Albumin combined with neutrophil-to-lymphocyte ratio score and outcomes in patients with acute coronary syndrome treated with percutaneous coronary intervention

Abstract

Background

Evidence about the association between albumin combined with neutrophil-to-lymphocyte ratio score (ANS) and survival outcomes in patients with acute coronary syndrome (ACS) undergoing percutaneous coronary intervention (PCI) is rare. This study aimed to evaluate the prognostic value of ANS in patients with ACS undergoing PCI by propensity score matching (PSM) analysis.

Patients and methods

Patients with ACS undergoing PCI were consecutively enrolled in this prospective cohort study from January 2016 to December 2018. The albumin and neutrophil-to-lymphocyte ratio cutoff values for predicting major adverse cardiovascular events (MACEs) were calculated using receiver operating characteristic curves. Survival analysis was performed using Kaplan–Meier estimates, the Cox proportional hazard regression models and PSM. The study endpoint was the occurrence of a MACE, which included all-cause mortality and rehospitalization for severe heart failure during follow-up.

Results

Overall, 1549 patients with adequate specimens were identified and assigned into different groups for comparison. Before and after PSM, the Kaplan–Meier curves showed that a higher ANS value was associated with a higher risk of MACEs (all P < 0.001). The multivariate Cox proportional hazard regression model showed that the ANS (per 1 score increase) [hazard ratio (HR), 2.016; 95% confidence interval (CI), 1.329–3.057; P = 0.001 vs. HR, 2.166; 95% CI, 1.344–3.492; P = 0.002] was an independent predictor for MACEs.

Conclusion

This study tentatively confirms that ANS may be a valuable clinical indicator to identify high-risk ACS patients after PCI. More high-quality prospective studies are needed in the future.

Introduction

Despite continuous advances in diagnostic and therapeutic modalities, the mortality rate of acute coronary syndrome (ACS), has remained high over the past decades [1,2]. ACS encompasses a range of conditions that result from a sudden reduction in blood flow to the heart muscle and includes unstable angina, non-ST-segment elevation myocardial infarction (NSTEMI) and ST-segment elevation myocardial infarction (STEMI) [3]. Primary percutaneous coronary intervention (PCI) of the infarct-related artery reduces mortality and myocardial infarction in patients [4]. However, considering the persistently high mortality rates, it is also critical to explore emerging risk factors for prognosis in patients with ACS.

Serum albumin is the most abundant plasma protein in the human body [5]. Additionally, its antithrombotic, anti-inflammatory and antioxidative effects have been widely examined. Several studies have shown that albumin can be an excellent predictor of risk in patients with coronary artery disease (CAD) [6–8]. Chronic inflammation is a key process in the pathogenesis of atherosclerosis [9]. Unstable plaque rupture in the coronary artery and thrombosis are the most important pathological basis of ACS [10]. With the widespread use of lipid-lowering drugs, ACS is sometimes caused by plaque erosion rather than rupture [11,12]. Neutrophils and lymphocytes are the primary cells involved in this process [13]. In previous studies, the neutrophil-to-lymphocyte ratio (NLR) has been found to be associated with CAD as a marker of increased systemic inflammation [14,15].

Some studies have recently reported that albumin combined with neutrophil-to-lymphocyte ratio score (ANS) is valuable for prognosis of patients with colorectal cancer and patients who undergo mandibulofacial reconstruction with a fibula flap [16,17]. In our previous study, we found that albumin combined with NLR score was an independent risk factor of CAD [18]. To the authors’ knowledge, however, the prognostic value of the ANS for patients with ACS undergoing PCI has never been scrutinized explicitly. Thus, this study aimed to evaluate the prognostic value of ANS in patients with ACS undergoing PCI.

Methods

Study design and population

Overall, 1773 patients with ACS undergoing PCI were consecutively enrolled in this prospective cohort study from January 2016 to December 2018 at The Affiliated Hospital of Chengde Medical University. The inclusion criteria were as follows: (1) age ≥18 years; (2) ACS clinical types of unstable angina, NSTEMI and STEMI; (3) coronary arteriography showing ≥50% stenosis in at least one of the left main, left anterior descending, left circumflex, right coronary or main branches and (4) underwent PCI (complete revascularization) for the first time. The exclusion criteria are the same as our previous study [19]. A total of 1549 patients with ACS undergoing PCI were included in the final analysis.

