The use of D-dimer in specific clinical conditions: A narrative review

European Journal of Internal Medicine 20 (2009) 441–446 Contents lists available at ScienceDirect European Journal of Internal Medicine j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e j i m Review article The use of D-dimer in specific clinical conditions: A narrative review E. Bruinstroop a,1, M.A. van de Ree a, M.V. Huisman b,⁎ a b Diakonessenhuis, Department of Internal Medicine, Bosboomstraat 1, 3582 KE Utrecht, The Netherlands Leiden University Medical Center, Section of Vascular Medicine, Department of General Internal Medicine–Endocrinology, Room c4-68, PO Box 9600, 2300 RC Leiden, The Netherlands a r t i c l e i n f o Article history: Received 5 August 2008 Received in revised form 5 November 2008 Accepted 18 December 2008 Available online 24 January 2009 Keywords: Algorithms Diagnosis Fibrin fibrinogen degradation products Pulmonary embolism Sensitivity and specificity Venous thrombosis a b s t r a c t The use of D-dimer in combination with a clinical decision rule has been widely investigated in pulmonary embolism and deep venous thrombosis. Although it has been shown to be safe in excluding venous thromboembolism, the clinician is often faced with specific situations in which the use of D-dimer is controversial. We review the best available evidence on these patients. We conclude that it is not safe to use D-dimer testing in patients with symptoms of a venous thromboembolism for over 14 days, patients receiving therapeutic heparin treatment and patients with suspected deep venous thrombosis during oral anticoagulant therapy. In these populations the levels of D-dimer can be lower then expected giving rise to false-negative results. It is safe to use D-dimer testing in combination with a clinical decision rule in patients of all ages, patients presenting with a suspected recurrent venous thromboembolism or inpatients with suspected pulmonary embolism. As patients with recurrent venous thromboembolism, elderly patients and inpatients have higher levels of D-dimer, D-dimer testing has a low specificity and the need for additional radiological testing is increased. © 2008 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved. 1. Introduction D-dimer testing has been widely investigated and results show that, when combined with a clinical decision rule (CDR), this test is an effective diagnostic tool to rule out deep vein thrombosis (DVT) as well as pulmonary embolism (PE) [1,2]. The use of D-dimer testing has resulted in a significantly reduced need for ultrasonography for DVT and ventilationperfusion (VQ) lung scan or CT scan in patients with suspected PE [1,2]. In the general population, quantitative D-dimer tests are highly sensitive, ranging from 94 to 99%. However, in several clinical scenarios D-dimer testing is controversial. Firstly, for ruling out venous thromboembolism (VTE) when symptoms have been present for several days before coming to clinical attention or in patients on anticoagulation. Secondly, for safely excluding recurrent VTE. Recent research has also focused on the use of D-dimer in elderly patients and inpatients. This review discusses the performance of D-dimer tests in these different clinical scenarios. The methods for this guideline are in accordance with the system of Harbour and Miller [3]. All articles on the key clinical question were revised and judged for a level of evidence. The level of evidence is determined by the study type and the methodological quality. All studies with an acceptable standard, relevant to the key clinical question were revised. Finally for every key clinical question a grade of recommendation was given derived from the levels of evidence of the different studies, the degree of extra- ⁎ Corresponding author. Tel: +31 71 5262085; fax: +31 71 5248140. E-mail addresses: e.bruinstroop@amc.uva.nl (E. Bruinstroop), m.v.huisman@lumc.nl (M.V. Huisman). 1 Present address: Department of Endocrinology and Metabolism, Academic Medical Center, University of Amsterdam, 1100 DD Amsterdam, the Netherlands. polation required forming the recommendation, the applicability of the data to the target patient group and the consistency of the data (Table 1). 