165: Robotics: Lobectomy

Lobectomy via minimally invasive video-assisted thoracic surgery (VATS) has proved to be a feasible and oncologically acceptable approach for non–small-cell lung cancer (NSCLC) and other isolated tumors and conditions. However, despite multiple studies showing clear benefits over a traditional thoracotomy approach, such as decreased length of stay, decreased short-term postoperative pain, and fewer complications,^1–3 VATS is still not accepted as the standard approach for anatomic resection, and is only slowly being implemented more widely. The explanation is likely multifactorial including (1) technical issues, such as two-dimensional imaging and limited maneuverability of instrumentation; (2) lack of adequate training; and (3) concerns about the consequences of major vascular injury with a closed chest approach.

To address the perceived technical limitations of conventional minimally invasive platforms, a master–slave robotic surgical system was developed (da Vinci Surgical System, Intuitive Surgical, Sunnyvale, California). The major advances were the three-dimensional visual system that reestablished binocular vision and wristed instrumentation capable of seven degrees of freedom enabling more natural bimanual movement for precise dissection. Initially, the system was approved by the Food and Drug Administration for cardiothoracic surgery because the original intent was to achieve true closed chest cardiac surgery. This, however, has not been fully realized. Instead, the most common applications that evolved were for pelvic procedures—prostatectomy and hysterectomy. Similarly, while use of robotics for general thoracic surgical procedures dates back to initial case reports in the early 2000s, it was not until 2004 and 2006 that actual series of robotic lobectomies were reported by Melfi et al. and Park et al., respectively.^4,5 These centers reported the initial technique and early perioperative experiences that demonstrated feasibility, safety, and concordance of outcomes with the largest series of VATS lobectomies. Subsequently, there has been a steadily increasing interest in robotic lobectomy with additional publications with greater numbers of patients and various modifications of the technique.^6–8

This chapter will focus on review of the general principles and clinical aspects of robotic lobectomy with an emphasis on patient selection, preoperative preparation, technical aspects, and perioperative outcomes.

The guiding principle that must be remembered when one is considering utilizing robotic surgical systems for any procedure is that the robot is a tool like any other in the art of surgery. It is up to the surgeon to use his or her best judgment as to whether its use is appropriate and in the best interest of the patient. Robotic procedures are simply minimally invasive procedures, that are performed with a different, perhaps more advanced technology that has unique advantages and disadvantages. In the case of pulmonary lobectomy the robotic approaches that have been described all conform to the consensus criteria of a standard VATS lobectomy put forth in the Cancer and Leukemia Group B (CALGB) prospective, multi-institutional registry study (CALGB 39802).⁹ For early stage NSCLC (node-negative, peripheral tumors ≤3 cm) this definition includes: absence of rib-spreading, minimal incision size (no greater than a 4–8 cm access incision with 0.5-cm port incisions), videoscopic guidance at all times, and traditional hilar dissection with individual ligation and division of lobar structures. Adhering to these principles the authors were able to demonstrate that VATS lobectomy is associated with acceptable morbidity and mortality. Similarly, multiple independent centers have demonstrated the feasibility and safety of robotic lobectomy^4–8 while adhering to these same universal aspects of minimally invasive thoracic surgery established for VATS lobectomy. Moreover, a recent multicenter study by Park et al.¹⁰ also demonstrated excellent long-term oncologic results of robotic lobectomy in the treatment of early NSCLC.

For anatomic lobectomy the author practices a VATS-based robotic approach with a small (3–4 cm), non–rib-spreading access incision whereas others advocate a complete portal approach.^7,8 While there are minor technical differences, the conduct of the procedure and utilization of the robotic technology for dissection are uniform. Major emphasis will be placed on the VATS-based technique.

Currently, much like with VATS lobectomy neither the American Board of Thoracic Surgery nor any governing surgical society, such as the American College of Surgeons, Society of Thoracic Surgery or American Association of Thoracic Surgery has any published guidelines for the training and accreditation of surgeons and operating room teams for the performance of robotic thoracic procedures. As a result, each hospital typically has developed its own policies. Hospitals uniformly mandate that surgeons attend an intensive, 2-day training course given by Intuitive Surgical® that is comprised of didactic instruction regarding the system components followed by simulation training for basic skills and cadaver-based training for specific procedures. It is critical for specialty-specific personnel – operating room nurses, surgical technicians, and bedside assistants – to be formally trained on the basics of system functioning, instrument changes, and position of the surgical cart. This is typically done by the robotic company representative. It is common, but not required for the prospective robotic surgeon to observe an established practitioner to become familiar with specific index procedures. This author cannot stress enough how critically important case observation is during the training and prior to implementation of robotics into treatment of patients.

Once the entire surgical team has received the appropriate training, institutions usually will allow implementation of the robotic system into procedures under the supervision and guidance of a case proctor, defined as a surgeon with documented clinical experience independently performing robotic procedures. The console surgeon is typically required to perform between three and as many as ten proctored cases before being granted independent robotic procedure privileges. Some hospitals require that eligible proctors themselves have performed a minimum number of cases, while the majority has no such requirement. In fact, most institutions do not mandate that the proctor be specialty-specific—thus, a robotic urologist may proctor a thoracic surgeon. While this may be in compliance with a specific institutional requirement, this type of implementation is ill advised. The ideal situation is for the training surgeon to observe an experienced robotic surgeon in their respective field and then enlist that individual, if possible, to serve as the case proctor. This maximizes the continuity of training and, consequently, patient safety during clinical implementation.

The theoretical benefit of utilizing robotic technology is to replicate what can be done through VATS or almost entirely what can be done through a thoracotomy. Patients eligible for robotic lobectomy include those with suspicious or biopsy-proven NSCLC or other pathologic tumors or disease processes confined to the lung and ipsilateral hemithorax. This should be verified through computed tomography (CT) of the chest and whole body positron emission tomography (PET/CT). For NSCLC, suspicious mediastinal nodal or extrathoracic disease warrants further invasive staging to identify patients with advanced disease requiring multimodality or systemic therapy only. Patients should have adequate cardiopulmonary status and performance status to tolerate lobectomy. Specifically, cardiac disease should be asymptomatic and stable on medication and preoperative pulmonary function tests should demonstrate a postoperative predicted forced expiratory volume in 1 second (FEV₁) and diffusion capacity (D_LCO) above 40% of predicted. Borderline postoperative predicted lung function should be further investigated by quantitative lung scanning and/or exercise testing. Smoking cessation for active smokers should be aggressively advocated.

As with any new surgical technique or approach, careful selection of initial cases is critical to success and progression. While scenarios such as large tumors (>5 cm), extensive hilar or mediastinal disease, postinduction therapy, chest wall invasion, extensive adhesions, and need for bronchial or vascular sleeve resection do not absolutely preclude a robotic approach, it is wise to avoid these conditions until a sufficient experience with most straightforward cases has been developed. Informed consent for the use of robotic assistance should be obtained as a distinct portion of the procedure.

Preparation of the Robot

The operating room technical staff sets up the robotic surgical system (surgical cart, surgeon's console, vision system) in the room (Fig. 165-1). In the beginning of the case the nursing staff power up the system, run the appropriate diagnostics, and drape the robotic arms and camera. This requires two individuals and typically takes 5 to 10 minutes for staff who are trained and are familiar with the process and occurs prior to or while the patient is undergoing induction of anesthesia and positioning.

Anesthesia Considerations

Standard methods of general anesthesia and single-lung ventilation are employed via either double-lumen endotracheal tube placement or bronchial blocker. The patient is placed in a maximally flexed, lateral decubitus position, and single-lung ventilation is initiated. Depending on the size of the operating room, it will often be necessary to move the table away from the anesthesia machine and angle the foot of the table away from the surgical cart (Fig. 165-2). This establishes enough space to dock the robot. Care must be taken to insure that sufficient length of the circuit tubing is available during this positioning, and the anesthesia team must be comfortable that there is adequate access to the patient's airway once docking of the robotic system has taken place.

