European Journal of Neurology
Volume 8 Issue 6 Page 659 - November 2001
A. Chaudhuri & P. O. Behan
We report here 39 cases in which
definite multiple sclerosis (MS) was precipitated or exacerbated by specific
hyperextension-hyperflexion cervical cord trauma. The worsening or onset
of the symptomatic disease bore a striking temporal relationship to the
focal injury. Our data suggests that central nervous system (CNS)-specific
acute physical trauma such as cervical cord hyperextension-hyperflexion
injury may aggravate latent clinical symptoms in MS. The deterioration
of MS bore no direct relationship with the severity of neck injury. Possible
pathogenic mechanisms of focal CNS-specific trauma aggravating the course
of asymptomatic or benign MS are discussed. This may have implications
in improving our understanding of the factors that may modify the clinical
course of MS.
Despite many years of intensive research, the precise cause of multiple sclerosis (MS) is still unknown. Recently, researchers have drawn attention to an underlying axonal pathology (Trapp et al., 1998) and the role of nitric oxide (McDonald, 2000) in the demyelinating injury. Focal trauma increases nitric oxide in the brain (Cobbs et al., 1997) and spinal cord (Yick et al., 1998). Most researchers generally accept that certain factors do modify the clinical course for MS by precipitating or aggravating the clinical symptoms. These include infections (Sibley et al., 1985), pregnancy (Confavreux et al., 1998), electrical injuries (Bamford et al., 1981), and penetrating and surgical wounds of the brain (Poser, 1994a). Vaccinations (Sibley and Foley, 1965) and acute emotional psychological stress (Franklin et al., 1988) may also adversely affect the course of MS although certain vaccinations may be safe in MS patients (Ascherio et al., 2001; Confraveux et al., 2001). It is also recognized that breakdown in the blood-brain barrier (BBB) is an early and obligatory event in the development of acute MS lesions (Broman, 1947; Poser, 1994b). As focal hyperextension-hyperflexion injuries to the cervical spinal cord have been shown to cause a severe disruption of the BBB both locally and generally in the brain in experimental animals (Ommaya et al., 1964; Goodman et al., 1976; Kobrine et al., 1976; Domer et al., 1979) and to affect cerebral metabolism in man (Otte et al., 1995), serious consideration therefore must be given to this type of injury for its effect on MS.
In recent years, we have documented 39 patients who developed symptomatic MS or in whom a stable disease with minimal disability was converted to a rapidly progressive form within some days to weeks after an acute hyperextension-hyperflexion injury to the cervical spinal cord. These cases are described here in order to focus attention on this unique occurrence. As the allelic variations of the APOE gene have been reported to influence the clinical course of MS (Chapman et al., 1999; Evangelou et al., 1999) and the recovery after head injury (Roses and Saunders, 1995), we also carried out APOE genotype determination in this group.
All patients in this series presented within 3 months from the date of their acute cervical cord injury (hyperextension-hyperflexion injury in automobile accidents and hyperextension-rotational injury in industrial or environmental accidents). The diagnosis of definite MS in all the cases was confirmed by the neurologists prior to their referral to one of us (POB). There were a total of 39 cases, 24 of which were of new onset. These latter cases had new disease, without any history of neurological symptoms and were previously in excellent health. There were 12 males and 12 females, whose ages ranged from 19 to 55 years with a mean age of 32 years. They were all in excellent physical health previously. The onset of the symptoms occurred within a range of 12 h-12 weeks post-trauma with the maximum number of new cases reaching their peak between 2 and 3 weeks. Their expanded disability status scale (EDSS) scores ranged from 1.5 to 8.5 with a median score of 5.0. A further 15 cases, three male and 12 female, were of mild MS which rapidly accelerated to a progressive form following their injury. Their ages ranged from 21 to 51 years with a mean of 38.6 years. Their pre-trauma EDSS score had a range of 1-3 with a mode of 1, whilst their post-trauma EDSS at the time of examination were between 3 and 8.5 with a median score of 5.5. The worsening of their MS, i.e. the onset of new symptoms occurred between 1 and 12 weeks post-trauma with a mean maximum incidence of 1-2 weeks.
