A total of 75 healthy volunteers (28 men and 47 omen; mean age: 32.72 ± 10.28 years, range: 20–62 years) participated in this study. In order not to artificially reduce the CI around the limits of agreement (21), only one eye per participant was included (the right eye of the participants was measured if they were born in even months, and the left eye was measured if the participant was born in odd months). The study protocol adhered to the tenets of the Declaration of Helsinki and was approved by the Regional Ethical Committee (Regional ethics committee, Stockholm 2011/874-31/2). Written informed consent from all participants was obtained after explaining the purpose, nature, and the possible consequences of the study.

The inclusion criteria to participate in this study were best corrected visual acuity better or equal to 0.1 logMAR and no history of amblyopia; no history of any ocular surgery; normal anterior and posterior segment (i.e., no significant opacities, irregularities, or pathologies); and intraocular pressure below 21 mmHg.

The participants underwent OCT imaging with the Canon OCT HS-100 at the Optometry Clinic (Karolinska Institute, Solna, Sweden). The OCT HS-100 performs up to 70,000 A-scans/s with an axial resolution of 3 μm, with a maximum scan width of 13 mm. This instrument allows the possibility to image the macula and optic disc individually (with a maximum scan width of 10 and 6 mm, respectively), or simultaneously with a 13-mm scan width. Table 1 summarizes the specifications of each scan mode. The main differences between the two scan modes are the B-scan density and scan area. From this point forward, the individual image scan mode of the macula or optic disc will be referred to with the term individual scan, and the term combined scan will be used to refer to the scan mode in which the macula and optic disc are imaged simultaneously.

In this study, nine scans were acquired in each eye (three individual scans of the macula, three individual scans of the optic disc, and three combined scans). The repeated measurements were taken under repeatability conditions (22, 23), and with sufficient breaks in between to ensure good patient cooperation. The same experienced examiner performed all OCT scans, and these were repeated in case of poor fixation, subject blink, or signal strength <7 (out of 10).

From both individual and combined scans, measurements from the macular and optic disc area were exported. From the macular area, volumetric measurements of the retinal layer thicknesses were exported. The retinal thickness was measured from the inner limiting membrane (ILM) to the retinal pigment epithelium (RPE). The Early Treatment Diabetic Retinopathy Study (ETDRS) map was used to evaluate the ILM–RPE thickness. The ETDRS map consists of a central circle, inner ring (nasal, superior, temporal, and inferior), and outer ring (nasal, superior, temporal, and inferior) with diameters of 1, 3, and 6 m, respectively. Panel A in Figure 1 shows a schematic diagram of the ETDRS map.

In addition to ILM–RPE thickness, ganglion cell complex thickness was also evaluated from retinal nerve fiber layer (RNFL), ganglion cell layer (GCL), and inner plexiform layer (IPL) thicknesses. This was exported in terms of RNFL–GCL–IPL and GCL–IPL thickness in eight sectors divided into two concentric circles with diameters of 5 and 10 mm. Panel B in Figure 1 shows the schematic representation of the ganglion cell complex map.

From the optic disc, the peripapillary RNFL (pRNFL) thickness around the optic nerve head was evaluated in 12 clock-hour sectors in a circle of 3.45 mm diameter centered at the optic disc. Panel C in Figure 1 shows a schematic representation of the clock positions. The optic nerve parameters (disc area, rim area, cup volume, and cup–disc ratio vertical and horizontal) were also obtained. All thicknesses were obtained using the automated segmentation algorithm from the OCT instrument, and no manual adjustments of the segmentation were allowed.

The baseline demographics of the participants and observations are summarized with descriptive statistics. The within subject SD (Sw), repeatability limits, and coefficient of variation (CoV) were used to describe the repeatability of the OCT HS-100 in both individual and combined scan modes. The Sw, which represents the repeatability of the measurements, was calculated with a one-way ANOVA. The repeatability limit was calculated as 1.96·2·Sw, and it represents the expected limits that 95% of the measurements should be within (21). The CoVs were calculated as the repeatability limit divided by the average thickness of that sector and was expressed in percentage.

The Bland–Altman test for repeated measurements was used to analyze the agreement between the individual and combined scan modes (24). All these calculations were performed for all the parameters evaluated.

