Robotic Eye Surgery: Past, Present, and Future

Robotic surgical systems have had an increasing presence in the surgical landscape over the past two decades. Although the most wellknown platform is the Da Vinci Surgical System (Intuitive Surgical, Sunnyvale, CA), there are hundreds of different devices in use for abdominal, urologic, pelvic, cardiovascular, and neurologic surgery [1]. Robot-assisted methods offer many advantages over conventional surgical techniques such as improved precision, reduced tremor, and amplified scale of motion (Table 1). It is no surprise that we are in the midst of a paradigm shift towards the integration of robotic augmentation of conventional surgical methods.


Introduction
Robotic surgical systems have had an increasing presence in the surgical landscape over the past two decades. Although the most wellknown platform is the Da Vinci Surgical System (Intuitive Surgical, Sunnyvale, CA), there are hundreds of different devices in use for abdominal, urologic, pelvic, cardiovascular, and neurologic surgery [1]. Robot-assisted methods offer many advantages over conventional surgical techniques such as improved precision, reduced tremor, and amplified scale of motion (Table 1). It is no surprise that we are in the midst of a paradigm shift towards the integration of robotic augmentation of conventional surgical methods.
Ophthalmic surgeries are commonly performed with a highmagnification, three-dimensional view through a surgical microscope. Precise manipulations must be performed using delicate handheld instruments to minimize collateral damage that might result in a poor visual outcome for the patient (Figure 1a). Robotic surgery is particularly attractive in these cases, because of the advantages listed above. Complex feedback loops (Figure 1b) present numerous challenges to investigators who have attempted the adaptation or development of robotic devices to accomplish intraocular surgical maneuvers. Here we review prior advancements, current innovations, and future directions of robotic eye surgery.

Previous Advancements in Robot-Assisted Ophthalmic Surgery
Historically, most innovations in the field have focused on either accomplishing single tasks or assisting in technically difficult portions of procedures. One of the earliest examples was a micro-manipulator developed by investigators in France in the late 1980's. This device was one of the first to attempt maintenance of a remote center of motion (RCM), a fundamental necessity in intraocular surgery to avoid tissue damage due to translational forces at points of entry. The Stereotaxical Micro-telemanipulator for Ocular Surgery (S.M.O.S) allowed for 4 degrees of freedom (DoF) including rotation and translation about the RCM. A carrier allowed for 3-dimensional movement during surgery. Precise movements could be completed at the expense of greatly increased surgical procedure time [2]. Over the next decade, several other groups described prototypes with the ability to perform dedicated microsurgical tasks in animal models, including cannulation of retinal vessels [3], drainage device implantation [4] intravascular drug delivery, and microvascular pressure measurement [5].
Japanese collaborators created a prototype robotic system based on the S.M.O.S. platform that is designed to aid in multiple steps of vitreoretinal surgery. Robotic assistance increased accuracy five to ten fold, independent of the surgeon. The system eliminated the interoperator variability in precision that is seen in similar tasks performed manually. These modifications facilitated successful surgical induction of a posterior vitreous detachment, retinal vessel sheathotomy using 25-gauge microscissors, and microcannulation of retinal vessels with a diameter of 100 microns in porcine eyes [6].
Investigators at Johns Hopkins University developed a steadyhand manipulator (SHM) for retinal microsurgery [7]. This device consists of an arm with tilt and roll mechanisms on an xyz-stage that is attached to a force/torque sensor [6] allowing the instrument to move at the direction of the surgeon with software-augmented proportional velocity. The design places the RCM at the sclera, minimizing undesirable tension on the eye wall. The SHM provides filtration of tremor that was demonstrated experimentally. The operators, with an average of 182 microns of tremor, were successful in cannulating80 micronchorioallanotoic veins in chicken embryos. Further innovations by this group include intra-operative retina registration that syncs with pre-operative imaging to guide treatment delivery [8]. Equally as impressive is the Micron, a microsurgical tool that reduces unintentional tremor while preserving eye-hand coordination. Surgeons experienced up to 52% reduction in error in three simple microscopic positioning tests [9].
For even more dexterous manipulations, a group from Columbia University recently proposed a theoretical multi-arm hybrid robotic system [10] and a 16-DoF system utilizing surgical tools inside of the eye itself [11]. Other investigators have also proposed intraocular robots as a minimally invasive mechanism to provide a high degree of mobility and precision in the placement of intravitreal inserts for treatment of exudative age related macular degeneration. Using micro robots with an outer diameter of less than 500 µm, the authors were able to achieve targeted placement of drug reservoirs in porcine eyes using wireless electromagnetic controls [12].

The Present Dilemma: Complete Procedures
In the past, ophthalmic robotic surgical systems have focused on single tasks, often those that are particularly delicate. Although much of this research is ongoing, interest is growing in designing devices that are able to perform the entire surgical procedure, with the goal of adding speed and efficiency without sacrificing precision. Requirements for this goal include sufficient range of motion, simultaneous surgical instrument manipulation, and capability of mid-surgery instrument switching.
The Da Vinci Surgical System, approved by the Food and Drug Administration in 2000, increased the number of general robotic surgical procedures by a factor of 15 in its first four years of use [13]. The system consists of two components: the mechanical robot apparatus with three or four arms with a dual channel endoscope ( Figure 2a) and a control console where the surgeon manipulates the robotic arms remotely while looking through a stereoscopic viewfinder (Figure 2b). Each robotic arm can tilt in two planes and pivot around a stable point of rotation.
In ophthalmic surgery, this system has been used to perform suture repair of a corneal laceration [14], complete a continuous capsulorhexis on the anterior lens capsule in cataract surgery, and perform a 3-port 25-gauge pars plan a vitrectomy (Figure 2c) in porcine eyes [15]. Several limitations of the Da Vinci Surgical System were noted. The robotic arms did not mirror the exact movements of human arms, preventing a perfectly round, curvilinear capsulorhexis that would be optimal for cataract surgery. In addition, the 5 cm distance between the RCM and the instrument tip (site of ocular penetration) limited motion and created undue tension on the external eye surface. The endoscope lacks retroillumination capabilities and its position inhibited peripheral vitreous gel removal.
To overcome these challenges, investigators have customized microsurgical systems for ophthalmic surgeries. By mounting a micro robot, the Hexapod Surgical System (HSS), to the Da Vinci macro robot, a remote center of motion at the site of ocular penetration can be achieved (Figure 3). The precision and dexterity of this approach was validated by successful insertion of a vitreous cutter through a sclerotomy in porcine eyes [16]. Another adaptation, known as the "Micro hand," was equipped with micro electromechanical systems (MEMS) technology. This device was designed to mimic a human hand and is pneumatically controlled, allowing titration of grasping force (Figure 4a). Four fingers, with a length of 4 mm each (Figure 4b), were  to maneuver caliper weights and manipulate fresh retinal tissue of porcine cadaver eyes at 60 psi of applied compressed air [17].
The Intraocular Robotic Interventional Surgical System (IRISS), a joint effort between the Jules Stein Eye Institute and the UCLA Department of Mechanical and Aerospace Engineering, is a dedicated microsurgical platform capable of performing complete ophthalmic procedures. The master-slave design features a remote console, similar to the Da Vinci and Eye RHAS (Eye Robot for Haptically Assisted Surgery) systems [18], which could facilitate telesurgery. The IRISS design includes a head-mounted "True Vision" (True Vision Displays, Inc., Cerritos, CA) stereoscopic visualization system, two joystick controls with tremor filtration and scaled motion, custom designed arms appropriately sized to accommodate commercially available instrumentation, and two closely approximated remote centers of motion to avoid stress on surrounding tissues ( Figure 5). In ongoing early trials, validation in porcine eyes has focused on three complex ocular procedures: lens capsulorhexis in cataract surgery, 23-gauge core vitrectomy, and retinal vein microcannulation.

The Future of Robot-assisted Eye Surgery: Automation and Integration
In the past several years, advances in ophthalmologic imaging modalities such as optical coherence tomography (OCT) and ultrasound biomicroscopy (UBM) have enhanced the surgeon's ability to localize pathology both pre-and intra-operatively [19]. Simultaneously, femtosecond laser devices have been optimized for wound construction, capsulorhexis creation, and nucleus breakdown during cataract surgery. Integration of a robotic system with OCT has been proposed for vitreoretinal surgery [20], and the addition of laser technology would facilitate automation of surgery to treat cataract, which is a leading cause of blindness worldwide. Laser refractive surgery, such as laser-assisted in situ keratomileusis (LASIK), has already become nearly completely automated and requires minimal intra-operative manipulation by the surgeon. For less standardized procedures, such as vitreoretinal surgery, robotic augmentation could ultimately increase efficiency, amplify scale to allow performance of otherwise difficult tasks (e.g. sub retinal delivery of medication or stem cells), decrease complication rates by reducing tremor and increasing precision, and permit telesurgical care in remote locations.
Several obstacles remain before robotic surgery will become clinical reality in ophthalmology. High cost, steep learning curve, and patient trust all present individual challenges. These hurdles are similar to those initially encountered by the proponents of minimally invasive laparascopic surgery, which is now a widely accepted technique [21]. Over two decades of evidence suggests that robotic devices may help facilitate higher quality of care and that research in the field should be continued.