Personalization of CM Injection Protocols in Coronary Computed Tomographic Angiography
Personalization of CM Injection Protocols in Coronary Computed Tomographic Angiography
Cardiac computed tomography (CCT) is one of the standard non-invasive imaging techniques allowing imaging of the heart and coronary arteries with a high temporal and spatial resolution. The high sensitivity and negative predictive value (NPV) of coronary CT angiography (CCTA) make it a valuable tool in the assessment of coronary artery disease (CAD) in patients with low to intermediate risk for CAD, especially to rule out CAD. This risk stratification can be done with help of multiple different risk-calculators (e.g. the updated Diamond-Forrester model by Genders et al. 2012). These calculators take different variables into account, e.g. advanced age, gender, blood pressure, diabetes mellitus (DM), lipid profile and smoking.
The aim of CCTA is a high diagnostic accuracy, which depends on both optimal intravascular enhancement (in Hounsfield Units; minimal 325 HU) and contrast-to-noise ratio (CNR). Optimal intravascular enhancement and CNR depend on different factors such as scan technique (e.g. tube voltage, tube potential), parameters of the administered contrast material (CM) and patient related factors (e.g. cardiac output (CO), body weight (BW)).
Patients with cardiac diseases often have multiple risk factors for developing contrast induced nephropathy (CIN), e.g. diabetes mellitus, advanced age, hypertension and chronic kidney disease. Although the relationship between CTA and CIN has recently come to discussion (AMACING trial; Nijssen et al. 2017), it is still desirable to minimise the CM volume used in these patients. One method to reduce the CM volume is to personalise the injection protocols. The personalisation of injection protocols to the individual patient is gaining more attention in the field of CT imaging. The goal is to individualise the injection protocols to a level, where the patient only receives the minimal amount of CM needed to acquire a diagnostic scan, while maintaining a diagnostic image quality. Many techniques are available and have been studied, e.g. adjustment of CM volume to scan protocol, CO, lean body weight (LBW) and BW.
However, no data is available on which of these is the most beneficial method for the personalisation of CM injection protocols. Therefore, the aim of this study is to assess the performance of three different personalized injection protocols (based on CO, LBW and BW) in CCTA with regard to image quality in comparison to previously used protocols in our department. We hypothesize that the personalized injection protocols will be non-inferior, provide a homogenous coronary enhancement (less non-diagnostic scans) in patients, and will account for a reduction of CM volume in our department in comparison to the previously used protocols.
Cardiac diseases still are the leading cause of death in the western world. Besides the high mortality of the cardiac disease itself, recent studies have shown an increasing evidence of the prognostic value of cardiac diseases, particularly coronary artery disease (CAD), in a broad variety of other diseases. Especially in patients with oncological pathologies undergoing radiotherapy or lung resection surgery [Ambrogi et al. 2003; Kanzaki et al. 2017; Khakoo et al. 2008; Basacaraju et al. 2002; Fajardo et al. 1972]. Cardiac computed tomography angiography (CCTA) is one of the standard non-invasive imaging techniques allowing imaging of the heart and coronary arteries with a high temporal and spatial resolution. The high sensitivity and negative predictive value (NPV) of CCTA make it a valuable tool in the assessment of CAD. Especially to rule out CAD in patients with low to intermediate risk for CAD [Hendel et al. 2006; Goldstein et al. 2011; Budoff et al. 2008; Meijboom et al. 2008; Roffi et al. 2016].
The aim of CCTA is to achieve a high diagnostic accuracy, which depends on both optimal intravascular enhancement (in Hounsfield Units (HU); minimal 325 HU) and contrast-to-noise ratio (CNR). Optimal intravascular enhancement and CNR depend on multiple factors, such as scan technique (e.g. tube voltage (kV) and tube potential), parameters of the administered contrast material (CM) (e.g. concentration, flow rate) and patient related factors (e.g. body weight (BW), heart rate or cardiac output [CO]) [Bae et al. 2004; Awai et al. 2004]. Dedicated CT protocols are necessary to image the heart and the coronary arteries. These protocols require a correlation to the patient's electrocardiogram (ECG) in order to minimize cardiac motion and deliver a sharp image of the heart and coronary arteries. Depending on the heart rate of the patient, the scan can either be a prospectively ECG-triggered high-pitch or adaptive sequence, or retrospectively ECG-gated helical scan.
Patients with cardiac diseases often have multiple risk factors for developing contrast induced nephropathy (CIN), e.g. diabetes mellitus (DM), advanced age, hypertension, cardiovascular disease and chronic kidney disease. Although the relationship between CTA and CIN has recently come to discussion (AMACING trial; Nijssen et al. 2017), it is still desirable to minimise the CM volume used in these patients. Multiple strategies have been studied to reduce CM volume in CTA imaging [Kok et al 2016; Kok et al. 2015; Kok et al. 2016; Kok et al. 2016, Mihl et al. 2016; Seehofnerova et al. 2015; Hendriks et al. 2016]. For instance, lowering tube voltage to achieve a higher attenuation, the latter explained by the fact that mean photon energy of lower tube voltage settings in the x-ray beam moves closer to the K-edge of iodine (33.2 keV).
The personalisation of injection protocols to the individual patient is gaining more attention in CT imaging, and the goal is to individualise the injection protocols to a level, where the patient only receives the minimal amount of CM needed to make a diagnostic scan with a diagnostic image quality. Besides the techniques mentioned above, other techniques are possible, e.g. adjustment according to patients CO, lean body weight (LBW), and BW.
Both blood volume and CO increase with an increasing BW. When CO increases, the distribution of CM is also increased. On one hand this results in a fastened test bolus arrival, on the other hand in a decreased and shortened intravascular attenuation profile (in comparison to a decreased CO) [Bae KT 2010]. Therefore, both timing and CM volume can be adjusted to CO in order to achieve a similar intravascular attenuation profile.
LBW is a measurement for body fat percentage. It is known that fatty tissue is not as well vascularised in comparison to muscle tissue. Thus with increasing BW due to fatty tissue, the blood volume does not increase linear to BW. For instance, patients with a high BW due to increased muscle tissue, need more CM volume compared to patients with a high BW due to fatty tissue. The concept of LBW accounts for this fact [Bae KT 2010].
When adjusting CM volume to BW, less CM volume is needed with a lower BW in comparison to a higher BW, due to the decreased blood volume and CO [Bae KT 2010].
The tube voltage, set by automated tube voltage selection (ATVS, CAREkV, Siemens, Berlin, Germany), needs to be taken into account as well. CAREkV is a software program which chooses the optimal tube voltage (kV) and tube current (mAs) settings for the individual patient, based on their topograms, while maintaining a diagnostic image quality. Lower kV settings result in a higher intravascular enhancement if the same amount of CM is used, thus giving the potential to decrease CM volume in lower kV settings.
Since no data exist on the optimal method of choice, the aim of this study is to assess the performance of three personalized injection protocols (CO, LBW and BW) in CCTA, in comparison to previously used protocols, with regard to image quality. Secondly we want to assess the presence and severity of CAD in patients undergoing CCTA with help of the coronary artery disease reporting and data system (CAD-RADS).
A total of 330 patients will be included in this prospectively observer blinded randomized controlled non-inferiority trial. All patients will be randomised in one of three groups (CO, LBW and BW) with help of a randomisation software program (ALEA). The control group will consist of 110 consecutive patients who will be included retrospectively.
Inclusion Criteria:
Exclusion Criteria: