Journal of Thrombosis and Haemostasis, 3: 26072611

HISTORICAL SKETCH

To bleed or not to bleed? is that the question for the PTT?

To cite this article: Kitchens CS. To bleed or not to bleed? is that the question for the PTT? J Thromb Haemost 2005; 3: 260711.

Summary. The activated partial thromboplastin time (PTT) is the grandchild of the Lee-White whole blood clot time (WBCT). Both tests were developed to assist the diagnostic process for patients who exhibited features consistent with hemophilia, i.e., the pretest probability was extremely high. Probably <0.1%1.0% of PTTs ordered in current medical practice t that indication with the most common indication now being routine admission order; i.e., a pretest probability that is extremely low. The question asked of the PTT has evolved from why does this patient bleed? to will this patient bleed? As the PTT was never intended to answer that question, one must be careful regarding interpretation of results of that test. As many situations not related to hemorrhage are associated with perturbations of the PTT, a prolonged PTT is not strongly predictive of hemorrhage nor does a normal PTT provide shelter against hemorrhagic risk. Keywords: hemostasis, laboratory testing, partial thromboplastin time, preoperative testing. We teach that laboratory tests are best ordered to answer a particular question. At a diagnostic or therapeutic branch point, one test result may favor one approach or action whereas another result will suggest a different path. A corollary is that one should not order a test unless one knows what one will do with either a positive or negative result. A nely focused test, such as an HIV viral load, yields only prognostic and therapeutic information. A broader pair of tests, such as serum ALT and AST, is traditionally ordered as a so-called liver function test, yet unexpected positive results may be the rst or only clue to muscle disease or even unsuspected thyroid disease underlying the myopathy. Some tests are ordered merely as a tradition or maybe even a habit, a practice that is difcult to defend. Arguably, the partial thromboplastic time (PTT) is a prime example of the last category. The PTT is one of the most frequently ordered tests on hospital admission. Not only is
Correspondence: C. S. Kitchens, Department of Hematology/ Oncology, University of Florida, FL, USA; Associate Chief of Sta for Education, North Florida/South Georgia Veterans Health System. Tel.: +1 352379 7486; fax: +1 352 374 6116; e-mail: craig. [email protected] 2005 International Society on Thrombosis and Haemostasis

rationale lacking to support that practice, but either consciously or subconsciously the PTT is held out to be a touchstone to resolve whether a given patient will hemorrhage or not, given any of a multitude of clinical scenarios. The PTT is thus tasked to do things for which it was never designed (Table 1). Historical background A brief historical review will buttress this hypothesis [1]. Tissue factor (TF) is generated and released by damage to tissue either constitutively elaborating TF (nearly all tissue) or cells (endothelium, monocytes, and macrophages) that can rapidly produce TF in response to inammatory cytokines. The TFdriven coagulation system in controlled in vitro laboratory testing is termed the extrinsic system, as it requires the addition of material (TF) extrinsic to collected blood to be initiated. This coagulation test, the prothrombin time (PT), likely mimics the initiation of in vivo coagulation. In the laboratory, TF and phospholipids (together a complete thromboplastin) are added in a controlled amount. This thromboplastin was prepared from rabbit or human brain until recently. Currently, this thromboplastin is synthetically produced using recombinant technology. The sensitivity of recombinant thromboplastin is now highly reproducible so that the international normalized ratio (INR) concept of testing is highly reliable. The intrinsic system requires nothing added to initiate the coagulation cascade. Factor XII (FXII), a zymogen, is activated not by a proenzyme-turned-enzyme, but by being molecularly spread out on physical contact with negatively charged (wettable) surfaces in such a way that the new tertiary rearrangement of its molecular structure results in exposure of its active enzymatic site [now termed activated factor XII (FXIIa)] which, in concert with high molecular weight kininogen (HMWK) and prekallikrein (PK), enzymatically activates FXI to activated FXI (FXIa). Enzymatic reactions, typically accelerated by phospholipid membrane complexes, complete the coagulation cascade, resulting in thrombin generation and a brin clot. Physiologic activators of FXII are subendothelial materials (such as collagen), as well as microbial negatively charged endotoxins in pathological situations. For several reasons, the PTT may not accurately reect in vivo hemostasis. (Table 2).

