Skip to main content
PMC home page
European Heart Journal logoLink to European Heart Journal
. 2022 Mar 11;43(19):1809–1828. doi: 10.1093/eurheartj/ehac102

Bone marrow activation in response to metabolic syndrome and early atherosclerosis

Ana Devesa 1,2,3, Manuel Lobo-González 4, Juan Martínez-Milla 5,6, Belén Oliva 7, Inés García-Lunar 8,9,10, Annalaura Mastrangelo 11, Samuel España 12,13, Javier Sanz 14,15, José M Mendiguren 16, Hector Bueno 17,18,19, Jose J Fuster 20,21, Vicente Andrés 22,23, Antonio Fernández-Ortiz 24,25,26, David Sancho 27, Leticia Fernández-Friera 28,29, Javier Sanchez-Gonzalez 30, Xavier Rossello 31,32,33, Borja Ibanez 34,35,36,✉, Valentin Fuster 37,38,✉
PMCID: PMC9113301  PMID: 35567559

Abstract

Aims

Experimental studies suggest that increased bone marrow (BM) activity is involved in the association between cardiovascular risk factors and inflammation in atherosclerosis. However, human data to support this association are sparse. The purpose was to study the association between cardiovascular risk factors, BM activation, and subclinical atherosclerosis.

Methods and results

Whole body vascular 18F-fluorodeoxyglucose positron emission tomography/magnetic resonance imaging (18F-FDG PET/MRI) was performed in 745 apparently healthy individuals [median age 50.5 (46.8–53.6) years, 83.8% men] from the Progression of Early Subclinical Atherosclerosis (PESA) study. Bone marrow activation (defined as BM 18F-FDG uptake above the median maximal standardized uptake value) was assessed in the lumbar vertebrae (L3–L4). Systemic inflammation was indexed from circulating biomarkers. Early atherosclerosis was evaluated by arterial metabolic activity by 18F-FDG uptake in five vascular territories. Late atherosclerosis was evaluated by fully formed plaques on MRI. Subjects with BM activation were more frequently men (87.6 vs. 80.0%, P = 0.005) and more frequently had metabolic syndrome (MetS) (22.2 vs. 6.7%, P < 0.001). Bone marrow activation was significantly associated with all MetS components. Bone marrow activation was also associated with increased haematopoiesis—characterized by significantly elevated leucocyte (mainly neutrophil and monocytes) and erythrocyte counts—and with markers of systemic inflammation including high-sensitivity C-reactive protein, ferritin, fibrinogen, P-selectin, and vascular cell adhesion molecule-1. The associations between BM activation and MetS (and its components) and increased erythropoiesis were maintained in the subgroup of participants with no systemic inflammation. Bone marrow activation was significantly associated with high arterial metabolic activity (18F-FDG uptake). The co-occurrence of BM activation and arterial 18F-FDG uptake was associated with more advanced atherosclerosis (i.e. plaque presence and burden).

Conclusion

In apparently healthy individuals, BM 18F-FDG uptake is associated with MetS and its components, even in the absence of systemic inflammation, and with elevated counts of circulating leucocytes. Bone marrow activation is associated with early atherosclerosis, characterized by high arterial metabolic activity. Bone marrow activation appears to be an early phenomenon in atherosclerosis development.

[Progression of Early Subclinical Atherosclerosis (PESA); NCT01410318].

Keywords: Subclinical atherosclerosis, Metabolic syndrome, Bone marrow, PET/MRI

Structured Graphical Abstract

Structured Graphical Abstract.

Structured Graphical Abstract

Open in a new tab

The hypothesis of the natural history of the inflammatory process involving the atherosclerotic plaque formation. Bone marrow (BM) is implicated in the atherosclerotic process long before the appearance of acute cardiovascular events. Cardiovascular risk factors trigger BM activation, initially in the absence of systemic inflammation. As BM activation progresses, it is accompanied by an increase in haematopoietic progenitor cells and an associated increase in inflammatory markers. The next step in the process is arterial inflammation, leading to an increase in atherosclerotic burden.

See the editorial comment for this article ‘Mischief in the marrow: a root of cardiovascular evil’, by Peter Libby et al., https://doi.org/10.1093/eurheartj/ehac149.

Introduction

The association between inflammation and atherosclerosis is well established,1 and mechanistic studies have demonstrated that inflammation is an essential mediator of all stages of atherosclerosis, from initiation to progression and the development of thrombotic complications.2,3 Circulating immune cells play a critical role in the build-up of atherosclerotic plaques by adhering to activated endothelium and infiltrating the arterial wall to become lesional cells.4 This association has led to the study of various anti-inflammatory therapies in the last years, with encouraging results that justify the use of some of them such as low-dose colchicine in selected, high-risk patients.5

The bone marrow (BM) is the primary site of haematopoiesis, and the proliferation and migration of haematopoietic progenitors are regulated by various physiological and pathological stimuli.1,2 After an acute cardiovascular event, BM is activated by sympathetic signalling, triggering an increased haematopoiesis, and the release of progenitor cells that activate spleen production of monocytes, aggravating atherosclerosis progression.6–9 Experimental studies suggest that increased BM haematopoietic activity may be a central link between cardiometabolic risk factors and exacerbated inflammation in atherosclerosis. In mice, hypercholesterolaemia and low HDL-cholesterol levels associated with elevated haematopoietic activity with increased monocytosis and neutrophilia.10,11 Moreover, murine models of obesity present marked monocytosis and neutrophilia, associated with BM myeloid progenitor proliferation and expansion.12,13 Diabetes mellitus has also been associated with increased circulating neutrophils and monocytes, reflecting the expansion of BM myeloid progenitors.14,15 Hypertension, driven by an overactive sympathetic activation, deteriorates haematopoietic cell niche in the BM which can contribute to atherosclerosis.16 In humans, it has been suggested that chronic stress accelerates haematopoiesis, giving rise to higher levels of inflammatory cells that might contribute to the atherosclerotic process.17 In addition, haematopoietic stem cell division rates are increased in subjects with atherosclerosis,18 and it has been suggested that the haematopoietic system might be chronically affected in these subjects.19

Despite the extensive pre-clinical data, human data to support the association between BM haematopoietic activation and cardiovascular risk factors are sparse.

In some tissues, the high metabolic activity can be detected by imaging techniques such as 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) imaging. This technique has been used to characterize BM activation after acute coronary syndrome,20,21 based on a higher 18F-FDG uptake related to an increased glucose consumption due to cellular proliferation, and vascular inflammation,20–22 which correlates with macrophage density as measured by histology and immunohistochemistry.23–25 Moreover, an association between amygdalar activity as evaluated by 18F-FDG PET/CT and cardiovascular disease events has been suggested to be mediated by increased BM activation.26 Hybrid 18F-FDG PET/magnetic resonance imaging (PET/MRI) is an accurate method for the detection of early atherosclerosis (characterized by high arterial metabolic activity) and for later stages of the disease (plaque detection), in asymptomatic middle-aged individuals.27 Here, we studied BM activation, detected by 18F-FDG PET/MRI in apparently healthy middle-aged individuals, and its association with cardiovascular risk factors and subclinical atherosclerosis.

Materials and methods

Study population

The study population consisted of participants in the PESA study [Progression of Early Subclinical Atherosclerosis—CNIC–Santander; (PESA) NCT01410318]28 who underwent whole body 18F-FDG PET/MRI.27 PESA is an observational prospective cohort study of 4184 asymptomatic employees at Santander Bank in Madrid. Participants were aged 40–54 years at enrolment (June 2010–February 2014). Exclusion criteria were previous cardiovascular disease, any condition reducing life expectancy or affecting study adherence, morbid obesity [body mass index (BMI) ≥ 40 kg/m2], or chronic kidney disease (estimated glomerular filtration rate <60 mL/min/1.73 m2). The main goal of the PESA study is to characterize atherosclerosis initiation and progression by means of serial multi-territory, multimodality non-invasive imaging,29 and paired biological sampling. A subgroup of PESA participants showing atherosclerosis on baseline vascular ultrasound, defined as being in the highest plaque tertile on vascular ultrasound and/or having any coronary artery calcification on CT, underwent baseline whole body 18F-FDG PET/MRI study to characterize arterial metabolic activity.27 Cardiovascular risk factors were assessed prospectively at enrolment using the 10-year risk algorithm based on Pooled Cohort Equations.30 Risk bands of <5, 5 to <7.5, and ≥7.5% were defined as low, intermediate, and high risk, respectively.31 A fasting blood test included blood count and biochemistry with the determination of systemic inflammation parameters.28 Blood count included leucocytes and their components (including neutrophils, lymphocytes, monocytes, eosinophils, and basophils), red blood cell count, haemoglobin, haematocrit, red blood cell width, and platelets. Leucocytosis was defined as a white blood cell count >10.5 × 103 cells/µL. Inflammation parameters included high-sensitivity C-reactive protein (hs-CRP), ferritin, erythrocyte sedimentation rate, fibrinogen, P-selectin, and vascular cell adhesion molecule-1 (VCAM-1). Insulin levels were also measured. Insulin resistance was measured by HOmeostatic Model Assessment for Insulin Resistance (HOMA-IR), calculated as [(fasting plasma glucose level × fasting insulin level)/405].32 In this study, metabolic syndrome (MetS) was defined when a participant met at least three of the following conditions: central obesity (waist circumference ≥88 cm in women and ≥102 cm in men);33 elevated plasma triglycerides (≥150 mg/dL); low plasma HDL-cholesterol (<40 mg/dL in men or <50 mg/dL in women); elevated fasting plasma glucose (≥100 mg/dL); and high blood pressure (systolic ≥130 mmHg and/or diastolic ≥85 mmHg).34 Dyslipidaemia was defined as total cholesterol ≥240 mg/dL, LDL-cholesterol ≥160 mg/dL, HDL-cholesterol <40 mg/dL, or use of lipid-lowering drugs.

The study protocol was approved by the institutional review board, and all participants provided written informed consent.

