Adjusting for clinical covariates improves the ability of B-type natriuretic peptide to distinguish cardiac from non-cardiac dyspnoea: a sub-study of HEARD-IT

Identification of hub genes in heart failure by integrated bioinformatics analysis and machine learning

Objective

To screen feature genes of heart failure patients through machine learning methods, in order to identify characteristic genes driving heart failure and investigate the progression of heart failure.

Methods

Heart failure patient samples were downloaded from the public database GEO (Gene Expression Omnibus), including the datasets GSE116250, GSE120895, and GSE59867. GSE116250 and GSE120895 were used as the testing set, while GSE59867 was used as the validation set. LASSO regression analysis and SVM-RFE were utilized to identify feature genes.

Results

Analysis showed that among the differentially expressed genes between normal and heart failure patients, 9 genes were upregulated and 10 genes were downregulated. ROC curve analysis in the training set showed that TAGLN and SGPP2 had AUC values greater than 0.7. Moreover, SDSL and SMTNL2 had even higher AUC values of greater than 0.9. However, further analysis in the validation set showed that only SDSL had an AUC value greater than 0.7. Western blot experiments, RT-PCR, and ISO-induced experiments confirmed that SDSL was highly expressed in heart failure patients and promoted heart failure progression. In addition, SDSL promoted PARP1 expression and knockdown of SDSL expression led to decreased Cleaved-PARP1 expression and reduced cardiomyocyte apoptosis. Conversely, overexpression of SDSL resulted in increased PARP1 expression and myocardial cell apoptosis. These results suggest that elevated expression of SDSL in cardiomyocytes from heart failure patients may be an important factor promoting the occurrence and development of heart failure.

Conclusions

Using machine learning methods and experimental validation, it has been demonstrated that SDSL is a driving gene in patients with heart failure, providing a new treatment direction for clinical treatment.

Identification of crucial genes related to heart failure based on GEO database

BACKGROUND

The molecular biological mechanisms underlying heart failure (HF) remain poorly understood. Therefore, it is imperative to use innovative approaches, such as high-throughput sequencing and artificial intelligence, to investigate the pathogenesis, diagnosis, and potential treatment of HF.

METHODS

First, we initially screened Two data sets (GSE3586 and GSE5406) from the GEO database containing HF and control samples from the GEO database to establish the Train group, and selected another dataset (GSE57345) to construct the Test group for verification. Next, we identified the genes with significantly different expression levels in patients with or without HF and performed functional and pathway enrichment analyses. HF-specific genes were identified, and an artificial neural network was constructed by Random Forest. The ROC curve was used to evaluate the accuracy and reliability of the constructed model in the Train and Test groups. Finally, immune cell infiltration was analyzed to determine the role of the inflammatory response and the immunological microenvironment in the pathogenesis of HF.

RESULTS

In the Train group, 153 significant differentially expressed genes (DEGs) associated with HF were found to be abnormal, including 81 down-regulated genes and 72 up-regulated genes. GO and KEGG enrichment analyses revealed that the down-regulated genes were primarily enriched in organic anion transport, neutrophil activation, and the PI3K-Akt signaling pathway. The upregulated genes were mainly enriched in neutrophil activation and the calcium signaling. DEGs were identified using Random Forest, and finally, 16 HF-specific genes were obtained. In the ROC validation and evaluation, the area under the curve (AUC) of the Train and Test groups were 0.996 and 0.863, respectively.

CONCLUSIONS

Our research revealed the potential functions and pathways implicated in the progression of HF, and designed an RNA diagnostic model for HF tissues using machine learning and artificial neural networks. Sensitivity, specificity, and stability were confirmed by ROC curves in the two different cohorts.

Biomarkers in diagnosing and treatment of acute heart failure

Acute heart failure (AHF) is a complex disorder involving different pathophysiological pathways. In recent years, there is an increased focus on biomarkers that help with diagnosis, risk stratification and disease monitoring of AHF. Finding a reliable set of biomarkers not only improves morbidity and mortality but it can also potentially reveal the new targets of therapy. In this paper, we have reviewed the biomarkers found useful for the diagnosis as well as for risk stratification and prognostication in patients with AHF. We have discussed the established biomarkers for AHF including cardiac troponins and natriuretic peptides and emerging biomarkers including adiponectin, mi-RNA, sST2, Gal-3, MR-proADM, OPG, CT-proAVP and H-FABP for the purposes of making diagnosis, their use as a guide of therapy or for determination of prognosis.

Quantifying how diagnostic test accuracy depends on threshold in a meta-analysis

Tests for disease often produce a continuous measure, such as the concentration of some biomarker in a blood sample. In clinical practice, a threshold C is selected such that results, say, greater than C are declared positive and those less than C negative. Measures of test accuracy such as sensitivity and specificity depend crucially on C, and the optimal value of this threshold is usually a key question for clinical practice. Standard methods for meta‐analysis of test accuracy (i) do not provide summary estimates of accuracy at each threshold, precluding selection of the optimal threshold, and furthermore, (ii) do not make use of all available data. We describe a multinomial meta‐analysis model that can take any number of pairs of sensitivity and specificity from each study and explicitly quantifies how accuracy depends on C. Our model assumes that some prespecified or Box‐Cox transformation of test results in the diseased and disease‐free populations has a logistic distribution. The Box‐Cox transformation parameter can be estimated from the data, allowing for a flexible range of underlying distributions. We parameterise in terms of the means and scale parameters of the two logistic distributions. In addition to credible intervals for the pooled sensitivity and specificity across all thresholds, we produce prediction intervals, allowing for between‐study heterogeneity in all parameters. We demonstrate the model using two case study meta‐analyses, examining the accuracy of tests for acute heart failure and preeclampsia. We show how the model can be extended to explore reasons for heterogeneity using study‐level covariates.

The diagnostic accuracy of the natriuretic peptides in heart failure: systematic review and diagnostic meta-analysis in the acute care setting

OBJECTIVES

To determine and compare the diagnostic accuracy of serum natriuretic peptide levels (B type natriuretic peptide, N terminal probrain natriuretic peptide (NTproBNP), and mid-regional proatrial natriuretic peptide (MRproANP)) in people presenting with acute heart failure to acute care settings using thresholds recommended in the 2012 European Society of Cardiology guidelines for heart failure.

DESIGN

Systematic review and diagnostic meta-analysis.

DATA SOURCES

Medline, Embase, Cochrane central register of controlled trials, Cochrane database of systematic reviews, database of abstracts of reviews of effects, NHS economic evaluation database, and Health Technology Assessment up to 28 January 2014, using combinations of subject headings and terms relating to heart failure and natriuretic peptides.

ELIGIBILITY CRITERIA FOR SELECTING STUDIES

Eligible studies evaluated one or more natriuretic peptides (B type natriuretic peptide, NTproBNP, or MRproANP) in the diagnosis of acute heart failure against an acceptable reference standard in consecutive or randomly selected adults in an acute care setting. Studies were excluded if they did not present sufficient data to extract or calculate true positives, false positives, false negatives, and true negatives, or report age independent natriuretic peptide thresholds. Studies not available in English were also excluded.

RESULTS

37 unique study cohorts described in 42 study reports were included, with a total of 48 test evaluations reporting 15 263 test results. At the lower recommended thresholds of 100 ng/L for B type natriuretic peptide and 300 ng/L for NTproBNP, the natriuretic peptides have sensitivities of 0.95 (95% confidence interval 0.93 to 0.96) and 0.99 (0.97 to 1.00) and negative predictive values of 0.94 (0.90 to 0.96) and 0.98 (0.89 to 1.0), respectively, for a diagnosis of acute heart failure. At the lower recommended threshold of 120 pmol/L, MRproANP has a sensitivity ranging from 0.95 (range 0.90-0.98) to 0.97 (0.95-0.98) and a negative predictive value ranging from 0.90 (0.80-0.96) to 0.97 (0.96-0.98). At higher thresholds the sensitivity declined progressively and specificity remained variable across the range of values. There was no statistically significant difference in diagnostic accuracy between plasma B type natriuretic peptide and NTproBNP.

CONCLUSIONS

At the rule-out thresholds recommended in the 2012 European Society of Cardiology guidelines for heart failure, plasma B type natriuretic peptide, NTproBNP, and MRproANP have excellent ability to exclude acute heart failure. Specificity is variable, and so imaging to confirm a diagnosis of heart failure is required. There is no statistical difference between the diagnostic accuracy of plasma B type natriuretic peptide and NTproBNP. Introduction of natriuretic peptide measurement in the investigation of patients with suspected acute heart failure has the potential to allow rapid and accurate exclusion of the diagnosis.

Obesity and the response to intensified diuretic treatment in decompensated heart failure: a DOSE trial substudy

BACKGROUND

Obesity could attenuate diuretic effectiveness in treatment of acute decompensated heart failure (HF).

