Article Text

Download PDFPDF

Central nervous system mechanisms in Sjögren’s syndrome
  1. O P van Bijsterveld1,
  2. A A Kruize2,
  3. R L A W Bleys3
  1. 1Oogcentrum Houten, Achterom 16b, 3995 EB Houten, The Netherlands
  2. 2Department of Rheumatology and Clinical Immunology, University Medical Centre Utrecht, F02-127, PO Box 85500, 3508 GA Utrecht, The Netherlands
  3. 3Rudolf Magnus Institute of Neuroscience, Department of Pharmacology and Anatomy, University Medical Centre Utrecht, STR02-208, PO Box 80039, 3508 TA Utrecht, The Netherlands
  1. Correspondence to: A A Kruize; a.kruize{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

A theory

The origin of the dry eye in Sjögren’s syndrome has hardly been a subject of discussion as it is assumed that lymphoplasmocytic cell infiltration of the tear gland deteriorates its function to such a degree that it causes dry eye. Several observations, however, may not support the concept that tear gland degeneration is the only causative factor in ocular dryness in Sjögren’s syndrome.

Sjögren’s syndrome is thought to be the consequence of a generalised autoimmune induced exocrinopathy resulting in localised symptoms including dry eyes and mouth, and generalised symptoms including fatigue, myalgia, and arthralgia.1 Rarely available epidemiological data suggest a prevalence varying between 1.5–4%, a phenomenon probably at least partly the result of the various classification criteria used.2–4 Important components in the pathogenesis are the emergence5 and persistence of autoimmune T cells6 and a subsequent failure of apoptosis of these activated cells,7 which might result in a persistent stimulation of B cells. Extensive lymphoplasmocytic infiltration of tear and salivary glands is thought to intervene with its secretory function, resulting in dry eyes and mouth.

Even in idiopathic keratoconjunctivitis sicca, usually not thought to be based on generalised autoimmune disease, a local immune driven inflammatory reaction in the tear gland is also thought to interfere with the functional unit comprising the ocular surface, tear gland, and interconnecting reflex arc.8,9 A critical decrease in androgen level leads to atrophy of the lacrimal glands.10 Resulting apoptotic fragments of the interstitial and acinar cells might act as a source of potential autoantigens that subsequently might be presented either by interstitial antigen presenting cells or acinar cells to CD4 cell antigen receptors and start an immune response.11 Several experimental and clinical observations cast doubt on the notion that tear gland degeneration is the only factor in deterioration of the tear flow in Sjögren’s syndrome.


Although in a murine model of Sjögren’s syndrome, extensive lymphoplasmocytic cell infiltration of the lacrimal glands was observed, these glands contained a large number of apparently unaffected acinar and ductal cells. Despite this, lacrimal function was impaired to such a degree that aqueous tear deficient keratoconjunctivitis developed.12 Immunohistochemical analysis using specific antibodies to markers of parasympathetic, sympathetic, and sensory nerves demonstrated that the density and pattern of visceromotor and sensory innervation of the non-infiltrated parts of the lacrimal glands were indistinguishable from that of age matched healthy control lacrimal glands.

Although not all parasympathetic markers available were investigated, this implies that the loss of the secretory function in Sjögren’s syndrome is probably not due to impaired peripheral autonomic innervation.12 However, in another experiment using the murine model of Sjögren’s syndrome, activation of nerves of lacrimal glands infiltrated with plasmolymphocytic cells, did not increase the release of acetylcholine, thus contributing to an impaired secretion of the glands.13


In a long term follow up study, 10 years after an initial diagnosis of Sjögren’s syndrome was made, the average tear function returned to normal values, but not in all patients.14 Thus, a reversible component in tear function was suggested.

In a recent study (Kruize, personal communication) 25 patients were classified as suffering from (primary) Sjögren’s syndrome, according to the EC criteria.4 Fourteen patients with relatively mild ocular signs and symptoms and 11 patients with rather severe ocular surface disease were compared with regard to severity of disease as indicated by serum immunoglobulin G. No differences in serum IgG between both patient groups were found (Table 1). Thus, the degree of dryness seems not to be linked directly to the severity of disease.

