Neutrophil and monocyte ratios to high-density lipoprotein cholesterol as biomarkers in non-dipping hypertension

ABSTRACT

Objective

To investigate the level and significance of neutrophils to high-density lipoprotein cholesterol ratio (NHR) and monocytes to high-density lipoprotein cholesterol ratio (MHR) in patients with non-dipping hypertension.

Methods

A total of 228 patients were retrospectively enrolled in the study. They were divided into the dipping hypertension group (n = 76), the non-dipping hypertension group (n = 77) and the control group (n = 75) according to 24-h ambulatory blood pressure monitoring system (ABPM) recordings. NHR and MHR were calculated and compared statistically. Receiver operating characteristic (ROC) curve analyses were performed for NHR and MHR. Binary logistic regression analyses were introduced to investigate the independent associations of NHR and MHR with non-dipping hypertension.

Results

The NHR and MHR were significantly higher in the non-dipping hypertension group compared with the control group (p = .001, p < .001, respectively) and the dipping hypertension group (p = .039, p = .003, respectively). According to ROC curve analyses, NHR>73.35 and MHR>7.54 were regarded as high-risk groups. The area under the curve (AUC) was 0.642 (p < .001) for NHR and 0.718 (p < .001) for MHR. In multivariate analysis, compared with NHR, only MHR was still recognized as a marker for detection of non-dipping hypertension (odds ratio [OR]: 1.208, 95% confidence interval [CI]: 1.076 to 1.356, p = .001).

Conclusions

Our data indicated that not NHR but MHR as new composite marker of inflammation and lipid metabolism may predict non-dipping hypertension to some extent.

KEYWORDS:

• Neutrophil
• monocyte
• high-density lipoprotein cholesterol (HDL-C)
• biomarkers
• non-dipping hypertension

Introduction

Hypertension is one of the most prevalent risk factors of cardiovascular diseases (CVD). Inflammation has been described in hypertension with many studies demonstrating elevated levels of inflammatory markers and organ damage (1,2). The circadian rhythm variability of blood pressure (BP) is closely related to stroke, cardiovascular events and all-cause mortality, which inserts an important impact on the prognosis of diseases (3–5). Individuals who have a 10% to 20% reduction in BP during the night are known as “dippers.” Non-dipping BP is defined as <10% fall of BP during the night (6). Non-dipping BP has been associated with increased CVD risk and target organ damage (7).

As pro-inflammatory cells, neutrophils and monocytes have been recently recognized to participate in the various stages of atherosclerosis and appear to drive the pro-inflammatory state. Neutrophils attract monocytes into atherosclerotic lesions, activate macrophages and promote foam cell formation and plaque destabilization (8). Moreover, activated monocytes secrete a variety of pro-inflammatory mediators and pro-oxidants, and interact with vascular endothelial cells and platelets to cause inflammatory reaction, thrombosis, endothelial dysfunction (9). Inversely, high-density lipoprotein cholesterol (HDL-C) has been considered as a protective factor against atherosclerosis. The major cardioprotective mechanisms include inhibiting the migration of monocyte macrophages, promoting cholesterol efflux, antioxidant properties, and anti-inflammatory effect in the endothelium (10,11). As composite biomarkers of inflammation and lipid metabolism, NHR and MHR had been shown to be related to the prevalence of metabolic syndrome (12), ischemic stroke (13), cardiovascular diseases (14–16). Recently, MHR had been reported to be associated with target organ damage in hypertension (17,18). However, there are few studies yet on the association between NHR and MHR and non-dipping BP. Accordingly, in this study, we investigated the roles of NHR and MHR to determine if they were reliable and valid biomarkers of non-dipping hypertension.

Methods

Study population

We retrospectively included patients between January 2020 and December 2021 from the Wuhan Third Hospital. Among the 228 patients, 153 were essential hypertension patients and 75 were normotensive patients. None of them were treated with antihypertensive medication. Hypertension was defined as systolic blood pressure (SBP) greater than or equal to 140 mmHg and/or diastolic blood pressure (DBP) greater than or equal to 90 mmHg. Exclusion criteria: secondary hypertension, obstructive sleep apnea syndrome, pulmonary hypertension, presence of active infection, myocardial infarction, congestive heart failure, or neurological disorders, hepatic and renal insufficiency, hematological diseases, endocrine system diseases and rheumatic immune diseases, malignant tumor and presence of pregnancy.

