4. What is the diagnostic value (i.e. sensitivity, specificity and/or inter-observer variability) of MRI as compared to echocardiography (or vice versa) for detection of asymptomatic cardiac systolic dysfunction in childhood and young adult cancer survivors?
First Author Year	Study Design Treatment era Years of follow-up	Participants	Diagnostic tests	Main outcomes	Addt’l remarks
Armstrong50 2012	Single-center cohort study (USA). Treatment era: nm. Years of follow-up since cancer diagnosis: mean 27.7 years (range 18.4–38.3).	134 adult childhood cancer survivors (cancer diagnosed before age 21 years) treated with chestdirected radiotherapy and/or anthracyclines for ALL (n=44), Hodgkin’s lymphoma (n=37), osteosarcoma (n=11), non-Hodgkin’s lymphoma (n=8), AML (n=6), neuroblastoma (n=3), Ewing sarcoma (n=2). Wilms tumour (n=2) and soft tissue sarcoma (n=1). 47 men / 67 women; mean age at diagnosis tumour 10.5 years (range 0.02–19); mean age at time of study 38.3 years (range 22.7–53.7). Treatment: Mean cumulative anthracycline dose 186 mg/m2 (range 0–803); 97 patients received anthracyclines. 37 patients received chest-directed radiotherapy (n=16 1–30 Gy and n=21 > 30Gy; no information on number of fractions).	Cardiac magnetic resonance imaging (analysis was supervised and/or performed by a single investigator); an abnormal test result was defined as LVEF<50% (n=16; prevalence 14%). 3D as well as a 2D echocardiogram with Doppler and time-motion mode (M-mode) (analysis was performed by a single investigator); an abnormal test result was defined as LVEF<50% (n=22/prevalence 19.3% with 3D echocardiography; n=6/prevalence 5.3% with biplane 2D echocardiography; n=8/prevalence 7% with apical 4-Chamber 2D echocardiography and n=24/prevalence 21.1% with Teichholz 2D echocardiography). Time between tests: within a 48-hour period.	Screening performance of echocardiography compared with cardiac magnetic resonance imaging (reference standard) for detection of an LVEF<50%: 3D echocardiography: Sensitivity 53% Specificity 86% Positive predictive value 36% Negative predictive value 92% Biplane 2D echocardiography: Sensitivity 25% Specificity 98% Positive predictive value 67% Negative predictive value 89% Apical 4-Chamber 2D echocardiography: Sensitivity 25% Specificity 96% Positive predictive value 50% Negative predictive value 89% Teichholz 2D echocardiography: Sensitivity 29% Specificity 79% Positive predictive value 17% Negative predictive value 88% Bland-Altman measures of agreement with cardiac magnetic resonance imaging: For 3D echocardiography (bias, 1%; Bland-Altman limits of agreement [± 1.96 standard deviation], −11.8% to 14.0%); For 2D echocardiography: 2D biplane (bias, −5.2%; −19.0% to 8.69%), 2D apical 4-chamber (bias, −5.4%; −22.1% to 11.4%), Teichholz M-mode (bias, −3.1%; −28.3% to 22.1%).	This study is an analysis of data from 5 pilot studies, convenience sampled from the larger St. Jude Lifetime Cohort Study (SJLIFE). Patients with an implanted medical device or a history of congenital heart disease were excluded. Of the 114 patients that completed the evaluation, 108 were previously undiagnosed with cardiomyopathy. Selection bias cannot be ruled out (692 survivors enrolled in the SJLIFE cohort were exposed to anthracyclines and/or chest radiotherapy of which 134 participated in the study). The risk of detection bias is unclear; nm if outcome assessors were blinded. Outcome/attrition bias cannot be ruled out (for 20 out of 134 survivors that agreed to participate (15%) cardiac magnetic resonance imaging could not be completed*).

Nm: not mentioned; ALL: acute lymphoblastic leukemia; AML: acute myeloid leukemia; Gy: Gray; LVEF: left ventricular ejection fraction

^*information provided in this table is for the 114 participants with results for all tests unless otherwise stated.