Cardiophrenic lymph nodes (CPLNs), also referred to as paracardiac or supradiaphragmatic lymph nodes, are located proximally above the diaphragm in the most basal area of the mediastinum. The region anterior to the pericardium normally houses zero to two visible lymph nodes. This group of parietal thoracic lymph nodes are classified as anterior pre-pericardiac lymph nodes and middle latero-pericardiac lymph nodes, or right and left, respectively, with the midline as a reference.19 The anterior group drains into the parasternal lymph nodes, and the middle (juxtophrenic) group drains into the anterior mediastinal chain.20 Although rare, metastatic ovarian cancer tends to spread to the intrathoracic cavity, and CPLNs are the most common site of metastasis.21

The definition of positive CPLNs has remained a topic of debate. In general, CPLNs >5 mm along the short axis are defined as suspicious. Previous studies of metastatic ovarian cancer have demonstrated a detection rate of 85% for CPLNs >7 mm along the short axis, and a specificity as high as 90% for CPLNs >10 mm along the short axis.22 23 Positron emission tomography (PET)/CT imaging may have additive value in pre-operative diagnosis compared with standard CT imaging; however, the sensitivity remains limited. A 2012 study by Hynninen et al found CPLN involvement in 14 of 30 patients with metastatic ovarian cancer, 6 of whom had abnormal lymph nodes on a conventional CT scan.24 Although currently debated, the use of PET/CT imaging to diagnosis positive CPLNs has been suggested to be highly specific and can be considered when the diagnosis is unclear.25

Prior to 2013, studies on CPLN resection in the setting of metastatic ovarian cancer were limited to small feasibility series investigating removal through video-assisted thoracic surgery (VATS). The most notable of these series was performed in 2009 by Lim et al and described an approach using VATS in nine patients with suspected CPLN metastases.26 Among all nine patients, the estimated blood loss was <20 mL, all visible intrathoracic disease was completely resected without major complications, and planned chemotherapy treatment was not delayed.26 This series ultimately demonstrated CPLN resection with VATS is both feasible and safe.26 In 2013, a series by Yoo et al was one of the first reports to describe the feasibility of and technique for transabdominal CPLN resection via an incised or resected diaphragm, rather than via VATS, by gynecologic oncologic surgeons.27 From 2008 to 2011, 11 patients with metastatic ovarian cancer underwent transabdominal CPLN resection, 10 of whom underwent complete gross resection and one of whom had an optimal cytoreduction (≤1 cm of residual disease) of the chest; no significant morbidity or mortality was associated with the intrathoracic portions of their surgical procedures.27

A 2017 review by Cowan et al described 54 patients who underwent primary cytoreductive surgery with CPLN resection for stage IIIB-IV high-grade epithelial ovarian cancer at a single institution from 2001 to 2013.28 At the time of publication, this study was the largest, most comprehensive account of patients who underwent CPLN resection. Among this cohort, the median diameter of enlarged CPLNs was 1.3 cm (range: 0.6–2.9 cm). Of the 54 patients, 48 (89%) underwent transdiaphragmatic resection, 6 (11%) underwent VATS, 51 (94%) had disease present in the nodes, and 3 (6%) had no disease present in the nodes. Nineteen patients (35%) had post-operative complications, including four patients (7%) who experienced a surgery-related complication.28 All patients had optimal cytoreduction (≤1 cm of residual disease) and 30 (56%) had complete gross resection.

The prognostic relevance of enlarged CPLNs in the setting of metastatic ovarian cancer has been widely established. A 2015 retrospective study by Raban et al compared 31 patients with stage III ovarian cancer who were diagnosed with enlarged CPLNs (≥10 mm short-axis diameter) on CT scan during initial evaluation to 41 controls; patients in the control group had stage III ovarian cancer and no evidence of enlarged CPLNs on initial imaging.20 Compared with controls, patients with enlarged CPLNs had lower rates of complete response to initial chemotherapy (45.2% vs 78.0%, p=0.004), a shorter disease-free interval (median, 9.0 vs 24.0 months, p=0.0097), and decreased overall survival (median, 31.7 vs 61.3 months, p=0.001).20 Similar findings were reported in a 2020 retrospective review by Luger et al. Of 178 patients with advanced-stage ovarian cancer, 89 (50%) had enlarged CPLNs (≥5 mm short-axis diameter) on initial imaging.29 The authors demonstrated that enlarged CPLNs on imaging was an independent prognostic factor for progression-free survival (HR: 2.14, 95% CI: 1.33 to 3.42) and overall survival (HR: 2.18, 95% CI: 1.16 to 4.08).29

