Can a novel clinical risk score improve pneumonia prediction in acute stroke care? A UK multicenter cohort study

Annexin A5 derived from lung alleviates brain damage after ischemic stroke

Ischemic stroke is a leading cause of disability and death worldwide. It is now accepted that brain interacts bidirectionally with other organs after brain diseases. However, factors that might mediate crosstalk between brain and other organs are still less reported. Here we reported that plasma level of Annexin A5, not Annexin A1 or A2, was upregulated in stroke patients when compared to controls. In normal mice, the highest levels of Annexin A5 were detected in lung tissues compared with other major organs and lowest level in brain. Moreover, Annexin A5 was increased in brain and decreased in lung after stroke in mice when compared to sham group. Fluorescence in situ hybridization (FISH) assay indicated that Annexin A5 could penetrate the blood-brain barrier (BBB). Treatment with Annexin A5 recombinant protein reduced the infarct volumes and improved neurological function after stroke in mice, while administration of anti-Annexin A5 increased the infarct sizes and aggravated neurological function. In a proof-of-concept analysis, patients with both ischemic stroke and lung diseases had a lower plasma Annexin A5 level than those with only ischemic stroke. Furthermore, Annexin A5 level in bronchoalveolar lavage fluid (BALF) was lower in patients with severe chronic obstructive pulmonary disease (COPD) when compared with those at a less severe grade of COPD, and level of Annexin A5 was positively correlated with forced expiratory volume in 1 s (FEV1) and PaO2. Our results suggest that Annexin A5 could alleviate infarct area and improve general neurological performance post cerebral ischemia. Increased Annexin A5 may derive from lung tissue and permeate across BBB to provide a neuroprotective function. Therefore, Annexin A5 may potentially serve as a therapeutic candidate for defending against IS-induced brain injury.

Association Between Blood-Brain Barrier Disruption and Stroke-Associated Pneumonia in Acute Ischemic Stroke Patients After Endovascular Therapy: A Retrospective Cohort Study

Background

Stroke, particularly due to large vessel occlusion (LVO), is a major cause of mortality and disability globally. Endovascular therapy (ET) significantly improves outcomes for acute ischemic stroke (AIS) patients, but complications such as stroke-associated pneumonia (SAP) increase mortality and healthcare costs. This study investigates the association between blood-brain barrier (BBB) disruption and the increased risk of SAP and explores the relationship between BBB disruption and medium-term functional outcomes.

Methods

The retrospective cohort study was performed on AIS patients enrolled between January 2019 to February 2023 who underwent ET. Patients were divided into two groups: BBB disruption and without BBB disruption. Multiple logistic regression model was conducted to measure the association between BBB disruption and SAP. Mediation analysis was used to estimate the potential mediation effects on the associations of BBB disruption with SAP. A restricted cubic spline (RCS) regression model was used to further outline the connection between the highest CT value of hyperattenuated lesions areas and the risk of SAP.

Results

The study included 254 patients who underwent endovascular therapy, with 155 patients in the BBB disruption group (exposure) and 99 patients in the without BBB disruption group (control). Multiple logistic regression analysis revealed a significantly increased risk of SAP in patients with BBB disruption (OR = 2.337, 95% CI: 1.118-4.990, p = 0.025). Furthermore, mediation analysis suggested that this association may be partly due to malignant cerebral oedema and haemorrhagic transformation. The study found an inverse L-shaped dose-response relationship between the maximum CT values of BBB disruption areas and the incidence of SAP. SAP partially mediated the association between BBB disruption and 3-month poor functional outcome.

Conclusion

BBB disruption are a potential risk factor for SAP. BBB disruption may affect short- and medium-term prognosis of patients after ET in part through SAP.

Systemic inflammatory response index as a predictor of stroke-associated pneumonia in patients with acute ischemic stroke treated by thrombectomy: a retrospective study

BACKGROUND

The predictive value of systemic inflammatory response index (SIRI) for stroke-associated pneumonia (SAP) risk in patients with acute ischemic stroke (AIS) treated by thrombectomy remains unclear. This study aimed to investigate the predictive value of SIRI for SAP in patients with AIS treated by thrombectomy.

METHODS

We included AIS patients treated by thrombectomy between August 2018 and August 2022 at our institute. We used multivariate logistic regression to construct the prediction model and performed a receiver operating characteristic curve analysis to evaluate the ability of SIRI to predict SAP and constructed a calibration curve to evaluate the prediction accuracy of the model. We evaluated the clinical application value of the nomogram using decision curve analysis.

RESULTS

We included 84 eligible patients with AIS in the analysis, among which 56 (66.7%) had SAP. In the univariate analysis, there were significant differences in sex (p = 0.035), National Institute of Health Stroke Scale score at admission ≥ 20 (p = 0.019) and SIRI (p < 0.001). The results of multivariable logistic analysis showed that the risk of SAP increased with the SIRI value (OR = 1.169, 95% CI = 1.049-1.344, p = 0.014). Age ≥ 60 (OR = 4.076, 95% CI = 1.251-14.841, p = 0.024) was also statistically significant. A nomogram with SIRI showed good prediction accuracy for SAP in AIS patients treated by thrombectomy (C-index value = 0.774).

CONCLUSIONS

SIRI is an independent predictor for SAP in patients with AIS treated by thrombectomy. A high SIRI value may allow for the early identification of patients with AIS treated by thrombectomy at high risk for SAP.

Bacterial stroke-associated pneumonia: microbiological analysis and mortality outcome

BACKGROUND

Stroke-associated pneumonia (SAP) considerably burden healthcare systems. This study aimed to identify predictors of developing SAP in acute ischemic stroke patients admitted to the Stroke Unit at Manial Specialized Hospital factors with microbiological causality and impact on 30-day mortality.

METHODS

This was a retrospective cohort study. All patients with acute ischemic stroke admitted to the Stroke Unit at Manial Specialized Hospital (from February 2021 to August 2023) were divided into the SAP and non-SAP groups. Detailed clinical characteristics and microbiological results were recorded.

RESULTS

Five hundred twenty-two patients diagnosed with acute ischemic stroke (mean age of 55 ± 10) were included. One hundred sixty-nine (32.4%) of stroke patients developed SAP; Klebsiella pneumoniae was the most commonly detected pathogen (40.2%), followed by Pseudomonas aeruginosa (20.7%). Bacteremia was identified in nine cases (5.3%). The number of deaths was 11, all of whom were diagnosed with SAP, whereas none from the non-SAP group died (P < 0.001). The binary logistic regression model identified three independent predictors of the occurrence of SAP: previous history of TIA/stroke (OR = 3.014, 95%CI = 1.281-7.092), mechanical ventilation (OR = 4.883, 95%CI = 1.544-15.436), and bulbar dysfunction (OR = 200.460, 95%CI = 80.831-497.143).

CONCLUSIONS

Stroke-associated pneumonia was reported in one-third of patients with acute ischemic stroke, adversely affecting mortality outcomes. Findings showed that the main predictors of SAP were bulbar dysfunction, the use of mechanical ventilation and previous history of TIA/stroke. More attention to these vulnerable patients is necessary to reduce mortality.

Development and validation of radiology-clinical statistical and machine learning model for stroke-associated pneumonia after first intracerebral haemorrhage

BACKGROUND

Society is burdened with stroke-associated pneumonia (SAP) after intracerebral haemorrhage (ICH). Cerebral small vessel disease (CSVD) complicates clinical manifestations of stroke. In this study, we redefined the CSVD burden score and incorporated it into a novel radiological-clinical prediction model for SAP.

MATERIALS AND METHODS

A total of 1278 patients admitted to a tertiary hospital between 1 January 2010 and 31 December 2019 were included. The participants were divided into training and testing groups using fivefold cross-validation method. Four models, two traditional statistical models (logistic regression and ISAN) and two machine learning models (random forest and support vector machine), were established and evaluated. The outcomes and baseline characteristics were compared between the SAP and non-SAP groups.

RESULTS

Among the of 1278 patients, 281(22.0%) developed SAP after their first ICH. Multivariate analysis revealed that the logistic regression (LR) model was superior in predicting SAP in both the training and testing groups. Independent predictors of SAP after ICH included total CSVD burden score (OR, 1.29; 95% CI, 1.03-1.54), haematoma extension into ventricle (OR, 2.28; 95% CI, 1.87-3.31), haematoma with multilobar involvement (OR, 2.14; 95% CI, 1.44-3.18), transpharyngeal intubation operation (OR, 3.89; 95% CI, 2.7-5.62), admission NIHSS score ≥ 10 (OR, 2.06; 95% CI, 1.42-3.01), male sex (OR, 1.69; 95% CI, 1.16-2.52), and age ≥ 67 (OR, 2.24; 95% CI, 1.56-3.22). The patients in the SAP group had worse outcomes than those in the non-SAP group.

CONCLUSION

This study established a clinically combined imaging model for predicting stroke-associated pneumonia and demonstrated superior performance compared with the existing ISAN model. Given the poor outcomes observed in patients with SAP, the use of individualised predictive nomograms is vital in clinical practice.

Comparison of the Clinical Outcomes Between Reperfusion and Non-Reperfusion Therapy in Elderly Patients with Acute Ischemic Stroke

Purpose

To investigate the benefit (90-day mRS score) and rate of major complications (early symptomatic intracranial hemorrhage-SICH) after reperfusion therapy (RT) (including intravenous thrombolysis -IVT and mechanical thrombectomy -MT) in patients over 80 years with acute ischemic stroke (AIS).

Patients and Methods

AIS patients aged over 80 admitted to Huizhou Central People's Hospital from September 2018 to 2023 were included in this study. Data on SICH, NIHSS, and mRS were analyzed. A good prognosis was defined as a mRS ≤ 2 or recovery to pre-stroke status at 90 days.

Results

Of 209 patients, 80 received non-RT, 100 received IVT and 29 underwent MT. The non-RT group had the lowest baseline NIHSS while the MT group had the highest (non-RT 6.0 vs IVT 12.0 vs MT 18.0, P <0.001). Higher NIHSS was associated with increased SICH risk (OR 1.083, P=0.032), while RT was not (OR 5.194, P=0.129). The overall SICH rate in the RT group was higher but not significantly different after stratification by stroke severity. Poor prognosis was associated with higher admission NIHSS, stroke due to large artery atherosclerosis (LAA) combined with cardioembolism (CE), and stroke-associated pneumonia (SAP) (OR 0.902, P<0.001; OR 0.297, P=0.029; OR 0.103, P<0.001, respectively). The RT group showed a greater reduction in NIHSS (delta NIHSS) than the non-RT group (non-RT 2.0 vs IVT 4.0 vs MT 6.0, P<0.005). For severe AIS, the IVT group had a better prognosis at 90 days (non-RT 0% vs IVT 38.2%, P=0.039). No 90-day mortality difference was found between groups after stratification.

Conclusion

Stroke severity, rather than RT, is an independent risk factor for SICH in AIS patients over 80. RT in severe stroke patients improves NIHSS at 90 days, suggesting RT is safe and effective in this demographic. Further studies with larger samples are required to confirm these findings.

