Prevalence and significance of extravascular incidental findings on computed tomographic angiography and magnetic resonance angiography

Abstract

Computed tomographic angiography (CTA) and magnetic resonance angiography (MRA) are routinely used to evaluate patients with vascular disease. They have the ability to detect unexpected non-vascular pathology. The purpose of this study was to determine the prevalence and significance of extravascular incidental findings in patients undergoing CTA or MRA. A retrospective review of 737 patients who underwent CTA and 184 patients who underwent MRA during a five-year period was performed. Incidental findings were classified as low, moderate or high significance findings. For patients with high significance extravascular findings, assessment of the rates of appropriate follow-up was conducted. Among the CTA patients, 539 (73.1%) had incidental findings. Low, moderate and high significance findings were discovered in 514 (69.7%), 95 (12.9%) and 41 (5.6%) patients, respectively. Twenty (48.8%) patients with high significance findings received appropriate follow-up investigations. Among the MRA patients, 95 (51.6%) had extravascular findings. Low, moderate and high significance findings were present in 80 (43.5%), 27 (14.7%), and 3 (1.6%) patients, respectively. Two (66.7%) patients with high significance findings were properly followed up. In conclusion, incidental findings on CTA and MRA are very common. A small percentage of these findings could be serious and were not all adequately followed-up in our study population. Referring physicians should be aware of the potential for serious incidental findings and manage them appropriately.

Keywords

computed tomographic angiography magnetic resonance angiography incidental finding peripheral arterial disease aneurysm

Introduction

Computed tomographic angiography (CTA) and magnetic resonance angiography (MRA) have evolved into effective and robust imaging modalities for the evaluation of aneurysmal disease, peripheral arterial occlusive disease and arterial–venous malformations. They are increasingly used for the evaluation of these patients. Both CTA^1–4 and MRA^5,6 have been shown to have diagnostic performance comparable with that of conventional digital subtraction angiography. In addition, these tests are non-invasive, have shorter exam time, do not require postprocedural monitoring and involve a lower radiation exposure.⁷ Multiplanar and three-dimensional postprocessing is possible allowing for improved assessment of complex vascular pathology and these techniques allow for evaluation of the vessel wall, as well as the lumen.

In addition to their abilities to evaluate vascular pathology, a further advantage of CTA and MRA is their ability to detect incidental extravascular findings, which refer to unanticipated observations that are unrelated to the original purpose of the scan. Appropriate follow-up of important incidental findings on CTA and MRA may lead to informed treatment decisions that significantly impact patients’ wellbeing. Previous studies of patients undergoing CT following abdominal aortic aneurysm repair,⁸ CT for renal colic,⁹ CT colonography for colon cancer screening,¹⁰ coronary CTA^11–14 or CTA of the abdominal aorta and its branches^15–20 have shown varied prevalence of incidental findings. To the best of our knowledge, with the exception of renal MRA,²¹ there have been no other studies to date that examined incidental findings on MRA studies. None of these studies formally evaluated the follow-up of these incidental findings. The purpose of this study was to determine the incidence and significance of extravascular incidental findings on peripheral vascular CTA and MRA and assess the adequacy of follow-up for important incidental findings.

Materials and methods

Study population

A retrospective review was conducted on all peripheral vascular CTA and MRA studies performed between March 2006 and May 2011 at our multisite, tertiary, academic medical center. Vascular territories imaged included the thoracic and abdominal aorta, visceral arteries, carotid arteries and upper and lower extremity arteries. Intracranial CTA and coronary CTA were excluded. All other studies were included regardless of patient age, gender or indication for the scan. The CTA and MRA examinations were initially interpreted by one of three fellowship trained vascular and interventional radiologists with 4–10 years of experience interpreting vascular imaging studies. The study was approved by the institutional research ethics board.

