Comparative analysis of radiomics and deep-learning algorithms for survival prediction in hepatocellular carcinoma.

To examine the comparative robustness of computed tomography (CT)-based conventional radiomics and deep-learning convolutional neural networks (CNN) to predict overall survival (OS) in HCC patients. Retrospectively, 114 HCC patients with pretherapeutic CT of the liver were randomized into a development (n = 85) and a validation (n = 29) cohort, including patients of all tumor stages and several applied therapies. In addition to clinical parameters, image annotations of the liver parenchyma and of tumor findings on CT were available. Cox-regression based on radiomics features and CNN models were established and combined with clinical parameters to predict OS. Model performance was assessed using the concordance index (C-index). Log-rank tests were used to test model-based patient stratification into high/low-risk groups. The clinical Cox-regression model achieved the best validation performance for OS (C-index [95% confidence interval (CI)] 0.74 [0.57-0.86]) with a significant difference between the risk groups (p = 0.03). In image analysis, the CNN models (lowest C-index [CI] 0.63 [0.39-0.83]; highest C-index [CI] 0.71 [0.49-0.88]) were superior to the corresponding radiomics models (lowest C-index [CI] 0.51 [0.30-0.73]; highest C-index [CI] 0.66 [0.48-0.79]). A significant risk stratification was not possible (p > 0.05). Under clinical conditions, CNN-algorithms demonstrate superior prognostic potential to predict OS in HCC patients compared to conventional radiomics approaches and could therefore provide important information in the clinical setting, especially when clinical data is limited.

Outcome After Conservative and Endovascular Treatment of Stanford Type B Aortic Intramural Hematomas - A Single-Center Retrospective Study.

OBJECTIVES: Aortic intramural hematoma (IMH) is a rare disease. Thus far, only limited data is available and the indications for conservative and endovascular treatment are not well defined. The aim of this study was to investigate clinical presentation, course, CT imaging features and outcome of patients with type B aortic IMHs. METHODS: We included all patients with type B IMHs between 2012 and 2021 in this retrospective monocentric study. Clinical data, localization, thickness of IMHs and the presence of ulcer-like projections (ULPs) was evaluated before and after treatment. RESULTS: Thirty five patients (20 females; 70.3 y ± 11 y) were identified. Almost all IMHs (n = 34) were spontaneous and symptomatic with back pain (n = 34). At the time of diagnosis, TEVAR was deemed indicated in 9 patients, 26 patients were treated primarily conservatively. During the follow-up, in another 16 patients TEVAR was deemed indicated. Endovascularly and conservatively treated patients both showed decrease in thickness after treatment. Patients without ULPs showed more often complete resolution of the IMH than patients with ULPs (endovascularly treated 90.9% (10/11) vs 71.4% (5/7); conservatively treated 71.4% (10/14) vs 33.3% (1/3); P = .207). Complications after TEVAR occurred in 32% and more frequently in patients treated primarily conservatively (37.5% vs 22.2%). No in-hospital mortality was observed during follow-up. CONCLUSIONS: Prognosis of IMH seems favourable in both surgically as well as conservatively treated patients. However, it is essential to identify patients at high risk for complications under conservative treatment, who therefore should be treated by TEVAR. In our study, ULPs seem to be an adverse factor for remodeling.

Development and Validation of a Model for Postpancreatectomy Hemorrhage Risk.

