Factors contributing to transient and persistent worsening renal function in elderly patients with acute decompensated heart failure

Abstract

Background

The aim of this study is to analyze the distinct characteristics and risk factors contributing to the development of worsening renal function (WRF) in elderly patients with acute decompensated heart failure (ADHF), while also examining the subset of persistent WRF among elderly patients with ADHF.

Methods

In this retrospective study, patients were stratified into two groups, non-WRF and WRF, based on alterations in creatinine levels. Subsequently, the WRF group was further delineated into transient and persistent WRF subgroups, contingent upon temporal changes in creatinine levels. There 218 elderly ADHF patients aged ≥ 60 years old, with an average age of (72.11 ± 7.94) years old. Among them, 66 patients with ADHF developed WRF during hospitalization, with an incidence of 30.28%. Among the 66 WRF patients, 20 cases were transient WRF and 46 cases were persistent WRF. The study revealed notable distinctions within the WRF cohort, characterized by heightened smoking prevalence, significantly elevated brain natriuretic peptide (BNP) levels at admission, diminished hematocrit (HCT) levels and more applications. ACEI (angiotensin-converting enzyme inhibitors) or ARB (angiotensin receptor inhibitor) or ARNI (angiotensin receptor neprilysin inhibitor) drugs. Within the persistent WRF subgroup, patients were characterized by advanced age, predominantly male gender distribution, heightened incidence of coronary artery disease, and significantly elevated systolic blood pressure (SBP), uric acid, BNP, and glycosylated hemoglobin levels compared to their transient WRF counterparts, while displaying lower D-dimer levels. In multivariate analysis, BNP level (OR = 1.001, 95%CI 1.000-1.001; P = 0.032), D-dimer level (OR = 1.013, 95%CI 1.003–1.024; P = 0.013), ACEI or ARB or ARNI drugs (OR = 19.656, 95%CI 8.698–30.763; P = 0.009) were positively correlated with the occurrence of WRF. HCT level was negatively correlated with the occurrence of WRF (OR = 0.850, 95%CI 0.744–0.970; P = 0.016). Systolic blood pressure (OR = 1.158, 95%CI 1.051–1.276; P = 0.003), D-dimer levels (OR = 0.998, 95%CI 0.997–0.999; P < 0.001) were associated with persistent WRF.

Conclusion

The findings of this study indicate a clear association between BNP, HCT, D-dimer levels and ACEI or ARB or ARNI drugs and the emergence of WRF in elderly individuals with ADHF. Furthermore, the study underscores that persistent WRF is notably linked with systolic blood pressure and D-dimer levels.

Background

The global burden of heart failure (HF) is escalating, affecting an estimated 26 million individuals worldwide [1]. Despite advancements in medical care, the five-year survival rate for patients experiencing acute decompensated heart failure (ADHF) remains dismally low at approximately 50% [2]. Notably, the intricate relationship between the heart and kidneys, where the kidneys receive about 25% of cardiac output, underscores their critical interdependence [3]. While renal dysfunction prevails in only 4.5% of the general population, its prevalence skyrockets to nearly 50% among patients with HF, leading to increased hospital costs, complications, and mortality rates [4].

Traditionally, renal dysfunction during hospitalization of patients with ADHF has been labeled as worsening of renal function (WRF), affecting a significant proportion, ranging from 18 to 40% [5, 6]. Furthermore, WRF emerges as an independent predictor of both short- and long-term mortality in patients with HF [7, 8]. However, conflicting evidence suggests that only persistent WRF bears prognostic significance in patients with HF following relief from congestive symptoms [9]. Compounding this, the rate of decline in estimated glomerular filtration rate accelerates in patients with HF, indicating a unique pathological trajectory [10, 11].

Despite the improved understanding of WRF in HF, there remains a notable gap in research focusing on its occurrence among elderly patients with ADHF. This study aims to delve into the characteristics, duration, and analysis of risk factors associated with WRF in this vulnerable population. By shedding light on these aspects, our research endeavors to equip clinicians with essential insights for optimizing care strategies in elderly patients with ADHF.

