Hysterectomy With Pelvic and Paraaortic Lymphadenectomy

Therapeutic Value of Lymphadenectomy

The value of nodal assessment to predict patterns of failure and prognosis has never been questioned. On the other hand, with regard to possible therapeutic implications, several nonrandomized studies have suggested that systematic pelvic and paraaortic lymphadenectomy may provide a survival benefit in patients at high risk of nodal metastasis compared with no lymphadenectomy. However, the findings of two prospective randomized trials published in 2009 (the ASTEC trial and the LINCE trial) negated the findings of previous retrospective series. In the ASTEC trial, 1400 patients who underwent surgery for apparent early-stage endometrial cancer (stage I) were randomized to pelvic lymphadenectomy or no lymphadenectomy. The same allocation had been used in the Italian LINCE trial conducted by Benedetti Panici and colleagues, in which 540 women were randomized 1:1 for pelvic lymphadenectomy versus no lymphadenectomy, with 30% of patients in the lymphadenectomy arm receiving additional paraaortic dissection. However, major criticisms have been made regarding the study designs of both trials, particularly because the number of lymph nodes removed in the lymphadenectomy arm was relatively low and because the decision regarding adjuvant treatment according to lymph node status was not part of the protocols. Moreover, both trials shared the bias of not effectively selecting patients at high risk for nodal metastasis (the ASTEC trial included all patients with endometrial cancer without exclusion of early-stage and low-risk disease). Also, the low percentage of patients with positive nodes in the two studies (9% in the ASTEC trial, 13% in the LINCE trial) had the effect of inevitably diluting the possible (if any) therapeutic benefit of lymphadenectomy.

A large retrospective trial (the SEPAL study [Survival Effect of Para-aortic Lymphadenectomy in Endometrial Cancer]) compared two different approaches—systematic pelvic versus systematic pelvic and paraaortic lymphadenectomy—and demonstrated a survival advantage in patients who underwent nodal dissection in both areas. This advantage was even more pronounced when the analysis was limited to patients with intermediate- or high-risk disease. However, even if the discrepancy in adjuvant treatment between patients who underwent pelvic and paraaortic lymphadenectomy (77% received adjuvant chemotherapy) and patients who underwent pelvic lymphadenectomy alone (45%) could represent an important bias of the study, a subsequent analysis performed in the same groups of patients focusing on the initial failure site demonstrated a higher recurrence rate for women who underwent pelvic without paraaortic lymphadenectomy (9.5% vs. 1.3%). Despite the lack of survival benefit reported in the two prospective trials, lymph nodal evaluation remains crucial in patients with intermediate- and high-risk disease to identify patients in whom adjuvant therapy may be avoided. This will ultimately prevent unnecessary overtreatment and inappropriate undertreatment.

Sentinel Nodes in Endometrial Cancer

In recent years, the possibility of performing sentinel node detection (SND) in patients diagnosed with endometrial cancer has been proposed. The theoretical advantages of this approach include the possibility of assessing nodal status with minimal or limited dissection of the retroperitoneal node-bearing areas. It appears evident that the omission of full lymphadenectomy not only reduces the overall operative time and surgical difficulty but also has the potential to minimize the rate of both intraoperative and postoperative lymphadenectomy-related complications. If proven reliable, SND may be proposed as a balanced compromise between no nodal dissection and full, systematic, extended, and time-consuming nodal dissections.

Historically, sentinel node identification has been proposed and validated in the management of melanoma, breast cancer, and vulvar cancer. For more than a decade, researchers have investigated the role of SND in endometrial cancer, but a decisive increase of the application of this technique to uterine malignancies has been achieved only recently with the use of indocyanine green, a fluorescent tracer that can be easily detected during a minimally invasive surgical procedure (whether laparoscopic or robotic) by using endoscopic cameras with a near infrared high-intensity light source for detecting fluorescence.

Indocyanine green has been shown to be superior to the two traditional tracers (i.e., technetium 99 and blue dye) in terms of both detection rate and bilateral optimal mapping. Three possible sites of dye injection have been proposed: cervical, peritumoral (under hysteroscopic guidance), and subserosal. At present, cervical injection is the most commonly used approach because of its intrinsic simplicity and higher detection rate compared with the other methods. However, some doubts have been raised about the ability of cervical injection to permit reliable assessment of the true lymphatic drainage of endometrial malignancies. It is hypothesized that tracer injected into the cervix may migrate, following the drainage of uterine venous vessels through the parametrium to the pelvic nodal basins. However, uterine lymphatic drainage is complex and also involves migration through the infundibulopelvic ligaments up to the high paraaortic nodes. The low rate of detection of sentinel nodes in the paraaortic area (approximately 5%) when the tracer is injected into the cervix seems to confirm these doubts. Nevertheless, results of the application of sentinel node techniques in endometrial cancer appear encouraging, with detection rates ranging from 60% to 100% and false-negative rates of less than 5% if strict protocols are followed.

A recent retrospective study combining data from Memorial Sloan Kettering Cancer Center and the Mayo Clinic reported the results of two different approaches to nodal assessment in low-risk endometrial cancer patients (low risk was defined as endometrioid histology and myometrial invasion <50%). At Memorial Sloan Kettering Cancer Center, an algorithm involving sentinel lymph node removal alone was followed from 2006 to 2013. Conversely, at the Mayo Clinic, systematic pelvic and paraaortic lymphadenectomy was routinely performed from 2004 to 2008 in a similar population of patients. The results of the study are of value and show that, despite a lower number of lymph nodes removed when the sentinel lymph node–only algorithm was followed, the overall percentage of positive nodes was higher at Memorial Sloan Kettering Cancer Center than at the Mayo Clinic (5.1% vs 2.6%; P = .03), probably because of ultrastaging techniques applied to sentinel nodes. Of interest, the 3-year disease-free overall survival rate was similar in the two institutions ( Fig. 9.1 ).

At present, there are no definitive data regarding the superiority of SND over complete nodal assessment in endometrial cancer patients. In particular, several areas require additional research and clarification:

• The role of SND in high-risk endometrial cancer patients.

• The impact of SND on adequately targeting postoperative treatment and in decreasing overall morbidity.

• The impact of SND on detection of micrometastases (0.2–2 mm) or isolated tumor cells (<0.2 mm), which are not usually discovered with traditional pathology protocols applied to systematic lymphadenectomy.

• Whether the increased diagnostic ability (and the parallel increase in patients undergoing adjuvant treatment) may mitigate at least in part the advantage provided by SND in terms of reducing unnecessary surgical morbidity.

• The role of SND without associated lymphadenectomy in patients with positive nodes; whether extensive removal of pelvic and paraaortic nodes in women with lymphatic spread of disease may improve survival is still a subject of debate.

Indications for Lymphadenectomy

Lymphatic spread of endometrial cancer is unlikely in certain patients with small primary tumors, superficial or absent myometrial invasion, and low-grade disease. As a consequence, it has been estimated that about 30% of affected women (deemed not at risk) have almost no risk of lymphatic spread. This allows the avoidance of lymphadenectomy in almost one-third of patients, without consequences on adjuvant treatment decision making and survival ( Fig. 9.2 ).

Recent studies have assessed the risk of pelvic and paraaortic nodal metastases among patients considered at risk (i.e., excluding women with the characteristics shown in Fig. 9.2 ). Results of these reports are summarized in Table 9.1 .

As shown in Table 9.1 , the risk of paraaortic metastasis in endometrial cancer patients with myometrial invasion of less than 50% is low (2.2%). On the other hand, when myometrial invasion is greater than 50%, the risk of nodal metastasis is high, including paraaortic dissemination, particularly in grade 3 endometrioid tumors. It is well known that paraaortic lymphadenectomy up to the level of the renal vessels is potentially associated with significant intraoperative and postoperative morbidity. Given that paraaortic dissemination of endometrioid endometrial cancer is uncommon, a study from the Mayo Clinic has aimed to identify a subgroup of women at low risk of metastases in this area and in whom paraaortic lymphadenectomy may be safely avoided. The main findings of this study are summarized in Table 9.2 .

It is interesting to note that, in general, less than 4% of women with endometrioid subtypes have metastasis or failure at the level of the paraaortic nodes. Patients who may safely forego paraaortic lymphadenectomy include those with endometrioid histology, no or superficial myometrial invasion, negative pelvic nodes, and no lymphovascular space involvement.

Complications of Pelvic and Paraaortic Lymphadenectomy

The exact role of lymphadenectomy in endometrial cancer cannot be accurately defined outside of the assessment and complications associated with the procedure. Historical series of open abdominal surgical procedures have shown that the performance of pelvic and paraaortic lymphadenectomy is associated with an increased incidence of vascular injuries, deep vein thrombosis, lymphocysts, and death from pulmonary embolism. A previous study analyzed the acute morbidity and mortality of 812 patients who underwent surgery for endometrial cancer, finding a higher incidence of vascular lacerations, deep vein thrombosis, and pelvic lymphocysts in women who underwent lymphadenectomy compared with those who underwent only hysterectomy and bilateral salpingo-oophorectomy. Nine (1.1%) of the 812 patients included in the study died after surgery; eight of the nine were in the group of patients who underwent lymphadenectomy. Of note, five of these eight postoperative deaths were associated with thromboembolism. The performance of lymphadenectomy was the most important independent variable associated with postoperative deep vein thrombosis and pulmonary embolism.

