Fast Marginal Likelihood Estimation of Penalties for Group-Adaptive Elastic Net.

Elastic net penalization is widely used in high-dimensional prediction and variable selection settings. Auxiliary information on the variables, for example, groups of variables, is often available. Group-adaptive elastic net penalization exploits this information to potentially improve performance by estimating group penalties, thereby penalizing important groups of variables less than other groups. Estimating these group penalties is, however, hard due to the high dimension of the data. Existing methods are computationally expensive or not generic in the type of response. Here we present a fast method for estimation of group-adaptive elastic net penalties for generalized linear models. We first derive a low-dimensional representation of the Taylor approximation of the marginal likelihood for group-adaptive ridge penalties, to efficiently estimate these penalties. Then we show by using asymptotic normality of the linear predictors that this marginal likelihood approximates that of elastic net models. The ridge group penalties are then transformed to elastic net group penalties by matching the ridge prior variance to the elastic net prior variance as function of the group penalties. The method allows for overlapping groups and unpenalized variables, and is easily extended to other penalties. For a model-based simulation study and two cancer genomics applications we demonstrate a substantially decreased computation time and improved or matching performance compared to other methods. Supplementary materials for this article are available online.

Multi-scale spatial modeling of immune cell distributions enables survival prediction in primary central nervous system lymphoma.

To understand the clinical significance of the tumor microenvironment (TME), it is essential to study the interactions between malignant and non-malignant cells in clinical specimens. Here, we established a computational framework for a multiplex imaging system to comprehensively characterize spatial contexts of the TME at multiple scales, including close and long-distance spatial interactions between cell type pairs. We applied this framework to a total of 1,393 multiplex imaging data newly generated from 88 primary central nervous system lymphomas with complete follow-up data and identified significant prognostic subgroups mainly shaped by the spatial context. A supervised analysis confirmed a significant contribution of spatial context in predicting patient survival. In particular, we found an opposite prognostic value of macrophage infiltration depending on its proximity to specific cell types. Altogether, we provide a comprehensive framework to analyze spatial cellular interaction that can be broadly applied to other technologies and tumor contexts.

Intraoperative conditions of patients undergoing pancreatoduodenectomy.

BACKGROUND: Postoperative pancreatic fistula (POPF) is a severe complication following pancreatoduodenectomy (PD). Previous research in colorectal surgery demonstrated suboptimal intraoperative conditions to be related with an increased risk of anastomotic leakage. Aim of this study was to evaluate the intraoperative condition of patients undergoing PD by both assessing whether these known intraoperative modifiable risk factors in colorectal surgery are also present during PD and by measuring compliance to intraoperative ERAS guidelines. Secondly, to determine the relation of these factors with POPF. MATERIALS AND METHODS: This prospective single center study included patients undergoing PD from 2016 to 2020. Parameters regarding the patient's general condition, local perfusion, oxygenation, surgical factors and ERAS elements were measured with a checklist intraoperatively, before the creation of the pancreatojejunal anastomosis. Uni- and multivariable logistic regression analyses were performed. RESULTS: 83 patients were included. POPF occurred in 27.7% (9.0% grade B, 10.0% grade C). Patients with POPF significantly had more other postoperative complications compared to patients without POPF (100% vs. 76.2%, p = 0.017). A suboptimal intraoperative condition was observed in 89.2%. Overall compliance to the intraoperative ERAS guideline was 0%. In univariable analysis, soft pancreatic tissue, pancreatic duct <3 mm, tumor location and intraoperative vasopressor administration were significantly associated with POPF. In multivariable analysis, only soft pancreatic tissue was independently associated with POPF (OR 13.627; 95% CI 1.656-112.157, p = 0.015). CONCLUSION: Awareness amongst surgeons and anesthesiologists should be created. The influence of these intraoperative factors on POPF should be further evaluated in future, larger studies.

Perioperative Hypoxemia and Postoperative Respiratory Events in Infants with Hypertrophic Pyloric Stenosis.

BACKGROUND: Normalization of metabolic alkalosis is an important pillar in the treatment of infantile hypertrophic pyloric stenosis (IHPS) because uncorrected metabolic alkalosis may lead to perioperative respiratory events. However, the evidence on the incidence of respiratory events is limited. We aimed to study the incidence of peroperative hypoxemia and postoperative respiratory events in infants undergoing pyloromyotomy and the potential role of metabolic alkalosis. MATERIALS AND METHODS: We retrospectively reviewed all patients undergoing pyloromyotomy between 2007 and 2017. All infants received intravenous fluids preoperatively to correct metabolic abnormalities close to normal. We assessed the incidence of perioperative hypoxemia (defined as oxygen saturation [SpO2] < 90% for > 1min) and postoperative respiratory events. Additionally, the incidence of difficult intubations was evaluated. We performed a multivariate logistic regression analysis to evaluate the association between admission or preoperative serum pH values, bicarbonate or chloride, and peri- and postoperative hypoxemia or respiratory events. RESULTS: Of 406 included infants, 208 (51%) developed 1 or more episodes of hypoxemia during the perioperative period, of whom 130 (32%) experienced it during induction, 43 (11%) intraoperatively, and 112 (28%) during emergence. About 7.5% of the infants had a difficult intubation and 17 required more than 3 attempts by a pediatric anesthesiologist. Three patients developed respiratory insufficiency and 95 postoperative respiratory events were noticed. We did not find a clinically meaningful association between laboratory values reflecting metabolic alkalosis and respiratory events. CONCLUSIONS: IHPS frequently leads to peri- and postoperative hypoxemia or respiratory events and high incidence of difficult tracheal intubations. Preoperative pH, bicarbonate, and chloride were bad predictors of respiratory events.

Trends in low vision service utilisation: A retrospective study based on general population healthcare claims.

PURPOSE: To identify parameters associated with the downward trend in the uptake of Low Vision Services (LVS) in the Netherlands. METHODS: A retrospective cohort study was conducted based on a Dutch national health insurance claims database (Vektis CV) of all adults (≥18 years) who received LVS from 2015 until 2018. Descriptive statistics were used to assess socio-demographic, clinical and contextual characteristics and other healthcare utilisation of the study population. General estimating equations trends in characteristics and healthcare utilisation were determined over time. RESULTS: A total of 49,726 unique patients received LVS, but between 2015 and 2018, the number of patients decreased by 15%. The majority was aged 65 years or older (53%), female (54%), had a middle (38%) or low (24%) socio-economic status and lived in urban areas (68%). Between 2015-2018, significant downward trends were found for treatment with intravitreal injections and lens-related diseases for LVS patients. For physical comorbidity, utilisation of ophthalmic care, low vision aids and occupational therapy, a significant upward trend was found over time. CONCLUSION: The decrease of Dutch LVS patients by 15% between 2015 and 2018 might be explained by a reduced distribution of patients treated with intravitreal injections and patients with lens-related diseases within the LVS. Compared to 2015, patients were more likely to have physical comorbidity, to see an ophthalmologist and to use low vision aids and occupational therapy in 2016, 2017 and 2018. This might indicate enhanced access to LVS when treated by ophthalmologists or within other medical specialties, or the opposite, i.e., less access when not treated within one of these medical specialties. Future research is needed to examine differences in patterns between LVS users and non-users further.

Somatostatin analogues for the prevention of pancreatic fistula after open pancreatoduodenectomy: A nationwide analysis.

BACKGROUND: Somatostatin analogues (SA) are currently used to prevent postoperative pancreatic fistula (POPF) development. However, its use is controversial. This study investigated the effect of different SA protocols on the incidence of POPF after pancreatoduodenectomy in a nationwide population. METHODS: All patients undergoing elective open pancreatoduodenectomy were included from the Dutch Pancreatic Cancer Audit (2014-2017). Patients were divided into six groups: no SA, octreotide, lanreotide, pasireotide, octreotide only in high-risk (HR) patients and lanreotide only in HR patients. Primary endpoint was POPF grade B/C. The updated alternative Fistula Risk Score was used to compare POPF rates across various risk scenarios. RESULTS: 1992 patients were included. Overall POPF rate was 13.1%. Lanreotide (10.0%), octreotide-HR (9.4%) and no protocol (12.7%) POPF rates were lower compared to the other protocols (varying from 15.1 to 19.1%, p = 0.001) in crude analysis. Sub-analysis in patients with HR of POPF showed a significantly lower rate of POPF when treated with lanreotide (10.0%) compared to no protocol, octreotide and pasireotide protocol (21.6-26.9%, p = 0.006). Octreotide-HR and lanreotide-HR protocol POPF rates were comparable to lanreotide protocol, however not significantly different from the other protocols. Multivariable regression analysis demonstrated lanreotide protocol to be positively associated with a low odds-ratio (OR) for POPF (OR 0.387, 95% CI 0.180-0.834, p = 0.015). In-hospital mortality rates were not affected. CONCLUSION: Use of lanreotide in all patients undergoing pancreatoduodenectomy has a potential protective effect on POPF development. Protocols for HR patients only might be favorable too. However, future studies are warranted to confirm these findings.