Figures 2, 3 show the repeatability limit and the CoVs results, respectively, for the ILM–RPE thickness (left panel), RNFL–GCL–IPL thickness (central panel), and GCL–IPL thickness (right panel) for both scan modes. For ILM–RPE thickness, the repeatability limit values for the individual scan were about two to three times less than the combined scan for all ETDRS sectors except the central sector, where the repeatability limit for the individual scan was five times less than the combined scan. The repeatability limit values were similar among the ETDRS sectors for the individual scan, whereas the values were heterogeneous for the combined scan. Concretely, the repeatability values for the individual scan ranged from 1.80 μm (outer nasal sector) to 2.60 μm (central sector), and for the combined scan, the repeatability values ranged from 3.50 μm (outer nasal sector) to 12 μm (central sector). Regarding the CoV values for the ILM–RPE thickness, the values corresponding to all ETDRS sectors are represented on the top of the line of equality. The CoVs for the individual scan are smaller than 1% for all sectors, and the CoVs for the combined scan range from 1 to 4.5%.

Table 2 summarizes the descriptive statistics of the ILM–RPE thicknesses obtained with each scan modality, and the agreement results between the individual and combined scans for each ETDRS sector. On average, the mean difference between the individual and combined scans was lower than the axial resolution of the instrument (3 μm) for all ETDRS sectors except the central one, where the mean difference value is −3.50 μm. The limits of agreement intervals were <20 μm for all ETDRS sectors, except the central, where the interval was 22 μm.

The repeatability limit for the individual scan mode was <3 μm in all eight sectors (Figure 2, central panel). This scan mode had lower repeatability values compared to the combined scan for all sectors, except one, the inner nasal superior where the values were similar. The repeatability limit values were homogeneous among all sectors with both scan modes. Concretely, the outcomes ranged from 1.9 to 2.9 μm for the short scan and from 2.4 to 4.2 μm for the large scan.

From the CoV results (Figure 3, central panel), the values corresponding to the combined scan were larger than those from the individual scan for all sectors except for the inner nasal superior sector, where the CoVs from both scan modes were the same. The CoVs obtained with the individual and combined scans ranged from 2 to 3%, and 2.5 to 4 %, respectively.

The descriptive statistics of the NFL–GCL–IPL and GCL–IPL thicknesses obtained with each scan modality, and the agreement results between both scan modes for each sector are included in Table 3. On average, the mean difference between the individual and combined scan modes was <1 μm for all sectors. The limits of agreement intervals were similar among all sectors, the minimum and maximum intervals being 5 and 8 μm, respectively.

The repeatability limit values obtained with the individual scan mode were 1.5 to 2.0 times lower than those obtained with the combined scan for all sectors (Figure 2, right panel). With the individual scan mode, the repeatability limit was similar among all sectors, and it was in all cases lower than 3 μm. The repeatability limit for the combined scan mode was different among the sectors. Concretely, the repeatability values for the outer sectors were two times lower than their corresponding inner sectors.

The combined scan had larger CoV values compared to the individual scan for the GCL–IPL thickness in all sectors (Figure 3, right panel). In addition, the outer sectors had lower CoV values compared to their respective inner sectors.

The mean difference between both scan modes was <1 μm for all sectors (Table 3). The limits of agreement intervals were <10 μm for all sectors. However, the values for the inner sectors were twice than that of the corresponding outer sectors.

The repeatability of the three consecutive measurements of pRNFL thickness for the individual and combined scan modes is represented in a polar plot (Figure 4). In both scan modes, the repeatability limit never exceeded 20 μm in any of the sectors. The repeatability limit for the individual scan was lower than that of the combined scan for all clock positions except for clock positions 4 and 10, where the values were almost the same. It can also be noticed that the values for the individual and combined scan are distributed like a vertically elongated ellipse. Sector 9 showed the best repeatability with both scan modes, with the repeatability value of 2.7 and 3.8 μm, respectively. Both scan modes showed the worst repeatability in sector 12, with a repeatability value of 12.30 μm for the individual scan and 18.22 μm for the combined scan.

Table 4 summarizes the descriptive statistics of the pRNFL thicknesses obtained with each scan modality, and the CoV and agreement results between the individual and combined scans for each clock sector position. The combined scan had larger CoV values compared to the individual scan in all clock sectors. The mean difference between both scan modalities never exceeded 7 μm in any of the clock position. Nonetheless, the limits of agreement interval were lessr than 20 μm for sectors 3, 4, 9, and 10, and larger than 20 μm for the other sectors. The shortest and widest intervals are obtained for sectors 9 (9 μm) and 12 (40 μm), respectively.

The descriptive statistics, repeatability, and agreement results for the optic nerve parameters obtained with the individual and combined scan modes are summarized in Table 5. The repeatability limit was <0.15 unit for all parameters independently of the scan mode. Nevertheless, the repeatability values obtained with the individual scan were lower than those obtained with the combined scan. Concretely, these values were about two to three times lower. The CoVs for disc and rim area were also lower in the individual scan mode.

Regarding the agreement, the mean difference between the individual and combined scans was close to zero for all optic nerve parameters. The limits of agreement interval were narrower than 0.15 units for all parameters, except the Rim area, where the agreement interval was 1.50 mm².