A clinical prediction model for estimating the risk of developing uveitis in patients with juvenile idiopathic arthritis

OBJECTIVE

To build a prediction model for uveitis in children with JIA for use in current clinical practice.

METHODS

Data from the international observational Pharmachild registry were used. Adjusted risk factors as well as predictors for JIA-associated uveitis (JIA-U) were determined using multivariable logistic regression models. The prediction model was selected based on the Akaike information criterion. Bootstrap resampling was used to adjust the final prediction model for optimism.

RESULTS

JIA-U occurred in 1102 of 5529 JIA patients (19.9%). The majority of patients that developed JIA-U were female (74.1%), ANA positive (66.0%) and had oligoarthritis (59.9%). JIA-U was rarely seen in patients with systemic arthritis (0.5%) and RF positive polyarthritis (0.2%). Independent risk factors for JIA-U were ANA positivity [odds ratio (OR): 1.88 (95% CI: 1.54, 2.30)] and HLA-B27 positivity [OR: 1.48 (95% CI: 1.12, 1.95)] while older age at JIA onset was an independent protective factor [OR: 0.84 (9%% CI: 0.81, 0.87)]. On multivariable analysis, the combination of age at JIA onset [OR: 0.84 (95% CI: 0.82, 0.86)], JIA category and ANA positivity [OR: 2.02 (95% CI: 1.73, 2.36)] had the highest discriminative power among the prediction models considered (optimism-adjusted area under the receiver operating characteristic curve = 0.75).

CONCLUSION

We developed an easy to read model for individual patients with JIA to inform patients/parents on the probability of developing uveitis.

Risk of Traumatic Brain Injuries in Infants Younger than 3 Months With Minor Blunt Head Trauma

STUDY OBJECTIVE

Infants with head trauma often have subtle findings suggestive of traumatic brain injury. Prediction rules for traumatic brain injury among children with minor head trauma have not been specifically evaluated in infants younger than 3 months old. We aimed to determine the risk of clinically important traumatic brain injuries, traumatic brain injuries on computed tomography (CT) images, and skull fractures in infants younger than 3 months of age who did and did not meet the age-specific Pediatric Emergency Care Applied Research Network (PECARN) low-risk criteria for children with minor blunt head trauma.

METHODS

We conducted a secondary analysis of infants <3 months old in the public use data set from PECARN's prospective observational study of children with minor blunt head trauma. Main outcomes included (1) clinically important traumatic brain injury, (2) traumatic brain injury on CT, and (3) skull fracture on CT.

RESULTS

Of 10,904 patients <2 years old, 1,081 (9.9%) with complete data were <3 months old; most (750/1081, 69.6%) sustained falls, and 633/1081 (58.6%) underwent CT scans. Of the 514/1081 (47.5%) infants who met the PECARN low-risk criteria, 1/514 (0.2%, 95% confidence interval [CI] 0.005% to 1.1%), 10/197 (5.1%, 2.5% to 9.1%), and 9/197 (4.6%, 2.1% to 8.5%) had clinically important traumatic brain injuries, traumatic brain injuries on CT, and skull fractures, respectively. Of 567 infants who did not meet the low-risk PECARN criteria, 24/567 (4.2%, 95% CI 2.7% to 6.2%), 94/436 (21.3%, 95% CI 17.6% to 25.5%), and 122/436 (28.0%, 95% CI 23.8% to 32.5%) had clinically important traumatic brain injuries, traumatic brain injuries, and skull fractures, respectively.

CONCLUSION

The PECARN traumatic brain injury low-risk criteria accurately identified infants <3 months old at low risk of clinically important traumatic brain injuries. However, infants at low risk for clinically important traumatic brain injuries remained at risk for traumatic brain injuries on CT, suggesting the need for a cautious approach in these infants.

Machine Learning-based Prediction Model for Treatment of Acromegaly With First-generation Somatostatin Receptor Ligands

CONTEXT

Artificial intelligence (AI), in particular machine learning (ML), may be used to deeply analyze biomarkers of response to first-generation somatostatin receptor ligands (fg-SRLs) in the treatment of acromegaly.

OBJECTIVE

To develop a prediction model of therapeutic response of acromegaly to fg-SRL.

METHODS

Patients with acromegaly not cured by primary surgical treatment and who had adjuvant therapy with fg-SRL for at least 6 months after surgery were included. Patients were considered controlled if they presented growth hormone (GH) <1.0 ng/mL and normal age-adjusted insulin-like growth factor (IGF)-I levels. Six AI models were evaluated: logistic regression, k-nearest neighbor classifier, support vector machine, gradient-boosted classifier, random forest, and multilayer perceptron. The features included in the analysis were age at diagnosis, sex, GH, and IGF-I levels at diagnosis and at pretreatment, somatostatin receptor subtype 2 and 5 (SST2 and SST5) protein expression and cytokeratin granulation pattern (GP).

RESULTS

A total of 153 patients were analyzed. Controlled patients were older (P = .002), had lower GH at diagnosis (P = .01), had lower pretreatment GH and IGF-I (P < .001), and more frequently harbored tumors that were densely granulated (P = .014) or highly expressed SST2 (P < .001). The model that performed best was the support vector machine with the features SST2, SST5, GP, sex, age, and pretreatment GH and IGF-I levels. It had an accuracy of 86.3%, positive predictive value of 83.3% and negative predictive value of 87.5%.

CONCLUSION

We developed a ML-based prediction model with high accuracy that has the potential to improve medical management of acromegaly, optimize biochemical control, decrease long-term morbidities and mortality, and reduce health services costs.

The predictors of clinical outcomes in brainstem arteriovenous malformations after stereotactic radiosurgery

The brainstem arteriovenous malformations (BS-AVMs) have a high morbidity and mortality and stereotactic radiosurgery (SRS) has been widely used to treat BS-AVMs. However, no consensus is reached in the explicit predictors of obliteration for BS-AVMs after SRS.To identify the predictors of clinical outcomes for BS-AVMs treated by SRS, we performed a retrospective observational study of BS-AVMs patients treated by SRS at our institution from 2006 to 2016. The primary outcomes were obliteration of nidus and favorable outcomes (AVM nidus obliteration with mRS score ≤2). For getting the outcomes more accurate, we also pooled the results of previous studies as well as our study by meta-analysis.A total of 26 patients diagnosed with BS-AVMs, with mean volume of 2.6 ml, were treated with SRS. Hemorrhage presentation accounted for 69% of these patients. Overall obliteration rate was 42% with mean follow-up of more than five years and two patients (8%) had a post-SRS hemorrhage. Favorable outcomes were observed in 8 patients (31%). Higher margin dose (>15Gy) was associated with higher obliteration (P = .042) and small volume of nidus was associated with favorable outcomes (P = .036). After pooling the results of 7 studies and present study, non-prior embolization (P = .049) and higher margin dose (P = .04) were associated with higher obliteration rate, in addition, the lower Virginia Radiosurgery AVM Scale (VRAS) was associated with favorable outcomes (P = .02) of BS-AVMs after SRS.In the BS-AVMs patients treated by SRS, higher margin dose (19-24Gy) and non-prior embolization were the independent predictors of higher obliteration rate. In addition, smaller volume of nidus and lower VRAS were the potential predictors of long-term favorable outcomes for these patients.

Systemic Inflammation Response Index and Systemic Immune-inflammation Index for Predicting the Prognosis of Patients with Aneurysmal Subarachnoid Hemorrhage

OBJECTIVES

Inflammatory response plays a pivotal role in the progress of aneurysmal subarachnoid hemorrhage (aSAH). As novel inflammatory markers, systemic inflammation response index (SIRI) and systemic immune-inflammation (SII) index could reflect clinical outcomes of patients with various diseases. The aim of this study was to ascertain whether initial SIRI and SII index were associated with prognosis of aSAH patients.

METHODS

A total of 680 patients with aSAH were enrolled. Their prognosis was evaluated with modified Rankin Scale (mRS) at 3 months, and unfavorable clinical outcome was defined as mRS score of 3-6. Receiver operating characteristic (ROC) curve analysis was performed to identify cutoff values of SIRI and SII index for predicting clinical outcomes. Univariate and multivariate regression analyses were performed to explore relationships of SIRI and SII index with prognosis of patients.

RESULTS

Optimal cutoff values of SIRI and SII index to discriminate between favorable and unfavorable clinical outcomes were 3.2 × 10⁹/L and 960 × 10⁹/L, respectively (P < 0.001 and 0.004, respectively). In multivariate analysis, SIRI value ≥ 3.2 × 10⁹/L (odds ratio [OR]: 1.82, 95% CI: 1.46-3.24; P = 0.021) and SII index value ≥ 960 × 10⁹/L (OR: 1.68, 95% CI: 1.24-2.74; P = 0.040) were independent predicting factors for poor prognosis after aSAH.

CONCLUSIONS

SIRI and SII index values are associated with clinical outcomes of patients with aSAH. Elevated SIRI and SII index could be independent predicting factors for a poor prognosis after aSAH.

Profiles and outcomes in patients with COVID-19 admitted to wards of a French oncohematological hospital: A clustering approach

OBJECTIVES

Although some prognostic factors for COVID-19 were consistently identified across the studies, differences were found for other factors that could be due to the characteristics of the study populations and the variables incorporated into the statistical model. We aimed to a priori identify specific patient profiles and then assess their association with the outcomes in COVID-19 patients with respiratory symptoms admitted specifically to hospital wards.

METHODS

We conducted a retrospective single-center study from February 2020 to April 2020. A non-supervised cluster analysis was first used to detect patient profiles based on characteristics at admission of 220 consecutive patients admitted to our institution. Then, we assessed the prognostic value using Cox regression analyses to predict survival.

RESULTS

Three clusters were identified, with 47 patients in cluster 1, 87 in cluster 2, and 86 in cluster 3; the presentation of the patients differed among the clusters. Cluster 1 mostly included sexagenarian patients with active malignancies who were admitted early after the onset of COVID-19. Cluster 2 included the oldest patients, who were generally overweight and had hypertension and renal insufficiency, while cluster 3 included the youngest patients, who had gastrointestinal symptoms and delayed admission. Sixty-day survival rates were 74.3%, 50.6% and 96.5% in clusters 1, 2, and 3, respectively. This was confirmed by the multivariable Cox analyses that showed the prognostic value of these patterns.

