|
|
||||||||
Ann Thorac Surg 2007;83:1737-1743
© 2007 The Society of Thoracic Surgeons
a Departments of Thoracic and Cardiovascular Surgery, St. Lukes Medical Center, Milwaukee, Wisconsin
c Department of Laboratory Medicine, St. Lukes Medical Center, Milwaukee, Wisconsin
d Department of Internal Medicine, Section of Hematology/Oncology, St. Lukes Medical Center, Milwaukee, Wisconsin
e Department of Quality Management, St. Lukes Medical Center, Milwaukee, Wisconsin
f Department of Cardiovascular Research, St. Lukes Medical Center, Milwaukee, Wisconsin
b Department of Anesthesiology, Duke University Medical Center, Durham, North Carolina
Accepted for publication December 11, 2006.
* Address correspondence to Dr Kress, Midwest Heart Surgery Institute, Ltd, 2901 W. Kinnickinnic River Pkwy, Suite 511, Milwaukee, WI 53215 (Email: dkress2003{at}msn.com).
| Drs Kress and Aronson disclose that they have a financial relationship with The Medicines Company.
|
| Abstract |
|---|
|
|
|---|
Methods: Between March 2002 and December 2004, 1114 (92%) of 1209 patients undergoing cardiac surgery with heparin were tested in an unselected manner for HPF4 antibodies. Main outcome measures were HPF4 antibody seropositivity and fatal and nonfatal adverse clinical outcomes after cardiac surgery.
Results: Of those screened, 60 (5.4%) of 1114 had positive HPF4 antibodies preoperatively. These patients had longer mean postoperative length of stay (14.0 days versus 9.8 days, p = 0.05), a higher incidence of prolonged (
96 hours) mechanical ventilation (20.3% versus 9.2%, p = 0.02), acute limb ischemia (5.1% versus 0.9%, p = 0.03), renal complications including dialysis (20.3% versus 10.5%, p = 0.03), and gastrointestinal complications (15.3% versus 5.9%, p = 0.01). Stepwise logistic regression analysis showed positive HPF4 antibody status to be an independent predictor for adverse outcome and was associated with a higher risk for renal complications, including dialysis (adjusted odds ratio 2.2; 95% confidence interval, 1.1 to 4.3), than was diabetes.
Conclusions: In this large patient series, the presence of HPF4 antibodies before surgical heparin administration was an independent and clinically significant risk factor for postoperative adverse events after cardiac surgery. An optimal preoperative cardiac surgery risk profile should include HPF4 antibody status.
Heparin-induced thrombocytopenia (HIT) is an immune-mediated complication of heparin administration that contraindicates further heparin exposure during cardiac surgical procedures. Overt heparin-induced thrombocytopenia and thrombosis (HITT) occurs in approximately 50% of HIT patients, with devastating consequences such as limb ischemia requiring amputation in 10% to 20%, myocardial infarction (MI), stroke, pulmonary embolism, and death in 20% to 30% [14]. HIT occurs in 5% of orthopedic surgical procedures, 3% of cardiac operations, 1% of vascular operations, and in 1% of medically managed patients who receive heparin [36].
Antibodies against heparin-platelet factor 4 (HPF4) represent an important component in the overall diagnosis of HIT [7]. A clinical diagnosis of HIT is often thought to be heralded by new onset thrombocytopenia in the presence of recent heparin administration with or without HPF4 antibodies [8]. HPF4 antibodies may constitute an independent risk factor. Before cardiac operations, 13% to 22% of patients test positive for HPF4 antibodies [9, 10], and HPF4 antibodies develop in as many as 50% of patients after cardiac operations, which are distinguished by the need for large systemic doses of heparin for anticoagulation [1015]. The presence of HPF4 antibodies has been reported to predict adverse cardiac events in some patient populations even in the absence of thrombocytopenia [9, 11, 16].
Given the large number of patients undergoing cardiac operations annually, it is important to identify populations that are at high risk for adverse outcomes. It is not known whether patients with HPF4 antibodies before the procedure, even in the absence of thrombocytopenia, may be at higher risk for certain adverse outcomes. To address this question, we conducted an observational study to determine the incidence of patients with a positive assay for HPF4 antibodies before operation and its relationship to fatal and nonfatal events in a nonselected cardiac surgical patient population exposed to heparin for procedural anticoagulation. During a 2.5-year period, our study compared postoperative outcomes in preoperative HPF4 antibody-positive and HPF4 antibody-negative patients undergoing coronary or valve procedures with cardiopulmonary bypass (CPB) and anticoagulated with heparin.
| Patients and Methods |
|---|
|
|
|---|
Patient Population
Patients scheduled to undergo coronary artery bypass grafting (CABG) alone or valve procedures with or without CABG using CPB were evaluated. This study was based on data obtained by a cardiovascular surgical group that routinely performs preoperative HPF4 antibody screening and platelet counts for all cardiac surgical patients.
All patients studied had similar intraoperative management for surgical anticoagulation, which included a bolus of porcine heparin (300 mg/kg) before CPB. The dose was adjusted to maintain an activated clotting time (ACT) of more than 400 seconds.
Laboratory Diagnostic Evaluation
The presence of HPF4 antibodies was determined by enzyme-linked immunosorbent assay (ELISA) (GTI Inc, Waukesha, WI) [17]. Assay results were considered positive if an optical density (OD) of more than 0.4 was observed.
HPF4 antibodies were tested at least once during the index hospitalization and were identified as positive if at least one test demonstrated an OD of more than 0.4. Preoperative antibody status was defined from blood samples obtained before the surgical procedure, on the day of the procedure, or within the first 4 days after the procedure, based on the assumption that HPF4 antibodies resulting from surgical heparin exposure are typically not detectable until after postoperative day 4 [12, 18]. In some instances, more than one blood sample was obtained from patients beyond 4 days after the procedure in accordance with the treating physicians discretion. Patients with a positive test result during this time period were identified as having a positive postoperative antibody status.
Patients with positive ELISA results were administered a confirmatory heparin antibody aggregation (HAAG) functional assay [19]. HAAG results were considered positive if aggregation exceeded 50%, indeterminate if aggregation was between 26% and 50%, and negative if platelet aggregation was less than 26%.
Platelet counts were assessed preoperatively at baseline and postoperatively. Baseline platelet count was determined up to 7 days before surgery. Postoperative thrombocytopenia was defined as a 50% drop in platelet count from baseline, or an absolute platelet count of less than 100,000 x 109/L from postoperative day 4 to day 8 or discharge, whichever came first.
Clinical Diagnostic Evaluation
Patient demographics, prior medical history, preoperative medical status, postoperative outcomes through hospital discharge, and discharge status were determined by patient history, medical examination, and laboratory evaluation. Results were recorded in the database maintained by St. Lukes Medical Center on an ongoing basis for contribution to the national STS database. For the purposes of this study, data from this patient series were retrieved from this local database and analyzed. Patient variables available for analysis included all data fields defined and specified by STS [20]. Examples of definitions are in Table 1.
|
2 test or Fischer exact test for categoric variables. For these tests, a two-tailed p < 0.05 was considered statistically significant. For stepwise logistic regression analysis, variables found to meet p < 0.10 by univariate analysis were entered into the logistic regression model and p < 0.08 was used as a stay criterion. This model was used to determine whether a positive preoperative HPF4 antibody status was independently associated with a specific adverse outcome. All other preoperative comorbidities were included in the model. Adjusted odds ratios (OR) and 95% confidence intervals (CI) were derived from the final logistic model.
Patients with positive ELISA HPF4 antibody test results were evaluated for the presence of thrombocytopenia when possible. Patients with positive preoperative HPF4 antibodies and thrombocytopenia were compared with patients with positive preoperative HPF4 antibodies without thrombocytopenia for adverse outcomes. Analyses were conducted for all patients undergoing cardiac surgery, which included CABG only (76%) or valve surgery with or without CABG (24%). The data were analyzed using the SAS statistical software (SAS Institute Inc, Cary, NC).
| Results |
|---|
|
|
|---|
The first blood sample for evaluation of HPF4 antibodies was obtained before or including the day of surgery in 1098 (98.6%) of 1114 of the patients. In 99.5% of patients, the first blood sample was obtained before and including the day of surgery or within the first 4 days after surgery. The first blood samples were obtained in some patients within the first 4 days after the procedure rather than before the procedure primarily because of urgent scheduling needs and other time-limiting factors unrelated to routine elective management. These results were considered to represent antibody status before the procedure because HPF4 antibody formation after cardiac surgical heparin exposure typically requires 4 days [18]. One blood sample was drawn for HPF4 antibody testing in 933 patients, and two or more blood samples were drawn for 181 patients.
Sixty (5.4%) of 1114 patients tested positive for HPF4 antibodies preoperatively, with 51 (85%) of 60 identified before and including the day of the operation and 9 identified within the first 4 days postoperatively. All 60 of these patients who were identified as having a positive preoperative antibody status had negative HAAG assays. Based on the negative HAAG results, anticoagulation protocols, postoperative monitoring, and management were not modified for these patients. One of the 60 patients had a strongly positive preoperative HPF4 antibody assay and was treated with lepirudin for surgical anticoagulation. This patient was excluded from the outcomes analysis.
An additional 13 patients tested positive for HPF4 antibodies after the defined preoperative time period and were characterized as postoperative seroconverters. Patients who received HPF4 antibody testing after the first 4 days postsurgery were universally evaluated because of a suspicion of clinical HIT. All postoperative seroconverters were found to have postoperative thrombocytopenia, and five were identified as having a positive HAAG analysis.
Patient Characteristics
Preoperative demographic data and medical risk factors for patients undergoing all subtypes of procedures combined are listed in Table 2. Profiles were similar in patients testing positive or negative for preoperative HPF4 antibodies, except for a higher incidence of arrhythmia, mitral insufficiency, congestive heart failure, and prior cerebrovascular accident in the HPF4 antibody-positive patient group. No patients who tested positive for HPF4 antibodies had preoperative thrombocytopenia. Among all patients evaluated in this study, only 15 had preoperative platelet counts of less than 100,000 x 109/L, all of whom tested negative for HPF4 antibodies.
|
|
|
Patient risk for adverse postoperative outcomes was correlated with preoperative HPF4 antibody status (Table 5). Adjusted odds ratios obtained by stepwise logistic regression analysis showed that preoperative HPF4 antibody-positive status was associated with significant risk for acute limb ischemia (OR, 4.9; 95% CI, 1.2 to 20.2), renal complications including dialysis (OR, 2.2; 95% CI, 1.1 to 4.3) and gastrointestinal complications (OR, 2.9; 95% CI, 1.3 to 6.6). As a predictor of risk, HPF4 antibody status was found to have a greater association than diabetes with renal dialysis.
|
Heparin was administered intraoperatively to 725 patients whose platelet counts were determined at baseline and at a postoperative interval between days 4 and 8. Thrombocytopenia within the first 4 days after surgery was considered reflective of other causes common to cardiac operations, but not related to heparin administration, because the development of thrombocytopenia after cardiac operations resulting from heparin exposure typically requires 4 days [18]. Among these 725 patients, 52 had positive preoperative HPF4 antibodies, with postoperative thrombocytopenia developing in 23. No difference was found in the incidence of any adverse postoperative outcome for the 23 preoperative HPF4 antibody-positive patients with postoperative thrombocytopenia compared with the 29 preoperative HPF4 antibody-positive patients without postoperative thrombocytopenia.
Overall, HIT developed in 36 patients, as evidenced by preoperative HPF4 antibodies and postoperative thrombocytopenia in 23 or postoperative seroconversion and thrombocytopenia in 13. In our series, postoperative HIT and HITT developed in 9 patients, as evidenced by exposure to heparin, thrombocytopenia, and overt thrombosis (Table 6).
|
| Comment |
|---|
|
|
|---|
We report outcomes for consecutive, unselected patients after cardiac operations (76% CABG only, 24% valve procedures with or without CABG) from a large single center using routine experience screening for HPF4 antibodies. Our data are consistent with the observations made by Bennett-Guerrero and colleagues [9], and indicate that morbidity is significantly associated with the presence of preoperative HPF4 antibodies defined by commonly used clinical screening criteria. This previously unrecognized relationship between heparin antibody formation and surgical outcome clearly has a significant impact on risk profiling. This observation, now recognized in a cardiac surgical population, may not be surprising.
Evidence of the validity of our study cohort is derived from the observation that the incidence of HIT and HITT is consistent with previously published reports of cardiac surgical patients exposed to heparin intraoperatively [4]. In this study the respective incidence of HIT and HITT was 3.2% and 0.8%.
The first indication that HPF4 antibodies independently affect outcome beyond the risk for HIT was reported by Mattioli and colleagues [7] in patients with unstable angina treated with unfractionated heparin. The incidences of major adverse ischemic outcomes (MI, stroke, death) in these patients were almost 50% greater in the group with HPF4 antibodies compared with the group without HPF4 antibodies. None of the patients studied had clinical HIT. Their study differed from ours with respect to patient population and timing for HPF antibody testing relative to heparin exposure (ie, on days 1 and 40 after the beginning of heparin therapy).
Williams and colleagues [16] found an association between HPF4 antibody positivity and 30-day death or MI (30% versus 11%) in a small subset of 218 patients without thrombocytopenia and treated with unfractionated heparin. Positive HPF4 antibody status was a strong independent predictor for 30-day MI (22% versus 6%). Calaitges and colleagues [21] reported that 7% of 106 patients scheduled to have peripheral arterial reconstruction surgery had preoperative HPF4 antibodies, and 21% had a positive antibody assay during their index hospitalization. Patients with positive HPF4 antibodies were 2.6 times more likely to have a thrombotic event during their index hospitalization compared with HPF4 antibody-negative patients. No patient with the antibody and a thrombotic event had clinical HIT.
Antibody conversion is not always associated with clinical evidence of heparin-dependent platelet activation or thrombocytopenia. In our study, none of the patients with preoperative HPF4 antibodies had a positive HAAG assay or preoperative thrombocytopenia. No difference in outcome was found between those preoperative HPF4 antibody-positive patients who developed postoperative thrombocytopenia and those who did not.
The presence of a positive preoperative HPF4 antibody independently identified patients at increased risk for adverse renal, gastrointestinal, respiratory, and ischemic limb outcomes. It is unclear whether the HPF4 antibodies are causative or correlative for adverse outcomes, and these relationships warrant further mechanistic study. In addition, a positive preoperative HPF4 antibody identified patients who may develop HIT postoperatively. In this series, HIT developed postoperatively in 44% of patients (23/52) who tested positive preoperatively for HPF4 antibodies in the absence of preoperative thrombocytopenia.
These data provide evidence that HPF4 antibody status should be considered for optimal risk assessment, particularly in todays clinical setting, because it is unusual for cardiac surgical patients not to have experienced one or more heparin exposures from prior cardiac catheterizations. Our findings showed that HPF4 antibody status is as important or in some instances more relevant than age, left ventricular ejection fraction, diabetes, hypertension, or other commonly recognized preoperative predictors of postoperative outcomes. In addition, positive preoperative HPF4 antibody status is an independent predictor of patients who will require additional hospital and intensive care resources.
This study has several limitations. It was performed by retrospective analysis, but the data were collected prospectively in a standardized and comprehensive database. Although this represents a large reported series for preoperative HPF4 antibody testing in an unselected surgical population, our sample size was underpowered for extensive subgroup analyses such as are performed in larger confirmatory trials.
Not all patients had laboratory evaluation for HPF4 antibodies or platelets at the exact same interval during index hospitalization, and some HPF4 antibody-positive test results were identified after the day of the procedure. We used a cutoff of 4 days after the operation to define preoperative positive antibody status on the assumption that HPF4 antibody formation after intraoperative heparin exposure would take longer than 4 days [18]. In these patients, it was not possible to precisely determine the day of seroconversion.
A binomial metric was used to define positive versus negative HPF4 antibody status. The values used to define a significant OD with the ELISA method (GTI or Diagnostica Stago, Inc, Parsippany, NJ) are variable, ranging between 0.4 and 1.0, and are typically based on studies of patients with clinical HIT [10, 14, 15, 22]. It is possible that subthreshold values of antibody titers could have physiologic effects other than overt HIT. We used a value exceeding 0.4 OD with the GTI- ELISA, which is within commonly accepted clinical standards.
Our results show a relationship between HPF4 antibody positivity before cardiac operations and certain adverse outcomes. This finding demonstrates the significance of preoperative HPF4 antibodies on a clinical spectrum that is separate from their ability to precipitate HIT and thrombosis. The presence of preoperative HPF4 antibodies not associated with thrombocytopenia is an important, independent, and predictive risk factor for postoperative adverse outcome after a cardiac operation. Further study is needed to define the best ways to treat (eg, anticoagulant protocols, postoperative monitoring and management) and minimize risk for HPF4 antibody-positive patients undergoing cardiac operations.
An optimal cardiac surgical risk index profile should consider HPF4 antibody status. Patients with positive HPF4 antibody status should be closely monitored postoperatively because their risk for adverse outcomes and increased resource utilization is high. Diagnostic and therapeutic methods to identify and minimize the impact of HPF4 antibodies during cardiac operations may be important for reducing risk in the overall cardiac surgical patient population.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
D. Paparella, G. Scrascia, A. Galeone, M. Coviello, G. Cappabianca, M. T. Venneri, B. Favoino, M. Quaranta, L. de Luca Tupputi Schinosa, and T. E. Warkentin Formation of anti-platelet factor 4/heparin antibodies after cardiac surgery: influence of perioperative platelet activation, the inflammatory response, and histocompatibility leukocyte antigen status. J. Thorac. Cardiovasc. Surg., December 1, 2008; 136(6): 1456 - 1463. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Greinacher and J. H. Levy HIT Happens: Diagnosing and Evaluating the Patient with Heparin-Induced Thrombocytopenia Anesth. Analg., August 1, 2008; 107(2): 356 - 358. [Full Text] [PDF] |
||||
![]() |
M. P. Mullen, D. L. Wessel, K. C. Thomas, K. Gauvreau, E. J. Neufeld, F. X. McGowan Jr, and J. A. DiNardo The Incidence and Implications of Anti-Heparin-Platelet Factor 4 Antibody Formation in a Pediatric Cardiac Surgical Population Anesth. Analg., August 1, 2008; 107(2): 371 - 378. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. M. Dellaria, P. E. Greilich, and R. Sarode Urgent Cardiac Surgery in a Patient with a Remote History of Heparin-Induced Thrombocytopenia: Use of the Anti-PF4/Heparin ELISA Confirmatory Test Anesth. Analg., August 1, 2008; 107(2): 379 - 381. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. E Dager Considerations for Drug Dosing Post Coronary Artery Bypass Graft Surgery Ann. Pharmacother., March 1, 2008; 42(3): 421 - 424. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Apostolidou, M. F. Sweeney, E. Missov, L. D. Joyce, R. John, and R. C. Prielipp Acute Left Atrial Thrombus After Recombinant Factor VIIa Administration During Left Ventricular Assist Device Implantation in a Patient with Heparin-Induced Thrombocytopenia Anesth. Analg., February 1, 2008; 106(2): 404 - 408. [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 |