Prognostic value of pre-hospitalization stress perfusion cardiovascular magnetic resonance to predict death in patients hospitalized for COVID-19

BACKGROUND: Inducible ischaemia is a strong marker of vascular vulnerability. Knowing the important role of the vascular tropism of COVID-19 to explain its severity, the presence of a prior inducible ischaemia may be a key pathogenetic determinant of COVID-19 severity. AIMS: To investigate the prognostic value of prior inducible ischaemia on stress cardiovascular magnetic resonance (CMR) to predict death in patients hospitalized for COVID-19. METHODS: We retrospectively analysed consecutive patients referred for stress perfusion CMR during 1/1/18–1/1/20 who were later hospitalized for COVID-19. The primary outcome was all-cause death, including in-hospital and post-hospitalization deaths, based on the electronic national death registry. RESULTS: Among the patients referred for stress CMR, 481 were hospitalized for COVID-19 (mean age 68.4 ± 9.6 years, 61.3% male) and completed the follow-up (median [interquartile range] 73 [36–101] days). There were 93 (19.3%) all-cause deaths, of which 13.7% occurred in hospital and 5.6% were post-hospitalization deaths. Age, male sex, hypertension, diabetes, known coronary artery disease (CAD), the presence of prior inducible ischaemia, the number of ischaemic segments, the presence of late gadolinium enhancement and left ventricular ejection fraction were significantly associated with all-cause death. In multivariable stepwise Cox regression analysis, age (hazard ratio [HR]: 1.04, 95% confidence interval [CI]: 1.01–1.07; P = 0.023), hypertension (HR: 2.77; 95% CI: 1.71–4.51; P < 0.001), diabetes (HR: 1.72; 95% CI: 1.08–2.74; P = 0.022), known CAD (HR: 1.78; 95% CI: 1.07–2.94; P = 0.025) and prior inducible ischaemia (HR 2.05; 95% CI: 1.27–3.33; P = 0.004) were independent predictors of all-cause death. CONCLUSIONS: In COVID-19 patients, prior inducible ischaemia by stress CMR during the 2 years preceding the COVID-19 pandemic was independently associated with all-cause death.