CONCLUSION

The cluster approach seems appropriate and pragmatic for the early identification of patient profiles that could help physicians segregate patients according to their prognosis.

Automatic Detection of Thyroid and Adrenal Incidentals Using Radiology Reports and Deep Learning

BACKGROUND

Computed tomography (CT) is commonly performed when evaluating trauma patients with up to 55% showing incidental findings. Current workflows to identify and inform patients are time-consuming and prone to error. Our objective was to automatically identify thyroid and adrenal lesions in radiology reports using deep learning.

MATERIALS AND METHODS

All trauma patients who presented to an accredited Level 1 Trauma Center between January 2008 and January 2019 were included. Radiology reports of CT scans that included either a thyroid or adrenal gland were obtained. Preprocessing included word tokenization, removal of stop words, removal of punctuation, and replacement of misspellings. A word2vec model was trained using 1.4 million radiology reports. Both training and testing reports were selected at random, manually reviewed, and were considered the gold standard. True positive cases were defined as any lesions in the thyroid or adrenal gland, respectively. Training data was used to create models that would identify reports that contained either thyroid or adrenal lesions. Our primary outcomes were sensitivity and specificity of the models using predetermined thresholds on a separate testing dataset.

RESULTS

A total of 51,771 reports were identified on 35,859 trauma patients. A total of 1,789 reports were annotated for training and 500 for testing. The thyroid model predictions resulted in a 90.0% sensitivity and 95.3% specificity. The adrenal model predictions resulted in a 92.3% sensitivity and a 91.1% specificity. A total of 240 reports were confirmed to have thyroid incidentals (mean age 69.1 yrs ± 18.9, 35% M) and 214 reports with adrenal incidentals (mean age 68.7 yrs ± 16.9, 50.5% M).

CONCLUSIONS

Both the thyroid and adrenal models have excellent performance with sensitivities and specificities in the 90s. Our deep learning model has the potential to reduce administrative costs and improve the process of informing patients.

Predicting in-hospital mortality in ICU patients with sepsis using gradient boosting decision tree

Sepsis is a leading cause of mortality in the intensive care unit. Early prediction of sepsis can reduce the overall mortality rate and cost of sepsis treatment. Some studies have predicted mortality and development of sepsis using machine learning models. However, there is a gap between the creation of different machine learning algorithms and their implementation in clinical practice.This study utilized data from the Medical Information Mart for Intensive Care III. We established and compared the gradient boosting decision tree (GBDT), logistic regression (LR), k-nearest neighbor (KNN), random forest (RF), and support vector machine (SVM).A total of 3937 sepsis patients were included, with 34.3% mortality in the Medical Information Mart for Intensive Care III group. In our comparison of 5 machine learning models (GBDT, LR, KNN, RF, and SVM), the GBDT model showed the best performance with the highest area under the receiver operating characteristic curve (0.992), recall (94.8%), accuracy (95.4%), and F1 score (0.933). The RF, SVM, and KNN models showed better performance (area under the receiver operating characteristic curve: 0.980, 0.898, and 0.877, respectively) than the LR (0.876).The GBDT model showed better performance than other machine learning models (LR, KNN, RF, and SVM) in predicting the mortality of patients with sepsis in the intensive care unit. This could be used to develop a clinical decision support system in the future.

Does the Rule of Thirds Adequately Detect Deficient and Excessive Acetabular Coverage?

BACKGROUND

Assessment of AP acetabular coverage is crucial for choosing the right surgery indication and for obtaining a good outcome after hip-preserving surgery. The quantification of anterior and posterior coverage is challenging and requires either other conventional projections, CT, MRI, or special measurement software, which is cumbersome, not widely available and implies additional radiation. We introduce the "rule of thirds" as a promising alternative to provide a more applicable and easy method to detect an excessive or deficient AP coverage. This method attributes the intersection point of the anterior (posterior) wall to thirds of the femoral head radius (diameter), the medial third suggesting deficient and the lateral third excessive coverage.

QUESTION/PURPOSE

What is the validity (area under the curve [AUC], sensitivity, specificity, positive/negative likelihood ratios [LR(+)/LR(-)], positive/negative predictive values [PPV, NPV]) for the rule of thirds to detect (1) excessive and (2) deficient anterior and posterior coverages compared with previously established radiographic values of under-/overcoverage using Hip2Norm as the gold standard?

METHODS

We retrospectively evaluated all consecutive patients between 2003 and 2015 from our institutional database who were referred to our hospital for hip pain and were potentially eligible for joint-preserving hip surgery. We divided the study group into six specific subgroups based on the respective acetabular pathomorphology to cover the entire range of anterior and posterior femoral coverage (dysplasia, overcoverage, severe overcoverage, excessive acetabular anteversion, acetabular retroversion, total acetabular retroversion). From this patient cohort, 161 hips were randomly selected for analysis. Anterior and posterior coverage was determined with Hip2Norm, a validated computer software program for evaluating acetabular morphology. The anterior and posterior wall indices were measured on standardized AP pelvis radiographs, and the rule of thirds was applied by one observer.

RESULTS

The detection of excessive anterior and posterior acetabular wall using the rule of thirds revealed an AUC of 0.945 and 0.933, respectively. Also the detection of a deficient anterior and posterior acetabular wall by applying the rule of thirds revealed an AUC of 0.962 and 0.876, respectively. For both excessive and deficient anterior and posterior acetabular coverage, we found high specificities and PPVs but low sensitivities and NPVs.

