The association between blood count based inflammatory markers and the risk of atrial fibrillation heart failure and cardiovascular mortality

Abstract

The prevalence and progression of cardiovascular disease (CVD) are significantly influenced by low-grade chronic inflammation. Our aim was to investigate the association of blood-count-based inflammatory markers with atrial fibrillation (AF), heart failure (HF), and cardiovascular mortality. We utilized prospective follow-up data from the UK Biobank, including 334,674 individuals (aged between 39 and 70 years) free of HF and AF at baseline. The exposures were blood-count-based inflammatory markers, including neutrophil-to-lymphocyte ratio (NLR), monocyte-to-lymphocyte ratio (MLR), systemic immune-inflammation index (SII), aggregate index of systemic inflammation (AISI), and systemic inflammatory response index (SIRI), which were computed using leukocyte counts (lymphocytes, monocytes, and neutrophils). The primary outcomes were the occurrence of AF, HF and cardiovascular-related deaths. Cox proportional hazards models were employed to determine hazard ratios (HRs) and their 95% confidence intervals (CIs) for assessing the longitudinal association between inflammatory markers and the development of cardiovascular events. Survival analysis was illustrated using Kaplan–Meier curves with the occurrence of AF, HF and CVD mortality as the endpoint. There were 16,994 incidences of AF and 7342 incidences of HF over a median follow-up of 10.4 years (range, 2.4–12.1). Additionally, there were 3434 deaths from CVD causes. A higher risk of cardiovascular mortality was significantly associated with NLR, MLR, SIRI, AISI, and SII in multivariable adjusted models (HR 1.44, 95% CI 1.30–1.58; HR 1.46, 95% CI 1.31–1.61; HR 1.57, 95% CI 1.41–1.75; HR 1.57, 95% CI 1.41–1.73; and HR 1.36, 95% CI 1.24–1.49, respectively). Furthermore, our analysis revealed positive associations between NLR, MLR, SIRI, and AISI and the risk for AF (HR 1.18, 95% CI 1.13–1.24; HR 1.19, 95% CI 1.14–1.25; HR 1.22, 95% CI 1.16–1.27; and HR 1.08, 95% CI 1.03–1.13, respectively), and HF (HR 1.42, 95% CI 1.33–1.52; HR 1.28, 95% CI 1.19–1.37; HR 1.45, 95% CI 1.35–1.56; and HR 1.32, 95% CI 1.23–1.41, respectively). Furthermore, the SII demonstrated a positive association with HF (HR 1.24, 95% CI 1.16–1.32). However, the association with AF was not statistically significant (HR 1.01, 95% CI 0.97–1.06). In summary, our research provides preliminary evidence that blood cell indices associated with the systemic inflammatory response, may serve as potential biomarkers for the risk of developing AF, HF and CVD mortality. Additionally, it is likely that a systemic inflammatory-immune response existed before the clinical diagnosis. Identifying high-risk populations based on the levels of inflammatory markers could help reduce the risk of death and prevent the onset of AF and HF. Further investigation is warranted to determine whether reducing inflammatory markers causally improves cardiovascular outcomes.

Introduction

Despite significant advances in the prevention and treatment of AF and HF, the prognosis in patients who suffer from embolic complications with AF and those who have been hospitalised on at least one occasion due to exacerbation of HF is still poor¹. It is increasingly recognized that the inflammatory-immune system plays a role in the pathophysiology of cardiovascular disease (CVD), including atherosclerosis², AF³, and HF^4,5. Research has demonstrated the independent association between a higher risk of incident CVD and unfavorable clinical outcomes, and circulating immune-inflammatory indicators such as C-reactive protein (CRP), interleukin 6 (IL-6), and leukocyte count^6–8. Recently, there has been interest in using inflammatory indicators as immunotherapeutic targets to prevent CVD or improve CVD outcomes^9,10.

