Article, Hematology

Risk of venous thromboembolism in patients with borderline quantitative D-dimer levels

Original Contribution

Risk of venous thromboembolism in patients with borderline quantitative D-dimer levels

Taku Taira MD a, Breena R. Taira MD b,?, Matt Carmen MD b,

Jasmine Chohan BA b, Adam J. Singer MD b

aBellevue/NYU Medical Center New York, NY 10016, USA

bDepartment of Emergency Medicine, Stony Brook University Medical Center, Stony Brook, NY 11794-8350, USA

Received 5 September 2008; revised 23 December 2008; accepted 22 January 2009

Abstract

Objective: The lower threshold for D-dimer in evaluating patients with low clinical risk of venous thromboembolism ranges from 200 to 500 ng/mL. We compared the rates of VTE in patients based on D-dimer values. We hypothesized that the rate of VTE in low-risk patients with D-dimer levels less than 500 would be less than 1%.

Methods: Study Design: This was a retrospective chart review: Setting: The study was performed in a academic, suburban emergency department (ED). Subjects: Emergency department patients with suspected VTE and D-dimer obtained were included in the study. D-dimer assay: The D-dimer assay is a quantitative instrumentation latex suspension of plasma specimens. Outcomes: Presence of VTE within 30 days of ED visit. Data Analysis: Assuming a 0% event rate in patients with D-dimer levels between 200 and 500 ng/mL, a sample of 450 patients would result in a 95% confidence interval upper limit of 0.6%. Results: There were 1270 ED patients with suspected VTE in which D-dimer levels were performed between October 2005 and October 2006. Patient mean age was 47.8 +- 19.3 years; 63.2% were female, 78.2% were white. Of all D-dimer levels, 497 (39.1%) were less than 200 ng/mL, 479 (37.7%) were between 200 and 500 ug/mL, and 294 (23.1%) were greater than 500 ng/mL. There were no VTE events diagnosed in any of the patients with D-dimer levels less than 200 ng/mL. Four patients with D-dimer levels between 200 and 500 ug/mL had a pulmonary embolism on computed tomography angiography. Of these 4 patients, 3 had moderate clinical risk based on Well’s criteria and one had a false-positive computed tomography. There were no cases of VTE in the remaining 475 patients (0%; 95% confidence interval 0%-0.6%).

Conclusion: The rate of confirmed VTE in low-risk patients with D-dimer levels between 200 and 500 ng/mL is very low. Low-risk patients with suspected VTE with D-dimer levels less than 500 ng/mL might not require additional testing.

(C) 2010

Introduction

Background and importance

* Corresponding author. Tel.: +1 631 444 8351.

E-mail address: [email protected] (B.R. Taira).

Pulmonary embolism is a common and potentially lethal disease accounting for more than 250 000 hospitalizations a

0735-6757/$ – see front matter (C) 2010 doi:10.1016/j.ajem.2009.01.023

D-Dimer

year in the United States [1]. Pulmonary embolism has a variable presentation, and it continues to be a challenging

Subjects and D-dimer methodology

451

diagnosis for the emergency physician with more than 50% of cases undiagnosed [2]. Because of the potential for both morbidity and mortality, both low clinical suspicion and noninvasive studies such as electrocardiogram, arterial blood gas, and chest x-ray are insufficient for ruling out the disease [1,3,4]. In addition, radiologic studies have the drawbacks associated with ionizing radiation and iodinated contrast.

Clinical decision rules, such as the Wells’ criteria, have been used to rule out pulmonary embolism without radiologic studies. The Wells’ criteria combine an explicit algorithm to determine the pretest probability of a pulmonary embolism with a highly sensitive D-dimer assay to rule out the disease [5-7].

Another challenge is the D-dimer assay itself. The D- dimer is a highly sensitive but nonspecific test [8]. Normal D-dimer levels vary with age, sex, population, and the assay being used. These factors contribute to controversy regarding the accepted reference range [9-14]. Previous studies have used both 200 and 500 ng/mL as the threshold for a positive result [13,15]. Institutional reference ranges vary. Our institution uses 200 ng/mL as the threshold for a positive D-dimer.