METHODS AND RESULTS

The DOSE trial randomized 308 subjects with acute HF to low- versus high-intensification intravenous diuretic therapy. We tested for statistical interactions between obesity and dosing strategy across clinical end points. After 72 hours of treatment, obese subjects (body mass index >30 kg/m(2); n = 173) had greater volume loss than nonobese subjects (n = 119) but similar improvements in dyspnea and freedom from congestion. Both groups had greater fluid loss with high-intensification treatment. Obese subjects had a higher incidence of worsening renal function (WRF) at 72 hours with low-intensification treatment, compared with nonobese subjects. In contrast, nonobese and obese subjects had similar incidence of WRF with high-intensification treatment. There were no differences between obese and nonobese subjects in time to discharge and 60-day freedom from death, emergency department visit, or rehospitalization.

CONCLUSIONS

The incidence of WRF was greater in obese than in nonobese subjects with low-intensification treatment. However, the frequency of WRF was equivalent in obese and nonobese subjects with high-intensification treatment. Additional studies are needed to assess whether obese patients with acute HF benefit from an initial high-intensification treatment strategy.

Prognostic utility of plasma neutrophil gelatinase-associated lipocalin in patients with acute heart failure: the NGAL EvaLuation Along with B-type NaTriuretic Peptide in acutely decompensated heart failure (GALLANT) trial

Aims

Neutrophil gelatinase-associated lipocalin (NGAL) is a measure of acute kidney injury. Renal dysfunction portends significant risk after discharge from acute heart failure (AHF). Thus, a sensitive marker of renal injury might also help to risk stratify HF patients.

Methods and results

GALLANT [NGAL EvaLuation Along with B-type NaTriuretic Peptide (BNP) in acutely Decompensated Heart Failure] was a multicentre, prospective study to assess the utility of plasma NGAL, alone and in combination with BNP, as an early risk marker of adverse outcomes. We studied 186 patients (61% male). There were 29 events (AHF readmissions and all-cause mortality) at 30 days (16%). Patients with events had higher levels of NGAL than those without (134 vs. 84 ng/mL, P < 0.001). The area under the receiver operating characteristic curve was higher for NGAL (0.72) than BNP (0.65), serum creatinine (0.57), or estimated glomerular filtration rate (eGFR; 0.55). In multivariable analyses, NGAL predicted events (P= 0.001), BNP approached significance (P= 0.052 and 0.070 without creatinine and GFR, respectively) while neither serum creatinine nor eGFR were significant. The addition of discharge NGAL over BNP alone improved classification by a net 10.3% in those with events and 19.5% in those without events, for a net reclassification improvement of 29.8% (P= 0.010). Subjects with both BNP and NGAL elevated were at significant risk [hazard ratio (HR) = 16.85, P= 0.006], as were subjects with low BNP and high NGAL (HR = 9.95, P= 0.036).

Conclusions

Plasma NGAL is a measure of kidney injury that at the time of discharge is a strong prognostic indicator of 30 days outcomes in patients admitted for AHF.

Clinical trial registration number: NCT 00693745

Brain natriuretic peptide-guided treatment does not improve morbidity and mortality in extensively treated patients with chronic heart failure: responders to treatment have a significantly better outcome

AIM

To determine whether brain natriuretic peptide (BNP)-guided heart failure (HF) treatment improves morbidity and/or mortality when compared with conventional treatment.

METHODS AND RESULTS

UPSTEP was an investigator-initiated, randomized, parallel group, multicentre study with a PROBE design. Symptomatic patients with worsening HF, New York Heart Association class II-IV, ejection fraction <40% and elevated BNP levels, were included. All patients (n= 279) were treated according to recommended guidelines and randomized to BNP-guided (BNP) or to conventional (CTR) HF treatment. The goal was to reduce BNP levels to <150 ng/L in younger patients and <300 ng/L in elderly patients, respectively. The primary outcome was a composite of death due to any cause, need for hospitalization and worsening HF. The study groups were well matched, including for BNP concentration at entry (mean: 808 vs. 899 ng/L; P= 0.34). There were no significant differences between the groups regarding either the primary outcome (P = 0.18) or any of the secondary endpoints. There were no differences for the pre-specified analyses; days out of hospital, and younger vs. elderly. A subgroup analysis comparing treatment responders (>30% decrease in baseline BNP value) vs. non-responders found improved survival among responders (P< 0.0001 for the primary outcome), and all of the secondary endpoints were also improved.

CONCLUSIONS

Morbidity and mortality were not improved by HF treatment guided by BNP levels. However, BNP responders had a significantly better clinical outcome than non-responders. Future research is needed to elucidate the responsible pathophysiological mechanisms in this sub-population.