|
|
||||||||
Ann Thorac Surg 2001;72:1311-1315
© 2001 The Society of Thoracic Surgeons
a MCP/Hahnemann University, Philadelphia, Pennsylvania, USA
Accepted for publication June 18, 2001.
Address reprint requests to Dr Thomas, Acute Stroke and Neurointensive Care Program, MCP/Hahnemann University, 15th and Vine Sts, M.S. 423, Philadelphia, PA 19102
e-mail: carole.e.thomas{at}drexel.edu
| Abstract |
|---|
|
|
|---|
Methods. Retrospective chart review of all patients who had LVADs implanted at our hospital from September 1993 until September 1997. Complications from the time of implantation until heart transplantation or death and functional outcome were assessed.
Results. Twenty-three patients had LVADs placed in the four-year period. Of 23 patients, 9 had neurologic complications after placement of the LVAD. These included four strokes, three seizures, and two cases of delirium. The 3 patients with seizures all died from multiorgan failure. All of the patients with strokes received a transplant. One patient with delirium died from multiorgan failure and another received a transplant. The most devastating medical complication was renal failure, which occurred in 7 patients and was associated with 100% mortality. All surviving patients with neurologic complications went on to transplant and good functional outcome.
Conclusions. Neurologic complications are common in patients with LVADs, occurring in 9 out of 23 patients in our series. Seizures are a poor prognostic indicator and were associated with 100% mortality. Strokes did not have a negative impact on outcome. Patients with delirium had a mixed outcome, which reflects the multifactorial nature of delirium. Further study needs to be done to limit the neurologic complications associated with LVADs and further improve outcomes.
| Introduction |
|---|
|
|
|---|
Although LVADs are an accepted "bridge to transplant" in many patients, there is ongoing concern about the morbidity associated with these devices [1]. Patients who have devastating neurologic complications while awaiting heart transplantation may no longer be eligible for transplant. This article addresses the neurologic complications that occur with LVADs, the risk factors for these complications, and what impact these complications have on the clinical outcome in a series of 23 patients.
| Material and methods |
|---|
|
|
|---|
Charts were reviewed for demographic information, general medical complications and neurologic complications from the use of LVADs. The demographic information obtained was: age, sex, race ethnicity; past medical history; past cardiac history; history of neurologic or psychiatric diseases; base line pre-LVAD history including etiology of heart failure, echocardiogram results, right heart catherization results, and medications; operating room records of events during LVAD placement; and post-LVAD laboratory tests, medications, and complications at 2-week intervals until cardiac transplant or death.
General medical complications of LVADs included: bleeding (requiring transfusion or causing hemodynamic instability), renal failure (creatinine doubling in a 24-hour period, with oliguria), respiratory failure (requiring intubation), infection (positive cultures), and adult respiratory distress syndrome (diffuse pulmonary infiltrates and hypoxemia resistant to supplemental oxygen).
Neurologic complications recorded included: stroke (computed tomograph of the brain consistent with ischemic or hemorrhagic lesion), transient ischemic attack (clinical findings of stroke that cleared within 24 hours and no brain lesion on computed tomography), seizures (electroencephalograph consistent with epileptiform activity or clinical findings), and delirium (depressed mental status and disorientation).
The left ventricular assist device used in our patients was the Novacor left ventricular assist device (LVAD). This is an electrically powered cardiac assist system that functions as a parallel chamber to the left ventricle. It provides circulatory assistance in the presence of left ventricular failure. It has a typical output of 5 to 7 L/min and a maximal stroke volume of 70 ml [5]. Implantation of the Novacor LVAD is accomplished under general anesthesia through a median sternotomy incision with the aid of cardiopulmonary bypass. The apex of the left ventricle is cored out, which permits the placement of a large cannula into the cavity. The other end of the cannula is inserted into the aorta. Deairing maneuvers are performed, which are critical to prevent air embolization. These maneuvers involve careful inspection for macroscopic air bubbles in the circuit itself, placement of the patient in Trendelenberg position when initiating LVAD pumping to decrease microscopic air bubbles to the brain, and venting the aortic root during LVAD ejections to capture any remaining emboli.
All patients with LVADs are anticoagulated by protocol. Within 12 hours of LVAD placement the patient is given lowmolecular weight dextran intravenously (IV). When the bleeding from chest drains is less than 30 mL/h for three consecutive hours (usually within 12 to 72 hours), a continuous IV heparin drip is started with a target partial thromboplastin time (PTT) of 1.5 to 2.0 times the control. Once the target PTT is achieved the dextran is discontinued. When the chest tubes are removed, administration of 81 mg of aspirin is begun. When the patient can take oral administration of medications, Coumadin (Du Pont Pharmaceuticals, Wilmington, DE) is started to achieve an INR of 2.0 to 3.5. Once the target INR is reached, heparin is discontinued and the patient remains on long-term Coumadin and aspirin.
| Results |
|---|
|
|
|---|
Of the 23 patients, 9 had neurologic complications. These included four strokes, three seizures, and two episodes of delirium. All of the patients with stroke went on to transplant and had good outcomes (either home independent or acute rehabilitation). All of the patients with seizures died. One patient with delirium died and another went on to transplant and short-term rehabilitation
Neurologic complications
Strokes
All 4 patients with strokes underwent computerized tomography or magnetic resonance imaging of the brain, echocardiography, and carotid Doppler examinations. Two patients were found to have heparin-induced thrombocytopenia by serotonin assay. Both of these patients had infarctions of a branch of the right middle cerebral artery, leaving them with mild weakness of the left face and arm. The third patient had an infarction in the midbrain, producing a Webers syndrome (third nerve palsy and contralateral hemiparesis). The fourth patient had atrial fibrillation when his stroke occurred; he had hemiparesis of the left face, arm, and leg from an infarction of the right internal capsule. Both the third and the fourth patients had subtherapeutic anticoagulation on Coumadin with INRs of 1.5 and 1.7 respectively. Strokes occurred from 4 to 32 days after LVAD placement.
Seizures
All seizures occurred within 24 hours of LVAD placement. One patient had a bleeding diathesis immediately after surgery, requiring 14 U of blood. He had a prolonged period of hypotension, began having seizures, and did not regain consciousness before he died. The second patient had a seizure before LVAD insertion that was thought to be secondary to hypoxic encephalopathy from cardiogenic shock. He went into status epilepticus in the postoperative period and did not recover. The third patient had a seizure in the postoperative period and went on to renal failure, septic shock, multisystem organ failure, and death. All 3 patients had generalized tonicclonic seizures without obvious focality. Although all 3 patients had electroencephalograms to document their seizures, none of the patients were able to have imaging studies after their seizures because of medical instability. Post mortem examinations were not performed on these patients.
Delirium
One patient was found to have a change in mental status and was encephalopathic 24 hours after LVAD placement. He progressed to multisystem organ failure and death. The other patient had pneumonia and pancreatitis at 2 weeks and was encephalopathic, a condition that improved when the underlying illnesses were treated. He went on to transplant.
Risk factors for neurologic complications
Two patients with stroke had heparin-induced thrombocytopenia (HIT) detected by serotonin assay. While the majority of patients (18 of 23) had thrombocytopenia at some point in their postoperative period, only those with thromboembolic complications were tested for HIT. Neither of the patients with HIT had systemic thromboembolic events. The other two patients with stroke had subtherapeutic anticoagulation, with INRs of 1.5 and 1.7. One of these patients also had atrial fibrillation. None of the stroke patients had significant carotid stenosis or prior evidence of left atrial or ventricular clot. Patients with seizures had evidence of multisystem organ failure including renal insufficiency, elevated liver enzymes, and pulmonary edema. There were no other clear-cut risk factors for seizures. Delirium was associated with a variety of toxic and metabolic derangements including pneumonia and renal insufficiency.
General medical complications
The most common medical complication was bleeding, occurring in 16 out of 23 patients (70%). Infections were the next most frequent (9 of 23, 40%), followed by renal failure (7 of 23, 30%). The medical complications were evenly distributed among patients who survived and went on to transplant and patients who died, with the exception of renal failure. Patients with this complication had a 100% mortality rate.
Timing of neurologic and medical complications
We divided the neurologic and medical complications into early (< 14 days after LVAD placement), and late (
14 days after LVAD placement). All but 1 patient had some complication while using the LVAD. Fifteen patients had early complications, with the majority (11 of 15, 73%) occurring exclusively within 24 to 48 hours of LVAD placement. This group of patients had a mortality rate of 67%. Among the 8 patients with late complications, only 1 died (12% mortality). The remainder went on to heart transplant. Patient data is summarized in Table 1.
|
| Comment |
|---|
|
|
|---|
Cerebral thromboembolism is the most common neurologic complication reported, ranging from 3% to 47% in different series [19]. Inadequate anticoagulation has been sited as the cause of this complication in most cases [7, 10, 11], although there are no reports to date describing the role of heparin-induced thrombocytopenia in these patients. More recent studies have reported a significantly reduced rate of stroke using advanced-design LVADs, which do not require anticoagulation for stroke prevention [9].
Our series found stroke to be the most common neurologic complication among patients with LVADs. Two of the 4 patients with stroke had clear-cut, heparin-induced thrombocytopenia diagnosed by serotonin assay. We believe that this entity needs to be considered in any patient with thrombocytopenia and thromboembolic events after LVAD placement. Our other 2 patients were found to have inadequate anticoagulation profiles when their strokes occurred. LVADs are known to activate the coagulation cascade, resulting in device-related thrombus formation. Unstable thrombus exposed to the shearing force of blood flow predisposes to thromboembolic events including stroke and end-organ or extremity ischemia [9]. LVADs that have less thrombogenic potential have decreased this complication, and are being used with increased frequency in heart failure patients awaiting transplant.
This study looked at other neurologic complications associated with LVADs. We found seizures were by far the most devastating neurologic complication, with a 100% mortality rate in our patients. We suggest that seizures are an indication of the overall viability of the patient. They occurred in patients with multisystem organ failure in all three instances. There were no cases of isolated seizures in an otherwise hemodynamically stable patient. It is likely that death was caused by hypoperfusion of all organs from either pump failure or hypovolemia. The seizures reflect this extreme hypoperfusion in all vascular beds, including the brain. In 2 of the patients there were signs of multiorgan failure before LVAD placement. It is possible that earlier placement of the LVAD may have resulted in preservation of organ perfusion and better outcomes.
Another neurologic complication in 2 of our patients was delirium. This was an indication that toxic or metabolic derangements were occurring, causing changes in mental status. These alterations in homeostasis included hypoperfusion in 1 patient who died 1 day after LVAD implantation. Delirium can also reflect more subtle problems that are potentially reversible, as in the case of the second patient.
Medical complications associated with LVADs were fairly common, occurring in all but 1 of our patients. Bleeding was our most common medical complication, a finding that was consistent with other studies [12]. Clearly the most devastating medical complication was renal failure. All patients who experienced acute renal failure after LVAD placement also had multisystem organ failure and died. This is another indication of a lack of perfusion of the end organs; either from pump failure or hypovolemia from massive hemorrhage. Several of these patients survived long enough to have dialysis but later died.
Other studies have found that the rate of complications increase with the length of time the LVAD is in place [3]. Our series of 23 patients had the most devastating complications in the first 14 days after LVAD placement, with most occurring in the first 24 to 48 hours. Seven of the 8 patients who had complications after 14 days went on to heart transplantation. This is encouraging news for the longer-term or permanent use of LVADs.
It is clear that LVADs can have an enormous impact on the survival and viability of heart failure patients. As the technology improves the complications of these devices will hopefully diminish. LVADs are now being used as permanent implants in experimental protocols. More data will be gathered about neurologic complications as experience is gained with these devices.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. R. Wilson, M. M. Givertz, G. C. Stewart, and G. H. Mudge Jr Ventricular assist devices the challenges of outpatient management. J. Am. Coll. Cardiol., October 27, 2009; 54(18): 1647 - 1659. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. A. Genovese, M. A. Dew, J. J. Teuteberg, M. A. Simon, J. Kay, M. P. Siegenthaler, J. K. Bhama, C. A. Bermudez, K. L. Lockard, S. Winowich, et al. Incidence and patterns of adverse event onset during the first 60 days after ventricular assist device implantation. Ann. Thorac. Surg., October 1, 2009; 88(4): 1162 - 1170. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Tsukui, A. Abla, J. J. Teuteberg, D. M. McNamara, M. A. Mathier, L. M. Cadaret, and R. L. Kormos Cerebrovascular accidents in patients with a ventricular assist device J. Thorac. Cardiovasc. Surg., July 1, 2007; 134(1): 114 - 123. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Knez and B. Rigler Invited commentary Ann. Thorac. Surg., May 1, 2007; 83(5): 1690 - 1690. [Full Text] [PDF] |
||||
![]() |
R. M. Lazar, P. A. Shapiro, B. E. Jaski, M. K. Parides, R. C. Bourge, J. T. Watson, L. Damme, W. Dembitsky, J. D. Hosenpud, L. Gupta, et al. Neurological Events During Long-Term Mechanical Circulatory Support for Heart Failure: The Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH) Experience Circulation, May 25, 2004; 109(20): 2423 - 2427. [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 |