Does myasthenia gravis affect the brain?

Abstract

Associations between myasthenia gravis (MG) and CNS functions have been made for over 80 years. An increased incidence of psychiatric disorders, epilepsy and multiple sclerosis as well as electroencephalographic (EEG) abnormalities and abnormal evoked responses have been noted in patients with MG. Descriptions of sleep and memory disturbances in MG patients appeared as knowledge accumulated about the role of brain cholinergic systems in sleep and memory. The inference of many of these studies has been that the alleged central cholinergic effects in MG were caused either by the anticholinesterases used to treat MG or by antibodies to muscle nicotinic acetylcholine receptor (nAchR) present in the serum and cerebrospinal fluid (CSF) of MG patients. The antigenic differences between muscle nAchR and neuronal nAchRs, together with the very low concentrations of muscle nAchR antibodies in the CSF, make highly unlikely the claims that CNS cholinergic systems are affected by these muscle antibodies in MG patients. Evoked response abnormalities, if indeed present, are more likely caused by peripheral than central mechanisms, and sleep abnormalities in MG also probably originate in the periphery rather than in the CNS, the result of hypoxia caused by oropharyngeal, intercostal and diaphragmatic muscle weakness which may worsen during sleep, especially during REM sleep. Such hypoxia may account for some of the EEG abnormalities noted in MG patients, but the association of MG with epilepsy appears to be either coincidental or the result of uncontrolled MG. Significant excessive daytime sleepiness resulting from sleep disturbances can also impair memory and the performance of MG patients on neuropsychological tests, as can the presence of mental depression. The psychological aspects of MG can be attributed to the expected consequences of a chronic but unpredictable neuromuscular disease involving weakness of breathing, swallowing, talking, limb and eye movement. Considering the number and variety of claims for direct CNS involvement in MG, the evidence for this is remarkably unconvincing. The quality of MG treatment, both physical and psychological, is a presently undefined variable which might help explain the diametrically opposed results which have been obtained in some of the studies reviewed. Adequate respiratory muscle strength during sleep is an often overlooked peripheral influence upon mental functioning and general well-being of MG patients.

Introduction

Physicians in the late 19th century who first differentiated myasthenia gravis (MG) as a disease separate from typical progressive bulbar paralysis expected to find in the central nervous system (CNS) of autopsied cases some pathological explanation for the characteristic clinical signs of fluctuating skeletal muscle weakness during life. They were perplexed when, as Samuel Wilks commented in 1877 about his case of myasthenic bulbar palsy [1], “The medulla oblongata was very carefully examined, and no disease was found. It appeared quite healthy to the naked eye, and the microscope discovered no manifest change in the tissue.” In 1901 Herrman Oppenheim [2] actually named the disease ‘Bulbarparalyse ohne anatomischen Befund’ (bulbar paralysis without anatomical findings).

Even in the 1920s some people remained convinced that the site of pathogenesis could be found in the CNS [3], [4] after the attention of most people in their search for pathology had shifted to muscle [5] or to the thymus [6]. Eventually it became clear by the middle of the 20th century that the most likely explanation for most of the clinical features of MG was a problem at the neuromuscular junction [7], [8]. A dramatic series of scientific advances in the early 1970s established that the manifestations of MG could be explained by an autoimmune attack upon the nicotinic acetylcholine receptors (nAchRs) of the postsynaptic membrane of skeletal muscle [9], [10], [11], [12].

However, the possibility that the brain is involved in some subtle way in MG has continued to intrigue physicians and their MG patients, even though gross disturbances of CNS functioning have never been evident clinically. The occasional association of MG with psychosis, multiple sclerosis or epilepsy has been noted for over 80 years [13]. Electro-physiological abnormalities in patients with MG have also suggested CNS involvement. As knowledge accumulated about cholinergic systems in the brain and their role in sleep and memory [14], disturbances of sleep and memory in MG patients have been described. The inference of many of these studies has been that central cholinergic effects in MG were caused either by the anticholinesterases used to treat MG or by antibodies to muscle nAchRs present in the serum and cerebrospinal fluid (CSF) of MG [15]. This paper takes a critical look at the surprising number of papers in English which claim that the CNS is affected by MG or its treatment.