CONCLUSION

We found a high probability for an excessive (deficient) acetabular wall when this intersection point lies in the lateral (medial) third, which would qualify for surgical correction. On the other hand, if this point is not in the lateral (medial) third, an excessive (deficient) acetabular wall cannot be categorically excluded. Thus, the rule of thirds is very specific but not as sensitive as we had expected.

LEVEL OF EVIDENCE

Level II, diagnostic study.

Predicting Delirium Risk Using an Automated Mayo Delirium Prediction Tool: Development and Validation of a Risk-Stratification Model

OBJECTIVE

To develop a delirium risk-prediction tool that is applicable across different clinical patient populations and can predict the risk of delirium at admission to hospital.

METHODS

This retrospective study included 120,764 patients admitted to Mayo Clinic between January 1, 2012, and December 31, 2017, with age 50 and greater. The study group was randomized into a derivation cohort (n=80,000) and a validation cohort (n=40,764). Different risk factors were extracted and analyzed using least absolute shrinkage and selection operator (LASSO) penalized logistic regression.

RESULTS

The area under the receiver operating characteristic curve (AUROC) for Mayo Delirium Prediction (MDP) tool using derivation cohort was 0.85 (95% confidence interval [CI], .846 to .855). Using the regression coefficients obtained from the derivation cohort, predicted probability of delirium was calculated for each patient in the validation cohort. For the validation cohort, AUROC was 0.84 (95% CI, .834 to .847). Patients were classified into 1 of the 3 risk groups, based on their predicted probability of delirium: low (≤5%), moderate (6% to 29%), and high (≥30%). In the derivation cohort, observed incidence of delirium was 1.7%, 12.8%, and 44.8% (low, moderate, and high risk, respectively), which is similar to the incidence rates in the validation cohort of 1.9%, 12.7%, and 46.3%.

CONCLUSION

The Mayo Delirium Prediction tool was developed from a large heterogeneous patient population with good validation results and appears to be a reliable automated tool for delirium risk prediction with hospitalization. Further prospective validation studies are required.

Precision Medicine Approach to Develop and Internally Validate Optimal Exercise and Weight-Loss Treatments for Overweight and Obese Adults With Knee Osteoarthritis: Data From a Single-Center Randomized Trial

OBJECTIVE

To apply a precision medicine approach to determine the optimal treatment regime for participants in an exercise (E), dietary weight loss (D), and D + E trial for knee osteoarthritis that would maximize their expected outcomes.

METHODS

Using data from 343 participants of the Intensive Diet and Exercise for Arthritis (IDEA) trial, we applied 24 machine-learning models to develop individualized treatment rules on 7 outcomes: Short Form 36 physical component score, weight loss, Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain/function/stiffness scores, compressive force, and interleukin-6 level. The optimal model was selected based on jackknife value function estimates that indicate improvement in the outcomes if future participants follow the estimated decision rule compared to the optimal single, fixed treatment model.

RESULTS

Multiple outcome random forest was the optimal model for the WOMAC outcomes. For the other outcomes, list-based models were optimal. For example, the estimated optimal decision rule for weight loss indicated assigning the D + E intervention to participants with baseline weight not exceeding 109.35 kg and waist circumference above 90.25 cm, and assigning D to all other participants except those with a history of a heart attack. If applied to future participants, the optimal rule for weight loss is estimated to increase average weight loss to 11.2 kg at 18 months, contrasted with 9.8 kg if all participants received D + E (P = 0.01).

CONCLUSION

The precision medicine models supported the overall findings from IDEA that the D + E intervention was optimal for most participants, but there was evidence that a subgroup of participants would likely benefit more from diet alone for 2 outcomes.

Development and validation of a clinical prediction rule for development of diabetic foot ulceration: an analysis of data from five cohort studies

INTRODUCTION

The aim of the study was to develop and validate a clinical prediction rule (CPR) for foot ulceration in people with diabetes.

RESEARCH DESIGN AND METHODS

Development of a CPR using individual participant data from four international cohort studies identified by systematic review, with validation in a fifth study. Development cohorts were from primary and secondary care foot clinics in Europe and the USA (n=8255, adults over 18 years old, with diabetes, ulcer free at recruitment). Using data from monofilament testing, presence/absence of pulses, and participant history of previous ulcer and/or amputation, we developed a simple CPR to predict who will develop a foot ulcer within 2 years of initial assessment and validated it in a fifth study (n=3324). The CPR's performance was assessed with C-statistics, calibration slopes, calibration-in-the-large, and a net benefit analysis.

RESULTS

CPR scores of 0, 1, 2, 3, and 4 had a risk of ulcer within 2 years of 2.4% (95% CI 1.5% to 3.9%), 6.0% (95% CI 3.5% to 9.5%), 14.0% (95% CI 8.5% to 21.3%), 29.2% (95% CI 19.2% to 41.0%), and 51.1% (95% CI 37.9% to 64.1%), respectively. In the validation dataset, calibration-in-the-large was -0.374 (95% CI -0.561 to -0.187) and calibration slope 1.139 (95% CI 0.994 to 1.283). The C-statistic was 0.829 (95% CI 0.790 to 0.868). The net benefit analysis suggested that people with a CPR score of 1 or more (risk of ulceration 6.0% or more) should be referred for treatment.

