Comparison of the diagnostic performance of the original and modified Wells score in inpatients and outpatients with suspected deep vein thrombosis

doi:10.1016/j.thromres.2011.02.008

Thrombosis Research

Volume 127, Issue 6, June 2011, Pages 535-539

https://doi.org/10.1016/j.thromres.2011.02.008 Get rights and content

Abstract

Introduction

The original and modified Wells score are widely used prediction rules for pre-test probability assessment of deep vein thrombosis (DVT). The objective of this study was to compare the predictive performance of both Wells scores in unselected patients with clinical suspicion of DVT.

Methods

Consecutive inpatients and outpatients with a clinical suspicion of DVT were prospectively enrolled. Pre-test DVT probability (low/intermediate/high) was determined using both scores. Patients with a non-high probability based on the original Wells score underwent D-dimers measurement. Patients with D-dimers < 500 μg/L did not undergo further testing, and treatment was withheld. All others underwent complete lower limb compression ultrasound, and those diagnosed with DVT were anticoagulated. The primary study outcome was objectively confirmed symptomatic venous thromboembolism within 3 months of enrollment.

Results

298 patients with suspected DVT were included. Of these, 82 (27.5%) had DVT, and 46 of them were proximal. Compared to the modified score, the original Wells score classified a higher proportion of patients as low-risk (53 vs 48%; p < 0.01) and a lower proportion as high-risk (17 vs 15%; p = 0.02); the prevalence of proximal DVT in each category was similar with both scores (7-8% low, 16-19% intermediate, 36-37% high). The area under the receiver operating characteristic curve regarding proximal DVT detection was similar for both scores, but they both performed poorly in predicting isolated distal DVT and DVT in inpatients.

Conclusion

The study demonstrates that both Wells scores perform equally well in proximal DVT pre-test probability prediction. Neither score appears to be particularly useful in hospitalized patients and those with isolated distal DVT.

Introduction

Deep vein thrombosis (DVT) is a common problem in ambulatory and hospitalized patients. Untreated DVT may lead to potentially fatal pulmonary embolism (PE). On the other hand, unjustified anticoagulation therapy poses a risk for bleeding [1]. Correct diagnosis and prompt treatment are therefore crucial. Unfortunately, symptoms and signs of DVT are unspecific. Less than 25% of patients with clinically suspected DVT do actually have the disease [2], [3], underscoring the importance of accurate diagnostic strategies. Several clinical prediction rules have been developed and validated in various populations to simplify and improve the diagnostic process of patients with suspected DVT [4], [5], [6], [7], [8], [9]. Diagnostic strategies based on combining pretest probability with D-dimers measurements have been shown to be safe and cost-effective [10], leading to a significant reduction of ultrasound examinations [5], [11], [12].

The best validated prediction rule is the Wells score, consisting of nine clinical items (Table 1) [6]. This score has been developed in ambulatory patients addressed to a tertiary care center for a suspected first episode of proximal and distal lower limb DVT [6], [13], [14]. It has been subsequently validated in the emergency department, and the hospital setting [4], [11], [15]. More recently, Wells and colleagues published a modified score, adding an item for previously documented DVT [5]. This modified Wells score has been validated in outpatients [5], and emergency department patients only [8]. The aim of this prospective cohort study was to compare the accuracy of the two scores in predicting proximal and isolated distal DVT in a broad, unselected population of ambulatory and hospitalized patients with suspected DVT.

Section snippets

Patients

Consecutive patients, ≥ 18 years of age, with clinically suspected lower limbs DVT addressed to the thrombosis consultation of the vascular medicine service of a teaching hospital were potentially eligible for the present study. Outpatients presenting to the emergency department or directly referred by their general practitioner and inpatients referred by their physician in charge were eligible.

