Ann Thorac Surg 2004;78:e9-e12
© 2004 The Society of Thoracic Surgeons
Case report
Use of LeVeen pleuroperitoneal shunt for refractory high-volume chylothorax
Dipin Gupta, MDa,
Kerry Ross, BSa,
Valentino Piacentino, III, PhDb,
Dawn Stepnowski, CRNPa,
James B. McClurken, MDb,
Satoshi Furukawa, MDb,
Daniel T. Dempsey, MD*a
a Department of Surgery, Philadelphia, Pennsylvania, USA
b Division of Cardiac and Thoracic Surgery, Temple University School of Medicine, Philadelphia, Pennsylvania, USA
Accepted for publication December 12, 2003.
* Address reprint requests to Dr Dempsey, Department of Surgery, Temple University School of Medicine, 3401 N Broad St, Philadelphia, PA 19140, USA
e-mail: daniel.dempsey{at}temple.edu
 |
Abstract
|
|---|
We present a case of intractable high-volume (> 2L/d) chylothorax after transhiatal esophagectomy treated successfully with the simultaneous insertion of both Denver (Denver Biomedical, Golden, CO) and LeVeen (Becton-Dickinson, Rutherford, NJ) pleuroperitoneal shunts. The patient initially had chemoradiotherapy for a T4N1 squamous cell carcinoma of the thoracic esophagus. Re-staging showed a dramatic shrinkage of tumor, and a transhiatal esophagectomy was performed. Sequential bilateral thoracotomies were performed on postoperative days 19 and 26 for attempted control of high-volume chylothorax, but these were unsuccessful. Subsequent pleuroperitoneal shunt insertion was used, which immediately controlled the effusion. A shunt study was performed shortly after hospital discharge, which showed an occluded Denver shunt and a patent LeVeen shunt. The patient succumbed to metastatic carcinoma 18 months after discharge, but no pleural effusion had recurred.
|
Introduction
|
|---|
High-volume chylothorax after cardiothoracic surgery can be a life-threatening complication. Persistent drainage of lymph fluid may cause malnutrition, fluid and electrolyte abnormalities, and profound suppression of immune responses. Inadequate drainage of rapidly accumulating lymph fluid may also cause pulmonary compromise. Nonoperative therapy, consisting of total parenteral nutrition, octreotide, and low-fat diet (or nothing by mouth), is usually curative in low-output situations; pleurodesis may be helpful in selected cases. Aggressive intervention is recommended for the more dangerous high-output lymph leaks. Surgical options include early reexploration and lymphatic or thoracic duct ligation, translymphatic embolization, and active pleuroperitoneal shunt (ie, Denver pleuroperitoneal shunt [Denver Biomedical]). We present a case of intractable high volume chylothorax after esophagectomy in an irradiated field, which was successfully treated with the simultaneous insertion of both an active (Denver pleuroperitoneal shunt [Denver Biomedical]) and a passive (LeVeen pleuroperitoneal shunt [Becton-Dickinson, Rutherford, NJ]). Rapid resolution of the chylothorax ensued with clinical and radiologic evidence that the passive shunt was the major contributor to success in this patient.
A 55-year-old man who was a smoker was referred to our institution for management of a mid-third esophageal squamous cell carcinoma. Endoscopic ultrasound and chest computed tomographic scan demonstrated invasion into the aorta with peri-hilar adenopathy and enlarged celiac lymph nodes (T4N1MX). He was treated with a course of cisplatin and 5-fluorouracil, as well as radiotherapy (41Gy). A repeat chest computed tomographic showed no evidence of tumor (Fig 1).
Based on the complete response to chemoradiotherapy and the patient's favorable body habitus, initially our operative plan was to attempt a transhiatal esophagectomy and resort to an Ivor-Lewis technique if resection was not bluntly possible. During exploration we noted dense mediastinal fibrosis, but no visible tumor. We were able to complete the dissection through a transhiatal approach, and we performed a cervical esophagogastrostomy and feeding jejunostomy. Pathologic analysis demonstrated no residual tumor in the esophagus and clear resection margins. Three of seven peri-esophageal lymph nodes and two of eight left gastric lymph nodes contained metastatic carcinoma.
