Introduction

Chylous ascites is a rare form of ascites resulting from an extravasation of milky chyle rich in triglycerides (usually > 110 mg/dL) into the peritoneal cavity, following obstruction or disruption of major lymphatic channels. It has been described in patients with congenital lymphatic anomalies, tuberculosis, nephrotic syndrome, cirrhosis, and malignancies [1,2,3]. Moreover, it can occur after a surgical trauma of cisterna chyli or lymphatic vessels during abdominal surgery. Although it is still rare, there is evidence of an increase of incidence in patients receiving gynecologic cancer surgery, probably due to retroperitoneal lymph-node dissection [1].

The most common findings associated with this rare complication are abdominal distension, nausea, vomiting and milky-appearing discharge from the vagina after gynecologic surgery, if an intraperitoneal drain is not placed [1,2,3,4].

Management of postoperative chylous ascites may be challenging. In the present study, we retrospectively reviewed all patients who underwent surgery for ovarian cancer to identify the incidence of this rare complication in our series and to describe its management.

Materials and methods

In the present study, we retrospectively reviewed all patients who underwent surgery for ovarian cancer at our Institution from October 2016 to November 2018.

Study data were retrospectively extracted using REDCap electronic data capture tools hosted at Policlinico Agostino Gemelli Foundation, IRCCS. This study protocol was approved by Internal Review Board (IRB).

Chylous ascites was defined as milky-white appearance or high triglyceride levels (> 110 mg/dL) in the fluid obtained from diagnostic paracentesis or less commonly from the peritoneal drainages. Women showing these characteristics were then selected for final analysis.

Clinico-pathological features, including the primary tumor histology, stage, grade, surgical procedure, type of lymphoadenectomy, median number of removed pelvic and para-aortic lymph nodes, and number of metastatic lymph nodes were collected and analyzed.

Conservative treatments included total parenteral nutrition, paracentesis and somatostatin followed by additional low-fat diet supplemented with medium-chain triglycerides as maintenance therapy for a minimum of 1 month. Complete clinical success was defined as the disappearance of the ascites and/or clinical symptoms, such as abdominal distension and dyspnea. Patients with progressive abdominal distension or with increasing or stable chylous drainage despite treatment after 30 days were classified as non-responsive.

The time to chylous ascites onset was defined as the interval between the surgical procedure and the appearance of chylous ascites. The time of resolution was defined as the time from initial diagnosis and the condition resolution (abdominal drainage tube of chylous ascites < 300 mL or ascites tending to be limpid fluid). White blood cell counts and body temperature measurements were performed to exclude infected ascites fluid (particularly bacterial peritonitis).

Results

Five hundred and forty-six patients were submitted to surgery for ovarian cancer in the study period. Among them, 298 women underwent pelvic and/or para-aortic lymphadenectomy. Chylous ascites occurred in eight patients, with an overall incidence of 1.4% in the entire population and a 2.68% in the subgroup of patients submitted to lymphadenectomy (Table 1): two for staging purposes and five during primary debulking surgery and 1 during interval debulking surgery.

Table 1 Characteristics of ovarian cancer patients with available information regarding early complications performed from October 2016 to November 2018 in our institution according to the type of lymphadenectomy
Full size table

All patients with chylous ascites had bilateral pelvic and para-aortic lymphadenectomy. Clinicopathologic features of all patients are shown in Tables 2 and 3. The median number of lymph nodes was 11.5 (range 1–28) in the pelvis and 17 (range 6–40) in the para-aortic area. A variety of approaches (i.e., laparoscopy, robotic, and open surgery) and methods, (i.e., standard or advanced bipolar cautery) were used to dissect the lymphatic tissue (Fig. 1). Additionally, absorbable suture or hemoclips were used to ligate all vascular and lymphatic channels during dissection. No hemostatic or sealant agents such as fibrin glue were routinely used after retroperitoneal lymph node dissection (RPLND). Median time from surgery to the development of symptoms was 7 days (range 6–9). Six women (75%) developed chylous ascites after discharge from the hospital and were re-admitted through emergency room (ER) for occurrence of symptoms, such as abdominal distension, bloating, and dyspnea.

Table 2 Characteristics of patients with ovarian cancer and chylus ascites
Full size table
Table 3 Patients’ characteristics and surgical–pathological factors
Full size table
Fig. 1
figure 1

Lombo-aortic lymphadenectomy

Full size image

Ultrasonography and paracentesis were used for diagnosis. All women received a percutaneous drainage that remained in place for a median of 9 days (range 7–12). A maximum output of 1000 mL was allowed per day. In 4 women the fluid was examined biochemically, and 110 mg/dL was accepted as the threshold level for diagnosis of chylous ascites.

The management of this complication was chosen after a nutritional evaluation according to the attending physician. All patients received total parenteral nutrition (TPN) with Olimel N4E 2000 mL (Baxter®) and somatostatin therapy with 0.2 mL 3 times per day for a median of 9 days (range 7–11). Median time of recovery was 15 days (range 7–16).

Patients with complete clinical success took minimum 1 month of additional diet therapy as maintenance therapy with low-fat diet supplemented with medium-chain triglycerides. Only one patient needed to additional TPN after discharge. No patient required surgical correction since all of them responded to conservative treatment with the longest response requiring 22 days.

Discussion

The incidence of chylous ascites after surgery for gynecologic malignancies is not well established. Han et al. reported an incidence of 0.32% after para-aortic lymphadenectomy and 0.077% after pelvic lymphadenectomy, following staging surgery for gynecological malignancies [5]. In the same year, Tulunay et al. found a higher incidence up to 2% of chylous ascites with staging surgery for gynecological malignancies, probably due to a more aggressive lymphadenectomy in the para-aortic region [6]. Indeed, the incidence of chylous ascites in Tulunay’s study is similar to ones reported after retroperitoneal lymphadenectomy for testicular cancer [7, 8].

Regarding ovarian cancer, recently Harter et al. published results of Lymphadenectomy in Ovarian Neoplasms (LION) trial where patients with advanced ovarian cancer and pre- and intra-operatively clinical negative lymph nodes were randomized intra-operatively to receive lymphadenectomy or not.

LION trial shows between the two arms no differences in term of overall survival and an increased surgical morbidity in the lymphadenectomy arm without mentioning the rate of chylous ascites. Other data regarding chylous ascites after lymphadenectomy for ovarian cancer are limited to case report/series showing a widespread heterogeneity.

Our analysis showed an incidence of 2.6% among patients submitted to lymphadenectomy confirming data previously reported in the Literature. In our series six patients (75%) developed chylous ascites after discharge showing that this condition usually has a late onset.

Considering the reduced indications for lymphadenectomy in OC, both for early stage (ESMO-ESGO Guidelines, April 2018), where it stands only to clarify the role of adjuvant chemotherapy, and for advanced stage, where it is accepted only for resection of bulky nodes (LION trial), a further significant decrease of this condition is likely to happen in the near future. Finally, the introduction of sentinel node evaluation even in early stage OC might ensure adequate staging, while reducing complication related to lymph node dissection (Selly protocol, NCT03563781).

Chylous ascites management remains controversial, due to the rarity of this condition. In the literature we found a widespread heterogeneity of clinical conducts. Conservative treatment includes low-fat diets that are supplemented with medium-chain triglyceride (MCT) as well as total parenteral nutrition, somatostatin therapy and paracentesis.

While serial paracentesis can provide resolution, it may lead to prolonged leakage and increase the risk of infection [9, 10]. Therefore, this procedure should be reserved for patients with severe abdominal discomfort and dyspnea due to massive chylous ascites. We did not perform serial paracentesis because we routinely used intraperitoneal drainages. Surgical intervention should be indicated only for patients who do not respond to conservative management. Most authors wait for 4–8 weeks before proceeding with surgical exploration [4]. Conservative treatment methods have been reported with variable success rates in different studies. In a study by Frey et al. 12 patients were treated with diet only. In this series, five of these patients needed recurrent paracentesis but did not require surgery [1]. In another study by Tulunay et al., 7 (29%) of 24 patients required surgical correction [6]. Conversely, Han et al. and Zhao et al. reported 100% success rates for the conservative treatment of postoperative chylous ascites [5, 11]. Our findings were consistent with Han et al. [5] and Zhao et al. [11], and all patients responded to conservative therapy within 2 weeks.

Otherwise, scientific scenario seems to turn its view toward new technologies for gynecologic pathologies [12,13,14,15,16,17,18,19,20,21,22,23,24], and a maximal effort in minimal invasive procedures by laparoscopy or robotic [25,26,27,28,29,30,31,32,33,34,35] or such as sentinel node [36,37,38] evaluation which could be the right way to obtain staging information erasing complication linked to lymph node dissection.

Several authors described the possibility to predict surgical and oncological outcomes to perform aggressive surgical procedures without increasing surgical morbidity [14, 39,40,41,42,43,44,45,46].

Considering all these aspects, quality of life represents a fundamental goal to reach, even if different factors play a role [47,48,49,50].

In our center, we applied a protocol shared with nutritionists and this conservative management seems to be able to control this complication.

Nevertheless, the best scheme and the correct timing are already unclear. However, a prospective multicenter randomized study able to confirm the superiority of conservative management (and which conservative management), is unlikely to be performed due to the rarity and future reduction of this condition. Therefore, this retrospective study, despite its evident biases, confirms that conservative management of chylous ascites is feasible and effective.