Although there is extensive evidence demonstrating the safety and feasibility of CPLN resection, there is limited literature investigating survival benefit following CPLN resection. A 2018 retrospective study by Lee et al suggested that resection of suspicious CPLNs may not have an effect on survival in patients with advanced ovarian cancer.30 This study, however, was not designed to compare survival outcomes, but rather to show the utility of PET/CT imaging for evaluating the extent of disease. In addition, the percentage of surgical procedures that achieved complete gross resection (27.8%) was lower compared with similar reports; however, in these other reports, any benefit of intrathoracic resection could have been attenuated by greater rates of intra-abdominal recurrences.21 28 30 In contrast, the study by Cowan et al demonstrated favorable survival outcomes in patients who underwent CPLN resection, with a median progression-free survival of 17.2 months (95% CI: 12.6 to 21.8 months) and a median overall survival of 70.1 months (95% CI: 51.2 to 89.0 months).28 Other evidence suggests that the intrathoracic recurrence rate is relatively low following intrathoracic resections at the time of cytoreductive surgery.31

Based on the current literature demonstrating that enlarged CPLNs contain metastatic ovarian cancer in approximately 90% of patients, and that resection of CPLNs carries minimal added morbidity, the removal of enlarged CPLNs should be considered in appropriately selected patients at the time of cytoreductive surgery.31 CPLN resection should be performed by well-trained gynecologic oncologists at highly specialized centers; if the primary surgeon is not trained in intrathoracic procedures, a thoracic surgeon should be involved in the care and planning of the surgical procedure.

The uptake of CPLN resection for affected patients with metastatic ovarian cancer reshaped cytoreductive surgery in this setting, which now includes treatment of mediastinal disease.32 Beginning in 2000, several case reports on mediastinal resections were published. The earliest, a report from Barcelona, Spain, by Montero et al, described a patient with low-grade serous carcinoma who underwent successful resection of an anterior mediastinal mass in the inferior retrosternum, with the assistance of a cardiothoracic surgery team.33 Less than 5 years later in 2005, Lu et al reported on a patient with recurrent mediastinal masses more than 20 years following primary management of high-grade serous ovarian carcinoma who underwent a successful resection.34 In 2009, two reports were published: Chase et al described a patient with a superior mediastinal mass who underwent hormonal treatment and palliative CyberKnife radiotherapy; and Ki et al described a patient with a mediastinal recurrence 4 years after primary management of high-grade serous ovarian carcinoma who underwent a complete resection and remained progression-free for 18 months.35 36

By 2022, there were eight reports of patients with ovarian cancer with metastasis to the mediastinum, all of whom underwent successful surgical resection.33–35 37–40 These reports demonstrated that thoracic cytoreduction involving the mediastinum in advanced ovarian cancer is feasible and has acceptable morbidity; however, the survival benefit could not be determined in these reports.40 In 2023, our institution reported on 22 patients who underwent surgical resection of mediastinal lymph nodes (not CPLNs); this cohort represented 22 (12%) of 178 patients who had undergone intrathoracic resection at our institution at the time of primary cytoreductive surgery between January 1, 2006 and September 30, 2021.21 Although outcomes were not stratified by location of intrathoracic disease, among all patients, the median progression-free survival was 33.6 months (95% CI: 24.7 to 61.9 months), the 3-year progression-free survival rate was 48.9% (95% CI: 41.2% to 56.2%), and median overall survival was 81.3 months (95% CI: 68.9 to 103 months). When stratified by residual disease status, median progression-free survival was 51.8 months when complete gross resection was achieved versus 16.7 months with residual disease (HR: 2.17, p<0.001), and median overall survival was 97.6 months when complete gross resection was achieved versus 65.9 months with residual disease (HR: 2.05, pp=0.003).21

These studies demonstrate that mediastinal metastasis of ovarian cancer can be completely resected in appropriately selected patients. Ultimately, multidisciplinary treatment including surgery can lead to improved long‐term prognosis in patients with mediastinal metastasis of ovarian serous carcinoma.