View larger version (83K):
[in this window]
[in a new window]
|
Fig 1. Computed tomographic scan after neoadjuvant therapy shows esophageal mass and peri-hilar adenopathy. Arrow depicts plane between aorta and esophageal mass.
|
|
Within 24 hours of operation, a large left pleural effusion in the patient had developed, which required insertion of a left tube thoracostomy that had initially drained 2.3 L. Worsening hypoxia and tachypnea necessitated elective reintubation on postoperative day 3. No ascites was present. Thoracostomy output continued to exceed 3.0 L per day, and chemical analysis of tube output confirmed chylothorax (pleural fluid TG, 176 mg/dL). Total parenteral nutrition was initiated and bowel rest continued.
The patient remained dependent on the ventilator and continued to drain at least 3.0 L per day from the left tube thoracostomy. On postoperative day 19, we performed a left thoracoscopy, thoracotomy, and tracheostomy. Chyle was seen emanating from multiple lymph channels in the upper mediastinum, but the thoracic duct was not identified. These lymphatics were ligated with hemostatic clips. Within 24 hours after this operation the left chest tube thoracostomy output was negligible, but insertion of a right tube thoracostomy was required for a new right pleural effusion (Fig 2).
Despite continued bowel rest and parenteral nutrition, right thoracostomy output remained at least 2.0 L per day. Seven days later the patient underwent right thoracotomy and further suture ligation of lymphatics. Use of tissue adhesives, Floseal (Fusion Medical Technologies, Fremont, CA) and Tisseel (Baxter Corp, Mississauga, Ontario, Canada) were also used. After the right thoracotomy was done, the right thoracostomy output increased to between 3.0 and 4.0 L per day. Serum albumin content was 1.7 mg/dL despite daily intravenous administration of 25 g.
View larger version (107K):
[in this window]
[in a new window]
|
Fig 2. New large right pleural effusion immediately after left thoracotomy and attempted ligation of thoracic duct leak.
|
|
After 8 additional days of continued high-volume right thoracostomy output, the patient then underwent placement of both a right Denver pleuroperitoneal shunt (Denver Biomedical) and a right LeVeen pleuroperitoneal shunt (Becton-Dickinson) under general anesthesia. The chest tube was removed at this time. The Denver shunt (Denver Biomedical) was pumped by the patient, house officers, and nursing staff approximately 500 times per day initially. Follow-up radiographic studies showed no pleural effusions. Enteral feeding with a medium-chain triglyceride formula was initiated. The patient was gradually weaned from the ventilator and was discharged 3 weeks later. At the time of discharge, he was tolerating an oral diet and was receiving supplemental enteral tube feedings. At that time, the shunt was pumped approximately 100 to 200 times per day.
At the 2-week follow-up the patient was doing well with no clinical evidence of pleural effusion. He denied pumping the Denver shunt (Denver Biomedical) after leaving the hospital. A shunt study performed 1 month after discharge showed no flow through the Denver shunt (Denver Biomedical) and a patent LeVeen shunt (Becton-Dickinson).
Six months after the esophagectomy, a new solitary liver metastasis was discovered, and chemotherapy using Carboplatin (Bristol-Myers Squibb, New York, NY) and Taxotere (Aventis Pharma, Ltd, UK) was administered. Eighteen months after the esophagectomy, the patient succumbed to progression of distant metastases. At that time he was eating and had no significant pleural effusion (Fig 3).
View larger version (119K):
[in this window]
[in a new window]
|
Fig 3. Follow-up chest roentgenogram 18 months after shunt insertion shows no recurrent pleural effusion. Thin arrow depicts LeVeen shunt (Becton-Dickinson, Rutherford, NJ). Thick arrow delineates Denver shunt (Denver Biomedical, Golden, CO).
|
|
|
Comment
|
|---|
This unfortunate transhiatal esophagectomy patient had a high output of chylothorax for 5 weeks, despite attempted treatment with a left thoracotomy, right thoracotomy, and bowel rest. Simultaneous placement into the draining right chest of both an active (Denver [Denver Biomedical]) and a passive (LeVeen [Becton-Dickinson]) pleuroperitoneal shunt resulted in immediate resolution of the chylothorax. Of interest, the clinical evidence strongly suggests that the passive LeVeen shunt (Becton-Dickinson), which we believe has not been previously reported to be useful as a pleuroperitoneal shunt, was the more effective method of treatment for this intractable high output chylothorax. The evidence to suggest this is twofold. First, the active Denver shunt (Denver Biomedical) was not routinely pumped more than 500 times per day, which would have evacuated only 1 L per day from the pleural space in a patient producing at least 3 L per day. After discharge, the Denver shunt (Denver Biomedical) was never pumped, yet no effusion recurred. Second, the radiologic study that we performed to assess shunt patency approximately 1 month after discharge showed that the passive LeVeen shunt (Becton-Dickinson) remained patent but the active Denver shunt (Denver Biomedical) had become occluded.
