Postoperative chylothorax is a rare complication in children who have undergone cardiac surgery for congenital heart disease. When it occurs, it can contribute to patient's morbidity or mortality. The general treatment strategy of postoperative chylothorax is conservative, but surgical reintervention may be indicated in refractory cases. Lymphatic duct ligation or pleurodesis has been considered as the proper surgical intervention (1, 2). With these surgical treatments, however, problems still remain, and the results are not always satisfactory. Furthermore, persistent chylothorax may cause the additional risks to the already compromised patient due to reopening the chest for lymphostasis. Therefore, a noninvasive treatment for postoperative chylothorax would be better for the high-risk patients with congenital heart defect. Over the last years, an additional conservative treatment option with somatostatin has been introduced as a new modality for treatment of postoperative chylothorax (3-5). Thus far, the results have been mostly reported in the literature as individual case reports on children.

In this study, we report our experience in the management of postoperative chylothorax, with emphasis on the efficacy of somatostatin therapy in infant and children with chylothorax after cardiac surgery.

From March 2002 to February 2003, nine patients developed chylothorax following operations for treatment of congenital cardiovascular disease at Samsung Medical Center. Chylothorax was diagnosed based on the presence of pleural fluid containing at least one of the following three characteristics: 1) milky appearance; 2) pleural fluid triglyceride levels greater than 110 mg/mL; 3) lymphocytic predominance in the fluid cellular component along with sterile cultures (6-8).

The conventional treatment had been undertaken upon recognition of chylothorax in all patients initially. The conventional treatment consisted of drainage of the pleural space by thoracentesis or tube thoracostomy, total parenteral nutrition, enteric rest or diet manipulation such as oral fat intake restricted to medium-chain triglycerides. If the amount of chylous leakage increased in 2 or 3 consecutive days after the conventional treatment, then intravenous somatostatin infusion was instituted. The initial starting dose was 3.5 µg/kg/hr and dosage was increased in a stepwise fashion up to 5, 7, and 10 µg/kg/hr and with a maximum dosage of 15 µg/kg/hr. Calories, electrolytes, proteins and clotting factors were monitored and replaced as necessary. Those patients submitted to cavo-pulmonary procedure (Fontan operation) were also treated with diuretics, inotropes and vasodilators in order to reduce systemic venous pressure overload. The duration of somatostatin administration was dependent on the patient's response. After cessation of the chylous leakage and resumption of normal dietary habits, we initiated a tapering of somatostatin administration by halving the dosage every day over a 3-day period. Side effects of somatostatin such as hypotension, loose stool, and glucose instability were carefully monitored on a regular basis.

All results are represented as means or median values with ranges.

Table 1 shows details of patient, age at time of surgery, the cardiac diagnosis, the type of operation performed, interval to diagnosis of chylothorax after cardiac operation and the maximal drainage volume per day. The patients ranged in age from 1 week to 4 yr old. They were diagnosed of chylothorax between postoperative day 2 and day 16. The maximal daily volume of drainage ranged from 10 to 127 mL/kg per day. Only one patient (patient 1) experienced chylopericardium combined with chylothorax.

Table 2 indicates result of the treatment and the final outcome for each case. The duration of somatostatin therapy varied from 7 to 32 days with a median duration of 18 days. A decrease in drainage was observed at an average of 3.1 days after the initiation of somatostatin, as the dose of somatostatin had increased enough to produce an effect (Fig. 1). All cases of chylothorax were successfully treated with conservative management combined with intravenous infusion of somatostatin. The median duration to reduce to 5 mL/kg per day of lymph drainage was 5 days (range 1 to 25 days). In patient 6 and patient 9, a rebound phenomenon of chylothorax was noted upon the reduction of somatostatin therapy, but the rebound effusion responded promptly when the dose of somatostatin was boosted (Fig. 2). No increase in drainage was observed when their normal diet was resumed. All nine patients had recovered completely from postoperative chylothorax by the use of somatostatin, and could avoid any invasive surgical interventions.

