Characterisation of Schlemm's canal cross-sectional area

Abstract

Purpose: To compare three methods of Schlemm's canal (SC) cross-sectional area (CSA) measurement.

Methods: Ten eyes (10 healthy volunteers) were imaged three times using spectral-domain optical coherence tomography (Cirrus HD-OCT, Zeiss, Dublin, California, USA). Aqueous outflow vascular structures and SC collector channel ostia were used as landmarks to identify a reference location within the limbus. SC CSA was assessed within a 1 mm segment (±15 frames of the reference, 31 frames in all) by three techniques. (1) Using a random number table, SC CSA in five random frames from the set of 31 surrounding the reference were measured and averaged. (2) The most easily visualised SC location (subjective) was measured, and (3) SC CSA was measured in all 31 consecutive B-scans, and averaged. (comprehensive average, gold standard). Subjective and random CSAs were compared with the comprehensive by general estimating equation modelling, and structural equation modelling quantified agreement.

Results: The average from five random locations (4175±1045 µm(2)) was not significantly different than that obtained from the gold standard comprehensive assessment (4064±1308 µm(2), p=0.6537). Subjectively located SC CSA (7614±2162 µm(2)) was significantly larger than the comprehensive gold standard SC CSA (p<0.0001). The average of five random frames produced significantly less bias than did subjective location, yielding a calibration line crossing the 'no-bias' line.

Discussion: Subjectively located SC CSA measurements produce high estimates of SC CSA. SC assessed by measuring five random locations estimate CSA was similar to the gold standard estimate.

Keywords: Aqueous humour; Imaging.

Figure 1

Schlemm's canal reconstruction. Cropped processed image showing a cross section of Schlemm's canal without (A) and with (B) the canal traced (yellow). The cross-sectional area of Schlemm's canal was measured as the area of the traced region. The circularity computed for this section is shown on the bottom right. Panel C illustrates the method to measure the variations in Schlemm's canal area using sequential cross sections.

Figure 2

(Top) Histogram of cross-sectional area (CSA) measurements as the number of measurements. (Bottom) Histogram of CSA B-scan to B-scan change calculations as the total number of calculations.

Figure 3

Measurements of Schlemm's canal (SC) cross-sectional area made at a single subjectively identified location are representative of, but larger than, those obtained in a comprehensive assessment of all B-scans within a 1 mm circumferential segment of SC.

Figure 4

Measurements of Schlemm's canal cross-sectional area made at five randomly selected locations are representative of those obtained in a comprehensive assessment with minimal bias.

Figure 5

Three dimensional visualisation of Schlemm's canal (SC) (Bottom) shows that two nearly adjacent B-scans (blue lines show approximate locations of the B-scans on top) may have radically different cross-sectional areas. The two B-scans (top) are separated along the circumferential direction of SC by a distance of 23 μm only.