|
Deep Vein Thrombosis: Will You Know It When You See It?
Deep vein thrombosis can mimic other, less serious diseases. The author discusses how prediction rules, lab tests, and radiologic studies can help you recognize this potentially deadly condition.
By Mel L. Anderson III, MD
A previously healthy 57-year-old woman presents to your emergency department complaining that she’s had pain, erythema, and swelling in her right leg for two days. She denies any trauma, fever, or dyspnea. Physical examination reveals tenderness in the right popliteal fossa with pitting edema of the right leg. Her calf circumference, measured 10 cm below the tibial tuberosity in both legs, is 3 cm greater in the right leg.
You suspect deep vein thrombosis (DVT), but other diseases have the same signs and symptoms. How do you differentiate DVT? As you will see in this article, which reviews the current literature on the diagnosis and initial management of acute DVT, using prediction rules, laboratory tests, and radiologic studies can help you recognize this disease.
COMMON AND POTENTIALLY SERIOUS CONDITION
Deep vein thrombosis is a common and potentially serious condition that may be encountered in an emergency department setting. Morbidity and mortality develop from associated pulmonary thromboembolism and post-thrombotic syndrome. Individual signs and symptoms lack sensitivity and specificity for diagnosis.
About 50 out of 100,000 patients a year are diagnosed with DVT, with the rate increasing to 200 per 100,000 in elderly populations. Most cases of DVT occur in the proximal deep veins of the lower extremities and are often related to an underlying hypercoagulable disorder, venous stasis, vascular injury, or a combination of these problems (see table below).
Pulmonary embolism frequently accompanies proximal DVT, with clinical features ranging from none to sudden death. By contrast, calf vein clots rarely lead to pulmonary embolism. However, one in five of these clots extends proximally and presents a subsequent risk for embolism. Mostly, DVT develops from the extension of calf vein clots.
After the acute phase of DVT, patients may develop chronic edema, pain, and inflammation of the involved leg, known as post-thrombotic syndrome. Accurate and timely diagnosis of DVT is necessary to offset the risk of this complication.
PREDICTION RULES
The patient presented at the beginning of this article had acute unilateral leg pain, erythema, and swelling—common in DVT. But because these signs and symptoms aren’t specific to DVT, prediction rules have been developed and prospectively validated to assist with the diagnosis. The challenge is to differentiate DVT from cellulitis, a ruptured Baker cyst, or other causes of lower extremity edema.
Although they are better known for pulmonary embolism prediction rules, Wells and colleagues also developed a DVT prediction rule that, when combined with the results of a negative D-dimer assay, may rule out further diagnostic testing. A score of two points or higher indicates that DVT is likely; a score of less than two indicates that DVT is unlikely (see table below).
The investigators also concluded that patients with a negative D-dimer assay coupled with a Wells score of less than two had a negative predictive value of 99.1% for DVT and could safely forego ultrasound testing or anticoagulant therapy.
All patients determined likely to have DVT by this prediction rule underwent ultrasonography and D-dimer testing. Those whose ultrasound tests were positive for DVT were treated. Those with negative ultrasounds and D-dimers weren’t treated and had no thromboembolic events when they were followed up three months later. Patients with negative ultrasounds but positive D-dimers weren’t treated but underwent repeat ultrasonography at one week, leading to the detection of proximal DVT in 3%. With most patients, this strategy resulted in DVT either being diagnosed or excluded on the day of presentation (see algorithm below).
 |
In addition, the number of ultrasounds performed on patients in this study was substantially decreased. Combined compression and color Doppler ultrasonography has a sensitivity greater than 95% and is highly specific for the diagnosis of DVT among symptomatic patients. However, the test detects only 30% of calf vein clots. A repeat ultrasound may detect proximal extension of calf vein clots.
Other prediction rules have been developed to help diagnose DVT. Both the Geneva score and revised Geneva score were created to standardize the pretest or clinical suspicion in diagnosing acute DVT. The goal was to produce a score using only objective measures and to avoid the subjective Wells criterion of “alternate diagnosis at least as likely as deep vein thrombosis.” The revised Geneva score still must be prospectively tested on patients with acute venous thromboembolism.
The Wells score has limitations. One large trial found that the negative predictive value of a low Wells score plus a negative D-dimer assay dropped from 99.1% to 97.1% in a primary care setting, where the prevalence of DVT is lower.
Most outcomes trials have used Wells criteria, so we have the best data on its operating characteristics when combined with the results of D-dimer testing. Consequently, Wells criteria remain the most reliable in day-to-day clinical use. The original Wells criteria published in 1997 consisted of nine items, but in 2003, these criteria were modified slightly. Recent surgery was changed from “within the previous 4 weeks” to “within the previous 12 weeks” and a new criterion of previously documented DVT was added.
At least seven types of D-dimer assays have been studied. Sensitivity analyses were conducted on all seven types and a systematic review was published in 2004. The ELISA and quantitative rapid ELISA tests provide the best sensitivity and negative likelihood ratios when compared with whole blood and latex agglutination assays. Sensitivity decreases from 96% to between 80% and 90% with whole blood and latex agglutination assays. Knowing which assay your institution uses will help you thoughtfully apply test results to patients.
