Article Text

This article has a correction. Please see:

”Paratrigeminal” paralysis of the oculopupillary sympathetic system
  1. P J Goadsby
  1. Correspondence to:
 Professor P J Goadsby, Institute of Neurology, Queen Square, London WC1N 3BG, UK;


Raeder described five patients with mixed features of trigeminal nerve pathology and oculosympathetic impairment, with or without other cranial nerve lesions. This constellation of clinical features drew the original author's attention to the paratrigeminal region as a likely site for the causative lesion in this syndrome. An analysis of the anatomy of the oculosympathetic innervation supports the view that a restricted lesion in the middle cranial fossa might cause the syndrome of trigeminal nerve involvement, neuralgic pain or sensory change, with ptosis or miosis, or both, but no anhidrosis. Such a paratrigeminal oculosympathetic syndrome (POSS) usefully reminds clinicians to pursue vigorously possible lesions of the middle cranial fossa with careful, and possibly repeated, imaging studies. Attaching the eponym Raeder's syndrome or Raeder's paratrigeminal neuralgia to this syndrome adds nothing valuable to the anatomical description (POSS), which might be preferred for clarity.

  • paratrigeminal paralysis
  • ocuopupillary sympathetic inervation
  • Raeser's syndrome

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.

In 1924 Raeder1 wrote a classic clinical-anatomical colocalisation paper that described five patients, one of whom had been reported in 1918. The patients had two key features: involvement of the trigeminal nerve and the oculosympathetic nerves. He sought to differentiate the restricted oculosympathetic findings from the classic Horner's syndrome: cervical sympathetic dysfunction characterised by ptosis, miosis, anhidrosis and enophthalmos. Since that time various terms have been employed, meanings defined, and classifications developed.2,3 It is against this background that one ventures into the Coliseum that is the history of medicine to play out the academic battle. An eponym is useful if it conveys a very clear meaning, or perhaps acknowledges some contribution that was so pivotal as to provide insight well beyond the describer's demise. Raeder made an interesting clinical observation that pointed out the likely localisation of a lesion adjacent to the trigeminal nerve in the middle cranial fossa. In the era before cranial imaging this was a wonderful piece of neurology; in the modern era the eponym is much less useful. Solomon and Lustig4 recently set out the clinical cases that in many respects have illustrated the trigeminosympathetic anatomy of the carotid artery, concluding that the use of the term Raeder's paratrigeminal neuralgia had become corrupted to the point of being useless by careless attribution of cases that did not respect the anatomy. Conditions such as carotid disease, particularly dissection, may give rise to pain and a Horner's syndrome,3 and cluster headache may lead to oculosympathetic loss and impaired sympathetic facial sweating5,6 In both situations forehead sweating may be impaired. I will set out the relevant anatomy of the paratrigeminal region, then review Raeder's patients, and finally make some suggestions for the postimaging era based on an anatomical/pathophysiological designation for such patients.


The key anatomical feature to be understood for Raeder's syndrome is the relation between the trigeminal nerve and the oculopupillary sympathetic fibres. The trigeminal nerve lies in the middle cranial fossa and in close proximity, paratrigeminally as Monrad-Krohn suggested, there are other cranial nerves. Most particularly for a short course the fibres that will innervate the levator palpebrae superioris, specifically Müller's muscle, and the pupilodilator fibres without the sudomotor fibres for the forehead. Therein is the anatomical lesson of Raeder.

Cranial sympathetic innervation

The sympathetic outflow ultimately arises in the hypothalamus and can be modulated by brain stem neurons. Preganglionic fibres arise from neurons in the lateral column of grey matter of the thoracic and upper two to three lumbar segments. The fibres emerge through the ventral roots of the corresponding spinal roots, passing into the spinal nerve trunks and their ventral rami. The fibres leave the ventral rami in the white rami communicantes to synapse in the corresponding ganglion or, in the circumstance of interest for the cervical sympathetic ganglia, to ascend before synapsing7 The sympathetic ganglia give rise to non-myelinated postganglionic fibres, which, for the head, arise as the internal carotid nerve from the superior cervical ganglion, which is, in effect, the rostralmost section of the sympathetic trunk. The cervical sympathetic ganglia destined to innervate the eye derive their input from the upper thoracic (T1) sympathetic white rami communicantes.

Oculosympathetic innervation

The internal carotid nerve ascends with the internal carotid artery dividing into medial and lateral branches in the bony carotid canal, and forming a plexus known as the carotid plexus. The medial plexus communicates with the trigeminal ganglion and abducens nerve, and the lateral plexus communicates with the oculomotor, trochlear, ophthalmic, and abducens nerves, as well as the ciliary ganglion. The dilator pupillae fibres travel with either of the ophthalmic artery, nasociliary or long ciliary nerves, and those for levator palpebrae superioris travel with the oculomotor nerve.

Sympathetic control of facial flushing

The cervical sympathetic pathway is responsible for facial sweating and flushing, among other roles. The oculosympathetic and facial flushing roles seem well differentiated functionally given that central lesions have been reported in which the facial sweating is spared in the presence of an oculosympathetic lesion.8 Indeed whereas oculosympathetic fibres arise from T1 those for sudomotor function arise at lower levels; certainly at T2/3.9 Sudomotor fibres destined for the forehead traverse the internal carotid plexus, whereas those for the remainder of the face traverse the external carotid plexus, excepting those for the cheek, which may take either route10 Sudomotor fibres destined for the forehead join the ophthalmic nerve in the region of the cavernous sinus reaching the skin through the supraorbital nerve11 Sweating of the forehead is, therefore, normally mediated by fibres that exit the internal carotid plexus relatively laterally when compared with the oculosympathetic fibres. It must be borne in mind that parasympathetic neurons may reinnervate the sweat glands,8 so that intact sweating may not necessarily imply sympathetic sudomotor integrity.

Trigeminal nerve and oculosympathetic outflow

As the internal carotid artery passes from being inferior to the trigeminal ganglion and oculomotor nerve, and before it lies medial to the anterior clinoid process, it must distribute from the lateral internal carotid plexus the fibres for the dilator pupillae and levator palpebrae superioris; the sympathetic fibres destined for the trigeminal ganglion being delivered from the medial plexus before that point, as described above. Thus a lesion restricted to the middle cranial fossa medial to the trigeminal ganglion and effectively lateral to the anterior clinoid, would produce trigeminal pain by direct irritation of the anteromedial ganglion or initial section of the ophthalmic division, and could interrupt oculosympathetic outflow. There would be no expected effect on facial sweating because the sudomotor fibres have exited before this point. Thus a paratrigeminal oculosympathetic syndrome arises.


Set against the anatomy it is appriopriate to consider Raeder's patients. Table 1 summarises the patients by dividing up the clinical symptoms and signs to highlight the trigeminal and oculosympathetic involvement, as well as other relevant clinical features. Several features emerge from such an analysis.

Table 1

Raeder's cases

Trigeminal involvement

Considering three possible types of trigeminal involvement—pain, sensory change and motor signs—what is constant in Raeder's patients? Only three patients had pain and only two had neuralgia. Thus the term Raeder's paratrigeminal neuralgia is fanciful. Three patients had sensory change; one patient had pantrigeminal loss indicating that the lesion had moved lateral to the unique paratrigeminal region, and similarly patient 1 had motor loss indicating a more extensive lesion.

Oculosympathetic loss

All five of Raeder's patients had miosis, and four of five had ptosis. Four of five also had normal sudomotor function. Thus oculosympathetic loss which must be peripheral was demonstrated but without anihydrosis.

Other signs

Four out of five of the original patients had other cranial nerves involved. It is interesting that one had cranial parasympathetic activation, most likely simply as a result of pain and activation of the trigeminal-autonomic reflex12

Status of Raeder's patients

An analysis of Raeder's patients demonstrates several issues. It is notable that only two of the patients had neuralgia and two had lesions not restricted to the paratrigeminal region. Thus perhaps one patient at most, patient 4, really illustrated in a pure fashion the anatomical construct that Raeder seemed to illustrate. An analysis of the patients and the relevant anatomy suggests that Raeder's paratrigeminal neuralgia might usefully be dropped from the textbooks with no substantive loss as the patients described did not really, perhaps save one, fulfil the anatomical requirements strictly.


How might one usefully draw on the lesson that Raeder sought to teach, essentially one of functional anatomy, and make a useful clinical rule? It could be suggested that Raeder made a first, and very creditable, attempt to describe a clinically useful syndrome based on the relevant anatomy (fig 1). Patients with oculosympathetic loss, miosis or ptosis, or both, with normal forehead sweating, and evidence of trigeminal involvement, either sensory change or neuralgic pain, are highly likely to have a lesion in the middle cranial fossa that is medial to the trigeminal ganglion (paratrigeminal). The anatomy of this syndrome is clear with dissociation of the oculosympathetic outflow from forehead sweating in the middle cranial fossa; the only caveat being that if there is parasympathetic innervation of sympathetically dennervated sweat glands, as seen in cluster headache,8 then the sweating in Raeder's patients would be pathological and much of the argument would be lost. Given that Raeder was unaware of this and, therefore, could not have tested for it, the patients do not provide absolute evidence for his argument. Perhaps the name of the syndrome should simply reflect the anatomy—paratrigeminal oculosympathetic syndrome—and thus fulfil the useful role of teaching. One might suggest that the term Raeder's paratrigeminal neuralgia is inaccurate, and Raeder's (paratrigeminal) syndrome does not teach the essential lesson in its name. Certainly, the unique combination of symptoms and signs should trigger an exhaustive search for a lesion with MRI. When no lesion is found the patient must be followed up, and imaging should be repeated at least once. Neurology evolves with the lessons of history to better modern practice; the lesson of Raeder is learnt and requires more transparent dissemination, including dropping the eponym. One might simply argue that to learn one thing—that is, the paratrigeminal oculosympathetic syndrome—is to learn the entire lesson, to learn a person's name adds nothing more in clinical terms. The use of such eponyms remains, however, a matter of taste and practice; and for Raeder should be tightly coupled to the anatomy.

Figure 1

Drawing of the anatomy of the paratrigeminal oculosympathetic syndrome. Sympathetic fibres from the superior cervical ganglion (SCG) travel with the internal carotid artery in the neck, branching to innervate the sweat glands, levator palpebrae superioris (specifically Müller's muscle), and the pupilodilator fibres. The pupilodilator fibres travel near the ophthalmic portion of the trigeminal nerve, so that in this paratrigeminal region (grey shade) lesions produce the classic syndrome. 3, oculomotor nerve; 4, trochlear nerve; 5, trigeminal nerve; Vg, trigeminal ganglion, with first (I), second (II), and third (III) branches.


I thank Dr Peter Drummond for helpful comments on the manuscript. PJG is a Wellcome Trust senior research fellow.


Supplementary materials

  • Goadsby, PJ. Paratrigeminal paralysis of oculopupillary sympathetic system.
    J Neurol Neurosurg Psychiatry 2002;72;297-299.

    During the production process the title was inadvertently shortened. The title should read:  Raeder's syndrome: paratrigeminal paralysis of oculopupillary sympathetic system.

Linked Articles

  • Correction
    BMJ Publishing Group Ltd