CONCLUSION

The clinical prediction rule is simple, using routinely obtained data, and could help prevent foot ulcers by redirecting care to patients with scores of 1 or above. It has been validated in a community setting, and requires further validation in secondary care settings.

A Novel Metric for Developing Easy-to-Use and Accurate Clinical Prediction Models: The Time-cost Information Criterion

BACKGROUND

Guidelines recommend that clinicians use clinical prediction models to estimate future risk to guide decisions. For example, predicted fracture risk is a major factor in the decision to initiate bisphosphonate medications. However, current methods for developing prediction models often lead to models that are accurate but difficult to use in clinical settings.

OBJECTIVE

The objective of this study was to develop and test whether a new metric that explicitly balances model accuracy with clinical usability leads to accurate, easier-to-use prediction models.

METHODS

We propose a new metric called the Time-cost Information Criterion (TCIC) that will penalize potential predictor variables that take a long time to obtain in clinical settings. To demonstrate how the TCIC can be used to develop models that are easier-to-use in clinical settings, we use data from the 2000 wave of the Health and Retirement Study (n=6311) to develop and compare time to mortality prediction models using a traditional metric (Bayesian Information Criterion or BIC) and the TCIC.

RESULTS

We found that the TCIC models utilized predictors that could be obtained more quickly than BIC models while achieving similar discrimination. For example, the TCIC identified a 7-predictor model with a total time-cost of 44 seconds, while the BIC identified a 7-predictor model with a time-cost of 119 seconds. The Harrell C-statistic of the TCIC and BIC 7-predictor models did not differ (0.7065 vs. 0.7088, P=0.11).

CONCLUSION

Accounting for the time-costs of potential predictor variables through the use of the TCIC led to the development of an easier-to-use mortality prediction model with similar discrimination.

Establishment of a clinical diagnostic model for gouty arthritis based on the serum biochemical profile: A case-control study

The disease progression of gouty arthritis (GA) is relatively clear, with the 4 stages of hyperuricemia (HUA), acute gouty arthritis (AGA), gouty arthritis during the intermittent period (GIP), and chronic gouty arthritis (CGA). This paper attempts to construct a clinical diagnostic model based on blood routine test data, in order to avoid the need for bursa fluid examination and other tedious steps, and at the same time to predict the development direction of GA.Serum samples from 579 subjects were collected within 3 years in this study and were divided into a training set (n = 379) and validation set (n = 200). After a series of multivariate statistical analyses, the serum biochemical profile was obtained, which could effectively distinguish different stages of GA. A clinical diagnosis model based on the biochemical index of the training set was established to maximize the probability of the stage as a diagnosis, and the serum biochemical data from 200 patients were used for validation.The total area under the curve (AUC) of the clinical diagnostic model was 0.9534, and the AUCs of the 5 models were 0.9814 (Control), 0.9288 (HUA), 0.9752 (AGA), 0.9056 (GIP), and 0.9759 (CGA). The kappa coefficient of the clinical diagnostic model was 0.80.This clinical diagnostic model could be applied clinically and in research to improve the accuracy of the identification of the different stages of GA. Meanwhile, the serum biochemical profile revealed by this study could be used to assist the clinical diagnosis and prediction of GA.

The predictive value of serum procalcitonin for non-invasive positive pressure ventilation in the patients with acute exacerbation of chronic obstructive pulmonary disease

This study aimed to estimate the value of serum procalcitonin (PCT) for non-invasive positive pressure ventilation (NIPPV) prediction in the patients with acute exacerbation of chronic obstructive pulmonary disease (AECOPD).A total of 220 AECOPD patients were divided into NIPPV group (n = 121) and control group (routine treatment, n = 99) based on the routine standards and physicians' experience. Logistic regression analysis was performed to identify the independent factors for NIPPV treatment. Additionally, the predictive values of the factors were measured through receiver operation characteristic (ROC) curve.NIPPV group and control group showed significant differences in respiratory rate (RR), PaO2, PaCO2, pH, oxygenation index, erythrocyte sedimentation rate (ESR), neutrophil, CRP (C-reaction protein), and PCT (P < .05 for all). PCT, CRP, PaCO2, RR, and neutrophil were independently correlated with NIPPV treatment in AECOPD. ROC curve showed that PCT had superior predictive value, with AUC of 0.899, the sensitivity of 86%, and the specificity of 91.9%. The cut-off value of serum PCT for NIPPV prediction was 88.50 ng/l.AECOPD patients who require NIPPV treatment frequently have high levels of PCT, CRP, PaCO2, RR and neutrophil. Serum PCT may be employed as an indicator for NIPPV treatment in AECOPD patients.

Early prediction models for extended-spectrum β-lactamase-producing Escherichia coli infection in emergency department: A protocol for systematic review and meta analysis

BACKGROUND

Resistance in gram-negative bacteria has gained great importance in recent decades and one reason is the rapid increase of extended spectrum β-lactamase (ESBL)-producing bacteria as a growing problem worldwide. The increasing proportion of ESBL-producing Enterobacteriaceae (ESBL-E) infections acquired in the emergency community is a new feature of ESBLs epidemiology. Early recognition of patients with extended-spectrum β-lactamase-producing Escherichia coli infection is important in the emergency department. To mitigate the burden on the healthcare system, while also providing the best possible care for patients, early recognition of the infection is needed.

