Jerry A. Shields

Conjunctival Lymphoid tumors: clinical analysis of 117 cases and relationship to systemic lymphoma

OBJECTIVE To investigate the clinical features of conjunctival lymphoid tumors and factors predictive of systemic lymphoma. DESIGN Retrospective, observational case series. SETTING Clinical practice of ocular oncology. PARTICIPANTS The participants included 117 consecutive patients with lymphoid tumors of the conjunctiva treated at the Oncology Service of Wills Eye Hospital between 1974 and 1999. MAIN OUTCOME MEASURE The main outcome measure was the development of systemic lymphoma. Cox proportional regression models were used to calculate the risk of eventual systemic lymphoma. Kaplan-Meier survival estimates were used to analyze the development of systemic lymphoma as a function of time. RESULTS There were 55 males (47%) and 62 females (53%); 110 patients (94%) were white and 7 patients (6%) were African AMERICAN: The mean age at ocular presentation was 61 years, and the conjunctival lymphoid infiltrate was unilateral in 72 patients (62%) and bilateral in 45 patients (38%). In 8 patients (7%), initial unilateral conjunctival disease evolved into bilateral involvement over a mean of 32 months. Additional sites of ocular involvement were found in 27 patients (23%) and included lymphoid tumor in the eyelid in 3 cases, orbit in 18, choroid in 5, and vitreous in 1. Systemic lymphoma was known to exist before ocular diagnosis in 16 patients; (14%; for a mean of 51 months) and was found subsequent to ocular diagnosis in 20 patients (17; at a mean of 21 months). Therefore, of 117 patients with conjunctival lymphoid infiltration, 36 (31%) had or eventually developed systemic lymphoma, and 81 (69%) did not manifest systemic lymphoma during the mean follow-up of 38 months. Of the 64 patients with unilateral conjunctival involvement, 11 (17%) manifested systemic lymphoma, and of the 53 patients with bilateral involvement, 25 (47%) manifested systemic lymphoma. By univariate analysis, the clinical factors at date first seen predictive of the presence or development of systemic lymphoma included location of the tumor at an extralimbal site (fornix or midbulbar conjunctiva; P = 0.02) and increasing number of conjunctival tumors (P = 0.02). Using Kaplan-Meier life table analysis of those 101 patients who had conjunctival lymphoid tumor(s) and no evident systemic lymphoma at presentation, systemic lymphoma was eventually discovered in 7% of patients at 1 year, 12% at 2 years, 15% at 5 years, and 28% at 10 years. Overall, only one patient (<1%) died of systemic lymphoma, at 28 months after the diagnosis of the ocular disease. CONCLUSIONS Lymphoid tumors of the conjunctiva are associated with systemic lymphoma in 31% of patients. Systemic lymphoma is found more often in those patients with forniceal or midbulbar conjunctival involvement and in those with multiple conjunctival tumors. Long-term systemic follow-up is advised, because related systemic lymphoma can manifest many years later.

Conjunctival Melanoma: Outcomes Based on Tumor Origin in 382 Consecutive Cases

PURPOSE To evaluate prognostic factors based on origin of conjunctival melanoma. DESIGN Interventional case series. PARTICIPANTS Three hundred eighty-two consecutive patients. METHODS Retrospective chart review. MAIN OUTCOME MEASURES Melanoma-related metastasis and death. RESULTS The melanoma arose from primary acquired melanosis (PAM; n = 284; 74%), from pre-existing nevus (n = 26; 7%), and de novo (n = 72; 19%). The mean tumor base was 11 mm for melanoma arising from PAM, 6 mm for melanoma arising from nevus, and 10 mm for those arising de novo. At 5 years (10 years), melanoma metastasis occurred in 19% (25%) in melanoma arising from PAM (P = 0.003), 10% (26%) in melanoma from nevus (P = 0.193), and 35% (49%) in those de novo. Factors predictive of metastasis by multivariable analysis included tumor origin de novo (P = 0.001), palpebral location (P<0.001), nodular tumor (P = 0.005), and orbital invasion (P = 0.022). At 5 years (10 years), melanoma-related death occurred in 5% (9%) in melanoma arising from PAM (P<0.001), 0% (9%) in melanoma arising from nevus (P<0.057), and 17% (35%) in those arising de novo. Factors predictive of death by multivariable analysis included tumor origin de novo (P<0.001), fornix location (P = 0.04), and nodular tumor (P = 0.001). CONCLUSIONS Melanoma arising de novo carries a higher risk of melanoma-related metastasis and death compared with those cases arising from PAM or nevus.

Surgical management of circumscribed conjunctival melanomas

Summary: Circumscribed conjunctival melanoma usually arises in the bulbar conjunctiva and less often in the forniceal or palpebral conjunctiva. After simple superficial removal, employed by many ophthalmologists, these tumors have an increased tendency toward local recurrence and distant metastasis. A surgical procedure designed to remove the tumors completely and minimize the chances of recurrence would be desirable. The authors employed a surgical approach to conjunctival melanoma excision, which they believe insures more complete tumor removal and decreases the chances of recurrence and metastasis. The surgical management of melanoma in the limbal region of the bulbar conjunctiva consists of localized alcohol epitheliectomy, removal of the mass by a partial lamellar scleroconjunctivectomy, and supplemental double freeze-thaw cryotherapy to the adjacent remaining conjunctiva by a specific technique. For tumors located in the forniceal or palpebral conjunctiva, wide surgical resection with alcohol treatment to the scleral base and cryotherapy to the surrounding conjunctiva is performed. A “no touch” technique is employed and direct manipulation of the tumor is strictly avoided in an effort to prevent tumor cell seeding into a new area. The technique currently employed has evolved from experience with circumscribed conjunctival melanoma excision during a 20-year period. About 80 patients had circumscribed conjunctival melanoma unassociated with appreciable primary acquired melanosis. Although it is not the purpose of this article on surgical technique to provided a detailed statistical analysis of the results, the authors currently believe that this technique should be employed in all cases of circumscribed lesions in which conjunctival melanoma is a diagnostic consideration. Incisional biopsy and frozen sections are generally not advisable. Preliminary observations suggest that this method decreases the chances of local recurrence.