Methods

Patient population and study procedures

This retrospective single-center study involved the analysis of data from 218 consecutive participants screened at a heart failure center between January 2019 and January 2020. Based on the incidence of WRF in previous studies ranging from 15 to 40%, the allowable error was set at 0.05P, and the sample size was calculated to be 196–369. Eligibility criteria required patients to have been admitted with a primary diagnosis of ADHF and to exhibit signs of volume overload within 24 h of admission. (1) Treatment protocols included the administration of angiotensin-converting enzyme inhibitors/angiotensin receptor blockers/angiotensin receptor-neprilysin inhibitors, beta-blockers, spironolactone, and other diuretic drugs, with careful consideration given to any contraindications. (2) All patients had significant volume load and all required intravenous diuretic therapy. (3) The hyperemic symptoms of all patients were improved at discharge compared with those at admission: the inpatient medical records of the patients were consulted, and the judgment was evaluated mainly on the basis of changes in clinical symptoms and signs. Clinical symptoms mainly included fatigue, shortness of breath, dyspnea, orthopnea, decreased activity tolerance, decreased appetite, abdominal distension and other symptoms at discharge. The main clinical indicators were disappearance of pulmonary rales and reduction or disappearance of serous effusion at discharge. Hepatic jugular vein distention or hepatic jugular venous reflux syndrome reduced or disappeared; Peripheral edema decreased or disappeared. Patients aged 60 years and older diagnosed with ADHF and classified as New York Heart Association (NYHA) class II–IV upon admission to the hospital were enrolled in this study. Patients were excluded if they presented with a history of chronic renal insufficiency (defined as an eGFR < 60 m l/min/1.73 m 2) drug-induced kidney injury, urinary tract obstruction, malignancy, contrast nephropathy, or other conditions predisposing to acute kidney injury, or if their hospital stay was ≤ 3 days. All patients had hospitalizations exceeding 7 days. Comprehensive demographic profiles, comorbidities, laboratory findings and the application of drugs were meticulously collected for each patient. Upon admission, blood pressure readings were systematically recorded, and blood assays were promptly conducted within two hours. These assays encompassed assessments of hematocrit (HCT), renal function parameters (including creatinine, blood urea nitrogen, uric acid [UA]), brain natriuretic peptide (BNP), hemoglobin, glycosylated hemoglobin (HbA1C), and D-dimer levels. Renal function was further evaluated 72 h post-admission, pre-discharge, and at 3–5 days intervals thereafter, ensuring a thorough assessment throughout the hospitalization period for each patient.

Definitions of WRF

WRF was characterized as either a creatinine elevation of at least 26.5 µmol/L or a 25% increase from the baseline level recorded at admission, in accordance with the Kidney Disease Improving Global Outcomes criteria [12], observed at any point during the hospital stay. Persistent WRF was identified as a sustained elevation beyond this threshold throughout the hospitalization period. Transient WRF, on the other hand, was defined as a creatinine increase of at least 26.5 µmol/L within 72 h from the commencement of loop diuretics, followed by a subsequent return to creatinine levels below this designated threshold.

Statistical analysis

Normality was assessed using the Shapiro–Wilk test within each group. Variables exhibiting a normal distribution are presented as mean ± standard deviation, while those not conforming to normality are presented as a median (interquartile range, IQR). Categorical variables are reported as counts and percentages. Group comparisons were conducted accordingly: Student’s t-test for normally distributed variables, Mann–Whitney U test for non-normally distributed variables, and the chi-squared test for categorical variables. Logistic regression analysis was employed to elucidate the parameters obtained upon admission associated with the development of WRF. Univariate factors with P values ≤ 0.05 were included in a multiple logistic regression model. All reported p-values were two-tailed, and statistical significance was defined as P < 0.05. Statistical analysis was performed using SPSS version 26.0.

Results

The study included 218 patients, with a mean age of 72.11 ± 7.94 years. Among these, 66 patients developed WRF during their hospitalization, representing a rate of 30.28%. Of these 66 WRF patients, 20 experienced transient WRF, while 46 had persistent WRF (Fig. 1). Comparing baseline data, the prevalence of smoking was notably higher in the WRF group (39 [59.1%] vs. 63 [41.4%], P = 0.016), with no significant disparities observed in other comorbidities. Moreover, there were no significant differences in renal function at admission; however, the WRF group exhibited significantly elevated BNP levels upon admission (823 [413–2257] vs. 488 [344–1158], P < 0.001), coupled with lower HCT levels (35.28 ± 7.87 vs. 37.49 ± 6.88, P = 0.038), and more ACEI or ARB or ARNI drugs were used in WRF group (51 [77.3%] vs. 83[54.6%], P = 0.002) (Table 1).

