Endovascular Cannulation With a Microneedle for Central Retinal Vein Occlusion

We developed a new surgical treatment in which a microneedle is used for retinal endovascular cannulation to treat eyes with central retinal vein occlusion by flushing thrombus out of the central retinal vein as it passes through the lamina cribrosa. The eyes of 12 consecutive patients (12 eyes) with central retinal vein occlusion were successfully treated using this novel treatment. At 24 weeks after surgery, 9 of 12 eyes had gained more than 15 letters in best-corrected visual acuity, and the mean decrease in central foveal thickness was 271.1 μm. Few complications were observed. The microneedle is stiff and sharp enough to facilitate retinal endovascular cannulation in eyes with central retinal vein occlusion. This new technique is a promising treatment of macular edema due to central retinal vein occlusion.

Although central retinal vein occlusion (CRVO) accounts for few retinal vein occlusion cases, it leads to severe vision loss.^1-3 Because CRVO is thought to be caused by thrombus within the central retinal vein in the lamina cribrosa, recanalization is a reasonable treatment strategy; however, no endovascular treatment has been established for vascular occlusions in the human retina.^4-8 Green et al⁹ demonstrated that CRVO is caused by thrombus in the central retinal vein. The occlusion of the major outflow channel of the retinal circulation in eyes with CRVO increases venous pressure, and this results in macular edema and hemorrhages. Therefore, thrombus removal or chemical thrombolysis is a reasonable approach to the treatment of CRVO.

In the past, surgeons used special glass cannulas to pierce dilated retinal veins, but the cannulas were too difficult to maneuver because of their transparency or frangibility. We developed a novel microneedle having an outer diameter of 50 μm for retinal vessel cannulation.¹⁰ Herein, we report the results of a prospective study of eyes with CRVO in which retinal endovascular cannulation was performed using the microneedle.

This study adhered to the tenets of the Declaration of Helsinki. The protocol was approved by the local institutional research ethics committees, and the data analysis was performed at the Yokohama City University Medical Center, Yokohama City, Japan. Between March 2010 and February 2011, patients with CRVO were screened for study eligibility. Twelve patients who met the eligibility criteria were included in the study, and their demographic characteristics are summarized in Table 1. The inclusion criteria were clinical and angiographic diagnosis of CRVO less than 12 weeks after the onset of CRVO, central foveal thickness exceeding 300 μm as measured by optical coherence tomography, and best-corrected visual acuity (BCVA) between 0 and 65 Early Treatment of Diabetic Retinopathy Study letters (Snellen VA equivalents, 20/1000 and 20/50, respectively). The exclusion criteria were glaucoma, retinal or disc neovascularization, any previous treatment of CRVO, vascular retinopathy due to other causes, and intraocular surgery during the previous 3 months. The primary end point was change in BCVA at 24 weeks after surgery compared with initial VA. The BCVA was measured by examiners who were masked to patient identification. The secondary end points were surgical complications, central foveal thickness alteration, and change in the foveal avascular zone as determined by digital imaging analysis and fluorescein angiography.

A fabricated microneedle having an outer diameter of 50 μm and an inner diameter of 20 μm is used to pierce the dilated retinal vein. The microneedle is made of stainless steel and is manufactured by laser assembly (Figure 1); it is sharp and stiff enough to pierce microvessels. When the microneedle is connected to a 10-mL syringe containing a distilled solution, the solution can be injected into a vessel. The volume and pressure of the solution injected are proportional to the pressure of the syringe connected to a viscous fluid control system, which is controlled by the surgeon's pedal.

The BCVA of 9 of 12 patients had improved by more than 15 letters at 24 weeks after surgery compared with the baseline value (Table 2). The mean VA had improved by 14.1 letters at 6 weeks, by 15.3 letters at 12 weeks, by 15.3 letters at 18 weeks, and by 16.3 letters at 24 weeks. The preoperative mean BCVA of 29.6 letters (20/250) had improved to 45.9 letters (20/125) at 24 weeks after surgery, and the mean decrease in central foveal thickness was 271.1 μm (Figure 3).

No neovascular glaucoma was observed in any of the patients when examined at 24 weeks. All surgical procedures were successful, as confirmed by the streamlined flow during the injection. Intraoperative complications developed in 2 patients and consisted of a mild vitreous hemorrhage in 1 eye and a small subretinal hemorrhage in 1 eye, neither of which impaired VA. No occurrences of retinal tears, endophthalmitis, retinal detachment, severe vitreous hemorrhage, or recurrence of macular edema were observed during the 24-week follow-up period.

We demonstrated that retinal endovascular cannulation performed by piercing the retinal vein and injecting saline solution through a microneedle is a feasible treatment of CRVO and is unaccompanied by any severe surgical complications. The microneedle we used is made of stainless steel and has an outer lumen of 50 μm and an inner lumen of 20 μm. Modern industrial technology has made it possible to fabricate needles with a tiny lumen by laser assembly. Based on the favorable results obtained in an experimental study¹⁰ of the feasibility of using the microneedle, it was applied to human retinal veins. With the microneedle, all eyes with CRVO were successfully treated in this study, without any severe intraoperative complications.

At the 24-week follow-up examination, 9 of 12 eyes had gained more than 15 letters in VA, and the mean change in all 12 eyes at 24 weeks was an increase of 16.3 letters. Because visual improvement is rarely observed during the natural course of ischemic CRVO,¹¹ we consider the high rate of visual improvement and the low rate of surgical complications achieved using our technique in this study to be satisfactory.

The mechanism of this surgical treatment is thought to be elimination of the vascular occlusion by flushing out thrombus,¹² although it is technically impossible to confirm the disappearance of thrombus in the lamina cribrosa. We speculated that our surgical treatment resulted in a reduction in ischemia, an improvement in retinal perfusion, and a suppression of vascular endothelial growth factor production.

Central retinal vein occlusion is associated with severe vision loss, and ischemic CRVO, in particular, has a poor visual prognosis.^2,3 The multicenter Central Vein Occlusion Study¹³ showed that prophylactic panretinal laser treatment is not a beneficial treatment of ischemic CRVO. Although several pharmacological treatment options exist for macular edema associated with CRVO,^14-17 a substantial proportion of patients with CRVO are left with visual impairment, so retinal vein cannulation may merit inclusion as an option to treat CRVO.

In conclusion, our technique using a microneedle facilitates retinal endovascular cannulation. The surgical procedure may lead to visual improvement in eyes with CRVO and is accompanied by few intraoperative or postoperative complications. Further long-term follow-up observations and studies are needed to confirm the efficacy of this treatment.

Correspondence: Kazuaki Kadonosono, MD, Department of Ophthalmology, Yokohama City University Medical Center, Urafune-cho 4-57, Minami-ku, Yokohama 232-0024, Japan (kado@med.yokohama-cu.ac.jp).

Submitted for Publication: September 11, 2012; final revision received November 19, 2012; accepted November 27, 2012.