Data collection and variable definition

Descriptions of data collection and variable definitions (e.g. diabetes, hypertension, dyslipidemia, ischemic stroke, definition of success in PCI and cardiogenic shock) are taken from our previous research [19]. Hypertension was defined as systolic blood pressure ≥140 mmHg (1 mmHg=0.133 kPa) and diastolic blood pressure ≥90 mm Hg at rest or a previous hypertension diagnosis with antihypertensive therapy. Diabetes was defined according to the following American Diabetes Association guidelines: a glycated hemoglobin value of ≥6.5%, a fasting plasma glucose value of ≥126 mg/dl (7.0 mmol/l), a 2-h plasma glucose value of ≥200 mg/dl (11.1 mmol/l) during a 75g oral glucose tolerance test, classic hyperglycemia symptoms (e.g. polyuria, polydipsia and weight loss) or hyperglycemic crisis with a random plasma glucose value of ≥200 mg/dl (11.1 mmol/l). In the absence of unequivocal hyperglycemia, the first three criteria were confirmed by repeat testing. Dyslipidemia was defined as a serum total cholesterol value of ≥5.18 mmol/l (200 mg/dl), a high-density lipoprotein cholesterol value of ≤1.04 mmol/l, a low-density lipoprotein cholesterol value of ≥3.37 mmol/l, a triglyceride value of ≥1.7 mmol/l or a previous dyslipidemia diagnosis and prescribed medication.

PCI was performed by experienced cardiologists using the Judkins technique with 6F right and left heart catheters. Procedural success was defined as a reduction to <30% in the percent diameter stenosis associated with thrombolysis in myocardial infarction grades 2 or 3. The angiographic characteristics of all patients were determined and reported by the PCI doctors’ team. Ischemic stroke was diagnosed by two experienced clinical neurologists from the Department of Neurology based on the recommendations of the WHO. Diagnoses were confirmed by evaluating clinical symptoms, neurological examination results and computed tomography or MRI findings. Cardiogenic shock was diagnosed as systolic blood pressure <90 mmHg with adequate volume and clinical hypoperfusion (e.g. cold extremities, oliguria, mental confusion, dizziness and narrow pulse pressure) or laboratory hypoperfusion (e.g. metabolic acidosis, elevated serum lactate and elevated serum creatinine).

Laboratory data

Fasting blood samples were collected within the first 24 h of admission before PCI. White blood cell, platelet, neutrophil and lymphocyte counts were assessed using an automatic hematology analyzer (Sysmex XE-2100; Sysmex, Kobe, Japan). This is consistent with our previous research [19]. Concomitantly, the calculation methods of NLR and ANS are also the same as our previous studies [18]. The NLR was calculated using the following formula: NLR = neutrophil count (10⁹/l)/lymphocyte count (10⁹/l) [14]. The ANS was calculated using the following data: the cutoff points for albumin and NLR were 4.07 g/dl [area under the curve (AUC) of albumin, 0.625; 95% confidence interval (CI), 0.553–0.696] and 2.60 (AUC of NLR, 0.640; 95% CI, 0.570–0.709), respectively, as defined by receiver operating characteristic curve analysis. For the ANS assessment, patients with serum albumin levels ≥4.07 g/dl and NLRs ≤2.60 were assigned a score of 0; patients with serum albumin levels ≥4.07 g/dl and NLRs >2.60 were assigned a score of 1; patients with serum albumin levels <4.07 g/dl and NLRs ≤2.60 were assigned a score of 1; and those with both hypoalbuminemia (albumin level <4.07 g/dl) and a high NLR level (>2.60) were assigned a score of 2. In this study, patients with an ANS of 0 were defined as a low ANS group. Patients with ANS scores of 1 and 2 were defined as high ANS groups.

Follow-up and study endpoints

The method of follow-up and the definition of the study endpoint were consistent with our previous study [19]. Follow-up data were collected via a review of electronic medical records or clinic visits at 1, 3, 6 and 12 months after the procedure and annually thereafter. The study endpoint was the development of a major adverse cardiovascular event (MACE), which included all-cause mortality and rehospitalization for severe heart failure during the follow-up period. All-cause mortality was defined as death from any cause. Severe heart failure was defined as the New York Heart Association classification class IV.

Propensity score matching

In this study, propensity score matching (PSM) was applied to balance potential confounders between different groups (e.g. patients with ANS of 0, 1 and 2). PSM analysis was conducted with the 1 : 2 nearest neighbor matching method. PSM was performed with a tolerance of 0.02. Covariates for PSM include confounding factors that affect the risk of the outcome, as well as risk factors unrelated to exposure. Matching covariates consisted of sex, age, dyslipidemia, hypertension, diabetes, ischemic stroke, smoking, family history of CAD, unstable angina, STEMI and NSTEMI.

Statistical analyses

The Kolmogorov–Smirnov test was employed to assess the normal distribution of the variables. Continuous normally distributed data were displayed as mean and SD, and continuous non-normally distributed data were displayed as median and 25th percentile and 75th percentile. All continuous variables were compared among patients with ANS 0, 1 and 2. The one-way analysis of variance and Kruskal–Wallis H test were used to analyze continuous variables with normal and non-normal distribution respectively. The relationship of continuous variables between the MACEs and non-MACEs groups was compared using the Student’s t-test for normally distributed continuous variables and using the Mann–Whitney U test for non-normally distributed variables. Meanwhile, all categorical variables were presented as numbers (%) and compared using the chi-square test.