Fig. 1.

The proportion of different WRF groups. Abbreviation: worsening of renal function (WRF)

Table 1.

The characteristic of non-WRF group and WRF group

Characteristic	Non-WRF (n = 152)	WRF (n = 66)	P
Age, y	72.28 ± 8.59	71.73 ± 6.25	0.593
Male sex, %	96 (63.2)	46 (69.7)	0.352
SBP, mmHg	125.56 ± 21.32	129.42 ± 17.51	0.197
Past history
Smoking history, %	63 (41.4)	39 (59.1)	0.016
Prior CAD, %	74 (48.7)	29 (43.9)	0.634
Prior MI, %	35 (23)	15 (22.7)	0.962
DM, %	67 (44.1)	34 (51.5)	0.104
Hypertension, %	87 (57.2)	40 (60.6)	0.643
Dyslipidemia, %	55 (36.2)	19 (28.8)	0.289
Hyperuricemia, %	37 (24.3)	23 (34.8)	0.111
Laboratory data
Creatinine, ummol/L	96.1(79.6,119.8)	98.4(85.7,125.6)	0.474
Urea, mmol/L	9.13(6.9,12.5)	8.0(5.8,12.8)	0.308
UA, ummol/L	429 (350,556)	462 (334,509)	0.140
HbA1C, %	6.84 ± 1.47	6.65 ± 1.73	0.408
HGB, g/L	116.88 ± 36.38	136.17 ± 39.34	0.113
HCT, %	37.49 ± 6.88	35.28 ± 7.87	0.038
BNP, pg/mL	488 (344,1158)	823 (413,2257)	< 0.001
D-dimmer, ng/mL	246 (139,442)	262 (188,519)	0.073
Drugs data
ACEI or ARB or ARNI, %	83(54.6)	51(77.3)	0.002
Spirolactone, %	86(56.6)	36(54.5)	0.882
SGLT2i, %	102(67.1)	41(62.1)	0.535

CAD: coronary heart disease; MI: myocardial infarction; DM: diabetes mellitus; SBP: systolic blood pressure; UA: uric acid; HbA1C: glycosylated hemoglobin; HGB: hemoglobin; HCT: hematocrit; BNP: brain natriuretic peptide; ACEI: angiotensin-converting enzyme inhibitors; ARB: angiotensin receptor inhibitor; ARNI: angiotensin receptor neprilysin inhibitor; SGLT2i: sodium-dependent glucose transporters 2 inhibitors

A comparison between the two groups with distinct types of WRF revealed significant discrepancies. Patients in the persistent WRF group were older and predominantly male compared to those in the transient WRF group (72.59 ± 7.14 vs. 69.75 ± 2.65, P = 0.022; 36 [78.3%] vs. 10 [50%], P = 0.022). Regarding comorbidities, patients in the persistent WRF group exhibited a higher prevalence of coronary artery disease (CAD) (24 [52.2%] vs. 5 [25%], P = 0.041). Concerning baseline data at admission, systolic blood pressure (SBP) (133.20 ± 16.45 vs. 120.75 ± 17.14, P = 0.007), UA (499 [359–536] vs. 372 [201–490], P = 0.008), BNP (499 [359–4861] vs. 388 [316–1820], P = 0.003), and HbA1C (6.90 ± 1.96 vs. 5.99 ± 0.55, P = 0.006) levels were notably higher in the persistent WRF group compared to the transient WRF group. However, D-dimer levels were significantly lower (219 [149–425] vs. 349 [235–2585], P = 0.004) (Table 2).

Table 2.

