Why the study?
Does unfractionated heparin reliably achieve therapeutic aPTT levels in patients with acute pulmonary embolism?
Does unfractionated heparin reliably achieve therapeutic aPTT levels in patients with acute pulmonary embolism?
Answers are grounded in this paper and cite their sources.
Unfractionated heparin for acute pulmonary embolism rarely achieves consistent therapeutic aPTT levels in the first 48 hours, reinforcing guideline recommendations to use low-molecular-weight heparin instead.
Acute pulmonary embolism (PE) is a “can’t miss” diagnosis in the emergency department (ED) that mandates investigation should clinical suspicion be aroused. As the growing availability of multidetector-row computed tomography pulmonary angiography (CTPA) has decreased the threshold for testing, the prevalence of PE has rapidly increased to as many as 112.3 cases per 100,000 U.S. adults.1 For PE management, there remains considerable debate regarding inpatient versus outpatient treatment, which patients benefit from advanced therapies, and who requires long-term anticoagulation. Nevertheless, therapeutic anticoagulation alone remains the mainstay of acute PE management for the vast majority of patients. Professional guidelines recommend prompt initiation of anticoagulation in acute PE, even before the diagnosis is confirmed in those highly suspected of the disease.2 Rapid initiation of therapeutic anticoagulation also improves patient outcomes. In a study of 400 consecutive patients diagnosed with acute PE in the ED, administration of therapeutic anticoagulation was associated with lower in-hospital and 30-day mortality rates. Achieving a therapeutic activated partial thromboplastin time (aPTT) within the first 24 hours led to 4.1 and 9.2% absolute rate reductions in in-hospital and 30-day mortality rates, respectively.3 The majority of patients with acute PE are still treated with heparin products, most commonly intravenous unfractionated heparin (UFH) or a subcutaneous low-molecular-weight heparin (LMWH). In the Multicenter Emergency Medicine Pulmonary Embolism in the Real World Registry (EMPEROR), UFH was the most commonly administered anticoagulant in the ED (47.8% of participants), and this rate exceeded that of all LMWHs or fondaparinux combined (37.0%).4 Commonly cited reasons for using UFH as a first-line agent include its rapid onset, short half-life, ability to “turn off” in case of thrombolysis, invasive procedure, or bleeding and the availability of a reversal agent. Despite these explanations, the preference for UFH is at odds with professional guidelines and published data. UFH, in contrast to LMWH, is notorious for its unpredictable pharmacokinetics from its small volume of distribution and variable half-life due to a wide-ranging molecular weight, two-phased elimination, and extensive binding to plasma proteins.5 As a result, UFH is within the therapeutic range only 22% of the time in patients with acute PE.6 Current guidelines recommend LMWH or fondaparinux over UFH in the majority of patients.2 The only exception is individuals with high-risk PE, defined as those with hemodynamic instability.2 Fewer than 5% of patients with PE are classified as high risk by this definition.7 Routine use of either systemic or catheter-delivered thrombolysis in low- or intermediate-risk PE is not indicated,2, 8 so ubiquitously administering UFH in anticipation of advanced therapies is unwarranted. This continued reliance on UFH also exposes patients to an increased risk of bleeding. In an observational study of 2,610 patients (3,066 hospitalizations) receiving therapeutic anticoagulation for any indication, UFH was associated with a nearly fivefold increased risk of moderate or severe bleeding compared to LMWH.9 Similarly, in a meta-analysis of 1,951 patients with PE from 12 randomized controlled trials, UFH was associated with a 49% increased risk of bleeding complications compared to LMWH, although this did not reach statistical significance.10 The current issue of Academic Emergency Medicine includes a contemporary analysis of UFH for acute PE in the setting of PE Response Team (PERT) consultation that adds importantly to this evidence base.11 This study included all patients at Massachusetts General Hospital with an acute PE and subsequent PERT activation over a 4.5-year period from 2012 to 2017. The PERT is a multidisciplinary team consulted for known and suspected PE felt to be at high risk.11 This consultation system rapidly facilitates evaluation of the patient, discussion of care options involving multiple specialties, and ultimately delineation of a cohesive care plan. The present study analyzed patients treated with UFH according to a hospital-wide ordering template embedded in the electronic medical record. Two dosing options were available: standard dosing, which involved a bolus of 80 U/kg followed by an infusion at 18 U/kg/hour, or an infusion-only regimen, which consisted of an infusion at 18 U/kg/hour without a bolus. Heparin dosing was titrated to target an aPTT of 60 to 80 seconds with repeat laboratory measures typically drawn every 6 