Non-invasive characterization of pleural and pericardial effusions using T1 mapping by magnetic resonance imaging.

AIMS: Differentiating exudative from transudative effusions is clinically important and is currently performed via biochemical analysis of invasively obtained samples using Light's criteria. Diagnostic performance is however limited. Biochemical composition can be measured with T1 mapping using cardiovascular magnetic resonance (CMR) and hence may offer diagnostic utility for assessment of effusions. METHODS AND RESULTS: A phantom consisting of serially diluted human albumin solutions (25-200 g/L) was constructed and scanned at 1.5 T to derive the relationship between fluid T1 values and fluid albumin concentration. Native T1 values of pleural and pericardial effusions from 86 patients undergoing clinical CMR studies retrospectively analysed at four tertiary centres. Effusions were classified using Light's criteria where biochemical data was available (n = 55) or clinically in decompensated heart failure patients with presumed transudative effusions (n = 31). Fluid T1 and protein values were inversely correlated both in the phantom (r = -0.992) and clinical samples (r = -0.663, P < 0.0001). T1 values were lower in exudative compared to transudative pleural (3252 ± 207 ms vs. 3596 ± 213 ms, P < 0.0001) and pericardial (2749 ± 373 ms vs. 3337 ± 245 ms, P < 0.0001) effusions. The diagnostic accuracy of T1 mapping for detecting transudates was very good for pleural and excellent for pericardial effusions, respectively [area under the curve 0.88, (95% CI 0.764-0.996), P = 0.001, 79% sensitivity, 89% specificity, and 0.93, (95% CI 0.855-1.000), P < 0.0001, 95% sensitivity; 81% specificity]. CONCLUSION: Native T1 values of effusions measured using CMR correlate well with protein concentrations and may be helpful for discriminating between transudates and exudates. This may help focus the requirement for invasive diagnostic sampling, avoiding unnecessary intervention in patients with unequivocal transudative effusions.

A Multicenter, Scan-Rescan, Human and Machine Learning CMR Study to Test Generalizability and Precision in Imaging Biomarker Analysis.

BACKGROUND: Automated analysis of cardiac structure and function using machine learning (ML) has great potential, but is currently hindered by poor generalizability. Comparison is traditionally against clinicians as a reference, ignoring inherent human inter- and intraobserver error, and ensuring that ML cannot demonstrate superiority. Measuring precision (scan:rescan reproducibility) addresses this. We compared precision of ML and humans using a multicenter, multi-disease, scan:rescan cardiovascular magnetic resonance data set. METHODS: One hundred ten patients (5 disease categories, 5 institutions, 2 scanner manufacturers, and 2 field strengths) underwent scan:rescan cardiovascular magnetic resonance (96% within one week). After identification of the most precise human technique, left ventricular chamber volumes, mass, and ejection fraction were measured by an expert, a trained junior clinician, and a fully automated convolutional neural network trained on 599 independent multicenter disease cases. Scan:rescan coefficient of variation and 1000 bootstrapped 95% CIs were calculated and compared using mixed linear effects models. RESULTS: Clinicians can be confident in detecting a 9% change in left ventricular ejection fraction, with greater than half of coefficient of variation attributable to intraobserver variation. Expert, trained junior, and automated scan:rescan precision were similar (for left ventricular ejection fraction, coefficient of variation 6.1 [5.2%-7.1%], P=0.2581; 8.3 [5.6%-10.3%], P=0.3653; 8.8 [6.1%-11.1%], P=0.8620). Automated analysis was 186× faster than humans (0.07 versus 13 minutes). CONCLUSIONS: Automated ML analysis is faster with similar precision to the most precise human techniques, even when challenged with real-world scan:rescan data. Assessment of multicenter, multi-vendor, multi-field strength scan:rescan data (available at www.thevolumesresource.com) permits a generalizable assessment of ML precision and may facilitate direct translation of ML to clinical practice.

Motion-corrected free-breathing LGE delivers high quality imaging and reduces scan time by half: an independent validation study.

Late gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR) sequences have evolved. Free-breathing motion-corrected (MOCO) LGE has potential advantages over breath-held (bh) LGE including minimal user input for the short axis (SAX) stack without breath-holds. It has previously been shown that MOCO-LGE delivers high image quality compared to bh-LGE. We sought to conduct an independent validation study to investigate real-world performance of bh-LGE versus MOCO-LGE in a high-throughput CMR center immediately after the introduction of the MOCO-LGE sequence and with elementary staff induction in its use. Four-hundred consecutive patients, referred for CMR and graded by clinical complexity, underwent CMR on either of two scanners (1.5 T, both Siemens) in a UK tertiary cardiac center. Scar imaging was by bh-LGE or MOCO-LGE (both with phase sensitive inversion recovery). Image quality, scan time, reader confidence and report reproducibility were compared between those scanned by bh-LGE versus MOCO-LGE. Readers had > 3 years CMR experience. Categorical variables were compared by χ2 or Fisher's exact tests and continuous variables by unpaired Student's t-test. Inter-rater agreement of LGE reports was by Cohen's kappa. Image quality (low score = better) was better for MOCO-LGE (median, interquartile range [Q1-Q3]: 0 [0-0] vs. 2 [0-3], P < 0.0001). This persisted when just clinically complex patients were assessed (0 [0-1] vs. 2 [1-4] P < 0.0001). Readers were more confident in their MOCO-LGE rulings (P < 0.001) and reports more reproducible [bh-LGE vs. MOCO-LGE: kappa 0.76, confidence interval (CI) 0.7-0.9 vs. 0.82, CI 0.7-0.9]. MOCO-LGE significantly shortened LGE acquisition times compared to bh-LGE (for left ventricle SAX stack: 03:22 ± 01:14 vs 06:09 ± 01:47 min respectively, P < 0.0001). In a busy clinical service, immediately after its introduction and with elementary staff training, MOCO-LGE is demonstrably faster to bh-LGE, providing better images that are easier to interpret, even in the sickest of patients.

