Application of Wells Criteria, in Combination With Serum D-dimer to Rule Out Deep Vein Thrombosis in Lower Extremities

1 Department of General and Vascular Surgery, Shohada-e-Tajrish Hospital, Shahid Beheshti University of Medical Sciences, Tehran, IR Iran
*Corresponding author: Mohammad Reza Sobhiyeh, Department of General and Vascular Surgery, Shohada-e-Tajrish Hospital, Shahid Beheshti University of Medical Sciences, Tehran, IR Iran, Tel:+98-21721144, Fax:+98-21721144, E-mail:
Scimetr. 2(1): e14770 , DOI: 10.5812/scimetr.14770
Article Type: Research Article; Received: Sep 10, 2013; Revised: Nov 1, 2013; Accepted: Dec 6, 2013; epub: Jan 7, 2013; collection: Jan 2013


Background: Wells et al. developed a clinical prediction rule to estimate the probability of deep venous thrombosis (DVT), using data of secondary care outpatients.

Objectives: Our goal is to validate the diagnostic accuracy of this rule, in combination with D-dimer test.

Patients and Methods: A cross-sectional study was performed with prospective data collection from September 2010 to December 2011. Data was collected in emergency department settings in Shohada-e Tajrish hospital, Tehran. One hundred and seventy seven consecutive patients presented with suggestive symptoms of DVT, were included. All patients underwent historytaking and physical examinations to measure the Wells rule score and D- dimer test by the agglutinin method. Leg ultrasonography was the reference gold standard to determine the presence of DVT.

Results: 14.75% of patients in the low-risk group had DVT; when it was combined with negative D-dimer test results, the Wells rule yielded a 4% prevalence of DVT in the low-risk group. The estimated Positive Predicted Value for patients in the high-risk group, based on the Wells criteria in combination with D-dimer testing, was 79.24%, while Negative Predicted Value for patients in low- risk group was calculated as 85.24%.

Conclusions: The present study aimed to introduce a cost-effective, reliable, and available modality to recognize patients afflicted with DVT. The Wells rule, alone or in combination with D-dimer testing, does not guarantee accurate assessment of risk in primary care patients in whom DVT is suspected. Using Wells criteria in low -risk group may decrease the need for ultrasonography in up to 28% of patients, and may be cost beneficial.

Keywords: Venous Thrombosis; Fibrin Fragment D; Ultrasonography, Doppler

1. Background

Deep vein thrombosis (DVT) is one of the important causes of mortality and morbidity, especially in hospitalized patients (1). The most hazardous complication of DVT is pulmonary embolism, resulting in 50 to 200 thousand deaths every year in the US. In prolonged cases of DVT, complications such as venous insufficiency, impose an economic burden on the health care system, which can cause worrying issues. Therefore, prompt diagnosis and treatment of DVT : involves having high significance and cost effectiveness (24). Long before the emergence of the new imaging techniques, diagnosis of DVT was solely based on historytaking and physical examinations. However, imminent progress in the imaging techniques in the recent years has nevertheless made them a widely used diagnostic tool even in cases whose diagnosis is clinically evident as well as in cases with a trivial diagnostic suspicion. Therefore, given the high prevalence of DVT, it does not seem reasonable to rely on expensive imaging methods (5). P.S. Wells was one of the firsts who tried to diagnose DVT independently of imaging techniques by introducing Wells clinical criteria to diagnose acute lower limb DVT, and thereafter adding to their accuracy by including the serum D-dimer test (6). Defining the clinical criteria for diagnosis of DVT is an important issue in the field of venous disease, and many researches are being focused on this topic. Mr. P.S. Wells first introduced his Criteria (Table 1) in 1995. It has been revised two times, but it is still the most accepted criteria (Table 1) (7). Not many research projects have been performed regarding the application of these criteria (7). Establishing the accuracy of these criteria may help to reduce the need for imaging.

Table 1.

Wells Clinical Criteria for Predicting the Pretest Clinical Probability of Deep Venous Thrombosis

2. Objectives

Therefore, the aim of this study was to determine the value of Wells criteria in combination with D-dimer test for the diagnosis of DVT.

3. Patients and Methods

We designed a cross sectional study for 15 months, between September 2010 and December 2011. All the patients presenting to the emergency department of Shohada-e Tajrish hospital with clinical highly suggestive symptoms of acute lower limb DVT (lower limb pain, new-onset edema), were included. The followings were the exclusion criteria of the study: patients below 18 years old, patients with documented positive history of previous DVT or pulmonary emboli (PE), patients undergoing prolonged anticoagulant therapy, and those with chronic DVT. Likewise, patients managed on an outpatient basis for DVT and were candidates for elective venous Doppler ultrasonography were also excluded. All patients underwent historytaking and physical examinations in E.D by a general surgery resident under the supervision of a vascular surgeon. All patients were evaluated for the presence or absence of standard Wells criteria (Table 1) by an experienced vascular surgeon in a double blind manner regarding the results of D-dimer test and color Doppler ultrasonography, and the results were recorded in a special form designed for this purpose.

Patients were divided into three groups based on their risk score interpretation as the following:

High risk: 3 points or above

Moderate risk: 1 to 2 points

Low risk: 0

D-dimer test was performed in all the patients who used the agglutination method. All patients then underwent venous Doppler ultrasonography as the reference gold standard test to determine the presence or absence of DVT. Venous Doppler ultrasonography was performed by an experienced radiologist in a double blind manner regarding the results of D-dimer testing and patients risk classification. The Ethics Committee of Shahid Beheshti University of Medical Sciences and Health Services approved the study (Date 16.4.2009, NO: 89).

3.1. Statistical Analysis

The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were calculated regarding Wells criteria alone and in combination with D-dimer test to predict the absence or presence of DVT (SPSS software, version 12.0 for Windows, SPSS, Inc., Chicago, Illinois).

4. Results

One hundred and ninety three consecutive patients with suggestive symptoms of DVT were included. One hundred and seventy of them presented with suggestive symptoms of DVT and entered the study, and 16 patients were excluded. Venous Doppler ultrasonography as our standard reference test, confirmed the presence of DVT in 67 of 177 patients (37.58%), as the group one, and ruled out DVT in 110 patients (62.14%), as the group two. The mean age for patients in the groups one and two were 53.23 years (49 ± 3.07) and 50.32 years (50 ± 4.2), respectively. Of the 177 patients, 84 (47.45%) were males and 93(52.54%) were females. Among those diagnosed with DVT, 30 (44.77%) were males and 37 (55.22%) were females. Wells criteria score was calculated in patients with and without DVT (Table 2).

Table 2.

Frequency of Wells Criteria

Patients were classified into three groups based on DVT risk assessment by Wells criteria: 53 patients in high- risk, 63 patients in medium- risk , and 61 patients in low- risk groups. Of the 53 patients in the high-risk group, 39 (73.58%) had DVT according to the standard reference test, while in the medium -risk and low- risk groups, 19 (30.15%) and 9 (14.75%) patients were diagnosed with DVT, respectively. Based on the above findings, the PPV for the high-risk group was 73.58%, and NPV for the low risk group was 85.24 %. On the other hand, D-dimer test accuracy was estimated using agglutination method for detection of DVT. Of the 67 patients diagnosed with DVT, 59 had positive results for D-dimer test, while among the 110 patients in whom DVT diagnosis was not confirmed, 17 had positive results for D- dimer test. Overall, the D-dimer test sensitivity in diagnosing DVT was calculated as 88.05% with a specificity of 84.54%, while PPV and NPV were 77.63% and 92.07%, respectively. The D-dimer results were assessed separately in the both groups, based on the Wells risk assessment. The results are shown in Table 3.

Table 3.

Frequency of DVT Cases Based on Groups and D-dimer Test Results

The sensitivity, specificity, NPV, and PPV of Wells criteria were calculated in combination with D-dimer results for each of the three groups with presence of DVT. The PPV for the high-risk group with positive D-dimer test results was 85.71%, and the NPV for the low- risk group with negative D-dimer test results was 96%.

5. Discussion

We assessed the validity of Wells criteria alone and in combination with D-dimer test, in diagnosing DVT. The aim of this study was to find out whether we could rule out DVT without imaging techniques, which would be much less costly. We grouped the patients according to their Wells criteria score, and the results showed that 74% of patients in the high-risk group, 30% in the medium risk group, and 15% in the low- risk group were afflicted with DVT. While in the Wells study it was 75%, 17%, and 3%, respectively (8) as shown in Table 4 (911).

