-
The American Journal of Cardiology Nov 2023Heart failure (HF) is often categorized by left ventricular (LV) ejection fraction (LVEF). A new category of HF characterized by supra-normal LVEF (>65%), named HF with... (Review)
Review
Heart failure (HF) is often categorized by left ventricular (LV) ejection fraction (LVEF). A new category of HF characterized by supra-normal LVEF (>65%), named HF with supra-normal ejection fraction (HFsnEF), has been recently proposed. Some studies reported that patients with supra-normal LVEF might have an increased risk of long-term major adverse cardiovascular events and U-shaped mortality patterns. Currently, the prognosis of HFsnEF is not well established but seems to be associated with an increased risk of long-term major adverse cardiovascular events. It has been reported that HFsnEF is more prevalent in women and is associated with higher prevalence of nonischemic HF, higher blood urea nitrogen plasma levels, lower levels of natriuretic peptides, and to be less likely treated with β blockers. The pathophysiology of HFsnEF would be associated with microvascular dysfunction because of microvascular inflammation or reduced coronary flow reserve, and low stroke volume index with smaller cardiac chamber dimensions and concentric LV geometry. In this study, we systematically reviewed published data on patients with s supra-normal LV function and reported its definition, proposed pathophysiology, phenotypes, diagnostic strategy, and prognosis.
Topics: Humans; Female; Ventricular Function, Left; Stroke Volume; Heart Failure; Prognosis; Inflammation; Ventricular Dysfunction, Left
PubMed: 37734305
DOI: 10.1016/j.amjcard.2023.08.169 -
Artificial intelligence for diagnostic and prognostic neuroimaging in dementia: A systematic review.Alzheimer's & Dementia : the Journal of... Dec 2023Artificial intelligence (AI) and neuroimaging offer new opportunities for diagnosis and prognosis of dementia. (Review)
Review
INTRODUCTION
Artificial intelligence (AI) and neuroimaging offer new opportunities for diagnosis and prognosis of dementia.
METHODS
We systematically reviewed studies reporting AI for neuroimaging in diagnosis and/or prognosis of cognitive neurodegenerative diseases.
RESULTS
A total of 255 studies were identified. Most studies relied on the Alzheimer's Disease Neuroimaging Initiative dataset. Algorithmic classifiers were the most commonly used AI method (48%) and discriminative models performed best for differentiating Alzheimer's disease from controls. The accuracy of algorithms varied with the patient cohort, imaging modalities, and stratifiers used. Few studies performed validation in an independent cohort.
DISCUSSION
The literature has several methodological limitations including lack of sufficient algorithm development descriptions and standard definitions. We make recommendations to improve model validation including addressing key clinical questions, providing sufficient description of AI methods and validating findings in independent datasets. Collaborative approaches between experts in AI and medicine will help achieve the promising potential of AI tools in practice.
HIGHLIGHTS
There has been a rapid expansion in the use of machine learning for diagnosis and prognosis in neurodegenerative disease Most studies (71%) relied on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset with no other individual dataset used more than five times There has been a recent rise in the use of more complex discriminative models (e.g., neural networks) that performed better than other classifiers for classification of AD vs healthy controls We make recommendations to address methodological considerations, addressing key clinical questions, and validation We also make recommendations for the field more broadly to standardize outcome measures, address gaps in the literature, and monitor sources of bias.
Topics: Humans; Alzheimer Disease; Prognosis; Artificial Intelligence; Neurodegenerative Diseases; Brain; Neuroimaging
PubMed: 37563912
DOI: 10.1002/alz.13412 -
Critical Care (London, England) Nov 2023Bacteria are the main pathogens that cause sepsis. The pathogenic mechanisms of sepsis caused by gram-negative and gram-positive bacteria are completely different, and... (Meta-Analysis)
Meta-Analysis
BACKGROUND
Bacteria are the main pathogens that cause sepsis. The pathogenic mechanisms of sepsis caused by gram-negative and gram-positive bacteria are completely different, and their prognostic differences in sepsis remain unclear.
METHODS
The PubMed, Web of Science, Cochrane Library, and Embase databases were searched for Chinese and English studies (January 2003 to September 2023). Observational studies involving gram-negative (G (-))/gram-positive (G (+)) bacterial infection and the prognosis of sepsis were included. The stability of the results was evaluated by sensitivity analysis. Funnel plots and Egger tests were used to check whether there was publication bias. A meta-regression analysis was conducted on the results with high heterogeneity to identify the source of heterogeneity. A total of 6949 articles were retrieved from the database, and 45 studies involving 5586 subjects were included after screening according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Twenty-seven high-quality studies and 18 moderate-quality studies were identified according to the Newcastle‒Ottawa Scale score. There was no significant difference in the survival rate of sepsis caused by G (-) bacteria and G (+) bacteria (OR 0.95, 95% CI 0.70-1.28). Subgroup analysis according to survival follow-up time showed no significant difference. The serum concentrations of C-reactive protein (CRP) (SMD = 0.39, 95% CI 0.02-0.76), procalcitonin (SMD = 1.95, 95% CI 1.32-2.59) and tumor necrosis factor-alpha (TNF-α) (MD = 0.31, 95% CI 0.25-0.38) in the G (-) bacterial infection group were significantly higher than those in the G (+) bacterial infection group, but there was no significant difference in IL-6 (SMD = 1.33, 95% CI - 0.18-2.84) and WBC count (MD = - 0.15, 95% CI - 0.96-00.66). There were no significant differences between G (-) and G (+) bacteria in D dimer level, activated partial thromboplastin time, thrombin time, international normalized ratio, platelet count, length of stay or length of ICU stay. Sensitivity analysis of the above results indicated that the results were stable.
CONCLUSION
The incidence of severe sepsis and the concentrations of inflammatory factors (CRP, PCT, TNF-α) in sepsis caused by G (-) bacteria were higher than those caused by G (+) bacteria. The two groups had no significant difference in survival rate, coagulation function, or hospital stay. The study was registered with PROSPERO (registration number: CRD42023465051).
Topics: Humans; Prognosis; Tumor Necrosis Factor-alpha; Sepsis; Bacterial Infections; Gram-Negative Bacteria; C-Reactive Protein; Bacteria; Gram-Positive Bacteria
PubMed: 38037118
DOI: 10.1186/s13054-023-04750-w -
Breast Cancer (Tokyo, Japan) Nov 2023HER2-low breast cancer (BC) is proposed to be a special population of patients with an immunohistochemistry (IHC) score of 1 + or 2 + and non-amplified in situ... (Meta-Analysis)
Meta-Analysis
BACKGROUND
HER2-low breast cancer (BC) is proposed to be a special population of patients with an immunohistochemistry (IHC) score of 1 + or 2 + and non-amplified in situ hybridization (ISH) results. The role and prognostic impact of HER2-low BC is still controversial. This meta-analysis aims to explore the prognostic difference between of HER2-low and HER2-zero characteristic in BC patients.
METHODS
A meta-analysis was conducted using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and eligible studies were search in PubMed, Web of Science and EMBASE databases. Quality assessment of included studies were performed by Quality in Prognostic Studies (QUIPS) tool. Hazard ratios (HRs) and corresponding 95% confidence interval (CI) for overall survival (OS) and disease-free survival (DFS) were pooled in a meta-analysis. Furthermore, subgroup analysis, sensitivity analysis, and analysis for publication bias were conducted.
RESULTS
Eighteen studies comprising a total of 93,317 patients were included for meta-analysis. BC patients with HER2-low characteristic have longer OS (HRs 0.87, 95% CI 0.81-0.93, p < 0.0001) and DFS (HRs 0.82, 95% CI 0.73-0.93, p = 0.001) compared to those with HER2-zero characteristic. Subgroup analysis indicate that the source of heterogeneity may come from the hormone receptor (HR) status group. Although, the publication bias was detected, sensitivity analysis and the trim-and-fill method analysis demonstrated the stability and reliability of the results.
CONCLUSION
HER2-low BC patients have longer OS and DFS compared to HER2-zero BC patients, and its prognostic value is consistent among different HR status patients. Whether HER2-low breast cancer is an independent subtype of breast cancer is still a subject of ongoing research, and more studies are needed to fully understand the molecular and clinical features of this subtype.
Topics: Humans; Female; Breast Neoplasms; Reproducibility of Results; Prognosis; Proportional Hazards Models; Disease-Free Survival
PubMed: 37470943
DOI: 10.1007/s12282-023-01487-w -
World Journal of Emergency Surgery :... Dec 2023To assess the efficacy of artificial intelligence (AI) models in diagnosing and prognosticating acute appendicitis (AA) in adult patients compared to traditional... (Review)
Review
BACKGROUND
To assess the efficacy of artificial intelligence (AI) models in diagnosing and prognosticating acute appendicitis (AA) in adult patients compared to traditional methods. AA is a common cause of emergency department visits and abdominal surgeries. It is typically diagnosed through clinical assessments, laboratory tests, and imaging studies. However, traditional diagnostic methods can be time-consuming and inaccurate. Machine learning models have shown promise in improving diagnostic accuracy and predicting outcomes.
MAIN BODY
A systematic review following the PRISMA guidelines was conducted, searching PubMed, Embase, Scopus, and Web of Science databases. Studies were evaluated for risk of bias using the Prediction Model Risk of Bias Assessment Tool. Data points extracted included model type, input features, validation strategies, and key performance metrics.
RESULTS
In total, 29 studies were analyzed, out of which 21 focused on diagnosis, seven on prognosis, and one on both. Artificial neural networks (ANNs) were the most commonly employed algorithm for diagnosis. Both ANN and logistic regression were also widely used for categorizing types of AA. ANNs showed high performance in most cases, with accuracy rates often exceeding 80% and AUC values peaking at 0.985. The models also demonstrated promising results in predicting postoperative outcomes such as sepsis risk and ICU admission. Risk of bias was identified in a majority of studies, with selection bias and lack of internal validation being the most common issues.
CONCLUSION
AI algorithms demonstrate significant promise in diagnosing and prognosticating AA, often surpassing traditional methods and clinical scores such as the Alvarado scoring system in terms of speed and accuracy.
Topics: Adult; Humans; Artificial Intelligence; Appendicitis; Prognosis; Algorithms; Machine Learning; Acute Disease
PubMed: 38114983
DOI: 10.1186/s13017-023-00527-2 -
Psycho-oncology Nov 2023To evaluate the evidence base for patient, oncological, and treatment prognostic factors associated with multiple mental wellbeing outcomes in prostate cancer patients. (Meta-Analysis)
Meta-Analysis Review
OBJECTIVES
To evaluate the evidence base for patient, oncological, and treatment prognostic factors associated with multiple mental wellbeing outcomes in prostate cancer patients.
METHODS
We performed a literature search of MEDLINE, EMBASE, and CINAHL databases including studies evaluating patient, oncological, or treatment factors against one of five mental wellbeing outcomes; depression, anxiety, fear of cancer recurrence, masculinity, and body image perception. Data synthesis included a random effects meta-analysis for the prognostic effect of individual factors if sufficient homogenous data was available, with a structured narrative synthesis where this was not possible.
RESULTS
A final 62 articles were included. Older age was associated with a reducing odds of depression (OR 0.97, p = 0.04), with little evidence of effect for other outcomes. Additionally, baseline mental health status was related to depression and increasing time since diagnosis was associated with reducing fear of recurrence, albeith with low certainty of evidence. However, few other patient or oncological factors demonstrated any coherent relationship with any wellbeing outcome. Androgen deprivation therapy was associated with increased depression (HR 1.65, 95% CI 1.41-1.92, p < 0.01) and anxiety, however, little difference was seen between other treatment options. Overall, whilst numerous factors were identified, most were evaluated by single studies with few evaluating masculinity and body image outcomes.
CONCLUSION
We highlight the existing evidence for prognostic factors in mental wellbeing outcomes in prostate cancer, allowing us to consider high-risk groups of patients for preventative and treatment measures. However, the current evidence is heterogenous with further work required exploring less conclusive factors and outcomes.
Topics: Male; Humans; Prostatic Neoplasms; Depression; Prognosis; Androgen Antagonists; Neoplasm Recurrence, Local; Quality of Life
PubMed: 37789603
DOI: 10.1002/pon.6225 -
Expert Review of Respiratory Medicine 2023To investigate the diagnostic and prognostic value of angiopoietin-2 (Ang-2) for acute respiratory distress syndrome (ARDS). (Meta-Analysis)
Meta-Analysis Review
BACKGROUND
To investigate the diagnostic and prognostic value of angiopoietin-2 (Ang-2) for acute respiratory distress syndrome (ARDS).
METHODS
Seven databases (4 English and 3 Chinese databases) were searched, the quality was evaluated by QUADAS-2 and GRADE profile. The bivariate model was employed to combine area under the curve (AUC), pooled sensitivity (pSEN) and pooled specificity (pSPE), the Fagan's nomogram was employed for evaluating clinical utility. This study was registered in PROSPERO (NO.CRD42022371488).
RESULTS
18 eligible studies comprising 27 datasets (12 diagnostic and 15 prognostic datasets) were included for meta-analysis. For diagnostic analysis, Ang-2 yielded an AUC of 0.82, with a pSEN of 0.78 and a pSPE of 0.74; in clinical utility analysis, a pretest probability of 50% regulated the post probability positive (PPP) of 75% and the post probability negative (PPN) of 23%. In prognostic analysis, Ang-2 yielded an AUC of 0.83, with a pSEN of 0.69, a pSPE of 0.81, and good clinical utility (a pretest probability of 50% regulated the PPP of 79% and the PPN of 28%). Heterogeneity existed in both diagnostic and prognostic analysis.
CONCLUSIONS
Ang-2 demonstrates promising diagnostic and prognostic capabilities as a noninvasive circulating biomarker for ARDS, especially in the Chinese population. It is advisable to dynamically monitor Ang-2 in critically ill patients both suspected and with confirmed ARDS.
Topics: Humans; Angiopoietin-2; Biomarkers; Critical Illness; Prognosis; Respiratory Distress Syndrome
PubMed: 37366084
DOI: 10.1080/17476348.2023.2230883 -
International Journal of Medical... Aug 2023As diagnostic and prognostic models developed by traditional statistics perform poorly in real-world, artificial intelligence (AI) and Big Data (BD) may improve the... (Review)
Review
BACKGROUND
As diagnostic and prognostic models developed by traditional statistics perform poorly in real-world, artificial intelligence (AI) and Big Data (BD) may improve the supply chain of heart transplantation (HTx), allocation opportunities, correct treatments, and finally optimize HTx outcome. We explored available studies, and discussed opportunities and limits of medical application of AI to the field of HTx.
METHOD
A systematic overview of studies published up to December 31st, 2022, in English on peer-revied journals, have been identified through PUBMED-MEDLINE-WEB of Science, referring to HTx, AI, BD. Studies were grouped in 4 domains based on main studies' objectives and results: etiology, diagnosis, prognosis, treatment. A systematic attempt was made to evaluate studies by the Prediction model Risk Of Bias ASsessment Tool (PROBAST) and the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD).
RESULTS
Among the 27 publications selected, none used AI applied to BD. Of the selected studies, 4 fell in the domain of etiology, 6 in the domain of diagnosis, 3 in the domain of treatment, and 17 in that of prognosis, as AI was most frequently used for algorithmic prediction and discrimination of survival, but in retrospective cohorts and registries. AI-based algorithms appeared superior to probabilistic functions to predict patterns, but external validation was rarely employed. Indeed, based on PROBAST, selected studies showed, to some extent, significant risk of bias (especially in the domain of predictors and analysis). In addition, as example of applicability in the real-world, a free-use prediction algorithm developed through AI failed to predict 1-year mortality post-HTx in cases from our center.
CONCLUSIONS
While AI-based prognostic and diagnostic functions performed better than those developed by traditional statistics, risk of bias, lack of external validation, and relatively poor applicability, may affect AI-based tools. More unbiased research with high quality BD meant for AI, transparency and external validations, are needed to have medical AI as a systematic aid to clinical decision making in HTx.
