Development of a Fiber-Optic Optical Coherence Tomography Probe for Intraocular Use

Abstract

Purpose: To evaluate the performance of a newly developed 23-G optical coherence tomography (OCT) probe in animal and human eyes.

Methods: The probe is a side-imaging OCT device with a scanning beam set 43° to the optical axis and a working distance of 1.5 to 2.0 mm. The performance of the OCT probe was tested during vitrectomy in porcine cadaver eyes and rabbit eyes in situ. Optical coherence tomography images of a normal retina, retinal break, optic disc, pars plicata of the ciliary body, and intraoperative surgical manipulations were recorded. The probe was also tested in a pilot study of clinical cases; intraoperative real-time OCT imaging was performed in three patients, including a 56-year-old woman with an epiretinal membrane.

Results: The OCT probe was able to delineate intraocular tissues, including the posterior retina, and even the most peripheral pars plicata in animal eyes. The OCT probe also successfully delineated intraoperative surgical maneuvers such as membrane peeling and the minute structures of the vortex veins, ora serrata, and vitreous incarceration in the scleral incision from the trocar with sufficient resolution in the patients. There were no complications resulting from its use.

Conclusions: The ability of this new 23-G OCT probe to obtain images of intraoperative manipulations from the most peripheral tissues in animal and patient eyes suggests that it could enable surgeons to make better decisions during the course of intraocular surgery.

Optical coherence tomography (OCT)¹ is an essential tool in the diagnosis and management of vitreoretinal diseases. The resolution of OCT instruments has greatly improved, and swept-source OCT (SS-OCT) has enabled clinicians to obtain images of the choroid and sclera. Vitreoretinal surgery has also improved with improved surgical equipment and operating microscopes. These advances have resulted in an increased number of vitreoretinal surgeries. Despite these advances, surgeons occasionally need to make decisions during the course of intraoperative surgery. An intraocular OCT probe could help determine the optimal surgical course. 

Currently, three types of intraoperative OCT instruments are used in vitreoretinal surgery: a hand-held,^2–9 microscope-integrated,^10–14 and probe-type OCTs.^15–18 The hand-held OCT is a miniaturized version of the desktop OCT, wherein it is held by the surgeon or attached to the ophthalmic microscope with a mounting system that stabilizes the device and allows for rapid and reproducible scanning. Hand-held OCTs are primarily applications of desktop OCTs, which mainly scan the posterior retina. However, their disadvantages include inferior accuracy in scanning the exact target site and difficulty in scanning the peripheral retina. 

The microscope-integrated OCT^10–14 system is placed between the objective lens and imaging optics of the microscope, and the optical path of the OCT is incorporated into the surgical microscope path. Real-time images taken by the integrated OCT through the RESCAN 700¹⁴ microscope (Carl Zeiss Meditec, Jena, Germany) have high resolution and can provide the surgeon with useful information regarding retinal configuration, mainly of the macula and posterior retina. Further, the intraoperative OCT image is displayed within the surgeon's view inside the microscope and does not require changing views during surgical procedures, such as looking at a monitor that is separate from the microscope. However, difficulty in scanning the most peripheral part of the retina is a limitation of this type of OCT. 

A probe-type OCT was developed and tested mainly in animals.^15–18 It is intraocularly inserted, and the scanning light is projected onto the targeted tissue anywhere within the eye. The probe must be small enough to be inserted through a trocar used for microincision vitrectomy (viz. 23- to 27-G trocars). 

We developed a 23-G intraoperative probe-type OCT that can be inserted through a 23-G trocar. This study aimed to determine this probe's ability to record high-resolution tomographic images of intraocular structures during surgery. 