Table 1

No statistical differences were found between two groups of Sjögren patients, classified according to the EC criteria, one group with severe and another with mild ocular signs and symptoms, with regard to severity of the generalised autoimmune disease as expressed by mean concentrations of serum IgG (g/l)


Some patients with serious chronic inflammatory disease of the conjunctiva may bitterly complain of dry eyes. Clinical assessments may reveal repeatedly low or very low Schirmer-I test values without any clinical evidence of trigeminal neuropathy. Paradoxically, these patients may show copious tearing after an emotional event, indicating that the central cortical and limbic efferent transmission of impulses to the superior salivatory nucleus are intact, as well as the parasympathetic route from this nucleus to the lacrimal gland.

All these clinical and experimental data consequently suggest a concomitant responsible component of the mechanisms responsible for perpetual deterioration of tear function in Sjögren’s syndrome. According to our theory, the central nervous system may be involved in reduction of lacrimation and in altered pain sensations, as will be suggested in the following paragraphs.


In the trigeminal nerve two distinguishable somaesthetic fibre systems carry nociceptive impulses from the peripheral nerve endings—one phylogenetically old system consists of non-myelinated, small diameter fibres, slow in conducting which do not adapt easily: the C fibre system; the other, phylogenetically new system of myelinated, large diameter fibres, fast in conducting, which do adapt rather easily: the A-delta fibre system.

Volleys of impulses generated after high intensity stimulation of the outer eye are carried by A-delta fibres to the central transmission cells. The stimulus exceeding a critical firing level of these cells will activate neural mechanisms responsible for perception of acute pain, resulting in behavioural patterns aimed at avoiding sensory and affective sensations. It will also activate the parasympathetic system resulting in tearing.

If the stimulus continues, however, volleys of impulses are also carried by the C fibre system. These fibre impulses are transmitted to the substantia gelatinosa of the spinal trigeminal nucleus especially and will be transmitted to higher centres to be perceived as “slow pain.”15


The substantia gelatinosa is the region in the spinal cord and the spinal trigeminal nucleus where many nociceptive impulses are processed after entering the central nervous system. This region contains densely packed small interneurons which interconnect by short fibres and are also connected with longer fibres of the spinal trigeminal tract. It is believed that the substantia gelatinosa acts as a synaptic transmission modulating system of nerve impulses from the peripheral trigeminal fibres to the central transmission cells. Summation through continuous stimulation of these cells by A-delta fibres is prevented by a negative feedback mechanism. However, under conditions of prolonged stimulation, the A-delta fibre system begins to adapt and gives way by transmission through the C fibre system. Volleys in small diameter fibres activate a positive feedback mechanism, which enhances the effect of the incoming impulses.16

Because of the positive feedback mechanism in the substantia gelatinosa there is a temporal summation of the incoming signals to the central transmission cells, a mechanism first suggested by Livingstone17 and called “wind-up.”18 The wind-up response is thought to underlie central sensitisation, an adaptation of afferent neurons to prolonged nociceptive stimuli as occurs in tissue damage and inflammation, which contributes to hyperalgesia and persistence of pain.18,19 When the output of the central transmission cells exceeds a certain level, a number of responses may be triggered through ascending systems. Nociceptive information uses the trigeminothalamic and trigeminoreticulothalamic tracts, as well as direct pathways to limbic structures such as the hypothalamus and amygdala. Subsequent activation of, particularly, the thalamus, the amygdala, the hypothalamus, the parabrachial nucleus, the brain stem reticular formation, as well as the parietal and frontal cortex, may occur.