Novel indicators of the systemic inflammatory response based on leukocyte counts (lymphocytes, monocytes, and neutrophils) have been proposed. These indicators are involved in the diagnosis and prognosis of a variety of inflammatory diseases^11–13. The blood-count-based inflammatory markers include the neutrophil-to-lymphocyte ratio (NLR), monocyte-to-lymphocyte ratio (MLR), systemic inflammation index (SII), systemic inflammatory response index (SIRI), and the aggregate index of systemic inflammation (AISI). For individuals who had ST elevation myocardial infarction¹⁴ or isolated CABG (coronary artery bypass graft surgery)¹⁵, SII may be one of the independent predictors of new-onset AF. Additionally, NLR was shown to be an independent predictor of all-cause mortality, long-term unfavorable outcomes in elderly patients with chronic HF and AF, and new-onset AF following surgery^16–18. The probability of death from all causes increased with elevated SII and SIRI¹⁹. Despite the fact that these ratios have been linked to both mortality and AF risk, most of these studies have looked into their predictive role in exceptional individuals, and it is still debatable or unclear what effect they have in predicting AF or HF in healthy populations^20,21. Discrepancies observed in previous findings are likely attributable to methodological concerns or flaws in study design. Additionally, a limited sample size and imbalanced baseline characteristics may contribute to an underestimation of the actual effect.

At present, there exists a paucity of understanding regarding the relationship between blood-count-based inflammatory markers and the onset of AF, HF, as well as the risk of CVD mortality within the general population. The objective of this prospective cohort study is to investigate these associations utilizing a substantial multi-center database, specifically the UK Biobank.

Study design and methods

Study design and participants

The UK Biobank project is a prospective cohort study that recruited about 500,000 participants from all over the United Kingdom between 2006 and 2010. The participants, aged between 37 and 73²², completed a touch-screen questionnaire during the recruitment process. This questionnaire collected data on lifestyle, health-related factors, and sociodemographic characteristics, and participants also provided their digitally signed consent. Additionally, participants provided blood samples and underwent various anthropometric assessments. The study monitored participants’ morbidity and mortality, linking health and medical data to provide follow-up information²³. The key outcomes of the study included the incidence of AF, HF, and CVD mortality. The International Classification of Diseases, 10th version (ICD-10), was used to code each participant’s diagnosis, as detailed in (Supplementary Table S2). CVD mortality was defined as a composite of death from disorders of the heart (I00-I09, I11, I13, I20-I51). Participants who died from CVD were excluded from the study at the time of death, while those who did not die from CVD were excluded either at the time of non-cardiovascular death or at the end of the study period. Mortality data were provided by the National Health Service (NHS) Central Register Scotland (Scotland) and the NHS Information Center (England and Wales). The survival period of participants was measured from the date of inflammatory-immune marker detection to the date of death, loss to follow-up, or the latest follow-up date (August 31, 2017), whichever came first.

We did not include the following individuals in the current study: (1) those with hematologic or autoimmune diseases, malignant cancer, renal failure, liver failure, pregnancy, or taking corticotrophin or glucocorticoids at baseline; (2) those with AF or HF at baseline; (3) those with C-reactive protein levels > 10 mg/L; (4) those with missing data for variables. In conclusion, a total of 334,674 individuals qualified for further examinations (Fig. 1).

Fig. 1.

Flow diagram for participants’ recruitment.

Systemic inflammatory indexes measurements

Participants’ peripheral blood samples were examined using a Beckman Coulter LH750 Hematology Analyzer at the UK Biobank central laboratory within 24 h after blood collection. The device reported thirty-one parameters, from which distinct populations of blood cells were extracted⁸. Five systemic inflammatory markers were determined based on counts of peripheral blood cells (platelets, neutrophils, lymphocytes, and monocytes): NLR = neutrophils/lymphocytes; MLR = monocytes/lymphocytes; SII = (neutrophils*platelets)/lymphocytes; SIRI = (neutrophils*monocytes)/lymphocytes; and AISI = (neutrophils*monocytes*platelets) /lymphocytes.

Covariates

The study’s covariates included: (1) sociodemographics, such as age, gender, and ethnicity; (2) biological factors, such as BMI, diastolic and systolic blood pressure; (3) behavioral factors, such as alcohol consumption, smoking, dietary habits and sedentary time; (4) socioeconomic factors, such as the Townsend deprivation index; (5) medication use, such as antihypertensive, lipid-lowering and anti-platelet medications; and (6) chronic health conditions, such as diabetes mellitus and coronary heart disease. Table S1 contains detailed explanations of each covariate.

Statistical analysis

The distribution of traditional risk factors, including age, gender, ethnicity, body mass index (BMI), systolic blood pressure (SBP), deprivation score, smoking habits, alcohol consumption, and pre-existing chronic conditions, was analyzed according to the quartiles of NLR, MLR, SII, SIRI, and AISI. Utilizing restricted cubic splines for each marker, constrained to values within their mean plus or minus four standard deviations, we examined the relationship between five systemic inflammation markers and the corresponding outcomes. Our analysis indicated that the quartile effectively encapsulated the data.

We used the χ2 test for categorical variables (reported as percentages) and the Wilcoxon rank-sum test for continuous data (presented as mean ± SD) to compare baseline characteristics. Hazard ratios (HR) and corresponding 95% confidence intervals (95% CIs) were employed in the Cox proportional hazards model to analyze the association between circulating immunological markers and the occurrence of AF, HF, and CVD mortality. Two incremental models were constructed: model 1 was unadjusted, while model 2 (fully adjusted) considered variables such as age, gender, ethnicity, alcohol consumption, smoking status, Townsend Index, BMI, systolic and diastolic blood pressure, lipid-lowering medications, and antihypertensive medications. The lowest quartile group served as the reference category. The survival probability of inflammatory indexes derived from complete blood count (CBC) was calculated using Kaplan–Meier curves, with death as the endpoint. By excluding participants whose outcome events occurred within two years of their recruitment, we conducted a sensitivity analysis to minimize the potential impact of reverse causality on the study outcomes. Furthermore, sensitivity analyses were performed to validate the robustness of the correlations. R version 4.1.0 and Stata 14 (StataCorp., College Station, TX, USA) were used for all statistical analyses. A p-value less than 0.05 on both sides was considered statistically significant.

Institutional review board approval

Under application number 60651, the UK Biobank resource was utilized for this study. The research protocol is available online and has been approved by the North West Multi-Centre Research Ethics Committee in the United Kingdom. All participants have provided electronic consent for the baseline and follow-up evaluations, which was conducted in accordance with the principles of the Declaration of Helsinki.

Results

General characteristics of the study population in the UK Biobank

The study included 334,674 participants from the UK Biobank who were initially free of AF and HF. Over a median follow-up period of 12.3 years, 16,994 participants (5.1%) developed AF. These individuals tended to be older men of white ethnicity, obese, frequent drinkers, unhealthy eaters, and had comorbidities like diabetes, coronary heart disease, and hypertension, but were less likely to engage in physical exercise. Additionally, 7342 participants (2.2%) developed HF, sharing similar characteristics with those who developed AF, except for having higher Townsend scores, being of non-white ethnicity, and being current smokers. Additionally, among the 3434 participants (1.0%) who experienced cardiovascular mortality, the baseline characteristics were predominantly similar to those of individuals with AF. Notably, this group demonstrated a higher prevalence of current smoking and more advantageous socioeconomic conditions. Furthermore, there was no statistically significant difference in alcohol consumption between the groups classified by cardiovascular mortality and those who did not experience such outcomes.. It is important to highlight that individuals diagnosed with AF, HF, or cardiovascular death demonstrated a greater likelihood of utilizing chronic disease medications, such as anti-hypertensive drugs (all P < 0.01) as shown in (Table 1).

Table 1.

Baseline characteristics of participants in the UK Biobank study.

	Incident atrial fibrillation			Incident heart failure			CVD mortality
	Yes	No	P value	Yes	No	P value	Yes	No	P value
	(n = 16994)	(n = 317680)		(n = 7342)	(n = 327332)		(n = 3434)	(n = 331240)
Characteristics
Age (years)	62.3 (6.1)	56.3 (8.1)	< 0.001	62.2 (6.4)	56.5 (8.1)	< 0.001	62.2 (6.4)	56.6 (8.1)	< 0.001
Sex, male (%)	10801 (63.6)	142780 (44.9)	< 0.001	4755 (64.8)	148826 (45.5)	< 0.001	2421 (70.5)	151160 (45.6)	< 0.001
Race, white ethnicity (%)	16602 (97.7)	301827 (95.0)	< 0.001	7007 (95.4)	311422 (95.1)	0.252	3311 (96.4)	315118 (95.1)	0.001
Body mass index (kg/m2)	28.5 (5.0)	27.1 (4.5)	< 0.001	29.2 (5.3)	27.1 (4.5)	< 0.001	28.7 (5.3)	27.1 (4.5)	< 0.001
Townsend deprivation index	−1.4 (3.0)	−1.5 (3.0)	0.269	−1.0 (3.2)	−1.5 (3.0)	< 0.001	−0.9 (3.3)	−1.5 (3.0)	< 0.001
Current smoker, n (%)	1728 (10.2)	31456 (9.9)	0.263	1103 (15.0)	32081 (9.8)	< 0.001	720 (21.0)	32464 (9.8)	< 0.001
Drinking ≥ 3 times/week (%)	8280 (48.7)	143347 (45.1)	< 0.001	3225 (43.9)	148402 (45.3)	0.017	1552 (45.2)	150075 (45.3)	0.909
Ideal healthy diet, n (%)	9723 (57.2)	178025 (56.0)	0.003	3987 (54.3)	183761 (56.1)	0.002	1812 (52.8)	185936 (56.1)	< 0.001
Sedentary behavior (h)	5.1 (2.4)	4.7 (2.4)	< 0.001	5.3 (2.7)	4.8 (2.4)	< 0.001	5.2 (2.8)	4.8 (2.4)	< 0.001
Diabetes (%)	610 (3.6)	4319 (1.4)	< 0.001	530 (7.2)	4399 (1.3)	< 0.001	256 (7.5)	4673 (1.4)	< 0.001
Coronary heart disease (%)	1290 (7.6)	7763 (2.4)	< 0.001	1009 (13.7)	8044 ( 2.5)	< 0.001	428 (12.5)	8625 ( 2.6)	< 0.001
Systolic blood pressure(mmHg)	143.7 (19.1)	137.2 (18.5)	< 0.001	144.6 (19.8)	137.4 (18.5)	< 0.001	146.4 (20.1)	137.5 (18.5)	< 0.001
Diastolic blood pressure (mmHg)	83.0 (10.5)	82.1 (10.1)	< 0.001	82.9 (10.9)	82.2 (10.1)	< 0.001	84.1 (11.3)	82.2 (10.1)	< 0.001
Antihypertensive drugs (%)	6668 (39.2)	55940 (17.6)	< 0.001	3410 (46.4)	59198 (18.1)	< 0.001	1507 (43.9)	61101 (18.4)	< 0.001
Cholesterol-lowering drugs (%)	5523 (32.5)	47764 (15.0)	< 0.001	3027 (41.2)	50260 (15.4)	< 0.001	1358 (39.5)	51929 (15.7)	< 0.001
Aspirin (%)	4825 (28.4)	38133 (12.0)	< 0.001	2617 (35.6)	40341 (12.3)	< 0.001	1151 (33.5)	41807 (12.6)	< 0.001

CVD cardiovascular disease.

Systemic inflammation markers and cardiovascular disease risk

In the initial risk assessment, positive correlations were identified between NLR, MLR, SIRI, and AISI levels and the likelihood of developing AF. The results obtained from Cox proportional hazards models indicated that, in comparison to the lowest quartile (quartile 1), the fully adjusted Hazard Ratio (HR) values for quartile 4 were 1.18 (95% Confidence Interval [CI] 1.13–1.24) for NLR, 1.19 (95% CI 1.14–1.25) for MLR, 1.22 (95% CI 1.16–1.27) for SIRI, and 1.08 (95% CI 1.03–1.13) for AISI. Furthermore, initial observations also revealed positive associations between NLR, MLR, SII, SIRI, and AISI levels and the risk of HF. Upon adjusting for various factors including age, sex, race, lifestyle habits, medical history, and medication use, the fully adjusted HR values for quartile 4 were 1.42 (95% CI 1.33–1.52) for NLR, 1.28 (95% CI 1.19–1.37) for MLR, 1.24 (95% CI 1.16–1.32) for SII, 1.45 (95% CI 1.35–1.56) for SIRI, and 1.32 (95% CI 1.23–1.41) for AISI, respectively. Similar risk assessments were also evident when examining the associations of these systemic inflammation markers with CVD mortality in both unadjusted and fully adjusted models (Table 2).

Table 2.

