Low‐dose heparin to low molecular weight heparin prophylaxis: in pursuit of excellence – a personal perspective
2005; Elsevier BV; Volume: 3; Issue: 2 Linguagem: Inglês
10.1111/j.1538-7836.2004.01031.x
ISSN1538-7933
Autores Tópico(s)Cardiac tumors and thrombi
Resumo'Nothing in the world can take the place of persistence. Talent will not; nothing is more common than unsuccessful men with talent. Genius will not; unrewarded genius is almost a proverb. Education will not; the world is full of educated derelicts. Persistence and determination alone are omnipotent. The slogan 'press on' has solved and always will solve the problems of the human race.' Calvin Coolidge Scientific history often bears testimony that seminal achievements frequently arise from serendipity. However, development of low-dose heparin and low molecular weight heparin (LMWH) prophylaxis represents long lasting, tenacious pursuit and hope under most challenging circumstances over the last 35 years. This review provides an account of personal prospective and recollection; recollections of the clinical issues of the time, and of the individuals who influenced me, as well as the difficulties I had to face in bringing my ideas to completion. Having graduated from the Ghandi Medical College in Bhopal (India) in 1960, and completed 1 year of compulsory internship, I traveled to the United Kingdom in 1961 with an aim of completing my postgraduate surgical training. I was successful in obtaining Fellowships from both the Royal College of Surgeons of England (FRCS) and Edinburgh (FRCSE) in 1964. Venous thromboembolism had for many years been the Cinderella of clinical medicine and surgery. My interest in this disease process has been largely due to its importance in relation to the operations and surgical conditions, and being frequently responsible for disability and death. It began when I was appointed as Pfizer Research Fellow at King's College Hospital and Medical School in 1965. At that time, Professor J.G. Murray, Head of the Department of Surgery, lost two patients in quick succession following surgery for large bowel and the post mortem confirmed the presence of fatal pulmonary embolism. When I first reviewed the autopsy data at King's, it became obvious that one of the commonest causes of death in surgical patients after operation was pulmonary embolism [1]. A review of the literature at this stage also highlighted four main issues which needed to be addressed. Can the diagnosis of deep vein thrombosis (DVT), particularly in early postoperative phase, be made earlier and more accurately? Can the occurrence of DVT be prevented by a simple form of prophylaxis which could be used in all high risk patients? If a patient did develop DVT, which form of treatment would give the best immediate results? Finally, if immediate treatment were effective, would it help prevent the late sequalae of DVT, the post-phlebitic syndrome? It also became apparent that a major effort would be required for any impact in reducing morbidity and mortality due to postoperative venous thromboembolism. I had the unique experience of learning the basic assays for measuring activation of blood clotting under the guidance of P. T. Flute, Reader in the Department of Haematology at that time at King's and particularly his ever patient senior technician, Miss Gillian Pannell, who gave invaluable technical advice. Under the guidance of J. W. Laws, Head of the Radiology Department, I developed a technique of ascending biplane and cine functional phlebography [2], and established the 125I-labeled fibrinogen test under the guidance of Dr S. Osborne, Head of the Department of Nuclear Medicine (see below). Thus I was able to create a unique research facility, the Thrombosis Research Unit, funded by the King's College Hospital Research Trust for investigating and treating patients suffering from venous thromboembolic disease. In 1967, I was appointed to the post of Lecturer in Surgery and continued my clinical research activities with great enthusiasm which laid foundations not only for accurate and sensitive techniques for diagnosis of asymptomatic DVT but also opened up a new era for developing an effective method for prophylaxis of postoperative venous thromboembolism, and defining the most effective treatment for established DVT. These contributions were soon recognized by my surgical colleagues who bestowed a unique honor on me by awarding a Hunterian Professorship and formed the basis of a prestigious lecture delivered at the Royal College of Surgeons of England on the 20th March, 1969 (Fig. 1). On the way to deliver a Hunterian Professor lecture at the Royal College of Surgeons of England on the 20th March 1969. Leading the procession is Sir Hadley Atkins, President of the College. In the past, one of the major difficulties in the management of DVT had been the lack of means for recognizing the condition accurately at an early stage. In almost one-half of the patients there were no clinical signs or symptoms referable to