Longitudinal Changes in Optic Nerve Head Blood Flow in Normal Rats Evaluated by Laser Speckle Flowgraphy

Abstract

Purpose: To investigate longitudinal changes in mean blur rate (MBR) measured by laser speckle flowgraphy (LSFG) in the rat optic nerve head (ONH), and the reproducibility of MBR.

Methods: Rats were dilated under general anesthesia. Intraocular pressure (IOP), blood pressure, ocular perfusion pressure (OPP), heart rate, and LSFG were measured 30 minutes later. Mean blur rate in the ONH was determined using LSFG-Micro and was subdivided into MBR of the total area (MA), vessel region (MV), and tissue region (MT). Mean blur rate measurements were repeated at 10, 11, 13, 19, and 20 weeks, then every 5 weeks until 60 weeks of age. Intrasession repeatability, intrasession reproducibility, and intersession reproducibility were evaluated.

Results: Coefficient of variation of MBR was 0.3 to 6.2%, 1.3 to 5.2%, and 5.8 to 30.4% for intrasession repeatability, intrasession reproducibility, and intersession reproducibility, respectively. Mean blur rate of the total area, MV, and MT increased similarly until 19 weeks of age, but stabilized thereafter until 60 weeks. Mean blur rate of the total area in the inferior quadrant was significantly higher than in the temporal quadrant from 19 to 55 weeks. These changes exceeded a range of corresponding coefficient of reproducibility. There were no significant changes in IOP, blood pressure, or OPP during the experimental period.

Conclusions: Mean blur rate in the rat ONH changed over time, increased from 10 to 19 weeks of age, then stabilized until 60 weeks. Mean blur rate of the total area exhibited regional differences: higher in the inferior quadrant than in the temporal quadrant. Laser speckle flowgraphy-Micro may provide reliable information for evaluating longitudinal changes of rat ONH blood flow.

Abnormal ocular blood flow is related to the etiology of ocular diseases such as diabetic retinopathy,¹ age-related macular degeneration,² retinal vein occlusion,³ and glaucoma.⁴ Therefore, the measurement of ocular blood flow provides vital information for investigating the pathophysiology of each disorder. To date, various instruments have been developed to measure ocular blood flow. Previous studies on blood flow in the optic nerve head (ONH) utilized laser Doppler flowmetry,⁵ laser Doppler velocimetry,⁶ hydrogen gas clearance method,⁷ and microsphere method.⁸ While laser Doppler flowmetry provides only relative values of velocity, volume, and flow, laser Doppler velocimetry can quantify ocular circulation but require a high level of experience and skill to precisely capture the same position and correct alignment. The hydrogen gas clearance method and microsphere method can provide exact blood flow values, but are highly invasive and limited to laboratory research. 

Laser speckle flowgraphy (LSFG; Softcare Co., Ltd., Fukutsu, Japan) is a noninvasive, quick, and easy method that can measure the laser speckle phenomenon and provide the relative velocity index of erythrocytes, mean blur rate (MBR), in the ONH, retina, and choroid. Mean blur rate has been used to investigate the changes in the ONH circulation not only in humans^9,10 but also in monkeys and rabbits.^11–14 Laser speckle flowgraphy was approved for clinical use as a medical device in Japan in 2008, and the device was also approved for clinical use in the United States by the Food and Drug Administration (K153239) in 2016. Laser speckle flowgraphy-Micro (Softcare Co., Ltd.) was developed in Japan in 2012, and is based on the laser speckle phenomenon combined with a microscope. This new technology is also a noninvasive, quick, and easy method, and was developed for use in laboratory experiments. 

Rats are commonly used as an experimental animal model of the mammalian visual system^15,16 because they share similar anatomy and developmental patterns with humans.^17,18 However, the characteristics of blood flow in ONH of normal rats are currently unknown. Therefore, we investigated the longitudinal changes in MBR to evaluate ONH blood flow in normal rats and examined the reproducibility of MBR measurements to determine whether LSFG-Micro is a useful tool for monitoring the changes of ONH blood flow in rats. 