Associations between the inflammatory-immune markers with atrial fibrillation, heart failure and CVD mortality.

	Atrial fibrillation(n = 16,994)				Heart failure(n = 7342)				CVD mortality(n = 3434)
	Model 1		Model 2		Model 1		Model 2		Model 1		Model 2
	HR (95% CI)	P	HR (95% CI)	P	HR (95% CI)	P	HR (95% CI)	P	HR (95% CI)	P	HR (95% CI)	P
NLR
1st	1.00		1.00		1.00		1.00		1.00		1.00
2nd	1.07 (1.02, 1.12)	0.006	1.02 (0.97,1.06)	0.517	1.14 (1.06, 1.22)	< 0.001	1.10 (1.02, 1.18)	0.010	1.14 (1.02, 1.26)	0.020	1.04 (0.93, 1.16)	0.493
3rd	1.19 (1.14, 1.24)	< 0.001	1.09 (1.04, 1.13)	< 0.001	1.28 (1.19, 1.37)	< 0.001	1.19 (1.11, 1.27)	< 0.001	1.36 (1.23, 1.51)	< 0.001	1.17 (1.05, 1.29)	0.004
4th	1.36 (1.30,1.42)	< 0.001	1.18 (1.13, 1.24)	< 0.001	1.62 (1.51, 1.72)	< 0.001	1.42 (1.33, 1.52)	< 0.001	1.93 (1.75, 2.12)	< 0.001	1.44 (1.30, 1.58)	< 0.001
MLR
1st	1.00		1.00		1.00		1.00		1.00		1.00
2nd	1.12 (1.06, 1.17)	< 0.001	1.02 (0.97, 1.07)	0.400	1.03 (0.95, 1.10)	0.502	0.98 (0.91, 1.05)	0.584	1.19 (1.06, 1.33)	0.002	1.08 (0.96, 1.21)	0.191
3rd	1.28 (1.23, 1.34)	< 0.001	1.08 (1.03, 1.14)	< 0.001	1.29 (1.11, 1.28)	< 0.001	1.06 (0.99, 1.14)	0.117	1.49 (1.34, 1.65)	< 0.001	1.14 (1.03, 1.28)	0.016
4th	1.56 (1.49, 1.63)	< 0.001	1.19 (1.14, 1.25)	< 0.001	1.57 (1.47, 1.68)	< 0.001	1.28 (1.19, 1.37)	< 0.001	2.40 (2.17, 2.64)	< 0.001	1.46 (1.31, 1.61)	< 0.001
SII
1st	1.00		1.00		1.00		1.00		1.00		1.00
2nd	0.93 (0.89, 0.97)	0.001	0.93 (0.89, 0.97)	< 0.001	0.98 (0.92, 1.04)	0.550	1.00 (0.93, 1.06)	0.887	0.97 (0.87, 1.07)	0.488	0.98 (0.89, 1.09)	0.744
3rd	0.98 (0.94, 1.02)	0.263	0.97 (0.93, 1.01)	0.165	1.08 (1.01, 1.15)	0.020	1.09 (1.02, 1.17)	0.008	1.04 (0.95, 1.15)	0.398	1.06 (0.96, 1.17)	0.223
4th	1.04 (1.00, 1.09)	0.062	1.01 (0.97, 1.06)	0.576	1.27 (1.19, 1.35)	< 0.001	1.24 (1.16, 1.32)	< 0.001	1.39 (1.27, 1.52)	< 0.001	1.36 (1.24, 1.49)	< 0.001
SIRI
1st	1.00		1.00		1.00		1.00		1.00		1.00
2nd	1.13 (1.08, 1.19)	< 0.001	1.02 (0.97, 1.07)	0.411	1.21 (1.12, 1.31)	< 0.001	1.10 (1.01, 1.19)	0.020	1.30 (1.16, 1.47)	< 0.001	1.06 (0.94, 1.19)	0.358
3rd	1.34 (1.29, 1.40)	< 0.001	1.10 (1.05, 1.16)	< 0.001	1.48 (1.37, 1.59)	< 0.001	1.21 (1.12, 1.30)	< 0.001	1.80 (1.61, 2.02)	< 0.001	1.23 (1.10, 1.37)	< 0.001
4th	1.67 (1.60, 1.74)	< 0.001	1.22 (1.16, 1.27)	< 0.001	2.07 (1.94, 2.22)	< 0.001	1.45 (1.35, 1.56)	< 0.001	3.09 (2.79, 3.42)	< 0.001	1.57 (1.41, 1.75)	< 0.001
AISI
1st	1.00		1.00		1.00		1.00		1.00		1.00
2nd	1.04 (1.00, 1.09)	0.069	0.98 (0.94, 1.03)	0.378	1.10 (1.03, 1.19)	0.008	1.05 (0.97, 1.12)	0.235	1.30 (1.17, 1.46)	< 0.001	1.14 (1.02, 1.28)	0.020
3rd	1.16 (1.11, 1.21)	< 0.001	1.02 (0.97, 1.07)	0.420	1.27 (1.18, 1.36)	< 0.001	1.11 (1.03, 1.19)	0.004	1.60 (1.43, 1.78)	< 0.001	1.24 (1.12, 1.39)	< 0.001
4th	1.34 (1.28, 1.40)	< 0.001	1.08 (1.03, 1.13)	< 0.001	1.72 (1.61, 1.83)	< 0.001	1.32 (1.23, 1.41)	< 0.001	2.43 (2.20, 2.69)	< 0.001	1.57 (1.41, 1.73)	< 0.001