The severity of soft tissue injury was assessed retrospectively on an arbitrary scale of 1-15 (Table 3). There was no apparent correlation between the individual severity of injury as measured in this arbitrary scale and the subsequent deterioration of MS symptoms. In this series, none of the cases had any cervical vertebral fracture, dislocation or spinal cord compression.
Table 1 Patient data
Table 2 Clinical characteristics of patients
Table 3 Characteristics of neck injury
Table 4 APOE studies
Table 5 Possible mechanisms as to how specific focal trauma can aggravate multiple sclerosis
Results are presented in Table 4. APOE genotyping was determined from the samples of buccal smears as previously described (Ilveskoski et al., 1998). Only 4 of the 27 patients for whom APOE genotypes were determined possessed an E4 allele and these were all heterozygous. The E4 frequency of the patients is therefore 0.07 compared with a figure for the general population of 0.15. There was thus no over-representation of the APOE-4 allele in our group of patients with exacerbation of MS.
The role of physical trauma on MS has been debated for a number of years. The current consensus appears to be that whilst general, non-specific physical trauma may not influence the course of MS adversely, the role of central nervous system (CNS)-specific moderately severe physical trauma in modifying the clinical course of MS is still unknown. In this case series, all the patients had developed new symptoms of MS within 3 months following the cervical spine hyperextension-hyperflexion injury. The 24 patients with the new onset of MS had been in perfect physical health with no evidence of any disturbances within the CNS. Fifteen cases were considered to have previous MS symptoms but few had any major disability (pre-trauma EDSS range 1-3, median 1.0) prior to their acute cervical cord trauma. The sex distribution of the latter group was skewed towards the female patients (M : F=1 : 4). We do not have any satisfactory explanation for this other than by chance.
It is important to stress that the trauma was of a uniform type in all cases, i.e. an acute hyperextension-hyperflexion focal injury to the cervical spinal cord. The majority of cases occurred in motor car accidents but some followed accidental injuries in the work environment. As seen from Tables 1 and 2, the onset of symptoms occurred as early as 12 h with the maximum number of cases reaching their peak between 2 and 3 weeks following the specific trauma. The clinical course of MS was generally more rapid and progressive than usual in these cases. There is some suggestion of an association of APOE-4 allele with a more rapidly progressive course in patients with MS (Chapman et al., 1999; Evangelou et al., 1999). However, the fact that the majority of our patients were APOE-4 negative might suggest that they should have had a more benign course. It is noteworthy that the first post-trauma MS symptoms (Table 2) in the majority of these patients localized to lower brain stem (INO, oscillopsia) and cervical spinal cord (e.g. arm paraesthesia, motor weakness of both legs). The onset of oscillopsia was dramatic in all the four cases after cervical injury. The severity of the soft tissue injury was mild to moderate in the vast majority of cases.