Table 4.

Frequency of DVT Based on Wells Criteria in Different Studies

Based on our findings, it can be concluded that Wells criteria alone is not a reliable diagnostic tool to assess the presence or absence of DVT. The above conclusion will become more significant by considering pulmonary embolism as a major complication of DVT. The number of patients in Wells study was more compared to this study (593 patients), and the most patients in their study were in the low- risk group and overall, only 16% of the patients included were diagnosed with DVT. The difference in the results could be due to the fact that Shohada-e Tajrish hospital is a vascular surgery referral center and the probability of DVT is higher in patients. Nevertheless, relying only on Wells criteria to rule out DVT in low risk groups without performing any further tests does not seem to be a logical option. However, when Wells criteria are combined with D-dimer test, the accuracy of the method will increase significantly. Of 61 patients in the low- risk group, 9 were diagnosed with DVT, and only 2 of them had a negative D-dimer test result. The NPV for negative D-dimer test result in the low -risk group was 96% (95% CI: 85.14%-99.30%), while in the Wells study was more than 99%, and he proposed that in patients categorized as the lowrisk group, D-dimer test should be performed, and in patients whose D-dimer test is also negative, DVT can be ruled out without performing further imaging (12).

The NPV of D- dimer test in the medium- risk group in our study was 92.5% (95% CI: 78.52%-98.04%). While D-dimer testing resulted in exclusion of 50 patients (28.24% of all patients) in the low-risk group from further ultrasonography, it is still significant, that in our study, four patients who had DVT would be missed. The reason behind this difference in the two studies could be due to sample size difference, or the accuracy of the laboratory method used for D-dimer testing. In the Wells study, D-dimer test was performed by ELISA method, while in our study it was performed by agglutination method. The findings in other studies are shown in Tables 5 and 6 (1213).

Table 5.

Results of Various Studies on Acute Lower Limb DVT

Table 6.

Missed Patients With DVT Based on Wells Criteria in Different Studies

Wells proposed that all patients in the low- risk group should undergo the D-dimer test, and those with negative results should not undergo any further investigations, but those with positive results should undergo a venous Doppler ultrasonography. He also suggested that for all patients in the medium risk group, D-dimer test should be performed, and if it had a positive result, the patient should undergo venous Doppler ultrasonography. If the test result was negative, further decision should be made according to the method of D-dimer test, as if D-dimer test was performed by ELISA method, there would be no need for further investigation, and if the test was performed by agglutination method, venous Doppler ultrasonography would be recommended. He also recommended performing a venous Doppler ultrasonography in the high-risk group (12). Although we reported a striking 28% reduction in unnecessary ultrasonographies, the 4% probability of missing DVT in low- risk patients has made this recommended approach both logical and cost- beneficial.

Therefore, in spite of low accuracy of the Wells criteria, we recommend the following approach, in which all the suggestive patients of acute lower limb DVT should be primarily evaluated and classified based on the Wells criteria, and those in the high and medium- risk groups for DVT should undergo a venous Doppler ultrasonography. in the low -risk group, performing a D-dimer test should be proceeded. The necessity of performing an ultrasonography is absolutely determined by the positive result of D-dimer test. As for the patients in the moderate- risk group, if D-dimer test is performed by ELISA method, negative results will suggest the termination of diagnostic process. Moreover, with the significant NPV of D-dimer test by agglutination method, in cases with strong clinical suspicion, when the initial sonography has negative results, and the patient is categorized as medium or low- risk; the next diagnostic step will be performing a D-dimer test instead of costly serial sonographies. Among our study limitations, the sample size in each group should be considered. Moreover, performing D-dimer test by ELISA method was not available. Patients, who were included in the study, were only selected from those admitted to the emergency department. Based on the above limitations, further studies are required using ELISA method for D-dimer testing with larger sample sizes and possibly recruiting patients from wider clinical settings who can fulfill the inclusion criteria.


This article was entirely derived from an under- graduate thesis successfully completed by Dr. Shahabedin under supervision of Dr. Mozafar, and with cooperation of the Vascular and Trauma Research Center, Shahid Beheshti University of Medical Science, Tehran, IR Iran.


