European Urology

Introduction

Lymph node dissection (LND) plays a critical role in all genitourinary tumors. Studies of prognostic factors in patients with muscle-invasive bladder cancer show the histologic status of regional lymph nodes as one of the most important predictors of survival [1]. Many advocate LND as an integral part of the curative intent of radical surgical therapy for invasive carcinoma of the bladder and not merely as a staging procedure [2], [3], and [4]. This strategy is reflected in several recent studies that have also suggested that the extent and completeness of an LND may improve the survival of patients with bladder cancer [3], [5], and [6]. Some advocate a more extended LND, which must include all lymph nodes in the boundaries of the aortic bifurcation and common iliac vessels (proximally), the genitofemoral nerve (laterally), the circumflex iliac vein and lymph node of Cloquet (distally), the hypogastric vessels (posteriorly) including the obturator fossa, the lymph nodes around and medial to the internal iliac artery, and the presacral lymph nodes bilaterally [3], [6], and [7]. An extended dissection may also reach superiorly to the level of the inferior mesenteric artery (IMA) [8], and [9]. Others advocate a more limited “standard LND,” in which the cephalad extent begins at the level of the common iliac bifurcation. The lateral and distal limits are similar to the extended dissection [10], [11], and [12]. Although data show that an extended LND is critical in the treatment of high-grade, invasive bladder cancer, no uniform consensus exists regarding the limits and boundaries of LND or the minimum number of lymph nodes that should be removed [3], [10], and [13]. Despite the previously mentioned argument, there is uniform agreement that when performing laparoscopic radical cystectomy (LRC) with a curative intent, it must be accompanied with adequate laparoscopic LND that mirrors open surgically established anatomical boundaries. A technically demanding part of LRC is performing an adequate laparoscopic LND. This challenge is most pronounced in the area of the common iliac vessels, the aortic bifurcation, and up to the IMA, where good exposure is difficult to achieve. In this paper, we present our technique of optimal exposure for LND during LRC.

Materials and methods

The first 10 patients in which LRC and LND were performed using our approach were evaluated. The extended template of LND included all lymph nodes in the boundaries of the IMA and aortic bifurcation (proximally), the genitofemoral nerve (laterally), the circumflex iliac vein and lymph node of Cloquet (distally), the hypogastric vessels (posteriorly) including the obturator fossa, the lymph nodes around and medial to the internal iliac artery, and the presacral lymph nodes bilaterally (Fig. 1).

Create Platinum Slide Series

Fig. 1 The field of dissection: region 1, paracaval; region 2, interaortocaval; region 3, para-aortic; region 4, common iliac; region 5, intercommon iliac (presacral); region 6, external iliac; region 7, obturator; region 8, internal iliac; region 9, internal iliac branches (medial to the internal iliac).

Description of the technique

Trocar arrangement

A modification of the classic five-trocar arrangement for laparoscopic radical pelvic surgery is used, with the optic and the right-lateral trocars repositioned more cephalically (Fig. 2). The optic trocar is introduced in the midline two finger widths above the umbilicus. The supraumbilical camera position is essential for adequate visualization of the upper limit of the LND; this is aided by a 30° lens that provides an adequate angle as well as a wider range of vision. The right-lateral trocar is placed also two finger widths above and medially to its original position for use by the surgeon's dominant hand; this facilitates performance of the high-extended LND up to the IMA.

Create Platinum Slide Series

Fig. 2 Trocar arrangement: (A) The classical five-trocar arrangement for radical pelvic surgery; (B) our modified trocar arrangement. The 11-mm optic trocar is repositioned supraumbilically and at the midline; the right lateral 5-mm trocar is also repositioned supraumbilically and medially.

Laparoscopic radical cystectomy

Surgery begins with the cystectomy. A classic LRC is performed with dissection and transection of the distal ends of both ureters between clips; development of the posterior and lateral planes; ligation and division of the exposed postrolateral pedicles of the bladder; development of the anterior plane; and, finally, division of the urethra after securing it. The specimen is freed and placed in an endobag for extraction through the incision required for the extracorporeal urinary diversion. We prefer to perform the LND after the bladder is removed, rendering the pelvis empty and the landmarks readily visible. Additionally, cephalad dissection along the proximal common iliac artery is facilitated by the fact that the transected ureter has already been removed from the operative field.

