Safety and Efficacy of Supradiaphragmatic Lymph Node Dissection in Advanced Ovarian Cancer

Introduction

Ovarian cancer is the most lethal gynecologic malignancy in the United States. Up to 75% of patients present with advanced-stage disease; for these patients, survival outcomes are dependent on multiple factors, including residual disease status after cytoreductive surgery. ¹ Previous studies have consistently demonstrated that no gross residual disease after surgical cytoreduction is associated with the longest overall survival (OS).^2–6 Recently, comprehensive debulking techniques have expanded to include resection of metastatic disease in the intrathoracic supradiaphragmatic lymph nodes (SDLNs), also known as cardiophrenic lymph nodes. SDLNs are located in the most-basal portion of the mediastinum between the heart and the diaphragm, and are classified as anterior prepericardiac, middle lateropericardiac, left, and right. ⁷

SDLNs can be accessed via a transdiaphragmatic approach or video-assisted thoracic surgery (VATS). In the transdiaphragmatic approach, after mobilization of the liver, a longitudinal incision is made in the direction of the diaphragm muscle fibers to access the pleural and cardiophrenic spaces.^7–9 The surgeon must avoid damaging the diaphragmatic vessels, nerves, and the hepatic veins. After entering the chest cavity and identifying the fat pad containing the suspected lymph nodes, resection is performed using surgical metal clips or a sealing cut device; care must be taken to avoid injury to the lung or pericardium.^7–9 The same technique is performed on the left side, which requires mobilization of the triangular ligament of the liver; care should be taken to avoid injury to the spleen or the left diaphragmatic vessels and nerve.

Alternatively, VATS is typically performed by a thoracic surgeon by creating a 2-cm chest-wall incision between the fifth to eighth intercostal spaces. ¹⁰ VATS and abdominal cytoreduction can be performed in a single step, or VATS can be performed upfront and the abdominal cytoreduction later in the two-step technique. ¹⁰

Although there is much evidence demonstrating the safety and prognostic relevance of SDLN resection, few studies have evaluated any associated survival benefit. This review provides the rationale for performing SDLN dissection in advanced-stage ovarian cancer, summarizes clinical outcomes after SDLN resection, and describes its role in the treatment algorithm.

Materials and Methods

A PubMed literature search was performed for January 2004 to December 2021, using the following search terms: advanced ovarian cancer; cardiophrenic lymph nodes; supradiaphragmatic lymph nodes; intrathoracic cytoreduction; lymph node resection; primary debulking; and secondary cytoreduction. A variety of prospective, retrospective, and cohort studies were identified. The references of retrieved articles were also considered if the references were relevant.

Eligible studies included patients with advanced ovarian, tubal, or peritoneal cancer who had undergone SDLN resection. As SDLNs can be removed via a transdiaphragmatic approach or VATS, studies that used either approach were included and eligible for analysis. Study objectives included evaluating the utility of SDLN dissection for diagnostic and treatment purposes and assessing the benefits of SDLN resection in relation to prognosis. The study results were reviewed for intraoperative or postoperative complications, progression-free survival (PFS) or OS data, and disease recurrence. The findings are summarized in Table 1.^7,11–13

Results

Thirty studies from 2004 to 2021 were identified. Of these 30 studies that included the performance of SDLN resection and provided postoperative follow-up, 4 reported adequate clinical outcome data, such as PFS, OS, and/or recurrence rates, and were included in our analysis. The four studies were published after 2016 (Table 1).^7,11–13

Garbi et al. assessed the feasibility of transdiaphragmatic resection of enlarged SDLNs in 22 patients with advanced, high-grade serous ovarian cancer (HGSOC) who had undergone upfront debulking surgery. ¹¹ Suspicious SDLNs had a short-axis dimension >5 mm on preoperative computed tomography (CT) scanning. Including all the debulking procedures, the average operative time was 333 minutes, and the average estimated blood loss was 1000 mL. The median number of SDLNs resected was 1.7 (range: 1–8). Twenty patients (90%) had residual disease ≤5 mm, of whom 8 (40%) had no residual disease. The final diagnosis of metastatic disease was confirmed in 21 (95%) of 22 patients who had SDLNs removed.