2608 C. S. Kitchens
Table 1 Reasons PTTs are ordered (in descending logical sequence yet ascending frequency of ordering) Screening for congenital abnormalities of the coagulation system Screening for acquired abnormalities of the coagulation system Monitoring factor replacement in congenital hemophilia Monitoring anticoagulant therapy Search for a lupus anticoagulant Part of a DIC battery Liver function test Preoperative evaluation Routine hospital admission order Ritual to ward away potential hemorrhage

Table 2 Steps in testing that leads one to suspect that the PTT may not accurately reect physiologic hemostasis Blood is withdrawn from its normal environment Calcium is extracted by an anticoagulant Cellular elements are removed Plasma may be frozen and stored Frozen plasma is then thawed Plasma is placed in a glass test tube Particulate substance, such as dirt or glass particles, is added Non-human phospholipid and calcium are added to initiate clotting Results are compared with those from static, congenital hemorrhagic defects Risk stratication is estimated by comparing with expected values

conrmation in patients who had hemorrhaged and were suspected of having hemophilia, what we now would call a high pretest probability. The WBCT served that role well, being able to conrm the diagnosis of what is now regarded as classic severe hemophilia (FVIII or IX deciency respectively, two diseases which were not to be distinguished from each other until the mid-twentieth century). The rst expansion or even adulteration for the use of this test was for monitoring heparin therapy, which was clinically initiated in the 1930s by several groups. As imperfect as it was, the WBCT was thus used for several decades. The rst improvement in the WBCT was made by the Chapel Hill group, led by Brinkhous [9,10]. With their renement, platelet-free plasma from decalcied (anticoagulated) whole blood could now be kept, stored, and studied as needed. These features plus improved standardization made the WBCT diagnostically obsolete. The new test was initiated by recalcication and addition of a partial thromboplastin prepared from the complete thromboplastin by ultracentrifugation which, we now know, spun down the phospholipidbound TF. This test became known as the partial thromboplastin time (PTT). The raison detre of the PTT was to discern moderate or even mild cases of hemophilia A and B as well as several newly discovered abnormalities in the intrinsic system from normal [10]. With its portability, now plasma could be stored, frozen, shipped and studied at reference laboratories. Expanding beyond hemophilia The concept of surface activation of FXII was expanded by observations made by Oscar Ratnoff in studying the rst patient subsequently determined to have FXII deciency. Appreciating that surface activation of FXII was enhanced by materials having a negative charge, Proctor and Rappaport [11] accelerated the PTT by adding materials such as particulate glass or kaolin (a dirt-like product). This test was called the PTTK (K for kaolin) as well as the aPTT (a for activated). Whereas the older PTT normal control value was approximately 100 s, with the aPTT, the normal control was truncated to approximately 35 s. Not only did the shortening of the normal range speed up bulk laboratory testing, but also by accelerating activation, any mild deciency was brought more into focus as the rate-limiting step, and therefore more subtle perturbations of the intrinsic system could be discerned. The term PTT will be used for the rest of this manuscript, as it is now in common usage, for the activated PTT. Other reasons for prolongation of the PTT that were associated with clinical hemorrhage were being discovered and investigated: hemorrhagic disease of the newborn, liver failure (especially if it was due to obstructive jaundice), von Willebrand disease, various circulating coagulation inhibitors, and disseminated intravascular coagulation (DIC). The perceived rm link between a prolongation of the PTT and clinical hemorrhage was at its peak for no other reason than the patients for which this test was ordered experienced hemorrhage and/or were receiving therapeutic heparin. The exception
2005 International Society on Thrombosis and Haemostasis

Ages ago, therapeutic phlebotomists noted that freshly let blood clotted seemingly on its own. Rudimentary theories were offered to support such observations: cooling of blood caused gelling; contact with air initiated clotting; or shed blood was no longer moving and therefore lost some vital inuence with the subsequent destiny to coagulate. Negating these observations rst was Hewson [2], who in the 16th century showed that cooling blood actually slows coagulation. Hunter [3] showed that clotting occurred just as fast in a vacuum as when in contact with air, and it was Hewson again who demonstrated that blood trapped within carefully ligated veins required many hours to clot, thus eroding the notion that motion itself served as a natural anticoagulant [2]. Hunter [3] was the rst to record that coagulation originated around the edges of the (wettable) metallic or earthen vessel into which let blood was collected. Lister [4] noted that blood clotted much slower in rubber than either metal or earthenware basins, while others noted that coating either of the latter two basins with oil greatly retarded the coagulation of blood. None other than Virchow [5] noted that a drop of (wettable) mercury injected into a vein caused nearly immediate thrombosis. New observations required new theories. Although the existence of hemophilia was known for centuries [6] and that sufferers of that disease hemorrhaged uncontrollably, surprisingly it was not until 1893 that Wright [7] noted that let blood of hemophiliacs clotted more slowly than normal blood. This single but simple observation led to rudimentary testing for hemophilia, which in 1913 culminated in the more-or-less standardized whole blood clotting time (WBCT) of Lee and White [8]. Thus, this test was initiated as a

PTT 2609

of course was Ratnoffs patient, John Hageman, who, despite an enormously prolonged PTT, experienced no hemorrhage. Ratnoffs observations relied on the PTT not for diagnostic reasons but to begin unraveling the physiology of clotting. Other cracks began to appear in the convenient PTT: hemorrhage link. Newly discovered deciencies of HMWK and PK had among the most prolonged of PTTs but these patients also exhibited no hemorrhage. In fact, all were found serendipitously by studying their blood for other reasons when their prolongation of the PTT was discovered. Some patients with exceedingly prolonged PTTs due to homozygous deciency of FXI do not experience hemorrhage while other patients with heterozygous deciency, having borderline or even high normal PTTs, do. DIC was recognized increasingly as a thrombotic as well as a hemorrhagic disorder, while patients with either von Willebrand disease or hepatic failure, there was little or no correlation between the degree of prolongation of PTT and risk of hemorrhage. The newly discovered lupus anticoagulant was characterized by prolonged PTT [12] but how iconoclastic! such patients not only did not hemorrhage, but also experienced thrombosis! [13]. Numerous prelaboratory and laboratory phenomena were added to the growing list further separating the link between the tell-all PTT and the foretelling of hemorrhagic risk. If a laboratory sample clotted too early, i.e., before the PTT timer was engaged, the resulting PTT appeared innite [14]. Heparin contamination from indwelling arterial or venous lines continued to haunt both the coagulation laboratories and the wards, especially the intensive care units [15]. Samples from patients
Table 3 The poor correlation of PTT results and hemorrhagic potential PTT Results Hemorrhage No Normal Normal patients

with hematocrit levels > 60% contained too little plasma to appropriately dilute the liquid citrate anticoagulant, which resulted in PTTs persistently prolonged until the defect was corrected in vitro by using the correct volume of plasma [16]. Samples poorly and slowly drawn may be partially activated with subsequent testing yielding prolonged PTTs [17]. Some plasma substances can interfere with and blur the sol-to-gel transformation of the clotting test so that the instrument cannot precisely detect the point of clot formation. These include free hemoglobin; high levels of lipids, immunoglobulins, or bilirubin [18]; and complexes of C-reactive protein and very low density lipoprotein (VLDL) [19]. On the other hand, patients may experience hemorrhaging from plasma defects (i.e., not counting patients with either platelet or vascular disorders) while effectively displaying normal PTTs. Such conditions included surgical (anatomical) bleeding, many patients heterozygous for FXI deciency [20], those with FXIII deciency, those with FVII deciency, and nally those patients with hyperbrinolysis such as may result from patients with acute promyelocytic leukemia (APL), victims of either western diamondback rattlesnake or eastern diamondback rattlesnake envenomation (each by different mechanisms) [21], or patients who are decient in alpha-2 plasmin inhibitor (a2PI) [22]. Patients administered low molecular weight heparin (LMWH) may experience considerable hemorrhage from that anticoagulant all the while having a normal PTT [23]. A newly recognized and increasingly encountered clinical event is categorized by prolonged PTs and PTTs. Such patients