METHODS

For the acquisition of required data of eligible prospective/retrospective cohort study or randomized controlled trials (RCTs), we will search for publications from PubMed, Web of science, EMBASE, Cochrane Library, Google scholar. Two independent reviewers will read the full English text of the articles, screened and selected carefully, removing duplication. Then we evaluate the quality and analyses data by Review Manager (V.5.4). Results data will be pooled and meta-analysis will be conducted if there's 2 eligible studies considered.

RESULTS

This systematic review and meta-analysis will evaluate the value of the early prediction models for Extended-spectrum β-lactamase-producing E coli infection in emergency department.

CONCLUSIONS

This systematic review and meta-analysis will provide clinical evidence for predicting Extended-spectrum β-lactamase-producing E coli infection in emergency department, inform our understanding of the value of the predictive model in predicting Extended-spectrum β-lactamase-producing E coli infection in emergency department in the early stage. The conclusions drawn from this study may be beneficial to patients, clinicians, and health-related policy makers.

STUDY REGISTRATION NUMBER

INPLASY202130049.

Risk stratification scores for hospitalization duration and disease progression in moderate and severe patients with COVID-19

BACKGROUND

During outbreak of Coronavirus Disease 2019 (COVID-19), healthcare providers are facing critical clinical decisions based on the prognosis of patients. Decision support tools of risk stratification are needed to predict outcomes in patients with different clinical types of COVID-19.

METHODS

This retrospective cohort study recruited 2425 patients with moderate or severe COVID-19. A logistic regression model was used to select and estimate the factors independently associated with outcomes. Simplified risk stratification score systems were constructed to predict outcomes in moderate and severe patients with COVID-19, and their performances were evaluated by discrimination and calibration.

RESULTS

We constructed two risk stratification score systems, named as STPCAL (including significant factors in the prediction model: number of clinical symptoms, the maximum body temperature during hospitalization, platelet count, C-reactive protein, albumin and lactate dehydrogenase) and TRPNCLP (including maximum body temperature during hospitalization, history of respiratory diseases, platelet count, neutrophil-to-lymphocyte ratio, creatinine, lactate dehydrogenase, and prothrombin time), to predict hospitalization duration for moderate patients and disease progression for severe patients, respectively. According to STPCAL score, moderate patients were classified into three risk categories for a longer hospital duration: low (Score 0-1, median = 8 days, with less than 20.0% probabilities), intermediate (Score 2-6, median = 13 days, with 30.0-78.9% probabilities), high (Score 7-9, median = 19 days, with more than 86.5% probabilities). Severe patients were stratified into three risk categories for disease progression: low risk (Score 0-5, with less than 12.7% probabilities), intermediate risk (Score 6-11, with 18.6-69.1% probabilities), and high risk (Score 12-16, with more than 77.9% probabilities) by TRPNCLP score. The two risk scores performed well with good discrimination and calibration.

CONCLUSIONS

Two easy-to-use risk stratification score systems were built to predict the outcomes in COVID-19 patients with different clinical types. Identifying high risk patients with longer stay or poor prognosis could assist healthcare providers in triaging patients when allocating limited healthcare during COVID-19 outbreak.

Ethnic Specific body fat percent prediction equation as surrogate marker of obesity in Ethiopian adults

BACKGROUND

Application of advanced body composition measurement methods is not practical in developing countries context due to cost and unavailability of facilities. This study generated ethnic specific body fat percent prediction equation for Ethiopian adults using appropriate data.

METHODS

A cross-sectional study was carried ifrom February to April 2015 among 704 randomly selected adult employees of Jimma University. Ethnic specific Ethiopian body fat percent (BF%) prediction equation was developed using a multivariable linear regression model with measured BF% as dependent variable and age, sex, and body mass index as predictor variables. Agreement between fat percent measured using air displacement plethysmography and body fat percent estimated using Caucasian prediction equations was determined using Bland Altman plot.

RESULTS

Comparison of ADP measured and predicted BF% showed that Caucasian prediction equation underestimated body fat percent among Ethiopian adults by 6.78% (P < 0.0001). This finding is consistent across all age groups and ethnicities in both sexes. Bland Altman plot did not show agreement between ADP and Caucasian prediction equation (mean difference = 6.7825) and some of the points are outside 95% confidence interval. The caucasian prediction equation significantly underestimates body fat percent in Ethiopian adults, which is consistent across all ethnic groups in the sample. The study developed Ethnic specific BF% prediction equations for Ethiopian adults.

CONCLUSION

The Caucasian prediction equation significantly underestimates body fat percent among Ethiopian adults regardless of ethnicity. Ethiopian ethnic-specific prediction equation can be used as a very simple, cheap, and cost-effective alternative for estimating body fat percent among Ethiopian adults for health care provision in the prevention of obesity and related morbidities and for research purposes.

Performance of Emergency Heart Failure Mortality Risk Grade in the Emergency Department

INTRODUCTION

The purpose of this study was to validate and assess the performance of the Emergency Heart Failure Mortality Risk Grade (EHMRG) to predict seven-day mortality in US patients presenting to the emergency department (ED) with acute congestive heart failure (CHF) exacerbation.