Implication for health policy/practice/research/medical education The Wells rule, alone or in combination with D-dimer test, does not guarantee accurate assessment of risk in primary care patients in whom DVT is suspected, but as using Wells criteria in low- risk group, may decrease the need for Doppler ultrasound in up to 28% of all patients suspected to have DVT, it might be cost-benefit. The present study aims to introduce a cost-effective, reliable and available modality to discover patients afflicted with Deep Vein Thrombosis.
Authors’ Contribution: Dr. Mohammad Ali Shahabodin, Dr. Mohammad Reza Sobhiyeh, Dr. Saran Lotfollahzadeh, Dr. Mohamad Ali Kalantar Motamedi who performed the study, collected data, performed the statistical analysis, and prepared the manuscript. Dr. Mohammad Mozafar who supervised the study, and participated in designing and conducting the study and manuscript preparation.
Financial Disclosure: None declared.
Funding/Support: None declared.


  • 1. Mohr DN, Silverstein MD, Heit JA, Petterson TM, O’Fallon WM, Melton LJ. The venous stasis syndrome after deep venous thrombosis or pulmonary embolism: a population-based study. Mayo Clin Proc. 2000;75(12):1249-56. [PubMed]
  • 2. Ouellette DavidW, Patocka Catherine. Pulmonary Embolism. Emergency medicine clinics of North America. 2012;30(2):329-375.
  • 3. Moheimani F, Jackson DE. Venous thromboembolism: classification, risk factors, diagnosis, and management. ISRN Hematol. 2011;2011:124610. [DOI] [PubMed]
  • 4. Kunisawa S, Ikai H, Imanaka Y. Incidence and prevention of postoperative venous thromboembolism: are they meaningful quality indicators in Japanese health care settings? World J Surg. 2012;36(2):280-6. [DOI] [PubMed]
  • 5. Beckman JA. Cardiology patient page. Diseases of the veins. Circulation. 2002;106(17):2170-2. [PubMed]
  • 6. Merli G. Diagnostic assessment of deep vein thrombosis and pulmonary embolism. Am J Med. 2005;118 Suppl 8A:3S-12S. [DOI] [PubMed]
  • 7. Wells PS, Anderson DR, Rodger M, Forgie M, Kearon C, Dreyer J, et al. Evaluation of D-dimer in the diagnosis of suspected deep-vein thrombosis. N Engl J Med. 2003;349(13):1227-35. [DOI] [PubMed]
  • 8. Wells PS, Anderson DR, Bormanis J, Guy F, Mitchell M, Gray L, et al. Application of a diagnostic clinical model for the management of hospitalized patients with suspected deep-vein thrombosis. Thromb Haemost. 1999;81(4):493-7. [PubMed]
  • 9. Dryjski M, O’Brien-Irr MS, Harris LM, Hassett J, Janicke D. Evaluation of a screening protocol to exclude the diagnosis of deep venous thrombosis among emergency department patients. J Vasc Surg. 2001;34(6):1010-5. [DOI] [PubMed]
  • 10. Schutgens RE, Ackermark P, Haas FJ, Nieuwenhuis HK, Peltenburg HG, Pijlman AH, et al. Combination of a normal D-dimer concentration and a non-high pretest clinical probability score is a safe strategy to exclude deep venous thrombosis. Circulation. 2003;107(4):593-7. [PubMed]
  • 11. Bucek RA, Koca N, Reiter M, Haumer M, Zontsich T, Minar E. Algorithms for the diagnosis of deep-vein thrombosis in patients with low clinical pretest probability. Thromb Res. 2002;105(1):43-7. [PubMed]
  • 12. Anderson DR, Wells PS, Stiell I, MacLeod B, Simms M, Gray L, et al. Management of patients with suspected deep vein thrombosis in the emergency department: combining use of a clinical diagnosis model with D-dimer testing. J Emerg Med. 2000;19(3):225-30. [PubMed]
  • 13. Oudega R, Hoes AW, Moons KG. The Wells rule does not adequately rule out deep venous thrombosis in primary care patients. Ann Intern Med. 2005;143(2):100-7. [PubMed]