Myocardial native T1 and extracellular volume with healthy ageing and gender.

Aims: To determine how native myocardial T1 and extracellular volume (ECV) change with age, both to understand aging and to inform on normal reference ranges. Methods and results: Ninety-four healthy volunteers with no a history or symptoms of cardiovascular disease or diabetes underwent cardiovascular magnetic resonance at 1.5 T. Mid-ventricular short axis native and post-contrast T1 maps by Shortened MOdified Look-Locker Inversion-recovery (ShMOLLI), MOdified Look-Locker Inversion Recovery (MOLLI) [pre-contrast: 5s(3s)3s, post-contrast: 4s(1s)3s(1s)2s] and saturation recovery single-shot acquisition (SASHA) were acquired and ECV by these three techniques were derived for the mid anteroseptum. Mean age was 50 ± 14 years (range 20-76), male 52%, with no age difference between genders (males 51 ± 14 years; females 49 ± 15 years, P = 0.55). Quoting respectively ShMOLLI, MOLLI, SASHA throughout, mean myocardial T1 was 957 ± 30 ms, 1025 ± 38 ms, 1144 ± 45 ms (P < 0.0001) and ECV 28.4 ± 3.0% [95% confidence interval (CI) 27.8-29.0], 27.3 ± 2.7 (95% CI 26.8-27.9), 24.1 ± 2.9% (95% CI 23.5-24.7) (P < 0.0001), with all values higher in females for all techniques (T1 +18 ms, +35 ms, +51 ms; ECV +2.7%, +2.6%, +3.4%). Native myocardial T1 reduced slightly with age (R2 = 0.042, P = 0.048; R2 = 0.131, P < 0.0001-on average by 8-11 ms/decade-but not for SASHA (R2 = 0.033 and P = 0.083). ECV did not change with age (R2 = 0.003, P = 0.582; R2 = 0.002, P = 0.689; R2 = 0.003, P = 0.615). Heart rate decreased slightly with age (R2 = 0.075, coefficient = -0.273, P = 0.008), but there was no relationship between age and other blood T1 influences (haematocrit, iron, high density lipoprotein-cholesterol). Conclusion: Gender influences native T1 and ECV with women having a higher native T1 and ECV. Native T1 measured by MOLLI and ShMOLLI was slightly lower with increasing age but not with SASHA and ECV was independent of age for all techniques.

Assessing for Cardiotoxicity from Metal-on-Metal Hip Implants with Advanced Multimodality Imaging Techniques.

BACKGROUND: High failure rates of metal-on-metal (MoM) hip implants prompted regulatory authorities to issue worldwide safety alerts. Circulating cobalt from these implants causes rare but fatal autopsy-diagnosed cardiotoxicity. There is concern that milder cardiotoxicity may be common and underrecognized. Although blood metal ion levels are easily measured and can be used to track local toxicity, there are no noninvasive tests for organ deposition. We sought to detect correlation between blood metal ions and a comprehensive panel of established markers of early cardiotoxicity. METHODS: Ninety patients were recruited into this prospective single-center blinded study. Patients were divided into 3 age and sex-matched groups according to implant type and whole-blood metal ion levels. Group-A patients had a ceramic-on-ceramic [CoC] bearing; Group B, an MoM bearing and low blood metal ion levels; and Group C, an MoM bearing and high blood metal-ion levels. All patients underwent detailed cardiovascular phenotyping using cardiac magnetic resonance imaging (CMR) with T2*, T1, and extracellular volume mapping; echocardiography; and cardiac blood biomarker sampling. T2* is a novel CMR biomarker of tissue metal loading. RESULTS: Blood cobalt levels differed significantly among groups A, B, and C (mean and standard deviation [SD], 0.17 ± 0.08, 2.47 ± 1.81, and 30.0 ± 29.1 ppb, respectively) and between group A and groups B and C combined. No significant between-group differences were found in the left atrial or ventricle size, ejection fraction (on CMR or echocardiography), T1 or T2* values, extracellular volume, B-type natriuretic peptide level, or troponin level, and all values were within normal ranges. There was no relationship between cobalt levels and ejection fraction (R = 0.022, 95% confidence interval [CI] = -0.185 to 0.229) or T2* values (R = 0.108, 95% CI = -0.105 to 0.312). CONCLUSIONS: Using the best available technologies, we did not find that high (but not extreme) blood cobalt and chromium levels had any significant cardiotoxic effect on patients with an MoM hip implant. There were negligible-to-weak correlations between elevated blood metal ion levels and ejection fraction even at the extremes of the 95% CI, which excludes any clinically important association. LEVEL OF EVIDENCE: Therapeutic Level II. See Instructions for Authors for a complete description of levels of evidence.