Objective: The objective of this study was to develop and evaluate an unfractionated heparin dosing nomogram based up on a chromogenic anti-Xa assay and to employ this nomogram for patients presenting with either venous or coronary thrombosis.
Design: A prospective, two-phase cohort study design.
Setting: Intensive care unit, coronary care unit, emergency department.
Participants: Patients presenting with coronary or venous thrombosis.
Interventions: A heparin dosing nomogram was developed and implemented. Pre-printed nomograms were completed by the prescribing physician. Nurses were responsible for performing dosage adjustments and drawing bloodwork as specified in the nomogram.
Main Outcome measures: The percentage of patients with a therapeutic anti-factor Xa chromogenic heparin assay level at 24 hours, the time required to exceed the minimum therapeutic threshold (>0.35 U/mL), time to obtain the therapeutic range, time to stabilization within the therapeutic range, and incidence of bleeds.
Results: In the first phase (the pilot phase) (n=7) the primary outcome was attained by 28.6% of patients. The majority of patients were supratherapeutic at 6, 12, and 24 hours. These results suggested that the nomogram dosing was too aggressive, and thus the nomogram was modified for the second phase. Nineteen patients were included in the assessment of the modified nomogram. The percentage of patients who were therapeutic vs. supratherapeutic at 24 hours was 36.8% and 57.9%, respectively. Thus, 94.7% of patients exceeded the therapeutic threshold at 24 hours. Fifteen patients (79%) were stabilized within the therapeutic range, with a mean time to stabilization of 25.0 +/- 8.6 hours. There were no reports of major bleeds.
Conclusions: The modified nomogram was successful in exceeding the therapeutic threshold both rapidly and safely. Most patients subsequently settled within the therapeutic range. This nomogram is now routinely used at our center.
J Inform Pharmacother 2001;4:100-9.
Several heparin dosing nomograms based on patient weight and activated partial thromboplastin time aPTT have been developed and compared with empiric dosing by physicians. The advantages of these nomograms include individualization of initial dosing, standardization of dosage adjustments and structured monitoring. Furthermore, patients who were dosed according to a nomogram had a greater chance of attaining a therapeutic aPTT with the first measurement at six hours, achieved a more rapid therapeutic aPTT, and exceeded the therapeutic threshold sooner compared to those receiving standard dosing. On the other hand, patients were more likely to experience a supratherapeutic aPTT within the first 24 hours. However, there was either no difference in the frequency of bleeding between standard and nomogram-based heparin dosing or no documentation of bleeding complications.
The anti-Xa chromogenic heparin assay (ACCUCOLOR(TM)) is based on the heparin-accelerated neutralization of the activated serine protease factor Xa. Our institution has been using the heparin assay since August of 1998. A therapeutic level is considered to be in the range of 0.35 to 0.67 U/mL. Unlike the aPTT, the anti-Xa heparin assay specifically measures the biologic activity of heparin and is thus unaffected by concomitant therapy with warfarin or thrombolytics or by altered levels of coagulation factors.1 Acute phase reactants, factor VIII and fibrinogen, which are usually increased in patients with thromboembolic disease, may lead to a decreased response of the aPTT. Antithrombin deficiency, which may manifest as heparin resistance, or increased levels of heparin binding proteins may also lead to a decreased response. Alternatively, patients with mild reductions of multiple coagulation factors have an increased response of the aPTT. The presence of lupus anticoagulants and deficiencies in factor XII, prekallikrein, and high-molecular-weight kininogen lead to prolongation of the baseline aPTT. The aPTT may also be unreliable in patients with overwhelming infections, myocardial infarctions, or severe liver disease.
Heparin assays may be reliable in patients presenting with a prolonged baseline aPTT. However, heparin assay results are affected by age, obesity, increased levels of heparin binding proteins, hepatic or renal disease, and general heparin resistance. It is also unknown whether heparin assays are superior to aPTT testing in patients with liver disease or consumptive coagulopathies. In addition, heparin assays have limited availability and are more expensive to perform than the aPTT test.
