2024 Volume 31 Issue 8 Pages 1225-1237
Aims: Vascular calcification is observed in advanced atherosclerotic lesions. Vascular calcification is considered to increase the risk of intraplaque hemorrhage and subsequent plaque destabilization; however, there is limited pathohistoological evidence of the association between vascular calcification and intraplaque hemorrhage. The aim of this study was to investigate the association between vascular calcification and intraplaque hemorrhage in the coronary arteries.
Methods: We examined 374 coronary arteries obtained from the autopsy samples of 126 deceased individuals. The vascular calcification levels of each artery were categorized into no calcification and quintiles of calcification area size among the arteries with calcification. Macrophage infiltration and neovascularization were also evaluated. The association of the calcification area, macrophage area, or number of vessels with the presence of intraplaque hemorrhage in the coronary arteries was estimated using a logistic regression analysis.
Results: Calcification lesions were observed in 149 coronary arteries. Arteries in the fourth quintile of calcification area size had a significantly greater likelihood of intraplaque hemorrhage than the arteries without calcification, after adjusting for confounders: odds ratio 13.13 (95% confidence interval: 2.97–58.16). After evaluating the influence of macrophage infiltration, the highest odds ratio of intraplaque hemorrhage was associated with the combination of large macrophage area and moderately sized calicification areas. The odds ratio of intraplaque hemorrhage additively increased with the combination of calcification and the number of vessels.
Conclusions: The present findings suggest that vascular calcification is significantly associated with intraplaque hemorrhage. The association between vascular calcification and intraplaque hemorrhage may decrease above a certain size of the calcification area.
Abbreviations and acronyms: CAC = coronary artery calcification, CKD = chronic kidney disease, CHD = coronary heart disease, eGFR = estimated glomerular filtration rate, HE = hematoxylin and eosin, HDL = high-density lipoprotein, OR = odds ratio, CI = confidence interval, MMP = matrix metalloprotease
Vascular calcificationis often observed in patients with advanced atherosclerotic lesions. Apoptotic cells, extracellular matrix, and necrotic core material may act as a nidus for microscopic calcium granules, which can subsequently expand to form larger lumps and plates of calcium deposits1). Coronary artery calcification (CAC) has been acknowledged to be more common in older people and in patients with diabetes mellitus and chronic kidney disease (CKD)1, 2).
Several clinical studies have reported a significant association between CAC and an increased risk of developing advanced coronary atherosclerosis and coronary heart disease (CHD)3-7). Clinical studies using calcium scores that have been estimated with computed tomography have shown that higher calcium scores are associated with major cardiac events and mortality5, 6). A meta-analysis reported the CAC score measured using electron-beam computed tomography to be an independent predictor of CHD events7). However, another clinical study conducted in patients with CHD showed that the degree of CAC at the site of the culprit coronary stenosis decreased with CHD severity (i.e., stable angina, unstable angina, and myocardial infarction)8). An intravascular ultrasound study reported that heavily calcified plaques are resistant to changes in the plaque volume9). Therefore, the influence of CAC on CHD events remains unclear.
Intraplaque hemorrhage is common in advanced coronary atherosclerotic lesions and it is considered to increase the risk of plaque destabilization10). Intraplaque hemorrhage is frequently observed in patients who die from a ruptured coronary artery11). Clinical studies using optical coherence tomography have revealed that the presence of intimal calcification, particularly spotty calcification, is significantly associated with plaque rupture5, 12). However, other studies have shown that calcification is associated with stable, rather than unstable plaque13). Therefore, the association between calcification and plaque rupture is still not fully understood.
The Hisayama Study is a prospective, population-based study of CVD risk factors in the Japanese people14). The study was characterized by autopsy verification of the cause of death in approximately 75% of patients who died15). We previously reported that CKD was associated with coronary atherosclerosis and intraplaque hemorrhage in the coronary arteries, and that individuals with CKD were more likely to have CAC than those without CKD16, 17). Therefore, we consider that it is worth using the same samples to evaluate the association between vascular calcification and intraplaque hemorrhage in coronary atherosclerosis. The aim of the present study was to investigate the size area of calcification that is most associated with the presence of intraplaque hemorrhage, using the coronary arteries obtained from autopsy specimens.
The Hisayama Study was established in 1961 in the town of Hisayama, a suburban community adjacent to Fukuoka City in the metropolitan area of Kyushu Island in southern Japan18). Hisayama has approximately 8000 inhabitants and the population size has been stable for 50 years. The study design and population have been described elsewhere16). Supplementary Fig.1 shows the detailed process used for the sample selection. Briefly, 1162 residents of Hisayama town died from January 1988 to November 2005; of these, 844 underwent autopsy examinations. We excluded individuals without health examination data within three years of death. A total of 482 autopsied cases were included. Among them, 126 autopsied cases (49 men and 77 women) were randomly selected from each estimated glomerular filtration rate (eGFR) category (eGFR levels ≥ 60, 45–59, 30–44, and ＜30 mL/min/1.73 m2) by matching for age at death and sex to evaluate coronary atherosclerosis. For these cases, 378 coronary arteries were selected to evaluate three coronary arteries per case (i.e., the right, left anterior descending, and left circumflex coronary arteries). After excluding one left anterior descending artery due to difficulty in evaluating the site of calcification and three arteries whose segments were not adequately defined, the remaining 374 coronary arteries were included in the present study.
eGFR, estimated glomerular filtration rate
The Kyushu University Institutional Review Board for Clinical Research approved the study design (approval number 2021-457). Written informed consent was obtained from the relatives of the autopsied individuals at the time of autopsy. All procedures performed in these studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee at which the studies were conducted and were in line with the 1964 Declaration of Helsinki.
Collection and Evaluation of Coronary ArteriesThe coronary arteries were obtained as previously described. Briefly, the heart tissue obtained at autopsy was immersed in 10% buffered formaldehyde. The right coronary artery (segment 1), left anterior descending coronary artery (segment 6), and left circumflex coronary artery (segment 11) were dissected free from the heart surface and then were embedded in paraffin. Sections from each block were serially stained with hematoxylin and eosin (HE), elastica-van Gieson, and Masson’s trichrome staining. Intraplaque hemorrhage was assessed using hematoxylin and eosin (HE)-stained and Masson’s trichrome-stained paraffin sections. We also measured the calcification area in each artery using H&E-stained specimens. To evaluate the region of calcification in the intima, we classified the intima into three intimal layer types: surface layer, deep layer, and whole layer. Calcification in the surface layer indicated that calcification existed within 50% of the distance from the lumen intima. Calcification in the deep layer indicates that calcification is present at over 50% of the distance from the lumen intima. Calcification in the entire layer indicates that calcification exists in the area from the surface layer to the deep layer in the intima. The plaque area was defined as the area between the internal elastic lamina and lumen, including the lipid core, macrophages, fibrous cap, and fibrous area. The plaque area was also measured. Intimal neovascularization (newly formed blood vessels) was identified as a circumferential brown product of endothelial cells labeled with anti-CD34 antibody as previously described17). Additionally, all coronary arteries were examined using immunohistochemistry for anti-CD68 antibody (KP-1, Dako, Carpinteria, CA, USA) and anit-CD34 antibody (NCL-END, Novocastra, Newcastle, UK). The CD68-positive area was captured and analyzed using the Scion Image software program version 4.0.3.2 (Scion Corporation, Frederick, MD, USA).
Kidney Function and Risk FactorsAt the health examination, study participants completed a questionnaire covering their medical history, use of drugs, smoking habits, and alcohol intake. In 2001, the serum creatinine concentration was measured using the Jaffé method. From 2002 to 2005, the creatinine concentrations were measured using the enzymatic method, and the values were converted to the Jaffé method according to a previously described equation19). eGFR was estimated using a simplified prediction equation derived from the 6-variable Modification of Diet in Renal Disease Study20). Diabetes was defined as glycated hemoglobin ＞6.5% or current use of anti-hyperglycemic drugs. The serum total cholesterol and high-density lipoprotein (HDL) cholesterol levels were measured enzymatically.
Statistical AnalysesAll analyses were performed on a per vessel basis. Data are presented as the mean±standard deviation and frequency for categorical variables. Differences and trends in continuous and categorical values across the calcification status were estimated using a linear or logistic regression model with generalized estimating equation methods, respectively, to deal with the correlations among the three main coronary arteries in a single person. Age- and sex-adjusted and multivariable-adjusted odds ratios (ORs) with corresponding 95% confidence intervals (CIs) were estimated using a logistic regression model with generalized estimating equation methods. The multivariable-adjusted model was adjusted for age, sex, eGFR, systolic blood pressure, diabetes, total cholesterol, HDL cholesterol, calcium, phosphate, hemoglobin concentration, current smoking, and current drinking. Continuous multivariable-adjusted associations of the calcification area and ORs with 95% CIs for intraplaque hemorrhage were plotted using restricted cubic splines with generalized estimating equation methods according to the macro formulated by the Harvard T.H. Chan School of Public Health21). We used three knots located at the 10th, 50th, and 90th percentiles of the calcified area. The value 0.88 mm2, which was the overall median calcification area, was chosen as the reference for each spline plot. Statistical analyses were conducted using the IBM SPSS software program version 22 (IBM Corp., Armonk, NY, USA), R software program version 3.5.2 (http://www.r-project.org), and SAS software program (version 9.4; SAS Institute Inc., Cary, NC, USA). A two-tailed P value of ＜0.05 was deemed to indicate statistical significance.
Of the 374 coronary arteries, 149 showed calcified lesions. All calcified lesions were present in the intima of the coronary artery. Table 1 shows the clinical characteristics of each artery according to the absence or presence of calcification. Generalized estimating equation methods were used to compare the differences between the groups because the clinical background of the three vessels from the same patient overlapped. Arteries with calcification were associated with a lower frequency among men and current smokers, and tended to be from patients with lower mean values of eGFR, higher serum phosphate levels, and a higher frequency of diabetes than those without calcification.
| Calcification (number of arteries) | Absence (225) | Presence (149) | P value |
|---|---|---|---|
| Age (y) | 83.9±7.1 | 85.8±6.2 | 0.050 |
| Men (%) | 46 | 30 | 0.011 |
| eGFR | 53.6±19.4 | 48.5±21.1 | 0.069 |
| Systolic blood pressure (mmHg) | 145±27 | 145±29 | 0.891 |
| Diastolic blood pressure (mmHg) | 75±12 | 73±12 | 0.303 |
| Diabetes (%) | 17 | 26 | 0.082 |
| Serum calcium (mg/dL) | 8.9±0.6 | 9.0±0.5 | 0.749 |
| Serum phosphate (mg/dL) | 3.3±0.6 | 3.4±0.5 | 0.065 |
| Serum total cholesterol (mg/dL) | 186±43 | 193±46 | 0.161 |
| Serum HDL cholesterol (mg/dL) | 56±16 | 55±15 | 0.694 |
| Hemoglobin (g/dL) | 11.6±2.0 | 11.6±2.0 | 0.981 |
| Serum albumin (g/dL) | 3.9±0.5 | 3.9±0.5 | 0.765 |
| Current smoking (%) | 47 | 28 | 0.003 |
| Current drinking (%) | 28 | 19 | 0.091 |
Abbreviations: eGFR, estimated glomerular filtration rate; HDL. High density lipoprotein
This analysis is performed on a per-vessel basis. Values expressed as a mean±standard deviation or frequency.
P values were estimated using a linear or logistic regression model with generalized estimating equation methods because the clinical background of the three vessels from the same subject overlapped.
Fig.1 shows representative images of coronary arteries with calcification and intraplaque hemorrhage. Intraplaque hemorrhage was occasionally observed in atherosclerotic lesions with calcification (Fig.1A–D). We investigated the association between calcification and intraplaque hemorrhage in each coronary artery. Each vessel was analyzed. The age- and sex-adjusted OR for the presence of intraplaque hemorrhage in arteries with calcification was 4.92 (95% CI, 2.03–11.91), as compared with arteries without calcification (Table 2). This association remained significant after adjusting for potential confounders: multivariable-adjusted OR 5.15 (95% CI, 2.11–12.55). We also examined the ratio of calcification to intraplaque hemorrhage according to coronary branches (Supplementary Table 1). The ratio of calcification to intraplaque hemorrhage was higher in the left circumflex arteries among the three coronary branches.
(A) A coronary artery with calcification and a massive hemorrhage (indicated by an ＊). The calcification area is 0.83 mm2. Arrowheads show the calcification area (A-D). Scale bars indicate 1.0 mm (A-D). (B) A coronary artery with moderate calcification and a moderate hemorrhage (＊). The calcification area is 1.68 mm2. (C) A coronary artery with moderate calcification, necrosis, and a small hemorrhage (＊). The calcification area is 3.24 mm2. (D) A coronary artery with a large calcification area. There is no intraplaque hemorrhage. The calcification area is 5.56 mm2. (E) Mean frequencies of intraplaque hemorrhage in the coronary arteries according to no calcification and quintiles of the calcification area size. Each bar represents the mean value±standard area. (F) The multivariable-adjusted model was adjusted for age, sex, estimated glomerular filtration rate, systolic blood pressure, diabetes, serum total cholesterol, serum high-density lipoprotein cholesterol, calcium, phosphate, hemoglobin concentration, current smoking, and current drinking. The solid line represents the adjusted odds ratio, and the dotted line represents the 95% confidence interval (CI). The calcification area at the highest odds ratio for intraplaque hemorrhage was 1.98 mm2.
| Calcification | Age- and sex-adjusted | Multivariable-adjusted＊ | ||||
|---|---|---|---|---|---|---|
| Odds ratio | 95% CI | P value | Odds ratio | 95% CI | P value | |
| Absence | 1.00 | reference | 1.00 | reference | ||
| Presence | 4.92 | 2.03 – 11.91 | ＜0.001 | 5.15 | 2.11 – 12.55 | ＜0.001 |
Abbreviation: CI, confidence interval
This analysis is performed on a per-vessel basis. The risk estimates were calculated using the logistic regression model with generalized estimating equation methods.
＊Adjusted for age, sex, estimated glomerular filtration rate, systolic blood pressure, diabetes, serum total cholesterol, serum HDL cholesterol, serum calcium, serum phosphate, hemoglobin, current smoking, and current drinking.
| coronary vessels | number | number of calcification | positive ratio of calcification (%) | P value | number of intraplaque hemorrhage | positive ratio of intraplaque hemorrhage (%) | P value |
|---|---|---|---|---|---|---|---|
| RCA | 126 | 46 | 36.5 | 6 | 4.8 | ||
| LAD | 123 | 64 | 52.0 | 0.002 | 13 | 10.6 | 0.007 |
| LCX | 125 | 39 | 31.2 | 5 | 4.0 |
Abbreviation: RCA, right coronary artery; LAD, left anterior descending artery; LCX, left circumflex artery
P values were estimated using chi-square test.
To examine which regions of calcification in the intima have the greatest likelihood of intraplaque hemorrhage, we classified the intima into surface, deep, and whole layers. Coronary arteries with intimal calcification in the deep layer had a higher likelihood of intraplaque hemorrhage than those with intimal calcification in the surface layer (Supplementary Table 2). Coronary arteries with intimal calcification in the whole layer had the greatest likelihood of intraplaque hemorrhage compared to those in the surface or deep layers and those with no intimal calcification (Supplementary Table 2).
| (Number of arteries) | Calcification (–) (225) | Cacification (+) | ||
|---|---|---|---|---|
| Surface layer (16) | Deep layer (39) | Whole layer (94) | ||
| Calcification area (mm2) | 0 | 0.37±0.09 | 0.50±0.11 | 1.62±0.11 |
| Frequency of intraplaque hemorrhage (%) | 2.67 | 6.25 | 10.26 | 13.83 |
| Age-and sex-adjusted odd ratio | 1.00 | 3.04 | 4.78 | 6.83 |
| (95% CI) | (reference) | (0.33–27.63) | (1.26–18.09) | (2.44–19.13) |
Abbreviation: CI, confidence interval
This analysis is performed on a per-vessel basis. Calcification area expressed as mean value±standard error.
Odds ratios with corresponding 95% confidence intervals were calculated using the logistic regression model with generalized estimating equation methods.
