Correlation of patient-reported outcome measures to performance-based function in critical care survivors: PREDICTABLE

BackgroundEstablishing sequela following critical illness is a public health priority; however, recruitment and retention of this cohort make assessing functional outcomes difficult. Completing patient-reported outcome measures (PROMs) via telephone may improve participant and researcher involvement; however, there is little evidence regarding the correlation of PROMs to performance-based outcome measures in critical care survivors.ObjectivesThe objective of this study was to assess the relationship between self-reported and performance-based measures of function in survivors of critical illness.MethodsThis was a nested cohort study of patients enrolled within a previously published study determining predictors of disability-free survival. Spearman's correlation (r_s) was calculated between four performance-based outcomes (the Functional Independence Measure [FIM], 6-min walk distance [6MWD], Functional Reach Test [FRT], and grip strength) that were collected during a home visit 6 months following their intensive care unit admission, with two commonly used PROMs (World Health Organization Disability Assessment Scale 2.0 12 Level [WHODAS 2.0] and EuroQol-5 Dimension-5 Level [EQ-5D-5L]) obtained via phone interview (via the PREDICT study) at the same time point.ResultsThere were 38 PROMs obtained from 40 recruited patients (mean age = 59.8 ± 16 yrs, M:F = 24:16). All 40 completed the FIM and grip strength, 37 the 6MWD, and 39 the FRT. A strong correlation was found between the primary outcome of the WHODAS 2.0 with all performance-based outcomes apart from grip strength where a moderate correlation was identified. Although strong correlations were also established between the EQ-5D-5L utility score and the FIM, 6MWD, and FRT, it only correlated weakly with grip strength. The EQ-5D overall global health rating only had very weak to moderate correlations with the performance-based outcomes.ConclusionThe WHODAS 2.0 correlated stronger across multiple performance-based outcome measures of functional recovery and is recommended for use in survivors of critical illness.     

Appraising the quality of guidelines for peripheral arterial catheters care: A systematic review of reviews

IntroductionCatheter-related bloodstream infections are among the most critical adverse events in critical patients with peripheral arterial catheters (ACs). Adherence to evidence-based guidelines can prevent and reduce arterial bloodstream infections.ObjectiveThe objectives of this study were to assess clinical practice guidelines for AC care and analyse methodological factors related to their development for effective dissemination and implementation in clinical practice.Review method usedThis was a systematic review of guidelines.Data sourcesWe searched PubMed, CINAHL, EMBASE, CNKI, and WANFANG databases from inception until September 2021 and evaluated websites of organisations that complied or produced guidelines.Review methodsA comprehensive list of guidelines for ACs care was included. We excluded incomplete guidelines, guidelines translated in other languages, duplicate publications, and summaries of multiple guidelines. Two reviewers independently extracted and collected the data, and three authors conducted quality assessments independently using the Appraisal of Guidelines for Research and Evaluation, Second Edition (AGREE II) tool. The intraclass correlation coefficient (two-way random) with a 95% confidence interval was used to evaluate the concordance between reviewers.ResultsOf the 738 total publications screened, seven were selected for evaluation. The concordance between observers was substantial (intraclass correlation coefficient >0.9, P < 0.001). Most guidelines (4/6) were developed in the United States and the United Kingdom. The median scores for the six domains were 89.0%, 65.5%, 58.0%, 86.0%, 65.0%, and 86.0%. The domains of stakeholder involvement, rigour of development, and applicability had the lowest scores. Guidelines by the United Kingdom's National Institute for Health and Care Excellence showed the highest quality.ConclusionsThe guidelines we included scored poorly on crucial domains (rigour of development, applicability, and stakeholder involvement). Most of the current recommendations on ACs were included in the guidelines for vascular catheter–related bloodstream infections. Therefore, targeted guidelines created specifically for ACs are warranted to reduce the incidence of catheter-related complications and ensure patient safety.     

Bounding box-based 3D AI model for user-guided volumetric segmentation of pancreatic ductal adenocarcinoma on standard-of-care CTs

ObjectivesTo develop a bounding-box-based 3D convolutional neural network (CNN) for user-guided volumetric pancreas ductal adenocarcinoma (PDA) segmentation.MethodsReference segmentations were obtained on CTs (2006–2020) of treatment-naïve PDA. Images were algorithmically cropped using a tumor-centered bounding box for training a 3D nnUNet-based-CNN. Three radiologists independently segmented tumors on test subset, which were combined with reference segmentations using STAPLE to derive composite segmentations. Generalizability was evaluated on Cancer Imaging Archive (TCIA) (n = 41) and Medical Segmentation Decathlon (MSD) (n = 152) datasets.ResultsTotal 1151 patients [667 males; age:65.3 ± 10.2 years; T1:34, T2:477, T3:237, T4:403; mean (range) tumor diameter:4.34 (1.1–12.6)-cm] were randomly divided between training/validation (n = 921) and test subsets (n = 230; 75% from other institutions). Model had a high DSC (mean ± SD) against reference segmentations (0.84 ± 0.06), which was comparable to its DSC against composite segmentations (0.84 ± 0.11, p = 0.52). Model-predicted versus reference tumor volumes were comparable (mean ± SD) (29.1 ± 42.2-cc versus 27.1 ± 32.9-cc, p = 0.69, CCC = 0.93). Inter-reader variability was high (mean DSC 0.69 ± 0.16), especially for smaller and isodense tumors. Conversely, model's high performance was comparable between tumor stages, volumes and densities (p > 0.05). Model was resilient to different tumor locations, status of pancreatic/biliary ducts, pancreatic atrophy, CT vendors and slice thicknesses, as well as to the epicenter and dimensions of the bounding-box (p > 0.05). Performance was generalizable on MSD (DSC:0.82 ± 0.06) and TCIA datasets (DSC:0.84 ± 0.08).ConclusionA computationally efficient bounding box-based AI model developed on a large and diverse dataset shows high accuracy, generalizability, and robustness to clinically encountered variations for user-guided volumetric PDA segmentation including for small and isodense tumors.Clinical relevanceAI-driven bounding box-based user-guided PDA segmentation offers a discovery tool for image-based multi-omics models for applications such as risk-stratification, treatment response assessment, and prognostication, which are urgently needed to customize treatment strategies to the unique biological profile of each patient's tumor.     

Tislelizumab plus chemotherapy as first-line treatment for recurrent or metastatic nasopharyngeal cancer: A multicenter phase 3 trial (RATIONALE-309)

Download : Download video (256MB)     

Ventilator weaning and extubation practices in critically ill children: An Australian and New Zealand survey of practice

ObjectivesWe aimed to (i) describe current weaning and extubation practices in children (protocols to identify weaning candidates, spontaneous breathing trials, and other aspects of care such as sedation weaning) and (ii) understand responsibilities for ventilation weaning decisions across Australia and New Zealand (ANZ).MethodsA cross-sectional survey of ANZ intensive care units who routinely intubate and ventilate children (<18 years) was conducted. We worked with the Australian and New Zealand Intensive Care Society Paediatric Study Group to identify units and potential respondents (senior nurse representative per unit) and to administer questionnaires. Survey questions (n = 35) examined current protocols, practices, unit staffing, and decision-making responsibilities for ventilation weaning and extubation. Open-ended questions examined respondents' experiences of weaning and extubation.ResultsA senior nursing respondent from 18/22 intensive care units (82%) completed the survey. Across units, most used sedation assessment tools (88%), and less often, sedation weaning tools (55%). Spontaneous awakening protocols were not used; one unit (5%) reported the use of a spontaneous breathing protocol. Two respondents reported that ventilation weaning protocols (11%) were in use, with 44% of units reporting the use of extubation protocols. Weaning and extubation practices were largely perceived as medically driven, with qualitative data demonstrating a desire from most respondents for greater shared decision-making.ConclusionIn ANZ, ventilation weaning and extubation practices are largely medically driven with variation in the use of protocols to support mechanical ventilation weaning and extubation in children. Our findings highlight the importance of future research to determine the impact of greater collaboration of the multidisciplinary team on weaning practices.     

Internal and external validation of machine learning–assisted prediction models for mechanical ventilation–associated severe acute kidney injury

BackgroundCurrently, very few preventive or therapeutic strategies are used for mechanical ventilation (MV)-associated severe acute kidney injury (AKI).ObjectivesWe developed clinical prediction models to detect the onset of severe AKI in the first week of intensive care unit (ICU) stay during the initiation of MV.MethodsA large ICU database Medical Information Mart for Intensive Care IV (MIMIC-IV) was analysed retrospectively. Data were collected from the clinical information recorded at the time of ICU admission and during the initial 12 h of MV. Using univariate and multivariate analyses, the predictors were selected successively. For model development, two machine learning algorithms were compared. The primary goal was to predict the development of AKI stage 2 or 3 (AKI-23) and AKI stage 3 (AKI-3) in the first week of patients’ ICU stay after initial 12 h of MV. The developed models were externally validated using another multicentre ICU database (eICU Collaborative Research Database, eICU) and evaluated in various patient subpopulations.ResultsModels were developed using data from the development cohort (MIMIC-IV: 2008–2016; n = 3986); the random forest algorithm outperformed the logistic regression algorithm. In the internal (MIMIC-IV: 2017–2019; n = 1210) and external (eICU; n = 1494) validation cohorts, the incidences of AKI-23 were 154 (12.7%) and 119 (8.0%), respectively, with areas under the receiver operator characteristic curve of 0.78 (95% confidence interval [CI]: 0.74–0.82) and 0.80 (95% CI: 0.76–0.84); the incidences of AKI-3 were 81 (6.7%) and 67 (4.5%), with areas under the receiver operator characteristic curve of 0.81 (95% CI: 0.76–0.87) and 0.80 (95% CI: 0.73–0.86), respectively.ConclusionsModels driven by machine learning and based on routine clinical data may facilitate the early prediction of MV-associated severe AKI. The validated models can be found at: https://apoet.shinyapps.io/mv_aki_2021_v2/.