, with a focus on cognition. In 1981, Kurtzke reported that only 5% of patients with MS suffered from cognitive impairment. The Kurtzke Expanded Disability Status Scale (EDSS) focused primarily on somatic disability measures. A decade later, data from formal neuropsychological studies indicated that cognitive involvement has been underreported in MS (Rao et al., 1991). Neuropsychological test results have shown that 34% to 65% of patients with MS have cognitive impairment (Nocentini et al., 2006).

Cognitive abnormalities affect patients across the disease spectrum and involve all MS subtypes. Some 49% to 53.7% of clinically isolated syndrome (CIS) and early relapsing-remitting multiple sclerosis (RRMS) patients show significant impairment in one or more cognitive domains that impacts their quality of life (Achiron and Barak, 2003; Glanz et al., 2007, 2010). A recent longitudinal study reported 29% of early PPMS patients (within 5 years from disease onset) as cognitively impaired (Penny et al., 2010; Ukkonen et al., 2009). The domains of verbal memory and attention/speed of information processing were the most affected, and premorbid IQ was a robust predictor of cognitive status in this cohort. A reported 19% of patients with a so-called benign MS variant have progressive cognitive disability and structural brain MRI changes similar to those with SPMS (Mesaros et al., 2009; Rovaris et al., 2009). It is increasingly recognized that low physical disability can coexist with significant cognitive disease. In a less common clinical presentation, patients may have a subacute or fulminant cognitive decline with signs of cortical dysfunction. This disability progresses rapidly and results in significant socio-occupational dysfunction, while physical disability remains minimal (Houtchens et al., [in press]).

In general, the most frequently reported abnormalities are with working memory, attention, and speed of information processing. Patients complain of memory loss, difficulties at work or with interpersonal relations, inability to multitask, and “mental fog and fatigue.” Comorbid depression, anxiety disorders, and emotional lability may further affect cognitive performance. Mild to moderate abnormalities are usually not apparent during a routine office visit, and simple screening tools for cognitive dysfunction are not yet available. A 5- to 15-minute battery of three tests assessing functions most commonly impaired in MS patients is now being validated (Portaccio et al., 2009). Longer neuropsychological testing batteries are designed to assess cognitive impairment in MS patients and are used primarily in research trials. The Brief Repeatable Battery of Neuropsychological Tests (BRB-N) (Rao et al., 1991) and Minimal Assessment of Cognitive Function in Multiple Sclerosis (MACFIMS) (Benedict et al., 2002) are the most widely used screening tools. Multiple Sclerosis Functional Composite (MSFC) also includes the Paced Auditory Serial Addition Test (PASAT) as a measure of sustained attention (Cutter et al., 1999).

Cognitive impairment in MS has been associated with a variety of MRI metrics of disease activity and progression. Early studies reported high correlations with the total T2 lesion burden and volume (Patti et al., 1998; Rao, 1989). Lesions located in frontal and parietal lobes showed stronger correlations with tasks of processing speed, attention, and verbal memory (Sperling et al., 2001). Global and regional atrophy plays a role in cognitive impairment. Thalamic atrophy appears to correlate particularly well with impairment on a variety of cognitive tests in MS patients (Houtchens et al., 2007). Contribution of cortical lesions (Bakshi et al., 2001; Zivadinov and Miagarb, 2009) and loss of cortical ribbon thickness (Calabrese et al., 2010) are important. Cognitively impaired patients have abnormalities in normal-appearing white and gray matter as measured by magnetization transfer ratio (MTR) that are not seen in cognitively intact MS patients (Amato et al., 2008).

Treatment of cognitive impairment in MS is challenging. Pharmacological therapy focuses on treatment of underlying disease as well as symptomatic manifestations of declining mental function, such as difficulties with attention, memory, and fatigue. Appropriate management of the comorbid psychiatric disease often improves cognitive performance. A recent randomized study of 469 patients showed that treatment with higher versus lower doses of subcutaneous interferon beta-1a was predictive of lower cognitive impairment at 3 years of follow-up (Patti et al., 2010). An earlier trial of intramuscular (IM) interferon beta-1a showed a 47% reduction in cognitive decline on PASAT testing over 2 years of follow-up (Fischer et al., 2000).

There is some evidence that treatment with l-amphetamine is associated with improved learning and memory in cognitively impaired MS patients (Benedict et al., 2008; Morrow et al., 2009). Modafinil may improve focused attention and dexterity and palliate fatigue. It has a favorable safety profile and can be used for symptomatic management of MS patients with severe fatigue and decreased focus. There is no convincing evidence that cholinesterase inhibitors improve memory in MS patients. A small study assessed the effects of donepezil on SPMS patients who were residents in a nursing home, documenting improvement in several cognitive domains after 8 weeks of therapy (Greene et al., 2000). There are additional anecdotal reports of efficacy in small numbers of subjects. Cognitive-behavioral therapy, family and individual counseling, strategies to improve day-to-day function, and necessary job modifications and accommodations can be of great help to MS patients suffering from cognitive decline.

Affective Disorders

Cross-sectional studies have shown some degree of affective disturbance in a significant number of patients with MS (Arnett et al., 2005). Depression is the most common manifestation and is in part secondary to the burden of having to cope with a chronic disease. However, it is more prevalent in MS than in other chronic diseases, suggesting an organic component as well. The lifetime risk of major depression in patients with MS is up to 50%, compared with 12.9% in patients with chronic medical conditions in another study. Some data indicate a comorbid association, presumably genetic, between bipolar illness and MS. Suicide rates are higher in patients with MS than in the general population or when compared to patients with other chronic illnesses (Bronnum-Hansen et al., 2005). Frontal or subcortical white-matter disease may also be a contributory causative factor. Euphoria, formerly considered to be common in MS, is actually infrequent and is usually associated with moderate or severe cognitive impairment and greater disease burden on MRI. However, emotional “dyscontrol” is quite common, and patients oscillate frequently between sad and happy states, at times without precipitant.

Cranial Nerve Dysfunction

Impairment of Sensory Pathways

Sensory manifestations are a frequent initial feature of MS and are present in almost every patient at some time during the course of disease. The sensory features can reflect spinothalamic, posterior column, or dorsal root entry zone lesions. The sensory symptoms are commonly described as numbness, tingling, pins and needles, tightness, coldness, itching, or swelling of limbs or trunk. Radicular pains, unilateral or bilateral, can be present, particularly in the low thoracic and abdominal regions, or a bandlike abdominal sensation may be described.

The most frequent sensory abnormalities on clinical examination are varying degrees of impairment of vibration and joint position sense, decrease of pain and light touch in a distal distribution in the four extremities, and patchy areas of reduced pain and light touch perception in the limbs and trunk. A bilateral sensory level, frequently ascending in character, is a more frequent finding than a hemisensory (Brown-Séquard) syndrome. Patients commonly report that the feeling of pinprick is increased or feels like a mild electric shock or that the stimulus spreads in a ripple fashion from the point at which it is applied. The sensory useless hand (the “numb clumsy hand” syndrome) is a characteristic but uncommon feature, consisting of an impairment of function secondary to a pronounced alteration of proprioception, without loss of power. A lesion of the relevant root entry zones or posterior columns in the spinal cord may be responsible in such cases.

Impairment of Motor Pathways

Corticospinal tract dysfunction is common in MS. Paraparesis or paraplegia occurs more frequently than significant weakness in the upper extremities. With severe spasticity, extensor or flexor spasms of the legs and sometimes the trunk may be provoked by active or passive attempts to rise from a bed or wheelchair. The physical findings include spasticity, usually more marked in the legs than in the arms. The deep tendon reflexes are exaggerated, sustained clonus may be elicited, and extensor plantar responses are observed. All of these manifestations are commonly asymmetrical. The Achilles reflex can be absent in lesions of the sacral segments of the spinal cord, with or without concomitant sphincter and sexual problems. Occasionally, reduced reflexes reflect hypotonia due to cerebellar pathway lesions. Amyotrophy, when observed, most frequently affects the small muscles of the hand; lesions of the motor root exit zones may produce muscle denervation secondary to axon loss.