We examined the association between the size of the calcified area and the frequency of intraplaque hemorrhage in 374 coronary arteries. In this analysis, the calcification size in each artery was categorized into no calcification and quintiles of the calcification area size among the arteries with calcification. The frequency of intraplaque hemorrhage increased significantly with the size of the calcified area (Fig.1E, P for trend ＜0.01). Table 3 shows the clinical characteristics of the arteries according to the size of the calcified area. Arteries with a larger calcification area were associated with a lower frequency of current smokers and tended to be associated with higher mean serum phosphate levels. A multivariable-adjusted analysis showed that arteries in the fourth quintile of the calcification area size had a significantly greater likelihood of intraplaque hemorrhage than arteries without calcification, after adjusting for confounders: OR 13.13 (95% CI: 2.97–58.16) (Table 4). To evaluate the size of the calcification area with the greatest likelihood of intraplaque hemorrhage, we performed a logistic analysis with restricted cubic splines (Fig.1F). The calcification area with the greatest likelihood of intraplaque hemorrhage was 1.98 mm2, and this association slightly decreased for areas larger than this value.
| (Number of arteries) |
Calcification (-) (225) |
Calcification area (mm2) | |||||
|---|---|---|---|---|---|---|---|
| Q1 (＜0.26) (29) |
Q2 (0.26 – 0.58) (30) |
Q3 (0.59 – 1.16) (30) |
Q4 (1.17 – 1.95) (30) |
Q5 (≥ 1.96) (30) |
p for trend | ||
| Age (y) | 83.9±7.1 | 85.5±6.0 | 83.6±6.2 | 86.8±6.1 | 88.4±5.2 | 84.6±6.4 | 0.057 |
| Men (%) | 46 | 21 | 37 | 20 | 27 | 43 | 0.120 |
| eGFR | 53.6±19.4 | 42.0±21.1 | 53.6±24.0 | 47.3±21.0 | 47.4±16.9 | 51.9±21.2 | 0.266 |
| Systolic blood pressure (mmHg) | 145±27 | 148±34 | 138±27 | 147±25 | 149±30 | 146±30 | 0.743 |
| Diastolic blood pressure (mmHg) | 75±12 | 71.9±12.0 | 73.7±15.4 | 71.4±10.5 | 74.8±12.1 | 75.3±12.4 | 0.731 |
| Diabetes (%) | 17 | 31 | 13 | 20 | 30 | 33 | 0.107 |
| Serum calcium (mg/dL) | 8.9±0.6 | 9.0±0.5 | 8.9±0.4 | 9.0±0.6 | 8.9±0.5 | 9.1±0.5 | 0.532 |
| Serum phosphate (mg/dL) | 3.3±0.6 | 3.3±0.5 | 3.4±0.5 | 3.5±0.6 | 3.5±0.4 | 3.4±0.5 | 0.071 |
| Serum total cholesterol (mg/dL) | 186±43 | 188±45 | 192±41 | 195±46 | 192±48 | 199±50 | 0.102 |
| Serum HDL cholesterol (mg/dL) | 56±16 | 54±12 | 57±17 | 53±16 | 55±16 | 54±16 | 0.624 |
| Hemoglobin (g/dL) | 11.6±2.0 | 11.5±2.1 | 11.8±1.8 | 10.8±1.8 | 11.7±2.2 | 12.3±2.0 | 0.599 |
| Serum albumin (g/dL) | 3.9±0.5 | 3.8±0.5 | 3.9±0.4 | 3.9±0.6 | 3.8±0.4 | 4.0±0.5 | 0.803 |
| Current smoking (%) | 47 | 31 | 40 | 20 | 23 | 23 | 0.0036 |
| Current drinking (%) | 28 | 14 | 27 | 10 | 17 | 30 | 0.298 |
Abbreviations: eGFR, estimated glomerular filtration rate; LDL, low density lipoprotein; HDL. High density lipoprotein
This analysis is performed on a per-vessel basis. Values expressed as a mean±standard deviation or frequency.
P values were estimated using a linear or logistic regression model with generalized estimating equation methods.
| Calcification area (mm2) | age and sex adjusted | P for trend | multivariable adjusted | P for trend | |||||
|---|---|---|---|---|---|---|---|---|---|
| odds ratio | 95% CI | P-value | odds ratio | 95% CI | P-value | ||||
| Calcification (-) | 1.00 | reference | ＜0.001 | 1.00 | reference | ＜0.001 | |||
| Q1 | ＜0.26 | 1.11 | 0.08–15.19 | 0.938 | 1.15 | 0.14–9.73 | 0.901 | ||
| Q2 | 0.26–0.58 | 2.53 | 0.54–11.90 | 0.241 | 2.68 | 0.49–14.58 | 0.254 | ||
| Q3 | 0.59–1.16 | 6.18 | 2.01–18.96 | 0.002 | 7.14 | 2.12–24.04 | 0.002 | ||
| Q4 | 1.17–1.95 | 11.10 | 2.89–42.55 | ＜0.001 | 13.13 | 2.97–58.16 | ＜0.001 | ||
| Q5 | ≥ 1.96 | 6.85 | 1.86–25.22 | 0.004 | 7.14 | 2.12–23.98 | 0.002 | ||
Abbreviation: CI, confidence interval
Q1-Q5 indicate quintile of calcification area. This analysis is performed on a per-vessel basis.
The risk estimates were calculated using the logistic regression model with generalized estimating equation methods.
＊Adjusted for age, sex, estimated glomerular filtration rate, systolic blood pressure, diabetes, serum total cholesterol, serum HDL cholesterol, serum calcium, serum phosphate, hemoglobin, current smoking, and current drinking.
We also measured the plaque area in the coronary arteries and analyzed the ratio of calcification in the plaque to intraplaque hemorrhage (Table 5). The third quintile of the calcification rate had a significantly greater likelihood of intraplaque hemorrhage than arteries without calcification according to the multivariable-adjusted analysis: OR 34.34, (95% CI 9.09-129.72). In the fourth and fifth quintiles, the association between the calcification rate and intraplaque hemorrhage decreased. These results were similar to the results of the relationship between the calcification area and intraplaque hemorrhage (Table 4).
| ratio of the calcification area in the plaque area (%) | age and sex adjusted | P for trend | multivariable adjusted＊ | P for trend | |||||
|---|---|---|---|---|---|---|---|---|---|
| odds ratio | 95% CI | P-value | odds ratio | 95% CI | P-value | ||||
| Calcification (-) | 1.00 | reference | ＜0.001 | 1.00 | reference | ＜0.001 | |||
| Q1 | ＜4.35 | 2.43 | 0.42–13.93 | 0.318 | 2.30 | 0.41–12.74 | 0.341 | ||
| Q2 | 4.35–8.97 | 1.07 | 0.11–9.98 | 0.952 | 1.17 | 0.24–5.78 | 0.850 | ||
| Q3 | 8.98–17.52 | 21.68 | 6.95–67.66 | ＜0.001 | 34.34 | 9.09–129.72 | ＜0.001 | ||
| Q4 | 17.53–28.29 | 3.15 | 0.67–14.70 | 0.952 | 2.10 | 0.45–9.76 | 0.343 | ||
| Q5 | ≥ 28.30 | 4.38 | 1.00–19.23 | 0.050 | 5.96 | 1.37–25.99 | 0.018 | ||
Abbreviation: CI, confidence interval
Q1-Q5 indicate quintile of calcification area. This analysis is performed on a per-vessel basis.
The risk estimates were calculated using the logistic regression model with generalized estimating equation methods.
＊Adjusted for age, sex, estimated glomerular filtration rate, systolic blood pressure, diabetes, serum total cholesterol, serum HDL cholesterol, serum calcium, serum phosphate, hemoglobin, current smoking, and current drinking.
Vascular inflammation due to macrophage infiltration in plaques is considered to be involved in the increased risk of intraplaque hemorrhage10). To examine the association between macrophage infiltration and calcification and the presence of intraplaque hemorrhage, we investigated the influence of the calcification area size on the association between the size of the macrophage area and the presence of intraplaque hemorrhage. The macrophage infiltration area was evaluated by immunohistochemistry with an anti-CD68 antibody. Fig.2 shows representative images of a coronary artery with calcification and macrophage infiltration. An anti-CD68-positive macrophage area was observed in the atherosclerotic lesions of the intima (Fig.2A, B). With the increased size of the macrophage infiltration area, the frequency of intraplaque hemorrhage significantly increased (Supplementary Fig.2). We evaluated the OR of intraplaque hemorrhage for the combination of calcification and macrophage area (Fig.2C). The highest OR for intraplaque hemorrhage was the combination of the largest macrophage area (≥ 200000 µm2) and moderate-sized calcification area (Q4). There was no interaction between the calcification and macrophage areas.
