Acute intraoperative heparin-induced thrombocytopenia (HIT) and thrombosis during coronary artery bypass grafting: Two case reports providing evidence for the role of preoperative LMWH in triggering sensitization.

INTRODUCTION: Systemic anticoagulation is necessary during cardiac surgery. To date, the only well established anticoagulation protocol involves the use of heparin. However, heparin can cause heparin-induced thrombocytopenia (HIT) a potentially life threatening immune-mediated thromboembolic syndrome. Until now, devastating consequences of HIT syndrome in patients undergoing heart surgery have been described, but only postoperatively. Here we report the development of HIT syndrome during cardiac revascularization by intra-operative heparin administration in two patients previously exposed to LMWH. PATIENTS/METHODS: We report on two patients who developed rapid and profound intravascular coagulation with severe thrombocytopenia (platelet count decreased from ≥250×109/L to 50×109/L) due to HIT development caused by heparin administration during coronary artery bypass graft surgery. In addition we report that fondaparinux, given intra-operatively in association with antithrombin, may be a suitable alternative anticoagulant for successfully preventing the devastating consequences of intra-operative HIT development. CONCLUSION: To our knowledge, this is the first report describing the development of acute intra-operative HIT, secondary to high-dose UFH administered for coronary revascularization, in which the unexpected presence of platelet-activating anti-PF4/heparin antibodies at surgery was explained by preoperative administration of a one-week course of LMWH but without any preoperative evidence for HIT.

Long term effects of tamoxifen. Biological effects of Tamoxifen Working Party.

A total of 153 breast cancer patients who participated in two trials of adjuvant tamoxifen and who had not recurred were recruited into a study of the long term effects of tamoxifen. There were 60 controls (no tamoxifen), 73 ex-users (mostly for 2 years) and 20 current users (median treatment duration 72 months) and the median follow-up time was 7 years. A wide ranging study of lipids, hormones, bone density and haemostasis was undertaken. When compared with controls, current users had lower cholesterol levels (especially low density cholesterol), and increased triglyceride levels. Thyroid hormones were higher and sex hormone binding globulin was almost doubled. Bone density was non-significantly higher, clotting times were slightly shorter and fibrinogen and antithrombin III levels were reduced. However few of these changes persisted in ex-users, suggesting that most of the biological effects of treatment are reversible on cessation of treatment. This is reassuring for potentially negative side-effects, but also indicates that potentially positive 'side-effects' such as cholesterol lowering only occur while on active treatment.

Influence of whole blood on standard curve for heparin measurement--possible heparin binding by red cells.

Measurement of heparin ex vivo is usually with reference to standard curve prepared with a "spiked" normal human plasma pool (NHP). When the calibration curve was prepared by addition of heparin to whole blood before plasma separation, although the linear relationship was maintained the slope was increased in comparison to the classical standard calibration curve. It is concluded that the preparation of the calibration curve by addition of heparin to NHP may give erroneously high heparin levels in treated patients' plasma, leading perhaps to inappropriate dosage. It was also observed that when heparin was added to blood of different haematocrit (prepared by addition of washed RBC to plasma) and plasma prepared, the subsequent APTT was decreased with the fall in haematocrit; suggesting that the laboratory monitoring of heparin treatment should take into account the patient's haematocrit.

The influence of DDAVP infusion on the coagulation and fibrinolytic response to surgery.

The response of components of the coagulation and fibrinolysis systems to infusion of DDAVP has been examined in patients undergoing elective surgery. In the DDAVP treated group there was a significant increase, compared to control, in plasminogen activator (by fibrin plates p less than 0.005, ECLT p less than 0.0125, by Student's t test) before operation. No difference between groups was seen by either methods in the activator levels in samples 24 h postoperation, whereas a significant drop (p less than 0.002) in protein C concentration was observed at this stage in the treated group. Levels of factor VIII components were significantly higher (p less than 0.005) than control at all stages of operation and a significant shortening (5 sec p less than 0.05) of the APTT was seen at all stages (apart from 24 h samples). DDAVP infusion therefore may exacerbate the hypercoagulable state observed in surgical patients without preventing the (post-operatively) fibrinolytic shutdown. Instead, infusion tends to produce fibrinolytic depletion at the key mid-operative stage.

