Poor Discrimination Research Articles

A Hybrid Spectral Library and Protein Sequence Database Search Strategy for Bottom-Up and Top-Down Proteomic Data Analysis.

Tandem mass spectrometry (MS/MS) is widely employed for the analysis of complex proteomic samples. While protein sequence database searching and spectral library searching are both well-established peptide identification methods, each has shortcomings. Protein sequence databases lack fragment peak intensity information, which can result in poor discrimination between correct and incorrect spectrum assignments. Spectral libraries usually contain fewer peptides than protein sequence databases, which limits the number of peptides that can be identified. Notably, few post-translationally modified peptides are represented in spectral libraries. This is because few search engines can both identify a broad spectrum of PTMs and create corresponding spectral libraries. Also, programs that generate spectral libraries using deep learning approaches are not yet able to accurately predict spectra for the vast majority of PTMs. Here, we address these limitations through use of a hybrid search strategy that combines protein sequence database and spectral library searches to improve identification success rates and sensitivity. This software uses Global PTM Discovery (G-PTM-D) to produce spectral libraries for a wide variety of different PTMs. These features, along with a new spectrum annotation and visualization tool, have been integrated into the freely available and open-source search engine MetaMorpheus.

Expanding Active Surveillance Criteria for Low- and Intermediate-risk Prostate Cancer: Can We Accurately Predict the Risk of Misclassification for Patients Diagnosed by Multiparametric Magnetic Resonance Imaging–targeted Biopsy?

Models predicting the risk of adverse pathology (ie, International Society of Urological Pathology [ISUP] grade group ≥3, pT3, and/or pN1) among patients operated by radical prostatectomy (RP) have been proposed to expand active surveillance (AS) inclusion criteria. We aimed to test these models in a set of 1062 low-risk and favorable intermediate-risk prostate cancer (PCa) patients diagnosed by multiparametric magnetic resonance imaging (MRI) and MRI-targeted biopsy. We hypothesized that the inclusion of radiological features into a novel model would improve patient selection. Performance was assessed using discrimination, calibration, and decision curve analysis (DCA). Available models were characterized by poor discrimination (areas under the receiver operating characteristic curve [AUCs] of 59% and 60%), underestimation of predicted risk on calibration plots, and a small amount of net benefit against a probability threshold of 40–50% at the DCA. The development of a novel model slightly improved discrimination (AUC of 63% vs 59%, p = 0.001, and 63% vs 60%, p = 0.07) and net benefit against threshold probabilities of ≥30%. This first multicenter study demonstrated the poor performance of models predicting adverse pathology and that implementation of MRI and MRI-targeted biopsy in this setting was not associated with a clear improvement in patient selection. Patients harboring low-risk or favorable intermediate-risk PCa and candidates for RP cannot be referred accurately to an AS program without a non-negligible risk of misclassification. Patient summaryWe tested prediction models that could expand the selection of prostate cancer patients for active surveillance. Models were inaccurate and associated with a high risk of misclassification despite the implementation of multiparametric magnetic resonance imaging and targeted biopsies.

Approach to systematically examine the usefulness of quality measures in practice: Minnesota’s nursing home quality indicators and scoring approach

BackgroundHealthcare quality measurement systems, which use aggregated patient-level quality measures to assess organisational performance, have been introduced widely. Yet, their usefulness in practice has received scant attention. Using Minnesota nursing...

POSITIVE END-EXPIRATORY PRESSURE (PEEP) INDEX: A NOVEL, BETTER DISCRIMINATION MARKER THAN PAO2/FIO2 FOR RESCUE MEASURES IN COVID-RELATED ARDS

POSITIVE END-EXPIRATORY PRESSURE (PEEP) INDEX: A NOVEL, BETTER DISCRIMINATION MARKER THAN PAO2/FIO2 FOR RESCUE MEASURES IN COVID-RELATED ARDS

TNF and HNRNPL Related Immunoregulatory Long non-coding RNA (THRIL) and long intergenic noncoding RNA-p21 (lincRNA-p21) as potential useful biomarkers for the diagnosis of tuberculosis

Early diagnosis is crucial in controlling tuberculosis globally and in developing countries with the emergence of drug-resistant Mycobacterium tuberculosis strains. Long non-coding RNAs (lncRNAs) are promising tuberculosis diagnostic biomarkers. Two lncRNA diagnostic markers, lncRNA THRIL and lincRNA-p21, were studied as tuberculosis diagnostic biomarkers. This cross-sectional study was conducted at the Center of Respiratory Diseases of LAQUINTINIE hospital and the National Veterinary Laboratory of Douala from December 2020 to August 2021. The ability of lncRNAs to distinguish between 19 healthy controls, 15 latent tuberculosis, and 21 active tuberculosis was estimated using quantitative polymerase chain reaction and Receiver Operating Characteristic curve analysis. Our analysis showed that lncRNA THRIL and lincRNA-p21 were significantly upregulated (P <0.05) in active and latent tuberculosis compared with healthy controls. LincRNA-p21 expression was significantly increased (P <0.05) in active tuberculosis compared with latent tuberculosis, whereas lncRNA THRIL was not significantly affected (P ≥0.05). Both lncRNA THRIL and lincRNA-p21 showed excellent performance in classifying latent tuberculosis and healthy controls (AUC = 92.86%). Furthermore, lncRNA THRIL was good at discriminating active tuberculosis from healthy controls (AUC = 89.79%), while lincRNA-p21 showed excellent discriminating performance (AUC = 100%). LncRNA THRIL was identified as a poor discriminator of latent tuberculosis from active tuberculosis (AUC = 64.28%), while lincRNA-p21 showed excellent diagnostic performance in this distinction (AUC = 92.86%). Our cross-sectional study suggests that lncRNA THRIL and lincRNA-p21 are promising tuberculosis diagnostic biomarkers that can differentiate between latent and active infection.

Association between cognition and color discrimination among Lebanese patients with schizophrenia

BackgroundPatients with schizophrenia (SCZ) exhibit poorer color discrimination than normal individuals. Although retinal abnormalities, as well as cortical and subcortical alterations, found in patients with SCZ have been suggested to cause this poor color discrimination, the impact of cognitive impairment remains to be determined. Dopamine (DA) and glutamate (Glu), known to be disrupted in SCZ, are also suggested to play a role in color discrimination. Our objective was to investigate the contribution of cognitive impairment to color discrimination deficits in SCZ and to examine if these deficits are correlated to SCZ symptoms.MethodsThis study includes 127 patients with SCZ between July and September 2021. The participants completed several questionnaires, specifically the Positive and Negative Syndrome Scale (PANSS), the Montreal Cognitive Assessment (MoCA) test, and the Farnsworth D-15 test, to assess the extent of SCZ symptoms, cognition, and color discrimination respectively.ResultsHigher cognition (Beta = − 0.279) was significantly associated with a lower total error score (TES). Moreover, a higher positive PANSS score (Beta = 0.217) was significantly associated with a higher TES. A multinomial regression analysis taking the type of color blindness as the dependent variable showed that female sex (ORa = 5.46) was significantly associated with a certain type of color blindness.ConclusionColor discrimination deficits in patients with SCZ may be due to the effect of cognitive impairment and/or SCZ itself.

A Color-Texture-Based Deep Neural Network Technique to Detect Face Spoofing Attacks

Abstract Given the face spoofing attack, adequate protection of human identity through face has become a significant challenge globally. Face spoofing is an act of presenting a recaptured frame before the verification device to gain illegal access on behalf of a legitimate person with or without their concern. Several methods have been proposed to detect face spoofing attacks over the last decade. However, these methods only consider the luminance information, reflecting poor discrimination of spoofed face from the genuine face. This article proposes a practical approach combining Local Binary Patterns (LBP) and convolutional neural network-based transfer learning models to extract low-level and high-level features. This paper analyzes three color spaces (i.e., RGB, HSV, and YCrCb) to understand the impact of the color distribution on real and spoofed faces for the NUAA benchmark dataset. In-depth analysis of experimental results and comparison with other existing approaches show the superiority and effectiveness of our proposed models.

Is the efficacy of cone beam computed tomography in the diagnosis of tooth ankylosis influenced by dose reduction protocols?