Levine et al. have compared a heparin-dosing nomogram, based on chromogenic anti-factor Xa heparin assay results, but not patient weight, to the aPTT for monitoring response to therapy in patients with heparin resistance. They concluded that many patients who were heparin resistant according to the aPTT were, in fact, therapeutic according to the heparin assay. There was a greater incidence of bleeds in the group monitored by the aPTT although this difference was not statistically significant.
Prior to this study at our institution, adjustments in heparin infusion rates in response to non-therapeutic heparin assays were made at the discretion of the physician. Nursing staff informed the physician of all heparin assay levels and awaited a new order before adjusting the infusion rate. It was felt that with the implementation of a heparin dosing nomogram based on patient weight and heparin assay results, therapeutic heparin assay levels could be achieved more quickly and more efficiently.
The objective of this study was to develop and evaluate an unfractionated heparin dosing nomogram based upon a chromogenic anti-Xa assay and to employ this nomogram for patients presenting with either venous or coronary thrombosis.
A prospective, two-phase cohort design was used. The first phase involved the enrollment of a small pilot patient population to initially assess the effectiveness of the nomogram in achieving therapeutic heparin assays. The results from this pilot study were then used to determine if any changes to the nomogram were warranted. The second phase involved validating the modified nomogram using a larger patient population. For this phase, it was decided to enroll only those patients presenting with acute coronary syndromes as very few venous thrombosis patients were being initiated on unfractionated heparin.
The pharmacy investigator met with several individuals before the nomogram was distributed to the emergency department (ED), intensive care unit (ICU), and coronary care unit (CCU). In these areas, the head nurse, nursing team leaders, and nursing staff were informed of the study. In addition, the study was explained to ICU and cardiology medical staff.
The study population consisted of patients admitted to The Ottawa Hospital (General Campus) with a diagnosis of either venous or coronary thrombosis and who were initiated on intravenous unfractionated heparin. Therapy was begun in either the ICU, CCU, or ED.
Inclusion criteria were as follows: patients greater than 18 years of age diagnosed with either coronary or venous thrombosis, admitted to either the ICU, CCU, or ED and expected to receive at least 24 hours of unfractionated heparin therapy. Exclusion criteria consisted of obesity (i.e. greater than or equal to 130% of ideal body weight), increased risk of bleeding, history of heparin-induced thrombocytopenia (HIT), documented allergy to heparin, or thrombocytopenia (platelet count less than 100 x 109 g/L prior to therapy).
Each patient was assessed for risk of a major bleed (see data collection). It was left up to the physician to determine a patient’s risk of bleeding and whether he/she should be initiated on the heparin nomogram.
Patient consent was not deemed necessary as no randomization occurred. Moreover, this study was conducted as a continuous quality assurance program for the ICU, rather than as a clinical research protocol.
The initial nomogram designed by the department of hematology was based on patient weight, anti-Xa heparin assay results, and indication for heparin, either venous or coronary thrombosis. Deep venous thrombosis (DVT) clots are generally known to be larger than coronary artery clots. Also, given the burden of thrombus and involved tissue in DVT patients, there may be greater activation of acute phase reactants, which are known to bind to heparin. Therefore, patients presenting with venous thrombosis may initially have increased heparin requirements and thus higher loading doses were prescribed.
Pre-printed order forms were distributed to the ED, CCU, and ICU to be completed by the prescribing physician. The physician determined the initial heparin bolus and infusion rate based on patient weight and indication for heparin therapy. Baseline laboratory tests (international normalized ratio (INR), complete blood count (CBC) and aPTT) were ordered as specified in the nomogram. Six hours after the initial heparin bolus, the anti-Xa heparin assay was determined. Based on the results of the first and subsequent heparin assays, nurses took appropriate action as directed by the nomogram without involving the physician. The CBC was repeated daily while the patient was receiving heparin therapy. Heparin assays were reordered 6 hours following each dosage adjustment and daily if no dosage adjustments were made. The therapeutic range of 0.35 to 0.67 U/ml for the heparin assay was the same as used by Levine et al.