Methods

OCT Probe Design and System Specifications

A diagram of the OCT system and probe design are shown in Figure 1 (NIDEK Co. Ltd., Aichi, Japan), and images taken by the OCT probe are shown in Figure 2. The OCT probe system consists of light sources and a balanced photodetector (BPD; Axsun Technologies Inc., Billerica, MA, USA). The basic unit is a SS-OCT that uses a wavelength-swept light source. The light emitted from the source is split by a fiber coupler into sampling and reference arms (Fig. 1a). The swept-source scanning light (980–1060 nm) is transmitted through a circulator and combined with an aiming light (532 nm) at a wavelength-division multiplexer in the sample arm, then projected from the probe onto the target (Fig. 1b). The reference light is reflected by a mirror; the scanning light is reflected by the target; these two light beams are detected by a BPD (Fig. 1a). Interference signals obtained from the interaction of the two beams are processed by a computer, and tomographic images of the target are created. The digitizer sampling rate is 500 Mega samples per second (12 bit). Each image is recorded in a polar coordinate system and transformed into a rectangular coordinate system. 

Figure 1

 

Diagram of the OCT probe system. (a) Schematic drawing of the complete system. (b) Configuration of the tip of the 23-G probe. ODL, optical delay line; PC, polarization controller; BPD, balanced photo detector; DAQ, data acquisition; M, motor unit; C, fiber coupler; A, attenuator; WDM, wavelength-division multiplexer.

Figure 2

 

Photographs of the OCT instrument and fiber-optic probe. (a) Main body of the OCT instrument. (b) Two-meter long fiber cable. (c) Hand piece of the 23-G OCT probe (bottom) and the 23-G vitreous cutter probe (top). (d) Aiming beam projected from the tip of the OCT probe. (e) Photograph of the side monitor. A small monitor (arrow) is attached to the side of the ocular lens of the operating microscope and provides the surgeon with real-time images.

The motor rotates a gear connected to another gear that rotates the 2-m long fiber inside the cable with a torque coil at 1800 rotations/min (Fig. 2b). The scanning light is transmitted to the handpiece through the fiber and reflected at an angle of 43° by a prism at the probe tip. Thus, the probe is a side-imaging–type probe (Figs. 1b, 2d). The power of the aiming beam is 25 μW, and the aiming beam appears curved on the target (Fig. 2d). This curved shape is due to the angle between the target and projected light, which is 43° to the axis. The probe's working distance is 1.5 to 2.0 mm. The B-scan range on the target is 2.5 mm at 1.5-mm distance from the probe, and the field of view is 110°. 

The central wavelength of the swept-source laser is 1060 nm, and the spectral width is 100 nm. The A-scan rate is 100 kHz, and each B-scan consists of 1024 A-scans with a scanning frequency of 30 Hz. The emitted power of the swept-source laser is 500 μW at the target. The power fluctuates with rotation with a 3-dB range. The theoretical axial resolution is 3.65 μm, and the transverse resolution is 80 μm in water. The measuring spot size is 80 μm at 1.5-mm distance from the lens. Image size is 1024 (H) × 688 (V) pixels, and imaging speed is 30 frames/s. 

Characteristics of the Obtained Images

Artifacts appear as transverse lines at the top and bottom of the images. These are produced by the optical reflection at the interface between a lens and junction of the optical fiber. One or both artifacts in the images were cropped to magnify the images. 

The scaling of the obtained images has an unusual characteristic specific to the imaging system. The vertical scaling is constant and can be theoretically calculated using the wavelength. However, the horizontal scaling is difficult to calculate because it is a variable scale ratio based on the distance between the probe and target. Thus, the images in this article have only vertical scale bars. 

Side Monitor Screen

A small monitor screen (2.1 [h] × 2.4 [w] cm; Fig. 2e) is attached to the outside of the ocular lens of the operating microscope on which real-time images of the OCT scan are displayed. This monitor is a modification of a head-mounted display (AiRScouter WD-100G/100A; Brother Industries, Ltd., Nagoya, Japan) and is connected to a computer. The resolution of the monitor is 800 × 600 pixels, with a super video graphics array. Surgeons need only move their eyes approximately 30° horizontally to look at the monitor. 

