Clinical relevance and scope of accidental extracoronary findings in coronary computed tomography angiography: A cardiac versus thoracic FOV study

https://doi.org/10.1016/j.ejrad.2009.01.038Get rights and content

Abstract

Objective

To assess the spectrum and clinical relevance of extracoronary findings in coronary CT angiography (CCTA), and to compare a small (cardiac) field of view (FOV) to a large (thoracic) FOV setting.

Material and methods

1084 consecutive patients (mean 57 years) with low-to-intermediate risk of coronary artery disease were enrolled. 542 CCTA scans were interpreted with small FOV (160–190 mm2) encompassing the cardiac region. In another 542 CCTA (patients matched for age and gender), read-out of an additional full FOV (>320 mm2) covering the thorax was performed. Clinical relevance of extracoronary findings was considered as either “significant” or “non-significant”. “Significant” findings were subclassified as either score 1: findings necessitating immediate therapeutic actions, or score 2: findings with undoubted clinical or prognostic relevance, requiring clinical awareness, follow-up or further investigations (non-urgent). “Non-significant” findings were assigned to either score 3: findings not requiring follow-up or further tests, or as score 4: irrelevant incidental findings.

Results

Significantly more patients with extracoronary findings were identified by using a full FOV with 43.2% (234/542) compared to a small FOV with 33.6% (182/542) (p = 0.001). Similarly, a higher total number of extracoronary findings (n = 394) was found on full FOV compared to small FOV (n = 250) (p < 0.001).

The detection rate of clinically significant findings was higher by using full FOV compared to small FOV (25.6% versus 15.4%) (p < 0.001), out of those 2.2% versus 1.8% of findings required immediate actions (score 1), and 23.4% versus 13.6% (p = 0.0001), respectively were of clinical relevance (non-urgent, score 2). The rate of malign findings was 0.2%, and of acute pulmonary embolism 0.1%.

More lung pathologies were observed by using full FOV compared to small FOV (22% versus 7%) (p < 0.0001), and the detection rate of intrapulmonary nodules increased by 2.1%. Prevalence of aortic valve calcification (n = 72) was 13.3%, out of those 7% had less than 2 cm2 aortic valve orifice area.

Conclusions

The interpretation of extracoronary findings on CCTA scans is mandatory given high prevalence of clinically significant findings by using a full “thoracic” FOV.

Introduction

Coronary multislice computed tomography angiography (CCTA) is an arising non-invasive imaging technique for the evaluation of the coronary arterial tree [1]. Unlike conventional angiographic methods, CCTA has the ability to provide morphological information about the entire heart and the thorax. Although typically confined to a small field of view (FOV) focused on the heart to enable optimal spatial resolution of the coronary arteries, data of surrounding thoracic organs are acquired complementary during CCTA.

Each anomalous finding that cannot be related pathophysiologically to the target disease is regarded as accidental [2]. To this date there is modest information regarding the clinical relevance of unexpected imaging findings in patients undergoing CCTA. Some discrepancies exist regarding the potential benefits of specifically reanalysing CCTA for extracardiac disease. Budoff et al. suggest that specifically overreading the data set for non-cardiac findings would most likely cause additional anxiety to the patient and lead to additional cost, without proven benefit [3]. Several other authors, such as Kirsch et al. [4] found this approach unsatisfactory and recommend thoroughly and methodically evaluating extracoronary incidental findings, and reporting on their clinical significance. Other studies [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15] report varying prevalences [2] of accidental findings with a wide range from 25% to 79%, however the majority report a concerning high number of clinically significant findings ranging from 4% to 56% [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15]. These variations can be attributed to different scanner technology (electron-beam computed tomography (EBCT) or multislice computed tomography (CT)), the examination technique (e.g. calcium scoring or CCTA), or the FOV setting applied (small or full field of view). Hence an arising issue is the effect of a small (“cardiac”) FOV versus a large (“thoracic”) FOV setting, not at least since reading thoracic CT scans necessitates specialized training in radiological diagnoses of thoracic organs not included in a cardiologist's curriculum.

The purpose of this study was to ancillary analyse the clinical significance and the spectrum of accidental findings in a large series of patients referred to CCTA, and to compare a small (“cardiac”) FOV with a full (“thoracic”) FOV setting.

Section snippets

Study design

The study population consisted of 1084 consecutive patients (56% males and 44% females) with a mean age of 58 ± 10.8 years (range, from 37 to 87 years). The patients were referred to CCTA for evaluation of suspected coronary artery disease [16] between November 2005 and March 2008. The CT scans of 542 patients (group A) were reconstructed with a small field of view restricted to the heart. In the following 542 consecutive patients (group B) an additional large FOV covering the entire thorax was

Results

There was no difference in gender and age between the groups with small (group A) and large (group B) FOV. There were 55% (300/542) males in group A and 57% (310/542) males in group B [p = n.s.], and the mean age was 58.3 ± 11.2 and 57.6 ± 10.2 years [p = 0.268], respectively.

Discussion

It is currently intensely discussed whether non-coronary findings on a CCTA examination should be reported or not. Only 9% of cardiology centers read the non-coronary part of the CCTA scan, but still 72% of cardiologists believe that the interpretation of non-cardiac findings requires input from the radiologists [20].

Our data show that the prevalence of clinically significant findings is high with 25.6%, and markedly higher by using a full “thoracic” FOV, supporting the interpretation of

Limitations

We have validated subsequent patient management only, but we did not monitor our patients over a long-term follow-up period. We acknowledge that the readings were performed by one experienced, board certified radiologist, and we did not perform an interobserver variability.

Minor pathologies related to the spine were not included in this data presentation.

Conclusion

It is reasonable to conclude, as approved by the recently published training guidelines from the ACC/ASNC/SCCT [28], that it is necessary to educate and train all individuals performing CCTA examinations in the recognition of unexpected non-coronary CT findings, or to establish a multidisciplinary approach of radiologists and cardiologists. Not at least a false negative rate of 13.5% was reported for emergency thoracic CT scans by Wechsler et al. [27], pointing out the importance of radiology

References (29)

  • J. Kirsch et al.

    Prevalence and significance of incidental extracardiac findings at 64-multidetector coronary CTA

    J Thorac Imaging

    (2007)
  • K.M. Horton et al.

    Prevalence of significant noncardiac findings on electron-beam computed tomography coronary artery calcium screening examinations

    Circulation

    (2002)
  • P. Hunold et al.

    Prevalence and clinical significance of accidental findings in electron-beam tomographic scans for coronary artery calcification

    Eur Heart J

    (2001)
  • J.G. Schragin et al.

    Non-cardiac findings on coronary electron beam computed tomography scanning

    J Thorac Imaging

    (2004)
  • Cited by (0)

    View full text