|Year : 1993 | Volume
| Issue : 3 | Page : 114-116
Frozen section diagnosis in ophthalmic pathology
Jyotirmay Biswas, Nirmala Subramaniam
Medical Research Foundation, Madras, India
Ocular Pathologist, Medical Research Foundation, 18 College Road, Madras - 600 006
Source of Support: None, Conflict of Interest: None
Frozen section diagnosis is extensively used in various branches of pathology, but its application in ophthalmic pathology was recognised only in the 1970s. We studied 10 sections of ocular and adenexal lesions by frozen section diagnosis, which included orbital lesions (4 cases), lid lesions (3 cases), and intraocular tissue (1 case). The time taken for processing ranged between 10 to 15 minutes. Diagnoses based on frozen section evaluation included lymphoma, mesenchymal chondrosarcoma, solar keratosis, compound naevus, silicone oil globules in cataractous lens, neurofibromatosis, pseudotumour, retinoblastoma, and chronic blepharitis. Although further histopathologic examination correlated well with the frozen section (100%) observations, the diagnosis was deferred in the case of naevus and reactive lymphoid hyperplasia. Our study shows that frozen section diagnosis in ophthalmic surgery is quite reliable and is particularly useful in ocular adenexal lesions
Keywords: Frozen section - Lymphoma - Mesenchymal chondrosarcoma - Solar keratosis - Compound naevus - Pseudotumour - Retinoblastoma - Neurofibromatosis - Chronic blepharitis.
|How to cite this article:|
Biswas J, Subramaniam N. Frozen section diagnosis in ophthalmic pathology. Indian J Ophthalmol 1993;41:114-6
Diagnosis by frozen section method is commonly employed in all surgical specialities.  Though first used in 1891, its application in ophthalmic pathology commenced only in the 1970s.  In recent years, however, there has been a significant increase in its application in ophthalmic surgery.. This technique is used to make reliable identification of the pathological nature and extent of lesions within a period of 15 to 20 minutes. Special stains for fat and certain immunohistochemical studies, particulary immunofluorescence, can also be performed on frozen sections.
| Materials and methods|| |
Ten specimens of ocular and adenexal lesions, including four orbital, four lid, one lens and one optic nerve lesion, were studied by frozen section method.
Whenever a frozen section diagnosis was required, the pathologist, having prior information on the clinical details of the case, received the specimen in the operation theatre itself. The specimen was then immediately brought to the laboratory, where it was placed on the object holder platform and snap frozen. Sections were taken rapidly, mounted on albuminised slides, and stained with hematoxylin and eosin technique which took about three to five minutes. The slides were immersed by several dips in solutions contained in staining jars in the order listed below : Xylene - Xylene - Absolute alcohol - Absolute alcohol - Absolute alcohol - Tap water - Hematoxylin (1 minute) - Tap water - Acid alcohol - Tap water - Ammonia water - Eosin (30 seconds) - Tapwater - 95% alcohol - Absolute alcohol - Xylene.
The material was then mounted on a slide and a cover slip was placed over it. The time required for complete processing was about 15 minutes.
| Results|| |
The results of frozen section examination were later confirmed by permanent paraffin sections. In our study, though the diagnosis was accurate in all ten specimens (100% correlation), diagnosis was deferred in the case of naevus and reactive lymphoid hyperplasia. The few difficulties encountered during the procedure included pale staining, ill-defined shape of nuclei, and inferior quality of the sections as compared to the corresponding paraffin sections. However, an experienced ophthalmic pathologist aware of such artifactitious changes can accurately interpret frozen sections without much difficulty in most cases.
Clinical presentations, frozen section and permanent section diagnosis of specimens examined are given in [Table - 1].
In case 1, although the presence of monomorphic lymphoid infiltrate strongly indicated the possiblity of lymphoma, diagnosis was deferred. A final diagnosis was arrived at after immunoperoxidase study on permanent paraffin sections using antibody for common leucocyte antigen, Pan T cell marker (UCHL 1) and Pan B cell marker (L 26).
In case 2, fine needle aspiration biopsy was done prior to frozen section study, which revealed mature chondrocytes indicating a possibility of chondro-sarcoma. Moreover, a CT scan indicated the possibility of chondrosarcoma. Cryostat sections showed islands of well-differentiated cartilage surrounded by undifferentiated mesenchymal cells which concurred with the picture of a typical mesenchymal chondrosarcoma.
| Discussion|| |
In recent years, frozen section diagnosis has become universally accepted as a rapid and reliable means of diagnosis. The advantages are: (1) a provisional pathological diagnosis is possible within 20 minutes whereas conventional processing would take 48 hours; (2) the procedure does not require any special equipment except the cryostat (transport time is minimised by keeping the cryostat close to the operating room); (3) the processing, sectioning, and staining is easily handled in any surgical pathology laboratory by histopathology technicians familiar with routine paraffin processing; (4) accurate assessment of tumour involvement at the resection margin, particularly in orbital and adenexal tumours ; and (5) in intraocular tumours such as retinoblastomas, the extent of optic nerve involvement by the tumour can be determined while the patient is still in the operating room.  If frozen section suggests incomplete removal, further excision can be done in the same sitting and frozen section repeated to ensure tumour-free margin.
The application of immunohistochemical methods adds a greater degree of specificity in morphologic and histochemical study in general pathology. This can be done either by the immunofluorescent technique wherein a fluorescein-labelled antibody is used or by immunoperoxidase stains using an enzymeconjugated antibody. The immunofluorescent technique can be applied only on cryostat sections because changes due to chemical fixation cause significant differences in antigenicity, with a corresponding loss of specificity. Immunoperoxidase technique works mainly on formalin-fixed paraffin-embedded tissue sections, but certain monoclonal antibodies, for example, those for T helper (CD 4) cells, T suppressor cells (CD 8) work on frozen sections due to better preservation of antigenicity. Frozen sections can also be used for staining fat, using a special staining procedure, oil-red-O. For example, it can be used to establish a diagnosis of sebaceous gland carcinoma where malignant tumour cells stain positive with oil-red-O stain. sub In case 7, oilred-O stain was done on frozen sections so as to confirm the presence of silicone oil globules in the lens tissue.
Our study shows that frozen section diagnosis in ophthalmic surgery is quite reliable and is particularly useful in ocular adenexal lesions. Frozen sections also lend themselves to several special procedures such as immunohistochemical staining using the immunofluorescent or immunoperoxidase techniques which use certain antibodies that require only fresh frozen tissue. Such special studies using frozen sections increase the diagnostic efficacy of pathologic examinations in Ophthalmology.
| References|| |
Sawady J, Berner JJ, and Siegler EE. Accuracy of and reasons for frozen sections - A correlative retrospective study. Human Pathology. 19: 1019-1023, 1988.
Karcioglu ZA, and Caldwell DR. Frozen section diagnosis in ophthalmic surgery. Surv. Ophthalmol. 28: 323-332, 1984.
Frank HJ. Frozen section control of excision of eyelid basal cell carcinomas : 8 1/2 years' experience. Br J Ophthalmol.73: 328-332, 1989.
Karcioglu ZA, Haik BG, and Gordon RA. Frozen section of the optic nerve in retinoblastoma surgery. Ophthalmology. 95:674-676, 1988.
Rao NA, Hidayat AA, Mclean IW, and Zimmerham LB. Sebaceous carcinomas of the ocular adenexa. Human pathology. 13: 113-122, 1982.
[Figure - 1], [Figure - 2], [Figure - 3]
[Table - 1]