Prolonged Some with FXI deciency FXII deciency HMWK deciency PK deciency LA Some with thrombin and FV alloantibodies from bovine thrombin Rattlesnake envenomation Technical abnormalities Erythrocytosis Heparin contamination Premature in vitro clot Partially activated sample Interfering substances FVIII deciency FIX deciency Some with FXI deciency Some with thrombin and FV alloantibodies from bovine thrombin Fibrinogen, FII, V, or X deciency DIC Hepatic insuciency Vitamin K deciency Therapeutic heparin or warfarin usage

Yes

FVII deciency FXIII deciency Most with vWD Some with FXI deciency Therapeutic LMWH administration Hyperbrinolysis Malignancies (APL, prostate) a2PI deciency

LA, lupus anticoagulant; HMWK, high molecular weight kininogen; PK, prekallikrein; DIC, disseminated intravascular coagulation; a2PI, alpha2-plasmin inhibitor; vWD, von Willebrand disease; APL, acute promyelocytic leukemia; LMWH, low molecular weight heparin. 2005 International Society on Thrombosis and Haemostasis

2610 C. S. Kitchens

may experience either no or serious hemorrhage. This eventuates from acquired alloantibodies against bovine thrombin and V from exposure to bovine brin glue frequently used as a sealant sprayed onto large, potentially hemorrhagic lesions at the termination of neurologic, vascular, or thoracic surgical cases [24]. Those alloantibodies may crossreact with their human counterparts thus lowering FV levels and inhibiting gene rated thrombin [24]. Role in preoperative testing Somewhere along the way, the PTT (and PT) became routinely ordered by surgeons and others in preparation for planned invasive procedures. Despite rampant adamant opinion, no study has ever determined, using evidence-based medicine (EBM) techniques at which PTT (or PT) an invasive procedure is unsafe or even may become unsafe. Numerous papers have shown that invasive procedures are safe in patients with moderately prolonged coagulation tests [2532]. Multiple studies have demonstrated that routine preoperative coagulation tests not only are irrational and wasteful [33,34] but also a prolongation of the very test used for screening is more often than not ignored or not acted upon. On questioning these practices, those continuing to order preoperative testing despite such information offer as reason for not changing their behavior is that they feel better with a normal coagulation battery on the chart [35,36]. Several panels of experts state that situations for routine preoperative coagulation testing that seem justied include patients with leukemia, hepatic disease, history or physical evidence of a bleeding disorder, those who are or were taking anticoagulant drugs, and patients who are undergoing particularly challenging hemostatic surgical stress such as cardiopulmonary bypass [37,38]. The most efcacious preoperative test to estimate hemorrhagic potential with a procedure remains the presence or absence of hemorrhage experienced with prior surgery or trauma [39]. In studying a cohort of 100 patients having prolonged PTTs, the cause of which was not previously known upon referral to this laboratory, we elucidated almost all of the reasons so far discussed [40]. Equally of interest, it was demonstrated that not the degree but the reason for the prolongation of the PTT best estimated risk for hemorrhage. A patient having a preoperative PTT of 60 s due to a lupus anticoagulant was at no risk for surgical hemorrhage, whereas another patient with a PTT of 40 s, due to a heretofore undiagnosed mild to moderate case of hemophilia, may experience major or even fatal hemorrhage (Table 3). Summary The PTT may be prolonged for any of multiple reasons. It cannot serve and was not intended to serve as an hemostatic oracle. Using pretest probabilities it serves very well as a screen for plasma-based hemostatic defects in patients who have experienced hemorrhage. It serves fairly well as a surrogate

marker for monitoring therapeutic heparin administration. Those ordering this test should be aware of its capabilities and limitations. References
1 Ratno OD. Why do people bleed?, chapter 18. In: Wintrobe MM, ed. Blood, Pure and Eloquent. New York: McGraw-Hill, 1980: 60057. 2 Hewson W. In: Gulliver G, ed. The Works of William Hewson, F.R.S. London: The Sydenham Society, 1846: 360. 3 Hunter J. A Treatise on the Blood, Inammation, and Gun-Shot Wounds. Philadelphia: Webster, 1817: 514. 4 Lister J. On the coagulation of blood. Proc R Soc Lond 1863; 12: 580 611. 5 Virchow R. Ueber die akute Entzu dung der Arterien. Arch Pathol Anat 1847; 1: 272378. 6 Rosner F. Hemophilia in the Talmud and Rabbinic writings. Ann Intern Med 1969; 70: 8337. 7 Wright AE. On the method of determining the condition of blood coagulability for clinical and experimental purposes, and on the eect of the administration of calcium salts in haemophilia and actual or threatened hemorrhage. Br Med J 1893; 2: 2235. 8 Lee RI, White PD. A clinical study of the coagulation time of blood. Am J Med Sci 1913; 145: 495503. 9 White GC. The partial thromboplastin time: dening an era in coagulation. J Thromb Hemost 2003; 1: 226770. 10 Langdell RD, Wagner RH, Brinkhous KM. Eect of anti-hemophilic factor on one-stage clotting tests: a presumptive test for hemophilia and a simple one-stage anti-hemophilic factor assay procedure. J Lab Clin Med 1953; 41: 63747. 11 Proctor RR, Rapaport SI. The partial thromboplastin time with kaolin. A simple screening test for rst stage plasma clotting factor deciencies. Am J Clin Path 1961; 36: 2129. 12 Conley CL, Rathbun HK, Morse II WI, Robinson Jr JE. Circulating anticoagulant as a cause of hemorrhagic diathesis in man. Bull Johns Hopkins Hospital 1948; 83: 28896. 13 Bowie EJW, Thompson Jr JH, Pascuzzi CA, Owen Jr CA. Thrombosis in systemic lupus erythematosus despite circulating anticoagulants. J Lab Clin Med 1963; 62: 41630. 14 Smith LG, Kitchens CS: Pseudo-prolongation of the partial thromboplastin time. Am J Clin Pathol 1983; 80: 7502. 15 Czapek EE. Iatrogenic prolonged APTT: a nondisease state. JAMA 1974; 227: 1304. 16 Koepke JA, Rodgers JL, Ollivier MJ. Pre-instrumental variables in coagulation testing. Am J Clin Pathol 1975; 64: 5916. 17 McPhedron P, Clyne LP, Ortoi NA, Gagnon PG, Sanders JF. Prolongation of the activated partial thromboplastin time associated with poor venipuncture technic. Am J Clin Pathol 1974; 62: 1620. 18 Johns CS. Coagulation instrumentation, Chapter 48. In: Rodak BF, ed. Hematology, Principles and Applications, 2nd edn. Philadelphia: WB Saunders, 2002: 75568. 19 Toh CH, Samis J, Downey D, Walker J, Becker L, Brufatto N, Tesidor L, Jones G, Houdijk W, Giles A, Koschinsky M, Ticknor LO, Paton R, Wenstone R, Nesheim M. Biphasic transmittance waveform in the APTT coagulation assay is due to the formation of a Ca++-dependent complex of C-reactive protein with very-low-density lipoprotein and is a novel marker of impending disseminated intravascular coagulation. Blood 2002; 100: 25229. 20 Kitchens CS. Factor XI: a review of its biochemistry and deciency. Semin Thromb Hemost 1991; 17: 5572. 21 Kitchens CS. Hemostatic aspects of envenomation by North American snakes. Hematol/Oncol Clin North Am 1992; 6: 118995. 22 Aoki N. Alpha-2-plasmin inhibitor. In: High KA, Robert HR, eds. Molecular Basis of Thrombosis and Hemostasis. New York: Marcel Decker, Inc., 1995: 54559.

2005 International Society on Thrombosis and Haemostasis

PTT 2611
23 Kitchens CS. Evaluation and treatment of bleeding associated with heparin and low-molecular-weight heparin administration. In: Alving BM, ed. Blood Components and Pharmacologic Agents in the Treatment of Congenital and Acquired Bleeding Disorders. Bethesda: AABB Press, 2000: 16784. 24 Zumberg MS, Waples JM, Kao JK, Lottenberg R. Management of a patient with a mechanical aortic valve and antibodies to both thrombin and factor V after repeat exposure to brin sealant. Am J Hematol 2000; 64: 5963. 25 McGill DB, Rakela J, Zinsmeister AR, Ott BJ. A 21-year experience with major hemorrhage after percutaneous liver biopsy. Gastroenterology 1990; 99: 1396400. 26 McVay PA, Toy PTCY. Lack of increased bleeding after liver biopsy in patients with mild hemostatic abnormalities. Am J Clin Pathol 1990; 94: 74753. 27 Ewe K. Bleeding after liver biopsy does not correlate with indices of peripheral coagulation. Dig Dis Sci 1981; 26: 38893. 28 McVay PA, Toy PTCY. Lack of increased bleeding after paracentesis and thoracentesis in patients with mild coagulation abnormalities. Transfusion 1001; 31: 16471. 29 Kozak EA, Brath LK. Do screening coagulation tests predict bleeding in patients undergoing beroptic bronchoscopy with biopsy? Chest 1994; 106: 7035. 30 Bjortuft O, Brosstad F, Boe J. Bronchoscopy with transbronchial biopsies: measurement of bleeding volume and evaluation of the predictive value of coagulation tests. Eur Respir J 1998; 12: 10257. 31 DeLoughery TG, Liebler JM, Simonds V, Goodnight SH. Invasive line placement in critically ill patients; do hemostatic tests matter? Transfusion 1996; 36: 82731. 32 Doerer ME, Kaufman B, Goldenberg AS. Central venous catheter placement in patients with disorders of coagulation. Chest 1996; 110: 1858. 33 Roizen MF. More preoperative assessment by physicians and less by laboratory tests. N Engl J Med 2000; 342: 2045. 34 Kitchens CS. Preoperative PTs, PTTs, cost-eectiveness, and health care reform. Radical changes that make good sense. Chest 1994; 106: 6612. 35 Schein OD, Katz J, Bass EB, Tielsch JM, Lubomski LH, Feldman MA, Petty BG, Steinberg EP. The value of routine preoperative medical testing before cataract surgery. N Engl J Med 2000; 342: 168 75. 36 Golub R, Cantu R, Sorrento JJ, Stein HD. Ecacy of preadmission testing in ambulatory surgical patients. Am J Surg 1992; 163: 565 71. 37 Kaplan EB, Sheiner LB, Boeckmann AJ, Roizen MF, Beal SL, Cohen SN, Nicoll CD. The usefulness of preoperative laboratory screening. JAMA 1985; 253: 357681. 38 Litaker D. Preoperative screening. Med Clin North Am 1999; 83: 1565 81. 39 Narr BJ, Warner ME, Schroeder DR, Warner MA. Outcomes of patients with no laboratory assessment before anesthesia and a surgical procedure. Mayo Clin Proc 1997; 72: 5059. 40 Kitchens CS. Prolonged activated partial thromboplastin time of unknown etiology. A prospective study of 100 consecutive cases referred for consultation. Am J Hematol 1988; 27: 3845.

2005 International Society on Thrombosis and Haemostasis