Patients

All choroidal, retinal, or chorioretinal biopsies performed at Moorfields Eye Hospital over a 15-year period from 1999 to 2014 were identified on the hospital’s computerized database. Those where the diagnosis of lymphoma was suspected were included in this series. Previous series from this hospital have been described and overlap some of these cases. Cases of vitreous biopsy alone were not included in this series. This study received a prospective international review board approval for this retrospective, noncomparative, consecutive diagnostic case series (Research and Development, Moorfields Eye Hospital NHS Foundation Trust).

Vitritis Grading

Severity of vitritis is divided into 2 categories, termed “ minimal ” or “ marked .” The distinction is based on the 1985 classifications by Nussenblatt. In this article minimal includes score 0 and 1, that is, nil and minimal or grade 0 and trace. All other categories are pooled as marked. The demarcation clinically as minimal implies no obstruction to features of the vessel margins and marked implies any obstruction to observation of clinical features of the fundus, including the vessels.

Surgical Techniques

The surgical approach was determined by the anatomic location of the lesion to be biopsied, whether primarily retinal or choroidal. The surgical techniques performed to obtain the biopsy specimen were previously described. General anesthesia was performed in all cases. An external approach was used in 1 case while an internal approach was used in 28 cases. The only technical difference regarding the internal approach relative to previous descriptions was the use of 23 gauge vitrectomy instead of 20 gauge in the more recent cases. At the time of the removal of the specimen from the eye, 1 of the ports was enlarged. Some patients had a formal diagnostic vitrectomy, including the collection of both an undiluted sample at the beginning of the surgery and the contents of the vitrectomy cassette, which was centrifuged to concentrate the cells for analysis.

Laboratory Techniques

Intraocular biopsies tend to be smaller than other types of biopsy and not always amenable to conventional handling techniques. The following is an overview of the general techniques applied to the cases in this series, but there may have been slight variations for individual cases.

Tissue specimens (for histology) were placed in 10% neutral buffered formalin at the time of surgery. Following fixation they were examined macroscopically, placed in tissue cassettes, and processed using automated equipment (the processor in use changed over time). The blocks were embedded in paraffin wax and 4 μm sections were cut. All specimens had at least 1 section stained with hematoxylin-eosin for initial examination, with other histochemical stains (such as periodic acid-Schiff for basement membrane) being requested based on specimen type and clinical information.

Any vitreous (cytology) specimens submitted along with the tissue specimens were handled using a variety of methods, depending on factors such as (but not limited to) the following: (1) clinical information; (2) whether the specimen was submitted fresh (unfixed) or in formalin; (3) whether any particulate material was seen in the specimen receptacle; and (4) volume of material received. Generally speaking, particulate material was fixed and processed as tissue, smear cytology preparations were made from fluid received, and a proportion of any fresh vitreous received was frozen to allow the opportunity for subsequent molecular studies, volume permitting.

Any immunohistochemistry required was performed on a Dako Autostainer (Agilent Technologies, Denmark) or, more recently, a Leica Bond (Leica Microsystem, Germany) immunostainer using retrieval methods, antibody dilutions, and protocols appropriate to the required antibody.