OCT Probe Images of a Roll of Cellophane Tape

Images of a roll of cellophane tape under various conditions such as a partial defect, split, cleft, and full-thickness tear were taken to demonstrate the ability of the SS-OCT probe to delineate these changes. 

OCT Imaging of Isolated Porcine Eyes

Isolated porcine eyes were used to determine whether the OCT probe was able to obtain quality images. After 23-G trocars were placed 3-mm posterior to the limbus and a core vitrectomy was performed, the 23-G OCT probe was inserted. The OCT probe tip was placed 1.5 to 2.0 mm from the retina, and the aiming beam was directed at a normal retina, retinal break, retinal vessels, optic disc, or the pars plicata of ciliary body. 

OCT Images of Rabbit Eyes

Animal experiments were performed as per the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. The experimental protocol was approved by the Institutional Animal Care and Use Committee of Nagoya University School of Medicine. 

Two Dutch rabbits weighing approximately 2.5 kg were anesthetized with an intramuscular injection of ketamine hydrochloride (25 mg/kg loading dose, then 10 mg/kg per hour) and 2 mg/kg xylazine. The pupil was dilated with a combination of 0.5% phenylephrine hydrochloride and 0.5% tropicamide. Operative procedures were performed similarly to that for the porcine eyes. Optical coherence tomography images of several intraocular tissues were obtained. Images were also taken of various surgical procedures, including touching the retina with a diamond-dusted membrane scraper and aspirating the vitreous cortex using a back-flush needle. 

Pilot Study on Patients

The pilot study was conducted at Nagoya University Hospital. This study was registered with the University Hospital Medical Information Network Clinical Trials Registry, and the trial registration number is UMIN000012822. The study was approved by the institutional review board of Nagoya University Hospital; informed consent was obtained from patients after the nature of the study and possible complications were explained. The study adhered to the tenets of the Declaration of Helsinki. 

A 56-year-old representative female patient with an epiretinal membrane (ERM) in her left eye and a visual acuity of 16/20 underwent vitrectomy using fiber optic OCT. A four-port, 23-G core vitrectomy with the creation of a posterior vitreous detachment was performed under chandelier illumination. The ERM and internal limiting membrane (ILM) were made more visible by triamcinolone acetonide. The OCT probe was inserted through a 23-G trocar from the left-hand side, and the OCT probe tip was placed close to the ERM during membrane peeling, with the forceps coming from the right-hand side. Due to the use of a chandelier light, the surgeon did not require further intraocular illumination. The aiming beam was projected toward the ERM with a working distance of 1.5 to 2.0 mm. Real-time OCT images of the ERM peeling were recorded. Then, the beam was aimed at a vortex vein, the ora serrata, insertion site of a trocar, a ciliary sulcus, and pars plicata of the ciliary body, and OCT images were obtained. 

Results

OCT Imaging of a Roll of Cellophane Tape

Images of a nine-layered roll of cellophane tape with different deformations are shown in Figure 3. The tape in Figure 3a is partially defective; the cut end was hyperreflective, and there was an acoustic shadow under the cut end. The image of the tape split between the third and fourth layers is shown in Figure 3b. The roll of tape was partially or full-thickness cut (Figs. 3c, 3d). 

Figure 3

 

Optical coherence tomography images of a roll of cellophane tape with various deformations. The various configurations of the tape, for example, partially defective (a), split (b), cleft (c), and with a full-thickness tear (d), can be resolved in the OCT images. The inset is a schema showing how the tape was deformed. Scale bars: 500 μm.

All objects in the image appears to be convex, although the roll of tape is almost flat at a scanning distance of 1.5 mm. Because the distance between the target and probe tip differs at different points, images are transformed into a rectangular coordinate system from a polar coordinate system, which makes the images dome-shaped. 