A coronary artery with calcification and intraplaque hemorrhage (indicated by an ＊). Arrowheads indicate the calcification areas (A and B). Scale bars indicate 1.0 mm (A and B). (B) Immunohistochemical staining of CD68 in serial sections of the coronary artery. The brown color shows the CD68-positive macrophage area. (C) Bars represent age- and sex-adjusted odds ratios of intraplaque hemorrhage according to quintiles of calcification and CD68-positive macrophage areas. ＊＊P＜0.01 versus reference group. (D) Bars represent age- and sex-adjusted odd ratios of intraplaque hemorrhage according to the presence of a calcification area and the number of vessels. ＊P＜0.05, ＊＊P＜0.01 versus the reference group.
Each bar represents mean value±standard error.
Neovascularization is observed in advanced atherosclerotic plaques and it has been reported to affect plaque vulnerability11, 17). We examined the influence of neovascularization on the association between calcification and intraplaque hemorrhage. The frequency of intraplaque hemorrhage significantly increased with an increased number of vessels in the intima (Supplementary Fig.3). We evaluated the OR of intraplaque hemorrhage in the presence of calcification and number of vessels (Fig.2D). The odds ratio of intraplaque hemorrhage was observed to additively increase in the presence of calcification and the number of vessels. There was no interaction between the presence of calcification and the number of vessels.
Each bar represents mean value±standard error.
In this study, we histopathologically examined the relationship between calcification and intraplaque hemorrhage in the coronary arteries. The results showed that calcification in the coronary arteries was associated with intraplaque hemorrhage. Larger vascular calcification was significantly associated with a greater likelihood of intraplaque hemorrhage; however, the calcification area with the greatest likelihood of intraplaque hemorrhage was 1.98 mm2; and the likelihood decreased slightly with an area greater than this value (Fig.1F). The combination of greater macrophage infiltration and moderately sized calcifications in the intima showed the highest likelihood of intraplaque hemorrhage (Fig.2C). Coronary arteries with intimal calcification in the deep or whole layer had a greater likelihood of intraplaque hemorrhage than arteries with intimal calcification in the surface layer. These findings enhance our understanding of the relationship between intimal calcification and intraplaque hemorrhage.
Intraplaque hemorrhage is often observed in advanced coronary atherosclerotic lesions and it is associated with atheroma progression and plaque vulnerability10, 22). Intraplaque hemorrhage is more frequent in the coronary arteries at the sites of the culprit lesion in patients dying from plaque rupture11). Based on this evidence, intraplaque hemorrhage is thought to trigger plaque venerability and thereby contribute to plaque rupture22). In contrast, previous studies have shown that autopsy samples scanned with high-resolution micro-computed tomography demonstrated an association between microcalcification and local tissue stress and plaque instability in three-dimensional models23). Therefore, intimal calcification in the coronary arteries may carry the risk of intraplaque hemorrhage and thus lead to a risk of plaque rupture. We demonstrated that intimal calcification was significantly associated with the frequency of intraplaque hemorrhage in the coronary arteries. It has been suggested that calcification could increase local stress in the fibrous cap of atheromas during systole24). This finding suggests that intimal calcification is a risk factor for plaque rupture. However, intravascular ultrasound studies have revealed that heavily calcified plaques are more resistant to changes in the atheroma volume9). A biochemical model indicated that calcification within ruptured or stable plaques did not increase fibrous cap stress25). These results indicate that early calcification near the junction of the plaque may increase stress at the interface between the calcification and atherosclerotic lesions, thus leading to plaque rupture. However, more extensive calcification can possibly eliminate these weaknesses and may reduce the risk of rupture26).
