To investigate if the effect of cardiac computed tomography (CT) vs. invasive coronary
angiography (ICA) on cardiovascular events differs based on smoking status.This pre-specified
subgroup analysis of the pragmatic, prospective, multicentre, randomised DISCHARGE
trial (NCT02400229) involved 3561 patients with suspected coronary artery disease
(CAD). The primary endpoint was major adverse cardiovascular events (MACE: cardiovascular
death, non-fatal myocardial infarction, or stroke). Secondary endpoints included an
expanded MACE composite (MACE, transient ischaemic attack, or major procedure-related
complications).Of 3445 randomised patients with smoking data (mean age 59.1 years
+ / - 9.7, 1151 men), at 3.5-year follow-up, the effect of CT vs. ICA on MACE was
consistent across smoking groups (p for interaction = 0.98). The percutaneous coronary
intervention rate was significantly lower with a CT-first strategy in smokers and
former smokers (p = 0.01 for both). A CT-first strategy reduced the hazard of major
procedure-related complications (HR: 0.21, 95% CI: 0.03, 0.81; p = 0.045) across smoking
groups. In current smokers, the expanded MACE composite was lower in the CT- compared
to the ICA-first strategy (2.3% (8) vs 6.0% (18), HR: 0.38; 95% CI: 0.17, 0.88). The
rate of non-obstructive CAD was significantly higher in all three smoking groups in
the CT-first strategy.For patients with stable chest pain referred for ICA, the clinical
outcomes of CT were consistent across smoking status. The CT-first approach led to
a higher detection rate of non-obstructive CAD and fewer major procedure-related complications
in smokers.This pre-specified sub-analysis of the DISCHARGE trial confirms that a
CT-first strategy in patients with stable chest pain referred for invasive coronary
angiography with an intermediate pre-test probability of coronary artery disease is
as effective as and safer than invasive coronary angiography, irrespective of smoking
status.ClinicalTrials.gov NCT02400229.• No randomised studies have assessed smoking
status on CT effectiveness in symptomatic patients referred for invasive coronary
angiography. • A CT-first strategy results in comparable adverse events, fewer complications,
and increased coronary artery disease detection, irrespective of smoking status. •
A CT-first strategy is safe and effective for stable chest pain patients with intermediate
pre-test probability for CAD, including never smokers.