Technique of CTA and MRA

Computed tomographic angiography

All CTA studies were performed on a 64-channel multidetector row CT system (Aquilion 64, Toshiba Medical Systems, Tokyo, Japan). For the abdominal aorta and lower extremities, contrast-enhanced images were performed from suprarenal abdominal aorta to the toes with 0.5 mm collimation, gantry rotation time of 0.5 seconds, helical pitch factor of 0.828, 120 kVp and automatic tube current modulation. Images were reconstructed to both 3 and 1 mm slice thickness. In patients with prior endovascular repair of abdominal aortic aneurysm (EVAR), an additional precontrast scan with 5 mm slice thickness and a delayed scan with 5 mm slice thickness acquired two minutes after initial scan were also performed.

In all patients, isosmolar, non-ionic iodinated cotrast material (Visipaque 320 mgI/mL, GE Healthcare, Milwaukee, WI, USA) was injected into an antecubital vein with a power injector (Medrad, Philadelphia, PA, USA) at a flow rate of 4 mL/second. This was followed with a 30 mL normal saline bolus at the same rate. Scan delay was individualized per patient, using bolus tracking software set to trigger the scan when the attenuation of the aorta at the level of the renal arteries reached 180 Hounsfield unit (HU). Only 120 mL of contrast was used for CTA runoff examinations of the upper and lower extremities and 80–100 mL was used for examinations of the abdominal aorta. Only 80 mL of contrast was used for carotid CTA.

MRA

All MRAs were performed on a 1.5-T MR system (Magnetom Avanto, Siemens, Erlangen, Germany) using standard 3D spoiled fast gradient-recalled echo sequences after bolus administration of intravenous gadolinium chelate (Gadovist, Bayer, Germany). Contrast dose was 0.2 mmol/kg and the contrast was injected at a rate of 1.5 mL/second via an magnetic resonance imaging-compatible power injector (Medrad Spectris, Medrad, Philadelphia,PA) and followed by a 20-mL normal saline bolus, injected at the same rate. Scans were initiated by the MR technologist, using fluoroscopic triggering when the contrast bolus was observed to arrive in the area of interest. Venous phase imaging was routinely performed using a 3D fat suppressed spoiled gradient echo sequence (VIBE, Siemens). All imaging of the chest and abdomen was performed with a breath hold. Specific imaging parameters were optimized for the body part and indication.

Data collection

Consecutive studies were identified by a search of the radiology information system (RIS) for examinations described as CTA or MRA of the head and neck, thorax, abdomen, pelvis and/or extremities. The reports were then evaluated by one reviewer and the following information extracted: patient demographics, study indication, type of procedure, vascular findings and extravascular findings. Both the RIS and the electronic patient record were reviewed to determine if appropriate imaging follow-up, further investigation or treatment was undertaken when suggested in the reports. Letters were sent to referring physicians notifying them of the incidental findings and asking if any follow-up was performed in the case where none was found in the local hospital records. Cases for which a response was not received from the referring physician were counted as no follow-up obtained.

Terms and measures

Extravascular incidental findings refer to unanticipated observations that are unrelated to the original purpose of the scan. These incidental findings were classified as low significance, moderate significance or high significance. Low significance findings were defined as those with little or no clinical relevance such as renal cysts and pulmonary granulomas. Findings of moderate significance were defined as those that were clinically important and possibly benign, but in need of further work-up for definitive diagnosis such as an indeterminate adrenal lesion. High significance findings were defined as those with potentially serious clinical implications such as acute medical conditions requiring treatment or malignancies such as pancreatitis or hepatocellular carcinoma. See Figures 1 and 2 for examples of highly significant incidental findings. If an incidental finding did not clearly fit in one of these categories, it was reviewed by two of the authors (with 4 and 11 years of experience). Any finding for which a follow-up study was recommended by the reporting radiologist was automatically classified as moderate or high significance.

Figure 1

Incidental finding of bone metastasis. (a) A 61-yearold man with complex vascular disease undergoing follow-up imaging postaxillofemoral bypass. Axial CT image through the proximal femurs shows development of a new pathologic fracture in the right intertrochanteric region (arrow). (b) CT chest performed three weeks later shows a speculated right upper lobe lung mass (arrow) with right pleural effusion. Biopsy confirmed the mass was a non-small cell lung carcinoma. CT, computed tomography

Figure 2

Incidental finding of hepatocellular carcinoma. A 56-year-old man with known cirrhosis being evaluated for left leg claudication. (a) Axial CT image through the liver shows incidental arterial enhancing mass (arrow) in segment 7 of the liver. (b) Arterial phase image from quadriphasic CT scan of the liver completed three months later shows increased size of the mass (arrow) with washout on delayed phase (c, arrow), in keeping with hepatocellular carcinoma. CT, computed tomography

Statistical analysis

Data was entered and analyzed in a Microsoft Excel spreadsheet (Microsoft Corp, Redmond, CA, USA). Incidence, significance and follow-up rates for incidental findings were calculated. In addition, within each modality, comparison of the demographics between patients who did not have incidental findings versus those who did, the prevalence of incidental findings in patients who underwent CTA or MRA for peripheral arterial occlusive disease (PAOD) versus other indications, and the incidence of incidental findings in post-EVAR CTAs versus other abdominal CTAs were performed. Finally, a direct comparison between CTA and MRA was made with respect to patient demographics, study indication, and prevalence of incidental findings. Continuous variables were compared using Student's t-test. Categorical variables were compared with χ² test. All tests were two-sided and a P value of <0.05 was considered statistically significant.

Results

A total of 737 patients who underwent CTA and 184 patients who underwent MRA were included in the study. Details regarding patient demographics, study indication and vascular territories imaged by the CTA and MRA studies can be found in Table 1.

Table 1

Patient demographics, study indications, and vascular territories imaged in the CTA and MRA studies

	CTAn (%)	MRA n (%)
Demographics
Number of patients	737	184
Male	524 (71.1)	90 (48.9)
Female	213 (28.9)	94 (51.1)
Age*	68.3 ± 12.6	50.6 ± 18.4

Study indication
PAOD	303 (41.1)	51 (27.7)
Aneurysmal disease	313 (42.5)	28 (15.2)
Arteriovenous malformation	2 (0.3)	29 (15.8)
Thrombotic/embolic disease	14 (1.9)	7 (3.8)
Arterial dissection	23 (3.1)	6 (3.3)
Renal transplant assessment	18 (2.4)	1 (0.5)
Suspected infection	7 (1.0)	–
Thoracic outlet syndrome	–	10 (5.5)
Klippel–Trenaunay syndrome	–	7 (3.8)
Takayasu arteritis	–	7 (3.8)
Other	43 (5.8)	37 (20.1)
None stated	14 (1.9)	1 (0.5)

Vascular territories
Head and neck (H + N)	20 (2.7)	–
Thorax (T)	10 (1.4)	18 (9.8)
Abdomen (A)	24 (3.2)	13 (7.1)
Abdomen and pelvis (A + P)	231 (31.3)	52 (28.2)
Upper extremity (UE)	12 (1.6)	6 (3.3)
Lower extremity (LE)	–	8 (4.3)
A + P + LE	341 (46.3)	78 (42.4)
T + A + P	75 (10.2)	7 (3.8)
T + A + P + LE	22 (3.0)	–
H + N + T + A + P + LE	2 (0.3)	2 (1.1)

CTA, computed tomographic angiography; MRA, magnetic resonance angiography

*Age is reported as the mean ± standard deviation

A total of 1140 and 159 extravascular incidental findings were found in 539 CTA and 95 MRA patients, respectively. Categorization and actual findings of these incidental lesions are listed in Table 2. Furthermore, grouping of these findings by the number of patients for the two modalities can be found in Table 3.

