Intraoperative imaging of ocular tissues for diagnostic and therapeutic applications has gained immense admiration in recent years.
The real time cross-sectional imaging, as well as three and four dimensional reconstruction abilities of intraoperative optical coherence tomography (iOCT), has enhanced our knowledge on many fronts in surgical maneuvers.
Using iOCT intraoperative valuation of specific anatomical, morphological, and depth detailed events in real time is possible in anterior segment, posterior segment, extraocular, and ocular surface surgeries.
Therefore, by incorporating this crucial information, our routine surgical exercises and patient outcomes can be enhanced in a comprehensive way.
Over the past 20 years, OCT has evolved to become an integral component of ophthalmic diagnosis, management, and therapeutic monitoring, with an estimated tens of millions of images captured per year.
During this time, improvements in OCT technology have enabled the transition from an outpatient imaging modality to the operating room.
Integration of OCT into the operating room was initially limited by the bulky equipment and design of clinical systems that required patients to be cooperative and positioned upright.
This hurdle was overcome through the development of the handheld OCT, introduced to the surgical suite in 2009, which allowed imaging of patients who were supine and under anesthesia.
This system was further updated to a microscope-mounted OCT configuration, which served to ameliorate previous critical limitations in scan reproducibility and precision.
The use of these microscope-mounted external intraoperative OCT (iOCT) systems enabled extensive study and clinical evaluation of iOCT during vitreoretinal surgery, including study of the potential positive impact on surgical decision-making, with key findings from the PIONEER study highlighting the potential role in confirming achievement of surgical objectives and guiding decisions regarding need for tamponade choice or restaining.
However, it was not until the advent of microscopeintegrated OCT systems that widespread utilization became feasible.
Following their initial development, the next major milestone was the FDA clearance of the first microscope-integrated system, which provided broad availability of efficient, real-time visualization of surgical manipulations and the cross-sectional surgical anatomy.
Current commercially available iOCT options include the Leica Envisu handheld system and Enfocus microscopeintegrated system, the Zeiss Rescan 700, and the HaagStreit Surgical iOCT system. This review will focus on recent advancements in iOCT technology and its current utility in retinal surgery.
Novel Visualization of the Surgical Field and OCT Datastream
There has been significant advancement from the conventional microscope-integrated systems that first made the commercialization and adaptation of iOCT to a wider audience possible.
Although a major innovative improvement, limitations to image quality and contrast with current systems have often resulted in the use of external monitors for OCT review.
Emerging technological evolution in surgical microscopes and the surgical theater has brought new digitally enabled options for high-definition, 3D stereoscopic visualization during posterior-segment surgery.
The integration of iOCT with these digitally enabled viewing systems frees the surgeon from the confines of the microscope oculars and allows for simultaneous viewing of the surgical field and OCT datastream on a heads-up display.
These microscope systems, combined with next-generation OCT engines, have even allowed for real-time 3D OCT visualization, which has been termed 4D MIOCT. With certain systems, datafusion software allows for additional overlay of surgeon-chosen functions onto the 4K display screen.
This integration allows real-time surgical parameters (eg, flow rates, infusion pressure, laser power) to be visualized on the active surgical image without blocking the surgeon’s view of the field.
Surgical efficiency, ergonomics, patient safety, and new surgeon education in vitreoretinal surgery may be enhanced with this technology, but the literature in this area remains in its infancy.
Currently, two 3D digital surgical visualization systems with capacity for iOCT integration are commercially available: Ngenuity (Alcon) combined with a microscope-integrated OCT system (Rescan 700; Zeiss) and the Artevo 800 (Zeiss) combined with the Rescan 700.
One of the largest prospective studies to date comparing a digitally enabled iOCT surgical platform (Ngenuity with Rescan 700) with a conventional microscope-integrated iOCT platform (Rescan 700) demonstrated several advantages of the digital display system.
Simultaneous surgical field-based visualization iOCT datastream was significantly higher in the digitally enabled cases compared to conventional. Importantly, the feasibility and utility of the heads-up, digitally enabled iOCT platform was similar to the conventional iOCT system.
Similar results were also found when evaluating a newer digitally enabled iOCT system (Artevo 800 with Rescan 700). Whether this provides increased surgical safety by maintaining focus on the surgical field at all times remains to be elucidated, and future studies are under way.
