Clinical Use of the Activated Partial Thromboplastin Time and Prothrombin Time for Screening A Review of the Literature and Current Guidelines for Testing Jerrold H. Levy, MD, FAHA, FCCMa,b,c,*, Fania Szlam, MMScd, Alisa S. Wolberg, PhD, FAHAe, Anne M. Winkler, MDf KEYWORDS Activated clotting time Anticoagulation Monitoring Activated partial thromboplastin time Prothrombin time Surgery KEY POINTS Although the activated partial thromboplastin time (aPTT), prothrombin time (PT), and international normalized ratio (INR) are widely used in routine preoperative testing, these hemostatic tests are not reliable predictors of perioperative bleeding in patients without known bleeding risk factors. Continued
Conflict of Interest Statements: Prof J.H. Levy serves on steering committees for Boehringer Ingelheim, CSL Behring AG, Grifols, Janssen Pharmaceuticals (Johnson & Johnson), Merck, and The Medicines Company. F. Szlam, Dr AM Winkler, and Dr A.S. Wolberg have no relevant conflict of interest to disclose. a Department of Anesthesiology, Duke University School of Medicine, 2301 Erwin Road, Durham, NC 27710, USA; b Department of Surgery, Duke University School of Medicine, 2301 Erwin Road, Durham, NC 27710, USA; c Department of Critical Care, Duke University School of Medicine, 2301 Erwin Road, Durham, NC 27710, USA; d Emory University Hospital, Emory University School of Medicine, 1364 Clifton Road, Atlanta, GA 30322, USA; e Department of Pathology and Laboratory Medicine, University of North Carolina at Chapel Hill, 815 Brinkhous-Bullitt Building, CB# 7525, Chapel Hill, NC 27599-7525, USA; f Department of Pathology & Laboratory Medicine, Emory University Hospital, 1364 Clifton Road, Atlanta, GA 30322, USA * Corresponding author. Duke University Medical Center, 2301 Erwin Road, 5691H HAFS, Box 3094, Durham, NC 27710. E-mail address: [email protected] Clin Lab Med - (2014) -–http://dx.doi.org/10.1016/j.cll.2014.06.005 labmed.theclinics.com 0272-2712/14/$ – see front matter Ó 2014 Elsevier Inc. All rights reserved.
2
Levy et al
Continued A preoperative bleeding history and physical examination are usually obtained in an attempt to identify important bleeding risk factors. However, current questionnaires used to assess bleeding history are notoriously poor at characterizing bleeding. In such cases, follow-up hemostatic testing may be appropriate. The aPTT is extensively used to monitor therapy with unfractionated heparin and other anticoagulant agents, including direct thrombin inhibitors, whereas the PT and INR are used to monitor the anticoagulant effects of warfarin and other vitamin K antagonists, and to adjust their dosages.
INTRODUCTION
Standard coagulation monitoring is used to manage hemostasis and bleeding in hospitalized patients, including those undergoing surgical procedures and during cardiac surgery with cardiopulmonary bypass (CPB). Multiple factors influence coagulation in hospitalized patients, especially in a perioperative setting. Such factors include type of procedure, previous surgery at the same anatomic site, degree of tissue injury, and the underlying state of the hemostatic system. Surgical procedures that may be associated with increased bleeding potential due to high tissue vascularity include tonsillectomy and vascular and cardiac surgery. In other cases, bleeding risk is associated with the potential for related adverse events, especially if bleeding occurs in the central nervous system or other closed spaces, such as in ophthalmic procedures. Multiple assessments are routinely undertaken to assess the risk of bleeding in a given patient, including patient history, history of prior procedures with excessive bleeding, and family history. However, laboratory testing is often also extensively used in this evaluation. Tests that are performed most commonly include the activated partial thromboplastin time (aPTT), prothrombin time (PT), international normalized ratio (INR), and activated clotting time (ACT) for patients during cardiac surgery. The ACT is more often used for monitoring anticoagulation for CPB. Although laboratory testing with the aPTT and PT is appropriate to monitor anticoagulation, and the tests evolved soon after the introduction of warfarin to determine hemostatic abnormalities, the value of these tests to predict bleeding in surgical patients is not well documented despite their widespread use. Their usefulness is further complicated by underlying bleeding disorders, test characteristics, and the potential for false-positive and false-negative results. In the presence of a lupus anticoagulant or factor XII deficiency, for example, the aPTT may be prolonged, but this prolongation is not associated with an increased risk of bleeding.1 A systematic review of the literature published in 1997 suggests that preoperative laboratory tests of hemostasis before elective surgery in patients without a positive personal history for abnormal bleeding are not helpful and rarely lead to a change in clinical management of the patient.2 Furthermore, guidelines on the preoperative assessment of bleeding risk state that hemostatic tests are poor predictors of bleeding and that routine tests in patients without a history indicative of bleeding are not generally recommended.3,4 Despite these perspectives, these coagulation tests are frequently used for clinical decision-making in hospitalized patients. The purpose of this review is to examine the clinical applications of the aPTT and PT tests and their role in assessing perioperative bleeding risk for perioperative screening. In addition, the use of the ACT will be considered for anticoagulation monitoring for CPB as it is used extensively in the hospital setting.
