Evaluation of D-dimer using VIDAS assay in the diagnosis of suspected deep vein thrombosis

Evaluation of D-dimer using VIDAS assay in the diagnosis of suspected deep vein thrombosis

Ayman Omar,MD; Nehad Zaid,MD Department of General Surgery,Vascular  Surgical Unit,

Menoufyia University Hospitals (Medical School), Shebin El-Kom, Egypt.

Abstract

Background: Accurate diagnosis of deep vein thrombosis minimizes  the risk of thrombo­ embolic  complications and averts the exposure of patients without thrombosis to the risks of anticoagulant  therapy. Recent studies have shown the safety of using D-dimer assay in ruling out suspected DVT.

Objective:  Evaluate the usefulness  of using D-dimer (VIDAS)  assay  compared  to duplex ultrasonography in  the  early   diagnosis of  suspected DVT  in  Egyptian sample size.

Patients and methods: One hundred patients with suspected deep vein thrombosis have got assessment  with (i) Wells' clinical probability  guide for DVT diagnosis, (ii) VIDAS D-dimer essay & (iii) duplex scanning. The results of D-dimer testing & duplex scanning were evaluated

& compared on the basis of patients' clinical features.

Results: The validity of D-dimer testing in diagnosing likely patients in relation to duplex ultrasound  showed  that D-dimer  test has reached  100% sensitivity,  91% specificity  & 99% accuracy in comparison to duplex study. Alternatively for unlikely patiens; D-dimer test showed

100%  sensitivity, 89%  specificity & 92%  accuracy in  comparison to  duplex study.

Conclusion: The incorporation of information  gathered from Wells' clinical score, duplex scanning &  D-dimer assay should be  helpful in  the  diagnosis of  suspected DVT.

Key words: Deep vein thrombosis, D-dimer assay, duplex ultrasonography.

Introduction:

The clinical assessment of patients  with suspected  DVT is often difficult  because  of the interplay between risk  factors  and  the nonspecific nature  of the physical  findings. Discordance is  often  present  between the clinical assessment and the results of objective testing as duplex scanning.

Many studies have confirmed the diagnostic sensitivity and  specificity of duplex ultrasonography for proximal vein thrombosis. Sensitivity of  duplex  ultrasonography for proximal vein DVT is 97%; but only 73% for calf vein. The negative predictive value (NPV) for proximal vein  DVT  is 99%. Overall specificity  is 95%.1 Duplex ultrasonography is also helpful to differentiate venous thrombosis from  hematoma, Baker cyst, abscess, and other causes ofleg pain and edema. However; duplex scanning might face difficulty

in diagnosis of calf  DVT,  venous  thrombi proximal to inguinal ligament or differentiating between old and new clots in recurrent DVT.2

The Wells clinical prediction guide (1997) quantifies the probability of DVT in patients into high-, moderate-, or low-risk categories.3

Combining  this with the results of objective

testing greatly simplifies the clinical workup of patients with suspected DVT. The Wells clinical prediction guide  incorporates risk factors, clinical signs,  and  the presence or absence of alternative diagnoses. The parameter of Wells Clinical  Score for DVT included:

-Active cancer (treatment ongoing, or within

6 mo or palliative) +1.

- Paralysis or recent plaster immobilization of the lower extremities +1.

- Recently bedridden for >3 d or major surgery

<4wk+l.

- Localized tenderness along the distribution

of the deep venous system +1.

- Entire leg swelling +1.

- Calf  swelling >3  em  compared with  the asymptomatic leg +1.

- Pitting edema (greater  in the symptomatic leg) +1.

- Previous DVT documented +1.

-Collateral superficial veins (Non-varicose)

+1.

• High probability >3

• Moderate probability 1 or 2

• Low probability <0

A score of two or higher indicates that the probability  of DVT is likely; a score of less than two indicates that probability of DVT is unlikely.4

Recent interest  has focused on the use of D-dimer in the diagnostic approach to DVT. D-dimer is a specific fibrin degradation product which results from the digestion of cross-linked fibrin  by  plasmin. Monoclonal antibodies specific for the D-dimer fragment are used to differentiate fibrin-specific clot from  non­ cross-linked fibrin and from fibrinogen. These specific attributes of the D-dimer antibodies account for their high sensitivity  for venous

thromboembolism.s

Many different D-dimer assays are available,

with varying sensitivities and specificities. Traditional enzyme-linked immunosorbent assays (ELISAs), although accurate, are time­ consuming and not practical for use. A rapid ELISA assay (VIDAS)  with high sensitivity was  validated in  a large  European trial.6

The aim of the present study is to evaluate the usefulness of using D-dimer (VIDAS) assay compared  to duplex  ultrasonography in the

diagnosis of suspected DVT in a sample of

Egyptian patients.

Patients and methods:

One hundred patients with suspected deep vein thrombosis and attending the Emergency Department  or Surgical Out-patient  clinic at Menoufyia University Hospitals from  July

2009  till July  2011 were  included into  the study. All patients underwent:

(i) Full clinical assessment whereby WellsAf

model of clinical probability scoring system for  DVT  diagnosis was considered (mentioned above).

(ii) D-dimer  laboratory test was done before heparin therapy  using  VIDAS  D-dimer essay ((Bio-Merieaux SA, Marcy-Etoile, France). It usually takes 35 minutes. The cutoff  value  of VIDAS was  500  ng/ml where  D-dimer  level 500  ng/ml  was considered positive for DVT.

(iii) Duplex scanning.

The results  of D-dimer  testing  & duplex scanning were evaluated & compared on the basis  of   patients'  clinical  features.

Results:

The study was conducted  on 100 patients presented with  suspected DVT.  Males comprised 36 patients wile females comprised

64 with mean age 46.26±13.96 (range 20-80 years). The majority of patients (96 patients) presented with suspected lower limbs? DVT (66 left lower limbs versus 30 right ones). Only

4 patients presented with suspected  DVT in their left upper limbs.These findings are shown in Table(l).

Table (1): Demography of studied patients (n = 100).

Seventy-six patients (76%) were considered likely  to have DVT whenever applying  the Wells  clinical  prediction guide (score2)

versus 24 patients (24%) were considered as unlikely to  have  DVT. The  percentage distribution of probability is shown inTable(2).

Table (2): The percentage distribution  of probability (n = 100).

Duplex scanning was positive in 74 patients (68 patients have  got  proximal deep  veins below the inguinal ligament while 6 patients have got isolated  calf veins thrombosis). 26 patients  were  free of thrombosis on duplex study  on  their  first  presentation. On  the contrary; 72 patients were positive and 28 were negative by testing with D-dimer VIDAS essay.

The impact ofWells' criteria on the findings of duplex scanning or D-dimer test  was equivocal. A little difference emerges between

the results of either objective test. Inthe likely patients; D-dimer test was positive  in 66/76 (86.8%) patients.Inthe unlikely patients; 8/24 (33.3%) patients were positive for the test. On the other hand; Duplex was positive in 65/76 (85.5%) patients in  the  likely group. Furthermore; duplex was positive in6/24 (25%) of the unlikely patients. The incorporation  of D-dimer test and Wells'  criteria is shown in Table(3).

Table (3): D-dimer test in likely and unlikely patients (n = 100).

*Mann Whitney-U test

The relationship between D-dimer & duplex in diagnosing suspected DVT was interesting. In the likely group (n = 76); 66 patients were positive by using D-dimer test while 65 patients were positive to duplex study. Subsequently;

10 patients were negative to D-dimer and 11 patients were negative  to duplex testing. On

the other hand; in the unlikely group (n = 24);

8 patients  were positive  to D-dimer  while 6 patients were positive to duplex. Also; in the unlikely group: 60 patients were negative to D-dimer test while 18 patients were negative to duplex study. These findings were summarized in Table(4).

Table (4): D-dimer versus duplex in the likely and unlikely patients.

The  validity of D-dimer testing in diagnosing likely patients in relation to duplex ultrasound showed that  D-dimer test  has reached 100% sensitivity, 91% specificity &

99% accuracy in comparison to duplex study. Furthermore; the  positive predictive value

estimated 98%, while the negative predictive value was 100%. Alternatively for unlikely patients; D-dimer test showed 100% sensitivity,

89% specificity & 92% accuracy in comparison

to duplex study. These findings were summarized in Table(5).

Discussion:

Suspected deep-vein thrombosis is a common condition, with a lifetime cumulative incidence of2-5 % .7 Current evidence suggests that patients  with clinically suspected DVT and a normal venous ultrasound result should have a repeat ultrasound examination at 1 week interval to safely exclude DVT and continue without anticoagulation.8  The incorporation of clinical features & duplex scanning could be  insufficient in  approaching successful diagnosis. Recently developed tests for plasma levels ofD-dimer, a fibrin degradation product, have  shown  high  sensitivity and  moderate specificity in diagnosing clinically suspected DVT.  Highly  sensitive tests  are  generally helpful in ruling out the presence of disease. The present study was conducted in order to evaluate the  usefulness of  using  D-dimer laboratory tests in the diagnosis of suspected DVT.9

The demography of patients included in the study does not seem to be different from that mentioned in literatures world-wide.The males to females ratio accounts for 1:1.7. This little increase in the females' incidence can  be attributed to the small sample size of the study patients. Also; DVT usually affects individuals older than 40 years.lO In our study the mean age of patients included was 46 years (range:

20-80)  years. This may be explained on the basis of the  interchange of habits & subsequently risk factors among different races.

All patients in the present study (n =100)

were evaluated for suspected DVT with duplex scanning and D-dimer assays. We have chosen VIDAS test ofD-dimer assays because of rapid

& accurate results. The  results of  both

evaluations were compared to each other on the  basis of  patients' clinical features.

All patients were  applied  to the clinical probability scoring system of Wells (mentioned earlier) where 76 patients (76%) were likely to develop DVT and 24 patients (24%) were unlikely to develop DVT.

In the likely  group: D-dimer  was +vein

66176 (86.8%) patients while duplex study was

+ve  for  DVT  in  65176  (85.5%) patients. Subsequently; D-dimer was-ve in 10176 (13.2%) patients, while duplex  was -ve for DVT in 11176 (14.5%) patients.

The results of using D-dimer test in the likely group showed an increase in its validation in the diagnosis of DVT compared to duplex study  where  D-dimer has reached 100% sensitivity and 91% specificity with accuracy reaching 99%.

In unlikely  group:  D-dimer test was +ve

in 8/24 (33.3%) patients, while duplex study was +vein 6/24 patients (25%). Subsequently; D-dimer  was-vein 16/24 (66.7%) patients, while duplex was -ve for DVT in 18/24 (75.5%) patients.

The results of using D-dimer test in the unlikely group  showed an increase in  its validation in the diagnosis ofDVT compared to duplex study where D-dimer  has reached

100%  sensitivity and 89% specificity with accuracy reaching 92%.

Data  gathered from  this study  strongly supports the use of a D-dimer assay in the clinical algorithm of  suspected DVT.  A negative D-dimer assay results rules out DVT inunlikely patients with low-to-moderate risk (Wells' score <2). Also; a negative result also obviates the need for surveillance  and serial testing in likely patients with moderate-to-high risk whenever associated negative ultrasonographic  findings.

Furthermore; the usage of D-dimer  assay