A common pattern of disease evolution seen in the spinal form of MS is an ascending pattern of weakness that begins with involvement of the lower extremities and spreads to involve first one upper extremity and then the other, beginning in the intrinsic hand muscles. Often there is an associated weakness of the trunk muscles, with abnormal posture and, much less often, involvement of respiratory muscles.

Impairment of Cerebellar Pathways

Cerebellar pathway impairment results in gait imbalance, difficulty performing coordinated actions with the arms, and slurred speech. Examination reveals the usual features of cerebellar dysfunction: dysmetria, decomposition of complex movements, and hypotonia, most often observed in the upper extremities. An intention tremor in the limbs and titubation of the head may be seen. Walking is impaired by ataxia. Ocular findings of nystagmus, ocular dysmetria, and frequent refixation saccades suggest cerebellar or cerebellovestibular connection dysfunction. Speech can be scanning or explosive in character. In severe cases, complete astasia (inability to stand), inability to use the arms because of a violent intention tremor, and virtually incomprehensible speech may occur.

Impairment of Bladder, Bowel, and Sexual Functions

The extent of sphincter and sexual dysfunction often parallels the degree of motor impairment in the lower extremities. The most common complaint related to urinary bladder dysfunction is urgency, usually the result of uninhibited detrusor contraction, reflecting a suprasegmental lesion. As the disease progresses, urinary incontinence due to urgency becomes more frequent. With involvement of sacral segments of the spinal cord, symptoms of bladder hypoactivity may evolve (e.g., decreased urinary flow, interrupted micturition, incomplete bladder emptying). An atonic dilated bladder that empties by overflow results from loss of perception of bladder fullness and is usually associated with urethral as well as anal and genital hypoesthesia, and sensory deficits in the sacral dermatomes. A dyssynergic voluntary sphincter that interrupts bladder emptying will lead to frequent, small-volume urinations combined with a large postvoid residual. Urinary tract infections are common in MS, especially in women, and may present in atypical patterns.

Constipation is more common than fecal incontinence and is mainly due to decreased general mobility and the tendency of MS patients to restrict their fluid intake in a misguided attempt to decrease urinary urgency and incontinence. Almost all patients with paraplegia require special measures to maintain regular bowel movements.

Sexual dysfunction, although frequently overlooked, occurs commonly in MS. Approximately 50% of patients become completely sexually inactive because of their disease, and an additional 20% become less sexually active. Men experience various degrees of erectile dysfunction, often with rapid loss of erection at attempted intercourse, whereas loss of ejaculation is less common. Sexual dysfunction can be the result of multiple problems, including the direct effects of lesions on the motor, sensory, and autonomic pathways within the spinal cord as well as psychological factors that affect libido: self-image, self-esteem, and fear of rejection by the sexual partner. Mechanical difficulties created by spasticity, paraparesis, and incontinence may further aggravate the problem.

Clinical features uniquely characteristic of MS (Table 54.2) include bilateral INO, the Lhermitte phenomenon, and paroxysmal attacks of motor or sensory phenomena such as paroxysmal diplopia, facial paresthesia, trigeminal neuralgia, ataxia, dysarthria, and tonic spasms. These paroxysmal attacks usually respond to low doses of anticonvulsants and frequently remit after several weeks to months, usually without recurrence. Heat sensitivity (e.g., the Uhthoff phenomenon) is a well-known occurrence in MS. Fatigue is also common and is usually described as physical exhaustion unrelated to the amount of activity performed. The degree of fatigue correlates poorly with the overall severity of disease or the presence of any particular symptom or sign. In contrast to the situation with cognitive deficits, no MRI findings correlate with fatigue or depression. Fatigue is often seen in association with an acute attack, may precede the focal neurological features of the attack, and may persist long after the attack has subsided.

Table 54.2 Common Clinical Features of Multiple Sclerosis

CNS, Central nervous system. PNS, peripheral nervous system.

Differential Diagnosis

The differential diagnosis of MS (Box 54.3) is quite limited in the setting of a young adult with two or more clinically distinct episodes of CNS dysfunction with at least partial resolution. Problems arise with atypical presentations, monophasic episodes, or progressive illness. The unusual nature of some sensory symptoms may result in a misdiagnosis of conversion disorder. The retrospective nature of inquiries also blurs details of prior events, making clear ascertainment of prior attacks difficult in some cases. A monophasic illness with symptoms attributable to one site of the CNS creates a large differential diagnosis that includes neoplasms, vascular events, and infections. Appropriate imaging studies may help clarify the situation, depending on the site of involvement and clinical progression. The most trouble arises with progressive CNS dysfunction, in which great care must be taken to exclude treatable etiologies (e.g., vitamin B₁₂ deficiency, compressive spinal cord lesions, arteriovenous malformations, cavernous angiomas, Arnold-Chiari malformation), infectious causes (syphilis, HTLV-1, human immunodeficiency virus [HIV]), and hereditary disorders (adult metachromatic leukodystrophy, adrenomyeloneuropathy, spinocerebellar disorders, CADASIL).

A common error is to overinterpret multiple hyperintense lesions on MRI as equivalent to MS. Clinical symptoms must be consistent with MS. A few white-matter lesions on T2-weighted MRI scans are not infrequent, particularly in the elderly, but do not indicate a diagnosis of MS. CNS vasculitides such as systemic lupus erythematosus (SLE), Sjögren disease, polyarteritis nodosa, syphilis, retroviral diseases, and Behçet disease may all produce multifocal lesions with or without a relapsing-remitting course. SLE can present as a recurrent neurological syndrome before the systemic manifestations of this disease declare themselves. Behçet syndrome is characterized by buccogenital ulcerations in addition to the multifocal neurological findings. CNS sarcoidosis can be mistaken for MS with multifocal neurological and MRI lesions. Although uncommon, ADEM must be considered in the differential diagnosis (see Acute Disseminated Encephalomyelitis in this chapter). An MS-like phenotype associated with mitochondrial gene defects has been described. More important than features characteristic for MS are features that should prompt the clinician to reconsider the diagnosis of MS—red flags indicating that another diagnosis is more likely (Charil et al., 2006). Features that should alert the clinician to the possibility of other diseases include (1) family history of neurological disease, (2) a well-demarcated spinal level in the absence of disease above the foramen magnum, (3) prominent back pain that persists, (4) symptoms and signs that can be attributed to one anatomical site, (5) patients who are older than age 60 or younger than age 15 at onset, and (6) progressive disease (see Table 54.3).

Course

The most characteristic clinical course of MS is the occurrence of relapses (Fig. 54.8), which can be defined as the acute or subacute onset of clinical dysfunction that usually reaches its peak from days to several weeks, followed by a remission during which the symptoms and signs usually resolve partially or completely. The minimum duration for a relapse has been arbitrarily established at 24 hours. Clinical symptoms of shorter duration are less likely to represent what is considered a true relapse (i.e., new lesion formation or extension of previous lesion size). Worsening of previous clinical dysfunction can occur concurrently with fever, infection, physical activity, or metabolic upset and last for hours to a day or more and is referred to as pseudorelapse. Summaries of many studies provide an average figure of 0.4 to 0.6 relapses per year. The attack rate in the placebo group in clinical studies ranges from 0.8 to 1.2 attacks per year. In general, relapses are more frequent during the first years of the disease and tend to wane in later years.

Fig. 54.8 The clinical courses of multiple sclerosis.

(Data from Lublin, F.D., Reingold, S.C., 1996. Defining the clinical course of multiple sclerosis: results of an international survey. National Multiple Sclerosis Society (USA) Advisory Committee on Clinical Trials of New Agents in Multiple Sclerosis. Neurology 46, 907-911.)

A course marked by relapses interspersed with periods during which the disease seems relatively dormant is termed relapsing-remitting. A small number of patients never experience a second relapse. The exact frequency of such benign disease is unknown but is thought to be about 5%. Autopsy studies found a significant number of cases with CNS pathology consistent with MS and yet no documented clinical evidence of such disease. Similarly, MRI studies have shown MS-like plaques in T2-weighted scans in patients who have never had a neurological episode. A standardization of terms has been agreed on to determine the pattern and course of the illness (Lublin and Reingold, 1996). Four categories of disease are described (see Fig. 54.8):

Two severity outcomes are also described: benign MS is disease in which the patient remains fully functional in all neurological systems 15 years after the disease onset; and malignant MS is disease with a rapid progressive course leading to significant disability in multiple neurological systems or death in a relatively short time after disease onset. It is defined by EDSS score of 6 (unilateral assistive device is needed for ambulation) reached within 5 years from disease onset.

The rate of clinical progression of MS is variable. The commonly used index of clinical disability, the Kurtzke EDSS, uses numbers ranging from 0 for normal examination and function to 10 for death caused by MS. This scale is nonlinear, with great emphasis on ambulation capabilities with scores above 4. Most MS populations have bimodal distributions of EDSS scores, with peaks at values of 1 and 6 (ambulation with unilateral assistance). In a cohort of patients followed for 25 years (in the pretreatment era), the following data emerged: 80% of the patients had reached the progressive phase by 25 years, 15% had died, 65% had reached EDSS 6 (requiring aids for walking), and 50% reached EDSS 6 within 16 years of onset. More recent studies suggest a somewhat slower course of progression (Pittock et al., 2004; Tremlett et al., 2006). The EDSS, although universally used in clinical trials, has a number of limitations. Even with special training and examiner blinding, interrater and intrarater variations in scoring are common. EDSS scores of 4 and higher depend almost entirely on the ability to walk; developing dementia, vision loss, and weakness of hands may pass undetected by the scoring once one reaches these levels. An obvious implication of these facts is that other outcome measures should be used as well, and that minor changes in EDSS score alone should not be overinterpreted. The MSFC is a clinical tool designed to avoid the problems encountered with the EDSS (Cutter et al., 1999). The MSFC consists of three parts: PASAT, nine-hole peg test (9HPT), and timed 25-foot walk (T25FW). These three measures take into account cognition, upper-extremity function, and lower-extremity function. A z-score is obtained for each measure, and a combined z-score is then derived. The MSFC has been validated in several clinical trials. The tests can be performed by a nonphysician and are highly reproducible and predictable.

Pregnancy in Multiple Sclerosis

MS preferentially affects women of childbearing age. Scant evidence and little agreement exist about the influence of pregnancy on disease activity and, conversely, the effects of MS on pregnancy outcomes. Pregnancy is recognized to induce changes in the maternal immune system, including both immunosuppression on a local level and a heightened state of immunocompetence on a global level. Several retrospective studies reported an overall increase in relapse rate during the postpartum period and a lower relapse rate during pregnancy itself (Brick et al., 1990; Weinshenker et al., 1989). Pregnancy in Multiple Sclerosis (PRIMS) was the first prospective study of 254 women (269 pregnancies). Subjects were followed for 2 years after delivery (Confavreaux et al., 1998; Vukusic et al., 2004, 2006). In the cohort, a pre-pregnancy rate of 0.7 relapses per year decreased to 0.2 per year in the third trimester. The relapse rate increased to 1.2 per year in the first 3 months postpartum. However, 72% of women did not experience any relapses during the study period. An increased relapse rate in the year before pregnancy, an increased relapse rate during pregnancy, and a higher EDSS score at the beginning of pregnancy correlated significantly with occurrence of a postpartum relapse. Epidural anesthesia and breastfeeding were not predictive of a subsequent relapse or disability progression. There are no contraindications to caesarean section or vaginal delivery in MS patients.

A prospective 5-year study compared the rate of progression in disability between childless women, women who had onset of MS after childbirth, and women who had onset before or during their pregnancy (Stenager, 1994). The rates of disability increased most rapidly in nulliparous women. Another recent study retrospectively examined childbirth’s effects on disability progression in 330 women with MS (D’hooghe, 2010). Women who gave birth after MS onset reached EDSS scores of 6 significantly later in the disease course than those who did not (median time to progression, 13-15 years versus 22-23 years). Overall, there is no compelling evidence of adverse effects of pregnancy on MS progression. However, potential risks of prolonged MS therapy discontinuation, level of baseline neurological disability, and other factors need to be taken into consideration when counseling patients on family planning.

Another related issue is whether the disease has any effects on pregnancy outcomes, risk of malformations, fetal birth weight, or duration of pregnancy. Some groups report no increased risk in incidence of pregnancy and labor and delivery–related adverse events in MS patients (Mueller et al., 2002). Other reports indicate higher rates of operative deliveries and induced labor as well as greater numbers of neonates with low birth weight or being small for gestational age (Dahl et al., 2005). Kelly and colleagues evaluated obstetric outcomes in women with MS, epilepsy, or pregestational diabetes mellitus (DM) and in healthy controls. MS patients had a 30% higher risk for cesarean delivery and 70% higher rate of intrauterine growth restriction (IUGR) than healthy women. There were no long-term adverse pediatric outcomes (Kelly et al., 2009).

There are variable data on the safety of immunomodulatory and immunosuppressant therapies used in MS, but most medications are not recommended for use in pregnant or lactating women. Intravenous (IV) steroids are used in MS for treatment of acute relapses. Their use is not associated with fetal malformations and is probably safe in the second and third trimesters of pregnancy. They should be reserved for treatment of serious exacerbations of the disease because they cross the placental barrier and may cause transient neonatal leukocytosis (dexamethasone) or immunosuppression (methylprednisolone).

Pregnancy-dose estriol treatment trials in MS have been conducted over the last 5 years. In a crossover phase I trial, patients were treated with 8 mg/day of estriol combined with 100 mg/day of progesterone, achieving estriol levels seen at 6 months of pregnancy. Subjects demonstrated significant decreases in interferon-γ levels in peripheral blood mononuclear cells and decreased gadolinium-enhancing lesions on monthly brain MRI. These effects were lost with cessation of treatment and reproduced following reinstitution of estriol therapy (Sicotte et al., 2002; Voskuhl et al., 2001). The possible benefits of high-dose estrogen therapy in MS have to be weighed against adverse events associated with chronic high-dose estrogen use such as carcinogenesis, thrombogenesis, and dysfunctional uterine bleeding. Further studies with estrogen therapies are currently ongoing.

Optic Neuritis

Patients often present with acute or subacute unilateral decrease or loss of vision. Central vision tends to be most affected, and orbital pain with eye movement is common. Neurological exam often reveals reduced visual acuity on the affected side, relative afferent papillary defect, and red color desaturation. When the acute ON lesion involves the head of the optic nerve, one may observe disc edema (papillitis), a finding more commonly seen in children than in adults. Fundoscopy may show red and inflamed optic nerve head. More often, the lesion of the optic nerve is retrobulbar, and funduscopic examination is normal in the acute stage. Chronic changes on funduscopy are seen as a pale optic nerve head. Patients may have enlarged central scotoma following an optic neuritis event. Differential diagnosis includes retinal or macular disease, compressive or infiltrative optic neuropathy, granulomatous disease, ischemic optic neuropathy, optic nerve tumor such as glioma or lymphoma, mitochondrial disease, and vasculitis. Visual evoked potentials (VEPs) show prolonged P100 wave latency on the affected side. The optic nerve may show enhancement on postcontrast MRI images. Five-year data from the original Optic Neuritis Treatment Trial revealed that the 5-year cumulative probability of developing CDMS was 30% and did not depend on treatment. However, MRI was a strong predictor; the 5-year risk of developing CDMS was 16% in patients with no brain MRI lesions and 51% in patients with three or more lesions. Immunomodulating treatment may be offered based on CHAMPS study. Treatment of optic neuritis with interferon beta-1a reduced conversion rate to CDMS by 50%.

