Value of D-Dimer in Risk Stratification for Thromboembolism in Patients With Atrial Fibrillation and Low CHA2DS2-VASc Score

Abstract

Atrial fibrillation (AF) can cause thrombi formation and subsequent emboli deposition in systemic arteries, leading to various organ ischemia and infarction. Anticoagulation therapy can reduce the risk of thrombus formation and embolization, and is initiated based on a patient’s risk score, which is frequently estimated with the CHA2DS2-VASc score. We present a case of thromboembolism (TE) where a low CHA2DS2-VASc score suggested a low-moderate risk of systemic embolization, but an elevated plasma D-dimer value prompted further investigation which revealed an intracardiac thrombus with renal embolism. The patient is a 63-year-old male with past medical history of hypertension and AF treated with ablation 2 years prior presenting with sharp right flank pain of 5-hour duration. Primary workup and imaging were unrevealing at the time, and a low CHA2DS2-VASc score was suggestive of aspirin therapy. However, an elevated D-dimer of 289 ng/mL and a transient increase in creatinine pointed to possible etiology of embolic origin. The diagnosis was confirmed with computed tomography (CT) with contrast and transesophageal echocardiogram, revealing renal infarcts and the source of the emboli, respectively. The patient was treated with heparin and transitioned to apixaban prior to discharge with full resolution of symptoms. Through this case, we wish to show D-dimer’s predictive value of TE, as well as its potential benefit in risk assessment in patients with AF.

Introduction

Atrial fibrillation (AF) is a major cause of morbidity and mortality. ¹ Abnormal fibrillation of the atria leads to the stasis of the blood resulting in thrombi formation, especially within the left atrial appendage. These thrombi can embolize, and deposit in systemic arteries, leading to renal, mesenteric, or splenic artery infarction, as well as embolic stroke.^1-6 Anticoagulation therapy can reduce the risk of thrombus formation and embolization, and is initiated based on the patient’s risk for systemic embolization, frequently estimated with the CHA2DS2-VASc score.^7,8 We present a case of thromboembolism where a low CHA2DS2-VASc score of 1 suggested a low-moderate risk of stroke or systemic embolization. However, a more patient-specific biomarker for thrombosis, plasma D-dimer, prompted further investigation into a possible intracardiac thrombus with renal cardioembolism.

Case

A 63-year-old male with a past medical history of hypertension and AF treated with ablation 2 years prior presented to the emergency department (ED) for acute, sharp right flank pain of 5 hour duration. The pain radiated from the back to the right abdomen, and he reported nausea and 1 episode of non-bloody emesis. He denied hematuria, history of renal stones, and was not on any home medications. He was afebrile, his heart rate was 92 bpm, blood pressure was 178/110, respiratory rate was 22, and O₂ sat was 100% on room air. On physical examination, he had tenderness on the right side of the abdomen and at the costovertebral angle. Labs in the ED were significant only for a mildly elevated B-type natriuretic peptide (BNP) of 169, and a Cr of 1.3 with a prior baseline of 1.1. Initial non-contrast computed tomography (CT) abdomen and pelvis revealed no nephrolithiasis, hydronephrosis, cholelithiasis, or cholecystitis. Electrocardiogram (EKG) showed AF with a rate of 80 bpm. Given the unremarkable non-contrast CT, low CHA2DS2-VASc score of 1, and otherwise asymptomatic presentation, the team questioned whether the patient could be discharged with aspirin therapy.

Though the workup for abdominal pain was unrevealing at this time, a transient increase in creatinine measured in the ED on hospital day 1 suggested possible renal injury. On hospital day 2, creatinine had trended back down to 1.16, close to his baseline. The patient’s D-dimer was elevated at 289 ng/mL, with the hospital cut-off value for a high D-dimer at 250 ng/mL. Contrast CT abdomen and pelvis was ordered to evaluate possible renal thromboembolic event based on the patient’s elevated D-dimer, right flank pain, and AF, despite low-moderate pre-test probability of thromboembolism given his CHA2DS2-VASc score.

Contrast CT revealed a classic picture of multiple wedge-shaped infarcts caused by repeated thromboemboli to the right kidney (Figure 1). Transesophageal echocardiogram (TEE) showed a mildly dilated left atrium (LA), with reduced LA appendage (LAA) velocities. Multiple thrombi in the LAA and in the adjacent atrial septal aneurysm were also identified (Figure 2). His left ventricular ejection fraction (LVEF) had fallen to 30% to 35%, compared to normal values in previous measurements. After initial treatment with heparin, anticoagulation therapy was transitioned to Apixaban prior to discharge. At the point of discharge, his flank pain had resolved, and the patient was stable.

Figure 1.

Contrast CT of abdomen and pelvis showing multiple wedge-shaped infarcts caused by repeated thromboembolism to the right kidney.

Figure 2.

Transesophageal echocardiogram showing a mildly dilated left atrium with reduced left atrial appendage velocities and a thrombus in the left atrium.