The age, sex, the body mass index (BMI), socio-demographic data and 24-h ABPM recordings of the 228 patients were recorded in detail. BMI was defined as the body mass divided by the square of the body height. For 24-h ABPM, TM −2430 device was used (AND Co., Ltd, Japan). Measurement interval was set to repeat every 30 min during daytime (8am–22pm) and every 60 min (22pm–8am) during nighttime. The effective measurement time of the whole day accounted for more than 80% of the total measurement time were selected as effective data. DBP<40 mmHg and SBP>220 mmHg were defined invalid. Nocturnal BP decrease rate = (mean SBP during the day – mean SBP during the night)/mean SBP during the day × 100%. Dipping hypertension was defined as the reduction of nighttime BP ≥ 10%, and non-dipping hypertension was defined as the decrease of nighttime BP < 10%. White blood cell (WBC), hemoglobin, platelet, neutrophile, lymphocyte, monocyte, red cell distribution width, glucose, hypersensitivity C-reactive protein (hs-CRP), transaminase, creatinine and lipid profile were recorded. NHR and MHR values were calculated by dividing the neutrophil and monocyte counts to HDL-C level. The study protocol was approved by the Ethics Committee of the Wuhan Third Hospital. The study was conducted in accordance with the Declaration of Helsinki. Because of the retrospective nature of the study, the requirement for informed consent from eligible patients was waived by the ethics committee.

The estimation of the sample size was done using Power and Sample Size online software, with 90% power (1-β = 0.9) and a α = 5% false-positive rate. According to the mean of the existing three groups of data, the sample size was calculated between each two groups. The required sample size was estimated to be the maximum number calculated.

Statistical analysis

All data were tested for normal distribution by Shapiro–Wilk test. Continuous and normally distributed variables were expressed as the mean ± SD, and skewed variables were expressed as median and interquartile range. Categorical variables were expressed as number with percentage. Comparisons between the three groups were performed using one-way ANOVA or Kruskal Wallis test for continuous variables and the chi-square test for categorical variables. Bonferroni method was applied for multiple comparisons of one-way ANOVA and Kruskal Wallis test as post hoc analysis. ROC curve analyses were used to find out the optimal cutoff points of NHR and MHR. Moreover, the AUCs of the two indexes were calculated. Univariate analysis for odds to non-dipping BP was performed by logistic regression for every confounder from our data. Any variable examined in the univariate analysis for which the p < .10 was contained in the forward stepwise multivariate model. Results were shown as odds ratio (OR) [95% confidence interval]. The diagnostic efficacy of the multivariate model as sensitivity, specificity, positive predictive value and negative predictive value were made with logistic regression analysis. Statistical analyses were performed using SPSS 26 (Chicago, IL, USA). All tests were two-sided. The value of p < .05 was considered to indicate statistical significance.

Results

Baseline characteristics

Table 1 presents the baseline characteristics of 228 participants finally included in this study, including 77 patients in the non-dipping hypertension group, 76 patients in the dipping hypertension group and 75 patients in the control group. The average age of this study population was 61.8 ± 12.7 years and 57% were women. The patients were matched for gender, age and BMI and were not significantly different.

Table 1. Baseline characteristics of the study population.