TREATING DVT
Initial treatment for acute DVT consists of anticoagulation with a heparin agent and warfarin. Low-molecular-weight heparins (LMWHs), such as enoxaparin and dalteparin, have made outpatient therapy for uncomplicated DVT relatively safe. If your patient is eligible for outpatient therapy, be sure he doesn’t have a condition that contraindicates the use of LMWHs (especially advanced renal disease or obesity). Also be sure he knows how to inject himself safely or has someone who can perform the injections for him, and stress the need for frequent blood work to guide warfarin titration (see box below). Patients ineligible for an LMWH must be hospitalized to receive unfractionated IV heparin according to a weight-based nomogram (see table below).
 |
 |
A recent inpatient trial found outcomes were similar in patients given subcutaneous (SC) unfractionated heparin and those on an LMWH given subcutaneously. In the trial, a weight-based regimen for total unfractionated heparin needed daily was divided into two SC injections a day. This trial raises the possibility of treating outpatients with SC unfractionated heparin, although the investigators pointed out that an outcomes study should be conducted to confirm the safety of their hypothesis.
Warfarin inhibits protein C and protein S along with other vitamin K-dependent coagulation factors, creating an initially hypercoagulable state. Consequently, heparin must be started a few hours before warfarin. Starting warfarin on the first day of heparin therapy is usually safe, provided a therapeutic partial thromboplastin time is attained if unfractionated heparin is being used or if at least an hour has elapsed since administering an LMWH subcutaneously.
At a minimum, five days of heparin therapy must be overlapped with warfarin and two days of both therapeutic heparin and therapeutic warfarin must be given before discontinuing heparin. This is critical. Even when a therapeutic international normalized ratio (INR) is achieved in two or three days, patients must have heparin through day five before the drug is discontinued.
Investigators have studied the initial dose of warfarin across various populations. To test the hypothesis that a 10-mg loading dose might decrease the total duration of heparin therapy required, Kovacs and Rodger compared two groups of patients over two days. The first group received an initial warfarin dose of 5 mg followed by nomogram-based dosing. The second group received an initial warfarin dose of 10 mg followed by nomogram-based dosing. The percentage of patients with a therapeutic INR by day five of heparin therapy increased from 46% in the 5-mg loading dose group to 86% in the 10-mg loading dose group. No differences greater than five were found between bleeding events or instances of an INR.
To safely implement the 10-mg loading dose, it’s critically important to follow the protocol’s nomogram for dosing on subsequent days, guided by an INR check on day three after only two 10-mg doses of warfarin.
If possible, refer your patients with acute DVT to a pharmacist-run anticoagulation clinic. Treatment at these clinics improves anticoagulation control, decreases bleeding and thromboembolic event rates, and reduces unnecessary hospitalizations and emergency department visits. Anticoagulation clinics also can be cost effective and improve patient and provider satisfaction.
Compression stockings have been shown prospectively to significantly decrease post-thrombotic syndrome following acute DVT. Prandoni and Lensing studied patients with a documented first DVT and randomized them at discharge to knee-high compression stockings (30- to 44-mm Hg compression at the ankle) on the affected leg versus no stockings. At six months, one year, and five years, the compression stockings reduced the incidence of post-thrombotic syndrome to 25% in the treatment group versus 50% in the control group. While the study couldn’t be blinded, the results are robust and the intervention is both inexpensive and well tolerated. Clinicians should consider prescribing compression stockings to all patients with acute DVT of the lower extremity.
RETURNING TO OUR PATIENT
The patient who was presented at the beginning of this article was determined to have a Wells score of 3, using the prediction rules, indicating that DVT was likely. In this instance, D-dimer test results wouldn’t alter the need to perform an ultrasound examination, which confirmed DVT in the popliteal system. The patient’s renal function was normal and she weighed less than 330 pounds.
Outpatient therapy with enoxaparin (an LMWH) was initiated at a dose of 1 mg/kg twice daily. Emergency department staff taught the patient how to inject the drug and observed her doing so successfully. A minimum of five days of enoxaparin along with 5-mg tablets of warfarin was prescribed. The patient was instructed to begin taking 10 mg of warfarin on nights one and two after discharge and to go for laboratory testing on the morning of day three. The test results would determine if she needed further nomogram-based titration of warfarin for nights three and beyond. The patient was also prescribed knee-high compression stockings and referred to the pharmacist-run anticoagulation clinic for ongoing care.
EFFECTIVE MEASURES
Patients with acute DVT are common in emergency department settings and, luckily, effective measures are available to diagnose and treat this potentially serious condition. A focused history and physical examination should lead to the calculation of a Wells score to guide further testing. Once a diagnosis is made, treatment with unfractionated heparin or an LMWH should begin unless contraindications to anticoagulation are present. Many patients are candidates for outpatient management, avoiding the costs and risks associated with hospitalization. Follow-up laboratory testing and warfarin dosing according to defined protocols increase the safety of outpatient therapy.
Suggested Reading
Buller HR, et al.: Antithrombotic therapy for venous thromboembolic disease: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 126(3 Suppl):401S, 2004.
Kovacs MJ, et al.: Comparison of 10-mg and 5-mg warfarin initiation nomograms together with low-molecular-weight heparin for outpatient treatment of acute venous thromboembolism. Ann Intern Med 138(9):714, 2003.
Prandoni P, et al.: Below-knee elastic compression stockings to prevent the post-thrombotic syndrome: a randomized controlled trial. Ann Intern Med 141(4):249, 2004.
Stein PD, et al.: D-dimer for the exclusion of acute venous thrombosis and pulmonary embolism: a systematic review. Ann Intern Med 140(8):589, 2004.
Wells P, et al.: Does this patient have deep venous thrombosis? JAMA 295(2):199, 2006.
Wells P, et al.: Evaluation of D-dimer in the diagnosis of suspected deep venous thrombosis. N Engl J Med 349 (13):1227, 2003. |
|