the limbs, and fatal pulmonary embolism might be the first indication of thrombosis. In the absence of sensitive and accurate techniques for screening large numbers of patients, it had not been possible to study the incidence of venous thrombosis in hospitalized patients, to define the 'high risk' group who might be considered to be at a greater risk of developing this condition. It had not been possible to study the natural history of the thrombotic process, the types of thrombi that are safe and free of complications and those that might give rise to pulmonary emboli, and, most importantly, no means were available for measuring with any precision the effects of prophylaxis or treatment of this condition. One of the possible methods which could be developed to detect venous thrombosis was to use fibrinogen labeled with a radioactive isotope. The concept of using isotopes in the detection of venous thrombosis was first introduced by Ambrus and others in 1957; they produced radioactive thrombi in experimental animals by injecting 131I-labeled fibrinogen followed by thrombin into artificially occluded vessel segments [3]. Previously McFarlane had demonstrated that iodinated proteins are degraded at the same rate as the corresponding unlabeled protein [4]. This led to the notion that fibrinogen labeled with radioactive iodine should behave in the same manner as endogenous fibrinogen with normal conversion into fibrin under the action of thrombin. This hypothesis was tested in experimental animals [5] and in a limited number of patients who presented with clinical features suggestive of DVT [6]. In 1965, Atkins and Hawkins working in the Department of Surgery at King's introduced isotope 125I as an alternative to 131I because of its many advantages for use in clinical practice [7]. Our earlier studies [8] and those performed by other investigators [9] published at the same time used venography to confirm the findings of Fibrinogen Uptake Test for detecting clinically silent thrombosis. However, the equipment used for monitoring the radioactivity in the limbs was heavy, cumbersome and very expensive, and required lengthy complicated calculations to analyze the results. My first task was to simplify the test procedure and the equipment used so that it could be adopted to screen a large number of patients and be performed at the patients' bedside [10]. I was joined by my two colleagues, Charles Flanc – a surgeon and visiting Research Fellow from Australia and Martin B. Clark, a physicist from the Department of Nuclear Medicine at King's. Our task was to define the most suitable technique for labeling fibrinogen, as four different techniques had been advocated including the iodine monochloride, the chloramine –T, electrolytic and the enzymatic (lactoperoxidase) method. Physicochemical properties and kinetics of clearance from the circulation of canine-labeled fibrinogen had indicated that the iodine monochloride method produced a suitable labeled fibrinogen without significantly altering the molecule of fibrinogen, and in experimental model of venous thrombosis it resulted in the greatest fibrinogen uptake and the highest thrombus-to-blood ratio. Furthermore, with increasing degrees of iodination with 125I, the radioactivity of the α-chain increased proportionately more than that of the β and γ chains, reaching a ratio of 2.0 for the α-chain, 2.1 for the β-chain, and 1.1 for the γ-chain, under the condition of iodination of 16 atoms of iodine/molecule of fibrinogen. We also investigated the effect of iodination on the ability of fibrinogen to be converted into cross-linked fibrin. An increasing degree of iodination had no effect on cross-linking. We then performed a study in postoperative patients and those with a suspected DVT and used venography to establish the value of Fibrinogen Uptake Test (FUT) in detecting DVT. My next task was to standardize the procedure for measuring radioactivity over the limbs [10]. In the initial studies the equipment used consisted of a detector system unit composed of a sodium iodide crystal (the thallium activated) optically coupled to a photomultiplier, a processing and display unit. We introduced the use of simple portable equipment which consisted of a hand held collimator and a battery-operated ratemeter which could be used at patients' bedside and adopted for investigating a large number of patients (Fig. 2). In the old method, the radioactivity at the marked positions in the thigh and calves was recorded as three 10-s counts (Fig. 3). The absolute counts were represented as a percentage of a standard radioactive source or heat counts. This involved lengthy and complex calculations. The ratemeter had the advantage that the reading obtained from the linear scale of the instrument was automatically represented as a percentage of heat counts, no calculation was necessary. Our next aim was to define the criteria that could be used for diagnosing DVT, based on a comparison of the findings of FUT and venography (Fig. 4). The