Early CT-Derived Heart Tissue Measure Flags Death Risk in Aortic Dissection

The finding, published in the Egyptian Journal of Radiology and Nuclear Medicine, comes from a retrospective single-center study of 100 patients and offers a possible new prognostic tool drawn from images that are already acquired as the gold standard for diagnosing this life-threatening condition.

Aortic dissection is a relatively rare but highly fatal emergency in which blood penetrates the aortic intima and enters the media layer, progressively separating the walls of the body’s largest artery. Cardiac complications, including acute aortic regurgitation, myocardial ischemia and congestive heart failure, are among the most common problems and increase mortality. Because computed tomography angiography is the standard imaging test for confirming the diagnosis, the study’s authors asked whether the same routine scan could yield additional prognostic information without any extra radiation or contrast.

The measurement in question is myocardial extracellular volume, or ECV, which represents the non-cardiomyocyte portion of the heart muscle—the interstitial space between cells filled with interstitial fluid and fibrous material. ECV values rise in conditions involving protein accumulation such as fibrosis or amyloidosis. While ECV quantification is well established with cardiac MRI, it can also be performed with CT by exploiting the distribution of iodinated contrast between the blood pool and the myocardium. Traditionally, however, CT-based ECV requires a dedicated delayed acquisition seven to ten minutes after contrast injection, or photon-counting detector CT with virtual non-contrast images—both of which add complexity, radiation exposure or contrast load.

To bypass that requirement, the team calculated what they call early myocardial ECV directly from the arterial phase of routine thoracoabdominal CTA scans, obtained roughly 20 to 35 seconds after contrast injection. Regions of interest of approximately 80 to 100 square millimeters were placed in the midventricular interventricular septum and in the left ventricular blood pool, avoiding papillary muscles, on both non-contrast and arterial-phase images. ECV was then computed as the differential attenuation ratio between myocardium and blood pool, corrected for the patient’s hematocrit. The researchers also measured myocardial relative enhancement, or MRE, defined as the ratio of myocardial to blood-pool attenuation on the arterial phase alone. All scans were acquired on a 64-slice multidetector scanner with ECG gating, using 100 to 120 milliliters of iohexol injected at 4 milliliters per second.

The retrospective cohort drew on 218 consecutive patients who underwent thoracoabdominal CTA for suspected aortic dissection at the institution between January 2018 and December 2022. Ninety-seven had dissection; 47 were excluded, 44 for missing unenhanced scans and 3 for artifacts, leaving 50 patients with confirmed dissection and 50 controls in whom dissection had been suspected but was ruled out by CTA. Each group contained 36 men, or 72 percent, and mean ages were similar at 58.34 plus or minus 11.8 years in the dissection group and 60.46 plus or minus 16.8 years in controls. Among the dissection patients, 25 had Stanford type A and 25 had Stanford type B disease. Twenty-four patients underwent surgery, five underwent endovascular aneurysm repair, and the remaining 42 percent were managed medically.

Five patients in the dissection group died, and four of the five deaths occurred among those with Stanford type A dissection. Coronary artery involvement was identified in 3 of the 25 type A patients; mortality occurred in one of those three and in two of the 22 type A patients without coronary involvement, a difference that was not statistically significant. Hematocrit levels did not differ between groups. Notably, when dissection patients were compared with controls as a whole, early ECV values were nearly identical—14.89 plus or minus 7.2 percent versus 15.09 plus or minus 7.3 percent—and MRE values likewise showed no significant difference. Early ECV also showed no correlation with age or sex.

The picture changed when patients were stratified by dissection type. Within the dissection group, early myocardial ECV was significantly higher in patients with Stanford type A dissection, at 17.47 plus or minus 7.9 percent, than in those with type B, at 11.73 plus or minus 5.1 percent, a difference with a p-value of 0.004. MRE moved in the opposite direction, lower in type A at 32.12 plus or minus 7.8 compared with 40.4 plus or minus 12.5 in type B, with p equal to 0.007. Across all 100 participants, early ECV and MRE showed a strong positive correlation, with a Spearman’s rho of 0.872 and a p-value below 0.001. The authors suggest that the higher ECV in type A disease, which involves the ascending aorta, may reflect hemodynamic stress on the heart given its proximity to the affected segment, as well as indirect mechanisms such as coronary involvement causing ischemia, regurgitant flow from valvular involvement, or hemopericardium leading to tamponade.

The mortality analysis focused exclusively on the 50 dissection patients. Using a stepwise binary logistic regression that included age, hematocrit, attenuation values at several locations, early ECV, MRE and Stanford classification, the model’s explanatory power improved from a Nagelkerke R-squared of 0.238 in the first step to 0.440 in the second, and the Hosmer–Lemeshow test suggested good fit. Stanford classification alone was not a significant predictor of mortality. In the second step, however, early ECV emerged as a significant predictor of death, with an odds ratio of 1.283 and a 95 percent confidence interval of 1.002 to 1.640, and a p-value of 0.048. In practical terms, each 1 percent increase in early ECV was associated with a 28.3 percent increase in the odds of mortality. The authors themselves characterize this as a weak association, noting that the confidence interval barely excludes 1.

Several important limitations temper the findings. The study was single-center and retrospective with a small sample, and only five mortality events occurred—so few that statistical power was limited and overfitting in the multivariable model is a real risk. Region-of-interest measurements were performed by a single reader, so interobserver variability could not be assessed. Moreover, the early arterial phase used for ECV quantification has not yet been validated as a clinical biomarker; the recommended standard is late iodine CTA performed seven to ten minutes after injection, and while recent work in breast cancer patients suggests strong correlations between ECV measured at 1, 3 and 7 minutes, the correlation between delayed-phase and early CTA-derived ECV remains to be established. Coronary involvement, which may itself influence myocardial ECV, was too rare in this cohort to assess its independent effect.

Even with those caveats, the study is, to the authors’ knowledge, the first to evaluate myocardial ECV in patients with aortic dissection, and it aligns with a growing body of work showing CT-derived ECV as a prognostic marker in other cardiovascular settings, including severe aortic stenosis, pulmonary embolism and therapy-related cardiotoxicity. For now, the authors urge caution, given the marginal statistical significance of the mortality association, and call for further research to establish the measure’s clinical relevance.

Subject of Research: Early myocardial extracellular volume from routine CT angiography as a prognostic marker for survival in aortic dissection

Article Title: Prognostic value of early myocardial extracellular volume assessment in survival of aortic dissection

Article References: Buz Yaşar, A., Hızal, M., & Erdem, K. (2026). Prognostic value of early myocardial extracellular volume assessment in survival of aortic dissection. Egyptian Journal of Radiology and Nuclear Medicine, 57(1), Article 229. https://doi.org/10.1186/s43055-026-01899-6

Image Credits: AI Generated

DOI: 10.1186/s43055-026-01899-6

Keywords: aortic dissection, myocardial extracellular volume, CT angiography, Stanford classification, mortality prediction, cardiac imaging, ECV quantification, myocardial relative enhancement, logistic regression, retrospective study, radiology, prognosis