Postoperative chest tubes for pleural effusion were placed in 8 patients (36%). Six patients (27%) required blood transfusions. Of the 3 patients who had recurrences, none had a recurrence in the cardiophrenic space. Although this study did not show the impact of transabdominal SDLN resection on OS in patients with stage-IV disease, it demonstrated the safety and feasibility of SDLN resection during primary cytoreduction clearly.

Cowan et al. evaluated SDLN resection in a larger cohort of 54 patients who had undergone primary cytoreductive surgery. ¹² A transdiaphragmatic approach was used in 89% of patients and VATS by a thoracic surgeon in 11% of them. ¹² All patients had suspicious SDLNs on CT, with a median diameter of 1.3 cm (range: 0.6–2.9 cm). Fifty-one (94%) of 54 patients had positive nodes. A median of 3 nodes were resected (range: 1–23). The median operative time was 450 minutes, and the average EBL was 1000 mL. Optimal cytoreduction was achieved in all patients, and 30 patients (55.6%) had no gross residual disease.

An intraoperative chest tube was placed in 51 patients (94%) and kept in for a median of 4 days. Of the 19 major postoperative complications, only 3 were possibly related to the SDLN resection—1 pulmonary embolism, 1 chylothorax, and 1 pleural effusion. The median time to adjuvant chemotherapy was 40 days (range: 19–205 days). Six (13%) of 45 recurrences occurred in the thoracic nodes (4 [9%] in the pleura or pericardium). Median PFS was 17.2 months, and median OS was 70.1 months. This study demonstrated that dissection and resection of enlarged SDLNs in carefully selected patients can confirm intrathoracic disease spread and help achieve maximal cytoreduction without delaying adjuvant treatment, thereby improving survival.

Lopes et al. assessed 24 patients with epithelial ovarian cancer who had undergone SDLN resection at the time of primary or secondary debulking. ⁷ In this cohort, SDLNs were the only detectable site of extra-abdominal disease, and they were considered suspicious if the short-axis size was ≥8 mm on preoperative CT or magnetic resonance imaging (MRI), or had uptake on positron emission tomography (PET/CT). Transdiaphragmatic SDLN resection was performed after complete abdominal cytoreduction.

The average size of the SDLNs was 14 mm (range: 8–25 mm). An average of 1.5 nodes were resected (range: 1–6). SDLNs were positive for metastatic disease in 21 patients (87%)—12 in the primary and 9 in the secondary cytoreductive setting. All patients who recurred initially had stage-IIIC disease and had underwent complete gross resection at primary cytoreduction. The average duration of SDLN resection was 28 minutes, and blood loss was minimal. Twenty-three patients had chest tubes placed for an average of 48–72 hours. Six patients had pulmonary atelectasis, but, otherwise, there were no major pulmonary complications.

Eighteen of the 21 patients developed recurrences, and 11 of these patients ultimately died of complications related to cancer. The median disease-free survival of the recurrence and primary debulking groups were 17 and 12 months, respectively. The median OS durations were 66 and 30 months, respectively. The authors concluded that resection of enlarged SDLNs should be performed in select cases to confirm metastatic intrathoracic disease and to achieve complete cytoreduction with the possibility of achieving longer survival. ⁷

Lee et al. evaluated the accuracy of PET/CT for preoperative SDLN detection and prognostic significance of resection. ¹³ After initial staging with fluorine-18 fluorodeoxyglucose ( ¹⁸ F-FDG) PET/CT, patients either underwent neoadjuvant chemotherapy followed by interval debulking surgery or primary debulking surgery. Patients eligible for primary debulking (n = 176) were classified into the following groups: stage-III disease; stage-IV disease with SDLN metastases; or stage-IV disease with other metastases. On multivariate analysis, no residual disease after primary debulking surgery was associated with a better prognosis. Initial stage-IV disease with SDLN or with other metastases were associated with poorer prognoses.

Among the 176 patients who had undergone primary debulking surgery, 50 (28.4%) had ¹⁸ F-FDG uptake in the SDLNs. The median size of the SDLNs was 7.65 mm. Twenty-five patients had undergone SDLN biopsy for diagnosis; the remaining 25 patients had undergone SDLN debulking via VATS or transdiaphragmatic excision during primary debulking surgery. Twenty-one (77.8%) of the 25 patients had pathologic confirmation of metastatic disease in the SDLNs after resection.

Of the entire cohort, 151 patients (51.2%) experienced recurrences—78.8% were intraperitoneal recurrences and <10% were thoracic recurrences. The group of patients with stage-IV disease and SDLN metastases, compared with the group with stage-III disease, had significantly poorer PFS (14 versus 18 months, respectively) and OS (31.5 versus 37.5 months, respectively). PFS and OS did not differ significantly between patients with stage-IV disease and SDLN metastases and those with stage-IV disease and other metastases. Among patients with confirmed SDLN metastases, those who had undergone thoracic debulking during primary surgery did not have improved PFS or OS compared with patients who had not undergone debulking. The researchers sought to determine prognosis associated with detection of SDLNs on PET/CT, and showed the negative impact on OS in patients with FDG-avid SDLNs.

Although this study failed to show a survival benefit for SDLN resection, the low complete gross resection rate of 27.8%, the more frequent intraperitoneal recurrence rate, and the lower median OS of the entire cohort and subsets precluded adequate evaluation of the survival impact of SDLN resection as part of a comprehensive debulking strategy. ¹³

Discussion

The findings of the current review are similar to those of Nasser et al., ¹⁴ demonstrating the feasibility and safety of SDLN dissection for managing advanced epithelial ovarian cancer whether using a transdiaphragmatic technique or VATS.^8,9,11,12, The most-common complications associated with SDLN dissection are pneumothorax, pleural effusion, atelectasis, wound healing disorder, and infection; however, postoperative morbidity related to SDLN dissection is minimal overall.^8,9,11,12,Although rare, pulmonary embolism, chylothorax, sepsis, and acute pericarditis have been reported.^8,12,15

Previous studies have demonstrated that the presence of supradiaphragmatic lymphadenopathy on preoperative CT is associated significantly with a lower complete response rate, shorter PFS, and shorter OS.^16–19 Raban et al. compared patients with stage-IIIC epithelial ovarian cancer and enlarged SDLNs to patients without enlarged SDLNs and showed the former group had a lower complete response rate to initial treatment (45.2% versus 78.0%), shorter median PFS (9.0 versus 24.0 months), and shorter median OS (31.7 versus 61.3 months), respectively. ¹⁸ Similarly, McIntosh et al. showed that patients with stage-III HGSOC with enlarged SDLNs compared to those without enlarged SDLNs had lower PFS (14 versus 23 months) and OS (44 versus 65 months), respectively. ¹⁷

Despite the known prognostic relevance of SDLN metastases, the optimal method for evaluation is debatable. Preoperative staging images are typically used to detect enlarged SDLNs, which contain histologically positive metastases 85%–94% of the time.^8,11–13,20 In many institutions, suspicious SDLNs on preoperative CT are defined as those ≥5 mm^12,14,16,18; however, cutoff size for positivity has ranged from 7 mm to 13 mm across studies.^8,12,20

The emerging data was reviewed regarding clinical outcomes associated with resection of enlarged SDLNs at the time of cytoreduction for ovarian cancer; the procedure was feasible and safe in this setting. All studies but 1 (Lopes et al.) ⁷ showed a relatively low intrathoracic recurrence rate after intrathoracic debulking.

Furthermore, in Cowan et al.'s study, patients who underwent SDLN resection had an impressive median PFS and OS of 17.2 months and 70.1 months, respectively. ¹² While the study by Lee et al. ¹³ did not show a survival benefit with SDLN resection, the aim of the study was to show the utility of PET/CT scans for evaluating the extent of disease; with a low complete gross resection rate, approximately half of that reported by Cowan et al. (27.8% versus 55.6%, respectively), any benefit of SDLN resection on survival was likely blunted by the high intra-abdominal recurrence rate.

Conclusions

The safety and feasibility of SDLN resection has been demonstrated in retrospective studies within the past few years. Whether the procedure leads to improved survival or whether it should be performed during primary or interval cytoreduction warrants prospective study. The overall goal of surgery, however, remains the complete gross resection of disease. Considering that enlarged SDLNs contain metastatic disease ∼90% of the time and considering that their removal leads to minimal morbidity, the current authors favor resection of suspicious SDLNs at the time of debulking surgery for advanced ovarian cancer.

SDLN should be performed by well-trained providers in highly specialized centers. A fellowship-trained gynecologic oncologist with experience in SDLN is required to perform the procedure safely. Thoracic surgeons should be involved in the care and planning of the surgery when the primary surgeon is not trained in intrathoracic procedures.