Bilateral Laparoscopic Postchemotherapy Retroperitoneal Lymph-Node Dissection in Nonseminomatous Germ Cell Tumors–a Comparison to Template Dissection

Introduction

T esticular cancer accounts for approximately 1.5% of all malignancies in men with 3–10 new diagnoses per 100,000 every year. ^1,2 The introduction of the platinum-based chemotherapy in advanced nonseminomatous germ cell tumor (NSGCT) improved the survival rates dramatically. ^3,4 Retroperitoneal lymph node dissection (RPLND) postchemotherapy is necessary whenever a residual mass is detected, indicating teratoma or even active tumor. However, imaging alone cannot distinguish between viable tumor, teratoma, and necrosis. Moreover, unresected teratoma or viable tumor will inevitably progress if untreated. Hence, the completeness of resection is required to provide an optimal clinical outcome. ⁵

Recently, it has been advocated that RPLND in stage II NSGCT might be safely restricted to a unilateral modified template in selected patients. Nevertheless, bilateral RPLND remains the standard procedure with regard to all oncological principles. ^6,7 Wood and colleagues showed that a contralateral tumor spread postchemotherapy occurs in nearly 8% in germ cell tumor patients. ⁸ Especially in high-volume residuals, the risk of contralateral spread is considerably elevated, and, therefore, a unilateral dissection is not recommended. ^9,10 However, in many cases, RPLND by laparotomy has been associated with a considerable morbidity and an overall complication rate of more than 20% mainly due to postoperative adhesions and fibrosis. Even in contemporary open series, peri- and postoperative morbidity is still an issue. A laparoscopic approach has been increasingly proposed to reduce the inherent morbidity of RPLND. Janetschek et al. showed that bilateral laparoscopic (L)-RPLND was superior to laparotomy with regard to overall complication rates and morbidity. ¹¹ Furthermore, recent studies have shown equal functional and oncological results of L-RPLND as compared with open series. ^{12 –14} However, most studies looking at L-RPLND mainly describe template dissection as being one of the most criticized points. Nevertheless, the question whether laparoscopic bilateral dissection was superior to unilateral dissection remains controversial. Critics have been pointing out that primary bilateral dissection could lead to more complications than a limited approach. Studies specifically addressing this issue are scarce. Therefore, we aim at comparing ipsiliateral L-RPLND after chemotherapy with bilateral L-RPLND in two matched cohorts of men with regard to operative morbidity and oncological outcome. Furthermore, we are highlighting a technical modification to prospectively control the great vessels, allowing for the resection of more complex residuals with limited vascular involvement.

Patients and Methods

We conducted a retrospective analysis, including only patients who had undergone platinum-based chemotherapy (2–4 cycles) and consecutive L-RPLND (n=39). Institutional review board approval was given in all cases (529/2012R). All patients with initial clinical stages of IIA and higher with a normalization of tumor markers after chemotherapy and retroperitoneal residuals >1 cm were considered for L-RPLND (Fig. 1A and B). Both seminomatous germ cell tumor (SGCT) and NSGCT were included, while SGCT patients only underwent surgery when there was a suspicion of vital tumor on imaging. Those patients with extremely high-volume disease and extensive vascular involvement potentially requiring vascular replacement underwent open RPLND. Patients with relapsing cancer or late recurrence were also not included here. Unilateral template resection had been done from 2002 until 2009. After 2009, a bilateral nerve-sparing approach was introduced. All cases after that were done bilaterally. All perioperative variables were analyzed, and all complications were reported using the standardized Clavien-grading. ¹⁵ Ejaculatory dysfunction was reported separately. Oncological parameters included the need for salvage treatment, rate of recurrence, and final histology. We compared men who underwent only template resection (A) with a group (B) that had primary bilateral L-RPLND ¹⁶ after chemotherapy in our institution.

OPEN IN VIEWER

FIG. 1.

(A) CT scan demonstrating inter-aortocaval mass; (B) CT scan showing para-aortic residual mass.