Variable	Non-dipping hypertension	Dipping hypertension	Controls	p-Value
Number	77	76	75	-
Sex, female (%)	55.80	57.90	57.30	0.971
Age (years)	61.18 ± 13.93	61.50 ± 12.12	62.83 ± 12.16	0.703
BMI (kg/m^2)	24.94 ± 3.67	24.80 ± 2.79	24.14 ± 2.51	0.228
WBC (10^3/mm3)	6.63 ± 1.30	6.14 ± 1.48	5.95 ± 1.38	0.008*
Neutrophil (×10^3/μl)	4.07 ± 1.12	3.75 ± 1.27	3.70 ± 1.21	0.109
Monocyte (×10^3/μl)	0.50 ± 0.14	0.43 ± 0.14	0.37 ± 0.11	<0.001*
Lymphocyte (×10^3/μl)	1.88 ± 0.56	1.78 ± 0.60	1.74 ± 0.56	0.335
Hemoglobin (g/dl)	13.90 ± 1.21	13.57 ± 1.39	13.25 ± 1.59	0.018*
RDW (%)	13.41(12.90–13.70)	13.40(12.73–13.60)	13.15(12.70–13.50)	0.070
Platelet (×10^3/μl)	215.67 ± 48.94	226.33 ± 56.61	219.00 ± 49.95	0.433
NHR	94.59(66.19–117.68)	79.76(51.26–95.96)	72.85(50.76–96.11)	0.001*
MHR	11.54(7.94–13.71)	9.19(5.92–11.12)	7.42(4.13–9.14)	<0.001*
Total cholesterol (mg/dl)	196.49 ± 42.87	200.15 ± 49.76	186.00 ± 45.63	0.150
Triglyceride (mg/dl)	271.14(127.95–312.37)	191.98(94.64–226.78)	182.03(85.60–180.00)	<0.001*
HDL-C (mg/dl)	47.04 ± 14.21	50.42 ± 11.33	54.38 ± 14.06	0.003*
LDL-C (mg/dl)	103.16 ± 29.00	112.89 ± 33.26	97.66 ± 31.81	0.011*
Creatinine (mg/dl)	0.76 ± 0.20	0.75 ± 0.20	0.72 ± 0.20	0.356
ALT (u/l)	22.06(12.50–27.00)	22.70(14.00–27.00)	21.53(13.00–23.00)	0.695
AST (u/l)	20.97(16.00–23.00)	22.17(17.00–23.00)	22.41(16.00–25.00)	0.495
hs-CRP (mg/l)	3.91(1.40–4.35)	4.42(1.20–4.50)	2.48(0.60–2.55)	0.004*
Glucose (mg/dl)	96.99 ± 12.30	96.65 ± 11.03	95.78 ± 11.57	0.805

BMI, the body mass index. WBC, white blood cell. RDW, red cell distributio width. NHR, neutrophil to HDL-C ratio. MHR, monocyte to HDL-C ratio. HDL-C, high-density lipoprotein cholesterol. LDL-C low-density lipoprotein cholesterol. AST, aspartate aminotransferase. ALT, alanine aminotransferase. Results were presented as mean ± standard deviation or median (25th percentile, 75th percentile). *Statistically significant at p < .05.

There was statistically significant difference between three groups in terms of WBC (p = .008), hemoglobin (p = .018), monocyte count (p < .001), triglyceride (p < .001), HDL-C (p = .003), low-density lipoprotein cholesterol (LDL-C) (p = .011), hs-CRP (p = .004), NHR (p = .001) and MHR (p < .001). Post analysis showed that the WBC level was higher in the non-dipping hypertension group than that in the control group with a statistical difference (p = .008). The hemoglobin was higher in the non-dipping hypertension group than that in the control group (p = .014). The monocyte count was significantly higher in the non-dipping hypertension group when compared with the dipping hypertension group (p = .004) and the control group (p < .001). The triglyceride concentration was significantly higher in the non-dipping hypertension group when compared with the control group (p < .001) and the dipping hypertension group (p = .022). There was significantly statistical difference of the HDL-C concentration between the non-dipping hypertension group and the control group (p = .001). The LDL-C concentration was highest in the dipping hypertension group and there was a statistical difference between the dipping hypertension group and the control group (p = .019). The hs-CRP value was higher in the non-dipping and the dipping hypertension group when compared with the control group (p = .004 and p = .009, separately). As shown in Figure 1a, NHR was higher in the non-dipping hypertension group when compared with the dipping hypertension group (p = .039) and the control group (p = .001). Figure 1b shows that MHR was significantly higher in the non-dipping hypertension group when compared with the dipping hypertension group (p = .003) and the control group (p < .001).

Figure 1. Comparison of NHR (a) and MHR (b) in the non-dipping hypertension group contrast to the control group and the dipping hypertension group. NHR, neutrophil to HDL-C ratio. MHR, monocyte to HDL-C ratio.

