Background
Chronic inflammatory disease (CID) is a complex multisystem disease characterized by chronic inflammation, which can lead to coronary microvascular dysfunction (CMD) and can also predispose to coronary artery calcium deposition, even in the absence of obstructive coronary artery disease.
Methods
Twenty-one patients with systemic lupus erythematosus (SLE; mean age, 60 ± 11 years), 21 patients with systemic sclerosis (SSc; mean age, 66 ± 11 years), 32 patients with rheumatoid arthritis (RA; mean age, 65 ± 9 years), and 23 control subjects with comparable traditional risk factors for coronary artery disease (mean age, 65 ± 10 years) were prospectively enrolled in the outpatient clinic. All study participants underwent transthoracic Doppler-derived echocardiography for coronary flow velocity reserve (CFVR) measurement in the left anterior descending coronary artery; CFVR < 2.5 defined CMD. Coronary artery calcium score in the left anterior descending coronary artery was also assessed by computed tomography.
Results
None of study participants had obstructive coronary artery disease. The prevalence of CMD was 26% in the control group, 67% in the SLE group, 76% in the SSc group, and 63% in the RA group ( P < .05, CID groups vs control group). CFVR was significantly lower in all three CID groups than in the control group (control group, 3.01 ± 0.72; SLE group, 2.23 ± 0.71; SSc group, 2.14 ± 0.54; RA group, 2.33 ± 0.62; P < .05, CID groups vs control group). In contrast, coronary artery calcium scores were similar in the four groups and had no relation to CMD. The odds ratios for CMD in patients with SLE, SSc, and RA were 16.70, 25.78, and 8.44 ( P < .05) after adjusting for age, body mass index, the presence or absence of anemia, and hemoglobin level. Multiple linear regression analysis showed that only the presence of CID was independently associated with reduced CFVR among all study participants.
Conclusions
CID strongly contributes to CMD identified by qualitative evaluation of CFVR independently of traditional coronary risk factors of atherosclerosis but does not predispose to coronary artery calcification.
Coronary microvascular dysfunction (CMD) is functional abnormality in the regulation of myocardial blood flow and may precede coronary atherosclerosis. Inflammation has been shown to have a pathogenic role in endothelial dysfunction and CMD in different clinical settings. Indeed, recent clinical studies have demonstrated that patients with various chronic inflammatory diseases (CIDs), including systemic lupus erythematosus (SLE), systemic sclerosis (SSc), and rheumatoid arthritis (RA), had a high prevalence of CMD. Chronic inflammation also plays an important role in the pathogenesis of coronary artery calcification, a structural marker for early atherosclerosis. However, the relationship among CID, CMD, and coronary artery calcification in patients without obstructive CAD it not fully understood. Coronary flow velocity reserve (CFVR) has been considered a useful diagnostic index for the functional and physiologic assessment of CMD. Transthoracic Doppler-derived echocardiography (TTDE) enables noninvasive, accurate, and reproducible measurements of CFVR. Computed tomography is an effective method for detecting and quantifying coronary artery calcification. In this study, we evaluated whether CID has specific pathophysiologic pathways that underlie both CMD and coronary artery calcification, using CFVR measurement by TTDE and the Agatston calcium score method by computed tomography.
Methods
Patient Selection
A total of 87 consecutive patients with CID aged >35 years were prospectively recruited between January 2011 and December 2013. Ten patients with renal insufficiency (serum creatinine > 2.0 mg/dL) and three patients with severe respiratory insufficiency were excluded. Therefore, 74 patients with CID, including SLE ( n = 21), SSc ( n = 21), and RA ( n = 32), were ultimately included in this prospective study. Twenty-three control subjects aged >35 years who had comparable risk factors for coronary artery disease were also recruited to achieve groups of similar age and sex. All study participants were recruited from the outpatient clinics of rheumatology and cardiology in Yokkaichi Hazu Medical Center, and in-hospital patients were excluded from the present study. Anemia was defined as a reduction in the concentration of hemoglobin in a sample of venous blood when compared with reference values (<12.0 g/dL for women, <13.0 g/dL for men). The ability to perform activities of daily living was measured using the Modified Health Assessment Questionnaire (range, 0–3; a score of 0 represents no impairment of function) for all three patient groups. Disease activity of RA was measured using the Disease Activity Score on the basis of the evaluation of 28 joints, which is a validated composite index containing a 28-joint count for tenderness, a 28-joint count for swelling, the erythrocyte sedimentation rate, and the patient’s overall assessment of well-being. Disease activity of SLE was assessed by the Systemic Lupus Erythematosus Disease Activity Index. Written informed consent was obtained from all study participants, and the protocol was approved for use by the Human Studies Subcommittee of Yokkaichi Hazu Medical Center.
CFVR Measurement
All study participants underwent complete routine transthoracic echocardiography with CFVR measurement using an HD15 system (Philips Medical Systems, Andover, MA). Left ventricular ejection fraction and wall thickness were measured by two-dimensional echocardiography. Patients were defined as having left ventricular hypertrophy if interventricular septal or posterior wall thickness was ≥12 mm. To estimate the presence or absence of pulmonary hypertension, peak tricuspid regurgitation velocity was measured in each study subject. Coronary flow velocity was measured both at baseline and during intravenous infusion of adenosine triphosphate (ATP; 0.14 mg/kg/min) in the left anterior descending coronary artery (LAD) using TTDE, as previously described, and CFVR < 2.5, calculated as the ratio of hyperemic to basal mean diastolic flow velocity, defined CMD. Two experienced and blinded investigators measured CFVR by tracing the contour of the spectral Doppler signal using the computer incorporated in the ultrasound system. All measurements for calculating CFVR were averaged over three cardiac cycles.
Coronary Artery Calcium Score Measurement
All study participants underwent two computed tomographic scans (coronary calcium scoring and angiography) using a 320-row scanner with 0.5-mm detector elements, gantry rotation time of 350 msec, and up to 16 cm of coverage in the z direction (Aquilion ONE; Toshiba Medical Systems, Otawara, Japan), as described previously. Coronary artery calcium score (CACS) in the LAD as well as total CACS of the three major arteries were assessed on a postprocessing workstation (SURECardio Scoring; Toshiba Medical Systems). Two blinded and experienced investigators assessed coronary artery stenosis on a commercially available workstation (ZIO STATION System 1000; Amin/ZIOSOFT, Tokyo, Japan). Patients were included only if they had normal or minimal narrowing (<25% luminal reduction) in the three major coronary arteries assessed using computed tomography coronary angiography. In case the degree of coronary artery stenosis was not assessable by computed tomographic coronary angiography, additional invasive coronary angiography was performed to establish a definitive diagnosis.
Statistical Analysis
Categorical variables are presented as percentage frequencies and were analyzed using χ 2 tests, Fisher exact tests, or Kruskal-Wallis tests as appropriate. Continuous normally distributed variables are expressed as mean ± SD and were compared using Student’s two-tailed unpaired t test. Continuous data not normally distributed are expressed as median and interquartile range (IQR) and were analyzed using Mann-Whitney U tests. The Kolmogorov-Smirnov test was used to check for a normal distribution. CFVR was examined using one-way analysis of variance with the Bonferroni correction to localize the source of any difference. CACS was examined using the Kruskal-Wallis test. Bivariate correlations between study variables were calculated using Spearman rank correlation coefficients. All statistical analyses were performed with SPSS version 19 (SPSS, Chicago, IL). Two-sided P values > .05 were considered to indicate statistical significance.
Results
Patient Characteristics
None of the patients, including control subjects, had obstructive CAD. Patient characteristics were similar among the four groups except for body mass index (BMI), prevalence of anemia, and hemoglobin level ( Table 1 ). BMI and prevalence of anemia were higher and hemoglobin level was lower in the SSc group than in the control group. The median disease durations in the SLE, SSc, and RA groups were 9 years (IQR, 3–13 years), 7 years (IQR, 6–8 years), and 8 years (IQR, 4–10 years), respectively ( P = NS). The median Modified Health Assessment Questionnaire scores in the SLE, SSc, and RA groups were 0 (IQR, 0), 0.13 (IQR, 0–0.38), and 0 (IQR, 0–0.41), respectively ( P = NS). The median Systemic Lupus Erythematosus Disease Activity Index value in the SLE group was 0 (IQR 0), and the mean Disease Activity Score in the RA group was 2.4 ± 0.9. The prevalence of current prednisolone use in the SLE group was higher than in the SSc and RA groups (95.2%, 52.4%, and 46.9%, respectively; P < .05, SSc and RA groups vs SLE group). The median daily doses of prednisolone in the SLE, SSc, and RA groups were 6 mg/dL (IQR, 5–10 mg/dL), 2 mg/dL (IQR, 0−5 mg/dL), and 0 mg/dL (IQR, 0–3 mg/dL), respectively. The prevalence rates of current nonsteroidal anti-inflammatory drugs in the SLE and SSc groups were lower than in the RA group (9.5%, 4.8%, and 40.6%, respectively; P < .05, RA group vs SLE group; P < .05, RA group vs SSc group). Methotrexate (MTX) was administered to 63% of patients in the RA group. Almost all study participants had no signs of pulmonary hypertension on echocardiography; only one patient in the SSc group and two patients in the RA group had peak tricuspid regurgitation velocities between 2.9 and 3.4 m/sec in the absence of additional echocardiographic variables suggestive of pulmonary hypertension, indicating possible pulmonary hypertension.
| Variable | Control ( n = 23) | SLE ( n = 21) | SSc ( n = 21) | RA ( n = 32) | P |
|---|---|---|---|---|---|
| Age (y) | 65 ± 10 | 60 ± 11 | 66 ± 11 | 65 ± 9 | .150 |
| Women | 78% | 81% | 71% | 72% | .855 ∗ |
| BMI (kg/m 2 ) | 24.3 ± 3.3 | 22.0 ± 3.3 | 21.4 ± 2.7 ‡ | 23.0 ± 3.6 | .024 |
| Heart rate (beats/min) | 66 ± 10 | 69 ± 7 | 67 ± 15 | 67 ± 8 | .777 |
| Systolic BP (mm Hg) | 135 ± 21 | 127 ± 18 | 126 ± 25 | 139 ± 18 | .067 |
| Diastolic BP (mm Hg) | 77 ± 11 | 77 ± 13 | 72 ± 11 | 78 ± 10 | .228 |
| Hypertension | 57% | 52% | 38% | 53% | .640 ∗ |
| Diabetes mellitus | 39% | 29% | 24% | 25% | .663 ∗ |
| Hyperlipidemia | 48% | 43% | 43% | 28% | .443 ∗ |
| Current smoker | 13% | 19% | 10% | 19% | .805 ∗ |
| Obesity | 4% | 0% | 0% | 6% | .471 ∗ |
| Anemia | 13% | 24% | 52% ‡ | 34% | .036 ∗ |
| White blood cells (×10 3 /μL) | 5.3 (4.65–6.9) | 5.8 (4.7–7.4) | 6.0 (4.8–9.7) | 5.4 (4.6–6.525) | .427 † |