Coagulation Monitoring: A Review
aPTT
The aPTT is a global coagulation screening test that is used for assessment of the coagulation status in patients with suspected acquired deficiencies of coagulation factors of the intrinsic and common pathways of the coagulation system. The test is affected by multiple factors, including the levels of factors VIII, IX, XI, XII, X, II, and fibrinogen. The aPTT is widely used for monitoring anticoagulation therapy with low levels of heparin (from 0.1 IU/mL to approximately 1 IU/mL). In a normal population, the aPTT varies, and this interindividual variability is reflected in a wide reference interval. The aPTT reference interval also differs between laboratories that use reagents with different factor sensitivities and different lipid compositions as well as different instrumentation. The aPTT reagent is a mixture of phospholipids and activators (eg, kaolin, silica, or ellagic acid). Studies have shown considerable differences in the responsiveness of various aPTT reagents to mild and moderate deficiencies of coagulation factors, particularly factors VIII and IX.5 Furthermore, in 1.5% to 3% of the population, aPTT is prolonged because of mild to severe factor XII deficiency; however, this prolongation is not associated with an increased risk of bleeding.1 In addition, laboratories determine a reference interval to encompass the central 95% of apparently healthy men and women with similar demographics to the hospitalized population.6 As a result, 5% of normal individuals will fall outside of the reference interval. Last, elevations of biological substances such as C-reactive protein have recently been published to interfere with commonly used aPTT reagents and will cause a false prolongation because of interferences with phospholipids, particularly phosphatidylcholine and phosphatidylethanolamine, which are commonly used in aPTT reagents.7 Clinical Uses of the aPTT
The aPTT is extensively used to monitor unfractionated heparin (UFH) therapy and other anticoagulant agents, including direct thrombin inhibitors. The limitations for this test include biological variability, insensitivity to some clinically important bleeding disorders (eg, factor XIII deficiency, a2-antiplasmin deficiency), variability in instrumentation and reagents, low sensitivity to common pathway deficiencies (fibrinogen, prothrombin), variability due to physiologic changes (eg, in pregnancy, physical stress, or trauma), clinically irrelevant prolongation due to certain factor deficiencies (eg, factor XII [one of the commonest causes of unexpected aPTT prolongation], prekallikrein, and high-molecular-weight kininogen deficiencies), and preanalytical errors such as improper specimen collection. PT AND INR
The PT measures the time required for clotting to occur after the addition of a source of tissue factor to recalcified citrated plasma in laboratory instruments and on point of care (POC) devices. The PT is measured by adding thromboplastin (a mixture of tissue factor, calcium, and phospholipid) to a patient’s citrated plasma sample, and clot formation is determined. It is used also as a screening assay to detect deficiencies of one or more coagulation factors (fibrinogen and factors II, V, VII, and X). The INR was introduced by the World Health Organization to overcome variability in PT results due to differing sensitivities of thromboplastin reagents produced by different manufacturers.8 The INR is the ratio of the patient’s PT value divided by the normal value (geometric mean PT value for non-anticoagulated patients), as determined by the local laboratory, raised to the International Sensitivity Index (ISI) value (usually between 1.0 and 2.0) for the reagent and analytical system used: INR 5 (PTpatient/PTgeomean)ISI.
3
4
Levy et al
Clinical Uses of the PT/INR
The PT/INR is used extensively to monitor the anticoagulant effects of warfarin and other vitamin K antagonists and to adjust their dosages. Clinically, it is an in vitro measure of the extrinsic and common coagulation pathways and should therefore detect deficiencies of factors II, V, VII and X, and very low fibrinogen concentrations. As with all coagulation tests, the PT/INR is limited by biological variability, insensitivity to many bleeding disorders (eg, factor XIII deficiency, a2-antiplasmin deficiency), variability in results due to differences in reagents and coagulation analyzers, and preanalytical errors. In addition, the PT results may be prolonged in young children. A study in healthy subjects categorized into 3 age groups (1–5 years, 6–10 years, and 11–18 years) showed no significant differences between the groups in terms of aPTT; however, children in the youngest age group had significantly higher mean PT compared with adults (P 5 .03).9 Table 1 summarizes a range of conditions that may be present based on the results of PT and aPTT testing. UTILITY OF aPTT AND PT: EVIDENCE FROM THE LITERATURE
Findings of previously published reviews of the literature on coagulation testing are summarized in Table 2.2,10–18 The overwhelming conclusion from these reviews is that routine preoperative hemostatic testing with aPTT and PT is not useful in asymptomatic patients with no known risk factors. Moreover, with aPTT and PT, the high rate of false-positive and false-negative results may lead to inappropriate precautionary measures or false reassurance, respectively.12 Individual studies assessing the value of preoperative aPTT and PT (and other tests) for predicting bleeding risk are summarized in Table 3 (prospective studies19–35) and Table 4 (retrospective studies36–48). In most cases, aPTT and PT were not shown to
Table 1 Summary of conditions that may be present based on the results of PT and aPTT testing PT Result
aPTT Result
Examples of Conditions That May Be Present
Prolonged
Normal
Normal
Prolonged
Decreased or defective factor VIII, IX XI, XII, prekalikrein, high-molecular-weight kininogen Type 3 vWD Presence of lupus anticoagulant
Prolonged
Prolonged
Decreased or defective fibrinogen, factor II, V, or X Severe liver disease Acute DIC
Normal
Normal or slightly prolonged
May indicate normal hemostasis; however, PT and aPTT can be normal in conditions such as mild deficiencies in other factors and in the mild form of vWD Further testing may be required to diagnose these conditions
Liver disease Vitamin K deficiency Decreased or defective factor VII Chronic, low-grade DIC Vitamin K antagonist (warfarin) therapy
Abbreviations: DIC, disseminated intravascular coagulation; vWD, von Willebrand disease.
Table 2 Published reviews of preoperative coagulation testing Reference
Age
Conclusions
Swedish Council on Technology Assessment,8,10 1991
Adult
Presurgical hemostatic tests are unnecessary in asymptomatic patients
Munro et al,2 1997
Adult/pediatric
Hemostatic tests have no value in predicting perioperative bleeding in the absence of clinical features
Peterson et al,11 1998
Adult/pediatric
In the absence of a history of excessive bleeding, the bleeding time fails as a screening test and is not indicated as a routine preoperative test
Chee & Greaves,12 2003
Adult/pediatric
Indiscriminate coagulation testing is not useful in a surgical or a medical setting, due to the limited sensitivity and specificity of the tests, and the high rate of false-positive and false-negative results
Eckman et al,13 2003
Adult/pediatric
For nonsurgical and surgical patients without synthetic liver dysfunction or a history of oral anticoagulant use, routine testing has no benefit in assessment of bleeding risk
Dzik,14 2004
Adult/pediatric
PT/INR had a poor predictive value for bleeding risk. Whether mild or moderate abnormalities of such tests have any clinically relevant predictive value as appropriate triggers for prophylactic transfusions before invasive procedures should be investigated in formal randomized clinical trials
Sie´ & Steib,15 2006
Adult/pediatric
Systematic preoperative screening is poorly efficient. It should be restricted to selected patients based on clinical history and physical examination
Kozek-Langenecker,16 2010
Adult/pediatric
Routine perioperative coagulation tests, including aPTT, PT, and INR, are poor predictors of bleeding and mortality. Such testing is still in use due to tradition rather than evidence
Adult/pediatric
Clinical assessment of hemostatic function before cardiac operations is both effective and efficient. It obviates routine laboratory testing and favors the introduction of blood conservation strategies early on during the process of care
Adult/pediatric
There is no difference in the rate of posttonsillectomy bleeding in patients with abnormal compared with normal preoperative coagulation tests
Any surgery
Klopfenstein,17 1996
Tonsillectomy Krishna & Lee,18 2001
Coagulation Monitoring: A Review
Cardiac surgery
5
6
Type of Surgery and Reference
Age (n)
Laboratory Tests
PTT/aPTT Predictive
PT Predictive
Other Findings/Conclusions
Eisenberg et al,19 1982
Age not stated (n 5 750)
PTT, PT
No
No
PT and PTT detected few unsuspected bleeding disorders preoperatively, and there were a large number of apparently false-positive results
Borzotta & Keeling,20 1984
Adult (n 5 83)
PTT, PT, PC, BT
Yes, with bleeding history
Yes, with bleeding history
Bleeding history should guide the selection of laboratory tests. Neither history nor laboratory tests alone provide insurance against hemorrhagic mishaps, but together they help the surgeon protect his or her charge from unexpected failure of hemostasis
Rohrer et al,21 1988
Adult/pediatric (n 5 282)
PTT, PT, PC, BT
No
No
Of 514 screening tests performed, 4.1% were abnormal but none identified a clinically significant coagulopathy. Preoperative screening tests for coagulopathies not suspected on the basis of detailed clinical information are unnecessary and should not be done
Macpherson et al,22 1993
Age not stated Study 1, preoperative screening (n 5 111, after excluding 45 patients who had taken aspirin in the previous week, and 3 with a history suggesting bleeding risk) Study 2, investigation of disproportionate hemorrhage (n 5 1872)
aPTT, PT, BT, PC
No
No
If patients have no relevant history and physical examination is negative, preoperative screening for coagulation defects would seem to be unnecessary
General/not specified
Levy et al
Table 3 Prospective studies assessing preoperative coagulation tests for predicting intraoperative and/or postoperative bleeding risk
Houry et al,23 1995
Adult, age 16–99 y (n 5 3242)
PTT, PT, PC, BT
No
No
Preoperative hemostatic screening tests should not be performed routinely but only in patients with abnormal clinical data
Koscielny et al,24 2004
Adult, age 17–87 y (n 5 5649)
aPTT, PT, PC, including PFA100 (1vWF and BT in patients with a positive bleeding history)
No
No
The sensitivity of the PFA-100: collagen-epinephrine test was highest (90.8%) in comparison with the other screening tests, with a high positive predictive value (81.8%) and a higher negative predictive value (93.4%)
Age not stated (n 5 156)
aPTT (5 per patient), PT (6 per patient), INR, PT ratio, aPTT ratio
Variable
Variable
Among 16 patients with increased bleeding during surgery, PT and aPTT test results (all performed with the same instrument in the same laboratory) varied markedly. Variability was not reduced by use of the PT or aPTT ratio, or INR, or by incorporation of a measure of PT or aPTT test sensitivity to factordeficient serum
Adult/pediatric, with congenital heart disease (n 5 235)
aPTT, PTT, PT, TT, PC
Yes
Yes
One or more test values were abnormal in 19% of subjects—a significantly higher incidence than that expected in a normal population. Prolonged PT, PTT, or aPTT were seen most frequently; in 6/8 such patients evaluated further, there were decreased levels of factor VII or IX. However, the authors noted that use of blood products during cardiac surgery was not statistically significantly different between patients with normal or abnormal tests, and normal preoperative coagulation tests did not exclude the presence of a major bleeding diathesis
Spinal Murray et al,25 1999
Cardiac
(continued on next page)
Coagulation Monitoring: A Review
Colon-Otero et al,26 1987
7
8
Levy et al
Table 3 (continued ) Type of Surgery and Reference
PTT/aPTT Predictive
PT Predictive
Other Findings/Conclusions
PTT, PT, BT, PC, Hct, packed RBC volume
No
No
Excessive postoperative blood transfusion was predicted by BT and RBC volume but not by PTT, PT, PC, or Hct (nor by recent use of heparin or aspirin)
Adult (n 5 897)
Post-CPB: ACT, aPTT, PT, TT, PC fibrinogen, fibrin/ fibrinogen degradation products, Duke’s earlobe BT
No
No
Because the predictive values of the tests are so low, it does not appear sensible to screen patients routinely using these clotting tests after CPB
Pediatric (n 5 494)
PTT, PT, Hct, PC, fibrinogen, TE (before and during CPB)
Yes
No
Platelet count provided maximum sensitivity (83%) and specificity (58%) for prediction of excessive blood loss; TE was the only variable associated with total products transfused
Adult (n 5 1546)
PT
N/D
No
Of 25 evaluable patients with abnormal PT values, the results were not predictive in 20 cases and 5 had indications for coagulation testing based on history or physical examination. In the absence of specific indications, routine preoperative PT testing before elective gynecologic operations does not contribute to patient care and should be eliminated
Age (n)
Laboratory Tests
Ferraris & Gildengorin,27 1989
Adult (n 5 159), grouped according to packed RBC transfusion: Group I, 5 units (n 5 139) Group II, >5 units (n 5 20)
Gravlee et al,28 1994
Williams et al,29 1999
Gynecologic Aghajanian & Grimes,30 1991
Tonsillectomy/adenoidectomy Adult/pediatric (n 5 52)
PTT, PT, PC, BT
Yes?
Yes?
Clinical history failed to detect any previously unrecognized coagulation disorder. PT was prolonged in 5.8%, PTT in 11.5%, and BT in 9.5%. Six patients (11.5%) were considered to have important laboratory abnormalities. Laboratory screening therefore improved preoperative detection of occult hemostatic defects and allowed for appropriate alterations in perioperative care
Burk et al,32 1992
Pediatric (n 5 1603)
CBC, aPTT, PT, BT
No
No
Laboratory abnormalities had a high specificity (0.99) and high negative predictive value (0.98) with a low sensitivity (0.03) and low positive predictive value (0.07) in predicting postoperative bleeding. The large number of false positive tests, coupled with the relative rarity of inherited and acquired coagulopathies, raises doubts about the overall value of routine screening
Close et al,33 1994
Adult/pediatric (n 5 96)
PT, aPTT
No
No
Routine measurement of aPTT and PT in asymptomatic patients is not useful for predicting postoperative bleeding
Kang et al,34 1994
Pediatric (n 5 1061)
PTT, PT, BT, PC
Yes
No
A coagulation profile that includes PTT and BT may be a valuable screening tool for children undergoing tonsillectomy and adenoidectomy. An initially abnormal coagulation profile may identify those more likely to bleed after surgery (22.6% vs 5.6%)
Zago´lski,35 2010
Adult/pediatric (n 5 222)
aPTT, PT/INR, PC
No
No
In healthy adults, coagulation test results are irrelevant for the course of tonsillectomy and postoperative bleeding
Abbreviations: BT, bleeding time; CBC, complete blood count; CPB, cardiopulmonary bypass; Hct, hematocrit; n, number of patients in the study; N/D, not discussed; PC, platelet count; PFA, platelet function analyzer; RBC, red blood cell; TE, thromboelastography; TT, thrombin time; vWF, von Willebrand factor.
