The number remaining at risk: an adjunct to the number needed to treat.

Although the number of patients needed to treat (NNT) to prevent an adverse clinical event is of great clinical value to practising physicians, it is limited in that it fails to provide a measure of prognosis among patients not achieving benefit. For example, if the NNT is 100, what is likely to happen to the other 99? The number remaining at risk (NRR), which is an index that enhances the value of the NNT, is described. The NRR is the ratio of the residual event rate among treated patients and the absolute reduction in outcome events (NRR = experimental event rate [EER]/control event rate [CER]-EER), where EER and CER are the event rates among experimental and control groups, respectively. This index represents the number of events likely to occur among the NNT, or the odds of experiencing an adverse outcome event as opposed to deriving benefit from therapy. The NRR is a simple index that can easily be calculated from the published results of a clinical trial. As an adjunct to the NNT, it provides a measure of the impact of therapy and the average prognosis of remaining patients.

Dalteparin as periprocedure anticoagulation for patients on warfarin and at high risk of thrombosis.

BACKGROUND: Patients taking warfarin and at high risk for thromboembolic complications have traditionally been hospitalized for two to three days to receive standard treatment with intravenous heparin both prior to and following procedures while their international normalized ratio (INR) is subtherapeutic. OBJECTIVE: To assess the feasibility of protocol implementation for outpatient anticoagulation with low-molecular-weight heparin to eliminate or reduce the length of hospital admission needed solely for anticoagulation. METHODS: Patients included were receiving warfarin for a prosthetic heart valve, mitral valve disease with atrial fibrillation, or recent episode of venous thromboembolism. Warfarin was discontinued four days prior to the procedure. Subcutaneous dalteparin 200 units/kg was given on the two mornings prior to the procedure and restarted 12-24 hours after the procedure until the INR was in the therapeutic range. Warfarin was reinitiated on the evening of surgery. RESULTS: Twenty-four patients underwent 26 procedures. There were two minor bleeding complications, and one patient experienced a transient ischemic attack. Patients received a median of five days of dalteparin. The INR returned to the therapeutic range on the median postoperative day 4. All patients avoided two days of hospitalization prior to the procedure (i.e., no patients needed to be admitted preoperatively for anticoagulation). A median of four days would have been required for the sole purpose of postoperative anticoagulation. CONCLUSIONS: Outpatient perioperative anticoagulation with dalteparin for high-risk patients requiring long-term oral anticoagulation appears feasible and warrants further study.

Body weight does not predict for anti-Xa levels after fixed dose prophylaxis with enoxaparin after orthopedic surgery.

Enoxaparin after joint arthroplasty is effective prophylaxis against venous thromboembolism. This is usually given as a fixed dose without monitoring of anti-Xa levels. This study assesses the relationship between trough anti-Xa levels, body weight, and venous thromboembolism. Consenting patients at three institutions were treated with Enoxaparin 30 mg subcutaneously bis in die postoperatively until discharge. Chromogenic anti-Xa levels were measured on the fifth postoperative day by the method of Stachrome (Diagnostica Stago). All patients had bilateral compression doppler ultrasonography on day 10 or discharge and were followed for 12 weeks for evidence of venous thromboembolism. Eleven patients developed objectively confirmed venous thromboembolism during the study. In this study, there was poor correlation between weight and anti-Xa levels. In addition, body weight and anti-Xa levels of patients who developed venous thromboembolism were compared to those who did not and there were no significant differences between the two groups. In conclusion, this study shows that there is poor correlation of trough anti-Xa levels with body weight. Recognizing the low overall event rate this study does not support the need to monitor anti-Xa levels or adjusting the dose according to weight.

A standard heparin nomogram for the management of heparin therapy.

A nomogram for the adjustment of heparin dosage was developed to standardize heparin therapy and to reduce delays in achieving and maintaining a therapeutic activated partial thromboplastin time (APTT) result. Fifty consecutive patients with acute venous thromboembolism had their continuous intravenous heparin therapy adjusted according to this heparin nomogram. The effect of the nomogram on heparin therapy in these patients was compared with data from 53 historical control patients. The proportion of patients in the nomogram group who reached a therapeutic APTT at 24 hours after the start of heparin therapy was 66%, which increased to 81% at 48 hours. In contrast, 37% and 58% of the control patients reached a therapeutic APTT at 24 and 48 hours, respectively. The percentage of therapeutic APTT results of the total number of APTT determinations was greater in the nomogram patients than controls. The use of this heparin nomogram resulted in (1) achieving a therapeutic APTT at 24 and 48 hours in a large proportion of patients and (2) reduced periods of inadequate anticoagulation and overanticoagulation during heparin therapy.

An evaluation of impedance plethysmography and 125I-fibrinogen leg scanning in patients following hip surgery.

Venous thromboembolism is a common post-operative complication in patients following hip surgery. 125I-fibrinogen leg scanning and impedance plethysmography (IPG), are often used in the detection of venous thrombi in such patients. Information on the sensitivity and specificity of these non-invasive tests for the diagnosis of venous thrombosis following hip surgery is relevant for both patient management and for choosing the appropriate outcome measure for clinical trials evaluating new prophylactic regimens. We determined the sensitivity and specificity of the IPG alone, the 125I-fibrinogen leg scan alone, as well as the combined use of the two tests from a retrospective analysis of 685 hip surgery patients who participated in clinical trials of anti-thrombotic prophylaxis. These patients were followed prospectively with non-invasive tests. Bilateral venography was attempted either when one or both screening tests became positive or on day 10-14 post-operatively if both screening tests remained negative. Adequate venography was obtained in 1,010 (73.7%) legs and thrombi were identified in 198 (19.6%) legs. The sensitivities of the IPG and leg scanning were 12.9% and 44.6% respectively; the corresponding specificities were 98.1% and 95.0%. The sensitivity of a positive result on one or both screening tests was 49.6% with a specificity of 93.9%. Therefore, leg scanning and IPG, even in combination, are not sufficiently accurate to be recommended as the only strategy for the diagnosis of venous thrombosis following hip surgery. Venography should be considered in all patients undergoing surveillance testing either when one or both of the screening tests become positive or on day 10-14 if the screening tests remain negative.

Human tissue-type plasminogen activator releases fibrinopeptides A and B from fibrinogen.

In five patients with venous thromboembolic disease treated with recombinant tissue-type plasminogen activator (rt-PA), there was a marked increase in the mean concentrations of fibrinopeptide A (from 0.6 to 5.9 nM; P less than 0.0001) and desarginine fibrinopeptide B (from 5.6 nM to 24.1 nM; P less than 0.01) 30 min after a bolus of rt-PA (0.6 mg/kg). Thrombin was unlikely to be responsible because the levels of desarginine fibrinopeptide B exceeded those of fibrinopeptide A and the changes occurred despite concomitant heparin therapy. The purpose of this study therefore, was to determine whether rt-PA directly releases the fibrinopeptides from fibrinogen. Incubation of rt-PA with heparinized plasma or purified fibrinogen resulted in time and dose-dependent release of both fibrinopeptide A and B. Contaminating thrombin was not responsible for this activity by the following criteria: the rate of rt-PA mediated fibrinopeptide B release was considerably faster than that of fibrinopeptide A, and fibrinopeptide release was unaffected by heparin, hirudin, or a monospecific antithrombin IgG. Aprotinin also had no effect on fibrinopeptide release, indicating that this activity was not plasmin mediated. Fibrinopeptide release was shown to be due to rt-PA because this activity was completely blocked by a monoclonal antibody against the enzyme. Further, the specificity of rt-PA for the thrombin cleavage sites on fibrinogen was confirmed by the demonstration that rt-PA released fibrinopeptide A or fibrinopeptide B from fibrinopeptide A or B-containing substrates, respectively. These studies thus demonstrate that (a) rt-PA releases fibrinopeptides A and B from fibrinogen thereby indicating that this enzyme is not specific for plasminogen, and (b) plasma fibrinopeptide A and desarginine fibrinopeptide B levels are not specific markers of thrombin action on fibrinogen in patients receiving rt-PA.

Usefulness of erythrocyte ferritin analysis in hereditary hemochromatosis.

A study was carried out to determine the usefulness of erythrocyte ferritin analysis in identifying homozygotes and heterozygotes in families affected with hereditary hemochromatosis, an autosomal recessive disorder. To select the subjects the genotypes of 60 people from 26 affected families were determined by HLA-A and HLA-B haplotyping. In addition, data for 12 homozygotes for whom erythrocyte ferritin values were available from the literature were included. Likelihood analysis was used to evaluate the diagnostic value of erythrocyte ferritin analysis alone and in combination with serum ferritin testing. An erythrocyte ferritin value of 150 ag/cell or higher combined with a serum ferritin level above the 90th percentile indicated homozygosity, whereas a value of less than 150 ag/cell and a serum ferritin level at or below the 90th percentile indicated that homozygosity could be ruled out with a high degree of confidence. The probability of heterozygosity rose to 92% when the erythrocyte ferritin value was between 29 and 149 ag/cell and to 98% when this result was combined with a serum ferritin level at or below the 90th percentile. Erythrocyte ferritin analysis in combination with serum ferritin testing is useful for identifying homozygotes and a proportion of heterozygotes in families affected with hemochromatosis.