Volumetric analysis of acute uncomplicated type B aortic dissection using an automated deep learning aortic zone segmentation model.

BACKGROUND: Machine learning techniques have shown excellent performance in three-dimensional medical image analysis, but have not been applied to acute uncomplicated type B aortic dissection (auTBAD) using Society for Vascular Surgery (SVS) and Society of Thoracic Surgeons (STS)-defined aortic zones. The purpose of this study was to establish a trained, automatic machine learning aortic zone segmentation model to facilitate performance of an aortic zone volumetric comparison between patients with auTBAD based on the rate of aortic growth. METHODS: Patients with auTBAD and serial imaging were identified. For each patient, imaging characteristics from two computed tomography (CT) scans were analyzed: (1) the baseline CT angiography (CTA) at the index admission and (2) either the most recent surveillance CTA or the most recent CTA before an aortic intervention. Patients were stratified into two comparative groups based on aortic growth: rapid growth (diameter increase of ≥5 mm/year) and no or slow growth (diameter increase of <5 mm/year). Deidentified images were imported into an open source software package for medical image analysis and images were annotated based on SVS/STS criteria for aortic zones. Our model was trained using four-fold cross-validation. The segmentation output was used to calculate aortic zone volumes from each imaging study. RESULTS: Of 59 patients identified for inclusion, rapid growth was observed in 33 patients (56%) and no or slow growth was observed in 26 patients (44%). There were no differences in baseline demographics, comorbidities, admission mean arterial pressure, number of discharge antihypertensives, or high-risk imaging characteristics between groups (P > .05 for all). Median duration between baseline and interval CT was 1.07 years (interquartile range [IQR], 0.38-2.57). Postdischarge aortic intervention was performed in 13 patients (22%) at a mean of 1.5 ± 1.2 years, with no difference between the groups (P > .05). Among all patients, the largest relative percent increases in zone volumes over time were found in zone 4 (13.9%; IQR, -6.82 to 35.1) and zone 5 (13.4%; IQR, -7.78 to 37.9). There were no differences in baseline zone volumes between groups (P > .05 for all). The average Dice coefficient, a performance measure of the model output, was 0.73. Performance was best in zone 5 (0.84) and zone 9 (0.91). CONCLUSIONS: We describe an automatic deep learning segmentation model incorporating SVS-defined aortic zones. The open source, trained model demonstrates concordance to the manually segmented aortas with the strongest performance in zones 5 and 9, providing a framework for further clinical applications. In our limited sample, there were no differences in baseline aortic zone volumes between patients with rapid growth and patients with no or slow growth.

Multicentric clinical evaluation of a computed tomography-based fully automated deep neural network for aortic maximum diameter and volumetric measurements.

OBJECTIVE: This study aims to evaluate a fully automatic deep learning-based method (augmented radiology for vascular aneurysm [ARVA]) for aortic segmentation and simultaneous diameter and volume measurements. METHODS: A clinical validation dataset was constructed from preoperative and postoperative aortic computed tomography angiography (CTA) scans for assessing these functions. The dataset totaled 350 computed tomography angiography scans from 216 patients treated at two different hospitals. ARVA's ability to segment the aorta into seven morphologically based aortic segments and measure maximum outer-to-outer wall transverse diameters and compute volumes for each was compared with the measurements of six experts (ground truth) and thirteen clinicians. RESULTS: Ground truth (experts') measurements of diameters and volumes were manually performed for all aortic segments. The median absolute diameter difference between ground truth and ARVA was 1.6 mm (95% confidence interval [CI], 1.5-1.7; and 1.6 mm [95% CI, 1.6-1.7]) between ground truth and clinicians. ARVA produced measurements within the clinical acceptable range with a proportion of 85.5% (95% CI, 83.5-86.3) compared with the clinicians' 86.0% (95% CI, 83.9-86.0). The median volume similarity error ranged from 0.93 to 0.95 in the main trunk and achieved 0.88 in the iliac arteries. CONCLUSIONS: This study demonstrates the reliability of a fully automated artificial intelligence-driven solution capable of quick aortic segmentation and analysis of both diameter and volume for each segment.

