Recent literature in Emergency Radiology introduced reproducible, quantitative imaging markers that may improve triage and early intervention in high-stakes scenarios. This month, we explore three studies that introduce novel metrics for assessing disease severity and predicting outcomes in strangulated bowel obstruction, hemodynamic instability after trauma, and spontaneous intracerebral hemorrhage.
1. Intestine wall-to-Liver CT Attenuation Ratio as a Predictor of Ischemic Bowel in Obstruction
Prompt identification of bowel ischemia in strangulated bowel obstruction (SBO) is critical for timely surgical intervention and improved patient outcomes. In this retrospective single-center study of 52 patients who underwent emergency surgery for suspected SBO, Fujishima et al. introduced the intestinal-to-liver CT attenuation ratio as a quantitative imaging biomarker.
This ratio is calculated by measuring the mean attenuation (in Hounsfield Units) of three least-enhanced points of the affected intestinal wall and dividing it by the mean attenuation of three points of the liver parenchyma on equilibrium-phase contrast-enhanced CT. The liver serves as an internal control due to its relatively stable and homogeneous enhancement pattern, making the ratio less susceptible to timing or systemic variability.
Patients who required intestinal resection (n=35) exhibited significantly lower attenuation values in suspected ischemic bowel walls and correspondingly lower intestinal-to-liver attenuation ratios compared to patients who did not require resection. A cutoff ratio of <0.40 yielded a 92% resection rate, compared to 44% for those above the threshold.
To assess the performance of this ratio as a predictive tool, the authors calculated the Area Under the Receiver Operating Characteristic Curve (AUROC) to be 0.886. For reference, an AUROC of 0.5 suggests no better than chance, while 1.0 indicates perfect prediction. As the ratio achieved an AUROC of 0.886, this suggests strong diagnostic performance in distinguishing patients with ischemic bowel requiring resection from those without.
This study provides compelling support for the integration of quantitative attenuation ratios into clinical workflows, particularly in equivocal SBO cases where subjective assessment may be insufficient. Importantly, the use of the liver as an internal reference may mitigate individual variation in contrast timing and systemic perfusion.
2. Shock Thyroid Imaging Score (STIS): A Prognostic Tool in Trauma Imaging
The concept of “shock thyroid” (CT findings of thyroid gland hypo-enhancement, heterogeneity, swelling, or perithyroidal fluid in the absence of direct injury) has been previously observed in trauma patients, but its prognostic relevance has not been systematically evaluated. Mueller et al. aimed to formalize this pattern into a quantifiable tool, the Shock Thyroid Imaging Score (STIS), and assess its association with (1) imaging features of the hypovolemic shock complex (HSC), (2) in-hospital mortality, and (3) predictive performance compared to the Glasgow Coma Scale (GCS) and systolic blood pressure (SBP) at presentation.
The STIS ranges from 0 to 8 and is calculated by assigning a score of 0, 1, or 2 to each of the following four CT features:
Thyroid enhancement (0 = normal, 1 = low, 2 = very low)
Thyroid heterogeneity (0 = none/mild, 1 = moderate, 2 = severe)
Thyroid swelling (0 = none, 1 = mild, 2 = striking)
Perithyroidal fluid (0 = none, 1 = mild, 2 = striking)
The individual component scores are summed to generate the final STIS. Higher scores reflect more severe imaging signs of systemic shock affecting the thyroid gland.
| STIS Score | Interpretation | In-Hospital Mortality | Associated Findings |
|---|---|---|---|
| 0 | Normal thyroid appearance on CT | ~13.6% mortality | Baseline; low incidence of HSC features |
| 1–2 | Mild abnormalities (e.g., mild heterogeneity or minimal hypoenhancement) | Intermediate risk | Increasing prevalence of subtle HSC signs |
| 3 | Moderate abnormalities (e.g., multiple mild findings or one striking feature) | Elevated risk | Higher rates of IVC flattening, adrenal changes, shock bowel |
| >3 | Severe findings (e.g., multiple striking abnormalities: very low enhancement, fluid) | 50% mortality | Strong association with full HSC and poor outcomes |
In this retrospective single-center study of 748 hemodynamically unstable trauma patients, higher STIS was significantly associated with both in-hospital mortality and features of the hypovolemic shock complex (HSC). Patients with STIS >3 demonstrated a 50% in-hospital mortality rate, compared to 13.6% in those with STIS = 0.
