C-reactive protein (CRP) levels predict adverse coronary events, but it is uncertain if they predict the burden or stability of vulnerable coronary plaques. In the Providing Regional Observations to Study Predictors of Events in the Coronary Tree study, 697 patients with acute coronary syndromes underwent percutaneous coronary intervention followed by 3-vessel angiography, gray-scale intravascular ultrasound (IVUS), and radiofrequency IVUS. Major adverse cardiac events (MACE) during 3 years of follow-up were adjudicated to initially treated culprit lesions or to untreated nonculprit lesions (NCLs). NCLs at greatest risk of causing subsequent MACE had plaque burden ≥70%, minimal luminal area ≤4.0 mm 2 , and/or thin-cap fibroatheroma morphology. Here, we examine the interaction of high-risk NCLs with CRP levels, which were measured at presentation, 1 month, and 6 months, then categorized at each time as normal (<3 mg/L), elevated (3 to 10 mg/L), or very elevated (>10 mg/L). We found that patients with elevated CRP levels at any time did not have more high-risk NCLs; however, untreated high-risk NCLs were more likely to cause subsequent MACE in patients with very elevated compared with normal 6-month CRP levels (for thin-cap fibroatheromas, 13.8% vs 1.9%, p = 0.0003; for lesions with minimal luminal area ≤4.0 mm 2 , 15.6% vs 2.2%, p <0.0001). As expected, patients with very elevated 6-month CRP levels had higher rates of subsequent NCL-related MACE (19.0% vs 7.2%, p = 0.039). In conclusion, the higher rates of NCL-related MACE in post–acute coronary syndrome patients with very elevated CRP levels may reflect greater instability of high-risk NCLs, rather than a larger burden of such lesions.
Chronic inflammation contributes to the formation and eventual destabilization of coronary plaques. The serum concentration of C-reactive protein (CRP), a nonspecific marker of inflammation, independently predicts the risk of coronary artery disease and acute coronary syndromes (ACS) but does not reliably correlate with the overall burden of atherosclerosis. High CRP levels may instead reflect the presence of “vulnerable” plaques at highest risk of acute thrombosis. This association would support current models of atherosclerosis, in which the leukocytes that infiltrate and destabilize plaques release cytokines that upregulate hepatic synthesis of CRP. Recent autopsy studies suggest that ACS occur after rupture of a thin-cap fibroatheroma (TCFA), which contains a thin fibrous cap overlying a large necrotic core. The Providing Regional Observations to Study Predictors of Events in the Coronary Tree (PROSPECT) study found that in patients presenting with ACS, the untreated nonculprit lesions (NCLs) most likely to cause subsequent major adverse cardiac events (MACE) had, in almost all cases, ≥1 of 3 high-risk features: TCFA phenotype by virtual histology intravascular ultrasound (VH-IVUS) imaging, plaque burden (PB) ≥70%, and minimal luminal area (MLA) ≤4.0 mm 2 . In this analysis from the PROSPECT study, we hypothesized that patients with elevated CRP levels would have a greater burden and/or more rapid progression of these high-risk lesions, which could explain their heightened risk for MACE.
Methods
The methods and principal results of the PROSPECT study have been described in detail. In brief, 697 patients with ACS were enrolled after successful, uncomplicated percutaneous coronary intervention (PCI). Patients were excluded if they had an ST-segment elevation myocardial infarction (STEMI) within 24 hours, cardiogenic shock, left ventricular ejection fraction ≤30%, serum creatinine concentration ≥2.5 mg/dl, <1-year life expectancy, or incomplete revascularization.
After revascularization, 3-vessel angiography and gray-scale and VH-IVUS were performed (Eagle Eye In-Vision Gold; Volcano Corporation, San Diego, California). All imaging analyses were performed offline and not used for procedural guidance. Patients were treated with aspirin indefinitely; clopidogrel for at least 1 year; and statins, β blockers, and angiotensin-converting enzyme inhibitors unless contraindicated. Patients were reevaluated at 1 month, 6 months, 1 year, and then yearly, with 3.4-year median follow-up.