criterion for diagnosing a thrombus was defined as an increase of 20 in the percentage reading on any day, compared with the reading at the same position on the previous day, provided the increase persisted for more than 24 h. Most patients considered to have a thrombus showed an unequivocal scan with a rise in relative activity of 40–120% or more. The accuracy of the test in detecting DVT was confirmed in subsequent trials; a remarkably close correlation between the [125I]fibrinogen uptake test and venography was observed with a 90% agreement in the limb series and 94% in the patient series. An important limitation of the test was its inability to detect thrombi in the pelvic veins and to a lesser extent, in the upper third of thigh veins since the proximity of the bladder containing radioactive urine or large arteries and other vascular structures gave an increased background count, which made the test less reliable in these situations. The studies which were published by other investigators confirmed our initial observation that FUT was an accurate method for detecting thrombi in the deep veins. However, a number of clinicians questioned the clinical significance of such thrombi and whether their treatment also reduced the mortality due to pulmonary embolism. This issue was resolved by analysis of the incidence of fatal pulmonary embolism, proved by autopsy, in groups of surgical patients undergoing major elective operations, where early detection and treatment of isotopic thrombi significantly reduced its frequency [11]. Ratemeter (Pitman Model 235 Isotope Localization Monitor) and scintillation counter. Machine is operated by two small batteries. (Reproduced from Kakkar et al. by the courtesy of the Editor of the Lancet). Positions at which radioactivity is measured. These are marked at intervals of 2 inches along the course of the femoral vein and on the posterior aspect of the calf. The criteria for diagnosing DVT: the count difference in the group with positive and negative venographic results. The point at which the two curves intersect (18%) represents the count difference at which incorrectly positive and incorrectly negative fibrinogen tests are minimal. Using FUT as a routine screening procedure, a number of our prospective studies showed that the disease occurs quite frequently in patients undergoing major abdominal, thoracic, orthopedic and other surgical procedures as well as chronically ill medical patients (Table 1). Few physicians would deny that venous thrombosis, difficult as it may be to diagnose, was also a common and benign disease. The extent to which intravascular thrombi might dissolve spontaneously was not known. Furthermore, information in the past had not been available regarding the type of thrombi that are safe and free of complication and those that might give rise to pulmonary emboli or were likely to damage valve and lead to the postphlebitic syndrome. This information about the natural history of DVT was important, as the treatment of venous thromboembolism available in the late 1960s was neither simple nor safe and hence could not be justified in every patient. Our study showed that in surgical patients, the majority of thrombi form in the calf veins; a surprisingly high proportion of these remained confined to this region and undergo spontaneous lysis and in only a small number do thrombi extend more proximally into the major veins. In the group of patients with extending thrombi in the popliteal and femoral veins, minor pulmonary embolism occurred in approximately 50%, only a small proportion proving fatal [12]. We also used the FUT for defining the 'high risk' group of patients who may be at greater risk of developing post operative DVT. The effect of various predisposing factors on the incidence of postoperative DVT was assessed in 200 patients undergoing major abdominal surgery [13]. The results confirmed that advancing age, obesity, previous venous thromboembolism, varicose vein and malignancy increased the likelihood of developing venous thromboembolism [13]. The information generated by the FUT regarding the high incidence of postoperative DVT, its natural history and the definition of the 'high risk' group laid the foundation for developing an effective method for prevention. It was apparent that in order to significantly reduce postoperative venous thromboembolism, a method would be required which would be applicable to all patients, easily adopted on a large scale and free of any complications. Though the search for such an effective method of prophylaxis against venous thromboembolism had been going on for nearly 90 years, a method which was effective in the total elimination of this condition had yet to be developed. This uninspiring state of affairs had been due to two main reasons. First, essential knowledge of the 'trigger' mechanism that initiates intravascular coagulation has been lacking. Second, the absence of a sensitive and accurate technique for assessment of the effects of prophylaxis had