Technique

The surgical dissection contained the removal of the entire residual masses (Fig. 2A) and the resection of the respective fields according to the boundaries described by Weissbach and Boedefeld. ¹⁶ In bilateral dissection, the ipsilateral side was dissected as radical as necessary depending on tumor volume and its relation to the postganglionic fibers of the respective sympathetic trunk.

OPEN IN VIEWER

FIG. 2.

(A) Inter-aortocaval residual mass; (B) paracaval identification of the sympathetic trunk; (C) postganglionic fibers behind the Vena cava; (D) final view of the great vessels.

The surgical access was trans-peritoneal in all cases. In template resection, we used three trocars in left-sided and four trocars in right-sided tumors using a respective flank position. Trocars were positioned in form of a triangle: the first trocar near to the navel, the second one near to the rectus abdominis muscle, and the third one nearly 4–5 cm above the navel. When necessary, one 5 mm trocar 1 cm was placed below the xiphoid for liver retraction. ³

In primary bilateral L-RPLND, a total of five trocars is needed using an identical placement as described earlier. As in the unilateral dissection, a full-flank position is required, starting with right-sided dissection. Then, the trocars are removed, and the incisions are protected with a temporary dressing. Subsequently, the patient is fully repositioned and re-draped to perform left-sided dissection.

Right L-RPLND

Right L-RPLND in tumors located on the right side, the ipsilateral resection contained the para-caval as well as the inter-aortocaval area between the hilum of the right kidney and the right iliacal vessels and furthermore, the resection of tissue localized anterior to the aorta and cranially to the inferior mesenteric artery.

After incision of the line of Toldt, colon and duodenum are reflected medially until the retroperitoneum is exposed (Kocher Maneuver). After the preparation of the spermatic, lumbar and renal vessels, and the ureter, the sympathetic trunk is prospectively identified (Fig. 2B). The dissection of the postganglionic nerves follows (Fig. 2C). Then, the lymphatic tissue on the inferior Vena cava (IVC) is dissected from the IVC after control of lumbar vessels by clipping. Afterward, postganglionic nerves are visualized and preserved. After clipping and transecting the lymph node packet en-bloc, the inter-aortocaval lymphnodes are transected. By rotation of the IVC, nerve sparing becomes easily feasible. The inter-aorto-caval lymph nodes are then removed until the aortic bifurcation. Therefore, a complete retroperitoneal resection means dissecting the entire tissue from the left renal vein to the inferior mesenteric artery. ^17,18

Left L-RPLND

The retroperitoneal space should be exposed by a peritoneal incision along the descending colon until the internal inguinal ring. After mobilization of the descending colon, the aorta can be identified. The preparation (and resection) of the spermatic vein, the left renal vein, the renal artery, and the ureter (lateral boundary) follows. Before dissecting the lymphatic tissue between the nerves, the sympathetic trunk and postganglionic nerves should be identified. The lymphatic tissue laterally of the aorta is removed and, finally, also the pre-aortic lymphatic tissue from the left renal vein to the inferior mesenteric artery ¹⁷ (Fig. 2D).

Prospective control of the great vessels

In case of complex retroperitoneal tumors with vascular involvement, both the aorta and the Vena cava are secured and snared with vessel loops below and above the residual mass. During the resection of tumors, adhering to the Vena cava cross-clamping using laparoscopic Bulldog-Clamps (Aesculap) was applied when required. The same maneuver was used to control more complex bleeding in both the aorta and the vena cava whenever needed (Fig. 3A). Suturing can then be done in a bloodless field, allowing for accurate stitches (Fig. 3B). In addition to a conventional set of laparoscopic instruments and a 30° telescope, we used a vessel-sealing device (BiCision; ERBE) to control both minor blood vessels and lymphatics. Larger structures and lympthatics on the boundaries of dissection were clipped (Hem-o-lok) to reduce lymphorrhea.

OPEN IN VIEWER

FIG. 3.

(A) Cross-clamping of the Vena cava using Bulldog-clamps; (B) laparoscopic suturing in a bloodless field.