Demographic data including age, actual body weight, sex, reason for initiation of heparin therapy, concomitant disease states, bleeding risk, and indication for heparin therapy as well as baseline laboratory values (INR, platelets, hemoglobin, and heparin assay), date, time and location of initiation of heparin therapy, initial bolus and infusion rate, and results of all heparin assays were recorded. Next, the following were determined: time to reach the therapeutic range (0.35-0.67 U/mL) following initiation of therapy, time to stabilization (as defined by the time to reach the first of two consecutive therapeutic heparin assays) following initiation of therapy, total number of adjustments required to achieve stabilization, duration of heparin therapy, adverse events, and compliance.
Risk factors for a major bleed were defined by at least one of the following criteria: major surgery in the preceding 7 days, coagulation defects (baseline INR >1.2), gastrointestinal bleed or cerebral hemorrhage in the previous 3 months, uncontrolled hypertension (SBP >180 mmHg and/or DBP >120mmHg), thrombolytic therapy in the preceding 7 days, or concurrent antiplatelet therapy (including NSAID and ASA).
Adverse events (type and action taken) were recorded as either a major or minor bleeding episode or other (i.e. thrombocytopenia). A major bleeding episode was defined as meeting one or more of the following criteria: a decrease in hemoglobin of more than 20 g/L, transfusion of two or more units of blood, or bleeding in the retroperitoneum, cranium, or prosthetic joint. A minor bleeding episode was defined as a bleed not meeting any of the criteria of a major bleed.
The primary outcome measure was the percentage of patients whose heparin assay results fell within the therapeutic range (0.35-0.67 U/mL) at 24 hours. Secondary outcome measures included the percentage of patients whose assay results fell within the therapeutic range at 6, 12, and 48 hours, the time required to exceed the minimum therapeutic threshold (>0.35 U/mL), the percentage of patients stabilized within the therapeutic range at 12, 24, and 48 hours, the time required to stabilization within the therapeutic range, the percentage of supratherapeutic assays (> 0.67 U/mL) at 6, 12, 24, and 48 hours, the mean number of adjustments per patient to achieve stabilization within the therapeutic range, and the success rate of achieving a therapeutic assay following a dosage adjustment.
Descriptive statistics were used to evaluate the data. Non-compliance with the nomogram was assessed by the number and severity of deviations from the nomogram. Criteria for excluding patients included any of the following: incorrect initial bolus and/or infusion rate, omission of first or subsequent heparin assays, incorrect dosage adjustment, first heparin assay drawn greater than 2 1/2 hours late or early, or any subsequent assay drawn greater than 4 hours late or 2 hours early.
During this phase, 39 patients were initiated on the nomogram. The majority were excluded from the analysis for the following reasons: 19 patients were excluded due to the temporary unavailability of the heparin assay from the hematology laboratory, 8 patients were excluded based on gross non-compliance with the nomogram, 4 patients were excluded based on obesity, and 1 patient was excluded due to the unavailability of the chart. The remaining 7 patients, all cardiology patients, were included in the initial analysis.
The mean duration of heparin therapy for these 7 patients was 75.2 +/- 91.4 hours (range 22.3-278.6 hours). Only 2 patients (28.6%) were within the therapeutic range at 24 hours and thus attained the primary outcome. At 6, 12, and 48 hours, the percentages of patients within the therapeutic range were 14.3%, 14.3%, and 66.7%, respectively. The remaining patients had supratherapeutic assays at these times. Overall, 40 heparin assays were performed. Of these assays, 27.5% were therapeutic and 65% were supra-therapeutic.
Time to exceed the therapeutic threshold was 6.4 +/- 1.4 hours (range 3.8- 8.3 hours). Three patients remained supratherapeutic for the duration of heparin therapy and thus never reached the therapeutic range. Of the 4 remaining patients who did achieve the therapeutic range, they did so with a mean time of 21.2 +/- 17.4 hours (range 3.8-44.7 hours). Only 1 patient was stabilized within the therapeutic range, with a time to stabilization of 41.7 hours. Length of heparin therapy in this patient was 278.6 hours.
One patient experienced a large hematoma on the left flank on the third day of heparin therapy with a drop in hemoglobin of 53 g/L, necessitating discontinuation of heparin therapy. This patient had also received treatment with alteplase on arrival to the hospital. The first two heparin assays were greater than 0.8 U/mL.
Based on these results, it was decided that the nomogram should be modified and made less aggressive. The revised nomogram retained the same heparin loading dose formula, however the initial maintenance dose infusion rates were decreased. (Figure 1) Furthermore, the second phase of the study was limited to cardiology patients with acute coronary syndromes since very few venous thrombosis patients were being treated with unfractionated heparin.
The nomogram used in the second phase of the study is shown in Figure 1.
Forty-four patients were initiated on the nomogram over approximately a one month period during the second phase of the study. In total, 25 patients had to be excluded from the data analysis for the following reasons: obesity, duration of heparin therapy less than 24 hours, indication for heparin other than coronary thrombosis, already receiving heparin, and marked non-compliance with the nomogram. Some patients may have been excluded based on more than one criteria. Data for remaining 19 patients were used to assess the efficacy of the nomogram (Table 1). Characteristics of these patients are presented in Table 1. A total of 95 assays was reported by the hematology laboratory. Of these, 47 (49.5%), 41 (43.2%), and 7 (7.4%) assays were therapeutic, supratherapeutic, and subtherapeutic, respectively.
Table 1. Patient characteristics during the second phase
| Age (y)*
|64.6 +/- 11.3 (44-80)
75.4 +/- 8.2 (67-89)10
|Reason for initiation of heparin:
Acute myocardial infarction
|Risk factors for a major bleed:
|Location of initiation of heparin:
|Concomitant disease states:
Coronary artery disease
Congestive heart failure
Prior stroke/Transient ischemic attack
Peripheral vascular disease
|Baseline laboratory values*
|32 +/- 6 (25-48)
1.11 +/- 0.20 (0.91-1.60)
244 +/- 72 (117-395)
131 +/- 15 (129-135)
*Data presented as mean +/- standard deviation, range in parentheses.
The primary outcome, the percentage of patients whose assay results fell within the therapeutic range (0.35-0.67 U/mL) at 24 hours, was 36.8% (Table 2). The percentage of patients whose assay results were therapeutic at 6, 12, and 48 hours was 15.8%, 21.1%, and 83.3%, respectively. The percentage of patients whose assays were supratherapeutic at 6, 12, 24, and 48 hours was 84.2%, 78.9%, 57.9%, and 8.3%, respectively (Table 2). Table 3 further outlines the degree to which the assays were supratherapeutic. The magnitude of elevation of the supratherapeutic assays tended to diminish with time. The mean duration of heparin therapy was 60.1 +/- 19.2 hours (range 25.6-97.3).
Table 2. Number of patients therapeutic, supratherapeutic, and subtherapeutic at 6, 12, 24, and 48 hours during second phase
|Therapeutic (%)||3 (15.8)||4 (21.1)||7 (36.8)||10 (83.3)|
|Supratherapeutic (%)||16 (84.2)||15 (78.9)||11 (57.9)||1 (8.3)|
|Subtherapeutic (%)||0||0||1 (5.3)||1 (8.3)|
Time to exceed the therapeutic threshold and to obtain the therapeutic range were 6.2 +/- 0.8 hours (range 5.3-8.0) and 23.0 +/- 10.0 hours (range 5.4-35.5), respectively. The mean number of dose adjustments required to achieve the first therapeutic assay was 2.1 +/- 1.3 (range 0-4). In total, 15 patients had assay results stabilized within the therapeutic range. The percentage of patients whose assay results stabilized within the therapeutic range by 12, 24, and 48 hours was 5.3%, 26.3%, and 75%, respectively (Table 3). The mean number of dosage adjustments required to achieve stabilization was 2.2 +/- 0.8 (range 0-4). The mean time to stabilization within the therapeutic range was 25.0 +/- 8.6 hours (range 5.4-35.5). However, of these 15 patients, only 11 remained within the therapeutic range until heparin therapy was discontinued. Of the remaining 4 patients, 3 had subtherapeutic results while 1 had supratherapeutic results. Of the 4 patients who were never stabilized in the therapeutic range, 2 had their heparin therapy discontinued after the first therapeutic assay, 1 received heparin for 25.6 hours and had only 1 assay drawn, and the other patient was subtherapeutic following both supratherapeutic and therapeutic results.
