Discriminatory power of trabeculectomy bleb internal reflectivity and morphology in surgical success using anterior segment optical coherence tomography

Abstract

Background: The post-operative evaluation of trabeculectomy blebs has traditionally relied on subjective clinical grading systems performed at the slit-lamp. This study explores the use of swept source anterior-segment optical coherence tomography (AS-OCT) to objectively measure bleb internal reflectivity and morphology, and to distinguish blebs with surgical success vs. failure.

Methods: Cross-sectional study of patients with glaucoma who had undergone trabeculectomy at least one year prior. Swept source AS-OCT was used to capture filtering blebs in the sagittal plane. Standardised regions of interests on the sagittal plane were segmented, and pixel intensity values and bleb height were measured. Receiver operating characteristic curves were used to examine the discriminatory ability of pixel intensity values and bleb morphology to classify blebs with surgical success or failure.

Results: 100 eyes of 65 patients were included, with a median post-operative follow up of 7.0 years (IQR 3.2-16 years). The proportion of complete success, qualified success and failure was 45%, 33%, and 22% respectively. The maximum bleb height was significantly greater in the blebs with complete success (1.74 vs. 1.25 vs. 1.23 mm in CS vs. QS vs. F, one-way ANOVA, p < 0.0001). Mean pixel intensity was significantly lower in blebs with complete success (150.8 vs. 157.4 vs. 167.4 in CS vs. QS vs. F, p = 0.0001). Bleb intensity standard deviation (AUC 0.81), maximal bleb height (AUC 0.76), mean pixel intensity (AUC 0.75) and minimum pixel intensity (AUC 0.75) offered the best discrimination between surgical success and failure.

Conclusions: Swept-source AS-OCT can be used to quantify bleb internal reflectivity and morphology, which can be used to distinguish between well vs. poorly functioning blebs. These parameters may assist surgeons in the objective evaluation of post-operative bleb outcomes.

Keywords: Glaucoma surgery; Mitomycin-C; Scleral flap; Sclerostomy; Surgical success; Surgical technique; Trabeculectomy.

Fig. 1

Enface (A and D) and sagittal AS-OCT images of a well-functioning trabeculectomy bleb with the overlying raster block shown in A and D, anatomical landmarks labelled in B and E, and surgical parameters of interest in C and F. The yellow line in panel A represents the sagittal slice overlying the sclerostomy and peripheral iridotomy, while the yellow lines in panel D represent the sagittal slice overlying the sclerostomy and mid-point of the scleral flap. Abbreviations: Anterior chamber (AC), Peripheral iridotomy (PI)

Fig. 2

Standardised region of interest (ROI) measuring 400 by 800 pixels on the sagittal AS-OCT slice overlying the peripheral iridotomy and bisecting the scleral flap. This ROI is used for subsequent automated quantification of pixel intensity values

Fig. 3

Box and Whisker plots (median, interquartile range and 5th to 95th percentile) and results of one-way ANOVA of bleb intensity values (mean, minimum, maximum, standard deviation and solidity of pixel intensity) and sagittal bleb area across the outcome groups of complete success (CS), qualified success (QS) and failure (F). Ns = non-significant. Asterisks denote statistical significance [ie. p < 0.05 (*), < 0.01 (**), < 0.001 (***), < 0.0001 (****)]

Fig. 4

Receiver operating characteristic curves of standard deviation (SD) of pixel values, mean intensity, minimum intensity and bleb height to classify blebs with surgical success vs. failure

Fig. 5

Box and Whisker plots (median, interquartile range and 5th to 95th percentile) and results of one-way ANOVA of scleral flap dimensions, sclerostomy dimensions and distance of the sclerostomy from the iridocorneal angle across the outcome groups of complete success (CS), qualified success (QS) and failure (F). Ns = non-significant. Asterisks denote statistical significance [ie. p < 0.05 (*), < 0.01 (**), < 0.001 (***), < 0.0001 (****)]

Fig. 6

Representative blebs classified as complete success (A), qualified success (B) and failure (C) with associated global mean pixel intensity values, intraocular pressure (IOP), number of medications and maximal bleb height shown. Note the eye with bleb C underwent secondary tube shunt surgery, therefore bleb C was defined as having failed