Receiver operating characteristic curves were used to calculate the cutoff values for albumin and NLR for predicting MACEs (i.e. all-cause mortality and rehospitalization for severe heart failure). Survival was estimated by generating Kaplan–Meier survival curves, and differences in survival outcomes among groups were investigated using the log-rank test. Cox proportional hazard regression models with the forward selection method were used to identify the independent prognostic factors for MACEs. All statistical analyses were performed using SPSS (Windows version 22.0; SPSS Inc., Chicago, Illinois, USA), R software (version 2.15.3, The R Project for Statistical Computing, Vienna, Austria) and GraphPad Prism 8.0 (GraphPad Software Inc., La Jolla, California, USA). All statistical tests were two-sided, with a significance level of P < 0.05.

Results

Baseline characteristics before propensity score matching

The MACEs group involved 61 patients; among them, 56 patients died of any cause and five patients required rehospitalization for severe heart failure (heart function level IV based on the New York Heart Association classification). Meanwhile, the non-MACEs group involved 1488 patients (Fig. 1). Table 1 presents the clinicodemographic patient characteristics. The number of patients aged ≥65 years; prevalence rates of smoking, dyslipidemia, heart failure, family history of CAD, unstable angina, STEMI, LVEF < 40%; single-vessel coronary disease and triple-vessel coronary disease; white blood cell count; platelet count; neutrophil count; lymphocyte count; monocyte count; albumin level and left ventricular end-diastolic diameter were significantly different among the ANS 0, 1 and 2 groups (all P < 0.05).

Fig. 1.

Patient enrollment flowchart including propensity score matching.

Table 1.

Comparison of baseline patient characteristics by albumin combined with neutrophil-to-lymphocyte ratio score in the total and matched cohort

Variables	Original cohort (n = 1549)					Matched cohort (n = 137)
	ANS = 0 (n = 391)	ANS = 1 (n = 712)	ANS = 2 (n = 446)	χ2/H	P value	ANS = 0 (n = 27)	ANS = 1 (n = 54)	ANS = 2 (n = 56)	χ2/H/F	P value
Clinicodemographic data
Male	281 (71.9)	533 (74.9)	344 (77.1)	3.065	0.216	14 (51.9)	42 (77.8)	44 (78.6)	7.633	0.022
Age ≥65 years	76 (19.4)	156 (21.9)	137 (30.7)	17.262	<0.001	13 (48.1)	27 (50.0)	31 (55.4)	0.498	0.780
Smoking	184 (47.1)	373 (52.4)	248 (55.6)	6.189	0.045	6 (22.2)	31 (57.4)	24 (42.9)	9.128	0.010
Dyslipidemia	254 (65.0)	425 (59.7)	201 (45.1)	38.063	<0.001	20 (74.1)	38 (70.4)	27 (48.2)	7.797	0.020
Hypertension	235 (60.1)	422 (59.3)	254 (57.0)	0.968	0.616	15 (55.6)	35 (64.8)	28 (50.0)	2.487	0.288
Diabetes	105 (26.9)	175 (24.6)	113 (25.3)	0.691	0.708	9 (33.3)	14 (25.9)	14 (25.0)	0.695	0.707
Ischemic stroke	50 (12.8)	101 (14.2)	69 (15.5)	1.231	0.540	3 (11.1)	13 (24.1)	10 (17.9)	2.044	0.360
HF	22 (5.6)	65 (9.1)	68 (15.2)	22.538	<0.001	2 (7.4)	6 (11.1)	12 (21.4)	3.742	0.154
CGS	2 (0.5)	13 (1.8)	9 (2.0)	3.760	0.153	1 (3.7)	1 (1.9)	3 (5.4)	0.961	0.619
Family history of CAD	69 (17.6)	98 (13.8)	50 (11.2)	7.230	0.027	3 (11.1)	2 (3.7)	2 (3.6)	2.499	0.287
UA	248 (63.4)	284 (39.9)	82 (18.4)	176.692	<0.001	13 (48.1)	18 (33.3)	7 (12.5)	12.942	0.002
STEMI	90 (23.0)	312 (43.8)	287 (64.3)	144.358	<0.001	3 (11.1)	28 (51.9)	37 (66.1)	22.187	<0.001
NSTEMI	53 (13.6)	116 (16.3)	77 (17.3)	2.313	0.315	11 (40.8)	8 (14.8)	12 (21.4)	6.988	0.030
Laboratory data
WBC count (109/l)	6.81 [5.75–8.39]	7.89 [6.40–10.32]	9.45 [7.40–11.80]	166.832	<0.001	7.00 [5.15–7.84]	8.23 [6.65–10.33]	8.58 [7.60–10.54]	14.726	0.001
Platelet count (109/l)	222.00 [185.00–262.00]	214.00 [179.00–248.75]	209.50 [176.00–243.25]	10.910	0.004	231.44 ± 63.87	219.98 ± 47.64	197.05 ± 46.39	0.635	0.427
Neutrophil count (109/l)	3.93 [3.36–4.87]	5.47 [4.12–8.28]	7.55 [5.63–9.97]	384.829	<0.001	3.97 [2.85–4.76]	5.78 [4.30–9.08]	6.98 [5.63–9.11]	33.522	<0.001
Lymphocyte count (109/l)	2.30 [1.81–2.82]	1.63 [1.20–2.16]	1.30 [0.96–1.68]	367.359	<0.001	2.34 [1.72–2.77]	1.44 [0.91–2.10]	1.26 [0.87–1.72]	30.439	<0.001
Monocyte count (109/l)	0.44 [0.33–0.56]	0.42 [0.31–0.56]	0.43 [0.31–0.64]	7.010	0.030	0.40 [0.33–0.51]	0.38 [0.26–0.54]	0.48 [0.34–0.65]	4.767	0.092
ALB (g/dl)	4.34 [4.21–4.52]	4.19 [3.93–4.39]	3.82 [3.64–3.97]	681.733	<0.001	4.31 [4.22–4.41]	4.19 [4.00–4.39]	3.83 [3.67–3.98]	74.117	<0.001
Cr (μmol/l)	67.50 [58.70–77.66]	67.35 [59.00–76.59]	67.20 [59.11–80.11]	1.929	0.381	61.00 [49.87–73.82]	69.21 [55.35–85.38]	72.30 [63.23–85.75]	10.412	0.005
Serum uric acid (μmol/l)	329.30 [262.90–384.60]	327.95 [266.75–392.58]	319.05 [261.28–375.00]	3.847	0.146	342.40 [255.90–380.00]	315.40 [267.48–359.78]	317.30 [249.05–379.75]	0.798	0.671
Echocardiography
LA (mm)	35.00 [32.00–37.00]	34.00 [32.00–37.00]	35.00 [32.00–38.00]	4.476	0.107	33.00 [31.00–35.00]	35.00 [32.00–38.00]	35.00 [32.25–38.00]	5.359	0.069
LVEDD (mm)	50.00 [47.00–53.00]	50.00 [47.00–54.00]	51.00 [48.00–55.00]	9.799	0.007	49.89 ± 6.08	50.93 ± 4.78	52.02 ± 5.64	0.731	0.394
LVEF < 40%	5 (1.5)	15 (2.3)	21 (5.1)	10.278	0.006	0 (0.0)	2 (3.7)	7 (12.5)	5.830	0.054
Coronary angiogram
1 vessel	134 (34.3)	232 (32.6)	119 (26.7)	6.574	0.037	9 (33.3)	13 (24.1)	13 (23.2)	1.082	0.582
2 vessel	115 (29.4)	237 (33.3)	142 (31.8)	1.745	0.418	4 (14.8)	19 (35.2)	17 (30.4)	3.675	0.159
3 vessel	142 (36.3)	243 (34.1)	185 (41.5)	6.423	0.040	14 (51.9)	22 (40.7)	26 (46.4)	0.950	0.622

Data are presented as n (%) or median [range].

ACS, acute coronary syndrome; ALB, albumin; ANS, albumin combined with neutrophil-to-lymphocyte ratio score; CAD, coronary artery disease; CGS, cardiogenic shock; CK-MB, creatine phosphokinase isoenzyme; Cr, creatinine; HF, heart failure; LA, left atrium; LVEDD, left ventricular end-diastolic diameter; LVEF, left ventricular ejection fraction; NSTEMI, non-ST-segment elevation myocardial infarction; PCI, percutaneous coronary intervention; STEMI, ST-segment elevation myocardial infarction; UA, unstable angina; WBC, white blood cell.

An elevated ANS was associated with increased white blood cell and neutrophil counts (both P < 0.05). There was a greater decrease in platelet, lymphocytes and albumin in the high ANS group than in the low ANS group (all P < 0.05). The patient characteristics in the MACEs and non-MACEs groups are shown in Table 2. The MACEs group was more likely to be older than 65 years and have a history of ischemic stroke, heart failure or cardiogenic shock (all P < 0.05). The prevalence of unstable angina was also higher in the MACEs group than in the non-MACEs group, although the difference was NS (P = 0.014). The neutrophil count and creatinine levels were higher and the lymphocyte count and albumin levels were lower in the MACEs group (all P < 0.05). There were also significant differences in left atrial diameter and LVEF of <40% between the MACEs and the non-MACEs groups (all P < 0.05).