certainly been a major hurdle. To a large extent, the last difficulty had been overcome, as stated above, by using the FUT. Using this test it was possible to determine, with greater accuracy than ever before, the true incidence of DVT, and the effectiveness of a specific regimen for the prevention of such thrombi. In the mid-1960s the main attempts to prevent postoperative venous thromboembolism were either directed towards the elimination of stasis in the deep vein or employed to counteract changes in blood coagulability. It was argued by some clinicians that conventional physiotherapy reduced venous thromboembolism, while others denied it. Unfortunately, these conclusions were based on physical signs alone, which were quite inadequate to diagnose the existence of venous thrombosis. We therefore re-assessed the effectiveness of intensive physiotherapy combined with elastic stockings in preventing the condition. Despite all the care and effort that was taken to protect the patient's legs and prevent stasis of blood in the deep veins during and after surgery, at least 25% of the patients, whether young or old, developed venous thrombosis [14]. This incidence was not significantly different from that in patients in whom no particular measures were taken to prevent this occurrence. We turned our attention to the available pharmacological method. Several studies had shown that the oral anticoagulant therapy, when properly employed (i.e. started well before surgery), was the most effective and proven method of preventing major venous thromboembolism. However, a major drawback of this form of therapy was the risk of serious bleeding both during and after surgery despite strict laboratory control of the dosage. The incidence of severe hemorrhage in the various reported studies had varied between 2.00 and 6.97% and mortality in the range of 0.8–1.0%[15]. This had undoubtedly contributed to the relatively low level of acceptance of this form of prophylaxis among surgeons in general, at least in the United States and the United Kingdom. A form of therapy that would both be effective and without the drawback of oral anticoagulant therapy would therefore meet a real clinical need. One such promising approach appeared to us to be the use of low-dose heparin given subcutaneously if it could be shown to prevent thrombosis without increasing the risk of bleeding. The rationale for the use of low-dose heparin prophylaxis was first put forward by De Takats, who had shown that small amounts of heparin would effectively block the coagulation process at an early stage [16]. Subsequently, Bauer in 1954 [17] and Leggenhager [18] recommended heparin prophylaxis for all high risk patients undergoing major surgery. Furthermore, Sharnoff et al. [19, 20] suggested that low-dose heparin prophylaxis administered before, during and after surgery may be effective in preventing postoperative thromboembolism. The regimen they recommended consisted of administration of 10 000 IU of heparin subcutaneously at midnight prior to surgery. In their view, this dose was effective in neutralizing the increased activity of pulmonary megankerocytes and thus providing an anticoagulant effect lasting approximately 12 h. If surgery extended beyond this point, additional heparin was administered subcutaneously during operation. At the completion of the operation, the coagulation time was repeated and usually 2500 IU of heparin were administered subcutaneously until the patient was fully active or discharged. The findings of an uncontrolled study published in 1970, suggested that this regimen was effective in preventing fatal pulmonary embolism. However, this study had several drawbacks: it was not a randomly allocated, controlled trial and the results were not suitable for statistical analysis [20]. Furthermore, with the regimen recommended by Sharnoff, the amount of heparin to be administered every 6 h had to be tailored by measuring the whole blood clotting time. It was obvious to us that such a regimen would be impractical for protecting large numbers of patients. We argued that for a prophylaxis to be adopted on a wide scale it must be well tolerated, devoid of side-effects, require no special monitoring and produce no bleeding in a clinical situation in which the patient was subjected to major trauma. This last requirement meant that any tendency to excessive intravascular coagulation was to be prevented without interfering with normal hemostasis. The first hurdle I encountered in developing a fixed regimen of low-dose heparin prophylaxis was to enlist the support of a pharmaceutical company to support the research by providing free supplies of heparin. Therefore, in 1969, I wrote to six companies manufacturing heparin at that time. Three companies replied that they could not see heparin being used for prophylaxis, two did not even bother to reply, but I was delighted to receive a reply from a French company called CHOAY Pharmaceuticals Ltd (Paris) indicating that they would like to meet me and discuss the details of the proposed study. I immediately made arrangements to travel to Paris and was met by two extremely polite and kind gentlemen – André Kher and Francis Toulemande – and to my delight they spoke perfect English. But my enthusiasm was soon to encounter strong opposition from Claude Raby who had joined CHOAY in 1962 and referred to himself as 'heparin paranoid'. His view was that heparin should only be used for treating patients with established DVT and that high doses should be personalized based on the findings of thromboelstography. However, Jean and Henri Choay and Edmond Vaiavel, senior executives in the CHOAY Pharmaceutical, listened very carefully to the hypothesis I had suggested for developing a regimen of fixed low-dose heparin for prophylaxis. I explained to them that during my recent visit to the United States I had visited S. Wessler and E. T. Yin of the Department of Medicine, Jewish Hospital of St Louis, St Louis, Missouri, and had discussed the rationale for using low-dose heparin. The conceptual basis for a fixed dose heparin prophylaxis was based on their recently published paper where they had identified a potent naturally occurring inhibitor to activated factor X (FXa) in human plasma and serum and the unexpected finding of a striking increase in the activated FXa-inhibitor activity by a small amount of heparin [21]. They were very keen to actively participate in a clinical trial that I had organized to assess the efficacy of low-dose heparin prophylaxis. Yin immediately visited King's to help us to set up an assay for measuring FXa inhibitor activity. After a lengthy discussion, Jean Choay agreed to support a limited study in surgical patients. We first assessed the effect of subcutaneous administration of different doses of heparin. When 5000 units of heparin were injected subcutaneously into a normal subject and clotting assays were carried out at intervals over a 12-h period, the heparin level rose within 30 to 60 min of the injection, reached a peak within 1–2 h, and returned to baseline within 10–12 h (Fig. 5). The partial thromboplastin time was within the normal throughout the 12-h period or slightly increased only at the time of peak heparin concentration. The prothrombin and thrombin times invariably remained at the preheparin level throughout the 12-h period. Earlier onset (i.e. 15 min), higher peak, and more prolonged heparin levels were obtained with subcutaneous injections of 7500 and 10 000 units of calcium heparin. At any given heparin dose, however, there were striking variations among individuals in the onset, peak, and duration of the heparin levels measurable in the plasma. Because of these findings and because heparin-tolerance tests were not to be done routinely, we decided to use a fixed dose and time schedule of 5000 units every 12 h. Typical in vivo plasma-heparin levels after administration of 5000 units of heparin subcutaneously to healthy volunteers and patients having major elective operations. Reprinted with permission from Elsevier (The Lancet 1972; 300: 101–06). The effectiveness of a fixed low-dose heparin prophylaxis was investigated in 53 consecutive patients over the age of 50 undergoing repair of inguinal hernia. Heparin (5000 IU) was given 2 h before and every 12 h after surgery for the next five postoperative days [22]. Heparin was administered in exactly the same way as described by Griffith and Boggs [23]. DVT was detected by means of the FUT in seven (26%) of the control patients, while this was significantly reduced (4%) in the 26 similar patients who received heparin before and after surgery [22]. Furthermore, there was no significant difference in the thrombin-clotting time between the two groups. Thus the antithrombotic doses of heparin used neither prolonged the thrombin-clotting time nor produced significant clinical bleeding in the operative wound or elsewhere. The administration of heparin was initiated prior to surgery; this might have been the key to prophylactic success. This study had certain defects. The number of patients was relatively small; nevertheless, they were well matched for age and other factors likely to affect the incidence of thrombosis. We chose this group of patients as it was felt that any major bleeding during surgery could be easily controlled, and postoperative bleeding in the form of major wound hematoma would be easily detected and may not have major untoward effects. Our interest in presenting these data at this time was threefold: first, the findings, indicated that a fixed low dose of heparin could potentially be life saving; second, it encouraged us to organize a prospective randomized clinical trial involving a large number of patients, and, lastly, it defined more accurately the possible rationale for using a fixed dose heparin for prophylaxis. At this stage the possible role of FXa in the regulation of intravascular clotting was being hotly debated by several research workers. It became apparent that the activity of an inhibitor of