|
|
||||||||
Ann Thorac Surg 2005;80:1864-1870
© 2005 The Society of Thoracic Surgeons
a Department of Pediatrics, Division of Pediatric Cardiology, University of Toronto, The Hospital for Sick Children, Toronto, Ontario, Canada
b Department of Surgery, Division of Cardiovascular Surgery, University of Toronto, The Hospital for Sick Children, Toronto, Ontario, Canada
Accepted for publication April 26, 2005.
* Address correspondence to Dr McCrindle, The Hospital for Sick Children, 555 University Ave, Toronto, ON M5G 1X8, Canada (Email: brian.mccrindle{at}sickkids.ca).
| Abstract |
|---|
|
|
|---|
METHODS: Hospital databases were used to identify chylothorax cases. Surgical databases were used to identify all patients undergoing cardiothoracic surgery. Medical records were reviewed, including daily records of drainage volumes and management.
RESULTS: From September 2000 to December 2002, there were 48 cases of chylothorax in 1,257 surgeriesan incidence of 3.8% (95% confidence interval: 2.8% to 4.8%). Overall mortality rate was similar, but cases had longer postoperative hospital stays (median, 22 versus 8 days; p < 0.001). Incidence of chylothorax was significantly higher with heart transplantation and Fontan procedures. Diagnosis was made at a median of 6 days after surgery. Duration of drainage was a median of 15 days, with 11 patients draining more than 30 days. Longer duration of drainage was associated with cavopulmonary anastomosis procedures and longer time to diagnosis of chylothorax. Nutritional management included low fat diet, enteral feeds enriched with medium-chain triglycerides, and parenteral nutrition. Five patients were treated with octreotide, 4 with thoracic duct ligation, and 1 with pleurodesis. Octreotide was associated with a variable effect on drainage. Thoracic duct ligation reduced, but did not stop drainage.
CONCLUSIONS: Chylothorax increases duration of hospitalization after cardiovascular surgery in children. Early diagnosis may reduce the duration of chylothorax. Nutritional strategies remain the cornerstone for management of postoperative chylothorax. The impact of octreotide and surgical intervention is limited when reserved for patients with severe or prolonged drainage.
| Introduction |
|---|
|
|
|---|
Previous studies of chylothorax in children have been limited to issues regarding conservative versus surgical management and have not adequately determined the risk factors for postoperative chylothorax and factors affecting clinical course [1, 2, 5, 8, 9].We sought to determine the incidence, risk factors, clinical spectrum, and impact on outcomes of chylothorax in children undergoing cardiothoracic surgery, while further describing our experiences with treatment.
| Patients and Methods |
|---|
|
|
|---|
Measurements
The medical records for all patients were reviewed. Data collected included demographics, clinical history, surgical course, complications, and postoperative medical and surgical interventions. Serial data from their hospital stay were gathered from daily records, and included weight, fluid intake and output, dietary intake, laboratory investigations, medications, and chest tube drainage. Diagnostic criteria for chylothorax were applied as described by Buttiker and colleagues [4].
Data Analysis
Data are described as frequencies and medians with ranges as appropriate. The incidence of chylothorax was calculated together with 95% confidence limits (CL), and characteristics between patients with and without chylothorax were compared using
2 and Kruskal-Wallis analysis of variance testing. Factors associated with duration and volume of chylous drainage were sought using mixed linear regression analysis for repeated or serial measurements.
| Results |
|---|
|
|
|---|
|
The incidence relative to the surgical procedure for those procedures performed more than 20 times during the study interval are shown in Table 2, and represent procedures performed in 45 of the chylothorax patients. The remaining 3 patients had rare procedures, including repair of total anomalous venous return, mitral valve replacement and mitral valve replacement with Konno procedure. Surgical approach was by median sternotomy in 47 patients, with 1 having a thoracotomy. Median duration of cardiopulmonary bypass was 126 minutes (range, 0 to 369) and median aortic cross-clamp time was 61 minutes (range, 0 to 469).
|
Reoperations were required during hospital admission for 5 patients, and included 2 pacemaker insertions and 3 revision surgeries. Median duration of initial mechanical ventilation was 2 days (range, less than 1 to 38), with 14 patients requiring reintubation and 1 patient dying without ever being extubated.
Characteristics of Chylothorax
Chylous pleural effusions were bilateral in 26 patients (54%), with 7 patients (15%) having an additional pericardial effusion. Median time to diagnosis of chylothorax was 6 days (range, 1 to 30) after surgery. One patient did not have chest tubes in place at the time of diagnosis as fluid testing results were obtained by samples from pleurocentesis. This patient never required chest tube placement for chylous drainage. Median drainage on the day of diagnosis was 21 mL/kg (range, 2 to 314 mL/kg).
Diagnosis of the chylothorax was evident from laboratory testing of the fluid demonstrating the presence of chylomicrons in 43 patients (90%) and triglyceride concentration above 1.1 mmol/L in 33 (69%; median, 1.38 mmol/L.; range, 0.33 to 6.92 mmol/L). Additional testing of fluid cell counts in 35 patients showed total white blood cell counts above 1,000 cells/mm3 in 22 (63%; median, 1,753 cells/mm3; range, 204 to 21,500 cells/mm3), with a percentage of lymphocytes above 80% in 21 patients (60%; median 89%; range, 8% to 97%). Overall, all but 1 of the 48 patients had at least one of the four fluid abnormalities. Although the fluid analysis for 1 patient did not reach any of the cutpoints and the fluid was negative for chylomicrons, the triglyceride level was 0.83 mmol/L and the white cell count was 700 cells/mm3 with 78% lymphocytes.
The characteristics of chylous drainage from patients after cavopulmonary connection procedures were not significantly different from those after other surgeries (Table 3).
|
|
Chest tube reinsertion was required in 23 patients, in 9 after drainage supposedly stopped for a median of 7 days (range, 1 to 24); these included 1 patient who had a cardiac arrest as a result of reaccumulation.
Patient and surgical characteristics independently associated with duration of drainage (after logarithmic transformation to normalize the distribution) included cavopulmonary anastomosis procedures (cavopulmonary shunt or Fontan procedure; parameter estimate 0.97; p < 0.001) and greater time to diagnosis of chylous drainage (after logarithmic transformation; parameter estimate 0.37; p = 0.009). Patients who drained more than 30 days included all 4 patients who had Norwood I procedure, 3 of the 6 Fontan patients, 2 of the 4 heart transplantation patients, 2 of the 4 cavopulmonary shunt patients, 1 patient who had repair of ventricular septal defect, and 1 patient who had tetralogy of Fallot repair. Furthermore, the duration of chylothorax after cavopulmonary connection procedure was longer than after other surgeries (Table 3).
Amount of Drainage
The maximum amount of drainage occurred at a median of postoperative day 2 (range, 0 to 30) and with a median of 37 mL/kg (range, 11 to 672 mL/kg) on that day. Patient and surgical characteristics independently associated with greater maximum drainage (after logarithmic transformation) included only bilateral chest drainage (parameter estimate 0.7; p = 0.003); longer duration of drainage almost reached statistical significance (p = 0.055). The maximum amount of drainage for chylothorax after cavopulmonary connection procedure was not significantly different from other surgeries (Table 3).
Lower daily drainage was associated with an increased duration from surgery after inverse transformation, indicating an initially rapid then slower decrease in drainage amount. Higher daily drainage was associated with those with longer total duration of drainage (after logarithmic transformation) and patients who had the Norwood I procedure. The impact of therapy directed at reducing chylous drainage was sought using mixed linear regression analysis of daily drainage (Table 5). The use of octreotide was associated with a greater rate of decline in daily drainage.
|
| Comment |
|---|
|
|
|---|
A higher incidence of chylothorax was observed in heart transplantation and Fontan procedures in our study. Conceptually, heart transplantation is associated with increased trauma to the chest cavity and Fontan or cavopulmonary anastomosis procedures will elevate superior vena cava pressure, both resulting in higher risk for chylothorax. The systemic venous hypertension can cause a backup of pressure into the thoracic duct, resulting in increased chyle loss. This loss is consistent with our observations of longer duration of chylous drainage after these procedures.
Knowing that cavopulmonary anastomosis procedures have a higher risk of prolonged pleural drainage may indicate that earlier, more aggressive therapy is indicated for these patients. Further studies should be conducted to evaluate whether earlier intervention can reduce hospitalization duration and improve prognosis.
Beghetti and associates [1] reported an increased incidence of chylothorax with Blalock-Taussig shunt procedures. In this study, isolated Blalock-Taussig shunt procedures were not evaluated because too few procedures were performed in the time interval studied. However, Norwood I procedures were associated with an increased risk of chylothorax and higher daily drainage (Table 5).
Four patients had postoperative complications of venous thrombus in the innominate, jugular, or superior vena cava, all related to central venous lines. Despite the theoretical increase in systemic venous hypertension, their drainage duration was not distinctly lengthened from other patients at 13, 20, 20, and 61 days, respectively. However, further study with clear documentation of central venous pressures is needed to clarify the importance of this risk factor in chylothorax patients.
Presentation and Diagnosis
Diagnostic criteria for pediatric chylothorax have been defined by Buttiker and coworkers [4]. In our study, 90% of patient samples had chylomicrons, but fewer met the other diagnostic criteria. In some patients, that was due to less consistent testing for these other criteria. Poor enteral nutrition at the time of diagnosis may have also played a role in decreasing chylomicron and triglyceride levels in the pleural fluid [4], thereby contributing to delays in diagnosis. As longer time to diagnosis is correlated with increased drainage duration, there may be a role for early retesting of pleural fluid after enteral fat challenge to reduce the time to diagnosis in cases with a high index of suspicion for chylothorax.
Treatment of Chylothorax
To date, there have been no randomized controlled clinical trials to provide evidence for best management of chylothorax. The primary modes of treatment include pleural space evacuation, low-fat diets, medium chain triglyceride enriched feeds, enteric rest, and parenteral alimentation [1, 2, 1214]. The limited numbers of patients studied in each series precludes interpretations as to the protocol most effective for treatment [1, 4, 10, 12, 13].
Nutritional Management Strategies
The use of medium-chain triglyceride enriched diets is based upon the understanding that enterocytes directly absorb medium chain fatty acids into the circulation. That would theoretically allow patients to be supplied with adequate nutrition while reducing lymphatic flow to allow healing of the damaged lymphatic vessels. In our series, most patients (34 of 48, 71%) had resolution of their drainage with only changes to enteral nutrition. Although spontaneous recovery is possible, these data help substantiate the use of low fat or medium chain triglyceride enriched diets as first-line therapy.
A more rigorous mode of management is the use of total parenteral nutrition. This strategy was attempted for short periods for 11 patients in our series who had previously been managed with enteral diet change in this study. However, 5 of the 11 patients eventually went on to require further interventions, including octreotide or thoracic duct ligation. From these results, it may be reasonable to start patients on total parenteral nutrition and octreotide therapy concurrently after abandoning enteral changes to more achieve rapid control of the drainage or to expedite the decision to undertake a surgical intervention.
Octreotide, a somatostatin analog, was used in 5 patients in this series as a second-line therapy. Recently, successful treatment has been noted in some case reports for chylothoraces of various etiologies [1519], but its mechanism of action is not well understood. It has been postulated that octreotide can reduce leakage by increasing lymphatic vessel contraction [15]. The patients that were given octreotide in our study had variable responses to the treatment, with no overall decreases in drainage over the treatment period in 4 of the 5 patients (Table 4). With the weak contractility of lymphatics, it has been thought that octreotide may be more beneficial for use with mild to moderate flow chylothoraces [15]. In our study, the variable efficacy may be due to the late application of octreotide in management of patients with relatively higher daily drainage of chyle, and perhaps other subtle differences in patient characteristics not yet understood.
Surgical Management Strategies
The surgical management for postoperative chylothorax most commonly involves ligation of the thoracic duct, commonly through thoracotomy. Three of the 4 patients undergoing duct ligation in this series subsequently died of other causes. Chest tubes were still draining in 2 patients, and the remaining patient had a pleurodesis and further octreotide therapy before drainage was stopped. The high mortality rate of these patients seems to indicate that thoracic duct ligation was reserved for patients with poor prognosis.
In summary, octreotide and surgical interventions were not applied effectively to stop drainage in this study population. When reserved for patients with severe or prolonged drainage, the usefulness of these more aggressive therapies appears to be limited. Further studies are needed to elaborate an evidence-based protocol for the management of unremitting chylothorax. Until such evidence is available, we have drafted and instituted a "care map" (Fig 1) to standardize the diagnosis and management of chylothorax within our institution.
|
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. Nayar, R. Parmar, S. Kulkarni, and K. M. Cherian Treatment of Plastic Bronchitis Ann. Thorac. Surg., May 1, 2007; 83(5): 1884 - 1886. [Abstract] [Full Text] [PDF] |
||||
![]() |
S.-y. Chan, W. Lau, W. H.S. Wong, L.-c. Cheng, A. K.T. Chau, and Y.-f. Cheung Chylothorax in Children After Congenital Heart Surgery Ann. Thorac. Surg., November 1, 2006; 82(5): 1650 - 1656. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Talwar, S. K. Choudhary, and B. Airan Diaphragmatic Fenestration for Resistant Chlyothorax Ann. Thorac. Surg., August 1, 2006; 82(2): 767 - 768. [Full Text] [PDF] |
||||
![]() |
G. J. Pelletier Invited commentary Ann. Thorac. Surg., November 1, 2005; 80(5): 1870 - 1871. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |