Intraocular surgery with melphalan irrigation for vitreous hemorrhage in an only eye with retinoblastoma.

PURPOSE: To illustrate the successful outcome of pars plana vitrectomy (PPV) with melphalan irrigation for vitreous hemorrhage (VH) with suspected viable retinoblastoma. Despite the high risk of intraocular surgery, it was performed to preserve the only potentially seeing eye with treated retinoblastoma. METHODS: Vitreous hemorrhage occurred in the only eye of a 4-year-old boy after treatment for recurrent multifocal group C retinoblastoma with systemic chemotherapy (carboplatin, etoposide, and cyclophosphamide; and vincristine, cyclophosphamide, and doxorubicin), ruthenium brachytherapy with plaque repositioning, cryotherapy, and external radiotherapy. The VH developed 8 months after repeated brachytherapy with subsequent intravitreal melphalan chemotherapy. The patient's parents refused to remove the eye. The fellow eye was enucleated earlier because of VH and secondary glaucoma, without histologic signs of a viable tumor. Pars plana lensectomy, 25-G vitrectomy with melphalan irrigation (5 µg/mL), and silicone oil tamponade were performed. RESULTS: No ophthalmoscopic or morphologic signs of a viable tumor were detected. Four months later, the silicone oil was removed. Visual acuity was 20/200 with aphakic correction. Follow-up for 34 months revealed no signs of tumor recurrence or dissemination. CONCLUSIONS: Despite the high risk of intraocular surgery, the need to preserve the only potentially seeing eye with treated retinoblastoma may require PPV. Thus, in unclear cases of VH with suspected viable tumor, PPV with intraocular melphalan irrigation, with caution, may be a reasonable procedure.

CORE needle biopsy of orbital tumors.

BACKGROUND: In some orbital mass lesions, histology may be the only way to establish an accurate diagnosis. Core needle biopsy (CNB) is widely performed in oncology, and unlike fine needle aspiration biopsy (FNAB), it can provide sufficient tissue sample for histology and immunohistochemistry. The purpose of this study was to evaluate the usefulness and possible complications of CNB of orbital tumors. METHODS: Fifty orbital lesions from 46 patients (age: 2-84 years) were biopsied using 20-gauge and 18-gauge semi-automated needles; 11 procedures were performed under ultrasound guidance. All 50 samples received routine histopathological examination and immunohistochemical analysis. RESULTS: Specimens diagnostically sufficient for histological analysis were obtained in all biopsies. The histopathological diagnosis was established in 94% of specimens: 30 tumors were malignant: lymphoma (20); rhabdomyosarcoma (six); lacrimal gland carcinoma (one); breast carcinoma metastasis (two); melanoma (one); three were benign; ten inflammatory; and four were orbital fibrosis. Three biopsies were nondiagnostic: two (lymphoma, angiofibroma) showed undetermined identification and one was a false-negative (lacrimal gland adenocarcinoma was misdiagnosed as fibrous tissue). The diagnoses were confirmed by excisional or incisional biopsy in 26 patients. The concordance rate in patients with a surgically confirmed diagnosis was 88%. The sensitivity, specificity and accuracy for differentiating malignant from benign lesions were 94%, 100%, and 96%, respectively. Two patients suffered complications: mild retrobulbar hematoma with no decreased visual acuity. There was no damage to the globe or optic nerve, motility disorder, or infection. CONCLUSIONS: CNB of orbital mass lesions is a safe procedure that provides a sufficient amount of tissue for histology and immunohistochemistry, and may be useful for improving the quality of the pathological diagnosis of orbital tumors.