FXa was markedly enhanced by heparin [24], and that the biological activities previously defined as 'FXa inhibitor', 'antithrombin III' and 'heparin cofactor activity', all belonged to a single blood protease inhibitor with broad specificity, but whose primary physiological substrate was activated FX. Moreover, it had also been demonstrated that 1 µg of activated FX inhibitor, by inhibiting 32 units of activated FX, can indirectly prevent the potential generation of 16 000 NIH units of thrombin [24]. These findings emphasized that the enzymatic coagulation sequence functions as a biological amplification system. Accordingly it would be reasonable to anticipate that, if hypercoagulability were treated before intravascular coagulation was initiated, less of the antithrombotic agent would be required than if therapy was begun after thrombin formation had occurred. Subsequent studies had shown that heparin binds to both antithrombin and thrombin, and inhibition of thrombin prevented feedback activation of FV and VIII. In addition, by releasing tissue-factor pathway inhibitor, heparin, which by forming a complex with FVIIa, also contributed to inhibitory activity against FXa. Very significant progress was achieved on many fronts in 1971; events during this year laid the foundations not only for further developments in low-dose heparin prophylaxis but also proved critical for research on venous thromboembolism particularly further expansion in the areas of epidemiology, pathogenesis prevention and management of this disease. A very notable development was expansion of our facilities at King's to undertake biochemical and coagulation studies in surgical patients. I was indeed most fortunate that Mike Scully joined the Thrombosis Research Unit to head these facilities and has remained my most trusted and loyal colleague during the last 33 years. Three other important events in July 1971 marked the beginning of national and international recognition of efforts and contributions by the staff of the Thrombosis Research Unit at King's in research on venous thromboembolism. My first exposure to an international gathering was during the second congress of the International Society on Thrombosis and Haemostasis held in Oslo, where I was invited to discuss the recent progress in diagnosis of venous thrombosis [25]. In the same month, I had the privilege of organizing the first international symposium on venous thromboembolism held at King's on 10 July 1971, attended by leading international experts on thrombosis research including S. Wessler, A. P. Fletcher, S. Sherry, D. P. Thomas, A. A. Sasahara, V. Gurewich and B. Sigel. This meeting provided a unique opportunity to discuss rationale and early results of low-dose heparin prophylaxis [26]. Very significant national and international recognition paved the way for my appointment as Senior Lecturer in Surgery and Consultant Vascular Surgeon at King's College School of Medicine. At the same time major financial support from the King's College Hospital Research Trust and a number of grants from various organizations provided for the expansion of our research activities. We then assessed the efficacy of low-dose heparin prophylaxis in high-risk patients undergoing major operations. This study was designed to answer four questions. Was fixed low-dose subcutaneous heparin safe and effective in patients undergoing major surgery when assessed in a double-blind, randomized control study? Could individuals or groups of surgical patients who would not respond to low-dose heparin be recognized in advance? How might heparin be accurately monitored in human blood when present in low concentrations, and, finally, was there an optimal regimen of subcutaneous heparin that could be applied to all surgical patients? We investigated 261 patients undergoing major surgery, of these 78 took part in a double-blind randomized study, and 183 high-risk patients received fixed low doses of subcutaneous heparin [27]. In the double-blind part of this comparative, randomized trial, patients either received 5000 units of calcium heparin or a placebo solution of gelatin of the same consistency; subcutaneous injections were given two hours before operation and thereafter 12 hourly for 7 days or longer if the patient was still confined to bed. The efficacy of this regimen was also assessed in another 183 consecutive patients, 133 of these underwent major elective surgical procedures and 50 had emergency operations for fracture of the neck of femur. In the double blind part the frequency of DVT determined by the FUT was 42% in the control group but only 8% in the patients receiving heparin. The difference was statistically significant (P < 0.001). A low frequency of thrombosis (9.7%) was also observed in 133 consecutive patients undergoing major elective general or orthopedic operations. Only in patients who suffered fractures of the femoral neck several hours before heparin could be administered were the results unsatisfactory, since 20
Referência(s)