Model 1 unadjusted.

Model 2 was adjusted for age,sex, race, smoking, alcohol consumption,sedentary time, healthy diet pattern, body mass index, Townsen index, systolic blood pressure, diastolic blood pressure, diabetes mellitus, coronary heart disease, anti-hypertension drugs, cholesterol lower drugs, and aspirin.

NLR neutrophil to lymphocyte ratio, MLR monocyte to lymphocyte ratio, SII systemic inflammation index, SIRI systemic inflammation response index, AISI the aggregate index of systemic inflammation, CVD cardiovascular disease.

In the context of survival analysis, we utilized four knots to model (fully adjusted) and subsequently visualized the relationships between systemic inflammation markers and the probability of survival by using Kaplan–Meier curves. We observed inverse relationships, except for the association between SII and AF as illustrated in (Fig. 2).

Fig. 2.

Kaplan–Meier curves stratified by inflammatory markers quartiles. CVD cardiovascular diseases.

Sensitivity analyses

The sensitivity analyses, which involved the exclusion of new cases occurring within a two-year period following spirometry, revealed that the systemic inflammation markers maintained similar patterns of association with the risk of AF, HF, and CVD mortality, as demonstrated in (Tables S3).

Discussion

In this longitudinal study of 334,674 adult subjects who had no prior history of AF or HF at baseline, heightened levels of systemic inflammation indicators (NLR, MLR, SIRI, and AISI) were linked to a higher likelihood of developing AF, HF, and cardiovascular mortality.