Multivariable analysis confirmed that STIS outperformed both Glasgow Coma Scale (GCS) and systolic blood pressure (SBP) in predicting mortality, with an odds ratio of 1.51 per unit increase in score. While reader variability was noted for some features, the findings suggest that STIS may serve as a reproducible adjunctive tool in early trauma triage, particularly given the universal usage of chest imaging in trauma protocols.
3. Spot Sign vs. Leakage Sign: Predicting Hematoma Expansion and Mortality in Spontaneous ICH
Intracerebral hemorrhage (ICH) is associated with significant early mortality, with hematoma expansion serving as a key prognostic determinant. This prospective, single-center study of 94 patients compared the spot sign (focal increased density within a hematoma on arterial phase imaging) and leakage sign (10 mm ROI on arterial and delayed phase imaging with positive sign being 10% increase in HU on delayed phase) as predictors of hematoma expansion, neurological deterioration, and death.
| Outcome | Leakage Sign OR (95% CI) | Spot Sign OR (95% CI) | Interpretation |
|---|---|---|---|
| Hematoma expansion | 9.27 (2.95–29.20) | 4.40 (1.47–13.13) | Patients with a leakage sign were ~9 times more likely to have expansion. |
| Neurological deterioration | 26.67 (1.62–47.39) | 7.93 (1.47–42.77) | Leakage sign was associated with ~27x odds of neurological decline. |
| In-hospital mortality | 7.56 (2.97–19.25) | 7.05 (2.60–17.60) | Both signs strongly predicted death, but leakage sign had slightly stronger effect. |
The leakage sign outperformed the spot sign in nearly all measured outcomes. These high odds ratios underscore the strong predictive value of the leakage sign, as patients with this finding were up to 26 times more likely to experience neurological decline and over 7 times more likely to die in the hospital. In contrast, while the spot sign remained a significant predictor, it consistently demonstrated lower odds ratios across outcomes. While both signs demonstrated high specificity, the leakage sign demonstrated greater sensitivity, likely due to its ability to capture ongoing extravasation not visible in a single arterial-phase snapshot. Patients with a positive leakage sign had a median survival of just 6 days, compared to 54 days for those with a spot sign alone and 110 days for those with neither (p < 0.001).
These findings support the routine inclusion of delayed-phase imaging in ICH protocols to improve early risk stratification by capturing the leakage sign. This ultimately may lend to a more proactive clinical approach with regards to ICU admission, surgical intervention, and prognostic counseling.
Conclusion
Together, these studies reflect a growing emphasis on standardized, quantifiable imaging metrics in emergency radiology. Whether assessing bowel viability, systemic shock response, or intracranial hematoma evolution, these tools offer objectivity, reproducibility, and predictive value beyond traditional clinical parameters.
As imaging takes a greater role in the trauma setting where clinical exams are limited and time-sensitive decisions must be made, radiologists have the opportunity to provide critical early triage and outcome stratification with quantifiable imaging metrics. Future prospective validation studies will be essential in refining these tools and expanding their utility across broader patient populations and pathologies.
1. Fujishima S, Tsujimoto H, Yaguchi Y, et al. Preoperative intestine-to-liver CT ratio: useful predictor of resection in strangulated obstruction. Emerg Radiol. Published online July 4, 2025. doi:10.1007/s10140-025-02369-8
2. Mueller MC, du Plessis J, Mohammed AME, et al. Promising initial evaluation: the Shock Thyroid Imaging Score (STIS) could predict mortality in hemodynamically unstable trauma patients. Emerg Radiol. Published online May 27, 2025. doi:10.1007/s10140-025-02356-z
3. González Domínguez MC, Fornell-Pérez R, Santana Suárez E, et al. Comparative assessment of the spot sign and leakage sign as predictive factors for spontaneous intracranial hematoma expansion. Emerg Radiol. Published online May 19, 2025. doi:10.1007/s10140-025-02352-3
