Identification of ocular surface squamous neoplasia by in vivo staining with methylene blue Jonel Steffen, 1 James Rice, 1 Karin Lecuona, 1 Henri Carrara 2 1 Division of Ophthalmology, University of Cape Town, Cape Town, Western Cape, South Africa 2 School of Public Health and Family Medicine, University of Cape Town, Cape Town, Western Cape, South Africa Correspondence to Dr Jonel Steffen, Division of Ophthalmology, University of Cape Town, Ward D4, Groote Schuur hospital, Private Bag, Observatory, Cape Town, Western Cape 7937, South Africa; j.steffen@uct.ac.za Received 11 July 2013 Revised 18 September 2013 Accepted 23 September 2013 Published Online First 24 October 2013 To cite: Steffen J, Rice J, Lecuona K, et al. Br J Ophthalmol 2014;98: 13–15. ABSTRACT Aim To evaluate the diagnostic accuracy of methylene blue used as a non-invasive in vivo stain to detect ocular surface squamous neoplasia (OSSN). Methods A test validation study was performed according to Standards for the reporting of diagnostic accuracy studies (STARD) guidelines on 75 consecutive patients who presented with ocular surface lesions suspicious of OSSN. Methylene blue 1% was instilled in vivo following local anaesthetic. Stain results were documented photographically and read by an independent observer. Lesions were excised at the same visit and evaluated histologically by pathologists who were blind to the stain results. Sensitivity, specificity, positive and negative predictive values were determined. Results Thirty-three patients had histologically malignant lesions, of which 32 stained with methylene blue, and 42 patients had benign or premalignant lesions, of which 21 stained with methylene blue. Methylene blue had a sensitivity of 97%, specificity of 50% and positive and negative predictive values of 60% and 96%, respectively. Conclusions The topical application of methylene blue is a simple, inexpensive, non-invasive diagnostic test that can be helpful in excluding malignant ocular surface lesions but cannot replace histology as gold standard for diagnosis of OSSN. INTRODUCTION Ocular surface squamous neoplasia (OSSN) is a spectrum of neoplastic disease of the conjunctiva or cornea ranging from dysplasia to invasive squa- mous cell carcinoma. 1 The condition was originally described in elderly Caucasian men 23 living in areas of high ultraviolet light exposure. 4 More recently, it has become more common in younger patients, especially in regions of high HIV prevalence such as sub-Saharan Africa. 5–7 OSSN typically presents as a slightly elevated, variably shaped lesion at the limbus in the interpal- pebral fissure. It tends to have large feeding blood vessels, grow rapidly and may have a leukoplakic, gelatinous or papillomatous appearance. 1 Small OSSN lesions may be difficult to distinguish clinic- ally from other benign lesions such as pinguecula and pterygia. 1 This necessitates excision biopsy of all small lesions suspicious of OSSN, leading to an increased surgical burden in resource-poor areas of the world where the prevalence of OSSN is increasing. 8 A number of non-invasive diagnostic tests have been evaluated in an attempt to identify which lesions require excision. These include impression cytology, 9 10 confocal microscopy, 11 ultra-high resolution optical coherence tomography 12 and in vivo toluidine blue stain. 13 These techniques are either expensive or require highly specialised per- sonnel making them unavailable in small units and developing countries. A simple, inexpensive and non-invasive diagnostic technique would therefore be helpful. Methylene blue is an acidophilic dye that is inex- pensive, widely available and safe to use on the eye. 14 15 It has been used as an in vivo stain to detect oral squamous cell carcinoma, 16 gastric car- cinoma 17 and bladder carcinoma, 18 as well as malignant changes in Barrett’s oesophagus 19 and ulcerative colitis. 20 The aim of our study is to determine the diagnostic accuracy of methylene blue 1% used as an in vivo stain to detect OSSN. METHODS A prospective diagnostic test validation study was conducted on 75 eyes of 75 consecutive patients who presented with previously untreated ocular surface lesions suspicious of OSSN to the Department of Ophthalmology, Groote Schuur hos- pital. The study was conducted between October 2009 and December 2011 and was approved by the Human Research Ethics Committee of the University of Cape Town. The Standards for the reporting of diagnostic accuracy studies (STARD) guidelines were followed in the design, conduct and reporting of the study. 21 Informed consent was obtained from each enrolled patient. HIV counsel- ling and testing was offered to all patients. All eyes underwent complete anterior segment examination. A topical anaesthetic, followed by one drop of methylene blue 1%, was instilled in the eye and remained in contact with the lesion for approximately 30 s. The eye was then rinsed with sterile water, and the amount of staining was docu- mented photographically. Photographs were reviewed by an independent observer who was blinded to the histology results. Any uptake of dye on the lesion, whether diffuse or focal, was docu- mented as a positive stain ( figure 1). All eyes underwent excisional biopsy of the lesion with the aid of an operating microscope on the same day. Specimens were oriented with respect to anatomical location and were submitted for histological examination by pathologists who were blinded to the stain results. Histological outcomes were classified as (1) malignant, which included severe dysplasia (carcinoma in situ) and invasive squamous cell carcinoma; and (2) benign and pre- malignant, which included mild dysplasia, moder- ate dysplasia, pterygium and pingueculum. Sensitivity, specificity, positive and negative pre- dictive outcomes were determined. Stata V.12 was Steffen J, et al. Br J Ophthalmol 2014;98:13–15. doi:10.1136/bjophthalmol-2013-303956 13 Global issues