Association Between TCBI (Triglycerides, Total Cholesterol, and Body Weight Index) and Stroke-Associated Pneumonia in Acute Ischemic Stroke Patients

Purpose

Stroke-associated pneumonia (SAP) usually complicates stroke and is linked to adverse prognoses. Triglycerides, total cholesterol, and body weight index (TCBI) is a new and simple calculated nutrition index. This study seeks to investigate the association between TCBI and SAP incidence, along with its predictive value.

Patients and Methods

Nine hundred and sixty-two patients with acute ischemic stroke were divided into SAP group and Non-SAP group. The TCBI was divided into three layers: T1, TCBI < 948.33; T2, TCBI 948.33-1647.15; T3, TCBI > 1647.15. Binary Logistic regression analysis was used to determine the relationship between TCBI levels and the incidence of SAP. Furthermore, restricted cubic splines (RCS) analysis was utilized to evaluate the influence of TCBI on the risk of SAP.

Results

TCBI in the SAP group was markedly lower compared to that in the Non-SAP group (P < 0.001). The Logistic regression model revealed that, using T3 layer as the reference, T1 layer had the highest risk for SAP prevalence (OR = 2.962, 95% CI: 1.600-5.485, P = 0.001), with confounding factors being controlled. The RCS model found that TCBI had a linear relationship with SAP (P for nonlinear = 0.490, P for overall = 0.004). Moreover, incorporating TCBI into the A2DS2 (Age, atrial fibrillation, dysphagia, sex, and severity) model substantially enhanced the initial model's predictive accuracy.

Conclusion

Low TCBI was associated with a higher risk of SAP. In clinical practice, TCBI has shown predictive value for SAP, contributing to early intervention and treatment of SAP.

Machine learning-based survival analysis approaches for predicting the risk of pneumonia post-stroke discharge

BACKGROUND

Post-stroke pneumonia (PSP) is common among stroke patients. PSP occurring after hospital discharge continues to increase the risk of poor functional outcomes and death among stroke survivors. Currently, there is no prediction model specifically designed to predict the occurrence of PSP beyond the acute stage of stroke. This study aimed to explore the use of machine learning (ML) methods in predicting the risk of PSP after hospital discharge.

METHODS

This study analyzed data from 5,754 hospitalized stroke patients. The dataset was randomly divided into a training set and a holdout test set, with a ratio of 80:20. Several clinical and laboratory variables were utilized as predictors and different ML algorithms were employed to model time-to-event data. The ML model's predictive performance was compared to existing risk-scoring systems. A model-agnostic method based on Shapley additive explanations was utilized to interpret the ML model.

RESULTS

The study found that 5.7% of the study patients experienced pneumonia within one year after discharge. Based on repeated 5-fold cross-validation on the training set, the random survival forest (RSF) model had the highest C-index among the various ML algorithms and traditional Cox regression analysis. The final RSF model achieved a C-index of 0.787 (95% confidence interval: 0.737-0.840) on the holdout test set, outperforming five existing risk-scoring systems. The top three important predictors were the Glasgow Coma Scale score, age, and length of hospital stay.

CONCLUSIONS

The RSF model demonstrated superior discriminative ability compared to other ML algorithms and traditional Cox regression analysis, suggesting a non-linear relationship between predictors and outcomes. The developed ML model can be integrated into the hospital information system to provide personalized risk assessments.

Development and validation of a predictive model for stroke associated pneumonia in patients after thrombectomy for acute ischemic stroke

Objective

This study aims to identify the risk factors associated with stroke-associated pneumonia (SAP) in patients who have undergone thrombectomy for acute ischemic stroke and to develop a nomogram chart model for predicting the occurrence of pneumonia.

Methods

Consecutive patients who underwent thrombectomy for acute ischemic stroke were enrolled from three hospitals at Taizhou Enze Medical Center. They were randomly divided into a training group and a validation group in a 7:3 ratio. The training group data was used to screen for effective predictive factors using LASSO regression. Multiple logistic regression was then conducted to determine the predictive factors and construct a nomogram chart. The model was evaluated using the validation group, analyzing its discrimination, calibration, and clinical decision curve. Finally, the newly constructed model was compared with the AIS-APS, A2DS2, ISAN, and PANTHERIS scores for acute ischemic stroke-associated pneumonia.

Results

Out of 913 patients who underwent thrombectomy, 762 were included for analysis, consisting of 473 males and 289 females. The incidence rate of SAP was 45.8%. The new predictive model was constructed based on three main influencing factors: NIHSS ≥16, postoperative LMR, and difficulty swallowing. The model demonstrated good discrimination and calibration. When applying the nomogram chart to threshold probabilities between 7 and 90%, net returns were increased. Furthermore, the AUC was higher compared to other scoring systems.

Conclusion

The constructed nomogram chart in this study outperformed the AIS-APS, A2DS2 score, ISAN score, and PANTHERIS score in predicting the risk of stroke-associated pneumonia in patients with acute ischemic stroke after thrombectomy. It can be utilized for clinical risk prediction of stroke-associated pneumonia in patients after thrombectomy for acute ischemic stroke.

Necrosectomy Versus Stand-Alone Suboccipital Decompressive Craniectomy for the Management of Space-Occupying Cerebellar Infarctions-A Retrospective Multicenter Study

BACKGROUND AND OBJECTIVES

Space-occupying cerebellar stroke (SOCS) when coupled with neurological deterioration represents a neurosurgical emergency. Although current evidence supports surgical intervention in such patients with SOCS and rapid neurological deterioration, the optimal surgical methods/techniques to be applied remain a matter of debate.

METHODS

We conducted a retrospective, multicenter study of patients undergoing surgery for SOCS. Patients were stratified according to the type of surgery as (1) suboccipital decompressive craniectomy (SDC) or (2) suboccipital craniotomy with concurrent necrosectomy. The primary end point examined was functional outcome using the modified Rankin Scale (mRS) at discharge and at 3 months (mRS 0-3 defined as favorable and mRS 4-6 as unfavorable outcome). Secondary end points included the analysis of in-house postoperative complications, mortality, and length of hospitalization.

RESULTS

Ninety-two patients were included in the final analysis: 49 underwent necrosectomy and 43 underwent SDC. Those with necrosectomy displayed significantly higher rate of favorable outcome at discharge as compared with those who underwent SDC alone: 65.3% vs 27.9%, respectively ( P < .001, odds ratios 4.9, 95% CI 2.0-11.8). This difference was also observed at 3 months: 65.3% vs 41.7% ( P = .030, odds ratios 2.7, 95% CI 1.1-6.7). No significant differences were observed in mortality and/or postoperative complications, such as hemorrhagic transformation, infection, and/or the development of cerebrospinal fluid leaks/fistulas.

CONCLUSION

In the setting of SOCS, patients treated with necrosectomy displayed better functional outcomes than those patients who underwent SDC alone. Ultimately, prospective, randomized studies will be needed to confirm this finding.

A clinically actionable and explainable real-time risk assessment framework for stroke-associated pneumonia

The current medical practice is more responsive rather than proactive, despite the widely recognized value of early disease detection, including improving the quality of care and reducing medical costs. One of the cornerstones of early disease detection is clinically actionable predictions, where predictions are expected to be accurate, stable, real-time and interpretable. As an example, we used stroke-associated pneumonia (SAP), setting up a transformer-encoder-based model that analyzes highly heterogeneous electronic health records in real-time. The model was proven accurate and stable on an independent test set. In addition, it issued at least one warning for 98.6 % of SAP patients, and on average, its alerts were ahead of physician diagnoses by 2.71 days. We applied Integrated Gradient to glean the model's reasoning process. Supplementing the risk scores, the model highlighted critical historical events on patients' trajectories, which were shown to have high clinical relevance.

Immune regulation of the gut-brain axis and lung-brain axis involved in ischemic stroke

Local ischemia often causes a series of inflammatory reactions when both brain immune cells and the peripheral immune response are activated. In the human body, the gut and lung are regarded as the key reactional targets that are initiated by brain ischemic attacks. Mucosal microorganisms play an important role in immune regulation and metabolism and affect blood-brain barrier permeability. In addition to the relationship between peripheral organs and central areas and the intestine and lung also interact among each other. Here, we review the molecular and cellular immune mechanisms involved in the pathways of inflammation across the gut-brain axis and lung-brain axis. We found that abnormal intestinal flora, the intestinal microenvironment, lung infection, chronic diseases, and mechanical ventilation can worsen the outcome of ischemic stroke. This review also introduces the influence of the brain on the gut and lungs after stroke, highlighting the bidirectional feedback effect among the gut, lungs, and brain.

Feasibility of a clinical-radiomics combined model to predict the occurrence of stroke-associated pneumonia

PURPOSE

To explore the predictive value of radiomics in predicting stroke-associated pneumonia (SAP) in acute ischemic stroke (AIS) patients and construct a prediction model based on clinical features and DWI-MRI radiomics features.

METHODS

Univariate and multivariate logistic regression analyses were used to identify the independent clinical predictors for SAP. Pearson correlation analysis and the least absolute shrinkage and selection operator with ten-fold cross-validation were used to calculate the radiomics score for each feature and identify the predictive radiomics features for SAP. Multivariate logistic regression was used to combine the predictive radiomics features with the independent clinical predictors. The prediction performance of the SAP models was evaluated using receiver operating characteristics (ROC), calibration curves, decision curve analysis, and subgroup analyses.

RESULTS

Triglycerides, the neutrophil-to-lymphocyte ratio, dysphagia, the National Institutes of Health Stroke Scale (NIHSS) score, and internal carotid artery stenosis were identified as clinically independent risk factors for SAP. The radiomics scores in patients with SAP were generally higher than in patients without SAP (P < 0. 05). There was a linear positive correlation between radiomics scores and NIHSS scores, as well as between radiomics scores and infarct volume. Infarct volume showed moderate performance in predicting the occurrence of SAP, with an AUC of 0.635. When compared with the other models, the combined prediction model achieved the best area under the ROC (AUC) in both training (AUC = 0.859, 95% CI 0.759-0.936) and validation (AUC = 0.830, 95% CI 0.758-0.896) cohorts (P < 0.05). The calibration curves and decision curve analysis further confirmed the clinical value of the nomogram. Subgroup analysis showed that this nomogram had potential generalization ability.

CONCLUSION

The addition of the radiomics features to the clinical model improved the prediction of SAP in AIS patients, which verified its feasibility.

Predictive value of cytokines combined with human neutrophil lipocalinin acute ischemic stroke-associated pneumonia

OBJECTIVE

To explore the predictive value of interleukin-6 (IL-6) combined with human neutrophil lipocalin (HNL) of stroke-associated pneumonia (SAP) in patients who were diagnosed with acute ischemic stroke (AIS).

METHODS

108patients were divided into two groups: pneumonia group (52 cases) and non-pneumonia group (56 cases), according to whether the patients developed SAP within 7 days of admission. General information was compared between the two groups, like age, gender, history of hypertension, diabetes mellitus, cardiovascular disease, dysphagia, smoking and alcoholhistory. Clinical data were recorded and compared, including lipid profile, interleukin-6 (IL-6), homocysteine (Hcy), National Institutes of Health Stroke Scale (NIHSS) score, and HNL. Multivariate Logistic regression analysis was used to screen the risk factors of AIS-AP, and the predictive value of IL-6 and HNL alone and in combination was evaluated by receiver operating characteristic curve (ROC curve).

RESULTS

Logistic regression analysis showed that dysphagia (OR,0.018; 95% CI, 0.001 ~ 0.427; P = 0.013), increased NIHSS scores(OR,0.012; 95% CI, 0.000 ~ 0.434; P = 0.016), and high levels of IL-6 (OR,0.014; 95% CI, 0.000 ~ 0.695; P = 0.032)and HNL (OR,0.006; 95% CI, 0.000 ~ 0.280; P = 0.009) were independent risk factors for SAP with significant difference (all P < 0.05). According to the ROC curve analysis of IL-6, the area under the curve (AUC) was 0.881 (95% CI: 0.820 ~ 0.942), and the optimal cutoff value was 6.89 pg/mL with the sensitivity of 73.1% and specificity of 85.7%. As for the ROC curve analysis of HNL, the AUC was 0.896 (95% CI: 0.839 ~ 0.954), and the best cutoff value was 99.66ng/mL with the sensitivity of 76.9% and specificity of 89.3%. The AUC of the combination of IL-6 and HNL increased to 0.952 (95% CI: 0.914 ~ 0.989), and the sensitivity and specificity increased to 80.8% and 92.9%, respectively.

CONCLUSION

In this research, the levels of IL-6 ≥ 6.89 pg/mL and HNL ≥ 99.66ng/mL were considered as risk factors for AIS patients complicated with SAP. The combined detection had higher predictive value for patients with SAP, which may help to identify who were in highrisk.

Prevalence of Stroke-Associated Pneumonia and Its Predictors Among Hyperglycaemia Patients During Acute Ischemic Stroke

BACKGROUND

Hyperglycaemia (HG) during an acute ischemic stroke (AIS) is not only associated with unfavourable functional outcomes but also associated with stroke-associated pneumonia (SAP). This study aimed to determine the prevalence of SAP among Malaysian patients with AIS and the predictors of SAP among patients with HG during AIS.

METHODS

This is a retrospective cross-sectional study that included patients with AIS admitted to Hospital Sultanah Nur Zahirah, Malaysia from 2017 to 2020. SAP was defined as infection with pneumonia during the first seven days after IS. HG was defined as a blood glucose level > 7.8 mmol/L within 72 h after admission. Patients with SAP were divided into two groups according to HG status. Multivariate logistic regression analysis was performed using SPSS software, version 22 (IBM Corp., Armonk, NY) to identify SAP predictors among patients with HG. Kaplan-Meier log-rank test was used to compare the survival rate from unfavourable functional outcomes between hyperglycaemic patients with and without SAP.

RESULTS

Among 412 patients with AIS, 69 (16.74%) had SAP. The prevalence of SAP among patients with HG and normoglycemia during AIS was 20.98%, and 10.65%, respectively. Age above 60 years, leucocytosis, and National Institute of Health Stroke Scale (NIHSS) > 14 on admission were independent predictors of SAP with aOR of 2.08 (95% CI;1.01-4.30), 2.83 (95% CI; 1.41-5.67), and 3.67 (95% CI; 1.53-8.80), respectively. No significant difference in unfavourable functional outcomes survival was found among patients with and without SAP (p = 0.653).

CONCLUSION

This study demonstrated the prevalence of SAP was higher among patients with HG compared to normoglycemia during AIS. The patient being old, leucocytosis and severe stroke upon admission predict the occurrence of SAP among patients with HG during AIS.

Development and validation of a nomogram for predicting pulmonary infections after Intracerebral hemorrhage in elderly people

OBJECTIVES

The purpose of this study was to develop and validate a nomogram for the prediction of pulmonary infections in elderly patients with intracerebral hemorrhage (ICH) during hospitalization in the intensive care unit (ICU).

METHODS

A total of 1183 elderly patients diagnosed with ICH were included from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database and randomly grouped into training (n=831) and validation (n=352) cohorts. Candidate predictors were identified using the least absolute shrinkage and selection operator (LASSO) regression. Meanwhile, the variables derived from the LASSO regression were included in the multivariate logistic regression analysis, the variables with P < 0.05 were included in the final model and the nomogram was constructed. The discriminatory ability was assessed by plotting the receiver operating curve (ROC) and calculating the area under the curve (AUC). The Performance of the model was assessed by calibration plots and the Hosmer-Lemeshow goodness-of-fit test (HL test). In addition, clinical decision curves assess the net clinical benefit.

RESULTS

The nomogram included chronic lung disease, dysphagia, mechanical ventilation, use of antibiotics, Glasgow Coma Scale (GCS), Logical Organ Dysfunction System (LODS), blood oxygen saturation (SpO2), white blood cell count (WBC) and prothrombin time (PT). The AUC of the predictive model was 0.905 (95 % CI: 0.877, 0.764) in the training cohort and 0.888 (95 % CI: 0.754, 0.838) in the validation cohort, which showed satisfactory discriminative ability. Second, the nomogram showed good calibration. Decision curve analysis showed that the predictive nomogram was clinically useful.

CONCLUSION

A prediction model for predicting pulmonary infections in elderly ICH patients was constructed. The model can help clinicians to identify high-risk patients as soon as possible and prevent the occurrence of pulmonary infections.

Development and validation of a nomogram model for prediction of stroke-associated pneumonia associated with intracerebral hemorrhage

BACKGROUND

We aimed to establish risk factors for stroke-associated pneumonia (SAP) following intracerebral hemorrhage (ICH) and develop an efficient and convenient model to predict SAP in patients with ICH.

METHODS

Our study involved 1333 patients consecutively diagnosed with ICH and admitted to the Neurology Department of the First Affiliated Hospital of Wenzhou Medical University. The 1333 patients were randomly divided (3:1) into the derivation cohort (n = 1000) and validation Cohort (n = 333). Variables were screened from demographics, lifestyle-related factors, comorbidities, clinical symptoms, neuroimaging features, and laboratory tests. In the derivation cohort, we developed a prediction model with multivariable logistic regression analysis. In the validation cohort, we assessed the model performance and compared it to previously reported models. The area under the receiver operating characteristic curve (AUROC), GiViTI calibration belt, net reclassification index (NRI), integrated discrimination index (IDI) and decision curve analysis (DCA) were used to assess the prediction ability and the clinical decision-making ability.

RESULTS

The incidence of SAP was 19.9% and 19.8% in the derivation (n = 1000) and validation (n = 333) cohorts, respectively. We developed a nomogram prediction model including age (Odds Ratio [OR] 1.037, 95% confidence interval [CI] 1.020-1.054), male sex (OR 1.824, 95% CI 1.206-2.757), multilobar involvement (OR 1.851, 95% CI 1.160-2.954), extension into ventricles (OR 2.164, 95% CI 1.456-3.215), dysphagia (OR 3.626, 95% CI 2.297-5.725), disturbance of consciousness (OR 2.113, 95% CI 1.327-3.362) and total muscle strength of the worse side (OR 0.93, 95% CI 0.876-0.987). Compared with previous models, our model was well calibrated and showed significantly higher AUROC, better reclassification ability (improved NRI and IDI) and a positive net benefit for predicted probability thresholds between 10% and 73% in DCA.

CONCLUSIONS

We developed a simple, valid, and clinically useful model to predict SAP following ICH, with better predictive performance than previous models. It might be a promising tool to assess the individual risk of developing SAP for patients with ICH and optimize decision-making.

Does stroke-associated pneumonia play an important role on risk of in-hospital mortality associated with severe stroke? A four-way decomposition analysis of a national cohort of stroke patients

BACKGROUND

Severe strokes and stroke-associated pneumonia (SAP) have long been associated with poorer patient health outcomes, for example, in-hospital mortality. However, it is unclear what role SAP plays in the risk of in-hospital mortality associated with a severe stroke at admission.

METHODS

Using the Sentinel Stroke National Audit Program data on stroke admissions (2013-2018) in England and Wales, we modeled the "total" effect for severe stroke on risk of in-hospital mortality. Through four-way decomposition methodology, we broke down the "total" observed risk into four components. The direct "severity on outcome only" effect, the pure indirect effect of severity mediated via SAP only, the interaction between severity and SAP when mediation is not present, and when mediation via SAP is present.

RESULTS

Of 339,139 stroke patients included, 9.4% had SAP and 15.6% died in hospital. Of SAP patients, 45% died versus 12% of non-SAP patients. The risk ratio for in-hospital mortality associated with severe versus mild/moderate stroke (i.e. total effect) was 4.72 (95% confidence interval: 4.60-4.85). Of this, 43%-increased risk was due to additive SAP interaction, this increased to 50% for "very severe" stroke. The remaining excess relative risk was due to the direct severity on outcome effect only, that is, there was no evidence here for a mediation effect via SAP.

CONCLUSION

SAP was associated with a higher mortality in severe stroke patients. Prioritizing SAP prevention in severe stroke patients may improve in-hospital survival. Our results suggest that in severe stroke patients avoiding SAP might result in an up to 43% reduction in mortality.

From Admission to Discharge: Predicting National Institutes of Health Stroke Scale Progression in Stroke Patients Using Biomarkers and Explainable Machine Learning

As a result of social progress and improved living conditions, which have contributed to a prolonged life expectancy, the prevalence of strokes has increased and has become a significant phenomenon. Despite the available stroke treatment options, patients frequently suffer from significant disability after a stroke. Initial stroke severity is a significant predictor of functional dependence and mortality following an acute stroke. The current study aims to collect and analyze data from the hyperacute and acute phases of stroke, as well as from the medical history of the patients, in order to develop an explainable machine learning model for predicting stroke-related neurological deficits at discharge, as measured by the National Institutes of Health Stroke Scale (NIHSS). More specifically, we approached the data as a binary task problem: improvement of NIHSS progression vs. worsening of NIHSS progression at discharge, using baseline data within the first 72 h. For feature selection, a genetic algorithm was applied. Using various classifiers, we found that the best scores were achieved from the Random Forest (RF) classifier at the 15 most informative biomarkers and parameters for the binary task of the prediction of NIHSS score progression. RF achieved 91.13% accuracy, 91.13% recall, 90.89% precision, 91.00% f1-score, 8.87% FNrate and 4.59% FPrate. Those biomarkers are: age, gender, NIHSS upon admission, intubation, history of hypertension and smoking, the initial diagnosis of hypertension, diabetes, dyslipidemia and atrial fibrillation, high-density lipoprotein (HDL) levels, stroke localization, systolic blood pressure levels, as well as erythrocyte sedimentation rate (ESR) levels upon admission and the onset of respiratory infection. The SHapley Additive exPlanations (SHAP) model interpreted the impact of the selected features on the model output. Our findings suggest that the aforementioned variables may play a significant role in determining stroke patients' NIHSS progression from the time of admission until their discharge.

