Abstract
Purpose: :
To determine the accuracy of Optical Coherence Tomography (OCT) in delineating margins of periocular Basal Cell Carcinomas (BCC). This will enable surgeons to completely excise tumor with minimal loss of healthy tissue.
Methods: :
In three excised BCCs the margins were identified using OCT and conventional histopathological examination and the results compared. A novel dual wavelength en face OCT/confocal system was assembled. B scan (cross section) and C scan (en face) OCT images were generated using wavelengths of 670nm and 1300nm. Lateral resolution was 15µm and depth resolution was 14µm at 670nm and below 9µm at 1300nm. The field of view ranged between 9 and 26 degrees. The axial position was selected with a precision of 1µm. Stacks of C–scan images were obtained at regular intervals of medial(m), lateral(l), superior(s) and inferior(i) edges of the tissue. A series of B scan images was taken along the width of the excised lesion. mTM, lTM, sTM, iTM was measured by superimposing a 1mm calibration stick onto one good quality C scan image of the stack for each orientation. Tumor depth (TD) was calculated using the B scan image showing the deepest margin of tumor. Quality was determined by 2 independent observers taking into consideration the contrast of the neighbouring confocal scan, the amount of tumor present in the field of view and the image quality affected by reflectivity and birefringence of the tissue. Following OCT imaging, histopathological slides of the excised tissue were prepared in axial sections at 4µm cuts and TM was determined using conventional optical microscopy.
Results: :
The mean difference in TM as measured by the 2 techniques is 0.13mm for mFM, lFM, sFM, iFM and TD. There is a strong correlation between the TM in all orientations as measured by the 2 techniques (Pearson correlation coefficient=0.95).
Conclusions: :
Modified OCT image mapping of BCCs appears at least equivalent to histopathological mapping. In some cases the higher resolution OCT imaging can accurately map clear margins within 0.1mm of the tissue edge, thereby surpassing the accuracy of conventional histopathology. This is an exciting new clinical application of a modified OCT machine with possibilities for one stage excision–reconstruction of eyelid BCCs. The surgeon could follow the 3D reconstructed image to accurately and selectively excise tumour tissue once the machine is further modified for in vivo imaging.
Keywords: imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • tumors • eyelid