Table 2

Categorization of extravascular incidental findings detected in patients who underwent CTA or MRA for a variety of clinical indications

Categorization and finding	Number offindingsCTA	MRA
Total number of incidental findings	1140	159
Low significance (little or no clinical relevance)
Benign renal findings (e.g. simple cysts and stones)	272	48
Benign hepatic or biliary findings (e.g. hemangioma and stones)	160	24
Benign bone changes (e.g. degenerative changes)	57	12
Benign intestinal changes (e.g. diverticulum)	177	3
Benign pulmonary findings (e.g. benign nodules)	159	2
Benign adrenal lesions (e.g. nodules <1 cm)	42	–
Benign splenic lesions (e.g. simple cysts)	28	1
Benign pancreatic lesions (e.g. simple cysts)	14	1
Benign prostate, uterine, ovarian or vaginal finding	19	14
Other benign findings	65	23
Total	993 (87.1%)	128 (80.5%)

Moderate significance (clinically important but likely benign)
Renal (complex/exophytic cyst and angiomyolipoma)	14	2
Hepatic (arterially enhancing/hypodense focus and cirrhosis)	10	1
Gallbladder/biliary (ductal dilation and adenomyomatosis)	10	1
Bone (compression fracture; enchondroma and Brown tumor)	4	2
Intestinal (diverticulitis; angiodysplasia and wall thickening)	9	1
Pulmonary (pulmonary edema; indeterminate nodule; inflammatory changes and hydropneumothorax)	15	–
Adrenal (nodule >1 cm and angiomyolipoma)	12	2
Cardiac (valve calcification and congestive heart failure)	7	–
Pelvic congestion syndrome	1	8
Other (e.g. thyoid nodule; varicocele; FMD; Tarlove cyst and non-specific intraperitoneal findings, splenomegaly)	24	11
Total	106 (9.3%)	28 (17.6%)

High significance (potentially serious)
Renal (possible renal cell carcinoma and hydronephrosis)	13	–
Hepatic (possible hepatocellular carcinoma and abscesses)	7	–
Pathological fracture concerning for metastasis	1	–
Intestinal (appendicitis and possible IBD, malignancy)	3	2
Pulmonary (possible malignancy and pulmonary embolism)	6	–
Adrenal nodule (HU > 10)	3	–
Pancreatic (possible malignancy and acute pancreatitis)	4	–
Pericardial effusion	1	–
Findings suspicious for lymphomatous process	3	1
Total	41 (3.6%)	3 (1.9%)

CTA, computed tomographic angiography; MRA, magnetic resonance angiography; FMD, fibromuscular dysplasia; IBD, inflammatory bowel disease

Table 3

Analysis of CTA and MRA extravascular incidental findings by patient*

	CTAn (% of total)	MRA n (% of total)
Total number of patients	737	184
One or more incidental findings	539 (73.1)	95 (51.6)
Low significance	514 (69.7)	80 (43.5)
Moderate significance	95 (12.9)	27 (14.7)
High significance	41 (5.6)	3 (1.6)
One or more moderate or high significance finding	130 (17.6)	29 (15.8)

*Some patients had multiple findings from more than one significance level

Compared with patients with extravascular findings, the patients who had no extravascular incidental findings (n = 198) discovered on CTA were significantly younger (mean age 64.9 ± 14.8 versus 69.6 ± 11.5 years, P < 0.001) and were more commonly women (34.8 versus 26.7%, P = 0.03). Similarly, in the MRA group, patients with no extravascular incidental findings were significantly younger (mean age 45.0 ± 17.8 versus 55.8 ± 17.4 years, P < 0.001). However, contrary to the CTA patients, no significant differences were seen in the proportion of women (50.6 versus 51.6%, P = 0.89) showing incidental findings.

Patients who underwent CTA or MRA to assess their PAOD were compared with the same studies conducted for other indications. Similar proportions of patients had one or more incidental findings discovered on CTA studies conducted for PAOD (217 [71.6%] patients) and other indications (322 [74.2%] patients) (P = 0.44). However, a significantly higher percentage of patients had one or more incidental findings discovered on MRAs conducted for PAOD (33 [64.71%] patients) than MRAs conducted for other indications (62 [46.6%] patients) (P = 0.03). Furthermore, post-EVAR CTAs (n = 112) were compared with other CTAs involving the abdomen (n = 576). No significant difference was found in the proportion of patients with one or more incidental findings on post-EVAR CTAs (86 [76.8%] patients) when compared with other abdominal CTAs (436 [75.7%] patients) (P = 0.80).

Several differences were noted when comparing CTA and MRA studies. First, the mean age of patients undergoing CTAs was significantly older than patients who had MRAs (68.3 ± 12.6 versus 50.6 ± 18.4 years, P < 0.0001). In addition, there was a significantly smaller percentage of females among patients who underwent CTAs than MRAs (28.9 versus 51.1%, P < 0.0001). The indications for CTA and MRA were also different. Aneurysmal disease and PAOD combined were the reason for ordering the study in 83.6% of all CTAs examined. For MRAs, the same two indications made up only 42.9% of the studies examined (P < 0.0001). Furthermore, incidental findings were discovered on a greater proportion of CTA studies than MRA studies (73.1 versus 51.6%, P < 0.0001).

With respect to follow-up investigations, among patients who underwent CTA, follow-up studies were recommended by the reporting radiologist in 90 out of the 130 patients (69.2%) who had one or more moderate or high significance incidental finding. Appropriate follow-up was completed in 23 of the 90 patients (25.6%). Among the 41 patients with highly significant extravascular findings, 20 (48.8%) of them had additional investigations. For patients who underwent MRA, additional evaluation was recommended by the reporting radiologist in 20 out of the 29 patients (69.0%) with at least one moderate or high significance finding, and appropriate follow-up was conducted in three of the 20 patients (15.0%). Among the three patients with high significance extravascular findings, two (66.7%) had additional investigations performed for the suspicious lesions.

Discussion

Our study shows that extravascular incidental findings are prevalent on CTA and MRA studies, being present in 73 and 52% of patients, respectively. Thirteen percent of CTA patients and 15% of MRA patients had moderate significance findings that were considered to be clinically important, but likely benign. Highly significant incidental findings were detected in 5.6% of patients who underwent CTA and 1.6% of MRA patients. CTAs conducted for patients with PAOD did not reveal more incidental findings than CTAs conducted for other indications. However, MRA did reveal more extravascular findings in PAOD patients (65 versus 47% in other indications, P = 0.03). The reason for the difference between CTA and MRA findings in this respect is unclear, but could be related to the difference in age between the two populations. Post-EVAR CTAs did not reveal more incidental lesions than other abdominal CTAs (77 versus 76% of patients with one or more incidental findings, respectively). This suggests that the additional venous phase scan in post-EVAR CTAs has no added value in detecting extravascular lesions versus arterial phase alone. Where recommendations for imaging follow-up or further evaluation were made on study, appropriate investigations were conducted in only 26% of CTA patients and 15% of MRA patients. Moreover, among patients with highly significant findings discovered on CTA and MRA, only 49 and 67%, respectively, had additional testing done to examine the incidental lesions. It is likely that a combination of factors contributed to the low follow-up rates. For instance, the referring physician may be more concerned about the vascular findings, and assumed that the primary care (or other) physician would further investigate the non-vascular findings. In addition, the incidental finding(s) may have been presumed to be known to the patient already. Moreover, many vascular specialists interpret imaging studies of their own patients. They may not always read the study report and hence potentially miss other pathologies reported and any follow-up investigations recommended. However, the actual reasons are difficult to establish with certainty.

Since the first study that dealt with extravascular findings on single-slice CTA,¹⁹ a small number of published papers have examined incidental findings on peripheral vascular CTA,^15–18 and only two of those studies were performed using 64-channel multidetector CT scanners.^15,18 The rate of extravascular findings of 73% observed in this study is consistent with other studies which report a prevalence of 47–87%.^15–18 More importantly, the percentage of highly significant findings reported by previous studies was as high as 15¹⁵ versus 6.5% found in this study. Several factors may have contributed to this difference. Firstly, the mean age of our study population was lower (68 versus 72 years¹⁵). In addition, the indications for CTA were not identical, since Naidu et al. included only CTA runoffs in their study, whereas our study also included abdomen and pelvis, head and neck, as well as upper extremity CTAs. Furthermore, the patient demographics of the two studies were not the same.