Importance: Postpancreatectomy hemorrhage (PPH) due to postoperative pancreatic fistula (POPF) is a life-threatening complication after pancreatoduodenectomy. However, there is no prediction tool for early identification of patients at high risk of late PPH. Objective: To develop and validate a prediction model for PPH. Design, Setting, and Participants: This retrospective prognostic study included consecutive patients with clinically relevant POPF who underwent pancreatoduodenectomy from January 1, 2009, to May 20, 2023, at the University Hospital Mannheim (derivation cohort), and from January 1, 2012, to May 31, 2022, at the University Hospital Dresden (validation cohort). Data analysis was performed from May 30 to July 29, 2023. Exposure: Clinical and radiologic features of PPH. Main Outcomes and Measures: Accuracy of a predictive risk score of PPH. A multivariate prediction model-the hemorrhage risk score (HRS)-was established in the derivation cohort (n = 139) and validated in the validation cohort (n = 154). Results: A total of 293 patients (187 [64%] men; median age, 69 [IQR, 60-76] years) were included. The HRS comprised 4 variables with associations: sentinel bleeding (odds ratio [OR], 35.10; 95% CI, 5.58-221.00; P < .001), drain fluid culture positive for Candida species (OR, 14.40; 95% CI, 2.24-92.20; P < .001), and radiologic proof of rim enhancement of (OR, 12.00; 95% CI, 2.08-69.50; P = .006) or gas within (OR, 12.10; 95% CI, 2.22-65.50; P = .004) a peripancreatic fluid collection. Two risk categories were identified with patients at low risk (0-1 points) and high risk (≥2 points) to develop PPH. Patients with PPH were predicted accurately in the derivation cohort (C index, 0.97) and validation cohort (C index 0.83). The need for more invasive PPH management (74% vs 34%; P < .001) and severe complications (49% vs 23%; P < .001) were more frequent in high-risk patients compared with low-risk patients. Conclusions and Relevance: In this retrospective prognostic study, a robust prediction model for PPH was developed and validated. This tool may facilitate early identification of patients at high risk for PPH.

Effect of Contrast Timing, Volume, and Flow Rate on Image Quality of Pulmonary Arteries in CTA.

PURPOSE: To evaluate the effect of contrast timing, contrast volume, and contrast flow rate on the image quality of pulmonary arteries in computed tomography angiography (CTA) and to assess if bolus-tracking region of interest (ROI) positioning in the left atrium, which is used for triple-rule-out CTA, allows for sufficient depiction of the pulmonary arteries. METHODS: In this retrospective single-center study, data were collected for patients who underwent thoracic CTA during a specific period. Two groups of 121 patients each were created based on bolus-tracking ROI positioning in the main pulmonary artery or left atrium using propensity score matching. Image quality of the pulmonary arteries was evaluated using quantitative and qualitative scores. Subgroups were formed to examine the influence of contrast volume and flow rate. Two radiologists determined if pulmonary embolism was present, if pulmonary embolism could be excluded with certainty, and from which level pulmonary embolism could be excluded with certainty. Interrater reliability also was evaluated. RESULTS: ROI positioning in the main pulmonary artery scored significantly higher compared with the left atrium. There was no significant difference in subgroups of patients who were examined with 60 mL or more contrast volume and less than 4 mL/s flow rate; scores were similar or better than in the overall study population. Pulmonary embolism was not able to be excluded with certainty for each 1 patient in these subgroups compared with a high percentage in the overall study population. DISCUSSION: ROI positioning in the left atrium in combination with the 60 mL or more contrast volume and less than 4 mL/s flow rate does not adversely affect depiction of the pulmonary arteries compared with conventional ROI positioning in the main pulmonary artery. CONCLUSION: When using 60 mL or more contrast volume and less than 4 mL/s flow rate, ROI positioning in the left atrium, which is used in triple-rule-out CTA, is sufficient for the assessment of pulmonary arteries.

Single-center initial experience with inner-branch complex EVAR in 44 patients.