Coagulation Monitoring: A Review
Bolger et al,31 1990
9
10
Levy et al
Table 4 Retrospective studies assessing preoperative coagulation tests for predicting intraoperative and/or postoperative bleeding risk Type of Surgery and Reference
Age (n)
Laboratory Tests
PTT/aPTT Predictive
PT Predictive Other Findings/Conclusions
Adult (n 5 12,338)
aPTT
No
N/D
aPTT had no ability to predict the occurrence or absence of hemorrhage in patients at low clinical risk, but was a moderate predictor in high-risk patients. The data justify limiting preoperative coagulation screening to patients with active bleeding, known or suspected bleeding disorders, liver disease, malabsorption, malnutrition, or other conditions associated with acquired coagulopathies, and patients whose procedures may interfere with normal coagulation
PT, PTT, BT
No
No
The value of preoperative BT was not a reliable test for assessing the risk of clinically significant perioperative bleeding. Patients with a bleeding history were more likely to have abnormal BT but there was no statistically significant association between abnormal BT and the other indicators of bleeding risk examined, or the occurrence of clinically significant perioperative bleeding
General/not specified Suchman & Mushlin,36 1986
Gewirtz et al,37 1996 Adult/pediatric (n 5 167)
Reddy et al,38 1999
PTT Age not stated (n 5 199): grouped according to PTT: Group 0,
Conflict of Interest Statements: Prof J.H. Levy serves on steering committees for Boehringer Ingelheim, CSL Behring AG, Grifols, Janssen Pharmaceuticals (Johnson & Johnson), Merck, and The Medicines Company. F. Szlam, Dr AM Winkler, and Dr A.S. Wolberg have no relevant conflict of interest to disclose. a Department of Anesthesiology, Duke University School of Medicine, 2301 Erwin Road, Durham, NC 27710, USA; b Department of Surgery, Duke University School of Medicine, 2301 Erwin Road, Durham, NC 27710, USA; c Department of Critical Care, Duke University School of Medicine, 2301 Erwin Road, Durham, NC 27710, USA; d Emory University Hospital, Emory University School of Medicine, 1364 Clifton Road, Atlanta, GA 30322, USA; e Department of Pathology and Laboratory Medicine, University of North Carolina at Chapel Hill, 815 Brinkhous-Bullitt Building, CB# 7525, Chapel Hill, NC 27599-7525, USA; f Department of Pathology & Laboratory Medicine, Emory University Hospital, 1364 Clifton Road, Atlanta, GA 30322, USA * Corresponding author. Duke University Medical Center, 2301 Erwin Road, 5691H HAFS, Box 3094, Durham, NC 27710. E-mail address: [email protected] Clin Lab Med - (2014) -–http://dx.doi.org/10.1016/j.cll.2014.06.005 labmed.theclinics.com 0272-2712/14/$ – see front matter Ó 2014 Elsevier Inc. All rights reserved.
2
Levy et al
Continued A preoperative bleeding history and physical examination are usually obtained in an attempt to identify important bleeding risk factors. However, current questionnaires used to assess bleeding history are notoriously poor at characterizing bleeding. In such cases, follow-up hemostatic testing may be appropriate. The aPTT is extensively used to monitor therapy with unfractionated heparin and other anticoagulant agents, including direct thrombin inhibitors, whereas the PT and INR are used to monitor the anticoagulant effects of warfarin and other vitamin K antagonists, and to adjust their dosages.
INTRODUCTION
Standard coagulation monitoring is used to manage hemostasis and bleeding in hospitalized patients, including those undergoing surgical procedures and during cardiac surgery with cardiopulmonary bypass (CPB). Multiple factors influence coagulation in hospitalized patients, especially in a perioperative setting. Such factors include type of procedure, previous surgery at the same anatomic site, degree of tissue injury, and the underlying state of the hemostatic system. Surgical procedures that may be associated with increased bleeding potential due to high tissue vascularity include tonsillectomy and vascular and cardiac surgery. In other cases, bleeding risk is associated with the potential for related adverse events, especially if bleeding occurs in the central nervous system or other closed spaces, such as in ophthalmic procedures. Multiple assessments are routinely undertaken to assess the risk of bleeding in a given patient, including patient history, history of prior procedures with excessive bleeding, and family history. However, laboratory testing is often also extensively used in this evaluation. Tests that are performed most commonly include the activated partial thromboplastin time (aPTT), prothrombin time (PT), international normalized ratio (INR), and activated clotting time (ACT) for patients during cardiac surgery. The ACT is more often used for monitoring anticoagulation for CPB. Although laboratory testing with the aPTT and PT is appropriate to monitor anticoagulation, and the tests evolved soon after the introduction of warfarin to determine hemostatic abnormalities, the value of these tests to predict bleeding in surgical patients is not well documented despite their widespread use. Their usefulness is further complicated by underlying bleeding disorders, test characteristics, and the potential for false-positive and false-negative results. In the presence of a lupus anticoagulant or factor XII deficiency, for example, the aPTT may be prolonged, but this prolongation is not associated with an increased risk of bleeding.1 A systematic review of the literature published in 1997 suggests that preoperative laboratory tests of hemostasis before elective surgery in patients without a positive personal history for abnormal bleeding are not helpful and rarely lead to a change in clinical management of the patient.2 Furthermore, guidelines on the preoperative assessment of bleeding risk state that hemostatic tests are poor predictors of bleeding and that routine tests in patients without a history indicative of bleeding are not generally recommended.3,4 Despite these perspectives, these coagulation tests are frequently used for clinical decision-making in hospitalized patients. The purpose of this review is to examine the clinical applications of the aPTT and PT tests and their role in assessing perioperative bleeding risk for perioperative screening. In addition, the use of the ACT will be considered for anticoagulation monitoring for CPB as it is used extensively in the hospital setting.
Coagulation Monitoring: A Review
aPTT
The aPTT is a global coagulation screening test that is used for assessment of the coagulation status in patients with suspected acquired deficiencies of coagulation factors of the intrinsic and common pathways of the coagulation system. The test is affected by multiple factors, including the levels of factors VIII, IX, XI, XII, X, II, and fibrinogen. The aPTT is widely used for monitoring anticoagulation therapy with low levels of heparin (from 0.1 IU/mL to approximately 1 IU/mL). In a normal population, the aPTT varies, and this interindividual variability is reflected in a wide reference interval. The aPTT reference interval also differs between laboratories that use reagents with different factor sensitivities and different lipid compositions as well as different instrumentation. The aPTT reagent is a mixture of phospholipids and activators (eg, kaolin, silica, or ellagic acid). Studies have shown considerable differences in the responsiveness of various aPTT reagents to mild and moderate deficiencies of coagulation factors, particularly factors VIII and IX.5 Furthermore, in 1.5% to 3% of the population, aPTT is prolonged because of mild to severe factor XII deficiency; however, this prolongation is not associated with an increased risk of bleeding.1 In addition, laboratories determine a reference interval to encompass the central 95% of apparently healthy men and women with similar demographics to the hospitalized population.6 As a result, 5% of normal individuals will fall outside of the reference interval. Last, elevations of biological substances such as C-reactive protein have recently been published to interfere with commonly used aPTT reagents and will cause a false prolongation because of interferences with phospholipids, particularly phosphatidylcholine and phosphatidylethanolamine, which are commonly used in aPTT reagents.7 Clinical Uses of the aPTT
The aPTT is extensively used to monitor unfractionated heparin (UFH) therapy and other anticoagulant agents, including direct thrombin inhibitors. The limitations for this test include biological variability, insensitivity to some clinically important bleeding disorders (eg, factor XIII deficiency, a2-antiplasmin deficiency), variability in instrumentation and reagents, low sensitivity to common pathway deficiencies (fibrinogen, prothrombin), variability due to physiologic changes (eg, in pregnancy, physical stress, or trauma), clinically irrelevant prolongation due to certain factor deficiencies (eg, factor XII [one of the commonest causes of unexpected aPTT prolongation], prekallikrein, and high-molecular-weight kininogen deficiencies), and preanalytical errors such as improper specimen collection. PT AND INR
The PT measures the time required for clotting to occur after the addition of a source of tissue factor to recalcified citrated plasma in laboratory instruments and on point of care (POC) devices. The PT is measured by adding thromboplastin (a mixture of tissue factor, calcium, and phospholipid) to a patient’s citrated plasma sample, and clot formation is determined. It is used also as a screening assay to detect deficiencies of one or more coagulation factors (fibrinogen and factors II, V, VII, and X). The INR was introduced by the World Health Organization to overcome variability in PT results due to differing sensitivities of thromboplastin reagents produced by different manufacturers.8 The INR is the ratio of the patient’s PT value divided by the normal value (geometric mean PT value for non-anticoagulated patients), as determined by the local laboratory, raised to the International Sensitivity Index (ISI) value (usually between 1.0 and 2.0) for the reagent and analytical system used: INR 5 (PTpatient/PTgeomean)ISI.
3
4
Levy et al
Clinical Uses of the PT/INR
The PT/INR is used extensively to monitor the anticoagulant effects of warfarin and other vitamin K antagonists and to adjust their dosages. Clinically, it is an in vitro measure of the extrinsic and common coagulation pathways and should therefore detect deficiencies of factors II, V, VII and X, and very low fibrinogen concentrations. As with all coagulation tests, the PT/INR is limited by biological variability, insensitivity to many bleeding disorders (eg, factor XIII deficiency, a2-antiplasmin deficiency), variability in results due to differences in reagents and coagulation analyzers, and preanalytical errors. In addition, the PT results may be prolonged in young children. A study in healthy subjects categorized into 3 age groups (1–5 years, 6–10 years, and 11–18 years) showed no significant differences between the groups in terms of aPTT; however, children in the youngest age group had significantly higher mean PT compared with adults (P 5 .03).9 Table 1 summarizes a range of conditions that may be present based on the results of PT and aPTT testing. UTILITY OF aPTT AND PT: EVIDENCE FROM THE LITERATURE
Findings of previously published reviews of the literature on coagulation testing are summarized in Table 2.2,10–18 The overwhelming conclusion from these reviews is that routine preoperative hemostatic testing with aPTT and PT is not useful in asymptomatic patients with no known risk factors. Moreover, with aPTT and PT, the high rate of false-positive and false-negative results may lead to inappropriate precautionary measures or false reassurance, respectively.12 Individual studies assessing the value of preoperative aPTT and PT (and other tests) for predicting bleeding risk are summarized in Table 3 (prospective studies19–35) and Table 4 (retrospective studies36–48). In most cases, aPTT and PT were not shown to
Table 1 Summary of conditions that may be present based on the results of PT and aPTT testing PT Result
aPTT Result
Examples of Conditions That May Be Present
Prolonged
Normal
Normal
Prolonged
Decreased or defective factor VIII, IX XI, XII, prekalikrein, high-molecular-weight kininogen Type 3 vWD Presence of lupus anticoagulant
Prolonged
Prolonged
Decreased or defective fibrinogen, factor II, V, or X Severe liver disease Acute DIC
Normal
Normal or slightly prolonged
May indicate normal hemostasis; however, PT and aPTT can be normal in conditions such as mild deficiencies in other factors and in the mild form of vWD Further testing may be required to diagnose these conditions
Liver disease Vitamin K deficiency Decreased or defective factor VII Chronic, low-grade DIC Vitamin K antagonist (warfarin) therapy
Abbreviations: DIC, disseminated intravascular coagulation; vWD, von Willebrand disease.
Table 2 Published reviews of preoperative coagulation testing Reference
Age
Conclusions
Swedish Council on Technology Assessment,8,10 1991
Adult
Presurgical hemostatic tests are unnecessary in asymptomatic patients
Munro et al,2 1997
Adult/pediatric
Hemostatic tests have no value in predicting perioperative bleeding in the absence of clinical features
Peterson et al,11 1998
Adult/pediatric
In the absence of a history of excessive bleeding, the bleeding time fails as a screening test and is not indicated as a routine preoperative test
Chee & Greaves,12 2003
Adult/pediatric
Indiscriminate coagulation testing is not useful in a surgical or a medical setting, due to the limited sensitivity and specificity of the tests, and the high rate of false-positive and false-negative results
Eckman et al,13 2003
Adult/pediatric
For nonsurgical and surgical patients without synthetic liver dysfunction or a history of oral anticoagulant use, routine testing has no benefit in assessment of bleeding risk
Dzik,14 2004
Adult/pediatric
PT/INR had a poor predictive value for bleeding risk. Whether mild or moderate abnormalities of such tests have any clinically relevant predictive value as appropriate triggers for prophylactic transfusions before invasive procedures should be investigated in formal randomized clinical trials
Sie´ & Steib,15 2006
Adult/pediatric
Systematic preoperative screening is poorly efficient. It should be restricted to selected patients based on clinical history and physical examination
Kozek-Langenecker,16 2010
Adult/pediatric
Routine perioperative coagulation tests, including aPTT, PT, and INR, are poor predictors of bleeding and mortality. Such testing is still in use due to tradition rather than evidence
Adult/pediatric
Clinical assessment of hemostatic function before cardiac operations is both effective and efficient. It obviates routine laboratory testing and favors the introduction of blood conservation strategies early on during the process of care
Adult/pediatric
There is no difference in the rate of posttonsillectomy bleeding in patients with abnormal compared with normal preoperative coagulation tests
Any surgery
Klopfenstein,17 1996
Tonsillectomy Krishna & Lee,18 2001
Coagulation Monitoring: A Review
Cardiac surgery
5
6
Type of Surgery and Reference
Age (n)
Laboratory Tests
PTT/aPTT Predictive
PT Predictive
Other Findings/Conclusions
Eisenberg et al,19 1982
Age not stated (n 5 750)
PTT, PT
No
No
PT and PTT detected few unsuspected bleeding disorders preoperatively, and there were a large number of apparently false-positive results
Borzotta & Keeling,20 1984
Adult (n 5 83)
PTT, PT, PC, BT
Yes, with bleeding history
Yes, with bleeding history
Bleeding history should guide the selection of laboratory tests. Neither history nor laboratory tests alone provide insurance against hemorrhagic mishaps, but together they help the surgeon protect his or her charge from unexpected failure of hemostasis
Rohrer et al,21 1988
Adult/pediatric (n 5 282)
PTT, PT, PC, BT
No
No
Of 514 screening tests performed, 4.1% were abnormal but none identified a clinically significant coagulopathy. Preoperative screening tests for coagulopathies not suspected on the basis of detailed clinical information are unnecessary and should not be done
Macpherson et al,22 1993
Age not stated Study 1, preoperative screening (n 5 111, after excluding 45 patients who had taken aspirin in the previous week, and 3 with a history suggesting bleeding risk) Study 2, investigation of disproportionate hemorrhage (n 5 1872)
aPTT, PT, BT, PC
No
No
If patients have no relevant history and physical examination is negative, preoperative screening for coagulation defects would seem to be unnecessary
General/not specified
Levy et al
Table 3 Prospective studies assessing preoperative coagulation tests for predicting intraoperative and/or postoperative bleeding risk
Houry et al,23 1995
Adult, age 16–99 y (n 5 3242)
PTT, PT, PC, BT
No
No
Preoperative hemostatic screening tests should not be performed routinely but only in patients with abnormal clinical data
Koscielny et al,24 2004
Adult, age 17–87 y (n 5 5649)
aPTT, PT, PC, including PFA100 (1vWF and BT in patients with a positive bleeding history)
No
No
The sensitivity of the PFA-100: collagen-epinephrine test was highest (90.8%) in comparison with the other screening tests, with a high positive predictive value (81.8%) and a higher negative predictive value (93.4%)
Age not stated (n 5 156)
aPTT (5 per patient), PT (6 per patient), INR, PT ratio, aPTT ratio
Variable
Variable
Among 16 patients with increased bleeding during surgery, PT and aPTT test results (all performed with the same instrument in the same laboratory) varied markedly. Variability was not reduced by use of the PT or aPTT ratio, or INR, or by incorporation of a measure of PT or aPTT test sensitivity to factordeficient serum
Adult/pediatric, with congenital heart disease (n 5 235)
aPTT, PTT, PT, TT, PC
Yes
Yes
One or more test values were abnormal in 19% of subjects—a significantly higher incidence than that expected in a normal population. Prolonged PT, PTT, or aPTT were seen most frequently; in 6/8 such patients evaluated further, there were decreased levels of factor VII or IX. However, the authors noted that use of blood products during cardiac surgery was not statistically significantly different between patients with normal or abnormal tests, and normal preoperative coagulation tests did not exclude the presence of a major bleeding diathesis
Spinal Murray et al,25 1999
Cardiac
(continued on next page)
Coagulation Monitoring: A Review
Colon-Otero et al,26 1987
7
8
Levy et al
Table 3 (continued ) Type of Surgery and Reference
PTT/aPTT Predictive
PT Predictive
Other Findings/Conclusions
PTT, PT, BT, PC, Hct, packed RBC volume
No
No
Excessive postoperative blood transfusion was predicted by BT and RBC volume but not by PTT, PT, PC, or Hct (nor by recent use of heparin or aspirin)
Adult (n 5 897)
Post-CPB: ACT, aPTT, PT, TT, PC fibrinogen, fibrin/ fibrinogen degradation products, Duke’s earlobe BT
No
No
Because the predictive values of the tests are so low, it does not appear sensible to screen patients routinely using these clotting tests after CPB
Pediatric (n 5 494)
PTT, PT, Hct, PC, fibrinogen, TE (before and during CPB)
Yes
No
Platelet count provided maximum sensitivity (83%) and specificity (58%) for prediction of excessive blood loss; TE was the only variable associated with total products transfused
Adult (n 5 1546)
PT
N/D
No
Of 25 evaluable patients with abnormal PT values, the results were not predictive in 20 cases and 5 had indications for coagulation testing based on history or physical examination. In the absence of specific indications, routine preoperative PT testing before elective gynecologic operations does not contribute to patient care and should be eliminated
Age (n)
Laboratory Tests
Ferraris & Gildengorin,27 1989
Adult (n 5 159), grouped according to packed RBC transfusion: Group I, 5 units (n 5 139) Group II, >5 units (n 5 20)
Gravlee et al,28 1994
Williams et al,29 1999
Gynecologic Aghajanian & Grimes,30 1991
Tonsillectomy/adenoidectomy Adult/pediatric (n 5 52)
PTT, PT, PC, BT
Yes?
Yes?
Clinical history failed to detect any previously unrecognized coagulation disorder. PT was prolonged in 5.8%, PTT in 11.5%, and BT in 9.5%. Six patients (11.5%) were considered to have important laboratory abnormalities. Laboratory screening therefore improved preoperative detection of occult hemostatic defects and allowed for appropriate alterations in perioperative care
Burk et al,32 1992
Pediatric (n 5 1603)
CBC, aPTT, PT, BT
No
No
Laboratory abnormalities had a high specificity (0.99) and high negative predictive value (0.98) with a low sensitivity (0.03) and low positive predictive value (0.07) in predicting postoperative bleeding. The large number of false positive tests, coupled with the relative rarity of inherited and acquired coagulopathies, raises doubts about the overall value of routine screening
Close et al,33 1994
Adult/pediatric (n 5 96)
PT, aPTT
No
No
Routine measurement of aPTT and PT in asymptomatic patients is not useful for predicting postoperative bleeding
Kang et al,34 1994
Pediatric (n 5 1061)
PTT, PT, BT, PC
Yes
No
A coagulation profile that includes PTT and BT may be a valuable screening tool for children undergoing tonsillectomy and adenoidectomy. An initially abnormal coagulation profile may identify those more likely to bleed after surgery (22.6% vs 5.6%)
Zago´lski,35 2010
Adult/pediatric (n 5 222)
aPTT, PT/INR, PC
No
No
In healthy adults, coagulation test results are irrelevant for the course of tonsillectomy and postoperative bleeding
Abbreviations: BT, bleeding time; CBC, complete blood count; CPB, cardiopulmonary bypass; Hct, hematocrit; n, number of patients in the study; N/D, not discussed; PC, platelet count; PFA, platelet function analyzer; RBC, red blood cell; TE, thromboelastography; TT, thrombin time; vWF, von Willebrand factor.
Coagulation Monitoring: A Review
Bolger et al,31 1990
9
10
Levy et al
Table 4 Retrospective studies assessing preoperative coagulation tests for predicting intraoperative and/or postoperative bleeding risk Type of Surgery and Reference
Age (n)
Laboratory Tests
PTT/aPTT Predictive
PT Predictive Other Findings/Conclusions
Adult (n 5 12,338)
aPTT
No
N/D
aPTT had no ability to predict the occurrence or absence of hemorrhage in patients at low clinical risk, but was a moderate predictor in high-risk patients. The data justify limiting preoperative coagulation screening to patients with active bleeding, known or suspected bleeding disorders, liver disease, malabsorption, malnutrition, or other conditions associated with acquired coagulopathies, and patients whose procedures may interfere with normal coagulation
PT, PTT, BT
No
No
The value of preoperative BT was not a reliable test for assessing the risk of clinically significant perioperative bleeding. Patients with a bleeding history were more likely to have abnormal BT but there was no statistically significant association between abnormal BT and the other indicators of bleeding risk examined, or the occurrence of clinically significant perioperative bleeding
General/not specified Suchman & Mushlin,36 1986
Gewirtz et al,37 1996 Adult/pediatric (n 5 167)
Reddy et al,38 1999
PTT Age not stated (n 5 199): grouped according to PTT: Group 0,