Chylothorax is a rare, but potentially serious complication of pediatric cardiac operations. It was estimated to occur at an incidence of 0.25% to 0.8% following cardiovascular procedures (9, 10). The standard treatment for postoperative chylothorax includes conservative therapy with pleural space evacuation, enteric rest or fat-free nutrition, and total parenteral nutrition (11). If this is not successful, operative treatment such as pleurodesis, ligation of the thoracic duct and pleuroperitoneal shunt is indicated (12, 13). However, these surgical treatments have some limitations. Pleurodesis is a painful procedure and the long-term results can vary. Ligation of the thoracic duct may fail as a result of anatomic variations of the thoracic duct or due to a failure for identifying the chylous leakage sites (6).

There have been several anecdotal reports in the literature related to using somatostatin for treating chylothorax as a complication of thoracic procedures in adults (3-5). Somatostatin is a peptide hormone that acts as a neurohormone as well as a paracrine agent, and it is present in the central nervous system, gastrointestinal tract and pancreas (14). The exact mechanisms involved for the drying effect of somatostatin on lymphatic leakage are not wholly understood. The effectiveness of somatostatin may be due to its ability to decrease the hepatic venous pressure gradient, to decrease the intestinal absorption of fats, to decrease the triglyceride concentration in the thoracic duct and to attenuate splanchnic blood flow (15-17). In addition, it could be that the lymphatic vessels contain somatostatin receptors, like the blood vessels in the splanchnic area, and that their constriction reduces the lymph production (18, 19).

In 1998, Rimensberger et al. (20) reported the successful use of somatostatin in children to control increasing chylous drainage under conservative treatment. In addition, the authors described several pediatric cases of postoperative chylothorax that were treated with somatostatin (21-23). In our study, 3 patients underwent surgical procedure for their congenital heart disease during the neonatal period. In our cases, no serious complications associated with unabated chylous drainage were encountered, which is in contrast to the previous reports (21, 22). This is probably due to the prompt use of intravenous somatostatin after the diagnosis of chylothorax during the postoperative period.

Opinions vary in the literature regarding the appropriate dose and type of administration for somatostatin for children with postoperative chylothorax. The starting dose was 3.5 µg/kg/hr according to the case report of Rimensberger et al. (20). Buettiker et al. suggest that chyle production began cessation at a dose of 10 µg/kg/hr, but they also tried increasing the dose up to 12 µg/kg/hr for chylothorax refractory to the conventional dose (21). In our study, we increased the dosage of somatostatin up to 15 µg/kg/hr in one case without observing any serious side effects. Somatostatin is currently used for the therapy of diverse pediatric diseases such as acromegaly, intractable diarrhea and gastrointestinal bleeding (24-29). Because this therapy is still experimental, further studies that take into consideration the duration and tolerable dosages are necessary to assess the safety of its administration in children.

Superior vena cava thrombosis is also well known to be a risk factor for the failure of non-operative management of chylothorax (1). It is usually recommended that such patients be considered for earlier operative intervention when this complication is associated with chylothorax. In our case, however, the patient's poor general condition led us to attempt somatostatin trial before any invasive steps. The satisfactory outcome in this patient may be speculated as the results of somatostatin and/or the development of collateral lymphatic channels.

The side effect of somatostatin is primarily related to the suppressed actions of gastrointestinal motility and secretion, and this includes hypotension, loose stool, malabsorption, nausea, flatulence, liver dysfunction and hyperglycemia (15). In all 9 patients, no significant side effects were observed that necessitated medical intervention or discontinuation of somatostatin.

The results of our case series are suggesting the value and safety of somatostatin in pediatric patients with postoperative chylothorax. The continuous infusion of somatostatin may be of particular benefit if the patient is in a poor general condition, and additional risks may be introduced with repeated surgery. A larger prospective study needs to be undertaken to verify our investigations.