Potential benefits of digital systems extend beyond intraoperative visualization. The ergonomic advantage of a 3D display system has been demonstrated in a few studies, with noted reduction in fatigue, back pain, and headaches.
Another potential benefit is the amount of light necessary for a given case. In a recent study, surgeons who performed vitreoretinal surgical intervention with the digitally enabled platforms needed significantly less endoillumination throughout the case.
Theoretical benefits to this exist, but future studies are needed to determine if there is translatable patient benefit.
Utilization Limitations in iOCT visualization during posteriorsegment surgery exist in part due to a lag in compatible surgical instruments.
Standard metallic instrumentation causes shadowing and light scattering that limits use and image acquisition during real-time surgical maneuvers.
These limitations are likely demonstrated in studies that have shown that most surgeons preferred to pause surgery momentarily to utilize iOCT in a static fashion with only 37% utilizing iOCT in real-time during intrasurgical maneuvers.
Instruments with semitransparent materials reduce light-scattering and improve viewing of the instrument tips and tissue. The development of iOCT-compatible instrument tips may provide greater utility during surgical manipulation of posterior-segment structures.
In cadaver eyes, these instruments, which included forceps, membrane scrapers, and a surgical pick, demonstrated better iOCT visualization. Another limitation in real-time surgical application of iOCT is the fast frame rate, high sampling density, and image stabilization required to produce the highest quality image.
To achieve this, current technology necessitates a small field of view that may hinder the performance of certain surgical maneuvers. Instrument tracking may solve this issue by automatically recentering the OCT scan in real time on the tip of the instrument.
Current iterations of microscope-integrated iOCT platforms have a reference arm tracking that improves image stabilization. Research into improved methods for tracking is ongoing and includes using microscopic tip tracking and prediction via machinelearning algorithms and instrument handle tracking using stereo cameras.
Another highly impactful area of research that may propel the flexibility and utility of this technology forward is the integration of high-speed systems that provide volumetric real-time OCT capability.
Currently, spectral-domain systems are limited to B-scans of variable orientation, which does not allow for live volumetric imaging. The development of new graphics-processing technologies and microscope integrated high-speed swept-source iOCT systems can produce live volumetric 4D imaging in real time.
Advanced image analysis platforms are greatly lacking in all commercial iOCT systems. New work in retinal segmentation and pathology feature extraction, including volumetric analysis of fluid and macular holes, provides new opportunities for understanding surgical pathology and the impact of surgical maneuvers.
In combination with next-generation precision medicine, this technology may help guide surgical decision-making (eg, confirming optimal subretinal dose delivery or predicting macular hole closure time).
iOCT volumetric analysis has already been shown to be useful in macular hole surgery and may prove to be critical in newer procedures aimed at delivering therapies into the subretinal space for gene therapy, stem cell therapy, and pharmacotherapy.
Intraoperative imaging has also identified quantitative alterations in EZ-RPE height and subretinal hyporeflectivity following membrane peeling that are associated with foveal normalization speed, possibly suggesting a surrogate between these variables and intrinsic tissue characteristics.
Quantification of these alterations may be translated into prediction models for postoperative behavior.
Membrane Peeling and Epiretinal Membranes
Over the past 20 years, vital dyes such as indocyanine green and brilliant blue G have played a critical role in identifying the internal limiting membrane (ILM) to facilitate membrane peeling in a variety of vitreoretinal disorders.
However, these stains, especially indocyanine green, have phototoxic potential at high concentrations and may increase procedure length.
Intraoperative OCT has been shown to be a feasible alternative to dye, producing comparable clinical outcomes by providing noninvasive, granular visualization of the retinal surface.
Key areas of clinical utility include identifying an area to initiate the peel, recognizing areas of residual membrane that require further focus, and confirming completion of the peel.
The PIONEER trial, a prospective multisurgeon study examining the feasibility of iOCT, demonstrated that intraoperative imaging findings prevented unnecessary surgical maneuvers in 9.2% of cases while identifying residual areas of peel in 12% of epiretinal membrane (ERM) surgeries.
Previous studies have demonstrated that visualization of ERM intraoperatively is possible in both normal and highly myopic eyes. The DISCOVER trial built on this foundation to determine the feasibility of iOCT-guided ERM surgery.
Compared to the conventional treatment option of vitrectomy with ILM peeling, iOCT-guided ERM removal without mandated ILM peeling resulted in no significant difference in surgical or visual outcomes, including reoperations over 12 months.
Intraoperative OCT has also been utilized in outcomes research as a way of identifying potential risk factors for intraretinal architectural changes after ERM surgery.
Leisser et al investigated subfoveal and extrafoveal hyporeflective zones on iOCT during membrane peel, reporting that these findings were not associated with postoperative OCT changes or visual outcomes.
The same group also concluded that intraoperative characteristics were not associated with postoperative intraretinal cystoid changes.
Similar results were noted in the PIONEER study where intraoperative expansion of the subretinal hyporeflective zones was not clearly linked to postoperative OCT changes.
Macroarchitectural intraoperative changes (eg, inner retinal elevation, full-thickness retinal elevations) were noted to have postoperative OCT alterations in only 15% of macroarchitectural alterations cases, which were primarily noted in areas of intraoperative inner retinal elevation.
Most of these transitioned to focal atrophy (64%) compared to inner retinal thickening in 36%.
Macular Hole The DISCOVER trial confirmed that iOCT is a feasible option for closure of macular holes, as demonstrated by a high closure rate corresponding with no intraoperative adverse events. Aside from visualization, iOCT has been used in research to investigate prognostic factors and identify predictive biomarkers for clinical outcomes.
Inoue et al concluded that individuals with residual fragments seen on iOCT had worse postoperative visual improvement at 3 and 6 months than those without residual membrane, whereas Kumar et al identified an iOCT finding termed the “hole-door sign” that correlates with closure of type 1 macular holes, although not with clinical vision findings.
Other research has been directed at identifying iOCT variables to aid in prediction of macular hole closure speed — an outcome that has implications for patient postoperative quality of life because it influences the length of time a patient has to maintain facedown positioning.
This study found that changes in iOCT tissue measurements after ILM peeling may be a proxy for tissue elasticity and mobility that may influence how the retina will approximate. Intraoperative OCT is particularly helpful in this instance due to the difficulty of postoperative transtamponade imaging.
Dynamic iOCT has also been reported to be useful in the inverted internal limiting membrane flap technique used for large macular holes (>400 µm).
The PIONEER study investigated 15 eyes with primary macula-involving retinal detachments (RDs) to explore iOCT characteristics during RD repair and their impact on postoperative anatomic and functional outcomes.
Although all patients were found to have subclinical subretinal fluid during surgery on iOCT following perfluorocarbon tamponade, the fluid spontaneously resolved by the time of postoperative imaging assessment.
In the DISCOVER trial, when specifically looking at outcomes of RD repair and impact on surgical decision making, surgeons claimed that iOCT feedback was valuable in 36% of cases.
Useful information for RD included confirmation of vitreous detachment or retinal reattachment, identification and evaluation of residual fluid post-perfluorocarbon liquid or occult macular hole, necessity for retinectomy, presence of preretinal/ subretinal membranes, and identification of retinal cysts.
Intraoperative OCT data provided greater value in complex repairs than in uncomplicated primary cases. Therapeutic Delivery With the emergence of subretinal gene therapy and artificial vision devices (eg, Argus by Second Sight), the role for iOCT-guided feedback may be even more critical in the future.
Utilizing iOCT, multiple factors that guide successful therapeutic delivery can be ascertained: bleb location, volumetric delivery of therapeutic, and potential sequelae (eg, macular hole).
One of the first described cases of visualization of subretinal therapeutics involved the delivery of subretinal tissue plasminogen activator in a subretinal hemorrhage case that provided excellent visualization of tissue plasminogen activator delivery and bleb expansion.
Intraoperative OCT has been used to determine subretinal injection dosing by volume and can provide essential anatomic information during subretinal maneuvers, potentially working to decrease surgical technique-related complications and aiding in the delivery of stem cell and gene therapies, which require crucial precision.
In addition, iOCT has been reported to aid in visualization of retinal tacking for the electrode array during placement of the Argus II retinal prosthesis for retinitis pigmentosa.
Intraoperative OCT has been an exciting and rapidly advancing area of technology that has meaningful applications in vitreoretinal surgery and patient outcomes.
Promising areas of improvement include virtual reality visualization, robotassisted surgical techniques, automated segmentation software, automated tracking and compatible instrumentation, and overall improvement in image acquisition and quality.
An integrative, digital surgical outfit that integrates seamless utilization of iOCT in real time at a highly reliable level is the ultimate goal, which is closer to realization every day.