Changes in Blood Flow on Optic Nerve Head After Vitrectomy for Rhegmatogenous Retinal Detachment

Abstract

Purpose: To determine the preoperative characteristics and the changes in retinal blood flow following vitrectomy in eyes with a rhegmatogenous retinal detachment (RRD).

Methods: Twenty-five–gauge vitrectomy without scleral bucking was performed on 31 eyes of 31 patients with macula-on RRD. The retinal blood flow on the optic nerve head (ONH) was assessed by laser speckle flowgraphy (LSFG), and the mean blur rate (MBR) and pulse waveform parameters before and at 10 days, 1, 2, 3, and 6 months after the surgery were examined. Eyes treated by scleral buckling, and eyes with an epiretinal membrane and cataract that underwent surgery were used as controls.

Results: The mean preoperative MBR-vessel on the ONH was significantly lower in eyes with RRD than in the fellow unaffected eyes (P < 0.001), but it was not significantly different from the operated eye and the fellow eye in the control group. A significant increase in the mean MBR-vessel on the ONH was observed following vitrectomy in eyes affected by RRD (P < 0.001), whereas no significant difference was observed in the fellow eye, the scleral buckling–treated eyes, and the control eyes. Of the eight pulse waveform parameters, only the flow acceleration index was significantly lower in eyes with a RRD than in the fellow eyes preoperatively, but then it significantly increased with time following vitrectomy. The changes in the MBR-vessel were not correlated with that of other parameters (e.g., the ocular perfusion pressure).

Conclusions: These results indicate that the retinal blood flow is reduced in eyes affected by RRD preoperatively, and can recover following successful RRD repair by vitrectomy.

A rhegmatogenous retinal detachment (RRD) is a separation of the sensory retina from the RPE caused by tears in the retina and can lead to a permanent loss of vision if untreated.¹ Rhegmatogenous retinal detachment is usually treated by either scleral buckling alone or by pars plana vitrectomy (PPV) with or without scleral buckling. The results of a randomized controlled trial indicated that both of these surgical procedures led to successful reattachment in phakic, pseudophakic, and aphakic patients with similar final reattachment rates.² 

Although scleral buckling is a well-established surgical treatment for RRD, it has been reported that the procedure can cause problems in the ocular circulation.^3–10 Diddie and Ernest⁷ used the microsphere technique and found that the encircling buckling method significantly decreased the retinal and choroidal circulation in rabbit eyes. There have also been several studies reporting that scleral buckling can reduce the retinal^5,6,8 and choroidal blood flow.^9–11 Compressional mechanisms have been cited as the cause of the reduced choroidal blood flow following scleral buckling.^7,12 Thus, the reduced ocular blood flow has been attributed to a direct obstruction of the choroidal venous drainage by the scleral buckling, and the time course of the ocular blood flow following scleral buckling is always affected by the procedures. 

The safety and effectiveness of PPV has improved by the improvements in the surgical instruments including micro incision instruments and wide-angle viewing systems.¹³ Because PPV without concomitant scleral buckling does not entail a compressional step, it is possible to evaluate the natural course of ocular blood flow before and after successful RRD repair. However, there have been only a few studies examining the effects of PPV on the ocular blood flow in eyes with RRD. Kimura et al.¹⁴ reported that the ocular microcirculation is normal 6 months after PPV for RRD. It has been reported that a gas tamponade might have subclinical adverse effects on the circulation in the region of the neuroretinal disk rim.¹⁴ 

A variety of techniques have been developed for measuring the retinal blood flow including fluorescein angiography,¹⁵ radioactive microspheres,¹⁶ hydrogen clearance,¹⁷ laser Doppler technique,^4,5 color Doppler ultrasonography,¹⁸ and the pulsatile technique.⁹ Time intensiveness and problems with reproducibility have hampered the widespread use of many of these measurement techniques. More recently, a Doppler optical coherence tomographic (OCT) technique,¹⁹ OCT angiography,²⁰ and optical microangiography (OMAG)^21,22 have been developed and used to measure the blood flow on the optic nerve head (ONH) and retina using high-resolution depth-resolved imaging with high reproducibility. However, these techniques still have inherent limitations for clinical use because of the time intensiveness of the procedures. 

Laser speckle flowgraphy (LSFG) is a noninvasive, real-time method that has been used to measure the relative blood flow on the ONH. The measurement duration is only 4 seconds and can be performed without the intravenous injection of any contrast agents.^23–25 A recent update of the software embedded in the most recent LSFG device (LSFG Analyzer, v. 3.1.62; Softcare Co., Ltd., Fukutsu, Japan) enables the recording of images synchronized with each cardiac cycle, and can then determine the various blood flow parameters at each heartbeat. Laser speckle flowgraphy can detect the speckle contrast pattern produced by the interference of the illuminating laser light that is scattered by the movement of erythrocytes in the blood vessels. This enables the device to calculate the relative blood flow in the vessels of the ONH and retina, which is expressed as the mean blur rate (MBR).^23–25 The “vessel extraction” function of the software then identifies the vessel and tissue areas on the ONH so that the MBR of each area can be assessed separately. The vessel area can be used to evaluate the blood flow in the retinal vessels excluding the choroidal blood vessels.²⁶ Aizawa et al.²⁷ reported that the coefficient of variation (COV) for determining the MBR was 3.4 for the ONH. Therefore, LSFG is considered to be suitable for measuring the blood flow rates on the ONH. 

However, there have been only a few reports describing the ocular blood flow in eyes affected by RRD before surgery and that following PPV. It is not easy to estimate the preoperative retinal blood flow in eyes affected by RRD because the retinal vessels are tortuous and ascend vertically to the surface to the RPE in the areas where the retina is detached. 

Thus, the aim of this study was to evaluate the blood flow on the ONH in the PPV-treated eyes with RRD by comparing the values with that of the fellow eyes and the scleral buckling–treated eyes. Laser speckle flowgraphy was used to determine the changes in the ONH blood flow before and after surgery. 

Methods

Ethics Statement

This was a retrospective, cross-sectional, single-center study, and the procedures used were approved by the Ethics Committee of the Nagoya University Hospital (Nagoya, Japan). The study was performed at the Nagoya University Hospital, and the study conformed to the tenets of the Declaration of Helsinki. 

Subjects

We reviewed the medical records of all patients who had undergone 25-G PPV for RRD at the Nagoya University Hospital from July 2013 to March 2015. The unaffected fellow eyes, patients who had undergone scleral buckling for a RRD and PPV for an epiretinal membrane (ERM) with phacoemulsification and aspiration for cataract were studied as controls. All of the patients signed an informed consent form before the surgery. 

Exclusion Criteria

The exclusion criteria included the presence of any optic disc abnormalities such as those related to glaucoma, optic disc atrophy, history of intraocular surgery, presence of vitreous hemorrhage, severe cataract, and absence of a posterior vitreous detachment (PVD) in eyes treated by PPV. Subjects were also screened for any medical condition that could influence the hemodynamic of the eye such as diabetes, hypertension, arrhythmia, and vascular diseases. 

All patients underwent a comprehensive ophthalmic examination as standard examination in our hospital including the measurements of the IOP and axial length, slit-lamp biomicroscopy, fundus photography with the Optomap camera (Optos plc., Dunfermline, Scotland), spectral-domain OCT (SD-OCT; Spectralis OCT, Heidelberg Engineering, Heidelberg, Germany), and LSFG before, and 10 days, and 4, 8, 12, and 24 weeks after the surgery. 

All patients were asked to abstain from alcoholic and caffeinated beverages on the morning of the day of the examination because the intake of alcohol and caffeine can influence the IOP^28,29 and blood pressure.^30,31 Thirty minutes before the LSFG examinations, 0.4% tropicamide/phenylephrine (Mydrin P; Santen Pharmaceutical Co., Ltd., Osaka, Japan) was used to dilate the pupil. The subjects rested for 10 to 15 minutes in a quiet, dark room before the examination, and all examinations were performed in the sitting position. The refractive error (spherical equivalent) was measured with an autorefractometer (KR8900; Topcon, Tokyo, Japan), and the axial lengths were measured by partial optical coherence interferometry (IOLMaster; Carl Zeiss Meditec, La Jolla, CA, USA). The IOP was measured with a handheld tonometer (Icare; Tiolat Oy, Helsinki, Finland). The systolic and diastolic blood pressures (SBP and DBP) were measured at the left brachial artery at the height of the heart in a sitting position with an automatic sphygmomanometer (CH-483C; Citizen, Tokyo, Japan). The mean arterial blood pressure (MAP), and mean ocular perfusion pressure (MOPP) were calculated as follows: MAP = DBP + 1/3 (SBP − DBP), MOPP = 2/3MAP − IOP.³² 

The size of the retinal tear and the extent of the retinal detachment were calculated as the ratio of retinal tear or extent of detachment area to the area of the entire fundus photograph taken with the Optomap camera. 

Laser Speckle Flowgraphy (LSFG)

The LSFG-NAVI (Softcare Co., Ltd.) instrument was used to determine the ONH blood flow. The principles of LSFG have been described in detail.^33–36 Briefly, this instrument consists of a fundus camera equipped with an 830-nm diode laser as the light source and a standard charge-coupled device sensor (750 width × 360 height pixels) as the detector. After switching on the laser, a speckle pattern appears due to the interference of the light scattered by the movements of the erythrocytes. The MBR is a measure of the relative blood flow velocity, and it is determined by examining the pattern of the speckle contrast produced by the erythrocytes in the ocular blood vessels. The MBR images are acquired at a rate of 30 frames/seconds over a 4-second period. The same site can be measured by using the auto-tracking system. To evaluate the circulation on the ONH, a circular marker was set surrounding the ONH (Fig. 1A). The “vessel extraction” function of the software then identified the vessel and tissue areas on the ONH so that the MBR of each could be assessed separately (Fig. 1B). The MBR of the vessel (MBR-vessel) and tissue areas (MBR-tissue) on the ONH were determined. The software in the instrument was able to track and compensate for the eye movements during the measurement period. The LSFG was measured three times at each time-point in all eyes. The average of the variables derived from the LSFG device was calculated. The ratio of the MBR of the affected eye to the fellow eye was used to evaluate the relationship between the retinal blood flow rate and the other variables. 

Figure 1

 

Representative composite color maps of the MBR as measured by LSFG. (A) Red color indicates a high MBR and the blue color indicates a low MBR. To measure the MBR of the blood flow on the ONH a circle was set around the ONH. (B) A binary formatted image for segmentation between the vessel (white area) and tissue (black area) areas. (C) Changes in the MBR on the ONH during one heartbeat. White arrow indicates one beat. (D) Pulse waves showing changes in the MBR, which are synchronized with the cardiac cycle for 4 seconds. The total number of frames in one scan is 118.

The laser speckle flowgraphy analyzer software allows the recording of images synchronized to each cardiac cycle and derives various heartbeat waveform parameters (Figs. 1C, 1D). The mean blur rate for the overall area of the ONH was used for the pulse waveform analyses. All eight parameters of the pulse waveform analysis of LSFG were studied.³⁷ 

Surgical Technique

Standard three-port PPV was performed with 25-G instruments after retrobulbar anesthesia with 2.5 mL of 2% lidocaine and 2.5 mL of 0.5% bupivacaine. None of the patients had concurrent scleral buckling surgery. Cataract surgery was performed in eyes with a cataract. To begin the PPV procedure, a trocar was inserted at an approximate angle of 30° parallel to the limbus. Once the trocar was past the trocar sleeve, the angle was changed to be perpendicular to the retinal surface. After creating the three ports, PPV was performed using the Constellation system (Alcon Laboratories, Inc., Fort Worth, TX, USA). 

In eyes with a RRD, the vitreous was removed as completely as possible, then fluid-air exchange and subretinal fluid drainage from the causative retinal tear(s) or iatrogenic hole were performed. Endophotocoagulation was applied to the causative retinal tear(s) or any iatrogenic holes. Then 20% sulfur hexafluoride (SF₆) was injected into the vitreous upon completion of the PPV. After the IOP was adjusted to normal levels, the cannulas were withdrawn. The sclera was pressed and massaged with an indenter or sutured with 8-0 vicryl to close the wound. 

