Effect of routine preoperative screening for aortic calcifications using noncontrast computed tomography on stroke rate in cardiac surgery: the randomized controlled CRICKET study

Abstract

Objectives: To evaluate if routine screening for aortic calcification using unenhanced CT lowers the risk of stroke and alters the surgical approach in patients undergoing general cardiac surgery compared with standard of care (SoC).

Methods: In this prospective, multicenter, randomized controlled trial, adult patients scheduled for cardiac surgery from September 2014 to October 2019 were randomized 1:1 into two groups: SoC alone, including chest radiography, vs. SoC plus preoperative noncontrast CT. The primary endpoint was in-hospital perioperative stroke. Secondary endpoints were preoperative change of the surgical approach, in-hospital mortality, and postoperative delirium. The trial was halted halfway for expected futility, as the conditional power analysis showed a chance < 1% of finding the hypothesized effect.

Results: A total of 862 patients were evaluated (SoC-group: 433 patients (66 ± 11 years; 74.1% male) vs. SoC + CT-group: 429 patients (66 ± 10 years; 69.9% male)). The perioperative stroke rate (SoC + CT: 2.1%, 9/429 vs. SoC: 1.2%, 5/433, p = 0.27) and rate of changed surgical approach (SoC + CT: 4.0% (17/429) vs. SoC: 2.8% (12/433, p = 0.35) did not differ between groups. In-hospital mortality and postoperative delirium were comparable between groups. In the SoC + CT group, aortic calcification was observed on CT in the ascending aorta in 28% (108/380) and in the aortic arch in 70% (265/379).

Conclusions: Preoperative noncontrast CT in cardiac surgery candidates did not influence the surgical approach nor the incidence of perioperative stroke compared with standard of care. Aortic calcification is a frequent finding on the CT scan in these patients but results in major surgical alterations to prevent stroke in only few patients.

Key points: • Aortic calcification is a frequent finding on noncontrast computed tomography prior to cardiac surgery. • Routine use of noncontrast computed tomography does not often lead to a change of the surgical approach, when compared to standard of care. • No effect was observed on perioperative stroke after cardiac surgery when using routine noncontrast computed tomography screening on top of standard of care.

Keywords: Cardiac surgical procedures; Preoperative care; Radiography; Stroke; Tomography, X-ray computed.

Fig. 1

Study flowchart. Screening data was available in two of the three participating centers

Fig. 2

Reasons for changing the surgical approach in both groups

Fig. 3

Sample images of CXR and noncontrast CT. Sample images of a 70-year-old male patient whose surgical approach was changed from surgical to transcatheter aortic valve replacement. 1: The posterior-anterior (1a) and lateral (1b) views of the preoperative CXR 2: Ascending aortic calcifications on the axial (2a) and sagittal (2b) plane of the noncontrast CT. The arrow in panel 1a indicates the aortic knob, where only modest calcification is seen. The arrow in panel 1b point at the ventral boundary of the aorta, where no clear calcifications seem to be present. The arrows in panels 2a and 2b indicate the extensive ventral calcifications, hampering aortic manipulation in this area