September Virtual Journal Club: Old folks and their D-dimer cutoffs

A higher D-dimer threshold safely rules-out pulmonary embolism in very elderly emergency department patients

Hernan Polo Friz, Lorenzo Pasciuti, Dario Francesco Meloni, Matteo Crippa, Giulia Villa, Mauro Molteni, Laura Primitz, Davide Del Sorbo, Giovanni Delgrossi, Claudio Cimminiello

Internal Medicine, Medical Department, Vimercate Hospital, Vimercate, Italy

Introduction

Venous thromboembolism (VTE), encompassing both deep vein thrombosis (DVT) and its main complication Pulmonary Embolism (PE), is associated with high morbidity and mortality [1,2]. Pulmonary Embolism is the third largest cause of cardiovascular death after coronary artery disease and stroke [1]. The assessment of clinical probability using prediction scores [3–5] or implicit clinical judgment [6] in combination with D-dimer and pulmonary Computed Tomography Angiography (CTA) are useful tools for the diagnosis of PE, in outpatients and inpatients. In the Emergency Department (ED) setting diagnostic strategies initially focus on identifying patients in whom PE can be safely ruled out[1,2]. A confirmed diagnosis occurs in a minority of patients investigated for suspicion of PE [7]. D-dimer is a highly sensitive test but with increasing age its specificity for PE decreases [8]. In general and in elderly population higher D-dimer thresholds have been proposed to increase the proportion of subjects in whom PE could be safely ruled-out, thus reducing the number of CTA, avoiding contrast agent and radiation risks [1,2]. So far few studies have explored the potential of higher D-dimer cutoff values in very elderly subjects (aged 80 or more years), and since most studies have included outpatients with suspected DVT, data on D-dimer diagnostic value in very elderly ED patients with suspected PE are limited. 

On the basis of data from retrospective studies a higher age-adjusted D-dimer threshold was proposed to exclude VTE in the elderly [9–11]. Other authors have tested different fixed higher level D-dimer cutoffs [12,13]. A fixed threshold could be simpler and easier to use in clinical practice than the age-adjusted one, but evidence is quite limited and studies on direct comparison of these two approaches are lacking. 

Therefore, we conducted this study to determine whether increasing the standard threshold of serum D-dimer to a higher fixed value and to the age-adjusted cutoff could increase the test specificity for the exclusion of PE without reducing the sensitivity, in the setting of elderly and very elderly patients presenting to the ED with clinically suspected PE.

Material and methods

This retrospective cohort study was conducted in the Vimercate Hospital, a 500-bed community hospital. Study population consisted of all patients aged more than 18 years evaluated in the ED for clinically suspected PE (index episode) from January 1, 2010 through December 31, 2012. Patients underwent quantitative D-dimer and pulmonary CTA as part of the standard praxis in the ED. Those receiving full-dose anticoagulation before index episode were excluded from the analysis.

All clinical data were obtained through the electronic database. Helical computerized tomography scans were performed on a Brilliance Philips CT scanner (Philips, Cleveland, OH, USA) which included 64- detector row capability. D-dimer was measured with particle enhanced immunoturbidimetric assay Innovance DDIMER on the Behring Coagulation System (BCS) analyzer (Siemens Medical Solutions Diagnostics, Deerfield, IL, USA). The normal value declared by the producer is less than 490 ng/mL [14].

Pulmonary embolism was ruled out or confirmed on the basis of a negative or positive CTA, that is the absence or presence of a filling defect in one or more pulmonary arteries up to sub-segmental arteries. All cases were interpreted at the time of the acquisition of images by certified radiologists belonging to the hospital team. Vital status of index episode was recorded.

Data analysis

A Receiver Operating Characteristic (ROC) curve was constructed with D-dimer values and the diagnosis of PE on the basis of CTA results. Sensitivity, Specificity, Positive (PPV) and Negative Predictive Value (NPV) were calculated for different D-dimer cutoff values. Considering ROC curve analysis and clinical feasibility criteria, two fixed (1000 and 1500 ng/mL, that is approximately doubling and tripling the standard cutoff value) higher D-dimer cutoffs were tested, along with standard (490 ng/mL) and age-adjusted (age x 10 ng/mL [9]) thresholds.

Index Episode crude mortality rates and the Number Needed to Test (NNT) were calculated for the different thresholds of D-dimer both in the whole population and in subjects aged 80 or more years. NNT was obtained by dividing 1 by the proportion of patients with negative D-dimer.

Statistical analysis was performed using software SPSS version 18.0, (SPSS, Chicago, IL, USA). Student’s t-test analysis was performed for continuous variables and Fisher exact and chi-square test for differences between categorical variables. P values b 0.05 were considered to be significant. Exact 95% Confidence Intervals (CI) around the observed incidences were calculated using the exact method and the method based on the approximation to a normal distribution, depending on the case.

Results

A total of 492 patients with suspected PE were evaluated. Nine cases were excluded for receiving full-dose anticoagulation before index episode and two cases for lacking of D-dimer test results due to technical problems. The remaining 481 patients (305 women, 63.4%, mean age ± SD 73.0 ± 16.1) constituted the study population, with an overall prevalence of confirmed PE of 22.5% (108 cases). In 2010, the number of CTA carried out for clinical suspicion of PE were 94 and the number of ED visits/year 69,289. In 2011: 165 and 70,623, and in 2012: 222 and 74,418 respectively (p b 0.001). The percentage of confirmed PE were 22.3%; 20.0%; and 24.3% in 2010, 2011 and 2012 (p = 0.601). In patients aged 80 or more years (n = 191) the number of diagnosed PE was 37(19.4% of total). Table 1 shows the main characteristics of the study population.

ROC curve of D-dimer values showed an area under the curve (AUC) of 0.749 (95%CI: 0.698-0.800) in the whole population and 0.739 (95%CI: 0.654-0.824) in patients aged 80 or more years, Fig. 1.

Values of Sensitivity, Specificity, PPV and NPV for the different tested cutoffs are shown in Table 2. In patients aged 80 or more years, compared with the standard 490 ng/mL threshold, both a fixed higher D-dimer (1000 ng/mL) and the age-adjusted cutoff value increase the specificity of D-dimer assay for the exclusion of PE without reducing sensitivity and maintaining a NPV of 100%. In these subjects the use of a cutoff of 1000 ng/mL would lead to a saving of 12 CTAs scans (6.3% of total) and the age-adjusted derived a saving of 10 (5.2%) CTAs (percentage difference: 1.05, p = 0.158).

NNT values in the whole population were 60 for the conventional cutoff, 8 for 1000 ng/mL and 17 for the age adjusted derived. In very elderly subjects NNT was incalculable for the conventional cutoff (0 cases), 16 and 20, respectively, Table 2.

In the whole population, index episode mortality was higher in subjects aged 80 or more years, Table 1. In patients with PE, index episode mortality was 6.5% and in all cases death occurred in subjects with D-dimer values above 1000 ng/mL and age adjusted derived thresholds. Index episode mortality for different D-dimer cutoffs is shown in Table 3.

Discussion

With increasing age specificity of D-dimer for acute PE decreases resulting in a lower percentage of elderly patients in whom the diagnosis can be ruled-out.

Our study shows that in very elderly patients (aged 80 or more years) presenting to the ED for suspected PE, the application of a higher fixed threshold (1000 ng/mL, that is approximately doubling the value of the standard commercial cutoff) increased the specificity of D-

dimer assay for the exclusion of PE without reducing the test sensitivity. This was also true for the age-adjusted derived D dimer cutoff, developed on the basis of data from retrospective studies and externally validated in different studies [10–12]. It further shows that the number of CTA scans avoided by using the higher fixed threshold is similar to using the age-adjusted derived, reducing unnecessary radiation and contrast media exposure and excessive healthcare costs.

Several other aspects deserve to be mentioned. First, the number of CTA carried out for clinical suspicion of PE increased dramatically in the last few years, confirming the results of observational studies performed elsewhere [15]. However, the percentage of confirmed PE remained unchanged over the years ranging between 20 and 24%.

Second, compared with the conventional cutoff, we found a reduction in the NNT for the 1000 ng/mL threshold and age-adjusted one, in the whole population and in the very elderly, data that supports the good performance of higher cutoffs. All subjects aged 80 or more years had a D-dimer value over the conventional cutoff (NNT was not calculable) confirming the little usefulness of this threshold in the very elderly population. Penaloza et al. [11] observed a decrease in NNT in patients aged more than 75 years. Douma et al. found a reduction in the NNT from 18 to 5 between patients aged more than 80 years, but with one patient with a D-dimer level below the age-adjusted cutoff having PE [9]. In very elderly subjects we did not found false negative cases using the 1000 ng/mL and age-adjusted D-dimer thresholds.

Third, we found a 6.5% of mortality during index PE episode, against the 3.0% reported in a large study by Ng et al. [16], probably due to the fact that ours is an older population. In an Italian population Pomero et al. reported an index episode mortality of 12.8%, but they considered not only patients with PE presenting to the ED, but hospitalized subjects as well [17]. In subject aged 80 or more years, with and without PE, no index episode deaths occurred in patients with D-dimer values under 1000 ng/mL and the age adjusted cutoff. In the whole population we found an index episode mortality of 3% in subject without PE with D-dimer between 490 ng/mL and 1000 ng/mL and under the age-adjusted threshold. That is, in younger patients D-dimer seems to be able to identify high risk subjects due to conditions other than PE.

The fact that our study was a monocentric one warrants homogeneity in performing D-dimer test, CTA scanning procedures and in clinical approach. In the other hand, multicentric studies, using different commercially available D-dimer assays are necessaries for a complete validation and generalization of these results. Other limitations are worthy of mention. First, retrospective design. Prospective studies are needed to better understand the role of D-dimer in algorithms to manage patients with suspected PE. Doubling the standard threshold of commercial D-dimer assays could be proposed to be test in future prospective studies in very elderly patients. Second, CTA definition of PE was stated on the basis of the report of a radiologist belonging to the hospital team, without a blinded external reading. Third, the use of the proposed cutoff as part of a PE ruling-out algorithm (e.g. doubling the standard D-dimer threshold in patients with low pre-test probability) probably would yield a lower number of saved imaging tests. Finally, at the moment we have no median and long term mortality data to better interpret D-dimer performance results.

We conclude that in very elderly subjects referring to the ED with clinically suspected PE, the application of both a simple and easy-to-remember fixed higher D-dimer cutoff (1000 ng/mL, that is approximately doubling the standard commercial value) and the age-adjusted derived threshold increase the specificity of D-dimer assay for excluding PE without reducing test sensitivity. The application of these higher thresholds would lead to a safely reduction in the number of pulmonary CTA in that population, with patient benefits and cost savings.

Questions

1. What is the appropriate algorithm for working up PE?

     a) Wells' criteria

     b) PERC rule

     c) Determine pre-test probability with Wells' criteria, then D-dimer (if low prob for PE) or CT chest pulm angio (if mod-high prob for PE)

     d) Scan'em, scan'em good.

     e) Clinical gestalt

2. What is the current literature-supported cutoff considered pathologic for D-dimer test for all ages? (Of note, Allina lab values for D-dimer are measured in mcg/mL, not ng/mL)

     a) 200ng/mL

     b) 450ng/mL

     c) 500ng/mL

     d) 650ng/mL

     e) one MILLION ng/mL!

3. True or False: Using the 500ng/mL D-dimer cutoff in elderly patients can result in more false positives and CT scans?

4. True or False: This study found a D-dimer cutoff of 1000 ng/mL maintained a sensitivity of 100% for patients >80 years old?

5. What was the age-adjusted strategy for D-dimer levels that this paper used to achieve a sensitivity of 98.2% for finding PE? (Of note, you will now find this age adjusted calculation when looking at normal D-dimer limits in Epic)

     a) 2 times patient's age

     b) 5 times the patient's age

     c) 8 times the patient's age

     d) 10 times the patient's age
     e) 10 times Warren's age

References

[1] Torbicki A, Perrier A, Konstantinides S, Agnelli G, Gali e` N, Pruszczyk P, et al. Guidelines on the diagnosis and management of acute pulmonary embolism: the Task Force for the Diagnosis and Management of Acute Pulmonary Embolism of the European Society of Cardiology (ESC). Eur Heart J 2008;29:2276–315.

[2] Hirsh J, Guyatt G, Albers GW, Harrington R, Schunemann HJ, American College of Chest Physicians. Executive summary: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008;133:71S–109S.

[3] Wells PS, Anderson DR, Rodger M, Ginsberg JS, Kearon C, Gent M, et al. Derivation of a simple clinical model to categorize patients probability of pulmonary embolism: increasing the models utility with the SimpliRED D-dimer. Thromb Haemost 2000;83:416–20.

[4] Wicki J, Perneger TV, Junod AF, Bounameaux H, Perrier A. Assessing clinical probability of pulmonary embolism in the emergency ward: a simple score. Arch Intern Med 2001;161:92–7.

[5] Ceriani E, Combescure C, Le Gal G, Nendaz M, Perneger T, Bounameaux H, et al. Clinical prediction rules for pulmonary embolism: a systematic review and metaanalysis. J Thromb Haemost 2010 May;8(5):957–70.

[6] Penaloza A, Verschuren F, Meyer G, Quentin-Georget S, Soulie C, Thys F, et al. Comparison of the unstructured clinician gestalt, the wells score, and the revised Geneva score to estimate pretest probability for suspected pulmonary embolism. Ann Emerg Med 2013;62(2):117–24.

[7] Le Gal G, Bounameaux H. Diagnosing pulmonary embolism: running after the decreasing prevalence of cases among suspected patients. J Thromb Haemost 2004;2(8):1244–6.

[8] 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(9):607–13.

[9] Douma RA, le Gal G, Söhne M, Righini M, Kamphuisen PW, Perrier A, et al. Potential of an age adjusted D-dimer cutoff value to improve the exclusion of pulmonary embolism in older patients: a retrospective analysis of three large cohorts. BMJ 2010;340:c1475.

[10] Jaffrelot M, Le Ven F, Le Roux PY, et al. External validation of a D-dimer age adjusted cutoff for the exclusion of pulmonary embolism. Thromb Haemost 2012;107(5):1005–7.

[11] Penaloza A, Roy PM, Kline J, et al. Performance of age-adjusted D-dimer cutoff to rule out pulmonary embolism. J Thromb Haemost 2012;10(7):1291–6.

[12] Schouten HJ, Koek HL, Oudega R, et al. Validation of two age dependent D-dimer cutoff values for exclusion of deep vein thrombosis in suspected elderly patients in primary care: retrospective, cross sectional, diagnostic analysis. BMJ 2012;344: e2985.

[13] Haas FJ, Schutgens RE, Biesma DH. An age-adapted approach for the use of D-dimers in the exclusion of deep venous thrombosis. Am J Hematol 2009;84(8):488–91.

[14] Coen Herak D, Milos M, Zadro R. Evaluation of the Innovance D-DIMER analytical performance. Clin Chem Lab Med 2009;47(8):945–51.

[15] Wiener RS, Schwartz LM, Woloshin S. When a test is too good: how CT pulmonary angiograms find pulmonary emboli that do not need to be found. BMJ Jul 2 2013;347:f3368.

[16] Ng AC, Chung T, Yong AS, Wong HS, Chow V, Celermajer DS, et al. Long-term cardiovascular and noncardiovascular mortality of 1023 patients with confirmed acute pulmonary embolism. Circ Cardiovasc Qual Outcomes 2011;4(1):122–8.

[17] Pomero F, Fenoglio L, Melchio R, Serraino C, Ageno W, Dentali F. Incidence and diagnosis of pulmonary embolism in Northern Italy: A population-based study. Eur J Intern Med 2013;S0953–6205(13):00012–5.