Initial Exploration and Docking of the Robot

Initial thoracic exploration is conducted with the robotic thoracoscope through a 12-mm trocar in the eighth intercostal space (ICS) just posterior to the anterior axillary line (Fig. 165-3) to verify tumor location, establish a tissue diagnosis if necessary, assess resectability and appropriateness of the robotic approach, and to place the additional incisions prior to docking. A 1-mm incision is placed posterior to the tip of the scapula in the ninth ICS just above the diaphragm. The 3- to 4-cm access incision is placed in the fourth or fifth ICS in the midaxillary line. A fourth incision may be employed posteriorly in the fifth or sixth ICS in line with the ninth ICS incision if so desired. Once the skin incisions have been made, the surgical cart is brought into position from the posterior aspect of the patient with the center column and camera arm angled over the scapula at an approximately 45-degree angle with respect to the longitudinal axis of the patient (Fig. 165-4). This allows for the field of dissection to include the hilar structures and the majority of the chest. When docking the surgical cart, it is important to avoid positioning the surgical cart too close to the patient and maintain adequate spacing between ports (handbreadth). This will eliminate instrument arm conflicts and maximize range of motion of the instruments.

Once the surgical cart is in position the camera arm is attached first to the trocar, and the robotic thoracoscope is introduced and secured to the camera arm. The 8-mm metallic robotic trocars are introduced through each of the other incisions and attached to their respective arms. This is accomplished under direct vision both from outside the patient and from within the patient's thorax. In the case of the access incision the trocar is placed in the midpoint of the incision with room above and below to introduce additional instruments (lung retractor, suction). Care must be taken to ensure that each instrument arm has full range of motion and does not collide with one another or with the patient (Fig. 165-5).

Once the trocars are in place and attached to the robotic arms, the surgical instruments are introduced under direct thoracoscopic vision. A Cadiere forceps is most commonly controlled by one hand for grasping tissue, and a cutting instrument (monopolar spatula, Maryland bipolar, monopolar hook) is used in the other hand. If the fourth arm is employed, it is typically used for a lung grasper or suction irrigator. After the instruments have been introduced the operating surgeon moves to the surgeon's console. The bedside assistant stands at the anterior aspect of the patient and provides additional exposure through the access incision.

Completely Portal Robotic Lobectomy (CPRL-4)

Cerfolio et al.⁸ have developed a robotic incision strategy that avoids an access incision under the premise that no exposure of the intrathoracic cavity to air may have additional benefits over and above lack of rib spreading. There are four robotic arm incisions, all placed in the sixth ICS spaced 9 to 10 cm apart beginning from the midaxillary line to the paraspinal area (Fig. 165-6). In addition, there is a fifth, nonrobotic 15-mm assistant access port through which the endovascular staplers are passed. The surgical cart is then brought in over the head of the patient, with the camera view replicating the traditional thoracotomy view.

Mediastinal and Hilar Lymph Node Dissection

Dissection is performed with the Cadiere forceps and the monopolar cautery spatula. Wherever possible, the entire nodal packet is removed without fracturing the nodes into fragments (Fig. 165-7). Large bronchial or lymphatic vessels can be clipped, and when indicated suspicious lymph nodes are sent for frozen section analysis to identify occult N2 disease. For a right upper lobectomy it is easier to perform the paratracheal node dissection after the specimen has been removed. Similarly, it is often advantageous to perform the subcarinal lymphadenectomy by retracting the stump of the lower lobe to elevate the mediastinum.

Hilar Dissection

If there are no contraindications to lobectomy, individual isolation of the hilar structures proceeds with dissection around the hilar vessels and bronchi performed through a combination of cautery, sharp, and blunt dissection. Complete removal and labeling, rather than sweeping of all regional nodal tissue is performed both for adequate staging and to facilitate isolation of the hilar structures. When either a vessel or the bronchus is mobilized sufficiently, the Cadiere forceps are used to isolate the structure, using the seven degrees of freedom to articulate the instruments at near right angles to do so (Fig. 165-8). Ligation and division of the named vessels and bronchus are performed with endovascular staplers introduced either through the posterior inferior or access incision. This requires temporary removal of one of the robotic trocars followed by replacement of the arm after stapler firing.

The precise order in which the structures are divided depends on the particular lobectomy being performed and the approach (anterior vs. posterior/fissural). Therefore, the specific steps for each lobectomy will not be reviewed in detail.

Division of the Fissure

Completion of the fissure is performed last, just prior to removal of the specimen. For upper lobectomy the entire fissure is divided with endoscopic staplers (Fig. 165-9); for middle and lower lobectomies the anterior portion of the fissure is often divided with electrocautery in the course of dissection of the hilar structures. The remaining posterior portion of the fissure is then completed using the endoscopic staplers.