Table 2.

Comparison of baseline patient characteristics between the major adverse cardiovascular events and nonmajor adverse cardiovascular events groups

Variables	Original cohort (n = 1549)		χ2/Z	P value	Matched cohort (n = 137)		χ2/Z/t	P value
	MACEs (n = 61)	Non-MACEs (n = 1488)			MACEs (n = 48)	Non-MACEs (n = 89)
Clinicodemographic data
Male (%)	44 (72.1)	1114 (74.9)	0.232	0.630	34 (70.8)	66 (74.2)	0.175	0.676
Age ≥65 years (%)	30 (49.2)	339 (22.8)	22.502	<0.001	25 (52.1)	46 (51.7)	0.002	0.965
Smoking (%)	29 (47.5)	776 (52.2)	0.499	0.480	21 (43.8)	40 (44.9)	0.018	0.893
Dyslipidemia (%)	36 (59.0)	844 (56.7)	0.126	0.723	31 (64.6)	54 (60.7)	0.202	0.653
Hypertension (%)	35 (57.4)	876 (58.9)	0.054	0.816	28 (58.3)	50 (56.2)	0.059	0.808
Diabetes (%)	14 (23.0)	379 (25.5)	0.196	0.658	13 (27.1)	24 (27.0)	<0.001	0.988
Ischemic stroke (%)	16 (26.2)	204 (13.7)	7.538	0.006	12 (25.0)	14 (15.7)	1.743	0.187
HF (%)	19 (31.1)	136 (9.1)	31.517	<0.001	14 (29.2)	6 (6.7)	12.578	<0.001
CGS (%)	8 (13.1)	16 (1.1)	55.683	<0.001	5 (10.4)	0 (0.0)	9.622	0.002
Family history of CAD (%)	5 (8.2)	212 (14.2)	1.781	0.182	4 (8.3)	3 (3.4)	1.584	0.208
UA (%)	15 (24.6)	599 (40.3)	6.010	0.014	10 (20.8)	28 (31.5)	1.757	0.185
STEMI (%)	33 (54.1)	656 (44.0)	2.379	0.123	27 (56.3)	41 (46.0)	1.293	0.255
NSTEMI (%)	13 (21.3)	233 (15.7)	1.402	0.236	11 (22.9)	20 (22.5)	0.004	0.953
Laboratory data
WBC count (109/l)	8.55 [7.16–11.24]	7.89 [6.34–10.32]	−1.714	0.087	8.48 [7.39–10.97]	7.86 [6.18–10.04]	−1.412	0.158
Platelet count (109/l)	212.00 [177.50–241.50]	215.00 [179.00–252.00]	−0.729	0.466	203.00 [176.00–244.25]	203.00 [178.50–243.50]	−0.345	0.730
Neutrophil count (109/l)	6.48 [4.69–9.21]	5.39 [3.93–8.00]	−2.529	0.011	6.42 [4.95–8.90]	5.39 [4.06–7.83]	−1.805	0.071
Lymphocyte count (109/l)	1.37 [0.90–1.77]	1.68 [1.25–2.29]	−3.465	0.001	1.30 [0.89–1.74]	1.64 [1.17–2.25]	−2.182	0.029
Monocyte count (109/l)	0.47 [0.32–0.61]	0.43 [0.32–0.57]	−0.815	0.415	0.49 [0.34–0.61]	0.40 [0.32–0.54]	−1.327	0.185
ALB (g/dl)	3.98 [3.71–4.20]	4.13 [3.87–4.36]	−3.305	0.001	3.93 ± 0.33	4.13 ± 0.31	3.525	0.001
Cr (μmol/l)	75.50 [63.05–88.70]	67.07 [58.78–77.78]	−3.198	0.001	76.56 [63.23–88.70]	67.18 [54.90–76.97]	−2.899	0.004
Serum uric acid (μmol/l)	333.10 [270.00–390.85]	326.00 [264.70–384.53]	−0.660	0.509	336.78 ± 93.96	311.99 ± 86.04	−1.558	0.122
Echocardiography
LA (mm)	35.00 [33.00–39.75]	35.00 [32.00–37.00]	−2.212	0.027	37.00 [33.25–40.00]	34.00 [31.50–37.00]	−2.595	0.009
LVEDD (mm)	50.00 [47.00–57.00]	50.00 [47.00–54.00]	−0.264	0.792	50.00 [47.00–57.00]	51.00 [48.00–54.00]	−0.398	0.691
LVEF < 40% (%)	8 (14.3)	33 (2.5)	26.467	<0.001	8 (16.7)	1 (1.1)	12.274	<0.001
Coronary angiogram
1 vessel (%)	16 (26.3)	469 (31.5)	0.762	0.383	12 (25.0)	23 (25.8)	0.012	0.914
2 vessel (%)	21 (34.4)	473 (31.8)	0.188	0.665	18 (37.5)	22 (24.7)	2.464	0.116
3 vessel (%)	24 (39.3)	546 (36.7)	0.177	0.674	18 (37.5)	44 (49.5)	1.794	0.180

Data are presented as n (%) or median [range].

ALB, albumin; CAD, coronary artery disease; CGS, cardiogenic shock; CK-MB, creatine phosphokinase isoenzyme; Cr, creatinine; HF, heart failure; LA, left atrium; LVEDD, left ventricular end-diastolic diameter; LVEF, left ventricular ejection fraction; MACEs, major adverse cardiovascular events; NSTEMI, non-ST-segment elevation myocardial infarction; STEMI, ST-segment elevation myocardial infarction; UA, unstable angina; WBC, white blood cell.

Propensity-matched analysis

After PSM, 137 matched patients were included in the analysis. Among them, 48 patients belonged to the MACEs group (43 patients died from any cause and five patients required rehospitalization for severe heart failure) and while 89 patients belonged to the non-MACEs group (Fig. 1). These patients were well matched in most of the baseline characteristics. There were no significant differences in age ≥65 years, heart failure, family history of CAD, unstable angina, platelet count, monocyte count, left ventricular end-diastolic diameter, LVEF < 40%, single-vessel coronary disease and triple-vessel coronary disease among the patients with ANS 0, 1 and 2 (all P > 0.05, Table 1). Meanwhile, there were significant differences in the proportion of male patients and the rates of smoking, dyslipidemia, unstable angina, STEMI and non-STEMI (all P < 0.05, Table 1). A high ANS was associated with high white blood cell count, neutrophil count and creatinine levels (all P < 0.05, Table 1). There was a greater decrease in lymphocyte count and albumin level in the high ANS group than in the low ANS group (both P < 0.05, Table 1). When analyzed according to the development of MACEs (Table 2), the MACEs group was significantly more likely to have a history of heart failure or cardiogenic shock (both P < 0.05). The creatinine level was higher and the lymphocyte count and albumin levels were lower in the MACEs group (all P < 0.05). There were significant differences in left atrial diameter and LVEF of <40% between the MACEs and the non-MACEs groups (all P < 0.05).

Survival analyses

The mean follow-up duration in the entire cohort was 2446 days. There were significant differences in overall survival (3 years: 98% vs. 97% vs. 91%; 5 years: 98% vs. 95% vs. 89%, P < 0.001) among the patients with ANS of 0, 1 and 2 (Fig. 2a). The Kaplan–Meier curves showed that a higher ANS value was associated with a higher risk of MACEs (log-rank P < 0.001; Fig. 2a). After PSM, the mean follow-up duration was 1469 days. There were significant differences in overall survival (1 year: 85% vs. 81% vs. 66%, 3 years: 85% vs. 74% vs. 39%, P < 0.001) among patients with ANS of 0, 1 and 2 (Fig. 2b). The Kaplan–Meier curves showed that a higher ANS value was associated with a higher risk of MACEs (log-rank P < 0.001; Fig. 2b).

Fig. 2.

Kaplan–Meier curves of cumulative survival by ANS in ACS patients undergoing PCI. Before propensity score matching (a). After propensity score matching (b). ACS, acute coronary syndrome; ANS, albumin combined with neutrophil-to-lymphocyte ratio score; PCI, percutaneous coronary intervention.

Cox proportional hazard regression analyses

Figure 3, Tables 3 and 4 present the results of the univariate and multivariate analysis using Cox proportional hazard regression models for the independent predictors for MACEs. In the entire cohort, age ≥65 years old [hazard ratio (HR), 3.295; 95% CI, 1.993–5.446; P < 0.001], cardiogenic shock (HR, 11.399; 95% CI, 5.399–24.066; P < 0.001), heart failure (HR, 4.624; 95% CI, 2.686–7.959; P < 0.001), ischemic stroke (HR, 2.206; 95% CI, 1.247–3.904; P = 0.007), LVEF < 40% (HR, 6.875; 95% CI, 3.241–14.583; P < 0.001), ANS (HR, 2.478; 95% CI, 1.676–3.665, P < 0.001) were influencing factors of MACEs in the univariate analysis. Multivariate analysis showed that age ≥65 years (HR, 2.706; 95% CI, 1.592–4.598; P < 0.001), cardiogenic shock (HR, 4.845; 95% CI, 1.863–12.600; P = 0.001), heart failure (HR, 2.080; 95% CI, 1.083–3.995; P = 0.028), ischemic stroke (HR, 1.842; 95% CI, 1.003–3.383; P = 0.049), LVEF < 40% (HR, 2.811; 95% CI, 1.159–6.815; P = 0.022) and ANS (HR, 2.016; 95% CI, 1.329–3.057; P = 0.001) were independent prognostic factors for MACEs (Fig. 3a and Table 3).