Methods

Animals

A total of nine male pigmented Brown-Norway rats, 10 weeks of age, were used in this study and monitored until 60 weeks of age. The rats were permitted free access to food and water and were maintained in cages in an environmentally controlled room with a 12-hour light–dark cycle. All animals were treated in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Experimental procedures were approved by the Committee on Animal Experimentation of Kanazawa University. 

Measurement of MBR in Rat ONH

Experiments were conducted on rats anesthetized by an intraperitoneal injection (65 mg/kg) of pentobarbital sodium (Somnopentil; Schering-Plough Animal Health, Omaha, NE, USA), and rats were placed on a heating pad (hoonnkunn; Midori-Shoukai, Tokyo, Japan). Right eyes were dilated with 0.4% tropicamide ophthalmic solution (Mydrin-M; Santen Pharmaceuticals Co., Ltd., Osaka, Japan). Thirty minutes after the induction of anesthesia, rats were positioned with the left eye facing downward and were placed into the stand (Figs. 1A, 1B). Hydroxyethyl cellulose gel (Scopisol; Senju Pharmaceutical Co. Ltd., Osaka, Japan) was applied to the right eye prior to placing a cover glass, which weighs only 1 g, over it; then LSFG was measured in ONH of the right eye. To estimate the ocular orientation, photographs of the fundus were taken using a handheld retinal camera (GENESIS-D; Kowa Co., Ltd., Nagoya, Japan) without anesthesia. Rats were placed on a horizontal table in a natural position and compared to photographs in the same orientation as the color map produced using LSFG-Micro (Figs. 2A, 2B). 

Figure 1

 

The LSFG-Micro device setup. (A) Image of the LSFG-Micro system. (B) Higher-magnification view of the rat head rest.

Figure 2

 

Representative images of the optic nerve head in rats from this study. (A) Representative image of a normal rat fundus. (B) Representative color-coded MBR map produced by LSFG-Micro from the same rat. The white circle indicates the rubber band placement. (C) Image used for automated segmentation between the vessels (white region) and tissue (black region) using LSFG Analyzer software. (D) Orientation of the optic nerve head quadrants. S, superior quadrant; N, nasal quadrant; I, inferior quadrant; T, temporal quadrant. (E) Schematic of the LSFG-Micro setup. The LSFG-Micro is set on a tilting handle with transverse and lateral movement to adjust the region in the color map.

Laser speckle flowgraphy-Micro consists of an ordinary charge-coupled device (600 × 480 pixels) camera equipped with a diode laser (wavelength 830 nm) attached to a microscope (SZ61TR; Olympus Corporation, Tokyo, Japan). A schematic of the imaging setup is shown in Figure 2E. The principle of LSFG-Micro is the same as that of LSFG in clinical use, and its methods and principle have been described previously.^19–21 The MBR represents a relative index of blood velocity, and is determined by the blurring of the speckle pattern formed by the interference of a laser scattered by the movement of blood cells. The MBR images were acquired continuously at a rate of 30 frames per second over a 4-second period. The field of view was 3.43 × 3.43 mm with a working distance of 110 mm. The MBR in different regions of the ONH were calculated using LSFG Analyzer software (version 3.1.14.0; Softcare Co., Ltd.). The MBR was divided into three parameters: MBR of the total area (MA) was calculated as the average MBR over the entire ONH; the MBR of the vessel region (MV) was the average of the vessel region; the MBR of the tissue region (MT) was calculated as the average MBR of the total ONH area minus the vessel region. Measurements were performed three times consecutively without adjusting the position of rats, and the average MBR was calculated. The margin of the ONH was identified by manual placement of a rubber band (1.37-mm diameter); the software then segmented out the vessel region using the automated definitive threshold (Figs. 2B, 2C). These parameters were evaluated for each quadrant: superior, nasal, inferior, and temporal (Fig. 2D). 

