Strategies for chylothorax after pulmonary resection and lymph node dissection for primary lung cancer

Background: Pulmonary resection with mediastinal lymph node dissection for treating primary lung cancer could sometimes causes chylothorax as a postoperative complication. This study examined the validity of treatments for chylothorax in our hospital. Methods: We evaluated 2,019 patients who underwent lobectomy, bilobectomy, or pneumonectomy with mediastinal lymph node dissection for primary lung cancer at Shizuoka Cancer Center Hospital, Shizuoka, Japan, between September 2002 and March 2018. The diagnostic criteria for postoperative chylothorax were that the drainage from the pleural drain was evidently white and turbid, or the pleural effusion contained a triglyceride level of >110 mg/dL. The clinical courses and treatments were retrospectively reviewed. Results: Postoperative chylothorax occurred in 37 patients (1.8%), 20 men and 17 women, with a median age of 70 y (33 to 80). Thirty-ve patients had a lobectomy and two patients had a bilobectomy. A low-fat diet was instituted to all patients; 35 cases improved with conservative treatment, and 2 cases required reoperation. Nine cases had a drainage volume ≥ 500 ml one day following the low-fat diet commencement, which was resolved with conservative treatment and decreased drainage was observed on the third day of treatment in seven of those cases. Conclusions: With the exception of cases with excessive drainage of ≥ 1000 mL in one day and systemic symptoms associated with chyle loss, conservative treatment could be successful even when the daily drainage volume exceeds 500 mL following a low-fat diet, provided the drainage volume decreases within three days.

commencement, which was resolved with conservative treatment and decreased drainage was observed on the third day of treatment in seven of those cases.
Conclusions: With the exception of cases with excessive drainage of ≥ 1000 mL in one day and systemic symptoms associated with chyle loss, conservative treatment could be successful even when the daily drainage volume exceeds 500 mL following a low-fat diet, provided the drainage volume decreases within three days.

Background
We sometimes experience chylothorax as a postoperative complication when performing lung resection with mediastinal lymph node dissection for primary lung cancer. It is caused by a thoracic duct injury and is characterized by the presence of chyle, composed of abundant triglycerides and chylomicrons, in the thoracic cavity [1]. Conservative treatment is successful in cases with minor thoracic duct damage. If conservative treatment is ineffective, the treatment strategy is gradually altered and eventually reoperation is considered. Chymic drainage of > 450 mL [2] or 500 mL [3,4] per day is reportedly an indicator for reoperation; however, this treatment is practiced as an empirical rule.
Few reports exist regarding the treatment strategy for chylothorax following pulmonary resection. Hence, we report the clinical course and the validity of treatments for postoperative chylothorax occurring in our hospital.

Methods
This study was approved by the institutional review board of Shizuoka Cancer Center (approval no. 56-2020-1-3), and all the patients provided informed consent.
Among the 2,019 patients who underwent lobectomy, bilobectomy, or pneumonectomy with mediastinal lymph node dissection for primary lung cancer between September 2002 and March 2018 at Shizuoka Cancer Center Hospital, Shizuoka, Japan, 37 patients with postoperative chylothorax were included in this study. The diagnostic criteria for chylothorax were that the drainage of the pleural drain was evidently white and turbid, or the pleural effusion contained a triglyceride level of > 110 mg/dL. The medical records of these patients were reviewed for age, sex, surgical procedures, pathologic ndings, amount of chest tube drainage, and treatment methods of the chylothorax. The pathological stage was determined according to the 7th edition of the Tumor(T) Node(N) Metastases (M) classi cation for lung cancer proposed by the International Association for the Study of Lung Cancer [5].
All the patients were initially treated conservatively with a low-fat diet (fat < 10-30 g/day), prohibiting any consumption of fat-containing food or drink, and the amount of chylous drainage uid was measured. Pleurodesis was performed by injecting a mixture of 10 Klinische Einheit units of OK-432 (a heat-and penicillin-treated lyophilized preparation of the SU strain of Streptococcus pyogenes A3; Picibanil; Chugai Pharmaceutical, Tokyo, Japan) and 100 mL isotonic saline into the thoracic cavity through a chest tube.
Chest tube removal was considered when the chest tube drainage uid was serous and the volume was at least < 200 mL per day.
In this study, we retrospectively analyzed the clinical courses and the adequacy of treatments for postoperative chylothorax.