| Hemoglobin (g/dL) | 13.3 ± 1.1 | 13.1 ± 1.2 | 12.3 ± 1.4 ‡ | 12.7 ± 1.3 | .028 |
| Hematocrit (%) | 40.6 ± 3.0 | 40.0 ± 3.2 | 38.1 ± 4.2 | 38.7 ± 3.6 | .071 |
| Platelets (×10 3 /μL) | 232 (181–247) | 184 (129–213) | 207 (174–245) | 218 (179–243) | .238 † |
| Serum creatinine (mg/dL) | 0.8 ± 0.3 | 0.8 ± 0.2 | 0.8 ± 0.3 | 0.7 ± 0.2 | .311 † |
| Cholesterol (mg/dL) | |||||
| Total | 193 ± 25 | 193 ± 20 | 184 ± 28 | 193 ± 36 | .676 |
| High-density lipoprotein | 49 ± 12 | 60 ± 13 | 60 ± 20 | 57 ± 14 | .050 |
| Low-density lipoprotein | 112 ± 25 | 104 ± 20 | 95 ± 24 | 110 ± 27 | .094 |
| Triglycerides (mg/dL) | 164 ± 98 | 133 ± 62 | 131 ± 54 | 123 ± 72 | .221 |
| FBG (mg/dL) | 120 ± 24 | 111 ± 18 | 103 ± 14 | 114 ± 26 | .076 |
| CRP (mg/L) | 0.40 (0.2–0.7) | 0.40 (0.1–1.5) | 0.30 (0.1–0.7) | 0.75 (0.3–2.1) | .082 † |
| BNP (pg/mL) | 23.3 (10.8–41.8) | 20.6 (12.4–37.2) | 20.5 (8–55.7) | 25.9 (12.3–44.4) | .935 † |
| Medications | |||||
| ACE inhibitors/ARBs | 48% | 38% | 24% | 28% | .318 ∗ |
| CCBs | 30% | 14% | 19% | 13% | .397 ∗ |
| β-blockers | 22% | 5% | 5% | 6% | .208 ∗ |
| Statins | 44% | 43% | 43% | 31% | .746 ∗ |
| LV ejection fraction (%) | 67 ± 10 | 67 ± 7 | 68 ± 9 | 68 ± 5 | .836 |
| LV hypertrophy | 17% | 5% | 10% | 9% | .596 ∗ |
∗ Calculated with the Fisher exact test.
† Calculated with the Kruskal-Wallis test.
CFVR Measurement
Hemodynamics and coronary flow velocities at baseline and during ATP infusion are shown in Table 2 . There were no significant differences in heart rate, systolic blood pressure, and diastolic blood pressure among the four groups at baseline and during ATP infusion. Mean diastolic flow velocity at baseline tended to be higher in the three CID groups than in the control group, with no statistically significant difference. CFVR was significantly lower in the SLE, SSc, and RA groups than in the control group to similar extents (control group, 3.01 ± 0.72; SLE group, 2.23 ± 0.71; SSc group, 2.14 ± 0.54; RA group, 2.33 ± 0.62; P < .05, CID groups vs control group) ( Figure 1 A). Although only 26% of subjects in the control group had CFVR < 2.5, 67% in the SLE group, 76% in the SSc group, and 63% in the RA group had CFVR < 2.5 ( Table 3 ). Nine patients (43%) in the SLE group, eight patients (38%) in the SSc group, and 11 patients (34%) in the RA group but none in the control group had CFVR < 2.0. Except for the prevalence of CID and prednisolone use, patient characteristics, including age, gender, BMI, blood pressure, prevalence of traditional risk factors for coronary artery disease, C-reactive protein (CRP) level, and medications, were similar between patients with CFVR < 2.5 and those with CFVR ≥ 2.5 among all study participants. There was no significant difference in CFVR between patients who were receiving and those who were not receiving MTX in the RA group.
| Variable | Control ( n = 23) | SLE ( n = 21) | SSc ( n = 21) | RA ( n = 32) | P |
|---|---|---|---|---|---|
| Heart rate (beats/min) | |||||
| Baseline | 66.3 ± 10.1 | 69.3 ± 7.4 | 66.8 ± 14.8 | 67.0 ± 8.4 | .777 |
| ATP infusion | 73.4 ± 11.3 | 78.1 ± 9.9 | 73.9 ± 15.2 | 72.3 ± 8.9 | .327 |
| Systolic BP (mm Hg) | |||||
| Baseline | 134.5 ± 21.2 | 126.8 ± 18.3 | 126.3 ± 24.7 | 139.4 ± 18.0 | .067 |
| ATP infusion | 117.1 ± 20.6 | 116.0 ± 22.0 | 113.7 ± 25.9 | 128.3 ± 18.2 | .056 |
| Diastolic BP (mm Hg) | |||||
| Baseline | 76.5 ± 11.0 | 77.0 ± 13.0 | 71.8 ± 11.0 | 78.2 ± 10.1 | .228 |
| ATP infusion | 64.4 ± 12.0 | 68.5 ± 15.0 | 61.7 ± 13.4 | 70.8 ± 14.1 | .091 |
| Mean DFV (cm/sec) | |||||
| Baseline | 17.1 ± 4.7 | 19.8 ± 5.5 | 22.1 ± 9.4 | 20.4 ± 6.3 | .091 |
| ATP infusion | 49.5 ± 10.8 | 44.4 ± 17.2 | 44.3 ± 13.1 | 45.9 ± 13.6 | .560 |
| Group | No CMD ( n = 41) | CMD ( n = 56) | Unadjusted | Adjusted | ||
|---|---|---|---|---|---|---|
| Odds ratio (95% CI) | P | Odds ratio (95% CI) | P | |||
| Control | 74 (17%) | 26 (6%) | — | — | — | — |
| SLE | 33 (7%) | 67 (14%) | 5.67 (1.55–20.79) | .009 | 16.70 (3.22–86.52) | .001 |
| SSc | 24 (5%) | 76 (16%) | 9.07 (2.31–35.65) | .002 | 25.78 (4.47–148.70) | <.001 |
| RA | 37 (12%) | 63 (20%) | 4.72 (1.46–15.28) | .010 | 8.44 (2.17–32.74) | .002 |
CACS Measurement
None of the study participants had fixed coronary artery stenosis with >25% luminal reduction in the three major coronary arteries assessed by computed tomographic coronary angiography and/or invasive coronary angiography. CACS in the LAD was similar in the four groups (control group, 30 [IQR, 0–225]; SLE group, 0 [IQR, 0–138]; SSc group, 0 [IQR, 0–111]; RA group, 0 [IQR, 26–136]; P = NS) ( Figure 2 B). The frequencies of a calcium score of 0 in the LAD in the control, SLE, SSc, and RA groups were 35%, 62%, 57%, and 34%, respectively ( P = NS). Total CACS of the three major arteries was also similar in the four groups (data not shown). In addition, neither CACS in the LAD nor total CACS of the three major arteries was statistically different between patients with CFVR < 2.5 and those with CFVR ≥ 2.5 (median CACS in the LAD, 4 [IQR, 0–188] and 10 [IQR, 0–217], P = .527; median total CACS, 11 [IQR, 0–267] and 10 [IQR, 0–217], P = .571) in patients with CID.
Risk Factors for CMD
Among all study participants, SLE, SSc, and RA were associated with an increased prevalence of CMD after controlling for age, BMI, the presence or absence of anemia, and hemoglobin level ( Table 3 ). In contrast, SLE, SSc, and RA were not associated with the presence of coronary artery calcification (adjusted odds ratios: SLE, 0.38 [95% CI, 0.88–1.66]; SSc, 0.30 [95% CI, 0.69–1.33]; RA, 1.03 [95% CI, 0.29–3.67]).Univariate and stepwise multivariate linear regression analyses showed that only CID and mean diastolic flow velocity at baseline were independently associated with CFVR among all study participants ( Table 4 ). On the other hand, hematocrit and disease duration independently contributed to mean diastolic flow velocity at baseline ( Table 5 ). Among all patients with CID, there were no relationships between CFVR and age, type of disease, disease duration, dosage of prednisolone, duration of prednisolone use, hematocrit, and CRP level. In addition, there was no significant correlation between CACS and CFVR in each CID group, for neither CACS in the LAD (SLE group: r = −0.233, P = .31; SSc group, r = −0.30, P = .19; RA group, r = −0.03, P = .65) nor total CACS (SLE group, r = −0.028, P = .91; SSc group, r = 0.28, P = .23; RA group, r = −0.19, P = .31).