Hybrid positron emission tomography/magnetic resonance imaging acquisition protocol and image analysis

The vascular PET/MRI protocol has been published previously.27 In brief, the protocol included high-resolution black-blood MRI of the carotid, iliac, and femoral arteries and co-registered PET/MRI of the carotid arteries, thoracic aorta, infrarenal abdominal aorta, iliac, and femoral arteries. Ilio-femoral MRI attenuation maps and lower-body PET were acquired at the start of the protocol. Magnetic resonance imaging data were analysed with VP Diagnostics software version 2.1.0 (Seattle, WA, USA) and PET/MRI data were analysed with Philips Fusion Viewer version 2.0 (Philips Healthcare).

As previously described, attenuation correction used MRI attenuation maps with a three-class tissue (soft tissue, lung, and air) validated segmentation technique.35 A transverse 575 mm field of view was used, and images were generated with a voxel size of 4.0 mm × 4.0 mm × 4.0 mm. Specific templates for PET images were added to the attenuation maps to correct for attenuation effects of the scanner bed. For abdominal region, the coil has a minimal attenuation and was not included in the attenuation maps. Similarly, not corrected PET data were used to extend the attenuation maps beyond the magnetic resonance field of view limits and this information was integrated in attenuation correction maps for the final reconstruction.36

A total of six vascular territories per participant were analysed in MRI: left/right carotid arteries, left/right iliac arteries, and left/right femoral arteries. The presence, number, and plaque volumes were defined for each territory, and recorded as a surrogate of total plaque burden for each individual.

Lumbar vertebrae L3 and L4 were analysed in fused PET/MRI images.21 Quantitative 18F-FDG uptake was measured in multiple slices in the coronal axis by drawing 3D regions of interest encompassing the contour of each vertebra, excluding the cortical bone. The maximal standardized uptake value (SUVmax) was calculated in these regions of interest (calculated as decay-corrected tissue radioactivity divided by body weight and injected dose).37 For each participant, BM SUVmax was calculated as the mean SUVmax of the lumbar vertebrae (L3 and L4). Bone marrow activation was defined as BM SUVmax above the median value (1.9).

Statistical analysis

Normally distributed continuous variables are expressed as mean ± SD, whereas non-normally distributed variables are expressed as median (Q1–Q3). The distribution of continuous variables was assessed with graphical methods. Categorical variables are expressed as n (%). Differences between BM activation were assessed by Student’s t-test or Wilcoxon signed-rank test and χ  2 or Fisher exact test, for continuous and categorical variables, respectively, as appropriate. Linear trends across groups according to quintiles of 18F-FDG uptake were evaluated with an extension of the non-parametric Wilcoxon rank-sum test. For multivariate analysis, ordinal logistic regression models were performed. To evaluate the associations of BM activation in the presence of confounders, Model 1 (adjusting for age and sex) and Model 2 (adjusting for age, sex, glucose levels before PET/MRI, smoking, haemoglobin, and hs-CRP) were created. To evaluate the association between BM activation in the presence of vascular uptake associations and plaque volume (mm3) (0 and tertiles), several models were generated: Model 1 (qualitative), adjusting for age, sex, hypertension, dyslipidaemia, diabetes, smoking, family history of cardiovascular disease, and obesity; Model 2 (quantitative), adjusting for age, sex, systolic blood pressure, diastolic blood pressure, LDL-cholesterol, HDL-cholesterol, diabetes, smoking, dyslipidaemia treatment, family history of cardiovascular disease, and BMI; and Model 3, which is Model 1 but excluding dyslipidaemia treatment.

For all endpoints, differences were considered statistically significant at P-values < 0.05. Statistical analyses were performed using Stata software version 15 (StataCorp, College Station, TX, USA).

Results

A total of 946 PESA participants underwent whole body 18F-FDG PET/MRI at baseline. The mean 18F-FDG dose was 292.3 ± 11.1 MBq, and the radiation exposure was 5.6 ± 0.2 mSv. The mean start time after 18F-FDG injection was 106 ± 15 min for lower-body PET and 132.9 ± 19.9 min for upper-body PET. Reasons for non-completion were physical intolerance in upper-body studies (8 PETs and 51 MRIs), technical issues with MRI attenuation maps (97 initial PETs), and poor image quality (70 iliac MRIs). Complete 18F-FDG PET/MRI studies were available for 755 (79.8%) participants, and lumbar vertebrae BM images were of good quality for 745 participants (78.8% of the total sample who underwent PET/MRI); these participants constituted the population for the present study.

Baseline characteristics of subjects in relation to bone marrow activation

The median (Q1–Q3) age was 50.5 years (46.8–53.6) and 83.8% were men. Baseline characteristics are represented in Table 1. Participants with BM activation (Figure 1) were more frequently men (87.6 vs. 80% in those subjects without BM activation, P = 0.005) and more frequently had MetS (22.2 vs. 6.7%, P < 0.001). Bone marrow activation showed a significant association with central obesity (41.1 vs. 10.7%, P < 0.001), hypertension (23.5 vs. 14.9%, P = 0.003), higher plasma triglyceride levels (105.5 vs. 87 mg/dL, P < 0.001), lower HDL-cholesterol (44.5 vs. 48.4 mg/dL, P < 0.001), and higher fasting glucose (93 vs. 89 mg/dL, P < 0.001) (Figure 2). The BM activation and non-activation groups showed no differences in age, family history of cardiovascular disease, smoking, or total and LDL-cholesterol. Bone marrow activation group had higher levels of glycated haemoglobin (HbA1c, 5.5 vs. 5.4%, P = 0.007), and insulin resistance measured by HOMA-IR (1.7 vs. 1.1%, P < 0.001). Insulin levels were significantly increased in the group with BM activation (7 vs. 4.5 µU/mL, P < 0.001). Bone marrow activation was associated with significantly higher numbers of leucocytes (6.00 × 103 vs. 5.77 × 103 cells/µL in the group without BM activation, P = 0.027), especially neutrophils (3.4 × 103 vs. 3.2 × 103 cells/µL, P = 0.029) and red blood cell counts (4.9 × 106 vs. 4.8 × 106 cells/µL, P < 0.001). The BM activation group also showed significant elevation of the systemic inflammation markers, including hs-CRP (0.13 vs. 0.08 mg/dL, P < 0.001), ferritin (138.6 vs. 107.6 ng/dL, P = 0.001), fibrinogen (268.6 vs. 260.8 mg/dL, P = 0.03), P-selectin (139.7 vs. 129.2 ng/dL, P = 0.004), VCAM-1 (686.3 vs. 623.4 ng/mL, P = 0.025), and red blood cell distribution width (14.7 vs. 14.6%, P = 0.055). The main between-group differences in baseline characteristics are summarized in Figure 2.

Table 1.

Study population characteristics stratified by bone marrow activation (above or below-median 18F-fluorodeoxyglucose uptake)

Total population No bone marrow activation Bone marrow activationa P-value
(n = 745) (n = 375) (n = 370)
Age, years 50.5 (46.8–53.6) 50.5 (47.0–53.8) 50.5 (46.7–53.5) 0.961
Men 624 (83.8) 300 (80.0) 324 (87.6) 0.005
Metabolic syndrome and components
 Metabolic syndrome 107 (14.4) 25 (6.7) 82 (22.2) <0.001
 Central obesity 192 (25.8) 40 (10.7) 152 (41.1) <0.001
 Triglycerides, mg/dL 98 (72–131) 87 (65–121) 105 (77–140) <0.001
 HDL-C, mg/dL 46.5 ± 11.4 48.4 ± 11.6 44.5 ± 10.7 <0.001
 Fasting glucose, mg/dL 91 (85–97) 89 (84–94) 93 (87–99) <0.001
 SBP, mmHg 120.5 ± 12.3 118.7 ± 11.6 122.3 ± 12.6 <0.001
 DBP, mmHg 75.3 ± 9.1 73.7 ± 8.3 76.9 ± 9.5 <0.001
Other cardiovascular risk factors
 Family history of CV disease, n (%) 154 (20.7) 73 (19.5) 81 (21.9) 0.414
 Current smoking (%) 197 (26.9) 111 (30.1) 86 (23.8) 0.054
 Hypertension 143 (19.2) 56 (14.9) 87 (23.5) 0.003
 Dyslipidaemia 440 (59.1) 203 (54.1) 237 (64.1) 0.006
 Diabetes 34 (4.6) 13 (3.5) 21 (5.7) 0.149
 BMI, kg/m2 27.2 ± 3.5 25.5 ± 2.8 28.8 ± 3.2 <0.001
 Weight, kg 81.4 ± 13.3 75.5 ± 11.3 87.4 ± 12.4 <0.001
 Waist circumference, cm 93.8 ± 10.8 89.0 ± 9.6 98.7 ± 9.7 <0.001
Treatment
 Antihypertensive therapy 100 (13.4) 40 (10.7) 60 (16.2) 0.026
 Lipid-lowering therapy 113 (15.2) 50 (13.3) 63 (17.0) 0.160
 Antidiabetic therapy 28 (3.8) 11 (2.9) 17 (4.6) 0.233
Biochemistry
 Total cholesterol, mg/dL 208.3 ± 33.5 207.6 ± 32.3 209.0 ± 34.7 0.592
 LDL-C, mg/dL 139.9 ± 30.1 139.1 ± 28.8 140.6 ± 31.4 0.488
 HbA1c, % 5.5 (5.2–5.7) 5.4 (5.2–5.7) 5.5 (5.3–5.7) 0.007
 HOMA-IR, % 1.3 (0.9–2.1) 1.1 (0.7–1.7) 1.7 (1.0–2.5) <0.001
 Insulin, µU/mL 5.7 (3.9–8.2) 4.5 (3.4–6.8) 7.0 (5.0–10.3) <0.001
Inflammatory markers
 hs-CRP, mg/dL 0.11 (0.06–0.19) 0.08 (0.05–0.16) 0.13 (0.07–0.23) <0.001
 Ferritin, ng/mLb 120.8 (63.0–204.0) 107.6 (56.4–194.1) 138.6 (74.8–214.1) 0.001
 Erythrocyte sedimentation rate (1 h), mm 5 (4–8) 5 (4–7) 6 (4–8) 0.175
 Fibrinogen, mg/dL 265.4 (236.5–295.1) 260.8 (234.5–294.6) 268.6 (239.6–299.2) 0.030
 P-selectin, ng/mL 134.7 (106.7–166.2) 129.2 (104.5–160.6) 139.7 (109.9–177.0) 0.004
 Vascular cell adhesion molecule-1, ng/mL 651.1 (519.6–820.0) 623.4 (509.8–791.9) 686.3 (540.6–849.0) 0.025
Blood count
 Leucocytes, 103 cells/µL 5.87 (4.99–7.05) 5.77 (4.94–6.95) 6.00 (5.15–7.13) 0.027
 Leucocytosis (>10.5 × 103 cells/µL) 17 (2.3) 8 (2.1) 9 (2.4) 0.785
 Red blood cell count, 106 cells/µL 4.84 (4.59–5.08) 4.78 (4.51–5.03) 4.89 (4.64–5.12) <0.001
 Red cell distribution width, % 14.6 (14.0–15.2) 14.6 (14.0–15.1) 14.7 (14.0–15.2) 0.055
 Haemoglobin, g/dL 15.0 (14.3–15.7) 14.9 (14.1–15.6) 15.1 (14.4–15.8) <0.001
 Haematocrit, % 44.2 (41.9–46.3) 43.9 (41.6–46.0) 44.5 (42.5–46.5) 0.002
 Platelet count, 103 cells/µL 225 (198–257) 226 (199–256) 224 (194–258) 0.686
 Segmented neutrophils, 103 cells/µL 3.31 (2.73–4.23) 3.21 (2.66–4.11) 3.40 (2.78–4.36) 0.029
 Lymphocytes, 103 cells/µL 1.86 (1.56–2.22) 1.84 (1.55–2.20) 1.88 (1.57–2.22) 0.465
 Monocytes, 103 cells/µL 0.41 (0.34–0.52) 0.41 (0.33–0.51) 0.42 (0.34–0.52) 0.201
 Eosinophils, 103 cells/µL 0.13 (0.08–0.20) 0.12 (0.08–0.20) 0.13 (0.08–0.21) 0.446
 Basophils, 103 cells/µL 0.05 (0.03–0.07) 0.05 (0.03–0.07) 0.05 (0.03–0.06) 0.689
 Neutrophil to lymphocyte ratio 1.79 (1.43–2.30) 1.75 (1.41–2.24) 1.83 (1.46–2.36) 0.178
SCORE risk score
 Low (<1%) 459 (64.2) 245 (67.1) 214 (61.1) 0.095
 Intermediate (1–5%) 254 (35.5) 120 (32.9) 134 (38.3) 0.131
 High (>5%) 2 (0.3) 0 (0.0) 2 (0.6) 0.148
ASCVD risk score
 Low 369 (61.7) 207 (66.1) 162 (56.8) 0.020
 Intermediate 109 (18.2) 50 (16.0) 59 (20.7) 0.135
 High 120 (20.1) 56 (17.9) 64 (22.5) 0.164
Arterial uptake (18F-FDG)
 Presence of uptake 358 (48.1) 135 (36.0) 223 (60.3) <0.001
 Number of uptakes 0 (0–2) 0 (0–1) 1 (0–2) <0.001
 SUVmax arterial uptake 1.38 (1.26–1.52) 1.30 (1.20–1.41) 1.46 (1.37–1.59) <0.001
Plaques by magnetic resonance
 Plaque presence 671 (90.1) 337 (89.9) 334 (90.3) 0.854
 Number of plaques 3 (2–5) 3 (1–5) 3 (2–5) 0.842
 Global plaque burden 365.6 (175.6–706.9) 348.5 (164.9–681.0) 375.8 (192.4–751.2) 0.313
Open in a new tab

Data are presented as n (%) or median (Q1–Q3).

a

Bone marrow activation was defined when the mean BM SUVmax was above the median value (SUVmax 1.9).

b

Measured in 622 of 745 individuals.

Figure 1.

Figure 1

Open in a new tab

Bone marrow uptake. Representative baseline 18F-fluorodeoxyglucose positron emission tomography/magnetic resonance imaging scans from participants. The left panel shows fused 18F-fluorodeoxyglucose positron emission tomography/magnetic resonance imaging in coronal (upper) and sagittal (lower) views. L3 and L4 vertebrae (white arrows) present high 18F-fluorodeoxyglucose uptake (visualized in blue). The right panel shows the same coronal and sagittal views; 18F-fluorodeoxyglucose uptake is not visualized. The bladder is visualized in red in both upper panels (red asterisk).

Figure 2.

Figure 2

Open in a new tab

Population with bone marrow activation. Characteristics of participants with baseline bone marrow 18F-fluorodeoxyglucose uptake compared with the group without bone marrow uptake. Participants with bone marrow 18F-fluorodeoxyglucose uptake were more frequently male and had a higher prevalence of metabolic syndrome and its components, central obesity, hypertension, and altered glucose metabolism. The bone marrow uptake group also had higher levels of systemic inflammatory markers (high-sensitivity C-reactive protein and ferritin) and showed higher counts of leucocytes and red blood cells.

To explore the association between the degree of BM activation and participant characteristics, we divided the population into BM-uptake quintiles according to SUVmax. The characteristics of each BM-uptake subgroup are presented in Table 2. The higher the degree of BM activation (i.e. the higher the BM-uptake quintile), the higher the percentage of men and the more frequent the presence of central obesity. The same pattern of increase across quintiles was observed for the frequency of MetS; hypertension; low HDL; and elevated fasting glucose, HbA1c, and HOMA-IR.

Table 2.

Study population characteristics stratified by bone marrow activation (18F-fluorodeoxyglucose uptake) quintile

BM-uptake Q1 BM-uptake Q2 BM-uptake Q3 BM-uptake Q4 BM-uptake Q5 P for trend
(n = 149) (n = 149) (n = 149) (n = 149) (n = 149)
Mean BM SUVmax 0.57–1.6 1.6–1.81 1.81–2.0 2.0–2.3 2.3–3.8
 Age, years 50.0 (46.7–53.0) 51.7 (47.5–54.3) 49.9 (45.6–52.8) 50.5 (47.0–53.8) 50.9 (47.1–53.8) 0.431
 Men 105 (70.5) 128 (85.9) 128 (85.9) 134 (89.9) 129 (86.6) <0.001
Metabolic syndrome and components
 Metabolic syndrome 7 (4.7) 12 (8.1) 20 (13.4) 18 (12.1) 50 (33.6) <0.001
 Central obesity 11 (7.4) 16 (10.7) 27 (18.1) 40 (26.8) 98 (65.8) <0.001
 Triglycerides, mg/dL 81 (61–113) 96 (72–128) 101 (68–128) 103 (73–133) 107 (88–147) <0.001
 HDL-C, mg/dL 50.8 ± 12.5 46.3 ± 11.3 47.0 ± 10.6 45.5 ± 10.7 42.8 ± 10.2 <0.001
 Fasting glucose, mg/dL 87 (83–92) 90 (83–95) 91 (84–98) 93 (88–97) 94 (87–101) <0.001
 SBP, mmHg 117.6 ± 12.4 119.8 ± 11.6 120.5 ± 11.3 121.2 ± 11.8 123.5 ± 13.5 <0.001
 DBP, mmHg 73.1 ± 9.1 74.3 ± 8.1 75.1 ± 7.9 75.4 ± 9.2 78.6 ± 10.1 <0.001
Other cardiovascular risk factors
 Family history of CV disease 28 (18.8) 26 (17.4) 30 (20.1) 38 (25.5) 32 (21.5) 0.201
 Current smoking 47 (32.2) 45 (30.6) 31 (21.2) 32 (22.1) 42 (28.6) 0.176
 Hypertension 21 (14.1) 25 (16.8) 24 (16.1) 30 (20.1) 43 (28.9) 0.001
 Dyslipidaemia 67 (45.0) 91 (61.1) 88 (59.1) 98 (65.8) 96 (64.4) 0.001
 Diabetes 4 (2.7) 2 (1.3) 11 (7.4) 6 (4.0) 11 (7.4) 0.026
 BMI, kg/m2 24.7 ± 3.0 25.9 ± 2.5 26.7 ± 2.7 28.1 ± 2.7 30.5 ± 3.2 <0.001
 Weight, kg 71.9 ± 12.1 77.3 ± 10.1 79.6 ± 10.6 85.2 ± 10.9 93.1 ± 12.0 <0.001
 Waist circumference, cm 85.7 ± 10.0 90.7 ± 8.9 92.9 ± 8.5 96.7 ± 8.5 103.0 ± 9.5 <0.001
Treatment
 Antihypertensive therapy 16 (10.7) 15 (10.1) 18 (12.1) 21 (14.1) 30 (20.1) 0.010
 Lipid-lowering therapy 14 (9.4) 20 (13.4) 25 (16.8) 32 (21.5) 22 (14.8) 0.043
 Antidiabetic therapy 3 (2.0) 1 (0.7) 11 (7.4) 5 (3.4) 8 (5.4) 0.057
Biochemistry
 Total cholesterol, mg/dL 205.8 ± 30.3 209.3 ± 34.2 210.0 ± 31.2 210.8 ± 36.4 205.6 ± 35.1 0.837
 LDL-C, mg/dL 136.4 ± 28.3 141.2 ± 29.9 141.9 ± 27.0 142.3 ± 32.9 137.5 ± 31.8 0.583
 HbA1c, % 5.4 (5.2–5.6) 5.4 (5.1–5.6) 5.5 (5.2–5.7) 5.5 (5.2–5.8) 5.5 (5.3–5.7) 0.005
 HOMA-IR, % 0.9 (0.6–1.3) 1.2 (0.8–1.9) 1.2 (0.8–1.9) 1.5 (1.0–2.1) 2.0 (1.3–2.9) <0.001
 Insulin, µU/mL 4.0 (2.8–5.8) 5.1 (3.8–7.5) 5.3 (3.8–7.5) 6.3 (4.3–8.6) 8.2 (6.0–11.5) <0.001
Inflammatory markers
 hs-CRP, mg/dL 0.07 (0.04–0.13) 0.10 (0.05–0.18) 0.10 (0.05–0.20) 0.12 (0.07–0.21) 0.15 (0.08–0.28) <0.001
 Ferritin, ng/mLa 90.2 (40.9–152.4) 127.7 (71.3–198.5) 122.4 (69.0–224.6) 137.5 (74.5–200.2) 154.9 (81.6–253.0) <0.001
 Erythrocyte sedimentation rate (1 h), mm 5 (4–8) 5 (4–7) 5 (4–7) 6 (4–8) 6 (4–8) 0.277
 Fibrinogen, mg/dL 255.4 (230.1–286.6) 266.2 (236.5–296.1) 265.7 (235.3–294.6) 264.4 (239.9–299.2) 277.2 (249.8–302.8) 0.001
 P-selectin, ng/mL 125.0 (102.8–156.1) 130.5 (105.9–166.2) 135.6 (106.9–166.7) 135.3 (106.7–167.2) 144.1 (115.4–180.5) 0.002
 VCAM-1, ng/mL 615.3 (503.8–777.6) 625.1 (499.1–790.5) 681.2 (530.4–859.5) 674.0 (518.6–849.0) 680.8 (555.1–821.7) 0.027
Blood count
 Leucocytes, 103 cells/µL 5.59 (4.91–6.72) 5.93 (4.95–7.07) 5.80 (4.91–6.75) 5.98 (5.17–7.13) 6.20 (5.35–7.48) 0.001
 Leucocytosis (>10.5 × 103 cells/µL) 0 (0) 5 (3.4) 4 (2.7) 4 (2.7) 4 (2.7) 0.225
 Red blood cell count, 106 cells/µL 4.71 (4.38–4.99) 4.83 (4.60–5.05) 4.85 (4.59–5.05) 4.90 (4.67–5.10) 4.90 (4.65–5.15) <0.001
 Red cell distribution width, % 14.6 (14.0–15.2) 14.6 (14.0–15.2) 14.5 (13.9–15.1) 14.7 (13.9–15.2) 14.8 (14.2–15.2) 0.089
 Haemoglobin, g/dL 14.7 (13.8–15.3) 15.1 (14.4–15.7) 15.0 (14.2–15.6) 15.1 (14.6–15.8) 15.2 (14.5–15.8) <0.001
 Haematocrit, % 43.0 (40.9–45.6) 44.5 (41.9–46.4) 44.2 (41.7–46.0) 44.6 (42.9–46.6) 44.6 (42.8–46.7) <0.001
 Platelets, 103 cells/µL 228 (199–253) 224 (196–256) 222 (196–259) 227 (200–257) 222 (193–262) 0.831
 Segmented neutrophils, 103 cell/µL 3.13 (2.60–3.77) 3.34 (2.72–4.26) 3.19 (2.71–3.92) 3.48 (2.61–4.42) 3.48 (2.97–4.50) 0.001
 Lymphocytes, 103 cells/µL 1.83 (1.62–2.16) 1.86 (1.55–2.17) 1.86 (1.54–2.23) 1.82 (1.55–2.22) 1.91 (1.64–2.28) 0.136
 Monocytes, 103 cells/µL 0.39 (0.32–0.49) 0.41 (0.33–0.53) 0.41 (0.33–0.50) 0.42 (0.35–0.51) 0.43 (0.35–0.53) 0.019
 Eosinophils, 103 cells/µL 0.12 (0.08–0.20) 0.12 (0.08–0.20) 0.12 (0.07–0.21) 0.13 (0.08–0.19) 0.14 (0.09–0.23) 0.361
 Basophils, 103 cells/µL 0.05 (0.03–0.07) 0.05 (0.03–0.06) 0.04 (0.03–0.06) 0.05 (0.03–0.06) 0.05 (0.03–0.07) 0.128
 Neutrophil to lymphocyte ratio 1.66 (1.37–2.15) 1.81 (1.46–2.25) 1.74 (1.37–2.30) 1.82 (1.42–2.45) 1.88 (1.51–2.27) 0.048
SCORE risk score
 Low (<1%) 106 (72.1) 88 (59.5) 102 (73.9) 82 (57.8) 81 (57.9) 0.003
 Intermediate (1–5%) 41 (27.9) 60 (40.5) 35 (25.4) 59 (41.5) 59 (42.1) 0.002
 High (>5%) 0 (0.0) 0 (0.0) 1 (0.7) 1 (0.7) 0 (0) 0.538
ASCVD risk score
 Low 96 (73.8) 69 (56.1) 84 (70.0) 57 (54.3) 63 (52.5) 0.001
 Intermediate 17 (13.1) 24 (19.5) 18 (15.0) 22 (20.9) 28 (23.3) 0.047
 High 17 (13.1) 30 (24.4) 18 (15.0) 26 (24.8) 29 (24.2) 0.040
Arterial uptake (18F-FDG)
 Presence of uptake 37 (24.8) 59 (39.6) 73 (49.0) 82 (55.0) 107 (71.8) <0.001
 Number of uptakes 0 (0–0) 0 (0–1) 0 (0–1) 1 (0–2) 1 (0–2) <0.001
 SUVmax arterial uptake 1.26 (1.14–1.32) 1.33 (1.24–1.43) 1.37 (1.27–1.47) 1.41 (1.32–1.51) 1.58 (1.46–1.69) <0.001
Plaques by magnetic resonance
 Plaque presence 127 (85.2) 136 (91.3) 138 (92.6) 134 (89.9) 136 (91.3) 0.166
 Number of plaques 3 (1–5) 3 (2–5) 3 (2–5) 3 (2–5) 3 (2–5) 0.855
 Global plaque burden, mm3 348.5 (164.9–669.9) 362.3 (166.8–699.5) 370.2 (177.8–688.8) 372.5 (175.6–753.5) 371.4 (204.8–753.3) 0.365
Open in a new tab

Data are presented as n (%) or median (Q1–Q3).

a

Measured in 622 of 745 individuals.

The increase in BM activation was also associated with elevated numbers of leucocytes (mainly neutrophils, with an increased neutrophil to lymphocyte ratio), and red blood cells, with numbers increasing progressively across quintiles. The distribution of relevant participant characteristics stratified by BM-uptake quintile is shown in Figure 3. The circulating systemic inflammation markers, hs-CRP, ferritin, fibrinogen, P-selectin, and VCAM-1 also steadily increased across BM-uptake quintiles (Table 2).

Figure 3.

Figure 3

Open in a new tab

Progression of bone marrow activation. Prevalence of metabolic syndrome, central obesity, hypertension, dyslipidaemia, and leucocytosis stratified by quintiles of bone marrow 18F-fluorodeoxyglucose uptake.

When adjusted for Model 1 (age and sex) and Model 2 (age, sex, glucose levels before PET/MRI, smoking, haemoglobin, and hs-CRP), BM activation remained significantly associated with MetS and its components, insulin levels, hs-CRP, leucocytes, and arterial uptake (Figure 4).

Figure 4.

Figure 4

Open in a new tab

Unadjusted and adjusted associations for different factors with bone marrow activation. Associations between different factors with bone marrow activation, expressed as odds ratio with its 95% confidence interval. The unadjusted estimates are presented in black, the estimates adjusted for Model 1 (adjusted for age and sex) are presented in blue, and the estimates adjusted for Model 2 (adjusted for age, sex, glucose levels before positron emission tomography/magnetic resonance imaging, smoking, haemoglobin, and high-sensitivity C-reactive protein) are presented in red.

When BM uptake was evaluated as a continuous variable, MetS and its components (particularly central obesity) and the presence of arterial uptake had the largest effect size on BM activation (see Supplementary material online, Table S1); this was consistent with the analysis of BM-uptake quintiles.

Association between bone marrow activation and early atherosclerosis

Arterial 18F-FDG uptake is a surrogate for high vascular metabolic activity due to macrophage accumulation, the precursor of atherosclerosis.27,38 Bone marrow activation was significantly associated with the presence of vascular 18F-FDG uptake (60 vs. 36%, P < 0.001). Participants with BM activation also had more sites of vascular 18F-FDG uptakes, and higher degree of uptake (vascular SUVmax) (Tables 1 and 2 and Figure 5).

Figure 5.

Figure 5

Open in a new tab

Relationship between bone marrow and vascular 18F-fluorodeoxyglucose uptake. The left and right panels show representative positron emission tomography/magnetic resonance imaging analysis of 18F-fluorodeoxyglucose uptake in lumbar vertebrae and vascular tissue, respectively. The chart shows increases in the presence of vascular uptake (left Y-axis) and in vascular-uptake SUVmax (right Y-axis) with increasing bone marrow uptake quintile.

We next explored the association between BM activation and vascular 18F-FDG uptake. Participants with BM activation and vascular 18F-FDG uptake tended to be older (51.2 vs. 49.4 years among those with BM activation and no vascular uptake, P < 0.001) and more frequently presented with MetS (26.9 vs. 15%, P = 0.007) (Table 3). The co-occurrence of BM activation and vascular 18F-FDG uptake showed a significant association with central obesity (48.9 vs. 29.3% in participants with BM activation without vascular 18F-FDG uptake, P < 0.001), smoking (28.6 vs. 16.2%, P = 0.007), higher total cholesterol (211.9 vs. 204.5 mg/dL, P = 0.045), lower HDL-cholesterol (43.1 vs. 46.7 mg/dL, P = 0.001), higher triglyceride (112 vs. 92 mg/dL, P < 0.001), and higher fasting glucose levels (94 vs. 92 mg/dL, P = 0.013), as well as with lower rates of individuals assigned to a low-risk group in the SCORE algorithm. There was no between-group difference in age, family history of cardiovascular disease, and LDL-cholesterol. The co-occurrence of BM activation and vascular 18F-FDG uptake was associated with elevated levels of HbA1c (5.5 vs. 5.4%, P = 0.042), plasma insulin (7.6 vs. 5.8 µU/mL, P < 0.001) and insulin resistance measured by HOMA-IR (1.9 vs. 1.4%, P < 0.001). Co-occurring BM activation and vascular 18F-FDG uptake were associated with significantly elevated numbers of leucocytes (6.22 × 103 vs. 5.83 × 103 cells/µL in the BM activation group without vascular 18F-FDG uptake, P = 0.023), especially monocytes (0.43 × 103 vs. 0.40 × 103 cells/µL, P = 0.007) and of red blood cells (4.91 × 106 vs. 4.86 × 106 cells/µL, P = 0.038). This trend was accompanied by significantly elevated markers of systemic inflammation in the BM activation plus vascular 18F-FDG uptake group, including hs-CRP (0.15 vs. 0.11 mg/dL, P < 0.001).