Our rationale for using two shunts simultaneously was to provide an adjunctive pathway for pleural fluid evacuation if the active Denver shunt (Denver Biomedical) (2 mL per pump cycle) was unable to control the high-volume of pleural fluid, or became occluded, or was not pumped as instructed. The LeVeen shunt (Becton-Dickinson) was not placed alone in the pleuroperitoneal position due to its inflow requirement of a +3cm H2O for valve opening. Despite this characteristic, we found the LeVeen shunt (Becton-Dickinson) to be functional in the pleuroperitoneal position, even after mechanical positive pressure ventilation was discontinued 6 days after insertion. The Denver shunt (Denver Biomedical) was not used and had become occluded within 1 month of hospital discharge.
This patient had multiple risk factors for thoracic ductal injury at resection. We encountered an unusual amount of mediastinal fibrosis that may have increased the risk of postoperative lymph leak. Despite prior supplemental enteral feeding during administration of chemotherapy, the patient remained relatively malnourished. Tissues were found to be very edematous and the thoracic duct was not well visualized at the time of esophagectomy. Some authors have suggested administering methylene blue or cream through the nasogastric tubes as adjunctive measures to increase chyle flow and to make the thoracic duct more visible during resection. Dougenis and colleagues [1] suggested routine ligation of the thoracic duct, but we have not used this technique. Bolger and colleagues [2] have suggested that the transhiatal approach itself is associated with a higher incidence of postoperative chylothorax than the transthoracic approaches. However, Orringer and colleagues' [3] last series update showed a chylothorax incidence of less than 1% after 1,085 transhiatal resections.
This case demonstrates therapy of an iatrogenic chylothorax with use of multiple techniques after failure of nonoperative therapy. Ferguson and colleagues [4] reported spontaneous resolution of postoperative chylothorax in only 3 of 13 patients treated conservatively (including chest drainage, reduction of chyle flow, and supplemental nutrition) on days 12, 25, and 42. Milsom and colleagues [5] reported a mortality rate of 83% in 6 patients treated conservatively and 14% in 14 patients treated operatively. Despite our initial attempts of conservative management, this patient continued with high thoracostomy output and mechanical ventilation requirement for 19 days until reexploration. Positive pressure ventilation alone has been found to assist with rapid resolution of chylothorax in an infant after cardiotomy [6], but we did not find this to be helpful. Orringer and colleagues [7] recommended early reexploration and thoracic duct ligation after encountering no mortality and an average hospital stay of 11 days in a group of 11 patients with postoperative chylothorax. Of note, their series includes 1 patient who underwent "prophylactic" thoracic duct ligation at the time of resection because of extensive tumor involvement with peri-esophageal tissues. Nonetheless this patient had a chyle leak develop from a suture hole in the cisterna chyli and required reexploration. Cope and associates [8] attempted percutaneous cannulation of the thoracic duct in 11 patients with chylothorax. Of the five patients (45%) in whom the thoracic duct was cannulated, 4 underwent successful coil embolization.
Weese and Schouten's [9] initial success with the use of a pleuroperitoneal shunt for the treatment of a malignant pleural effusion has been confirmed by other authors [10–12] using the Denver shunt (Denver Biomedical). In a similar manner, Little and colleagues [13], Murphy and colleagues [14], and Milsom and colleagues [5] have reported success with the Denver shunt (Denver Biomedical) in patients with chylothorax. Little and colleagues' [13] series includes 29 patients of whom 5 had poor results, 4 had temporary benefit, and 20 had excellent results. The only 2 patients with postoperative chylothorax in this series had resolution of their effusions and their shunts were removed. Murphy and colleagues's [14] experience includes 16 patients with chylothorax (6 were postoperative) who underwent shunt implantation, with 12 patients achieving excellent results. Ten of these shunts were removed. Milsom and colleagues [5] details 8 postoperative patients (only 1 adult) with chylothorax who were treated with this shunt, and 6 of these patients were cured.
