Prediction of the characteristics of aggressiveness of pancreatic neuroendocrine neoplasms (PanNENs) based on CT radiomic features.

OBJECTIVES: To predict tumor grade (G1 vs. G2/3), presence of distant metastasis (M+), metastatic lymph nodes (N+), and microvascular invasion (VI) of pancreatic neuroendocrine neoplasms (PanNEN) based on preoperative CT radiomic features (RFs), by applying a machine learning approach aimed to limit overfit. METHODS: This retrospective study included 101 patients who underwent surgery for PanNEN; the entire population was split into training (n = 70) and validation cohort (n = 31). Based on a previously validated methodology, after tumor segmentation on contrast-enhanced CT, RFs were extracted from unenhanced CT images. In addition, conventional radiological and clinical features were combined with RFs into multivariate logistic regression models using minimum redundancy and a bootstrap-based machine learning approach. For each endpoint, models were trained and validated including only RFs (RF_model), and both (radiomic and clinicoradiological) features (COMB_model). RESULTS: Twenty-five patients had G2/G3 tumor, 37 N+, and 14 M+ and 38 were shown to have VI. From a total of 182 RFs initially extracted, few independent radiomic and clinicoradiological features were identified. For M+ and G, the resulting models showed moderate to high performances: areas under the curve (AUC) for training/validation cohorts were 0.85/0.77 (RF_model) and 0.81/0.81 (COMB_model) for M+ and 0.67/0.72 and 0.68/0.70 for G. Concerning N+ and VI, only the COMB_model could be built, with poorer performance for N+ (AUC = 0.72/0.61) compared to VI (0.82/0.75). For all endpoints, the negative predictive value was good (≥ 0.75). CONCLUSIONS: Combining few radiomic and clinicoradiological features resulted in presurgical prediction of histological characteristics of PanNENs. Despite the limited risk of overfit, external validations are warranted. KEY POINTS: • Histology is the only tool currently available allowing characterization of PanNEN biological characteristics important for prognostic assessment; significant limitations to this approach exist. • Based upon preoperative contrast-enhanced CT images, a machine learning approach optimized to favor models' generalizability was successfully applied to train predictive models for tumor grading (G1 vs. G2/3), microvascular invasion, metastatic lymph nodes, and distant metastatic spread. • Moderate to high discriminative models (AUC: 0.67-0.85) based on few parameters (≤ 3) showing high negative predictive value (0.75-0.98) were generated and then successfully validated.

Preoperative assessment of microvessel density in nonfunctioning pancreatic neuroendocrine tumors (NF-PanNETs).

BACKGROUND: Hypervascularization is a typical feature of pancreatic neuroendocrine tumors, and it frequently allows their recognition at imaging studies. However, the density of microvessels in pancreatic neuroendocrine tumors changes according to their biological behavior, and a low microvessel density is associated with higher disease aggressiveness. The primary aim was to investigate the relationship between microvessel density and aggressiveness of nonfunctioning pancreatic neuroendocrine tumors. The secondary aim was to evaluate the ability of contrast-enhanced computed tomography and contrast-enhanced endoscopic ultrasound in predicting tumor microvessel density. METHODS: The patients who underwent surgery for nonfunctioning pancreatic neuroendocrine tumors (n = 66) with an available preoperative contrast-enhanced computed tomography (n = 39) and/or contrast-enhanced endoscopic ultrasound (n = 37) performed at San Raffaele Hospital (2016-2020) were included. The tumor vascularization was assessed by CD-34 staining, contrast-enhanced computed tomography, and contrast-enhanced endoscopic ultrasound. Median microvessel density (165 microvessels/mm2) was chosen as the cutoff to define low microvessel density and high microvessel density. RESULTS: The patients with a low microvessel density showed a significantly higher frequency of nodal metastases (P = .026), G2-G3 tumors (P = .022), and death domain-associated protein/α-thalassemia/mental retardation syndrome X-linked loss (P = .011) compared to patients with high microvessel density. The contrast-enhanced computed tomography tumor density in the arterial phase was significantly higher in patients with high microvessel density compared to those with low microvessel density (P = .016). The patients with a low microvessel density showed a significantly higher frequency of contrast-enhanced endoscopic ultrasound arterial hypoenhancement (P = .042) and late washout (P = .034). Contrast-enhanced computed tomography arterial hypoenhancement (P = .007) and contrast-enhanced endoscopic ultrasound late washout (P = .048) independently predicted a low microvessel density in the patients who underwent contrast-enhanced computed tomography and contrast-enhanced endoscopic ultrasound, respectively. CONCLUSION: A low microvessel density represents a marker of aggressiveness in the patients with nonfunctioning pancreatic neuroendocrine tumors. Contrast-enhanced computed tomography and contrast-enhanced endoscopic ultrasound are reliable and easily available tools for preoperative assessment of microvessel density.

Prediction of Early Distant Recurrence in Upfront Resectable Pancreatic Adenocarcinoma: A Multidisciplinary, Machine Learning-Based Approach.

Despite careful selection, the recurrence rate after upfront surgery for pancreatic adenocarcinoma can be very high. We aimed to construct and validate a model for the prediction of early distant recurrence (<12 months from index surgery) after upfront pancreaticoduodenectomy. After exclusions, 147 patients were retrospectively enrolled. Preoperative clinical and radiological (CT-based) data were systematically evaluated; moreover, 182 radiomics features (RFs) were extracted. Most significant RFs were selected using minimum redundancy, robustness against delineation uncertainty and an original machine learning bootstrap-based method. Patients were split into training (n = 94) and validation cohort (n = 53). Multivariable Cox regression analysis was first applied on the training cohort; the resulting prognostic index was then tested in the validation cohort. Clinical (serum level of CA19.9), radiological (necrosis), and radiomic (SurfAreaToVolumeRatio) features were significantly associated with the early resurge of distant recurrence. The model combining these three variables performed well in the training cohort (p = 0.0015, HR = 3.58, 95%CI = 1.98-6.71) and was then confirmed in the validation cohort (p = 0.0178, HR = 5.06, 95%CI = 1.75-14.58). The comparison of survival curves between low and high-risk patients showed a p-value <0.0001. Our model may help to better define resectability status, thus providing an actual aid for pancreatic adenocarcinoma patients' management (upfront surgery vs. neoadjuvant chemotherapy). Independent validations are warranted.

Predictors of Pneumothorax/Pneumomediastinum in Mechanically Ventilated COVID-19 Patients.

OBJECTIVE: To determine the incidence, predictors, and outcome of pneumothorax (PNX)/pneumomediastinum (PMD) in coronavirus disease 2019 (COVID-19) acute respiratory distress syndrome (ARDS). DESIGN: Observational study. SETTING: Tertiary-care university hospital. PARTICIPANTS: One hundred sixteen consecutive critically ill, invasively ventilated patients with COVID-19 ARDS. INTERVENTIONS: The authors collected demographic, mechanical ventilation, imaging, laboratory, and outcome data. Primary outcome was the incidence of PNX/PMD. Multiple logistic regression analyses were performed to identify predictors of PNX/PMD. MEASUREMENTS AND MAIN RESULTS: PNX/PMD occurred in a total of 28 patients (24.1%), with 22 patients developing PNX (19.0%) and 13 developing PMD (11.2%). Mean time to development of PNX/PMD was 14 ± 11 days from intubation. The authors found no significant difference in mechanical ventilation parameters between patients who developed PNX/PMD and those who did not. Mechanical ventilation parameters were within recommended limits for protective ventilation in both groups. Ninety-five percent of patients with PNX/PMD had the Macklin effect (linear collections of air contiguous to the bronchovascular sheaths) on a baseline computed tomography scan, and tended to have a higher lung involvement at intensive care unit (ICU) admission (Radiographic Assessment of Lung Edema score 32.2 ± 13.4 v 18.7 ± 9.8 in patients without PNX/PMD, p = 0.08). Time from symptom onset to intubation and time from total bilirubin on day two after ICU admission were the only independent predictors of PNX/PMD. Mortality was 60.7% in patients who developed PNX/PMD versus 38.6% in those who did not (p = 0.04). CONCLUSION: PNX/PMD occurs frequently in COVID-19 patients with ARDS requiring mechanical ventilation, and is associated with increased mortality. Development of PNX/PMD seems to occur despite use of protective mechanical ventilation and has a radiologic predictor sign.