Table 3.

Characteristics of population with bone marrow activation according to the presence or absence of vascular 18F-fluorodeoxyglucose uptake

BM activation without vascular uptake BM activation and vascular uptake P-value
(n = 147) (n = 223)
Age, years 49.4 (45.4–52.9) 51.2 (48.0–54.1) <0.001
Men 123 (83.7) 201 (90.1) 0.065
Metabolic syndrome and its components
 Metabolic syndrome 22 (15.0) 60 (26.9) 0.007
 Central obesity 43 (29.3) 108 (48.9) <0.001
 Triglycerides, mg/dL 92 (68–128) 112 (86–147) <0.001
 HDL-C, mg/dL 46.7 ± 11.7 43.1 ± 9.8 0.001
 Fasting glucose, mg/dL 92 (85–98) 94 (88–101) 0.013
 SBP, mmHg 120.4 ± 11.4 123.6 ± 13.2 0.018
 DBP, mmHg 75.6 ± 8.7 77.8 ± 9.9 0.029
Other cardiovascular risk factors
 Family history of CV disease 29 (19.7) 52 (23.3) 0.414
 Current smoking 23 (16.2) 63 (28.6) 0.007
 Hypertension 31 (21.1) 56 (25.1) 0.372
 Dyslipidaemia 85 (57.8) 152 (68.2) 0.043
 Diabetes 6 (4.1) 15 (6.7) 0.282
 BMI, kg/m2 27.9 ± 3.3 29.5 ± 3.0 <0.001
 Weight, kg 84.1 ± 12.3 89.5 ± 12.1 <0.001
 Waist circumference, cm 95.4 ± 9.3 100.8 ± 9.3 <0.001
Treatment
 Antihypertensive therapy 22 (15.0) 38 (17.0) 0.596
 Lipid-lowering therapy 25 (17.0) 38 (17.0) 0.993
 Antidiabetic therapy 6 (4.1) 11 (4.9) 0.702
Biochemistry
 Total cholesterol, mg/dL 204.5 ± 31.3 211.9 ± 36.5 0.045
 LDL-C, mg/dL 137.1 ± 27.1 143.0 ± 33.7 0.077
 HOMA-IR, % 1.4 (1.0–2.1) 1.9 (1.2–2.9) <0.001
 HbA1c, % 5.4 (5.2–5.7) 5.5 (5.3–5.8) 0.042
 Insulin, µU/mL 5.8 (4.3–8.0) 7.6 (5.4–11.5) <0.001
Inflammatory markers
 hs-CRP, mg/dL 0.11 (0.05–0.18) 0.15 (0.08–0.29) <0.001
 Ferritin, ng/mLa 131.38 (61.13–207.27) 151.23 (81.63–224.11) 0.390
 Erythrocyte sedimentation rate (1 h), mm 5 (4–7) 6 (4–8) 0.084
 Fibrinogen, mg/dL 264.4 (237.3–291.5) 273.1 (240.9–302.8) 0.178
 P-selectin, ng/mL 139.9 (111.0–177.0) 139.3 (107.7–177.2) 0.981
 Vascular cell adhesion molecule-1, ng/mL 645.2 (512.4–861.2) 712.4 (557.1–838.8) 0.379
Blood count
 Leucocytes, 103 cells/µL 5.83 (4.99–6.89) 6.22 (5.23–7.33) 0.023
 Leucocytosis (>10.5 × 103 cells/µL) 2 (1.4) 7 (3.1) 0.277
 Red blood cell count, 106 cells/µL 4.86 (4.59–5.05) 4.91 (4.68–5.18) 0.038
 Red cell distribution width, % 14.6 (14.0–15.2) 14.7 (14.0–15.3) 0.130
 Haemoglobin, g/dL 15.0 (14.4–15.6) 15.1 (14.5–15.9) 0.107
 Haematocrit, % 44.3 (41.6–46.1) 44.7 (42.9–46.7) 0.047
 Platelet count, 103 cells/µL 228 (194–259) 221 (193–258) 0.452
 Segmented neutrophils,103 cell/µL 3.30 (2.69–4.17) 3.45 (2.86–4.50) 0.072
 Lymphocytes,103 cells/µL 1.83 (1.55–2.13) 1.92 (1.58–2.29) 0.097
 Monocytes, (103 cells/µL 0.40 (0.32–0.49) 0.43 (0.36–0.54) 0.007
 Eosinophils,103 cells/µL 0.12 (0.08–0.19) 0.14 (0.08–0.23) 0.326
 Basophils,103 cells/µL 0.04 (0.03–0.06) 0.05 (0.03–0.07) 0.085
 Neutrophil to lymphocyte ratio 1.79 (1.42–2.36) 1.85 (1.48–2.38) 0.498
SCORE risk score
 Low (<1%) 100 (71.9) 114 (54.0) 0.001
 Intermediate (1−5%) 39 (28.1) 95 (45.0) 0.001
 High (>5%) 0 (0.0) 2 (1.0) 0.250
ASCVD risk score
 Low 83 (72.8) 79 (46.2) <0.001
 Intermediate 15 (13.2) 44 (25.7) 0.010
 High 16 (14.0) 48 (28.1) 0.005
Plaques by magnetic resonance
 Plaque presence 122 (83.0) 212 (95.1) <0.001
 Number of plaques 2 (1–4) 4 (2–5) <0.001
 Global plaque burden, mm3 284.5 (125.2–519.5) 448.4 (229.4–819.0) <0.001
Plaques by 2D vascular ultrasound
 Plaque presence 117 (81.2) 203 (94.4) <0.001
 Number of plaques 2 (1–4.5) 4 (2–7) <0.001
Plaques by 3D vascular ultrasound
 Plaque presence 90 (67.2) 180 (87.4) <0.001
 Global plaque burden, mm3 30.3 (0–128.6) 100.8 (37.8–214.4) <0.001
Open in a new tab

Data are presented as n (%) or median (Q1–Q3).

a

Measured in 622 of 745 individuals.

More advanced stages of atherosclerosis are characterized by full-grown plaques. We observed that subjects with BM activation plus vascular 18F-FDG uptake had a higher prevalence of plaques than those with BM activation but without vascular 18F-FDG uptake (95.1 vs. 83%, P < 0.001). Similarly, the BM activation plus vascular 18F-FDG uptake group had more plaques (4 vs. 2, P < 0.001) and a higher plaque burden (448.4 vs. 284.5 mm3, P < 0.001). These associations remained significant after adjusting for classical risk factors (odds ratio 2.33, 95% confidence interval 1.54–3.52, P < 0.001 in the fully adjusted model) (Figures 6 and 7). This association was consistent when atherosclerosis was evaluated by 2D and 3D vascular ultrasound; subjects with BM activation plus 18F-FDG uptake had a higher prevalence and number of plaques and higher plaque burden when compared with those with BM activation but without vascular uptake.

Figure 6.

Figure 6

Open in a new tab

Bone marrow activation in the presence of vascular 18F-fluorodeoxyglucose uptake is associated with higher atherosclerotic plaque volume. Participants with co-occurring bone marrow activation and vascular 18F-fluorodeoxyglucose uptake had a significantly higher plaque burden than those with bone marrow activation but no vascular 18F-fluorodeoxyglucose uptake. The upper panel shows atherosclerotic plaque volume (mm3) in the group with bone marrow activation and vascular uptake (orange bar) and in the group with bone marrow activation without vascular uptake (blue bar). The mid-panel shows the comparison of adjusted odds ratios and 95% confidence interval for the different models. *In Model 3, 105 participants taking lipid-lowering therapies were eliminated. The lower panel shows representative magnetic resonance images of atherosclerotic plaques.

Figure 7.

Figure 7

Open in a new tab

Positron emission tomography/magnetic resonance imaging analysis of the atherosclerotic plaque 18F-fluorodeoxyglucose uptake. The upper row shows a carotid atherosclerotic plaque with 18F-fluorodeoxyglucose uptake. The lower row shows a femoral atherosclerotic plaque without 18F-fluorodeoxyglucose uptake.

Bone marrow activation in the absence of systemic inflammation

To assess whether BM activation occurred as part of a systemic inflammatory reaction, we studied the subgroup of 402 participants showing no systemic inflammation (below-median hs-CRP). The characteristics of this subpopulation stratified by BM activation are summarized in Table 4. Participants with BM activation were more frequently male (86.4 vs. 75.8%, P = 0.009), and were more frequently positive for MetS (17.9 vs. 3.8%, P < 0.001) and its components, hypertension (25.3 vs. 12.9%, P = 0.001), and diabetes (7.4 vs. 2.5%, P = 0.020). This group also showed higher elevations in fasting glucose (92 vs. 88 mg/dL, P < 0.001), and insulin resistance index (HOMA-IR 1.54 vs. 0.97%, P < 0.001) and plasma insulin (6.1 vs. 4.3 µU/mL, P < 0.001); and presented lower levels of HDL-cholesterol (46.1 vs. 49.7 mg/dL, P = 0.002) and higher levels of triglycerides (97 vs. 83 mg/dL, P = 0.010). The BM activation group showed a higher elevation of erythropoiesis (red blood cell count 4.90 × 106 vs. 4.76 × 106 cells/µL, P = 0.001); however, leucocyte numbers did not differ between inflammation-free participants with and without BM activation. In this subgroup without systemic inflammation, BM activation was significantly associated with higher arterial metabolic activity: more prevalence of vascular 18F-FDG uptake (51.9 vs. 31.2%, P < 0.001), and more sites of vascular 18F-FDG uptake and higher vascular SUVmax.

Table 4.

Bone marrow activation in the subpopulation without systemic inflammation (below-median high-sensitivity C-reactive protein)

Total No bone marrow uptake Bone marrow uptakeb P-value
(n = 402) CRP < median CRP < median
(n = 240) (n = 162)
Age, years 50.1 (46.7–53.2) 50.1 (46.2–53.2) 50.3 (46.7–53.2) 0.657
Men 322 (80.1) 182 (75.8) 140 (86.4) 0.009
Metabolic syndrome and its components
 Metabolic syndrome 38 (9.5) 9 (3.8) 29 (17.9) <0.001
 Central obesity 73 (18.2) 17 (7.1) 56 (34.6) <0.001
 Triglycerides, mg/dL 88 (64–118) 83 (62–115) 97 (71–128) 0.010
 HDL-C, mg/dL 48.2 ± 11.4 49.7 ± 11.4 46.1 ± 11.1 0.002
 Fasting glucose, mg/dL 89 (84–96) 88 (83–94) 92 (85–98) <0.001
 SBP, mmHg 119.4 ± 11.8 117.7 ± 11.3 121.9 ± 12.1 <0.001
 DBP, mmHg 74.6 ± 8.9 72.9 ± 8.4 77.2 ± 9.1 <0.001
Other cardiovascular risk factors
 Family history of CV disease 88 (21.9) 50 (20.8) 38 (23.5) 0.533
 Current smoking 92 (23.3) 65 (27.4) 27 (17.1) 0.017
 Hypertension 72 (17.9) 31 (12.9) 41 (25.3) 0.001
 Dyslipidaemia 218 (54.2) 118 (49.2) 100 (61.7) 0.013
 Diabetes 18 (4.5) 6 (2.5) 12 (7.4) 0.020
 BMI, kg/m2 26.4 ± 3.3 25.1 ± 2.8 28.2 ± 3.1 <0.001
 Weight, kg 79.0 ± 13.4 74.0 ± 11.4 86.4 ± 12.8 <0.001
 Waist circumference, cm 91.5 ± 10.8 87.4 ± 9.5 97.5 ± 9.7 <0.001
Treatment
 Antihypertensive therapy 50 (12.4) 23 (9.6) 27 (16.7) 0.035
 Lipid-lowering therapy 59 (14.7) 31 (12.9) 28 (17.3) 0.225
 Antidiabetic therapy 15 (3.7) 5 (2.1) 10 (6.2) 0.034
Biochemistry
 Total cholesterol, mg/dL 205.6 ± 33.3 206.2 ± 30.9 204.7 ± 36.6 0.646
 LDL-C, mg/dL 137.5 ± 30.2 137.7 ± 28.0 137.2 ± 33.2 0.869
 HbA1c, % 5.4 (5.2–5.7) 5.4 (5.2–5.7) 5.4 (5.2–5.7) 0.404
 HOMA-IR, % 1.15 (0.77–1.86) 0.97 (0.67–1.52) 1.54 (0.98–2.23) <0.001
 Insulin, µU/mL 5.1 (3.6–7.2) 4.3 (3.2–6.2) 6.1 (4.3–8.2) <0.001
Inflammatory markers
 hs-CRP, mg/dL 0.06 (0.04–0.08) 0.06 (0.04–0.08) 0.07 (0.04–0.08) 0.080
 Ferritin, ng/mLa 107.7 (56.5–197.9) 101.7 (46.2–187.5) 122.8 (70.5–227.4) 0.015
 Erythrocyte sedimentation rate (1 h), mm 5 (4–6) 5 (4–6) 5 (2–7) 0.311
 Fibrinogen, mg/dL 252.0 (229.4–275.4) 252.5 (229.8–276.1) 250.9 (226.6–275.1) 0.926
 P-selectin, ng/mL 131.1 (103.4–162.0) 125.8 (103.2–160.2) 139.3 (103.7–168.0) 0.071
 Vascular cell adhesion molecule-1, ng/mL 638.3 (512.8–816.6) 620.4 (507.8–788.8) 699.9 (514.4–862.5) 0.113
Blood count
 Leucocytes, 103 cells/µL 5.59 (4.86–6.64) 5.55 (4.83–6.73) 5.63 (4.93–6.47) 0.575
 Leucocytosis, (>10.5 × 103 cells/µL) 2 (0.5) 2 (0.8) 0 (0) 0.244
 Red blood cells, 106 cells/µL 4.81 (4.51–5.06) 4.76 (4.46–5.02) 4.90 (4.62–5.15) 0.001
 Red cell distribution width, % 14.5 (14.0–15.1) 14.5 (14.0–15.0) 14.6 (14.0–15.2) 0.245
 Haemoglobin, g/dL 14.9 (14.0–15.6) 14.8 (13.9–15.5) 15.1 (14.2–15.9) 0.002
 Haematocrit, % 43.9 (41.6–46.3) 43.5 (41.1–45.8) 44.5 (42.2–46.9) 0.005
 Platelets, 103 cells/µL 219 (192–255) 224 (196–256) 214 (189–244) 0.074
 Segmented neutrophils, 103 cell/µL 3.10 (2.56–3.80) 3.09 (2.55–3.88) 3.12 (2.59–3.77) 0.839
 Lymphocytes, 103 cells/µL 1.83 (1.54–2.17) 1.82 (1.52–2.18) 1.87 (1.55–2.16) 0.471
 Monocytes, 103 cells/µL 0.39 (0.33–0.48) 0.39 (0.33–0.48) 0.39 (0.33–0.47) 0.678
 Eosinophils number (1000 cells/µL) 0.12 (0.08–0.19) 0.11 (0.08–0.19) 0.12 (0.07–0.20) 0.547
 Eosinophils, 103 cells/µL 0.04 (0.03–0.06) 0.05 (0.03–0.06) 0.04 (0.02–0.06) 0.094
 Neutrophil to lymphocyte ratio 1.68 (1.37–2.10) 1.71 (1.37–2.13) 1.66 (1.36–2.08) 0.431
SCORE risk score
 Low (<1%) 265 (68.6) 165 (70.2) 100 (66.2) 0.410
 Intermediate (1–5%) 121 (31.4) 70 (29.8) 51 (33.8) 0.410
 High (>5%) 0 (0.0) 0 (0.0) 0 (0.0) —
ASCVD risk score
 Low 216 (66.9) 139 (69.5) 77 (62.6) 0.201
 Intermediate 55 (17.0) 33 (16.5) 22 (17.9) 0.748
 High 52 (16.1) 28 (14.0) 24 (19.5) 0.191
Plaques by magnetic resonance
 Plaque presence 357 (88.8) 213 (88.8) 144 (88.9) 0.965
 Number of plaques 3 (1–5) 3 (1–5) 3 (1–4) 0.708
 Global plaque burden 328.7 (162.4–616.0) 321.6 (150.2–604.3) 347.3 (180.2–628.0) 0.647
Arterial uptake (18F-FDG)
 Presence of uptake 159 (39.5) 75 (31.2) 84 (51.9) <0.001
 Number of uptakes 0 (0–1) 0 (0–1) 1 (0–2) <0.001
 SUVmax arterial uptake 1.33 (1.23–1.45) 1.28 (1.18–1.37) 1.44 (1.32–1.54) <0.001
Open in a new tab

Data are presented as n (%) or median (Q1–Q3).

a

Measured in 622 of 745 individuals.

b

Bone marrow activation was defined when the mean BM SUVmax was above the median value (SUVmax 1.9).

Discussion

The present study analysed BM activation in a population of apparently healthy individuals from the PESA study who underwent whole body 18F-FDG PET/MRI. The main study findings are as follows: (i) BM activation is associated with MetS and its individual components, as well as with elevated numbers of leucocytes, and systemic inflammation; (ii) BM activation is associated with high arterial metabolic activity (high arterial 18F-FDG uptake), a surrogate for macrophage infiltration (i.e. the precursor of atherosclerosis); (iii) the association between BM activation and MetS, and between BM activation and arterial 18F-FDG uptake is maintained even in the absence of systemic inflammation; and (iv) co-occurring BM activation and vascular 18F-FDG uptake was significantly associated with more advanced stages of atherosclerosis (higher plaque prevalence, number of plaques, and plaque burden). To the best of our knowledge, this is the first demonstration of an association between BM activation and metabolic factors linked to atherosclerosis in humans.

The relationship between BM activation and acute cardiovascular events has been described extensively, both in experimental models and in clinical studies. In mice, acute myocardial infarction is followed by activation of the sympathetic nervous system that increases BM activation signals. These stimuli trigger the release of progenitor cells from the BM and activate monocyte production in the spleen,6,7 which together with local plaque macrophage proliferation produce a rapid turnover that facilitates atherosclerosis progression.39

Experimental studies have demonstrated that increased haematopoietic activity in BM plays a central role in the association between cardiovascular risk factors, vascular inflammation, and atherosclerosis formation. Studies in mice have demonstrated that low HDL-cholesterol levels and hypercholesterolaemia are associated with an increase in BM myelopoietic activity which leads to increased neutrophilia and monocytosis.10,11 Elevated HDL-cholesterol levels have an anti-atherogenic role based on the suppression of BM myeloid proliferation. In conditions such as obesity, inflamed adipose tissue increases BM haematopoietic cells proliferation, leading to an exacerbated inflammation and associated disease processes.12,13 Hyperglycaemia and diabetes have been associated with an increased production of inflammatory myeloid cells in the BM, which exacerbate diabetes mellitus-associated complications including atherosclerosis.14,15 Sympathetic activation present in hypertension has been demonstrated to modulate BM haematopoiesis with the increase of myeloid cells and contributing to atherosclerosis and cardiovascular disease.

However, evidence of the association between BM metabolic activity and cardiovascular risk factors in humans is lacking.16

In this PESA subcohort of middle-aged healthy participants, BM activation indexed as lumbar vertebrae 18F-FDG uptake was associated with the presence of MetS and its components, with higher frequencies detected for central obesity, hypertension, low HDL-C, triglycerides, and altered glucose metabolism in these participants. Bone marrow activation was also associated with increased haematopoiesis and systemic inflammation, assessed from circulating hs-CRP, ferritin, fibrinogen, P-selectin, and VCAM-1 levels. Notably, the association between BM activation and MetS was maintained even in the absence of systemic inflammation (Table 4). The activation of BM in the presence of MetS even in the absence of systemic inflammation suggests that an association between them exists. Indeed, cardiovascular risk factors, including lifestyle factors, have been shown to contribute to haematopoiesis activation.40 In line with this finding, we found an association between BM activation and increased haematopoiesis, even in the absence of systemic inflammation (Table 4). The increase in leucocyte numbers and red blood cell counts was slight but still significant. Bone marrow activation showed a significant association with high arterial metabolic activity (a precursor of atherosclerosis) indexed by arterial 18F-FDG uptake. These results suggest that BM activation is an early phenomenon occurring in response to MetS that contributes to the early stages of atherosclerosis (Structured Graphical Abstract). It has been shown that BM activation causes the release of haematopoietic progenitors into the blood stream as a response to different stimuli including acute cardiovascular events.6 In the absence of such an event, as in the asymptomatic PESA population studied here, BM activation in response to cardiovascular risk factors triggers the release of haematopoietic progenitors and starts the inflammatory process that leads to atherosclerosis initiation and progression. In this regard, haematopoietic progenitor cells mobilized from the BM during the atherosclerotic process and seeded in the spleen, contribute to leucocyte production.40

Metabolic syndrome has been associated with several lipid abnormalities such as elevated triglyceride levels, low HDL-cholesterol levels, and increased proportion of small dense LDL particles, despite optimal LDL-cholesterol levels.41,42 Consistent with this, in our population, the group with BM activation presented higher triglyceride and lower HDL-cholesterol levels, but there was no association with LDL-cholesterol and total cholesterol levels.

In our population, we have observed a myeloid-bias haematopoiesis in patients with BM activation (Table 1). This is consistent with experimental findings, which have demonstrated that cardiovascular risk factors are linked to an increased BM myeloid proliferation.10–13

Bone marrow activation is associated with early atherosclerosis, characterized by high arterial metabolic activity (arterial 18F-FDG uptake). Moreover, BM activation in the presence of arterial 18F-FDG uptake is associated with more advanced stages of atherosclerosis, characterized by higher plaque burden, suggesting that high arterial metabolic activity is a prerequisite to trigger atherogenesis.

Taken together, our results suggest that, in the presence of cardiovascular risk factors, mainly those associated with MetS, there is an increase in BM metabolic activity that contributes to early atherosclerosis by increasing inflammatory cell proliferation. Further research is needed to define the mechanisms involved in this process in order to identify targets to prevent subclinical atherosclerosis progression and its clinical complications.

Study limitations

The study was subject to selection bias because the subcohort undergoing whole body 18F-FDG PET/MRI was selected from the total PESA cohort based on the presence of subclinical atherosclerosis on vascular ultrasound; this design could also lead to collider bias, which also applies when looking at the non-inflammation subgroup. However, if we had studied participants with no evidence of atherosclerosis, associations might have been even stronger. Moreover, the cross-sectional nature of the study precludes a definite conclusion about causal relationship between risk factors, BM activation, and atherosclerosis. Based on pathology studies, the progression of areas with 18F-FDG uptake to fully grown plaques is thought to be part of the natural history of atherosclerosis, but it has not been formally demonstrated yet.

The availability of data on markers of systemic inflammation allows us to validate that the association between cardiometabolic risk factors and increased BM activity is not simply a reflection of systemic inflammation. Around 20% of participants invited to participate declined enrolment or had MRI contraindications, comparable to similar studies.43 As previously reported,27 the first 100 PET studies could not be used due to inaccuracies in MRI-based attenuation and reconstruction; however, feasibility was almost 100% thereafter.

Conclusions

In middle-aged apparently healthy individuals, BM activation, identified as 18F-FDG uptake, is associated with cardiovascular risk factors, mainly MetS, and its components. This association is present even in the absence of systemic inflammation. Bone marrow activation is associated with early atherosclerosis, characterized by high arterial metabolic activity (18F-FDG uptake).

Supplementary material

Supplementary material is available at European Heart Journal online.

Supplementary Material

ehac102_Supplementary_Data
Click here for additional data file. (38.1KB, docx)

Acknowledgements

Simon Bartlett provided English editing. Braulio Pérez Asenjo and Belén Arroyo helped with imaging acquisition and interpretation.

Contributor Information

Ana Devesa, Centro Nacional de Investigaciones Cardiovasculares (CNIC), c/Melchor Fernández Almagro 3, Madrid 28029, Spain; Cardiology Department, IIS-Fundación Jiménez Díaz University Hospital, Madrid, Spain; Icahn School of Medicine at Mount Sinai, New York, NY, USA.

Manuel Lobo-González, Centro Nacional de Investigaciones Cardiovasculares (CNIC), c/Melchor Fernández Almagro 3, Madrid 28029, Spain.

Juan Martínez-Milla, Centro Nacional de Investigaciones Cardiovasculares (CNIC), c/Melchor Fernández Almagro 3, Madrid 28029, Spain; Cardiology Department, IIS-Fundación Jiménez Díaz University Hospital, Madrid, Spain.

Belén Oliva, Centro Nacional de Investigaciones Cardiovasculares (CNIC), c/Melchor Fernández Almagro 3, Madrid 28029, Spain.

Inés García-Lunar, Centro Nacional de Investigaciones Cardiovasculares (CNIC), c/Melchor Fernández Almagro 3, Madrid 28029, Spain; Cardiology Department, Hospital Ramón y Cajal, Madrid, Spain; CIBER de Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain.

Annalaura Mastrangelo, Centro Nacional de Investigaciones Cardiovasculares (CNIC), c/Melchor Fernández Almagro 3, Madrid 28029, Spain.

Samuel España, Centro Nacional de Investigaciones Cardiovasculares (CNIC), c/Melchor Fernández Almagro 3, Madrid 28029, Spain; Departamento de Estructura de la Materia, Física Térmica y Electrónica, Universidad Complutense de Madrid, IdISSC, Madrid, Spain.

Javier Sanz, Centro Nacional de Investigaciones Cardiovasculares (CNIC), c/Melchor Fernández Almagro 3, Madrid 28029, Spain; Icahn School of Medicine at Mount Sinai, New York, NY, USA.

José M Mendiguren, Banco de Santander, Madrid, Spain.

Hector Bueno, Centro Nacional de Investigaciones Cardiovasculares (CNIC), c/Melchor Fernández Almagro 3, Madrid 28029, Spain; CIBER de Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain; Cardiology Department, Hospital Universitario 12 de Octubre, and i+12 Research Institute, Madrid, Spain.

Jose J Fuster, Centro Nacional de Investigaciones Cardiovasculares (CNIC), c/Melchor Fernández Almagro 3, Madrid 28029, Spain; CIBER de Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain.

Vicente Andrés, Centro Nacional de Investigaciones Cardiovasculares (CNIC), c/Melchor Fernández Almagro 3, Madrid 28029, Spain; CIBER de Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain.

Antonio Fernández-Ortiz, Centro Nacional de Investigaciones Cardiovasculares (CNIC), c/Melchor Fernández Almagro 3, Madrid 28029, Spain; CIBER de Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain; Hospital Clínico San Carlos, Universidad Complutense, IdISSC, Madrid, Spain.

David Sancho, Centro Nacional de Investigaciones Cardiovasculares (CNIC), c/Melchor Fernández Almagro 3, Madrid 28029, Spain.

Leticia Fernández-Friera, Centro Nacional de Investigaciones Cardiovasculares (CNIC), c/Melchor Fernández Almagro 3, Madrid 28029, Spain; Hospital Universitario HM Montepríncipe-CIEC, Madrid, Spain.

Javier Sanchez-Gonzalez, Philips Healthcare, Madrid, Spain.

Xavier Rossello, Centro Nacional de Investigaciones Cardiovasculares (CNIC), c/Melchor Fernández Almagro 3, Madrid 28029, Spain; CIBER de Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain; Cardiology Department, Hospital Universitari Son Espases-IDISBA, Palma de Mallorca, Spain.

Borja Ibanez, Centro Nacional de Investigaciones Cardiovasculares (CNIC), c/Melchor Fernández Almagro 3, Madrid 28029, Spain; Cardiology Department, IIS-Fundación Jiménez Díaz University Hospital, Madrid, Spain; CIBER de Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain.

Valentin Fuster, Centro Nacional de Investigaciones Cardiovasculares (CNIC), c/Melchor Fernández Almagro 3, Madrid 28029, Spain; Icahn School of Medicine at Mount Sinai, New York, NY, USA.

Funding

The PESA study is funded by the CNIC and Santander Bank. The present study was partially funded by an intramural grant CNIC-Severo Ochoa to D.S. and B.I. B.I. is supported by the European Commission (H2020-HEALTH 945118 and ERC-CoG 819775). The CNIC is supported by the ISCIII, the Ministry of Science and Innovation, and the Pro CNIC Foundation. CNIC is a Severo Ochoa Center of Excellence (CEX2020-001041-S).

Conflict of interest: J.S.-G. is a Philips employee. All other authors declare no conflicts of interest.