ROC curve analysis

ROC curves were analyzed in search of the cutoff values of NHR and MHR for determining non-dipping hypertensive patients. The results are presented in Figure 2. For NHR, the AUC was 0.642 (0.567–0.716; p < .001) (Figure 2a). The cutoff point was observed at 73.35, with a sensitivity of 68.80% and a specificity of 55.00%. Accordingly, the AUC for MHR was calculated as 0.718 (0.650–0.786; p < .001) (Figure 2b). The cutoff value for MHR was found to be 7.54, with a sensitivity of 84.40% and a specificity of 50.30%.

Figure 2. ROC curve analyses of NHR (a) and MHR (b). NHR, neutrophil to HDL-C ratio. MHR, monocyte to HDL-C ratio.

Univariate and multivariate logistic regression analysis

As shown in Table 2, univariate analyses were used to evaluate the effects of all baseline confounders on the identification of non-dipping hypertension. WBC (OR:1.353, 95% CI: 1.104 to 1.657, p = .004), neutrophil (OR:1.271, 95% CI: 1.010 to 1.601, p = .041), monocyte (OR:1.005, 95% CI: 1.003 to 1.008, p < .001), hemoglobin (OR:1.025, 95% CI: 1.005 to 1.045, p = .016), triglyceride (OR:1.002, 95% CI: 1.000 to 1.003, p = .017), HDL-C (OR:0.969, 95% CI: 0.947 to 0.991, p = .006), NHR (OR:1.031, 95% CI: 1.006 to 1.021, p = .001) and MHR (OR:1.190, 95% CI: 1.107 to 1.278, p < .001) were found to may be risk factors for non-dipping hypertension. The Box–Tidwell method was used to test whether there were linear relationships between continuous independent variables and dependent variables. There existed collinearity between neutrophil and NHR, as well as between monocyte and MHR. There was no collinearity between NHR and MHR. Finally, the independent variables except for neutrophils and monocytes were included in multivariate logistic regression analyses. Moreover, the sensitivity of the model was 32.5% and the specificity was 89.4%. The positive predictive value was 61.0%, and the negative predictive value was 72.2%. Among the eight independent variables included in the model, only MHR was statistically significant and was an independent predictor of non-dipping hypertension (OR: 1.223, 95% CI: 1.088 to 1.376, p = .001).

Table 2. Univariate logistic regression analysis and multivariate logistic regression analysis related to non-dipping hypertension.

Variable	Univariate; OR (95% CI)	p-Value	Multivariate; OR (95% CI)	p-Value
Sex, female	1.075(0.618 to 1.870)	0.978	-
Age (years)	0.994(0.973 to 1.016)	0.583	-
BMI	1.052(0.961 to 1.151)	0.271	-
WBC	1.353(1.104 to 1.657)	0.004*	1.149(0.787 to 1.677)	0.473
Neutrophil	1.271(1.010 to 1.601)	0.041*
Monocyte	1.005(1.003 to 1.008)	<0.001*
Lymphocyte	1.418(0.876 to 2.295)	0.155
Hemoglobin	1.025(1.005 to 1.045)	0.016*	1.011(0.989 to 1.034)	0.326
RDW	1.171(0.880 to 1.559)	0.280	-
Platelet	0.997(0.992 to 1.003)	0.335	-
NHR	1.013(1.006 to 1.021)	0.001*	0.991(0.973 to 1.010)	0.359
MHR	1.190(1.107 to 1.278)	<0.001*	1.223(1.088 to 1.376)	0.001*
Total cholesterol	1.002(0.996 to 1.007)	0.603	-
Triglyceride	1.002(1.000 to 1.003)	0.017*	1.000(0.999 to 1.002)	0.558
HDL-C	0.969(0.947 to 0.991)	0.006*	1.007(0.968 to 1.048)	0.732
LDL-C	0.998(0.989 to 1.007)	0.628	-
Creatinine	1.970(0.495 to 7.844)	0.336	-
ALT	0.976(0.978 to 1.021)	0.976	-
AST	0.261(0.944 to 1.016)	0.261	-
hs-CRP	1.009(0.959 to 1.062)	0.720	-
Glucose	1.006(0.982 to 1.030)	0.633	-

BMI, the body mass index. WBC, white blood cell. RDW, red cell distributio width. NHR, neutrophil to HDL-C ratio. MHR, monocyte to HDL-C ratio. HDL-C, high-density lipoprotein cholesterol. LDL-C low-density lipoprotein cholesterol. AST, aspartate aminotransferase. ALT, alanine aminotransferase. *Statistically significant at p < .05.

