|
|
||||||||
Ann Thorac Surg 2003;78:689-691
© 2003 The Society of Thoracic Surgeons
a Division of Cardiothoracic Surgery, Columbia University, College of Physicians and Surgeons, New York, New York, USA
Accepted for publication June 23, 2003.
* Address reprint requests to Dr Naka, Division of Cardiothoracic Surgery, Columbia University, College of Physicians and Surgeons, Milstein 7-435, 177 Fort Washington Ave, New York, NY 10032, USA
e-mail: yn33{at}columbia.edu
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
A 55-year-old male (Case 1 in Table 1) underwent the HeartMate (Thoratec Corporation, Pleasanton, CA) vented electric left ventricular assist device (LVAD) implantation for chemo- and radiation-induced dilated cardiomyopathy for Hodgkin's disease. He was extubated on postoperative day 4. On postoperative day 5, a sudden decrease in his LVAD flows from 5.0 L/min to 3.5 L/min was noted. Bleeding from the previous chest tube and the LVAD driveline exit sites was noticed as well as bloody fluid in the LVAD pocket drains. He became hypotensive and the LVAD flow remained low despite volume resuscitation. The chest roentgenogram revealed an increased right pleural effusion. The hematocrit dropped from 33.1% to 22.1%. Reexploration of the chest was performed.
|
Table 1 summarizes the characteristics of the 5 patients. Past medical histories were unremarkable. Only one patient (Case 3) had previous cardiac surgery (coronary artery bypass) and one (Case 5) had a history of heparin-induced thrombocytopenia (HIT). All patients were on aspirin and one patient (Case 2) was on an additional intravenous heparin drip for a small cerebrovascular accident. The laboratory tests for coagulation were unremarkable in most of the patients (Table 1). No patient required right ventricular assist device implantation.
Clinical manifestations included decreased LVAD flow (4 patients), hypotension (3 patients), and bleeding from the LVAD driveline or previous chest tube sites (2 patients). Hemothorax or increased right pleural effusion were noted by chest roentgenogram in 4 patients and transthoracic echocardiography was effective in identifying a pericardial effusion in 3 patients. A decrease in hematocrit was observed in all cases (from 30.3% ± 3.9% to 20.4% ± 1.1%, p < 0.005 by paired Student's t test).
In 4 patients, in whom bleeding occurred from 5 to 15 days after surgery, reoperative median sternotomy was chosen for reexploration because no minimal adhesions were expected. In contrast, for one patient whose bleeding occurred after 69 days of LVAD implantation, a right anterolateral thoracotomy was chosen because of the adhesions and difficulty of dissection via median sternotomy.
In 3 patients, the distal branch of the RIMA, in the space of Larre, was the site of bleeding, whereas in 2 patients, the middle portion (at the second or third intercostal space) of the RIMA was the site of bleeding. It was evident, on inspection, that the repeated friction with the LVAD outflow graft caused erosion to the RIMA itself or its distal branches.
The RIMA or its branches were ligated, clipped, or cauterized, depending upon the size of the vessel, thus achieving satisfactory hemostasis. In one case, in which right thoracotomy was performed, the outflow graft was wrapped with bovine pericardium to prevent further erosion of the graft to the vessel. In all cases, the chest was closed primarily and there was no recurrence of bleeding.
All 4 patients, in whom bleeding occurred in early postoperative period (5 to 15 days), underwent heart transplantation and were discharged home.
One patient (in whom bleeding occurred after 69 days of LVAD implantation) did not survive to heart transplantation. As mentioned previously, this patient had HIT and experienced significant postoperative bleeding after LVAD implantation, requiring massive blood product transfusion with resultant prolonged postoperative course including renal and hepatic failure. Due to the sepsis and profound multiorgan failure, therapy was withdrawn 139 days after implantation of the LVAD. However, this failure to survive to heart transplantation was not related specifically to the bleeding episode from the RIMA, which occurred 70 days before the withdrawal from support.
| Comment |
|---|
|
|
|---|
Although it was evident that the repeated friction with the LVAD outflow graft caused erosion into the RIMA itself or its distal branches, it is not clear why this complication occurred only in these patients. The body surface area (BSA) of the five patients ranged from 1.72 to 2.00 m2 (mean 1.82). Therefore, these patients were not smaller than the average HeartMate LVAD recipient. It is known that the thoracic index (ratio of posteroanterior diameter to lateral diameter of the chest) is smaller in females than in males [5]. Since all five patients were male, the shape of the thorax may have contributed; however,
-square test analysis revealed that this was not statistically different (5/210 vs 0/49, p = 0.28).
The length of the outflow graft could also have contributed to this complication. If the graft is oversized, a sharp curvature or more movement during the device systole may cause repetitious erosion of the tissue overlying the RIMA. Finally, 4 out of 5 incidents occurred after patients were transferred from the ICU and had begun to ambulate. Therefore, increased activities might have changed the position of the device and outflow graft, subsequently causing mechanical abrasion of the RIMA. Although we do not have a definitive solution to prevent this complication, care must be taken in selecting the proper length of the outflow graft and in confirming the position of the outflow graft before chest closure.
Although this is a rare complication, erosion of the RIMA with subsequent bleeding is potentially lethal if not treated expeditiously. Therefore, it is important for the medical personnel to be vigilant in preventing this complication. Once identified, surgical hemostasis (ie, chest reexploration) should be performed immediately.
In conclusion, bleeding from the RIMA in HeartMate LVAD recipients occurred in the relatively early postoperative period. The initial manifestations were a sudden decrease in LVAD flow, hypotension, bleeding from the LVAD driveline or chest tube sites, and a drop in hematocrit. Chest roentgenogram and transthoracic echocardiography were effective in identifying hemothorax and cardiac tamponade. If identified and treated in a timely fashion, the prognosis of these patients is not altered by this complication.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Guha, S. Munjampalli, V. Bandi, M. Loebe, G. Noon, and W. Lunn Pleural Effusion After Ventricular Assist Device Placement: Prevalence and Pleural Fluid Characteristics Chest, August 1, 2008; 134(2): 382 - 386. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Chumnanvej, M. J. Wood, T. E. MacGillivray, and M. F. V. Melo Perioperative Echocardiographic Examination for Ventricular Assist Device Implantation Anesth. Analg., September 1, 2007; 105(3): 583 - 601. [Abstract] [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 |