Multicenter Development and Prospective Validation of eCARTv5: A Gradient Boosted Machine Learning Early Warning Score.

Rationale: Early detection of clinical deterioration using early warning scores may improve outcomes. However, most implemented scores were developed using logistic regression, only underwent retrospective internal validation, and were not tested in important patient subgroups. Objectives: To develop a gradient boosted machine model (eCARTv5) for identifying clinical deterioration and then validate externally, test prospectively, and evaluate across patient subgroups. Methods: All adult patients hospitalized on the wards in seven hospitals from 2008- 2022 were used to develop eCARTv5, with demographics, vital signs, clinician documentation, and laboratory values utilized to predict intensive care unit transfer or death in the next 24 hours. The model was externally validated retrospectively in 21 hospitals from 2009-2023 and prospectively in 10 hospitals from February to May 2023. eCARTv5 was compared to the Modified Early Warning Score (MEWS) and the National Early Warning Score (NEWS) using the area under the receiver operating characteristic curve (AUROC). Measurements and Main Results: The development cohort included 901,491 admissions, the retrospective validation cohort included 1,769,461 admissions, and the prospective validation cohort included 46,330 admissions. In retrospective validation, eCART had the highest AUROC (0.835; 95%CI 0.834, 0.835), followed by NEWS (0.766 (95%CI 0.766, 0.767)), and MEWS (0.704 (95%CI 0.703, 0.704)). eCART's performance remained high (AUROC ≥0.80) across a range of patient demographics, clinical conditions, and during prospective validation. Conclusions: We developed eCARTv5, which accurately identifies early clinical deterioration in hospitalized ward patients. Our model performed better than the NEWS and MEWS retrospectively, prospectively, and across a range of subgroups.

Validation of SeptiCyte RAPID to Discriminate Sepsis from Non-Infectious Systemic Inflammation.

(1) Background: SeptiCyte RAPID is a molecular test for discriminating sepsis from non-infectious systemic inflammation, and for estimating sepsis probabilities. The objective of this study was the clinical validation of SeptiCyte RAPID, based on testing retrospectively banked and prospectively collected patient samples. (2) Methods: The cartridge-based SeptiCyte RAPID test accepts a PAXgene blood RNA sample and provides sample-to-answer processing in ~1 h. The test output (SeptiScore, range 0-15) falls into four interpretation bands, with higher scores indicating higher probabilities of sepsis. Retrospective (N = 356) and prospective (N = 63) samples were tested from adult patients in ICU who either had the systemic inflammatory response syndrome (SIRS), or were suspected of having/diagnosed with sepsis. Patients were clinically evaluated by a panel of three expert physicians blinded to the SeptiCyte test results. Results were interpreted under either the Sepsis-2 or Sepsis-3 framework. (3) Results: Under the Sepsis-2 framework, SeptiCyte RAPID performance for the combined retrospective and prospective cohorts had Areas Under the ROC Curve (AUCs) ranging from 0.82 to 0.85, a negative predictive value of 0.91 (sensitivity 0.94) for SeptiScore Band 1 (score range 0.1-5.0; lowest risk of sepsis), and a positive predictive value of 0.81 (specificity 0.90) for SeptiScore Band 4 (score range 7.4-15; highest risk of sepsis). Performance estimates for the prospective cohort ranged from AUC 0.86-0.95. For physician-adjudicated sepsis cases that were blood culture (+) or blood, urine culture (+)(+), 43/48 (90%) of SeptiCyte scores fell in Bands 3 or 4. In multivariable analysis with up to 14 additional clinical variables, SeptiScore was the most important variable for sepsis diagnosis. A comparable performance was obtained for the majority of patients reanalyzed under the Sepsis-3 definition, although a subgroup of 16 patients was identified that was called septic under Sepsis-2 but not under Sepsis-3. (4) Conclusions: This study validates SeptiCyte RAPID for estimating sepsis probability, under both the Sepsis-2 and Sepsis-3 frameworks, for hospitalized patients on their first day of ICU admission.

Development and external validation of deep learning clinical prediction models using variable-length time series data.

