Double region of interest timing bolus technique following endovascular aortic repair: Short-term prognosis analysis

Abstract

Objectives

To compare the incidence rate of reintervention in patients with and without complication findings at aortic computed tomography using double region of interest timing bolus (DRTB) method after endovascular stent placement of the aorta.

Methods

We included 40 patients who underwent computed tomography of the aorta using DRTB method after endovascular stent placement. DRTB method allows to scan the aorta with a short injection time of 9 s by synchronizing the scan speed to the aortic flow. Complication findings at computed tomography were defined as endoleak, rupture, occlusion, and infection. The primary endpoint was reintervention, which was defined as any of the following three events: conversion to open repair, graft revision, or secondary intervention.

Results

The mean contrast medium during computed tomography angiography was 38.6 ± 3.9 mL. Complication findings at computed tomography were present in 10 patients (25%): endoleak (n = 9) and infection (n = 1). During a median follow-up of 7 months (interquartile range, 4–11 months), two patients experienced reintervention. Kaplan–Meier curves by complication findings showed that event rate at 6 months was significantly higher in patients with complication findings than in patients without (20% vs 0%, p = 0.01). No patients without complication findings at computed tomography experienced reintervention.

Conclusions

No complication findings at computed tomography after intervention of the aorta resulted in good prognosis in patients who underwent aortic computed tomography using DRTB method.

Keywords

Aortic disease computed tomography angiography endovascular therapy

Introduction

Computed tomography (CT) is a useful modality for assessing post-operative complications of endovascular aneurysm repair.¹ Although aneurysm diameter could be measured by unenhanced CT, contrast medium injection is necessary to diagnose complications such as endoleaks, in-stent stenosis/occlusion, and infection. Patients with aortic diseases occasionally have renal disorders and the amount of contrast medium should be decreased without hampering the image quality.² Recent advances in CT technology allowed to perform CT of the aorta using low tube voltage or dual-energy technique with smaller amount of contrast medium than the conventional methods.^3,4 Furthermore, double region of interest (ROI) timing bolus (DRTB) technique reduced the contrast medium for CT angiography of the aorta to 40 mL, while significantly increasing the aortic enhancement than the conventional method.⁵ CT of the aorta could be performed with an injection time of 9 s by synchronizing the scan speed to the aortic flow. However, the short injection time might overlook complications after endovascular stent placement such as endoleaks. Therefore, the purpose of the present study was to compare the incidence rate of reintervention and mortality in patients with and without complication findings at aortic CT using DRTB method after endovascular stent placement of the aorta.

Materials and methods

Patients

Patients who underwent CT of the aorta entered in a prospective registry. The local ethics committee approved this study, and all patients provided written informed consent (UMIN Clinical Trials Registry 000030497). We initially included 63 patients in this registry between February 2018 and January 2019 with a prior history of endovascular stent placement of the aorta due to abdominal aortic aneurysm and/or occlusion. The indications for aneurysm repair were as follows: aneurysm diameter ≥5.0 cm, saccular aneurysm, an associated iliac aneurysm ≥3.0 cm, and rapidly expanding aneurysm (>5 mm expansion in <6 months).^6,7 We excluded the following patients: unable to give consent due to emergency treatment (n = 7), declined to participate (n = 7), and protocol deviation (n = 9). Therefore, the final study population consisted of 40 patients.

In order to compare contrast medium amount, renal function, and frequency of endoleak patients, we included 40 consecutive patients who underwent CT of the aorta using the conventional method between February and October 2019. Renal function after the CT exam was not routinely collected as a clinical practice unless the patient was suspected of renal dysfunction.

A 20-gauge catheter was placed at the right antecubital vein and contrast medium was injected using a power injector (Dual Shot GX7; Nemoto Kyorindo, Tokyo, Japan). A spiral flow tube (Nemoto spiral flow; Nemoto Kyorindo, Tokyo, Japan) was used as a connecting tube for saline flush.⁸

CT data acquisition: DRTB method

All scans were performed using a 64-row CT (Somatom Definition AS+; Siemens Healthineers, Forchheim, Germany). First, a non-enhanced scan from the chest to pelvis was performed with the following parameters: tube voltage, 80 or 100 kVp; reference mAs, 600 or 340 mAs; collimation, 64 × 0.6 mm; gantry rotation time, 500 ms; helical pitch 0.85. The tube voltage was set to 80 kVp when the maximal tube current did not exceed the limit using automatic exposure control with a helical pitch of 0.85. Otherwise, a 100 kVp scan was performed. The length of the aortic root to the top of the aortic arch (Figure 1(a), L0), chest scan length (Figure 1(a), L1), and the abdominal scan length (Figure 1(a), L2) were recorded.⁵

Figure 1.

Timing bolus scan was performed at the aortic root level (a, asterisk). Time density curves at the ascending and descending (b) aorta were drawn (c). The distance from the aortic root to the top of the aortic arch (a, L0), chest scan length (a, L1), and abdominal scan length (a, L2) were measured on non-enhanced images. The difference in time to peak between the ascending and descending aorta (c, T1), and the time to peak of the descending aorta (c, T2) were recorded using the time density curve. Three-dimensional image of the aorta (d) is acquired with 40 mL of contrast medium. Non-enhanced (e), arterial phase (f), and delayed phase (g) axial images show a type II endoleak (f, g, arrow).

A timing bolus scan was performed at the level of the aortic root with a tube voltage of 80 or 100 kVp and a current of 60 or 30 mAs. We used 7 and 9 mL of iopamidol 370 mg iodine/mL (Iopamiron 370; Bayer, Osaka, Japan) and a 30 mL saline flush, with an injection speed of 3.6 and 4.5 mL/s when the tube voltage was 80 and 100 kVp, respectively (Table 1). The monitoring scan started 12 s after injection of the contrast medium with an interval of 1 s. The reconstructed image was transferred to a workstation (Synapse VINCENT, ver 5.2; Fujifilm Medical, Tokyo, Japan) for time density curve analysis. Two ROIs were placed at the ascending and descending aorta (Figure 1(b)), respectively, and the difference in time to peak between the two curves (Figure 1(c), T1) and the time to peak of the descending aorta (Figure 1(c), T2) were recorded. The flow of the thoracic aorta (F1) was estimated as follows

F 1 = \frac{2 \times L 0}{T 1}

Table 1.

Contrast medium injection protocol.

	80 kVp		100 kVp
	Amount (mL)	Speed (mL/s)	Amount (mL)	Speed (mL/s)
Test injection	7	3.6	9	4.5
Main bolus
ΔDesc <100 HU	36	4.0	45	5.0
ΔDesc 100–175 HU	33	3.6	40	4.5
ΔDesc >175 HU	30	3.2	36	4.0

ΔDesc, peak enhancement at the descending aorta during test injection; HU: Hounsfield units.

As the flow of the abdominal aorta (F2) is slower than the thoracic aorta,⁹ we estimated F2 as follows

F 2 = 0.6 \times F 1

The peak enhancement of the descending aorta was also recorded. The total amount of contrast medium and the injection speed during aortic CT acquisition ranged from 30 to 45 mL and 3.2 to 5.0 mL/s, respectively, followed by a 50 mL saline flush (Table 1). This amount was determined based on the results of a previous study to achieve a target enhancement of 380 Hounsfield units (HU).⁵ The scan start timing was set as T2 and the helical pitch was adjusted to scan the aorta with the following scan time

Scan time = \frac{L 1}{F 1} + \frac{L 2}{F 2} + C

where C is a constant to adjust for abdominal stents (C = 1 s) and abdominal aortic aneurysms (C = 1 s for diameter 40–49 mm, C = 2 s for diameter ≥50 mm).⁵

Finally, a delayed phase scan was performed with the same scan parameters of the plain scan. The scan initiated 16 s after the end of the early phase.