Results

We completed 39 L-RPLNDs without any conversion to an open intervention. Early re-interventions were not necessary. Mean age was 31.2 years (18–62). Interestingly, more patients of the bilateral group had a clinical stage II (13 versus 20, p=< 0.001); whereas more patients of the unilateral group had a clinical stadium III (6 versus 2, p=0.0357). Operative time ranged from 125 to 370 minutes (mean 221) in unilateral and 186 to 397 minutes (mean 270) in bilateral dissection (p=0.12). No peri-operative blood substitutions were necessary in both groups. There were no deviations from the normal intra and- postoperative course in 36 cases. There was one Grade II complication (bleomycin-induced pneumonitis) in group A and 1 grade III complication (chylous ascites) in group B. One patient in group A (mean age 30.8 years; 20–49) had bleomycin-induced pneumonitis with a prolonged hospitalization (14 days) but without any persistent pulmonary dysfunction. Chylous ascites (Grade III) was drained via a single puncture and then followed conservatively until it had completely resolved. In one patient during the bilateral L-RPLND, a partial nephrectomy with pedicle clamping was performed because of a renal tumor suspected to be a residual mass. There were no differences regarding hospitalization time between both groups (p=0.1). In group A, the histological examination revealed mostly teratoma (Table 1). We found significantly more patients with embryonal carcinoma in the initial histology in the unilateral L-RPLND group than in the bilateral group (17 versus 8, p=0.0022). In two men with viable tumor, a consecutive high-dose chemotherapy was necessary. In one man, an enhanced lymph node had been detected after 12 weeks. Therefore, a second-look laparoscopy was performed with necrosis in the histological report. In group B (mean age 31.5 years; 18–62), in 11 patients, necrosis was found in the final histological examination, 7 patients had teratoma, and one had active disease (Table 1). The viable tumor was found bilaterally. In one patient, no tumor was found. There were no symptomatic lymphoceles or cases of lower-limb lymphostasis. A normal and unrestricted ejaculatory function could be maintained in all men with nerve sparing in both groups after the full observation period. Two men in group B reported retrograde ejaculation. These patients did not have full contralateral nerve sparing because of the location and volume of the residual mass (p 0.4872). The median follow up in group A was 24 months (Range 4–38) and 13 months (Range 3–37) in group B. In group A, there were two recurrences: After 11 months, one patient with viable tumor developed a relapse and another did so 3 months after L-RPLND. The latter died after having received salvage high-dose chemotherapy. Hence, in-field recurrence was noted in 2 of 19 patients in group A. Furthermore, three patients of group A had residuals in their lungs <1 cm. Nevertheless, a tumor excision was not performed, because the final histological report revealed necrosis. In group B, one patient required salvage treatment with high-dose chemotherapy and autologous stem-cell transplantation as well as a resection of outside field suprahilar residuals because of a recurrence 4 months after the L-RPLND. In one patient, a whole brain irradiation was performed due to brain metastases. There were no in-field recurrences in group B. However, with regard to the overall rate of disease, recurrence was not significant (HR=1.84; 95% CI 0.17–39.92; p-value=0.6109).

Discussion

Due to the advent of a multi-modal therapy, excellent cure rates can be achieved in testicular germ cell cancer. The mainstays of this success are early diagnosis, careful staging, platinum-based chemotherapy, radiotherapy, and, ultimately, retroperitoneal surgery. There is a common consensus that successful chemotherapy for advanced disease should be consolidated by RPLND and a resection of residual tumors exceeding a size of 1 cm. Surgery helps in identifying viable tumor and in removing teratoma that is not susceptible to chemotherapy. However, the Indiana group showed a total complication rate of 20.7% in patients who underwent a RPLND after chemotherapy. The overall mortality rate was 0.8%. ¹⁹ Postoperative adverse events such as chylous ascites and lymphoceles emerged in 2% and 1.6%. ²⁰ In the last few decades, the laparoscopic approach has been established to reduce morbidity, and more and more data were published about the superiority of L-RPLND in selected patients in whom the approach was performed by a center with high experience. It is well known that L-RPLND minimizes blood loss and hospitalization. Furthermore, convalescence is faster, and the cosmetic results are more appealing. Another important aspect is also lower postoperative pain compared with open surgery. ¹⁸

However, the extent of node dissection in this setting needs to be discussed. Template dissection is a compromise between morbidity and oncological efficacy, bilateral L-RPLND might be associated with elevated short- and long-term morbidity.

Janetschek argued that in clinical stage IIB and IIC testis cancer unilateral resection of residuals is not recommended because of an elevated risk of a possible contralateral tumor spread, ⁹ thus favoring bilateral dissection. Wood and colleagues showed that a contralateral tumor spreading postchemotherapy occurs in nearly 8% in germ cell tumor patients. ⁸ In our series of primary bilateral L-RPLND, we had one patient with bilaterally located active residuals. This underlines the oncological significance and the necessity of bilateral RPLND. In addition to that, we had two cases of in-field recurrence in the unilateral group but none after bilateral dissection. Furthermore, others argue that bilateral RPLND of any residual mass larger than 1 cm for NSGCT is the standard of care. ⁶ Heidenreich et al. also reported that bilateral RPLND should be standard in high-volume residual disease. ⁷ Underwood and Kim underlined the applicability of the bilateral approach in clinical stage I and II germ cell tumors. ¹³

The modified template resection may minimize retrograde ejaculatory rates with the disadvantage of an increased risk of local recurrence. In this context, Stephenson and Klein also favor a nerve-sparing bilateral RPLND in low-stage NSGCT. ²¹ The loss of antegrade ejaculation is a commonly reported long-term complication. ²² Nerve-sparing or template dissection may reduce the risk of ejaculatory dysfunction. In terms of the elevated risk of ejaculatory dysfunction, we could not report any ejaculatory problems in patients with a full nerve-sparing L-RPLND. Complex retroperitoneal tumors adhering to the Vena cava can be dissected by using laparoscopic Bulldog-Clamps for prospective vascular control. In case of complex bleeding in both the aorta and the Vena cava, this technique can be useful to successfully complete L-RPLND. To prevent and reduce other complications such as the incidence of chylous ascites and lymphoceles, we did not use drains to avoid lymphatic fistula formation. In addition, Hem-o-lok clips can be used to reduce lymphorrhea. Oral fluid intake started immediately after surgery, while food intake started only after appropriate bowel movements. All men were put on an MCT-diet (medium chain triglyceride) to reduce lymphorrhea and to prevent chylous ascites. The MCT-diet had to be maintained for another 3 weeks after discharge from the hospital. In the conducted study, we have seen ascites in one case that resolved spontaneously after a single puncture. Postchemotherapy primary bilateral L-RPLND in stage II and III testicular cancer patients with full contralateral nerve-sparing is feasible with a similar complication rate but with additional oncological safety, respecting the known data about the risk of a contralateral spread especially in clinical stage IIC patients. In our study, we had a relatively high number of IIC tumors in the bilateral group (7/19). Even these complex cases can be successfully cleared using L-RPLND with prospective vascular control. Importantly, we showed that the known increased oncological efficacy of bilateral L-RPLND is not leading to an increase in both short- and long-term morbidity. Due to the short follow-up, the durability of the presented data is limited. To confirm the long-term oncological superiority of bilateral L-RPLND, a longer follow-up is required. Other study limitations that have to be acknowledged are the retrospective nature and the small sample size. Bilateral L-RPLND represents an additional step in developing laparoscopic techniques for the management of NSGCT. Due to the relative complexity of L-RPLND in the postchemotherapy setting, the procedure remains limited to institutions with extensive laparoscopic experience.

Conclusions

Bilateral L-RPLND with full contralateral nerve sparing is feasible and apparently has a similar morbidity as compared with template dissection. The potential risk of contralateral spread, especially in more extensive retroperitoneal disease, is specifically addressed using a primary bilateral approach. Hence, the risk of retroperitoneal recurrence may be reduced by primary bilateral dissection. Additional studies with extended follow-up are necessary to confirm our findings and to investigate a potentially superior oncological safety.