OCT Imaging of Animal Ocular Tissue

Retinal tissues could be clearly distinguished from the choroid in the isolated porcine and in situ rabbit eyes (Figs. 4a, 4b, respectively). The retina appears as a two-layered tissue in porcine eyes (Fig. 4a). An intentional retinal tear could be seen as a full-thickness retinal defect (Fig. 4d) and micro retinal tear (Fig. 4c, white arrow) with retinal detachment in the porcine eye. A retinal break (Fig. 4d, white arrow) with enhanced reflectivity from the choroid beneath the break (Fig. 4d, yellow arrow) can be seen. Cross-sectional images of retinal vessels can be seen as prominent structures with acoustic shadows (Figs. 4e, 4f). Cross-sectional OCT images of the optic disc of a porcine eye (Fig. 4g) and rabbit eye (Fig. 4h) show a depression in the optic disc (*). The surface of the vitreous cortex is clearly delineated as a single line by the triamcinolone acetonide particles (Fig. 4h, white arrow). Images of the pars plicata of the ciliary body can be seen in Figures 4i and 4j and Supplementary Video S1. The pars plicata near the zonule of Zinn is seen as a corrugated structure with deep grooves (Fig. 4i) and shallow grooves near the ora serrata (Fig. 4j). A real-time OCT image of a diamond-dusted membrane scraper touching the retinal surface of a rabbit retina can be seen in Figures 5a and 5b and Supplementary Video S2. The instrument tip is coated with diamond powder, which appears as highly reflective objects on the scraper (Fig. 5a, white arrows), and the retinal surface is depressed when the tip touches the retina (Fig. 5b, arrow). An OCT image of a back-flush needle aspirating the vitreous cortex of a rabbit can be seen in Figures 5c and 5d and Supplementary Video S3. The retina is detached, and the vitreous cortex is coated with triamcinolone acetonide (Figs. 5c, 5d, white arrow). The needle tip has a silicone tube that appears as a hollow tube (Figs. 5c, 5d, yellow arrow). The vitreous cortex is aspirated up by the silicone tube, and the detached retina can be seen being dragged toward the tube. 

Figure 4

 

Optical coherence tomography images of retinal tissue at the posterior pole (left column, isolated porcine eye; right column, rabbit eye). (a, b) Normal retina. Retinal tissue can be clearly distinguished from the choroid. (c, d) Retinal tear (white arrow). The subretinal space is also clearly discernible. Even a tiny retinal tear can be detected (c). Enhanced reflectivity in the choroid beneath the tear (yellow arrow) can be seen, because there is no absorption by the retinal tissue. (e, f) Cross-sectional image of the retinal vessels (arrow). Each vessel has an acoustic shadow beneath it. The inset (f) is a schema showing how the OCT probe scanned the tissue. (g, h) Optical coherence tomography images of the optic disc. The optic disc can be clearly seen in an isolated porcine eye (g) and a rabbit eye (h). The visibility of the vitreous cortex is increased by triamcinolone acetonide (h, white arrow). Inset (h): a schema showing how the OCT probe scanned the tissue. Asterisks represent the optic disc. (i, j) Optical coherence tomography images of the pars plicata of the ciliary body in an isolated porcine eye (Supplementary Video S1). The pars plicata near the zonule of Zinn is corrugated with deep grooves (i) and with shallow grooves near the ora serrata (j). Ret, retina; Ch, choroid. Scale bars: 500 μm.

Figure 5

 

Real-time OCT image of a diamond-dusted membrane scraper touching the rabbit retina (a, b; and Supplementary Video S2) and of a back-flush needle aspirating the vitreous cortex in a rabbit eye (c, d; and Supplementary Video S3). Scale bars: 500 μm. The OCT images clearly show how the tip of the diamond-dusted membrane scraper touches (a) and compresses (b) the retina. The tip of the scraper is coated with diamond powder, which appears as highly reflective objects (a; white arrow). Optical coherence tomography can show a real-time image when the back-flush needle (c, d; yellow arrow) aspirates the vitreous cortex (c, d; white arrow) on a detached retina. The vitreous cortex is sucked into the tube of the back-flush needle, and the detached retina is dragged toward the needle (d). The quality of the retinal image beneath the back-flush needle is low because the scanning light is obstructed by the needle (c).

OCT Images From Pilot Clinical Case

Preoperative OCT images (Fig. 6a) from a patient showed thickened ERM and fovea. Optical coherence tomography images show a partially peeled ERM with the ILM (Figs. 6c, 6d, white arrow; Supplementary Video S4). A vortex vein with scleral indentation (Fig. 7b, white arrow) shows a structure with tributaries (Fig. 7a, yellow arrow; Supplementary Video S5), and an OCT scan taken perpendicular to the ora serrata (Figs. 7c, 7d) demonstrates cystoid changes (Fig. 7c, yellow arrow) in the tissue (Supplementary Video S6). An OCT scan wherein the trocar was removed (Figs. 7e, 7f, white arrows) shows that the incised wound was slanted (Fig. 7e, white arrow) because the trocar blade was inserted at an angle oblique to the sclera for self-sealing. The OCT image demonstrates vitreous incarceration in the scleral incision (Fig. 7e, yellow arrow; Supplementary Video S7). The OCT probe was inserted through a side port to the ciliary sulcus (Figs. 7g, 7h), and an OCT image was obtained showing the spatial relationship with the iris (Fig. 7g, white arrow), ciliary body (Fig. 7g, yellow arrow), and intraocular lens (Fig. 7g, open arrow; Supplementary Video S8). A transverse section of the pars plicata of the ciliary body showed a corrugated structure with deep grooves (Fig. 7i; Supplementary Video S9), and a longitudinal section revealed its relationship with the haptics of the intraocular lens (Fig. 7j, yellow arrow) and ciliary body (Fig. 7j, white arrow; Supplementary Video S10). 

Figure 6

 

Clinical data from a pilot case of a 56-year-old patient with an ERM and a visual acuity of 16/20. (a) Desktop OCT image shows an ERM and macular edema. (b) Intraoperative microscope image with the OCT probe aimed at the ERM and internal limiting membrane (ILM) under chandelier illumination. (c, d) Optical coherence tomography images showing a half-peeled ERM with the ILM. Particles of triamcinolone acetonide are seen on the retina and the ERM (Supplementary Video S4). Scale bars: 500 μm.

Figure 7

 

Clinical data of a pilot case with an ERM. (a) Optical coherence tomography image of a vortex vein shows a structure with tributaries (yellow arrow) (Supplementary Video S5). (b) Intraoperative microscope image of a vortex vein. The sclera was indented from the outside to obtain this image. (c) An image from a perpendicular scan of the ora serrata showing cystoid changes (yellow arrow) (Supplementary Video S6). (d) Intraoperative microscope image with the tip of the probe aimed at the ora serrata (white arrow). (e) An OCT scan of the insertion site of a trocar after its removal, showing vitreous incarceration (yellow arrow) into the slanted incision of the sclera (white arrow) (Supplementary Video S7). (f) Intraoperative image of the incision site where the trocar was removed (white arrow). (g) An OCT scan of the ciliary sulcus (▵) showing the spatial relationships among the iris (white arrow), ciliary body (yellow arrow), and intraocular lens (open arrow) (Supplementary Video S8). (h) The probe was inserted through a side port, and the image of the ciliary sulcus was obtained. (k) A transverse section of the pars plicata of the ciliary body shows a corrugated structure with deep grooves (Supplementary Video S9). (j) A longitudinal section of the pars plicata of the ciliary body clearly shows the spatial relationship between the haptics of the intraocular lens (yellow arrow) and ciliary body (white arrow) (Supplementary Video S10). Post, posterior; Ant, anterior. Scale bars: 500 μm.

There were no complications such as retinal damage or postoperative infections resulting from the intraoperative manipulations using the fiber optic probe. 

Discussion

A 23-G, fiber optic, side-imaging OCT probe was developed and its performance was evaluated in isolated porcine eyes, in situ rabbit eyes, and clinical human cases. The image quality obtained using this OCT system was sufficient to detect minute structures of the retina, choroid, optic disc, and ciliary body in all three types of eyes. All procedures were performed without any complications. 

The usefulness of intraoperative OCT has been reported.^2–15 For instance, in eyes with vitreous opacities that obscure the fundus, the macular configuration can be intraoperatively detected using a fiber optic OCT probe after opacities in the vitreous are removed. This allows the surgeon to judge whether ERM or ILM removal is necessary (e.g., in eyes with a macular pseudohole or edema). Furthermore, triamcinolone acetonide or dyes such as indocyanine green or brilliant blue G can be used to increase the visibility of the ILM to ensure that it has been completely peeled from the appropriate areas. Intraoperative OCT should provide surgeons with another option evaluating the completion of the removal. In eyes with retinal detachment, the success of the surgery relies on the preoperative and intraoperative detection of all retinal breaks and their closure. However, detecting these is sometimes difficult. Recent advances in intraoperative observation with a wide-viewing system and chandelier lighting made it easier to detect peripheral retinal breaks; however, biopsy by OCT is more sensitive for detecting them than en face surgical views. Fiber optic OCT probes have the advantage of allowing observation of the peripheral retina and ciliary body. 

Fiber optic OCT probes can be classified into those with side-imaging modes and with forward-imaging modes.¹⁹ The side-imaging OCT is a system with a scanning beam projected perpendicular to the probe axis. These probes are suitable for tissues within tubular organs, for example, the gastrointestinal tract,^20–24 vascular system,^25–29 respiratory system,^21,30–33 and urinary tract,³⁴ because the perpendicular angle of the scanning laser can make more efficient use of such a limited space and can obtain images over 360°. Further, three-dimensional imaging^23,31 can also be obtained because circumferential scanning allows continuous images of tubular organs. Forward-imaging OCT systems have a scanning laser that projects straight from the probe tip. These systems are useful for OCT-guided biopsy because the target tissue is on the same axis as the probe, and the depth can be recorded before the biopsy. Our probe is a side-imaging–type probe with the scanning laser at 43° angle to the axis. Its advantage is that the optic fiber rotation allows the laser to be directed anywhere within the eye such as the posterior retina, peripheral retina near the trocar, ciliary body, and tissues in the anterior chamber. 

Reducing the probe size is critical because small-incision vitrectomy has become standard. A 30-G (0.31 mm) OCT probe was developed for bronchopulmonary³³ and skeletal muscle surgeries.³⁵ Thirty-gauge probes are smaller than those used in small-gauge vitrectomy systems; however, they may be dangerous for ophthalmic use because they are needle-shaped. For ophthalmic use, a 21-G (0.82 mm) probe for retinal observation¹⁵ and a 1.6-mm probe for Schlemm's canal surgery¹⁶ were developed. These diameters are too large for small-gauge vitrectomy. More recently, a 25-G OCT probe for corneal and retinal imaging with an axial resolution of 19 μm and a 1310-nm central wavelength laser was reported.¹⁷ The axial resolution was lower than that of our probe (3.65 μm), but sufficient to detect rough configurations. Further, a forward-imaging 25-G OCT with an excellent axial resolution of 4 to 6 μm and lateral resolution of 25 to 35 μm was reported¹⁸; however, the central wavelength of the scan beam was 870 nm, which may not allow deeper structure imaging. Our OCT probe has axial and lateral resolutions of 3.65 and 80 μm, respectively, and the central wavelength of the swept-source laser is 1060 nm. The poor lateral resolution could be due to several reasons, including a reduction in the performance of the optical devices because of downsizing the gauge size relative to the laser beam size, low-contrast images caused by a loss of light in the optical system, and motion artifacts caused by the coil rotation. The image quality may not be sufficient for a precise analysis of retinal thickness. However, it should give surgeons adequate information to make a decision on how to proceed with intraoperative procedures. On the other hand, one advantage of this probe is that the swept-source laser's long wavelength (1060 nm) enables surgeons to obtain detailed information on deeper retinal and choroidal tissues. 

