Article Text

Download PDFPDF

Linear scleroderma with orbital involvement: follow up and magnetic resonance imaging
  3. A L BAERT
  1. Department of Radiology
  2. University Hospitals, Leuven, Belgium
  3. Department of Ophthalmology
  4. University Hospitals, Leuven, Belgium
  1. Department of Radiology
  2. University Hospitals, Leuven, Belgium
  3. Department of Ophthalmology
  4. University Hospitals, Leuven, Belgium
  1. Ph Demaerel, MD, Department of Radiology, University Hospitals, Herestraat 49, B-3000 Leuven, Belgium.

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.

Editor,—Scleroderma is a chronic connective tissue disease of unknown aetiology.1 Linear scleroderma is an uncommon dermatological disorder which produces an unilateral and localised band-like linear atrophy of the skin and its underlying structures.

Orbital involvement has been described, but is very rare.12


A 21-year-old woman first presented to an ophthalmologist because of a progressive narrowing of the right palpebral fissure over a period of 3 years with loss of lashes of the eyelids. Further ophthalmic examination was normal. An orbital computed tomogram (CT) was normal.

Three years later there was additional enophthalmia and an entropion of the lower right eyelid was seen (Fig 1). Visual acuity was 8/10 in the right eye and 10/10 in the left. Visual field examination and colour vision were normal.

Figure 1

Right eye: note the narrowing of the palpebral fissure and the loss of lashes of the eyelids.

With the exception of a localised area of parietotemporal alopecia on the right side, no other skin defects could be detected. General examination revealed no muscular pain, no arthralgia, no fatigue, and no other systemic symptoms. Laboratory examination of muscle enzymes, tissue antibodies and complement factors was normal. There were no inflammatory factors and thyroid tests were within normal limits.

On an orbital CT the right eye was displaced laterally by a band-like mass. The medial rectus muscle and oblique superior muscle could not be identified. Muscle and conjunctival biopsy only showed some oedema in these structures.

The presumed diagnosis of localised scleroderma was confirmed by a dermatological examination. Initially no treatment was established.

An orbital magnetic resonance image (MRI) confirmed the CT findings but also demonstrated that the right superior rectus muscle was involved in the mass (Fig 2A and B). A new biopsy only showed muscular and fibroid tissue without signs of inflammation.

Figure 2

(A,B). Coronal T1 weighted MRI (TR/TE = 520/15 ms). The superior rectus muscle, the superior oblique muscle, and the medial rectus muscle on the right side appear enlarged and cannot be distinguished from each other (A) (arrowheads). The orbital apex is not involved (B) (arrow). (C) Axial T2 weighted MRI (TR/TE = 2200/80 ms). (D, E) Coronal T1 weighted MRI (TR/TE = 500/15 ms). (F) Coronal T1 weighted MRI after gadopentetate dimeglumine injection (TR/TE = 500/15 ms). The orbital abnormalities return a low signal on T2 weighted images (C) (arrows). The apex is involved and the entire orbital fat is infiltrated (D,E). Note that the optic nerve is only visible after the contrast administration (arrow) (F).

Two years after the first admission the patient complained of increasing right sided headache, photophobia, and progressive decrease of the visual acuity on the right side from 8/10 to 3/10. Visual field examination and colour vision were normal. Biomicroscopic examination showed a punctate keratitis on the right side. Funduscopy revealed a hyperaemic optic disc. Follow up MRI showed a progressive increase in the size of the orbital mass with extension into the apex. The optic nerve could only be identified following intravenous gadopentetate dimeglumine injection (Fig 2C–F). A canthotomy and an eyelid plasty were performed for restoration of the right upper eyelid (Fig3).

Figure 3

Clinical photograph taken after the two minor surgical procedures shows the progression of the disease.

The unilateral loss of vision progressed slowly to finger counting at 1 metre the following year. The visual fields showed normal peripheral limits, but colour vision was absent in the right eye. Cortisone therapy (Medrol, 32 mg/day) and radiotherapy (4 Gy on the right orbit, followed by 20 Gy in fractions of 2 Gy) were initiated, but the patient complained of increasing pain in and around the right eye. The patient became blind on the right side. Ophthalmic examination showed a mydriatic non-reactive pupil and a pale optic disc on the right side. Two follow up MR examinations did not show a further progression of the orbital abnormalities.


Scleroderma may occur as a systemic disease or as a localised form. Localised scleroderma presents in three clinical forms: generalised, morphoea (atrophic and sclerotic skin lesions), and linear scleroderma. In the linear form the lesions are arranged in a band-like linear distribution and may involve and fix the deeper layers of the skin and underlying structures. If it involves the frontoparietal region it is called scleroderma ‘en coup de sabre’ because of the depression that extends below the level of the skin. In patients with morphoea or linear scleroderma, the disease progresses slowly and may last from a few months to many years, but the prognosis is worst in the en coup de sabre type. According to some authors, it may lead to homolateral facial hemiatrophy (Parry–Romberg syndrome).1 However, the relation between linear scleroderma en coup de sabre and Parry–Romberg syndrome remains unclear.3

Olver et al have reported a patient with bilateral acquired Brown’s syndrome and combined lichen sclerosus et atrophicus and morphoea.4 They suggest that the deep subdermal fibrosis may limit the passive movement of the superior oblique tendon. The clinical features of Brown’s syndrome were not present in our patient, but can be seen in localised scleroderma.

The cutaneous pathological features depend upon the stage of the disease: the striking accumulation of inflammatory cells among collagen bundles of the lower dermis and among the fat cells and collagen bundles of the subcutaneous tissue is followed by the replacement of the subcutaneous tissue by hyalinised connective tissue.1

Involvement of the orbit and the eye is uncommon but has been described.12 Involvement of the extrinsic eye muscles is extremely rare and only CT findings have been reported.2

The clinical findings in this case are typical of localised linear scleroderma: the age and sex of the patient, the slow progression, the distribution and nature of the superficial defects, and the unsuccessful therapy. This patient has a significant induration of the scalp but there is no extension below the forehead and is therefore not considered as a Parry–Romberg syndrome. The term pseudotumour is misleading and inappropriate because of the presence of enophthalmia and the absence of mass effect.56 The latter can be explained by the atrophy of orbital fat that is only partially replaced by fibrosis. Orbital fat atrophy is a basic mechanism of enophthalmia, and its relation to linear scleroderma has already been reported.78 The thickening of the orbital muscles in this case can probably be secondary to orbital fat atrophy and fibrosis.2

Although the imaging findings are not specific, important information can be obtained. A precise localisation of the orbital abnormalities, a clinicoradiological correlation and the follow up of the abnormalities can be achieved.

The CT findings are non-specific.9 In this patient the lesion is of the chronic and infiltrative or diffuse type. It does not invade or distort the globe and there is no bone erosion.56

MRI findings in patients with a non-specific orbital inflammatory condition have been described as hypointense to fat and isointense to muscle on T1 weighted images. On T2 weighted images the abnormalities were isointense or slightly hyperintense to fat in nine out of 10 cases.10 We found similar signal characteristics in this patient. These appearances were markedly different from those in other diseases, including malignancy and haematoma, which appeared markedly hyperintense to fat on T2 weighted images.10The use of gadopentetate dimeglumine was helpful in assessing the compression by the lesion on the optic nerve.

To our knowledge, this is the first report on follow up of localised linear scleroderma by MRI. The abnormalities on MRI correlate with the clinical progression. MRI might be better to distinguish between orbital fat atrophy, oedema, and fibrosis but more experience is needed to know if MRI can distinguish the type and the activity of an inflammatory condition.