Purpose: The use of inner-branch aortic stent grafts in the treatment of complex aortic pathologies aims at broad applicability and stable bridging stent sealing compared to other endovascular technologies. The objective of this study was to evaluate the early outcomes with a single manufacturer custom-made and off-the-shelf inner-branched endograft in a mixed patient cohort. Methods: This retrospective, monocentric study between 2019 and 2022 included 44 patients treated with inner-branched aortic stent grafts (iBEVAR) as custom-made device (CMD) or off-the-shelf device (E-nside) with at least four inner branches. The primary endpoints were technical and clinical success. Results: Overall, 77% (n = 34) and 23% (n = 10) of the patients (mean age 77 ± 6.5 years, n = 36 male) were treated with a custom-made iBEVAR with at least four inner branches and an off-the-shelf graft, respectively. Treatment indications were thoracoabdominal pathologies in 52.2% (n = 23), complex abdominal aneurysms in 25% (n = 11), and type Ia endoleaks in 22.7% (n = 10). Preoperative spinal catheter placement was performed in 27% (n = 12) of patients. Implantation was entirely percutaneous in 75% (n = 33). Technical success was 100%. Target vessel success manifested at 99% (178/180). There was no in-hospital mortality. Permanent paraplegia developed in 6.8% (n = 3) of patients. The mean follow-up was 12 months (range 0-52 months). Three late deaths (6.8%) occurred, one related to an aortic graft infection. Kaplan-Meier estimated 1-year survival manifested at 95% and branch patency at 98% (177/180). Re-intervention was necessary for a total of six patients (13.6%). Conclusions: Inner-branch aortic stent grafts provide a feasible option for the treatment of complex aortic pathologies, both elective (custom-made) and urgent (off-the-shelf). The technical success rate is high with acceptable short-term outcomes and moderate re-intervention rates comparable to existing platforms. Further follow-up will evaluate long-term outcomes.

Impact of the COVID-19 pandemic on therapeutic interventional oncology procedures and diagnostic CT/MRI examinations at a German university hospital.

PURPOSE: The COVID-19 pandemic led to the implementation of severe restrictions on public life in Germany and a reduction in the number of non-COVID patients presenting for care. The aim of this study was to measure the impact on the number of therapeutic interventional oncology procedures in relation to diagnostic imaging studies at a high-volume radiology department. MATERIALS AND METHODS: The numbers of therapeutic interventional oncology procedures and diagnostic CT/MRI examinations for the years 2010 to 2021 were extracted using the hospital information system. Monthly data from January 2010 to December 2019 were used to build forecasting models for the timeframe from January 2020 to December 2021. Real procedure numbers were compared with predicted numbers to calculate residual differences, which were considered statistically significant if the real number was outside the 95 % confidence interval (p < 0.05). RESULTS: During the first German lockdown (March/April 2020), the number of outpatient CT/MRI examinations decreased significantly, with a less pronounced decrease of overall CT/MRI numbers. The second German lockdown (January-May 2021) led to lower than predicted outpatient CT numbers, whereas outpatient MRI numbers in part even exceeded predicted numbers and overall CT/MRI numbers stayed within confidence limits. The lockdowns had a more pronounced negative effect on the number of oncological MRI examinations compared to CT examinations. The number of therapeutic interventional oncology procedures showed no significant decrease during both lockdowns. CONCLUSION: Lockdown measures had minor impact on the number of therapeutic interventional oncology procedures, possibly due to a shift from more resource-intensive therapies like surgery towards interventional oncology. The overall numbers of diagnostic imaging decreased during the first lockdown, while the second lockdown had less negative impact. The number of oncological MRI examinations was affected most severely. To avoid adverse outcomes, specific protocols for patient management during future pandemic outbreaks should be implemented and continuously adapted. KEY POINTS: · COVID-19 lockdowns had minor effect on therapeutic interventional oncology procedures.. · Numbers of diagnostic outpatient imaging procedures dropped markedly, especially during the first lockdown.. · The number of oncological MRI examinations showed a significant decrease during both lockdowns.. CITATION FORMAT: · Nebelung H, Radosa CG, Schön F et al. Impact of the COVID-19 pandemic on therapeutic interventional oncology procedures and diagnostic CT/MRI examinations at a German university hospital. Fortschr Röntgenstr 2023; 195: 707 - 712.

Plasma extracellular vesicle messenger RNA profiling identifies prognostic EV signature for non-invasive risk stratification for survival prediction of patients with pancreatic ductal adenocarcinoma.