Myelopathic Syndromes

Neuroimaging

Magnetic Resonance Imaging

MRI is the preferred imaging modality for both making the diagnosis of MS and longitudinal follow-up of patients. MRI is based on relaxation properties of water in tissues and is sensitive to T2 (transverse) and T1 (longitudinal) relaxation rates of protons. Gadolinium diethylenetriamine pentaacetic acid (DTPA) is a paramagnetic contrast agent. It crosses the disrupted BBB, indicating increased vascular permeability in association with inflammation. Gadolinium (Gd) enhancement is best seen on spin-echo T1-weighted imaging. These fundamental properties of MR image acquisition provide a sensitive measure of pathology; areas of brain inflammation, demyelination, and loss of axons can be especially well seen with this technique. More than 95% of patients with MS have T2 and fluid-attenuated inversion recovery (FLAIR) abnormalities (Fig. 54.9). New lesions occur many times more often than new clinical attacks in RRMS. In the brain, typical lesions are located in the periventricular white matter, in and near the corpus callosum, the deep white matter, the centrum semiovale, as well as in cortical and deep gray-matter structures. Juxtacortical lesions affecting U-fibers are commonly seen in MS. Posterior optic radiation lesions are also frequently present. Posterior fossa lesions are present in the cerebellum, middle cerebral peduncles, and in areas adjacent to the fourth ventricle.

Fig. 54.9 Magnetic resonance imaging studies in a 29-year-old man. A, Multifocal lesions in the centrum semiovale (proton density image). B, Multiple, at times confluent, white-matter lesions abutting the lateral ventricles (proton density image). C, Lesions distributed in a radiating fashion from the corpus callosum (Dawson fingers). Significant cerebral atrophy with ventriculomegaly and cortical atrophy is also noted (T2-weighted image).

Characteristic cerebral lesions, or plaques, are focal and discrete, have an ovoid appearance, and are oriented perpendicularly to the plane of the lateral ventricles. These are classically referred to as Dawson fingers and are thought to represent perivenular inflammation, seen pathologically in MS plaques.

Lesion Differentiation Based on Mri Appearance

MRI is a helpful tool to distinguish between acute, subacute, and chronic MS lesions. Acute lesions of less than 8 weeks duration often show Gd contrast enhancement on T1-weighted sequences, indicating inflammation and BBB disruption. Enhancement patterns can appear as concentric open or complete rings, or as patchy or homogenously enhancing lesions. Ring-enhancing lesions are associated with significant tissue destruction (Minneboo, 2005). They may also be associated with a bright signal on diffusion-weighted imaging (DWI). Contrast-enhancing lesions may precede the appearance of T2 lesions (hyperacute MS plaque) or be associated with a bright T2 or FLAIR signal (acute MS plaque). Acute MS lesions tend to be larger in size, with less well-defined margins. Infrequently, acute T2 bright lesions can disappear on subsequent scans, indicating reversible tissue inflammation and edema. At times, acute MS lesions may show an associated dark signal on T1 noncontrast scan (acute T1 lesion), consistent with edema and demyelination. Subacute MS plaques may no longer show contrast enhancement but may continue to show bright DWI abnormality. Chronic MS plaques appear hyperintense on T2 or FLAIR sequences, are usually smaller, and have sharper margins. This implies smaller, patchy areas of demyelination and gliosis with lesion evolution. Persistent and profound T1 hypointensity (chronic T1 lesion) usually reflects irreversible tissue damage such as axonal loss and permanent demyelination (Neema, 2007). Deep gray-matter structures often appear hypointense on T2-weighted images. Iron deposition reduces T2 relaxation time and is postulated to be a cause for this finding. While T2 hypointense signal in basal ganglia showed independent association with whole and regional brain atrophy and physical and cognitive impairment in MS, it remains unclear whether it is an epiphenomenon or a direct contributor to secondary damage (Bakshi, 2002; Brass, 2006).

Role of Mri in Multiple Sclerosis Diagnostic Criteria

In 2001, the McDonald Criteria formalized the role of MRI in the diagnosis of MS. A change in MRI scan at 3 months following a clinical event can be substituted for a clinical attack and satisfy dissemination in time criteria necessary for an MS diagnosis. Three out of the four factors had to be true to satisfy MRI criteria for dissemination in space:

One Gd-enhancing lesion or nine T2-hyperintense lesions.

At least one infratentorial lesion.

At least one juxtacortical lesion.

At least three periventricular lesions.

In 2005, these criteria were revised. Dissemination in space criteria were changed to allow any cord lesions to be substituted for a brain lesion. A cord lesion now carries the same weight as an infratentorial lesion. In addition, a new T2 lesion seen more than 30 days after a clinical event, and a new Gd-enhancing lesion seen more than 3 months after a clinical event could satisfy dissemination in time criteria (see Table 54.4).

Two typical and verifiable clinical events separated in time and space can still be used to satisfy an MS diagnosis, even without meeting MRI criteria. Improved and simplified criteria are highly specific and more sensitive than previous Posner criteria to correctly identify patients with MS.

Predictive Value Of Mri In Conversion To Clinically Definite Multiple Sclerosis

Studies suggest that MRI may provide some prediction of the risk of conversion to CDMS based on the presence and number of T2 lesions within 1 to 5 years following CIS. Increasingly higher numbers of T2 lesions result in greater likelihood of conversion to CDMS and higher disability scores at 5 years. Two long-term follow-up studies of 10 to 14 years show that most patients with CIS and MRI abnormalities will develop CDMS (70%-80%), while only 20% of patients with normal cerebral MRI scans will be diagnosed with MS over this time period (Brex et al., 2002; Optic Neuritis Study Group, 2003; O’Riordan et al., 1998). The concept of radiologically isolated syndrome was introduced recently (Bourdette, 2009; Lebrun, 2009; Miller, 2008) and describes a subset of individuals with no clinical symptoms or neurological events, but with imaging findings on MRI suggestive of demyelination. The reported rate of conversion to CIS in this group was 33% in 5 years. This finding may well pioneer a concept of “preclinical” MS (Okuda et al., 2009). A recent retrospective analysis of patients with RIS determined that the presence of asymptomatic spinal cord lesions carries a higher risk of conversion to CIS within 1 year (88% of patients in this category converted), and this risk is independent of the number or location of brain lesions. The utility of treating this group of patients with immunomodulating therapies—or indeed, our ability to predict and time the onset of clinical disease—may have unpredictable ethical implications. Therefore, caution should be exercised in offering brain imaging to asymptomatic individuals until further information about managing this group becomes available.

Predictive Value Of Conventional Mri Metrics In Multiple Sclerosis Progression

Cerebral MRI abnormalities do not always correlate with clinical disease state in an individual patient. The so-called clinical/MRI paradox of MS is well described but not well understood. Some patients with dramatic MRI scans showing a large number of T2 lesions, T1 “black holes,” and atrophy may have quite mild clinical disease, minimal physical disability, and near-normal cognition. Alternatively, some patients with very few MRI lesions may require assistive ambulation devices and have marked cognitive impairment. Possible explanations over the years have included the occurrence of large and obvious lesions in clinically silent areas of the brain, while small lesions eluding detection may target critical spinal cord pathways. Lower field-strength MRI may miss lesions of clinical relevance in the brainstem, cortex, or basal ganglia, while larger lesions may lack functional correlates and reflect merely an increase in tissue water content without impairment of neural networks.