Fig. 3.

Forest graphs based on Cox proportional hazards regression models with ANS included as a risk factor for MACEs. (a) Model 1: entire cohort. (b) Model 2: propensity score matching. ANS, albumin combined with neutrophil-to-lymphocyte ratio score; MACEs, major adverse cardiovascular events.

Table 3.

Univariate and multivariate Cox proportional hazards regression analyses of predictive factors of major adverse cardiovascular events

Variables	Cohort (n = 1549)
	Univariate analysis		Multivariate analysis
	Hazard ratio (95% CI)	P value	Hazard ratio (95% CI)	P value
Age ≥65 years	3.295 (1.993–5.446)	<0.001	2.706 (1.592–4.598)	<0.001
CGS	11.399 (5.399–24.066)	<0.001	4.845 (1.863–12.600)	0.001
HF	4.624 (2.686–7.959)	<0.001	2.080 (1.083–3.995)	0.028
Ischemic stroke	2.206 (1.247–3.904)	0.007	1.842 (1.003–3.383)	0.049
LVEF < 40%	6.875 (3.241–14.583)	<0.001	2.811 (1.159–6.815)	0.022
ANS (per 1 score increase)	2.478 (1.676–3.665)	<0.001	2.016 (1.329–3.057)	0.001

ACS, acute coronary syndrome; ANS, albumin combined with neutrophil-to-lymphocyte ratio score; CGS, cardiogenic shock; HF, heart failure; LVEF, left ventricular ejection fraction; MACEs, major adverse cardiovascular events; PCI, percutaneous coronary intervention.

Table 4.

Univariate and multivariate Cox proportional hazards regression analyses of predictive factors of major adverse cardiovascular events after propensity score matching

Variables	Propensity score matching (n = 137)
	Univariate analysis		Multivariate analysis
	Hazard ratio (95% CI)	P value	Hazard ratio (95% CI)	P value
Male sex	1.294 (0.690–2.427)	0.421	2.489 (1.256–4.932)	0.009
HF	4.003 (2.127–7.532)	<0.001	2.454 (1.133–5.315)	0.023
Family history of CAD	2.646 (0.945–7.410)	0.064	4.723 (1.586–14.062)	0.005
LVEF < 40%	6.703 (3.044–14.762)	<0.001	4.215 (1.622–10.954)	0.003
Triple-vessel coronary disease	1.474 (0.821–2.646)	0.194	1.978 (1.089–3.593)	0.025
ANS (per 1 score increase)	2.239 (1.416–3.540)	0.001	2.166 (1.344–3.492)	0.002

ACS, acute coronary syndrome; ANS, albumin combined with neutrophil-to-lymphocyte ratio score; CAD, coronary artery disease; HF, heart failure; LVEF, left ventricular ejection fraction; MACEs, major adverse cardiovascular events; PCI, percutaneous coronary intervention.

After PSM, the results demonstrated that male sex (HR, 1.294; 95% CI, 0.690–2.427; P = 0.421), heart failure (HR, 4.003; 95% CI, 2.127–7.532; P < 0.001), family history of CAD (HR, 2.646; 95% CI, 0.945–7.410; P = 0.064), LVEF < 40% (HR, 6.703; 95% CI, 3.044–14.762; P < 0.001), triple-vessel coronary disease (HR, 1.474; 95% CI, 0.821–2.646; P = 0.194) and ANS (HR, 2.239; 95% CI, 1.416–3.540; P = 0.001) were influencing factors of MACEs in the univariate analysis. In the multivariate analysis, male sex (HR, 2.489; 95% CI, 1.256–4.932; P = 0.009), heart failure (HR, 2.454; 95% CI, 1.133–5.315; P = 0.023), family history of CAD (HR, 4.723; 95% CI, 1.586–14.062; P = 0.005), LVEF < 40% (HR, 4.215; 95% CI, 1.622–10.954; P = 0.003), triple-vessel coronary disease (HR, 1.978; 95% CI, 1.089–3.593; P = 0.025) and ANS (HR, 2.166; 95% CI, 1.344–3.492; P = 0.002) were independent predictors of MACEs (Fig. 3b and Table 4).

Discussion

The prognostic value of the ANS in patients with ACS undergoing PCI has not been clarified to date. This study found that a high ANS was an independent risk factor for predicting MACEs. Patients with a high ANS had a greater risk of clinically adverse cardiovascular events than patients with no risk factors.