Exclusion criteria were as follows: clinical suspicion of PE, treatment with therapeutic

Results

Between May 4, 2007 and May 31, 2009, 447 consecutive patients with clinically suspected DVT of the lower limbs were screened. 149 patients were excluded because at least 1 of the predefined exclusion criteria was present (clinical suspicion of PE: 39; treatment with therapeutic anticoagulation for more than 48 hours prior to inclusion: 31; planned long-term anticoagulation for a diagnosis other than VTE: 22; pregnancy: 17; life expectancy less than 3 months: 5; unwillingness or inability to

Discussion

Our results demonstrate that original and modified Wells scores are similarly accurate in determining pre-test probability in patients with clinical suspicion of lower limb DVT. While both scores performed well in patients with proximal DVT, the discriminatory power in patients with isolated distal DVT was much lower. Both scores performed rather poorly in inpatients.

Because the modified Wells score comprises an additional variable (previously documented DVT), it classified significantly more

Conflict of interest statement

No conflict of interest to declare.

Acknowledgements

The authors wish to thank Monique Salvi for her excellent technical assistance as well as all the physicians of the division of angiology for performing the ultrasound examinations. This study was financially supported by a grant from Cardiomet.

References (27)

P.S. Wells et al.
Value of assessment of pretest probability of deep-vein thrombosis in clinical management
Lancet
(1997)
J. Constans et al.
Comparison of four clinical prediction scores for the diagnosis of lower limb deep venous thrombosis in outpatients
Am J Med
(2003)
R.M. Subramaniam et al.
Diagnosis of lower limb deep venous thrombosis in emergency department patients: performance of Hamilton and modified Wells scores
Ann Emerg Med
(2006)
N. Perone et al.
Comparison of four strategies for diagnosing deep vein thrombosis: a cost-effectiveness analysis
Am J Med
(2001)
D.R. Anderson et al.
Combined use of clinical assessment and d-dimer to improve the management of patients presenting to the emergency department with suspected deep vein thrombosis (the EDITED Study)
J Thromb Haemost
(2003)
P.S. Wells et al.
Accuracy of clinical assessment of deep-vein thrombosis
Lancet
(1995)
C. Ambid-Lacombe et al.
Excellent performances of Wells' score and of the modified Wells' score for the diagnosis of proximal or distal deep venous thrombosis in outpatients or inpatients at Toulouse University Hospital: TVP-PREDICT study
J Mal Vasc
(2009)
W. Blattler et al.
Diagnosis of deep venous thrombosis and alternative diseases in symptomatic outpatients
Eur J Intern Med
(2004)
S.M. Schellong
Distal DVT: worth diagnosing? Yes
J Thromb Haemost
(2007)
J. Cornuz et al.
Clinical prediction of deep venous thrombosis using two risk assessment methods in combination with rapid quantitative D-dimer testing
Am J Med
(2002)

J. Hirsh et al.

Management of deep vein thrombosis and pulmonary embolism. A statement for healthcare professionals. Council on Thrombosis (in consultation with the Council on Cardiovascular Radiology), American Heart Association

Circulation

(1996)

S. Goodacre et al.

Meta-analysis: The value of clinical assessment in the diagnosis of deep venous thrombosis

Ann Intern Med

(2005)

P.S. Wells et al.

Does this patient have deep vein thrombosis?

JAMA

(2006)

Cited by (32)