In eyes with an ERM, the ERM and ILM were peeled from the retina using ILM-peeling forceps (25+ Grieshaber Revolution DSP ILM forceps; Alcon Laboratories, Inc.) after core vitrectomy. Next, air was injected into the vitreous cavity to close the scleral port if needed. 

In cataract surgery, a 2.4-mm wide self-sealing superior sclerocorneal tunnel was created at 12 o'clock, and a continuous curvilinear capsulorhexis was performed. The lens nucleus was removed and the residual cortex was aspirated with an irrigation/aspiration (I/A) tip. Next, a foldable acrylic IOL was implanted into the capsular bag. 

In segmental scleral buckling procedure, retinal breaks were identified in all patients and were treated by transscleral cryotherapy. Mattress sutures were placed 7.0 to 7.5 mm apart with 4-0 supramid (Kono, Chiba, Japan) for the segmental buckle and a silicone sponge (Mira No. 506; Mira, Inc., Waltham, MA, USA) was sutured as an explant in all cases. Neither the scleral dissection, extraocular muscle disinsertion, nor the concomitant encircling was required for any patients. Subretinal fluid drainage was performed if necessary. 

Statistical Analyses

We evaluated the changes in MBR-vessel by a mixed-model method to incorporate appropriate covariates between repeated measured values over time. Specifically, we assumed the following model:  

 

i(subject) = 1,…,31, j(time) = 0, 10 (days), 1, 2, 3, 6 (months) where y_ij is the MBR at time j of subject i. a_i is a subject-specific random effect, G_i = 1 represents the affected eye and 0 the unaffected eye. The function f(t_j:b), which represents a fixed effect of time on the refraction, was specified as a polynomial function:  

 

The b parameters represent the fixed-time effects and interaction between the time and group effects, respectively. The order of polynomials in f(t_j,G_i:b) was selected on the basis of the Akaike information criteria (AIC). For the residual term ε_ij of the refraction value, we assumed a heterogeneous compound symmetrical structure within patients.  

Chi-square tests were used to compare categorical data, and independent t-tests were used to compare normally distributed data. Repeated 1-way ANOVA with post hoc Bonferroni corrections was used to evaluate changes in the MBR, IOP, MOP, and pulse waveform parameters. We used paired t-tests to compare normally distributed data between the affected eyes and the fellow eyes at specific time-points. Pearson correlation coefficient analysis was used to determine the correlations between the ratio of preoperative MBR-vessel in the affected eyes to those in the fellow eyes and the independent variables such as the size of retinal tear, extent of RD, preoperative MOPP, heart rate, IOP, age, visual acuity, and axial length. All statistical analyses were performed with SAS9.4 (SAS, Inc., Cary, NC, USA). A P less than 0.05 was considered statistically significant. 

Results

Patient Demographics and Surgical Characteristics

One hundred thirty-six eyes of 136 patients underwent PPV in our department for the repair of a RRD between July 2013 and March 2015. Of these, 105 eyes were excluded for macula-off RRD (n = 76), presence of proliferative vitreoretinopathy grade C or worse (n = 2), vitreous hemorrhage (n = 8), glaucoma (n = 4), diabetic retinopathy (n = 2), or an inability to attend regular follow-up visits (n = 13). In the end, 31 eyes with macula-on RRD (mean age, 55.8 ± 10.6 years) were studied. 

The demographics and surgical procedures to treat the RRD and the control patients are shown in Table 1. Forty-six eyes of 46 patients with macula-on RRD that underwent scleral buckling (mean age, 44.6 ± 17.4 years), 15 eyes of 15 patients with an ERM (mean age, 72.0 ± 9.1 years), and 15 eyes of 15 patients that underwent cataract surgery (mean age, 72.5 ± 11.9 years) were studied as controls. There were significant differences in the age (P < 0.001), axial length (P < 0.001), and surgical procedures (P < 0.001) among the groups. 

Table 1

 

Clinical Characteristics of Subjects

 

In the RRD treated by PPV group, PPV alone was performed on 20 eyes (PPV alone) and PPV combined with phacoemulsification and IOL implantation was performed on 11 eyes (combined surgery). 

