Retrograde Intrarenal Surgery in the Future: Robotics

57
Retrograde Intrarenal Surgery in the Future: Robotics


Anup Patel


Consultant Urological Surgeon, London, UK


Introduction


As the world enters the fourth industrial revolution, there will be rapid evolution in the areas of artificial intelligence (AI), robotics, three‐dimensional (3D) printing, nanotechnology, genetics, and biotechnology. “Robots” have been part of human society since before Leonardo Da Vinci described gears, wheels, and pulleys to animate a robotic knight for a Milan pageant in the 1490s. Robot, a derivation of the Czech word “robota” from a 1920 play entitled R.U.R. (Rossumovi Univerzalini Roboti) by Czech author Karel Capek, meant forced labor or serf. Capek described fictional human‐like machines that continuously performed repetitive hard, dull, or dangerous complicated tasks for their human counterparts without feeling. The machines eventually became “resentful,” rebelled, and killed all the humans. Consequently, over time, the word “robot” has encompassed several meanings including that of a brutal human being who has become insensitive or machine‐like because of overwork and mistreatment. In 1939, after meeting Ernest and Otto Binder at the Queens Science Fiction Society, who had published a story entitled I, Robot about a sympathetic misunderstood robot called Adam Link (who was motivated by love and honor), and following a conversation a year later with John W. Campbell, Isaac Asimov introduced his Three Laws of Robotics in a short story called “Runaround.”



  1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.
  2. A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.
  3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws.

    A fourth law was later added:


  4. A robot may not harm humanity, or, by inaction, allow humanity to come to harm.

Since then, robots have been remade in mankind’s image, and popularized by the televisual industry. Before Azimov, Edgar Allen Poe (1839) related the story of a wounded soldier whose body was rebuilt with synthetic parts. During the early part of the space race (1960) this was termed a cybernetic organism, for it was envisaged that blending technology with astronauts’ bodies could help endurance and survival in space. This led to a plethora of medical technologies such as pacemakers, insulin pumps, cochlear implants, prosthetics, and exo‐suits, all of which are already integrated into human society.


In today’s world, the real issues regarding robots’ future roles center over debate as to whether they will primarily augment or replace current human functions. Autonomous robotic technology has increasingly replaced traditional tedious repetitive industrial human manufacturing roles, performing dull, dirty, and dangerous tasks, leaving their human counterparts in the shade for both accuracy and efficiency. Simultaneously, programmable and master/slave‐controlled drone technology, automated driverless vehicles (cars and airplanes), and crop spraying and seed dispersal farming drones have all begun to impact different aspects of daily human life, while also replacing humans in situations of inherent task‐associated danger. Where sustained precision, reliability, reduced costs, and greater productivity are desired, advanced sensor robotic technology can easily augment or supplant human performance without fatigue. More human functions are or will be threatened in the coming years, and some have even predicted that the ultimate supremacy of robotic over human development will be reached in only two more decades. It is a potential revolution akin to the Internet, where robots can fulfill diverse roles in society as human substitutes, companions, and partners (co‐bots).


For medical applications, a key issue is whether robotic machines are better, or simply enhance human performance by lacking or reducing fatigability, allowing greater precision and efficiency over longer durations. This is particularly true for master/slave devices where expensive development costs can be offset by time saved, greater task performance precision, and increased productivity, and perhaps also consistency, particularly in repetitive or endurance‐based tasks. Ideally, a master/slave human–robot partnership should enhance the capability of each individual component part provided the master can maintain concentration throughout the task. In interventional urology, robotics has already made great inroads through the use of devices that are master‐independent, such as the PUMA560 [1] (1988; arm with six degrees of freedom used for brain biopsy and later for transurethral resection of prostate) and, more recently, the Cyberknife for precision radiotherapy. Zeus and Aesop (US Food and Drug Administration [FDA] approved in 1990; Computer Motion, Santa Barbara, CA, USA), were bought and quashed by Intuitive Surgical (Sunnyvale, CA, USA) in favor of their Da Vinci™, a next‐generation master/slave console‐controlled robotic device with endo‐wrist technology (FDA approved in 2000) [2, 3]. It and the Sensei‐Hansen device (Hansen Medical, Mountain View, CA, USA) [4, 5] were their urological successors, incorporating console‐based 3D optical visualization with magnification, motion control and scaling, tremor reduction, and a greater range of movement and ergonomics compared to traditional instruments. Although it is beyond the scope of this chapter to examine cost‐benefits, this aspect of all new technologies will be amplified in an era of economic austerity coupled to higher healthcare demands from chronic illness, fueled by obesity and ageing pandemics, and falling birth rates in the world’s developed nations.