Development of a neighborhood drivability index and its association with transportation behavior in Toronto.

BACKGROUND: Car driving is a form of passive transport that is associated with an increase in physical inactivity, obesity, air pollution and noise. Built environment characteristics may influence transport mode choice, but comprehensive indices for built environment characteristics that drive car use are still lacking, while such an index could provide tangible policy entry points. OBJECTIVE: We developed and validated a neighbourhood drivability index, capturing combined dimensions of the neighbourhood environment in the City of Toronto, and investigated its association with transportation choices (car, public transit or active transport), overall, by trip length, and combined for residential neighbourhood and workplace drivability. METHODS: We used exploratory factor analysis to derive distinct factors (clusters of one or more environmental characteristics) that reflect the degree of car dependency in each neighbourhood, drawing from candidate variables that capture density, diversity, design, destination accessibility, distance to transit, and demand management. Area-level factor scores were then combined into a single composite score, reflecting neighbourhood drivability. Negative binomial generalized estimating equations were used to test the association between driveability quintiles (Q) and primary travel mode (>50% of trips by car, public transit, or walking/cycling) in a population-based sample of 63,766 Toronto residents enrolled in the Transportation Tomorrow Survey (TTS) wave 2016, adjusting for individual and household characteristics, and accounting for clustering of respondents within households. RESULTS: The drivability index consisted of three factors: Urban sprawl, pedestrian facilities and parking availability. Relative to those living in the least drivable neighbourhoods (Q1), those in high drivability areas (Q5) had a significantly higher rate of car travel (adjusted Risk Ratio (RR): 1.80, 95%CI: 1.77-1.88), and lower rate of public transit use (RR: 0.90, 95%CI: 0.85-0.94) and walking/cycling (RR: 0.22, 95%CI: 0.19-0.25). Associations were strongest for short trips (<3 km) (RR: 2.72, 95%CI: 2.48-2.92), and in analyses where both residential and workplace drivability was considered (RR for car use in high/high vs. low/low residential/workplace drivability: 2.18, 95%CI: 2.08-2.29). CONCLUSION: This novel neighbourhood drivability index predicted whether local residents drive or use active modes of transportation and can be used to investigate the association between drivability, physical activity, and chronic disease risk.

Oncological Outcomes After Anastomotic Leakage After Surgery for Colon or Rectal Cancer: Increased Risk of Local Recurrence.

OBJECTIVE: The aim of this study was to evaluate oncological outcome for patients with and without anastomotic leakage after colon or rectal cancer surgery. SUMMARY OF BACKGROUND DATA: The role of anastomotic leakage in oncological outcome after colorectal cancer surgery is still topic of debate and impact on follow-up and consideration for further treatment remains unclear. METHODS: Patients included in the international, multicenter, non-inferior, open label, randomized, controlled trials COLOR and COLOR II, comparing laparoscopic surgery for curable colon (COLOR) and rectal (COLOR II) cancer with open surgery, were analyzed. Patients operated by abdominoperineal excision were excluded. Both univariate and multivariate analyses were performed to investigate the impact of leakage on overall survival, disease-free survival, local and distant recurrences, adjusted for possible confounders. Primary endpoints in the COLOR and COLOR II trial were disease-free survival and local recurrence at 3-year follow-up, respectively, and secondary endpoints included anastomotic leakage rate. RESULTS: For colon cancer, anastomotic leakage was not associated with increased percentage of local recurrence or decreased disease-free-survival. For rectal cancer, an increase of local recurrences (13.3% vs 4.6%; hazard ratio 2.96; 95% confidence interval 1.38-6.34; P = 0.005) and a decrease of disease-free survival (53.6% vs 70.9%; hazard ratio 1.67; 95% confidence interval 1.16-2.41; P = 0.006) at 5-year follow-up were found in patients with anastomotic leakage. CONCLUSION: Short-term morbidity, mortality, and long-term oncological outcomes are negatively influenced by the occurrence of anastomotic leakage after rectal cancer surgery. For colon cancer, no significant effect was observed; however, due to low power, no conclusions on the influence of anastomotic leakage on outcomes after colon surgery could be reached. Clinical awareness of increased risk of local recurrence after anastomotic leakage throughout the follow-up is mandatory.Trial Registration: Registered with ClinicalTrials.gov, number NCT00387842 and NCT00297791.