Mobilization of the sigmoid colon

The essential step for optimal exposure is mobilization of the sigmoid colon. This step is often performed during dissection of the ureters, as the plane for colon mobilization can be readily identified. Mobilization begins with dissection of the lateral attachments of the sigmoid colon from the neighboring ureter deep down into the pelvis. The posterior attachments (mesocolon) of the colon are then freed beginning at the sacral promontory and continuing caudally down to the rectosigmoid junction (Fig. 3). This plane opens easily, as long as dissection remains close to the colon. This dissection is an extension of the window developed in the sigmoid mesocolon for passage of the left ureter to the right side during the urinary diversion. A similar dissection of the sigmoid colon is duplicated on the right side. Exposure of the anterior aorta and mobilization of the sigmoid colon are then continued until the origin of the IMA, which is considered the upper limit of the LND.

Create Platinum Slide Series

Fig. 3 Mobilization of the sigmoid colon: (A) schematic diagram; (B) endoscopic view. U = ureter; C = colon.

Retraction of the mobilized colon

To aid in exposure during LND, a braided umbilical tape (4 mm × 61 cm [BARD, Tempe, AZ, USA]) is passed across the undersurface of the mobilized colon at the level of the sacral promontory and its free ends are delivered out through the left-lateral trocar. Using a trocar obturator, the trocar is removed and the tape is repositioned outside the trocar lumen. The trocar is then reintroduced with the tape now passing alongside the trocar, without hindering further use of the trocar (Fig. 4). Traction on the tape with external fixation allows retraction of the mobilized colon toward the left side and exposure of the lower aorta and inferior vena cava, the left and right common iliac vessels, and the complete right pelvic side wall.

Create Platinum Slide Series

Fig. 4 Retraction of the mobilized colon using an umbilical tape: (A) passing the tape across the undersurface of the mobilized colon; (B) delivery of both ends of the tape out through the trocar; (C) using a trocar obturator to deliver the tape outside the trocar lumen; (D) final position of the tape beside the trocar.

Lymph node dissection

Dissection starts with exposure of the IMA (proximal boundary of the template). The paracaval, interaortocaval, and para-aortic lymph nodes are dissected, followed by lymph nodes along the right common iliac artery. The lymph nodes overlying the external iliac vessels are then dissected, starting at the lateral border of dissection (the genitofemoral nerve) and continuing distally up to the lymph node of Cloquet and the medial surface of the external iliac vein. Following skeletonization of the external iliac vessels circumferentially, their medial retraction exposes the obturator nerve at its exit from the medial surface of the psoas muscle (posterior boundary of the template). The vein is then retracted laterally to dissect out the obturator group of nodes. Lymph nodes along the right hypogastric vessels (medial boundary of the template) are dissected down to the clipped vesical pedicle. Finally, the subobturator and presacral (posterior boundary of the template) lymph packages are dissected. A similar dissection is done on the left side down to the iliac bifurcation, as the position of the retracted colon to the left side hinders further dissection. Each lymphatic package is removed separately within a specimen bag through the midline 12-mm port. Dissection of the previously mentioned regions is performed from the right side with the colon retracted to the left, using the right- and left-lateral trocars (Fig. 5). Exposure of the remaining left template is accomplished by reintroduction of the free ends of the umbilical tape into the abdomen and alongside the most right-lateral trocar to retract the colon toward the right. Subsequent dissection of the remainder of the left template is accessible through the left-lateral and midline trocars. A complete dissection within an extended template can be accomplished using this simple and reproducible technique of optimal exposure for LND following LRC (Fig. 6).

Create Platinum Slide Series

Fig. 5 Left retraction of the mobilized colon exposes the complete right side as well as the left side down to the left iliac bifurcation, which is dissected from right to left: (A) schematic diagram; (B) endoscopic view. The dissected inferior mesenteric artery lies at the cranial border of the template.

Create Platinum Slide Series

Fig. 6 Endoscopic view after completing the extended lymph node dissection.

Results

From August 2006 to September 2008, 10 laparoscopic cystectomies with an extended LND and extracorperal diversion were performed at our institution. Average patient age was 63.5 yr at an average body mass index (BMI) of 25.3 (range: 21.8–29.1). Patient characteristics are detailed in Table 1. Mean total operative time was 512.5 min (range: 420–660), with a mean operative time of 143 min (range: 115–165) for the extended LND, including the time needed for mobilization and retraction of the sigmoid colon. Adequate exposure was successful in all patients. Mean drop in postoperative hemoglobin was 1.82 mg%. The average number of lymph nodes examined was 25.5 (range: 19–32). Table 2 lists lymph node distribution. Overall, four lymph nodes were positive for metastatic disease, with no extracapsular extension. No positive margins were detected in any specimens examined. Table 3 lists the operative and pathologic outcomes. At a mean follow-up of 14.8 mo (range: 4–30), no patient had local pelvic or lymph node recurrences or recurrences at the site of lymph node extraction, including the patient with positive nodes. In four patients (40%), distant metastasis was detected at an average of 8.5 mo (range: 2–27); all four patients received adjuvant chemotherapy (Gemsar combination). Two patients died of metastatic disease, at 14 mo and 25 mo postoperatively, without any evidence of pelvic or lymph node recurrence. Four patients (40%) had minor complications in the form of pelvic lympoceles (three patients) or prolonged bowel ileus (one patient), and all were successfully managed conservatively. One patient developed an intestinal fistula, detected at 3 wk postoperatively. During surgical exploration, the fistula was detected at the terminal ileum, which was probably due to inadvertent bowel injury during dissection of the severe bowel adhesions, as a result of a previous laparoscopic nephroureterectomy with prolonged bladder leakage. Another patient developed stenosis of the left ureterointestinal anastomosis at 28 d postoperatively; the patient was treated with a left nephrostomy and refused reintervention.

Discussion

The surgical treatment of patients with invasive bladder cancer is based on wide excision of all localized, perivesical, and regional disease. LND was included as part of the radical cystectomy procedure to enhance local regional disease control and possibly to improve overall survival. In addition to the potential therapeutic benefit, pathologic evaluation of the regional lymphatics increases the accuracy of surgical staging and allows for the identification of those patients who are in the greatest need of adjuvant chemotherapy. Lymph node metastases are found in 20–25% of patients who undergo pelvic LND during radical cystoprostatectomy for bladder cancer [2], [3], [4], [10], [13], and [14], which is associated with an increased risk of local recurrence and disease progression. There is complete accord that any radical cystectomy should be accompanied by an adequate LND; however, the anatomic extent of dissection remains controversial. While the medial, lateral, and distal borders have remained unchanged, the proximal border has varied from the bifurcation of the common iliac artery up to the distal aorta. Early mapping studies showed that the incidence of involvement of obturator, external iliac, common iliac, hypogastric, and perivesical nodes was 74%, 65%, 19%, 17%, and 16%, respectively [15]. Other studies show that primary lymph node metastases exist beyond the limits of a standard LND. The specific distribution of nodal metastases was prospectively evaluated in a study by Steven and Poulsen [16]. They found that 34% of their patients with lymph node metastases had nodal involvement in the periaortic, common iliac, and presacral regions and that if dissection had been restricted to the standard template, 34% of the lymph node metastases would have been left behind and 6.25% of cases would also have been misclassified as lymph node metastases negative. The importance of an extended LND was also corroborated in a recent stage-specific lymph node metastasis mapping study was reported by Vazina and colleagues [7]. A total of 176 patients underwent an extended LND with radical cystectomy; 43 patients (24.4%) had pathologic lymph node involvement. Although the most common sites of nodal metastases were the external iliac and hypogastric/obturator regions, 5.1% of patients had presacral nodal involvement and 9% had disease above the common iliac bifurcation. Collectively, these studies support the application of an extended LND with a cephalad extent of dissection that includes the lymph nodes around the common iliac vessels and bifurcation of the aorta, including the presacral nodal packet. Some authors even recommend removal of distal para-aortic and paracaval lymph nodes [8], [9], and [17].

The oncologic result of LRC should not be compromised by an inadequate LND that does not fulfill the present standard of open LND. The problem during laparoscopic LND is not the task of removing the lymph nodes within an extended template, once accessible, but rather the difficulty in achieving this access. This is most evident at the cranial border of the template. Several factors contribute to this problem: The rectosigmoid colon and, to some extent, the small bowel hinder the approach to the lower aorta, aortic bifurcation, and common iliac vessels. Furthermore, a 0° optic positioned at or below the level of the umbilicus would have to be directed cranially toward the surgeon for visualization of the cranial border of the template (Fig. 7). This severely compromises hand–eye coordination, which is essential during LND in the typical cranial-to-caudal direction. The ergonomics of movement would also be disrupted if the instruments of dissection were also directed toward the surgeon, with the dominant arm at maximal extension. These obstacles make the simplest surgical tasks at the area of the cranial nodes almost impossible, due to the standard position of the optic trocar and a working trocar at or below the level of the umbilicus. Our approach overcomes all of the aforementioned problems. First, the optic trocar is positioned two finger widths above the umbilicus and, together with a 30° lens, allows visualization of the upper border of the template with the optic maintained in its normal direction, preventing any compromise in coordination (Fig. 8). Second, the most right-lateral trocar is repositioned more cephalically and medially. This trocar, which is used by the assistant throughout most of the procedure, is reserved for use by the surgeon's dominant hand during LND. Finally, retraction of the mobilized colon to the left provides access to the complete right side as well as to the left side down to the left iliac bifurcation, which is dissected from right to left. The remainder of the left template is exposed by retraction of the colon in a similar manner toward the right side. The rationale behind developing this approach was to ensure a complete and thorough LND within an extended template, especially at the cephalad boundary; saving valuable intraoperative time was not our main focus. The time needed for manipulation of the sigmoid colon was 10–15 min in all cases. Many surgeons advocate performing the LND prior to the cystectomy to better identify and secure the vesical pedicles. We delay the LND until after the LRC but identify the vesical pedicles during dissection of the ureters. Extending the peritoneal incision for identification of the ureters from a point lateral to the medial umbilical ligament to the crossing of the ureter with the iliac bifurcation aids in exposure of the internal iliac artery down to the vesical pedicles, which are easily secured following development of the lateral and posterior bladder planes.

Create Platinum Slide Series

Fig. 7 Schematic diagram of the range of vision at the cranial border offered by a 0° telescope at or below the umbilicus. Notice the unusual direction of the optic.

Create Platinum Slide Series

Fig. 8 Schematic diagram of the range of vision at the cranial border offered by a 30° telescope above the umbilicus with the optic maintained in a normal direction.

A limitation of this study was the small sample size, with a limited number of patients with positive lymph nodes. One of the primary end points was to achieve adequate exposure, assessed by exposing the larger part of the template without additional bowel retraction and bowel interference during the LND. Although this finding is subjective, adequate exposure is a parameter that can only be assessed subjectively by the surgeon intraoperatively. Furthermore, the completeness of our LND was supplemented with additional data (intraoperative pictures, number of lymph nodes retrieved, and absence of early lymph node recurrence). In our series, an adequate exposure for LND was feasible in all of our patients. Although we did not encounter patients with a high BMI, we propose the addition of a second umbilical tape to further retract the colon though a left-cephalad trocar at the level of the umbilicus in such patients. This would further lighten the burden of a bulky sigmoid during LND at the level of the common iliac vessels, aortic bifurcation, and proximal aorta bilaterally, as dissection of these areas is performed exclusively from the right side.

Conclusions

The aim of this publication is to present a simple and reproducible technique for adequate exposure of the largest template during LND following LRC. Although controversy remains as to the exact extent of LND, this technique would provide adequate exposure not only to the accessible pelvic lymph nodes but also to the cephalad extent of LND, whether it be the common iliac or aortic bifurcation, or even up to the IMA.