BACKGROUND: The prognosis of pancreatic ductal adenocarcinoma (PDAC) is one of the most dismal of all cancers and the median survival of PDAC patients is only 6-8 months after diagnosis. While decades of research effort have been focused on early diagnosis and understanding of molecular mechanisms, few clinically useful markers have been universally applied. To improve the treatment and management of PDAC, it is equally relevant to identify prognostic factors for optimal therapeutic decision-making and patient survival. Compelling evidence have suggested the potential use of extracellular vesicles (EVs) as non-invasive biomarkers for PDAC. The aim of this study was thus to identify non-invasive plasma-based EV biomarkers for the prediction of PDAC patient survival after surgery. METHODS: Plasma EVs were isolated from a total of 258 PDAC patients divided into three independent cohorts (discovery, training and validation). RNA sequencing was first employed to identify differentially-expressed EV mRNA candidates from the discovery cohort (n = 65) by DESeq2 tool. The candidates were tested in a training cohort (n = 91) by digital droplet polymerase chain reaction (ddPCR). Cox regression models and Kaplan-Meier analyses were used to build an EV signature which was subsequently validated on a multicenter cohort (n = 83) by ddPCR. RESULTS: Transcriptomic profiling of plasma EVs revealed differentially-expressed mRNAs between long-term and short-term PDAC survivors, which led to 10 of the top-ranked candidate EV mRNAs being tested on an independent training cohort with ddPCR. The results of ddPCR enabled an establishment of a novel prognostic EV mRNA signature consisting of PPP1R12A, SCN7A and SGCD for risk stratification of PDAC patients. Based on the EV mRNA signature, PDAC patients with high risk displayed reduced overall survival (OS) rates compared to those with low risk in the training cohort (p = 0.014), which was successfully validated on another independent cohort (p = 0.024). Interestingly, the combination of our signature and tumour stage yielded a superior prognostic performance (p = 0.008) over the signature (p = 0.022) or tumour stage (p = 0.016) alone. It is noteworthy that the EV mRNA signature was demonstrated to be an independent unfavourable predictor for PDAC prognosis. CONCLUSION: This study provides a novel and non-invasive prognostic EV mRNA signature for risk stratification and survival prediction of PDAC patients.

ICU patients with infectious complications after abdominopelvic surgery: Is thoracic CT in addition to abdominal CT helpful?

BACKGROUND: The aim of this study was to assess the usefulness of adding thoracic CT to abdominal CT in intensive care unit (ICU) patients with signs of infection after abdominopelvic surgery. METHODS: 143 thoracoabdominal CTs of ICU patients with signs of infection after abdominopelvic surgery were retrospectively reviewed for thoracic pathologies. It was determined if pathologic findings were visible only on thoracic CT above the diaphragmatic dome or also on abdominal CT up to the diaphragmatic dome. All thoracic pathologies visible only above the diaphragmatic dome were retrospectively analyzed by an ICU physician in terms of clinical relevance. Diagnostic and therapeutic efficacy of thoracic CT were assessed with regard to an infectious focus and to other pathologic findings. RESULTS: 297 pathologic thoracic findings were recorded. 26 of the 297 findings could only be detected on images obtained above the diaphragmatic dome (in 23 of 143 CTs). A change in patient management was initiated due to only one of the 26 supradiaphragmatic findings. Diagnostic efficacy of thoracic CT in addition to abdominal CT to identify an infectious focus was 3.5% (95%-CI: 0.5-6.5%) and therapeutic efficacy was 0.7% (95%-CI: 0-2.1%). With regard to all pathologic thoracic findings, diagnostic efficacy was 16.1% (95%-CI: 10.1-22.1%) and therapeutic efficacy remained at 0.7%. CONCLUSIONS: Additional thoracic CT to detect an infectious focus in ICU patients after abdominopelvic surgery leads to identification of the focus in only 3.5% and to changes in patient management in only 0.7%. Other relevant findings are more common (16.1%), but very rarely affect patient management.

Robotic Esophagectomy Compared With Open Esophagectomy Reduces Sarcopenia within the First Postoperative Year: A Propensity Score-Matched Analysis.