Our results were in general agreement with previous studies of the relationship between inflammatory biomarkers and the cardiovascular outcomes. NLR and MLR are simple, rapid, nonspecific, and cost-effective methods for identifying systemic inflammation, served as indicators of innate immune responses that are primarily driven by acute myeloid activity. These measures have been associated with chronic, lymphocyte-mediated immunological memory, as evidenced by lymphocyte counts. An elevation in both MLR and NLR may be associated with a reduction in lymphocyte levels, and signify an immunological dysregulation, potentially indicative of an active or subclinical acute inflammatory process, alongside a compromised immune response to pathogens^24,25. As emerging biomarkers, the superiority of SII, SIRI and AISI have been identified in numerous diseases, including cancers, cardiovascular diseases, sacroiliitis and diabetic nephropathy. In comparison to conventional inflammatory markers, these three systemic inflammation metrics have been shown to provide more reliable assessments of inflammatory status and exhibit enhanced predictive capabilities and prognostic significance in numerous studies²⁶. Research has shown a non-linear positive association between systemic inflammation markers such as SIRI, AISI and the occurrence of paroxysmal atrial fibrillation (PAF), and these markers have also exhibited favorable sensitivity and specificity in identifying the presence of PAF²⁷. SIRI is also a convenient and effective measurement for predicting the presence of AF in patients with ischemic stroke²⁸. MLR, NLR and SII demonstrated the ability to forecast the recurrence of AF following surgical intervention²⁹. In patients undergoing elective off-pump coronary artery bypass grafting (OPCABG), a study revealed that NLR, and SII were independent predictors of new-onset atrial fibrillation (NOAF), which highlight the important role of inflammation in the development of post-OPCABG arrhythmic problems³⁰. Additionally, by pooling data from 12 trials and 9,262 patients, a meta-analysis uniquely addressed the AF predictive value of NLR in cardiac surgery³¹. A prospective study revealed a strong association between SII and SIRI and the risk of all-cause mortality and cardiovascular mortality in the general population^32–34. In another retrospective cohort study of 2834 women, NLR, MLR, SIRI, and AISI were significantly associated with all-cause mortality and CVD mortality in postmenopausal women with osteoporosis or osteopenia³⁵. In individuals with chronic kidney disease (CKD), there exists a positive association between MLR and the likelihood of mortality. Moreover, MLR demonstrates superior predictive accuracy for assessing mortality risk compared to other systemic inflammatory biomarkers³⁶. Longitudinal cohort studies, systemic review and meta-analysis have revealed a association between elevated NLR, SII, SIRI levels and heightened mortality risk in individuals with HF^37–39. A observational study found that NLR were higher in HF patients than in age-sex matched controls. However, it was not deemed reliable enough to accurately predict the development of HF due to its limited sensitivity and specificity⁴⁰. MLR is strongly related to HF markers and independently predicts HF hospitalizations in patients with coronary artery disease (CAD)⁴¹. Unlike the aforementioned studies, we have introduced preliminary findings regarding the relationship between newly identified biomarkers of systemic inflammation and the occurrence of AF, HF, and CVD mortality in individuals without pre-existing health conditions, highlighting the importance of addressing systemic inflammation for better prevention strategies.

Notably, though SII is strongly linked to cardiovascular and all-cause mortality³² and the predictive capacity of SII for major cardiovascular events in patients with CAD who have undergone coronary intervention exceeds that of traditional risk factors⁴², our analysis revealed that the relationship between SII and AF was not statistically significant. SII, the index in consideration of platelets, neutrophils, and lymphocytes, might be a reliable and representative indicator to predict the coagulation and inflammation risk of CVDs events³². Unlike single blood cell count, which can be affected by factors such as alterations in body fluids, elevated SII levels frequently indicate the presence of neutrophilia, thrombocytosis, and lymphocytopenia⁴³. The interrelated effects of these three distinct cell lineages and population-specific factors may diminish the predictive accuracy for AF.

Emerging evidence suggests several mechanisms of inflammation in the pathogenesis of AF and HF, including oxidative stress, apoptosis, fibrosis, endothelial dysfunction, platelet activation, and coagulation cascade activation^24,25. Patients with AF or HF demonstrate a phenomenon of systemic inflammation, which causes and accelerates the electrical and structural remodeling of the myocardium, leading in AF substrate development and cardiac dysfunction²⁶. These can be demonstrated by increasing serum levels of inflammatory biomarkers, increased expression of inflammatory biomarkers in cardiac tissues, and favorable outcomes of anti-inflammatory drugs in humans with AF and HF or in animal models^27,28. According to the information above, identifying useful and accurate biomarkers associated with AF and HF may help us better understand the genesis and prognosis of these illnesses. An expanding corpus of research suggests that inflammation plays a significant role in cardiac electrophysiology through various mechanisms, which encompass both direct effects on the heart and indirect influences mediated by systemic changes associated with cytokines. The identified mechanisms include: 1) the prolongation of ventricular action potential duration (APD) due to modulation of membrane ion channels; 2) dysfunction of intracellular calcium handling proteins, resulting in spontaneous diastolic calcium release from the sarcoplasmic reticulum (SR); 3) alterations in gap junction functionality through modifications in connexins; 4) the promotion of cardiac fibrosis; 5) the impact of fever, which induces temperature-related changes in the biophysics of cardiac ion channels; 6) the activation of the cardiac sympathetic nervous system; 7) the inhibition of cytochrome P450 enzymes in the liver, which increases the bioavailability of various medications, including those that prolong the QT interval; and 8) the stimulation of aromatase activity in adipose tissue, leading to an enhanced conversion of androgens to estrogens, which is associated with decreased testosterone levels and an elevated risk of long-QT syndrome (LQTS) and torsades de pointes (TdP) in male patients⁴⁴. Inflammatory cytokines have the capacity to modify atrial electrophysiology and structural substrates, consequently heightening the susceptibility to atrial fibrillation (AF). Furthermore, inflammation influences calcium homeostasis and connexin expression, both of which are linked to the initiation of AF and the variability in atrial conduction⁴⁵.