In 2001, Franchi and colleagues published a detailed analysis of postoperative complications after pelvic lymphadenectomy performed by means of laparotomy for the surgical staging of endometrial cancer. Results of this study are summarized in Table 9.3 .

Also in the study by Franchi and colleagues, lymphadenectomy was associated with a higher incidence of postoperative complications. It must be emphasized that the removal of more than 14 pelvic nodes was the only factor independently associated with the development of at least one complication and that the removal of more than 19 lymph nodes was the only factor independently associated with the development of two postoperative complications.

The advent of minimally invasive procedures has dramatically reduced morbidity. Two landmark studies on a large number of patients who underwent extensive laparoscopic lymphadenectomy were published in 2004 to 2006. The first study included 650 patients operated on by Schneider and Kohler’s group in Jena, Germany, and the second study included 1000 patients of Querleu’s group in Lille and Toulouse, France. Both studies demonstrated that, in referral centers, laparoscopic lymphadenectomy is a safe procedure and that it is associated with a high nodal yield and a very low rate of intraoperative and postoperative complications.

The rate of major vascular complications was 1.1% in the study by Kohler and colleagues, with a slightly higher prevalence of venous (57.1%) over arterial injuries (42.9%). Of note, four of the seven vascular complications observed were resolved laparoscopically with no need for conversion to an open procedure. In the report by Querleu and colleagues, the rate of intraoperative and early postoperative complications and the percentage of patients in whom lymphocyst formation occurred were 2.0%, 2.9%, and 7.1%, respectively. Of the 20 intraoperative complications, 11 (1.1% of the entire cohort) were vascular injuries (4 during pelvic lymphadenectomy and 7 during paraaortic lymphadenectomy), 3 were bowel lesions, 3 were ureteral injuries, and 3 were nerve injuries (intraoperative obturator nerve transection). Table 9.4 shows the incidence and type of complications in the two cited studies. Table 9.5 focuses on the incidence and characteristics of major vascular complications during laparoscopic pelvic and paraaortic lymphadenectomy.

In general, looking at the number of complications in the largest published series from tertiary care centers, laparoscopic procedures appear to be associated with a decrease in the overall rate of surgical complications. Among the most common (although usually not life-threatening) postoperative complications of lymphadenectomy are lymphocyst formation, lymphorrhea, or lymphedema. A study conducted at the University of Insubria compared the incidence of these types of lymphatic complications in women undergoing laparoscopic versus open procedures. The findings of that analysis showed that the incidence of lymphocysts is significantly lower if minimally invasive surgery is performed compared with standard laparotomy, whereas no significant differences between the two approaches were reported in terms of lymphorrhea and lymphedema. Among the possible explanations for these results, the most plausible is that laparoscopy is associated with decreased formation of adhesions and consequently a lower likelihood of lymphatic fluid entrapment after lymph node dissection. A recent study performed at the Mayo Clinic enrolling more than 1000 consecutive surgically treated patients suggested a significant correlation between lymphadenectomy and lymphedema, with an attributable risk of 23% for patients who underwent lymphadenectomy compared with hysterectomy alone. Moreover, pelvic and paraaortic lymphadenectomy, as well as open surgical procedures, have been shown not only to affect the patients’ quality of life but also to be strongly associated with a marked increase in costs of approximately US $4500 (calculated as year-2010 Medicare dollars) per patient undergoing staging for endometrial cancer.

Vascular Anatomic Variations in the Paraaortic Area

One of the major contributors to complications during paraaortic lymphadenectomy is the impact of vascular anomalies. Although the pelvis is seldom affected, the paraaortic area is not infrequently the site of possible anatomic variants that are not always recognized preoperatively. Authors reporting the findings of several series of in vivo and cadaveric dissections have indicated a consistent incidence of vascular anomalies in the paraaortic region, ranging from 17% to 44%. Anomalies of the renal vessels are the most commonly described. An accessory left lumbar vein draining into the left renal vein is the most common, followed by the polar renal arteries and circumaortic left renal vein. A precaval right renal artery has been described in about 5% of patients. Retroaortic renal veins are less common, as are duplicated venae cavae ( Fig. 9.3 and Table 9.6 ).

Tips and Tricks to Avoid Vascular Injuries During Lymphadenectomy

• Carefully review preoperative imaging studies before the operation.

• The anterior aspect of the aorta is an area that is much less likely to be affected by anatomic variation. Most anatomic variations arise from the lateral aspects of the vessel.

• Paraaortic node dissection should start at the anterior aspect of the aorta, particularly during minimally invasive procedures.