In Sjögren’s syndrome, the amount of tear flow is initially decreased through infiltration of the tear gland with subsequent loss of function, which causes increased friction between the outer ocular tissues. Dryness of the cornea may cause exfoliation of the superficial corneal epithelium leaving corneal erosions, resulting in considerable ocular discomfort, which is chronic in nature, because of the character of Sjögren’s syndrome. Because of the continuous stimulation caused by the chronic inflammatory reaction of the conjunctiva and the exfoliative changes of the cornea epithelium in Sjögren’s syndrome, repeated constant C fibre stimuli result in “slow pain,” a diffuse, burning sensation of the conjunctiva, so characteristic of the dry eye in Sjögren’s syndrome.

This relatively unchecked C fibre input can also easily lead to summation (wind-up), which can augment responses of spinal trigeminal nucleus neurons and lead to central sensitisation. More important even: we hypothesise that this C fibre input can induce a frequency conditioned reversal of the expected response by inhibiting the parasympathetic system. Frequency conditioned reversal of responses of the autonomic nervous system have been described for the cardiovascular system.20 It is uncertain which circuitry is responsible for this inhibition. Considering the connections between ascending pathways and limbic structures (hypothalamus and amygdala) on one hand, and between limbic structures and the peri-aqueductal grey on the other hand,19 we hypothesise that the inhibition process takes place in this circuitry, the hypothalamus being important to coordinate autonomic functions. As a consequence there is reduction of tearing and this we believe is the reversible and additional factor in Sjögren’s dry eye.


In chronic inflammatory conditions of Sjögren’s dry eye, summation of peripheral input may occur since C fibre activity starts to dominate the A-delta fibre activity resulting in an active inhibition of the parasympathetic system at the central level in the peri-aqueductal grey area of the limbic system. There have been reports of peripheral sympathetic and/or parasympathetic dysfunction,21–27 but none of these reports are incompatible with the concept of C fibre induced inhibition of tear flow in Sjögren’s syndrome.

Also, in clinical practice, including persistent and consequent care of a supportive ophthalmologist, long term, low dose treatment of chronic inflammatory disease of the conjunctiva, consisting of twice a day a drop of 0.025–0.1% dexamethasone, markedly improved subjective and objective conditions in Sjögren patients. This treatment allowed a number of patients to discontinue any local medication, including tear substitutes. In case of the complication of a herpes simplex infection of the eye, a maximal oral dose of antiviral agents during 3 weeks, in addition to tapering down the local steroid treatment, not lower than an equivalent of a once a day drop of 0.1% dexamethasone, may be necessary.

Furthermore, in our long term follow up study, many, but not all, Sjögren patients had subjective rather than objective improvement of tear function. The most important contribution to the understanding of the alleviation of discomfort and pain, without an objective improvement under these conditions, is the emphasis on central neural mechanisms, as will be explained now.


In a previous section it was pointed out that afferent fibres in the medulla oblongata and spinal cord, transporting nociceptive information, activate many brain centres, among them peri-aqueductal grey and brainstem raphe nuclei. Descending systems originating in these centres modulate the sensitivity of the substantia gelatinosa, the primary processing centre of nociceptive impulses. Limbic brain structures influence the peri-aqueductal grey, which explains altered pain sensations during stressful events. For example, some soldiers severely wounded at the Anzio beachhead in the second world war experienced no pain, probably because they realised they had escaped from the ordeal of the battlefield alive.28 Central nervous system activities subserving emotion, attention, and memories of previous experience are thus able to exert control over the sensory input through modulating the synaptic transmission.29 In our view this mechanism underlies subjective improvement of discomfort as a result of consequent care of a supportive ophthalmologist. The central control is complex and may be very selective, however, which may explain individual differences.


The central nervous system may have an important role in the cascade of events that occur in Sjögren’s syndrome. Our presented theory of repeated constant C fibre stimuli, in combination with frequency conditioned inhibition of the central autonomic centre for tear flow, has a striking power to explain clinical and experimental discrepancies of the concept of infiltration and destruction of the tear gland as a single cause in tear flow depression. In addition, central neural mechanisms may also explain those cases where patients have subjective rather than objective improvements of complaints.

Note in Proof

A theory