METHODS

We performed a retrospective chart review on patients presenting to the ED with acute CHF exacerbation between January 2014-January 2016 across eight EDs in New York. We identified patients using codes from the International Classification of Diseases, 9th and 10 Revisions, or who were diagnosed with CHF in the ED. Inclusion criteria were patients ≥ 18 years of age who presented to the ED for acute CHF. Exclusion criteria included the following: end-stage renal disease related heart failure; < 18 years of age; pregnancy; palliative care; renal failure; and "do not resuscitate" directive. The primary outcome was seven-day mortality. We used mixed-effects logistic regression models to estimate C-statistics and continuous net reclassification index for events and nonevents.

RESULTS

We identified 3,320 ED visits associated with suspected CHF among 2,495 unique patients. Of the 3,320 ED visits, 94.7% patients were admitted to the hospital and 3.4% were discharged. The median age was 78.6 (interquartile range 68.01 - 86.76). There was an overall seven-day mortality of 2%, an inpatient mortality rate of 2.4%, and no mortality among the discharge group. Adding EHMRG to the risk prediction model improved the C-statistic (from 0.748 to 0.772) and led to a higher degree of reclassification for both events and nonevents.

CONCLUSION

The EHMRG can be used as a valuable and effective screening tool in the US while considering disposition decision for patients with acute CHF exacerbation. Emergency medical services transport and metolazone use is much higher in the US population as compared to the Canadian population. We observed minimal to no short-term mortality among discharged CHF patients from the ED.

Identifying COVID-19 cases in outpatient settings

Case identification is an ongoing issue for the COVID-19 epidemic, in particular for outpatient care where physicians must decide which patients to prioritise for further testing. This paper reports tools to classify patients based on symptom profiles based on 236 severe acute respiratory syndrome coronavirus 2 positive cases and 564 controls, accounting for the time course of illness using generalised multivariate logistic regression. Significant symptoms included abdominal pain, cough, diarrhoea, fever, headache, muscle ache, runny nose, sore throat, temperature between 37.5 and 37.9 °C and temperature above 38 °C, but their importance varied by day of illness at assessment. With a high percentile threshold for specificity at 0.95, the baseline model had reasonable sensitivity at 0.67. To further evaluate accuracy of model predictions, leave-one-out cross-validation confirmed high classification accuracy with an area under the receiver operating characteristic curve of 0.92. For the baseline model, sensitivity decreased to 0.56. External validation datasets reported similar result. Our study provides a tool to discern COVID-19 patients from controls using symptoms and day from illness onset with good predictive performance. It could be considered as a framework to complement laboratory testing in order to differentiate COVID-19 from other patients presenting with acute symptoms in outpatient care.

Clinical predictors of delayed neurological sequelae in charcoal-burning carbon monoxide poisoning

BACKGROUND

The main objective of the treatment of acute carbon monoxide (CO) poisoning is to prevent delayed neurological sequelae (DNS). However, today there is still no objective screening tool to identify patients at high risk of developing DNS. The aim of this study was to identify clinical factors that could predict DNS after acute charcoal-burning CO poisoning.

METHODS

This prospective observational study was conducted from September 1, 2019 to August 31, 2020 in a single academic medical center. Patients older than 18 years of age suffering from charcoal-burning CO poisoning were included in the study. After acute recovery, patients were followed up for six weeks to investigate for DNS development. The clinical predictors of DNS were determined using a multivariate logistic regression model.

RESULTS

Of the 217 patients-113 males (52.1%), median age 37.0 (27.5-51.5) years-included, 49 (22.6%) developed DNS. The multivariate logistic regression analysis revealed the independent predictors of DNS as a lower initial Glasgow Coma Scale (GCS) score (adjusted odds ratio (AOR): 0.73, 95% confidence interval (CI): 0.62-0.87), a longer duration of CO exposure (AOR: 2.18, 95% CI: 1.65-2.88), and the presence of acute brain lesions with high signal intensity on diffusion-weighted imaging (AOR: 5.22, 95% CI: 1.50-18.08). The created multivariate regression model predicted DNS development with high accuracy (area under the curve: 0.93, 95% CI: 0.89-0.97).

CONCLUSION

A low initial GCS score, longer exposure to CO and abnormal findings on diffusion-weighted magnetic resonance imaging can assist in the early identification of patients at high risk of DNS development.

A Value Framework for the Assessment of Diagnostic Technologies: A Proposal Based on a Targeted Systematic Review and a Multistakeholder Deliberative Process in Latin America

OBJECTIVES

there are very few value frameworks (VFs) to assess health technologies that are focused on diagnostic tests; they usually do not reflect a multistakeholder process; and they are all developed in high-income countries. Our project performed a targeted systematic review, with the objective of proposing an evidence-based, up-to-date VF informed by a multinational multistakeholder group working in the health technology assessment (HTA) space.

METHODS

(1) A targeted systematic review, with the aim to identify existing VFs and their dimensions; and (2) generation a VF proposal through a mixed-methods, qualitative-quantitative approach.

