Published in the February 2009 issue of Today’s Hospitalist
When it comes to managing patients with venous thromboembolic disease, how should you risk stratify patients with pulmonary embolism? Who is the best candidate for low-molecular-weight heparin vs. unfractionated heparin? And in which patients should you suspect heparin-induced thrombocytopenia?
At a recent hospital medicine conference, hospitalist and anticoagulation expert Tracy Minichiello, MD, associate professor of medicine at the University of California, San Francisco, and chief of the San Francisco VA’s Anticoagulation Services, addressed the most common questions she hears from hospitalists on anticoagulation therapy. Here’s an overview of her advice.
Stratifying risk in PE patients
Most hospitalists now use validated scoring systems to help determine pre-test probability when they suspect pulmonary embolism (PE), said Dr. Minichiello. Those systems help guide both the choice of diagnostic testing and the interpretation of those tests.
But the concept of risk stratification once the diagnosis is made is new to many practitioners, she pointed out. Citing the most recent guidelines on antithrombotic and thrombolytic therapy published in a June 2008 supplement of Chest, Dr. Minichiello said that the assessment of all patients with PE should include rapid-risk stratification to identify those patients at higher risk of adverse outcome and death. Current strategies include echocardiogram to look for right ventricular dysfunction; assessment of RV size by CT scan; serum biomarkers; and a clinical risk score.
“This is really important, and not just as an academic exercise,” Dr. Minichiello said. “It can help you decide how to triage patients within the hospital, how soon to discharge them and if they are candidates for more aggressive therapy.”
Echocardiography has emerged as the principal risk-stratification tool in acute PE, said Dr. Minichiello. Using echocardiography to assess RV dysfunction does identify patients who may be able to be discharged early because they are at lower risk for early decompensation and who with submassive PE may be candidates for thrombolysis.
However, echocardiolography is not always available, and it entails getting an additional diagnostic test. Another way to stratify risk, Dr. Minichiello explained, is to ask the radiologist who interprets the patient’s CT angiogram to report the RV:LV ratio. If the ratio is less than 1, the patient’s risk of 30-day mortality from PE is “extremely low, perhaps obviating the need for additional testing.”
While researchers have evaluated serum biomarkers for their predictive value in patients with PE, she said that the best evidence recommends against using an abnormal troponin or BNP alone to guide management, due to low specificity. However, both measures have very good negative predictive value, Dr. Minichiello explained. “If the levels are normal,” she said, “the risk of adverse outcomes and elevated 30-day mortality is lower.”
Finally, a recently developed risk-scoring system, based purely on easily obtainable clinical variables, helps identify patients at low risk of adverse outcomes. (The scoring system was described in the Jan. 23, 2006, Archives of Internal Medicine.) Dr. Minichiello suggested that physicians who cannot risk stratify using an echocardiogram or CT angiogram might want to include this risk score when assessing PE patients.
In terms of when it is safe to discharge these patients, there are no available guidelines. That is why Dr. Minichiello is a proponent of assessing RV function in patients who otherwise feel well and are being prepared for rapid discharge.
“I don’t think it is prudent to quickly send home a patient with submassive PE who is at elevated risk of short-term adverse outcome,” she said. “After all, you cannot offer rescue thrombolysis once the patient is at home. ”
Initial PE treatment
With regards to fondaparinux for initial treatment, Dr. Minichiello said that the drug is clearly effective in treating DVT and PE. However, its very long half life “about 17 hours “and its lack of reversibility make it a better agent for the outpatient setting.
Low-molecular-weight heparin (LMWH) offers significant benefits over unfractionated heparin in terms of a more predictable dose effect, less bleeding and ease of transition to the outpatient arena. But Dr. Minichiello stressed that in certain patients, unfractionated heparin is clearly preferred.
Because LMWHs and anti-Xa inhibitors are primarily eliminated by the kidneys, these agents can accumulate and potentially increase bleeding risk in patients with renal impairment. “I would avoid,” she said, “the use of LMWH and fondaparinux in patients with end-stage renal failure and those with a creatinine clearance of less than 30.”
Patients at weight extremes may also be less than ideal candidates for LMWHs. For people at extremely low weight “women who weigh less than 45 kilos and men who weigh less than 55 kilos “there is a “less predictable dose effect that can lead to accumulation and increased risk for bleeding.”
For obese patients, Dr. Minichiello said, the problem is that there are very little data in patients over 150 kg. Unfractionated heparin is also the preferred agent in cases of massive PE and in patients with submassive PE when thrombolysis is being considered.
It’s not uncommon, she pointed out, for hospitalists to receive a patient who was started on unfractionated heparin in the emergency department but will be transitioned to LMWH once admitted. “Our service is often asked how to make this transition,” she said.
In an ideal world “one with adequate nursing staff “the best strategy would be to stop heparin one to two hours before starting the LMWH. But Dr. Minichiello said that because she can’t be sure that a nurse will get back to a patient in that short amount of time, she usually starts the LMWH at the same time she orders the unfractionated heparin to be discontinued.
And how do you transition a patient from LMWH to unfractionated heparin? Dr. Minichiello said she recommends discontinuing the LMWH and starting a weight-based IV unfractionated heparin drip, without a bolus, one or two hours before the next LMWD dose is due.