Analysis of Repeatability and Reproducibility of MBR Measurements

Intrasession repeatability was defined as measurement variability without resetting the position of the rat's head. The coefficient of variation (COV) and coefficient of repeatability (CR) were calculated from MBR of three consecutive measurements at each week. Intrasession reproducibility was defined as measurement variability after resetting the rat's head position on the restrainer, and was evaluated at 13 and 20 weeks of age. Mean blur rate measurements were repeated three times separated by breaks for head adjustment, and COV and CR were calculated. 

Intersession reproducibility was defined as measurement variability between two consecutive measurement weeks. The COV and CR were calculated using the average of three consecutive measurements of MBR at each week. 

Measurement of IOP and Blood Pressure

Intraocular pressure, blood pressure, ocular perfusion pressure (OPP), and heart rate were measured 30 minutes after the induction of anesthesia. Before LSFG measurements (prior to the application of hydroxyethyl cellulose gel and a cover glass), IOP was measured with a handheld tonometer (TonolabTV02; M.E. Technica, Tokyo, Japan) in the right eye of each animal (mean of three measurements per eye). Blood pressure and heart rate were measured at the tail using an automatic sphygmomanometer (BP-98; Softron, Tokyo, Japan). Ocular perfusion pressure was calculated using the formula OPP = 2/3 mean blood pressure − IOP. 

Statistical Analysis

The data are expressed as the mean ± standard deviation (SD). One-way repeated-measures analysis of variance (ANOVA) was used to analyze the temporal changes in MBR from the baseline. Differences in the superior, nasal, inferior, and temporal MBR ratios at different time points were analyzed by repeated-measures ANOVA. The COV and CR, defined as 1.96 × within-subject standard deviation, were calculated according to the methods outlined by Bland and Altman.²² Differences of P < 0.05 were considered to be statistically significant. 

Results

Longitudinal Changes in MBR in ONH of Normal Rats

The longitudinal changes in MBR in ONH are shown in Figure 3 in the representative color map produced using LSFG-Micro. The mean MBR changes over time (MA, MV, and MT) are shown in Figure 4. Mean blur rate (MA, MV, and MT) increased until 19 weeks of age, but was stable thereafter until 60 weeks of age. The longitudinal changes were similar between different quadrants. Regarding the comparisons between different quadrants in each week, MA in the inferior quadrant was significantly higher than in the temporal quadrant from 19 to 55 weeks of age (Fig. 5). 

Figure 3

 

Representative color map produced by LSFG-Micro showing the longitudinal changes in optic nerve head blood flow in the same rat at 10 (A), 11 (B), 13 (C), 19 (D), 20 (E), 25 (F), 30 (G), 35 (H), 40 (I), 45 (J), 50 (K), 55 (L), and 60 (M) weeks of age. S, superior quadrant; N, nasal quadrant; I, inferior quadrant; T, temporal quadrant. The color scale is shown at the bottom right.

Figure 4

 

Longitudinal changes in MBR in normal rats. (A) MA, (B) MV, (C) MT. Data show the mean ± SD. n = 9. Significant changes compared to initial values are indicated (^†P < 0.01, ^‡P < 0.001; repeated-measures ANOVA).

Figure 5

 

Changes in MBR from each quadrant in normal rats at 10 (A), 11 (B), 13 (C), 19 (D), 20 (E), 25 (F), 30 (G), 35 (H), 40 (I), 45 (J), 50 (K), 55 (L), and 60 (M) weeks of age. Data show the mean ± SD. n = 9. Difference in the superior, nasal, inferior, and temporal MBR ratios at each time point was analyzed. (*P < 0.05, **P < 0.01, ***P < 0.001; repeated-measures ANOVA).

Repeatability and Reproducibility of MBR Measurements

Coefficient of variation of intrasession repeatability was 1.0 to 3.7%, 0.3 to 4.0%, and 1.7 to 6.2% in MA, MV, and MT, respectively (Table 1). The CR was 0.13 to 0.29, 0.28 to 0.72, and 0.15 to 0.33 in MA, MV, and MT, respectively. Coefficient of variation of intrasession reproducibility per quadrant at 13 weeks of age (20 weeks of age) was 1.9 to 2.4% (1.3–5.3%), 2.5 to 2.8% (1.7 to 2.5%), and 4.1 to 5.2% (3.7–4.5%) in MA, MV, and MT, respectively (Table 2). The CR per quadrant at 13 weeks of age (20 weeks of age) was 0.19 to 0.27 (0.15–0.67), 0.54 to 0.58 (0.41–0.64), and 0.24 to 0.27 (0.24–0.30) in MA, MV, and MT, respectively. Coefficient of variability of intersession reproducibility was 7.0 to 25.3%, 5.8 to 26.0%, and 7.1 to 30.4% in MA, MV, and MT, respectively (Table 3). The CR was 0.95 to 2.88, 1.70 to 6.17, and 0.61 to 1.86 in MA, MV, and MT, respectively. 

Table 1

 

Intrasession Repeatability of MBR in the Optic Nerve Head of Normal Rats (n = 9)

 

Table 2

 

Intrasession Reproducibility of MBR in the Optic Nerve Head of Normal Rats (n = 6)

 

Table 3

 

Intersession Reproducibility of MBR in the Optic Nerve Head of Normal Rats (n = 9)

 

IOP and Blood Pressure

Figure 6 shows the mean changes in IOP, blood pressure (systolic blood pressure, diastolic blood pressure, and mean blood pressure), OPP, and heart rate over time. Intraocular pressure, blood pressure (systolic, diastolic, and mean blood pressure), and OPP did not significantly change during the experimental period. Heart rate was significantly reduced from 11 weeks of age (P < 0.05). A cover glass, which was gently placed on the eye, did not cause IOP changes after LSFG measurements. 

Figure 6

 

Longitudinal changes in intraocular pressure (IOP), blood pressure, ocular perfusion pressure (OPP), and heart rate in normal rats. (A) Longitudinal changes in IOP. (B) Longitudinal changes in systolic blood pressure, diastolic blood pressure, and mean blood pressure. (C) Longitudinal changes in OPP. (D) Longitudinal changes in heart rate. Data show the mean ± SD. n = 9. Significant changes compared to initial values are indicated (*P < 0.05, ^‡P < 0.001; repeated-measures ANOVA).

Discussion

This study is the first to evaluate blood flow in normal rat ONH using the LSFG method. Laser speckle flowgraphy makes it easy to reimage the same position, provides a wide field of view, and can be completed within several seconds using this simple apparatus. Laser speckle flowgraphy has the advantage that it can be applied to both humans and animals. However, MBR measured by LSFG is a relative, not absolute, value of velocity data. Recent animal studies have shown that MBR closely correlates with actual blood flow in the ONH measured by the microsphere method in monkeys¹¹ and by hydrogen gas clearance method in rabbits.^13,14 Therefore, MBR changes should reflect the actual changes of blood flow in the ONH. Consequently, several clinical studies that examined between-subject comparisons of MBR in human ONH have been published.^23–25 

However, LSFG measurements are affected by laser speckle signal bias due to the influence of various factors including retinal pigment epithelial scattering, laser beam reflectance, thickness of the vascular wall, and target tissue absorption. Therefore, interpretation of MBR comparison between different subjects, different eyes, or even different retinal locations must be done with caution. In this study, we evaluated the longitudinal changes in the same retinal location, optic disc, in the same eye. We did not observe any remarkable changes in the appearance of the fundus and media opacity, including cataract, that may affect MBR measurements during the study period. Therefore, the tissue reflectance did not have a significant effect on our results. However, the biometric dimensions such as axial length do change with age. For example, longer axial length is associated with smaller lateral magnification and should result in a wider area of LSFG imaging.²⁶ If the image area centered on the ONH is enlarged, the retinal thickness and vascular density decrease, and MBR values may change accordingly. However, as shown in Figure 3, the imaging area of LSFG was almost identical throughout the study period. Therefore, the longitudinal changes in biometric dimensions were unlikely to modify MBR values through the alteration of lateral magnification of LSFG images. 

