Correlation between the Measurement of Transverse Diameter in the Proximal Neck on Computed Tomography and on Aortography before Endovascular Treatment of Infrarenal Aortic Aneurysm
Article Outline
The aim of this study was to determine the correlation between the measurement of transverse diameter of the proximal neck on computed tomographic angiography (CTA) and graduated catheter aortography in patients who are candidates for endovascular graft placement in order to replace, if both measurements are equivalent, aortography for CTA alone. Preoperative dual-slice CTA and graduated catheter aortography were performed in 35 consecutive patients with infrarenal aortic aneurysm within 10 days. Transverse proximal neck diameters were measured on a true axial section on CTA reconstructions and on aortographic images, always 6 mm distal from the most inferior main renal artery. Mean, median, and standard deviation were obtained and the measurements correlated for each patient using Pearson's correlation and linear regression analysis. A significant difference in proximal neck transverse diameter measurements was found between graduated catheter aortography and CTA in all cases. CTA values were a mean of 1.74 mm higher than aortography values. Pearson's correlation indicates a strong correlation between both techniques, and a regression equation determines the predictive value of aortography on the basis of CTA values. Estimation of the transverse diameter of the proximal neck on aortography on the basis of that obtained on CTA allows us to affirm that CTA could be used as the sole method for the preoperative selection of appropriate endograft size in patients with infrarenal aortic aneurysm.
INTRODUCTION
An abdominal aortic aneurysm is a focal irreversible enlargement of the abdominal aorta, >3 cm in its largest transverse diameter, that involves its whole wall.1 It is the most frequent aortic pathology and affects approximately 5% of males between 65 and 75 years old.2 The natural evolution of an abdominal aneurysm is its continued expansion, with the consequent increase of the risk of rupture. The mortality of this complication is as high as 90%. That is why when the aneurysm exceeds 4.5 or 5 cm in diameter, its prophylactic treatment is recommended.3
Surgical repair by exposure of the aortic abdominal aneurysm from an anterior approach is the traditional treatment. Nevertheless, in 1991, Parodi et al.4 described the endovascular repair of an infrarenal abdominal aortic aneurysm for the first time. It consists of introducing a catheter via a small vascular access remote from the aneurysm, usually a surgical femoral cutdown. This catheter supports a self-expandable stent graft, which is deployed, with the use of imaging guidance, into the appropriate position in the aorta. In the beginning, this percutaneous technique was performed only in patients with high surgical risk, but currently it is increasingly becoming a less invasive alternative to open surgery.5, 6 Endovascular repair has been demonstrated to be as successful as conventional open surgical repair in exclusion of abdominal aortic aneurysms from the arterial pressure. The mortality derived from both techniques is also similar, even if we take into account that these patients have a higher associated comorbidity, which is usually the main reason for the use of this form of treatment instead of open surgery.7
Once the aneurysm is diagnosed, its extension and progressive enlargement must be evaluated with ultrasound, computed tomography (CT), or magnetic resonance imaging (MRI).5, 8 At our institution, when endovascular repair is decided, the preoperative evaluation of candidates includes a contrast-enhanced CT scan or contrast-enhanced MR angiography, as well as angiography using a graduated pigtail catheter that has markers over a 20–25 cm distance in order to select the right size of the endograft.6
Before implantation of the stent graft, some anatomical features of the aorta and the iliac arteries must be evaluated, such as their length and diameters, to decide the proper vascular access and the design of the stent graft.8 The required measurements are represented in Figure 1: the maximum diameter of the aneurysm and its length, the distance between the most inferior main renal artery and the iliac bifurcation, the length and diameter of the common iliac arteries, the diameter of the external iliac arteries and the length and diameter between the most inferior main renal artery and the origin of the aneurysm.9, 10 This is the proximal neck.
Fig. 1.
Preoperative measurements. A a, b Length and diameter of the proximal neck. (c, d) Length and diameter of the aneurysm. (e, f) Common iliac artery diameters. B g Distance from the most inferior renal artery to the bifurcation. C h Common iliac arteries' length.
The accuracy of these quantitative dimensions is crucial for the successful outcome of endovascular treatment because selection of the appropriate stent-graft diameter and length is based on them.10, 11 Two of the most important measurements are the diameters of the proximal and distal necks, where the stent graft will be anchored. They are essential to ensure appropriate exclusion of the aneurysm and prevent its rupture.8, 11 Long-term success of the procedure will depend on the graft remaining correctly placed in the distal and proximal attachment sites.11
A minimum length for the proximal neck of 10–15 mm,9 although some authors10 require 15–25 mm, and a maximum width of 26–32 mm8, 10 are recommended, depending on the device being used.