Thoracic Endovascular Aortic Repair With Additional Distal Bare Stents in Type B Aortic Dissection Does Not Prevent Long-Term Aneurysmal Degeneration.

PURPOSE: TEVAR (thoracic endovascular aortic repair) + PETTICOAT (Provisional ExTension to Induce COmplete ATtachment) technique has been selectively employed since 2005 at our institution during endovascular treatment of type B aortic dissection (TBD). The aim of this study is to evaluate the long-term (>5 years) clinical results and the evolution of aortic volume. MATERIALS AND METHODS: All the patients receiving an endovascular treatment for TBD with the PETTICOAT technique were collected in a prospectively maintained database and follow-up computed tomography scan were retrospectively analyzed. Study endpoints included short- and long-term clinical success (absence of need for reintervention) and any major adverse event. The volumes of thoracic and abdominal aorta at long-term follow-up were also analyzed. RESULTS: Twenty-eight patients received a TEVAR + PETTICOAT and were followed up (median follow-up 85 months). Primary 30-day clinical success rate was 82% with an adverse event rate of 31%; 4 type I endoleak and 1 retrograde dissection were recorded. Secondary mid-term clinical success was 96% while the long-term clinical success rate was 79%. Six cases (21%) received either an open repair or an endovascular repair for a significant distal aortic enlargement at follow-up. With regards to volumetric analysis, an increase of overall (thoracic and abdominal) aortic volume was observed in 8 cases mainly related to an increase (mean: +31%) of the abdominal volume that was observed in 11 cases. CONCLUSIONS: PETTICOAT technique does not protect from long-term significant aneurysmal degeneration that may require aortic open or endovascular reinterventions. Aortic growth occurs mainly in the bare-stented aorta and thus, life-long surveillance is advisable in these patients.

Large false lumen area is a predictor of failed false lumen volume reduction after stent-graft repair in type B aortic dissection.

PURPOSE: To investigate the predictors of failed false lumen (FL) volume reduction at 12 months after stent-graft implantation in patients with type B aortic dissection. METHODS: The retrospective analysis comprised 38 patients (25 men; mean age 60±12 years) with double-barrel type B aortic dissection (9 acute) treated with thoracic endovascular aortic repair (TEVAR) and evaluated with serial computed tomography (CT) scans up to 12 months. Aortic volume changes were determined. Based on FL volume change at 1 year after stent-graft implantation, patients were dichotomized according to the presence or absence of FL volume reduction. Clinical and CT variables were compared between groups to determine risk factors of failed FL volume reduction. A major adverse event (MAE) was defined as death or reintervention. RESULTS: Patients were followed for 4.2±2.8 years. FL volume reduction (+FLVR) occurred in 27 (71%) patients, whereas 11 (29%) patients had no FL volume reduction (-FLVR). The MAE-free survival rate was significantly higher in the +FLVR patients than in the -FLVR group (88.9% vs. 27.3%, respectively; p=0.001). Chronicity of dissection, location of tear site, or the maximum total aortic lumen area was not associated with failure to achieve FL volume reduction. However, the maximum preprocedure FL area was significantly lower in the +FLVR group than in the -FLVR group (12.6±6.6 vs. 21.0±11.4 cm(2), respectively; p=0.041) and was an independent predictor for failed FL volume reduction (odds ratio 1.3, 95% confidence interval 1.02 to 1.70, p=0.031). CONCLUSION: Failed FL volume reduction after TEVAR was associated with a significantly increased rate of mortality or reintervention during follow-up. A larger preprocedure maximum FL area was a predictor of failed FL volume reduction after TEVAR in type B dissection.

Prognostic impact of early aortic volume changes at hospital discharge in patients with acute type B aortic dissection.