Table 3. Magnitude of elevation of the supratherapeutic assays at 6,12, 24, and 48 hours during second phase
|0.68 – 0.74 U/mL||4||6||9||1|
|0.75 – 0.79 U/mL||6||5||2||0|
The success rate of achieving a therapeutic assay following one dosage adjustment was 43.2%. In the 16 patients who were not therapeutic at 6 hrs (all were supratherapeutic), the first dosage adjustment was successful in 3, leading to a subsequent therapeutic heparin assay. The remaining 81.2% of assays remained supratherapeutic after dosage adjustment.
There were no major bleeds among the patients included in the data analysis. However, 8 patients experienced minor bleeds and/or bruising, including the 5 patients who had received treatment with alteplase (tPA). These 5 patients had at least one other risk factor for a bleed (e.g. daily ASA). One chronic ASA user also had an initial INR of 1.53. The remaining 3 patients who had not received tPA had also had at least one risk factor for a bleed (daily ASA +/- initial INR > 1.2). Types of bleeds included: rectal bleed (1 patient), epistaxis (2 patients), and bleeding at new/old intravenous site (3 patients). There were no reports of heparin-induced thrombocytopenia.
Complete compliance with the nomogram was achieved in 5 patients. In the remaining 14 patients, a total of 27 minor deviations occurred.
The nomogram used in the first phase of the study was successful in rapidly exceeding the therapeutic threshold. However, there was a high percentage of supratherapeutic assays, and not all patients reached the therapeutic range. Nonetheless, the results from this small patient sample were valuable as they indicated that the nomogram to be used in the second phase should be made less aggressive.
The modified nomogram used in the second phase of the study proved to be effective in rapidly and safely exceeding the therapeutic threshold as 36.8% of patients attained the primary outcome and the mean time to obtain the therapeutic range was 23.0 hours. Furthermore, the majority of patients were stabilized within the therapeutic range with a mean time of 25.0 hours. One patient was stabilized by as quickly as 5.4 hrs since the first heparin assay was drawn in the late evening, approximately 7 hours before the daily heparin assay, drawn in the early morning.
In general, patients in the second phase tended to be supra-therapeutic with the first 1 to 3 assays before achieving the therapeutic range. The percentage of patients supra-therapeutic at 24 hours was 57.9%. Despite these results, there was no evidence of any major bleeds. In fact, studies involving patients with venous thrombosis have not clearly demonstrated that excessive anticoagulation with heparin during the first few days of therapy is associated with an increased risk of bleeding.15 However, there is sufficient proof that serious bleeding may result despite therapeutic anticoagulation with heparin.16 In the first phase of the study, 1 patient experienced a major bleed. However, this patient had also received treatment with alteplase which may have been contributory. Despite the many supra-therapeutic assays and the concomitant use of alteplase in 5 patients, there were no major bleeds associated with the modified nomogram used in the second phase.
In one study which examined the use of an aPTT-based heparin nomogram in coronary patients, 72.5% of patients had assays which were within the therapeutic range at 24 hours as compared to 36.8% of patients in the present study. The discrepancy in results may be explained by the fact that the previous study made dosage adjustments based on units/kg/hour. As compared to the present study, this method of dosage adjustment may be more precise in achieving the therapeutic range at 24 hours. With the aPTT-based nomogram, only 72.5% exceeded the therapeutic threshold within 12 hours whereas all patients in the present study exceeded the therapeutic threshold with a mean time of 6.15 hours. Furthermore, 33% of patients in the previous study had an initial supratherapeutic aPTT as compared to 84.2% of patients in the present study. However, the present study aimed to rapidly exceed the therapeutic threshold and thus used higher loading doses which may explain this higher percentage of initially supratherapeutic assays.
In a second study which evaluated the use of an aPTT-based nomogram in a coronary care unit, the times to exceed the therapeutic threshold and to achieve a therapeutic aPTT were 6.7 hours and 20.8 hours, respectively. These values are comparable to the times of 6.15 and 22.96 hours found in the present study. In contrast, at 24 hours, more patients were therapeutic and fewer were supra-therapeutic than in the present study, 51% vs. 36.8% and 38.3% vs. 57.9%, respectively. These differences may be explained by the less aggressive dosing used in the previous study, a fixed bolus of 5000 units followed by an initial infusion rate of 15 units/kg/hour.
