Tolosa-Hunt Syndrome

In 1954, Edvardo Tolosa of Barcelona described a 47-year-old male with signs and symptoms of severe retro-orbital and supraorbital pain, followed later by paralysis of the third, fourth, and sixth cranial nerves, and a diminished corneal reflex. The patient had had a similar episode 3 years previously, but had spontaneously recovered. Angiography showed a suggestive narrowing of the carotid siphon just distal to the cavernous sinus. The patient died 3 days after a negative intracranial exploration. At autopsy, ``The intracavernous portion of the carotid artery was wrapped in granulomatous tissue which did not obstruct the lumen of the sinus.'' The process had also invaded the adjacent cranial nerves.

Hunt et al. (1961) reported six examples of the ``painful ophthalmoplegia'' syndrome and set forth the following criteria by which it can be recognized. (1) Pain may precede the ophthalmoplegia by several days, or may not appear until some time later. The pain is not a throbbing, paroxysmal hemicrania but is a steady pain behind the eye and in the brow, often described as ``gnawing'' or ``boring.'' (2) Neurologic involvement is not confined to the third cranial nerve, but may include the fourth and sixth cranial nerves as well as the first (and occasionally the second) division of the trigeminal nerve. The optic nerve and the oculosympathetic fibers may be involved, resulting in diminished vision in the former instance and a small, reactive (or nonreactive if the oculomotor nerve is also involved) pupil in the latter. (3) The symptoms last for weeks or months. (4) Spontaneous remission occurs, sometimes with residual neurologic deficit. (5) Attacks recur at intervals of months or years. (6) Exhaustive studies, including angiography and surgical exploration, have produced no evidence of involvement of structures outside the cavernous sinus.

Hunt et al. (1961), in a study, of the histologic material from Tolosa's (1954) case, noted that: ``The region where the carotid artery enters the cranial cavity by penetrating the outer wail of the cavernous sinus is the site of a low-grade, non-specific, inflammatory process. This process is characterized by proliferation of fibroblasts and by infiltration of the septa and wall of the sinus with lymphocytes and plasma cells. There is no necrosis, and no polymorphonuclear cells are present. No fibrinoid collagen degeneration is observed. The adventitia of the carotid artery, as well as the smaller vessels, is only incidentally infiltrated by the chronic inflammatory cells.'' Hunt et al. found no evidence for a primary arteritis. The granulomatous lesion had engulfed the abducens nerve and all small connecting branches of the carotid plexus and the ophthalmic division of the trigeminal nerve. Myelin sheaths were fully preserved, and no inflammatory changes were seen within the perineurium. They considered the possibility of infection arising within the sphenoid sinus but emphasized that the etiology of the syndrome was still unknown.

Schatz and Farmer (1972) reported four patients with painful ophthalmoplegia, two of whom underwent craniotomy. One patient was found to have a soft, granular lesion in the dura of the lateral wall of the cavernous sinus. A biopsy specimen from this region showed dense connective tissue and a granulomatous response with epithelioid cells, giant cells, and scattered plasma cells with necrosis but no caseation. In the second patient who underwent intracranial exploration, a soft tissue mass was seen elevating the oculomotor nerve, depressing the abducens nerve, and extending to the Gasserian ganglion from the cavernous sinus. A biopsy specimen from this mass showed chronic, granulomatous tissue with plasma cells, lymphocytes, and minute areas of necrosis. In addition to presenting the pathology in their cases, Hunt et al. (1961) as well as Schatz and Farmer emphasized the dramatic response of their patients to systemic corticosteroid therapy.