Discussion

Intracardiac thrombi and subsequent embolization are a major cause of renal infarction, especially in patients with pre-existing AF. ³ Lodging of emboli in the renal artery can lead to serious renal complications due to ischemia and infarction. D-dimer measures fibrin degradation products and the activity of thrombin, making it a useful marker for thrombosis. ⁹ Current clinical practice uses high negative predictive value of D-dimer to exclude diagnoses such as deep vein thrombosis (DVT) and pulmonary embolism (PE) in low-risk patients. ¹⁰ While it is not a definite rule-out test for these conditions, its usefulness and high sensitivity has made it a mainstay in the diagnosis of DVT and PE when pretest probabilities are low.^11,12 Unfortunately, due to its poor specificity, the D-dimer test is not routinely used when there is a high pretest probability of other thromboembolic events. Conditions such as trauma, infection, and recent surgery can also lead to increased levels.^13-15 In AF, the use of D-dimer as a biomarker for assessment of thromboembolic risk has not been well established. However, multiple studies have shown elevated D-dimer is related to an increased risk of thrombus formation in patients with AF.^16-18

Virchow’s triad for clot formation is all present in AF. Stasis of blood due to irregular LA contraction, endocardial damage especially within the LAA, and irregular levels of clotting and coagulation factors are all seen in AF.^19,20 These factors increase the risk of clot formation in the LA and LAA, and subsequent embolization. By definition, D-dimer should be a good measure of risk assessment for AF patients because the embolic sequelae are the product of coagulation and fibrinolysis—for both of which D-dimer has been established to be a gold standard of measurement. ²¹ A study of 509 individuals with non-valvular atrial fibrillation (NVAF) treated with warfarin showed that those with D-dimer >150 ng/mL had significantly increased risk of transient ischemic attack and stroke than those with <150 ng/mL, after controlling for other risk factors. Other markers of coagulation such as prothrombin fragment 1+2, platelet factor 4, and β-thromboglobulin were not effective at predicting thromboembolic events. ²² This case supports previous literature on the utility of D-dimer as an independent predictor of thrombus formation and embolic events. More case control studies exploring the relationship between D-dimer and TE in patients with AF should be done to further our understanding.

Another point that we wish to show with this case is its potential benefit in embolic risk assessment for patients with AF. Systemic embolization to other organs may also occur in AF. One of the most worrisome outcomes is stroke due to its devastating consequences, so anticoagulation therapy is essential to avoid these complications. In patients with AF, thromboembolic risk assessment is typically assessed via the CHA2DS2-VASc scoring system. Factors considered are older age, sex, history of congestive heart failure, hypertension, thromboembolism, vascular disease, and diabetes are factored into the calculation. ²³ Because anticoagulation therapy increases risk of intracranial hemorrhage and other bleeding complications, ²⁴ a clinical risk assessment such as the CHA2DS2-VASc score is used to determine if therapy has more benefit than the risk. For men, scoring stratifies scores of zero into low risk, 1 low-moderate risk, and 2 or more into moderate-high risk; for women all the cutoffs are increased by 1 point. A high score indicates the benefit of anticoagulation outweighs the bleeding risks due to the high likelihood of embolic complications. Low-risk patients most likely do not need anticoagulation therapy, while moderate-high risk patients will likely benefit from therapy. However, the benefit of anticoagulation in patients of low-moderate risk category is based on individual risk factors. Among the non-sex factors, an age of 65 to 74 has the strongest effect on risk.^25,26

While risk assessment with CHA2DS2-VASc is useful because it provides a recommendation based on a snapshot of a patient’s history, it is not tailored to the patient’s current presentation and status. This case supports the utility of the D-dimer assay as an additional component of risk assessment in patients with AF who have low-moderate pretest probability risk.

The ARISTOTLE trial involved 14 878 individuals with AF treated with either warfarin or apixaban and assessed their D-dimer levels, clinical risk scores such as CHA2DS2-VASc, and adverse cardiovascular events. D-dimer levels were associated with higher risk of thromboembolic events, both for stroke and for other systemic emboli, as well as for all-cause death. More importantly, the study found that including D-dimer as a component of a modified CHA2DS2-VASc in risk assessment improved its predictive value for TE. Furthermore, on its own, the D-dimer value was determined to have a higher predictive value for cardiovascular and all-cause death than clinical risk scores. ²⁷ The patient’s risk of TE was low-moderate based on his history, but his clinical presentation and elevated D-dimer pointed to the etiology of his presentation as thromboembolism and subsequent renal infarct. As suggested in the ARISTOTLE study, this case highlights the utility D-dimer could contribute toward standard risk assessment scoring systems.

Conclusion

D-dimer is a simple blood test with a high negative predictive value to exclude a thrombus. In clinical practice, it is useful in ruling out DVT and PE. This case, along with other studies such as the ARISTOTLE trial, shows D-dimer’s high predictive value for thromboembolism. Furthermore, a positive D-dimer test may have an incremental value in risk stratification of patients with a low-moderate risk of TE as determined by traditional CHA2DS2-VASc score. D-dimer can easily be measured in various clinical settings, allowing physicians easy access, and may prove to be an excellent rule-out test for TE. As a biomarker, it is also highly individualized to the patient and provides a snapshot of the current condition of the patient. The D-dimer value has been shown to be a strong independent predictor of embolism and could provide additional information to standard clinical risk assessment scores. This test could be used by clinicians when TE is suspected to help clinicians provide better individualized care for patients with AF. Future studies including randomized control trials that determine the role of D-dimer in thromboembolic risk stratification in patients with AF should be performed.