ANESTH ANALG CARDIOVASCULAR ANESTHESIA MCCRATH ET AL. 15772005;100:1576 –83 TEG® MA PREDICTS THROMBOTIC COMPLICATIONScontinue to support this concept (2,3). Hypercoagula- index, which has been shown to be predictive in as-ble states, including both inherited and acquired con- sessing risk of major cardiac complications, includingditions, have been shown to be a potential mechanism MI, after noncardiac surgery (14). Risk factors in-in postoperative ischemic stroke (4 –7). cluded in this risk scoring system are high-risk type of Classically, in the medical (nonsurgical) setting, cor- surgery, presence of ischemic heart disease, history ofonary atherosclerosis and plaque rupture are thought congestive heart failure, history of cerebrovascularto be the main factors in the initial pathogenesis of the disease, preoperative use of insulin therapy for diabe-majority of MIs. The role of the coagulation system in tes, and preoperative serum creatinine more thanacute coronary syndromes has been well recognized in 2.0 mg/dL.the field of cardiology, hence the widespread use of Study patients received routine surgical and an-prophylactic therapies, such as aspirin and heparin, in esthetic care for their procedures. Patients receivednonsurgical patients. The process of undergoing a ma- standard postoperative care including routine post-jor surgical procedure has been demonstrated to in- operative thromboprophylaxis with lower extremityduce an acquired postoperative hypercoagulable state. pneumatic compression devices or subcutaneousPotential mechanisms for this prothrombotic state in- (SQ) heparin administration. Orthopedic surgicalclude increased platelet activation, decreased fibrino- patients received standard postoperative warfarinlysis, and decreased anticoagulants (antithrombin III). dosing. During each patient’s postoperative courseNo studies have assessed the role of postoperative until discharge from the hospital, study personnelhypercoagulability on MI after major surgery. blinded as to the TEG® results assessed the patient Most assays of clotting (e.g., prothrombin time, ac- for the presence of thrombotic complications. Thesetivated partial thromboplastin time) can identify a complications included DVT, PE, ischemic stroke,specific factor deficiency; however, their performance and MI. New onset MI was assessed using the Jointin plasma (rather than whole blood) and the addition European Society of Cardiology/American Collegeof buffered solutions limits their relevance to overall of Cardiology definition (15). DVT was diagnoseddynamic clot formation in whole blood. In contrast, based on new clinical findings resulting in changesthromboelastography (TEG®) was designed specifi- in treatment and positive venous duplex ultrasoundcally to assess overall clotting kinetics and strength in scanning. PE was diagnosed based on clinical find-whole blood. TEG® has been successfully used in clin- ings along with ventilation/perfusion scanning orical settings to detect hypercoagulable states. TEG® pulmonary computed tomography (CT) angiogra-has been increasingly used in the assessment of post- phy. Ischemic stroke was confirmed after clinicaloperative hypercoagulability for a variety of surgical findings with head CT.procedures (8 –13). Within 2 h after completion of a procedure, 0.5 mL Therefore, we designed a prospective cohort study of whole blood was sampled from an indwelling in-to test the hypothesis that a hypercoagulable state at travascular catheter. According to the manufacturer’sthe end of surgery as revealed by TEG® is associated guidelines, celite-activated TEG® using a Throm-with an increased incidence of postoperative throm- boelastograph Coagulation Analyzer (TEG®; Haemo-botic complications, including MI. scope Corporation, Skokie, IL) was performed on the blood samples within 4 min of collection, obviating the need for citrated blood. Maximum amplitude (MA) from the TEG® tracing, representing clotMethods strength, was recorded for each patient. MA is a directAfter IRB approval and informed consent, 240 patients function of the maximum dynamic properties of fibrinundergoing major elective noncardiac surgery at and platelet bonding via GPIIb/IIIa and represents theColumbia-Presbyterian Medical Center between July ultimate strength of the fibrin clot. The upper limit ofand October 2002 were enrolled in a prospective normal for MA in celite-activated TEG® is 68 mm (16).blinded cohort study. Patients were identified for en- No relevant reference standard exists for dynamicrollment in the study based on the severity of the clotting of whole blood.surgical procedure to be performed; only patients un- The objective of this study was to test the hypothesisdergoing surgery requiring inpatient admission were that a postoperative hypercoagulable state as deter-selected for enrollment. No patients undergoing am- mined by TEG® is associated with an increased likeli-bulatory or emergency surgery were included in the hood of postoperative thrombotic complications in-study. cluding MI. Patients were divided into control and Patients providing informed consent underwent hypercoagulable (MA Ͼ68 mm) groups based onpreoperative evaluations including detailed medical TEG® MA values, and the incidence of thrombotichistories, physical examinations, and laboratory test- complications were recorded for each group. A Fish-ing. A cardiac risk score was assigned to each patient er’s exact test was conducted to test the significance ofusing the established Revised Goldman cardiac risk differences between groups. This analysis was also
1578 CARDIOVASCULAR ANESTHESIA MCCRATH ET AL. ANESTH ANALG TEG® MA PREDICTS THROMBOTIC COMPLICATIONS 2005;100:1576 –83conducted examining the occurrence of new postop- setting. Both the TEG® variable MA at the end of surgeryerative MI. For all analyses a significance level of P ϭ (odds ratio [OR], 1.14; 95% confidence interval [CI], 1.03–0.05, two-tailed, was assumed. 1.27; P ϭ 0.014) and the established Goldman Risk Score Given the observed association between the TEG® (OR, 2.47; 95%CI, 1.01– 6.04; P ϭ 0.047) were univariatevariable MA and postoperative MI and the availability predictors of postoperative MI. The association betweenof an established risk score for MI (14), a multivariate age and postoperative MI did not achieve significance (Panalysis was performed to determine if MA predicted ϭ 0.1028). In a multivariate analysis, MA at the end ofpostoperative MI independent of known risk factors surgery independently predicted postoperative MI (OR,using the Revised Goldman Risk Index. The indepen- 1.16; 95%CI, 1.03–1.20; P ϭ 0.013) over and above thedent variables included MA and Revised Goldman effects of Goldman Risk Score (OR, 2.39; 95%CI, 1.02–Risk Index, and the dependent variable was postop- 5.61; P ϭ 0.046). This overall model was significant at theerative MI. In addition, use of a dichotomous cutpoint P ϭ 0.008 level (model 2 9.784; df ϭ 2). Forcing thefor MA was analyzed using a cutpoint of 68 mm, independent variable age into this model still resulted inwhich is the test’s upper limit of normal. statistical significance (P ϭ 0.0158) for the variable MA at predicting MI. In addition, use of a dichotomous cut- point for MA was also analyzed using a cutpoint of 68 mm, which is the test’s upper limit of normal. ThisResults analysis revealed a robust effect corresponding with anOverall, 240 patients participated in the study. Twelve OR of 6.6 (P Ͻ 0.01).postoperative thrombotic complications occurred in 10patients (MI, n ϭ 6; DVT, n ϭ 2; PE, n ϭ 2; cerebrovas-cular accident, n ϭ 2). Perioperative variables are shown Discussion(Table 1) for patients with (n ϭ 95) and without (n ϭ 145) Little attention has been