Correlation of acute pulmonal embolism with D-dimer levels and the diameter of the pulmonary trunk in thoracic multislice computed tomography. A single-centre retrospective analysis of 100 patients.

PURPOSE: The aim of this retrospective study was to evaluate the correlation between D-dimer levels in positive thromboembolic thoracic computed tomography (CT) with the diameter of the pulmonary trunk and to study the relation between the D-dimer and the uni- or bilateralism of the lesions and the presence of pulmonal trunk involvement. We also analysed gender-specific differences in patients with and without dilatation of the pulmonal trunk.
MATERIAL AND METHODS: A total of 100 acute care patients (50 men and 50 women) with positive thromboembolic multiple detector computed tomography of the thorax, performed on two modern CT scanners, were retrospectively studied. All thoracic CTs were evaluated by two expert radiologists, with attention paid to the diameter of the pulmonary trunk and the correlation of D-dimer level with the uni-or bilateralism of the lesions. We also analysed sex-specific correlations. All patients underwent multislice computed tomography-examination after applying 70 ml iodinated non-ionic contrast media. Graphpad Prism 8.1.1 software was used for statistical data.
RESULTS: The "strongest" weak correlation resulted between D-dimer levels and the axial diameter of the pulmonal trunk. Considering the correlation between the axial diameter of the pulmonal trunk and gender-related distributions, we found that female patients had higher axial diameters than men. Another weak relationship, almost zero, was found between the D-dimer level and gender. Regarding the correlation between the uni- or bilateralism of thromboembolism and the D-dimer levels, we also found a weak correlation.
CONCLUSIONS: This retrospective study showed that D-dimer levels, the diameter of the pulmonal trunk, the location, and gender-related distributions have almost no correlation and are not significantly predictive in imaging.

Introduction

Acute pulmonal embolism (PE) is a very common diagnosis in acute care and can cause diverse symptoms, ranging from mild respiratory distress to sudden death. In acute care units the annual incidence rate is from 0.2 to 0.8 /1000 [1,2]. Today, pulmonary angio-computed tomography (CT) after contrast administration is the gold standard method. The aim of the diagnosis of acute pulmonary embolism is to have the dose of the contrast agent adapted to the body weight, with the associated improved resolution of the smaller lung vessels [3].

In this retrospective single centre study, we tried to ascertain if there is a correlation between D-dimer levels in positive acute thromboembolic thoracic CT and the axial diameter of the pulmonary trunk (Figure 1), and if there is a cut-off D-dimer level regarding the uni-or bilateralism of the lesions and with the presence of pulmonal trunk involvement.

Figure 1

An example of measuring the pulmonal trunk

Today’s clinicians use diagnostic scores to identify risk stratification like the classic Wells score (WS), modified WS, simplified WS, revised Geneva score (GS), simplified GS, and the YEARS score. In our Institution clinicians use clinical signs and measurement of the D-dimer level in peripheric blood to confirm the suspicion of acute PE. D-dimer is a fibrin cleavage product that results from cleavage of cross-linked fibrin by factor XIIIa. Its low-level plasma expression expresses the control of haemostasis by coagulation factors and fibrinolytic system. Increased D-dimer values serve as activation markers of haemostasis, a condition that is present in any thrombotic process. Normal values preclude significant activation. The exact quantification of D-dimer levels has an important role in guiding therapy [4]. But which levels of D-dimer are pathologic? The D-dimer level has a low specificity but a very high sensitivity. It can be elevated in many other pathologies, after surgery, but also after bone fractures or infections. Many previous studies have shown that the D-dimer test is highly sensitive (> 95%) in acute deep venous thrombosis or pulmonary embolism [5]. In our hospital levels under 500 ng/ml fibrinogen equivalent units (FEU) are considered normal.

Another key in the pathophysiology of acute PE is the dilatation of the main pulmonal trunk (MPA). In a study from 2012 [6] the authors found no correlation in women between age and MPA in a healthy reference group. But which clinical significance has the dilatation of the MPA? There are many other causes of MPA dilatation, such as pulmonary arterial hypertension, Eisenmenger syndrome, high altitude, congenital cardiac shunting, pulmonal valvular stenosis, congenital, infectious, rheumatologic/vasculitis, connective tissue diseases, and idiopathic or traumatic causes. The most common cause is pulmonary hypertension (PH). Increased intrapulmonary pressures lead to vascular thickening and dilation with increased vascular wall shear stress, as well as increased collagen and elastin deposition in the adventitia layer of the pulmonal trunk. The continuous high pressure in the MPA causes cell activation and vascular remodelling in hypoxia-induced PH [7]. Another key in understanding pulmonal thromboembolism is to analyse the correlation between MPA-dilatation and the side involvement, i.e. if there is a predilection of involving the right or the left side in trunk dilation and non-trunk dilatation.

Material and methods

In this retrospective study, we considered 100 patients who underwent thoracic computed tomography angiography (CTA) with contrast administration at our Institution in the period January 2015 – January 2019 after admission to the Emergency Department. All patients presented clinical signs of PE and underwent D-dimer examination levels in acute care. The range of blood D-dimer levels were between 300 and 39.958 ng/ml (Figure 2). The diameter of the main pulmonary artery (MPA) ranged from 21 to 50 mm. We considered a normal reference of 29 mm in men and 27 mm in women [6]. Retrospectively we also analysed whether there were other associated reasons for high D-dimer levels.

Figure 2

The distribution shows that the most D-dimer levels are in the range under 10,000 ng/ml

All of the authors confirm that the research was performed according to the principles of the Declaration of Helsinki. Informed consent was not obtained.

Imaging protocol

Multiple detector computed tomography (MPCT) scans were performed in our department with two multislice CT-scanners. All the examinations were performed on a dual-source CT scanner (Somatom Definition Flash; Siemens Healthcare, Forchheim, Germany and Somatom Drive; Siemens Healthcare, Forchheim, Germany) with patients lying supine on the table with their arms at the side of their body, with spiral acquisition. Scanning parameters are given in Table 1. Thoracic CTAs were obtained after intravenous administration of 70 ml of 350 mg iodine/ml iodinated contrast material (Iobitridol, Xenetix 350, Guerbet, France) at a flow rate of 4 ml/s, followed by a 50-ml saline flush through an 18-gauge catheter placed in an antecubital vein using an automatic power injector (Medrad Stellant, Bayer); a bolus-tracking technique was adopted, with the region of interest (ROI) placed in the pulmonal trunk (threshold 100 Hounsfield units [HU]). We analysed only axial images with a slice thickness of 1 mm.

Table 1

Scanning and reconstruction parameters for pulmonal computed tomography angiography

Factor	SOMATOM definition flash	SOMATOM definition drive
Scanning technique	Spiral	Spiral
Scan direction	Craniocaudal	Craniocaudal
kVp	Care kV (ref. kV 120)	Care kV (ref. kV 120)
Care-DOSE reference	170 mA	150 mA
Collimation	128 × 0.6 mm	128 × 0.6 mm
Rotation time	0.28 sec	0.28 sec
Pitch	2.2	2.5

Interpretation

Two radiologists with approximately 10 years of experience retrospectively analysed the CTA scans, measuring the axial diameter of the pulmonal trunk at level of pulmonal artery bifurcation. We selected this method because of its fast and daily feasibility in routine praxis.

Statistical analysis

All statistical calculations were performed using Graphpad Prism 8.1.1 software. To calculate the correlations, we used Pearson (r) and the Spearman (p) correlation statistic methods.

Results

Considering the correlation between D-dimer levels and the axial diameter of the pulmonal trunk, we found a weak correlation, with a Pearson correlation r value of 0.2141 (Table 2). Using the Spearman correlation, we also found a weak value (p = 0.2871). The relationship between the D-dimer level and gender (Figure 3) gave a negative Pearson score of –0.05 (almost zero correlation).

Table 2

Pearson correlation

Pearson correlation r	Trunk diameter (mm) vs. D-dimer (ng/ml)
r	0.2141

Figure 3

Correlation between sex and D-dimer levels

Considering the correlation between the axial diameter of the pulmonal trunk and gender-related distributions (Figure 4), we found that female patients had higher axial diameters (58.3% above 27 mm) in PE versus male patients (46% above 29 mm), which means that women have a higher risk of developing pulmonal hypertension in PE than men [8]. Hormonal influence must be considered in the pathophysiology of trunk dilatation. Considering the correlation between the location (uni- or bilateralism) and the D-dimer levels, we found a weak correlation with a Pearson value of 0.2572 and a Spearman value of 0.2789. 64% of the study population had bilateral pulmonal involvement, 31% had unilateral involvement, and only 3% had a bilateral PE with trunk involvement (Figure 5). In the search for associated diseases or conditions that could lead to an increase in D-dimer levels, we found neoplastic diseases in 11 patients (11%) (four patients with a non-surgically treated lung tumour in chemotherapy and one in immunotherapy, one patient with a cerebral neoplasm in follow-up after surgical therapy and chemotherapy, and five patients with a recent diagnosis of an adenocarcinoma of the colon). In 35 patients (35%) we found a coexisting chronic atrial fibrillation. Twenty-six patients (26%) showed renal pathologies like mild chronic renal failure. In nine patients (9%) there was a chronic congestive cardiac failure. Associated deep vein thrombosis was found in 46 patients (46%). This indicates the presence of coexisting and overlapping conditions in these patients, which also increase D-dimer levels (Tables 3 and 4).

Figure 4

Correlation between diameter and gender. The red points indicate trunk dilatation, the blue ones indicate the normal range diameters

Figure 5

Distribution of location of the pulmonal embolism

Table 3

The study parameters

D-dimer(ng/ml)	Location	Gender	Trunk-diameter(mm)
1393	Bilateral	M	25
2238	Bilateral	F	38
5308	Unilateral	F	24
1940	Bilateral	M	26
1981	Unilateral	M	24
2356	Bilateral	M	23
30,388	Bilateral	F	25
5170	Unilateral	M	25
1492	Bilateral	F	24
6415	Unilateral	F	26
4259	Bilateral	M	31
17,803	Bilateral with trunk involvement	F	42
1378	Unilateral	M	24
916	Bilateral	F	24
7597	Bilateral	F	33
1654	Unilateral	F	25
1845	Bilateral	M	23
18,836	Bilateral	M	31
4870	Bilateral	M	24
10,374	Unilateral	M	27
21,965	Bilateral	F	26
13,858	Unilateral	M	24
2040	Bilateral	F	27
1095	Unilateral	M	25
3349	Bilateral	F	29
1315	Unilateral	F	24
11,519	Unilateral	F	29
39,958	Bilateral	F	31
4022	Bilateral	F	23
1511	Bilateral	F	24
1893	Bilateral	F	24
6281	Bilateral	F	28
2376	Bilateral	M	24
30,167	Bilateral	M	25
2052	Bilateral	M	25
2122	Bilateral	M	24
12,174	Bilateral	F	26
13,267	Bilateral	M	31
1286	Bilateral	F	31
6506	Bilateral	M	24
950	Unilateral	F	26
4766	Bilateral	F	24
1602	Bilateral	F	25
12,135	Bilateral	M	25
2410	Bilateral	M	33
1721	Unilateral	M	23
4318	Bilateral	M	24
4518	Bilateral	M	24
1602	Bilateral	M	26
4400	Bilateral	F	31
4649	Unilateral	F	28
6378	Bilateral	F	25
1621	Bilateral	M	25
24,340	Bilateral	M	31
6411	Bilateral	F	26
4318	Unilateral	M	32
4343	Bilateral	F	26
3519	Unilateral	M	27
1386	Unilateral	M	28
3038	Bilateral	M	29
2628	Bilateral	F	26
6109	Unilateral	F	27
1131	Bilateral	F	26
848	Unilateral	M	23
10,890	Bilateral with trunk involvement	M	36
11,451	Bilateral	M	33
1351	Unilateral	M	28
21,662	Bilateral	M	26
6689	Bilateral	F	36
4400	Unilateral	F	25
4400	Unilateral	M	25
2028	Bilateral	M	34
300	Unilateral	M	28
9200	Bilateral with trunk involvement	M	29
10,230	Bilateral	F	25
9615	Bilateral	F	25
1200	Bilateral	F	26
7200	Bilateral	F	36
4000	Bilateral	F	27
4300	Bilateral	M	24
4100	Bilateral	M	24
5902	Bilateral	F	31
4100	Bilateral	M	28
3125	Bilateral	F	36
4807	Unilateral	M	35
5386	Bilateral	F	31
8258	Bilateral	F	31
3402	Unilateral	F	24
4000	Bilateral	M	31
1030	Unilateral	M	26
3464	Bilateral	F	27
3719	Bilateral	M	23
3015	Bilateral	M	25
7314	Bilateral	F	26
2807	Unilateral	F	24
5355	Unilateral	F	23
1785	Unilateral	M	25
1646	Unilateral	F	25
2839	Bilateral	F	27
2060	Unilateral	M	28

Table 4

Minimum values, 25% percentile, median, 75% percentile and maximum values

Factor	Trunk diameter(mm)	D-dimer(ng/ml)
Total number of patients	100	100
Minimum value	23	300
25th percentile	24	1904.75
Median	26	4061
75th percentile	29	6483.25
Maximum value	42	39,958

Conclusions

The clinical presentation of acute PE is varied. The most frequent clinical symptoms are chest pain, tachycardia, hypotension, dyspnoea, cough, and haemoptysis [9]. But what can a radiologist expect from the CT examination? Can there be any correlation between clinical and radiological parameters? In this retrospective study we tried to find correlations between different radiological and clinical parameters in the imaging “outcome’, i.e whether parameters like D-dimer, diameter of MPA, uni- or bilateralism of PE, and gender-related factors are radiologically significant. Considering that not every centre has a dual-source CT, we analysed parameters that are simple in their clinical and radiological evaluation. Assuming D-dimer levels under 500 ng/ml to be negative, we found one of 100 patients (1%) with a D-dimer-level of 300 ng/ml with PE.

A study from 2014 [10] proposed an age-adjusted D-dimer cut-off, but the results were not satisfactory. In other studies several scores for quantification of the clot were proposed, such as the CT severity score and CT obstruction index developed by Mastora et al. and Qanadli et al., respectively, but these are not routinely used in clinical practice [11,12].

The different correlation types show low correlation statistical data. The “strongest” weak correlation was between D-dimer levels and the axial diameter of the pulmonal trunk. Considering the correlation between the axial diameter of the pulmonal trunk and gender-related distributions, we found that female patients had higher axial diameters than men. We confirmed that women have a higher risk of developing pulmonal hypertension in PE than men [8].

Another weak relationship (almost zero) was found between D-dimer level and gender. Regarding the correlation between uni- or bilateralism of thromboembolism and the D-dimer levels, we found a weak correlation. The limitations of this retrospective study are, in our opinion, the use of only one clinical parameter like the D-dimer level to assess acute PE. We did not include diagnostic scores to identify risk stratification, such as the classic Wells score (WS), modified WS, simplified WS, revised Geneva score (GS), simplified GS, and the YEARS score.

Other drawbacks of this study are the exclusion of the age of the patients and the fact that a dilatation of MPA can have other causes, like cardiac shunts, arterial hypertension, Eisenmenger syndrome, high altitude, pulmonal valvular stenosis, infections, and connective tissue diseases, as well as idiopathic or traumatic causes. Another limitation is the absence of a control group and the presence of coexisting pathologies that increase D-dimer levels.

This retrospective study showed that D-dimer levels, the diameter of the pulmonal trunk, its location, and gender-related distributions have almost no correlation and are not significantly predictive in imaging.

Disclosure

The authors declared no conflicts of interest.