Randomized trials are deemed to be the most scientifically rigorous study design to investigate the efficacy of treatment while minimizing systematic bias [20]. PSM analysis is statistical method of adjusting for potential baseline confounding variables to simulate a randomized controlled trial design. PSM analysis has increasingly appeared in cardiovascular research [20–23]. However, we consider that the advantages of one method do not outweigh the other. Therefore, we used Cox proportional hazard regression model analysis and propensity scoring methods control the confounding variables. To our best knowledge, the present study is the first to use both Cox proportional hazard regression analyses and PSM to investigate the association between the ANS and the clinical outcomes of ACS patients undergoing PCI.

Several reports have recently shown that hypoalbuminemia often occurs due to inflammation, but can also be caused by hepatocyte damage and low albumin synthesis, dietary insufficiency of amino acids or high excretion of albumin [24]. In a prospective study, Eckart et al. [25] found that the status of low serum albumin in patients with acute illness was significantly affected by inflammation. Bretschera et al. [26] reported that albumin concentrations should be used to assess disease severity, but not to assess nutritional status or diagnose malnutrition. Thus, low serum albumin levels also appear to be associated with a high risk of CAD [25,27]. In the current study, the albumin level was lower in the MACEs group both before and after PSM (Table 2). Our results seem to support the above opinion.

Numerous studies have shown the association between biomarkers of inflammation and cardiovascular events induced by atheroma [28,29]. In clinical practice, complete blood cell counts have been widely used to reflect the inflammatory status of patients owing to its convenience and economy [30–32]. Our previous study found that the systemic immune-inflammatory index (formula: platelet count × neutrophil count/lymphocyte count) and derived neutrophil-lymphocyte ratio [formula: neutrophil count/(white blood cell count – neutrophil count)] had similar diagnostic values for MACEs [19]. These elevated markers were linked to aggravated inflammation and poor ACS outcomes.

Albumin combined with NLR score was first reported to be an inflammatory marker in a prognostic study of oral cavity squamous cell carcinoma [33]. There are several novel studies found that albumin and the NLR play critical roles in the immune and systemic inflammation status of patients. Huang et al. [34] reported that albumin and NLR were associated with the overall survival of patients with thymic epithelial tumors. Li et al. [35] demonstrated that low levels of albumin and high levels of NLR were identified as risk factors for poor prognoses patients who underwent surgical treatment for endometrial cancer. Therefore, we concluded that ANS combined with albumin and NLR increased the prognostic value of inflammatory biomarkers.

Based on our previous study, we used the same method to determine the cutoff points for albumin and NLR were 4.07 g/dl and 2.60, respectively [18]. Kaplan–Meier analysis both before and after PSM indicated that the ANS was a powerful prognostic factor for the clinical outcomes of ACS patients undergoing PCI (Fig. 2). In the entire cohort, multivariate Cox proportional hazard regression analyses showed that the HR of the ANS in patients with ACS undergoing PCI per 1 score increase was 2.016 (95% CI, 1.329–3.057) (Table 3, Fig. 3a). After PSM, multivariate Cox proportional hazards regression analyses showed that the HR of the ANS in patients with ACS patients undergoing PCI per 1 score increase was 2.166 (95% CI, 1.344–3.492) (Table 4, Fig. 3b). These findings supported that a higher ANS was significantly associated with an increased risk of MACEs in patients with ACS patients undergoing PCI.

Study limitations

This study has some limitations. First, our data were from a single center, and selection bias may have occurred. Second, although we used PSM analysis, inconsistencies in the baseline characteristics and bias are possible. Third, as postdischarge ANS changes were not recorded, we cannot entirely rule out the possibility of residual or unmeasured confounding factors. Fourth, our sample size was relatively small. The findings should therefore be interpreted with caution given these limitations. More high-quality prospective studies are needed in the future.

Conclusion

A higher ANS is independently associated with a higher risk of MACEs in patients with ACS undergoing PCI. ANS may be a valuable clinical indicator to identify high-risk ACS patients after PCI.

Acknowledgements

The authors are grateful for the assistance of doctors and nurses of the cardiology research team at the Affiliated Hospital of Chengde Medical University.

This work was supported by the Natural Science Foundation of Hebei Province (Grant number H2021406071) to L.S. and the Government funded clinical medicine talent training project (Grant number ZF2023252) to Y.Z.

All authors contributed to preparing the manuscript, and approved the final version. Authors contributed as follows. Study design: C.W., W.F. and L.S.. Acquisition of data: Y.Z.. Analysis and interpretation of data: Q.S., Y.L., X.W. and J.L. Drafting of the manuscript: C.W. and L.S. Critical revision of the manuscript for important intellectual content: C.W. and L.S. Statistical analysis: C.W. and L.S. Supervision: L.S.

The authors have no ethical conflicts to disclose. This prospective single-center cohort study was approved by the Institutional Review Board of the Affiliated Hospital of Chengde Medical University (Number: LL2021036) and was conducted in accordance with the Declaration of Helsinki. All participants provided written informed consent.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Conflicts of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.