To evaluate the diagnostic efficacy of cone beam computed tomography (CBCT) examinations acquired with varying dose protocols in the diagnosis of simulated tooth ankylosis (TA). Tooth ankylosis was simulated in 15 of 30 teeth, and CBCT examinations were acquired at 3 mA levels (8, 6.3, and 5) and 3 voxel sizes (0.08, 0.125, and 0.2 mm). Four radiologists independently assessed the presence of TA using a 5-point scale. Sensitivity, specificity, and area under the curve (AUC) obtained through receiver operating characteristic analysis were compared among mA levels and voxel sizes using two-way analysis of variance (α=0.05). Intra- and interexaminer reliability levels were assessed with the weighted kappa examination. Sensitivity was low (0.32-0.49), and specificity was reasonably high (0.71-0.83). Mean values of AUC were low, ranging between 0.54 and 0.67, which reveals poor overall discrimination between health and disease. The detection of TA was not significantly influenced by mA level or voxel size (P > .05). Intra- and interexaminer agreements ranged from slight to moderate (0.160-0.535) and from fair to substantial (0.236-0.697), respectively. Despite the limited efficacy of CBCT for TA detection, when indicated as a complementary examination, mA reduction should be considered for dose optimization purposes.

Quality of multiple-choice questions in medical internship qualification examination determined by item response theory at Debre Tabor University, Ethiopia

BackgroundAssessment of cognitive competence is a major element of the internship qualification exam in undergraduate medical education in Ethiopia. Assessing the quality of exam items can help to improve the validity of assessments and assure stakeholders about the accuracy of the go/no decision to the internship. However, we know little about the quality of exam items utilized to ascertain fitness to join the medical internship. Therefore, this study aimed to analyze the quality of multiple-choice questions (MCQs) of the qualification exam administered to final-year medical students at Debre Tabor University (DTU), Ethiopia.MethodsA psychometric study was conducted to assess the qualities of 120 randomly selected MCQs and 407 distractors. Item characteristics were estimated using the item response theory (IRT) model. T-test, one-way ANOVA, and chi-square tests were run to analyze the univariate association between factors. Pearson’s correlation test was done to determine the predictive validity of the qualification examination.ResultOverall, 16, 51, and 33% of the items had high, moderate, and low distractor efficiency, respectively. About two-thirds (65.8%) of the items had two or more functioning distractors and 42.5% exhibited a desirable difficulty index. However, 77.8% of items administered in the qualification examination had a negative or poor discrimination index. Four and five option items didn’t show significant differences in psychometric qualities. The qualification exam showed a positive predictive value of success in the national licensing examination (Pearson’s correlation coefficient = 0.5).ConclusionsThe psychometric properties of the medical qualification exam were inadequate for making valid decisions. Five option MCQs were not better than four options in terms of psychometric qualities. The qualification examination had a positive predictive validity of future performance. High-stakes examination items must be properly created and reviewed before being administered.

A-220 Hooper Visual Organization Test: Item Analyses and Scale Performance in a Nonclinical Adult Sample

Abstract Objective: The 30-item Hooper Visual Organization Test (HVOT) measures visuospatial synthesis. While satisfactory reliability estimates have been reported, some items show poor discrimination and items do not appear to be ordered by increasing difficulty level. Little psychometric evidence, however, has been gathered with healthy nonclinical adults in North America. Examination of item performance in healthy adult samples aids in understanding what might be clinically noteworthy in neurological or psychiatric patients. Thus, we examined HVOT total score performance, item difficulty and discrimination, internal consistency reliability, and correlations of HVOT total scores with demographic and neuropsychological variables in a healthy adult sample in Canada. Method: Participant were 71 cognitively intact, non-depressed men and women ages 21-82 years (M=51.5, SD=18.3), with 9-21 years of education (M=15.0, SD=2.79), recruited from the general community. Participants were tested individually on various neuropsychological tests (HVOT, MMSE, Block Design, ROCF, Trails, Five Point Test, FAS/Animals, Vocabulary). Results: Items ranged in difficulty (0.16-1.00) but tended to be easy (M=0.82). Item discrimination ranged from -.17 to.47. Cronbach’s alpha was 0.62. Total scores ranged from 17-29.5 out of 30 (M=25.4) and correlated significantly with age, MMSE, Block Design, ROCF, Trails, and Animals (r=|.29|-|.44|). Conclusions: Our results suggest that some HVOT items are too easy and/or show poor discrimination. Items are not well-ordered by difficulty. We discuss some notable divergences from proposed re-orderings from clinical samples. Demographic results fit with the literature; correlations with other measures extend previous findings. Our results serve as a guide for revisions and a useful baseline for clinical comparison.

Factors affecting talker discrimination ability in adult cochlear implant users

IntroductionReal-world speech communication involves interacting with many talkers with diverse voices and accents. Many adults with cochlear implants (CIs) demonstrate poor talker discrimination, which may contribute to real-world communication difficulties. However, the factors contributing to talker discrimination ability, and how discrimination ability relates to speech recognition outcomes in adult CI users are still unknown. The current study investigated talker discrimination ability in adult CI users, and the contributions of age, auditory sensitivity, and neurocognitive skills. In addition, the relation between talker discrimination ability and multiple-talker sentence recognition was explored. MethodsFourteen post-lingually deaf adult CI users (3 female, 11 male) with ≥1 year of CI use completed a talker discrimination task. Participants listened to two monosyllabic English words, produced by the same talker or by two different talkers, and indicated if the words were produced by the same or different talkers. Nine female and nine male native English talkers were paired, resulting in same- and different-talker pairs as well as same-gender and mixed-gender pairs. Participants also completed measures of spectro-temporal processing, neurocognitive skills, and multiple-talker sentence recognition. ResultsCI users showed poor same-gender talker discrimination, but relatively good mixed-gender talker discrimination. Older age and weaker neurocognitive skills, in particular inhibitory control, were associated with less accurate mixed-gender talker discrimination. Same-gender discrimination was significantly related to multiple-talker sentence recognition accuracy. ConclusionAdult CI users demonstrate overall poor talker discrimination ability. Individual differences in mixed-gender discrimination ability were related to age and neurocognitive skills, suggesting that these factors contribute to the ability to make use of available, degraded talker characteristics. Same-gender talker discrimination was associated with multiple-talker sentence recognition, suggesting that access to subtle talker-specific cues may be important for speech recognition in challenging listening conditions.

Muscle B mode ultrasound and shear-wave elastography in idiopathic inflammatory myopathies (SWIM): criterion validation against MRI and muscle biopsy findings in an incident patient cohort

BackgroundB mode ultrasound (US) and shear wave elastography (SWE) are easily accessible imaging tools for idiopathic inflammatory myopathies (IIM) but require further validation against standard diagnostic procedures such as MRI and muscle biopsy.MethodsIn this prospective cross-sectional study we compared US findings to MRI and muscle biopsy findings in a group of 18 patients (11 F, 7 M) with active IIM (dermatomyositis 6, necrotising autoimmune myopathy 7, inclusion body myositis 4, overlap myositis 1) who had one or both procedures on the same muscle. US domains (echogenicity, fascial thickness, muscle bulk, shear wave speed and power doppler) in the deltoid and vastus lateralis were compared to MRI domains (muscle oedema, fatty infiltration/atrophy) and muscle biopsy findings (lymphocytic inflammation, myonecrosis, atrophy and fibro-fatty infiltration). A composite index score (1–4) was also used as an arbitrary indicator of overall muscle pathology in biopsies.ResultsIncreased echogenicity correlated with the presence of fatty infiltration/atrophy on MRI (p = 0.047) in the vastus lateralis, and showed a non-significant association with muscle inflammation, myonecrosis, fibrosis and fatty infiltration/atrophy (p > 0.333) Severe echogenicity also had a non-significant association with higher composite biopsy index score in the vastus lateralis (p = 0.380). SWS and US measures of fascial thickness and muscle bulk showed poor discrimination in differentiating between pathologies on MRI or muscle biopsy. Power Doppler measures of vascularity correlated poorly with the presence of oedema on MRI, or with inflammation or fatty infiltration on biopsy. Overall, US was sensitive in detecting the presence of muscle pathology shown on MRI (67–100%) but showed poorer specificity (13–100%). Increased echogenicity showed good sensitivity when detecting muscle pathology (100%) but lacked specificity in differentiating muscle pathologies (0%). Most study participants rated US as the preferred imaging modality.ConclusionsOur findings show that US, in particular muscle echogenicity, has a high sensitivity, but low specificity, for detecting muscle pathology in IIM. Traditional visual grading scores are not IIM-specific and require further modification and validation. Future studies should continue to focus on developing a feasible scoring system, which is reliable and allows translation to clinical practice.

A New Many-Objective Evolutionary Algorithm Based on Generalized Pareto Dominance.

In the past several years, it has become apparent that the effectiveness of Pareto-dominance-based multiobjective evolutionary algorithms deteriorates progressively as the number of objectives in the problem, given by M , grows. This is mainly due to the poor discriminability of Pareto optimality in many-objective spaces (typically M ≥ 4 ). As a consequence, research efforts have been driven in the general direction of developing solution ranking methods that do not rely on Pareto dominance (e.g., decomposition-based techniques), which can provide sufficient selection pressure. However, it is still a nontrivial issue for many existing non-Pareto-dominance-based evolutionary algorithms to deal with unknown irregular Pareto front shapes. In this article, a new many-objective evolutionary algorithm based on the generalization of Pareto optimality (GPO) is proposed, which is simple, yet effective, in addressing many-objective optimization problems. The proposed algorithm used an "( M-1 ) + 1" framework of GPO dominance, ( M-1 )-GPD for short, to rank solutions in the environmental selection step, in order to promote convergence and diversity simultaneously. To be specific, we apply M symmetrical cases of ( M-1 )-GPD, where each enhances the selection pressure of M-1 objectives by expanding the dominance area of solutions, while remaining unchanged for the one objective left out of that process. Experiments demonstrate that the proposed algorithm is very competitive with the state-of-the-art methods to which it is compared, on a variety of scalable benchmark problems. Moreover, experiments on three real-world problems have verified that the proposed algorithm can outperform the others on each of these problems.

Probing deaf oral linguistic competence with minimal morphosyntactic pairs

The COnVERSA test (Test di Comprensione delle Opposizioni morfo-sintattiche VERbali attraverso la ScritturA) is a tool for grammatical competence assessment in Italian. It uses informal grammaticality (binary) judgments on written linguistic minimal pairs and it has been developed for testing deaf children, for which no suitable comprehension test was available. Here we discuss the results obtained both from a normally developing hearing children group (NC, 6-10 y.o.) and a deaf children population (DF, 7-16 y.o.). We will focus on children’s minimal morphosyntactic pairs sensitivity in three domains: (i) argument structure, (ii) question formation, and (iii) pronominal usage. We will then compare these results with those obtained on agreement-based dependencies (Chesi et al., 2023). Our findings support a complexity metric, based on two structural dimensions (functional height and intervention), which is used to guide a dynamic administration modality, in the end significantly reducing the assessment time without decreasing the discriminative power of the test. Our data confirm the poor discriminative ability of DF (especially those without a cochlear implant) for configurations targeting the highest functional domains and involving intervention.

Novel Radiomics-Clinical Model for the Noninvasive Prediction of New Fractures After Vertebral Augmentation

To investigate the noninvasive prediction model for new fractures after percutaneous vertebral augmentation (PVA) based on radiomics signature and clinical parameters. Data from patients who were diagnosed with osteoporotic vertebral compression fracture (OVCF) and treated with PVA in our hospital between May 2014 and April 2019 were retrospectively analyzed. Radiomics features were extracted from T1-weighted magnetic resonance imaging (MRI) of the T11-L5 segments taken before PVA. Different radiomics models was developed by using linear discriminant analysis (LDA), multilayer perceptron (MLP), and stochastic gradient descent (SGD) classifiers. A nomogram was constructed by integrating clinical parameters and Radscore that calculated by the best radiomics model. The model performance was quantified in terms of discrimination, calibration and clinical usefulness. Four clinical parameters and 16 selected radiomics features were used for model development. The clinical model showed poor discrimination capability with area under the curves (AUCs) yielding of 0.522 in the training dataset and 0.517 in the validation dataset. The LDA, MLP and SGD classifier-based radiomics model had achieved AUCs of 0.793, 0.810, and 0.797 in the training dataset, and 0.719, 0.704, and 0.725 in the validation dataset, respectively. The nomogram showed the best performance with AUCs achieving 0.810 and 0.754 in the training and validation datasets, respectively. The decision curve analysis demonstrated the net benefit of the nomogram was higher than that of other models. Our findings indicate that combining clinical features with radiomics features from pre-augmentation T1-weighted MRI can be used to develop a nomogram that can predict new fractures in patients after PVA.

Transcriptomic profiling identifies host-derived biomarker panels for assessing cerebral malaria

Cerebral malaria (CM) is a rare but fatal form of severe malaria sequelae. Host-derived biomarkers have the potential to assist in the early assessment, diagnosis, and prognosis of CM. However, previously reported host-derived CM biomarkers have limited clinical utility despite their ability to discriminate between CM and non-CM conditions, either due to their specificity to other non-malaria conditions or consist of many biomarkers for test implementation. Here, we describe the identification of host-derived transcriptomic signatures that address these limitations using whole-blood transcriptomic profiling, weighted gene co-expression analysis, and random forest classification models. We report three panels; PADI4 + ANGPT2, SLC2A3 + ANGPT2, and SLC2A3 + CA4, that distinguished CM from other malaria manifestations with high accuracy and are poor discriminators of non-malaria conditions that share similar blood-based signatures with severe malaria. Also, these transcripts showed elevated expression in the blood during CM. Our results provide preliminary evidence to support the potential of these biomarker panels in assisting early assessment, diagnosis, and prognosis of CM. Further validations are needed in large and diverse malaria-infected cohorts to establish the clinical utility of these biomarker panels.

Utility of the IMPACT score for predicting heart transplant mortality. Analysis on a contemporary cohort of the Spanish Heart Transplant Registry.

The Index for Mortality Prediction After Cardiac Transplantation (IMPACT) score was derived and validated as a predictor of mortality after heart transplantation (HT). The primary objective of this work is to externally validate the IMPACT score in a contemporary Spanish cohort. Spanish Heart Transplant Registry data were used to identify adult (>16 years) HT patients between January 2000 and December 2015. Retransplantation, multiorgan transplantation and patients in whom at least one of the variables required to calculate the IMPACT score was missing were excluded from the analysis (N=2810). Median value of the IMPACT score was five points (IQR: 3, 8). Overall, 1-year survival rate was 79.1%. Kaplan-Meier 1-year survival rates by IMPACT score categories (0-2, 3-5, 6-9, 10-14, ≥15) were 84.4%, 81.5%, 79.3%, 77.3%, and 58.5%, respectively (Log-Rank test: p<.001). Performance analysis showed a good calibration (Hosmer-Lemeshow chi-square for 1 year was 7.56; p=.47) and poor discrimination ability (AUC-ROC .59) of the IMPACT score as a predictive model. In a contemporary Spanish cohort, the IMPACT score failed to accurately predict the risk of death after HT.

Cochlear Implantation in Advanced Otosclerosis: Rising to the Challenges of Surgery and Decision making

Introduction: Cochlear implantation as well as stapedotomy followed by use of hearing aid are acceptable modes of surgically rehabilitating patients with far advanced otosclerosis. Surgical challenges of CI include difficulties associated with electrode insertion and facial nerve stimulation. Improvement in speech discrimination scores and overall satisfaction with stapedotomy and hearing aid use are reportedly poor in many patients, yet being a low cost procedure it may be used as initial management in a subset of patients.&#x0D; Case Report: 46 year old patient with diffuse confluent retrofenestral otoscerosis underwent cochlear implantation. He was mapped using behavioral thresholds as despite intracochlear electrode position no neural response was recordable per-operatively as well as in the postoperative period. Perimodiolar electrodes and sodium flouride therapy were used to overcome problems of FNS. 18 months post CI the patient has good audiologic outcomes (CAP 7) without any FNS.&#x0D; Conclusion: High resolution computed tomography, air bone gap and speech discrimination scores are important in formulating treatment plan in patients with far advanced otosclerosis. Early cochlear implantation can be considered in patients with poor speech discrimination scores and extensive cochlear lesions. Facial nerve stimulation can be prevented by adequate pre-operative planning.&#x0D; Bangladesh J Otorhinolaryngol 2022; 28(1): 112-117

Development and Internal Validation of a New Prognostic Model Powered to Predict 28-Day All-Cause Mortality in ICU COVID-19 Patients-The COVID-SOFA Score.

Background: The sequential organ failure assessment (SOFA) score has poor discriminative ability for death in severely or critically ill patients with Coronavirus disease 2019 (COVID-19) requiring intensive care unit (ICU) admission. Our aim was to create a new score powered to predict 28-day mortality. Methods: Retrospective, observational, bicentric cohort study including 425 patients with COVID-19 pneumonia, acute respiratory failure and SOFA score ≥ 2 requiring ICU admission for ≥72 h. Factors with independent predictive value for 28-day mortality were identified after stepwise Cox proportional hazards (PH) regression. Based on the regression coefficients, an equation was computed representing the COVID-SOFA score. Discriminative ability was tested using receiver operating characteristic (ROC) analysis, concordance statistics and precision-recall curves. This score was internally validated. Results: Median (Q1–Q3) age for the whole sample was 64 [55–72], with 290 (68.2%) of patients being male. The 28-day mortality was 54.58%. After stepwise Cox PH regression, age, neutrophil-to-lymphocyte ratio (NLR) and SOFA score remained in the final model. The following equation was computed: COVID-SOFA score = 10 × [0.037 × Age + 0.347 × ln(NLR) + 0.16 × SOFA]. Harrell’s C-index for the COVID-SOFA score was higher than the SOFA score alone for 28-day mortality (0.697 [95% CI; 0.662–0.731] versus 0.639 [95% CI: 0.605–0.672]). Subsequently, the prediction error rate was improved up to 16.06%. Area under the ROC (AUROC) was significantly higher for the COVID-SOFA score compared with the SOFA score for 28-day mortality: 0.796 [95% CI: 0.755–0.833] versus 0.699 [95% CI: 0.653–0.742, p < 0.001]. Better predictive value was observed with repeated measurement at 48 h after ICU admission. Conclusions: The COVID-SOFA score is better than the SOFA score alone for 28-day mortality prediction. Improvement in predictive value seen with measurements at 48 h after ICU admission suggests that the COVID-SOFA score can be used in a repetitive manner. External validation is required to support these results.

Development of a mathematical formula and online tool to calculate the potential maximum flap width to allow for primary anterolateral thigh donor-site closure in Caucasians.

Primary anterolateral thigh (ALT) flap donor-site closure is crucial to achieve patient satisfaction, avoid burdensome secondary surgeries, and avert poor outcomes. Only vague maximum flap width recommendations have been suggested, which fall short of acknowledging individual patient habitus and thigh morphology. Therefore, we aimed at identifying a user-friendly preoperative calculation of maximum flap width for primary closure. A total of 429 ALT free flaps performed between 2009 and 2020 were analyzed. A total of 350 donor-sites were closed primarily (82%) and 79 (18%) were split-thickness skin-grafted (STSG). Patient demographics including sex, age, and BMI, operative details, and flap characteristics were compared to assess their impact on the outcome variable. Receiver operating characteristic (ROC) curves were plotted for all significant predictors discriminating between closure and STSG. Areas under the curve (AUCs) were calculated for each parameter combination and optimal cutoffs were determined using Youden's Index. Sex, age, BMI, and flap width alone were poor discriminators. Dividing flap width by BMI and logarithmized BMI yielded AUCs of 0.91 and 0.94, respectively. Including patient sex yielded the best fitting regression model (χ2 = 251.939, p< .0001) increasing the AUC to 0.96 (95% CI: 0.93-0.98, p< .0001). The optimal cutoff value discriminated between primary closure and STSG with 90% sensitivity and 89% specificity. An online calculator of patient-individual maximum ALT width was then programmed. Sex and BMI are reliable predictors of successful primary ALT donor-site closure in Caucasians. We devised a novel formula for calculating patient-individual maximum ALT widths preoperatively, predicting failure of primary closure with 90% sensitivity in our cohort, available at: https://kitteltaschenbuch.com/altwidth/calculate.htm.