Severely low testosterone in males with COVID-19: A case-control study.

BACKGROUND: Circulating androgens could have a relevant pathobiological role in clinical outcomes in men with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection (COVID-19). OBJECTIVES: We aimed to assess: (a) circulating sex steroids levels in a cohort of 286 symptomatic men with laboratory-confirmed COVID-19 at hospital admission compared to a cohort of 281 healthy men; and (b) the association between serum testosterone levels (tT), COVID-19, and clinical outcomes. MATERIALS AND METHODS: Demographic, clinical, and hormonal values were collected for all patients. Hypogonadism was defined as tT ≤9.2 nmol/l. The Charlson Comorbidity Index (CCI) was used to score health-significant comorbidities. Severe clinical outcomes were defined as patients either transferred to intensive care unit (ICU) or death. Descriptive statistics and multivariable linear and logistic regression models tested the association between clinical and laboratory variables and tT levels. Univariable and multivariable logistic regression models tested the association between tT and severe clinical outcomes. RESULTS: Overall, a significantly lower levels of LH and tT were found in patients with COVID-19 compared to healthy controls (all p < 0.0001); conversely, healthy controls depicted lower values of circulating E2 (p < 0.001). Testosterone levels suggestive for hypogonadism were observed in 257 (89.8%) patients at hospital admission. In as many as 243 (85%) cases, hypogonadism was secondary. SARS-CoV-2 infection status was independently associated with lower tT levels (p < 0.0001) and greater risk of hypogonadism (p < 0.0001), after accounting for age, BMI, CCI, and IL-6 values. Lower tT levels were associated with higher risk of ICU admission and death outcomes (all p ≤ 0.05), after accounting for clinical and laboratory parameters. CONCLUSIONS: We unveil an independent association between SARS-CoV-2 infection status and secondary hypogonadism already at hospital admission, with lower testosterone levels predicting the most severe clinical outcomes.

Initial chest radiographs and artificial intelligence (AI) predict clinical outcomes in COVID-19 patients: analysis of 697 Italian patients.

OBJECTIVE: To evaluate whether the initial chest X-ray (CXR) severity assessed by an AI system may have prognostic utility in patients with COVID-19. METHODS: This retrospective single-center study included adult patients presenting to the emergency department (ED) between February 25 and April 9, 2020, with SARS-CoV-2 infection confirmed on real-time reverse transcriptase polymerase chain reaction (RT-PCR). Initial CXRs obtained on ED presentation were evaluated by a deep learning artificial intelligence (AI) system and compared with the Radiographic Assessment of Lung Edema (RALE) score, calculated by two experienced radiologists. Death and critical COVID-19 (admission to intensive care unit (ICU) or deaths occurring before ICU admission) were identified as clinical outcomes. Independent predictors of adverse outcomes were evaluated by multivariate analyses. RESULTS: Six hundred ninety-seven 697 patients were included in the study: 465 males (66.7%), median age of 62 years (IQR 52-75). Multivariate analyses adjusting for demographics and comorbidities showed that an AI system-based score ≥ 30 on the initial CXR was an independent predictor both for mortality (HR 2.60 (95% CI 1.69 - 3.99; p < 0.001)) and critical COVID-19 (HR 3.40 (95% CI 2.35-4.94; p < 0.001)). Other independent predictors were RALE score, older age, male sex, coronary artery disease, COPD, and neurodegenerative disease. CONCLUSION: AI- and radiologist-assessed disease severity scores on CXRs obtained on ED presentation were independent and comparable predictors of adverse outcomes in patients with COVID-19. TRIAL REGISTRATION: ClinicalTrials.gov NCT04318366 ( https://clinicaltrials.gov/ct2/show/NCT04318366 ). KEY POINTS: • AI system-based score ≥ 30 and a RALE score ≥ 12 at CXRs performed at ED presentation are independent and comparable predictors of death and/or ICU admission in COVID-19 patients. • Other independent predictors are older age, male sex, coronary artery disease, COPD, and neurodegenerative disease. • The comparable performance of the AI system in relation to a radiologist-assessed score in predicting adverse outcomes may represent a game-changer in resource-constrained settings.