Rupel and Brown’s endourological legacy of cystoscopic nephroscopy for renal stones at open surgery in 1941 [6] has been the evolution of a wide range of extracorporeal, ureteroscopic, percutaneous, and laparoscopic procedures, all embraced by urologists worldwide. The dominant large/branched renal stone treatment remains percutaneous nephrolithotomy (PCNL). Nevertheless, after two decades of evolution, classical retrograde intrarenal surgery (cRIRS) is now the most dominant endourological stone treatment modality overall. Expertise and allied technological growth has projected cRIRS to the mainstream of complex endourological stone and soft tissue treatment in the upper urinary tract. For the first time, cRIRS rightly challenges the dominant alternative of PCNL treatment of larger intrarenal calculi [712]. However, cRIRS for large stones remains a technically challenging procedure requiring specific expert endo‐skill sets, real‐time problem solving capabilities [1315], and both physical and mental endurance. Even when ureteral access sheaths (UAS) are used (almost mandatory in this setting), and the stone(s) can be accessed and completely laser‐fragmented, the single‐sitting stone‐free rate is limited by the operator’s stamina during long dusting and “pop‐corning” procedures, and the size and number of stone fragments that can safely be removed down the ureter in a safe, timely manner (in turn dependent on stone composition and hardness, and collecting system anatomy). Additional limiting factors are current flexible ureterorenoscope (FUR) design, a moving target (renal respiratory excursion) during fragmentation, and awkward calyceal anatomy, particularly if several such cases are scheduled per day. This has resulted in ≥50% secondary procedure rates to render patients stone‐free. Furthermore, the surgeon depends on assistants for key task performance, to activate lasers (standby‐ready and energy‐rate settings adjustments), deploy nitinol baskets and capture fragments, maintain and enhance irrigant flow, aspirate, and to draw/inject contrast, while manipulating and targeting the ureterorenoscope tip. All operating room staff are exposed to radiation over a longer period (range of 1.7–56 μSv [1618]). Meanwhile, less invasive evolutionary PCNL techniques (mini, ultra‐mini, super‐mini, and micro‐PCNL [1924]) have also evolved to challenge cRIRS, to gain parity or ascendancy in the current larger renal stone treatment paradigm, particularly in Asia, so further technological and technique evolution is mandated.


During the last two decades, significant changes in both semirigid and FUR design, specifically tip and shaft size reduction along with configuration changes, have facilitated easier natural orifice access into the upper urinary tract for the majority of users (albeit at the cost of increased fragility and repair costs). Greater direction and range of tip deflection (both single and dual active flexion/extension motion range) has made navigation through the intrarenal collecting system easier, allowing more urologists to reach and inspect >95% of the entire collecting system (the most challenging part being the lower antero/medial calyx, especially in a dilated pelvicalyceal system which limits use of both active and secondary passive deflection).


In 2012, ELMED (Ankara, Turkey) started developing a procedure‐specific robot‐assisted RIRS (RA‐RIRS) device [25]. The IDEAL (idea, development, evaluation, assessment, long‐term study) framework for surgical innovation stages was embraced from prototype to the commercially available Avicenna Roboflex (IDEAL stage 1), and early clinical experience with treatments performed by different experienced endourologists (IDEAL stages 2–4) was reported as follows.


Device development


The basic Avicenna Roboflex design from prototype (Figure 57.1a) to successive versions (Figure 57.1b–d) consisted of a master control console (MCC) seating the surgeon and a manipulator arm (MA), which can now be coupled to any standard commercially available FUR using a custom‐made model‐specific adapter. During the early developmental phase, incremental improvements were made to the size and design of the MCC, the joysticks to control advancement, rotation, and tip deflection, and to develop a system to remotely advance‐retract the laser fiber.

Image described by caption.
Image described by caption.
Image described by caption.
Image described by caption.

Figure 57.1 (a) Avicenna Roboflex version 1 prototype for RA‐RIRS: simple circular holder for FURS control and laser fiber movement, controlled from console with two small four‐way joysticks (left for forward/backward and laser fiber in/out, right for rotation clockwise/counterclock wise, and deflection up/down). The joysticks were later modified with bigger, thicker handles. (b) Similar manipulator arm with longer horizontal movement (forward/backward movement) for easier basket application, and smaller control console with height adjustable seat. Control of the left joystick movements were unchanged, but right joystick had a thick, rotatable handle to control FUR rotation and there was another individually rotatable disk incorporated into the top of the joystick handle to control up/down deflection. (c) Version 3: new manipulator arm design (with high‐precision robotic components) with open‐type endoscope holder and new master control console (MCC). Rotation movements over 440° were designed for smooth precision following hand movement. Special endoscope holder was designed for easy docking with deflection control unit. A force‐torque deflection mechanism sensing system was developed to protect the device against high‐deflection torsion. Horizontal movement (forward/backward) was set to a speed of 22 mm/second (adjustable from 0.5 to 22 mm/second). The MCC was designed with two joysticks to control rotation and forward/backward movement (left joystick), and deflection movement on the right joystick. Laser fiber movement and vertical movement (according to patient tabletop position) were controlled from the touch screen. Later, a central precision control cylinder was added to the center of the MCC for fine tip deflection control as a modification of that version. (d) Version 4 incorporating a new interchangeable endoscope holder system to accommodate all FUR brands/models. The right joystick controlled rotation and two‐stage speed control for forward/backward movements and incorporated a thumb wheel (“magic wheel”) for tip deflection. Precision or scale of all rotation and tip deflection movements were selectable from the MCC touch screen. A new MCC foot‐pedal unit was incorporated to control any laser or fluoroscopy foot pedals. Additional visual guidance was provided on the video monitor screen to indicate the spatial position of the FUR (such as horizontal distance, rotation angle, and deflection angle).


Master control console design features


The current Avicenna Roboflex (version 5) MCC (Figure 57.2a) is connected to the MA unit, and its monitor is connected as a slave to the operating room stack monitor. It has key‐control on‐switch, with separate red push‐button emergency stop (Figure 57.2b). The MCC reclining seat has adjustable height, head, and armrests. Six users can store individual seat positions and all preferred custom deflection‐scaling settings in the system memory (Figure 57.3). The comfortably seated surgeon activates wheel locks, selects endoscope type, brightness, time, background screen color (one of six options), rotation precision and direction, deflection precision, and horizontal plane movement speed via the set‐up menu. Moreover, they can also select low‐ (0.5×) or high‐precision (1.5×) modes, timer stopwatch start or reset options, and define target kidney side (right or left) on the touch‐screen panel (also displayed on the center right of the monitor screen). The MCC touch screen also allows control of laser fiber actuator, irrigation flow rate, short bursts of flush, and abdominal compression belt inflation to reduce renal respiratory movement (Figure 57.4). Below this on the monitor screen is the endoscope deflection and rotation icon, and above it are laser fiber and irrigation flow rate indication bars (Figure 57.5). The rotation display circle containing the endoscopic view is in the center. The surgeon controls two joysticks (with adjustable padded wrist support) to manipulate the FUR, which is mounted on the MA component. The right‐hand vertical joystick bulb’s “magic wheel” controls fine tip deflection akin to the handpiece of any standard FUR (Figure 57.6a), but with additional tenfold motion scaling. The deflection direction (upwards, downwards) can be programmed for logical deflection (down = tip down and up = tip up) or counterintuitive deflection (down = tip up and vice versa). The new horizontal left joystick (Figure 57.6b) allows clockwise and counterclockwise MA rotation, as well as advancement and retraction of the endoscope shaft with millimeter per second adjustment, using +/– buttons. The rotation speed, scale (2, 1, 0.5×) and MA rotation direction (clockwise or counterclockwise), and advancement can be regulated by console touch‐screen control. Warning display box alerts inform the user about rotating the MA to the mounting position (90° clockwise), continuing from a particular position, or returning to the neutral position.

Image described by caption and surrounding text.

Figure 57.2 (a) Version 5 MCC unit with key control start, and improved adjustable seat design, headrest, and wrist support bar as an ergonomically sound comfortable surgical workstation. Horizontal left joystick for horizontal and rotational movement control, and right vertical joystick with magic wheel to control fine tip deflection (all with adjustable precision from MCC touch screen). (b) Red emergency stop button.

Two screens displaying the seat positions and all preferred custom deflection‐scaling settings in the system memory for six users.

Figure 57.3 System memory for six users.