Changes in MBR in Eyes With RRD, ERM, and Cataract

The findings in a representative case of RRD treated by PPV are shown in Figure 2. This was a 48-year-old woman who had macula-on RRD in the right eye which was successful repaired. The MBR-vessel before surgery was 25.4 arbitrary units (AU), which increased to 41.6 AU at week 24 after the surgery (Fig. 2A). Although the peak time of the MBR was similar within one cycle before and after the surgery, a higher maximum MBR and a greater increase of the MBR to the peak was observed with increasing time after the PPV (Fig. 2B). 

Figure 2

 

Color map of MBR obtained by LSFG. (A) Composite color before vitrectomy (left), 10 days (middle), and 24 weeks (right) after the vitrectomy. The MBR of the vessel areas of the ONH increases after surgery, whereas the MBR of the tissue areas does not change significantly. (B) The change in the MBR on the ONH in one heart beat in a patient with a RRD before (left), 10 days (middle), and 24 weeks (right) after vitrectomy. MV, mean blur rate of vessel areas; MT, mean blur rate of tissue areas.

Preoperatively, the mean MBR-vessel on the ONH was significantly lower in eyes affected by RRD than in the fellow unaffected eye and this held until 10 days postoperatively in the RRD treated by PPV group (Fig. 3A; Table 2). A significant increase was observed in the mean MBR-vessel of the ONH from 33.7 ± 6.7 AU before surgery to 39.2 ± 6.1 AU at week 24 (P < 0.001) in the RRD eyes. There was no significant change over time in the unaffected fellow eyes. No significant difference was found in the mean MBR-issues on the ONH between the RRD and unaffected eyes before and after the surgery (Fig. 3A; Table 2). There was no significant difference in the MBR-tissue over time in both the RRD and the unaffected eyes (Fig. 3B; Table 2). 

Figure 3

 

Changes in the MBR of the vessel areas of the ONH of eyes with RRD. (A) The mean MBR of the vessel areas of the ONH is significantly lower in eyes with a RRD than in the fellow unaffected eye preoperatively and until 10 days postoperatively. The mean MBR of the vessel areas of the ONH at 3 and 6 months after vitrectomy is significantly increased relative to that before the surgery in an eye with a RRD. No significant difference is observed in the fellow eye. (B) No significant difference in the mean MBR of the tissue area in eyes affected by RRD and the fellow eye before and after the surgery. There is no significant difference in the MBR of the tissue area over time in both the RRD and fellow eyes.

Table 2

 

Change in MBR, IOP, and Circulating Parameters in Eyes With RRD Treated by PPV

 

The MBR-vessel on the ONH of eyes that underwent PPV alone was compared with that in eyes that had had combined surgery. There were no significant differences in the MBR-vessel or the tissues on the ONH between these two groups before and after surgery. However, a significant increase was observed in the MBR-vessel with increasing time in both groups (Table 3). 

Table 3

 

Comparison of Vitrectomy Alone and Combined Surgery Regarding the MBR Before and After Surgery in Eyes With RRD Treated by PPV

 

Composite color maps of the MBR in a representative eye treated by scleral buckling are shown in Figure 4A. While the mean MBR-vessel and MBR-tissue on the ONH did not change significantly with time in eyes treated by scleral buckling (Figs. 4B, 4C), the mean MBR-vessel on the ONH was significantly lower in eyes affected by RRD than in the fellow unaffected eye before and after surgery, even in the final observation time point 24 weeks after surgery (Fig. 4B). 

Figure 4

 

Composite color map and the changes in the MBR on the ONH in the eye with RRD treated by scleral buckling. The composite color map before scleral buckling (left), 10 day (middle), and 24 weeks (right) after the scleral buckling (A). (B) The mean MBR of the vessel areas of the ONH is significantly lower in eyes with a RRD than in the fellow unaffected eye preoperatively and postoperatively. There is no significant change in the MBR of the vessel area in the operated and the fellow eyes during the follow-up period. (C) There is no significant change in the MBR of the tissue area of the ONH in the operated and the fellow eyes during the follow-up period.

Composite color maps of the MBR in a representative eye with an ERM and an eye with a cataract are shown in Figures 5A and 5B. The mean MBR-vessel and MBR-tissue on the ONH did not change significantly with time in eyes with an ERM and with a cataract (Figs. 5C, 5D). In addition, no significant differences were detected between eyes with an ERM or with a cataract and the fellow eye at any time-point. 

Figure 5

 

Composite color map and the changes in the MBR on the ONH in an eye with an ERM and eye with a cataract. The composite color maps before vitrectomy (left), 10 day (middle), and 24 weeks (right) after the vitrectomy in eye with an ERM (A) and with a cataract (B). (C) There is no significant change in the MBR of the vessel area in eyes with an ERM and with a cataract during the follow-up period. (D) There is no significant change in the MBR of the tissue area of the ONH in eyes with an ERM and with a cataract.

Changes in Pulse Waveform Parameters in Eyes With RRD, ERM, and Cataract

Of the eight pulse waveform parameters, only the mean flow acceleration index (FAI) was significantly lower in eyes with a unilateral RRD than in the fellow eyes preoperatively in eyes with RRD treated by PPV (P = 0.001; Fig. 6, Table 4). A significant increase was observed in the FAI following surgery in eyes with a RRD (P < 0.001), but no significant change was observed in the fellow eye over time (Fig. 6B). 

Figure 6

 

Flow acceleration index in eyes with a RRD and normal fellow eye. (A) Flow acceleration index was calculated from the maximum change among all frames (1/30 seconds) in a rising curve. (B) The mean FAI is significantly lower in eyes with a RRD than in the fellow eyes preoperatively. A significant increase is observed in the FAI following surgery in eyes with a RRD (P < 0.001), whereas no significant difference is observed over time in the fellow eye.

Table 4

 

Preoperative Value of MBR and Pulse Waveforms in Eyes With RRD Treated by PPV

 

There were no significant changes in the other seven waveform parameters in eyes with a RRD. In addition, no significant changes were observed in all of the waveform parameters in the unaffected fellow eye of the patients with unilateral RRD treated by PPV, eyes with RRD treated by scleral buckling and eyes with an ERM or a cataract. 

Correlation Between MBR and Other Parameters in Eyes With RRD Treated by PPV

The preoperative parameters in eyes with a RRD are shown in Table 5. The MBR ratio of the affected eye to the fellow eye was 82.2 ± 15.8%. The relative size of the retinal tear was 0.74 ± 0.65%, and the extent of the retinal detachment was 25.6 ± 11.6% of the entire photographed retina. The flare intensity was 12.3 ± 20.2 photon count/ms in eyes with a RRD. 

Table 5

 

Preoperative Characteristic in Eyes With RRD Treated by PPV

 

Pearson correlation coefficient analyses showed that there was no significant correlation between the preoperative MBR ratio and other parameters (Table 6). 

Table 6

 

Results of Pearson Correlation Coefficient Between the Ratio of Eyes Affected by RRD to Fellow Eye in Preoperative MBR Vessel and Other Variables in Eyes With RRD Treated by PPV

 

The trend of the changes in the MBR on the ONH was not correlated with the BCVA, IOP, MOPP, and flare (Table 7). 

Table 7

 

Correlation Between MBR Vessel and Other Factors With Time in Eyes With RRD Treated by PPV

 

Discussion

Our results showed that the mean preoperative MBR-vessel on the ONH in RRD eyes was lower than that of the fellow unaffected eyes, and it increased significantly with increasing time after the PPV. The trend of the changes in the MBR-vessel on the ONH over time was not correlated with the IOP or the MOPP. In addition, only the FAI of the eight pulse waveform parameters in RRD eyes was significantly lower than that of the fellow eyes preoperatively, and then it increased significantly with increasing time after the PPV. On the other hand, while the mean preoperative MBR-vessel on the ONH in the scleral buckling-treated RRD eyes was lower than that of the fellow unaffected eyes, it did not change with time after scleral buckling. 