The reference standard technique to obtain an accurate measurement of the aortic dimensions has been aortography with a calibrated catheter with radiopaque markers at 1 cm intervals. Even though the validity of preoperative evaluation with CT has been recently reported,12 the endovascular repair planning must be much more accurate because errors in measurements as small as 2 mm can lead to important complications.6, 13
Specific software based on CT and MRI images has been recently developed and demonstrated to be superior to angiography in the selection of the appropriate endograft length and in the evaluation of iliac access. In some cases, these methods have been used as the sole imaging procedures in the planning of endovascular treatment.13 The accuracy of the measurements has also been proven with a method that determines the center of the vascular lumen and calculates in a semiautomatic way the diameter and length of an endograft of previously known dimensions.14
The accuracy among the parameters obtained with different imaging techniques, including intravascular ultrasonography, digital subtraction angiography, and helicoidal CT, has been compared.15 Contrast-enhanced MRI and multislice CT angiography with a specific measurement software program have been compared by other studies.16
The evolution of the proximal neck diameter over time, to assess whether it enlarges or not after endograft implantation, has been evaluated in some reports.17, 18, 19, 20 However, we have found only one study21 that compared the measurements obtained with CT angiography to those obtained with calibrated catheter aortography. This study is based on seven patients and three phantoms, with variable aortic tortuosity. There are few references that compare the measurement of the proximal neck transverse diameter between the two techniques.
As previously noted, the accuracy of this measurement is crucial because once the selected device is released into the aorta it can be neither captured nor replaced.6
Attachment of the endograft to the aortic wall is done by contact, with fixation of either autoexpandable devices or ballon-dilatation devices, which is different from surgical anastomosis. The stent graft must have enough pressure to remain anchored and prevent blood flow between it and the aortic wall but not so much as to produce rupture of it.22
The aim of this study was to assess the correlation between measurements of the proximal neck obtained with calibrated catheter angiography and dual-slice helical CT angiography in patients who are candidates for endovascular aortic graft placement. If the measurements obtained by these techniques are equivalent or have a constant relationship, then selection of the appropriate endograft could be based exclusively on preoperative CT angiography. This way, angiography, which is an invasive technique and undoubtedly has some risks, could be replaced, at least in some cases, by CT.
MATERIALS AND METHODS
Patients
In the last 2 years, 210 patients have been treated with endovascular repair at our institution. Most of them belong to the fifth sanitary area of the community of Madrid. All of them were evaluated by both vascular surgeons and radiologists from La Paz University Hospital. We performed a prospective study with 35 of these patients, who were candidates for endograft implantation, including a preoperative dual-slice CT angiography and calibrated digital substraction aortography. The study lasted approximately 5 months.
Only those patients who underwent endovascular infrarenal abdominal aneurysm repair as the treatment of choice and had both imaging techniques performed within 10 days of each other were included. All patients had an aneurysm diameter >4.5 cm.
Thoracic aneurysms, thoracoabdominal aneurysms, and urgent endovascularly repaired aneurysms, as well as those patients from any other institutions who had a previous CT angiography, were excluded.
Dual-Slice Helical CT Protocol
CT scans were performed with a commercially available CT scanner (Asteion, Toshiba, Tokyo, Japan) that is equipped with a double array of detectors. Z-axis coverage ranged from the level of the adrenal gland region to the bifurcation of the femoral arteries. A total of 120–150 mL of nonionic 30% contrast material was administered intravenously through an antecubital vein with an automated injector at a rate of 3–5 mL/sec. Acquisition was automatically performed with the sure-start technique, and the threshold to start scanning was set at 140 UH in the proximal abdominal aorta. Craniocaudal helical CT was performed in a single breath-hold with the following parameters: 120 kVp, 200 mAs, 3 mm (x2) collimation; table speed of 12 mm per rotation; and pitch of 4. Scanning time ranged 18-23 sec and the length of the acquisition ranged 250-400 mm. Images were in a range of 150–250 and reconstructed using an independent workstation (Vitrea, Vital Images, Minnetonka, Minnesota). Multiplanar reformations (MPRs) and three-dimensional (3D) and maximum-intensity-projection (MIP) reconstructions were performed for each case. In cases with an important anterior or lateral aortic angulation, the measurement was performed in an oblique or double oblique reconstruction in order to obtain the true axial section (Fig. 2).
Fig. 2.
Double oblique reconstruction to obtain the true axial section.
Calibrated Aortographic Technique
Digital subtraction aortographic (DSA) examinations were performed via a transfemoral artery approach using a catheter with contrast (Integris 3000, Philips, Zoeken, Netherlands), injected in the abdominal aorta for a total of 30 mL per series. Anteroposterior and lateral projections from the renal arteries to the level of the hypogastric arteries were obtained in each patient.
Data Analysis
For CTA image analysis, one experienced radiologist measured the transverse proximal neck diameter in the true axial sections, with a scale of 1:150. Measurements were obtained from inner wall to inner wall, at a distance of 6 mm from the most inferior main renal artery. Accesory renal arteries were not considered (Fig. 3).
Fig. 3.
Transverse diameter of the proximal neck.
For calibrated catheter aortographic analysis, measurements were performed by one experienced radiologist on a magnified radiological image (Fig. 4).
Fig. 4.
Calibrated-catheter digital substraction aortography.
Statistical Analysis
A pilot study of 35 patients was performed to estimate the preliminary values of the correlation between the two techniques. The results have a power of >90% at p < 0.05 to indicate significant correlation (r = 0.9). Data were analyzed with the statistical software program SPSS 9.0 (SPSS Inc., Chicago, IL).
Qualitative data were described as absolute frequencies and relative percentages and quantitative values as mean, median, and standard deviation (SD; minimum, maximum) depending on their distribution.
For comparison of the measurements obtained by the two techniques, Student's t-test for paired data was applied.
Pearson's correlation was used to assess the strength of the relationship between the two measurements in each case, and a linear regression analysis was applied to estimate the diameter on angiography based on the value of CT angiographic (CTA) measurements. All statistical techniques were considered bilateral and significant at p < 0.05.
RESULTS
Measurements in centimeters of the proximal neck diameters from both preoperative CTA and DSA are shown in Table I.
Table I. Comparative values on CT and aortography
| CTA (mm) | Aortography (mm) |
|---|---|
| 20.7 | 20.2 |
| 32.5 | 30 |
| 22.9 | 22 |
| 20.7 | 20 |
| 17 | 15 |
| 17 | 16 |
| 22.3 | 17 |
| 23.5 | 16 |
| 25.2 | 23 |
| 18.2 | 19 |
| 22 | 20 |
| 17.4 | 18 |
| 22.4 | 21 |
| 21.4 | 22 |
| 21.1 | 20 |
| 22 | 17 |
| 24 | 22 |
| 19.2 | 19 |
| 30.7 | 30 |
| 22 | 20 |
| 26 | 22 |
| 21.7 | 18 |
| 23 | 20 |
| 20 | 19 |
| 26.8 | 22 |
| 24.8 | 27 |
| 22.3 | 22.5 |
| 20 | 20 |
| 28 | 25 |
| 23 | 20 |
| 23.5 | 22.8 |
| 36 | 35 |
| 27 | 25 |
| 28.5 | 28.6 |
| 31.2 | 29 |
The mean (±SD) transverse proximal neck diameters were, by CTA, 23.5 (± 4.4) mm (median 21.4, range 17-36) and, by DSA, 21.8 (± 4.5) mm (median 20.2, range 15-35) (Table II). The values for the two techniques were significantly different, with p < 0.01, which exceeds the interobserver error. A mean difference of 1.74 mm was found between the two techniques [95% confidence interval (CI) 1.06-2.4, p < 0.001] and was higher for CTA in all cases except one.
Table II. Quantitative values on CT and aortography
| Axial CTA | Aortography | |
|---|---|---|
| Mean | 23.5 | 21.8 |
| SD | 4.4 | 4.5 |
| Median | 22.4 | 20.2 |
| Minimum | 17.00 | 15.00 |
| Maximum | 36 | 35 |
| n | 35 | 35 |
The correlation based on Pearson's coefficient for paired data was 0.901, which indicates a strong correlation between measurements for the two techniques.
The linear regression analysis obtained to estimate DSA values based on CTA measurements is as follows: A = 0.04 + (0.924 × CTA) where CTA is the neck diameter on CTA, 0.04 is a constant parameter, 0.924 is the correction factor, and A is the neck measurement on DSA estimated on the basis of a measurement known from CTA.
DISCUSSION
All patients in our study were males, as occurs in most series, since abdominal aortic aneurysm prevalence is higher for males than for females.2
In our results, different measurements were obtained from CTA and DSA in all cases. These differences are statistically significant with a power >90%.
In most cases, higher measurements are obtained on CTA than on DSA, with a mean value of 1.74 mm higher on CTA, which exceeds the interobserver error. This result indicates a percentage difference of approximately 10%, higher on CTA. Whether CTA values are magnified when compared to DSA values or more accurate is a fact that can be proven only by comparing them with a stent graft of a previously known size. What we have found is that in all cases CTA values are more similar to the real size of the implanted endograft than DSA values.
Lutz et al.16 assessed the accuracy and reliability of volumetric measurement with an automated analysis software tool in the preoperative determination of all aortic dimensions on the basis of CTA and MRA data sets.
DSA underestimates the neck diameter in the same way as it does with length. This occurs because DSA measures the vessel lumen alone, unlike CTA, which measures from wall to wall including a possible thickness of the wall if it exists.14 In fact, it does not seem probable that CTA magnifies the measurements, taking into account that the measured segment must be a normal and undilated aortic portion.
For the proximal neck diameter, Broeders et al.23 found that graft sizes would have been too small in 62% of cases if they had been determined only with intra-arterial angiography.
Isokangas et al.14 compared the dimensions obtained from CTA with those of two phantoms of known dimensions. The mean fractional error for all diameter measurements combined was 0.017 ± 0.011. Although this error is slightly superior for the transverse diameter than for the length (0.009 ± 0.0006), in view of the stent-graft sizing, errors of <1% in length measurement are quite acceptable and do not affect device selection.
In studies where both preoperative measures are obtained, CTA is preferred for the transverse neck diameter, whereas DSA is considered a better option for the length by some authors.6 Nevertheless, CTA can measure the length within the exact center of the artery, whereas length measured with aortography depends on the location of the marker catheter and, subsequently, on the specific morphology of the aorta. CTA also allows the selection of a true axial section from an oblique or double oblique reconstruction in the aorta direction.
We must keep in mind that an undermeasurement of the length can be corrected during the endovascular repair by enlarging the endograft but an undermeasurement of the transverse diameter has no possible solution.
Wyers et al.13 used CTA as the sole preoperative imaging technique in the repair planning of abdominal aortic aneurysm: 196 patients were followed for a mean of 18 months and showed no higher incidence of long-term complications, with in-hospital mortality of zero. These authors13 affirm that the endograft length prediction and the iliac access are superior and more accurate on CTA than on DSA.
The stent-graft size is assumed to be approximately 20% more than the external diameter of the vessel. This way a good seal and attachment is achieved and primary endoleaks due to incomplete seal, endograft migration secondary to the sac enlargement, and neck expansion are prevented.6, 11
We believe that the endograft could be selected using only CTA measurements. The DSA value can be calculated from the linear regression equation (A = 0.04 + [0.924 × CTA]). This equation estimates the measurement that should have been obtained on DSA based on the high correlation between the measurements on both techniques and could allow the elimination of preoperative arteriography. The appropriate endograft size could then be chosen based on CTA values, and approximately only a 10% oversized graft should be implanted.
There is evidence of a progressive enlargement of the neck over time.17, 18 It is currently unclear which are the causal factors of this enlargement. They may be the overestimation of the endograft or the natural history of the abdominal aortic aneurysm.20 Even May et al.24 affirm that endovascular repair prevents dilatation of the proximal neck. What seems probable is that excessive overestimation could lead to rupture due to an accelerated expansive force to reach the endograft size.
CTA is believed to be a reliable method in the postoperative control of these patients.25, 26 CTA is also superior to DSA in the evaluation of complications and follow-up of endografts.
This way, preoperative DSA, which is an invasive technique and undoubtedly has some risks, could be used only in intraoperative endovascular repair control.15
Multislice CTA is becoming more widespread as the sole preoperative method in the anatomical evaluation15 and in the assessment of abdominal aortic aneurysm dimensions.13 As it provides identical resolution reconstructions in all planes, the coronal reconstruction can be equivalent to the DSA obtained during the procedure. Estimation of the transverse diameter on DSA on the basis of that obtained on CTA allows us to affirm that CTA could be used as the sole method, or at least a reliable method, for the preoperative selection of the appropriate endograft size, as it is currently in the postoperative control and diagnosis of complications.5, 27
A larger patient population would be needed in order to increase the power of Pearson's correlation and, subsequently, improve the regression equation precision.
REFERENCES
- . Normal aortoiliac diameters by CT . J. Comput. Assist Tomogr . 1988;12:602–603
- . Oxford screening program for abdominal aortic aneurysm in men aged 65 to 74 years . Lancet . 1988;10:613–615
- Changing incidence of abdominal aortic aneurysms: a population-based study . Am J Epidemiol . 1984;120:379–386
- . Transfemoral intraluminal graft implantation for abdominal aortic aneurysms . Ann Vasc Surg . 1991;5:491–499
- . Endovascular repair of abdominal aortic aneurysms . AJR Am J Roentgenol . 2001;177:607–614
- Endovascular repair of abdominal aortic aneurysms: current status and future directions . AJR Am J Roentgenol . 2000;175:289–302
- Concurrent comparison of endoluminal versus open repair in the treatment of abdominal aortic aneurysms: analysis of 303 patients by life table method . J Vasc Surg . 1998;27:213–221
- . Imaging of abdominal aortic aneurysm before and after endoluminal stent-graft repair . Eur J Radiol . 2001;39:3–15
- . Selecting stent grafts for the endovascular treatment of abdominal aortic aneurysms . J Cardiovasc Surg . 2003;44:511–518
- Endovascular aortic aneurysm operations . Arch Surg . 2002;137:211–216
- . The choice of stent-graft for endovascular repair of abdominal aortic aneurysm . J Cardiovasc Surg . 2003;44:519–525
- Abdominal aortic aneurysm: pretherapy assessment with dual-slice helical CT angiography . AJR Am J Roentgenol . 2000;174:181–187
- Endovascular repair of abdominal aortic aneurysm without preoperative arteriography . J Vasc Surg . 2003;38:730–738
- . Accuracy of computer-aided measurements in endovascular stent-graft planning: an experimental study with two phantoms . Invest Radiol . 2003;38:164–170
- Prediction of aortoiliac stent-graft length: comparison of measurement methods . Radiology . 2001;220:475–483
- Evaluation of aortoiliac aneurysm before endovascular repair: comparison of contrast-enhanced magnetic resonance angiography with multidetector row computed tomographic angiography with an automated analysis software tool . J Vasc Surg . 2003;37:619–626
- Aneurysm neck diameter after endovascular repair of abdominal aortic aneurysms . J Vasc Interv Radiol . 2002;13:887–892
- . Is proximal aortic neck dilatation after endovascular aneurysm exclusion a cause for concern? . J Vasc Surg . 2001;33:39–45
- . Imaging of abdominal aortic aneurysms after endoluminal repair . Semin. Ultrasound CT MR . 1999;20:16–24
- . Dilatation of the infrarenal aneurysm neck after endovascular exclusion of abdominal aortic aneurysm . J Endovasc Surg . 1998;5:195–200
- . Prediction of aortoiliac stent graft length: comparison of a semiautomated computed tomography angiography method and calibrated aortography . J Comput Assist Tomogr . 2003;27:284–288
- Leakages after endovascular repair of aortic aneurysms: classification based on findings at CT, angiography, and radiography . Radiology . 1999;213:767–772
- . Preoperative sizing of grafts for transfemoral endovascular aneurysm management: a prospective comparative study of spiral CT angiography, arteriography, and conventional CT imaging . J Endovasc Surg . 1997;4:252–261
- . Endoluminal repair of abdominal aortic aneurysm prevents enlargement of the proximal neck: a 9-year life-table and 5-year longitudinal study . J Vasc Surg . 2003;37:86–90
- . Endovascular repair of abdominal aortic aneurysm: value of postoperative follow-up helical CT . AJR Am J Roentgenol . 1995;165:1473–1479
- Analysis of initial CT findings after endovascular repair of abdominal aortic aneurysm . Radiology . 2001;220:157–160
- . Imaging of aortic stent-grafts and endoleaks . Radiol Clin North Am . 2002;40:799–833
PII: S0890-5096(06)61467-2
doi:10.1007/s10016-006-9077-0
© 2006 Annals of Vascular Surgery, Inc. Published by Elsevier Inc All rights reserved.
