Materials and methods

SUBJECTS

From February 1996 to July 1996, 12 patients (mean age 52 years; range 21–7; four males and eight females) with unilateral, localised, quadrant rhegmatogenous retinal detachment repaired by uncomplicated scleral buckling and encircling procedures were consecutively enrolled. The entry criteria for the study were retinal detachment localised to the periphery of the fundus and not involving the macula, no other known retinal vascular disease, visual acuity correctable to at least 0.4, adequate pupillary dilatation, and clear media. Eyes which would need exoplants crossing the midline were not included in the study. The protocol of the study was approved by the ethics review committee of the University of Tokyo School of Medicine and written informed consent was obtained from each participant after full explanation of the procedure. No patient had any known haematological disorder. The primary surgeon (SE) was notified of the patient's enrolment into the study after the surgery.

Surgery was performed under local anaesthesia for all 12 patients by one of the authors (SE). Retinal breaks were identified in all patients and were treated with transscleral cryotherapy. A Mira No 240 silicone encircling element (Mira, Inc, Waltham, MA, USA), in conjunction with a silicone tyre exoplant, was used in all patients. The encircling element was secured to the sclera 13 mm posterior to the limbus with 5-0 Surgiron mattress sutures placed in the four quadrants between the muscles and fastened with a Mira No 270 sleeve with moderate tightness. The extent of the exoplant was from 1.5 to 2.5 quadrants and was consistent with the size and location of retinal breaks and co-existing lattice degeneration. Subretinal fluid was drained externally in every patient. Extraocular muscle disinsertion was not required for any patient. There were no intraoperative complications. At the end of the procedure, indirect ophthalmoscopy revealed retinal reattachment along with the absence of central retinal artery pulsations in all patients. In none of the patients was the eye believed to be excessively firm at the end of the operation. The height of the scleral buckle also was roughly judged at this point by the angle of bending the retinal vessel onto the segmental buckle by the primary surgeon (SE); about 30° to 60° designated as medium and 60° to 90° as high. (designated as low, medium, or high). The type and location of each exoplant was drawn in the chart after the procedure.

INSTRUMENT

A fundus camera was equipped with a diode laser (wavelength 808 nm) and image sensor (100 × 100 pixels). The fundus where the laser beam was focused was observed by means of an infrared CCD camera. The scattered laser light was imaged on an image sensor corresponding to a field of 1.06 × 1.06 mm (45 degree visual angle of the fundus camera) in the human fundus, on which a speckle pattern appeared. The difference between the average of the speckle intensity (I_mean) and the speckle intensity for successive scannings of the image speckles at the pixels on the sensor plane was calculated, and the ratio of I_mean to this difference was defined as NB.12 14 NB is nearly equivalent to the reciprocal of the speckle contrast defined by Fercher and Briers.16 17 The NB was calculated using a logic board every 0.125 seconds successively for a maximum of 7 seconds and divided into 50 colour coded levels, which were displayed as colour graphics on a colour monitor showing the two dimensional variation of the NB level over the field of interest.18 The average NB level (NB_av) in any rectangular field of interest on a displayed colour map can be calculated and the change of NB_av over 7 seconds can be monitored at the same time. The maximum exposure of the retina with the present apparatus is below the permissible variable limits of American National Standard Institute (ANSI).18

NB MEASUREMENT IN HUMAN ONH AND CHOROID RETINA

The pupil was dilated with one drop of Mydrin M (0.4% tropicamide, Santen Pharmaceutical, Osaka, Japan) 30 minutes before measurements were taken. During the measurement, the subject was asked to watch a target light; an electrocardiogram was monitored simultaneously. The image speckles from the measurement field located in the temporal site of the ONH, corresponding to a field of 0.72 × 0.72 mm (30 degree visual angle of the fundus camera), were recorded. The average NB across the largest rectangular field in the measurement field free of visible surface vessels further averaged over three cardiac pulses when fixation was satisfactory, was calculated to obtain mean NB_av from the ONH (NB_ONH). The size of this rectangular field varied among subjects to avoid visible surface vessels, ranging from 0.15 × 0.22 mm (20 × 30 pixels) to 0.37 × 0.48 mm (50 × 65 pixels).

Image speckles from a field corresponding to a field of 1.06 × 1.06 mm (45 degree visual angle of the fundus camera) were recorded and the average NB across the largest rectangular field free of visible surface retinal vessels was averaged as described above to obtain the mean NB_av from the choroid and retina (NB_ch-ret). Although large choroidal vessels can be seen on the display, their contours were not as discrete as those of surface retinal vessels and the density of choroidal vessels are greater than surface visible retinal vessels. Therefore, we did not avoid visible choroidal vessels, while we did avoid visible retinal vessels. The size of this rectangular field varied among subjects to avoid surface visible retinal vessels, ranging from 0.48 × 0.53 mm (45 × 50 pixels) to 0.64 × 0.80 mm (60 × 75 pixels). Image speckles from the macula, corresponding to a field of 1.06 × 1.06 mm (45 degree visual angle of the fundus camera), were recorded and the average NB across the foveal area free of retinal vessels was averaged as above to obtain the mean NB_av from the foveal area (NB_fovea). The size of this square field was 0.32 × 0.32 mm (30 × 30 pixels).

The movement of the subject's eye in any direction during the measurement period was checked by the method previously described.14 It was also checked by inspecting the colour map and the time course plot of NB_av taken every 0.125 seconds; when there was no eye movement during measurement, visible surface vessels did not change position on the colour map and the time course plot of NB_av exhibited periodic fluctuations synchronised with cardiac pulse.18 On the other hand, when eye movement occurred, visible surface vessels in the colour map changed position according to the eye movement. The coefficients of reproducibility for 5 minute and 24 hour interval measurements in normal humans were 11.7% and 13.0% for NB_ONH and 8.7% and 9.7% for NB_ch-ret, respectively.18

Image speckles from all subjects were digitally stored as colour maps. NB_av was calculated by a blinded investigator (MN) for each subject in the same position and size of the rectangular field free of visible surface vessels by referring to a polaroid fundus picture and colour map.

STUDY PROTOCOL

The both eyes of a patient underwent identical assessment at each time point of measurement. The baseline measurements were carried out between 1100 and 1300 one day before surgery. After the pupil was dilated by one drop of Mydrin M, the NB_ONH and NB_fovea in both eyes were measured. The NB_ch-ret was measured in the eye with the retinal detachment in an area where the retina was not detached, outside the vascular arcade, and close to the area where the buckle was going to be located (the buckled side), and an area 180° from the buckled side (unbuckled side) (Fig 1). Thereafter, the NB_ch-ret on the mirror image side of the unoperated contralateral eye in relation to the buckled or unbuckled side of the operated eye was measured. Polaroid fundus photographs were taken to record the sites of measurement. The IOP in both eyes and brachial arterial blood pressure were also measured. It took 5–10 minutes to complete all procedures in one eye.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Measurement field of normalised blur (NB) in human choroid and retina and optic nerve head (ONH). Image speckles from a field on the buckled side (A) and unbuckled side (B) with no discrete retinal vessels visible in the operated eye and those on the mirror image sides of the unoperated contralateral eye were recorded to measure NB in the choroid and retina. Image speckles from fovea and a field in temporal ONH, free of visible surface vessels, were recorded to measure NB in the fovea (C) and that in ONH tissue (D).

During the prospective follow up period of 12 weeks, patients were examined 1, 4, and 12 weeks after the surgery and at each visit, brachial arterial blood pressure, and the IOP, NB_ONH, NB_fovea, and NB_ch-ret in both eyes were measured as described above. These NB measurements were performed in the same area as measured before surgery, referring to a fundus photograph and colour map of NB obtained before surgery to determine the preoperative locations. All measurements were performed between 1100 and 1300. The ocular perfusion pressure (OPP) was also calculated by:

OPP=2/3BP_m − IOP      (1) where BP_m was mean brachial arterial blood pressure. All patients were monitored for evidence of anterior segment ischaemia or other postoperative complications.

DATA ANALYSIS

Statistical significance was determined with Student'st test for paired variates. Analysis of variance (ANOVA) for the data obtained by sequential measurements was also used.

Results

Two of the 12 patients, who could not visit the hospital on the scheduled postoperative day, were excluded from the study and the following analysis was performed in 10 patients (Table 1). In both the operated and unoperated contralateral eyes, there was no significant change in IOP from the baseline and no bilateral differences at any of the time points (paired t test, p >0.1). There was no significant change from the baseline at any time point in mean blood pressure or OPP (paired t test, p >0.1; Table 2).

There was no significant difference in the baseline NB_ch-ret, NB_fovea, and NB_ONH at any of the corresponding sites between the operated and the unoperated contralateral eyes (Table 3). There was a significant approximately 40% decrease from the baseline in the NB_ch-ret on the buckled side of the operated eye (Table 4), while no significant change was observed in NB_ch-ret on the mirror image side of the unoperated contralateral eye. Further, no significant change was observed in the NB_ONH and NB_fovea, in both eyes, and NB_ch-ret on the unbuckled side of the operated eye and that on the mirror image side of the unoperated contralateral eye (Table 4). There were significant bilateral differences in the change in the NB_ch-ret from the baseline (ΔNB_ch-ret ) on the buckled side (ANOVA of repeated measurements, p<0.0001), but not the unbuckled side (Table 4). The bilateral difference was not significant for the change in both NB_fovea and NB_ONH from the baseline (ΔNB_fovea and ΔNB_ONH, respectively; Table4).

Discussion

METHOD

NB is primarily a quantitative index of tissue blood velocity.12 14 In rabbits, when the ocular blood flow rate was artificially changed by increasing the IOP12 14or by intravenous injection of a calcium antagonist,15averaged NB levels over the sampling field, NB_av, had a good correlation with the blood flow rate determined using the microsphere method in the retina or choroid.12 14 15These findings suggest that the NB also correlates with the blood flow rate in the choroid or retina, at least under certain conditions. According to Koelle et al,19the penetration depth of near infrared laser (wavelength 811 nm) in the cat optic nerve exceeds 1 mm. When ONH blood flow rate in rabbits was artificially changed by a small amount of endothelin-1 or by inhalation of 10% carbon dioxide, NB_av obtained from ONH tissue strongly correlated with the blood flow rate in the ONH determined using the hydrogen gas clearance method where a hydrogen electrode was inserted into the ONH tissue to a depth of approximately 0.7 mm.13 Although blood flow rate measured by the hydrogen gas clearance method does not directly reflect the movement of red blood cells as the NB_av measured by the laser speckle method, these findings suggest that the NB correlates with the blood flow rate also in the ONH, at least under certain conditions.

Unlike rabbit eyes, in which retinal vasculature exists only in the medullary wing area, the NB obtained from the fundus in human eyes is affected by both retinal and choroidal circulation.20 In the case of Dutch rabbits, no significant difference was observed between the NB_av obtained using the diode laser from the medullary field where no discrete vessels were observed and the NB_av from the extramedullary field, within an IOP range between 10 and 80 mm Hg.12 This finding suggests that the contribution of retinal microcirculation to the NB obtained from the fundus using the diode laser is very minor in rabbits. Measurement of NB_av obtained from the posterior fundus of monkeys using the diode laser before and after obstruction of a retinal artery (experimental branch retinal artery occlusion) suggested that the contribution of retinal microcirculation to NB_av obtained with the diode laser was about 20% or less in monkeys.21These results suggest that the contribution of choroidal circulation to the NB_av obtained in the present study from the human fundus with no discrete retinal vessels visible is probably much greater than the contribution of retinal microcirculation. The NB obtained from the foveal area is supposed to be affected only by choroidal circulation because the area is free of retinal capillaries. It must be noted that the laser speckle method does not give an absolute value for the tissue blood velocity, but a relative value, because NB is also influenced by the reflectivity of the laser beam in the tissue. Therefore, the NB_av obtained from the ONH, foveal area, or mid-peripheral fundus should not be directly compared and may not be suited for interindividual comparison. On the other hand, this method yields good reproducibility in the same tissue of an individual,18 and its change relative to the baseline correlates well with the relative change of blood flow determined using the microsphere method12 14 15 22 or hydrogen gas clearance method.13 These characteristics and the simplicity of the measurements makes this method very useful in following the time course of change in an individual tissue.

EFFECT OF SCLERAL BUCKLING PROCEDURE

The present study indicated that segmental scleral buckling procedures with encircling elements significantly decreased NB_ch-ret on the buckled side. Using bidirectional laser Doppler in conjunction with retinal photography, Ogasawaraet al  6 reported that in the patients who had undergone uncomplicated scleral buckling procedures the blood flow rates through the major temporal retinal arteries were on average 50% lower in the surgically treated eyes than in the unoperated contralateral eyes. If we assume that the contribution of retinal microcirculation to the NB_ch-ret obtained in the present study is approximately 20% according to the result reported by Isono,21 it seems difficult to explain a 40% decrease in the NB_ch-ret on the buckled side after scleral buckling only by its effect on retinal circulation. A decrease in choroidal tissue blood velocity after the scleral buckling procedure seems very likely. This finding is consistent with previous reports using the microsphere23 or hydrogen gas clearance method24 in rabbits that choroid and retina blood flow buckled with encircling elements was significantly decreased23 or that choroidal blood flow was reduced at the posterior and peripheral sides of the encircling band.24 Ohkubo25 found signs of reduced retinal perfusion in regions of reattached retina that were indented by scleral buckling procedures. Compressional mechanisms have been cited as the cause of reduced choroidal blood flow following scleral buckling procedures.9 23 24 Thus, the reduced blood velocity in the choroid was probably attributed to direct obstruction of the choroidal venous drainage due to scleral buckling and/or local effects of cryotherapy.

The important finding in the present study is that the NB_ch-ret on the unbuckled side, the NB_fovea, and NB_ONH showed no significant change in spite of the significant decrease in NB_ch-ret on the buckled side. This result may not be consistent with those of previous reports suggesting substantial decrease in the posterior segment circulation on the whole at various intervals after buckling procedure5-11—that is, 50% decrease in the flow rates in the major temporal retinal arteries,6 35%–50% decrease in flow velocities in the central retinal artery,7 31% decrease in the flow velocity in the ophthalmic artery,8 and 53% decrease in the ocular pulse.11 These previous reports, however, did not study the circulatory change in different parts of the fundus. Findings of the present study suggested that the segmental scleral buckling procedure with encircling elements does not cause clinically significant circulatory changes elsewhere on the buckled side, including the ONH and the foveal area where postoperative vision might be affected. The discrepancy between the present result and previous reports might be partly due to that the reduction of blood flow rate depends upon the degree of constriction of the encircling band, and/or the extent of the exoplant as reported by Mano24 or Diddieet al.23 In the present study, the encircling element was fastened with a moderate tightness and the extent of the exoplant was 1.5 or 2.5 quadrants. If the encircling element was fastened more tightly and/or the exoplant was more than three quadrants, it may be possible that not only the NB_ch-ret on the unbuckled side, but also NB_fovea, and/or NB_ONH are also affected significantly. In all of the present subjects, retinal detachment was successfully treated without any complications and it seems very unlikely that buckling and/or fastening of the encircling element were unreasonably mild. As discussed above, the present NB_ch-retis thought to mainly reflect choroidal circulation, and partly retinal circulation. Therefore, the present result does not exclude the possibility that the retinal blood flow was slightly decreased also on the unbuckled side, as Ogasawara et al 6 reported, while the NB_ch-retshowed no significant change. Whether the NB_fovea or NB_ONH was also unaltered after segmental scleral buckling and encircling procedure in patients with pre-existing compromise of ocular tissue circulation awaits future study.

In summary, the present study demonstrated in otherwise healthy human eyes that the segmental scleral buckling procedure with encircling elements fastened with a moderate tightness causes a decrease in the blood velocity only in the choroid and/or retina on the buckled side, but little change in other areas of the fundus, including the ONH, foveal area, and fundus on the unbuckled side.