There is existing literature supporting the concept that specific CNS trauma may precipitate or aggravate MS. In a personal study of 255 cases, Keschner (1950) had observed that specific focal trauma had to be to the skull or vertebral column aggravates MS symptoms with a close temporal relationship. Lord Brain studied 17 patients in whom the clinical features showed the co-existence of cervical spondylitic myelopathy and MS (Brain and Wilkinson, 1957). Two of these patients died and at autopsy plaques were present throughout the cord and brain, the most conspicuous demyelinating lesions being found in the cervical cord in the region of the spondylitic compression. Oppenheimer (1978) studied the spinal cords at autopsy of 18 patients with MS. He paid particular attention to the cervical enlargement, finding that demyelination occurred in the cervical bulb twice as commonly as in other parts of the spinal cord axis, with a preponderance of fan-shaped lesions in the lateral columns. He attributed this to mechanical stresses communicated to the cord via the denticulate ligaments during the flexion of the spine and postulated that these repetitive physical stresses caused the breakdown of the BBB, providing the nidus for demyelination. Both McAlpine (1946) and Miller (1964) had reported their experience of cases where MS exacerbations occurred after specific physical trauma. Gonsette in examining the brains of patients with MS who had been subjected to stereotactic neurosurgery, found a direct anatomical relationship between new fresh plaques and the path of the trocar (Gonsette et al., 1966). He emphasized that these plaques were fresh and inferred that mechanical destruction of the cerebral parenchyma led to a breakdown of the BBB, thus allowing putative myelinotoxic substances egress into the brain to form new plaques. His argument is supported by the fact that demyelination was not demonstrated from the same techniques used in other patients who did not have MS. This finding was later confirmed by Reichert et al. (1975). Electrical injuries in animals and man also cause breakdown of the BBB and there is evidence to support that electrical injuries induce the symptoms of new disease or aggravate existing MS. One of Millers patients had developed sensory symptoms within 6 days of electrical injury in the same limb only to be followed later by the development of clinically definite MS. Electrical injury was associated with a statistically significant risk of MS exacerbations in Sibleys prospective study of non-specific physical trauma and MS (Sibley et al., 1991). Electro-convulsive therapy (ECT) increases the permeability of the BBB and dramatic clinical deterioration of symptoms have been reported in 20% of all MS patients receiving ECT for depression (Mattingly et al., 1992). Furthermore, widespread spinal cord demyelination has been found to occur in patients after electrocution (Levine et al., 1975) and lightning injuries (Davidson and Deck, 1988).
The critical role of changes in BBB influencing the clinical course of MS was evident from a series of clinical research carried out in Oxford in the 1950s. Initially, patients with tuberculous meningitis were treated with PPD intrathecally with the aim of creating a sterile meningitis and thereby increase, because of the resultant breakdown of BBB, the delivery of the systemic antituberculous drugs within the CNS. The Oxford workers noticed an apparent success with this treatment (intrathecal PPD) in MS patients who also suffered from tuberculous meninigitis (Smith et al., 1955). This formed the basis of a controlled trial of intrathecal tuberculin (PPD) in MS (Kelly and Jellinek, 1961). However, MS patients in this and other trials treated with intrathecal tuberculin did not improve, and on the contrary, deteriorated both in their physical symptoms and also in serial psychometric tests (Tables 3 and 4).
Subsequent research has clearly established that an abnormal BBB plays a critical role in the initiation and progression of demyelination (Broman, 1947; Poser, 1986, 1994b). The effect of this increased vessel permeability may be more generalized as it is known that retinal vessels may also be leaky during an attack of MS (Poser, 1986; Lightman et al., 1987; Gay and Esiri, 1991). Our hypothesis is that acute hyperextension-hyperflexion injuries of the neck will at the very least produce a local breakdown of the BBB as in animal studies, including non-human primates, massive breakdown of the BBB not only of the cord itself but also of the brain follows experimental induction of whiplash injuries (Ommaya et al., 1964; Goodman et al., 1976; Kobrine et al., 1976; Domer et al., 1979). Ommaya and others had shown by means of injecting sodium fluorescein that BBB alterations occurred at a distance from the site of trauma. In their study, a blow to the occipital area altered the BBB in the medulla and in the cervical spinal cord when the monkey did not wear a collar (Ommaya et al., 1964). Korbines group observed that when rhesus monkeys were exposed to moderately severe trauma to the cervical spinal cord, extravasation of Evans blue dye and radio-iodinated human serum albumen from small cerebral vessels was observed (Kobrine et al., 1976). Domer et al. (1979) injected intravenous technetium (Tc99m) pertechnate to show a significant increase in BBB permeability in the brains of animals subjected to whiplash injury that had not been subjected to head trauma. These experiments further demonstrated that the increased permeability occurred not only in the cervical cord but also throughout the brain explaining why in some of our cases lesions were found in the occipital lobes and periventricular regions. The only prospective epidemiological study showing lack of significance of physical trauma and MS (Sibley et al., 1991) did not address the specific question of acute, moderately severe, cervical cord trauma (as in all of our cases) and were not sufficiently powerful to have a good chance of producing a significant result if, indeed, there were a casual connection between CNS-specific focal trauma (e.g. acute cervical cord hyperextension-hyperflexion injury) and MS.
There are many instances of pathologically verified lesions within the brain where new MS plaques were observed surrounding the specific areas where BBB had broken down. This was seen in the post-mortem material in patients with congophilic angiopathy (Heffner et al., 1976), a condition where amyloid is deposited at the sites of BBB breakdown and also in patients dying after sterotactic brain surgery (Gonsette et al., 1966; Reichert et al., 1975) and open brain surgery (Russell, 1946). It is not impossible, therefore, that any external factor that can influence the integrity of BBB of an individual with a diathesis to develop MS will have the potential to trigger the disease symptoms. Other than the breakdown of BBB, CNS-specific focal trauma may also induce neuronal nitric acid synthase (Grünewald and Beal, 1999) which may act alone or synergistically with cytokines to produce neuronal injury. Mechanisms by which CNS-specific trauma like whiplash injury may unmask MS are probably multifactorial (Table 5) and these are not mutually exclusive. The cervical region is the commonest site of spinal cord involvement in MS and spinal cord atrophy provides the best correlate of the degree of disability as measured by Kurtzkes EDSS scale (Loseff et al., 1996). Thus, it would only seem logical that rapid progression of disability was a direct consequence of the cervical cord disease in our cases.
A question that may be asked is whether the association of specific trauma in this series of MS patients reflected a recall bias on the part of the patients who were involved in accidents. However, because of the nature of injury sustained (acute cervical hyperextension-hyperflexion) and the close temporal relationship of the symptoms (often days and weeks), it seems improbable that the recall bias was the only explanation for this association. We could not identify either the specific mechanism(s) that could cause clinical deterioration of MS cases after focal trauma or any individual characteristics of our patients that made them more vulnerable to the effect of acute cervical cord injury. We accept the limitation of our study in that it is observational and based on personal case series. However, a randomized control study of CNS-specific acute, focal trauma in MS would be impractical and ethically impossible.
In our paper, we have sought to draw the attention of the readers to the role of certain CNS-specific focal trauma (cervical cord injury, stereotactic brain surgery like thalamotomy and electrical injury) in precipitating the symptoms of undeclared MS and adversely affecting the course of benign MS. Like infection, which will trigger MS symptoms only in a proportion of patients [10% (McAlpine et al., 1965)-48% (Sibley and Foley, 1965)], cervical cord hyperextension-hyperflexion injury is likely to unmask or worsen the natural course of MS in a subgroup of affected patients with an underlying diathesis. This may be important because the prevalence of asymptomatic (silent) MS has been estimated to be about 25% of that diagnosed in vivo (Engell, 1989).
We must make it clear that we do not propose physical trauma in any form causes MS per se. Physiologically, CNS-specific trauma produces focal breaches in the BBB and induces metabolic changes by activating the stress response. In addition, focal trauma also enhances the expression of nitric oxide synthase in the CNS microvasculature (Cobbs et al., 1997). In susceptible individuals, these effects might unleash critical changes in the levels of pro-inflammatory cytokines and nitric oxide, thus triggering MS symptoms ab initio or aggravating symptoms of pre-existing latent disease. The mechanism of MS is still unknown and research must focus on the role of traumatic breakdown of BBB, stress and nitric oxide pathways in the MS symptom exacerbation for a better understanding of this common, disabling neurological disorder.
We thank James AR Nicoll (Department of Neuropathology, University of Glasgow) for carrying out the APOE genotyping. AC is supported by the Barclay Research Trust held at the University of Glasgow.
1 Ascherio A, Zhang
SM, Hernan Medium Aet al. (2001). Hepatitis B vaccination and the risk
of multiple sclerosis. N Engl J Med 344:327-332.
2 Bamford CR, Sibley WA, Thies Cet al. (1981). Trauma as an etiologic and aggravating factor in multiple sclerosis. Neurol 31:1229-1234.
3 Brain R&Wilkinson M (1957). The association of cervical spondylosis and disseminated sclerosis. Brain 80:456-478.
4 Broman T (1947). Supravital analysis of disorders in the cerebral vascular permeability. Two cases of multiple sclerosis. Acta Psychiat Scand Suppl 46:58-71.
5 Chapman J, Sylatiev C, Nisipeanu P, Korczyn AD (1999). Preliminary observations on APOE epsilon4 allelle and progression of disability in multiple sclerosis. Arch Neurol 56:1484-1487.
6 Cobbs CS, Fenoy A, Bredt DS, Noble LJ (1997). Expression of nitric oxide synthase in the cerebral microvasculature after traumatic brain injury in the rat. Brain Res 751:336-338.
7 Confavreux C, Hutchinson M, Hours MMet al. (1998). Rate of pregnancy related relapse in multiple sclerosis. N Engl J Med 339:285-291.
8 Confraveux C, Suissa S, Saddier Pet al. (2001). Vaccinations and the risk of relapse in multiple sclerosis. N Engl J Med 344:319-326.
9 Davidson GS&Deck JH (1988). Delayed myelopathy following lightning strike: a demyelinating process. Acta Neuropathol 77:104-108.
10 Domer FR, Lin YK, Chandran KB, Krieger KW (1979). Effect of hyperextension-hyperflexion (whiplash) on the function of the blood-brain barrier of Rhesus monkeys. Exp Neurol 63:304-310.
11 Engell T (1989). A clinical patho-anatomical study of clinically silent multiple sclerosis. Acta Neurol Scand 79:428-430.
12 Evangelou N, Jackson M, Beeson D, Palace J (1999). Association of APOE 4 allelle with disease activity in multiple sclerosis. J Neurol Neurosurg Psychiat 67:203-205.
13 Franklin GM, Nelson LM, Heaton RK, Burks JS, Thompson DS (1988). Stress and its relationship to acute exacerbations in multiple sclerosis. J Neurol Rehab 2:7-11.
14 Gay D&Esiri M (1991). Blood- brain barrier damage in acute MS plaques. Brain 114:557-572.
15 Gonsette R, Andre-Balisaux G, Delmotte P (1966). La perméabilité des cérébraux. Démyélinisation expérimentale provoquée par des substances agissant sur la barriére hémato-encéphalique. Acta Neurol Psychiatr Belgica 66:247-262.
16 Goodman J, Bingham W, Hunt W (1976). Ultrastructural blood-brain barrier alterations and edema formation in acute spinal cord trauma. J Neurosurg 44:418-424.
17 Grünewald T&Beal MF (1999). NOS knockouts and neuroprotection. Nature Med 5:1354-1355.
18 Heffner RR Jr, Porro RS, Olson ME, Earle KM (1976). A demyelinating disorder associated with cerebrovascular amyloid angiopathy. Arch Neurol 33:501-506.
19 Ilveskoski E, Lehtimaki T, Erkinyuntii T, Koivula T, Karhunen PJ (1998). Rapid apolipoprotein E genotyping from mailed buccal swabs. J Neuroscience Meth 79:5-8.
20 Kelly RE&Jellinek EH (1961). Intrathecal tuberculin in disseminated sclerosis: a controlled trial. BMJ 2:421-424.
21 Keschner M (1950). The effect of injuries and illness on the course of multiple sclerosis. Res Public Assoc Res Nervous Mental Dis 28:533-547.