The recruitment of monocytes to the intima is the initial step in the development of atherosclerosis27). Monocytes acquire macrophage characteristics and express scavenger receptors. Tissue macrophages and foam cells produce cytokines in response to the noxious effects of oxidized lipoproteins. Inflammatory mediators found in atheromas include interleukin 1β, tumor necrosis factor-α, CD40 ligand, and expressed matrix metalloprotease (MMP). The dissolution of the collagenous matrix in the fibrous cap by MMP renders this structure weak, friable, and susceptible to rupture when exposed to hemodynamic stress. The biochemical model above suggests that a greater lipid area, which induces more MMP, is associated with increased vulnerability in the vessel wall and may lead to plaque rupture25). These results suggest that atherosclerotic plaques, due to the accumulation of foamy macrophages, might be weak against hemodynamic stress, thus leading to plaque rupture. Intimal neovascularization is also associated with plaque progression and is a source of intraplaque hemorrhage and related to vulnerability to rupture11). Neovascularization and calcification was also found to additively increase the risk of intraplaque hemorrhage (Fig.2D).
At autopsy, some plaques consisted of fibrous and sometimes calcified tissue without extracellular lipid pools or a necrotic core1). The necrotic core can completely calcify over time, and calcifications can constitute most of the plaque volume1). Therefore, calcification may occur as a result of plaque regression28). However, we do not know why the risk of intraplaque hemorrhage does not linearly increase with calcification size. The healing process in atheroma progresses through overlapping phases of acute inflammation, resolving inflammation, proliferation and remodeling28). These balances may affect the incidence of intraplaque hemorrhage in atheromas with calcifications. A larger atheroma with calcification may thus contribute to the occurrence of intraplaque rupture (Fig.2C). A smaller atheroma with calcification was associated with a lower risk of intraplaque hemorrhage (Fig.2C). The phase of inflammation and size of the atheroma may affect the relationship between calcification and the incidence of intraplaque hemorrhage. A higher level of resolving inflammation may contribute to the enlargement of calcification without intraplaque hemorrhage in fibrocalcific plaques.
The strength of this study is that it consisted of a large number of autopsy samples from a population-based cohort study. We confirmed the health examination data of each individual after adjusting for confounding factors. This study is associated with several limitations. First, this was a cross-sectional study; thus, we could not determine the causality between calcification and intraplaque hemorrhage in the coronary arteries. Second, information regarding the severity or duration of cardiovascular disease risk factors was not included. Third, this study was based on autopsies, and the individual participants’ ages tended to be very advanced. Therefore, the findings of this study may not apply to the general population. Fourth, no data were available on medication use. Fifth, the autopsy cases in this study were based on previous studies16, 17), that evaluated the relationship between atherosclerotic lesions and kidney function. The selection of previous studies may also have caused some bias in this study. Nevertheless, the information gained in this study is considered to improve our overall understanding of the pathogenesis of calcification in the coronary arteries.
Intimal calcification in the coronary arteries was found to be associated with the presence of intraplaque hemorrhage, which is possibly linked to plaque vulnerability. Small-sized calcifications in the intima may increase the risk of plaque rupture; however, the risk of intraplaque hemorrhage may decrease above a certain size of the calcification area.
We thank the residents of Hisayama for their participation in the survey, and the staff of the Division of Health of Hisayama for their cooperation in this study. We sincerely thank the Department of Anatomic Pathology, Graduate School of Medical Sciences, Kyushu University, who provided insight and expertise in analyzing the autopsy findings, which greatly assisted in the research. statistical analyses using the SAS software program version 9.4 were performed using computers offered under the category of General Projects by the Research Institute for Information Technology, Kyushu University. We thank Edanz (https://jp.edanz.com/ac) for editing this manuscript.
This study was supported in part by the Ministry of Education, Culture, Sports, Science, and Technology of Japan (JSPS KAKENHI Grant Numbers JP21H03200, JP19K07890, JP20K10503, JP20K11020, JP21K07522, JP21K11725, JP21K10448, JP18K17925, JP20590342, JP22590892, and JP23590400), the Health and Labour Sciences Research Grants of the Ministry of Health, Labour and Welfare of Japan (JPMH20FA1002), Grand for pathophysiological research conference in chronic kidney disease, and by the Japan Agency for Medical Research and Development (JP21dk0207053).
All authors declare that they have no competing interests.
T. Nakano contributed to the study design, statistical analysis, data interpretation, and manuscript drafting. H.K. contributed to statistical analyses. J.H. and K.M. and T.K. contributed to the critical revision of the manuscript. T. Ninomiya contributed to study supervision and critical revision of the manuscript. All the authors critically reviewed the draft of the manuscript and approved its final version.