References

  • 1. Libby  P. Inflammation in atherosclerosis. Nature  2002;420:868–874. [DOI] [PubMed] [Google Scholar]
  • 2. Libby  P, Loscalzo  J, Ridker  PM, Farkouh  ME, Hsue  PY, Fuster  V, et al.  Inflammation, immunity, and infection in atherothrombosis: JACC review topic of the week. J Am Coll Cardiol  2018;72:2071–2081. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Poller  WC, Nahrendorf  M, Swirski  FK. Hematopoiesis and cardiovascular disease. Circ Res  2020;126:1061–1085. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Swirski  FK, Libby  P, Aikawa  E, Alcaide  P, Luscinskas  FW, Weissleder  R, et al.  Ly-6Chi monocytes dominate hypercholesterolemia-associated monocytosis and give rise to macrophages in atheromata. J Clin Invest  2007;117:195–205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Visseren  FLJ, Mach  F, Smulders  YM, Carballo  D, Koskinas  KC, Bäck  M, et al.  2021 ESC Guidelines on cardiovascular disease prevention in clinical practice. Eur Heart J  2021;42:3227–3337. [DOI] [PubMed] [Google Scholar]
  • 6. Dutta  P, Courties  G, Wei  Y, Leuschner  F, Gorbatov  R, Robbins  C, et al.  Myocardial infarction accelerates atherosclerosis. Nature  2012;487:325–329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Cremer  S, Schloss  M, Vinegoni  C, Brody  F, Zhang  S, Rohde  D, et al.  Diminished reactive hematopoiesis and cardiac inflammation in a mouse model of recurrent myocardial infarction. J Am Coll Cardiol  2020;75:901–915. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Courties  G, Herisson  F, Sager  HB, Heidt  T, Ye  Y, Wei  Y, et al.  Ischemic stroke activates hematopoietic bone marrow stem cells. Circ Res  2015;116:407–417. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Courties  G, Frodermann  V, Honold  L, Zheng  Y, Herisson  F, Schloss  MJ, et al.  Glucocorticoids regulate bone marrow B lymphopoiesis after stroke. Circ Res  2019;124:1372–1385. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Tolani  S, Pagler  TA, Murphy  AJ, Bochem  AE, Abramowicz  S, Welch  C, et al.  Hypercholesterolemia and reduced HDL-C promote hematopoietic stem cell proliferation and monocytosis: studies in mice and FH children. Atherosclerosis  2013;229:79–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Yvan-Charvet  L, Pagler  T, Gautier  EL, Avagyan  S, Siry  RL, Han  S, et al.  ATP-binding cassette transporters and HDL suppress hematopoietic stem cell proliferation. Science  2010;328:1689–1693. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Nagareddy  PR, Kraakman  M, Masters  SL, Stirzaker  RA, Gorman  DJ, Grant  RW, et al.  Adipose tissue macrophages promote myelopoiesis and monocytosis in obesity. Cell Metab  2014;19:821–835. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Singer  K, DelProposto  J, Lee Morris  D, Zamarron  B, Mergian  T, Maley  N, et al.  Diet-induced obesity promotes myelopoiesis in hematopoietic stem cells. Mol Metab  2014;3:664–675. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Nagareddy  PR, Murphy  AJ, Stirzaker  RA, Hu  Y, Yu  S, Miller  RG, et al.  Hyperglycemia promotes myelopoiesis and impairs the resolution of atherosclerosis. Cell Metab  2013;17:695–708. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Hoyer  FF, Zhang  X, Coppin  E, Vasamsetti  SB, Modugu  G, Schloss  MJ, et al.  Bone marrow endothelial cells regulate myelopoiesis in diabetes mellitus. Circulation  2020;142:244–258. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Al-Sharea  A, Lee  MKS, Whillas  A, Michell  DL, Shihata  WA, Nicholls  AJ, et al.  Chronic sympathetic driven hypertension promotes atherosclerosis by enhancing hematopoiesis. Haematologica  2019;104:456–467. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Heidt  T, Sager  HB, Courties  G, Dutta  P, Iwamoto  Y, Zaltsman  A, et al.  Chronic variable stress activates hematopoietic stem cells. Nat Med  2014;20:754–758. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Heyde  A, Rohde  D, McAlpine  CS, Zhang  S, Hoyer  FF, Gerold  JM, et al.  Increased stem cell proliferation in atherosclerosis accelerates clonal hematopoiesis. Cell  2021;184:1348–1361.e22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. van der Valk  FM, Kuijk  C, Verweij  SL, Stiekema  LCA, Kaiser  Y, Zeerleder  S, et al.  Increased haematopoietic activity in patients with atherosclerosis. Eur Heart J  2017;38:425–432. [DOI] [PubMed] [Google Scholar]
  • 20. Emami  H, Singh  P, Macnabb  M, Vucic  E, Lavender  Z, Rudd  JHF, et al.  Splenic metabolic activity predicts risk of future cardiovascular events: demonstration of a cardiosplenic axis in humans. JACC Cardiovasc Imaging  2015;8:121–130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Kim  EJ, Kim  S, Kang  DO, Seo  HS. Metabolic activity of the spleen and bone marrow in patients with acute myocardial infarction evaluated by 18F-fluorodeoxyglucose positron emission tomograpic imaging. Circ Cardiovasc Imaging  2014;7:454–460. [DOI] [PubMed] [Google Scholar]
  • 22. Tahara  N, Kai  H, Yamagishi  Si, Mizoguchi  M, Nakaura  H, Ishibashi  M, et al.  Vascular inflammation evaluated by [18F]-fluorodeoxyglucose positron emission tomography is associated with the metabolic syndrome. J Am Coll Cardiol  2007;49:1533–1539. [DOI] [PubMed] [Google Scholar]
  • 23. Nahrendorf  M, Frantz  S, Swirski  FK, Mulder  WJM, Randolph  G, Ertl  G, et al.  Imaging systemic inflammatory networks in ischemic heart disease. J Am Coll Cardiol  2015;65:1583–1591. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Taqueti  VR, Di Carli  MF, Jerosch-Herold  M, Sukhova  GK, Murthy  VL, Folco  EJ, et al.  Increased microvascularization and vessel permeability associate with active inflammation in human atheromata. Circ Cardiovasc Imaging  2014;7:920–929. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Teague  HL, Ahlman  MA, Alavi  A, Wagner  DD, Lichtman  AH, Nahrendorf  M, et al.  Unraveling vascular inflammation: from immunology to imaging. J Am Coll Cardiol  2017;70:1403–1412. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Tawakol  A, Ishai  A, Takx  RAP, Figueroa  AL, Ali  A, Kaiser  Y, et al.  Relation between resting amygdalar activity and cardiovascular events: a longitudinal and cohort study. Lancet  2017;389:834–845. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Fernández-Friera  L, Fuster  V, López-Melgar  B, Oliva  B, Sánchez-González  J, Macías  A, et al.  Vascular inflammation in subclinical atherosclerosis detected by hybrid PET/MRI. J Am Coll Cardiol  2019;73:1371–1382. [DOI] [PubMed] [Google Scholar]
  • 28. Fernández-Ortiz  A, Jiménez-Borreguero  LJ, Peñalvo  JL, Ordovás  JM, Mocoroa  A, Fernández-Friera  L, et al.  The progression and early detection of subclinical atherosclerosis (PESA) study: rationale and design. Am Heart J  2013;166:990–998. [DOI] [PubMed] [Google Scholar]
  • 29. Ibanez  B, Fernández-Ortiz  A, Fernández-Friera  L, García-Lunar  I, Andrés  V, Fuster  V. Progression of Early Subclinical Atherosclerosis (PESA) study. J Am Coll Cardiol  2021;78:156–179. [DOI] [PubMed] [Google Scholar]
  • 30. Fernández-Friera  L, Peñalvo  JL, Fernández-Ortiz  A, Ibañez  B, López-Melgar  B, Laclaustra  M, et al.  Prevalence, vascular distribution, and multiterritorial extent of subclinical atherosclerosis in a middle-aged cohort the PESA (Progression of Early Subclinical Atherosclerosis) study. Circulation  2015;131:2104–2113. [DOI] [PubMed] [Google Scholar]
  • 31. Arnett  DK, Blumenthal  RS, Albert  MA, Buroker  AB, Goldberger  ZD, Hahn  EJ, et al.  2019 ACC/AHA guideline on the primary prevention of cardiovascular disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol  2019;74:e177–e232. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Pisprasert  V, Ingram  KH, Lopez-Davila  MF, Munoz  AJ, Garvey  WT. Limitations in the use of indices using glucose and insulin levels to predict insulin sensitivity: impact of race and gender and superiority of the indices derived from oral glucose tolerance test in African Americans. Diabetes Care  2013;36:845–853. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. World Health Organization . Obesity: Preventing and Managing the Global Epidemic: Report of a WHO Consultation. Geneva: WHO; 1999. https://apps.who.int/iris/handle/10665/42330. [PubMed] [Google Scholar]
  • 34. Swarup  S, Goyal  A, Grigorova  Y, Zeltser  R. Metabolic syndrome. StatPearls Publ, 2020. https://www.ncbi.nlm.nih.gov/books/NBK459248/. [PubMed] [Google Scholar]
  • 35. Schulz  V, Torres-Espallardo  I, Renisch  S, Hu  Z, Ojha  N, Börnert  P, et al.  Automatic, three-segment, MR-based attenuation correction for whole-body PET/MR data. Eur J Nucl Med Mol Imaging  2011;38:138–152. [DOI] [PubMed] [Google Scholar]
  • 36. Mota-Cobian  A, Alonso-Farto  JC, Fernández-Friera  L, Sánchez-González  J, López-Melgar  B, Jiménez-Borreguero  LJ, et al.  The effect of tissue-segmented attenuation maps on PET quantification with a special focus on large arteries. Rev Esp Med Nucl Imagen Mol (Engl Ed)  2018;37:94–102. [DOI] [PubMed] [Google Scholar]
  • 37. Bucerius  J, Hyafil  F, Verberne  HJ, Slart  RHJA, Lindner  O, Sciagra  R, et al.  Position paper of the cardiovascular committee of the European Association of Nuclear Medicine (EANM) on PET imaging of atherosclerosis. Eur J Nucl Med Mol Imaging  2016;43:780–792. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Abdelbaky  A, Corsini  E, Figueroa  AL, Fontanez  S, Subramanian  S, Ferencik  M, et al.  Focal arterial inflammation precedes subsequent calcification in the same location. Circ Cardiovasc Imaging  2013;6:747–754. [DOI] [PubMed] [Google Scholar]
  • 39. Ye  YX, Calcagno  C, Binderup  T, Courties  G, Keliher  EJ, Wojtkiewicz  GR, et al.  Imaging macrophage and hematopoietic progenitor proliferation in atherosclerosis. Circ Res  2015;117:835–845. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Nahrendorf  M, Swirski  FK. Lifestyle effects on hematopoiesis and atherosclerosis. Circ Res  2015;116:884–894. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Paredes  S, Fonseca  L, Ribeiro  L, Ramos  H, Oliveira  JC, Palma  I. Novel and traditional lipid profiles in metabolic syndrome reveal a high atherogenicity. Sci Rep  2019;9:11792. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Ginsberg  HN, Zhang  YL, Hernandez-Ono  A. Metabolic syndrome: focus on dyslipidemia. Obesity (Silver Spring)  2006;14:41S–49S. [DOI] [PubMed] [Google Scholar]
  • 43. Fayad  ZA, Mani  V, Woodward  M, Kallend  D, Abt  M, Burgess  T, et al.  Safety and efficacy of dalcetrapib on atherosclerotic disease using novel non-invasive multimodality imaging (dal-PLAQUE): a randomised clinical trial. Lancet  2011;378:1547–1559. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ehac102_Supplementary_Data
Click here for additional data file. (38.1KB, docx)

Articles from European Heart Journal are provided here courtesy of Oxford University Press

RESOURCES