Mean Ocular Perfusion Pressure Effect During Pars Plana Vitrectomy on the Foveal Avascular Zone: A Pilot Study

Abstract

Purpose: To evaluate the effect of mean ocular perfusion pressure on the foveal avascular zone (FAZ) area in eyes with successful retinal detachment (RD) surgery. Methods: This prospective pilot fellow eye–controlled study measured the intraoperative mean ocular perfusion pressure in eyes having surgery for rhegmatogenous RD (RRD). Postoperatively, the FAZ area was measured; the change was calculated as the difference in the FAZ area between the operated eye and the contralateral control eye. Results: The study comprised 8 patients with a mean age (±SD) of 53.38 ± 13.92 years. The mean superficial FAZ area was not different between operated eyes and control eyes, while the deep FAZ area was significantly larger in operated eyes. There was a strong negative correlation between the mean ocular perfusion pressure and the change in the deep FAZ area (Spearman ρ, −0.73; P = .04); the correlation between the mean ocular perfusion pressure and the change in the superficial FAZ area was not significant (Spearman ρ, −0.24; P = .57). A significant linear regression was found between the mean ocular perfusion pressure and the change in the deep FAZ area (R² = 0.388). The predicted enlargement of the deep FAZ area was 0.03 mm with every 1 mm Hg decrease in the mean ocular perfusion pressure. Conclusions: Lower intraoperative mean ocular perfusion pressure is associated with enlargement of the deep FAZ area in eyes having successful RRD surgery.

Introduction

Rhegmatogenous retinal detachment (RRD) is a frequently encountered pathology caused by a full-thickness retinal tear that allows vitreous into the subretinal space. ¹ Among the various interventions for RRD repair, pars plana vitrectomy (PPV) is one of the most commonly performed, with primary success rates ranging from 64% to 96%. ² However, it has been observed that despite successful postoperative anatomic attachment of the retina, visual acuity (VA) does not improve beyond a certain level in some eyes. ³ The visual prognosis of RRD is related to the integrity of the foveal structure and dimensions of the foveal avascular zone (FAZ).^4–8

Many causes for a poor visual prognosis after a successful PPV have been reported. The most popular theory is direct damage to the retinal nerve fiber layer or perfusion of the optic nerve head, while an earlier supposition was a dehydrating retinal injury during a fluid–air exchange.^8–10 However, the exact mechanism has yet to be elucidated.

Rossi et al ¹¹ found a significant acute and prolonged drop in the mean ocular perfusion pressure during the course of vitrectomy, which is a possible reason for unexplained visual deficits after PPV. However, to our knowledge, no report has provided quantitative evidence for the relationship between the mean ocular perfusion pressure and visual deficits after PPV. Therefore, this study evaluated the effect of the mean ocular perfusion pressure on the FAZ in eyes with successful RD surgery after PPV.

Methods

This prospective case-control pilot study was performed at the Ophthalmology Department, Shifa International Hospital, Islamabad, between April 2020 and December 2020. Ethical approval for this study was obtained from the hospital’s institutional review board (IRB#050-870-2020). Written informed consent was obtained from all patients before the study.

Patients with a unilateral primary uncomplicated RRD planned for PPV with gas tamponade were included in the study. Exclusion criteria were active diabetes or a history of diabetes; hypertension; dyslipidemia; stroke; retinal vascular disease, dystrophy, or choroidal neovascularization; intraocular inflammation; and ocular surgery, including intraocular injections and laser treatment. Patients who did not achieve retinal attachment in a single surgical procedure were also excluded.

All eyes had a complete ophthalmic examination on the first postoperative day, 2 weeks postoperatively, and then monthly.

Surgical Technique

Eyes included in the study had a 25-gauge PPV (Constellation Vitrectomy System, Alcon Laboratories) by the same vitreoretina surgeon (M.A.A.). A noncontact, wide-angle viewing system (BIOM, Oculus Inc) was used for fundus visualization. A modified retrobulbar block with 1% lignocaine and 0.5% bupivacaine was used for anesthesia.

After core and peripheral vitrectomies were performed, endolaser, cryotherapy, or both were used to seal the breaks. Toward the end of surgery, intraocular gas (sulfur hexafluoride [SF₆], perfluoroethane [C₂F₆], or perfluoropropane [C₃F₈]) was used as a tamponade agent, followed by a subconjunctival antibiotic and corticosteroid injection. In each eye, surgery began at an infusion pressure of 25 mm Hg, with 40 mm Hg being used during fluid–air exchange. The infusion pressure was reduced to 20 mm Hg during wound closure and injection of gas.

Mean Ocular Perfusion Pressure Measurements

For patients who met the inclusion criteria, the perioperative infusion pressure displayed on the vitrectomy surgical monitor was recorded every 3 minutes. The supine blood pressure was recorded every 3 minutes. At the end of the surgery, the patients’ mean ocular perfusion pressure was calculated from the mean infusion pressure, mean systolic pressure, and mean diastolic pressure using the following equations:

$$\mathrm{MAP}=2/3\mathrm{DNBP}+1/3\mathrm{SNBP}$$

where MAP is the mean arterial (brachial) pressure in the supine position, DNBP is the diastolic noninvasive (brachial) blood pressure, and SNBP is the systolic noninvasive (brachial) blood pressure.

$$\mathrm{MOAP}=115/130(\mathrm{MAP})$$

where MOAP is the mean ophthalmic artery pressure in the supine position and MAP is the mean arterial pressure in the supine position.

$$\mathrm{MOPP}\mathrm{=}\mathrm{MOAP}-\mathrm{MIP}$$

where MOPP is the mean ocular perfusion pressure, MOAP is the mean ophthalmic artery pressure in the supine position, and MIP is the mean infusion pressure.

OCTA Imaging and FAZ Area Calculation

For all patients with successful single-procedure RRD repair, optical coherence tomography angiography (OCTA) was performed after the gas tamponade was completely absorbed (3 to 12 weeks postoperatively). Fovea-centered 4.5 mm × 4.5 mm OCTA scans (320 × 320 clusters of 4 repeated B-scans) were acquired of the operated eyes and fellow control eyes using swept-source OCTA (DRI Triton, Topcon) and the built-in OCTARA algorithm. En face OCTA images of the superficial vascular plexus and deep capillary plexus were generated using the machine’s default segmentations. Images with motion, shadow, or projection artifacts and segmentation errors were repeated until satisfactory results were obtained. The same masked retinal grader (H.A.K.) manually segmented and measured the deep FAZ area and superficial FAZ area after contrast enhancement using ImageJ (version 1.51p, US National Institutes of Health). Figure 1 and Figure 2 show the segmented superficial FAZ area and deep FAZ area, respectively.

Figure 1.

Foveal avascular zone segmentation in the superficial vascular plexus on optical coherence tomography angiography.

Figure 2.

Foveal avascular zone segmentation in the deep capillary plexus on optical coherence tomography angiography.

Statistical Analysis

Data were analyzed using SPSS software (version 21.0, IBM). Quantitative data are presented as the mean ± SD, while categorical data are presented as counts (percentages). The FAZ area in the superficial vascular plexus and deep capillary plexus was compared between the operated eyes and control eyes as well as between operated eyes with macula-off RRD and those with macula-on RRD using an independent-sample t test.

Considering the high interocular symmetry and correlation with the fellow eye, the difference between the control eye and operated eye was considered a change induced by the mean ocular perfusion pressure in the operated eye and calculated as follows: Change in the FAZ area = FAZ area in control eyes − FAZ area in operated eyes. The difference was calculated for the deep FAZ area and superficial FAZ area as the change in the deep FAZ area and in the superficial FAZ area. The effect of the mean ocular perfusion pressure on the change in the FAZ area was determined using the Spearman rank correlation. For all tests, a P value < .05 was considered significant.

Results

The final analysis included 8 patients (6 men [75%]) with a mean age of 53.4 ± 13.9 years. Surgery was performed in the right eye in 5 patients and the left eye in 3 patients. Six eyes (75%) had a macula-off RRD, and 2 eyes had a macula-on RRD. The mean operating time was 47.5 ± 13.6 minutes. Three eyes (37.5%) had SF₆ gas tamponade, while C₂F₆ gas and C₃F₈ gas was used in 3 eyes (37.5%) and 2 eyes (25%), respectively. Table 1 shows the patients’ demographics and clinical parameters. Table 2 shows the duration of the RRD and the preoperative and postoperative VA.

Table 1.

Demographics and Clinical Characteristics.

		Pressure (mm Hg)				Foveal Avascular Zone Area (mm)
		Noninvasive Brachial Blood				Superficial			Deep
Eye	Age (Y)/Sex	Systolic	Diastolic	Mean Infusion	Mean Ocular Perfusion	Operated Eye	Control Eye	Difference	Operated Eye	Control Eye	Difference
1	70/M	140.0	103.0	41.1	60.8	0.492	0.282	0.210	0.611	0.584	0.027
2	33/M	114.4	76.6	26.9	52.0	0.339	0.338	0.001	0.384	0.299	0.085
3	55/M	163.6	98.6	46.7	59.6	0.288	0.275	0.013	0.522	0.488	0.034
4	52/F	164.8	93.3	57.8	45.8	0.528	0.486	0.042	1.110	0.944	0.166
5	43/M	127.0	77.4	23.5	59.6	0.319	0.256	0.063	0.336	0.288	0.048
6	76/F	151.0	91.4	50.6	48.0	0.623	0.228	0.395	0.915	0.247	0.668
7	50/M	149.6	91.2	42.6	55.3	0.252	0.280	−0.028	0.318	0.387	−0.069
8	48/M	130.2	89.4	35.3	55.8	0.153	0.235	−0.082	0.258	0.191	0.067

Table 2.

Duration of RRD and Preoperative and Postoperative Visual Acuity.

	RRD Duration (Days)	LogMAR Visual Acuity
Eye		Preoperative	Postoperative
1	2	0.50	0.00
2	14	0.9	0.10
3	10	1.00	0.4
4	4	1.00	1.00
5	8	0.90	0.30
6	7	1.00	0.30
7	14	0.10	0.00
8	10	0.10	0.10

Abbreviation: RRD, rhegmatogenous retinal detachment.

The perioperative overall mean infusion pressure was 40.56 ±  11.6 mm Hg, the mean systolic noninvasive (brachial) blood pressure was 142.6 ± 17.9 mm Hg, and the mean diastolic noninvasive (brachial) blood pressure was 90.1 ± 9.22 mm Hg. The mean ocular perfusion pressure was 54.6 ± 5.58 mm Hg. Only 2 eyes had a perioperative mean ocular perfusion pressure lower than the physiologic range of 50 to 60 mm Hg. In these 2 eyes, the mean systolic noninvasive (brachial) blood pressure was toward the higher end at 164.8 mm Hg and 151.8 mm Hg, respectively, with subsequent increases in mean infusion pressure readings of 57.8 mm Hg and 50.6 mm Hg, respectively, which led to an overall decrease in the mean ocular perfusion pressure.

The mean superficial FAZ area and deep FAZ area were not different between operated eyes and control eyes. Eyes with a macula-off RRD had a larger superficial FAZ area (mean, 0.43 ± 0.14 mm) than eyes with a macula-on RRD (mean, 0.20 ± 0.07 mm) (t = 2.2; df, 6; P = .04). The mean deep FAZ area was also significantly larger in eyes with a macula-off detachment than in eyes with a macula-on detachment (0.647 ±  0.03 mm vs 0.288 ± 0.042 mm) (t = 2.78; df, 6; P = .04).

The mean change in the superficial FAZ area and deep FAZ area was 0.077 ± 0.154 mm and 0.129 ± 0.228 mm, respectively. There was a strong negative correlation between the mean ocular perfusion and deep FAZ area (Spearman ρ, −0.73; P = .04); however, the correlation between the mean ocular perfusion pressure and change in the superficial FAZ area was not significant (Spearman ρ, −0.24; P = .57). A significant linear regression was found between the mean ocular perfusion and the change in the deep FAZ area (R² = 0.388), and the predicted enlargement of the deep FAZ area was 0.03 mm with every 1 mm Hg decrease in the mean ocular perfusion pressure (Figure 3).

Figure 3.

Effect of mean ocular perfusion pressure on the deep FAZ area.

Abbreviation: FAZ, foveal avascular zone.

Conclusions

The fovea is an exceptionally visually sensitive area of the retina that is responsible for central vision. It is primarily composed of cone photoreceptors with elongated outer segments and the FAZ, which is enclosed by a network of closely connected capillaries comprising the superficial vascular plexus and deep capillary plexus of the retinal vasculature.^12,13

The size of the fovea differs among healthy individuals; however, this difference does not affect visual function except in retinal pathologies where the visual prognosis can be correlated with the size of the FAZ.^13–15 In RRD, hypoxia in the detached retina occurs at the outer layers as a result of deprivation of crucial nutrients by the subretinal fluid (SRF) as well as the inner retinal layers caused by the relative hypoxia of the deep capillary plexus resulting from a relatively low perfusion pressure. ⁸ Other reported vascular abatements in the detached retina include increased vascular permeability caused by inflammatory mediators in the SRF, vascular spasms, and vascular occlusion.^16–18