LekCheck: A Prospective Study to Identify Perioperative Modifiable Risk Factors for Anastomotic Leakage in Colorectal Surgery.

OBJECTIVE: To assess potentially modifiable perioperative risk factors for anastomotic leakage in adult patients undergoing colorectal surgery. SUMMARY BACKGROUND DATA: Colorectal anastomotic leakage (CAL) is the single most important denominator of postoperative outcome after colorectal surgery. To lower the risk of CAL, the current research focused on the association of potentially modifiable risk factors, both surgical and anesthesiological. METHODS: A consecutive series of adult patients undergoing colorectal surgery with primary anastomosis was enrolled from January 2016 to December 2018. Fourteen hospitals in Europe and Australia prospectively collected perioperative data by carrying out the LekCheck, a short checklist carried out in the operating theater as a time-out procedure just prior to the creation of the anastomosis to check perioperative values on 1) general condition 2) local perfusion and oxygenation, 3) contamination, and 4) surgery related factors. Univariate and multivariate logistic regression analysis were performed to identify perioperative potentially modifiable risk factors for CAL. RESULTS: There were 1562 patients included in this study. CAL was reported in 132 (8.5%) patients. Low preoperative hemoglobin (OR 5.40, P < 0.001), contamination of the operative field (OR 2.98, P < 0.001), hyperglycemia (OR 2.80, P = 0.003), duration of surgery of more than 3 hours (OR 1.86, P = 0.010), administration of vasopressors (OR 1.80, P = 0.010), inadequate timing of preoperative antibiotic prophylaxis (OR 1.62, P = 0.047), and application of epidural analgesia (OR, 1.81, P = 0. 014) were all associated with CAL. CONCLUSIONS: This study identified 7 perioperative potentially modifiable risk factors for CAL. The results enable the development of a multimodal and multidisciplinary strategy to create an optimal perioperative condition to finally lower CAL rates.

18F-FDG PET baseline radiomics features improve the prediction of treatment outcome in diffuse large B-cell lymphoma.

PURPOSE: Accurate prognostic markers are urgently needed to identify diffuse large B-Cell lymphoma (DLBCL) patients at high risk of progression or relapse. Our purpose was to investigate the potential added value of baseline radiomics features to the international prognostic index (IPI) in predicting outcome after first-line treatment. METHODS: Three hundred seventeen newly diagnosed DLBCL patients were included. Lesions were delineated using a semi-automated segmentation method (standardized uptake value ≥ 4.0), and 490 radiomics features were extracted. We used logistic regression with backward feature selection to predict 2-year time to progression (TTP). The area under the curve (AUC) of the receiver operator characteristic curve was calculated to assess model performance. High-risk groups were defined based on prevalence of events; diagnostic performance was assessed using positive and negative predictive values. RESULTS: The IPI model yielded an AUC of 0.68. The optimal radiomics model comprised the natural logarithms of metabolic tumor volume (MTV) and of SUVpeak and the maximal distance between the largest lesion and any other lesion (Dmaxbulk, AUC 0.76). Combining radiomics and clinical features showed that a combination of tumor- (MTV, SUVpeak and Dmaxbulk) and patient-related parameters (WHO performance status and age > 60 years) performed best (AUC 0.79). Adding radiomics features to clinical predictors increased PPV with 15%, with more accurate selection of high-risk patients compared to the IPI model (progression at 2-year TTP, 44% vs 28%, respectively). CONCLUSION: Prediction models using baseline radiomics combined with currently used clinical predictors identify patients at risk of relapse at baseline and significantly improved model performance. TRIAL REGISTRATION NUMBER AND DATE: EudraCT: 2006-005,174-42, 01-08-2008.

