Epibulbar Seeding at the Site of a Transvitreal Fine-Needle Aspiration Biopsy

Transocular fine-needle aspiration biopsy (FNAB) with analysis of the aspirate is a diagnostic technique that can be helpful in selected cases of suspected intraocular tumors.¹ In most cases, the improvement in clinical diagnostic imaging permits an accurate diagnosis,² but in selected cases, histological confirmation is necessary to rule out other malignancies. The FNAB was introduced into ophthalmology by Schyberg in 1975 for the diagnosis of orbital neoplasm.³ In 1979, Jakobiec et al⁴ proposed the use of FNAB in the evaluation of intraocular tumors. Since then, this technique has been used in a large number of patients without any evidence of local or systemic spread. In this article, we report the first described case of a local epibulbar seeding at the scleral pars plana puncture site after a transvitreal FNAB.

A 61-year-old man complained of foreign body sensation and progressive visual loss in his left eye for the last 4 months. His medical history was remarkable for a bladder carcinoma that had been treated 6 years previously with transurethral resection and intravesical bacillus Calmette-Guérin therapy. At the time of admission, his visual acuity was 20/20 OD and 1/20 OS. Dilated fundus examination disclosed a temporal inferior amelanotic mass with an associated serous detachment (Figure 1). Ultrasonography revealed a dome-shaped postequatorial choroidal mass. A B-scan demonstrated a moderate to high acoustic solidity without appreciable choroidal excavation or orbital shadowing (Figure 2A). An A-scan showed a high initial spike and median to high internal reflectivity (Figure 2B). The tumoral base was 16 × 18 mm and the maximal elevation was 6.2 mm. The distance to the optic disc was 1 mm. Although clinical appearance was compatible with an uveal melanoma, a metastatic carcinoma could not be completely ruled out. Systemic evaluation did not reveal more clinical findings or distant dissemination. An FNAB of the tumor was performed to confirm the diagnosis.

Under retrobulbar anesthesia, we placed an infusion with fiber optic light at the inferior temporal pars plana. A scleral puncture was then performed at the superior nasal pars plana using a 25-gauge needle attached to a flexible plastic tube, which was connected to a 10-mL syringe. The needle was guided using the operating microscope and a retinal wide field view system. The needle was passed into the lesion and gentle aspiration was performed. Finally, the suction was released before the needle was removed without any complication except a local hemorrhage.

The cytological diagnosis confirmed the presence of atypical melanotic cells. The Papanicolaou stain of the aspirate showed highly pleomorphic cells with round nuclei, prominent nucleoli, and a different nuclei-cytoplasm ratio (Figure 3A). Immunohistochemical stain revealed an intense positivity for S100 and HMB-45 (Figure 3B) consistent with melanoma.

A 22-COMS (Collaborative Ocular Melanoma Study) iodine 125 plaque was used for treatment. The total activity of the plaque was 69.60 μCi. The total dose was 85.29 Gy with a dose rate of 92.71 Gy per hour at the tumoral apex. The duration of treatment was 92 hours. Eight months later, the patient consulted for a small and painless orange subconjunctival mass on the superior nasal quadrant at the location of the pars plana puncture (Figure 4). A complete excision was made with triple freeze-thaw cryotherapy application on the scleral surface. The histopathologic examination showed a subconjunctival mass (Figure 5A) with closely packed cells with numerous mitoses, prominent nucleoli, and melanic pigment (Figure 5B) compatible with a local melanoma growth in the puncture area. The histological margins were free.

Transvitreal FNAB is an invasive procedure useful in the diagnosis of intraocular lesions. Although in most cases clinical findings permit an accurate diagnosis,¹ sometimes histological confirmation is required. Important limitations of this technique are the difficulty in obtaining sufficient material for cytological examination,¹ theoretical local complications, and spread of the tumor. On the other hand, few complications have been described.¹ A local hemorrhage in the site of biopsy, in almost all cases, is controlled by increasing the intraocular pressure. Vitreous and subretinal hemorrhage occurred in 9% to 13% of cases with transient visual loss. Other theoretical problems, such as retinal detachment, local recurrences, or systemic dissemination from tumor seeding, have not been previously described.

In nonocular tumors, iatrogenic dissemination of tumoral cells along the needle track has been found immediately after an FNAB, but only a few cases of local seeding with metastasis have been reported.⁵ Ryd et al⁶ demonstrated local seeding risk in an experimental evaluation with solid and ascitic-growing tumors in mice, where almost 10² to 10⁴ cells contaminated the needle track in FNAB. The high rate of seeding demonstrated in this study contrasts with the low percentage of local clinical recurrences described in the literature. In experimental studies, the transplanted inoculum probably has a higher proportion of viable malignant tumoral cells and is more aggressive than human tumors.

In contrast, no cases of local recurrences after FNAB in ocular tumors have been described before. This fact had been explained by Jakobiec et al⁴ because the aspiration is through the vitreous or the aqueous humor. This indirect transocular approach from the opposite side of the lesion would make a mechanical washout of tumoral cells from the needle surface. However, several studies^7,8 have demonstrated tumoral seeding cells after transscleral and transvitreal FNAB in enucleated tumor eyes. Glasgow et al⁷ revealed that a transvitreal approach contained significantly fewer cells than transscleral aspirates. The few malignant cells along the needle track seem to disappear spontaneously and could be insufficient for tumor growth.⁹ In addition, experimental animal studies have shown the difficulty in transplanting tumor cells in vitreous,¹⁰ and most of the malignant ocular tumors are treated by irradiation or excision after diagnosis, which probably prevents implantation of metastasis in most cases.

In conclusion, transvitreal FNAB is a useful and safe tool in clinical diagnosis based on previous reports. However, there is a potential risk of vascular dissemination and a real risk of local malignant seeding as described in this report. For this reason, this technique should only be used when there is a reasonable doubt about the diagnosis and when the expected results may modify management of the lesion.

Correspondence: Dr Caminal, Hospital de Bellvitge, Ophthalmology Department, calle 18, Feixa Llarga s/n, 08907 L’Hospitalet de Llobregat, Barcelona, Spain (jmcaminal@csub.scs.es).

Submitted for Publication: September 8, 2004; final revision received February 9, 2005; accepted February 15, 2005.

Financial Disclosure: None.