Post-Stroke Pneumonia in Real-World Practice: Background, Microbiological Examination, and Treatment

Post-stroke pneumonia (PSP) has an impact on acute ischemic stroke (AIS). Although predictive scores for PSP have been developed, it is occasionally difficult to predict. Clarifying how PSP was treated after its onset in clinical practice is important. Admitted patients with AIS over a 2-year period were retrospectively reviewed. Of 281 patients with AIS, 24 (8.5%) developed PSP. The integer-based pneumonia risk score was higher in patients with PSP. The onset of PSP was frequently seen up to the 4th day of hospitalization. Of patients with PSP, sputum examination yielded Geckler 4 or 5 in only 8.3%. Angiotensin-converting enzyme inhibitor (ACE-I) was more frequently administered to patients with PSP; however, all these cases were started with ACE-I following PSP onset. Nasogastric tubes (NGTs) were inserted in 16 of the patients with PSP, of whom 11 were inserted following PSP onset. Multivariate analysis showed that PSP onset was a poor prognostic factor independent of the female sex, urinary tract infection, and National Institutes of Health Stroke Scale. PSP treatment would benefit from the administration of antimicrobials and ACE-I, as well as NGT insertion. To select effective agents for PSP and evaluate the indications for NGT insertion, further case studies are needed.

Systemic Immune-Inflammation Index and Long-Term Mortality in Patients with Stroke-Associated Pneumonia

Background

Systemic immune inflammation has been investigated as a prognostic marker of different diseases. This study is designed to assess the association of systemic immune-inflammation index (SII) with long-term mortality of stroke-associated pneumonia (SAP) patients.

Methods

Patients aged ≥18 years with SAP were selected from the Nanjing Stroke Registry Program in China. We retrospectively evaluated systemic immune-inflammation response with SII and pneumonia severity with the pneumonia severity index and the confusion, uremia, elevated respiratory rate, hypotension, and aged 65 years or older score. To explore the correlation between SII and mortality in SAP patients, multivariable Cox regressions and competing risk regressions were conducted. Mediation analysis was also performed to assess the role of pneumonia severity.

Results

Among 611 patients in the SAP population, death occurred in 164 patients (26.8%) during the median follow-up of 3.0 (1.2-4.6) years. In multivariate analysis, higher SII scores could predict increased mortality in patients with SAP (adjusted hazard ratio 2.061; 95% confidence interval, 1.256-3.383; P = 0.004), and the association was mediated by pneumonia severity. Moreover, adding SII to traditional models improved their predictive ability for mortality.

Conclusion

Our study displayed that SII was characterized in SAP patients with different prognoses. Elevated SII scores increased the risk of mortality. Further research is required for the clinical practice of the index among SAP patients.

A Clinical Prediction Model Based on Post Large Artery Atherosclerosis Infarction Pneumonia

BACKGROUND AND PURPOSE

Stroke-associated pneumonia (SAP) has been found as a common complication in acute ischemic stroke (AIS) patients. Large artery atherosclerosis (LAA) infarct is a major subtype of AIS. This study aimed to build a clinical prediction model for SAP of LAA type AIS patients.

METHODS

This study included 295 patients with LAA type AIS. Univariate analyses and logistic regression analyses were conducted to determine the independent predictors for the modeling purpose. Nomogram used receiver operating characteristics to assess the accuracy of the model, and the calibration plots were employed to assess the fitting degree between the model and the practical scenario. One hundred and five patients were employed for the external validation to test the stability of the model.

RESULTS

From the univariate analysis, patients' ages, neutrophil-to-lymphocyte ratios, National Institute of Health Stroke scale (NIHSS) scores, red blood cell, sex, history of coronary artery disease, stroke location and volume-viscosity swallow test showed statistical difference in the development group for the occurrence of SAP. By incorporating the factors above into a multivariate logistic regression analysis, patients' ages, neutrophil-to-lymphocyte ratios, NIHSS, and volume-viscosity swallow test emerged as the independent risk factors of the development of SAP. The nomogram based on the mentioned 4 variables above achieved a receiver operating characteristic of 0.951 and a validation group of 0.946.

CONCLUSIONS

The proposed nomogram is capable of predicting predict the occurrence of SAP in LAA type AIS patients, and it may identify high-risk patients in time and present information for in-depth treatment.

Preventive ceftriaxone in patients at high risk of stroke-associated pneumonia. A post-hoc analysis of the PASS trial

BACKGROUND

Infections complicate the acute phase of stroke in one third of patients and especially pneumonia is associated with increased risk of death or dependency. In randomized trials of stroke patients, preventive antibiotics reduced overall infections, but did not reduce pneumonia or improve outcome. This may be explained by broad selection criteria, including many patients with a low risk of pneumonia. To assess the potential of selection of patients at high risk of pneumonia, we performed a post-hoc analysis in the Preventive Antibiotics in Stroke Study (PASS).

METHODS

PASS was a multicentre phase 3 trial in acute stroke patients who were randomized to preventive ceftriaxone for four days within 24 hours or standard care. For this analysis patients were divided based on the ISAN risk score for pneumonia as follows: low (0-6), medium (7-14) and high (15-21). Primary outcomes were pneumonia rate during admission as judged by the treating physician, and by an independent committee; secondary outcomes were overall infections and unfavorable outcome (modified Rankin Scale ≥3). We adjusted with multivariable regression for possible confounders: age, stroke subtype and severity, pre-stroke dependency and diabetes.

RESULTS

Pneumonia occurred more frequently in higher risk groups (25.7% (high), 9.0% (medium) 1.5%, (low)). The absolute difference in pneumonia rate between patients treated with ceftriaxone or standard care increased with the ISAN score (low: 0.5%, medium: 1.2%, high: 10.1%). After adjustment ceftriaxone reduced overall infections in the low and medium groups, not in the high-risk group. There was a trend towards reduction of pneumonia as judged by the committee (3.7% vs 13.6%, aOR = 0.164, p = 0.063) in the high-risk group.

CONCLUSIONS

This post-hoc analysis of PASS confirmed higher rates of pneumonia with higher ISAN scores, and suggests that in acute stroke patients with an ISAN score of ≥15, preventive ceftriaxone for four days may reduce pneumonia rate.

A Novel Index in the Prediction of Pneumonia Following Acute Ischemic Stroke

BACKGROUND

The aim of our study was to search for predictive factors and to develop a model (index) for the risk of pneumonia following acute ischemic stroke.

MATERIAL AND METHODS

This study is an analysis of prospectively collected data from the neurology department of a district general hospital in Poland, comprising 1001 patients suffering from an acute ischemic stroke. Based on the medical data, the formula for the prediction of pneumonia was calculated.

RESULTS

Multivariate assessment for pneumonia occurrence was performed using the new PNEUMOINDEX. The study showed a significant increase in pneumonia risk with an increasing PNEUMOINDEX (OR non-adjusted = 2.738, p < 0.001). After accounting for age and comorbidities as confounders, the effect of the Index on pneumonia changed marginally (OR = 2.636, p < 0.001).

CONCLUSIONS

This study presents factors that show a significant association with the occurrence of pneumonia in patients with acute ischemic stroke. The calculated PNEUMOINDEX consists of data obtained at admission, namely NYHA III and IV heart failure, COPD, generalized atherosclerosis, NIHHS score on admission, and CRP/Hgb ratio, and shows high prediction accuracy in predicting hospital-acquired pneumonia in ischemic stroke patients.

Low triiodothyronine syndrome is associated with stroke-associated pneumonia

OBJECTIVE

Stroke-associated pneumonia (SAP) is the most common early consequence in patients suffering from an acute ischaemic stroke (AIS). The purpose of this study was to explore the possible relationship between low triiodothyronine (T3) syndrome and SAP in stroke patients.

METHODS

This study recruited 2460 consecutive AIS patients. SAP was defined according to the modified Centers for Disease Control and Prevention criteria for hospital-acquired pneumonia. The thyroid hormones levels were measured within 24 h after admission. Low T3 syndrome was characterized as T3 below the lower limit of the reference interval accompanied by normal TSH levels.

RESULTS

Among the total patients, 336 (13.7%) patients were diagnosed with SAP. SAP in individuals with low T3 syndrome was substantially greater (p < .001) as compared to those without low T3 syndrome. After adjusting for possible confounders, low T3 syndrome (adjusted odds ratio [aOR] = 1.59; 95% confidence interval [CI], 1.20-2.09; p = .001) remained significant in our logistic model. Patients with low T3 syndrome had a higher risk of severe SAP (aOR = 2.17, 95% confidence interval [CI] 1.38-3.44; p = .001).

CONCLUSION

Low T3 syndrome, independent of recognized risk factors, is a possible risk factor for in-hospital SAP, which can help clinicians in the early detection and treatment of high-risk patients.

Patient's care bundle benefits to prevent stroke associated pneumonia: A meta-analysis with trial sequential analysis

Background

Patient's care bundle has been found to have a beneficial effect on refractory diseases, but the preventive effect of this strategy on stroke-associated pneumonia (SAP) remains unclear. The purpose of this meta-analysis was to determine the role of the patient's care bundle in the prevention of SAP.

Methods

A systematic search was conducted in five electronic databases to identify randomized controlled trials (RCTs) published before January 31, 2022. The incidence of SAP and aspiration and the length of hospital stay were assessed. Random pair-wise meta-analysis was conducted using Review Manager 5.4, and trial sequential analysis (TSA) was also performed.

Results

Twenty eligible RCTs involving 1916 patients were included for data analysis. Pooled results suggested that patient's care bundle was associated with significantly lower incidence of SAP (risk ratio [RR], 0.37; 95% CI, 0.29-0.46; p < 0.001; I2 = 0%) and aspiration (RR, 0.23; 95% CI, 0.15-0.35; p < 0.001; I2 = 0%). Meanwhile, patient's care bundle also significantly shortened the length of hospital stay for general patients (mean difference [MD], -3.10; 95% CI, -3.83 to -2.37; p < 0.001; I2 = 16%) and the length of intensive care unit (ICU) stay for patients with severe stoke (MD, -4.85; 95% CI, -5.86-3.84; p < 0.001; I2 = 0%). Results of TSA confirmed that none of the findings could be significantly reversed by future studies.

Conclusions

The patient's care bundle effectively prevents the occurrence of SAP and aspiration and shortens the hospital stay of stroke patients. However, it is necessary to design more high-quality studies to further validate our findings and investigate their applicability in other geographical regions.

A nomogram based on nutritional status and A2DS2 score for predicting stroke-associated pneumonia in acute ischemic stroke patients with type 2 diabetes mellitus: A retrospective study

Background

Stroke-associated pneumonia (SAP) commonly complicates acute ischemic stroke (AIS) and significantly worsens outcomes. Type 2 diabetes mellitus (T2DM) may contribute to malnutrition, impair innate immunity function, and increase the probability of SAP occurrence in AIS patients. We aimed to determine early predictors of SAP in AIS patients with T2DM and to construct a nomogram specifically for predicting SAP in this population by combining the A²DS² score with available nutrition-related parameters.