In general, brain T2 lesions show rather weak correlation with short-term disability progression in MS. Combining T2 lesion volume at baseline with whole brain atrophy measures, as well as looking at T2 lesion volume changes in the first 2 years after the diagnosis, improves predictive value of MRI metrics for short and long-term physical and cognitive disability progression. T1 hypointensities correlate strongly with neurological disability, especially in SPMS and PPMS patients (Bitsch et al., 2001; Truyen et al., 1996; Zivadinov et al., 2007). Total T1 hypointense lesion volume, as well as the number of newly formed T2 hyperintense and T1 Gd contrast positive lesions, have been used as secondary or supportive outcome measures in MS-related clinical trials with new therapeutic agents. Although Gd-enhancing lesions predict short-term clinical relapse and indicate continuous disease activity, they are often clinically silent. Taken alone, they do not reliably predict progression of neurological disability. An MRI-based disease severity scale (MRDSS) was developed by Bakshi et al., consisting of T2 hyperintense lesion volume, the ratio of T1 (hypointense) to T2 lesion volume, and whole brain atrophy (normalized whole brain volume). MRDSS showed the largest effect size in differentiating RRMS and SPMS groups, compared to the individual MRI components, and predicted the risk of developing sustained progression 3 years later (Bakshi, 2008).

It is currently not possible to reliably distinguish MS subtype based on conventional MRI metrics. Few differences are appreciated in MRI T2 lesion load and volume between CIS patients and RRMS patients, although patients with benign disease have a lower rate of brain atrophy than patients with early MS (Gauthier, 2009). SPMS patients tend to have greater reduction in brain volume and diffuse changes in normal-appearing white matter (NAWM) (Bakshi, 2008). PPMS patients have fewer and smaller size lesions in the brain and greater numbers of T2 lesions in the spine, coupled with more spinal cord atrophy. Newer MRI techniques may hold promise in understanding specific imaging characteristics of MS subsets and of the clinical/MRI paradox (Fig. 54.10).

Fig. 54.10 Changes in magnetic resonance imaging scans with duration of disease. A-C, Comparison of three scans from patients with different disease duration, indicating the appearance of atrophy and ventricular dilation with time. D, As brain atrophy appears, it is common to observe that the number of gadolinium-enhancing lesions declines.

New Imaging Techniques and Approaches

Magnetic Resonance Spectroscopy

Magnetic resonance spectroscopy (MRS) is a tool that derives MRI signal from multiple metabolites. Many brain metabolite concentrations are not normal in MS patients, change over time, and can be measured with this technique. For example, N-acetylaspartate (NAA) is a marker of neuronal metabolism that is decreased in MS patients, suggesting a loss of neuronal integrity (Fig. 54.11). A high choline (Cho) peak is indicative of an increase in membrane turnover, as can be seen in demyelination and remyelination. Increased creatine (Cr) usually indicates an increase in cell density. Increased myoinositol (mI) indicates an increase in the number of glial cells. MRS gives insight into the biochemical composition of MS lesions and NAWM. In active, newly formed MS plaques, NAA is slightly decreased and Cre, Cho, and mI are increased. Lactate (Lac) peak is also increased, suggesting energy failure. Over time, the initial surge in lipids, Cho, Lac, and Cre may return to near normal. This suggests some degree of recovery. In contrast to active plaques, chronic MS lesions show considerable reduction in NAA, normal glutamate peaks, and elevation in mI. Interestingly, NAWM MRS patterns are quite similar to active acute MS plaques, and these changes can be detected some months before appearance of active and T2 bright lesions. MRS may also show changes in the earliest stages of MS. Patients with CIS have abnormalities in both NAWM and gray matter. Studies showed an early increase in mI signal even when the other metabolites remain normal. MRS measures are more robust in predicting cognitive and physical disability in MS patients in small research trials. (Bonneville, 2002; De Stefano, 1998; Mathiesen, 2006). Clinical utility of this technique, however, is limited owing to inconsistent reproducibility and lack of standardization of this tool among various platforms.

Fig. 54.11 A-B, Areas of confluent demyelination in a severely affected progressive multiple sclerosis patient. Outlines show volume of interest selected for spectroscopy study. C, Spectra obtained with proton spectroscopy in this patient. D, Spectra of normal brain. Note reduced height of N-acetylaspartate (NAA) peak, with resultant reduction in the ratio of NAA to creatine (Cr). CHO, Choline; LA, lactic acid.

Magnetization Transfer Imaging

Magnetization transfer imaging (MTI) measures interactions between protons in free fluids and tissues and protons in bound form (Bakshi, 2008). Injury to the CNS decreases the exchange of mobile protons and can be measured by a change in magnetization transfer ratio (MTR). Low MTR reflects damage to myelin and axons (Fillipi, 2007). Variable MTR reductions have been shown in acute and chronic MS lesions and are also seen in NAWM and gray matter. T2 hyperintense lesions that correspond to T1 black holes on conventional MRI are associated with the most profound decrease in MTR and with greater disability progression (Filippi, 2007). MTR signal shows variability based on disease subtype and is reduced more dramatically in SPMS and PPMS patients compared to RRMS or CIS patients (De Stefano, 2006). Individual MS lesions can be followed with this tool, assessing the degree of demyelination and subsequent remyelination (Chen, 2007). Further, MTR histogram analysis may show differences between clinically defined disease phenotypes. MTR may help differentiate more benign MS cases from more disabling disease (Traboulsee, 2003). MTR “brain maps” are now routinely obtained in larger academic MS centers, although their application as surrogate outcome measures in randomized clinical trials or their clinical utility is still limited.

Diffusion Tensor Imaging

In the white matter of the CNS, the diffusivity of water molecules is much less hindered along the long axis of ordered axons than perpendicular to the long axis, where molecules take a more tortuous route. The diffusivity is anisotropic—it varies with direction of fiber orientation. Diffusion tensor (DT) is a measurement of this process, which is estimated in each MRI voxel, a technique known as DT MRI. This technique allows displaying the orientation of white matter tracts in vivo, reconstruction of three-dimensional trajectories of white-matter tracts, and segmentation of tracts and lesions. Tracking through gray- and white-matter connections allow visualization of gray-matter connections and complex subcortical white-matter tracts that appear most involved in MS. Inflammation, demyelination, and loss of axonal structure increases diffusivity of water molecules in damaged tissues. This is measured by the apparent diffusion coefficient (ADC), which is increased in chronic MS lesions and NAWM, or fractional anisotropy (FA), which is decreased. Significant increase in ADC, as seen in T1 black holes correlates with a higher degree of axonal destruction and demyelination (Filippi, 2001; Werring, 2000). Superacute enhancing lesions sometimes show a temporary increase in ADC. Gray- and white-matter DT changes correlate with clinical disease progression in longitudinal and cross-sectional studies (Rovaris, 2005; Tavazzi, 2007).

High-Field-Strength Mri

High (3T) and ultra-high (7T) MRI scanners greatly improve sensitivity for detecting T2 and Gd-enhancing lesions. Cerebral lesion volume also increases with higher field strength (Bachmann, 2006; Sicotte, 2003), and cortical lesions are easier to detect. Moreover, more patients fulfill the diagnostic criteria for CDMS when studied with 3T or higher MRI because of improved lesion detection in both supra- and infratentorial compartments (Sicotte, 2003). High-field-strength MRI has the potential to improve the quality of MTR, MRS, DTI, and other new imaging techniques. Patients’ safety and comfort remain of concern; patients report greater discomforts such as increase in body temperature and sensory symptoms associated with higher-field-strength studies (Bakshi, 2008).

