D-Dimer: a Role in Ruling out Pulmonary Embolism in an Emergency Care Department

Emergency Care Deparment, Primary Healthcare Centre Banja Luka, the Republic of Srpska, Bosnia and Herzegovina. Centre for Biomedical Research, Faculty of Medicine, University of Banja Luka, Banja Luka, the Republic of Srpska, Bosnia and Herzegovina. Primary Healthcare Centre Modriča, Modriča, the Republic of Srpska, Bosnia and Herzegovina. Department of Anaesthesiology and Intensive Therapy, Institute for Cardiovascular Diseases “Dedinje“, Belgrade, Serbia.


Introduction
Methods PE is one of the most urgent conditions in medicine. The success of treatment largely depends on timely diagnosis. PE primarily needs to be thought of, and early diagnosis and treatment of patients is possible if PE is suspected on the basis of detailed anamnestic data. Many risk factors can raise suspicion about PE, however it can happen without any of the predisposing factors (up to 30 %). 9 Clinical manifestation of PE is also nonspecific, so dyspnoea, as the most common symptom/sign of PE, occurs in only 50 % of the clinically confirmed PEs. 10,11 A clinical probability for PE is estimated on the basis of clinical presentation by using the Wells clinical decision rule or the revised Geneva score. [12][13][14][15][16] The electrocardiographic (ECG) changes are miscellaneous and usually nonspecific (most common is sinus tachycardia, followed by the right bundle branch block -RBBB, turning of heart axis to the right, SI QIII TIII, P pulmonale, simultaneous inversion of T waves in inferior and right precordial leads), but in 18 % of patients ECG is normal. 17,18 Nonspecific changes or lack of them can also be found during the physical check-up, chest X-ray and laboratory tests. 7,19,20 Gold standard for the diagnosis of PE is computed tomography (CT)-pulmonary angiography and/or ventilation-perfusion (VQ) scan. 21,22 D-dimer is a fibrin degradation product, a small protein fragment present in the blood after fibrinolysis. Role of D-dimer is to exclude the PE diagnosis. 23-25 D-dimer plasma levels are elevated in patients with acute thrombosis because of the simultaneous activation of coagulation and fibrinolysis. 26 Negative predictive D-dimer value is high (95-98 %) and PE with normal values of D-dimer is unlikely. On the other hand, fibrin is produced in other conditions including cancer, inflammation, bleeding, trauma, necrosis and surgical intervention. [27][28][29] Therefore, positive predictive D-dimer value is low and elevated D-dimer le-vels are not useful in confirmation of PE. 30,31 This implies recommendations that the patients with highly suspected PE should immediately undergo CT-pulmonary angiography and skip the D-dimer testing. 9,32,33 The aim of this study was to determine to what extent D-dimer values help to differentiate PE in emergency care departments and therefore help in making a decision to transfer patients to a higher referential level or not. Furthermore, the aim was to determine whether the D-dimer values correlated with the values of revised Geneva score.
The protocol of the study was approved by the Ethics Committee of the Primary Medical Centre Banja Luka and all the efforts were undertaken in order to keep the anonymity of the included patients.
Following the protocol of the Emergency Care Department of the City of Banja Luka, data were found for all the patients in whom the D-dimer was tested in 2018. Gender has been recorded (female/male), age, symptoms, D-dimer level (mg/L), working diagnosis, and data whether the patient was referred to the hospital treatment or not. Patients' anonymity was preserved, while only gender and age were recorded, excluding other personal data.
The D-dimer values have been analysed by a quantitative, latex-enhanced immunoturbidimetric immunoassay on Cobas h-232 system. Although D-dimer has only one cut-off value, the authors were interested in finding to what extent a D-dimer value affected the physician's decision, ie, if there was a difference whether a D-dimer value was slightly elevated or if it was four times or more higher than referential value. Therefore, a value of D-dimer was recorded in two ways, as a numeric value and as a value in one of three categories: < 0.5 mg/L, ≥ 0.5-2 mg/L and ≥ 2 mg/L. Using the anamnestic data, symptoms, clinical signs, age and gender of the patient, for each patient the revised Geneva score was calculated. The parameters set for Geneva score are listed in

Results
Sixty-eight tests were done in the Emergency Care Department of the City of Banja Luk a in 2018. Of 68 tests, 41 were negative (60.29 %), meaning that it was less than 0.5 mg/L; from 0.5 up to 2 mg/L there were 19 tests (13.23 %), and over 2 mg/L there were 8 tests (11.76 %). Patients' characteristics are shown in Table 2.
The mean age of the patients was 60.3 (± 17.4) years, with 45.5 % patients being older than 65. Levels of D-dimer were statistically significantly higher in elderly people (Kruskal-Wallis test: χ 2 = 12.96, p = 0.002). Also, by analysing the ratio of positive and negative values of the

Discussion
In 2018, 68 D-dimer analyses were performed at the Emergency Care Department of the City of Banja Luka. The assumption is that the need for analyses was greater, but unfortunately analyses are not continuously available. For this reason, it was decided to analyse only patients who had done the D-dimer test, because if all patients who were suspected on PE or with dyspnoea had been analysed, an unrealistically low percentage of patients in whom D-dimer was determined would had been obtained.
Two thirds of the tests were negative (< 0.5 mg/L). A high percentage of patients who are D-dimer-negative is logical, since the D-dimer is primarily used to rule out the diagnosis of PE and deep vein thrombosis. 30 The D-dimer was significantly higher in elderly patients, which is in accordance with the results of other studies. There are also recommendations to adjust the cut-off value of the D-dimer according to age by adding 0.1 to 0.5 mg/L for every 10 years of age for people over 50. [34][35][36] This is thought to increase the specificity of the D-dimer in the elderly. [37][38][39] Clinical studies indicate that the D-dimer values are slightly higher in women, which is in accordance with the results of this study. As was the case with the results of other researchers, the difference was not significant in the present study either. The difference in D-dimer values in men and women is considered to be of no clinical significance and it is not recommended to correct the cut-off values based on gender. 39 However, some researchers believe that D-dimer values should be adjusted for both gender and age, with multiple cut-off values, in order to significantly improve the specificity of D-dimer testing. 40 In contrast to the D-dimer values, PE itself is slightly more common in men. 41 The results of this study show that women were slightly more frequently diagnosed with PE (30 % : 20 % of patients), which could be explained by the fact that no cut-off corrections were made for women and therefore there was a higher percentage of false-positive tests in female patients.
The Wells score and the Geneva score were introduced in an attempt to adequately suspect or exclude PE based on history, symptoms and clinical presentation. 42 Clinical trials indicate that the significance of the revised Geneva score is primarily in the exclusion of PE combined with low D-dimer values. 43,44 The results of this study indicate that the revised Geneva score values did not influence physicians' decision not to refer such patients for further treatment, but only the D-dimer values. Analysis from other studies also implies underuse of clinical decision rules. 45 There was a correlation between the revised Geneva score and the D-dimer, but it was not statistically significant. However, given a relatively small sample, it would be assumed that with a sufficiently large sample statistical significance would be reached. It is also the authors' belief that when deciding whether to refer a patient for further hospital treatment or not, physicians should consider the revised Geneva score, especially if its value is < 3. 46, 47 Intermediate-risk PE patients present a diagnostic and therapeutic dilemma. 48 Besides, research results indicate that in older, high-risk patients, the Wells scores are more in correlation with the diagnosis of PE than the revised Geneva score. [49][50][51] The D-dimer values significantly aided the physicians' decision to refer the patient to further hospital treatment under the diagnosis of PE. When D-dimer values were < 0.5 mg/L, twothirds of patients were not referred for further hospital treatment and the others were referred under some other diagnosis. Two patients, despite the negative D-dimer values, were referred under the diagnosis of PE and this is, in fact, an example of poor clinical practice. The reason is that if the low D-dimer value does not alter the physician's opinion that it is PE or not, the D-dimer should not been tested at all. On the other hand, half of the patients with D-dimer values > 0.5 and all patients with values > 2 mg/L were referred for further treatment and diagnosis.

Conclusion
The very purpose of the D-dimer and revised Geneva score is to exclude PE and reduce unnecessary imaging diagnostic procedures, such as CT-pulmonary angiography and similar. The fact that the test results are available after 20 minutes should also not be overlooked, given that the speed of diagnosis and patient treatment is one of the key factors in working in an emergency care department.