Strengths and limitations

Our study’s main strength is, in fact, the sizable sample from UK Biobank, which made it possible to evaluate the specific connections between AF and HF and systemic inflammatory markers with enough statistical power and a low chance of random error. Another advantage is that systemic inflammatory biomarkers, such as NLR, MLR, SIRI, SII, and AISI, are easily obtained, inexpensive, and routinely obtained through full blood testing. These biomarkers show promise in helping clinicians identify patients at high risk and make better medical decisions in clinical practice. Nevertheless, there are several limitations with our study. First, a single measurement of the systemic inflammatory biomarkers was conducted, and dynamic alterations might have occurred throughout the follow-up. Extra information might be obtained through dynamic monitoring. Second, in an observational research, residual confounding may still exist and cannot be totally ruled out, and a causal relation of these systemic inflammatory biomarkers with the predisposition of incident AF and HF, as well as potential interactions between these inflammatory markers cannot be established. Also, absolute risks from this study might not be estimated. Third, the conventional methodologies employed in our investigation, including the Kaplan–Meier method and the Cox proportional hazards model, treat competing events, such as non-cardiovascular (non-CVD) mortality, as instances of “deletion.” This approach may lead to an overestimation of the incidence rates of atrial fibrillation (AF), heart failure (HF), and cumulative cardiovascular disease (CVD) mortality, thereby introducing bias in the estimation of the effects of certain risk factors and potentially misidentifying high-risk populations. In forthcoming research, we intend to implement the following methodologies to appropriately address the issue of competing risks: Cumulative Incidence Function (CIF) or Fine-Gray Subdistribution Hazard Model.

Conclusions

This research elucidates the relationship between biomarkers indicative of systemic inflammation and the probability of AF, HF and CVD mortality. The ratio of blood cells may function as a novel biomarker for systemic inflammation, potentially serving as a predictive tool for the risk of developing AF, HF and CVD mortality. The assessment of novel systemic inflammatory biomarkers for the purpose of identifying high-risk populations may facilitate the prevention of AF and HF, as well as diminish mortality risk.Further investigation is necessary to comprehensively grasp the intricate network of relationships that blood cell ratios have in the realm of cardiac pathology, and to substantiate the potential utility of these ratios as therapeutic targets, risk indicators, or disease markers.

Abbreviations

CVD

Cardiovascular diseases

AF

Atrial fibrillation

HF

Heart failure

CRP

C-reactive protein

IL-6

Interleukin 6

CABG

Coronary artery bypass graft surgery

BMI

Body mass index

SBP

Systolic blood pressure

DBP

Diastolic blood pressure

NLR

Neutrophil to lymphocyte ratio

PLR

Platelet to lymphocyte ratio

MLR

Monocyte to lymphocyte ratio

SII

Systemic inflammation index

SIRI

Systemic inflammation response index

AISI

The aggregate index of systemic inflammation

HR

Hazard ratio

OR

Odds ratio

CI

Confidence interval

IPAQ

International physical activity questionnaire

Author contributions

YL and MLH contributed equally to this work and are joint first authors. ZLX designed the study. XJC and YPB designed data collection tools, monitored data collection for the whole trial, wrote the statistical analysis plan. YL conducted the data analysis. MLH and ZLX drafted and revised the manuscript. YL and LY acquired and interpreted the data. All authors read and approved the final manuscript.

Funding

This work was funded by the National Natural Science Foundation of China (No. 82000339).

Data availability

The datasets used during the current study from UK Biobank are available on application at www.ukbiobank.ac.uk/register-apply.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

Ethics approval for the UK Biobank study was obtained from the North West Centre for Research Ethics Committee (11/NW/0382). All participants have provided electronic informed consent for the baseline and follow-up evaluations.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-94507-y.