OBJECTIVES: To compare and externally validate popular deep learning model architectures and data transformation methods for variable-length time series data in 3 clinical tasks (clinical deterioration, severe acute kidney injury [AKI], and suspected infection). MATERIALS AND METHODS: This multicenter retrospective study included admissions at 2 medical centers that spanned 2007-2022. Distinct datasets were created for each clinical task, with 1 site used for training and the other for testing. Three feature engineering methods (normalization, standardization, and piece-wise linear encoding with decision trees [PLE-DTs]) and 3 architectures (long short-term memory/gated recurrent unit [LSTM/GRU], temporal convolutional network, and time-distributed wrapper with convolutional neural network [TDW-CNN]) were compared in each clinical task. Model discrimination was evaluated using the area under the precision-recall curve (AUPRC) and the area under the receiver operating characteristic curve (AUROC). RESULTS: The study comprised 373 825 admissions for training and 256 128 admissions for testing. LSTM/GRU models tied with TDW-CNN models with both obtaining the highest mean AUPRC in 2 tasks, and LSTM/GRU had the highest mean AUROC across all tasks (deterioration: 0.81, AKI: 0.92, infection: 0.87). PLE-DT with LSTM/GRU achieved the highest AUPRC in all tasks. DISCUSSION: When externally validated in 3 clinical tasks, the LSTM/GRU model architecture with PLE-DT transformed data demonstrated the highest AUPRC in all tasks. Multiple models achieved similar performance when evaluated using AUROC. CONCLUSION: The LSTM architecture performs as well or better than some newer architectures, and PLE-DT may enhance the AUPRC in variable-length time series data for predicting clinical outcomes during external validation.

SARS-CoV-2 infection elucidates unique features of pregnancy-specific immunity.

Pregnancy is a risk factor for increased severity of SARS-CoV-2 and other respiratory infections. The mechanisms underlying this risk have not been well-established, partly due to a limited understanding of how pregnancy shapes immune responses. To gain insight into the role of pregnancy in modulating immune responses at steady state and upon perturbation, we collected peripheral blood mononuclear cells (PBMC), plasma, and stool from 226 women, including 152 pregnant individuals (n = 96 with SARS-CoV-2 infection and n = 56 healthy controls) and 74 non-pregnant women (n = 55 with SARS-CoV-2 and n = 19 healthy controls). We found that SARS-CoV-2 infection was associated with altered T cell responses in pregnant compared to non-pregnant women. Differences included a lower percentage of memory T cells, a distinct clonal expansion of CD4-expressing CD8 + T cells, and the enhanced expression of T cell exhaustion markers, such as programmed cell death-1 (PD-1) and T cell immunoglobulin and mucin domain-3 (Tim-3), in pregnant women. We identified additional evidence of immune dysfunction in severely and critically ill pregnant women, including a lack of expected elevation in regulatory T cell (Treg) levels, diminished interferon responses, and profound suppression of monocyte function. Consistent with earlier data, we found maternal obesity was also associated with altered immune responses to SARS-CoV-2 infection, including enhanced production of inflammatory cytokines by T cells. Certain gut bacterial species were altered in pregnancy and upon SARS-CoV-2 infection in pregnant individuals compared to non-pregnant women. Shifts in cytokine and chemokine levels were also identified in the sera of pregnant individuals, most notably a robust increase of interleukin-27 (IL-27), a cytokine known to drive T cell exhaustion, in the pregnant uninfected control group compared to all non-pregnant groups. IL-27 levels were also significantly higher in uninfected pregnant controls compared to pregnant SARS-CoV-2-infected individuals. Using two different preclinical mouse models of inflammation-induced fetal demise and respiratory influenza viral infection, we found that enhanced IL-27 protects developing fetuses from maternal inflammation but renders adult female mice vulnerable to viral infection. These combined findings from human and murine studies reveal nuanced pregnancy-associated immune responses, suggesting mechanisms underlying the increased susceptibility of pregnant individuals to viral respiratory infections.

Glutamate-dependent arginine biosynthesis requires the inactivation of spoVG, sarA, and ahrC in Staphylococcus aureus.

Genome sequencing has demonstrated that Staphylococcus aureus encodes arginine biosynthetic genes argDCJBFGH synthesizing proteins that mediate arginine biosynthesis using glutamate as a substrate. Paradoxically, however, S. aureus does not grow in a defined, glutamate-replete medium lacking arginine and glucose (CDM-R). Studies from our laboratory have found that specific mutations are selected by S. aureus that facilitate growth in CDM-R. However, these selected mutants synthesize arginine utilizing proline as a substrate rather than glutamate. In this study, we demonstrate that the ectopic expression of the argDCJB operon supports the growth of S. aureus in CDM-R, thus documenting the functionality of this pathway. Furthermore, suppressor mutants of S. aureus JE2 putA::Tn, which is defective in synthesizing arginine from proline, were selected on CDM-R agar. Genome sequencing revealed that these mutants had compensatory mutations within both spoVG, encoding an ortholog of the Bacillus subtilis stage V sporulation protein, and sarA, encoding the staphylococcal accessory regulator. Transcriptional studies document that argD expression is significantly increased when JE2 spoVG sarA was grown in CDM-R. Lastly, we found that a mutation in ahrC was required to induce argD expression in JE2 spoVG sarA when grown in an arginine-replete medium (CDM), suggesting that AhrC also functions to repress argDCJB in an arginine-dependent manner. In conclusion, these data indicate that the argDCJB operon is functional when transcribed in vitro and that SNPs within potential putative regulatory proteins are required to alleviate the repression.IMPORTANCEAlthough Staphylococcus aureus has the capability to synthesize all 20 amino acids, it is phenotypically auxotrophic for several amino acids including arginine. This work identifies putative regulatory proteins, including SpoVG, SarA, and AhrC, that function to inhibit the arginine biosynthetic pathways using glutamate as a substrate. Understanding the ultimate mechanisms of why S. aureus is selected to repress arginine biosynthetic pathways even in the absence of arginine will add to the growing body of work assessing the interactions between metabolism and S. aureus pathogenesis.

Causes, Diagnostic Testing, and Treatments Related to Clinical Deterioration Events among High-Risk Ward Patients.

OBJECTIVE: Timely intervention for clinically deteriorating ward patients requires that care teams accurately diagnose and treat their underlying medical conditions. However, the most common diagnoses leading to deterioration and the relevant therapies provided are poorly characterized. Therefore, we aimed to determine the diagnoses responsible for clinical deterioration, the relevant diagnostic tests ordered, and the treatments administered among high-risk ward patients using manual chart review. DESIGN: Multicenter retrospective observational study. SETTING: Inpatient medical-surgical wards at four health systems from 2006-2020 PATIENTS: Randomly selected patients (1,000 from each health system) with clinical deterioration, defined by reaching the 95th percentile of a validated early warning score, electronic Cardiac Arrest Risk Triage (eCART), were included. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Clinical deterioration was confirmed by a trained reviewer or marked as a false alarm if no deterioration occurred for each patient. For true deterioration events, the condition causing deterioration, relevant diagnostic tests ordered, and treatments provided were collected. Of the 4,000 included patients, 2,484 (62%) had clinical deterioration confirmed by chart review. Sepsis was the most common cause of deterioration (41%; n=1,021), followed by arrhythmia (19%; n=473), while liver failure had the highest in-hospital mortality (41%). The most common diagnostic tests ordered were complete blood counts (47% of events), followed by chest x-rays (42%), and cultures (40%), while the most common medication orders were antimicrobials (46%), followed by fluid boluses (34%), and antiarrhythmics (19%). CONCLUSIONS: We found that sepsis was the most common cause of deterioration, while liver failure had the highest mortality. Complete blood counts and chest x-rays were the most common diagnostic tests ordered, and antimicrobials and fluid boluses were the most common medication interventions. These results provide important insights for clinical decision-making at the bedside, training of rapid response teams, and the development of institutional treatment pathways for clinical deterioration. KEY POINTS: Question: What are the most common diagnoses, diagnostic test orders, and treatments for ward patients experiencing clinical deterioration? Findings: In manual chart review of 2,484 encounters with deterioration across four health systems, we found that sepsis was the most common cause of clinical deterioration, followed by arrythmias, while liver failure had the highest mortality. Complete blood counts and chest x-rays were the most common diagnostic test orders, while antimicrobials and fluid boluses were the most common treatments. Meaning: Our results provide new insights into clinical deterioration events, which can inform institutional treatment pathways, rapid response team training, and patient care.

Metformin treatment reduces inflammation, dysmyelination and disease severity in a mouse model of multiple sclerosis, experimental autoimmune encephalomyelitis.

Multiple sclerosis (MS) is an autoimmune disease characterized by inflammation, death or damage of oligodendrocytes, and axonal degeneration. Current MS treatments are non-curative, associated with undesired side-effects, and expensive, highlighting the need for expanded therapeutic options for patients. There is great interest in developing interventions using drugs or therapeutics to reduce symptom onset and protect pre-existing myelin. Metformin is a well-tolerated drug used to treat Type 2 diabetes that has pleiotropic effects in the central nervous system (CNS), including reducing inflammation, enhancing oligodendrogenesis, increasing the survival/proliferation of neural stem cells (NSCs), and increasing myelination. Here, we investigated whether metformin administration could improve functional outcomes, modulate oligodendrocyte precursor cells (OPCs), and reduce inflammation in a well-established mouse model of MS- experimental autoimmune encephalomyelitis (EAE). Male and female mice received metformin treatment at the time of EAE induction ("acute") or upon presentation of disease symptoms ("delayed"). We found that acute metformin treatment improved functional outcomes, concomitant with reduced microglia numbers and decreased dysmyelination. Conversely, delayed metformin treatment did not improve functional outcomes. Our findings reveal that metformin administration can improve EAE outcomes when administered before symptom onset in both sexes.

Better together: utilizing an interprofessional course and escape room to educate healthcare students about opioid use disorder.

BACKGROUND: The aim of the present study was to determine the impact of an innovative interprofessional educational activity on healthcare professional students' learning. The educational activity targeted student knowledge of opioid use disorder (OUD) and perceptions of working with an interprofessional team while caring for patients with OUD. METHODS: Students from nursing, pharmacy, physician assistant, dentistry, social work, and medicine programs were recruited to participate in the interprofessional educational activity. The educational experience included seven asynchronous modules and a virtual synchronous escape room. Prior to the educational programming, participants completed a pre-survey that assessed their knowledge and attitudes towards working on an interprofessional team and perceptions of patients with OUD. The asynchronous modules were required in order to participate in the escape room and each module contained its own pre/post quiz to assess student knowledge. RESULTS: A total of 402 students participated in the course. Prior to participating in the course, students disagreed that they had extensive educational experience with SUD (2.45 ± 0.79). The students displayed significant improvement in the knowledge based areas after completing the seven asynchronous modules. The largest significant area of knowledge-based improvement was seen in treatment of OUD where on the pre-quiz 65.54 ± 20.21% were answered correctly compared to 95.97 ± 9.61% on the post-quiz. Participation in the escape room significantly changed the students' perceptions of working in interprofessional teams while managing patients with OUD. Of the eleven perception variables assessed, seven showed a significant increase in the post-survey. Following the escape room, participants also strongly agreed that they now would refer patients to colleagues in other disciplines. CONCLUSIONS: An interprofessional educational experience including both an asynchronous course and virtual synchronous escape room can increase participant knowledge around OUD and may improve student perceptions of working with an interprofessional team and caring for patients with OUD.

Development and external validation of multimodal postoperative acute kidney injury risk machine learning models.

Objectives: To develop and externally validate machine learning models using structured and unstructured electronic health record data to predict postoperative acute kidney injury (AKI) across inpatient settings. Materials and Methods: Data for adult postoperative admissions to the Loyola University Medical Center (2009-2017) were used for model development and admissions to the University of Wisconsin-Madison (2009-2020) were used for validation. Structured features included demographics, vital signs, laboratory results, and nurse-documented scores. Unstructured text from clinical notes were converted into concept unique identifiers (CUIs) using the clinical Text Analysis and Knowledge Extraction System. The primary outcome was the development of Kidney Disease Improvement Global Outcomes stage 2 AKI within 7 days after leaving the operating room. We derived unimodal extreme gradient boosting machines (XGBoost) and elastic net logistic regression (GLMNET) models using structured-only data and multimodal models combining structured data with CUI features. Model comparison was performed using the receiver operating characteristic curve (AUROC), with Delong's test for statistical differences. Results: The study cohort included 138 389 adult patient admissions (mean [SD] age 58 [16] years; 11 506 [8%] African-American; and 70 826 [51%] female) across the 2 sites. Of those, 2959 (2.1%) developed stage 2 AKI or higher. Across all data types, XGBoost outperformed GLMNET (mean AUROC 0.81 [95% confidence interval (CI), 0.80-0.82] vs 0.78 [95% CI, 0.77-0.79]). The multimodal XGBoost model incorporating CUIs parameterized as term frequency-inverse document frequency (TF-IDF) showed the highest discrimination performance (AUROC 0.82 [95% CI, 0.81-0.83]) over unimodal models (AUROC 0.79 [95% CI, 0.78-0.80]). Discussion: A multimodality approach with structured data and TF-IDF weighting of CUIs increased model performance over structured data-only models. Conclusion: These findings highlight the predictive power of CUIs when merged with structured data for clinical prediction models, which may improve the detection of postoperative AKI.

Metformin Improves Functional Outcomes, Activates Neural Precursor Cells, and Modulates Microglia in a Sex-Dependent Manner After Spinal Cord Injury.

Spinal cord injury (SCI) results in devastating patient outcomes with few treatment options. A promising approach to improve outcomes following SCI involves the activation of endogenous precursor populations including neural stem and progenitor cells (NSPCs) which are located in the periventricular zone (PVZ), and oligodendrocyte precursor cells (OPCs) found throughout the parenchyma. In the adult spinal cord, resident NSPCs are primarily mitotically quiescent and aneurogenic, while OPCs contribute to ongoing oligodendrogenesis into adulthood. Each of these populations is responsive to SCI, increasing their proliferation and migration to the site of injury; however, their activation is not sufficient to support functional recovery. Previous work has shown that administration of the FDA-approved drug metformin is effective at promoting endogenous brain repair following injury, and this is correlated with enhanced NSPC activation. Here, we ask whether metformin can promote functional recovery and neural repair following SCI in both males and females. Our results reveal that acute, but not delayed metformin administration improves functional outcomes following SCI in both sexes. The functional improvement is concomitant with OPC activation and oligodendrogenesis. Our data also reveal sex-dependent effects of metformin following SCI with increased activation of NSPCs in females and reduced microglia activation in males. Taken together, these findings support metformin as a viable therapeutic strategy following SCI and highlight its pleiotropic effects in the spinal cord.