It was not easy to measure the preoperative retinal blood flow at the detached retinal area because the retinal vessels are tortuous and ascend vertically on the surface of the RPE in a bullous detached retina and each retinal vessel has a different diameter and blood flow velocity. Thus, these alterations in the retinal morphology make it difficult to determine whether the preoperative retinal blood flow is reduced in the detached area compared with that in the normal attached retina. Accordingly, we evaluated the preoperative retinal blood flow for the vascular areas of the ONH in eyes with a RRD and compared it with that in the fellow unaffected eyes. MBR-vessel on the ONH is separated from MBR using the “vessel extraction” function of the software (i.e., which can be dominantly expressed as retinal blood flow). Our results showed that mean preoperative MBR-vessel on the ONH in RRD eyes was lower than that of the fellow eyes preoperatively. 

The normality of the preoperative ocular blood flow in eyes with a RRD is still controversial. There have been several studies that reported that the blood flow in the central retinal artery and the ophthalmic artery in eyes affected by RRD was normal before surgery.^38,39 On the other hand, Eshita et al.⁸ reported that the retinal microcirculation in the macular area was altered in RRD eyes, and the degree of alteration was correlated with the extent of the RRD. Our results are in general agreement with their results in terms of the retinal circulation in eyes affected by RRD before both of PPV and scleral buckling. However, there was no significant correlation between the reduction in the retinal blood flow and the extent of the RRD in our patients. All of the PPV-treated RRD eyes had a PVD, which resulted in the bullous and tortuous retinal detachment, and the height and flection of the detachment varied among the patients. The discrepancy between our findings and their results is unclear, but might be caused by the characteristics of the eyes studied, measurement methods, and measured areas. 

Earlier studies showed that PPV resulted in an increase in the retinal oxygen concentration,⁴⁰ which led to vasoconstriction by the autoregulation of the retinal blood flow. These changes resulted in a decrease in the retinal blood flow.⁴¹ On the other hand, a recent study reported that these changes are transient, and the blood flow returns to the previtrectomy state after 150 minutes.⁴² Our results demonstrated that there were no significant differences in the MBR-vessel on the ONH in eyes with an ERM or a cataract following the surgery. In fact, eyes with an ERM underwent PPV combined with cataract surgery, and eyes with a cataract had no significant change in the MBR-vessel on the ONH. These findings indicated that it is not likely that the PPV affected the blood flow on the ONH following surgery. 

Interestingly, the lower preoperative MBR-vessel on the ONH significantly increased following PPV in eyes with a RRD. The change of the MBR-vessel on the ONH over time was similar in spite of the differences in the surgical procedures for PPV alone and for the combined surgery. These results are consistent with finding that the MBR-vessel on the ONH in eyes with a cataract did not change following surgery as was also observed in the control eyes. 

Although the MOPP can affect the ocular blood flow, there was no significant correlation between the MBR-vessel on the ONH and the MOPP. In addition, no other parameters (e.g., IOP and flare) were significantly correlated with the MBR-vessel on the ONH with increasing postoperative times. 

The FAI in the RRD eyes was lower than that of the fellow eye but it increased following PPV along with the increase of the MBR-vessel in eyes with successful RRD repair. The FAI is calculated from the maximum change of all frames (1/30 seconds) in a rising curve expressing the speed which the blood flow is elevated. These results suggest that the normal blood flow was disturbed in the tortuous vessels running vertically in eyes with bullous retinal detachment before surgery, and the blood flow became normal postoperatively because the retina was attached and flattened. 

There have been only a few studies that reported an increase in the retinal blood flow after successful reattachment of a RRD. In most of these reports, the RRD was reattached by scleral buckling. Scleral buckling causes compressive forces on the peripheral vasculature by the indentation, which result in a decrease of ocular blood flow. In addition, the procedures of scleral buckling reported are not consistent (e.g., use of an exoplant^8,10 or implant,^5,11 and concomitant use of these with the encircling procedures^5,8,10,11). Thus, a decrease in the rate of ocular blood flow and the time course of blood flow varies among the different studies.^5,8,10,11 Our results showed that the mean MBR-vessel on the ONH did not change with time in eyes with RRD treated by segmental scleral buckling without encircling, while the preoperative MBR-vessel in RRD eyes was lower than that of the fellow unaffected eyes. The preoperative situation was the same but the postoperative progress was different between the PPV-treated and the scleral buckling eyes with a RRD. 

Taken together, our findings indicate that the blood flow of the vascular areas of the ONH, representing retinal blood flow, decreases in eyes affected by RRD and returns to the level before the onset of RRD following PPV. In addition, the scleral buckling procedures might mask the real time course of ocular blood flow following successful RRD repair. 

There are several reasons why the retinal blood flow is reduced in eyes with a RRD. First, the detached sensory retina should not require as much oxygen as the attached retina because the photoreceptors are the main consumers of oxygen and they do not function well in the situation of a retinal detachment. Experimental studies have demonstrated a loss of the outer segments of the photoreceptors due to a RD, thereby disrupting the normal outer segment renewal and leading to outer segment shortening and eventual degeneration of the inner segments.^43–46 

A detachment of the neural retina from the RPE induces a variety of changes in several cell types (e.g., photoreceptors, Müller cells, and other retinal neurons).⁴⁷ Recent OCT studies have shown disruptions of the photoreceptor microstructures and the integrity of the outer retinal zones in cases of macula-off RRD.^48–51 These changes are accompanied by a thinning of several retinal layers in eyes with a RRD.⁵² If the same volume of retinal blood flow with the same concentration oxygen passes into the detached retina, the retina is fed with a relatively higher oxygen concentration than the normal attached condition. The situation would then be similar to hyperoxia, which should lead to arterial vasoconstriction by autoregulation of the retinal blood flow resulting in a decrease of retinal blood flow. After successful RRD repair, the photoreceptor layer thickness and the foveal area thickness increase in parallel with an improvement of vision.⁵² These changes should cause the photoreceptors to return to normal oxygen consummation. 

Second, the normal blood flow is disturbed by the tortuous vessels running vertically in eyes with a bullous retinal detachment before surgery which results in a lower MBR-vessel on and FAI in the ONH. After the PPV, the retina is flattened and the blood flow rate recovers to the normal state resulting in an increase in the MBR-vessel on the ONH. There should also be an increase in the FAI. 

In the tissue areas of the ONH, the MBRs was not significantly different between the RRD eyes and the fellow eyes at all experimental times, and no significant differences were observed with time in all of the groups. These results confirm that PPV itself does not affect the MBR on the ONH. 

This study has several limitations. First, the diameters of the retinal vessels were not measured. Ogasawara et al.⁵ reported that the retinal arteries become slightly narrower following scleral buckling but no significant change was observed as opposed to the changes in the blood flow velocity. These results suggest that the changes of the retinal vessel diameter before and after surgery might be small. A second limitation is the effect of PPV for retinal blood flow should be determined in eyes that have undergone PPV alone. However, there was no significant difference in the MBR-vessel on the ONH in eyes with cataract and ERM, which underwent combined surgery. These findings indicate that the MBR-vessel on the ONH in RRD eyes would not be affected by PPV. Third, we did not evaluate the height or volume of the retinal detachment. The functional recovery after macula-off RRD is correlated with the height of the macular detachment,⁵³ which would suggest that these parameters might affect retinal blood flow. And fourth, the size of the sample was not large enough to make strong conclusions. Further studies with a larger number of patients are needed to confirm our findings. 

In conclusion, our results showed a significant reduction of MBR-vessel on the ONH in eyes affected by RRD preoperatively, and the MBR-vessel increased following successful RRD repair by PPV. There was no significant correlation between the increase in the MBR-vessel on the ONH and other factors. These findings indicate that the retinal blood flow is reduced in eyes affected by RRD preoperatively and can recover following successful RRD repair by PPV. 

Acknowledgments

Supported by a Grant-in-Aid for Scientific Research (C) (TI; Tokyo, Japan) and a Grant-in-Aid for Scientific Research (B) (HT; Tokyo, Japan). 

Disclosure: T. Iwase, None; M. Kobayashi, None; K. Yamamoto, None; K. Yanagida, None; E. Ra, None; H. Terasaki, None