Cardiac computed tomography as a complete functional tool

This editorial refers to ‘Association of baseline and change in global longitudinal strain by computed tomography with post transcatheter aortic valve replacement outcomes’ by M. Fukui et al., pp. 476-84.

The advanced age of patients with aortic stenosis (AS) often masks the development or interpretation of symptoms due to the presence of co-morbidities and self-limited levels of activity.¹ When combined with the reluctance of both patients and physicians to use exercise stress testing, these factors can lead to an underestimation of symptomatic status.¹ Thus, imaging plays a key role in assessing AS haemodynamic severity and progression,² as well as its consequences on left ventricular (LV) remodelling and function.³ Fukui et al.⁴ examine the reproducibility and prognostic value of baseline global longitudinal strain (GLS) obtained from modern cardiac computed tomography (CCT) performed with dual-source equipment in a substantial cohort of mostly severely symptomatic AS patients being evaluated for transcatheter aortic valve replacement (TAVR) across a wide spectrum of clinical risk. Out of 446 consecutive patients, CT Angiography-LVGLS was attainable at baseline in 431 (97%), representing an increased feasibility in comparison to GLS reported for speckle tracking echocardiography (STE). Patients with a worse CT Angiography-LVGLS value (both as a continuous variable and using a detected threshold of −18.2%) had an independent higher risk of the composite end-point. Importantly, the addition of baseline CT Angiography-LVGLS to the conventional clinical and echocardiographic assessment incrementally improves the prognostic value of the models. Furthermore, the authors evaluate the change in CT Angiography-LVGLS after TAVR and its association with outcomes in 383 (88%) patients. The overall improvement in CT Angiography-LVGLS indeed averaged patients having preserved (158) or improved (51) LVGLS, while LVGLS worsened in 36 and did not improve in 136. However, the composite outcome of the latter subgroup is worse only than that of patients with preserved LVGLS at baseline, but not when compared with those with worsening or improving LVGLS.

From the technical standpoint, the study of Fukui et al. shows the potential of CCT in the functional/dynamic evaluation of LV which is not something new.⁵ In fact, CCT at its origins was performed with retrospective electrocardiogram (ECG)-gating technique^6–9 because the temporal resolution was insufficient, and it was necessary to acquire all phases of the cardiac cycle to find the least motion artefacts for adequate visualization of coronary arteries. In that period, functional assessment of heart chambers performed of 64-slice CT equipment was already validated against Cardiac Magnetic Resonance with very good results.⁵ This technique was progressively abandoned over time in favour of prospective ECG triggering to reduce radiation dose. However, with the latest generations of CT equipment implementing the high temporal resolution, lower kV capabilities, high coverage, iterative reconstructions, and so forth, the issue of radiation dose has been progressively managed in a quite effective way allowing to reintroduce retrospective ECG-gating as a standard especially in conditions in which functional evaluation of valves, ventricle, and other moving structure might be relevant for diagnostic and operating purposes (Supplementary data online, Videos S1–S5).

Surely the potential of 4D CCT with its high spatial resolution and its robustness can allow more than linear evaluations. 3D and 4D quantitative assessment of morphological features can be performed quite seamlessly and also introduced into Artificial Intelligence engines able to develop fast, reliable, and reproducible parametrization of imaging features.

The paper by Fukui et al.⁴ thus provides a state-of-the art assessment of LV function by CCT adding further evidence on the relevance of GLS as a determinant of the natural history of AS.^3,10 Noteworthy, virtually all the patients with reduced GLS display a LV ejection fraction (EF) <60% supporting the notion that LVEF < 50% may be an incorrect threshold for intervention.^3,10 What is left unexplored in this article, is if a software-independent assessment of longitudinal LV function by mitral annual plane systolic excursion (MAPSE) is as effective as CT Angiography-LVGLS (which needs a dedicated software to be performed) to favour the clinical implementation of multi-parametric CT Angiography in AS for clinical purposes.¹¹

Unfortunately, due to the clinical characteristics of the study cohort, we cannot derive any hint regarding the potential clinical allocation of these findings in the real world, to understand when the utilization of this hi-tech approach should be implemented in the elderly apparently asymptomatic individuals met in clinical practice. Though we understand that patients in whom STE is unreliable may benefit from CT Angiography-GLS, Fukui et al. cannot explore the incremental value of this technology over a comprehensive scoring system including progression of the disease, LV hypertrophy and subclinical function (by STE or MAPSE), biomarkers, and exercise testing.¹² This novel, functional tool, needs to be assessed in the large clinical arena of ‘asymptomatic’ individuals in whom the assessment of clinical severity remains problematics. Fukui et al.⁴ must be congratulated, however, to provide a clearcut evidence that CT imaging in AS should not consist of an anatomic (and coronary) evaluation anymore and that this novel technology is ready for clinical implementation.

Supplementary data

Supplementary data are available at European Heart Journal - Cardiovascular Imaging online.

Conflict of interest: none declared.

The opinions expressed in this article are not necessarily those of the Editors of EHJCI, the European Heart Rhythm Association or the European Society of Cardiology.

References