RESULTS

From 73 citations identified, 20 met our inclusion criteria and served to provide the initial list of dimensions for our VF. An initial list of criteria and subcriteria for a preliminary VF was proposed. After a full-day deliberative face-to-face meeting with 30 relevant stakeholders from seven Latin American countries and the United Kingdom, the final VF was defined, consisting of 15 criteria: five "essential or core," six highly relevant, three moderately relevant, and one of low relevance. Barriers and facilitators of value assessment of diagnostic technologies were also discussed.

CONCLUSIONS

We propose a VF oriented to diagnostic technologies based on a targeted systematic review and a participatory process with key HTA stakeholders. It is the first to be produced in a lower and middle income setting but can also be potentially useful in other contexts aimed to assist decision-making processes with these particularly complex health technologies.

Development and validation of the ISARIC 4C Deterioration model for adults hospitalised with COVID-19: a prospective cohort study

BACKGROUND

Prognostic models to predict the risk of clinical deterioration in acute COVID-19 cases are urgently required to inform clinical management decisions.

METHODS

We developed and validated a multivariable logistic regression model for in-hospital clinical deterioration (defined as any requirement of ventilatory support or critical care, or death) among consecutively hospitalised adults with highly suspected or confirmed COVID-19 who were prospectively recruited to the International Severe Acute Respiratory and Emerging Infections Consortium Coronavirus Clinical Characterisation Consortium (ISARIC4C) study across 260 hospitals in England, Scotland, and Wales. Candidate predictors that were specified a priori were considered for inclusion in the model on the basis of previous prognostic scores and emerging literature describing routinely measured biomarkers associated with COVID-19 prognosis. We used internal-external cross-validation to evaluate discrimination, calibration, and clinical utility across eight National Health Service (NHS) regions in the development cohort. We further validated the final model in held-out data from an additional NHS region (London).

FINDINGS

74 944 participants (recruited between Feb 6 and Aug 26, 2020) were included, of whom 31 924 (43·2%) of 73 948 with available outcomes met the composite clinical deterioration outcome. In internal-external cross-validation in the development cohort of 66 705 participants, the selected model (comprising 11 predictors routinely measured at the point of hospital admission) showed consistent discrimination, calibration, and clinical utility across all eight NHS regions. In held-out data from London (n=8239), the model showed a similarly consistent performance (C-statistic 0·77 [95% CI 0·76 to 0·78]; calibration-in-the-large 0·00 [-0·05 to 0·05]); calibration slope 0·96 [0·91 to 1·01]), and greater net benefit than any other reproducible prognostic model.

INTERPRETATION

The 4C Deterioration model has strong potential for clinical utility and generalisability to predict clinical deterioration and inform decision making among adults hospitalised with COVID-19.

FUNDING

National Institute for Health Research (NIHR), UK Medical Research Council, Wellcome Trust, Department for International Development, Bill & Melinda Gates Foundation, EU Platform for European Preparedness Against (Re-)emerging Epidemics, NIHR Health Protection Research Unit (HPRU) in Emerging and Zoonotic Infections at University of Liverpool, NIHR HPRU in Respiratory Infections at Imperial College London.

Mixed-Methods Study to Predict Upstaging of DCIS to Invasive Disease on Mammography

BACKGROUND. The incidence of ductal carcinoma in situ (DCIS) has steadily increased, as have concerns regarding overtreatment. Active surveillance is a novel treatment strategy that avoids surgical excision, but identifying patients with occult invasive disease who should be excluded from active surveillance is challenging. Radiologists are not typically expected to predict the upstaging of DCIS to invasive disease, though they might be trained to perform this task. OBJECTIVE. The purpose of this study was to determine whether a mixed-methods two-stage observer study can improve radiologists' ability to predict upstaging of DCIS to invasive disease on mammography. METHODS. All cases of DCIS calcifications that underwent stereotactic biopsy between 2010 and 2015 were identified. Two cohorts were randomly generated, each containing 150 cases (120 pure DCIS cases and 30 DCIS cases upstaged to invasive disease at surgery). Nine breast radiologists reviewed the mammograms in the first cohort in a blinded fashion and scored the probability of upstaging to invasive disease. The radiologists then reviewed the cases and results collectively in a focus group to develop consensus criteria that could improve their ability to predict upstaging. The radiologists reviewed the mammograms from the second cohort in a blinded fashion and again scored the probability of upstaging. Statistical analysis compared the performances between rounds 1 and 2. RESULTS. The mean AUC for reader performance in predicting upstaging in round 1 was 0.623 (range, 0.514-0.684). In the focus group, radiologists agreed that upstaging was better predicted when an associated mass, asymmetry, or architectural distortion was present; when densely packed calcifications extended over a larger area; and when the most suspicious features were focused on rather than the most common features. Additionally, radiologists agreed that BI-RADS descriptors do not adequately characterize risk of invasion, and that microinvasive disease and smaller areas of DCIS will have poor prediction estimates. Reader performance significantly improved in round 2 (mean AUC, 0.765; range, 0.617-0.852; p = .045). CONCLUSION. A mixed-methods two-stage observer study identified factors that helped radiologists significantly improve their ability to predict upstaging of DCIS to invasive disease. CLINICAL IMPACT. Breast radiologists can be trained to better predict upstaging of DCIS to invasive disease, which may facilitate discussions with patients and referring providers.