If the lower threshold for an abnormal D-dimer were increased from 200 to 500ng/mL without changing the sensitivity, there would be decrease in the amount of diagnostic imaging and subsequent false-positive results.

Goals of this study

The goal of this study was to determine the rate of VTE in low-risk patients with an intermediate D-dimer between 200 and 500 ng/mL. We hypothesized that a change in the threshold for a positive D-dimer level from 200 to 500 ng/mL in low-risk patients would reduce the number of false positives and unnecessary additional testing while maintain- ing a false-negative rate for VTE of less than 1%. This rate was considered acceptable because even the gold standard, pulmonary angiogram, has a false-negative rate greater than 1% [6].

Methods

Study design

A retrospective, computerized database and chart review was performed to determine the rate of VTE in emergency department (ED) patients with suspected VTE.

Setting

The study was performed in an academic, suburban ED with an annual census of 75,000 visits per year in a medical center that has central laboratory testing for D-dimer.

Subjects included all ED patients with suspected VTE who had a D-dimer level obtained. The D-dimer assay is a quantitative instrumentation latex suspension of plasma specimens, the immunoturbidometric method, known com- mercially as HemosIL D-Dimer assay (Instrumentation Laboratories, Lexington, Mass). This assay is comparable to an enzyme-linked immunosorbent assay [16].

Measures

Standardized extraction of data from computerized records was used to obtain Demographic and clinical data, including D-dimer levels, the usage of a definitive test for VTE such as ultrasound or computed tomography (CT), and the presence or absence of VTE on the definitive test. Wells’ criteria [2] were calculated for each patient to determine the risk of VTE. Use of imaging (ultrasound or CT) was left to the discretion of the treating physician.

Outcomes

The primary outcome was the presence of VTE within

30 days of ED visit during which the D-dimer was performed. Chart review and/or telephone follow-up was performed for all patients with D-dimer levels between 200 and 500 ng/mL with VTE to verify the diagnosis and treatment of their VTE. We also determined whether any of the study patients were readmitted to our hospital for VTE within the subsequent 30 days.

Data analysis

Descriptive statistics were used to characterize the population and false-negative rates were calculated for the cutoff values of 200 and 500 ng/mL. Continuous data are summarized as means and 95% confidence intervals (CIs), and binomial data are summarized as the percent frequency of occurrence. Data analysis was performed using SPSS 15.0 for Windows (SPSS Inc, Chicago, Ill). Assuming a 0% event rate in patients with D-dimer levels between 200 and 500 ug/mL, a sample of 450 patients would result in a 95% CI upper limit of 0.6%.

Results

A total of 1270 patients had D-dimer testing between October 2005 and October 2006. The mean age was 47.8 +-

19.3 years. Of the patients, 63.2% were female and 78.2% were white. Four hundred ninety-seven (39.1%) patients had D-dimer values less than 200 ng/mL. Four hundred seventy- nine (37.7%) patients had D-dimer levels between 201 and 499 ng/mL, and 294 (23.1%) of patients had D-dimer level of more than 500 ng/mL.

452 T. Taira et al.

Table 1 Diagnostic test characteristics of various D-dimer levels

Cutoff at 500 ng/mL

(95% CI)

Cutoff at 200 ng/mL

(95% CI)

Sensitivity (%)

100 (80-100)

100 (80-100)

Specificity (%)

78.0 (75.6-80.3)

39.9 (37.1-42.7)

Negative predictive

100 (99.5-100)

100 (99.0-100)

value (%)

Positive predictive

6.8 (4.3-10.5)

2.6 (1.6-4.1)

value (%)

Of the 479 patients with D-dimer values between 201 and 499 ng/mL, 153 patients underwent CT angiography and 26 underwent lower extremity duplex ultrasonography. Four patients were diagnosed with venous thromboembolism. Of these 4 patients, 3 had a high pretest probability as determined by Wells’ criteria. The remaining patient was diagnosed with and treated for a pulmonary embolus. However, subsequent review of the CT angiogram by 2 board-certified radiologists specializing in chest CT angio- grams determined that the reading was a false positive. There were no cases of VTE diagnosed in the remaining 475 patients with a D-dimer level between 201 and 499 ng/mL. Thus, the false-negative rate of an intermediate D-dimer level between 201 and 499 ng/mL in low-risk patients was 0% (95% CI, 0.0%-0.6%). The test characteristics of the various D-dimer thresholds are noted in Table 1.

Discussion

A missed diagnosis of pulmonary embolism is associated with both morbidity and mortality. Because of the insuffi- ciency of low clinical suspicion [3] and noninvasive studies such as electrocardiogram, arterial blood gas, and chest x-ray and the inherent risks of radiologic studies, emergency physicians rely on clinical decision rules using the D-dimer test. Wells’ criteria are the most commonly used clinical decision rule. These criteria risk-stratify patients by the presence or absence of clinical signs of Deep vein thrombosis , lack of a likely alternative diagnosis, tachycardia, immobilization, previous DVT or pulmonary embolism (PE), hemoptysis, and malignancy [2]. In Wells’ study, the combination of a Low pretest probability and a highly sensitive D-dimer assay has good negative predictive value [2]. In our study, the HemosIL D-Dimer assay has test characteristics comparable to other assays [13]. The high sensitivity of the D- dimer confirms its utility as a screening test.

Our population had 4 patients with D-dimer values between 201 and 499 ng/mL and a diagnosis of venous thromboembolism. Of these patients, 3 had a high pretest probability as determined by Wells’ criteria and thus were inappropriate for diagnostic evaluation with D-dimer testing.

The remaining patient was a 39-year-old woman who presented with fever, cough, shortness of breath, and chest pain. The patient was at low risk by Wells’ criteria and we are not sure why a D-dimer test was obtained. The D-dimer level, however, was 257 ng/mL. The initial interpretation of her CT angiogram by the on-call radiology resident was a small alveolar infiltrate and there was no evidence of pulmonary embolism. The next morning the scan was re-read as positive for a subsegmental pulmonary embolism and the patient was called back, admitted, and anticoagulated with heparin and subsequently coumadin. One year later, 2 additional board- certified radiologists specializing in chest CT imaging subsequently reviewed this CT and both read the CT as no evidence of pulmonary embolism. As a result this patient was determined to have erroneously been diagnosed with PE. This case illustrates the potential danger of using a low threshold for diagnosing VTE. Although the patient did not have a serious adverse event, she was put at an unnecessary risk of bleeding and was admitted and treated for 6 months with anticoagulation unnecessarily.

Although the clinical algorithms for evaluating patients with suspected VTE were created to reduce testing, recent data show a paradoxical increase testing for pulmonary embolism with no change in the rate of positive imaging [17] The increase in imaging is may be due to physician’s sense of obligation to obtain a definitive test when the level is above a laboratory threshold despite a low clinical suspicion. In our population, the increase in laboratory threshold would have lead to a one third decrease in the amount of diagnostic imaging without an increase in false-negative results.

Our study demonstrates that in patients presenting with symptoms suspicious for pulmonary embolism, the threshold for a positive D-dimer can be safely increased from 200 to 500 ng/mL without altering the negative predictive value of the test. In our study population, an increase in reference range would have resulted in no instances of a falsely negative D-dimer while reducing the number of patients requiring subsequent testing by one third.

Finally, our findings are supported by the use of the 500 ng/mL threshold in major recent studies of PE. In a recent study of the effectiveness of a combination of clinical decision rule, D-dimer and CT, 500 ng/mL was used as the threshold for an abnormal value [18]. Many studies before this had been criticized for retrospectively choosing a threshold value that contributed to the variation in definitions of a “positive” D-dimer [13].

Limitations

Our study is limited in that it was a single institution study and follow-up was limited to 30 days after the ED visit. Furthermore, we did not review any Medical Examiner reports or death certificates for cases of patients who died of VTE. We also did not obtain follow-up on patients who did not subsequently visit our institution. We used only one

D-Dimer 453

particular type of D-dimer assay; and of those that had D- dimer values obtained, only one third of those with D-dimer values were subject to confirmatory testing. Finally, a single radiologist reviewed all radiologic studies.

Conclusion

In our study, no confirmed cases of venous thromboem- bolism in patients with low pretest probability and a D-dimer value between 201 and 499 were identified. Our findings suggest that low-risk patients with a D-dimer level less than 500 ng/mL may not require further evaluation. Raising the threshold for a positive D-dimer has the potential to reduce numbers of confirmatory radiologic studies obtained leading to a subsequent decrease in the cost and radiation exposure as well as the number of false positives.

References

  1. Goldhaber SZ. Pulmonary embolism. N Engl J Med 1998;339(2): 93-104.
  2. Wells PS, Anderson DR, Rodger M, et al. Excluding pulmonary embolism at the bedside without diagnostic imaging: management of patients with suspected pulmonary embolism presenting to the emergency department by using a simple clinical model and d-dimer. Ann of Inter Med 2001;135(2):98-107.
  3. Value of the Ventilation/perfusion scan in acute pulmonary embolism. Results of the prospective investigation of pulmonary embolism diagnosis (PIOPED). The PIOPED Investigators. JAMA 1990;263 (20):2753-9.
  4. Hampson NB. Pulmonary embolism: difficulties in the clinical diagnosis. Semin Respir Infect 1995;10(3):123-30.
  5. Wells PS, Anderson DR, Rodger M, et al. Derivation of a simple clinical model to categorize patients probability of pulmonary embolism: increasing the models utility with the SimpliRED D- dimer. Thromb Haemost 2000;83(3):416-20.
  6. Kruip MJ, Slob MJ, Schijen JH, van der Heul C, Buller HR. Use of a clinical decision rule in combination with D-dimer concentration in diagnostic workup of patients with suspected pulmonary embolism: a prospective management study. Arch Inter Med 2002;162(14):1631-5.
  7. Ten Wolde M, Hagen PJ, Macgillavry MR, et al. Non-invasive diagnostic work-up of patients with clinically suspected pulmonary embolism; results of a management study. J Thromb Haemost 2004;2 (7):1110-7.
  8. Brown MD, Rowe BH, Reeves MJ, Bermingham JM, Goldhaber SZ. The accuracy of the Enzyme-linked immunosorbent assay D-dimer test in the diagnosis of pulmonary embolism: a meta-analysis. Ann Emerg Med 2002;40(2):133-44.
  9. Gallagher E. Clots in the lung. Ann Emerg Med 2000;35(2):181-7.
  10. Mauron T, Baumgartner I, Z’Brun A, et al. SimpliRED D-dimer assay: comparability of capillary and citrated venous whole blood, between- assay variability, and performance of the test for exclusion of deep vein thrombosis in symptomatic outpatients. Thromb Haemost 1998;79(6): 1217-9.
  11. Becker DM, Philbrick JT, Bachhuber TL, Humphries JE. D-dimer testing and acute venous thromboembolism. A shortcut to accurate diagnosis? Arch Inter Med 1996;156(9):939-46.
  12. Kelly J, Rudd A, Lewis RR, Hunt BJ. Plasma D-dimers in the diagnosis of venous thromboembolism. Arch Inter Med 2002;162(7): 747-56.
  13. Heim SW, Schectman JM, Siadaty MS, Philbrick JT. D-dimer testing for deep venous thrombosis: a metaanalysis. Clin Chem 2004;50(7): 1136-47.
  14. Kline JA. New diagnostic tests for pulmonary embolism. Ann Emerg Med 2000;36(3):280-1.
  15. Farrell S, Hayes T, Shaw M. A negative SimpliRED D-dimer assay result does not exclude the diagnosis of deep vein thrombosis or pulmonary embolus in emergency department patients. Ann Emerg Med 2000;35(2):121-5.
  16. Moerloose P, Vanrusselt M, Reber G, Arnout J. Performances of the HemosIL D-dimer HS assay for the exclusion of venous thromboem- bolism. J Thromb Haemost 2005;3(10):2361-3.
  17. Kabrhel C, Matts C, McNamara M, Katz J, Ptak T. A highly sensitive ELISA D-dimer increases testing but not diagnosis of pulmonary embolism. Acad Emerg Med 2006;13(5):519-24.
  18. Christopher Study Investigators. Effectiveness of managing suspected pulmonary embolism using an algorithm combining clinical prob- ability, D-dimer testing, and computed tomography. JAMA 2006;295 (2):172-9.