Ultrasound Biomicroscopy of Patients With Hemorrhagic Complications Following Intravitreal Anti–Vascular Endothelial Growth Factor Injection

Abstract

Purpose: To describe a case series of patients who experienced hemorrhagic complications following intravitreal anti–vascular endothelial growth factor (anti-VEGF) injections. Methods: In a small case series of 3 patients with retinal disorders, imaging with ultrasound biomicroscopy (UBM) was performed within 1 day following anti-VEGF injection. Results: The first patient had a mild vitreous hemorrhage, and UBM demonstrated a suspected needle track through the pars plana; the hemorrhage cleared spontaneously. The second patient had a dense vitreous hemorrhage as well as hyphema, and UBM demonstrated a suspected needle track through the pars plicata; the hemorrhage did not clear and the patient was subsequently treated elsewhere with pars plana vitrectomy. The third patient had a subconjunctival hemorrhage but no vitreous hemorrhage, and UBM demonstrated no definite evidence of a needle track. Conclusion: In some patients with vitreous hemorrhage following anti-VEGF injection, a suspected needle track may be imaged with UBM within 1 day following injection.

Introduction

Intravitreal injections of anti–vascular endothelial growth factor (anti-VEGF) agents have become standard treatment for many retinal disorders, including neovascular age-related macular degeneration (AMD), diabetic retinopathy, and retinal vein occlusions. ¹ Some of the observed complications following intravitreal anti-VEGF injections include endophthalmitis, noninfectious intraocular inflammation, and retinal detachment. ² Another complication of intravitreal anti-VEGF injections is vitreous hemorrhage.

In the current case series, we present 3 patients with retinal disorders who received treatment with intravitreal anti-VEGF injections. In all 3 patients, hemorrhagic complications occurred. We describe the findings from imaging including ultrasound biomicroscopy (UBM) 1 day after the injection.

Methods

Review of 3 patients who experienced hemorrhagic complications and were imaged with UBM 1 day after they had received an anti-VEGF injection.

Results

Case 1

A 90-year-old man with a history of neovascular AMD in both eyes received treatment with anti-VEGF injections for many years and returned to our clinic for a scheduled follow-up visit. The patient’s medications included aspirin 81 mg daily, as well as allopurinol, amlodipine, atorvastatin, candesartan, montelukast, omeprazole, and Age-Related Eye Disease Study 2 (AREDS2) nutritional supplements. His corrected visual acuity (VA) was 20/40 OD and 20/50 OS. His intraocular pressure (IOP) was 16 mm Hg in each eye. Slit lamp examination was significant for signs of pseudophakia bilaterally. Dilated fundus examination showed characteristics consistent with a diagnosis of active neovascular AMD in both eyes. The patient was treated with intravitreal injections of faricimab (Vabysmo, Genentech) administered using a 32-gauge needle in both eyes.

Immediately after the anti-VEGF injection, the patient reported seeing blood in his right eye. Examination of the right eye with indirect ophthalmoscopy was notable for a mild vitreous hemorrhage, with no retinal breaks.

When the patient returned to the clinic the next day, his VA was 20/50 OD. Fundus photography confirmed the presence of a mild vitreous hemorrhage in his right eye (Figure 1A). B-scan echography further confirmed the vitreous hemorrhage, with no retinal breaks (Figure 1B). UBM imaging with a 50 MHz probe demonstrated a hypoechoic line through the pars plana, which was deemed to represent the needle track, and there was evidence of hyperechoic tissue extending from the vitreous toward the sclera, which was deemed to represent vitreous incarceration (Figure 1C).

Figure 1.

Case 1, a 90-year-old man with a history of neovascular age-related macular degeneration in both eyes treated with intravitreal anti–vascular endothelial growth factor injections. The patient was assessed 1 day after injection. (A) A fundus photograph (right eye) demonstrates mild vitreous hemorrhage. (B) B-scan echography (right eye) demonstrates vitreous hemorrhage with no retinal breaks. (C) Ultrasound biomicroscopy (right eye) demonstrates a hypoechoic line through the pars plana, deemed to represent the needle track (green arrow). There is also hyperechoic tissue extending from the vitreous toward the sclera, representing vitreous incarceration.

The patient was observed over several weeks. At 5 weeks following the anti-VEGF injection, the patient’s vitreous hemorrhage had cleared spontaneously and was completely resolved. In addition, his VA had improved to 20/30 OD.

Case 2

A 79-year-old woman with a history of neovascular AMD in both eyes received treatment with anti-VEGF injections for many years and returned to our clinic for a scheduled follow-up visit. The patient’s medications included apixaban and clopidogrel, as well as amlodipine, balsalazide, dapagliflozin, dulaglutide, evolocumab, olmesartan, and AREDS2 nutritional supplements. Her corrected VA was 20/200 OD and 20/30 OS. Her IOP was 8 mm Hg OD and 10 mm Hg OS. Slit lamp examination was significant for signs of pseudophakia bilaterally. Dilated fundus examination showed characteristics consistent with a diagnosis of active neovascular AMD in both eyes. The patient was treated with intravitreal injections of faricimab administered using a 32-gauge needle in both eyes.

Later that evening, she emergently reported an acute worsening of vision in her left eye. The patient was examined that evening, 9 hours after the injections. Her VA was light perception OS. Her IOP was 28 mmHg OS. Slit lamp examination demonstrated the presence of hyphema in her left eye. Dilated fundus examination demonstrated that the left eye had a vitreous hemorrhage, with no view to the fundus. The patient was treated with brimonidine 0.2% in the left eye and asked to return to the clinic the following morning.

The next morning, her VA OS had improved to hand motion. In addition, her IOP had improved to 11 mm Hg OS. B-scan echography demonstrated a vitreous hemorrhage with no retinal breaks in her left eye (Figure 2A). UBM demonstrated the presence of hyphema (Figure 2B). In addition, there was a partial hypoechoic line through the pars plicata, which was deemed to represent a needle track (Figure 2C). There was no definite evidence of vitreous incarceration. The patient was informed of the risks and benefits of treatment with pars plana vitrectomy but was advised to allow 1 or 2 weeks for the blood to clear spontaneously before undergoing the procedure.

Figure 2.

Case 2, a 79-year-old woman with a history of neovascular age-related macular degeneration in both eyes treated with intravitreal anti–vascular endothelial growth factor injections. The patient was assessed 1 day after injection. (A) B-scan echography (left eye) demonstrates vitreous hemorrhage with no retinal breaks. Ultrasound biomicroscopy (left eye) demonstrates (B) hyphema and also (C) an incomplete hypoechoic line through the pars plicata, deemed to represent the needle track (green arrow).

Approximately 2 weeks later, the patient elected to return to her home in another state, where she underwent pars plana vitrectomy and continued her anti-VEGF injections. She returned to our clinic for a follow-up visit 1 year later, by which time her VA had improved to 20/40 in the left eye.

Case 3

A 69-year-old woman with a history of central retinal vein occlusion (CRVO) as well as macular edema in the left eye received anti-VEGF injections for 2 years and returned to our clinic for a scheduled follow-up visit. The patient’s medications included ibuprofen as needed. Her corrected VA was 20/25 OD and 20/30 OS. Her IOP was 12 mm Hg OD and 14 mm Hg OS. Slit lamp examination demonstrated the presence of nuclear sclerosis in both eyes. Dilated fundus examination showed characteristics consistent with a diagnosis of CRVO with active macular edema in her left eye. The patient was treated with intravitreal injections of aflibercept 2 mg (Eylea, Regeneron) administered using a 32-gauge needle in the left eye.

On the evening following the anti-VEGF injection, the patient reported seeing a white scotoma as well as having symptoms of pain in her left eye. On examination the next day, her VA was 20/30 OS and her IOP was 15 mm Hg OS. Slit lamp examination demonstrated the presence of a subconjunctival hemorrhage at the injection site in her left eye. Examination by dilated fundus photography demonstrated a clear vitreous, with no evidence of hemorrhage or retinal tear. External photography demonstrated a subconjunctival hemorrhage (Figure 3A). Fundus photography showed characteristics consistent with a diagnosis of CRVO (Figure 3B). UBM showed no definite evidence of a needle track or vitreous incarceration (Figure 3C).

Figure 3.

Case 3, a 69-year-old woman with a history of central retinal vein occlusion and macular edema in the left eye treated with intravitreal anti–vascular endothelial growth factor injections. The patient was assessed 1 day after injection. (A) An external photograph (left eye) demonstrates subconjunctival hemorrhage at the injection site. (B) A fundus photograph (left eye) demonstrates central retinal vein occlusion with no evidence of vitreous hemorrhage. (C) Ultrasound biomicroscopy imaging (left eye) through the area of the injection site (green arrow) demonstrates no definite evidence of needle track or vitreous incarceration.

