Bone marrow activation in response to metabolic syndrome and early atherosclerosis

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: Bone marrow; Metabolic syndrome; PET/MRI; Subclinical atherosclerosis.

Structured Graphical Abstract

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.

Figure 1

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

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.

Figure 3

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

Figure 4

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.

Figure 5

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.

Figure 6

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

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.