Lakke (1962) published a superb review of the so-called superior orbital fissure syndrome and presented convincing evidence that many examples of this syndrome are clinically indistinguishable from the inflammatory cavernous sinus syndrome of Tolosa or the ``painful ophthalmoplegia'' syndrome of Hunt et al. Lakke described a 47-year-old man who was admitted to the Neurosurgical Clinic in Utrecht with violent right-sided, stabbing headaches, radiating from the right retro-orbital region along the right side of the skull to the neck. This patient developed progressive paresis of all of his right-sided ocular motor nerves and depression of his right corneal reflex. His right pupil, though small, reacted normally to a direct light stimulus. A neurosurgical exploration revealed a thin layer of grayish red granulation tissue on the lateral wall of the cavernous sinus and on the dura covering the lesser wing of the sphenoid. Histologic examination of a biopsy specimen showed inflammatory tissue containing polymorphonuclear cells. The dura mater from the region of the superior orbital fissure was necrotic in some places and lined with granulation tissue. The histologic diagnosis was pachymeningitis of undetermined etiology. We agree with Lakke that painful inflammatory involvement of the tissues near the superior orbital fissure cannot be distinguished from inflammatory involvement of the tissues that comprise and occupy the cavernous sinus unless there is involvement of the second division of the trigeminal nerve, in which case a cavernous sinus syndrome may be diagnosed with some assurance. The spared pupil in some of our cases, in Lakke's case, and in a few of the cases that he summarized from the literature suggests a mechanism of oculomotor paresis related to the ischemic ophthalmoplegia of diabetes mellitus or giant cell arteritis (Trobe et al., 1978).

Although granulomatous inflammation of the cavernous sinus may produce a painful ophthalmoplegia, it should be clear that other lesions that involve the structures within the superior orbital fissure or cavernous sinus may also produce painful ophthalmoplegia that is often responsive to systemic corticosteroid therapy. Thomas and Yoss (1970) studied 102 patients with intracranial parasellar lesions and found that neither the mode of onset of symptoms nor the sequence of evolution or pattern of the neurologic deficit is characteristic of the etiology of the underlying lesion, whether it is neoplastic, aneurysmal, or inflammatory. These investigators found evidence of both spontaneous and steroid-induced remissions of symptoms and signs, sometimes of prolonged duration, in cases with tumors and aneurysms, and stressed the need for complete neuroradiologic investigations in patients with the syndrome of painful ophthalmoplegia. Other investigators (Fowler et al., 1975; Hunt, 1976; Coppeto and Hoffman, 1981; Kline and Galbraith, 1981) have also emphasized the similarity of symptoms and signs caused by inflammatory and noninflammatory lesions of the cavernous sinus and the need for careful neuroradiologic investigations in all patients.

Radiologic abnormalities have been described in a number of cases of Tolosa-Hunt syndrome. Plain roentgenographic findings consist primarily of bone erosion in the sellar and parasellar areas (Schatz and Farmer, 1972; Polsky et al., 1979). Venography has been performed on several patients and often shows obstruction of the third portion of the ipsilateral superior ophthalmic vein as well as obstruction within the ipsilateral cavernous sinus (Milstein and Morretin, 1971; Hallpike, 1973; Sondheimer and Knapp, 1973; Spirn et al., 1975; Aron-Rosa et al., 1978; Gulliksen and Krarup, 1978; Takeoka et al., 1978; Cohn et al., 1979; Muhletaler and Gerlock, 1979; Joshita et al., 1980). Some investigators have emphasized that venography may be abnormal when all other neuroradiologic tests are unremarkable; however, we would emphasize that with the rapidly advancing technical achievements in neuroradiology, it is probably inappropriate to single out a ``best'' test for this type of disorder.

Arteriographic findings in the Tolosa-Hunt syndrome include irregular narrowing, flattening, and displacement of the intracavernous portion of the internal carotid artery, at times suggesting a mass in this region (Tolosa, 1954; Mathew and Chandy, 1970; Schatz and Farmer, 1972; Hallpike, 1973; Sondheimer and Knapp, 1973; Roca, 1975; Aron-Rosa et al., 1978; Takeoka el al., 1978; Cohn et al., 1979) as well as hypervascularity with the suggestion of an orbital mass (Rosenbaum et al., 1979). Kettler and Martin (1975) identified an arterial stationary wave phenomenon in a patient with Tolosa-Hunt syndrome and constriction of the internal carotid siphon. According to Takeoka et al., the vascular changes observed in patients with Tolosa-Hunt syndrome appear to resolve with systemic corticosteroid therapy (see below).
Computed tomography has been performed in some cases of Tolosa-Hunt syndrome. Aron-Rosa et al. (1978) performed computerized tomography in seven patients with only one patient showing any abnormality. In our experience, however, thin section scanning techniques using intravenous drip and bolus enhancement of both axial and direct or indirect coronal images almost always reveal increased density in the region of the involved cavernous sinus and superior orbital fissure.