The prespecified primary end point was the occurrence of MACE, a composite of cardiac death, cardiac arrest, myocardial infarction, or rehospitalization due to unstable or progressive angina. When MACE occurred, follow-up angiography was performed when clinically indicated to determine whether an originally treated culprit lesion or a previously untreated NCL was responsible. If follow-up angiography was not performed, the responsible lesion was classified as “indeterminate.” An independent committee adjudicated all clinical events without knowledge of CRP levels or baseline imaging characteristics.
All imaging analyses were performed at independent core laboratories at the Cardiovascular Research Foundation using prespecified techniques and without access to clinical data. Angiographic qualitative and quantitative measurements were obtained for the entire coronary tree using Medis CMS software, version 7.0 (Medis, Leiden, the Netherlands). Angiographic lesions were defined as those with ≥30% diameter stenosis; for each, the reference vessel diameter, minimal luminal diameter, and diameter stenosis were determined.
Gray-scale and VH-IVUS analyses were performed using QCU-CMS software (Medis), pcVH 2.1 software (Volcano), and qVH software (Cardiovascular Research Foundation). The external elastic membrane and luminal borders were contoured for each frame, permitting quantification of the cross-sectional areas (CSAs) of the external elastic membrane, the lumen, the plaque and media (CSA of the external elastic membrane minus the CSA of the lumen), PB (plaque and media CSA divided by the external elastic membrane CSA), and MLA. An intravascular ultrasound (IVUS) lesion was defined as possessing a PB of ≥40% in ≥3 consecutive frames.
Each IVUS lesion was further classified using VH-IVUS as a TCFA, thick-cap fibroatheroma, pathologic intimal thickening, fibrotic plaque, or fibrocalcific plaque. Gray-scale and VH-IVUS frames were co-registered to the angiographic roadmap using fiduciary branches for alignment, as previously described.
Serum CRP concentrations were measured at presentation and at 1 and 6 months of follow-up using a high-sensitivity assay at an independent core laboratory (Clinical Reference Laboratory, Lenexa, Kansas). On the basis of laboratory cut-off values for abnormality, as well as previous studies of inflammation and cardiovascular risk, CRP level was categorized at each time point as normal (<3 mg/L), elevated (3 to 10 mg/L), or very elevated (>10 mg/L).
Patients were stratified based on CRP levels and baseline demographics, imaging characteristics, and outcomes compared. In all comparisons, a p value of <0.05 was considered statistically significant.
Binary variables are summarized as percentages and were compared using the chi-square or Fisher’s exact test. Continuous variables are summarized as medians with interquartile ranges and were compared using the Kruskal-Wallis test. The relation between CRP level and subsequent MACE was determined using landmark analysis (i.e., only examining events occurring after the CRP measurement). Patients were censored at the last complete follow-up (in the case of drop out) or death but not for other events. Thus, in analysis of survival free from NCL-related MACE, patients were censored if they were lost to follow-up but not if they experienced MACE related to a culprit or indeterminate lesion.
Outcomes are displayed as time-to-event event curves, summarized as Kaplan-Meier estimates, and compared using log-rank tests. Multivariate outcomes analysis using Cox proportional hazards regression included the following covariates: CRP level (as a time-varying covariate), age, gender, body mass index, waist circumference, insulin-dependent diabetes, hypertension, hyperlipidemia, low-density lipoprotein cholesterol level, previous PCI, number of TCFAs, mean PB (across all lesions), and mean MLA. Only patients with complete covariate data were included in the analysis. A supremum test confirmed no violation of the proportional hazards assumption in this model (all p >0.1).
Results
673 patients had baseline CRP values; 571 had follow-up CRP values at 1 month, and 462 had follow-up CRP values at both 1 and 6 months ( Figure 1 ). The median CRP values were 7.2 mg/L (2.5 to 18.9) at presentation, 1.7 mg/L (0.8 to 4.0) at 1 month, and 1.4 mg/L (0.7 to 3.0) at 6 months.
Patients with higher initial CRP levels were more likely to have presented with STEMI or non-STEMI rather than unstable angina, and they were also more likely to have multiple culprit vessels ( Table 1 ). In contrast, patients with higher 6-month CRP levels were more likely to be smokers and to have diabetes, hypertension, heart failure, and higher Framingham risk scores ( Table 1 ). One-month CRP levels had a weaker association with baseline characteristics ( Supplementary Table S1 ); however, at all time points higher CRP levels were associated with greater body mass index and waist circumference. CRP levels were not associated with the use of aspirin, statin, and other medications at any time ( Supplementary Table S2 ).
| Patient Characteristics | CRP Level (mg/L) at Presentation (n = 673) | CRP Level (mg/L) at 6 Months (n = 462) | ||||||
|---|---|---|---|---|---|---|---|---|
| <3 (n = 190) | 3–10 (n = 206) | >10 (n = 277) | p | <3 (n = 345) | 3–10 (n = 87) | >10 (n = 30) | p | |
| Baseline Characteristics | ||||||||
| Age (years) | 59.2 [52.5–67.4] (n = 190) | 56.8 [50.1–65.6] (n = 206) | 58.1 [50.2–67.5] (n = 277) | 0.15 | 58.8 [51.7–66.6] (n = 345) | 57.6 [50.1–67.7] (n = 87) | 64.6 [58.0–70.0] (n = 30) | 0.05 |
| Men | 146/190 (76.8%) | 158/206 (76.7%) | 208/277 (75.1%) | 0.88 | 270/345 (78.3%) | 68/87 (78.2%) | 23/30 (76.7%) | 0.98 |
| White | 119/127 (93.7%) | 147/154 (95.5%) | 187/205 (91.2%) | 0.28 | 210/221 (95.0%) | 64/68 (94.1%) | 15/17 (88.2%) | 0.50 |
| Body mass index (kg/m 2 ) | 26.3 [23.9–29.6] (n = 188) | 28.0 [25.7–31.1] (n = 206) | 28.7 [25.8–32.3] (n = 275) | <0.0001 | 27.1 [24.6–30.3] (n = 344) | 28.7 [26.2–32.9] (n = 87) | 28.2 [26.4–33.8] (n = 30) | 0.002 |
| Waist circumference (cm) | 95.5 [90.0–104.0] (n = 158) | 101.6 [93.0–109.0] (n = 166) | 101.6 [95.0–111.8] (n = 239) | <0.0001 | 97.0 [91.4–105.0] (n = 284) | 103.0 [95.8–114.3] (n = 78) | 103.5 [96.0–119.4] (n = 26) | 0.0003 |
| Diabetes mellitus | 37/190 (19.5%) | 32/205 (15.6%) | 46/275 (16.7%) | 0.58 | 45/344 (13.1%) | 21/87 (24.1%) | 11/30 (36.7%) | 0.0005 |
| Hypertension requiring medication | 85/189 (45.0%) | 98/205 (47.8%) | 123/273 (45.1%) | 0.80 | 151/342 (44.2%) | 43/86 (50.0%) | 21/30 (70.0%) | 0.02 |
| Dyslipidemia requiring medication | 86/176 (48.9%) | 86/190 (45.3%) | 96/243 (39.5%) | 0.15 | 136/318 (42.8%) | 38/77 (49.4%) | 14/29 (48.3%) | 0.53 |