The characteristic of transient-WRF group and persistent-WRF group

Characteristic	Transient-WRF (n = 20)	Persistent-WRF (n = 46)	P value
Age, y	69.75 ± 2.65	72.59 ± 7.14	0.022
Male sex, %	10 (50)	36 (78.3)	0.022
SBP, mmHg	120.75 ± 17.14	133.20 ± 16.45	0.007
Past history
Smoking history, %	12 (60)	27 (58.7)	0.921
Prior CAD, %	5 (25)	24 (52.2)	0.041
Prior MI, %	5 (25)	10 (21.7)	0.771
DM, %	8 (40)	26(56.5)	0.083
Hypertension, %	11 (55)	29 (63.1)	0.539
Dyslipidemia, %	6 (30)	13 (28.3)	0.886
Hyperuricemia, %	7 (35)	16 (34.8)	0.986
Laboratory data
Creatinine, ummol/L	102.3(66.5,113.7)	98.4(85.7,146.1)	0.124
Urea, mmol/L	7.6 (5.6,11.9)	8.0(5.8,14.3)	0.624
UA, ummol/L	372 (201,490)	499 (359,536)	0.008
HbA1C, %	5.99 ± 0.55	6.99 ± 1.96	0.006
HGB, g/L	120.50 ± 37.80	142.98 ± 65.18	0.551
HCT, %	36.05 ± 11.37	34.94 ± 5.86	0.684
BNP, pg/mL	388 (316,1820)	499 (359,4861)	0.003
D-dimmer, ng/mL	349 (235,585)	219 (149,425)	0.004
Drugs data
ACEI or ARB or ARNI, %	15(75)	36(78.3)	0.661
Spirolactone, %	11(55)	25(54.3)	0.812
SGLT2i, %	10(50)	21(45.7)	0.179

In the correlation analysis, BNP, HCT, and D-dimer levels at admission and application of ACEI or ARB or ARNI were found to be associated with the occurrence of WRF in both univariate and multifactorial analyses (1.001 [1.000–1.001], P = 0.032; 0.850 [0.744–0.970], P = 0.016; 1.013 [1.003–1.024], P = 0.013; 19.656 [8.698–30.763], P = 0.009; respectively) (Table 3). Univariate analysis of the WRF group demonstrated that gender, SBP, history of CAD, UA, and D-dimer levels were all correlated with the occurrence of persistent WRF. Conversely, in the multifactorial analysis, only SBP and D-dimer levels were identified as associated with the occurrence of persistent WRF (1.158 [1.051–1.276], P = 0.003; 0.998 [0.997–0.999], P < 0.001) (Table 4).

Table 3.

Baseline predictors of WRF at univariate and multivariate analysis

Characteristic	Univariable		Multivariable
	OR (95%CI)	P value	OR (95%CI)	P value
Age, y	0.991 [0.955,1.028]	0.634	-	-
Male sex, %	1.342 [0.722,2.494]	0.353	-	-
SBP, mmHg	1.009 [0.995,1.024]	0.197	-	-
Past history
Smoking history, %	2.041 [1.134,3.671]	0.017	-	-
Prior CAD, %	0.790 [0.446,1.400]	0.419	-	-
Prior MI, %	0.983 [0.494,1.957]	0.962	-	-
DM, %	1.619 [0.904,2.897]	0.105	-	-
Hypertension, %	1.149 [0.638,2.072]	0.643	-	-
Dyslipidemia, %	0.713 [0.318,1.335]	0.290	-	-
Hyperuricemia, %	1.662 [0.888,3.113]	0.112	-	-
Laboratory data
Creatinine, ummol/L	1.004 [0.997,1.011]	0.240	-	-
Urea, mmol/L	0.979 [0.917,1.046]	0.532	-	-
UA, ummol/L	1.000 [0.999,1.001]	0.602	-	-
HbA1C, %	0.941 [0.770,1.149]	0.548	-	-
HGB, g/L	1.003 [0.997,1.009]	0.269	-	-
HCT, %	0.960 [0.922,0.998]	0.042	0.850 [0.744,0.970]	0.016
BNP, pg/mL	1 [1,1.001]	0.008	1.001 [1.000,1.001]	0.032
D-dimmer, ng/mL	1.001 [1,1.001]	0.007	1.013 [1.003,1.024]	0.013
Drugs data
ACEI or ARB or ARNI, %	2.827[1.463,5.459]	0.002	19.656[8.698,30.763]	0.009
Spirolactone, %	0.921[0.515,1.647]	0.781	-	-
SGLT2i, %	0.804[0.441,1.467]	0.477	-	-

Table 4.