The patient subsequently returned to our clinic 11 weeks later, at which time her VA was 20/40 OS. Slit lamp examination of the left eye showed that the subconjunctival hemorrhage had resolved, and dilated fundus examination showed stable CRVO with clear vitreous.

Conclusion

Vitreous hemorrhage following intravitreal anti-VEGF injection is uncommon. In a 2008 case series involving a total of 4303 injections of bevacizumab (Avastin, Genentech), vitreous hemorrhage occurred at a rate of 0.023%, or 1 in 4303 injections. ³ In a 2011 review of the Medicare 5% claims database, in which a total of 40,903 anti-VEGF injections were analyzed, 94 cases of vitreous hemorrhage occurred, at a rate of 0.23% or ~1 in 435 injections. ⁴ In a 2013 case series involving a total of 1710 anti-VEGF injections, 3 cases of vitreous hemorrhage occurred, at a rate of 0.18% or ~1 in 570 injections. ⁵ In a 2022 case series involving a total of 73,286 intravitreal injections (of which 97.7% were anti-VEGF injections), there were 18 patients who experienced a vitreous hemorrhage, occurring at a rate of 0.025% or ~1 in 4000 injections. ⁶

In contrast, vitreous incarceration following intravitreal injection may be relatively common. In a prospective case series analyzing a total of 39 injections of ranibizumab (Lucentis, Genentech), the investigators performed UBM imaging 1 day after the injection, which demonstrated the presence of vitreous incarceration in 6 patients (15.3%). ⁷

Based on these previously published case series, in addition to the 3 cases presented here, certain observations can be made. The rate of vitreous incarceration may be as high as 15%, while the rate of vitreous hemorrhage has been reported to be between 0.023% and 0.23%.^3,4,6 In addition, the reported rate of rhegmatogenous retinal detachment following intravitreal injection has been reported to be between 0.0084% and 0.025%.^8–10 Thus, most patients who experience vitreous incarceration following intravitreal injection do not have subsequent vitreous hemorrhage or retinal detachment. Furthermore, vitreous hemorrhage following intravitreal injection may be attributable to trauma to the pars plana or pars plicata, and therefore retinal detachment does not necessarily result.

In 2 of the 3 cases presented, UBM imaging showed a partial or complete hypoechoic line that was interpreted as a needle track. These hypoechoic lines appear larger than what might be anticipated for a 32-gauge needle, which has an outer diameter of ~0.24 mm. It is possible that the UBM image demonstrated fluid within the needle track, which could represent incarcerated vitreous. Alternatively, it is possible that the injection could have been initiated while the needle was only partly through the sclera, and the UBM image was showing some of the fluid from the injection. (All 3 patients were injected by the same physician.)

The third patient, who did not have a vitreous hemorrhage, did not have a suspected needle track on UBM, which would support the hypothesis that the UBM image actually showed incarcerated vitreous. However, in this third patient, UBM imaging was performed through the subconjunctival hemorrhage at the injection site, and it is possible that this subconjunctival hemorrhage created sound attenuation, which may have disrupted the signal and obscured a possible needle track.

The first patient, who had a mild vitreous hemorrhage that resolved spontaneously, had a suspected needle track through the pars plana, a relatively avascular structure. The second patient, who had a dense vitreous hemorrhage with hyphema, had a suspected needle track through the pars plicata, a relatively more vascularized structure. Also of note, the first patient was taking 81 mg of aspirin, while the second patient was taking both apixaban and clopidogrel, and the third patient was taking only ibuprofen intermittently. Therefore, the observations in this series suggest that 2 risk factors for vitreous hemorrhage following anti-VEGF injection are antiplatelet or anticoagulant treatment and injection through the pars plicata.

Vitreous hemorrhage following intravitreal anti-VEGF injection could occur through a variety of mechanisms, including trauma to a blood vessel within the ciliary body, incarceration of vitreous with subsequent traction, temporary hypotony due to delayed closure of the wound, or other possibilities. A prospective series of patients imaged with UBM at 1 day following intravitreal injection might provide additional information, by demonstrating how the sclera typically appears following injection.