Impaired robotic surgical visualization: archaic issues in a modern operating room

While robotic-assisted surgery (RAS) has been revolutionizing surgical procedures, it has various areas needing improvement, specifically in the visualization sector. Suboptimal vision due to lens occlusions has been a topic of concern in laparoscopic surgery but has not received much attention in robotic surgery. This study is one of the first to explore and quantify the degree of disruption encountered due to poor robotic visualization at a major academic center. In case observations across 28 RAS procedures in various specialties, any lens occlusions or “debris” events that appeared on the monitor displays and clinicians’ reactions, the cause, and the location across the monitor for these events were recorded. Data were then assessed for any trends using analysis as described below. From around 44.33 h of RAS observation time, 163 debris events were recorded. 52.53% of case observation time was spent under a compromised visual field. In a subset of 15 cases, about 2.24% of the average observation time was spent cleaning the lens. Additionally, cautery was found to be the primary cause of lens occlusions and little variation was found within the spread of the debris across the monitor display. This study illustrates that in 6 (21.43%) of the cases, 90% of the observation time was spent under compromised visualization while only 2 (7.14%) of the cases had no debris or cleaning events. Additionally, we observed that cleaning the lens can be troublesome during the procedure, interrupting the operating room flow.


Introduction
Robotic-assisted surgery (RAS) has rapidly developed a foothold in today's operating rooms (ORs) due to landmark technological innovations that yield tangible advantages over laparoscopic and open surgery [1].These advantages expand the benefits of minimally invasive surgery (MIS) over traditional surgeries and allow for smaller incisions, lower risk of infection, and shorter hospital stays while simultaneously enhancing surgeons' ability to perform more complex procedures that require meticulous detail [1].RAS reaches beyond the laparoscopic platform by offering increased dexterity through the use of robotic instruments that follow the motion of the human wrist and tackle the limited maneuvering of the straight-stick laparoscope [1][2][3].Three-dimensional vision is also a critical advancement that improves upon the hurdles of hand-eye coordination, depth perception, and twodimensional view offered by laparoscopic surgery [1,3,4].With these benefits, RAS is able to transform medical and surgical applications through the completion of complex procedures and tasks that were previously difficult to achieve [4].Despite the purported advantages of robotic technology, RAS is not without limitations that may impede future growth.The most commonly known disadvantage of RAS includes the lack of haptic feedback and decreased sense of touch inside the body [4,5].This makes surgeons more reliant on visualization during a procedure and places high importance on the maintenance of visual clarity in the operating field [3,6].Unfortunately, during both laparoscopic and robotic procedures, the lens is prone to visual occlusions, which range in severity and are generally caused by fogging, cauterization, and/or fluid accumulations [6][7][8].
Although the literature is scarce regarding a deeper understanding of the impact of visual occlusions in RAS, the topic has been explored in laparoscopic surgery.For example, Yong et al. [8] found that nearly 40% of procedure time is performed under impaired vision.In addition, some procedures are converted to open laparotomy due to visualization problems caused from lens debris events [9].Poor visual clarity has also been found to decrease dexterity and hand-eye coordination and spur a sense of frustration [5,8,10].Furthermore, camera cleaning and changing of the camera lens can cause disruptions to the surgical environment flow and can serve as risk factors for errors and compromised safety in the operating room (OR) [11,12].
There appears to be limited research focusing on the consequences of lens occlusions and poor visualization in the RAS [13,14].Nezhat's et al. found that maintenance of visual clarity is more difficult for RAS because the system occasionally switches between the 30° degree and 0° camera, both of which are also bulkier than the laparoscopic camera [7].FloShield™ (Minimally Invasive Devices, Inc.) introduced a cleaner specifically designed for the robotic scope.Moreover, Insuflow™ (Lexion Medical) has tested their device in some RAS cases [5,7].However, such solutions have shown limited clinical impact, and more research needs to be conducted to understand and quantify how visual occlusions play a role in the outcome of an RAS procedure [14].To shed a light on the degree of surgical disruption, this study quantifies the degree of disruption due to visual acuity in RAS at a major academic institution.

Methods
This observational study includes visual case observations during n = 28 RAS procedures using the Intuitive Surgical daVinci XI performed at Dell Seton Medical Center at the University of Texas at Austin.The cases were across various specialties such as general surgery, urologic surgery, and obstetrics and gynecology (OB/GYN) and were performed by six surgeons.Tracked metrics regarding visual clarity included: (i) the total operative time spent under clear vision, suboptimal vision, and cleaning time, (ii) number of cleaning events, (iii) number of lens occlusion (or "debris") events, (iv) reaction/decision of clinicians in response to a lens occlusions event, (v) cause of lens occlusion, and (vi) the location distribution of debris across the monitor.General information of the case including type of case, the surgeon operating through the case, the scope lens angle (0 or 30 degrees) and any switches in the scope, and instruments utilized throughout the procedure was recorded.There were three observers in total.A standardized data tracking sheet was utilized to verify uniformity in observations.This study was deemed IRB exempt.

Definitions
Debris events were defined as any time a new discernible distortion was observed on the lens in relation to the previous state of lens clarity.Clear vision was defined as the monitor being clear with no lens occlusion and/or visual distortion present, as determined by the naked eye.Time under impaired or "suboptimal" vision was defined as time spent performing surgery with such distortion present (i.e., time spent with any distortion left uncleaned/ignored by the surgeon).Of note, this includes the time spent cleaning the lens, which includes the insertion and withdrawal of the scope.In the observed cases, start time was recorded as the time when the first appearance of a lens occlusion occurred, and end time was the time of the last lens occlusion or cleaning event during the time of observation.Additionally, the data recorded were collected as seen from the bedside monitor displays versus inside the console.In total, 44.50 h of RAS observation time was recorded.The lens cleaning technology used in each case was Clearify Visualization System™ (Medtronic) [15].

Results
Table 1 shows a summary of lens occlusions across the 28 cases, of which there were 163 debris events.With an average case observation time of 95.00 min, an average of 44.46 min was spent under suboptimal vision with an average 5.82 different debris events per case.An average of 52.53% of the case was spent under suboptimal vision.Ratio of clear vision and suboptimal vision to total time were compared utilizing a t test assuming equal variances.The total operating time under impaired vision was not statistically longer than the total operating time under clear vision (p = 0.574).
A time progression chart was built to represent the progress of debris events across a case and to visualize clinicians' reactions to those events (Fig. 1).Lens debris events and the surgeons' reactions were noted throughout the observation period.Consecutive blue dots represent uncleaned occlusions accumulating across the lens and visual field throughout the surgery.Green squares indicate the decision to clean the scope lens.It is worth mentioning that the time represented in the figure is not representative of total operative time as the start time was recorded as the instance when the first appearance of a lens occlusion occurred.Also, the end time was defined as the time of the last lens occlusion or cleaning event during the time of observation.Cleaning times were also recorded for a subset of 15 cases (Table 2).A maximum of 6.37 min of the total observation time (for a case of 139.90 min) was spent cleaning the scope.On average, the clinician spent 2.13 min per case cleaning the scope.Additionally, scope cleaning took an average of 36.09s per cleaning event (range 13-150 s).
For a subset of 102 recorded debris events, the debris spread was recorded visually as it appeared on the screen, and the monitor display was divided into 16 cells (a 4 × 4 grid) (Fig. 2).The debris spread was recorded in terms of the percentage covering each cell in the 4 × 4 grid.The   To assess whether the amount of debris coverage of the monitor influenced the surgeon's decision to clean the lens, the monitor display before cleaning occurred was analyzed.For 39 events, cells that contained debris were counted using a 4 × 4 grid within each cell, and the frequency of monitor coverage that led to cleaning was calculated and shown in Fig. 3.
In the later cases, the cause of the lens occlusion (cautery, blood/tissue/fat, irrigation) was tracked and recorded for 13 cases to find the most common factor contributing to the compromised vision (Fig. 4).

Discussion
Although research on the effects of visual occlusions during minimally invasive surgery has been predominantly focused on laparoscopic surgery, the robotic OR also has similar problems when trying to maintain visual acuity [7,8].This study found that a considerable amount (52.7%) of the total case observation time was spent under compromised visualization.Although the calculated p value states that there is no statistically significant difference between the clear vision and suboptimal vision time across each case, it should be noted that 52.7% of the total case observation time spent under compromised vision is still a concern.
Six (21.43%) of the cases showed that over 90% of observation time was spent under suboptimal vision, suggesting there are certain scenarios in which lens obscuration is a substantial problem.The progression chart shown in Fig. 1 is a critical representation of the overall findings in the OR and graphically displays surgeons' decision-making output when encountering debris events.This figure shows that 129 out of the 163 (79.14%) debris events were not cleaned within the first 2 minutes of being encountered.Additionally, it can be seen that 19 (67.86%) of the cases ended with a dirty lens, and 96 out of the 163 (58.90%) of the lens occlusion events were seen as ignored (meaning they were not followed by a cleaning event).Alternatively, only two (7.15%) of the cases showed that under 10% of the observation time was spent under suboptimal vision and in these cases, 100% of the observation time was spent under clear vision with no cleaning or debris events occurring.Of note, these two cases were hysterectomies both performed by a surgeon specializing in OB/GYN.
Based on the subset of cleaning analysis across 15 cases, the average total cleaning time was 2.13 min making up 2.23% of the entire average case observation time.While this makes up a small percentage of total time, we did find that times varied on level of assistant experience.Having OR technicians with less experience led to higher cleaning times and more errors when cleaning the scope.Whether or not this affected patient safety is beyond the scope of this study, but it did add slightly to the overall operative time.
As shown in Fig. 2, debris spread was also tracked to see which locations have the most accumulation of debris.Coverage of the monitor display with debris was fairly consistent across each grid cell indicated by the low standard deviation.The cells with the maximum difference from the mean are located on the bottom corners.Similarly, Fig. 3 illustrates the amount of the debris coverage of the monitor tolerated before the cleaning was performed.It was observed that 17 cleaning events (43.60%) were correlated to instances in which more than 50% of the monitor was covered by debris.As seen in Fig. 4, blood fluids (tissue/fat/blood) were shown to be the most frequent (59.4%) cause of debris accumulation on the scope lens followed by cautery (27.5%).Bodily fluids were grouped together because it was difficult to distinguish between these during the case.Finding the common causes of debris may allow for a better understanding of how to prevent/address lens occlusions while providing insights on what features future cleaning technologies should target.
Lens occlusions is a significant problem in ORs that can result in interruptions and disruptions to the surgical flow during an operation that may subsequently worsen surgeons' mental focus [12].To react to these interruptions, surgeons and OR staff can either ignore the lens occlusions and work with impaired vision or take time to clean the scope.Surgeons have to make a real-time decision based on the state of surgery and the benefit of cleaning to not waste additional time or resources.Although the possible effects lens occlusions have on patient safety are currently unknown, it is generally accepted that over time, these decisions may exacerbate decision fatigue [16].Patient safety is likely tied to surgeon visualization and amount of decision fatigue.However, minimizing decision fatigue is reason enough to strive to minimize lens occlusions during operations.
Despite our findings, there are a few limitations that should be noted.These include the heterogeneity of the cases observed and being a single institution study.The heterogeneity of cases introduces a wide variety of variables such as the length of the case, complexity of the case, surgical staff involved, and the experience of personnel, which speaks to the large standard deviation for suboptimal time.However, the heterogeneity of the cases leads to a more robust and diversified sample set.Additionally, it should be noted the monitor display has three different display screens (one for each eye of the lens and the surgeon console).The surgeon console includes combined views of the two lenses and added 3D visualization, which may result in lens occlusions appearing somewhat different from the external bedside screens.The surgeon also has the ability to toggle between eye views which may improve visualization versus the assistant monitors.Comparisons of the display of the console to the monitor to see how lens occlusions appear to the operator of the console and the rest of the team relying solely on the monitor displays is also needed to better understand surgeons' decision-making process regarding cleaning relative to lens obscuration thresholds.

Conclusions
By analyzing 28 RAS procedures, we demonstrated that lens occlusions affecting visualization is a major problem in RAS in which over 50% (52.53%) of the case duration had impaired vision.It is concerning that surgeons are performing complex minimally invasive surgeries across various surgical specialties with a compromised visual field.Future areas of research will include analyzing the effects of experience of the surgeon and severity of the debris on willingness to clean.Additionally, clinical and monetary impacts of compromised visualization and cleaning will be assessed in the future.

Fig. 1 a
Fig. 1 a Progression of debris events in vision in 28 cases over time.In this figure, the start time was recorded as the time instance when the first appearance of a lens occlusion occurred, and the end time was defined as the time of last lens occlusion or cleaning event during

Fig. 2
Fig. 2 Heatmap of the distribution of debris on the monitor display divided into 16 squares

Fig. 3 Fig. 4
Fig. 3 Histogram of debris coverage of the monitor before cleaning was performed

Table 1
Metrics regarding time spent working under suboptimal vision and debris events in each case across 28 cases

Table 2
Metrics regarding cleaning time of the scope considering both individual cleaning events and case totals for a subset of 15 cases