Assessment of measurement reproducibility is a prerequisite for any imaging device. High reproducibility is mandatory to detect small changes and is a determinant of the usefulness of the instrument. In human studies using a current model of LSFG, Aizawa et al.²⁷ reported that COV values of intrasession reproducibility were 2.9 ± 2.1%, 1.9 ± 1.1%, and 2.1 ± 1.1% for MA, MV, and MT in normal subjects, respectively. When evaluated per quadrant, average COV values were 3.4 to 4.7%, 2.3 to 2.8%, and 2.9 to 3.6% for MA, MV, and MT, respectively. In their study, MBR measurements were repeated three times with resetting of the face position each time, which was the same as for the intrasession reproducibility evaluated at 13 and 20 weeks of age in our rat study. The COV values for intrasession reproducibility per quadrant obtained in this study were comparable to the human data. 

Statistically significant differences in MBR do not necessarily mean true or critical differences. At least, the magnitude of differences should exceed the measurement variability to be regarded as meaningful differences. Regarding regional differences, MA in the inferior quadrant was significantly higher than in the temporal quadrant from 19 to 55 weeks of age. The statistically significant regional variation of MA at 20 weeks of age (1.35 ± 0.85) exceeded the intrasession CR of either inferior (0.56) or temporal (0.21) quadrants. Mean blur rate increased more than 3, 6, and 2 over time in MA, MV, and MT, respectively. The magnitude was larger than the intersession reproducibility at 20 weeks of age or older when MBR reached a stable level. These results indicate that the regional variation and the longitudinal changes of MBR in rat ONH can be regarded as actual phenomena. 

Regarding possible regional variation of blood flow in rat ONH, Young and Lund²⁸ show that the largest number and density of RGCs, which mediate pupilloconstriction in response to luminance changes, were found in the inferior and nasal quadrants of the retina. Secondly, Wallace et al.²⁹ discovered that a major function of the rat visual system is to provide the animal with comprehensive overhead surveillance for predator detection. Thus, the relatively larger MBR in the inferior quadrant compared to the temporal quadrant of ONH may correspond to the importance of the upper visual field for rats. 

The effects of age on ONH blood flow has been addressed only in normal human subjects using laser Doppler flowmetry. Rizzo et al.³⁰ reported that Doppler broadening, which is directly proportional to the speed of red blood cells in the capillary, increased with age from 16 to 30 years, then decreased with age between 30 and 76 years in humans. The limitation of the study was that the relative values of laser Doppler flowmetry were not derived from longitudinal observation. Nevertheless, the result is in agreement with our data in rats, given that MBR increased until 19 weeks of age, an age equivalent to approximately 20 to 30 years of age in humans.³¹ 

The present study has some limitations. First, it is unknown when stable ONH blood flow begins to decrease in normal rats. We monitored longitudinal changes in MBR of ONH until 60 weeks of age. Due to the death of some rats caused by various factors, we were unable to maintain a statistically relevant number of rats beyond 60 weeks of age. Second, the LSFG-Micro Analyzer software cannot examine additional target parameters, such as waveform analysis, because the heart rate in rats is too fast. The relationship between pulse waveform parameters in ONH circulation is referred to as blowout time (BOT) and has been reported in human adults. Shiba et al.^32,33 reported that age was significantly and negatively correlated with the BOT (age range, 29–80 years). Finally, given that close correlations of MBR with absolute values of ONH blood flow measured by other methods have been proven only in rabbits and monkeys, the applicability to rats remains to be elucidated for future studies comparing between different eyes or different rats. Nevertheless, the significant longitudinal changes and regional variations of MBR in ONH of the same rat eye should reflect actual blood flow in ONH given the favorable measurement reproducibility. 

In conclusion, MBR in the rat ONH changed over time, increased from 10 to 19 weeks of age, then stabilized until 60 weeks. Mean blur rate of the total area exhibited regional differences; it was higher in the inferior quadrant than in the temporal quadrant. Laser speckle flowgraphy-Micro may provide reliable information for evaluating longitudinal changes in rat ONH blood flow. 

Acknowledgments

Disclosure: Y. Wada, None; T. Higashide, None; A. Nagata, None; K. Sugiyama, None