Recently, a pharmacoeconomic study was carried out to determine the costs of monitoring unfractionated heparin therapy by the anti factor Xa assay compared to the aPTT test. Although the 96-hour assay cost using the heparin assay was 16% higher than with the aPTT test, the group monitored with the heparin assay had significantly fewer tests and dosage adjustments. This reduction in the number of tests and dosage adjustments compensated for most of the increased costs associated with the assay. The investigators concluded that, given the anti-Xa heparin assay advantages over the aPTT test in certain groups of patients and the modestly increased costs, the assay should be used to monitor unfractionated heparin therapy in most large hospitals.
The present study did not consider the possible interaction between heparin and intravenous nitroglycerin, leading to decreased heparin efficacy as demonstrated by the aPTT results. This interaction is of questionable significance and if it does occur, it is usually at higher doses of nitroglycerin. Furthermore, the interaction has been reported only with the aPTT test, not with the heparin assay. Even if such an interaction occurred in our study population, the present study focused on the timeframe required to achieve the therapeutic range, rather than the exact results of the heparin assays.
Limitations of this study included the small sample size. Unfortunately, the hematology laboratory was unable to perform heparin assays for approximately two thirds of the initial phase due to a lack of reagent from the supplier. Furthermore, approximately 36% of patients initiated on the heparin nomogram during the second phase of the study had to be excluded from the data analysis due to non-compliance with the nomogram. Perfect compliance was achieved in only 5 patients. There was reluctance on the part of physicians to prescribe the recommended loading doses and initial infusion rates as dictated by the nomogram as these differed significantly from the customary 5000 unit loading dose and 1000 unit/hour maintenance infusion. In total, 7 patients were excluded based on either an incorrect loading dose, initial infusion rate, or both. The tendency also existed to make dosage adjustments based on non-therapeutic aPTT results, despite therapeutic heparin assays. Other instances of non-compliance included assays either drawn at inappropriate times or omitted altogether, inappropriate dosage adjustments or dosage adjustments not done when indicated, and missing heparin assay results. The lack of compliance may have been due to nursing time constraints and laboratory delays in providing heparin assay results. In addition, the inability of an investigator able to monitor active patients on a 24 hour basis would have played a key role in increasing compliance with the nomogram. However, it may be argued that since the evaluation of the nomogram was conducted as an effectiveness trial rather than as an efficacy trial, the low compliance rate more accurately reflects clinical practice.
Secondly, the lack of a control group does not permit any direct comparisons to be made between nomogram and empiric dosing in terms of efficacy and safety. Thirdly, this study was designed to use heparin assay results, and not patient outcomes, as endpoints. Lastly, this study examined the use of the nomogram in non-obese cardiology patients. Thus, extrapolation of these results to obese patients as well as to patients receiving heparin for indications other than coronary thrombosis may not be feasible. For example, as previously mentioned, patients initially presenting with deep venous thrombosis often have increased heparin requirements. Patients at increased risk of bleeding, such as stroke patients, may suffer more serious consequences if they should experience a bleed.
Positive feedback regarding the nomogram was received from both nursing and cardiology staff. Nurses appreciated the autonomy of making dosage adjustments without the need to contact cardiology staff. In turn, cardiology staff was satisfied that patients were attaining the therapeutic range and that there were no major bleeds attributed to heparin therapy. In addition, other services, including vascular surgery and family medicine, expressed interest in the nomogram. The nomogram which was implemented throughout the hospital was slightly modified. It was decided to treat a heparin assay of greater than 0.8 U/mL more aggressively by holding the infusion for 1 hour and then decreasing the rate by 250 U/hr, rather than 200 U/hr.
The final nomogram was successful in rapidly and safely exceeding the therapeutic threshold. Although evaluated only in cardiology patients, the use of this nomogram was implemented throughout the hospital for use in non-obese patients at low risk of bleeding presenting with either coronary or venous thrombosis.