Postoperative increases in the FAZ at the superficial vascular plexus and deep capillary plexus after successful PPV have been reported in the literature, with a more marked difference in the macula-off group than in the macula-on group, indicating ischemic damage at the macular capillary plexus.^3,8,19,20 However, Francisconi et al ²¹ recently found that the postoperative FAZ area in eyes that had pneumatic retinopexy was not statistically different from that in fellow control eyes. Similarly, the superficial FAZ area and deep FAZ area were not statistically different between contralateral control eyes and operated eyes in our study; that is, there was a significant negative relationship between the deep FAZ area and the mean ocular perfusion pressure. Each 1 mm Hg reduction in the mean ocular perfusion pressure was associated with enlargement of the deep FAZ area of 1.5 ± 0.72 mm, while the change in the superficial FAZ area was unrelated. The current literature shows that changes in the deep FAZ area precede and are more pronounced than those in the superficial FAZ area in response to vascular/ischemic events,^22,23 possibly because the smaller capillary bed in the deep capillary plexus is more susceptible to ocular perfusion than the larger vessels of the superficial vascular plexus.

To our knowledge, there are no studies of the effects of the mean ocular perfusion pressure on retinal ischemia in the literature. Our study hypothesized that the FAZ size in operated eyes correlates negatively with the mean ocular perfusion pressure during PPV, and our study results show a moderate negative correlation between the 2 parameters. Increased FAZ size has been correlated with a poor visual prognosis, especially at the deep capillary plexus level. ⁸ Therefore, our study adds to the significance of the perioperative infusion pressure and systemic pressures on postoperative retinal ischemia and thus on the visual prognosis.

The mean ocular perfusion pressure is responsible for the sustenance of all intraocular structures and is calculated as a differential between arterial pressure and intraocular pressure (IOP). Although a low mean ocular perfusion pressure can cause irreversible damage to the optic nerve, an acute increase in pressure can obstruct the retrograde transport of neurotrophic factors. ²⁴ During PPV, the IOP increases as a result of different maneuvers; thus, the systemic pressure decreases from the effects of sedation and the patient’s supine posture, which results in a low mean ocular perfusion pressure. This low mean ocular perfusion pressure is the cause of unexplained visual field defects in the postoperative period, even after a successful repair. ¹¹

To achieve the ideal mean ocular perfusion pressure perioperatively (50 to 60 mm Hg), blood pressures should be kept in the normal range (120/80 mm Hg) and infusion pressures should be kept on the lower side (22 to 32 mm Hg). Theoretically, by increasing the infusion pressure, blood pressures have to be decreased, and vice versa, to achieve the ideal mean ocular perfusion pressure. However, infusion pressures have to be increased to control perioperative bleeding; therefore, blood pressures should be slightly toward the lower side, or at least in the normal range, to achieve the ideal mean ocular perfusion pressure.

The mean ocular perfusion pressure in our study was 54.6 ± 5.58 mm Hg. Only 2 eyes (#4 and #6) had a perioperative mean ocular perfusion pressure lower than the physiologic range of 50 to 60 mm Hg. ²⁴ Interestingly, in these 2 eyes, the mean systolic noninvasive (brachial) blood pressure was toward the higher end at 164.8 mm Hg and 151.8 mm Hg, respectively, with subsequent increases in mean infusion pressure readings of 57.8 mm Hg and 50.6 mm Hg, respectively, leading to an overall decrease in the mean ocular perfusion pressure. Both patients were known to be normotensive. Surgical anxiety might have played a role in the increases, as previously suggested in the literature. ²⁵ These higher systemic pressures in turn led to an increased incidence of intraoperative retinal bleeding that resulted from the use of higher infusion pressure and thus the lower mean ocular perfusion pressure. Our results signify the importance of both pressures in vitreoretinal surgery. We should give equal importance to perioperative systemic pressures and infusion pressures as well as surgical maneuvers to achieve the best possible outcomes for our patients.

Our study has several limitations, including a small sample, short follow-up, and lack of longitudinal data. An ideal approach to study the effect of the mean ocular perfusion pressure on FAZ parameters would be to compare the postoperative OCTA with the OCTA before the onset of RRD. However, it is not possible to anticipate which patient will develop an RRD. Considering a high interocular symmetry in the FAZ area, we used the FAZ area in the contralateral control eye to calculate the change in the operated eye. Our findings suggest that a lower mean ocular perfusion pressure is associated with enlargement of the deep FAZ; however, our results are from a pilot study and require further validation by studies that include a large number of patients.