INTRODUCTION: Sarcopenia is a known risk factor for adverse outcomes after esophageal cancer (EC) surgery. Robot-assisted minimally invasive esophagectomy (RAMIE) offers numerous advantages, including reduced morbidity and mortality. However, no evidence exists to date comparing the development of sarcopenia after RAMIE and open esophagectomy (OE). The objective was to evaluate whether the development of sarcopenia within the first postoperative year after esophagectomy is associated with the surgical approach: RAMIE versus OE. METHODS: A total of 168 patients with EC were analyzed who either underwent total robotic or fully open Ivor Lewis esophagectomy in a propensity score-matched analysis. Sarcopenia was assessed using the skeletal muscle index (cm2/m2) and psoas muscle thickness per height (mm/m) on axial computed tomography scans during the first postoperative year; in total 540 computed tomography scans were evaluated. RESULTS: After 1-to-1 propensity score matching for confounders, 67 patients were allocated to RAMIE and OE groups, respectively. Skeletal muscle index in the OE group was significantly lower compared with the RAMIE group at the third (43.2 ± 7.6 cm2/m2 versus 49.1 ± 6.9 cm2/m2, p = 0.001), sixth (42.7 ± 7.8 cm2/m2 versus 51.5 ± 8.2 cm2/m2, p < 0.001) and ninth (43.0 ± 7.0 cm2/m2 versus 49.9 ± 6.6 cm2/m2, p = 0.015) postoperative month. Similar results were recorded for psoas muscle thickness per height. CONCLUSIONS: To our knowledge, this study is the first to suggest a substantial benefit of RAMIE compared with open esophagectomy in terms of postoperative sarcopenia. These results add further evidence to support the implementation of the robotic approach in multimodal therapy of EC.

Value of Clinical Information on Radiology Reports in Oncological Imaging.

Radiological reporting errors have a direct negative impact on patient treatment. The purpose of this study was to investigate the contribution of clinical information (CI) in radiological reporting of oncological imaging and the dependence on the radiologists' experience level (EL). Sixty-four patients with several types of carcinomas and twenty patients without tumors were enrolled. Computed tomography datasets acquired in primary or follow-up staging were independently analyzed by three radiologists (R) with different EL (R1: 15 years; R2: 10 years, R3: 1 year). Reading was initially performed without and 3 months later with CI. Overall, diagnostic accuracy and sensitivity for primary tumor detection increased significantly when receiving CI from 77% to 87%; p = 0.01 and 73% to 83%; p = 0.01, respectively. All radiologists benefitted from CI; R1: 85% vs. 92%, p = 0.15; R2: 77% vs. 83%, p = 0.33; R3: 70% vs. 86%, p = 0.02. Overall, diagnostic accuracy and sensitivity for detecting lymphogenous metastases increased from 80% to 85% (p = 0.13) and 42% to 56% (p = 0.13), for detection of hematogenous metastases from 85% to 86% (p = 0.61) and 46% to 60% (p = 0.15). Specificity remained stable (>90%). Thus, CI in oncological imaging seems to be essential for correct radiological reporting, especially for residents, and should be available for the radiologist whenever possible.

Diagnostic performance of 18F-fluorodeoxyglucose-PET/MRI versus MRI alone in the diagnosis of pelvic recurrence of rectal cancer.

PURPOSE: To compare the diagnostic performance of 18F-fluorodeoxyglucose-PET/MRI and MRI in the diagnosis of pelvic recurrence of rectal cancer. METHODS: All PET/MRIs of patients in the follow-up of rectal cancer performed between 2011 and 2018 at our institution were retrospectively reviewed. Recurrence was confirmed/excluded either by histopathology or imaging follow-up (> 4 months). Four groups of readers (groups 1/2: one radiologist each, groups 3/4: one radiologist/one nuclear medicine physician) independently interpreted MRI and PET/MRI. The likelihood of recurrence was scored on a 5-point-scale. Inter-reader agreement, sensitivity, specificity, PPV/NPV and accuracy were assessed. ROC curve analyses were performed. RESULTS: Fourty-one PET/MRIs of 40 patients (mean 61 years ± 10.9; 11 women, 29 men) were included. Sensitivity of PET/MRI in detecting recurrence was 94%, specificity 88%, PPV/NPV 97% and 78%, accuracy 93%. Sensitivity of MRI was 88%, specificity 75%, PPV/NPV 94% and 60%, accuracy 85%. ROC curve analyses showed an AUC of 0.97 for PET/MRI and 0.92 for MRI, but the difference was not statistically significant (p = 0.116). On MRI more cases were scored as equivocal (12% versus 5%). Inter-reader agreement was substantial for PET/MRI and MRI (0.723 and 0.656, respectively). CONCLUSION: 18F-FDG-PET/MRI and MRI are accurate in the diagnosis of locally recurrent rectal cancer. Sensitivity, specificity, PPV, NPV and accuracy are comparable for both modalities, but PET/MRI increases readers' confidence levels and reduces the number of equivocal cases.