focussed on characterizingincreased MA at the end of surgery. Of note, age, gender, the incidence and clinical relevance of a hypercoagu-body mass index, Goldman Cardiac Risk score, type of lable state on patient outcome in the perioperativeanesthetic, and surgical procedure type were similar be- period. We have completed a prospective study of 240tween groups. Postoperative prophylactic anticoagula- patients undergoing a diverse group of surgical pro-tion management was also similar between groups. The cedures and found that a hypercoagulable state, asincidence of thrombotic complications was significantly manifested by increased MA at the end of surgery, ismore frequent in patients in the increased MA group associated with postoperative thrombotic complica-(Fig. 1). Moreover, all postoperative MIs were observed tions, including MI.in patients with a hypercoagulable state as manifested It is increasingly recognized that postoperativeby increased MA (Fig. 2). The sensitivity and specificity thrombotic complications are common and consumeof the increased TEG® MA value for all thrombotic com- considerable amounts of health care resources. In-plications included was 80% and 62%, respectively, deed, the significant incidences of postoperative MIwhereas for MI alone the sensitivity and specificity were and other thrombotic complications occurring in our100% and 61%, respectively. The positive predictive study are consistent with the rates documented in thevalue and negative predictive value of the increased literature (1,14,17–23). Hypercoagulable states, such asTEG® MA value for all thrombotic complications in- those shown to be induced by surgery, have been citedcluded were 8% and 99%, respectively, whereas for MI as a factor in the development of the thrombotic com-alone these were 6% and 100%, respectively. plications we studied, including ischemic stroke (4 – 6), Perioperative variables for patients with and without MI (7,24), DVT, and venous thromboembolism (2,3).a thrombotic complication (Table 2) and MI (Table 3) are Despite these well-known links, studies developingshown. Of note, postoperative hospital length of stay preoperative risk stratification or assessment systemswas significantly longer in patients with a thrombotic typically do not include information concerning coag-complication or MI. Patients who experienced a throm- ulation and, specifically, hypercoagulable states, as abotic complication or MI were just as likely to have risk factor for perioperative morbidity and mortality.received SQ heparin or coumadin postoperatively for Given the infrequent occurrence of heritable or ac-thromboprophylaxis. Interestingly, none of the patients quired hypercoagulable states existing without previ-who suffered a postoperative MI had received aspirin in ous clinical manifestations, screening for such states inthe postoperative period (Table 3). the clinical or perioperative setting is rarely recom- This study was not designed to rigorously develop mended. This is a result of the small yield and result-and validate a predictive model of postoperative MI. ant poor cost-effectiveness of the extensive battery ofNevertheless, we were interested in ascertaining tests for cases such as Factor V Leiden mutation, pro-whether the TEG® variable MA at the end of surgery tein C, protein S, and antithrombin III deficiencies, orwas an independent predictor of postoperative MI or antiphospholipid antibodies and hyperhomocystine-merely closely linked to known predictors of MI in this mia (25). However, TEG® has shown a promising
ANESTH ANALG CARDIOVASCULAR ANESTHESIA MCCRATH ET AL. 15792005;100:1576 –83 TEG® MA PREDICTS THROMBOTIC COMPLICATIONSTable 1. Perioperative Variables in Patients with End of Surgery Maximum Amplitude (MA) Յ 68 Versus MA Ͼ 68 MA Յ 68 MA Ͼ 68 Variable (n ϭ 145) (n ϭ 95) Age (yr) 58 Ϯ 16 (59) 61 Ϯ 16 (63) Gender (% male) 48 42 Body mass index 28 Ϯ 8 (27) 29 Ϯ 8 (27) Goldman cardiac risk 0.78 Ϯ 0.62 (1.0) 0.74 Ϯ 0.80 (1.0) % general anesthetic 85 84 % regional anesthetic 10 16 % combined technique 5 0 Intraoperative LR (L) 3.2 Ϯ 1.6 (2.9) 3.2 Ϯ 1.6 (3.0) Neurosurgery (%) 12 19 General surgery 27 15 Gynecologic surgery 7 14 Vascular surgery 21 15 Spinal surgery 12 13 Orthopedic surgery 9 21 Urology/ENT surgery 12 5 Postoperative SQ heparin (%) 21 24 Postoperative coumadin (%) 8 11 Postoperative aspirin (%) 7 7 Postoperative any anticoag. (%) 37 43 Postoperative length of stay (days) 5.1 Ϯ 4.7 (3.0) 5.6 Ϯ 4.9 (4.0) Postoperative thrombotic complication 2 (1.4) 8 (8.4)* Postoperative MI 0 (0) 6 (6.3)† Values are mean Ϯ sd and median in parenthesis or n (%) where applicable. LR ϭ lactated Ringer’s solution; SQ ϭ subcutaneous; Postoperative any coag. ϭ received SQ heparin, coumadin, or aspirin postoperatively; MI ϭ myocardialinfarction; ENT ϭ ear, nose, throat. * P ϭ 0.0157 compared with the MA Յ68 group; † P ϭ 0.0035 compared with the MA Յ68 group.Figure 1. Patients with a confirmed thrombotic complication (deepvein thrombosis, pulmonary embolism, myocardial infarction, cere- Figure 2. Patients with a confirmed myocardial infarctionbrovascular accident) by presence or absence of increased maxi- by presence or absence of increased maximum amplitude.mum amplitude. Crosshatched bar ϭ control group; solid bar ϭ MA Crosshatched bar ϭ control group; solid bar ϭ MA Ͼ68 group.Ͼ68 group. *P ϭ 0.0157. *P ϭ 0.0035. TEG® results were also normal on conventionalability to detect known or established hypercoagula- testing.ble states during pregnancy and postpartum (26 –28) Although a larger formal investigation is neededor during the presence of cancer (29 –32). Furthermore, to establish TEG® as a widespread screening test fora study by Handa et al. (33) suggested that TEG® prothrombotic states, TEG® clearly shows an abilitycould play a valuable role in screening patients with to detect hypercoagulable states produced by sur-suspected prothrombotic states. In this study of 103 gery. Potential causes for this hypercoagulable statepatients, 49 were found to have hypercoagulable or include surgical trauma, systemic inflammation, tis-borderline TEG® results. Of these 49 patients, 31 (63%) sue factor expression, platelet activation, and crys-were then identified as having a defined prothrom- talloid administration.botic abnormality by conventional testing for acquired Despite the limitation that TEG® cannot specify theor heritable hypercoagulable states. Furthermore, mechanism for hypercoagulable states, it is nonethe-100% of the study’s 31 patients found to have normal less a valuable tool for assessing the presence of these
1580 CARDIOVASCULAR ANESTHESIA MCCRATH ET AL. ANESTH ANALG TEG® MA PREDICTS THROMBOTIC COMPLICATIONS 2005;100:1576 –83Table 2. Perioperative Variables in Patients with and without a Postoperative Thrombotic Complication (DVT, PE, MI,and/or CVA) Thrombotic Thrombotic Complication—NO Complication—YES Variable (n ϭ 230) (n ϭ 10) Age (yr) 59 Ϯ 16 (61) 68 Ϯ 19 (75) Gender (% male) 44 70 Body mass index 28 Ϯ 8 (27) 27 Ϯ 5 (26) Goldman cardiac risk 0.75 Ϯ 0.67 (1.0) 1.1 Ϯ 1.2 (1.0) % general anesthetic 85 90 % regional anesthetic 12 10 % combined technique 3 0 Intraoperative LR (L) 3.2 Ϯ 1.6 (3.0) 2.7 Ϯ 1.3 (2.0) Neurosurgery (%) 14 30 General surgery 23 0 Gynecologic surgery 10 10 Vascular surgery 18 20 Spinal surgery 13 10 Orthopedic surgery 14 20 Urology/ENT surgery 9 10 Postoperative SQ heparin (%) 22 30 Postoperative coumadin (%) 9 10 Postoperative aspirin (%) 7 0 Postoperative any anticoag. (%) 39 40 MA at end of surgery 66 Ϯ 9 (66) 71 Ϯ 9 (70) R at end of surgery 7.1 Ϯ 5.7 (6.0) 8.0 Ϯ 4.4 (7.5) Postoperative length of stay (days) 4.9 Ϯ 3.9 (3.0) 16.9 Ϯ 10.2 (16.0) Values are mean Ϯ sd and median in parenthesis