| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 2004;127:548-554
© 2004 The American Association for Thoracic Surgery
Cardiopulmonary support and physiology |
a Third Department of Cardiac Surgery, Onassis Cardiac Surgery Center, Athens, Greece
b Department of Anesthesiology, Onassis Cardiac Surgery Center, Athens, Greece
c Coagulation and Hemostasis Laboratory, Onassis Cardiac Surgery Center, Athens, Greece
Read at the Eighty-third Annual Meeting of The American Association for Thoracic Surgery, Boston, Mass, May 4-7, 2003.
Received for publication April 29, 2003; revisions received August 13, 2003; accepted for publication August 18, 2003.
* Address for reprints: George M. Palatianos, MD, Onassis Cardiac Surgery Center, 356 Sygrou Avenue, 176 74 Athens, Greece
palatianos{at}otenet.gr
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionConclusionDiscussionReferences |
|---|
METHODS: Among 1518 patients who underwent cardiac surgery between June 1998 and May 2001, 32 (2.1%) presented with platelet counts less than 150,000/mm3 preoperatively or a history of prolonged (>3 days) intravenous exposure to heparin or both. These 32 patients were evaluated with an enzyme-linked immunosorbent assay for antibodies against heparin-platelet factor 4 complex. Platelets of patients with detected antibodies were tested with the prostacyclin analog iloprost for inhibition of heparin aggregation and determination of the inhibiting concentration and corresponding intravenous infusion rate of iloprost. Patients with antibodies received heparin after complete platelet inhibition with iloprost infusion. Hypotension was prevented or treated with intravenous noradrenaline. Ten randomly selected patients with similar preoperative characteristics, no previous extended exposure to heparin, and normal platelet counts served as controls.
RESULTS: Ten of the 32 patients (group A, 31.3%) and none of the controls had antibodies against heparin-platelet factor 4 complex. Patients in group A underwent surgery with iloprost (6-24 ng · kg-1 · min-1) and had their blood pressure maintained at greater than 95 mm Hg with norepinephrine infusion (1-4 µg · kg-1 · min-1). Operative mortality was zero. There were no thrombotic complications or bleeding requiring exploration. One patient in group A bled 1310 mL/6 hours but did not need exploration. There was no difference in postoperative blood loss and morbidity between groups. Platelet counts were reduced by 12.5% ± 8.7% (group A) and 38.1% ± 15.2% (control) (P < .001) 1 hour postoperatively and reached preoperative values by the fifth postoperative day.
CONCLUSIONS: Immune heparin-induced thrombocytopenia can be detected preoperatively among patients with a low platelet count or a history of prolonged heparin exposure or both. Cardiac surgery can be safely undertaken using iloprost-induced platelet inhibition during heparinization.
|
Heparin-induced thrombocytopenia and thrombosis (HITT), or heparin-induced thrombocytopenia type II, is an immunologic disorder that affects patients who receive prolonged (>3 days) heparin anticoagulation.1-3 When heparin enters the blood, it forms a complex with platelet factor (PF) 4. Patients with HITT have been sensitized against the heparin-PF4 complex, producing mainly immunoglobulin G antibodies against it.4 On reexposure to heparin, these antibodies bind to the heparin-PF4 complex at the platelet surface receptor Fc, causing platelet activation, release of procoagulant active microparticles and more PF4, and, consequently, platelet aggregation. The newly released PF4 reacts with the circulating heparin and eventually with the endothelial surface-attached heparin-like molecules, promoting reaction with the specific antibodies.2,3 This leads to activation of more platelets and to a vicious cycle that causes platelet aggregation, thromboembolic complications, and bleeding diathesis in association with thrombocytopenia.3,5 Patients with HITT antibodies are at high risk (
50%) for experiencing thromboembolic complications, bleeding, and sudden death on reexposure to heparin. Death, limb loss, myocardial infarction, coronary artery graft occlusion, mesenteric infarction, deep venous thrombosis, pulmonary embolism, stroke, intracardiac mural thrombosis, bilateral adrenal hemorrhagic infarcts, and bleeding have occurred intraoperatively or early postoperatively in patients with HITT receiving heparin.1,3,5-8
The management of patients with HITT antibodies who require cardiac surgery is challenging, because heparin anticoagulation is an integral part of cardiac operations performed with or without extracorporeal circulation. A standard approach to these patients has not yet been established, although several therapeutic options have been proposed on the basis of using anticoagulants other than heparin.1,3 However, experience with the Food and Drug Administration-approved alternative anticoagulants is limited, specific antidotes are not available, and special tests not readily available are required to monitor the effectiveness of the alternatives.3,5,6
In an effort to prevent serious perioperative thromboembolic and bleeding complications in our patients, we established a protocol for the identification and management of HITT-sensitive patients who need cardiac surgery with full heparinization. The purpose of this project was 2-fold: the preoperative recognition of patients with heparin-sensitive antiplatelet antibodies and the prevention of platelet activation by temporary platelet function inhibition during heparinization. Our management protocol was based on reports by Addonizio and associates,9,10 Kappa and collaborators,11,12 and Palatianos and colleagues (on the use of the prostacyclin analog iloprost [ZK 36374]).13 Addonizio and associates9,10 and Kappa and associates11,12 reported platelet preservation with iloprost during cardiopulmonary bypass (CPB) and showed that temporary platelet inhibition with iloprost allowed safe heparin administration in patients with HITT undergoing cardiac or vascular operations.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionConclusionDiscussionReferences |
|---|
Anesthesia protocol
Anesthesia was induced with midazolam (0.05-0.075 mg/kg of body weight), etomidate (0.3 mg/kg), and fentanyl (10-15 µg/kg). Neuromuscular block was induced by intravenous pancuronium or rocuronium (0.15 mg/kg) and maintained by continuous infusion of cisatracurium (1.5-2.0 µg · kg-1 · min-1). Anesthesia was maintained using sevoflurane 0.5% to 2.0% in oxygen/air and additional boluses of fentanyl as needed or a continuous infusion of remifentanil (0.2-1.0 µg · kg-1 · min-1).
Each patient had a 3-lumen central venous catheter, pulmonary artery catheter, and radial arterial catheter positioned before anesthesia induction. During surgery, there was continuous monitoring of the central venous, pulmonary artery, and arterial pressures; urinary output; electrocardiogram; and nasopharyngeal and rectal temperatures. Tissue oxygenation was monitored with pulse oximetry, and arterial blood gases were frequently tested.
Surgical protocol and postoperative management
After anesthesia induction, iloprost infusion was started at 3 ng · kg-1 · min-1 intravenously and was increased progressively (doubled) every 5 minutes up to the dose corresponding to the concentration of iloprost, which in vitro inhibited platelet aggregation to heparin. A HIPA test was performed 5 minutes after reaching the target dose. Once complete inhibition of platelet aggregation to heparin was demonstrated, the patients were given porcine heparin for full heparinization (heparin sodium, 300 IU/kg intravenously). Norepinephrine was infused at 0.05 µg · kg-1 · min-1 to prevent or treat hypotension (systolic blood pressure < 100 mm Hg). For cases requiring CPB, the activated coagulation time (ACT) was maintained at more than 480 seconds throughout perfusion. Additional heparin was administered if needed. CPB was established with a roller pump (Stöckert-Shiley Instruments, Munich, Germany), a hollow fiber membrane oxygenator (Quadrox, Jostra, Hirrlingen, Germany), and an arterial filter (HBF 40, Jostra) at 32°C to 33°C for cases undergoing coronary revascularization procedures and at 28°C for valve surgery cases. Cardioprotection was achieved with induced cardioplegic arrest. Cold (8°C) blood cardioplegic solution was infused retrograde through the coronary sinus and antegrade through the aortic root. Surface-heparinized extracorporeal circuits were not used in these patients. Off-pump procedures were performed using the Octopus 3 flexible stabilization system (Medtronic, Inc, Minneapolis, Minn) under systemic heparinization (heparin dose 150-200 IU/kg, ACT > 350 seconds). On termination of CPB, or after construction of the bypass grafts for the off-pump coronary bypass cases, protamine sulfate (Leo Pharmaceutical Products) was administered to neutralize the heparin and return the ACT to normal values. The patients were transferred to the intensive care unit (ICU), ventilated, continuously monitored for arterial and pulmonary artery pressures and electrocardiogram, and administered iloprost infusion. After demonstration of the absence of circulating heparin with the Hepcon HMS/HMS Plus blood analyzer (Medtronic, Inc), the iloprost infusion was reduced to half and was further halved every 30 minutes until it was discontinued after the first hour in the ICU. Iloprost was used solely for platelet inhibition according to the protocol dose. It was not modified to treat blood pressure changes. These conditions were treated as needed with the standard medications used in the immediate postoperative period. Blood losses from the chest tubes were recorded every hour.
Postoperative management was performed according to a standard protocol. If there was increased bleeding diathesis, red blood cells, platelets, and fresh frozen plasma were administered as needed to maintain the hematocrit above 26% in hemodynamically stable patients or to correct prolonged coagulation parameters. No heparin was administered to the patients thereafter. The intravenous lines were flushed with heparin-free saline, and any need for anticoagulation was achieved with acenocoumarin (Sintrom, Novartis-Pharma, Basel, Switzerland). Platelet counts were obtained preoperatively, 1 hour after surgery, and on the first and fifth postoperative days, and were corrected for hemodilution according to hematocrit values.
Statistical methods
The analysis of the data collected for the study was performed using the following statistical methods, depending on the nature of the data analyzed:
Longitudinal data (ie, measurements on the same variable taken at consecutive time points for each patient) were evaluated by repeated-measures analysis of variance (RM-ANOVA). All other continuous variables measured at a single time point (eg, duration of ICU stay and amounts of blood loss), aggregates of the longitudinal data, and percent reduction of platelet counts from the preoperative values were evaluated by the t test and its nonparametric equivalent (Wilcoxon rank-sum test). Changes from baseline at each time point of measurement within each group (eg, no intergroup comparisons) were evaluated using the Student paired t test and the Wilcoxon signed-rank test. The association between length of perfusion and platelet count reduction was assessed with Spearman's correlation. In general, during the analysis, the parametric methods were used after testing for normality. If the test rejected the normality assumption, then the nonparametric method was used. The P values reported are accompanied by the type of test they are derived from. All calculations were performed using the SASJ statistical package (SAS Institute, Inc, Cary, NC).
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionConclusionDiscussionReferences |
|---|
|
|
|
|
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionConclusionDiscussionReferences |
|---|
Our protocol addresses the preoperative recognition of HITT-sensitive patients and their intraoperative management for prevention of activation of the HITT process. In this protocol, HITT is diagnosed preoperatively by screening all patients who present with low platelet counts (<150,000/mm3) or a history of prolonged heparin therapy or thrombosis regardless of platelet count. Currently, in our practice, all our patients scheduled for surgery undergo a preadmission complete blood count and platelet count within the frame of a preadmission evaluation. No patient with a low platelet count is scheduled for surgery without prior investigation of his or her thrombocytopenia. In our laboratory, the special diagnostic workup can be completed within 5 to 6 hours. The determination of the iloprost concentration that inhibits HIPA is usually performed within 1 hour before heparinization for CPB. Therefore, with proper mobilization of the involved services, this protocol can be applied to cases requiring urgent surgery. Since we started using this protocol, we have not seen any of the unexpected postoperative hard-to-manage bleeding that we used to face sporadically. In our experience, 31.3% of patients with thrombocytopenia who presented for cardiac surgery tested positive for HITT antibodies. Patients testing positive for HIPA are further tested with ELISA for HITT antibodies. In patients with HITT antibodies, surgery is postponed for approximately 3 months if their condition allows it. If the antiplatelet antibodies disappear from the patient's plasma, surgery proceeds with standard CPB and heparin. Any further exposure postoperatively to either unfractionated or low molecular weight heparin is strictly avoided.3,5,16-18 In case the patient's condition forbids delay of surgery, we use iloprost for temporary platelet inhibition (as described previously) to prevent HITT-related complications. Again, heparin is not used postoperatively in any form. This protocol allows the safe use of heparin during CPB in patients with HITT antibodies.
Iloprost, a stable analog of prostacyclin with a half-life of 30 minutes, is effective in completely inhibiting platelets by binding to specific platelet receptors, activating adenyl cyclase, and increasing the intracellular cyclic adenosine monophosphate levels. In addition, iloprost acts on the arterial smooth muscle with a vasodilatory effect that may lead to hypotension.19 At appropriate infusion rates, iloprost allows safe heparin administration in patients with HITT antibodies provided there is close monitoring of hemodynamics and careful use of alpha agonists (norepinephrine and phenylephrine) to prevent or treat hypotension.9-13 Our data support the hypothesis that heparin may be used with impunity in patients with HITT when their platelets are completely inhibited with iloprost and their hemodynamics are closely monitored. The main side effect of iloprost is hypotension secondary to vasodilation. However, in our experience, hypotension during iloprost infusion was readily controlled with norepinephrine infusion (1-4 µg · kg-1 · min-1) or phenylephrine boluses. In this report, we present our first 10 patients in whom HITT antibodies were detected preoperatively. Our overall hospital experience to date with patients who test positive for HITT is satisfactory. This protocol has been used successfully for the identification and management of these patients so they can undergo cardiac surgery without the thrombotic and hemorrhagic complications related to their sensitivity to heparin.
|
Conclusion |
|---|
|
TopAbstractMethodsResultsDiscussionConclusionDiscussionReferences |
|---|
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionConclusionDiscussionReferences |
|---|
You successfully used iloprost in 10 patients identified with HIT, and you followed a similar successful series by Addonizio and colleagues and Kappa and colleagues from the University of Pennsylvania nearly 15 years ago. They found, as you did, that the vasodilatory side effects of iloprost can be reasonably handled in a safe fashion and that iloprost prevents platelet aggregation leading to bleeding and thrombotic complications. Unfortunately, iloprost is not available in the United States.
I have 3 questions, and there are 3 slides. You did not mention the serotonin release assay in your presentation, which is perhaps the oldest and still the most specific and sensitive test for HIT as far as indicating the development of thrombotic complications. You have not used this test. Why?
We found recently that many patients who test positive for heparin antibodies with the ELISA, test negative with the serotonin release assay. We have successfully proceeded with CPB using a 1-time heparin dose without adjuvant or alternative means of anticoagulation in this small group of patients. Do you believe that patients with a negative functional assay such as serotonin or platelet aggregation, yet a positive heparin antibody with the ELISA, define a different group of patients who can be treated safely in the traditional fashion?
Finally, given that iloprost is not available in the United States, and it is interesting to conjecture why not, what is your advice for the U.S. surgeon faced with a patient with HIT? Can you comment on the alternative anticoagulation strategies I have listed, such as alternative prostacyclin analogs (eg, epoprostenol), direct thrombin inhibitors (there are several now), defibrinating agents (eg, ancrod and GPIIb or IIIa receptor inhibitors), or the heparinoids?
Dr Palatianos. There is no 100% sensitive assay to identify the patients who will have clinical HITT. We used the ELISA assay because it was available in our hospital. Indeed, there are some patients who test negative and may have clinical HITT, and there are patients who test positive for antibodies and will not have clinical HITT after heparinization for surgery. To improve diagnostic accuracy, we also test platelet aggregation to heparin. The serotonin release assay is a highly sensitive method for detection of HITT antibodies. However, it is technically demanding and not readily available in every hospital. When available, it should be used to identify patients prone to have HITT complications.
Patients who test positive for HITT antibodies with the ELISA test and test negative with the serotonin release assay should be restudied. If the ELISA assay remains positive, I think they should be treated as HITT prone.
Although iloprost is available in Europe, its use for prevention of HITT is not included among its accepted indications. We use it within the frame of the clinical study. There are recent reports promising that newer medications such as thrombin inhibitors or heparin substitutes may become available in the near future. However, a generalized experience with them is lacking.
I think it is important to identify these patients preoperatively. Carefully investigating the patients who come for surgery with low platelet counts or a history of prolonged heparin therapy may help in identifying patients with HITT antibodies. In general, we test the occasionally seen patient with thrombocytopenia with a bleeding time and perhaps with platelet aggregation studies. If these tests are normal, we should not stop there. Some of these patients may have HITT antibodies. With our protocol, we detected HITT antibodies in 31.3% of these patients. Once we identify the patients, if their condition allows us to wait, postponing surgery is the best thing to do until the antibodies clear, usually within 2 to 3 months. In case surgery cannot be postponed, the patients should be managed with a specific protocol addressing this serious and potentially dangerous problem.
Dr Paschalis Tossios (Cologne, Germany). We know that these patients are sometimes very sick and that they are very difficult to handle when the diagnosis is confirmed preoperatively. In my own experience, heparin-induced thrombocytopenia and associated complications occur more often postoperatively, and they might occur with venous and arterial thromboembolic complications, even with normal platelet counts. Do you have a protocol for when heparin-induced thrombocytopenia is diagnosed during the postoperative course?
Dr Palatianos. Basically, heparin, either unfractionated or low molecular weight, should not be given to these patients postoperatively, not even through a heparin flush to keep arterial or venous lines open. Of course, HITT antibodies may be present with a normal platelet count.
We have not observed any thrombosis in our patients. In case of thrombosis, any heparin preparation that the patient is receiving should be stopped, and some antithrombotic agent could be used. Of course, the site of thrombosis, clinical picture, timing, and general condition of the patient direct the treatment. For this condition, prevention is the best treatment. By making the platelets insensitive to the stimuli with iloprost, we prevent platelet activation and release reaction, aggregation, and thrombosis.
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionConclusionDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Choi and R. Brull Neuraxial Techniques in Obstetric and Non-Obstetric Patients with Common Bleeding Diatheses Anesth. Analg., August 1, 2009; 109(2): 648 - 660. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Dyke, G. Aldea, A. Koster, N. Smedira, E. Avery, S. Aronson, B. D. Spiess, and A. M. Lincoff Off-Pump Coronary Artery Bypass With Bivalirudin for Patients With Heparin-Induced Thrombocytopenia or Antiplatelet Factor Four/Heparin Antibodies Ann. Thorac. Surg., September 1, 2007; 84(3): 836 - 839. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. Levy, K. A. Tanaka, and M. J. Hursting Reducing Thrombotic Complications in the Perioperative Setting: An Update on Heparin-Induced Thrombocytopenia Anesth. Analg., September 1, 2007; 105(3): 570 - 582. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Koster, C. M. Dyke, G. Aldea, N. G. Smedira, H. L. McCarthy II, S. Aronson, R. Hetzer, E. Avery, B. Spiess, and A. M. Lincoff Bivalirudin During Cardiopulmonary Bypass in Patients With Previous or Acute Heparin-Induced Thrombocytopenia and Heparin Antibodies: Results of the CHOOSE-ON Trial Ann. Thorac. Surg., February 1, 2007; 83(2): 572 - 577. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. B. Cines, J. B. Bussel, R. B. McMillan, and J. L. Zehnder Congenital and Acquired Thrombocytopenia Hematology, January 1, 2004; 2004(1): 390 - 406. [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 |
, No iloprost; 
, iloprost 60 pg/mL (3 ng · kg-1 · min-1); 
, 120 pg/mL (6 ng · kg-1 · min-1); •, iloprost, 240 pg/mL (12 ng · kg-1 · min-1).