Use of low molecular weight heparin in pregnancy.

In a controlled study of 15 pregnant patients undergoing therapeutic termination of pregnancy, seven received subcutaneously 5,000 anti-FXa units of low molecular weight (LMW) heparin 15 and 3 h prior to the termination, and eight patients acted as controls. Paired maternal and fetal blood samples were taken (before or immediately after the termination) for assay of heparin activity by a chromogenic anti-FXa method sensitive to levels of 0.02 anti-FXa U/ml. LMW heparin was detected in all maternal samples of the test patients but was not detected in any of the fetal samples. The use of LMW heparin as a thromboprophylactic agent was then evaluated in 11 patients who were known to have a severe thromboembolic tendency, had suffered recurrent miscarriages and had responded poorly to conventional anticoagulation (oral anticoagulant, conventional heparin). All patients receiving LMW heparin in thromboprophylactic doses completed uneventful pregnancies and gave birth to healthy babies (three for the first time) without complication. Bone density scans performed in all patients shortly after the delivery showed normal mineral mass. We conclude that LMW heparin does not cross the placental barrier, and in addition offers satisfactory antithrombotic protection for both maternal and placental circulation. In addition, this study provides preliminary data from 11 patients suggesting LMWH may not give rise to maternal osteoporosis, a finding that now needs further investigation.

A dose ranging study to evaluate dermatan sulphate in preventing deep vein thrombosis following total hip arthroplasty.

Dermatan sulphate catalyses thrombin inhibition by heparin cofactor II; it has a lower haemorrhagic to antithrombotic ratio than that of heparin in animal models. Consecutive patients aged forty years or more, electively undergoing total hip replacement under general anaesthesia, were randomly allocated to one of three dosage regimens of dermatan sulphate (MF701, Mediolanum Farmaceutici) given intramuscularly. These were 200 mg once daily (n = 50), 200 mg twice daily (n = 52) and 300 mg twice daily (n = 51), administered from twenty-four hours pre-operatively until the tenth postoperative day. The overall incidence of DVT assessed by bilateral venography was 53%, 51% and 34% respectively (Chi-square test for trend p = 0.06). The incidence of major proximal DVT was 10.6%, 8.5% and 2.1% respectively. Pulmonary embolism (PE) and bleeding were assessed in all 153 patients. There was one case of PE in each dose group. The incidence of bleeding episodes, volume of blood lost and blood transfusion requirements were low and showed no increase with increasing dose. The patients were followed up 4-8 weeks after discharge. We conclude that the two lower doses were subtherapeutic in this population, however dermatan sulphate given 300 mg twice daily, proved to be efficacious with an incidence of proximal major DVT of 2.1% and a low incidence of bleeding complications. A trial of dermatan sulphate 300 mg twice daily compared to standard prophylactic agents is needed.

The effects of oestrogen administration on the plasma free protein S and C4b-binding protein.

This study was undertaken to evaluate the effects of oestrogen administration (low dose as an oral contraceptive or higher dose as a hormone replacement therapy) on the levels of plasma free protein S and C4b-binding protein. The participants were 59 women aged 18-49 years, divided into 2 groups: A and B. Group A was composed of 22 post-menopausal women on a hormonal replacement therapy programme (HRT) consisting of 2 mgs daily oestradiol valerate for 21 days. Group B was divided into subgroup B1: 18 women who had been on oral contraceptive for at least one year and subgroup B2 (control): 17 women who were not pregnant and not taking any oral contraceptive. In this study were also included two young women who both suffered from severe thromboembolic disease a few months after initiation of oral contraceptive. The first was 25 years old, with congenital moderately decreased prekallikrein (activity and antigen 40% and 45% respectively) and the second was a 21 year-old woman with congenital moderately decreased plasminogen activity and antigen 45%). In both cases, family members with similarly reduced levels of prekallikrein (PK) and plasminogen (PLG) respectively were free from any thromboembolic disease and had normal protein S levels. In Group A, 22 women at the end of the first cycle of treatment, had lower levels of free protein S (p less than 0.001) than before the initiation of HRT. In subgroup B1, the levels of free protein S were found to be significantly lower than in subgroup B2 (p less than 0.001).(ABSTRACT TRUNCATED AT 250 WORDS)