where applicable. LR ϭ lactated Ringer’s solution; SQ ϭ subcutaneous; Postoperative any coag. ϭ received SQ heparin, coumadin, or aspirin postoperatively; MA ϭthrombolestographic variable maximum amplitude; R ϭ thrombolestographic variable R; DVT ϭ deep vein thrombosis; PE ϭ pulmonary embolism; MI ϭmyocardial infarction; CVA ϭ cerebrovascular accident; ENT ϭ ear, nose, throat.states. The utility of TEG® in the measurement of and lower extremity Doppler sonography postopera-postoperative hypercoagulability is evident in many tively and, again, limited occurrence of postoperativestudies within a variety of noncardiac surgical spe- complications prevented significant statistical analy-cialty fields including neurosurgery (13), orthopedic sis. In addition, a study by Traverso et al. (36) includedsurgery (8), abdominal surgery (9,34), and vascular 100 patients undergoing elective abdominal surgery,surgery (35). Indeed, our study included a variety of and in the patients randomized to receive no postop-the aforementioned noncardiac surgical specialties erative heparin thromboprophylaxis, TEG® MA valueand documented 95 of the 240 study patients (40%) as showed the ability to predict the occurrence of DVTexhibiting a postoperative hypercoagulable state. with a sensitivity of 72.2% and specificity of 69% (37). Despite the established link between the occurrence Finally, Wilson et al. (8) performed TEG® every otherof thrombotic complications and the induction of hy- day in 250 patients having undergone proximal fem-percoagulability by major surgery and the ability of oral fracture repair and showed that patients sufferingTEG® to measure hypercoagulable states and specifi- from postoperative DVT had a significantly highercally postoperative hypercoagulability, no large, well level of hypercoagulability as measured by TEG® thandesigned studies have assessed the association be- did those who did not suffer DVT.tween postoperative hypercoagulability as measuredby TEG® and the occurrence of a multitude of throm- Our study is the first to demonstrate an associationbotic complications in a diverse group of surgical pa- between hypercoagulability as measured by TEG®tients. Caprini et al. (34) attempted to study the asso- and postoperative thrombotic complications in pa-ciation of TEG® and postoperative DVT in a series of tients undergoing a diverse set of surgical procedures.100 patients undergoing laparoscopic cholecystec- The aforementioned previous studies were limited bytomy. TEG® measurements were measured postoper- power or methodological limitations, including focusatively and patients were followed for the occurrence on a single type of surgical procedure (e.g., orthope-of DVT; however, only one DVT complication was dic) and limiting thrombotic complications to DVTobserved and thus no statistical association could be only. Our study included 240 patients undergoing ademonstrated. Abrahams et al. (13) studied 46 patients variety of major noncardiac surgical procedures whoundergoing neurosurgery with TEG® measurements were followed postoperatively until discharge. Several
ANESTH ANALG CARDIOVASCULAR ANESTHESIA MCCRATH ET AL. 15812005;100:1576 –83 TEG® MA PREDICTS THROMBOTIC COMPLICATIONSTable 3. Perioperative Variables in Patients with and without a Postoperative Myocardial Infarction Myocardial Myocardial Infarction—NO Infarction—YES Variable (n ϭ 234) (n ϭ 6) Age (yr) 59 Ϯ 16 (61) 70 Ϯ 21 (75) Gender (% male) 45 67 Body mass index 28 Ϯ 8 (27) 27 Ϯ 6 (26) Goldman cardiac risk 0.75 Ϯ 0.66 (1.0) 1.33 Ϯ 1.51 (1.0) % general anesthetic 85 83 % regional anesthetic 12 17 % combined technique 3 0 Intraoperative LR (L) 3.2 Ϯ 1.6 (3.0) 2.2 Ϯ 0.7 (2.0) Neurosurgery (%) 15 33 General surgery 23 0 Gynecologic surgery 10 0 Vascular surgery 18 17 Spinal surgery 12 17 Orthopedic surgery 13 33 Urology/ENT surgery 9 0 Postoperative SQ heparin (%) 22 33 Postoperative coumadin (%) 9 17 Postoperative aspirin (%) 7 0 Postoperative any anticoag. (%) 39 50 MA at end of surgery 66 Ϯ 9 (66) 74 Ϯ 5 (72) R at end of surgery 7.2 Ϯ 5.7 (6.0) 6.5 Ϯ 3.3 (6.0) Postoperative length of stay (days) 5.1 Ϯ 4.4 (3.0) 15.0 Ϯ 9.5 (15.0) Values are mean Ϯ sd and median in parenthesis where applicable. LR ϭ lactated Ringer’s solution; SQ ϭ subcutaneous; Postoperative any coag. ϭ received SQ heparin, coumadin, or aspirin postoperatively; MA ϭthrombolestographic variable maximum amplitude; R ϭ thrombolestographic variable R; ENT ϭ ear, nose, throat.conclusions can be drawn from our results. First, post- identifying patients at high risk for thrombotic com-operative TEG® MA values in the hypercoagulable plications. It is interesting to speculate that individualsrange are associated with a higher risk of postopera- affected with hypercoagulable states postoperativelytive thrombotic complications including PE, DVT, MI, might benefit from a more intensive anticoagulationand ischemic stroke. These findings, along with the treatment, perhaps involving platelet-specific drugs.aforementioned research linking thrombotic complica- A potential limitation of our study is the lack of ations with hypercoagulability, suggest that surgical preoperative TEG® measurement for comparisonpatients are at high risk for hypercoagulability and with the postoperative TEG® values. The goal of ourthat this plays an important role in the pathogenesis of study, however, was to show that a postoperativethrombotic complications. Second, patients with in- hypercoagulable state as manifested by TEG® is as-creased MA values in the postoperative setting are at sociated with the occurrence of thrombotic compli-particularly higher risk for postoperative MI. Despite cations, regardless of whether or not the postoper-preoperative risk adjustment based on an established ative hypercoagulable state is a result of the surgerycardiac risk index for noncardiac surgery, patients itself or to a preexisting heritable or acquired pro-with hypercoagulable TEG® MA values had a signif- thrombotic condition.icantly more frequent incidence of postoperative MI. Our study was not designed or powered to compareGiven the morbidity and mortality of postoperative the associations between TEG® MA and thromboticMI, patients identified with hypercoagulable TEG® complications between subgroups of patients under-MA values postoperatively may benefit from aspirin going specific surgical procedures. This could be con-in the immediate perioperative period. Although post- sidered a limitation of the study. However, the inci-operative aspirin has been shown to decrease the in- dence of thrombotic complications including MI alonecidence of postoperative MI and stroke after coronary were not localized to a specific procedure type, sug-bypass (38), it is typically held in the immediate peri- gesting applicability of these results to most patientsoperative period to improve hemostasis. In light of undergoing major surgery.Mangano et al.’s report (38) involving cardiac surgical A significant consideration in any study assessingpatients, it is interesting to note that none of the pa- postoperative complications is proper risk stratificationtients with postoperative MI in our study received and adjustment of patients to minimize the likelihoodaspirin postoperatively. The high sensitivity of in- that patients suffering postoperative complications sim-creased TEG® MA values validates it as a tool for ply had increased preoperative morbidities. We used the
1582 CARDIOVASCULAR ANESTHESIA MCCRATH ET AL. ANESTH ANALG TEG® MA PREDICTS THROMBOTIC COMPLICATIONS 2005;100:1576 –83revised Goldman cardiac risk index for noncardiac sur- 5. Hart R, Hindman B. Mechanisms of perioperative cerebral in- farction. Stroke 1982;13:766 –73.gery to risk-adjust for the presence of postoperative MI. 6. Bushnell CD, Goldstein LB. Diagnostic testing for coagulopa-There is no well-established risk index for the occurrence thies in patients with ischemic stroke. Stroke 2000;31:3067–78.of postoperative DVT, PE, and stroke in the setting of 7. Moster ML. Coagulopathies and arterial stroke. J Neuroophthal-noncardiac surgery. The focus of proposed risk indices mol 2003;23:63–71. 8. Wilson D, Cooke EA, McNally MA, et al. Changes in coagula-for DVT and venous thromboembolism have focused on bility as measured by thrombelastography following surgery forprophylactic treatments to be used with given risk fac- proximal femoral fracture. Injury 2001;32:765–70.tors rather than the incidence of events associated with 9. Mahla E, Lang T, Vicenzi MN, et al. Thromboelastography forthese risk factors. monitoring prolonged hypercoagulability after major abdomi- nal surgery. Anesth Analg 2001;92:572–7. The length of postoperative TEG® monitoring has 10. Gibbs NM, Crawford GP, Michalopoulos N. Thrombelasto-been scrutinized by previous studies examining post- graphic patterns following abdominal aortic surgery. Anaesthoperative hypercoagulability measured by TEG®. In Intensive Care 1994;22:534 – 8.studies measuring TEG® values from the immediate 11. Bell CR, Cox DJ, Murdock PJ, et al. Thrombelastographic eval- uation of coagulation in transurethral prostatectomy. Br J Urolpostoperative period to between 3 and 7 days postop- 1996;78:737– 41.eratively, TEG® MA levels were increased immedi- 12. Arcelus JI, Traverso CI, Caprini JA. Thromboelastography forately after surgery and remained increased over the the assessment of hypercoagulability during general surgery.subsequent measurement days (8,9,39). Therefore, it is Semin Thromb Hemost 1995;21 Suppl 4:21– 6. 13. Abrahams JM, Torchia MB, McGarvey M, et al. Perioperativepossible that little benefit is gained from the repeated assessment of coagulability in neurosurgical patients usingmeasurement of TEG® over a prolonged postoperative thromboelastography. Surg Neurol 2002;58:5–11.period. Our results strongly suggest that most (if not 14. Lee TH, Marcantonio ER, Mangione CM, et al. Derivation and prospective validation of a simple index for prediction of car-all) hypercoagulable states associated with MI can be diac risk of major noncardiac surgery. Circulation 1999;100:diagnosed with a simple diagnostic test at the end of 1043–9.surgery. If this holds true in subsequent studies, it 15. Alpert JS, Thygesen K, Antman E, Bassand JP. Myocardial in-would obviate the need for more complex and lengthy farction redefined: a consensus document of The Joint European Society of Cardiology/American College of Cardiology Com-screening procedures. Of note, the MIs in our study mittee for the redefinition of myocardial infarction. J Am Collwere observed on postoperative days 1 (n ϭ 4), 2 (n ϭ Cardiol 2000;36:959 – 69.1), and 4 (n ϭ 1), which argues for the relevance of 16. Srinivasa V, Gilbertson LI, Bhavani-Shankar K. Thrombo-measurements obtained immediately after surgery. elastography: where is it and where is it heading? Int Anes- thesiol Clin 2001;39:35– 49. Finally, we did not perform screening diagnostic 17. Mangano DT, Goldman L. Preoperative assessment of patientstests on all patients for the thrombotic complications with known or suspected coronary disease. N Engl J Med 1995;we included in the study, including MI, so we cannot 333:1750 – 6.exclude the occurrence of clinically occult complica- 18. Mangano DT. Perioperative cardiac morbidity. Anesthesiology 1990;72:153– 84.tions that could have been revealed by diagnostic tests 19. Larsen SF, Olesen KH, Jacobsen E, et al. Prediction of cardiacalone. Although we do not know the overall relevance risk in non-cardiac surgery. Eur Heart J 1987;8:179 – 85.of these occult complications, length of stay was con- 20. Ashton CM, Petersen NJ, Wray NP, et al. The incidence of perioperative myocardial infarction in men undergoing noncar-sistently longer among those patients we found to diac surgery. Ann Intern Med 1993;118:504 –10.have complications, demonstrating the potential clin- 21. Knapp RB, Topkins MJ, Artusio JF Jr. The cerebrovascular acci-ical relevance of the complications we did diagnose. dent and coronary occlusion in anesthesia. JAMA 1962;182: In summary, increased postoperative TEG® MA val- 332– 4. 22. Huag CA, Dale WA. Major surgery in old people. Arch Surgues immediately after surgery were associated with a 1952;64:421–37.higher risk of postoperative thrombotic complications, 23. Wilder RJ, Fishbein RH. Operative experience with patientsincluding MI. On the basis of our findings, a randomized over 80 years of age. Suvr Med (Sofiia) 1961;113:205–12.clinical trial is being planned to assess the utility of TEG® 24. Falk E, Shah PK, Fuster V. Coronary plaque disruption. Circu- lation 1995;92:657–71.to improve the safety and efficacy of antithrombotic 25. Eckman MH, Erban JK, Singh SK, Kao GS. Screening for the riskstrategies in patients undergoing major surgery. for bleeding or thrombosis. Ann Intern Med 2003;138:W15–24. 26. Sharma SK, Philip J, Wiley J. Thromboelastographic changes in healthy parturients and postpartum women. Anesth Analg 1997;85:94 – 8. 27. Steer PL, Krantz HB. Thromboelastography and Sonoclot anal-References ysis in the healthy parturient. J Clin Anesth 1993;5:419 –24. 1. Bick RL, Haas S. Thromboprophylaxis and thrombosis in med- 28. Rai R, Tuddenham E, Backos M, et al. Thromboelastography, ical, surgical, trauma, and obstetric/gynecologic patients. He- whole-blood haemostasis and recurrent miscarriage. Hum Re- matol Oncol Clin North Am 2003;17:217–58. prod 2003;18:2540 –3. 2. Rosendaal FR. Venous thrombosis: a multicausal disease. Lancet 29. Caprini JA, Zuckerman L, Cohen E, et al. The identification of 1999;353:1167–73. accelerated coagulability. Thromb Res 1976;9:167– 80. 3. Bick RL, Kaplan H. Syndromes of thrombosis and hypercoagu- 30. Francis JL, Francis DA, Gunathilagan GJ. Assessment of hyper- lability. Congenital and acquired causes of thrombosis. Med coagulability in patients with cancer using the Sonoclot Ana- Clin North Am 1998;82:409 –58. lyzer and thromboelastography. Thromb Res 1994;74:335– 46. 4. Kelley RE. Stroke in the postoperative period. Med Clin North 31. Hathaway WE, Hays T. Hypercoagulability in childhood can- Am 2001;85:1263–76. cer. J Pediatr Surg 1975;10:893–9.
ANESTH ANALG CARDIOVASCULAR ANESTHESIA MCCRATH ET AL. 15832005;100:1576 –83 TEG® MA PREDICTS THROMBOTIC COMPLICATIONS32. Raina S, Spillert CR, Greenstein SM, Lazaro EJ. Effect of surgery 37. Salooja N, Perry DJ. Thrombelastography. Blood Coagul Fibri- on tumor-induced accelerated coagulation in a rat carcinoma. nolysis 2001;12:327–37. J Surg Res 1985;38:138 – 42. 38. Mangano DT, Multicenter Study of Perioperative Ischemia Re-33. Handa A, Cox DJ, Pasi KJ. Thromboelastography: an effective search Group. Aspirin and mortality from coronary bypass screening test for prothrombotic states. Br J Surg 1998;85:90. surgery. N Engl J Med 2002;347:1309 –17.34. Caprini JA, Arcelus JI, Laubach M, et al. Postoperative hyper- 39. Gibbs NM, Crawford GP, Michalopoulos N. Postoperative coagulability and deep-vein thrombosis after laparoscopic cho- changes in coagulant and anticoagulant factors following ab- lecystectomy. Surg Endosc 1995;9:304 –9. dominal aortic surgery. J Cardiothorac Vasc Anesth 1992;6:35. Butler MJ. Thrombelastography during and after elective ab- 680 –5. dominal surgery. Thromb Haemost 1978;39:488 –95.36. Traverso CI, Arcelus JI, Gomez E, et al. Prospective assessment of the risk of deep vein thrombosis in elective abdominal surgery: predictive role of thromboelastography. Thromb Hae- morrh Disorders 1993;7:9 –15.