Measuring Cerebral Volume Loss

Progressive MS-related cerebral atrophy has been documented with various MRI techniques for over a decade. The rate of atrophy is estimated to be between 0.6% and 1.35% per year (Bermel, 2006). Semi-automated (atlas-based) and fully automated (voxel-based) segmentation tools are used in imaging research and clinical trials to assess loss of cerebral volume in MS. The brain parenchymal fraction (BPF), defined as the ratio of brain parenchymal volume to the total volume within the brain surface contour, is used to measure whole brain atrophy (Rudick, 1999). In clinical practice, cerebral volume loss is immediately apparent through progressive enlargement of CSF spaces including lateral ventricles and subarachnoid spaces (gyri and sulci) seen on conventional MRI. Atrophy in MS seems to correlate with the volume of T2 lesions and with changes in normal-appearing white and gray matter. In recent years, topographic differentiation of atrophy within specific cerebral compartments seems to show an important association with cognitive and physical disability in MS. Gray matter is affected by atrophy more profoundly than white matter, and deep gray-matter nuclei are even more susceptible. While cerebral tissue loss in MS is an important factor, our ability to reliably measure it with the least interobserver variability remains suboptimal. Confounding factors include effects of aging, osmotic agents, and even antiinflammatory treatments—all of which can decrease cerebral water content and result in a skewed finding of atrophy progression.

Spinal Cord Imaging

Over 90% of MS patients have spinal cord lesions at some point in their disease course, and 30% of patients presenting with CIS other than transverse myelitis have (asymptomatic) disease in the spinal cord (Dalton, 2003). Lesions are especially common in the cervical cord opposite the C2 or C3 vertebra. They typically involve less than two contiguous segments of the spinal cord and less than half of the transverse diameter of the cord. Spinal cord lesions can be in the form of discrete isolated plaques or manifest as more confluent affected areas, especially in SPMS or PPMS patients. It is less common to have T1 hypointensities in the spinal cord. Usually, lesions in the spinal cord produce neurological symptoms, and cord atrophy has a strong correlation with neurological disability (Losseff, 1996).

Cerebrospinal Fluid Analysis

CSF findings alone neither make nor exclude the diagnosis of MS, but they remain important in atypical clinical syndromes, atypical or nondiagnostic MRI findings, or unusual clinical manifestations such as a course of progressive neurological impairment without history of relapses. CSF does not show any gross abnormalities in MS; it is clear, colorless, and under normal pressure. Cell counts are typically normal but may be slightly elevated in 15% to 20% of patients. The predominant cells are T lymphocytes. Significant pleocytosis should raise suspicion of another etiology. Although CSF protein is also usually normal, it is common to see mild elevations of albumin in 20% to 30% of MS patients. Determining the presence of OCBs is the most important diagnostic test. Isoelectric focusing followed by immunoblot is the preferred test for OCB detection. These bands represent excess antibody produced by one or more clones of plasma cells. The pattern of banding remains relatively stable in an individual patient throughout the course of the disease. However, 20% to 30% of patients with confirmed CDMS do not have OCBs at any given point in time. Presence of OCBs in a patient with CIS predicts a higher rate of conversion to CDMS and may be used to convince a reluctant patient to start disease-modifying therapy. Myelin basic protein (MBP) in the CSF is a marker of tissue damage and has been used as a measure of CNS myelin breakdown. The levels of the protein may be quite increased in MS patients, but the specificity of these findings is not known. An abnormality in CSF IgG production (as measured by the IgG index, or as a percentage of total protein or albumin) is found in over 90% of patients with confirmed MS (Table 54.5).

Evoked Potentials

EPs are CNS electrical events generated by peripheral stimulation of a sensory organ and are useful to determine abnormal function that may be clinically unapparent. Detecting a subclinical lesion at a site remote from the region of clinical dysfunction supports the multifocal disease portion of the diagnostic criteria for MS. The three most commonly used EPs are VEPs, somatosensory evoked potentials (SSEPs), and brainstem auditory-evoked responses (BAER). The role of EPs in MS was recently reviewed, and only VEPs were thought to be useful to determine increased risk for MS (Gronseth, 2000). Using pattern shift VEPs, abnormalities (P100 wave prolongation) are detected in over 90% of patients with a history of optic neuritis, even in a setting of complete restoration of vision (Table 54.6).

Optic Coherence Tomography

Optic coherence tomography (OCT) can measure the retinal nerve fiber layer (RNFL) using a process analogous to ultrasound imaging, with light instead of sound. The RNFL is devoid of myelin and contains axons that converge to form the optic nerve. An acute lesion of optic neuritis can cause RNFL atrophy by acute axonal transection and consequently by retrograde degeneration (Rashid, 2008). OCT can be used to noninvasively quantify axonal damage following an optic neuritis event. There is a correlation between optic nerve atrophy and RNFL thinning, suggesting that OCT may be useful in clinical trials aiming at neuroprotection.

Recurrent Optic Neuropathy

There are patients whose entire clinical illness is confined to the optic nerves. These individuals may have sequential involvement of one nerve, then the other, or they may have simultaneous bilateral vision loss, a state that is quite uncommon in classic MS. In some instances, cranial MRI demonstrates scattered intracerebral lesions, or a CSF examination shows OCBs, attesting to some degree of dissemination of the lesions. Children are more likely than adults to have recurrent or simultaneous optic neuropathy. Rarely, slowly progressive optic neuropathy similar to that seen with optic nerve sheath tumors such as meningioma occurs. Sarcoidosis is commonly a diagnostic consideration in patients with bilateral optic neuropathy.

Devic Disease (Neuromyelitis Optica)

A combination of bilateral optic neuropathy and myelopathy characterize Devic disease, or neuromyelitis optica, which many authorities now classify as a separate entity rather than a variant of MS. The myelopathy tends to be more severe than typically occurs with MS and carries less likelihood of recovery. The neuropathological features at autopsy are those of a much more severe necrotic lesion of the cord. In some patients, the optic neuropathy and the myelopathy occur at the same time; in others one or the other component is delayed. Antibody to the aquaporin-4 water channel is present at high frequency in patients with the clinical characteristics of Devic disease (Weinshenker et al., 2006). Further recent work has expanded the clinical spectrum of neuromyelitis optica beyond the typical presentation (Wingerchuk et al., 2006).

Slowly Progressive Myelopathy

A syndrome of slowly progressive spinal cord dysfunction can present a major diagnostic challenge. If there are no sensory signs or symptoms, the entity known as primary lateral sclerosis, a form of motor neuron disease, may be the cause. HTLV-1 infection, vitamin B₁₂ deficiency, and HIV infection all can be excluded by appropriate testing. Spinal dural arteriovenous fistula can cause a steadily or stepwise progressive myelopathy, usually in the mid- to lower spinal segments. Adrenomyeloneuropathy, an X-linked recessive disorder, also should be considered.

A number of patients do not fit into these categories, and their spinal MRI results are repeatedly negative. VEPs, CSF OCBs, and MRI of the head show no sign of demyelination elsewhere. No firm diagnosis is possible. Minor clues that MS is present may be furnished by a Lhermitte sign that has come and gone or by undue sensitivity to elevated temperature. Progressive myelopathy caused by MS is part of the PPMS group and carries the poor prognosis typical of it.

Acute Tumor-like Multiple Sclerosis (Tumefactive Multiple Sclerosis)

Some patients with demyelinating disease present with a large acute lesion of one hemisphere (Fig. 54.12) or rarely in other locations such as the spinal cord. Mass effect may occur, with compression of the lateral ventricle and shift across the midline. The clinical abnormalities in such patients are variable. They may be slight, even in a patient with a massive lesion, whereas confusion, hemiparesis, or neglect syndrome may be seen in another patient with a lesion that appears no different. Much of the T2 bright lesion volume is often due to edema and may be rapidly responsive to corticosteroids. (This change with corticosteroids also may occur with glioma or CNS lymphoma and is therefore not a useful diagnostic criterion.) Biopsy may be necessary to establish a correct diagnosis.