2. Time between onset of symptoms and diagnosis Clinical scenario: A 48 year old female presents to the emergency department with calf pain. She reports that the pain has started 10 days ago, and has not decreased since. The D-dimer value is below cut-off and the clinical decision rule concludes DVT is unlikely. Could the level of D-dimer have dropped below cut-off due the late presentation? Variable delays between symptom onset and D-dimer testing have been allowed in the major studies of the diagnostic algorithm for VTE. For DVT, patients have been included ranging from 3 days up to 43 days [4]. Wells et al. [1] included patients with a mean duration of symptoms of 7.8 days, without specifying the maximum. For PE, van Belle et al. [2] reported a mean duration of symptoms of 2 days (IQR: 1–5 days), but there was no upper time limit for inclusion. Few studies have investigated the relation between duration of symptoms and D-dimer sensitivity (Table 2). D'Angelo et al. [5] studied a total of 103 consecutive patients with symptoms ranging from 1 to 15 days. One patient with confirmed DVT involving two venous segments with symptoms lasting from 15 days had a normal level of D-dimer (220 ng/mL). The median levels of Ddimer in patients with confirmed DVT decreased significantly over time, from 8760 ng/mL (2990–10,800) in patients with duration of symptoms less than three days compared to 2140 ng/mL (220–4780) in patients with duration of symptoms between seven and fifteen days. de Bastos et al. [6] retrospectively evaluated the duration of symptoms in 335 adults with suspected DVT or PE. One hundred 0953-6205/$ – see front matter © 2008 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.ejim.2008.12.004 442 E. Bruinstroop et al. / European Journal of Internal Medicine 20 (2009) 441–446 Table 1 Grades of recommendation, adjusted from Harbour and Miller [3]. Table 3 Characteristics of studies on measurement of D-dimer after start of heparin treatment. Grade of Recommendation Study Confirmed VTE (%) Test (cut-off) Speiser et al. [9] 13 DVT (100) Minnema et al. [10] 28 DVT/PE (100) Amelsberg et al. [11] 14 DVT (100) DVTENOX study group [12] 93 DVT (100) Meissner et al. [13] Peternel et al. [14] 61 DVT (100) 59 DVT (100) Kuruvilla et al. [15] 104 DVT (87) 16 PE (13) ELISA Dimertest Agen, New Jersey, USA (quantitative) EIA Dimertest Agen, Australia (quantitative) ELISA Boehringer, Germany (quantitative) ELISA Asserachrom, Diagnostica Stago, France (quantitative) ELISA (400 ng/mL) TintElize D-dimer, Biopool (100 ng/mL) Amelung coagulometer, sigma diagnostics (200 ng/mL) A B C D At least one high quality meta-analysis, systematic review, or RCT rated as with a very low risk of bias and directly applicable to the target population or A systematic review of RCTs or a body of evidence consisting principally of well conducted meta-analyses, systematic reviews of RCTs, or RCTs with a low risk of bias, directly applicable to the target population and demonstrating overall consistency of results. A body of evidence including high quality systematic reviews of case–control or cohort studies or high quality case–control or cohort studies with a very low risk of confounding, bias, or chance and a high probability that the relationship is causal, directly applicable to the target population and demonstrating overall consistency of results or Extrapolated evidence from high quality or well conducted metaanalyses, systematic reviews of RCTs, or RCTs with a low to very low risk of bias. A body of evidence including well conducted case–control or cohort studies with a low risk of confounding, bias, or chance and a moderate probability that the relationship is causal directly applicable to the target population and demonstrating overall consistency of results or Extrapolated evidence from high quality systematic reviews of case–control or cohort studies or high quality case–control or cohort studies with a very low risk of confounding, bias, or chance and a high probability that the relationship is causal. Case–control or cohort studies with a high risk of confounding, bias, or chance and a significant risk that the relationship is not causal or Non-analytic studies, e.g. case reports, case series or Expert opinions or Extrapolated evidence from well conducted case–control or cohort studies with a low risk of confounding, bias, or chance and a moderate probability that the relationship is causal percent sensitivity for D-dimer testing was obtained when symptom duration was fifteen days or less. In patients with symptoms for more than fifteen days, four patients had a VTE. Two of these patients had normal levels of D-dimer with symptom duration of 30 and 60 days. Together, these small studies underscore the recommendation of the British Committee for Standards in Haematology [7], that D-dimer testing should be used with caution in patients presenting with symptoms for over 14 days. In these patients further diagnostic imaging should be done to exclude VTE (Grade of Recommendation: D). 3. Measurement of D-dimer after start of heparin treatment Clinical scenario: A 70 year old patient presents at the emergency department with symptoms of calf pain, which started yesterday evening. Based on the clinical suspicion, the primary care physician has already given a therapeutic dose of low-molecular weight heparin. Should one proceed directly to diagnostic imaging? Several small studies have evaluated the effect of therapeutic heparin, on D-dimer levels (Table 3). Couturaud et al. [8] reviewed three articles on the relation between heparin treatment and the decline of D-dimer levels after one day. In all three studies unfractionated heparin was given at a therapeutic dose and D-dimer was measured before and after one day of Table 2 Characteristics of studies on time between onset of symptoms and diagnosis. Study Patients (n) Confirmed VTE (%) Test (cut-off) D'Angelo et al. [5] 103 22 DVT (21) de Bastos et al. [6] 335 10 DVT (3) 19 PE (6) VIDAS D-dimer, bio Merieux, France (quantitative) VIDAS D-dimer, bio Merieux, France (500 ng/mL) treatment. Two studies with a total of 41 patients with DVT or PE [9,10] showed a decline in D-dimer levels after intravenous heparin of 30% and 40% after one day. In Amelsberg et al. [11] 14 patients with DVT received subcutaneous heparin and the value of D-dimer increased 10% after one day. The DVTENOX study group [12] monitored D-dimer levels at presentation and on the following days after subcutaneous Enoxaparin or continuous intravenous unfractionated heparin treatment. In the whole population (n = 93) there was a sharp decrease measured on the second day (median −2500 ng/mL) with a continuous decline of D-dimer levels on days six and ten. There was no significant difference between the two treatment groups. The mean value of D-dimer remained elevated after ten days (1800 ± 300 ng/mL). In Meissner et al. [13] D-dimer also showed a rapid decrease after initial presentation independent of initial treatment (unfractionated heparin, low-molecular-weight heparin, warfarin or no treatment). However, on the third and seventh day D-dimer levels were still elevated in respectively 94% and 92% of the 61 patients with DVT. In Peternel et al. [14] 59 consecutive patients were randomized to receive unfractionated heparin (UH) or low-molecular-weight heparin (LMWH). Compared to the day when heparin therapy was started, evaluations on days three and seven showed a non-significant decrease of D-dimer levels in both groups, but all determinations remained above the cut-off value. No significant differences were observed between UH and LMWH groups. In general, these studies show that in most patients the levels of D-dimer remain elevated during the first days of treatment. However, there is an initial rapid decrease in D-dimer, which can cause the value of D-dimer to drop below cut-off. This is likely due to a rapid reduction in thrombin activity and subsequent generation of new fibrin [9]. Thereafter, the slower decline of D-dimer may result in a persistently elevated level of D-dimer, which can last for months after diagnosis of VTE [13,15]. The results were independent of the therapy used. We conclude that when therapeutic heparin therapy has been started prior to diagnostic testing with the current available evidence it is mandatory to perform radiological tests since the combination of a low clinical suspicion with a normal D-dimer test cannot safely rule out venous thromboembolism (Grade of recommendation: D). 4. Measurement of D-dimer during oral anticoagulation Clinical scenario: A 72 year old woman is using fenprocourmon because of atrial fibrillation. Ten days ago she had a serious knee injury and has been immobilized since then. Her left lower leg has been swollen and painful for one day, and you think of a deep venous thrombosis as a possible cause. Can you use D-dimer testing even though she is using a coumarin derivative? Traditionally, patients receiving oral anticoagulation have been excluded from the major diagnostic studies. In such patients, a reduced capacity for thrombin generation can be expected, with possibly a lower level of D-dimer. It was shown that patients treated with OAT for a VTE have a significantly lower level of D-dimer compared to untreated patients after discontinuation of OAT [16]. 443 E. Bruinstroop et al. / European Journal of Internal Medicine 20 (2009) 441–446 However, patients present with symptoms of venous thromboembolism while receiving oral anticoagulant therapy (OAT). One study has specifically investigated this population (Table 4). Aguilar et al. [17] investigated the usefulness of D-dimer testing in combination with a CDR in 70 patients with suspected DVT on OAT. Four patients with normal D-dimer levels at presentation were diagnosed with confirmed DVT during follow-up. This resulted in a lower sensitivity (69.2%; 95% CI: 42.2–87.3) than in the normal population. We conclude from this small study, that it is not safe to use Ddimer testing to exclude deep venous thrombosis in patients already on OAT. These patients should always undergo compression ultrasound (Grade of recommendation: D). Table 5 Characteristics of studies on diagnosis of recurrent venous thromboembolism. Study Patients (n) Confirmed VTE (%) Test (cut-off) Rathbun et al. [20] 300 55 DVT (18) Aguilar et al. [21] 105 47 DVT (45) Le Gal et al. [22] 308 124 PE (40) Söhne et al. [23] 480 129 PE (27) Nijkeuter et al. [24] 234 65 PE (28) 5. Diagnosis of recurrent venous thromboembolism Clinical scenario: A 30 year old female presents at the emergency department with symptoms of a pulmonary embolism. She had one previous deep venous thrombosis one year ago and is not currently using OAT. Can you safely exclude PE in this patient with a normal D-dimer? D-dimer has been reported to be elevated in a large amount of the patients one month after completion of a six month anticoagulation drug course [18]. Therefore, it is controversial whether patients with suspected recurrence of VTE can be diagnosed with the same algorithm as a first VTE [19]. Several studies have been performed on PE and DVT recurrence (Table 5). Wells et al. [1] have included a variable of ‘a previously documented DVT’ in their clinical model for predicting the pretest probability of deep vein thrombosis. The combination of this clinical model in combination with D-dimer testing has shown to successfully omit patients from ultrasound testing. Rathbun et al. [20] prospectively studied 300 patients with suspected recurrent DVT. They investigated whether the STA-Liatest D-dimer (Diagnostica Stage) could be used as a single test, without a clinical decision rule, to exclude recurrent VTE. One of the patients with a normal D-dimer at presentation had a confirmed VTE during follow-up. However, it is unclear if this represents the actual amount of false-negatives as VTE could not be excluded in 6 patients. Aguilar et al. [21] enrolled 105 outpatients with suspected recurrent DVT and measured D-dimer levels in combination with clinical probability. D-dimer was normal in 17.1% of the cases, and combined with a low clinical probability, a hundred percent sensitivity was reached. Van Belle et al. [2] have investigated the use of a clinical decision rule in combination with D-dimer testing in the exclusion of PE. In the clinical decision rule, the variable ‘previous pulmonary embolism or deep vein thrombosis’ is included. This diagnostic management strategy is associated with low risk of subsequent fatal and nonfatal VTE. Le Gal et al. [22] analyzed data from two outcome studies with suspected PE in 308 patients with previous VTE. Compared to a large group of subjects with no previous VTE, the incidence of confirmed PE was approximately twice as high in patients with previous VTE. After a 3-month follow-up there were no false-negative D-dimer results. It was noticed that the chance of a normal D-dimer test was 2-fold lower in patients with previous VTE, independent of older age, active malignancy, fever and recent surgery. Söhne et al. [23] investigated 480 patients with suspected PE and a previous VTE. One patient with suspected recurrent VTE and a normal D-dimer had a confirmed VTE during follow-up, resulting in a three month risk of 1.1% (95% CI: 0.03%–5.7%). The number needed to test for one negative result was 5.1. Nijkeuter et al. [24] has also investigated D-dimer testing in combination with a CDR in 234 patients with suspected recurrent PE. In the case of likely PE after the clinical decision rule or abnormal levels of D-dimer, a CT was performed. Of the patients that scored PE unlikely after the use of the clinical decision rule with a normal value of D-dimer, none returned with a symptomatic recurrent VTE. Based on these findings, D-dimer in combination with a clinical decision rule is safe to exclude PE or DVT in the same diagnostic strategy as for patients with first presentation. However the utility of the D-dimer test in excluding PE or DVT without further radiological testing is reduced (Grade of Recommendation: B). 6. Age and level of D-dimer Clinical scenario: A 75 year old man presents to the emergency department with symptoms of DVT. You expect D-dimer to be elevated in this older patient. Is it still useful to do D-dimer testing? Several studies have shown that D-dimer levels significantly increase with age [25]. Therefore, the age at which D-dimer testing can be used safely and cost-effectively is unclear. Over the last years, a few studies have investigated the relation between age and D-dimer levels (Table 6). Table 6 Characteristics of studies on age and level of D-dimer. Study Tardy et al. [26] Study Patients (n) Confirmed VTE (%) Test (cut-off) Aguilar et al. [17] 70 13 DVT (19) STA Liatest, Diagnostica Stago, France (0.4 ng/ml) Patients (n) Confirmed VTE (%) Test (cut-off) 96 40 PE (42) Righini et al. [27] 1029 280 PE (27) Söhne et al. [23] 482 131 PE (27) Harper et al. [28] 1897 Schutgens et al. [29] Carrier et al. [30] Table 4 Characteristics of study on measurement of D-dimer during oral anticoagulation. STA-Liatest, Diagnostica Stage (470 ng/mL) STA Liatest, Diagnostica Stago, France (400 ng/mL) VIDAS D-dimer, bioMerieux, France (500 ng/mL) VIDAS D-dimer, bioMerieux, France or Tinaquant assay, Roche Diagnostica, Germany (500 ng/mL) VIDAS D-dimer, bioMerieux, France or Tinaquant assay, Roche Diagnostica, Germany (500 ng/mL) 482 68 PE (4) 80 DVT (4) 10 both (1) 317 DVT (66) 2696 400 DVT (15) ELISA Asserachrom, Diagnostica Stago, France (500 ng/mL) ELISA Asserachrom, Diagnostica Stago, France and VIDAS D-dimer, bio Merieux, France (500 ng/mL) VIDAS D-dimer, bioMerieux, France or Tinaquant assay, Roche Diagnostica, Germany (500 ng/mL) VIDAS D-dimer, bio Merieux, France (500 ng/mL) Tinaquant assay, Roche Diagnostica, Germany (500 ng/mL) SimpliRED assay, Agen Biomedical Ltd., Australia (qualitative), IL-Test, Instrumentation Laboratory, USA (200 ng/mL), MDA D-Dimer, Trinity Biotech, USA (500 ng/mL) 444 E. Bruinstroop et al. / European Journal of Internal Medicine 20 (2009) 441–446 Tardy et al. [26] studied 96 consecutive outpatients older than 70 years old with suspected PE. D-dimer testing yielded a sensitivity of 100% and a specificity of only 14.3%. A larger study was performed by Righini et al. [27], who has divided 1029 patients into six age groups (b40; 40–49; 50–59; 60–69; 70–79; ≥80). The prevalence of PE increased heavily with age, from 24% in the youngest group, to 73% in the oldest group. There was only one false negative D-dimer result among the patients with confirmed pulmonary embolism in the age group 70–79 years. This resulted in an overall sensitivity of 99.6%, concluding that it can be used in all age groups. However, a normal Ddimer test result was present in 58% of patients younger than 40 years, but only in 5% of patients over 80 years. Söhne et al. [23] emphasized this conclusion for PE in 482 patients over 75 years with a sensitivity of 97% (95% CI: 84–100) and a very high number needed to test for one negative of 10.6 (7.9–13.5). Harper et al. [28] has studied both DVT and PE in 1897 unselected patients, which were divided into four age groups (b40; 40–60; 60–80; N80). They have also concluded that specificity decreased heavily with older age, from 70.5% in the youngest group to 4.5% in the oldest group. Regarding DVT, Schutgens et al. [29] included 812 patients in a retrospective post-hoc analysis of a multicentre prospective management study. Patients were divided into quartiles according to their age (18–46.7; 46.8–59.8; 59.9–73.8; N73.8). The prevalence of DVT was the lowest in the oldest quartile. In the total population only one DVT was missed in a patient of 48 years. Therefore for the oldest quartile the sensitivity and negative predictive value were both 100%. However, specificity decreased significantly with age from 49.2% in the youngest quartile to 17.4% in the oldest quartile. Recently, Carrier et al. [30], investigated the use of a clinical decision rule in combination with D-dimer testing using three different assays. 15.5% of the 2696 patients were older than 60 years. They concluded that negative predictive values of a low or unlikely CDR score in combination with a normal D-dimer can safely be used to exclude DVT in all age groups, including the elderly. All studies conclude that older patients have higher values of Ddimer, possibly due to higher incidence of co-morbidity. It has been hypothesized that the cut-off level of D-dimer should therefore be higher in the elderly [28,31]. Righini et al. [31] concluded that this rapidly increases the number of false-negatives and is therefore unsafe. We conclude that D-dimer testing is indicated in all age groups. Ddimer testing is inexpensive, and may render further invasive diagnostic tests unnecessary. Studies have shown that it is safe to exclude PE and DVT in older patients with a normal D-dimer value using standard cutoff values. However, the prevalence of normal D-dimer values decreases with age, which means that more radiological investigations are necessary to exclude VTE (Grade of Recommendation: B). 7. Diagnostic strategy in inpatients Clinical Scenario: A 60 year old male is admitted to the hospital with gastro-enteric disease. Four days after admission the patient suffers from respiratory symptoms, possibly due to a pulmonary embolism. Should the same diagnostic strategy be followed in this patient, as patients presenting to the emergency department? More co-morbidity can also be expected in patients admitted to the hospital (inpatients), providing a sample with elevated D-dimer levels. Factors that can contribute to a higher level of D-dimer are extensive inflammation, malignancy, surgery and liver disease [32– 35]. Hospitalized patients are also at higher risk for developing VTE with an increased mortality rate [36]. Several studies addressed the question whether D-dimer testing can be used in inpatients presenting with symptoms of PE (Table 7). Miron et al. [37] studied 114 consecutive patients admitted with surgical or medical conditions other than VTE. Only five patients with suspected PE had a normal D-dimer level, with no false negatives. However, the specificity was only 7%. Schrecengost et al. [38] included 233 patients on medical and surgical floors, as well as intensive care Table 7 Characteristics of studies on diagnostic strategy in inpatients. Study Patients (n) Confirmed VTE (%) Test (cut-off) Miron et al. [37] 114 Schrecengost et al. [38] 233 41 PE (36) 8 DVT (3) 17 PE (7) 2 both (1) 21 PE (10) 20 DVT (10) 4 both (2) Brotman et al. [39] 203 Söhne et al. [40] 207 160 PEa Kruip et al. [41] 605 155 PE (26) Kearon et al. [42] 556 108 PE (20) VIDAS D-dimer, bio Merieux, France (500 ng/mL) VIDAS D-dimer, bio Merieux, France (500 ng/mL) ELISA Dimertest Gold EIA, American Diagnostics, California (90 ng/mL) and others Tinaquant D-dimer, Roche Diagnostica, Germany (500 ng/mL) VIDAS D-dimer, bioMerieux, France or Tinaquant assay, Roche Diagnostica, Germany (500 ng/mL) SimpliRED assay, Agen Biomedical Ltd., Australia (qualitative) a Confirmed VTE in total study population of inpatients (n=207) and outpatients (n=540). unit patients, for whom D-dimer testing was requested to rule out PE. It has to be noted that no standard diagnostic protocol with a CDR was followed. The sensitivity of D-dimer testing alone was 89% with a specificity of 20%. Brotman et al. [39] investigated 203 hospitalized patients referred for radiographic imaging with a differential diagnosis of PE or DVT. ELISA D-dimer testing yielded the best results with a sensitivity of 96% and a specificity of 23%. Specificity declined with age and in patients hospitalized for over three days. Söhne et al. [40] studied D-dimer testing combined with a CDR in elderly inpatients, with 43% of the study population of 747 patients older than 65 years. Sensitivity in the group with low clinical probability and a normal Ddimer value was 91%. Compared to outpatients, only half of the inpatients had a low clinical probability and a normal D-dimer. The lowest specificity is seen in inpatients older than 75 years (15%). Kruip et al. [41] has also evaluated a diagnostic strategy including a clinical decision rule and D-dimer testing in 605 hospitalized patients. Compared to outpatients, inpatients were older, had higher prevalences of risk factors (e.g. immobility, surgery) and comorbid conditions (e.g. cancer, heart failure, COPD). In total only 10% of the inpatients were classified with an unlikely CDR-score and a normal Ddimer and were omitted from further investigations. During follow-up no venous thromboembolic events or death due to PE occurred in this group. Kearon et al. [42] prospectively investigated 562 inpatients in a randomized controlled trial to evaluate if it is safe to withhold additional diagnostic testing in patients with suspected PE and normal D-dimer levels. Compared to outpatients, inpatients were older, had a higher prevalence of cancer, a lower prevalence of low clinical probability of pulmonary embolism according to the CDR, a lower prevalence of normal D-dimer test results among these patients with low clinical probability, and a higher overall prevalence of PE. This study concludes that it is safe to withhold additional testing in inpatients with a low pretest clinical probability after a CDR and normal D-dimer results. However, the clinical utility is greater in outpatients as there are more than twice the amount of patients with low pretest clinical probability and normal D-dimer levels than for inpatients. We conclude that inpatients with suspected PE and a low clinical probability combined with a normal D-dimer level can safely be withheld from further diagnostic testing. Although the prevalence of inpatients with a low clinical probability and normal D-dimer level is lower than in outpatients, it must be emphasized that all of these patients can be withheld from CT scanning (Grade of Recommendation: B). E. Bruinstroop et al. / European Journal of Internal Medicine 20 (2009) 441–446 8. Conclusion The D-dimer levels are of value in excluding VTE in patients with a normal D-dimer in combination with low probability outcome of a clinical decision rule. In this review we aimed at assisting the clinician by reviewing the best available evidence on specific clinical conditions. The grades of recommendation have to be taken into account. In this review, three conditions are discussed in which the value can be lower then normally expected and give potentially false negative results. This includes patients with symptoms lasting for more than 14 days, during the first days of heparin treatment and during oral anticoagulation treatment. In the diagnosis of recurrence, older patients and inpatients the D-dimer level can possibly be higher than expected, potentially giving false positive results and making the diagnostic strategy including D-dimer less cost effective. These conditions therefore are possible pitfalls for the clinician. Based on the current evidence D-dimer should not be used when the onset of symptoms is more then two weeks before presentation or while using therapeutic heparin or OAT treatment. In the case of diagnosis of recurrent VTE, D-dimer can be used in the same diagnostic strategy as for first presentations, including the use of a clinical decision rule. In all age groups this diagnostic strategy can also safely be used. With inpatients D-dimer testing together with a clinical prediction model is safe to exclude PE. 9. Learning points D-dimer cannot safely be used in: • Patients with symptoms of VTE for over 14 days. Radiological examination should be performed. • Patients with suspected VTE receiving therapeutic heparin treatment. Radiological examination should be performed. • Patients with suspected DVT receiving OAT. Radiological examination should be performed. D-dimer combined with a clinical decision rule can safely be used in: • Patients presenting with recurrent VTE. A lower specificity should be expected. • Patients with suspected VTE of all age groups. A lower specificity should be expected in the elderly. • Inpatients with suspected PE. A lower specificity should be expected. Acknowledgement We would like to thank N. Pedersen for his helpful advice. References [1] 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;349:1227–35. [2] van Belle A, Büller HR, Huisman MV, Huisman PM, Kaasjager K, Kamphuisen PW, et al. Effectiveness of managing suspected pulmonary embolism using an algorithm combining clinical probability, D-dimer testing, and computed tomography. JAMA 2006;295:172–9. [3] Harbour R, Miller J. A new system for grading recommendations in evidence based guidelines. BMJ;323:334–6. [4] van der Graaf F, van den Borne H, van der Kolk M, de Wild PJ, Janssen GW, van Uum SH. Exclusion of deep venous thrombosis with D-dimer testing—comparison of 13 D-dimer methods in 99 outpatients suspected of deep venous thrombosis using venography as reference standard. Thromb Haemost 2000;83:191–8. [5] D'Angelo A, D'Alessandro G, Tomassini L, Pittet JL, Dupuy G, Crippa L. Evaluation of a new rapid quantitative D-dimer assay in patients with clinically suspected deep vein thrombosis. Thromb Haemost 1996;75:412–6. [6] de Bastos M, de Bastos MR, Bogutchi T, Carneiro-Proietti AB, Rezende SM. Duration of symptoms and D-dimer testing in the ruling-out of venous thromboembolism. Thromb Haemost 2006;4:2079–80. 445 [7] Keeling DM, Mackie IJ, Moody A, Watson HG. The diagnosis of deep vein thrombosis in symptomatic outpatients and the potential for clinical assessment and D-dimer assays to reduce the need for diagnostic imaging. Br J Haematol 2004;124:15–25. [8] Couturaud F, Kearon C, Bates SM, Ginsberg JS. Decrease in sensitivity of D-dimer for acute venous thromboembolism after starting anticoagulant therapy. Blood Coagul Fibrinolysis 2002;13:241–6. [9] Speiser W, Mallek R, Koppensteiner R, Stumpflen A, Kapiotis S, Minar E, et al. D-dimer and TAT measurement in patients with deep venous thrombosis: utility in diagnosis and judgement of anticoagulant treatment effectiveness. Thromb Haemost 1990;64:196–201. [10] Minnema MC, ten Cate H, van Beek EJ, van den Ende A, Hack CE, Brandjes DP. Effects of heparin therapy on fibrinolysis in patients with pulmonary embolism. Thromb Haemost 1997;77:1164–7. [11] Amelsberg A, Zurborn KH, Gärtner U, Kiehne KH, Preusse AK, Bruhn HD. Influence of heparin treatment on biochemical markers of an activation of the coagulation system. Thromb Res 1992;66:121–31. [12] Markers of hemostatic system activation in acute deep venous thrombosisevolution during the first days of heparin treatment. The DVTENOX Study Group. Thromb Haemost 1993;70:909–14. [13] Meissner MH, Zierler BK, Bergelin RO, Chandler WC, Manzo RA, Strandness Jr DE. Markers of plasma coagulation and fibrinolysis after acute deep venous thrombosis. J Vasc Surg 2000;32:870–80. [14] Peternel P, Terbizan M, Tratar G, Bozic M, Horvat D, Salobir B, et al. Markers of hemostatic system activation during treatment of deep vein thrombosis with subcutaneous unfractionated or low-molecular weight heparin. Thromb Res 2002;105:241–6. [15] Kuruvilla J, Wells PS, Morrow B, MacKinnon K, Keeney M, Kovacs MJ. Prospective assessment of the natural history of positive D-dimer results in persons with acute venous thromboembolism (DVT or PE). Thromb Haemost 2003;89:284–7. [16] Ombandza-Moussa E, Samama MM, Horellou MH, Elalamy I, Conard J. Potential use of D-dimer measurement in patients treated with oral anticoagulant for a venous thromboembolic episode. Int Angiol 2003;22:364–9. [17] Aguilar C, Del Villar V. Diagnostic value of D-dimer in outpatients with suspected deep venous thrombosis receiving oral anticoagulation. Blood Coagul Fibrinolysis 2007;18:253–7. [18] Palareti G, Cosmi B, Legnani C, Tosetto A, Brusi C, Iorio A, et al. D-dimer testing to determine the duration of anticoagulation therapy. N Engl J Med 2006;355: 1780–9. [19] Scarvelis D, Wells PS. Diagnosis and treatment of deep-vein thrombosis. CMAJ 2006;175:1087–92. [20] Rathbun SW, Whitsett TL, Raskob GE. Negative D-dimer result to exclude recurrent deep venous thrombosis: a management trial. Ann Intern Med 2004;141:839–45. [21] Aguilar C, Del Villar V. Combined D-dimer and clinical probability are useful for exclusion of recurrent deep venous thrombosis. Am J Hematol 2007;82:41–4. [22] Le Gal G, Righini M, Roy PM, Sanchez O, Aujesky D, Perrier A, et al. Value of D-dimer testing for the exclusion of pulmonary embolism in patients with previous venous thromboembolism. Arch Intern Med 2006;166:176–80. [23] Söhne M, Kruip MJ, Nijkeuter M, Tick L, Kwakkel H, Halkes SJ, et al. Accuracy of clinical decision rule, D-dimer and spiral computed tomography in patients with malignancy, previous venous thromboembolism, COPD or heart failure and in older patients with suspected pulmonary embolism. J Thromb Haemost 2006;4: 1042–6. [24] Nijkeuter M, Kwakkel-van Erp H, Söhne M, Tick LW, Kruip MJ, Ullmann EF, et al. Clinically suspected acute recurrent pulmonary embolism: a diagnostic challenge. Thromb Haemost 2007;97:944–8. [25] Rumley A, Emberson JR, Wannamethee SG, Lennon L, Whincup PH, Lowe GD. Effects of older age on fibrin D-dimer, C-reactive protein, and other hemostatic and inflammatory variables in men aged 60–79 years. J Thromb Haemost 2006;4:982–7. [26] Tardy B, Tardy-Poncet B, Viallon A, Lafond P, Page Y, Venet C, et al. Evaluation of D-dimer ELISA test in elderly patients with suspected pulmonary embolism. Thromb Haemost 1998;79:38–41. [27] Righini M, Goehring C, Bounameaux H, Perrier A. Effects of age on the performance of common diagnostic tests for pulmonary embolism. Am J Med 2000;109:357–61. [28] Harper PL, Theakston E, Ahmed J, Ockelford P. D-dimer concentration increases with age reducing the clinical value of the D-dimer assay in the elderly. Intern Med J 2007;37:607–13. [29] Schutgens RE, Haas FJ, Biesma DH. Reduced efficacy of clinical probability score and D-dimer assay in elderly subjects suspected of having deep vein thrombosis. Br J Haematol 2005;129:653–7. [30] Carrier M, Le Gal G, Bates SM, Anderson DR, Wells PS. D-dimer testing is useful to exclude deep vein thrombosis in elderly outpatients. Thromb Haemost 2008;6:1072–6. [31] Righini M, de Moerloose P, Reber G, Perrier A, Bounameaux H. Should the D-dimer cut-off value be increased in elderly patients suspected of pulmonary embolism? Thromb Haemost 2001;85:744. [32] Kinasewitz GT, Yan SB, Basson B, Comp P, Russell JA, Cariou A, et al. Universal changes in biomarkers of coagulation and inflammation occur in patients with severe sepsis, regardless of causative micro-organism. [ISRCTN74215569]Crit Care 2004;8(2):R82–90. [33] Raimondi P, Bongard O, de Moerloose P, Reber G, Waldvogel F, Bounameaux H. D-dimer plasma concentration in various clinical conditions: implication for the use of this test in the diagnostic approach of venous thromboembolism. Thromb Res 1993;69(1):125–30. [34] Boldt J, Huttner I, Suttner S, Kumle B, Piper SN, Berchthold G. Changes of haemostasis in patients undergoing major abdominal surgery—is there a difference between elderly and younger patients? Br J Anaesth 2001;87(3):435–40. 446 E. Bruinstroop et al. / European Journal of Internal Medicine 20 (2009) 441–446 [35] Vukovich T, Teufelsbauer H, Fritzer M, Kreuzer S, Knoflach P. Hemostasis activation in patients with liver cirrhosis. Thromb Res 1995;77(3):271–8. [36] Zakai NA, Wright J, Cushman M. Risk factors for venous thrombosis in medical inpatients: validation of a thrombosis risk score. Thromb Haemost 2004;2: 2156–61. [37] Miron MJ, Perrier A, Bounameaux H, Moerloose P, Slosman DO, Didier D, et al. Contribution of noninvasive evaluation to the diagnosis of pulmonary embolism in hospitalized patients. Eur Respir J 1999;13:1365–70. [38] Schrecengost JE, LeGallo RD, Boyd JC, Moons KG, Gonias SL, Rose Jr CE, et al. Comparison of diagnostic accuracies in outpatients and hospitalized patients of D-dimer testing for the evaluation of suspected pulmonary embolism. Clin Chem 2003;49:1483–90. [39] Brotman DJ, Segal JB, Jani JT, Petty BG, Kickler TS. Limitations of D-dimer testing in unselected inpatients with suspected venous thromboembolism. Am J Med 2003;114:276–82. [40] Söhne M, Kamphuisen PW, van Mierlo PJ, Büller HR. Diagnostic strategy using a modified clinical decision rule and D-dimer test to rule out pulmonary embolism in elderly in- and outpatients. Thromb Haemost 2005;94:206–10. [41] Kruip MJ, Söhne M, Nijkeuter M, Kwakkel-Van Erp HM, Tick LW, Halkes SJ, et al. A simple diagnostic strategy in hospitalized patients with clinically suspected pulmonary embolism. J Intern Med 2006;260: 459–66. [42] Kearon C, Ginsberg JS, Douketis J, Turpie AG, Bates SM, Lee AY, et al. An evaluation of D-dimer in the diagnosis of pulmonary embolism: a randomized trial. Ann Intern Med 2006;144:812–21.