Smith and Taxdall (1966) suggested the use of prednisone (60---80 mg per day) as a preliminary diagnostic (as well as therapeutic) test, before subjecting the patient with painful ophthalmoplegia to the expense and discomfort of neuroradiologic studies. According to these investigators, if pain and ophthalmoplegia subside within 24---48 hours, a diagnosis of inflammatory painful ophthalmoplegia is assured. While this may be true in the majority of cases, several points must be kept in mind. First, as Thomas and Yoss (1970) and others have emphasized, noninflammatory lesions causing painful ophthalmoplegia may appear to respond dramatically to systemic corticosteroids. We believe that if a patient with a presumed Tolosa-Hunt syndrome does not respond to systemic administration of corticosteroids, the syndrome is almost certainly not inflammatory in nature; however, the reverse is not always true. Second, in our experience, even in true inflammatory painful ophthalmoplegia, although pain may be immediately relieved following institution of systemic corticosteroids, the ophthalmoplegia may require days, weeks, or months to resolve and may never resolve completely. Third, the availability of computed tomographic scanning and magnetic resonance imaging, as well as digital subtraction angiography allow the physician to perform a sensitive neuroradiologic evaluation of the cavernous sinus and orbital apex regions without significant discomfort or risk to the patient. Finally, we would emphasize that the Tolosa-Hunt syndrome may be exquisitely sensitive to systemic corticosteroids in most cases, but other cases may require large doses of steroids to achieve any improvement and also may require slow and careful tapering to avoid a ``rebound'' phenomenon.

In addition to tumors, aneurysms, and inflammatory lesions, vascular processes may produce painful ophthalmoplegia. Painful ophthalmoplegia has been described in patients with syphilis, giant cell (temporal) arteritis, diabetes mellitus (Jabs et al., 1981), rheumatoid arthritis (Roger and Alliez, 1935; Dornan et al., 1979), and systemic lupus erythematosus (Evans and Lexow, 1978). Thus, in addition to a complete neuroradiologic evaluation, patients with painful ophthalmoplegia may require a serologic test for syphilis (STS), an erythrocyte sedimentation rate, a temporal artery biopsy, a glucose tolerance test, a rheumatoid factor, and an antinuclear antibody titer.

A superb review of painful ophthalmoplegia in general and the Tolosa-Hunt syndrome in particular has been written by Kline (1982) and should be consulted by the reader interested in this subject. Please also see the differential diagnosis of painful ophthalmoplegia outlined in Table 3.

Table 3. Differential Diagnosis of Painful Ophthalmoplegia with Cranio-orbital Lesions.*

Diagnostic Category Orbital Inflammatory, infectious Orbital cellulitis Inflammatory, noninfectious Vascular Neoplasm at orbital apex Non-orbital Neoplastic Primary intracranial Cranial Secondary Metastatic Infectious Inflammatory Vascular Traumatic Other

Diagnostic Possibilities Bacterial Adjacent sinusitis Penetrating trauma Septic embolus Viral Herpes Zoster Fungal Mucor, aspergillus Thyroid-related orbitopathy Idiopathic (orbital pseudotumor) Orbital hemorrhage Arteriovenous malformation Primary Secondary Extension of sinus or intracranial lesion Metastasis Pituitary adenoma Meningioma Craniopharyngioma Chondroma Multiple myeloma Lymphoma Nasopharyngeal carcinoma Breast adenocarcinoma Prostatic adenocarcinoma Lung carcinoma Bacterial, viral, fungal Tolosa-Hunt syndrome Herpes zoster Wegener’s granulomatosis Temporal arteritis Sarcoidosis Aneurysm Carotid-carvernous fistula Dural-cavernous fistula Cavernous-sinus thrombosis Fractures, hematoma Diabetic cranial polyneuropathy

*Broad diagnostic categories are shown, with a limited list of diagnostic possibilities. From Case Records of the Massachusetts General Hospital, Case 4-1993, The New England Journal of Medicine, 328(4):266-275, 1993.
 

 

Return to Other Headaches