In-hospital brain natriuretic peptide and N-terminal prohormone brain natriuretic peptide variations are predictors of short-term and long-term outcome in acute decompensated heart failure

A time series driven model for early sepsis prediction based on transformer module

Sepsis remains a critical concern in intensive care units due to its high mortality rate. Early identification and intervention are paramount to improving patient outcomes. In this study, we have proposed predictive models for early sepsis prediction based on time-series data, utilizing both CNN-Transformer and LSTM-Transformer architectures. By collecting time-series data from patients at 4, 8, and 12 h prior to sepsis diagnosis and subjecting it to various network models for analysis and comparison. In contrast to traditional recurrent neural networks, our model exhibited a substantial improvement of approximately 20%. On average, our model demonstrated an accuracy of 0.964 (± 0.018), a precision of 0.956 (± 0.012), a recall of 0.967 (± 0.012), and an F1 score of 0.959 (± 0.014). Furthermore, by adjusting the time window, it was observed that the Transformer-based model demonstrated exceptional predictive capabilities, particularly within the earlier time window (i.e., 12 h before onset), thus holding significant promise for early clinical diagnosis and intervention. Besides, we employed the SHAP algorithm to visualize the weight distribution of different features, enhancing the interpretability of our model and facilitating early clinical diagnosis and intervention.

Single measurement of serum N-terminal pro-brain natriuretic peptide: the best predictor of long-term mortality in patients with chronic systolic heart failure

BACKGROUND

Although high serum natriuretic peptide (NP) has long been associated with mortality prediction, it was usually tested under acute heart failure (HF) conditions and periods of analysis were short. This may explain the lack of consensus when its routine measurement for mortality prediction is contemplated. Here we evaluated, at the first clinic visit of chronic systolic HF patients, the usefulness of a single serum NP assessment for long-term mortality prediction.

METHODS

In 279 consecutive patients with chronic systolic HF, serum NT-proBNP was routinely measured once during the first clinic visit. We analyzed correlations between recorded mortality and the NT-proBNP finding, along with several known clinical echocardiographic, electrocardiographic and laboratory parameters recorded at that visit.

RESULTS

During average follow-up of 34±21months 59 (21%) patients died. Serum NT-proBNP was the strongest of the tested predictors of mortality [hazard ratio 3.76, 95% Cl (1.20-11.80), p=0.008]. Nearly seven years later, mortality was still higher in patients with higher initial serum NT-proBNP (p<0.001).

CONCLUSIONS

Compared to many other traditional prognostic parameters tested at the same time, the single serum NT-proBNP finding was the strongest predictor of long-term mortality. These results may justify its routine use for this purpose.

B-type natriuretic peptide predicts an ischemic etiology of acute heart failure in patients with stage 4-5 chronic kidney disease

OBJECTIVES

The non-invasive differentiation of ischemic and non-ischemic acute heart failure (AHF) not resulting from acute myocardial infarction is difficult and has therapeutic and prognostic implications. The aim of this study was to assess whether plasma B-type natriuretic peptide (BNP) can identify ischemic etiology in patients with stage 4-5 chronic kidney disease (CKD) presenting with AHF.

DESIGN AND METHODS

We prospectively analyzed 61 patients. The diagnosis of ischemic AHF was confirmed by coronary angiography or stress myocardial perfusion imaging. Plasma levels of BNP were measured at admission (BNP1) and 48 h after admission (BNP2).

RESULTS

The mean age of the study patients was 67 years. In these patients, 70.5% had diabetes and 47.5% had dialysis-dependent CKD; 28 of these patients (45.9%) had an ischemic etiology with significantly higher concentrations of BNP1 and BNP2 than did patients without ischemia. The area under the receiver operating characteristic curve was 0.755 (P=0.001) for BNP1 and 0.868 (P<0.001) for BNP2 to detect ischemic etiology of AHF. Plasma BNP1 >2907 ng/L (odds ratio [OR], 10.9; 95% confidence interval [CI] 2.5-48.4; P=0.002) and BNP2 >2322 ng/L (OR 93.1, 95% CI 7.0-1238.7; P=0.001) were independently associated with an ischemic etiology of AHF.

CONCLUSIONS

Plasma BNP may represent a clinically useful non-invasive tool for identification of ischemic etiology of AHF in patients with stage 4-5 CKD.

The diagnostic utility of brain natriuretic peptide in heart failure patients presenting with acute dyspnea: a meta-analysis

Heart failure with normal ejection fraction (HFNEF) accounts for approximately 50% of heart failure (HF) cases. To establish the utility of brain natriuretic peptide (BNP) in differentiating HF-related severe dyspnea from non-HF-related acute dyspnea, we used an estimation formula (eF) that was obtained from a series of three meta-regressions. We selected 60 out of 2721 case-control and follow-up studies that were published from 1998 to 2010. The heart failure levels (HFLs) were assessed using the New York Heart Association (NYHA) criteria. Random-effects meta-regression analyses of the natural logarithm (ln) of the BNP odds ratio (OR) were performed on the HFLs. The ln of the median BNP values (lnmBNP) was meta-regressed over the laboratory method (LM). A third meta-regression was performed on the HFLs to account for only the lnmBNP in the homogeneous LM subgroups. To determine the eF, the data from the diseased and control subjects were combined. The Bland-Altman method was used to detect eF bias. The overall BNP(OR) in the subgroup with severe HF was 35. The lnmBNP analysis showed that LM was a significant heterogeneity factor in the meta-regression (slope −0.38; CI −0.59 to −0.16). The meta-regression of lnmBNP on the HFL resulted in the following calculation for eF: estimated HFL (eHFL)=(lnmBNP–3.157)/0.886. The Bland-Altman test revealed no significant difference (0.0997; 95% CI −2.84 to 3.06) between HFL and eHFL. The severe eHFL showed a 78% accuracy. Based on the eF obtained from this meta-analysis, the BNP outcomes were shown to reliably diagnose severe dyspnea in HF and differentiate this condition from non-HF acute dyspnea.

Acute study of clinical effectiveness of nesiritide in decompensated heart failure: nesiritide redux

Nesiritide, a synthetic drug form of human B-type natriuretic peptide, is approved for the early treatment of dyspnea in acute decompensated heart failure. Meta-analyses suggested a risk of worsening renal insufficiency and mortality with its use. Therefore, the Acute Study of Clinical Effectiveness in Decompensated Heart Failure (ASCEND-HF) was designed as a prospective, multicenter, double-blind, randomized trial to examine the use of nesiritide in this common, morbid, and often lethal clinical condition. Two coprimary end points, dyspnea and 30-day hospital readmission or death, were chosen to examine symptomatic response and objective outcomes, respectively. Preliminary reports from ASCEND-HF investigators suggest no significant improvement in symptoms or clinical outcomes, although no adverse effect on mortality or renal function was noted. We recommend the continued use of nesiritide in acute decompensated heart failure as an individualized case-based therapy to those patients who meet criteria for treatment and are expected to receive benefit from its use.