Radioembolization versus portal vein embolization for contralateral liver lobe hypertrophy: effect of cirrhosis.

PURPOSE: Preoperative hypertrophy induction of future liver remnant (FLR) reduces the risk of postoperative liver insufficiency after partial hepatectomy. One of the most commonly used methods to induce hypertrophy of FLR is portal vein embolization (PVE). Recent studies have shown that transarterial radioembolization (TARE) also induces hypertrophy of the contralateral liver lobe. The aim of our study was to evaluate contralateral hypertrophy after TARE versus after PVE taking into account the effect of cirrhosis. METHODS: Forty-nine patients undergoing PVE before hemihepatectomy and 24 patients with TARE as palliative treatment for liver malignancy were retrospectively included. Semi-automated volumetry of the FLR/contralateral liver lobe before and after intervention (20 to 65 days) was performed on CT or MRI, and the relative increase in volume was calculated. Cirrhosis was evaluated independently by two radiologists on CT/MRI, and interrater reliability was calculated. RESULTS: Hypertrophy after PVE was significantly more pronounced than after TARE (25.3% vs. 7.4%; p < 0.001). In the subgroup of patients without cirrhosis, the difference was also statistically significant (25.9% vs. 8.6%; p = 0.002), whereas in patients with cirrhosis, the difference was not statistically significant (18.2% vs. 7.4%; p = 0.212). After PVE, hypertrophy in patients without cirrhosis was more pronounced than in patients with cirrhosis (25.9% vs. 18.2%; p = 0.203), while after TARE, hypertrophy was comparable in patients with and without cirrhosis (7.4% vs. 8.6%; p = 0.928). CONCLUSION: TARE induces less pronounced hypertrophy of the FLR compared to PVE. Cirrhosis seems to be less of a limiting factor for hypertrophy after TARE, compared to PVE.

Coronary computed tomography angiography (CCTA): effect of bolus-tracking ROI positioning on image quality.

OBJECTIVES: The aim of the study was to evaluate the effect of bolus-tracking ROI positioning on coronary computed tomography angiography (CCTA) image quality. METHODS: In this retrospective monocentric study, all patients had undergone CCTA by step-and-shoot mode to rule out coronary artery disease within a cohort at intermediate risk. Two groups were formed, depending on ROI positioning (left atrium (LA) or ascending aorta (AA)). Each group contained 96 patients. To select pairs of patients, propensity score matching was used. Image quality with regard to coronary arteries as well as pulmonary arteries was evaluated using quantitative and qualitative scores. RESULTS: In terms of the coronary arteries, there was no significant difference between both groups using quantitative (SNR AA 14.92 vs. 15.46; p = 0.619 | SNR LM 19.80 vs. 20.30; p = 0.661 | SNR RCA 24.34 vs. 24.30; p = 0.767) or qualitative scores (4.25 vs. 4.29; p = 0.672), respectively. With regard to pulmonary arteries, we found significantly higher quantitative (SNR RPA 8.70 vs. 5.89; p < 0.001 | SNR LPA 9.06 vs. 6.25; p < 0.001) and qualitative scores (3.97 vs. 2.24; p < 0.001) for ROI positioning in the LA than for ROI positioning in the AA. CONCLUSIONS: ROI positioning in the LA or the AA results in comparable image quality of CT coronary arteriography, while positioning in the LA leads to significantly higher image quality of the pulmonary arteries. These results support ROI positioning in the LA, which also facilitates triple-rule-out CT scanning. KEY POINTS: • ROI positioning in the left atrium or the ascending aorta leads to comparable image quality of the coronary arteries. • ROI positioning in the left atrium results in significantly higher image quality of the pulmonary arteries. • ROI positioning in the left atrium is feasible to perform triple-rule-out CTA.