PIP: This study was undertaken to evaluate the effects of estrogen administration (low dose as an oral contraceptive [OC[ or higher dose as hormone replacement therapy (HRT) on the levels of plasma free protein S and C4b-binding protein. The participants were 59 women ages 18-49 years, divided into 2 groups. Group A was composed of 22 postmenopausal women on an HRT program consisting of 2 mg daily estradiol valerate for 21 days. Group B was divided into subgroup B1--18 women who had been on OCs for at least 1 year--and subgroup B2--a control group of 17 women who were not pregnant and not taking any OCs. In this study were also included 2 young women who both suffered from severe thromboembolic disease a few months after initiation of OCs. The 1st was 25 years old, with congenital moderately decreased prekallikrein (PK); (activity and antigen 40% and 45%, respectively) and the 2nd was a 21-year old woman with congenital moderately decreased plasminogen (PLG) activity and antigen 45%. In both cases, family members with similarly reduced levels of PK and PLG respectively were free from any thromboembolic disease and had normal protein S levels. In Group A, 22 women at the end of the 1st cycle of treatment had lower levels of free protein S (p0.001) than before HRT initiation. In subgroup B1, the levels of free protein S were found to be significantly lower than in subgroup B2 (p0.001). Although the C4b-binding protein was 20% lower in subgroup B1 than in the other subgroup, this does not represent a significant difference. Since an association between OC use and incidence of venous thromboembolism without predisposition to thrombosis has been consistently observed in case-control and cohort studies, the authors conclude that estrogen reduces free protein S in both group A and subgroup B1; i.e., the group receiving higher dosages as an HRT for a period of 21 days, and the subgroup B1 receiving low doses as an OC over 12 months, whose mild deficiency in a factor may predispose the thrombosis heretofore dormant.

The influence of LMW heparin on the coagulation and fibrinolytic response to surgery.

The response of components of the coagulation and fibrinolytic systems has been examined in patients undergoing minor and major elective surgery and receiving or not receiving subcutaneously administered prophylactic low molecular weight (LMW) heparin. Blood samples were withdrawn pre-operation (PO), post-anaesthesia (PA), during operation (DO) and 24 hours post-operation. Release of plasminogen activator occurred post anaesthesia at a time when factor VIII components were unchanged or decreased. Induction of anaesthesia decreased plasminogen (p less than 0.005) fibrinogen (p less than 0.02) ATIII (p less than 0.002) and fast a2-antiplasmin (p less than 0.005). During operation plasminogen activator release reached a peak in association with a moderate increase in factor VIII components. Heparin treatment produced a prolongation of APTT at DO (p less than 0.05) and at 24hr (p less than 0.002) stages, this prolongation being apparently unrelated to its concentration but did not prevent the rise of factor VIII components observed at 24 hr (p less than 0.02). Prekallikrein (p less than 0.05) and antithrombin III levels (p less than 0.05) were significantly higher whereas fast a2-antiplasmin (p less than 0.002) was significantly lower at 24 hours in patients undergoing major operation and treated with heparin. Protein C levels fell significantly at 24 hours in the untreated patient (p less than 0.014) and in the heparin treated group the fall was greater than the control group (p less than 0.007). At 24 hours other haemostatic and fibrinolytic components were unaffected by heparin treatment.

The influence of tamoxifen in vivo on the main natural anticoagulants and fibrinolysis.

The long-term effects of tamoxifen on alterations in haemostasis which could lead towards thrombosis were investigated in 149 women who were disease-free for at least 5 years following mastectomy for breast cancer. All participants were randomized to receive tamoxifen as a post-surgical adjuvant treatment (89 patients, treated group) or not (60 patients, controls) for at least 2 years. 5.62% of the cases treated with tamoxifen suffered a venous thrombosis, while no thromboembolism was reported in the control group. No significant differences were observed between groups in the global clotting times, fibrinogen, fibrinolytic factors, or in the concentration of the main natural anticoagulants, antithrombin III (AT-III), protein C(PC) and protein S (PS). However, when the treated group was sub-divided, current users (n = 18) of the drug (median treatment duration 72 months) had significantly lower AT-III (P < 0.05) and PC (P < 0.05) activities, together with higher levels of plasminogen activity (P < 0.01) and tissue plasminogen activator antigen (P < 0.01), compared with 71 ex-users (who mostly received treatment for 2 years) and controls. We conclude that long-term treatment with tamoxifen for 2 or more years tends to reduce both AT-III and PC, a situation possibly predisposing towards thrombosis. Monitoring haemostasis in tamoxifen-treated breast cancer patients is therefore advisable.