Specimen Removal

The completed lobectomy specimen should be placed in a durable laparotomy sac and removed through the access incision in the case of a VATS-based incision strategy or by enlarging one of the port incisions in a completely portal approach. In the event that the primary tumor is large the access incision may have to be enlarged to remove the specimen without inadvertently fracturing the ribs.

Termination of the Procedure

Once the specimen has been removed and the systematic lymphadenectomy or sampling has been completed the surgical arms can be undocked from the trocars, and the cart can be moved away from the patient. The trocars should be removed, and a single drainage chest tube placed through the anterior inferior camera incision and positioned with the tip at the apex of the chest posteriorly. The lung should be reinflated under direct thoracoscopic vision with the robotic scope placed in the access incision. The remaining wounds are closed in a standard fashion (Fig. 165-10).

Standard postoperative lobectomy management should be undertaken. In virtually all cases, patients should be extubated in the operating room and brought initially to the postanesthesia care unit. Chest tubes may be left to water seal unless the immediate postoperative radiograph demonstrates a large air space. Patient-controlled analgesia should be initiated through the use of an epidural or a peripheral narcotic combined with either intraoperative intercostal blocks or continuous subpleural local anesthetic infusion. Patients may be transferred to a surgical floor with telemetry. Early ambulation and chest physiotherapy is critical to prevent hypoventilatory atelectasis and pneumonitis. Removal of the chest tube should be performed once there is no evidence of air leak and fluid drainage is sufficiently diminished. Commonly, patients can be discharged once the chest tube is discontinued provided there are no other concomitant complications and pain control on oral medication is sufficient.

Potential intraoperative and postoperative complications are no different with a robotic approach to lobectomy than VATS lobectomy. Major perioperative morbidity and mortality are consistent with the largest and best series of VATS lobectomies.¹⁰ However, there are unique aspects to robotic thoracic procedures and lobectomy that need to be considered.

1. Lack of tactile feedback. The robotic arms do not impart haptic feedback to the console (operating) surgeon. Therefore, both the console surgeon and the bedside assistant must be constantly vigilant about inadvertent injury to either the external patient or surrounding internal structures by the instrument arms or the instruments themselves. Externally, care must be taken to insure that the arms do not compress any part of the patient with undue force for a prolonged period of time. Internally, the surgeon must pay close attention to the visual feedback to prevent direct traction injury and compression injury by the shaft or heel of the instruments to adjacent structures.

2. Visual magnification. The robotic thoracoscope by design has greater magnification compared with the optics of the conventional camera. This is quite beneficial when working in a narrow, confined space, but results in decreased overall perspective view. This can be a disadvantage in the chest when one is attempting to delineate anatomic boundaries, such as lobar versus segmental structures or location of the minor fissure. It is critical to zoom out to the farthest extent possible when a greater overall view is required.

3. Hemorrhage. While the threat of catastrophic hemorrhage during major pulmonary resection is not unique to robotic procedures, absence of the operating surgeon at the bedside for immediate intervention is a necessary condition of the current master–slave robotic system. There are several strategies to maximize safety and minimize patient morbidity in the event of significant vascular injury. First, anticipation of a potential injury and maintaining a low threshold for conversion is key. Second, there must be a sponge stick ready and available for the bedside assistant to use for temporary tamponade of a bleeding vessel. Third, the bedside staff must be poised for rapid instrument removal and undocking of the surgical cart in preparation for conversion. With proper training this can be executed in less than 1 minute, and use of robotic technology should never impede timely management of potentially catastrophic bleeding.

Robotic lobectomy is a feasible, safe, and oncologically sound surgical treatment for early-stage lung cancer. The technique is reproducible across multiple centers and yields results consistent with the best seen with conventional VATS. It should not be considered experimental, but an accepted minimally invasive thoracic surgical technique. Successful and safe implementation into clinical practice requires preparation and commitment on an institutional and multidisciplinary team level. The future directions for study of this technology include further refinement of the technique, validation of the adequacy of the oncologic results, and determining methods to compare it with conventional VATS and thoracotomy techniques.

The overall setup for robotic lobectomy and the unique considerations through the eye of an experienced “insider” make this chapter particularly valuable to surgeons interested in robotic thoracic surgery.