Assessment of plasma D-dimer as a diagnostic and prognostic aid for abdominal aortic aneurysm

INTRODUCTION

The abdominal aorta is the main vessel supplying blood to the abdomen, pelvis and legs. Weakening and dilation of this artery is a condition known as abdominal aortic aneurysm (AAA). Screening studies suggest that AAA is present in 1-5 % of men aged >65 years [1].

AAA can lead to aortic rupture, which is usually fatal and responsible for a considerable number of deaths in the Western World, although recent reports suggest that incidence may be falling [2]. Patients with AAA are also at increased risk of other cardiovascular events, such as myocardial infarction and stroke [3].

Identification of AAAs is usually by incidental imaging since in most cases they are asymptomatic. Other means to identify patients with AAAs could facilitate early identification of at-risk individuals for cardiovascular risk management and treatment of their AAA as currently indicated [4].

Current guidelines for AAA advocate open or endovascular repair of AAAs measuring >55 mm in maximum diameter [5]. Smaller AAAs are monitored by repeat imaging at 3-24 months intervals, depending on the initial AAA diameter [5].

Most patients with AAA receive prolonged monitoring involving repeat imaging and physician consultation, with up to 70 % of patients eventually requiring AAA surgery [6].

Currently, initial AAA diameter is the only measure suggested to determine the frequency of monitoring interval for small AAAs. Additional predictors of outcome may provide a much more streamlined and individualized management pathway for patients with small AAAs [4].

Circulating biomarkers have been suggested as a potential means to aid both identification and monitoring of small AAAs. A number of previous small studies suggested that plasma D-dimer was elevated in patients with AAAs [7]. This manuscript provides a summary of a recent study which suggested that measurement of plasma D-dimer could be used to aid diagnosis and prognosis of AAA.

The work has been previously published in full [8].

THE STUDY

Study subjects were recruited from two sources [9, 10]:

Clinical risk factors were defined based on history or previous diagnosis of smoking, hypertension, diabetes, coronary heart disease (CHD) and dyslipidemia. Waist-hip ratio (WHR) was measured in the HIMS subjects.

The following variables were assessed: age, sex (referral group only), CHD, smoking, diabetes, hypertension, dyslipidemia, WHR, creatinine, C-reactive protein (CRP) and D-dimer. The effect of D-dimer on AAA prevalence adjusted for other risk factors was assessed. In cases where more than one cut-off of a single variable proved to be of relevance, the odds ratios (ORs) were expressed relative to the baseline of the lowest cut-off.

DIAGNOSTIC AND PROGNOSTIC VALUE OF D-DIMER

In a multiple logistic regression model, where all risk factors were examined, plasma D-dimer proved to be dominant, see Table I and Table II.

TABLE I: Risk factors for AAA in the population group [8]

TABLE II: Risk factors for AAA in the referral group [8]

Average yearly increase in AAA diameter was positively correlated with plasma D-dimer (r=0.39, p<0.001). Plasma D-dimer was independently associated with AAA progression after adjusting for other risk factors, including initial AAA diameter.

Plasma D-dimer plus initial AAA diameter were able to predict groups of patients with median yearly increase in AAA diameter as disparate as 0.2 and 2.5 mm/year [8].

CONCLUSIONS

Currently there are no circulating markers in clinical use for aiding diagnosis or prognosis for AAA [4]. Previous studies of AAA biomarkers have tended to concentrate on small groups of subjects and failed to employ tools to assess potential diagnostic and prognostic ability of markers [4].

The current study suggests that plasma D-dimer may have potential to aid identification and management of AAA. The exact cut-off of D-dimer to employ for either of these uses will likely depend on which D-dimer assay is used and on what the aim of the assessment is and the population being analyzed.

Thus if the aim is to screen in order to identify any subjects with possible AAAs, a low cut-off, such as >90 ng/mL, may be required, accepting there will be a number of false positives, for example in patients with DVT and aortic dissection [12,13].

However, if the most overall accurate and economic use is required, a higher cut-off will be needed. Using D-dimer plus the other risk factors, which made up our logistic regression model, a sensitivity of 85 % was achieved along with a specificity of 67 %. This equates to 15 false negatives per 100 patients with AAA and 33 false positives per 100 patients without AAA.

Development of risk-stratification groups is probably a more useful use of D-dimer plus clinical risk factors as illustrated by our classification and regression tree analysis described in the full publication [8]. Here multiple D-dimer cut-offs were employed along with clinical risk factors and CRP to identify groups with disparate AAA risk.

The novel finding of the current study was the association of plasma D-dimer with AAA progression. This finding remained after the adjustment for AAA diameter. It has been suggested that AAA thrombus is involved in aneurysm progression as a result of promoting inflammation and aortic-wall degradation [14].

D-dimer concentrations could reflect the “activity” of thrombus and thus provide a measure of on-going aortic-wall destruction. Further evidence is required and it is likely that randomized trials will address the effect of anti-platelet agents in limited AAA progression in the future.

In conclusion this work suggests that D-dimer can add to both a diagnostic and prognostic algorithm for AAA. The exact level of D-dimer employed and its economic value need further assessment in other populations.

Read more about d-dimer:

• Causes of increased D-dimer
• Age-adjustment of the D-dimer cut-off value to improve the exclusion of thromboembolic events in older patients
• The clinical use of a D-dimer assay
• Comparing D-dimer assays using likelihood ratios and Fagan nomograms
• D-dimer assays - pitfalls of analytical comparisons
• The role of D-dimer in diagnosis of venous thromboembolism – a review