Baseline predictors of persistent-WRF at univariate and multivariate analysis

Characteristic	Univariable		Multivariable
	OR (95%CI)	P value	OR (95%CI)	P value
Age, y	1.087 [0.985,1.200]	0.097	-	-
Male sex, %	3.600 [1.172,11.057]	0.025	-	-
SBP, mmHg	1.053 [1.011,1.097]	0.012	1.158 [1.051,1.276]	0.003
Past history
Smoking history, %	0.947 [0.325,2.762]	0.921
Prior CAD, %	3.273 [1.020,10.500]	0.046	-	-
Prior MI, %	0.833 [0.243,2.854]	0.772	-	-
Atrial fibrillation, %			-	-
DM, %	2.559 [0.872,7.508]	0.087	-	-
Hypertension, %	1.396 [0.481,4.049]	0.540	-	-
Dyslipidemia, %	0.919 [0.291,2.908]	0.886	-	-
Hyperuricemia, %	1.990 [0.329,2.979]	0.986	-	-
Laboratory data
Creatinine, ummol/L	1.015 [1.000,1.029]	0.050	-	-
Urea, mmol/L	1.085 [0.958,1.229]	0.197	-	-
UA, ummol/L	1.005 [1.001,1.009]	0.012	-	-
HbA1C, %	1.539 [0.952,2.488]	0.079	-	-
HGB, g/L	1.002 [0.994,1.011]	0.613	-	-
HCT, %	0.982 [0.916,1.052]	0.597	-	-
BNP, pg/mL	1.002 [1.000,1.004]	0.081	-	-
D-dimmer, ng/mL	0.999 [0.998,1.000]	0.051	0.998 [0.997,0.999]	< 0.001
ACEI or ARB or ARNI, %	0.577[0.143,2.323]	0.439
Spirolactone, %	1.161[0.049,1.528]	0.503	-	-
SGLT2i, %	2.308[0.718,7.420]	0.161	-	-

Discussion

This study represents a pioneering effort in delineating the clinical characteristics of WRF among elderly patients with ADHF, alongside an in-depth analysis of the distinctive clinical profiles associated with different forms of WRF. Furthermore, the study delves into the underlying factors contributing to the onset of WRF. Notably, the research underscores the pervasive occurrence of WRF among elderly patients with ADHF, with notable associations identified between pertinent factors such as D-dimer levels, BNP concentrations, and HCT levels upon hospitalization, and application of ACEI or ARB or ARNI and the development of WRF. Moreover, within the group of elderly patients with ADHF experiencing WRF, the study reveals marked variations in clinical phenotypes across different types of WRF, implicating a multifaceted interplay of factors in their manifestation. Of particular significance is the recognition of the initial SBP level as a salient determinant in this context.

Despite significant advances in medical technology, the prognosis for patients with ADHF remains poor, with a 30-day readmission rate of up to 20% and an in-hospital mortality rate of 5–6% [13, 14]. One of the strongest predictors of adverse outcomes in ADHF is renal dysfunction [15]. Some studies have shown that up to 47% of patients with ADHF develop renal impairment [16]. Worsening renal function may mirror low cardiac output, renal hypoperfusion, or renal venous congestion, which may lead to worse clinical outcomes [17]. The proportion of patients with WRF in the present study was as high as 30.28%. WRF is important in terms of its preventability and modifiability, unlike baseline renal function, which is usually not modifiable [18, 19]. Therefore, early identification of the characteristics and risk factors of patients with WRF is essential. This focus on clinical management is crucial for preventing the occurrence of WRF. The severity of the underlying renal insufficiency is one of the strongest risk factors for WRF, reflecting the likelihood of such differences due to reduced renal reserve and impaired ability of the kidneys to cope with stress [20]. Therefore, patients with severe renal insufficiency were excluded from this study. At the same time, early identification of patients at risk of renal dysfunction and a robust definition and better understanding of its cause and consequences is necessary [21].

This study revealed higher levels of BNP and lower HCT levels upon admission and higher application of ACEI or ARB or ARNI in the WRF group among patients with ADHF compared to the non-WRF group. Patients with HF undergo dynamic physiological changes attributed to low cardiac output and/or altered venous return, subsequently impacting renal function via sympathetic activation and/or activation of the renin–angiotensin–aldosterone system (RAAS) [22]. BNP serves as a marker reflecting the degree of congestion in patients with ADHF, strongly correlating with prognosis and serving as a robust indicator of cardiovascular risk [23, 24]. Although all admissions involved patients with NYHA class II–IV HF, it is conceivable that individuals with more severe congestion were predisposed to developing WRF. Patients in the WRF group exhibited lower admission HCT levels, suggesting a potential association with a higher relative volume in this cohort.

Numerous studies now advocate for the dynamic monitoring of HCT changes as a means to assess the response of patients with HF to diuretic therapy, offering valuable insights into renal injury [25]. However, it is crucial to acknowledge that while HCT assessment primarily reflects the relative decrease in plasma volume between two time points, it does not provide an absolute measure of plasma volume [26, 27]. Nevertheless, despite this limitation, HCT remains a valuable indicator for evaluating WRF and responsiveness to diuretics. This assertion is bolstered by evidence from a sub-analysis of the Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheterization Effectiveness trial, which confirms the association between HCT levels and the occurrence of WRF [21, 28]. In a recent study from Fuwai Hospital, 3661 patients with heart failure were enrolled. An increase in HCT was associated with the occurrence of WRF, but not with the risk of 1-year death in a subgroup of patients with impaired renal function (HR, 0.90; 95%CI: 0.64–1.26; P = 0.545). In HF patients, the prognostic value of HCT varied with renal function [29]. In HF patients with normal renal function, HCT level should be considered as an effective reference index of plasma volume, which should be paid attention to in clinical practice. The results of this study showed that a higher baseline HCT may be a protective factor for WRF. HCT can represent the current volume of patients to a certain extent, and a higher HCT can indicate that the current circulatory congestion of patients is relatively mild. However, this study did not evaluate the relative change of HCT, and further study is needed.

The impact of WRF on the prognosis of patients with ADHF remains a contentious topic within the medical community. A recent meta-analysis conducted by Damman et al. [4] suggested a robust correlation between renal dysfunction and adverse outcomes in ADHF. However, an opposing perspective posits that WRF may simply reflect the activation of the RAAS and alterations in renal hemodynamics during infusion therapy, without directly influencing patient prognosis [30, 31]. Contrary to this notion, studies such as that of Valente [17] have highlighted that worsening renal function during diuresis improvement and HF status enhancement might actually signal a more favorable prognosis for patients with ADHF. WRF that occurs in patients with AHF or ADHF after diuretic treatment with improvement of hyperemic symptoms does not predict worse prognosis of HF patients and is currently referred to as pseudo-WRF [17]. However, it should also be warned that not all WRF occurring in the case of improvement of hyperemic symptoms after diuretic treatment are pseudoWRF, which is not associated with poor prognosis. A total of 6112 patients with heart failure in RELAX-AHF-2 study were enrolled and found that hemodynamic abnormalities at admission and continuous use of diuretics during hospitalization could not explain the occurrence of WRF. Finally, it was found that the incidence of WRF increased gradually with the increase of LVEF during AHF hospitalization. Patients with high LVEF are more likely to develop WRF, which is not related to hemodynamics and the effect of decongestive treatment. WRF was associated with an increased risk of post-discharge events except for patients in the lowest LVEF quartile group [32].

Recent investigations have focused on the timing of WRF development during hospitalization. Ruocco et al. delineated two distinct subtypes of WRF: persistent and transient. Through multivariable analysis, they discerned that only persistent WRF exhibited a significant association with poor prognosis (HR 1.61 [1.02–2.57], P = 0.04) [33]. In two large cohorts of patients with AHF, WRF in the first 4 days was not associated with worse outcomes when patients had a good diuretic response [34]. Noteworthy distinctions were observed in the clinical profiles of patients with persistent WRF, characterized by a predominance of males, advanced age, and elevated levels of various biomarkers including SBP, HbA1C, BNP, and UA upon admission, along with a higher prevalence of comorbid CAD. These findings suggest that patients with persistent WRF tend to manifest poorer overall health, often compounded by metabolic and vascular-related comorbidities. Moreover, these patients may exhibit heightened RAAS activity, increased sympathetic tone, and a more pronounced inflammatory response. The elevated SBP observed in the persistent WRF group also hints at compromised arterial elasticity, suggesting an augmented reliance on preload. Consequently, any reduction in preload could precipitate a more pronounced decline in renal perfusion [35, 36].