Methods

A total of 1,330 consecutive AIS patients with T2DM were retrospectively recruited. The patients were randomly allocated to the training (n = 887) and validation groups (n = 443). Univariate and multivariate binary logistic regression analyses were applied to determine the predictors of SAP in the training group. A nomogram was established according to the identified predictors. The areas under the receiver operating characteristic curve (AUROC) and calibration plots were performed to access the predictive values of the nomogram. The decision curve was applied to evaluate the net benefits of the nomogram.

Results

The incidence of SAP was 9% and 9.7% in the training and validation groups, respectively. The results revealed that the A²DS² score, stroke classification, Geriatric Nutritional Risk Index, hemoglobin, and fast blood glucose were independent predictors for SAP. A novel nomogram, A²DS²-Nutrition, was constructed based on these five predictors. The AUROC for A²DS²-Nutrition (0.820, 95% CI: 0.794–0.845) was higher than the A²DS² score (0.691, 95% CI: 0.660–0.722) in the training group. Similarly, it showed a better predictive performance than the A²DS² score [AUROC = 0.864 (95% CI: 0.828–0.894) vs. AUROC = 0.763 (95% CI: 0.720–0.801)] in the validation group. These results were well calibrated in the two groups. Moreover, the decision curve revealed that the A²DS²-Nutrition provided an additional net benefit to the AIS patients with T2DM compared to the A²DS² score in both groups.

Conclusion

The A²DS² score, stroke classification, Geriatric Nutritional Risk Index, hemoglobin, and fast blood glucose were independent predictors for SAP in AIS patients with T2DM. Thus, the proposed A²DS²-Nutrition may be a simple and reliable prediction model for SAP occurrence in AIS patients with T2DM.

Comparison of six risk scores for stroke-associated pneumonia in patients with acute ischemic stroke: A systematic review and Bayesian network meta-analysis

Background

Stroke-associated pneumonia (SAP) is one of the major causes of death after suffering a stroke. Several scoring systems have been developed for the early prediction of SAP. However, it is unclear which scoring system is more suitable as a risk prediction tool. We performed this Bayesian network meta-analysis to compare the prediction accuracy of these scoring systems.

Methods

Seven databases were searched from their inception up to April 8, 2022. The risk of bias assessment of included study was evaluated by the QUADAS-C tool. Then, a Bayesian network meta-analysis (NMA) was performed by R 4.1.3 and STATA 17.0 software. The surface under the cumulative ranking curve (SUCRA) probability values were applied to rank the examined scoring systems.

Results

A total of 20 cohort studies involving 42,236 participants were included in this analysis. The results of the NMA showed that AIS-APS had excellent performance in prediction accuracy for SAP than Chumbler (MD = 0.030, 95%CI: 0.004, 0.054), A2DS2 (MD = 0.041, 95% CI: 0.023, 0.059), ISAN (MD = 0.045, 95% CI: 0.022, 0.069), Kwon (MD = 0.077, 95% CI: 0.055, 0.099) and PANTHERIS (MD = 0.082, 95% CI: 0.049, 0.114). Based on SUCRA values, AIS-APS (SUCRA: 99.8%) ranked the highest.

Conclusion

In conclusion, the study found that the AIS-APS is a validated clinical tool for predicting SAP after the onset of acute ischemic stroke.

Systematic review registration

https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=292375, identifier: CRD42021292375.

Application of machine learning and natural language processing for predicting stroke-associated pneumonia

Background

Identifying patients at high risk of stroke-associated pneumonia (SAP) may permit targeting potential interventions to reduce its incidence. We aimed to explore the functionality of machine learning (ML) and natural language processing techniques on structured data and unstructured clinical text to predict SAP by comparing it to conventional risk scores.

Methods

Linked data between a hospital stroke registry and a deidentified research-based database including electronic health records and administrative claims data was used. Natural language processing was applied to extract textual features from clinical notes. The random forest algorithm was used to build ML models. The predictive performance of ML models was compared with the A²DS², ISAN, PNA, and ACDD⁴ scores using the area under the receiver operating characteristic curve (AUC).

Results

Among 5,913 acute stroke patients hospitalized between Oct 2010 and Sep 2021, 450 (7.6%) developed SAP within the first 7 days after stroke onset. The ML model based on both textual features and structured variables had the highest AUC [0.840, 95% confidence interval (CI) 0.806-0.875], significantly higher than those of the ML model based on structured variables alone (0.828, 95% CI 0.793-0.863, P = 0.040), ACDD⁴ (0.807, 95% CI 0.766-0.849, P = 0.041), A²DS² (0.803, 95% CI 0.762-0.845, P = 0.013), ISAN (0.795, 95% CI 0.752-0.837, P = 0.009), and PNA (0.778, 95% CI 0.735-0.822, P < 0.001). All models demonstrated adequate calibration except for the A²DS² score.

Conclusions

The ML model based on both textural features and structured variables performed better than conventional risk scores in predicting SAP. The workflow used to generate ML prediction models can be disseminated for local adaptation by individual healthcare organizations.

Reduced plasma levels of RGM-A predict stroke-associated pneumonia in patients with acute ischemic stroke: A prospective clinical study

Background

Stroke-induced immunodepression syndrome is considered the major etiology of stroke-associated pneumonia (SAP). Repulsive guidance molecule A (RGM-A) is an immunomodulatory protein that is closely related to inflammation and immune responses. To explore the relationship between RGM-A and SAP and facilitate the early identification of patients at high risk of developing SAP, we investigated the predictive value of RGM-A in SAP.

Methods

We enrolled 178 patients with acute ischemic stroke (AIS) and finally analyzed 150 patients, among whom 69 had SAP and 81 had non-SAP. During the same period, 40 patients with community-acquired pneumonia and 40 healthy participants were included as controls. SAP was defined according to the modified US Centers for Disease Control and Prevention criteria. Blood samples were collected at 24 h, 48 h, 3 days, 4 to 7 days, and 8 to 14 days after stroke onset. An enzyme-linked immunosorbent assay was used to detect the plasma levels of RGM-A and interleukin-6.

Results

The plasma RGM-A levels were significantly decreased in both patients with community-acquired pneumonia and those with AIS, and the decline was most pronounced in patients with SAP (P < 0.001). RGM-A started to decline within 24 h after stroke in the SAP group, and the lowest levels were detected on day 3 and days 4 to 7 (P < 0.001). The RGM-A levels in the SAP group were lower than those in the non-SAP group at all blood collection time points (P < 0.05). In the logistic regression analyses, RGM-A was a protective factor for SAP after adjusting for confounders (adjusted odds ratio = 0.22, 95% confidence interval = 0.091–0.538, P = 0.001). Receiver operating characteristic curve analysis showed that the area under the curve for RGM-A was 0.766 (0.091–0.538; P = 0.001), the cutoff value was 4.881 ng/mL, and the sensitivity and specificity were 80.00 and 76.36%, respectively.

Conclusions

We demonstrated that reduced plasma levels of RGM-A might help in the early identification of high-risk patients with SAP and predict the occurrence of SAP in patients with AIS. RGM-A might provide new clues to a potential alternative therapy for SAP.

Novel machine learning models to predict pneumonia events in supratentorial intracerebral hemorrhage populations: An analysis of the Risa-MIS-ICH study

Background

Stroke-associated pneumonia (SAP) contributes to high mortality rates in spontaneous intracerebral hemorrhage (sICH) populations. Accurate prediction and early intervention of SAP are associated with prognosis. None of the previously developed predictive scoring systems are widely accepted. We aimed to derive and validate novel supervised machine learning (ML) models to predict SAP events in supratentorial sICH populations.

Methods

The data of eligible supratentorial sICH individuals were extracted from the Risa-MIS-ICH database and split into training, internal validation, and external validation datasets. The primary outcome was SAP during hospitalization. Univariate and multivariate analyses were used for variable filtering, and logistic regression (LR), Gaussian naïve Bayes (GNB), random forest (RF), K-nearest neighbor (KNN), support vector machine (SVM), extreme gradient boosting (XGB), and ensemble soft voting model (ESVM) were adopted for ML model derivations. The accuracy, sensitivity, specificity, and area under the curve (AUC) were adopted to evaluate the predictive value of each model with internal/cross-/external validations.

Results

A total of 468 individuals with sICH were included in this work. Six independent variables [nasogastric feeding, airway support, unconscious onset, surgery for external ventricular drainage (EVD), larger sICH volume, and intensive care unit (ICU) stay] for SAP were identified and selected for ML prediction model derivations and validations. The internal and cross-validations revealed the superior and robust performance of the GNB model with the highest AUC value (0.861, 95% CI: 0.793–0.930), while the LR model had the highest AUC value (0.867, 95% CI: 0.812–0.923) in external validation. The ESVM method combining the other six methods had moderate but robust abilities in both cross-validation and external validation and achieved an AUC of 0.843 (95% CI: 0.784–0.902) in external validation.

Conclusion

The ML models could effectively predict SAP in sICH populations, and our novel ensemble model demonstrated reliable robust performance outcomes despite the populational and algorithmic differences. This attempt indicated that ML application may benefit in the early identification of SAP.

Braden scale for predicting pneumonia after spontaneous intracerebral hemorrhage

OBJECTIVE

Stroke-associated pneumonia is an infection that commonly occurs in patients with spontaneous intracerebral hemorrhage and causes serious burdens. In this study, we evaluated the validity of the Braden scale for predicting stroke-associated pneumonia after spontaneous intracerebral hemorrhage.

METHODS

Patients with spontaneous intracerebral hemorrhage were retrospectively included and divided into pneumonia and no pneumonia groups. The admission clinical characteristics and Braden scale scores at 24 h after admission were collected and compared between the two groups. Receiver operating characteristic curve analysis was performed to assess the predictive validity of the Braden scale. Multivariable analysis was conducted to identify the independent risk factors associated with pneumonia after intracerebral hemorrhage.

RESULTS

A total of 629 intracerebral hemorrhage patients were included, 150 (23.8%) of whom developed stroke-associated pneumonia. Significant differences were found in age and fasting blood glucose levels between the two groups. The mean score on the Braden scale in the pneumonia group was 14.1±2.4, which was significantly lower than that in the no pneumonia group (16.5±2.6), p<0.001. The area under the curve for the Braden scale for the prediction of pneumonia after intracerebral hemorrhage was 0.760 (95%CI 0.717-0.804). When the cutoff point was 15 points, the sensitivity was 74.3%, the specificity was 64.7%, the accuracy was 72.0%, and the Youden's index was 39.0%. Multivariable analysis showed that a lower Braden scale score (OR 0.696; 95%CI 0.631-0.768; p<0.001) was an independent risk factor associated with stroke-associated pneumonia after intracerebral hemorrhage.

CONCLUSION

The Braden scale, with a cutoff point of 15 points, is moderately valid for predicting stroke-associated pneumonia after spontaneous intracerebral hemorrhage.

15-Hydroxyprostaglandin Dehydrogenase Is a Predictor of Stroke-Associated Pneumonia

Background and Purpose

Stroke is a serious fatal and disabling disease. Stroke-associated pneumonia (SAP) is the most common complication of stroke, which may further aggravate the stroke. The prevention and early prediction of SAP is a key clinical strategy. 15-hydroxyprostaglandin dehydrogenase (15-PGDH) is involved in pneumonia, while its relationship with SAP has yet to be determined. Therefore, we investigated the predictive value of 15-PGDH for SAP and visualized their relationship.

Methods

Stroke patients were recruited and divided into SAP group and Non-SAP group. Baseline demographic and clinical data were obtained from the medical record system, blood samples were collected to detect relevant variables and 15-PGDH levels. Patient characteristics were compared with a t-test. Binary logistic regression analysis was performed to determine the predictive value of 15-PGDH for SAP. Restricted cubic splines (RCS) were performed to visualize the relationship between 15-PGDH and SAP risk. Finally, the SAP patient characteristics between the severe group and mild group were compared.

Results

50 patients were enrolled and divided into SAP group (n = 26) and Non-SAP group (n = 24). 15-PGDH in the SAP group was lower than that in the Non-SAP group (0.258 ± 0.275 vs. 0.784 ± 0.615, p = 0.025). Binary logistic regression analysis revealed that the lower 15-PGDH, the higher the risk of SAP (OR = 0.04, 95%CI, 0.010–0.157, p < 0.001). The RCS model showed the L-shaped relationship between 15-PGDH and SAP.

Conclusions

In stroke patients, serum 15-PGDH is a valuable biomarker for predicting SAP. There is an L-shaped relationship between the level of 15-PGDH and the risk of SAP.

Application of machine learning in liver transplantation: a review

BACKGROUND

Machine learning (ML) has been increasingly applied in the health-care and liver transplant setting. The demand for liver transplantation continues to expand on an international scale, and with advanced aging and complex comorbidities, many challenges throughout the transplantation decision-making process must be better addressed. There exist massive datasets with hidden, non-linear relationships between demographic, clinical, laboratory, genetic, and imaging parameters that conventional methods fail to capitalize on when reviewing their predictive potential. Pre-transplant challenges include addressing efficacies of liver segmentation, hepatic steatosis assessment, and graft allocation. Post-transplant applications include predicting patient survival, graft rejection and failure, and post-operative morbidity risk.

AIM

In this review, we describe a comprehensive summary of ML applications in liver transplantation including the clinical context and how to overcome challenges for clinical implementation.

METHODS

Twenty-nine articles were identified from Ovid MEDLINE, MEDLINE Epub Ahead of Print and In-Process and Other Non-Indexed Citations, Embase, Cochrane Database of Systematic Reviews, and Cochrane Central Register of Controlled Trials.

CONCLUSION

ML is vastly interrogated in liver transplantation with promising applications in pre- and post-transplant settings. Although challenges exist including site-specific training requirements, the demand for more multi-center studies, and optimization hurdles for clinical interpretability, the powerful potential of ML merits further exploration to enhance patient care.

ICH-LR2S2: a new risk score for predicting stroke-associated pneumonia from spontaneous intracerebral hemorrhage

Purpose

We develop a new risk score to predict patients with stroke-associated pneumonia (SAP) who have an acute intracranial hemorrhage (ICH).

Method

We applied logistic regression to develop a new risk score called ICH-LR2S2. It was derived from examining a dataset of 70,540 ICH patients between 2015 and 2018 from the Chinese Stroke Center Alliance (CSCA). During the training of ICH-LR2S2, patients were randomly divided into two groups – 80% for the training set and 20% for model validation. A prospective test set was developed using 12,523 patients recruited in 2019. To further verify its effectiveness, we tested ICH-LR2S2 on an external dataset of 24,860 patients from the China National Stroke Registration Management System II (CNSR II). The performance of ICH-LR2S2 was measured by the area under the receiver operating characteristic curve (AUROC).

Results

The incidence of SAP in the dataset was 25.52%. A 24-point ICH-LR2S2 was developed from independent predictors, including age, modified Rankin Scale, fasting blood glucose, National Institutes of Health Stroke Scale admission score, Glasgow Coma Scale score, C-reactive protein, dysphagia, Chronic Obstructive Pulmonary Disease, and current smoking. The results showed that ICH-LR2S2 achieved an AUC = 0.749 [95% CI 0.739–0.759], which outperforms the best baseline ICH-APS (AUC = 0.704) [95% CI 0.694–0.714]. Compared with the previous ICH risk scores, ICH-LR2S2 incorporates fasting blood glucose and C-reactive protein, improving its discriminative ability. Machine learning methods such as XGboost (AUC = 0.772) [95% CI 0.762–0.782] can further improve our prediction performance. It also performed well when further validated by the external independent cohort of patients (n = 24,860), ICH-LR2S2 AUC = 0.784 [95% CI 0.774–0.794].

Conclusion

ICH-LR2S2 accurately distinguishes SAP patients based on easily available clinical features. It can help identify high-risk patients in the early stages of diseases.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-022-03389-5.

Preventive antibiotic therapy in acute stroke patients: A systematic review and meta-analysis of individual patient data of randomized controlled trials

Introduction

Infection after stroke is associated with unfavorable outcome. Randomized controlled studies did not show benefit of preventive antibiotics in stroke but lacked power for subgroup analyses. Aim of this study is to assess whether preventive antibiotic therapy after stroke improves functional outcome for specific patient groups in an individual patient data meta-analysis.

Patients and methods

We searched MEDLINE (1946–7 May 2021), Embase (1947–7 May 2021), CENTRAL (17th September 2021), trial registries, cross-checked references and contacted researchers for randomized controlled trials of preventive antibiotic therapy versus placebo or standard care in ischemic or hemorrhagic stroke patients. Meta-analysis was performed by a one-step and two-step approach. Primary outcome was functional outcome adjusted for age and stroke severity. Secondary outcomes were infections and mortality.

Results

4197 patients from nine trials were included. Preventive antibiotic therapy was not associated with a shift in functional outcome (mRS) at 3 months (OR1.13, 95%CI 0.98–1.31) or unfavorable functional outcome (mRS 3–6) (OR0.85, 95%CI 0.60–1.19). Preventive antibiotics did not improve functional outcome in pre-defined subgroups (age, stroke severity, timing and type of antibiotic therapy, pneumonia prediction scores, dysphagia, type of stroke, and type of trial). Preventive antibiotics reduced infections (276/2066 (13.4%) in the preventive antibiotic group vs. 417/2059 (20.3%) in the control group, OR 0.60, 95% CI 0.51–0.71, p < 0.001), but not pneumonia (191/2066 (9.2%) in the preventive antibiotic group vs. 205/2061 (9.9%) in the control group (OR 0.92 (0.75–1.14), p = 0.450).

Discussion and conclusion

Preventive antibiotic therapy did not benefit any subgroup of patients with acute stroke and currently cannot be recommended.

Association of Platelet-to-Lymphocyte Ratio with Stroke-Associated Pneumonia in Acute Ischemic Stroke

A common consequence of acute ischemic stroke (AIS), stroke-associated pneumonia (SAP), might result in a poor prognosis after stroke. Based on the critical position of inflammation in SAP, this study aimed to explore the correlation between platelet-to-lymphocyte ratio (PLR) and the occurrence of SAP. We included 295 patients with acute ischemic stroke, 40 with SAP, and 255 without SAP. The area under the receiver operating characteristic curve was used to determine the diagnostic value of SAP risk factors using binary logistic regression analysis. The comparison between the two groups showed that age, the baseline National Institutes of Health Stroke Scale (NIHSS) score, and the proportion of dysphagia, atrial fibrillation, and total anterior circulation infarct were higher, and the proportion of lacunar circulation infarct was lower in the SAP group (P < 0.001). In terms of laboratory data, the SAP group had considerably greater neutrophil counts and PLR, while the non-SAP group (P < 0.001) had significantly lower lymphocyte counts and triglycerides. Binary logistic regression analysis revealed that older age (aOR = 1.062, 95% CI: 1.023–1.102, P = 0.002), atrial fibrillation (aOR = 3.585, 95% CI: 1.605–8.007, P = 0.019), and PLR (aOR = 1.003, 95% CI: 1.001–1.006, P = 0.020) were independent risk factors associated with SAP after adjusting for potential confounders. The sensitivity and specificity of PLR with a cutoff value of 152.22 (AUC: 0.663, 95% CI: 0.606–0.717, P = 0.0006) were 57.5% and 70.6%, respectively. This study showed that high PLR is an associated factor for SAP in AIS patients. Increased systemic inflammation is linked to SAP in ischemic stroke. Inflammatory biomarkers that are easily accessible may aid in the diagnosis of high-risk SAP patients.

Low serum transthyretin levels predict stroke-associated pneumonia

BACKGROUND AND AIMS

Stroke-associated pneumonia (SAP) is commonly seen in ischemic stroke patients. Low transthyretin levels are found to be correlated with stroke. This study aims to investigate the potential relationship between transthyretin levels and SAP.

METHODS AND RESULTS

In total, 920 patients were involved in our study. Serum transthyretin levels were measured within 24 h at admission. We defined SAP according to the modified Centers for Disease Control criteria. In the study population, 123 (13.4%, 77 men, 46 women) were diagnosed with SAP. In the multivariable analysis, we found that serum transthyretin levels were significantly lower in SAP compared with non-SAP patients (231 ± 80 vs. 279 ± 75; P < 0.001) after adjusting for confounders. Meanwhile, we discovered that low transthyretin levels (≤252 mg/L) were independently associated with the development of SAP (OR 3.370; 95% CI: 1.763-6.441; P < 0.001). Moreover, patients with SAP had a worse clinical outcome than those without SAP at discharge. In addition, dysphagia, leukocyte count and NLR (neutrophil-to-lymphocyte ratio) were also found to be associated with SAP.

CONCLUSION

We found that low transthyretin levels significantly increased the risk of SAP. Patients with high risk of developing SAP could be early identified and prevented timely.

Predicting Poststroke Pneumonia in Patients With Anterior Large Vessel Occlusion: A Prospective, Population-Based Stroke Registry Analysis

Objective

To assess predictive factors for poststroke pneumonia (PSP) in patients with acute ischemic stroke (AIS) due to large vessel occlusion (LVO) of the anterior circulation, with special regard to the impact of intravenous thrombolysis (IVT) and endovascular treatment (EVT) on the risk of PSP. As a secondary goal, the validity of the A²DS², PNEUMONIA, and ISAN scores in LVO will be determined.

Methods

Analysis was based on consecutive data for the years 2017 to 2019 from the prospective inpatient stroke registry covering the entire federal state of Hesse, Germany, using the Kruskal-Wallis test and binary logistic regression.

Results

Data from 4,281 patients with LVO were included in the analysis (54.8% female, median age = 78 years, range = 18–102), of whom 66.4% (n = 2,843) received recanalization therapy (RCT). In total, 19.4% (n = 832) of all LVO patients developed PSP. Development of PSP was associated with an increase in overall in-hospital mortality of 32.1% compared with LVO patients without PSP (16.4%; p < 0.001). Incidence of PSP was increased in 2132 patients with either EVT (n = 928; 25.9% PSP incidence) or combined EVT plus IVT (n = 1,204; 24.1%), compared with 2,149 patients with IVT alone (n = 711; 15.2%) or conservative treatment only (n = 1,438; 13.5%; p < 0.001). Multivariate analysis identified EVT (OR 1.5) and combined EVT plus IVT (OR 1.5) as significant independent risk factors for PSP. Furthermore, male sex (OR 1.9), age ≥ 65 years (OR 1.7), dysphagia (OR 3.2) as well as impaired consciousness at arrival (OR 1.7) and the comorbidities diabetes (OR 1.4) and atrial fibrillation (OR 1.3) were significantly associated risk factors (each p < 0.001). Minor stroke (NIHSS ≤ 4) was associated with a significant lower risk of PSP (OR 0.5). Performance of risk stratification scores varied between A²DS² (96.1% sensitivity, 20.7% specificity), PNEUMONIA (78.2% sensitivity and 45.1% specificity) and ISAN score (98.0% sensitivity, 20.0% specificity).

Conclusion

Nearly one in five stroke patients with LVO develops PSP during acute care. This risk of PSP is further increased if an EVT is performed. Other predictive factors are consistent with those previously described for all AIS patients. Available risk stratification scores proved to be sensitive tools in LVO patients but lack specificity.

A2DS2 Score Combined With Clinical and Neuroimaging Factors Better Predicts Stroke-Associated Pneumonia in Hyperacute Cerebral Infarction

Objective

To investigate the predictors of stroke-associated pneumonia (SAP) and poor functional outcome in patients with hyperacute cerebral infarction (HCI) by combining clinical factors, laboratory tests and neuroimaging features.

Methods

We included 205 patients with HCI from November 2018 to December 2019. The diagnostic criterion for SAP was occurrence within 7 days of the onset of stroke. Poor outcome was defined as a functional outcome based on a 3-months MRS score >3. The relationship of demographic, laboratory and neuroimaging variables with SAP and poor outcome was investigated using univariate and multivariate analyses.

Results

Fifty seven (27.8%) patients were diagnosed with SAP and 40 (19.5%) developed poor outcomes. A²DS² score (OR = 1.284; 95% CI: 1.048–1.574; P = 0.016), previous stroke (OR = 2.630; 95% CI: 1.122–6.163; P = 0.026), consciousness (OR = 2.945; 95% CI: 1.514–5.729; P < 0.001), brain atrophy (OR = 1.427; 95% CI: 1.040–1.959; P = 0.028), and core infarct volume (OR = 1.715; 95% CI: 1.163–2.528; P = 0.006) were independently associated with the occurrence of SAP. Therefore, we combined these variables into a new SAP prediction model with the C-statistic of 0.84 (95% CI: 0.78–0.90). Fasting plasma glucose (OR = 1.404; 95% CI: 1.202–1.640; P < 0.001), NIHSS score (OR = 1.088; 95% CI: 1.010–1.172; P = 0.026), previous stroke (OR = 4.333; 95% CI: 1.645–11.418; P = 0.003), SAP (OR = 3.420; 95% CI: 1.332–8.787; P = 0.011), basal ganglia-dilated perivascular spaces (BG-dPVS) (OR = 2.124; 95% CI: 1.313–3.436; P = 0.002), and core infarct volume (OR = 1.680; 95% CI: 1.166–2.420; P = 0.005) were independently associated with poor outcome. The C-statistic of the outcome model was 0.87 (95% CI: 0.81–0.94). Furthermore, the SAP model significantly improved discrimination and net benefit more than the A²DS² scale, with a C-statistic of 0.76 (95% CI: 0.69–0.83).

Conclusions

After the addition of neuroimaging features, the models exhibit good differentiation and calibration for the prediction of the occurrence of SAP and the development of poor outcomes in HCI patients. The SAP model could better predict the SAP, representing a helpful and valid tool to obtain a net benefit compared with the A²DS² scale.

High Fibrinogen to Albumin Ratio: A Novel Marker for Risk of Stroke-Associated Pneumonia?

Background: Stroke-associated pneumonia (SAP) is associated with poor prognosis after acute ischemic stroke (AIS).

Purpose: This study aimed to describe the parameters of coagulation function and evaluate the association between the fibrinogen-to-albumin ratio (FAR) and SAP in patients with AIS.

Patients and methods: A total of 932 consecutive patients with AIS were included. Coagulation parameters were measured at admission. All patients were classified into two groups according to the optimal cutoff FAR point at which the sum of the specificity and sensitivity was highest. Propensity score matching (PSM) was performed to balance potential confounding factors. Univariate and multivariate logistic regression analyses were applied to identify predictors of SAP.

Results: A total of 100 (10.7%) patients were diagnosed with SAP. The data showed that fibrinogen, FAR, and D-dimer, prothrombin time (PT), activated partial thromboplastin time (aPTT) were higher in patients with SAP, while albumin was much lower. Patients with SAP showed a significantly increased FAR when compared with non-SAP (P < 0.001). Patients were assigned to groups of high FAR (≥0.0977) and low FAR (<0.0977) based on the optimal cut-off value. Propensity score matching analysis further confirmed the association between FAR and SAP. After adjusting for confounding and risk factors, multivariate regression analysis showed that the high FAR (≥0.0977) was an independent variable predicting the occurrence of SAP (odds ratio =2.830, 95% CI = 1.654–4.840, P < 0.001). In addition, the FAR was higher in the severe pneumonia group when it was assessed by pneumonia severity index (P = 0.008).

Conclusions: High FAR is an independent potential risk factor of SAP, which can help clinicians identify high-risk patients with SAP after AIS.

Stroke-associated pneumonia: A bibliometric analysis of worldwide trends from 2003 to 2020

Stroke-associated pneumonia (SAP) is a spectrum of pulmonary infections in patients within 7 days of stroke. Which is one of the most common complications after stroke and is significantly associated with a poor prognosis of stroke. To the best of our knowledge, a bibliometric method was not previously used to analyze the topic of SAP; we aim to describe the situation and evolution of SAP from 2003 to 2020, and to discuss the research hotspots and frontiers.A total of 151 articles were retrieved from the Scopus database. Bibliometric analysis was used to explore the dynamic trends of articles and the top subject areas, journals, institutes, citations, and co-keywords. VOS viewer software (version 1.6.15) was used to graphically map the hot topics of SAP based on the co-keywords.A total of 151 articles were identified. Articles have increased over the recent years and faster in the last 2 years (55 articles, 36.4%), the majority of subject areas are medicine (124 articles, 82.1%) and neuroscience (38 articles, 25.2%). The "Journal Of Stroke And Cerebrovascular Diseases" with 15 articles has been scored as the first rank followed by "Plos One." Regarding the geographical distribution of articles, China is the most productive country with 50 articles (33.1%), others are more prominent in Europe, and most institutes are universities. Citations have increased over time, the main country of the top five highly cited published articles are Germany and before 2008. The co-keywords are mainly divided into four aspects: risk factors, predictive scores, preventions, and outcomes.This study could provide practical sources for researchers to find the top subject areas, journals, institutes, citations, and co-keywords. Moreover, the study could pave the way for researchers to be engaged in studies potentially lead to more articles in this field.

Monocyte-to-lymphocyte ratio as a predictor of stroke-associated pneumonia: A retrospective study-based investigation

BACKGROUND AND PURPOSE

Early prediction of stroke-associated pneumonia (SAP) is significant in clinical practice, as it is frequently challenging due to delays in typical clinical manifestations and radiological changes. The monocyte-to-lymphocyte ratio (MLR) has been proposed as an indicator of systemic inflammation and infection. However, none of these studies have focused on the predictive value of the MLR for SAP. We investigated the predictive value of MLR for SAP and investigated its relationship with disease severity.

METHODS

In this retrospective study, we assessed 399 consecutive patients with acute stroke. SAP was defined according to the modified Centers for Disease Control and Prevention criteria. The severity of pneumonia was rated using the pneumonia severity index (PSI). MLR was calculated by dividing absolute monocyte counts by absolute lymphocyte counts.

RESULTS

Among all the patients, SAP occurred in 116 patients (29.1%). White blood cell (WBC), neutrophil, monocyte, and MLR levels in the SAP group were higher than those in the non-SAP group, while lymphocyte levels were lower (p < .05). Multivariable regression analysis revealed that the MLR (OR = 7.177; 95% CI = 1.190-43.292, p = .032) remained significant after adjusting for confounders. The ROC curve showed that the AUC value of MLR for SAP was 0.779, the optimal cutoff value of MLR for SAP was 0.388, with a specificity of 64.7% and sensitivity of 81.3%. The MLR levels were significantly higher in the severe pneumonia group when assessed by PSI (p = .024) than in the mild group. The AUC value of MLR was 0.622 (95% CI = 0.520-0.724, p = .024) in the severe pneumonia group. The optimal cutoff value of MLR was 0.750, with a specificity of 91.0% and a sensitivity of 33.0%.

CONCLUSIONS

Our study shows that a high MLR is an independent risk factor for SAP and has a predictive value for severe pneumonia in patients with SAP.

Prediction of stroke-associated pneumonia by the A2DS2, AIS-APS, and ISAN scores: a systematic review and meta-analysis

Background: Different scoring systems (A2DS2, AISAPS, ISAN) have been designed to predict the risk of in-hospital stroke-associated pneumonia (SAP). Studies have assessed the accuracy of these scores for predicting SAP. We performed this meta-analysis to consolidate the evidence on the predictive accuracies for SAP of the A2DS2, AISAPS, and ISAN scores.Materials and methods: We conducted a systematic search for all studies reporting the SAP predictive accuracy of A2DS2, AISAPS, or ISAN scores in the databases of PubMed Central, SCOPUS, MEDLINE, Embase, and Cochrane from inception until December 2020. We used the STATA software for the meta-analysis.Results: We included 19 studies with 35 849 patients. The pooled score sensitivities were 78% (95% CI, 71%-83%) for A2DS2, 79% (95% CI, 77%-81%) for AISAPS, and 79% (95% CI, 77%-81%) for ISAN. The pooled score specificities were 73% (95% CI, 65%-80%) for A2DS2, 74% (95% CI, 69%-79%) for AISAPS, and 74% (95% CI, 69%-79%) for ISAN. We found significant heterogeneity for all the scoring systems based on the chi-square test results and an I² statistic > 75%. We performed meta-regression to explore the source of heterogeneity and found that patient selection (p< 0.05) and reference standards (p< 0.05) in the sensitivity model, index test standards (p< 0.05), flow and timing of tests (p< 0.01) in the specificity model, and mean age (p < 0.001) in the joint model were the source of heterogeneity.Conclusions: To summarize, we found that A2S2, AISAPS and ISAN have moderate predictive accuracy for SAP with A2S2 having a stable cutoff value.

Usefulness of the Neutrophil-to-Lymphocyte Ratio as a Predictor of Pneumonia and Urinary Tract Infection Within the First Week After Acute Ischemic Stroke

Background: A high Neutrophil-to-Lymphocyte ratio (NLR) in patients with acute ischemic stroke (AIS) has been associated with post-stroke infections, but it's role as an early predictive biomarker for post-stroke pneumonia (PSP) and urinary tract infection (UTI) is not clear.

Aim: To investigate the usefulness of NLR obtained within 24 h after AIS for predicting PSP and UTI in the first week.

Methods: Clinical and laboratory data were retrieved from the University Hospital Brussels stroke database/electronic record system. Patients were divided into those who developed PSP or UTI within the first week after stroke onset and those who didn't. Receiver operating characteristics (ROC) curves and logistic regression analysis were used to identify independent predictors.

Results: Five hundred and fourteen patients were included, of which 15.4% (n = 79) developed PSP and 22% (n = 115) UTI. In univariate analysis, NLR was significantly higher in patients who developed PSP (4.1 vs. 2.8, p < 0.001) but not in those who developed UTI (3.3 vs. 2.9, p = 0.074). Multiple logistic regression analysis for PSP showed that NLR, male gender, dysphagia, and stroke severity measured by the National Institutes of Health Stroke Scale (NIHSS), were independent predictors of PSP. For NLR alone, the area under the curve (AUC) in the ROC curve was 0.66 (95% CI = 0.59–0.73). When combining NLR ≥ 4.7 with age >75 years, male gender, NIHSS > 7, and dysphagia, the AUC increased to 0.84 (95% CI = 0.79–0.89).

Conclusion: The NLR within 24 h after AIS appears to have no predictive value for post-stroke UTI, and is only a weak predictor for identifying patients at high risk for PSP. Its predictive value for PSP appears to be much stronger when incorporated in a prediction model including age, gender, NIHSS score, and dysphagia.

Use of Pulmonary Computed Tomography for Evaluating Suspected Stroke-Associated Pneumonia

OBJECTIVES

Accurate and timely diagnosis of pneumonia complicating stroke remains challenging and the diagnostic accuracy of chest X-ray (CXR) in the setting of stroke-associated pneumonia (SAP) is uncertain. The overall objective of this study was to evaluate the use of pulmonary computed tomography (CT) in diagnosis of suspected SAP.

MATERIALS AND METHODS

Patients with acute ischemic stroke (IS) or intracerebral hemorrhage (ICH) were recruited within 24h of clinically suspected SAP and underwent non-contrast pulmonary CT within 48h of antibiotic initiation. CXR and pulmonary CT were reported by two radiologists. Pulmonary CT was used as the reference standard for final diagnosis of SAP. Sensitivity, specificity, positive and negative predictive values (PPV and NPV), and diagnostic odds ratio (OR) for CXR were calculated.

RESULTS

40 patients (36 IS, 4 ICH) with a median age of 78y (range 44y-90y) and a median National Institute of Health Stroke Scale score of 13 (range 3-31) were included. All patients had at least one CXR and 35/40 patients (88%) underwent pulmonary CT. Changes consistent with pneumonia were present in 15/40 CXRs (38%) and 12/35 pulmonary CTs (34%). 9/35 pulmonary CTs (26%) were reported normal. CXR had a sensitivity of 58.3%, specificity of 73.9%, PPV of 53.8 %, NPV of 77.2 %, diagnostic OR of 3.7 (95% CI 0.7 - 22) and an accuracy of 68.5% (95% CI 50.7% -83.1%).

DISCUSSION

CXR has limited diagnostic accuracy in SAP. The majority of patients started on antibiotics had no evidence of pneumonia on pulmonary CT with potential implications for antibiotic stewardship.

CONCLUSIONS

Pulmonary CT could be applied as a reference standard for evaluation of clinical and biomarker diagnostic SAP algorithms in multi-center studies.

Variation of stroke-associated pneumonia in stroke units across England and Wales: A registry-based cohort study

Background

Pneumonia is common in stroke patients and is associated with worse clinical outcomes. Prevalence of stroke-associated pneumonia varies between studies, and reasons for this variation remain unclear. We aimed to describe the variation of observed stroke-associated pneumonia in England and Wales and explore the influence of patient baseline characteristics on this variation.

Methods

Patient data were obtained from the Sentinel Stroke National Audit Programme for all confirmed strokes between 1 April 2013 and 31 December 2018. Stroke-associated pneumonia was defined by new antibiotic initiation for pneumonia within the first seven days of admission. The probability of stroke-associated pneumonia occurrence within stroke units was estimated and compared using a multilevel mixed model with and without adjustment for patient-level characteristics at admission.

Results

Of the 413,133 patients included, median National Institutes of Health Stroke Scale was 4 (IQR: 2–10) and 42.3% were aged over 80 years. Stroke-associated pneumonia was identified in 8.5% of patients. The median within stroke unit stroke-associated pneumonia prevalence was 8.5% (IQR: 6.1–11.5%) with a maximum of 21.4%. The mean and variance of the predicted stroke-associated pneumonia probability across stroke units decreased from 0.08 (0.68) to 0.05 (0.63) when adjusting for patient admission characteristics. This difference in the variance suggests that clinical characteristics account for 5% of the observed variation in stroke-associated pneumonia between units.

Conclusions

Patient-level clinical characteristics contributed minimally to the observed variation of stroke-associated pneumonia between stroke units. Additional explanations for the observed variation in stroke-associated pneumonia need to be explored which could reduce variation in antibiotic use for stroke patients.

Impact of a learning health system on acute care and medical complications after intracerebral hemorrhage

INTRODUCTION

Patients with stroke often experience pneumonia during the acute stage after stroke onset. Oral care may be effective in reducing the risk of stroke-associated pneumonia (SAP). We aimed to determine the changes in oral care, as well as the incidence of SAP, in patients with intracerebral hemorrhage, following implementation of a learning health system in our hospital.

METHODS

We retrospectively analyzed the data of 1716 patients with intracerebral hemorrhage who were hospitalized at a single stroke center in Japan between January 2012 and December 2018. Data were stratified on the basis of three periods of evolving oral care: period A, during which conventional, empirically driven oral care was provided (n = 725); period B, during which standardized oral care was introduced, with SAP prophylaxis based on known risk factors (n = 469); and period C, during which oral care was risk-appropriate based on learning health system data (n = 522). Logistic regression analysis was performed to evaluate associations between each of the three treatment approaches and the risk of SAP.

RESULTS

Among the included patients, the mean age was 71.3 ± 13.6 years; 52.6% of patients were men. During the course of each period, the frequency of oral care within 24 hours of admission increased (P < .001), as did the adherence rate to oral care ≥3 times per day (P < .001). After adjustment for confounding factors, a change in the risk of SAP was not observed in period B; however, the risk significantly decreased in period C (odds ratio 0.61; 95% confidence interval 0.43-0.87) compared with period A. These associations were maintained for SAP diagnosed using strict clinical criteria or after exclusion of 174 patients who underwent neurosurgical treatment.

CONCLUSIONS

Risk-appropriate care informed by the use of learning health system data could improve care and potentially reduce the risk of SAP in patients with intracerebral hemorrhage in the acute stage.

Development and performance assessment of novel machine learning models to predict pneumonia after liver transplantation

Background

Pneumonia is the most frequently encountered postoperative pulmonary complications (PPC) after orthotopic liver transplantation (OLT), which cause high morbidity and mortality rates. We aimed to develop a model to predict postoperative pneumonia in OLT patients using machine learning (ML) methods.

Methods

Data of 786 adult patients underwent OLT at the Third Affiliated Hospital of Sun Yat-sen University from January 2015 to September 2019 was retrospectively extracted from electronic medical records and randomly subdivided into a training set and a testing set. With the training set, six ML models including logistic regression (LR), support vector machine (SVM), random forest (RF), adaptive boosting (AdaBoost), extreme gradient boosting (XGBoost) and gradient boosting machine (GBM) were developed. These models were assessed by the area under curve (AUC) of receiver operating characteristic on the testing set. The related risk factors and outcomes of pneumonia were also probed based on the chosen model.

Results

591 OLT patients were eventually included and 253 (42.81%) were diagnosed with postoperative pneumonia, which was associated with increased postoperative hospitalization and mortality (P < 0.05). Among the six ML models, XGBoost model performed best. The AUC of XGBoost model on the testing set was 0.734 (sensitivity: 52.6%; specificity: 77.5%). Pneumonia was notably associated with 14 items features: INR, HCT, PLT, ALB, ALT, FIB, WBC, PT, serum Na⁺, TBIL, anesthesia time, preoperative length of stay, total fluid transfusion and operation time.

Conclusion

Our study firstly demonstrated that the XGBoost model with 14 common variables might predict postoperative pneumonia in OLT patients.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12931-021-01690-3.

Blood Biomarker Panels for the Early Prediction of Stroke-Associated Complications

Background Acute decompensated heart failure (ADHF) and respiratory tract infections (RTIs) are potentially life-threatening complications in patients experiencing stroke during hospitalization. We aimed to test whether blood biomarker panels might predict these complications early after admission. Methods and Results Nine hundred thirty-eight patients experiencing ischemic stroke were prospectively recruited in the Stroke-Chip study. Post-stroke complications during hospitalization were retrospectively evaluated. Blood samples were drawn within 6 hours after stroke onset, and 14 biomarkers were analyzed by immunoassays. Biomarker values were normalized using log-transformation and Z score. PanelomiX algorithm was used to select panels with the best accuracy for predicting ADHF and RTI. Logistic regression models were constructed with the clinical variables and the biomarker panels. The additional predictive value of the panels compared with the clinical model alone was evaluated by receiver operating characteristic curves. An internal validation through a 10-fold cross-validation with 3 repeats was performed. ADHF and RTI occurred in 19 (2%) and 86 (9.1%) cases, respectively. Three-biomarker panels were developed as predictors: vascular adhesion protein-1 >5.67, NT-proBNP (N-terminal pro-B-type natriuretic peptide) >4.98 and d-dimer >5.38 (sensitivity, 89.5%; specificity, 71.7%) for ADHF; and interleukin-6 >3.97, von Willebrand factor >3.67, and d-dimer >4.58 (sensitivity, 82.6%; specificity, 59.8%) for RTI. Both panels independently predicted stroke complications (panel for ADHF: odds ratio [OR] [95% CI], 10.1 [3-52.2]; panel for RTI: OR, 3.73 [1.95-7.14]) after adjustment by clinical confounders. The addition of the panel to clinical predictors significantly improved areas under the curve of the receiver operating characteristic curves in both cases. Conclusions Blood biomarkers could be useful for the early prediction of ADHF and RTI. Future studies should assess the usefulness of these panels in front of patients experiencing stroke with respiratory symptoms such as dyspnea.