Results
A total of 37 cases (20 men, 17 women; median age, 70 y; age range, 33-80 y) with postoperative chylothorax were observed among 2,019 patients (1.8%). Two patients underwent a bilobectomy and 35 patients underwent a lobectomy. Lobectomy on the right side was performed in 25 patients, including 20 upper, 3 lower, and 2 middle lobectomies. Lobectomy on the left side was performed in 10 patients, including 8 upper and 2 lower lobectomies. In addition, selective and systematic mediastinal lymph node dissections were performed in 17 and 20 cases, respectively. Right upper lobectomy was the most common; however, the same result was obtained when comparing the frequency in each resected lobe.
The patient characteristics are summarized in Table 1. [insert Table 1 here] No patient received induction therapy. Complete video-assisted thoracoscopic surgery was performed in 1 patient (3%). The pathologic lymph node metastases were stage pN0 in 33 patients (89%), pN1 in 1 patient (3%), and pN2 in 3 patients (8%).  Data are presented as median (range) or number. a : Some patients had more than one comorbidity. b : The frequency is shown for each resected lung lobe.
When comparing the areas of the dissected lymph nodes, in all right-sided operations, superior mediastinal lymph node dissection was performed, and no chylothorax was noted in the group in which only the inferior mediastinal lymph node was dissected. In the left-sided operations, chylothorax occurred in the superior mediastinal lymph node dissection only, the inferior mediastinal lymph node dissection only, and both groups (Table 1).
Thirty-ve patients (95%) were relieved with conservative treatment. No patient was treated with complete oral intake cessation and total parenteral nutrition. When the standard for drainage was set to 500 mL on the rst day following the introduction of a low-fat diet, 28 cases were < 500 mL and 9 cases were ≥ 500 mL. Reoperation was performed in 2 of these 9 cases; and both cases showed a drainage volume ≥ 1000 mL per day. Leak points of chyle were revealed in front of the trachea at the lower end of the right superior mediastinum and root of the left common carotid artery at the upper edge of the aorta, respectively. The injuries were covered using a TachoComb brinogen tissue sealing sheet (CSL Behring, Tokyo, Japan). In both the cases, the thoracic duct was ligated at a level slightly above the diaphragm. After reoperation, the chest tube drainage decreased, and the chylothorax was treated at a median of 5 days (range, 4-6) following reoperation and a median of 6.5 days (range, 5-8) following the initial surgery.
In the other 7 cases, conservative treatment provided relief, based on the assumptions that the turbidity of the pleural effusion did not worsen over time, and the drainage volume decreased within 3 days. Pleurodesis was performed in 2 of the 7 cases. Chest tubes of these 7 cases were removed at a median of 5 days (range, 5-9) following the initial surgery (Fig. 1). There was 1 case in which the daily drainage volume exceeded 500 mL on the third day, and it took 9 days until the chest drain could be removed (  In addition, conservative treatment was successful in all 28 patients whose drainage did not exceed 500 mL on the rst day, their chest tubes were removed at a median of 4 days (range, 1-12) following the initial surgery (Fig. 1). Pleurodesis was performed in 3 cases.
Thirty-day death did not occur in any of the patients. Finally, 4 patients developed arrhythmia and 1 acute renal failure. All of these patients were treated conservatively.

Discussion
Chylothorax, although relatively rare, is known as a serious complication and has a variety of causes. In thoracic surgery, chylothorax is caused by a thoracic duct injury due to mediastinal lymph node dissection, especially when performed for primary lung cancer. The incidence of postoperative chylothorax following lung resection is estimated to be 0.25-3% [3,6] and is known to occur more often in the right thoracic cavity [7]. Previous studies have reported that lobectomy, right side surgery, robotic surgery, and pathological N2 disease are the most common causes of chylothorax [2]. In addition, Uchida et al. [8] stated that the most frequent chyle leak point was the #4R lymph node.
To date, there have been multiple discussions regarding the treatment methods for postoperative chylothorax [2][3][4][9][10][11]. The principles of conservative treatment include e cient pleural drainage, su cient residual lung expansion, and pleurodesis for reducing the dead space in the thoracic cavity. Furthermore, we attempted to close the stula by reducing lymphatic ow in the thoracic duct using a low-fat diet, fasting, and total parenteral nutrition [12]. The rst choice of treatment is conservative treatment, and it is reported that approximately 70-80% of the cases are relieved by conservative treatment alone [4].
Since chyle contains triglycerides, lymphocytes, proteins, and electrolytes, the loss of a large amount of chyle results in malnutrition and impaired immune function [4]. Therefore, for cases in which conservative treatment is ineffective, surgical treatments involving closure of the damaged site of the thoracic duct or ligation of the thoracic duct are considered. Regarding surgical indications, Shimizu et al. [4] argued that reoperation should be performed if drainage uid ≥ 500 mL is obtained on the rst day following fasting and total parenteral nutrition. In addition, Takuwa et al. [3] concluded that reoperation should be considered if there is drainage ≥ 500 mL per day after starting a low-fat diet (≤ 10 g).
In our study, conservative treatment was successful if the amount of drainage was < 500 mL on the rst day after starting a low-fat diet. Furthermore, when the drainage on the rst day was ≥ 500 mL, the cases in which the drainage tended to decrease within 3 days recovered by conservative treatment. This suggests that even in cases in which the chyle leakage exceeds 500 mL daily, if a patient's general condition is maintained, the course could be observed conservatively at least until the third day. There was no noticeable difference in the duration of chest drainage between the surgery group and the conservative treatment group, and the maintenance period did not exceed 2 weeks even in the conservative treatment group. However, even if the drainage decreased by the third day, in the case where the drainage still exceeds 500 mL per day, the drainage period may be prolonged. Treatment progression chart The treatment progression when the rst day drainage standard is set to 500 mL following the introduction of a low-fat diet. The period until drain removal is counted from the day when chylothorax is diagnosed. Data are presented as median (range).

Figure 2
Treatment ow chart Treatment ow chart for cases where the amount of drainage exceeds 500 mL on the rst day following the introduction of a low-fat diet for postoperative chylothorax.