Pulmonary embolism and deep vein thrombosis: Similar but different
2021, Thrombosis Research
Citation Excerpt :
In a study comparing PE and DVT, PE patients more commonly were older females and had less often prior VTE, recent surgery, or cancer but more often chronic lung disease, heart failure, or renal insufficiency [1]. Both symptoms and clinical signs of DVT and PE are nonspecific [25]. Only approximately 25% of the patients with clinical signs or symptoms of DVT, actually do have the disease [25].
Pulmonary embolism (PE) and deep vein thrombosis (DVT), the two clinical manifestations of venous thromboembolism (VTE), constitute a major global burden of cardiovascular disease. They are often referred to as one disease but several patient characteristics, risk factors, real-world treatment, and clinical outcomes may differ substantially between PE and DVT alone.
We conducted a narrative review of the state-of-the-art literature on the topic of PE and DVT alone using PubMed, Google scholar, and MEDLINE databases and the most established international consensus statement guidelines for the management of VTE, focusing on the recommendations for diagnosis and treatment but also including epidemiological and clinical characteristics of VTE, highlighting similarities and differences between PE and DVT alone.
Several patient characteristics, risk factors, clinical manifestations, and outcomes differ substantially between PE versus DVT alone. Nevertheless, recommendations for both diagnosis and treatment are strikingly similar in the current guidelines for the management of DVT and PE, except for the indication for advanced reperfusion therapies.
The differences in risk factors, clinical manifestations, and clinical outcomes between patients with PE versus DVT alone are only marginally addressed in the current consensus guidelines. More data is needed allowing proposal of evidence-based adjustments in the diagnostic and therapeutic strategies for these two manifestations of VTE. Tailored risk stratification and individualized management strategies for patients with PE and DVT alone may lead to a better prognosis, less recurrence and complications, and possibly to a gain of quality-adjusted life years in patients with VTE.
Literature review of distal deep vein thrombosis
2021, Journal of Vascular Surgery: Venous and Lymphatic Disorders
Although distal deep vein thrombosis (DDVT) has been more frequently diagnosed with the availability of better ultrasound imaging quality, the data on the best method to manage DDVT have been conflicting. The aim of the present review was to summarize the current and evidence-based recommendations for the diagnosis and management of DDVT and to provide a summary of the most recent societal guideline recommendations.
A literature review of DDVT was performed. The PubMed databases were queried for articles on the epidemiology, risk factors, diagnosis, and management of DDVT.
The prevalence of isolated DDVT has been reported in a broad range. The reported risk factors include older age, active malignancy, a low degree of mobility, acute infection, and atrial fibrillation. With more evidence, anticoagulation therapy was found to be associated with a reduced risk of recurrent venous thromboembolism (VTE) and/or thrombus propagation compared with conservative management. However, anticoagulation was associated with an increased risk of bleeding in a number of studies. The rate of VTE recurrence ranged from 7% to 23% during a follow-up period ranging from 3 months to 8 years. The significant risk factors for VTE recurrence included cancer, older age, an unprovoked event, and inpatient status.
Few studies have addressed the diagnosis and management of DDVT. Further research is needed to standardize the best approach to diagnose and treat DDVT.
Diagnosis of deep vein thrombosis
2020, Revista Clinica Espanola
La trombosis venosa profunda (TVP) forma parte del espectro clínico de la enfermedad tromboembólica venosa (ETV), cuya incidencia anual se estima en 1-2 episodios por 1.000 personas y constituye la tercera causa de mortalidad cardiovascular en los países desarrollados.
Establecer un diagnóstico certero de la TVP es primordial para evitar complicaciones agudas, como la embolia pulmonar, y crónicas asociadas al síndrome postrombótico. Actualmente están bien establecidos los algoritmos diagnósticos para la TVP de las extremidades inferiores. Estos incluyen modelos de probabilidad clínica, que permiten establecer el riesgo de presentar la enfermedad con base en los antecedentes del paciente y en hallazgos clínicos, la determinación del dímero D, producto de degradación de la fibrina con un elevado valor predictivo negativo, y pruebas de imagen para confirmar el diagnóstico.
La ecografía venosa por compresión es la técnica de elección en la actualidad, al tratarse de una prueba no invasiva de fácil realización y que permite evaluaciones seriadas. Hay mucha evidencia acumulada de que en pacientes con baja o intermedia probabilidad clínica y un dímero D negativo (< 500 ng/ml) se puede excluir de forma segura el diagnóstico de TVP sin exploraciones adicionales. No hay un consenso tan claro sobre la necesidad de una exploración proximal o completa de toda la extremidad. Otras técnicas también pueden ser utilizadas, como la venografía por resonancia magnética o la tomografía axial computarizada en fase venosa, aunque estas no deben sustituir a la ecografía de compresión como primera prueba diagnóstica. Hay otras circunstancias especiales en las que el diagnóstico es más problemático y no existen algoritmos diagnósticos tan consolidados, como la TVP durante el embarazo, el diagnóstico de una retrombosis o la TVP que afecta a las extremidades superiores.
Deep vein thrombosis (DVT) is part of the clinical spectrum of venous thromboembolism disease (VTE), whose estimated annual incidence rate is 1-2 episodes per 1000 individuals and represents the third leading cause of cardiovascular mortality in developed countries.
Establishing an accurate diagnosis of DVT is essential for preventing acute complications (such as pulmonary embolism) and chronic complications associated with post-thrombotic syndrome. Currently, there are well-established diagnostic algorithms for lower extremity DVT, which include clinical probability models that help establish the risk of experiencing the disease based on the patients’ history, clinical findings, D dimer measurements, fibrin degradation product tests with a high negative predictive value and imaging tests to confirm the diagnosis.
Venous compression ultrasonography is currently the technique of choice because it is a non-invasive, easy-to-administer test that can make serial evaluations. There is much accumulated evidence that DVT can be safely ruled out in patients with a low or intermediate clinical probability and a negative D dimer (<500 ng/mL) without performing additional examinations. The consensus is not as clear about the need for a proximal or complete examination of the entire extremity. Other techniques may also be employed, such as magnetic resonance venography and venous phase computed axial tomography, although these should not be a substitute for compression ultrasonography as the initial diagnostic test. There are other special circumstances in which the diagnosis is more problematic and there are no diagnostic algorithms as consolidated, such as DVT during pregnancy, diagnosing rethrombosis and DVT that affects the upper extremities.
Diagnosis of deep-vein thrombosis
2018, Thrombosis Research
Citation Excerpt :
Accordingly, primary care guidelines for the management of patients with suspected DVT endorse the use of the Wells rule [10]. In hospitalized patients, the Wells rule performs poorly, since the probability of DVT in low-risk patients is disappointingly high, ranging between 6% and 12% (Table 3) [15,18]. Similarly, in patients with isolated distal DVT the rule displays unsatisfactory sensitivity, as the probability of DVT in low-risk patients ranges between 8% and 14% [13,15,17].
Cost-effectiveness of the deep vein thrombosis diagnosis process in primary care
2016, Atencion Primaria
Analizar el coste-efectividad de la aplicación de algoritmos diagnósticos en pacientes con sospecha de un primer episodio de trombosis venosa profunda (TVP) en la atención primaria, en comparación con la derivación sistemática a centros especializados.
Estudio observacional, transversal y analítico.
Pacientes seleccionados en urgencias hospitalarias derivados desde la atención primaria para completar estudio y diagnóstico.
Se reclutó a 138 candidatos con clínica compatible con un primer episodio de TVP; 22 fueron excluidos (sin informe de derivación, clínica de más de 30 días, anticoagulados y TVP previa), incluyéndose finalmente a 116 pacientes, un 61% mujeres, de 71 años edad media.
Variables de las escalas de probabilidad clínica de Wells y Oudega, dímero-D (portátil y hospitalario), ecografía-Doppler y costes directos generados por los 3 circuitos analizados: derivación sistemática de todos los pacientes, derivación según escala de Oudega o de Wells.
En el 18,9% se confirmó el diagnóstico de TVP. Las 2 escalas de probabilidad clínica presentaron una sensibilidad del 100% (IC del 95%: 85,1-100) y una especificidad alrededor del 40%. Con la aplicación de las escalas, se hubiesen podido evitar con total seguridad un tercio de las derivaciones a urgencias hospitalarias (p < 0,001) y se hubieran podido disminuir los costes del proceso diagnóstico en 8.620 € según Oudega y 9.741 € según Wells, por cada 100 pacientes atendidos.
La aplicación de algoritmos diagnósticos en las sospechas de TVP permitiría al médico de atención primaria una orientación más resolutiva y coste-efectiva del proceso diagnóstico.
To analyse the cost effectiveness of the application of diagnostic algorithms in patients with a first episode of suspected deep vein thrombosis (DVT) in Primary Care compared with systematic referral to specialised centres.