Association Between Race/Ethnicity and COVID-19 Outcomes in Systemic Lupus Erythematosus Patients From the United States: Data From the COVID-19 Global Rheumatology Alliance.

OBJECTIVE: To determine the association between race/ethnicity and COVID-19 outcomes in individuals with systemic lupus erythematosus (SLE). METHODS: Individuals with SLE from the US with data entered into the COVID-19 Global Rheumatology Alliance registry between March 24, 2020 and August 27, 2021 were included. Variables included age, sex, race, and ethnicity (White, Black, Hispanic, other), comorbidities, disease activity, pandemic time period, glucocorticoid dose, antimalarials, and immunosuppressive drug use. The ordinal outcome categories were: not hospitalized, hospitalized with no oxygenation, hospitalized with any ventilation or oxygenation, and death. We constructed ordinal logistic regression models evaluating the relationship between race/ethnicity and COVID-19 severity, adjusting for possible confounders. RESULTS: We included 523 patients; 473 (90.4%) were female and the mean ± SD age was 46.6 ± 14.0 years. A total of 358 patients (74.6%) were not hospitalized; 40 patients (8.3%) were hospitalized without oxygen, 64 patients (13.3%) were hospitalized with any oxygenation, and 18 (3.8%) died. In a multivariable model, Black (odds ratio [OR] 2.73 [95% confidence interval (95% CI) 1.36-5.53]) and Hispanic (OR 2.76 [95% CI 1.34-5.69]) individuals had higher odds of more severe outcomes than White individuals. CONCLUSION: Black and Hispanic individuals with SLE experienced more severe COVID-19 outcomes, which is consistent with findings in the US general population. These results likely reflect socioeconomic and health disparities and suggest that more aggressive efforts are needed to prevent and treat infection in this population.

Effectiveness of Bupropion and Varenicline for Smokers With Baseline Depressive Symptoms.

INTRODUCTION: Smokers with concurrent depression are less likely to achieve abstinence, even with pharmacotherapy. The purpose of this secondary data analysis was to evaluate if the presence of any depressive symptoms at baseline alters the effectiveness of bupropion and varenicline for smoking cessation. AIMS AND METHODS: Eligible participants were enrolled via the internet and randomized 1:1 to receive a 12-week supply of either bupropion (n = 465) or varenicline (n = 499). Depressive symptoms were assessed using the Patient Health Questionnaire (PHQ-2). Follow-up surveys were conducted at weeks 4, 8, 12, 26, and 52 to assess self-reported quit. The primary outcome was 7-day point prevalence abstinence at 12 weeks follow-up (end-of-treatment). RESULTS: Participants who endorsed any depressive symptoms (PHQ-2 > 0; n = 280) were less likely to be quit at end-of-treatment compared to participants who did not endorse any symptoms (PHQ-2 = 0; n = 684) (OR = 0.56, 95% CI: 0.38 to 0.8, p = .003). Within the varenicline group, quit outcomes did not differ between those with and without depressive symptoms (21.3% vs. 26.9%, respectively). Within the bupropion group, however, those with symptoms had a significantly reduced quit rate compared to those without symptoms (7.0% vs. 17.3%, respectively). CONCLUSIONS: The presence of even one symptom of depression at the start of a quit attempt may adversely affect quit outcomes. Patients should be assessed for depressive symptoms when planning to quit smoking as it may inform the approach to treatment. However, future studies are needed to confirm these findings. IMPLICATIONS: Findings from the current study illustrate the importance of evaluating baseline sub-clinical depressive symptoms before a quit attempt using first-line pharmacotherapies. This secondary analysis of a large-scale randomized trial suggests that bupropion may be less effective for those with baseline depressive symptoms while varenicline may be equally effective for those with and without depressive symptoms.

Association between mortality and critical events within 48 hours of transfer to the pediatric intensive care unit.

Introduction: Critical deterioration in hospitalized children, defined as ward to pediatric intensive care unit (PICU) transfer followed by mechanical ventilation (MV) or vasoactive infusion (VI) within 12 h, has been used as a primary metric to evaluate the effectiveness of clinical interventions or quality improvement initiatives. We explore the association between critical events (CEs), i.e., MV or VI events, within the first 48 h of PICU transfer from the ward or emergency department (ED) and in-hospital mortality. Methods: We conducted a retrospective study of a cohort of PICU transfers from the ward or the ED at two tertiary-care academic hospitals. We determined the association between mortality and occurrence of CEs within 48 h of PICU transfer after adjusting for age, gender, hospital, and prior comorbidities. Results: Experiencing a CE within 48 h of PICU transfer was associated with an increased risk of mortality [OR 12.40 (95% CI: 8.12-19.23, P < 0.05)]. The increased risk of mortality was highest in the first 12 h [OR 11.32 (95% CI: 7.51-17.15, P < 0.05)] but persisted in the 12-48 h time interval [OR 2.84 (95% CI: 1.40-5.22, P < 0.05)]. Varying levels of risk were observed when considering ED or ward transfers only, when considering different age groups, and when considering individual 12-h time intervals. Discussion: We demonstrate that occurrence of a CE within 48 h of PICU transfer was associated with mortality after adjusting for confounders. Studies focusing on the impact of quality improvement efforts may benefit from using CEs within 48 h of PICU transfer as an additional evaluation metric, provided these events could have been influenced by the initiative.

Indolent T-Cell Lymphoproliferative Disease: A Rare Case of a Benign Lymphoma of the Gastrointestinal Tract With Extra-Gastrointestinal Involvement.

Indolent T-cell lymphoproliferative disease of the gastrointestinal (GI) tract is an exceedingly rare benign proliferation of clonal and mature-appearing lymphoid cells originating from the GI tract. We discuss the case of a 52-year-old woman with indolent T-cell lymphoproliferative disease of the GI tract manifesting as chronic diarrhea and profound weight loss. Interestingly, the patient also had extra-GI involvement of her disease process, which has not been previously reported. Our patient was managed with steroids with improvement in symptoms and weight gain. We provide a review of the literature to highlight the importance of early recognition and intervention of this disease entity.

Risk of severe COVID-19 outcomes associated with rheumatoid arthritis and phenotypic subgroups: a retrospective, comparative, multicentre cohort study.

Background: Rheumatoid arthritis has been associated with severe COVID-19, but few studies have investigated how phenotypes of rheumatoid arthritis affect these associations. We aimed to investigate the associations between rheumatoid arthritis and phenotypes of interstitial lung disease, serostatus, and bone erosions with COVID-19 severity. Methods: We did a retrospective, comparative, multicentre cohort study at two large health-care systems (Mayo Clinic [19 hospitals and affiliated outpatient centres] and Mass General Brigham [14 hospitals and affiliated outpatient centres]) in the USA. Consecutive patients with rheumatoid arthritis meeting the 2010 American College of Rheumatology/European League Against Rheumatism classification criteria and who had COVID-19 between March 1, 2020, and June 6, 2021, were matched 1:5 on age, sex, and calendar date with patients without rheumatoid arthritis (comparators). Data were received from electronic health records from Mayo Clinic and Mass General Brigham. We examined subgroups of patients with rheumatoid arthritis by phenotypic features: rheumatoid arthritis-associated interstitial lung disease, seropositivity (for anti-cyclic citrullinated peptide, rheumatoid factor, or both), and bone erosions. Severe COVID-19 was a composite of hospitalisation or death. We used Cox regression to estimate hazard ratios (HR) for severe COVID-19, comparing rheumatoid arthritis and subgroups to the comparator group. Findings: We identified 582 patients with rheumatoid arthritis and 2875 matched comparators, all of whom had COVID-19 within the study dates. The mean age of those with rheumatoid arthritis was 62 [SD 14] years, 421 (72%) of 582 were women and 161 (28%) were men, 457 (79%) were White, 65 (11%) were Hispanic or Latino, and 41 (7%) were Black. Among patients with rheumatoid arthritis, 50 (9%) of 582 had interstitial lung disease, 388 (68%) of 568 were seropositive, and 159 (27%) of 582 had bone erosions. Severe COVID-19 occurred in 126 (22%) of 582 patients with rheumatoid arthritis versus 363 (13%) 2875 in the comparator group. Patients with rheumatoid arthritis had an HR of 1·75 (95% CI 1·45-2·10) for severe COVID-19 versus the comparator group. Patients with rheumatoid arthritis-associated interstitial lung disease had an HR of 2·50 (1·66-3·77) versus the comparator group for severe COVID-19. The risk for severe COVID-19 was also higher in patients with rheumatoid arthritis who were seropositive (HR 1·97 [95% CI 1·58-2·46]) or had erosive disease (1·93 [1·41-2·63]) than for those in the comparator group. Interpretation: Patients with rheumatoid arthritis have an increased risk of severe COVID-19 across phenotypic subgroups, especially among patients with interstitial lung disease. These findings suggest that rheumatoid arthritis with interstitial lung disease, or its treatment, might be a substantial contributor to severe COVID-19 outcomes for patients with rheumatoid arthritis. Funding: None.

Environmental and societal factors associated with COVID-19-related death in people with rheumatic disease: an observational study.

Background: Differences in the distribution of individual-level clinical risk factors across regions do not fully explain the observed global disparities in COVID-19 outcomes. We aimed to investigate the associations between environmental and societal factors and country-level variations in mortality attributed to COVID-19 among people with rheumatic disease globally. Methods: In this observational study, we derived individual-level data on adults (aged 18-99 years) with rheumatic disease and a confirmed status of their highest COVID-19 severity level from the COVID-19 Global Rheumatology Alliance (GRA) registry, collected between March 12, 2020, and Aug 27, 2021. Environmental and societal factors were obtained from publicly available sources. The primary endpoint was mortality attributed to COVID-19. We used a multivariable logistic regression to evaluate independent associations between environmental and societal factors and death, after controlling for individual-level risk factors. We used a series of nested mixed-effects models to establish whether environmental and societal factors sufficiently explained country-level variations in death. Findings: 14 044 patients from 23 countries were included in the analyses. 10 178 (72·5%) individuals were female and 3866 (27·5%) were male, with a mean age of 54·4 years (SD 15·6). Air pollution (odds ratio 1·10 per 10 µg/m3 [95% CI 1·01-1·17]; p=0·0105), proportion of the population aged 65 years or older (1·19 per 1% increase [1·10-1·30]; p<0·0001), and population mobility (1·03 per 1% increase in number of visits to grocery and pharmacy stores [1·02-1·05]; p<0·0001 and 1·02 per 1% increase in number of visits to workplaces [1·00-1·03]; p=0·032) were independently associated with higher odds of mortality. Number of hospital beds (0·94 per 1-unit increase per 1000 people [0·88-1·00]; p=0·046), human development index (0·65 per 0·1-unit increase [0·44-0·96]; p=0·032), government response stringency (0·83 per 10-unit increase in containment index [0·74-0·93]; p=0·0018), as well as follow-up time (0·78 per month [0·69-0·88]; p<0·0001) were independently associated with lower odds of mortality. These factors sufficiently explained country-level variations in death attributable to COVID-19 (intraclass correlation coefficient 1·2% [0·1-9·5]; p=0·14). Interpretation: Our findings highlight the importance of environmental and societal factors as potential explanations of the observed regional disparities in COVID-19 outcomes among people with rheumatic disease and lay foundation for a new research agenda to address these disparities. Funding: American College of Rheumatology and European Alliance of Associations for Rheumatology.

Identifying infected patients using semi-supervised and transfer learning.

OBJECTIVES: Early identification of infection improves outcomes, but developing models for early identification requires determining infection status with manual chart review, limiting sample size. Therefore, we aimed to compare semi-supervised and transfer learning algorithms with algorithms based solely on manual chart review for identifying infection in hospitalized patients. MATERIALS AND METHODS: This multicenter retrospective study of admissions to 6 hospitals included "gold-standard" labels of infection from manual chart review and "silver-standard" labels from nonchart-reviewed patients using the Sepsis-3 infection criteria based on antibiotic and culture orders. "Gold-standard" labeled admissions were randomly allocated to training (70%) and testing (30%) datasets. Using patient characteristics, vital signs, and laboratory data from the first 24 hours of admission, we derived deep learning and non-deep learning models using transfer learning and semi-supervised methods. Performance was compared in the gold-standard test set using discrimination and calibration metrics. RESULTS: The study comprised 432 965 admissions, of which 2724 underwent chart review. In the test set, deep learning and non-deep learning approaches had similar discrimination (area under the receiver operating characteristic curve of 0.82). Semi-supervised and transfer learning approaches did not improve discrimination over models fit using only silver- or gold-standard data. Transfer learning had the best calibration (unreliability index P value: .997, Brier score: 0.173), followed by self-learning gradient boosted machine (P value: .67, Brier score: 0.170). DISCUSSION: Deep learning and non-deep learning models performed similarly for identifying infection, as did models developed using Sepsis-3 and manual chart review labels. CONCLUSION: In a multicenter study of almost 3000 chart-reviewed patients, semi-supervised and transfer learning models showed similar performance for model discrimination as baseline XGBoost, while transfer learning improved calibration.

Development of a Prediction Model for COVID-19 Acute Respiratory Distress Syndrome in Patients With Rheumatic Diseases: Results From the Global Rheumatology Alliance Registry.

OBJECTIVE: Some patients with rheumatic diseases might be at higher risk for coronavirus disease 2019 (COVID-19) acute respiratory distress syndrome (ARDS). We aimed to develop a prediction model for COVID-19 ARDS in this population and to create a simple risk score calculator for use in clinical settings. METHODS: Data were derived from the COVID-19 Global Rheumatology Alliance Registry from March 24, 2020, to May 12, 2021. Seven machine learning classifiers were trained on ARDS outcomes using 83 variables obtained at COVID-19 diagnosis. Predictive performance was assessed in a US test set and was validated in patients from four countries with independent registries using area under the curve (AUC), accuracy, sensitivity, and specificity. A simple risk score calculator was developed using a regression model incorporating the most influential predictors from the best performing classifier. RESULTS: The study included 8633 patients from 74 countries, of whom 523 (6%) had ARDS. Gradient boosting had the highest mean AUC (0.78; 95% confidence interval [CI]: 0.67-0.88) and was considered the top performing classifier. Ten predictors were identified as key risk factors and were included in a regression model. The regression model that predicted ARDS with 71% (95% CI: 61%-83%) sensitivity in the test set, and with sensitivities ranging from 61% to 80% in countries with independent registries, was used to develop the risk score calculator. CONCLUSION: We were able to predict ARDS with good sensitivity using information readily available at COVID-19 diagnosis. The proposed risk score calculator has the potential to guide risk stratification for treatments, such as monoclonal antibodies, that have potential to reduce COVID-19 disease progression.

Regulating Endogenous Neural Stem Cell Activation to Promote Spinal Cord Injury Repair.

Spinal cord injury (SCI) affects millions of individuals worldwide. Currently, there is no cure, and treatment options to promote neural recovery are limited. An innovative approach to improve outcomes following SCI involves the recruitment of endogenous populations of neural stem cells (NSCs). NSCs can be isolated from the neuroaxis of the central nervous system (CNS), with brain and spinal cord populations sharing common characteristics (as well as regionally distinct phenotypes). Within the spinal cord, a number of NSC sub-populations have been identified which display unique protein expression profiles and proliferation kinetics. Collectively, the potential for NSCs to impact regenerative medicine strategies hinges on their cardinal properties, including self-renewal and multipotency (the ability to generate de novo neurons, astrocytes, and oligodendrocytes). Accordingly, endogenous NSCs could be harnessed to replace lost cells and promote structural repair following SCI. While studies exploring the efficacy of this approach continue to suggest its potential, many questions remain including those related to heterogeneity within the NSC pool, the interaction of NSCs with their environment, and the identification of factors that can enhance their response. We discuss the current state of knowledge regarding populations of endogenous spinal cord NSCs, their niche, and the factors that regulate their behavior. In an attempt to move towards the goal of enhancing neural repair, we highlight approaches that promote NSC activation following injury including the modulation of the microenvironment and parenchymal cells, pharmaceuticals, and applied electrical stimulation.

Characteristics associated with poor COVID-19 outcomes in individuals with systemic lupus erythematosus: data from the COVID-19 Global Rheumatology Alliance.

AIM: To determine characteristics associated with more severe outcomes in a global registry of people with systemic lupus erythematosus (SLE) and COVID-19. METHODS: People with SLE and COVID-19 reported in the COVID-19 Global Rheumatology Alliance registry from March 2020 to June 2021 were included. The ordinal outcome was defined as: (1) not hospitalised, (2) hospitalised with no oxygenation, (3) hospitalised with any ventilation or oxygenation and (4) death. A multivariable ordinal logistic regression model was constructed to assess the relationship between COVID-19 severity and demographic characteristics, comorbidities, medications and disease activity. RESULTS: A total of 1606 people with SLE were included. In the multivariable model, older age (OR 1.03, 95% CI 1.02 to 1.04), male sex (1.50, 1.01 to 2.23), prednisone dose (1-5 mg/day 1.86, 1.20 to 2.66, 6-9 mg/day 2.47, 1.24 to 4.86 and ≥10 mg/day 1.95, 1.27 to 2.99), no current treatment (1.80, 1.17 to 2.75), comorbidities (eg, kidney disease 3.51, 2.42 to 5.09, cardiovascular disease/hypertension 1.69, 1.25 to 2.29) and moderate or high SLE disease activity (vs remission; 1.61, 1.02 to 2.54 and 3.94, 2.11 to 7.34, respectively) were associated with more severe outcomes. In age-adjusted and sex-adjusted models, mycophenolate, rituximab and cyclophosphamide were associated with worse outcomes compared with hydroxychloroquine; outcomes were more favourable with methotrexate and belimumab. CONCLUSIONS: More severe COVID-19 outcomes in individuals with SLE are largely driven by demographic factors, comorbidities and untreated or active SLE. Patients using glucocorticoids also experienced more severe outcomes.