Fig. 54.12 A, Magnetic resonance imaging scan showing a large white-matter lesion, initially without mass effect, in a young woman with right hemiplegia developing over 5 days. B, Marked increase in size of lesion, with associated edema and mass effect 1 month later while on corticosteroids and at a time when clinical deficits were improving. Biopsy showed evidence of demyelination.

In one series of 31 patients, the prognosis was good, most patients recovered well clinically, and their lesion volume rapidly cleared. In 24, the demyelinating lesion was solitary, whereas in the others, one or more satellite nodules existed. Six of the patients were older than age 57. In follow-up, 28 of the patients did not develop additional evidence of demyelinating activity during a period from 9 months to 12 years. Other authorities have reported a higher rate of recurrent disease, in particular a conversion to more ordinary types of MS, both clinically and by MRI scan criteria.

Symptomatic Treatment

Treatment Strategies

For some patients, MS is a disease with one or two acute neurological episodes with no further evidence of disease activity. In others, it is a chronic relapsing or progressive disease with an unpredictable clinical course that generally spans 20 to 30 years, during which time neurological disability accumulates. Treatment of MS, as with other diseases, is now based on the results of prospective well-controlled clinical trials. Most of these trials have been designed to establish efficacy but have not followed patients in controlled fashion for longer than 2 or 3 years, and have therefore only given hints about long-term results of treatment. Long-term placebo-controlled trials are not ethically possible once efficacy is established. Patients in clinical practice may differ markedly from those who have been treated in clinical trials, yet therapeutic decisions must be made, and these trials provide best evidence. Table 54.7 outlines one of the possible treatment paradigms.

Table 54.7 Multiple Sclerosis Treatment Strategies

IM, Intramuscular; IV, intravenous; MRI, magnetic resonance imaging; SQ, subcutaneous.

Treatment of Acute Attacks

Acute attacks are typically treated with corticosteroids. Indications for treatment of a relapse include functionally disabling symptoms with objective evidence of neurological impairment. Thus, mild sensory attacks are typically not treated. Treatment with short courses of IV methylprednisolone, 500 to 1000 mg daily for 3 to 5 days, with or without a short prednisone taper, has commonly been used. A 1993 randomized therapeutic trial in ON demonstrated that patients treated with oral prednisone alone were more likely to suffer recurrent episodes of ON compared with those treated with IV methylprednisolone followed by oral prednisone. Furthermore, definite MS developed in 7.5% of the IV methylprednisolone group, 14.7% of the oral prednisone group, and 16.7% of the placebo group over a 2-year period. Development of disability, even when the diagnosis of MS had been made, was very rare, reemphasizing the need for follow-up periods of decades and the sometimes benign nature of MS presenting with ON. These data support the use of high-dose IV methylprednisolone for acute MS attacks. High-dose IV methylprednisolone is accompanied by relatively few side effects in most patients, although psychiatric changes, predilection for infections, gastrointestinal disturbances, anaphylactoid reactions, and an increased incidence of fractures may occur. The immunological mechanisms of high-dose corticosteroids include reduction of CD4⁺ cells and decrease in cytokine release from lymphocytes, including TNF, IFN-γ, and decreased MHC complex class II expression. Corticosteroids have been shown to decrease IgG synthesis in the CNS. Intravenous methylprednisolone may decrease the entry of cells into the brain.

Very high-dose oral methylprednisolone (3676 mg) was studied and resulted in no serious adverse events; there was no statistical difference between the clinical outcomes in oral versus IV formulation treatment groups.

Disease-Modifying Treatments

Treatment of Progressive Disease

A number of clinical trials in the past investigated treatments for chronic progressive MS. Unfortunately, these were done prior to publication of the disease course definitions listed in this chapter. Therefore, they tended to include a mixture of patient types, which could potentially obscure a beneficial effect of a therapy on a particular disease type. This is most important in considering trials that mix both primary and secondary progressive patients because the latter may be more amenable to therapy than the former. Furthermore, some patients with RRMS with stepwise progression might have been included in these trials.

Most past clinical trials in progressive MS have used chemotherapeutic (cytostatic) agents. These all have the disadvantage of producing generalized immunosuppression and have considerable potential systemic toxicity. In addition, the toxic and long-term residual effects of these agents tend to limit or preclude chronic or repetitive dosing, further restricting their usefulness.

The most successful trial for progressing patients was performed in Europe using the chemotherapeutic agent, mitoxantrone (Novantrone), which produced a reduction in the progression of disability and a considerable decrease in relapse rate in patients with worsening RRMS, PPMS, and SPMS over a 2-year period (Hartung et al., 2002). Improvement in MRI measures of disease was also demonstrated. At 3 years, an appreciable benefit was still evident, even though the dosing had been stopped at 2 years. The toxicity profile was acceptable, but the agent has dose-related cardiac toxicity that limits lifetime dosing to 140 mg/m². Other side effects include nausea, alopecia, and neutropenia. Secondary leukemias have been uncommonly reported.

A number of studies of cyclophosphamide treatment of progressive MS have been conducted, with some suggesting a benefit and others not. Comparison between trials is always hazardous, and various induction protocols have been used, some with the addition of steroids or plasmapheresis. Nevertheless, many anecdotal reports of success have led to use of this agent in rapidly progressive cases.

Azathioprine (Imuran) has been studied in several clinical trials. A modest effect on relapses has been shown but no convincing effect on progressive disease. Systemic toxicity is an important consideration with this agent, although less than with cyclophosphamide.

Methotrexate has been used to treat other autoimmune diseases (e.g., rheumatoid arthritis). A study of patients with progressive MS was conducted using a dose regimen similar to rheumatoid arthritis treatment: 7.5 mg orally per week (Goodkin et al., 1995). This double-blind placebo-controlled study revealed a modest beneficial effect of methotrexate on a composite scale of upper-extremity function but showed no effect on ambulation. Toxicity was minimal. Methotrexate is not commonly used in MS.

Chronic administration of corticosteroids in MS has not been demonstrated to be effective and has considerable potential morbidity. Some have advocated the periodic use of pulses of steroids in secondary progressive disease. One approach involved treatment of progressive MS patients with methylprednisolone, 500 mg IV, every other month for up to 2 years. There was no convincing effect of this treatment approach. Another study administered 5-day pulses of 1 gram daily methylprednisolone every 4 months for 3 years, and then every 6 months for 2 years in RRMS patients. Compared to patients who received IV methylprednisolone only for acute exacerbations, the more frequently treated group had less brain atrophy, fewer black holes on MRI, and less disability as measured by EDSS (Miller et al., 2000). There was no treatment effect on annual relapse rate or T2 lesion load.

IVIG, studied in a large European trial of SPMS, demonstrated no beneficial effect on any clinical outcome and little effect on MRI metrics (Hommes et al., 2004).

There are several anecdotal reports of the use of bone marrow/stem cell transplantation in MS, but no controlled data. Absent controlled trials, single case studies and even multiple cases with an international registry are unlikely to yield convincing data. The procedure often produces serious morbidity, and thus far, rather worrisome mortality in MS patients. A multinational effort is now underway to establish protocols and procedures for use of bone marrow/stem cell transplantation in selected MS patients. This treatment method should only be carried out in a clinical trial setting.