Observational, cross-sectional, analytical study.
Patients from hospital emergency rooms referred from Primary Care to complete clinical evaluation and diagnosis.
A total of 138 patients with symptoms of a first episode of DVT were recruited; 22 were excluded (no Primary Care report, symptoms for more than 30 days, anticoagulant treatment, and previous DVT). Of the 116 patients finally included, 61% women and the mean age was 71 years.
Variables from the Wells and Oudega clinical probability scales, D-dimer (portable and hospital), Doppler ultrasound, and direct costs generated by the three algorithms analysed: all patients were referred systematically, referral according to Wells and Oudega scale.
DVT was confirmed in 18.9%. The two clinical probability scales showed a sensitivity of 100% (95% CI: 85.1 to 100) and a specificity of about 40%. With the application of the scales, one third of all referrals to hospital emergency rooms could have been avoided (P<.001). The diagnostic cost could have been reduced by € 8,620 according to Oudega and € 9,741 according to Wells, per 100 patients visited.
The application of diagnostic algorithms when a DVT is suspected could lead to better diagnostic management by physicians, and a more cost effective process.
Clinical factors that influence the pre-test diagnostic probability of deep vein thrombosis in out-patients
2015, Angiologia
Los signos y síntomas de la trombosis venosa profunda (TVP) en las extremidades suelen ser inespecíficos y a menudo influidos por la comorbilidad del paciente.
Análisis de variables clínicas de los pacientes con sospecha clínica de un primer episodio de TVP en Atención Primaria.
Estudio observacional transversal y analítico de pacientes ambulatorios con sospecha de un primer episodio de TVP en extremidades inferiores. Se correlacionan los factores de riesgo y comorbilidades con la confirmación diagnóstica de TVP mediante ecografía-doppler.
Se estudió a 116 pacientes, de los cuales en 22 (18,9%) se confirmó el diagnóstico. Las comorbilidades más frecuentes fueron la hipertensión, la hipercolesterolemia y la diabetes tipo 2. Solo la presencia de un diagnóstico alternativo probable y la inmovilización prolongada se asociaron con la exclusión o el diagnóstico final de TVP, respectivamente. Las enfermedades del aparato locomotor conllevaron una mayor probabilidad de falsa sospecha de TVP y el único signo clínico que se correlacionó con el diagnóstico fue la tumefacción de la extremidad afecta.
En la práctica clínica, la tumefacción o el antecedente de parálisis de la extremidad nos ha de hacer sospechar firmemente una TVP. Sin embargo, la presencia de un posible diagnóstico alternativo prácticamente descartará dicha enfermedad.
Deep vein thrombosis (DVT) has unspecific signs and symptoms that are often influenced by the comorbidity of the patient.
To analyze the clinical features and the associated comorbidity of patients with symptoms compatible with a first DVT in a Primary Care setting.
Observational, cross-sectional and analytical study of patients with a first episode of suspected DVT. The risk factors and comorbidities are correlated with the DVT diagnostic confirmation by ultrasound.
A total of 116 patients were included, of whom 22 had DVT (18,9%) confirmed by ultrasound. The most frequent associated comorbidities were hypertension, hypercholesterolemia, and type 2 diabetes mellitus. Only the presence of an alternative diagnosis more likely than DVT and the lower extremity immobilization were associated with the exclusion or confirmation of the diagnosis, respectively. Musculoskeletal diseases were correlated with a greater probability of a false suspicion of DVT. The only sign associated with the final diagnosis was the swelling of lower limbs.
In clinical practice, swelling or a history of limb paralysis should lead to the suspicion of DVT. However, the presence of a possible alternative diagnosis could rule out this disease.

View all citing articles on Scopus

View full text

Regular ArticleComparison of the diagnostic performance of the original and modified Wells score in inpatients and outpatients with suspected deep vein thrombosis

Abstract

Introduction

Methods

Results

Conclusion

Introduction

Section snippets

Patients

Results

Discussion

Conflict of interest statement

Acknowledgements

Lancet

Am J Med

Ann Emerg Med

Am J Med

J Thromb Haemost

Lancet

J Mal Vasc

Eur J Intern Med

J Thromb Haemost

Am J Med

Management of deep vein thrombosis and pulmonary embolism. A statement for healthcare professionals. Council on Thrombosis (in consultation with the Council on Cardiovascular Radiology), American Heart Association

Circulation

Meta-analysis: The value of clinical assessment in the diagnosis of deep venous thrombosis

Ann Intern Med

Does this patient have deep vein thrombosis?

JAMA

Regular Article
Comparison of the diagnostic performance of the original and modified Wells score in inpatients and outpatients with suspected deep vein thrombosis