Ambulatory blood pressure monitoring system data

As shown in Table 3, the ABPM data among the three groups were compared. For 24-hour mean SBP, the difference between the control group and non-dipping hypertension group was statistically significant (p < .001), as well between the control group and dipper blood pressure group (p < .001). However, there was no significant difference in blood pressure between non-dipping group and dipping group (p = 1.000). For 24-hour mean DBP and daytime mean SBP, the difference between the control group and non-dipping hypertension group or dipping group was statistically significant (p < .001). Nevertheless, there was no statistical difference between non-dipper hypertension and dipper hypertension in 24-hour mean DBP and daytime mean SBP (p = 1.000). For daytime mean DBP, the difference between the control group and non-dipping hypertension group or dipping group was statistically significant (p < .001). There was no statistical difference between non-dipping group and dipping group (p = .481) yet. The difference between the control group and non-dipping group or dipping group was statistically significant in night mean SBP and night mean DBP (p < .001). Also there was a statistical difference between non-dipping group and dipping group in night mean SBP (p = .017). However, there was no statistical difference between non-dipping group and dipping group in night mean DBP (p = .479). For night SBP decrease percentage, the difference between the non-dipping group and dipping group or control group was statistically significant (p < .001). There was no statistical difference between dipping group and control group (p = .097). For night DBP decrease percentage, the difference between the non-dipping group and dipping group or control group was statistically significant (p < .001). There was no statistical difference between dipping group and control group (p = 1.000).

Table 3. ABPM data of three groups.

BP	Non-dipping	dipping	Control	P value
24-hour mean SBP	153.79(145.00–159.50)	151.50(141.75–155.25)	127.00(122.75–135.25)	<0.001
24-hour mean DBP	86.43(78.00–94.00)	88.08(79.50–95.50)	75.68(71.00–81.00)	<0.001
Daytime mean SBP	154.60(146.00–160.50)	155.81(145.75–159.50)	130.60(126.75–139.25)	<0.001
Daytime mean DBP	86.94(78.00–95.00)	90.88(80.75–100.00)	78.08(73.00–83.00)	<0.001
Night mean SBP	151.60(141.00–160.00)	137.04(128.75–140.50)	113.44(108.50–121.00)	<0.001
Night mean DBP	84.75(76.00–91.00)	79.08(72.50–87.50)	66.66(62.00–71.25)	<0.001
Night SBP decrease percentage	2.40(−2.94–5.51)	12.64(9.94–14.96)	17.77(14.63–21.81)	<0.001
Night DBP decrease percentage	3.03(−1.93–5.51)	13.04(10.47–14.76)	14.61(11.25–18.82)	<0.001

SBP, systolic blood pressure; DBP, diastolic blood pressure. P value represented the comparison with the control group. *Statistically significant at p < .05.

Discussion

Hypertension is one of the main causes of CVD and leads an increased risk of damage to multiple target organs including brain, heart and kidney. It is very important to diagnose hypertension in time for treatment. Blood pressure has a circadian rhythm. Generally, the blood pressure at night is 10–20% lower than that during the day. But in some hypertensive patients the blood pressure at night is less than 10% lower. This type of BP is called non-dipping hypertension.

Non-dipping hypertension is commonly defined as <10% fall of night-time BP according to 24-h ABPM. Non-dipping hypertensive status has been associated with increased risk of metabolic disorders, obesity and fatal and non-fatal cardiovascular events (19–21). Therefore, early identification of non-dipping hypertension is of significance. Although ABPM is known as an effective means to identify non-dipping hypertension, there is no access to these facilities in the vast community health service centers in China. Nevertheless, each community health service center will organize residents to participate in a physical examination every year. These items include height and weight measurement, brachial artery blood pressure measurement, blood routine examination and biochemical examination such as liver and kidney function, blood lips and blood glucose, etc. Therefore, there is a certain clinical practical value in finding new markers to predict non-dipper hypertension. Neutrophil/high-density lipoprotein (NHR) and monocyte/high-density lipoprotein ratios (MHR) are conventional blood indicator, cheap and easy to obtain. Recently, increasing numbers of studies have focused on MHR and NHR on cardiovascular and cerebrovascular diseases. For instance, MHR was reported higher and might be an independent factor that could predict outcomes in patients (n = 411) with mechanical thrombectomy for acute ischemic stroke (22). In addition, MHR was considered an independent predictor of all-cause mortality and major adverse cardiac events in coronary artery disease patients undergoing PCI (n = 673) (23). Similarly, NHR was relevant to prognosis in elderly patients with acute myocardial infarction (n = 528) (15). Consequently, we hypothesized that NHR and MHR could be predictors of non-dipping hypertension.