| Current smoker | 79/188 (42.0%) | 99/201 (49.3%) | 142/275 (51.6%) | 0.12 | 143/339 (42.2%) | 55/86 (64.0%) | 13/29 (44.8%) | 0.001 |
| Prior revascularization | 29/189 (15.3%) | 25/206 (12.1%) | 19/277 (6.9%) | 0.01 | 35/344 (10.2%) | 11/87 (12.6%) | 5/30 (16.7%) | 0.48 |
| Framingham risk score | 7.0 [5.0–9.0] (n = 188) | 7.0 [5.0–9.0] (n = 206) | 7.0 [4.0–9.0] (n = 274) | 0.29 | 7.0 [5.0–9.0] (n = 342) | 7.5 [5.0–10.0] (n = 86) | 8.0 [7.0–9.0] (n = 30) | 0.006 |
| Heart failure | 0/188 (0.0%) | 4/206 (1.9%) | 8/275 (2.9%) | 0.07 | 0/343 (0.0%) | 1/87 (1.1%) | 4/30 (13.3%) | <0.0001 |
| Initial presentation | ||||||||
| ST-segment elevation myocardial infarction | 56/190 (29.5%) | 51/206 (24.8%) | 98/277 (35.4%) | 0.04 | 111/345 (32.2%) | 27/87 (31.0%) | 4/30 (13.3%) | 0.10 |
| Non-ST-segment elevation myocardial infarction | 119/190 (62.6%) | 148/206 (71.8%) | 173/277 (62.5%) | 0.06 | 220/345 (63.8%) | 55/87 (63.2%) | 24/30 (80.0%) | 0.19 |
| Unstable angina pectoris | 15/190 (7.9%) | 7/206 (3.4%) | 6/277 (2.2%) | 0.008 | 14/345 (4.1%) | 5/87 (5.7%) | 2/30 (6.7%) | 0.67 |
| Multiple culprit vessels | 42/189 (22.2%) | 54/206 (26.2%) | 95/277 (34.3%) | 0.01 | 98/344 (28.5%) | 31/87 (35.6%) | 8/30 (26.7%) | 0.40 |
Patients with very elevated, elevated, and normal initial CRP levels had a comparable number of angiographic and IVUS-detected NCLs, as well as a similar number of TCFAs and lesions with MLA ≤4.0 mm 2 ( Table 2 ). IVUS-detected NCLs in patients with very elevated initial CRP levels had a larger MLA and were less likely to have PB ≥70%; however, the differences were small and not present among NCLs causing subsequent MACE.
| Variable | C-Reactive Protein (mg/L) | p | ||
|---|---|---|---|---|
| <3 | 3–10 | >10 | ||
| Angiographic findings | ||||
| Patient-level | n = 190 | n = 206 | n = 277 | |
| Number of untreated lesions | 2.0 [1.0–4.0] | 2.0 [1.0–4.0] | 2.0 [1.0–4.0] | 0.84 |
| All untreated lesions | n = 517 | n = 518 | n = 724 | |
| Length (mm) | 7.90 [5.31–12.24] | 7.89 [5.14–12.14] | 8.03 [5.23–11.63] | 0.89 |
| Minimal luminal diameter (mm) | 1.39 [1.04–1.83] | 1.43 [1.09–1.82] | 1.43 [1.08–1.85] | 0.60 |
| Diameter stenosis (%) | 37.3 [31.4–47.4] | 37.5 [32.2–46.5] | 37.2 [31.6–45.7] | 0.60 |
| Untreated lesions with subsequent events | n = 24 | n = 32 | n = 47 | |
| Length (mm) | 9.38 [6.74–16.57] | 11.63 [7.5–16.43] | 10.14 [8.15–14.76] | 0.85 |
| Minimal luminal diameter (mm) | 1.62 [0.92–2.02] | 1.39 [0.91–1.70] | 1.62 [1.25–2.11] | 0.11 |
| Diameter stenosis (%) | 37.1 [21.9–58.6] | 40.8 [28.1–54.9] | 34.1 [16.5–52.9] | 0.25 |
| Gray-scale IVUS findings | ||||
| Patient-level | n = 190 | n = 206 | n = 277 | |
| Number of untreated lesions | 4.5 [3.0–6.0] | 5.0 [4.0–6.0] | 5.0 [3.5–6.0] | 0.27 |
| All untreated lesions | n = 861 | n = 966 | n = 1309 | |
| MLA (mm 2 ) | 5.67 [4.15–7.74] | 5.91 [4.30–7.94] | 5.91 [4.37–8.26] | 0.03 |
| MLA ≤4.0 mm 2 | 190 (22.1%) | 187 (19.4%) | 247 (18.9%) | 0.17 |
| Plaque burden (%) | 56.4 [49.2–63.7] | 55.4 [48.6–63.1] | 55.2 [49.3–62.1] | 0.12 |
| Plaque burden ≥70% | 90 (10.5%) | 94 (9.7%) | 95 (7.3%) | 0.02 |
| Untreated lesions with subsequent events | n = 15 | n = 15 | n = 23 | |
| MLA (mm 2 ) | 3.74 [2.90–4.97] | 3.95 [3.16–5.17] | 3.93 [3.40–4.80] | 0.88 |
| MLA ≤4.0 mm 2 | 8 (53.3%) | 8 (53.3%) | 13 (56.5%) | 0.97 |
| Plaque burden (%) | 70.0 [58.8–76.5] | 70.9 [66.0–74.7] | 69.2 [63.3–73.5] | 0.70 |
| Plaque burden ≥70% | 8 (53.3%) | 8 (53.3%) | 9 (39.1%) | 0.59 |
| VH IVUS findings | ||||
| Patient-level | n = 190 | n = 206 | n = 277 | |
| Number of thin-cap fibroatheromas | 1.0 [0.0–2.0] | 1.0 [0.0–2.0] | 1.0 [0.0–2.0] | 0.63 |
| Number of thick-cap fibroatheromas | 2.0 [1.0–3.0] | 2.0 [1.0–3.0] | 2.0 [1.0–3.0] | 0.90 |
| Number of pathological intimal thickenings | 1.0 [0.0–2.0] | 1.0 [0.0–3.0] | 1.0 [0.0–3.0] | 0.07 |
| All untreated lesions | n = 768 | n = 873 | n = 1149 | |
| Thin-cap fibroatheroma | 191 (24.9%) | 182 (20.8%) | 259 (22.5%) | 0.15 |
| Thick-cap fibroatheroma | 304 (39.6%) | 318 (36.4%) | 425 (37.0%) | 0.37 |
| Pathological intimal thickening | 253 (32.9%) | 336 (38.5%) | 423 (36.8%) | 0.06 |
| Untreated lesions with subsequent events | n = 12 | n = 14 | n = 23 | |
| Thin-cap fibroatheroma | 6 (50.0%) | 5 (35.7%) | 14 (60.9%) | 0.33 |
| Thick-cap fibroatheroma | 3 (25.0%) | 7 (50.0%) | 8 (34.8%) | 0.40 |
| Pathological intimal thickening | 3 (25.0%) | 2 (14.3%) | 0 (0.0%) | 0.06 |
Patients with very elevated, elevated, and normal 6-month CRP levels again had a comparable number of angiographic and IVUS-detected NCLs, as well as a similar likelihood of IVUS-detected NCLs having PB ≥70%, TCFA phenotype, or MLA ≤4.0 mm 2 ( Table 3 ). Angiographic lesions in patients with very elevated levels had a smaller median diameter stenosis and larger minimal luminal diameter; however, the differences were small and not present among NCLs causing subsequent MACE.