The disadvantage of this system is that surgeons need to move their eyes approximately 30° horizontally without moving their heads when looking at the side screen. Thus, they lose their microscope view during observation. The probe's working distance is 1.5 to 2.0 mm, and the surgeons need to keep a safe distance between the target tissue and probe tip by holding the probe with their hands to avoid hitting the tissue. However, we believe it is not difficult to maintain the correct distance if they alternate between observing the screen and microscope view. If the side monitor is outside the surgical field, the surgeon needs to raise his or her head to look at the image. This movement may make it more difficult to maintain the correct distance because they cannot help losing their attention and cannot avoid moving the probe; consequently, the probe could potentially touch the target tissue. The most desirable design may be that of RESCAN 700, a microscope-integrated OCT because the real-time OCT image is displayed within the same view as the surgical view in the microscope, thus avoiding the surgeons' loss of their microscope view. However, it is very technically difficult to incorporate the real-time image of the fiber-optic OCT probe into the surgical view. Thus, we believe that the side monitor screen in our system is the best way to display the real-time image. To reduce the risk of hitting the tissue with the probe, the probe's working distance should be increased. 

The probe's working distance is 1.5 to 2.0 mm, and the focus (i.e., “spot size”) can vary depending on this distance. When the working distance is shorter or longer than 1.5 mm, the spot size gradually increases, and the recovery rate of the returning light decreases, leading to lower quality images. Because the side-viewing probe scans a circumferential trajectory, images at the edges of the field of view are out of focus and have poor resolution due to the longer working distance compared with those at the center of the field of view. For instance, the image quality shown in Figure 5 is lower at the edges. However, careful interpretation is necessary because the image quality also decreases when an instrument blocks the scanning light, as seen in Figure 5c. The surgeon was able to keep the retina in the probe's focal plane; however, the focus and image quality deteriorated when the hand holding the probe moved. This is a limitation of this probe. 

Another limitation of this study is that many specifications of the OCT probe and system are confidential because of the nondisclosure agreement between NIDEK Co. Ltd. and the suppliers, including specifications of the balanced detector, torque coil, gradient-index optics, lens protection material, outer sheath, and optical properties of the OCT. 

In conclusion, we successfully developed a fiber optic, 23-G, side-imaging SS-OCT probe that can be inserted through a 23-G trocar. This probe can provide high-quality images of targeted ocular tissues, such as retinal tears, the ILM, retinal detachments, and ciliary body, in animals and humans. It should enable surgeons to determine the configurations of diseased structures and select suitable treatment procedures. 

Acknowledgments

The authors thank Yasuhiro Higashijima (NU SYSTEM, Inc., Nagoya, Aichi, Japan) and Santec Corporation (Komaki, Aichi, Japan) for their technical support in the development of fiber optic OCT. The authors also thank Duco Hamasaki of the Bascom Palmer Eye Institute of the University of Miami for his critical discussion and final manuscript revisions. 

Supported by grants from the Program to support development of medical equipment and devices to solve unmet medical needs (Tokyo, Japan); JSPS KAKENHI, Grant No. 23390401 and 25462710 (Tokyo, Japan); and the Translational Research Network Program supported by the Japan Agency for Medical Research and Development, AMED, Grant No. 15lm0103009j004 (Tokyo, Japan). 

This study was registered with the University Hospital Medical Information Network Clinical Trials Registry (UMIN-CTR); the trial registration number is UMIN000012822. 

Disclosure: T. Asami, NIDEK Co. Ltd. (R); H. Terasaki, NIDEK Co. Ltd. (F, R), P; Y. Ito, NIDEK Co. Ltd. (R); T. Sugita, None; H. Kaneko, None; J. Nishiyama, None; H. Namiki, None; M. Kobayashi, None; N. Nishizawa, None