The mechanism of NHR or MHR in predicting non-dipper hypertension is unknown. It may be involved in endothelial cell inflammation with monocytes and neutrophils, resulting in increased arterial stiffness and decreased arterial compliance. Vascular stiffness reduces baroreflex function, resulting in an increase in blood pressure variability and a decrease in blood pressure drop at night. Hyperlipidemia and inflammation are interrelated, inducing the increase of monocytes, which make these cells more likely to migrate to atherosclerotic lesions (24). At the same time, neutrophil homeostasis is also disturbed by hyperlipidemia and ultimately involved in the process of atherosclerosis (8). Neutrophils attract monocytes to atherosclerotic plaque through granule protein. Monocytes migrate to inflammatory sites and convert into macrophages, which phagocytize oxidized LDL-C and turn into foam cells (9). HDL-C has been suggested to protect endothelial cell functions by prevention of oxidation of LDL-C, anti-inflammatory, anti-apoptotic and anti-thrombotic effects as well as endothelial repair processes. NHR and MHR simultaneously reflect inflammatory status and lipid metabolism.

In our experiment, we found that NHR and MHR in non-dipping hypertension group were higher than those in dipping hypertension group and control group with statistical significant difference. In univariate logistic regression analysis, WBC, hemoglobin, triglyceride, HDL-C, monocytes, neutrophils, NHR and MHR were considered as risk factors for non-dipping hypertension. However, only MHR was considered as a possible independent risk factor for non-dipping hypertension through further multivariate logistic regression analysis. The AUC of MHR was not excellent as 0.718 (0.650–0.786; p < .001) although it was greater than 0.7. The specificity was low as 0.503 similar to previously reported (25). The reason might be the confounding factors not considered in the retrospective study data. On the other hand, perhaps more subjects should be included.

From the above results, it was inferred that MHR could be used as a predictor of non-dipping hypertension, but its diagnostic value is not particularly good. Although the area under the AUC curve of MHR was not excellent and its specificity was not high, MHR was higher in non-dipper hypertension than other groups, which served as a reminder in clinical practice to pay more attention to BP of these individuals who need better BP control.

Limitations

We have to admit that there were some limitations in present study. Firstly, confounding bias in this retrospective study existed objectively. As a conventional inflammatory marker, our study did not observe that hs-CRP was statistically significant among the three groups. On the one hand, it may be caused by the selection bias of the retrospective study, although we tried to strictly include patients following the inclusion criteria and exclusion criteria. On the other hand, hs-CRP might not be sensitive in predicting non-dipping hypertension. Further prospective studies are needed to confirm this conclusion. Secondly, we did not compare the lipid ratios with other conventional inflammatory markers, for instance, the ratios of neutrophil to lymphocyte (NLR), platelet to lymphocyte (PLR), monocyte to lymphocyte (MLR), lymphocyte to HDL-C (LHR), triglyceride to HDL-C (THR), and LDL-C to HDL-C. Moreover, inflammatory markers such as TNF-α, and IL-6 were needed to be considered. Thirdly, complications of non-dipping hypertension are needed to be studied in the future, such as carotid intima thickening and left ventricular hypertrophy.

Conclusion

In conclusion, we observed that NHR and MHR, as composite biomarkers of inflammation and lipid metabolism, were increased in non-dipping hypertension when compared with dipping hypertension and normal BP. However, not NHR but MHR might have a diagnostic value for non-dipping hypertension patients through logistic regression analysis.

Acknowledgments

Funding: This work was supported by the 2014 clinical medical research project of Wuhan Health and Family Planning Commission under Grant number W×14B12.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The work was supported by the 2014 clinical medical research project of Wuhan Health and Family Planning Commission [WX14B12].