| Variable | C-Reactive Protein (mg/L) | p | ||
|---|---|---|---|---|
| <3 | 3–10 | >10 | ||
| Angiographic findings | ||||
| Patient-level | n = 345 | n = 87 | n = 30 | |
| Number of untreated lesions | 2.0 [1.0–4.0] | 3.0 [1.0–4.0] | 2.5 [1.0–3.0] | 0.97 |
| All untreated lesions | n = 933 | n = 236 | n = 83 | |
| Length (mm) | 8.13 [5.52–12.09] | 8.16 [5.24–12.12] | 6.63 [4.94–10.72] | 0.08 |
| Minimal luminal diameter (mm) | 1.38 [1.03–1.78] | 1.52 [1.15–1.93] | 1.48 [1.20–1.86] | 0.03 |
| Diameter stenosis (%) | 38.3 [32.4–47.6] | 36.6 [31.5–44.0] | 35.3 [28.5–43.1] | 0.006 |
| Untreated lesions with subsequent events | n = 46 | n = 19 | n = 13 | |
| Length (mm) | 10.31 [7.19–16.57] | 12.85 [9.67–15.72] | 10.26 [7.51–14.33] | 0.80 |
| Minimal luminal diameter (mm) | 1.60 [0.93–2.04] | 1.59 [1.19–1.93] | 1.59 [0.84–2.58] | 0.83 |
| Diameter stenosis (%) | 35.9 [17.1–54.4] | 42.2 [32.5–55.5] | 17.4 [10.2–56.2] | 0.19 |
| Gray-scale IVUS findings | ||||
| Patient-level | n = 345 | n = 87 | n = 30 | |
| Number of untreated lesions | 5.0 [4.0–6.0] | 5.0 [4.0–6.0] | 5.0 [3.0–6.0] | 0.90 |
| All untreated lesions | n = 1644 | n = 396 | n = 141 | |
| MLA (mm 2 ) | 5.87 [4.26–8.00] | 5.61 [4.23–7.60] | 6.09 [4.54–8.20] | 0.47 |
| MLA ≤4.0 mm 2 | 320 (19.5%) | 84 (21.2%) | 32 (22.7%) | 0.52 |
| Plaque burden (%) | 55.5 [49.0–62.9] | 56.9 [49.4–64.0] | 55.3 [49.6–63.1] | 0.25 |
| Plaque burden ≥70% | 139 (8.5%) | 46 (11.6%) | 17 (12.1%) | 0.07 |
| Untreated lesions with subsequent events | n = 21 | n = 10 | n = 6 | |
| MLA (mm 2 ) | 3.74 [3.27–5.31] | 3.67 [2.93–4.36] | 3.53 [3.49–3.70] | 0.53 |
| MLA ≤4.0 mm 2 | 11 (52.4%) | 7 (70.0%) | 5 (83.3%) | 0.58 |
| Plaque burden (%) | 69.2 [66.5–71.7] | 70.7 [69.3–73.5] | 64.4 [49.0–75.2] | 0.50 |
| Plaque burden ≥70% | 10 (47.6%) | 6 (60.0%) | 2 (33.3%) | 0.58 |
| VH IVUS findings | ||||
| Patient-level | n = 345 | n = 87 | n = 30 | |
| Number of thin-cap fibroatheromas | 1.0 [0.0–1.0] | 0.0 [0.0–1.0] | 1.0 [0.0–2.0] | 0.86 |
| Number of thick-cap fibroatheromas | 2.0 [1.0–3.0] | 1.0 [1.0–2.0] | 2.0 [1.0–3.0] | 0.41 |
| Number of pathological intimal thickenings | 1.0 [0.0–3.0] | 2.0 [1.0–3.0] | 1.0 [1.0–2.0] | 0.37 |
| All untreated lesions | n = 1487 | n = 350 | n = 140 | |
| Thin-cap fibroatheroma | 312 (21.0%) | 72 (20.6%) | 29 (20.7%) | 0.98 |
| Thick-cap fibroatheroma | 566 (38.1%) | 117 (33.4%) | 58 (41.4%) | 0.17 |
| Pathological intimal thickening | 563 (37.9%) | 150 (42.9%) | 44 (31.4%) | 0.05 |
| Untreated lesions with subsequent events | n = 18 | n = 10 | n = 6 | |
| Thin-cap fibroatheroma | 8 (44.4%) | 4 (40.0%) | 4 (66.7%) | 0.56 |
| Thick-cap fibroatheroma | 9 (50.0%) | 4 (40.4%) | 1 (16.7%) | 0.35 |
| Pathological intimal thickening | 1 (5.6%) | 1 (10.0%) | 1 (16.7%) | 0.70 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