The relationship between guilt feelings, conflicts with staff and satisfaction with care in relatives of nursing home residents with dementia: A longitudinal analysis.

INTRODUCTION: Family caregivers often experience guilt after nursing home placement. The aim of the present study was to describe family caregivers' guilt over time and assess the impact of conflicts with staff and satisfaction with care on guilt. METHOD: Data of 222 family caregivers at three assessments during one-year follow-up were used. In addition to caregivers' guilt and the variables conflicts with staff and satisfaction with the care, potential confounders were measured: sociodemographic data, clinical characteristics of the person with dementia, and caregiver burden. Linear mixed model analyses were performed to examine the longitudinal relationships between variables. RESULTS: Guilt remained stable over time. Unadjusted models showed that conflicts with staff were positively associated with guilt (ß = 0.11; p < 0.001; 95% CI: 0.06 to 0.16) and satisfaction with care showed a negative association with guilt (ß = -0.10; p< 0.05; 95% CI: -0.18 to -0.01). After adjusting for the confounders, only the positive association of guilt with conflicts with staff was similar as in the unadjusted analysis (ß = 0.11; p < 0.001; 95% CI: 0.05 to 0.16), whereas satisfaction with care was not significantly associated with guilt in the adjusted analyses (ß = -0.07; p = 0.10; 95% CI: -0.16 to 0.01). DISCUSSION: More conflicts with staff are associated with stronger guilt feelings. Guilt feelings are experienced by caregivers even after the admission of the person with dementia, and they remain stable over time. Further studies should focus on how to address guilt in family caregivers of people with dementia living in nursing homes.

Diagnostic Performance of [18F]FDG PET in Staging Grade 1-2, Estrogen Receptor Positive Breast Cancer.

Positron emission tomography using [18F]fluorodeoxyglucose (FDG PET) potentially underperforms for staging of patients with grade 1-2 estrogen receptor positive (ER+) breast cancer. The aim of this study was to retrospectively investigate the diagnostic accuracy of FDG PET in this patient population. Suspect tumor lesions detected on conventional imaging and FDG PET were confirmed with pathology or follow up. PET-positive lesions were (semi)quantified with standardized uptake values (SUV) and these were correlated with various pathological features, including the histological subtype. Pre-operative imaging detected 155 pathologically verified lesions (in 74 patients). A total of 115/155 (74.2%) lesions identified on FDG PET were classified as true positive, i.e., malignant (in 67 patients) and 17/155 (10.8%) lesions as false positive, i.e., benign (in 9 patients); 7/155 (4.5%) as false negative (in 7 patients) and 16/155 (10.3%) as true negative (in 14 patients). FDG PET incorrectly staged 16/70 (22.9%) patients. The FDG uptake correlated with histological subtype, showing higher uptake in ductal carcinoma, compared to lobular carcinoma (p < 0.05). Conclusion: Within this study, FDG PET inadequately staged 22.9% of grade 1-2, ER + BC cases. Incorrect staging can lead to inappropriate treatment choices, potentially affecting survival and quality of life. Prospective studies investigating novel radiotracers are urgently needed.

MRI Natural History of the Leukodystrophy Vanishing White Matter.

Background In vanishing white matter (VWM), a form of leukodystrophy, earlier onset is associated with faster clinical progression. MRI typically shows rarefaction and cystic destruction of the cerebral white matter. Information on the evolution of VWM according to age at onset is lacking. Purpose To determine whether nature and progression of cerebral white matter abnormalities in VWM differ according to age at onset. Materials and Methods Patients with genetically confirmed VWM were stratified into six groups according to age at onset: younger than 1 year, 1 year to younger than 2 years, 2 years to younger than 4 years, 4 years to younger than 8 years, 8 years to younger than 18 years, and 18 years or older. With institutional review board approval, all available MRI scans obtained between 1985 and 2019 were retrospectively analyzed with three methods: (a) ratio of the width of the lateral ventricles over the skull (ventricle-to-skull ratio [VSR]) was measured to estimate brain atrophy; (b) cerebral white matter was visually scored as percentage normal, hyperintense, rarefied, or cystic on fluid-attenuated inversion recovery (FLAIR) images and converted into a white matter decay score; and (c) the intracranial volume was segmented into normal-appearing white and gray matter, abnormal but structurally present (FLAIR-hyperintense) and rarefied or cystic (FLAIR-hypointense) white matter, and ventricular and extracerebral cerebrospinal fluid (CSF). Multilevel regression analyses with patient as a clustering variable were performed to account for the nested data structure. Results A total of 461 examinations in 270 patients (median age, 7 years [interquartile range, 3-18 years]; 144 female patients) were evaluated; 112 patients had undergone serial imaging. Patients with later onset had higher VSR [F(5) = 8.42; P < .001] and CSF volume [F(5) = 21.7; P < .001] and lower white matter decay score [F(5) = 4.68; P < .001] and rarefied or cystic white matter volume [F(5) = 13.3; P < .001]. Rate of progression of white matter decay scores [b = -1.6, t(109) = -3.9; P < .001] and VSRs [b = -0.05, t (109) = -3.7; P < .001] were lower with later onset. Conclusion A radiologic spectrum based on age at onset exists in vanishing white matter. The earlier the onset, the faster and more cystic the white matter decay, whereas with later onset, white matter atrophy and gliosis predominate. © RSNA, 2021.

Whole-body MRI versus an FDG-PET/CT-based reference standard for staging of paediatric Hodgkin lymphoma: a prospective multicentre study.

OBJECTIVES: To assess the concordance of whole-body MRI (WB-MRI) and an FDG-PET/CT-based reference standard for the initial staging in children with Hodgkin lymphoma (HL) METHODS: Children with newly diagnosed HL were included in this prospective, multicentre, international study and underwent WB-MRI and FDG-PET/CT at staging. Two radiologists and a nuclear medicine physician independently evaluated all images. Discrepancies between WB-MRI and FDG-PET/CT were assessed by an expert panel. All FDG-PET/CT errors were corrected to derive the FDG-PET/CT-based reference standard. The expert panel corrected all reader errors in the WB-MRI DWI dataset to form the intrinsic MRI data. Inter-observer agreement for WB-MRI DWI was calculated using overall agreement, specific agreements and kappa statistics. Concordance for correct classification of all disease sites and disease stage between WB-MRI (without DWI, with DWI and intrinsic WB-MRI DWI) and the reference standard was calculated as primary outcome. Secondary outcomes included positive predictive value, negative predictive value and kappa statistics. Clustering within patients was accounted for using a mixed-effect logistic regression model with random intercepts and a multilevel kappa analysis. RESULTS: Sixty-eight children were included. Inter-observer agreement between WB-MRI DWI readers was good for disease stage (κ = 0.74). WB-MRI DWI agreed with the FDG-PET/CT-based reference standard for determining disease stage in 96% of the patients versus 88% for WB-MRI without DWI. Agreement between WB-MRI DWI and the reference standard was excellent for both nodal (98%) and extra-nodal (100%) staging. CONCLUSIONS: WB-MRI DWI showed excellent agreement with the FDG-PET/CT-based reference standard. The addition of DWI to the WB-MRI protocol improved the staging agreement. KEY POINTS: • This study showed excellent agreement between WB-MRI DWI and an FDG-PET/CT-based reference standard for staging paediatric HL. • Diffusion-weighted imaging is a useful addition to WB-MRI in staging paediatric HL. • Inter-observer agreement for WB-MRI DWI was good for both nodal and extra-nodal staging and determining disease stage.

The Dutch Incidence of Infantile Hypertrophic Pyloric Stenosis and the Influence of Seasons.

INTRODUCTION: Studies report contradicting results on the incidence of infantile hypertrophic pyloric stenosis (IHPS) and its association with seasons. We aim to assess the IHPS incidence in the Netherlands and to determine whether seasonal variation is present in a nationwide cohort. MATERIALS AND METHODS: All infants with IHPS hospitalized in the Netherlands between 2007 and 2017 were included in this retrospective cohort study. Incidence rates per 1,000 livebirths (LB) were calculated using total number of LB during the matched month, season, or year, respectively. Seasonal variation based on month of birth and month of surgery was analyzed using linear mixed model and one-way ANOVA, respectively. RESULTS: A total of 2,479 infants were included, of which the majority was male (75.9%). Median (interquartile range) age at surgery was 34 (18) days. The average IHPS incidence rate was 1.28 per 1,000 LB (variation: 1.09-1.47 per 1,000 LB). We did not find a conclusive trend over time in IHPS incidence. Differences in incidence between season of birth and season of surgery were not significant (p = 0.677 and p = 0.206, respectively). CONCLUSION: We found an average IHPS incidence of 1.28 per 1,000 LB in the Netherlands. Our results showed no changing trend in incidence and no seasonal variation.

Machine learning-based analysis of [18F]DCFPyL PET radiomics for risk stratification in primary prostate cancer.

PURPOSE: Quantitative prostate-specific membrane antigen (PSMA) PET analysis may provide for non-invasive and objective risk stratification of primary prostate cancer (PCa) patients. We determined the ability of machine learning-based analysis of quantitative [18F]DCFPyL PET metrics to predict metastatic disease or high-risk pathological tumor features. METHODS: In a prospective cohort study, 76 patients with intermediate- to high-risk PCa scheduled for robot-assisted radical prostatectomy with extended pelvic lymph node dissection underwent pre-operative [18F]DCFPyL PET-CT. Primary tumors were delineated using 50-70% peak isocontour thresholds on images with and without partial-volume correction (PVC). Four hundred and eighty standardized radiomic features were extracted per tumor. Random forest models were trained to predict lymph node involvement (LNI), presence of any metastasis, Gleason score ≥ 8, and presence of extracapsular extension (ECE). For comparison, models were also trained using standard PET features (SUVs, volume, total PSMA uptake). Model performance was validated using 50 times repeated 5-fold cross-validation yielding the mean receiver-operator characteristic curve AUC. RESULTS: The radiomics-based machine learning models predicted LNI (AUC 0.86 ± 0.15, p < 0.01), nodal or distant metastasis (AUC 0.86 ± 0.14, p < 0.01), Gleason score (0.81 ± 0.16, p < 0.01), and ECE (0.76 ± 0.12, p < 0.01). The highest AUCs reached using standard PET metrics were lower than those of radiomics-based models. For LNI and metastasis prediction, PVC and a higher delineation threshold improved model stability. Machine learning pre-processing methods had a minor impact on model performance. CONCLUSION: Machine learning-based analysis of quantitative [18F]DCFPyL PET metrics can predict LNI and high-risk pathological tumor features in primary PCa patients. These findings indicate that PSMA expression detected on PET is related to both primary tumor histopathology and metastatic tendency. Multicenter external validation is needed to determine the benefits of using radiomics versus standard PET metrics in clinical practice.

Respiratory problems owing to severe metabolic alkalosis in infants presenting with hypertrophic pyloric stenosis.

OBJECTIVE: Uncorrected metabolic alkalosis in infantile hypertrophic pyloric stenosis (IHPS) could lead to perioperative apnea. However, the precise incidence of preoperative respiratory problems and the association with metabolic alkalosis are unknown. Therefore, we aimed to determine the incidence of preoperative respiratory problems in IHPS and to assess the association with metabolic alkalosis. METHODS: We retrospectively reviewed all patients diagnosed with IHPS during 2007-2017. Respiratory problems were classified as present or absent. With multivariate logistic regression we analyzed the association between bicarbonate and respiratory problems, corrected for gestational age and birth weight. RESULTS: We included 459 infants, of whom 23 developed preoperative respiratory problems (5.0%). Infants with preoperative respiratory problems were more often female (43.5% vs. 13.3% p = 0.001) and had significantly higher median serum levels of bicarbonate (32.0 mmol/L vs. 30.0 mmol/L), base excess (6.5 mmol/L vs. 5.3 mmol/L) and pCO2 (6.4 kPa vs. 5.9 kPa), compared to infants without respiratory problems. Multivariate analysis of serum bicarbonate and presence of respiratory problems showed an OR of 2.18 per 10 mmol/L (95% CI 1.21-4.71) (p = 0.009). The optimal bicarbonate cutoff point was 25.7 mmol/L (sensitivity 100%, specificity 13.4%). CONCLUSION: IHPS with metabolic alkalosis potentially results in preoperative respiratory problems. A lower bicarbonate target before surgery might be recommended and respiratory monitoring should be considered. LEVEL OF EVIDENCE: Level IV.