[Optical coherence tomography versus ultrasound biomicroscopy of conjunctival and eyelid lesions].

BACKGROUND

The value of ultrasound biomicroscopy in the diagnosis of conjunctival and eyelid lesions is still unknown.

METHOD: For the examination of conjunctival and eyelid lesions, we used an anterior segment OCT (4optics, Lübeck) and an ultrasound biomicroscope (Humphrey, Zeiss, Oberkochen) with high frequency transducer (30 MHz). Included in this study were all patients with conjunctival and eyelid lesions, which were consecutively operated on between December 2002 and January 2003. Histological examination of the excised tissue was performed.

RESULTS

38 tumours of 35 patients, aged 8 - 93 years, were examined. 13 tumours were conjunctival lesions and 25 tumours were eyelid lesions. Histological examination displayed the presence of a pterygium (8/38), a seborrhoic keratosis (7/38), a cyst of the eyelid (5/38), a basal cell carcinoma of the eyelid (4/38), a compound naevus of the conjunctiva (4/38), a chalazion (3/38), a primary acquired melanosis (1/38), an actinic keratosis (1/38), a naevus (1/38), a cavernous haemangioma (1/38), a melanoma in situ (1/38), a foreign body (1/38), and an epidermoid cyst (1/38). Using both ultrasound biomicroscopy and optical coherence tomography (OCT) we were able to demonstrate a cystic tumour in the five patients (13 %) with a cyst of the eyelid. With OCT and with ultrasoundbiomicroscopy in all patients with pterygium a slit was found and in patients with compound naevus very small cystic structures were seen, but with OCT the imaging was more reliable. In patients with solid tumours the definite diagnosis could not be differentiated by ultrasound biomicroscopy or OCT alone. Using OCT, assessment of the margins of the tumours (particularly in depth) was impossible or uncertain.

CONCLUSION

Compared to ultrasound biomicroscopy OCT is able to show very small cystic structures more distinctly. For assessing the margins of the tumour ultrasound biomicroscopy is the better tool.