The outcomes of this study underscore the correlation between elevated SBP and D-dimer levels upon admission and the subsequent development of persistent WRF in elderly patients with ADHF. The escalating prevalence of both HF and hypertension with advancing age is worth noting [37]. The elderly HF population typically exhibits compromised vascular elasticity and frequently presents with concurrent hypertension [37]. An in-depth analysis of 6,544 patients with acute heart failure (AHF) in the RELAX-AHF-2 (Relaxin in Acute Heart Failure-2) study disclosed a significant association between a rapid SBP decline early in the course of the condition and the onset of WRF. Crucially, the primary determinant of this early SBP decline was identified as the initial SBP level [38]. Previous investigations primarily focused on the impact of D-dimer on WRF in patients with diabetes mellitus or diabetic nephropathy [38], D-dimer serves as a pivotal indicator of fibrinolysis within the body. It reflects the activation of the body’s coagulation and fibrinolytic system, thereby inducing a pathological hypercoagulable state of the blood and predisposing individuals to thrombosis. Furthermore, microangiopathy can inflict damage upon endothelial vessels, thereby impeding glomerular function [39, 40]. Consequently, to some extent, D-dimer may offer insights into the pathological changes associated with early glomerular injury. In addition, elderly ADHF patients are often complicated with atrial fibrillation, and these patients are treated with warfarin or new oral anticoagulants. Previous studies have shown that some of the deterioration of renal function may be due to the use of anticoagulants. Warfarin can promote the deterioration of renal function when the anticoagulant level is too high. In contrast, newer oral anticoagulants appear to protect against renal injury better than warfarin, but the precise mechanism is not demonstrated. The application of these drugs may have a partial effect on the coagulation function of patients [41]. Notably, in this study, D-dimer levels exhibited a marked elevation in the transient WRF group, a phenomenon warranting further investigation. Given the paucity of international studies exploring the factors linked to different types of renal impairment, the relationship between D-dimer and various manifestations of WRF remains unverified.

This study has some limitations. First, this was a single-center retrospective study and the small sample size of this study which may have affected the generality of the results. Second, some confounding factors could not be excluded as the small sample size limit prevented us from conducting appropriate subgroup analysis or corresponding adjustment, including the use of RAAS inhibitors, atrial fibrillation, and use of SGLT2 inhibitors. Third, in terms of the etiology of heart failure, the etiology of heart failure in some patients is still unclear in practice, and the causes of heart failure in some patients may be influenced by many factors, so the original data are missing. Forth, the degree of change in creatinine was not studied further. Fifth, the novel coronavirus pandemic led to a complex situation, which could lead to biased results. Furthermore, our study did not specifically address patients with Grade 2–3 worsening renal function, nor did we further stratify patients based on the severity of creatinine changes. Future research should address these limitations through multicenter, prospective studies with larger sample sizes, clearer patient selection criteria, and detailed stratification of renal function severity.

Conclusion

In conclusion, the data analysis reveals a significant correlation between BNP and HCT levels and the onset of WRF in elderly patients with ADHF. Notably, there were discernible differences in the clinical profiles of patients experiencing various types of WRF. Furthermore, persistent WRF was found to be closely associated with systolic blood pressure and D-dimer levels. These findings underscore the importance of alerting clinicians to these critical associations for more informed patient care and management strategies.

Abbreviations

ADHF

acute decompensated heart failure

WRF

worsening renal function

HF

heart failure

HCT

hematocrit

CAD

coronary artery disease

SBP

systolic blood pressure

UA

uric acid

BNP

brain natriuretic peptide

HbA1C

glycosylated hemoglobin

NYHA

New York Heart Association

ACEI

angiotensin-converting enzyme inhibitors

ARB

angiotensin receptor inhibitor

ARNI

angiotensin receptor neprilysin inhibitor

SGLT2i

sodium-dependent glucose transporters 2 inhibitors

Author contributions

YQS, LML and ZYW conceived the idea and conceptualised the study. ZYW, QL and XD collected the data. LML and ZYW analysed the data. YQS and JZD drafted the manuscript, then YQS and JZD reviewed the manuscript. All authors read and approved the final draft.

Funding

No external funding received to conduct this study.

Data availability

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The retrospective study was approved by ethics committee of the Beijing Anzhen Hospital. This study was conducted in accordance with the declaration of Helsinki. Written informed consent was obtained from all subjects.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.