BACKGROUND: Early prediction of aorta-related events is important for determining subsequent treatment strategies in patients with acute aortic dissection. However, most studies evaluated long-term aortic growth rates by annual assessment. The purpose of our study was to determine whether the in-hospital growth rate of aortic volume was associated with aorta-related events. METHODS: We studied 116 patients with uncomplicated type B acute aortic dissection. We analyzed whether changes in aortic volume were associated with aorta-related events during a 5-year follow-up. According to the growth rate from admission to discharge, patients were divided into two groups: Increase >0 (aortic volume: n = 59, aortic diameter: n = 43) and Reduction ≤0 (aortic volume: n = 57, aortic diameter: n = 73) in maximum aortic diameter or aortic volume. The primary endpoint was the discriminative ability of the growth rate of aortic volume for aorta-related events. RESULTS: According to the evaluation of aortic volume changes, the Increase group had significantly higher aorta-related event rates than those in the Reduction group (49.2 % vs. 3.5 %, respectively; p < 0.001). Receiver operating characteristics analysis showed that the growth rate of aortic volume had a clearly useful discrimination, with an area under the curve of 0.84, whereas the discriminative ability of the growth rate of maximum aortic diameter was poor (area under the curve: 0.53). Multivariate analysis showed that the growth rate of aortic volume from admission to discharge was an independent predictor of aorta-related events (hazard ratio, 26.3; 95 % confidence interval, 2.04-286.49; p = 0.001). CONCLUSIONS: In-hospital evaluation of aortic volume was helpful to predict long-term aorta-related events in patients with uncomplicated type B acute aortic dissection.

3D Ultrasound Measurements Are Highly Sensitive to Monitor Formation and Progression of Abdominal Aortic Aneurysms in Mouse Models.

Background: Available mouse models for abdominal aortic aneurysms (AAAs) differ substantially in the applied triggers, associated pathomechanisms and rate of vessel expansion. While maximum aortic diameter (determined after aneurysm excision or by 2D ultrasound) is commonly applied to document aneurysm development, we evaluated the sensitivity and reproducibility of 3D ultrasound to monitor aneurysm growth in four distinct mouse models of AAA. Methods: The models included angiotensin-II infusion in ApoE deficient mice, topical elastase application on aortas in C57BL/6J mice (with or without oral administration of ß-aminoproprionitrile) and intraluminal elastase perfusion in C57BL/6J mice. AAA development was monitored using semi-automated 3D ultrasound for aortic volume calculation over 12 mm length and assessment of maximum aortic diameter. Results: While the models differed substantially in the time course of aneurysm development, 3D ultrasound measurements (volume and diameter) proved highly reproducible with concordance correlation coefficients > 0.93 and variations below 9% between two independent observers. Except for the elastase perfusion model where aorta expansion was lowest and best detected by diameter increase, all other models showed high sensitivity of absolute volume and diameter measurements in monitoring AAA formation and progression by 3D ultrasound. When compared to standard 2D ultrasound, the 3D derived parameters generally reached the highest effect size. Conclusion: This study has yielded novel information on the robustness and limitations of semi-automated 3D ultrasound analysis and provided the first direct comparison of aortic volume increase over time in four widely applied mouse models of AAA. While 3D ultrasound generally proved highly sensitive in detecting early AAA formation, the 3D based volume analysis was found inferior to maximum diameter assessment in the elastase perfusion model where the extent of inflicted local injury is determined by individual anatomical features.

3D morphometric analysis of ascending aorta as an adjunctive tool to predict type A acute aortic dissection.

BACKGROUND: Acute type A aortic dissection (AAAD) is a pathological process that implicates the ascending aorta and represents a surgical emergency burdened by high mortality if not promptly treated in the first hours of onset. Despite best efforts, the annual incidence rates of aortic dissection has remained stable over the past decades. We measured aortic dimensions (aortic diameters, area, length and volume) using 3D multiplanar reconstruction imaging with the purpose of refining the risk- morphology for AAAD. METHODS: Computerized tomography angiography studies of three groups were compared retrospectively: patients affected by AAAD (AAAD group; n=71), patients affected by aortic aneurysm and subsequently subjected to ascending aorta replacement (Aneurysm, n=77) and a healthy aorta's group (Control, n=75). RESULTS: Mean diameters of AAAD (4.9 cm) and Aneurysm (5.1 cm) aortas were significantly larger than those of the control group (3.4 cm). In AAAD patients, an ascending aorta diameter greater than 5.5 cm was observed in 18% of patients. Multiple comparisons showed statistically significant differences among mean of the ratio of aortic root area to height between the three groups (P<0.001). In frontal and sagittal planes, the length of the ascending aorta was significantly greater in patients affected by aortic pathology (AAAD and aneurysm) than in the control group (P<0.001). Significant differences were confirmed when indexing the aortic length to patient's height and BSA, and the aortic volume to patient's BSA. CONCLUSIONS: Maximum transverse diameter, considered separately, is not the best predictor of aortic dissection. In our opinion, the introduction into clinical practice of measurements of the area, length, and volume of the aorta, as absolute or indexed values, could improve the selection of patients who would benefit from preventive surgical aortic replacement.