Image described by caption.
Image described by caption.

Figure 57.4 The MCC touch screen with various functions in the set‐up menu: adaptation to FUR mode (i.e. American or non‐American), change of rotation, deflection precision, horizontal advancement/retraction speed, laser fiber advancement and retraction, irrigation flow rate adjustment, compression belt inflation‐deflation, and degree of rotation, deflection, and horizontal insertion display, along with timer and stop watch.

Image described by caption and surrounding text.

Figure 57.5 MMCC with slave monitor screen for endoscopic image within central rotation indicator circle, progress bars for irrigation flow rate and laser fiber extrusion distance from tip (mm) on right upper, side of kidney right central, FUR tip deflection indicator right lower, and insertion rate progress bar left of center.

Image described by caption and surrounding text.
Image described by caption and surrounding text.

Figure 57.6 (a) Magic wheel atop bulb of right‐hand MCC control joystick for fine scaled up‐and‐down endoscope tip deflection. (b) Left‐hand horizontal control joystick of MCC for clockwise and counterclockwise rotation, and in‐and‐out endoscope movement.


The assistant introduces the laser fiber at the start with the endoscope tip in a “0” straight position, with the tip cladding just visible endoscopically, and the touch‐screen “eye button” in the laser fiber box is pressed. The zero button is then activated to level the fiber tip with the work‐channel exit point. The endoscope can now be safely maneuvered to the stone before advancing the fiber tip 1.8 mm for safe firing. The MCC floor‐mounted red right foot pedal controls laser firing, while the white left pedal controls the fluoroscopy (Figure 57.7a), both aided by compressed air cabling to a foot‐pedal coupler unit (Figure 57.7b). The laser fiber foot pedal is automatically disabled when the fiber tip is within 1.6 mm of the endoscope tip (the red warning bar is a useful safety feature).

Image described by caption and surrounding text.

Figure 57.7 (a) MCC floor‐mounted red right foot pedal (for laser firing control) and white left pedal (for fluoroscopy control). (b) Compressed air cable activated foot‐pedal coupler unit.


Manipulator arm design


The MA consists of computer‐controlled motor systems, a robotic arm, which holds and moves the endoscope, and the lower stabilizing arm to which vertical supports are attached in two places. The arm height can be adjusted according to the patient’s anatomy. The endoscope hand piece with appropriately placed deflection lever is fixed into the arm’s customized coupler housing and secured by double clip holders after draping (Figure 57.8). The distal vertical stabilizer is fixed to the preplaced UAS hub, while the proximal one stabilizes the straight endoscope shaft approximately 5 cm further back (Figure 57.9a). A UAS is virtually mandatory for Avicenna Roboflex procedures, due to lack of haptic feedback, reducing risk of inadvertent ureteral injury. The custom‐made coupler accepts any FUR, although initial development was made using a digital instrument (Flex‐XC; Karl Storz, Tuttlingen, Germany). The MA can be rotated bidirectionally by 210° allowing overlap for a total 440° compared to 120° for humans. Miniaturized hand‐piece adaptor motors allow endoscope steering lever movement for deflection and enable motion scaling of the cable‐based endoscope tip movement mechanism. Due to FUR design and cabling inertia, a short lag period between robot control and endoscope tip reaction exists. This is vital for the user to understand and remember, for there is a tendency to continue MCC deflection if no immediate visible screen reaction is seen, leading to delayed deflection overshoot. The same lag occurs for the same reason when switching deflection direction. Normally, a 10° manual lever deflection moves the tip 60°. With Avicenna Roboflex, FUR tip control precision is eightfold greater than manual operation. The thumb‐wheel scaling precision, which can be low or high, or on a scale of 1–10 as set on the MCC, allows 10° of thumb movement to result in 3–30° of tip deflection. The applied forces were limited to 1 N/mm2 for safety to minimize the risk of collecting system and endoscope injury. Moreover, the coupling hand‐piece adapter accepts placement of a micromotor‐driven actuator system, which is connected to the instrument’s working‐channel for precise laser fiber advancement (Figure 57.9b).

Image described by caption and surrounding text.

Figure 57.8 The MA custom endoscope coupler and motorized actuator housing, with FUR hand piece fixed in the manipulator coupler and stabilized by double clip holder after draping.

Image described by caption.

Figure 57.9 (a) The proximal stabilizer supports the straight endoscope shaft, while the distal stabilizer fixes the UAS shaft in the same plane. (b) The laser fiber (and irrigation tube) is attached to the endoscope by a motorized actuator for fine laser fiber movements.


Experimental evaluation


For all MCC steering, and movement reliability functions, the Avicenna Roboflex was tested in a validated flexible ureterorenoscopy in vitro training model (Minnesota University Kidney Model [25]) containing artificial stones. This enabled safe prototype(s) simulations to evaluate key parameters and all device functions. It was also used for hands‐on training of all participating surgeons.


In vitro studies


Avicenna Roboflex showed safe, stable maneuverability, with easy translation of MCC functions when in both bench training models and pig ureters. Prototype ergonomics were already good, but incremental improvements to wrist movements, and fine‐tuning of deflection were made.


Surgical technique


image With the patient under general anesthesia, supine in lithotomy, intravenous diuretic (lasix) and antibiotic prophylaxis are given, and manual cystoscopy, retrograde pyelography, and placement of a safety guidewire can be performed. The ureteral orifice is accessed with a semirigid ureteroscope, inspecting the ureteral lumen up to/beyond the ureteropelvic junction, thereby achieving maximal optical ureteral orifice and ureteral dilation, leaving behind a working wire as needed. A suitable size (10–12/12–14 Fr) and length (35–46 cm) UAS is then advanced into the upper ureter under fluoroscopic control as needed, confirming ureteral integrity postplacement with further pyelography via the access sheath obturator. Next, the FUR is inserted into the UAS lumen and its hand piece is fixed in the sterile plastic‐draped robotic‐arm coupling adaptor, where it is securely locked with securing clips. Thereafter, checks are made for a stable, secured, straight FUR shaft, before console steering can begin. Using the MCC touch screen and control levers, the target stone(s) is endoscopically visualized and recorded with fluoroscopic snapshot (master console floor left foot pedal). If the laser fiber was not inserted at the start, the endoscope is placed in a neutral straight retracted position with the console zero button to guarantee safe laser fiber insertion without risking work channel damage. The MCC provides a memory function to guide the scope back to its previous position near the stone surface after laser fiber insertion. The inserted laser fiber tip with cladding should be just visible near the endoscope tip and the eye button on the master control laser fiber box is activated for memory entry. Only then can the fiber be retracted to a position level with the scope tip (zero button) and advanced from there by the console touch screen controls using the actuator device fitted to the endoscope work channel. The fiber cannot be retracted inadvertently into the scopes working channel and energy activated, where it might damage the instrument. Once the fiber is advanced (as shown by bicolored bar in MCC laser fiber panel) with precision to the optimal distance from the stone surface, laser‐induced lithotripsy can be initiated, ideally “dusting” the stone surface with meandering rotation and small up‐and‐down deflection movements of the laser fiber tip in the millimeter range using the left hand rotation control lever and/or the right hand pin‐wheel control, or by fragmentation (various techniques). Any commercially available holmium:YAG laser unit can be used, but one that allows higher‐frequency application at low energy is recommended for optimal dusting. Smaller fiber sizes allow sufficient scope tip deflection with lowest risk of fiber fracture in the work channel at maximum energy and frequency. Once fragmentation has started, higher energy frequency gives a “popcorn effect” for fragment pulverization when dusting is not possible. Sometimes, with fiber tip in the calyx lumen center, this can even be done “hands‐free” if time‐consuming (see Video 57.1). Fiber tip erosion must be monitored, as it reduces energy transfer over time, and judicious appropriate cleaving with fiber re‐introduction is undertaken as required. To introduce nitinol baskets for fragment retrieval at the end, the laser actuator has to be driven back to its preset zero position with the endoscope tip straight, and the fiber exchanged for the basket, once the laser console is on standby (assistant). This might be time‐consuming especially if there are multiple fragments, and hence ablation by stone dusting is preferred until a suitably sized fragment is retrieved at the end for stone analysis. The need for JJ stent placement at the end of the procedure is not mandated, but left to the operator’s discretion.


Published clinical experience


Patients and outcomes


In the first peer‐reviewed study, Ankara Medical School ethics commission approval was secured, and 81 patients were treated (IDEAL stage 2) (Table 57.1

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Aug 5, 2020 | Posted by in UROLOGY | Comments Off on Retrograde Intrarenal Surgery in the Future: Robotics

Full access? Get Clinical Tree

Get Clinical Tree app for offline access