22 Kobrine A, Doyle T, Rizzoli H (1976). The effect of antihistamines on experimental posttraumatic edema of the spinal cord. Surg Neurol 5:307-309.
23 Levine NS, Atkins A, McKeel DW, Peck SD, Pruitt BA (1975). Spinal cord injury following electrical accidents: case reports. J Trauma 15:459-463.
24 Lightman S, McDonald WI, Bird ACet al. (1987). Retinal venous sheathing in optic neuritis: its significance for the pathogenesis of optic neuritis. Brain 110:405-414.
25 Loseff NA, Webb SL, O'Riordan JIet al. (1996). Spinal cord atrophy and disability in multiple sclerosis. A new reproducible and sensitive MRI method with potential to monitor disease progression. Brain 119:701-708.
26 Mattingly G, Baker K, Zorumski CF, Figiel GS (1992). Multiple sclerosis and ECT: possible value of gadolinium-enhanced magnetic resonance scans for identifying high risk patients. J Neuropsychist Clin Neurosci 4:145-151.
27 McAlpine D (1946). The problem of disseminated sclerosis. Brain 69:233-250.
28 McAlpine D, Compston ND, Lumsden CE (1965). Multiple Sclerosis. Edinburgh, Livingstone, pp. 74-75.
29 McDonald WI (2000). Relapses, remission and progression of multiple sclerosis. N Engl J Med 343:1486-1487.
30 Miller H (1964). Trauma and multiple sclerosis. Lancet 1:848-850.
31 Ommaya A, Rockoff D, Baldwin M (1964). Experimental concussion: a first report. J Neurosurg 21:249-265.
32 Oppenheimer DR (1978). The cervical cord in multiple sclerosis. Neuropath Appl Neurobiol 4:151-162.
33 Otte A, Mueller-Brand J, Fierz L (1995). Brain SPECT findings in late whiplash syndrome. Lancet 345:1513-1514.
34 Poser C (1986). The pathogenesis of multiple sclerosis: a critical reappraisal. Acta Neuropathol 71:1-10.
35 Poser CM (1994a). The role of trauma in the pathogenesis of multiple sclerosis: a review. Clin Neurol Neurosurg 96:103-110.
36 Poser CM. (1994b). The role of the blood-brain barrier in the pathogenesis of multiple sclerosis. In: Salvati S, ed. A Multidisciplinary Approach to Myelin Diseases II. Plenum Press, New York, pp. 221-229.
37 Reichert T, Hassler R, Mundinger F (1975). Pathologic-anatomical findings and cerebral localization in stereotactic treatment of experimental motor disturbances in multiple sclerosis. Confina Neurol 37:24-40.
38 Roses AD&Saunders A (1995). Head injury, amyloidß and Alzheimers disease. Nature Med 7:603-604.
39 Russell DS (1946). The problem of demyelination. London Hospital Gazette49: May Clinical Supplement.
40 Sibley WA, Bamford CR, Clark K (1985). Clinical viral infections and multiple sclerosis. Lancet 1:1313-1315.
41 Sibley WA, Bamford CR, Clark K, Smith MS, Laguna JF (1991). A prospective study of physical trauma and multiple sclerosis. J Neurol Neurosurg Psychiat 54:584-589.
42 Sibley WA&Foley JM (1965). Infection and immunization in multiple sclerosis. Ann New York Acad Sci 122:457-468.
43 Smith HV, Hughes IE, Hunter G (1955). The blood-cerebrospinal fluid barrier in tuberculous meningitis and allied conditions. J Neurol Neurosurg Psychiat 24:101-117.
44 Trapp BD, Paterson J, Ransohoff RMet al. (1998). Axonal transection in the lesions of multiple sclerosis. N Engl J Med 338:278-285.
45 Yick L, Wu W, So K, Wong S (1998). Time course of NOS expression and neuronal death in Clarkes nucleus following traumatic injury in adult rat spinal cord. Neurosci Lett 241:155-158.