hours. The primary outcome was the proportion of individuals with an aPTT within this range during each 6-hour time period of treatment. Of the 716 patients with PERT activation, 505 (70.5%) were treated with UFH and were otherwise eligible. The median age was 66 years with a median BMI of 29.3 kg/m2. As intended with PERT activation, this was an ill population. Nearly half of participants had evidence of right heart strain on CTPA, classifying these cases as submassive PE.12 A total of 28.3% were classified as having a high-risk or massive PE, although only 7.9% had evidence of hemodynamic collapse. As some manifestation of hemodynamic instability is the sine qua non of massive PE,12 the reason for this inconsistency is not explained. Not surprisingly, the incidence of acute PE portended poor outcomes, and 30-day mortality was 5.7% overall. Despite their illness and the fact that the majority of patients had submassive or massive PE, few individuals required advanced therapies. Rates of systemic thrombolysis, catheter-based intervention, and surgical embolectomy were 4.0, 5.9, and 2.0%, respectively. Bleeding occurred in 3.0% of individuals during 30 days of follow-up. The results for the primary outcome are sobering and highlight the fickle pharmacokinetics of UFH. No more than 34.9% of individuals treated with the standard dosing regimen had a therapeutic aPTT within the first 48 hours of therapy. The rate was as low as 13.9% within the first 6 hours. The proportion of individuals with subtherapeutic aPTT values ranged from 48.2% to 60.6%, and the proportion with supratherapeutic values was 11.0% to 34.8%. Individuals treated with the infusion-only protocol, presumably a group at higher risk of bleeding, had higher rates of supratherapeutic aPTTs (9.1%–37.4%). Perhaps most damning, only 1.3 and 0.7% of patients receiving the standard-dosing and infusion-only regimens, respectively, had consistently therapeutic aPTT values in the first 12 hours. From hours 12 to 48, not a single patient was consistently in the therapeutic aPTT range. Similarly, with the standard-dosing regimen, only 57.4% of patients had a single therapeutic aPTT in the first 48 hours; the rate was only 62.1% in those on the infusion-only protocol. The authors should be applauded for this contemporary addition to the growing literature on UFH for PE. In our experience, frequent reassessment of patients with acute PE on therapeutic anticoagulation in the first 24 to 48 hours is critical to determine who requires advanced therapies. As the present analysis demonstrates, however, UFH makes this nearly impossible, as fewer than 2% of patients are appropriately anticoagulated in that time frame. These data should also reassure providers that very few patients with submassive and massive PE ultimately receive advanced therapies, even at a large tertiary facility capable of delivering all such therapies. Similarly, not all such interventions, such as catheter-delivered thrombolysis for submassive PE, need to be performed on an emergent basis. Therefore, routine use of the therapeutically inferior UFH in anticipation of procedures that rarely occur is unnecessary. Using UFH over LMWH to reduce the risk of bleeding is also unwarranted. More than one-third of individuals on UFH had aPTT values in the supratherapeutic range, and a dosing regimen without a UFH bolus performed even worse than one with a bolus. Such a strategy falsely reassures clinicians while unnecessarily exposing patients to further risk. As the authors note, there are several limitations to the present analysis. This was a single-center study at a tertiary care facility, limiting the generalizability of the study, although our own clinical experience matches the present data. Some institutions use a higher upper limit for a therapeutic aPTT, but this would still not account for the large incidence of subtherapeutic aPTTs. Other institutes dose UFH based on anti-Xa levels, but this comes with additional cost and runs counter to professional guidelines.2 Outcome rates in terms of bleeding, thrombosis, and death were too low to examine the impact of dosing strategies on each. Additionally, data comparing clinical events for those on LMWH to UFH are also not available. Nonetheless, these data, along with previous reports, clearly demonstrate that routine use of UFH for acute PE needs to be curtailed. UFH is neither associated with less bleeding nor better outcomes than LMWH or direct oral anticoagulants. For other acute cardiovascular conditions, such as myocardial infarction, we pride ourselves on the prompt initiation of guideline-recommended therapies supported by clinical data. Acute PE management, especially in the PERT era, should be no different. Since existing data and professional guidelines are clearly being disregarded, we recommend facility-wide algorithms to direct appropriate anticoagulant use for acute PE. We hope that these data prompt organizations such as the PERT Consortium to lead this charge.
No takes yet. Share an insight, caveat, or question.
Aday et al. (2019) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: