Presence and extent of cardiac computed tomography angiography defined coronary artery disease in patients presenting with syncope

This was an observational single-centre study analysing 142 retrospectively collected consecutive patients presenting with syncope at the outpatient cardiology department between May 2007 and April 2015 who were referred for CCTA within their diagnostic workup. Patients were selected if they met the definition of syncope, which was defined as transient loss of consciousness due to global cerebral hypoperfusion with rapid onset, short duration and spontaneous complete recovery [2].

General exclusion criteria for CCTA examination were: haemodynamic instability, pregnancy, renal insufficiency (defined as glomerular filtration rate <45 ml/min/1.73 m) and known severe allergic reactions regarding iodine.

This study was approved by the Institutional Review Board (METC 15-4-091) and complies to the ethical guidelines of the 1975 Declaration of Helsinki. Written informed consent was waived, because the data were anonymously recorded and analysed in accordance with guidelines of our Review Board.

Data regarding syncope type, age, gender, cardiovascular risk factors, medication use, additional diagnostic testing and CCTA results were collected from the patients’ charts and electronic medical records.

The type of syncope was ascertained by chart review by two reviewers (N.M., T.D.), blinded for CCTA results. The definitive syncope type was ascertained by consensus between the two reviewers. The following pathophysiological classification and sub-classification was used to adequately define the syncope types in each individual patient [2]:

Subsequently, patients with reflex syncope, orthostatic syncope and syncope of unknown cause were classified as having ‘non-cardiac syncope’, which led to two main syncope categories: cardiac and non-cardiac syncope.

Diabetes mellitus was defined as fasting glucose levels of ≥7 mmol/l or treatment with either diet intervention, oral glucose lowering agent or insulin [14]; smoking was defined as current smoking. A positive family history was defined as having a first-degree relative with a history of myocardial infarction or sudden cardiac death before the age of 60.

The PROCAM risk score was determined according to the following parameters: age, LDL cholesterol, smoking, HDL cholesterol, systolic blood pressure, family history of premature myocardial infarction, diabetes mellitus, and triglycerides [15]. This risk score predicts the absolute 10-year risk for the occurrence of an acute coronary event (fatal or non-fatal myocardial infarction or acute coronary death). A score <10% is estimated as low risk, 10–20% as intermediate risk, and >20% as high risk.

To further study the impact of cardiac syncope symptoms on detection of CAD, we compared the cardiac syncope patients with stable chest pain patients. The cardiac syncope patients were compared (1:2 ratio) with stable chest pain patients, referred for CCTA from the outpatient cardiology clinic to compare the prevalence and extent of CAD in cardiac syncope patients with stable chest pain patients. Matching was based upon age, gender, smoking status, diabetes mellitus type 2 and systolic blood pressure (±10 mm Hg).

Scans were performed using a 64-slice multidetector computed tomography scanner (Brilliance 64; Philips Healthcare, Best, The Netherlands) or a 2^nd generation dual-source CT scanner (Somatom Definition Flash, Siemens Healthcare, Forchheim, Germany). Data acquisition parameters for the Brilliance 64 were a 64 × 0.625 mm slice collimation, a gantry rotation time of 420 ms and a tube voltage of 80 or 120 kV depending on patients’ height and weight. Data acquisition parameters for the Somatom Definition Flash were a 2 × 128 × 0.600 mm slice collimation, a gantry rotation time of 280 ms and a tube voltage of 100 or 120 kV depending on patients’ height and weight. A non-contrast enhanced scan was performed using 120 kV and 3 mm slice thickness to determine the CCS [16]. CCTA was performed using 75–120 ml of contrast agent (Xenetix 350; Guerbet, France or Ultravist 300; Bayer Healthcare, Berlin, Germany) injected in the antecubital vein at a rate of 5.2–7.4 ml/s followed by 40 ml intravenous saline at the same flow rates.

Scan protocols differed between both CT scanners. For the 64-slice scanner, a prospectively gated ‘Step and shoot’ protocol was used in patients with stable heart rate <65 bpm. In patients with a heart rate >65 bpm, a retrospectively gated ‘Helical’ protocol was used with dose modulation. For the 2^nd generation dual-source CT scanner, a prospectively gated high-pitch spiral ‘Flash’ protocol was used in patients with a stable heart rate <60 bpm. In patients with a stable heart rate between 60–90 bpm, a prospectively gated axial ‘Adaptive sequence’ protocol was used. In patients with a heart rate >90 bpm or with an irregular heart rhythm, a retrospectively gated ‘Helical’ protocol was used. Dose modulation was switched on in all three protocols using tube current modulation (CARE Dose4D, Siemens Healthcare, Forchheim, Germany).

Patients received 50 mg Metoprolol tartrate orally (AstraZeneca, Zoetermeer, the Netherlands), two hours before CCTA, unless contra-indicated. If indicated, an additional dose of 5–20 mg Metoprolol was administered intravenously to lower the heart rate. All patients received nitroglycerine (Pohl-Boskamp, Hohenlockstedt, Germany) sublingually in a dose of 0.8 mg just prior to CCTA.

The Agatston method was used to define the CCS [16]. CCTA’s were independently analysed by a cardiologist and a radiologist, both experienced in the assessment of CCTA. In case of disagreement, consensus was reached by discussion. The assessment was performed using the source images on the provided software (Cardiac Comprehensive Analysis, Philips Healthcare or Syngo CT 2010A, Siemens Healthcare). The coronary artery tree was analysed for the presence and extent of CAD, according to the 16-segment classification of the American Heart Association [17]. Plaques were defined as visible structures within or adjacent to the coronary artery lumen, which could be clearly distinguished from the vessel lumen and the surrounding pericardial tissue. The degree of stenosis was visually defined and classified as no CAD (no luminal stenosis), mild (<50% luminal stenosis), moderate (50–70% luminal stenosis), severe (>70% luminal stenosis). A segment involvement score was defined by counting all coronary segments with plaques (irrespective of degree of stenosis), which resulted in a score ranging from 0–16 [18]. A segment stenosis score was the sum of the lesion severity in all 16 coronary segments, resulting in a score ranging from 0–48 [18].

Data were analysed using SPSS version 23.0 (SPSS Inc., Chicago, IL, USA). Continuous variables were checked whether they were normally distributed using box plots, histograms or by computing skewness and kurtosis. Continuous data were reported as means and standard deviations (SDs) if normally distributed. If data were not normally distributed, continuous data were reported as medians and boundaries of interquartile ranges (IQR). Proportions (%) were used for categorical values.

All p-values were 2‑sided, and a p-value below 0.05 was considered statistically significant.

Between May 2007 and April 2015, 142 consecutive patients presented with syncope at the outpatient cardiology clinic who were referred for CCTA. The distribution of syncope classifications was as follows: 49 (35%) cardiac syncope, 63 (44%) reflex (neurally-mediated) syncope, 10 (7%) orthostatic syncope and 20 (14%) syncope of unknown cause. Subsequently, patients with reflex syncope, orthostatic syncope and syncope of unknown cause were classified as having ‘non-cardiac syncope’ leading to two main categories, with 49 (35%) cardiac syncope patients and 93 (65%) non-cardiac syncope patients.

The baseline characteristics of the study population are described in Table 1. When compared to non-cardiac syncope patients, only age showed a statistically significant difference between the cardiac and non-cardiac syncope groups (mean (SD): 60 (13) versus 54 (12); p = 0.002). The Online Supplemental Material provides further detailed clinical information regarding the syncope patients.

The distribution of CAD presence and extent in cardiac and non-cardiac syncope patients was as follows: 72% versus 48% any CAD; 31% versus 26% mild CAD; 8% versus 14% moderate CAD and 33% versus 7% severe CAD (Table 2).

Fig. 1 visualises CAD prevalence and distribution of CAD in patients with cardiac and non-cardiac syncope. CAD presence and extent was significantly higher in patients with cardiac syncope in comparison with non-cardiac syncope patients (Table 1; p = 0.001). Additionally, all conventional CT parameters including CCS, segment involvement and segment stenosis score were significantly higher in patients with cardiac syncope compared to non-cardiac syncope (80 [0–387] versus 0 [0–79]; 2 [0–5] versus 0 [0–2]; 3 [0–7] versus 0 [0–3]) (all p-values ≤0.004; Table 2).

Fig. 2 displays the presence and extent of CAD in both cardiac syncope patients and matched chest pain controls. Patients with cardiac syncope showed a higher prevalence of any CAD (72% versus 63%, respectively) and percentage of severe luminal stenosis (33% versus 19%). Interestingly, no statistically significant difference was observed between cardiac syncope patients and chest pain controls regarding overall prevalence of CAD presence and its extent (p = 0.133).

Table 2 shows that all other conventional CT parameters including CCS, segment involvement score and segment stenosis score were comparable between cardiac syncope patients and chest pain controls (all p-values ≥0.272).

Meticulous history taking is essential in arriving at a diagnosis in patients with syncope. Once cardiac syncope is suspected, a wide range of different diagnostic tests can be considered in patients presenting with syncope [2, 6]. However, there is no guidance regarding anatomical imaging techniques to detect CAD due to lacking evidence describing the extent and nature of CAD in patients with cardiac syncope. The relationship between cardiac ischaemia and syncope is multiple, including induction of non-sustained ventricular arrhythmias and sinoatrial or atrioventricular block, or by triggering e. g. the Bezold-Jarisch-reflex causing severe bradycardia and hypotension. Apart from a direct relationship, indirect mechanisms may be important, including old myocardial infarction with ventricular remodelling as a basis for re-entrant or adrenergic ventricular tachycardia; likewise, atrial remodelling leading to late onset atrioventricular nodal tachycardia or atrial tachycardias may at times occur with well-known haemodynamic compromise eliciting syncope at the beginning of the attack [19]. In all of the above-mentioned aetiologies, CAD may be causal or contributory for syncope whereby one could conjecture that interventional and vascular prophylactic management may help to reduce further syncope in these patients. In all other cases, CAD presence should be considered as coincidental wherefore vascular prophylactic vascular management may not be indicated in the management of syncope.

Within the European Society of Cardiology guidelines for the diagnosis and management of syncope, exercise stress testing is only recommended in patients with suspected exercise-induced syncope, which concerns a rare condition [2]. Concurrently, ischaemia evaluation is recommended by the American College of Cardiology/American Heart Association (ACC/AHA) for patients with syncope and an intermediate-to-high risk for coronary heart disease or known CAD, but such strategy may underdiagnose the presence of non-obstructive CAD, which still is associated with high major adverse cardiac event rates [6, 20, 21]. On the other hand, a previous report revealed a low-diagnostic yield for stress myocardial perfusion imaging across all risk categories in syncope patients without known CAD [22].