Successful use of the LeVeen peritoneo-venous shunt (Becton-Dickinson) for pleural effusion associated with ascites has been described by LeVeen and colleagues [15] and Hobbs and colleagues [16]. They used the shunt to resolve simultaneous ascites and hydrothorax in patients with presumed diaphragmatic defects, and LeVeen and colleagues describe a concomitant thoracic chemical pleurodesis.
In conclusion, we summarize the use of a LeVeen shunt (Becton-Dickinson) in the pleuroperitoneal position in the treatment of intractable postesophagectomy high-volume chylothorax. We found this technique to be feasible, well-tolerated, and effective, and we recommend that it be considered as a last resort therapy in patients with high-volume chylothorax in whom thoracic duct ligation has failed.
|
References
|
|---|
- Dougenis D., Walker W.S., Cameron E.W., et al. Management of chylothorax complicating extensive esophageal surgery. Surg Gynec Obstet 1992;174:501-506.
- Bolger C., Walsh C.N., Tanner W.A., et al. Chylothorax after oesophagectomy. Br J Surgery 1991;78:587-588.[Medline]
- Orringer M.B., Marshall B., Iannettoni M.D. Transhiatal esophagectomy: clinical experience and refinements. Ann Surg 1999;230(3):392-400.[Medline]
- Ferguson M.K., Little A.G., Skinner D.B. Current concepts in the management of postoperative chylothorax. Ann Thorac Surgery 1985;40:542-545.[Abstract]
- Milsom JW, Kron IL, Rheuban KS, et al. Chylothorax: an assessment of current surgical management. J Thorac Cardiovasc Surg 1985;89:221–7
- Ragosta K.G., Alfieris G. Chylothorax: a novel therapy. Crit Care Med 2000;28(4):1208-1209.[Medline]
- Orringer M.B., Bluett M., Deeb G.M. Aggressive treatment of chylothorax complicating transhiatal esophagectomy without thoracotomy. Surgery 1988;104:720-726.[Medline]
- Cope C., Salem R., Kaiser L.R. Management of chylothorax by percutaneous catheterization and embolization of the thoracic duct: prospective trial. J Vasc Interv Radiol 1999;10(9):1248-1254.[Medline]
- Weese J.L., Schouten J.T. Pleural peritoneal shunts for the treatment of malignant pleural effusions. Surg Gynecol Obstet 1982;154:391-392.[Medline]
- Cimchowski G.E., Joyner L.R., Fardin R., et al. Pleuroperitoneal shunting for recalcitrant pleural effusions. J Thorac Cardiovasc Surg 1986;92:866-870.[Abstract]
- Little A.G., Ferguson M.K., Golomb H.M., et al. Pleuroperitoneal shunting for malignant pleural effusions. Cancer 1986;58:2740-2743.[Medline]
- Tsang V., Fernando H.C., Goldstraw P. Pleuroperitoneal shunt for recurrent malignant pleural effusions. Thorax 1990;45:369-372.[Abstract/Free Full Text]
- Little A.G., Kadowaki M.H., Ferguson M.K., et al. Pleuroperitoneal shunting -alternative therapy for pleural effusions. Ann Surg 1988;208(4):443-450.[Medline]
- Murphy M.C., Newman B.M., Rodgers B.M. Pleuroperitoneal shunts in the management of persistent chylothorax. Ann Thorac Surg 1989;48:195-200.[Abstract]
- LeVeen H.H., Piccone V.A., Hutto R.B. Management of ascites with hydrothorax. Am J Surg 1984;148:210-211.[Medline]
- Hobbs C.L., Mullen J.L., Rosato E.F. Peritoneovenous shunt for hydrothorax associated with ascites. Arch Surg 1982;117:1233-1234.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
J. E. Heffner and J. S. Klein
Recent Advances in the Diagnosis and Management of Malignant Pleural Effusions
Mayo Clin. Proc.,
February 1, 2008;
83(2):
235 - 250.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
