Non-contrast-enhanced magnetic resonance imaging: Objective figures in differentiation between acute and chronic deep venous thrombosis in the lower extremities

Abstract

Background

Deep vein thrombosis is a severe health problem. Treatment options may differ between acute and chronic deep vein thrombosis. Thus, distinguishing acute from chronic deep vein thrombosis is essential for patients with deep vein thrombosis.

Triggered angiography non-contrast enhanced is an innovative magnetic resonance imaging protocol that may provide objective evidence in differentiating acute from chronic deep vein thrombosis.

Method

We prospectively collected information on consecutive patients who had been evaluated through triggered angiography non-contrast enhanced magnetic resonance imaging for venous pathology in their lower extremities at a vascular wound care center in a tertiary hospital between April 2017 and January 2020. Patients included were divided into two groups with the onset time cutoff point of 21 days. All were undergone non-contrast-enhanced magnetic resonance imaging evaluation. Non-contrast-enhanced magnetic resonance imaging images were evaluated by a radiologist, and lower extremity venous thrombosis, collateral-vein development, and subcutaneous honeycombing were emphasized. Cohen’s kappa coefficient was used to measure interrater agreement between the development of collateral veins, subcutaneous honeycombing, and symptom onset over 21 days.

Results

Interrater agreement analysis revealed that the development of collateral veins was substantially correlated with the onset of symptoms over 21 days (Table 1). Additionally, the development of subcutaneous honeycombing detected through triggered angiography non-contrast enhanced magnetic resonance imaging also substantially agreed with the onset of symptoms over 21 days (Table 2).

Conclusion

The diagnostic power of triggered angiography non-contrast enhanced magnetic resonance imaging in deep vein thrombosis is rival to current gold standard, color Doppler sonography. Triggered angiography non-contrast enhanced magnetic resonance imaging provides objective information on onset timing in patients with deep vein thrombosis that could differentiate acute from chronic deep vein thrombosis and provides guidance for treatment planning.

Keywords

Magnetic resonance imaging non-contrast venography TRiggered Angiography Non-Contrast Enhanced thrombosis

Background

Deep vein thrombosis (DVT) is a severe health problem that may lead to conditions ranging from uncomfortable venous claudication to life-threatening pulmonary embolism (PE).^1–3 Treatment options may differ between acute and chronic DVT, especially given the newly developed thrombolysis technique, catheter-directed thrombolysis (CDT).^4,5 Thus, distinguishing acute from chronic DVT is essential for patients with DVT. However, few objective diagnostic tools provide information on the onset of DVT.

TRiggered Angiography Non-Contrast Enhanced (TRANCE) is a magnetic resonance imaging (MRI) protocol that detects the vascular signals in the systolic and diastolic phases and depicts the arterial and venous structures by exploiting the signal discrepancy between the two phases. The resolution of TRANCE MRI is sufficient for imaging newly developed collateral veins and subcutaneous honeycombing in patients with DVT. In this study, we exploit this information to differentiate acute from chronic DVT.

Methods

Study design and participants

The Institutional Review Board (IRB) of Chang Gung Memorial Hospital approved this study (IRB number: 201700389B0). We prospectively collected information from patients who underwent lower extremity examinations at the tertiary hospital vascular wound care center from April 2017 and September 2019. Patients were eligible for inclusion in the study if they presented with symptoms of DVT and were clinically indicated for computed tomography angiography (CTA). Exclusion criteria were pregnancy and MRI contraindications (e.g., non-MRI-compatible device implant). In addition, patients with poor compliance and patients with multiple comorbidities that prevented them from lying down for the 25-min MRI protocol were excluded. Because ultrasonography (US) had been considered as the standard first-line tool for evaluating lower limb swelling, especially with suspicious of DVT, all participants were evaluated by a trained cardiologist using Doppler US, and then evaluated by an expert radiologist using non-contrast-enhanced MRI.

We defined subcutaneous honeycombing in MRI as the presence of vessel dilatation caudal to the location of the thrombosed vein. Collateral-vein development was defined as either cross-pelvic collateral or visible dilatation of the superficial vein across the thrombosed site compared with the unaffected limb.^6–9

MRI acquisition

All participants underwent non-contrast-enhanced MRI using TRANCE technique, which was performed on a 1.5-T MR scanner (Philips Ingenia, Philips Healthcare, Best, The Netherlands). Patients underwent imaging acquisition in supine position with use of peripheral pulse unit triggering. Imaging venography through TRANCE technique (TRANCE-MRV [magnetic resonance venography]) were evaluated by three-dimensional (3D) spin-echo (TSE) short tau inversion recovery (STIR) at systole period with the following parameters: repetition time, 1 beats; echo time, 85; inversion recovery delay time, 160; voxel size, 1.7 × 1.7 × 4 mm; field of view, 360 × 320. With systolic triggering, the arteries were black because of flow void effect. STIR provides extra background suppression because fat and bones are also suppressed. The result was a 3D data set with only venous structures. The quantitative flow (Q-Flow) scan was routinely performed to determine the appropriate trigger delay times for systolic and diastolic triggering. Imaging acquisition of three levels (abdomen, pelvis–thigh, and knee–leg–foot) was performed and the data were merged for reconstructing the whole lower extremity veins (including inferior vena cava). The imaging acquisition time takes 25-min in this TRANCE-MRV protocol. (Imaging principles please see Figure 1. Non-contrast of venous and arterial MRI please see Figure 2.)

Figure 1.

Summary principle of TRANCE-MRI technique. All images of the arterial system were assessed by 3D TSE sequences of systolic and diastolic phases. During systole, arterial blood flows rapidly and causes flow voids. Subtracting two phase scans will constitute a 3D data set with only arteries. Another image of the venous system was assessed during systole by 3D TSE STIR. STIR provides additional background suppression as signals of fat and bone are also inhibited.STIR: short tau inversion recovery; TSE: turbo spin-echo.

Figure 2.

Acute and chronic deep venous thrombosis (DVT). (a) A case of acute DVT of the right leg. MRV showing partial obliteration of the right anterior and posterior tibial vein. (b) Another case of chronic DVT of the left leg. MRV showing obliteration of the superficial femoral vein and popliteal vein. Besides, the development of collaterals (arrows) and lymphoedema (arrowheads) was noticed. MRV showed thrombosis of the posterior tibial vein and medial collateral veins in the left lower extremity.

Patient grouping and statistical analysis

The examination results of Doppler US and non-contrast-enhanced MRI were collected for statistical analysis. The results of the Doppler ultrasound examination for detecting DVT are considered to be true conditions as it is already present in the gold standard, then, the sensitivity, specificity, and accuracy of TRANCE-MRI for detecting DVT were calculated. Cohen’s kappa coefficient was used to measure the inter-rater agreement between US and MRI. Patients whose diagnose of DVT were established in both US and MRI were divided into two groups: ≤21 days and >21 days, depending on the time from the first day of symptoms to the day TRANCE MRI was performed. The cutoff point was set because the maximal delay of CDT is 21 days. Chi-square test and Fisher’s exact test were used to assess the correlation among collateral vein, subcutaneous honeycombing, and the chronicity of DVT.

Results

Between June 2017 and September 2019, 80 patients were enrolled in this study and were evaluated through TRANCE MRI for DVT at a vascular center in a tertiary hospital. Cohen’s kappa coefficient was used to measure interrater agreement between these two diagnostic tools (Table 3). It resulted that the Cohen’s kappa coefficient was 0.70, which indicated that the diagnosis of TRANCE MRI and color Doppler sonography was highly correlated in terms of DVT.

We subsequently analyzed the 46 patients whose diagnosis of DVT was agreed in both TRANCE MRI and color Doppler sonography. (Please see Figure 2 for venous thrombosis on TRANCE MRI). The demographic and descriptive data are shown in Table 4. All patients presented with symptoms of swelling in the lower extremities, pain, or local heat in the affected leg. Men accounted for 50% of the patients. The average age was 67.3 ± 11.7 years (Table 4). Clinical onset of the venous lesion was used to divide patients into two groups: $\leq$ 21 days and $>$ 21 days, depending on the time from the first day of symptoms to the day TRANCE MRI was performed. Eleven patients (24%) were allocated to the $\leq$ 21-days group and 35 patients (76%) were allocated to the $>$ 21-days group. Thirty-four (74%) had collateral veins from distal to proximal bypassing the thrombosed veins. Thirty-three patients (97%) were in the $>$ 21-days group (Table 4). Additionally, 36 patients (78%) had subcutaneous honeycombing according to TRANCE MRI. (Please see Figures 3 and 4 for subcutaneous honeycombing and collateral veins on TRANCE MRI.) Thirty-four patients (94%) were in the $>$ 21-days group (Table 4). Chi-square test revealed that the development of collateral veins was substantially correlated with the onset of symptoms over 21 days (Fisher’s exact test <0.05) (Table 1). Additionally, the development of subcutaneous honeycombing detected through TRANCE MRI also substantially agreed with the onset of symptoms over 21 days (Fisher’s exact test <0.05) (Table 2). Sensitivity and specificity for the development of collateral veins to the onset > 21 days of DVT were 97% and 83%, respectively. Whereas that for the development of subcutaneous honeycombing was 94% and 90%, respectively.

Figure 3.

Cases of subcutaneous honeycombing (a) A case of diffuse subcutaneous honeycombing of the both lower extremities. (b and c) Another case of left thigh subcutaneous honeycombing showed swelling of the left thigh with superficial honeycombing and net-like lesions.

Figure 4.

Cases of collateral veins. (a and b) A case of compression (*) on left common iliac vein with collateral veins (arrow) across the medial aspect of left thigh. (c) Another case of thrombosis of left popliteal vein with collateral veins (arrowhead) across the lateral aspect of left knee.

Table 1.

Development of collateral veins was substantially correlated with the onset of symptoms over 21 days.

	Collateral on TRANCE
	No	Yes
Onset
$\leq$ 21 days	10	1
$>$ 21 days	2	33

TRANCE: TRiggered Angiography Non-Contrast Enhanced.

Table 2.

The development of subcutaneous honeycombing detected through TRANCE MRI also substantially agreed with the onset of symptoms over 21 days.

	Subcutaneous honeycombing on TRANCE
	No	Yes
Onset
$\leq$ 21 day	9	2
$>$ 21 days	1	34

MRI: magnetic resonance imaging; TRANCE: TRiggered Angiography Non-Contrast Enhanced.

Table 3.

Cohen’s kappa coefficient was used to measure interrater agreement between these two diagnostic tools.

	DVT on TRANCE
	Positive	Negative
DVT on sonography
Positive	46	4
Negative	7	23
$κ = 0.70$

DVT: deep vein thrombosis; TRANCE: TRiggered Angiography Non-Contrast Enhanced.

Table 4.

The demographic and descriptive data.

Variables
Gender–Male	23(50%)
Age (years)	6 $7.3 \pm$ 11.7
Onset of venous lesion
$\leq$ 21 days	11 (24%)
$>$ 21 days	35 (76%)
Collateral veins detected by TRANCE-MRI	34 (76%)
Onset from venous lesion $>$ 21 days	33 (97%)
Subcutaneous honeycombing detected by TRANCE-MRI	36 (78%)
Onset from venous lesion $>$ 21 days	34 (94%)

MRI: magnetic resonance imaging; TRANCE: TRiggered Angiography Non-Contrast Enhanced.

Discussion

Venous pathology in the lower extremities may be suspected in patients with engorged calf veins, unhealed wounds, and asymmetrically swollen legs.¹⁰ US is the current mainstream screening tool and gold standard diagnostic tool.¹¹ Although it is convenient and noninvasive, it is operator-dependent; furthermore, detecting higher venous lesions in iliac veins is difficult. The alternative diagnostic tool was conventional venography.¹² However, it is time-consuming and invasive and the use of contrast media threatens patients’ renal functions. CT venography is an alternative to venography in diagnosis of DVT, but it still exposes patients to contrast media and radiation. Notwithstanding the advantages and disadvantages of these examination techniques, none provide adequate evidence to differentiate acute DVT from chronic DVT.

With the development of therapy for DVT, definite treatment such as CDT, systemic thrombolysis emerged. Treatment plans depend on the chronicity of the disease. In particular, because of newly developed thrombolysis management, the scientific consensus is that patients with an onset $\leq$ 21 days should receive treatment.^4,13,14 Differentiating acute from chronic DVT is essential.

In 2014, Dharmarajah et al. also published a systemic review on the aging technique for DVT.¹⁵ US and MRI were the two dominant techniques applied. Elastography was one of the ultrasound-related technique described.^16–21 Through measuring the elasticity of thrombosed vein, aging of clot was found to grow harder with time which may help determining the onset of DVT. Nevertheless, as high cost-effectiveness as it is, ultrasound-related examination remained operator dependent in quality. Most researches remained to be in vitro or animal studies. Reproductivity was the main concern. As for MRI-based techniques, existing method to determine the age of thrombosis was presented by Saha et al. with the technology of magnetic resonance T1 relaxation time.²² As TRANCE, neither the technique we presented in this study nor does the technique of magnetic resonance T1 relaxation time require contrast media. Froehlich et al. presented a contrast-MRI technique in 1997, which utilized the rim-center ratio to differentiate the clot age.²³ The application of magnetic transfer rate (MTR) and diffusion-weighted MRI (DW-MRI) on determining clot age was described by Phinikaridou et al. in 2013.²⁴ Protein and erythrocyte were found more condensed in older clots and this could be detected with MTR and DW-MRI. Nuclear examinations were also helpful in establishing the timeline of disease onset. Bates et al. and Brighton et al. both utilized Technetium to label certain biomarker, GPIIb/IIIa receptor antagonist and rt-PA respectively. Different uptake ratio was found in different age of clot.^25,26

In comparison to all the existing aging techniques, we would like to not only determine the chronicity of DVT, but we also want to analyze the flow of thrombosed veins. Rather than aiming on the age of clot, we had a different approach on the topic. We studied the focal pathological change along the natural history of DVT. Collateral veins development and subcutaneous honeycombing were two most significant change with time in DVT patients.

TRANCE MRI was the suitable tool of our choice for the analysis of venous flow rate and velocity. In our patient group, there were certain variation of venous flow rate in thrombosed veins and there was a trend that the severity of the DVT symptoms was related to poor flow of the thrombosed veins. This might give us a hint that the treatment plans might not only relate to the chronicity of the disease but also the flow rate of the thrombosed veins. Owing to the risk of CDT, certain acute DVT patients with relative fair flow of thrombosed veins might not need CDT. Rather, traditional anticoagulation should be adequate and a safer choice for them. While this might be a preliminary idea seen in the data of flow rate and velocity of vein yielded by TRANCE MRI, further analysis and helpful result to clinical practice could be expected.

We hereby present another choice for determining the chronicity of DVT in this paper, TRANCE MRI. The principle of TRANCE MRI technique is that various blood flow velocities have different signal intensities on the 3D TSE sequence. High signal intensity, as well as bright color, indicates a slow velocity, such as venous blood flow and diastolic arterial blood flow. High-velocity systolic arterial blood flow results in a flow void effect and is dark and low signal intensity. TRANCE MRI employs a high resolution and can identify isolated vascular structures, such as arteries or veins. Presenting only the venous structure without the accompanying arterial structure is difficult to achieve through conventional MRI or CT with use of contrast medium because the proper acquisition time is short and variable. We determined an additional advantage of TRANCE: subcutaneous honeycombing and development of collateral veins were clearly outlined in imaging.

As described in 2012 by Lu et al., the MRI protocol of fast-spin echo image with fat suppression was capable of revealing subcutaneous lymphatic structure.²⁷ The MRI protocol that we used was very much identical as the study by Lu et al. The extensive reticular pattern at epifascia was also compatible with our finding described as subcutaneous honeycombing. Therefore, though on the basis of noninvasive and radiation-free examination, we did not perform the gold standard examination for diagnosis of lymphedema, lymphoscintigraphy, we were fairly confident that the subcutaneous honeycombing that we found on TRANCE MRI was actually lymphedema.

The two pathological changes, collateral vein development and subcutaneous honeycombing, were more prevalent in patients with chronic DVT over 21 days and might be depicted by TRANCE MRI.²⁸ The TRANCE MRI should therefore provide the information on whether the patient should receive CDT.

We set the cutoff point of clinical onset time at 21 days and divided patients into two groups accordingly. Collateral veins and subcutaneous honeycombing were studied using TRANCE MRI in each patient to ascertain the relationship of their development with chronological onset. Cohen’s coefficient revealed that TRANCE MRI facilitated differentiating the patients’ chronological onset at the cutoff point of 21 days in the study of collateral-vein development and subcutaneous honeycombing. The results provide objective onset timing of DVT and provide guidance on the choice of options for patients.

Study limitation

The major limitation of the study was that it was a nonrandomized study with limited number of patients enrolled. Because of the number of patients, we could study only the quality rather than quantity of collateral veins. Standardization of the evaluation of the anatomy of the collateral veins was also absent. The same problem occurred in the investigation of subcutaneous honeycombing.

Furthermore, we have not utilized the result of the examination to direct the choice of treatment. Thus, we could not analyze the patient’s prognosis.

At last, we wish to analyze the flow of thrombosed veins with refined MRI protocol so that we could further understand whether certain cut point of venous flow could direct the choice of treatment in DVT patients.

Conclusion

The diagnostic power of TRANCE MRI in DVT is rival to current gold standard, color Doppler sonography. Furthermore, TRANCE MRI also provides objective information on onset timing in patients with DVT that could differentiate acute from chronic DVT and provides guidance for treatment planning.

Key questions

Is TRANCE MRI a diagnostic tool equivalent to current gold standard, Doppler sonography, in terms of deep vein thrombosis?

Is TRANCE MRI a reliable diagnostic tool on determining the chronicity of DVT?

Key findings and take-home message

The diagnostic power of TRANCE MRI in DVT is rival to current gold standard, color Doppler sonography.

TRANCE MRI provides objective information on the chronicity of DVT which could direct the treatment option.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical approval

The Institutional Review Board (IRB) of Chang Gung Memorial Hospital approved this study (IRB number: 201700389B0).

Guarantor

YKH.

Contributorship

YKH and CCK researched literature and conceived the study. CWC, YHT, and SCW were involved in protocol development. CCK, YKH, and CWC gained ethical approval, patient recruitment, and data analysis. CCK, CWC, and YHT wrote the first draft of the manuscript. All authors reviewed and edited the manuscript and approved the final version of the manuscript.

Acknowledgements

ORCID iD

Yao-Kuang Huang

References

De Backer

Epidemiology of chronic venous insufficiency. Angiology 1997; 48: 569–576.

Fowkes

Evans

Lee

AJ.

Prevalence and risk factors of chronic venous insufficiency. Angiology 2001; 52: S5–S15.

Beebe-Dimmer

, et al. The epidemiology of chronic venous insufficiency and varicose veins. Ann Epidemiol 2005; 15: 175–184.

Saha

, et al. Contemporary management of acute and chronic deep venous thrombosis. Br Med Bull 2016; 117: 107–120.

Wittens

, et al. Editor’s choice—management of chronic venous disease: clinical practice guidelines of the European Society for Vascular Surgery (ESVS). Eur J Vasc Endovasc Surg 2015; 49: 678–737.

Divittorio

Bluth

Sullivan

MA.

Deep vein thrombosis: diagnosis of a common clinical problem. Ochsner J 2002; 4: 14–17.

Ennis

Elmes

RJ.

Radionuclide venography in the diagnosis of deep vein thrombosis. Radiology 1977; 125: 441–419.

Kafeza

, et al. A systematic review of clinical prediction scores for deep vein thrombosis. Phlebology 2017; 32: 516–531.

Rossi

Agnelli

Current role of venography in the diagnosis of deep-vein thrombosis. Minerva Cardioangiol 1998; 46: 507–514.

10.

Hirsh

Hull

Raskob

GE.

Clinical features and diagnosis of venous thrombosis. J Am Coll Cardiol 1986; 8: 114B–127B.

11.

Needleman

, et al. Ultrasound for lower extremity deep venous thrombosis: multidisciplinary recommendations from the Society of Radiologists in ultrasound consensus conference. Circulation 2018; 137: 1505–1515.

12.

Bettmann

, et al. Contrast venography of the leg: diagnostic efficacy, tolerance, and complication rates with ionic and nonionic contrast media. Radiology 1987; 165: 113–116.

13.

Comerota

AJ.

The ATTRACT trial: rationale for early intervention for iliofemoral DVT.

Perspect Vasc Surg Endovasc Ther 2009; 21: 221–224; quiz 224–225.

14.

Notten

, et al. Ultrasound-accelerated catheter-directed thrombolysis versus anticoagulation for the prevention of post-thrombotic syndrome (CAVA): a single-blind, multicentre, randomised trial. Lancet Haematol 2020; 7: e40–e49.

15.

Dharmarajah

, et al. Aging techniques for deep vein thrombosis: a systematic review. Phlebology 2015; 30: 77–84.

16.

Emelianov

, et al. Triplex ultrasound: elasticity imaging to age deep venous thrombosis. Ultrasound Med Biol 2002; 28: 757–767.

17.

Aglyamov

, et al. Model-based reconstructive elasticity imaging of deep venous thrombosis. IEEE Trans Ultrason Ferroelectr Freq Control 2004; 51: 521–531.

18.

Geier

, et al. Ultrasound elastography for the age determination of venous thrombi. Evaluation in an animal model of venous thrombosis. Thromb Haemost 2005; 93: 368–374.

19.

Rubin

, et al. Clinical application of sonographic elasticity imaging for aging of deep venous thrombosis: preliminary findings. J Ultrasound Med 2003; 22: 443–438.

20.

Rubin

, et al. Sonographic elasticity imaging of acute and chronic deep venous thrombosis in humans. J Ultrasound Med 2006; 25: 1179–1186.

21.

Xie

, et al. Correspondence of ultrasound elasticity imaging to direct mechanical measurement in aging DVT in rats. Ultrasound Med Biol 2005; 31: 1351–1359.

22.

Saha

, et al. Magnetic resonance T1 relaxation time of venous thrombus is determined by iron processing and predicts susceptibility to lysis. Circulation 2013; 128: 729–736.

23.

Froehlich

, et al. “Bull’s-eye” sign on gadolinium-enhanced magnetic resonance venography determines thrombus presence and age: a preliminary study. J Vasc Surg 1997; 26: 809–816.

24.

Phinikaridou

, et al. In vivo magnetization transfer and diffusion-weighted magnetic resonance imaging detects thrombus composition in a mouse model of deep vein thrombosis. Circ Cardiovasc Imag 2013; 6: 433–440.

25.

Bates

, et al. Imaging characteristics of a novel technetium Tc 99m-labeled platelet glycoprotein IIb/IIIa receptor antagonist in patients With acute deep vein thrombosis or a history of deep vein thrombosis. Arch Intern Med 2003; 163: 452–456.

26.

Brighton

Janssen

Butler

SP.

Aging of acute deep vein thrombosis measured by radiolabeled 99mTc-rt-PA. J Nucl Med 2007; 48: 873–878.

27.

, et al. Magnetic resonance lymphography at 3T: a promising noninvasive approach to characterise inguinal lymphatic vessel leakage. Eur J Vasc Endovasc Surg 2012; 43: 106–111.

28.

Kurstjens

, et al. Abdominal and pubic collateral veins as indicators of deep venous obstruction. J Vasc Surg Venous Lymphat Disord 2016; 4: 426–433.