Long-term administration of heparin and heparin fractions and osteoporosis in experimental animals.

To test whether heparin-induced osteoporosis is influenced by the molecular weight of heparin, 24 male rabbits received single daily subcutaneous injections of either physiological saline (controls, n = 5), low molecular weight heparin (LMWH, n = 7), conventional heparin (UFH, n = 7) or high molecular weight heparin (HMWH, n = 6). Heparin was administered in supratherapeutic daily dosages for 120 days (750 anti-FXa units/kg for 90 days and 1500 anti-FXa units/kg for another 30 days). Studied variables were: serial analysis of serum calcium, albumin, phosphate and alkaline phosphatase, measurement of the cortical thickness of the femur (radiographically), tibial and trabecular bone density (both by cross-sectional analysis) and femoral fragility. Observed changes in blood biochemistry associated with bone metabolism were not correlated to any of the treatments. Compared with the controls, a reduction in cortical and trabecular bone density was seen with UFH (P < 0.05) and HMWH (P < 0.01) but not with LMWH. Femoral fragility was also significantly increased (P < 0.002) by HMWH. In conclusion, LMWH did not cause toxic skeletal effects as opposed to HMWH which clearly did, and UFH which induced some osteoporotic changes.

The influence of venous occlusion on the status of protein S.

In the present study the effect of perturbation of the vessel wall by venous occlusion on protein S release, was evaluated in 10 healthy male volunteers. A 9% (P less than 0.01) increase in free protein S and a 7% (NS) in total protein S over the baseline were observed 7 min after venous occlusion, whereas von Willebrand factor antigen (vWF:Ag)--a known release product of vascular endothelium--increased by 11% (P less than 0.01). In addition, the possibility of free protein S release from unstimulated platelets was investigated in pre- and post-occlusion samples incubated at room temperature for 24 h as whole blood or platelet-rich plasma. An increase in the mean free protein S was observed in the pre-occlusion samples which peaked after 4 h of incubation--121% at time 0 to 130% (P less than 0.05) of the normal pooled plasma whereas total protein S increased by only 4% (NS). In the post-occlusion samples, a similar peak in mean free protein S levels was observed after 12 h of incubation--132% at time 0 increased to 142% (P less than 0.01) of the normal human plasma pooled and the total protein S was increased by 8.55 (P less than 0.05). Small but significant increases in vWF:Ag levels were observed at 2 h and 6 h for the pre- and post-occlusion samples, respectively. No changes in C4b-binding protein were observed throughout the study. We conclude that, firstly, venous occlusion causes release of protein S in addition to vWF:Ag and, secondly, unstimulated platelets release protein S in addition to vWF:Ag.

Congenital thrombophilia among patients with venous thromboembolism.

During a 3-year period we studied 393 adult patients (382 of whom were unrelated) with a history of acute venous thromboembolism. A congenital deficiency state known to predispose to thrombosis was found in 27.2%. Of these, most were due to deficiencies of protein C (9.2%), protein S (7.6%), antithrombin III (5%) or to increased plasma PAI-1 concentration (3.1%) which, in the absence of any known factor that predisposes towards thrombosis, results in a diminished fibrinolytic activity. There was a characteristic pattern between the age of onset (mean 34 years) of thrombosis and individual protein deficiency. Thrombosis appeared spontaneously in 73% of cases with recurrence in 80%. In contrast, in the remaining unrelated patients, 138 (35.1%) in whom venous thromboembolism was secondary and occurred at a mean age of 43 years, and in the other 140 (35.6%) who suffered thromboembolism spontaneously at a later age (mean age 55), there was no permanent protein deficiency state or alteration in fibrinolytic activity and thrombosis recurrence was lower (53.6% and 20.7% respectively). Of the 393 patients, deep vein thrombosis was the most common manifestation; however, in congenital thrombophilia, thrombosis of visceral vessels and Raynaud's syndrome (6%) were also detected.