Topics: Humans; Artificial Intelligence; Big Data; Heart Transplantation; Prognosis; Retrospective Studies
PubMed: 37285695
DOI: 10.1016/j.ijmedinf.2023.105110 -
Clinical Biochemistry Nov 2023Sudden sensorineural hearing loss (SSNHL) is defined as hearing loss of more than 30 dB in less than 72 h. SSNHL is a frequent complaint and an emergency in... (Meta-Analysis)
Meta-Analysis Review
Sudden sensorineural hearing loss (SSNHL) is defined as hearing loss of more than 30 dB in less than 72 h. SSNHL is a frequent complaint and an emergency in otolaryngology. Various biomarkers have been used to determine the prognosis of SSNHL. This systematic review and meta-analysis aims to evaluate the relationship between the different biomarkers and the prognosis of SSNHL. We searched English-language literature up to October 2022 in four databases, including PubMed, Google Scholar, Cochrane, and Science Direct. This search was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. This study was reported in the International Prospective Register of Systematic Reviews (PROSPERO) database (ID = CRD42022369538). All studies examining the role of neutrophil to lymphocyte ratio (NLR) concluded that higher NLR is associated with a worse prognosis. The results of studies regarding the relationship between platelet to lymphocyte ratio (PLR) and tumor necrosis factor (TNF) are controversial. Other factors shown to be associated with SSNHL include Glycated hemoglobin (HbA1C), blood glucose, iron levels, serum endocan, salusin-beta, and bone turnover biomarkers. This meta-analysis showed that PLR, NLR, and neutrophils were significantly different between recovered and non-recovered patients. PLR, NLR, and neutrophil count are reliable tools to assess the prognosis of patients with SSNHL.
Topics: Humans; Biomarkers; Hearing Loss, Sensorineural; Hearing Loss, Sudden; Lymphocytes; Neutrophils; Prognosis
PubMed: 37944628
DOI: 10.1016/j.clinbiochem.2023.110684 -
Kidney360 Aug 2023The Oxford classification of IgA nephropathy defined five features scored subjectively in renal biopsies, identified by the initials MESTC. Two large studies with...
KEY POINTS
The Oxford classification of IgA nephropathy defined five features scored subjectively in renal biopsies, identified by the initials MESTC. Two large studies with independent observers showed reproducibility was moderate for T, moderate or poor for M and S, and poor for E and C. In multivariate analyses including clinical features, T was related to 58% of outcomes, with no correlation of MESTC with 24% of outcomes.
BACKGROUND
The Oxford classification of IgA nephropathy defined five prognostic features scored subjectively in renal biopsies: mesangial cellularity (M), endocapillary hypercellularity (E), segmental sclerosis (S), interstitial fibrosis/tubular atrophy (T), and (fibro)cellular crescents (C). Pathological scoring systems should be reproducible and have prognostic value independently of clinical features. Reproducibility of the classification was not previously investigated in a systematic review, and the most recent systematic reviews of prognostic value were in 2017.
METHODS
This systematic review followed PRISMA 2020 guidelines. MEDLINE, PUBMED, and EMBASE databases were searched using the terms “IgA nephropathy” and “Oxford.” Eligible papers applied the classification and mentioned statistical analysis of interobserver reproducibility and/or included multivariate analysis of outcomes related to individual Oxford scores and clinical features, including treatment with corticosteroids or other immunosuppressive drugs.
RESULTS
There were 99 suitable papers before September 23, 2022. Of 12 papers that mentioned reproducibility, only six reported statistics for MEST/MESTC scoring. Four of these were small studies and/or had observers at the same institution. These were considered less representative of application of the classification than two large studies with independent observers, in which agreement was moderate for T, either moderate or poor for M and S, and poor for E and C. In 92 papers with 125 multivariate analyses of various outcomes, the commonest Oxford element associated with outcomes was T (73 of 125, 58%), with no correlation of any element with outcomes in 30 analyses (24%). Treatment with immunosuppression was often related to scores, particularly C and E, without consistent relations between Oxford scores and outcomes in immunosuppressed patients.
CONCLUSIONS
This systematic review showed limitations of the Oxford classification in practice, particularly the moderate or poor reproducibility of scores. T was the Oxford score most often related to clinical outcomes, but even this was not consistently reliable as a prognostic indicator.
Topics: Humans; Glomerulonephritis, IGA; Prognosis; Reproducibility of Results; Kidney; Glomerular Filtration Rate
PubMed: 37357346
DOI: 10.34067/KID.0000000000000195