Aortic volume determines global end-diastolic volume measured by transpulmonary thermodilution.

BACKGROUND: Global end-diastolic volume (GEDV) measured by transpulmonary thermodilution is regarded as indicator of cardiac preload. A bolus of cold saline injected in a central vein travels through the heart and lung, but also the aorta until detection in a femoral artery. While it is well accepted that injection in the inferior vena cava results in higher values, the impact of the aortic volume on GEDV is unknown. In this study, we hypothesized that a larger aortic volume directly translates to a numerically higher GEDV measurement. METHODS: We retrospectively analyzed data from 88 critically ill patients with thermodilution monitoring and who did require a contrast-enhanced thoraco-abdominal computed tomography scan. Aortic volumes derived from imaging were compared with GEDV measurements in temporal proximity. RESULTS: Median aortic volume was 194 ml (interquartile range 147 to 249 ml). Per milliliter increase of the aortic volume, we found a GEDV increase by 3.0 ml (95% CI 2.0 to 4.1 ml, p < 0.001). In case a femoral central venous line was used for saline bolus injection, GEDV raised additionally by 2.1 ml (95% CI 0.5 to 3.7 ml, p = 0.01) per ml volume of the vena cava inferior. Aortic volume explained 59.3% of the variance of thermodilution-derived GEDV. When aortic volume was included in multivariate regression, GEDV variance was unaffected by sex, age, body height, and weight. CONCLUSIONS: Our results suggest that the aortic volume is a substantial confounding variable for GEDV measurements performed with transpulmonary thermodilution. As the aorta is anatomically located after the heart, GEDV should not be considered to reflect cardiac preload. Guiding volume management by raw or indexed reference ranges of GEDV may be misleading.

The effect of ticagrelor on growth of small abdominal aortic aneurysms-a randomized controlled trial.

AIMS: To evaluate if ticagrelor, an effective platelet inhibitor without known non-responders, could inhibit growth of small abdominal aortic aneurysms (AAAs). METHODS AND RESULTS: In this multi-centre randomized controlled trial, double-blinded for ticagrelor and placebo, acetylic salicylic acid naïve patients with AAA and with a maximum aortic diameter 35-49 mm were included. The primary outcome was mean reduction in log-transformed AAA volume growth rate (%) measured with magnetic resonance imaging (MRI) at 12 months compared with baseline. Secondary outcomes include AAA-diameter growth rate and intraluminal thrombus (ILT) volume enlargement rate. A total of 144 patients from eight Swedish centres were randomized (72 in each group). MRI AAA volume increase was 9.1% for the ticagrelor group and 7.5% for the placebo group (P = 0.205) based on intention-to-treat analysis, and 8.5% vs. 7.4% in a per-protocol analysis (P = 0.372). MRI diameter change was 2.5 mm vs. 1.8 mm (P = 0.113), US diameter change 2.3 mm vs. 2.2 mm (P = 0.778), and ILT volume change 12.9% vs. 10.4% (P = 0.590). CONCLUSION: In this RCT, platelet inhibition with ticagrelor did not reduce growth of small AAAs. Whether the ILT has an important pathophysiological role for AAA growth cannot be determined based on this study due to the observed lack of thrombus modulating effect of ticagrelor. TRIAL REGISTRATION: The TicAAA trial is registered at the US National Institutes of Health (ClinicalTrials.gov) #NCT02070653.