Interferon beta-1b has been used in two studies of SPMS. The first was a European multicenter controlled trial of interferon beta-1b in 718 patients with SPMS treated with either interferon beta-1b, 8 MIU subcutaneously every other day, or placebo. The study was planned for 3 years but was stopped after enrollees had completed 2 years because an interim analysis demonstrated efficacy (Kappos et al., 1998). Analysis of this study revealed that patients treated with interferon beta-1b had significant delays in progression of disability. On average, this delay amounted to 9 to 12 months. Similarly, patients receiving interferon beta-1b had a delay in time to becoming wheelchair bound of about 9 months, compared to placebo patients. Similar results were seen for relapse rate reduction, MRI analyses, hospitalizations, and steroid use. A second study of interferon beta-1b was performed in North America. In this study, SPMS patients were treated with placebo, interferon beta-1b, 8 MIU every other day, or interferon beta-1b 5 MIU/m² every other day (this group had an average dose of 9.6 MIU). As opposed to the European study, in this study showed no effect on the primary outcome assessment of reduction in time to worsening disability was found (Kappos et al., 2004). Other outcome measures such as reduction in relapse rate, MRI activity, and lesion load favored the treated groups. Further analysis of these two studies suggested that the marked difference in their outcomes might relate in part to differences in the degree of MS disease activity between the groups. Although the entrance criteria were very similar, the actual groups differed in several ways (Kappos et al., 2004). This highlights the difficulty in trying to compare results across studies, even those that attempt to evaluate the same problem with the same type of subject.

Another European/Canadian study (King et al., 2001) of high doses of interferon beta-1a (Rebif) failed to achieve its primary outcome measure (time to confirmed worsening of EDSS). Other outcome measures such as relapse rate and MRI measures were favorable.

A study with interferon beta-1a (Avonex) dosed at 60 mg weekly (twice the standard dose of this agent in RRMS) succeeded in achieving its primary outcome measure of a significant benefit as measured by the MSFC (Cohen et al., 2002). There was also benefit on relapse rate and MRI metrics. However, no benefit was seen on time to disability as measured by EDSS; therefore, this agent did not receive regulatory approval to treat SPMS.

In summary, there are two reasonably successful treatments of SPMS, mitoxantrone and perhaps interferon beta. There are no available treatments for PPMS. Little testing has been done on PRMS, but there were some PRMS patients in the mitoxantrone study, and thus it would be appropriate to use that agent in PRMS.

History

Acute disseminated encephalomyelitis (ADEM) is a demyelinating syndrome that often occurs in association with an immunization or vaccination (postvaccination encephalomyelitis) or systemic viral infection (parainfectious encephalomyelitis). This disorder is almost always monophasic. Recurrent cases have been described, but their nature and distinction from MS is unclear. ADEM is characterized pathologically by perivascular inflammation, edema, and demyelination within the CNS, and clinically by rapid development of focal or multifocal neurological dysfunction. The most precise clinical and pathological observations regarding ADEM are derived from case studies in which there has been a close link between the specific viral infection or vaccine and the syndrome. The syndromes arising after acute measles infection or rabies vaccine administration can be considered the prototypes of the illness (Box 54.4).

Laboratory Features

The hallmark lesions of ADEM are perivascular inflammation and surrounding demyelination within the CNS (Figs. 54.13 and 54.14). Vessel necrosis is frequently observed. The demyelinating aspect may be minimal or widespread, with coalescence of the multiple lesions. Some meningeal reaction may also be apparent. Reports of MRI studies, largely from apparent sporadic cases of ADEM, describe multifocal CNS lesions initially indistinguishable from those observed in MS. A feature of ADEM is that the majority of the T2 lesions enhance, suggesting they were of recent onset, consistent with a monophasic illness. Unlike MS, in ADEM, after several weeks, lesions show at least partial resolution without the appearance of new lesions. In some cases lesions can persist. The usual CSF formula is normal pressure, little or modest increase in cell count (<100 cells/mL), and a moderate increase in protein. Well-documented cases exist with totally normal CSF. Cases with high cell counts, including some polymorphonuclear cells and high protein values, occur and seemingly reflect a more necrotizing disease process. The high counts usually return to normal within a few days. The CSF Ig content is not usually increased, and OCB patterns occur less commonly than in MS.

Fig. 54.13 Lesion of acute disseminated encephalomyelitis showing demyelination and macrophage infiltration (arrows) (hematoxylin and eosin stain; bar = 50 µm).

(Courtesy Dr. S. Carpenter.)

Fig. 54.14 Perivenular zone of demyelination in brainstem of patient with acute disseminated encephalomyelitis (Heidenhain Woelke stain; bar = 100 µm).

The diagnosis of ADEM can usually be made with confidence in the setting of a clear-cut antecedent event strongly associated with the disorder, such as measles infection or vaccination. The occurrence of an acute focal or multifocal CNS syndrome subsequent to a nonspecific viral illness creates a wider differential diagnosis, depending on the age of the patient and the clinical manifestations of the disease. The following would be included in this differential diagnosis:

Treatment

The current favored therapy for ADEM is high-dose corticosteroids, although no controlled clinical studies have been conducted. Immunosuppressants such as cyclophosphamide have been used in refractory cases, as has plasmapheresis. Data on the various other immunotherapies described in this chapter are lacking.

Acute Hemorrhagic Leukoencephalitis

Acute hemorrhagic leukoencephalitis is a rare entity that represents a hyperacute form of ADEM, in a parallel fashion to the hyperacute forms of EAE. The most frequent antecedent history is that of an upper respiratory infection. Given the nonspecific nature of the antecedent event and the lack of a specific diagnostic clinical laboratory test, the exact incidence and full clinical spectrum of the disorder can only be estimated and are based largely on descriptions of autopsy-proven cases. The clinical manifestations mimic ADEM but are more abrupt in their development and more severe. They include focal or multifocal signs, seizures, and obtundation. Relapse following initial recovery has been described. Fever is common. The CSF usually demonstrates increased pressure, protein, and both white and red cells. Peripheral white blood cell count is usually increased. MRI usually shows large and confluent FLAIR bright lesions with petechial hemorrhages and may yield additional information on lesion evolution on subsequent scans.

The differential diagnosis of this syndrome includes entities that present as rapidly evolving focal cerebral disorders with fever and obtundation, such as brain abscess and herpes simplex encephalitis, in addition to those syndromes considered in the section on ADEM.

Chronic or Recurrent Forms of Postinfectious and Postvaccination Encephalomyelitis

The clinical hallmark of ADEM occurring after measles infection or rabies vaccination is its course, so relapses occurring during the recovery phase could well represent physiological conduction blocks rather than true reactivation of immune-mediated mechanisms. However, recurrent encephalomyelitis cases have been described, particularly in pediatric age groups. Recurrent cases in adults, if they occur, would be difficult to distinguish from MS. The existence of chronic or recurrent encephalomyelitis cases might be considered to be parallel to the peripheral neuritis syndromes—acute nonrelapsing Guillain-Barré syndrome being the prototype—but in which chronic and recurrent cases are reported. The basis of these latter syndromes and their relation to classic Guillain-Barré syndrome remain unresolved. Animal models of encephalomyelitis (EAE) and peripheral neuritis (experimental allergic neuritis) can both be induced in relapsing form if one selects animals of critical age and specific genetic background.

Combined Central and Peripheral Demyelinating Disease

The existence of combined central and peripheral demyelinating disease as a postinfectious disorder or as a complication after administration of vaccines not known to contain PNS and CNS tissue remains highly questionable based on available epidemiological evidence. Cases of combined PNS and CNS involvement after swine influenza vaccination were reported, but some of these cases had subacute progressive courses. To date, no convincing in vitro immune sensitization to either viral or neural antigens has been shown in such cases. Reports of combined central and peripheral demyelination syndromes do exist in which onion bulb formation in the PNS is demonstrated, indicating recurrent demyelination and remyelination. Some, although not all, of these patients have clinical features consistent with MS. Whether such combined demyelination represents a chance occurrence of two processes or a distinct disease is unresolved.

Site-Restricted Forms of Postinfectious Demyelinating Disorders