CT acquisition: Conventional method

The acquisition parameters for conventional method were as follows: tube voltage, 120 kVp; reference mAs, 250 mAs; collimation, 64 × 0.6 mm; gantry rotation time, 500 ms; helical pitch 0.85. First, a non-enhanced scan was performed. Contrast medium with an iodine concentration of 370 mg iodine/mL (Iopamiron 370; Bayer) was used in all patients. The total amount was determined by the body weight (1.7×body weight (kg)) with a maximum amount of 100 mL. Contrast medium was injected for 25 s followed by a 30 mL saline flush. The aortic phase scan started 6 s after the descending aorta reached 150 HU above the baseline using bolus tracking method. Finally, a delayed phase scan was performed 90 s after the injection of contrast medium started.

CT analysis

Two radiologists (13 and 4 years of experience) independently assessed the presence of postoperative complication findings defined as following¹: endoleak, rupture, occlusion, and infection. When discrepancy existed between the readers, the final diagnosis was determined by consensus. The axial image data of the non-enhanced, early, and delayed phase images were used for analysis.

The maximal short axis of the aneurysm was measured in 30 patients with a follow-up CT. The measurement was performed at axial images of the baseline and follow-up CT. The difference in the diameter was also calculated.

Follow-up and endpoints

Follow-up information was obtained by clinical visits at 3-month intervals up to April 2019. All reported events were verified by hospital records or, if possible, by direct contacts with the primary physician. The primary endpoint was reintervention, which was defined as any of the following three events: conversion to open repair, graft revision, or secondary intervention.⁶ Graft revision was defined as a procedure that involved placement of a new endograft component to repair the primary implant. Secondary interventions were defined as procedures such as embolization for treating endoleaks and balloon angioplasty/stenting for limb kinking or thrombosis. The secondary endpoint was all-cause mortality.

Statistical analysis

Continuous variables are shown as means ± SD and categorical variables as number unless otherwise indicated. Student’s t-test was used to compare continuous variables. Fisher’s exact test was used to compare categorical variables. Kaplan–Meier survival analysis was used to estimate the distribution of time to events in different groups. Differences in time-to-event curves were compared with the log-rank statistic.

All statistical analyses were performed using JMP software (ver 12.2.0; SAS, Cary, NC). In all analyses, p <0.05 was taken to indicate statistical significance.

Results

Patient and scan characteristics

Indication for endovascular stent placement was repairing of abdominal aortic aneurysm in 39 patients (98%) and aortic occlusion in the remaining 1 patient (2%). CT was performed during the admission for aortic treatment with a median duration between aortic treatment and CT scan of 5.5 days (interquartile range, 5–8 days).

Male patients accounted for 83% and the mean age was 78.1 ± 7.3 years (Table 2). Most patients had hypertension (70%) and the presence of dyslipidemia (43%) and diabetes mellitus (8%) was lower. Mean estimated glomerular filtration rate (eGFR) was 67.3 ±26.1 mL/min/1.73 m² and 15 patients (38%) had chronic kidney disease (eGFR <60 mL/min/1.73 m²). Although we do not have data about renal function after the CT exam, no patients were reported/suspected of contrast-induced nephropathy. Histories of previous coronary artery bypass grafting and percutaneous coronary intervention were present in 4 (10%) and 10 (25%) patients, respectively. The tube voltage was 80 kVp in 10 patients (25%) and 100 kVp in 30 patients (75%). The mean contrast medium during CT angiography and the total dose including the test bolus were 38.6 ± 3.9 and 47.1 ± 4.7 mL, respectively.

Table 2.

Patient and scan characteristics.

Number of patients	40
Male	33 (83)
Age (years)	78.1 ± 7.3
Body weight (kg)	58.8 ± 11.7
BMI (kg/m²)	22.3 ± 3.5
eGFR (mL/min/1.73 m²)	67.3 ± 26.1
<60 mL/min/1.73 m²	15 (38)
Risk factor
PVD	5 (13)
Post CABG	4 (10)
Post PCI	10 (25)
EF <50%	8 (20)
Angina	3 (8)
Comorbidities
Hypertension	28 (70)
Diabetes	3 (8)
Dyslipidemia	17 (43)
Smoking (current/ex)	13 (33)
Implant
Aorfix	2 (5)
Endurant	26 (65)
Excluder	11 (28)
Epic	1 (2)
Tube voltage (kVp)
80	10 (25)
100	30 (75)
Contrast medium (mL)
CTA	38.6 ± 3.9
Test + CTA	47.1 ± 4.7
Injection speed (mL/s)	4.3 ± 0.5

Data are means ± SD or n (%).

BMI, body mass index; CABG, coronary artery bypass grafting; CTA, computed tomography angiography; EF, ejection fraction; eGFR, estimated glomerular filtration rate; PCI, percutaneous coronary intervention; PVD, peripheral vascular disease.

Complication findings at CT

Complication findings at CT were present in 10 patients (25%): endoleak (n = 9) and infection (n = 1) (Figures 2 and 3). The diagnosis of the two observers agreed in 38 patients (95%). Each observer missed one endoleak, which were determined as complication findings after consensus reading.

Figure 2.

CT images of a 76-year-old male patient with postoperative EVAR and F-F bypass. Three-dimensional (a) and axial non-contrast (b), early phase (c), and delayed phase (d) images show occlusion of the left limb and enhancement of the thickened adventitia. The total amount of contrast medium was 30 mL with 80 kVp acquisition. Infection of the graft was suspected, followed by open repair.

Figure 3.

CT images of a 72-year-old male patient with postoperative stent placement due to occlusion of the aorta. Three-dimensional image (a) shows residual stenosis at the distal iliac artery. Axial non-contrast (b), early phase (c), and delayed phase (d) images show endoleak around the stent (arrow). The total amount of contrast medium was 40 mL with 100 kVp acquisition. The patient developed total occlusion of the left limb 9 months after the initial CT (e, arrow).

Outcomes

The median follow-up was 7 months (interquartile range, 4–11 months). A total of two patients experienced reintervention: open repair (n = 1) (Figure 2) and secondary intervention (n = 1) (Figure 3). Both patients had complication findings at CT. No patients without complication findings at CT experienced reintervention (Table 3). Kaplan–Meier curves by complication findings showed that event rate at 6 months was significantly higher in patients with complication findings than in patients without (20% vs 0%, p = 0.01) (Figure 4(a)).

Table 3.

Post CTA reintervention and death cumulative incidence rate.

Outcomes	Complication finding at CTA
Outcomes	Yes	No	P
Reintervention	20% (2)	0% (0)	0.02^*
Mortality	10% (1)	0% (0)	0.09

Data are rate (number of events).

*Statistically significant, p <0.05.

CTA, computed tomography angiography.

Figure 4.

Kaplan–Meier curves for reintervention (a) and all-cause mortality (b) according to complication findings at CT.

One patient with an endoleak at CT scan died due to sepsis 8 months later. No patient died without complication findings at CT (Table 3). Kaplan–Meier curves by complication findings showed that the mortality rate at 6 months did not significantly differ between patients with and without complication findings (10% vs 0%, p = 0.09) (Figure 4(b)).

Change in arterial diameter

The median duration between baseline and follow-up CT scans was 6 months (interquartile range, 4–10 months). No significant difference was observed between patients with and without complication findings at CT in baseline (39.6 ± 10.7 vs 42.3 ± 9.9 mm, p = 0.53) and follow-up (39.9 ± 11.2 vs 39.9 ± 8.8 mm, p = 0.99) diameters (Table 4). The diameter difference between scans did not show significant difference (0.3 ± 1.4 vs –2.4 ± 3.9 mm, p = 0.07).

Table 4.

Change in arterial diameter by complication finding at CTA.

Arterial diameter (mm)	Complication finding at CTA
Arterial diameter (mm)	Yes	No	P
Baseline	39.6 ± 10.7	42.3 ± 9.9	0.53
Follow-up	39.9 ± 11.2	39.9 ± 8.8	0.99
Difference	0.3 ± 1.4	–2.4 ± 3.9	0.07

Data are means ± SD.

CTA, computed tomography angiography.

Comparison with conventional method

The injected contrast medium was significantly higher using the conventional method than the DRTB method (84.9 ± 12.6 vs 38.6 ± 3.9 mL, p <0.001). Estimated GFR of DRTB and control groups were 67.3 ± 26.1 and 62.1 ± 19.2 mL/min/1.73 m², respectively, without significant difference (p = 0.34). Endoleak was detected in three patients using the conventional method. The frequency was lower than patients in the DRTB group (22.5% vs 7.5%) but the difference was not significant (p = 0.11). The diagnosis of endoleak agreed in all patients using the conventional method.

Discussion

Reducing the amount of contrast medium during CT angiography is important to prevent contrast-induced nephropathy because patients with aortic disease occasionally have renal disorders.² According to the European Society of Urogenital Radiology Guidelines on Contrast Agents, eGFR <45 mL/min/1.73 m² before intra-arterial and eGFR <30 mL/min/1.73 m² before intra-venous contrast medium injection is a risk factor of acute kidney injury.¹⁰ Although no patients experienced contrast-induced nephropathy in the present cohort, it is important to use the lowest dose of contrast medium consistent with a diagnostic result.¹⁰ Concerns might remain for overlooking important complications after stent treatment of the aorta such as endoleaks. In this study, endoleak was detected in 23% of the patients and graft infection was diagnosed in one patient despite the small amount of contrast medium of 39 mL. Importantly, no patients without complication findings at CT experienced adverse events during a median follow-up duration of 7 months. Although no direct comparison of the frequency to detect endoleak was performed between DRTB and conventional methods, the frequency of endoleak using DRTB method was not inferior to conventional method.

Diagnosis of endoleaks is important because some types of endoleaks, especially types I and III, might cause severe adverse events and need treatment. CT is an established method to diagnose complications after stent treatment.¹ Typical CT protocols include three phases: non-enhanced, early phase, and delayed phase. Most endoleaks appear at early phase scans, but some endoleaks could only be detected at delayed phase scans.¹¹ Moreover, the typing of endoleaks is occasionally difficult because the sampling rate of CT scan is much lower than the rate at fluoroscopy during invasive angiograms. Dynamic volumetric CT, which scans the site of stent placement in multiple phases (12–16 phases), might help to improve the confidence rate of the typing of endoleaks.¹² However, the short scan coverage with a maximum of 16 cm is a shortcoming of this protocol. The optimal sampling rate and scan range for diagnosis for complications need further investigation.

Majority of endoleaks appear early after intervention, but some endoleaks occur in delayed periods.^13,14 A previous study showed that 13.6% of patients after stent placement of the aorta experienced delayed endoleaks (≥12 months after intervention) and non-type II was dominant.¹³ Late type II endoleak (>30 days after intervention) occurred in 17% of patients, out of which 22% of patients had sac expansion.¹⁴ The follow-up period of 7 months in the present study might be short, but the guarantee period of no complication findings at CT is not fully assessed.

Low–low CT protocol, which is a combination of contrast medium dose reduction with low tube voltage scanning, has become available with the advance in CT technologies. A split bolus technique which divides the contrast bolus to perform a single-phase acquisition during a mixed early and delayed phase could further reduce the radiation dose by omitting multiple-phase scanning.¹⁵ Dual-energy CT is another approach to reduce the amount of contrast medium, while maintaining the image quality.^4,16 As much as 70% reduction in iodine dose with a total amount of 15 g iodine could be achieved using dual-energy CT.⁴ One major strength of this technique is that the appropriate monochromatic image could be reconstructed out of multiple energy levels.¹⁶ Another strength is that metal artifacts from stents might be reduced using monochromatic images.¹⁷ Combination of dual-energy CT and DRTB method might further reduce the amount of contrast medium necessary for CT of the aorta.

We acknowledge the following limitations. First, this was a single center study and further multicenter study is necessary to confirm the results. Second, the number of patients included in this study and the number of events were small. Therefore, we could not calculate the hazard ratio for complication findings at CT. Third, the follow-up duration was short, but complications after stent placement might occur at delayed periods.

In conclusion, no complication findings at CT with a mean contrast medium injection of 39 mL using DRTB method after intervention of the aorta resulted in good prognosis during a 7-month period. DRTB method would help to reduce the contrast medium dose during CT scan to assess complications after stent placement of the aorta.

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.

ORCID iDs

Nobuo Tomizawa

Kodai Yamamoto

References

Ilyas

Shaida

Thakor

, et al. Endovascular aneurysm repair (EVAR) follow-up imaging: The assessment and treatment of common postoperative complications. Clin Radiol 2015; 70: 183–196.

Vasconcelos

Vrtiska

Foley

, et al. Reducing iodine contrast volume in CT angiography of the abdominal aorta using integrated tube potential selection and weight-based method without compromising image quality. Am J Roentgenol 2017; 208: 552–563.

Wei

Gao

, et al. Use of low tube voltage and low contrast agent concentration yields good image quality for aortic CT angiography. Clin Radiol 2016; 71: 1313.e5–1313.e10.

Shuman

O’Malley

Busey

, et al. Prospective comparison of dual-energy CT aortography using 70% reduced iodine dose versus single-energy CT aortography using standard iodine dose in the same patient. Abdom Radiol (NY) 2017; 42: 759–765.

Tomizawa

Ito

Nakao

, et al. Double ROI timing bolus technique to perform aortic CT angiography with a 9-second contrast injection duration. Am J Roentgenol 2019; 213: 1–8.

Hye

Janarious

Chan

, et al. Survival and reintervention risk by patient age and preoperative abdominal aortic aneurysm diameter after endovascular aneurysm repair. Ann Vasc Surg 2019; 54: 215–225.

Lal

Zhou

, et al. Predictors and outcomes of endoleaks in the Veterans Affairs Open Versus Endovascular Repair (OVER) Trial of Abdominal Aortic Aneurysms. J Vasc Surg 2015; 62: 1394–1404.

Tomizawa

Hayakawa

Inoh

, et al. Spiral flow tube for saline flush in coronary CT angiography: Initial experience. Int J Radiol 2016; 3: 90–94.

Bae

KT.

Intravenous contrast medium administration and scan timing at CT: Considerations and approaches. Radiology 2010; 256: 32–61.

10.

Thomsen

HS.

ESUR guidelines on Contrast Agents, version 10. Eur Soc Urogenit Radiol 2018; 10: 0–45.

11.

McDonald

Paik

Fairman

, et al. Outcomes of type 2 endoleaks detected on venous phase CT arteriography. Diagn Interv Imaging 2018; 99: 225–229.

12.

Hou

Zhu

Zhang

, et al. Dynamic volumetric computed tomography angiography is a preferred method for unclassified endoleaks by conventional computed tomography angiography after endovascular aortic repair. J Am Heart Assoc 2019; 8: 16.

13.

Park

K-M

Jeon

, et al. Evaluation of delayed endoleak compared with early endoleak after endovascular aneurysm repair. J Vasc Interv Radiol 2018; 29: 203–209.

14.

Kumar

Cowled

Boult

, et al. Type II endoleak after endovascular aneurysm repair: Natural history and treatment outcomes. Ann Vasc Surg 2017; 44: 94–102.

15.

Böning

Rotzinger

Kahn

, et al. Tailored CT angiography in follow-up after endovascular aneurysm repair (EVAR): Combined dose reduction techniques. Acta Radiol 2018; 59: 1316–1325.

16.

Liu

, et al. Feasibility of spectral imaging with low-concentration contrast medium in abdominal CT angiography of obese patients. Int J Clin Pract 2016; 70: B37–B43.

17.

Boll

Merkle

Paulson

, et al. Coronary stent patency: Dual-energy multidetector CT assessment in a pilot study with anthropomorphic phantom. Radiology 2008; 247: 687–695.