Using explainable machine learning to identify patients at risk of reattendance at discharge from emergency departments.

Short-term reattendances to emergency departments are a key quality of care indicator. Identifying patients at increased risk of early reattendance could help reduce the number of missed critical illnesses and could reduce avoidable utilization of emergency departments by enabling targeted post-discharge intervention. In this manuscript, we present a retrospective, single-centre study where we created and evaluated an extreme gradient boosting decision tree model trained to identify patients at risk of reattendance within 72 h of discharge from an emergency department (University Hospitals Southampton Foundation Trust, UK). Our model was trained using 35,447 attendances by 28,945 patients and evaluated on a hold-out test set featuring 8847 attendances by 7237 patients. The set of attendances from a given patient appeared exclusively in either the training or the test set. Our model was trained using both visit level variables (e.g., vital signs, arrival mode, and chief complaint) and a set of variables available in a patients electronic patient record, such as age and any recorded medical conditions. On the hold-out test set, our highest performing model obtained an AUROC of 0.747 (95% CI 0.722-0.773) and an average precision of 0.233 (95% CI 0.194-0.277). These results demonstrate that machine-learning models can be used to classify patients, with moderate performance, into low and high-risk groups for reattendance. We explained our models predictions using SHAP values, a concept developed from coalitional game theory, capable of explaining predictions at an attendance level. We demonstrated how clustering techniques (the UMAP algorithm) can be used to investigate the different sub-groups of explanations present in our patient cohort.

Up-regulation of chemokine C-C ligand 2 (CCL2) and C-X-C chemokine 8 (CXCL8) expression by monocytes in chronic idiopathic urticaria.

The disturbed cytokine-chemokine network could play an important role in the onset of diseases with inflammatory processes such as chronic idiopathic urticaria (CIU). Our main objectives were to evaluate the relation between proinflammatory chemokine serum levels from CIU patients and their response to autologous skin test (ASST) and basophil histamine release (BHR). We also aimed to assess the chemokine secretion by peripheral blood mononuclear cells (PBMC) upon polyclonal stimulus and to evaluate chemokine C-C ligand 2/C-X-C chemokine 8 (CCL2/CXCL8) and Toll-like receptor-4 (TLR-4) expression in monocytes. We observed significantly higher serum levels of the CXCL8, CXCL9, CXCL10 and CCL2 in CIU patients compared to the healthy group, regardless of the BHR or ASST response. The basal secretion of CCL2 by PBMC or induced by Staphylococcus aureus enterotoxin A (SEA) was higher in CIU patients than in the control group, as well as for CXCL8 and CCL5 secretions upon phytohaemagglutinin stimulation. Also, up-regulation of CCL2 and CXCL8 mRNA expression was found in monocytes of patients upon SEA stimulation. The findings showed a high responsiveness of monocytes through CCL2/CXCL8 expression, contributing to the creation of a proinflammatory environment in CIU.

Statin effects on regulatory and proinflammatory factors in chronic idiopathic urticaria.

Immunological dysfunction has been described to occur in chronic idiopathic urticaria (CIU), most notably in association with an inflammatory process. Some pharmacological agents as statins--drugs used in hypercholesterolaemia--display a broad effect on the immune response and thus should be tested in vitro in CIU. Our main objectives were to evaluate the effects of statins on the innate and adaptive immune response in CIU. Simvastatin or lovastatin have markedly inhibited the peripheral blood mononuclear cells (PBMC) proliferative response induced by T and B cell mitogens, superantigen or recall antigen. Simvastatin arrested phytohaemaglutinin (PHA)-induced T cells at the G0/G1 phase, inhibiting T helper type 1 (Th1), Th2, interleukin (IL)-10 and IL-17A cytokine secretion in both patients and healthy control groups. Up-regulation of suppressor of cytokine signalling 3 (SOCS3) mRNA expression in PHA-stimulated PBMCs from CIU patients was not modified by simvastatin, in contrast to the enhancing effect in the control group. Statin exhibited a less efficient inhibition effect on cytokine production [IL-6 and macrophage inflammatory protein (MIP)-1α] induced by Toll-like receptor (TLR)-4, to which a statin preincubation step was required. Furthermore, statin did not affect the tumour necrosis factor (TNF)-α secretion by lipopolysaccharide (LPS)-stimulated PBMC or CD14+ cells in CIU patients. In addition, LPS-activated PBMC from CIU patients showed impaired indoleamine 2,3-dioxygenase (IDO) mRNA expression compared to healthy control, which remained at decreased levels with statin treatment. Statins exhibited a marked down-regulatory effect in T cell functions, but were not able to control TLR-4 activation in CIU patients. The unbalanced regulatory SOCS3 and IDO expressions in CIU may contribute to the pathogenesis of the disease.

Impaired IFN-α secretion by plasmacytoid dendritic cells induced by TLR9 activation in chronic idiopathic urticaria.

BACKGROUND: Understanding the early events of the immune response, through the activation of plasmacytoid dendritic cells (pDC) by Toll-like receptor (TLR)9-sensing, could contribute to the evaluation of immune dysregulation in chronic idiopathic urticaria (CIU). OBJECTIVES: We decided to investigate innate immunity in CIU and the mechanisms implicated in the modulation of interferon (IFN)-α production by pDC upon TLR9 activation. METHODS: Patients with CIU (n = 31) and healthy control subjects (HC, n = 36) were enrolled in the study. Leucocytes cultured with the TLR9 ligand, CpG type A, or with inhibitory-oligodeoxynucleotide (ODN) were used to determine IFN-α secretion by enzyme-linked immunosorbent assay. Enumeration of pDC, intracellular IFN-α and signal transducers and activators of transcription protein (STAT) (1 and 4) phosphorylation were assessed by flow cytometry. TLR9 and regulatory factor-7 mRNA transcripts were evaluated by real-time polymerase chain reaction. Evidence of pDC in the skin lesions of patients was analysed with immunohistochemistry staining. RESULTS: The findings show a decreased IFN-α secretion induced by CpG A by leucocytes, due to the diminished IFN-α expression on pDC in CIU. It was mediated by TLR9-activation since inhibitory-ODN further suppressed TLR9-induced IFN-α secretion. A normal pDC percentage and degree of activation by the expression of costimulatory molecules was observed in CIU, with the rare presence of pDC in the skin lesion. In addition, an increased constitutive STAT1 phosphorylation on nonstimulated lymphocytes and a downregulation of TLR9 mRNA transcripts after CpG A activation were verified in patients with CIU. CONCLUSIONS: The findings showed an innate immune response in CIU disturbed by impairment of the pDC response to TLR9 activation.

Activated status of basophils in chronic urticaria leads to interleukin-3 hyper-responsiveness and enhancement of histamine release induced by anti-IgE stimulus.

BACKGROUND: Basophils and mast cells are the main target cells in chronic idiopathic urticaria (CIU). Besides the basopenia, intrinsic defects of the anti-IgE cross-linking signalling pathway of basophils have been described in CIU. OBJECTIVES: We sought to investigate the profile of expression of activation markers on basophils of patients with CIU and to explore the effect of interleukin (IL)-3 priming upon anti-IgE cross-linking stimuli through expression of activation markers and basophil histamine releasability. METHODS: Evaluation of the surface expression of FcepsilonRIalpha, CD63, CD203c and CD123 on whole blood basophils of patients with CIU undergoing autologous serum skin test (ASST) was performed by flow cytometry. The effect of pretreatment with IL-3 in the anti-IgE response was analysed by the expression of basophil activation markers and histamine release using enzyme-linked immunosorbent assay. RESULTS: Blood basophils of patients with CIU were reduced in number and displayed increased surface expression of FcepsilonRIalpha, which was positively correlated with the IgE serum levels. Upregulation of expression of both surface markers CD203c and CD63 was verified on basophils of patients with CIU, regardless of ASST response. High expression of IL-3 receptor on basophils was detected only in ASST+ patients with CIU. Pretreatment with IL-3 upregulated CD203c expression concomitantly with the excreting function of blood basophils and induced a quick hyper-responsiveness to anti-IgE cross-linking on basophils of patients with CIU compared with healthy controls. CONCLUSIONS: Basophils of patients with CIU showed an activated profile, possibly due to an in vivo priming. Functionally, basophils have high responsiveness to IL-3 stimulation, thereby suggesting that defects in the signal transduction pathway after IgE cross-linking stimuli are recoverable in subjects with chronic urticaria.

Spina bifida occulta of S1 is not an innocent finding.

One thousand two hundred patients (600 women and 600 men) aged 18 to 72 years were referred for computed tomographic examination of the lumbosacral spine (L3-S1) after low-back pain or sciatica. Patients with spinal abnormalities other than spina bifida occulta (SBO)-S1 and findings other than posterior herniation of intervertebral disc were not included in this study. All of the patients underwent conventional radiographs of the lumbosacral spine. The incidence of SBO-S1 was higher in younger age groups and decreased with age. Patients with SBO-S1 showed a higher incidence of posterior disc herniation, which increased with age. This can be explained by instability of the base of the lumbar spine caused by SBO-S1, which produces a predisposition to posterior disc herniation. The results were statistically significant.