Limited data are available on the surgical management of metastatic ovarian cancer within the lung parenchyma and/or pleura; however, literature on this topic has been growing in recent years. The largest series to date is a retrospective, multicenter study in Japan published in 2020.41 In this study, Kanzaki et al described 1508 patients who underwent pulmonary resection between January 2006 and December 2015 to treat lung nodule metastasis from tumors of various organs. Among all patients, six underwent surgical resections for metastatic ovarian cancer; three of these patients underwent a wide-wedge resection, two underwent a segmentectomy, and one underwent a lobectomy.41 At the time of publication, two patients had no evidence of disease — one had achieved long-term disease-free survival of 61 months and the other had surgery 5 months prior. Three patients achieved mid- to long-term survival and were alive with disease ranging from 38 to 61 months following surgery.41 One patient, who had synchronous metastasis present throughout the intrathoracic region, died of disease 2 months following surgery.41

Prior to the study by Kanzaki et al, Kimura et al (1995) described a patient with ovarian cancer who underwent a lobectomy/wide-wedge resection and recurred 14 months after; Toishi et al (2011) described a patient who underwent a lobectomy for ovarian cancer and had disease recurrence 11 months later; and Kita et al (2015) described a patient who underwent a segmentectomy for ovarian cancer metastasis and had disease recurrence 12 months later.42–44 Most recently, our institution reported on 14 patients who underwent surgical resection of lung parenchyma/pleural metastasis. This cohort represented 7.9% (14/178) of patients who had undergone intrathoracic resection at our institution at the time of primary cytoreductive surgery; patient-specific outcomes were not reported.21 Overall, studies have demonstrated that resection of ovarian cancer metastasis involving the lungs/pleura is feasible and can provide favorable outcomes in properly selected patients; however, evidence suggests that patients with wide-spread or synchronous pulmonary metastasis from ovarian cancer are not good candidates for intrathoracic resection.41

As intrathoracic surgical procedures for the treatment of metastatic ovarian cancer have become more established, the use of VATS in this setting has also increased. Potential benefits of VATS include visualization of pleura, the potential to biopsy or resect pleural-based tumor nodules, possible modification/abbreviation of the abdominal procedure, lysis of adhesions, improved distribution of pleurodesis substance, and a lower failure rate. Another benefit is that VATS is performed under the same anesthesia as a laparotomy (dual lumen endotracheal tube).45 46

In 2020, our institution performed a study investigating 100 patients with advanced ovarian cancer with suspected intrathoracic metastasis who underwent VATS.47 The results supported using VATS in the triage of patients with moderate-to-large pleural effusions to primary cytoreductive surgery versus neoadjuvant chemotherapy with possible interval cytoreductive surgery. Our findings suggest patients who do not have macroscopic intrathoracic disease seen on VATS can be offered primary cytoreductive surgery.47 Patients who have intrathoracic macroscopic disease that cannot be resected or ablated should be offered neoadjuvant chemotherapy followed by intrathoracic cytoreduction, if necessary.47 Our findings also demonstrate that intrathoracic cytoreduction could benefit appropriate patients only if all macroscopic disease can be eliminated (both in the intrathoracic and abdominal regions). These findings led us to change our prior management algorithm, which was developed in 2010,48 to the recently modified algorithm shown in Figure 1.

Newly developed institutional algorithm for use of video-assisted thoracic surgery (VATS) in patients with suspected ovarian carcinoma with moderate-to-large pleural effusions.