Experience of ventriculo-peritoneal shunt insertion in late presenting congenital hydrocephalus.

OBJECTIVE: To evaluate the outcomes of insertion of ventriculo-peritoneal shunts in paediatric neglected hydrocephalus. METHODS: The quasi-experimental study was conducted at the Combined Military Hospital, Rawalpindi, from January 2012 to June 2014, and comprised infants of both genders who presented late with congenital hydrocephalus, having fronto-occipital circumference more than 98 percentile of matched age group. Pre-operative assessment was based upon detailed history, clinical examination, laboratory investigations and computed tomography scan of head. After insertion of ventriculo-peritoneal shunt, patients were followed up for outcomes. Data was analysed using SPSS 17. RESULTS: Of the 30 infants, 12(40%) were girls and 18(60%) were boys. Overall mean age was 7.73±1.41 months (range: 5-10 months). Mean fronto-occipital circumference was 54.30±3.08. Cerebrospinal fluid infection was documented in 12(40%). Abdominal wound complications were observed in 7(23.3%) infants. Ventriculo-peritoneal shunt was removed in 18(60%) and eventually replaced in 18(60%). In-hospital mortality on account of complications was encountered in 13(43.3%). Correlation of fronto-occipital circumference to mortality was significant (p=0.001). CONCLUSIONS: To shunt or not to shunt remains a dilemma for poor-risk infants, but timely reporting of infants with hydrocephalus, proper case selection may improve the outcome of surgical intervention.

Early autologous cranioplasty after decompressive hemi-craniectomy for severe traumatic brain injury.

OBJECTIVE: To evaluate the outcome of early replacement of autologous bone flap for decompressive hemicraniectomy in severe traumatic brain injury patients. METHODS: The observational cross-sectional prospective study was conducted at the Neurosurgical Unit of the Combined Military Hospital, Rawalpindi, Pakistan, from July 2011, to June 2014, and comprised patients who underwent cranioplasty after decompressive hemicraniectomy for trauma. Patients over 20 years of age and of either gender were included. Cranioplasty was timed in all these patients using native bone flap preserved in the abdominal wall after decompressive craniectomy. Parameters recorded were mortality, wound infection, subdural collection, wound dehiscence, ventriculomegaly, bone resorption, cosmetic deformity and neurological outcome. SPSS 17 was used for data analysis. RESULTS: Of the 30 patients in the study, 28(93.3%) were males. The overall mean age was age 32.03±8.01 years (range: 20-48 years). Mean cranioplasty time was 66.2±11.50 days (range: 44-89 days). Major infection necessitating bone flap removal was found in 1(3.33%) patient, while minor scalp wound infections, treated with antibiotics and dressings were found in 2(6.66%). Cosmetic18 deformity was seen in 3(10%). Improved neurological outcome was noted in 21(70 %) patients; 6(20%) survived with a moderate to severe disability and 3(10%) remained in a vegetative state. No mortality was found after the procedure. CONCLUSIONS: Early autologous bone replacement for decompressive hemicraniectomy in severe traumatic brain injury patients offered cost-effective, acceptable surgical and improve dneurological outcome.