Making the transition to warfarin
When transitioning these patients to warfarin, the key is to be cautious and to start with low doses. Compared to outpatients, Dr. Minichiello explained, many hospitalized patients have underlying conditions (such as heart or liver disease, febrile illnesses or poor nutrition) or are taking other drugs ” including most antibiotics “that can interfere with warfarin metabolism.
“Keep 5 mg in your head as your starting dose,” she said. “The hospital is no place for a loading dose.” For elderly, frail and Asian patients, she added, the starting dose should be even lower: 2.5 mg. She noted that African-American patients, on the other hand, often require higher doses.
An online tool is now available that can help hospitalists select a starting dose of warfarin. The tool bases its dose estimates on a number of clinical factors and will also consider genotype information, if it is available, although you don’t need that information to use the tool.
Dr. Minichiello noted, however, that you shouldn’t transition all patients with venous thromboembolic disease to warfarin after initial treatment with an injectable anticoagulant. Patients with cancer-related DVT should continue on LMWH for a minimum of between three and six months, due to the reduced risk of recurrence with LMWH compared to warfarin therapy.
“Owning” IVC filters
To help prevent PE in patients with DVT, inferior vena cava (IVC) filter placement has become increasingly common. Dr. Minichiello cautioned, however, that these implanted mechanical devices have their limitations and only sparse data supporting their use.
While most experts agree that IVC filters are indicated in patients with acute DVT who have an immediate contraindication to anticoagulation, the only randomized, controlled trial examining the benefit of these devices was in patients already on anticoagulation. All other indications, including the use of IVC filters in patients with free-floating iliac thrombosis and in patients with submassive PE, “are controversial,” she said. Dr. Minichiello also noted that complications “which include insertion site thrombosis, DVT and migration “are not uncommon.
Given concerns about long-term complications, retrievable filters are now quite commonly used. “But retrievable filters are only temporary if practitioners remember to remove them,” she said.
This can be a challenge, particularly for hospitalists who see patients for only a short time and rarely follow patients after discharge. “If you order an IVC filter to be placed,” she said, “you should really own it until it comes out, or make sure you specifically hand off that responsibility, just as you would the follow-up of any other critical test or procedure. Develop a process for follow-up to be sure that removal actually happens. ”
It’s also important, she said, for temporary IVC filters to be removed as soon as possible; the longer a device stays in, the greater the chance that it will become thrombosed and/or be more difficult to remove. Dr. Minichiello warned, however, that if a patient received a filter because of a temporary contraindication to anticoagulation, physicians should give a few doses of anticoagulation once the therapy is deemed safe prior to removing the filter.
“Before you pull the filter out, make sure the patient can tolerate anticoagulation,” Dr. Minichiello said. She will typically do a “three-day test to see that the patient has no signs or symptoms of bleeding on full-intensity anticoagulation.”
She also encouraged physicians to start anticoagulation as soon as deemed safe because IVC filters do not prevent recurrent thrombosis or all PE.
Deborah Gesensway is a freelance health care writer based in Sierra Madre, Calif.
Does your patient have HIT?
ONE OF THE MOST DREADED complications of anticoagulation therapy is heparin-induced thrombocytopenia (HIT), a disorder characterized by a drop in platelet count and by paradoxically catastrophic clotting.
According to Tracy Minichiello, MD, chief of San Francisco VA’s Anticoagulation Services, clinicians need to develop a rational, evidence-based approach to determining the likelihood of HIT in a patient on heparin with a falling platelet count. She said she recommends using the “4Ts” approach first proposed by Theodore E. Warkentin, MD: timing, thrombocytopenia, thrombosis and oTher diagnoses.
First, consider timing. For patients without previous exposure to heparin, it takes approximately five days to develop antibodies, so thrombocytopenia will typically be seen between days five and 10. However, in patients who have had previous exposure and may have pre-formed antibodies, typically within the past three months, the platelet count can fall within 24 hours of exposure.
Thrombocytopenia that precedes initiation of heparin or occurs within the first few days is usually due to another cause, except in patients who have been previously exposed. Before you write this falling platelet count off to another drug or to DIC, go back and see if there’s any chance they had heparin within the last three months or if they had been hospitalized in that time period. “In this case, you need to suspect HIT,” Dr. Minichiello added. “Timing is very important.”
Next, consider the degree of thrombocytopenia. HIT is associated with a fall in platelet count of 50% from baseline. “People can miss this when patients have an elevated baseline platelet count and have a 50% drop, but are still not thrombocytopenic,” she warned.
Always think about HIT when a person develops a thrombosis while on heparin. Patients with HIT have a markedly increased risk of developing thrombosis, so once the diagnosis is made, all patients should undergo ultrasound to look for occult DVT.
Finally, consider alternative diagnoses. “If there is no alternative diagnosis that is more likely,” she said, “that increases the likelihood of true HIT.”
Studies have shown that patients on prophylactic doses of heparin are much less likely to develop HIT, Dr. Minichiello explained. The patients at higher risk for HIT are those who receive full intensity anticoagulation, particularly unfractionated heparin, and who have undergone orthopedic surgery or cardiopulmonary bypass. And monitoring for HIT is critical, Dr. Minichiello pointed out. For patients on heparin therapy, physicians should examine their platelet count at least every other day to allow for earlier identification and initiation of potentially limb- and life-saving therapy with direct thrombin inhibitors.
“If you suspect this disorder, stop all heparin, including flushes and heparin coated lines,” she said. “And call for a hematology consult STAT.”