In this study, MRA identified extravascular findings in a smaller percentage of patients compared with CTA. This decreased identification of incidental findings is likely due to differences in the patient population that underwent MRA. This may also be due to the improved background suppression in MRA resulting in a decreased ability to evaluate non-vascular structures. MRA patients had a lower mean age (51 versus 68 years for CTA patients, P < 0.0001), as well as the different and more diverse mixture of indications for undergoing MRA (43% of patients undergoing MRA had aneurysmal disease or PAOD compared with 84% of patients undergoing CTA, P < 0.0001). Based on our results, it is difficult to recommend CTA or MRA over one another as being superior for detection of extravascular findings.

The only other study to date that examined extravascular incidental findings on MRA looked exclusively at renal artery MRA.²¹ Extravascular findings were discovered in 58% of patients in that study, which is similar to the 52% found in this study. The incidence of potentially serious clinical findings could not be compared due to the lack of clearly defined categories of significance in that study. A difference would be expected given that our study included all types of peripheral vascular MRA and the lower mean age of patients in our study (66²¹ versus 51 years, respectively).

The results of this study highlight the importance of thoroughly interpreting the entire CTA or MRA study, not just focusing on the vasculature. Although the majority of incidental findings (greater than 80% in both modalities) were considered to have low significance, unsuspected lesions with potentially serious implications were identified in up to 6% of cases. Despite the detection of potentially serious extravascular findings, subsequent follow-up or referral was often not made, which may have significant negative impact on patient outcome, particularly when potentially treatable or life-threatening pathology may be present.^15–17,21 It is of significant concern that appropriate follow-up was conducted in only a portion of these patients – 49% and of CTA patients and 67% of MRA patients. This issue was also identified in two other studies.^15,16 This suggests the need for increased awareness among referring physicians about the importance of reviewing the report of the CTA or MRA study and acting on any recommendations made. In addition, it suggests that the standard MRA or CTA report may not be an adequate form of communication in the setting of a significant extravascular finding.

There are several limitations to this study. The retrospective nature of the study design means that the CTA and MRA studies were not conducted specifically for the purposes of detecting incidental findings. In addition, each study was only reviewed by a single reader at the time of initial reporting and the results were obtained from the initial interpretations without additional review of the images. Hence, it is likely that the prevalence of incidental findings was underestimated. However, this reflects ‘real-life’ practice and would therefore be a reasonable estimate of the rate of findings discovered in these studies performed across a spectrum of imaging practices. Another limitation is the lack of standardization in the classification of findings to various degrees of clinical significance between different studies. This makes comparisons and across such studies challenging. Even within this current study, how incidental findings are classified can have a significant impact on the percentages of low, moderate and high clinical significance findings. For example, it is plausible to include all findings whereby additional investigations were recommended by the reporting radiologist as high significance findings. This would significantly alter the number of moderate and high significance findings, as well as the follow-up rates. Ultimately, we decided, with reference to published studies, on a classification system based on the actual findings themselves, which is less subjective, but by no means perfect. Furthermore, due to the wide geographical area of the referral base to our hospital, it is impossible to determine if some of these pathologies were followed up appropriately in the referring institutions or were previously investigated. Since cases for which a response was not received from the referring physician were counted as no follow-up obtained, it is entirely possible that the follow-up rates were underestimated.

In conclusion, our findings confirm that CTA and MRA detect significant extravascular pathology that would not have been possible with conventional catheter-based angiography. Although most of these findings are clinically benign, a small percentage could be life-threatening. Follow-up investigation rates for potentially serious lesions were suboptimal for both CTA and MRA. This highlights the need for increased awareness among referring physicians and an improved communication tool for vascular specialists that would improve appropriate investigation and management of suspicious incidental findings.

Footnotes

Acknowledgements: This study is supported by the Canadian Society for Vascular Surgery.

Conflicts of interest: None declared.

References

Met

Bipat

Legemate

Diagnostic performance of computed tomography angiography in peripheral arterial disease: a systematic review and meta-analysis. JAMA 2009;301:415–24.

Sun

. Diagnostic accuracy of multislice CT angiography in peripheral arterial disease. J Vasc Interv Radiol 2006;17:1915–21.

Kock

Adriaensen

Pattynama

DSA versus multi-detector row CT angiography in peripheral arterial disease: randomized controlled trial. Radiology 2005;237:727–37.

Heijenbrok-Kal

Kock

Hunink

. Lower extremity arterial disease: multidetector CT angiography meta-analysis. Radiology 2007;245:433–9.

Menke

Larsen

. Meta-analysis: accuracy of contrast-enhanced magnetic resonance angiography for assessing steno-occlusions in peripheral arterial disease. Ann Intern Med 2010;153:325–34.

Schoenberg

Rieger

Weber

High-spatial-resolution MR angiography of renal arteries with integrated parallel acquisitions: comparison with digital subtraction angiography and US. Radiology 2005;235:687–98.

Rubin

Schmidt

Logan

Multi-detector row CT angiography of lower extremity arterial inflow and runoff: initial experience. Radiology 2001;221:146–58.

Harthun

Lau

. The incidence of pulmonary neoplasms discovered by serial computed tomography scanning after endovascular abdominal aortic aneurysm repair. J Vasc Surg 2011;53:738–41.

Messersmith

Brown

Barry

. The prevalence and implications of incidental findings on ED abdominal CT scans. Am J Emerg Med 2001;19:479–81.

10.

Yee

Kumar

Godara

Extracolonic abnormalities discovered incidentally at CT colonography in a male population. Radiology 2005;236:519–26.

11.

Johnson

Dennis

Dowe

. Extracardiac findings on coronary CT angiograms: limited versus complete image review. AJR Am J Roentgenol 2010;195:143–8.

12.

Kirsch

Araoz

Steinberg

Prevalence and significance of incidental extracardiac findings at 64-multidetector coronary CTA. J Thorac Imaging 2007;22:330–4.

13.

Lazoura

Vassiou

Kanavou

Incidental non-cardiac findings of a coronary angiography with a 128-slice multi-detector CT scanner: should we only concentrate on the heart?

Korean J Radiol 2010;11:60–8.

14.

Venkatesh

You

Landry

Extracardiac findings in cardiac computed tomographic angiography in patients at low to intermediate risk for coronary artery disease. Can Assoc Radiol J 2010;61:286–90.

15.

Naidu

Hara

Brandis

Incidence of highly important extravascular findings detected on CT angiography of the abdominal aorta and the lower extremities. AJR Am J Roentgenol 2010;194:1630–4.

16.

McDougal

Valentine

Josephs

Computed tomographic angiography has added value in patients with vascular disease. J Vasc Surg 2006;44:998–1001.

17.

Iezzi

Cotroneo

Filippone

Extravascular incidental findings at multislice CT angiography of the abdominal aorta and lower extremity arteries: a retrospective review study. Abdom Imaging 2007;32:489–94.

18.

Belgrano

Pozzi Mucelli

Spadacci

Prevalence of extravascular collateral findings during 64-slice CT angiography of the abdominal aorta and lower limbs. Radiol Med 2010;115:983–96.

19.

Katz

Jorgensen

Rubin

. Detection and follow-up of important extra-arterial lesions with helical CT angiography. Clin Radiol 1999;54:294–300.

20.

Maizlin

Barnard

Gourlay

Economic and ethical impact of extrarenal findings on potential living kidney donor assessment with computed tomography angiography. Transpl Int 2007;20:338–42.

21.

Glockner

. Incidental findings on renal MR angiography. AJR Am J Roentgenol 2007;189:693–700.