Left ventricular (LV) remodeling after myocardial infarction (MI) indicates a high risk of heart failure and death, but LV remodeling remains difficult to predict. Biomarkers may help to refine risk stratification for a more personalized medical approach. They may also shed light on the pathophysiologic processes involved. We performed a systematic review of the published evidence about the association of circulating biomarkers with LV remodeling after MI. We selected 59 publications. Overall, these studies examined 112 relations between 52 different biomarkers and LV remodeling. The biomarkers most consistently associated with LV remodeling were involved in extracellular matrix turnover or neurohormonal activation: matrix metalloproteinase-9, collagen peptides, and B-type natriuretic peptide. This review underscores the vitality of the research on LV remodeling but concludes that the ideal biomarker has not yet been identified. To reach this goal, future studies will have to be larger, have standardized imaging end points, and include replication populations to define optimal cutoffs for LV remodeling prediction. Cardiovascular magnetic resonance appears to be the best technique for LV remodeling assessment but its current availability may be a concern for recruitment for multicenter studies. Recent technologic advances will probably yield new candidate biomarkers of LV remodeling. Tests are necessary to determine whether a multimarker approach would significantly improve risk prediction.
Biomarkers are biological variables that can provide information about a condition of interest. In cardiac diseases, biomarkers may include demographic features, cardiac imaging findings, and genetic polymorphisms. The term is most often applied, however, to circulating serum or plasma analytes beyond those used in routine hematology and biochemistry tests. Over the previous decade, many studies have reported relations between circulating biomarkers and left ventricular (LV) remodeling after myocardial infarction (MI). The aim of these studies is twofold: (1) to improve pathophysiologic knowledge about LV remodeling and (2) to identify markers that can be used in clinical practice to refine risk stratification after MI. In view of an absence of a comprehensive review article on this topic, we systematically searched for, reviewed, and assessed published evidence on the association of circulating biomarkers with LV remodeling after MI.
Methods
We conducted a computerized Medline search of published articles through November 2011. We searched for “ventricular remodeling AND myocardial infarction,” “ventricular remodeling AND biomarkers,” and “myocardial infarction AND heart failure AND biomarkers.” Bibliographies of all relevant articles were searched manually for additional articles. Next, we performed a computerized search of the Journal Citation Reports through November 2011 to retrieve all articles citing any of the articles identified in the first step. Only English-language articles reporting original data were eligible for inclusion in the study. Review articles were searched for additional references but were not included in the analysis.
An article was judged relevant if it was a cohort study or a clinical trial of patients admitted with acute MI and it reported the measurement of ≥1 biomarker and LV remodeling data in ≥30 patients. We included studies that analyzed biomarkers in peripheral venous samples but did not consider those measuring transcardiac production of biomarkers or those used to assess infarct size such as creatine phosphokinase or cardiac troponin. Studies with a follow-up of <1 month after MI were not included. We included studies that reported LV volumes or LV diameters as indicators of LV remodeling. Because variability of the data reported (morphologic and biological variables on their original continuous scales or dichotomized into 2 groups) precluded a formal meta-analysis, relations between biomarkers and LV remodeling are presented in Table 1 as positive if a high level of the biomarker was significantly associated (p <0.05) with increased LV remodeling, negative if a low level of the biomarker was significantly associated (p <0.05) with increased LV remodeling, and none in the absence of any significant association. To visualize the association of a given biomarker with LV remodeling, data were grouped by biomarker; thus, publications that assessed >1 biomarker appear >1 time in Table 1 . Biomarkers are presented with their Gene Ontology classification.
| Biomarker | Number of Patients | Timing of Blood Sampling | Method for Assessment of LV Remodeling | Timing of Assessment of LV Remodeling | Correlation With LV Remodeling | Reference |
|---|---|---|---|---|---|---|
| GO: 0004222 metallopeptidase activity | ||||||
| Granzyme B | 33 | baseline, 2 wk | CVG | 6 mo | positive | Kondo et al |
| Matrix metalloproteinase-2 (MMP-2) | 60 | baseline | Echo | 6 wk | none | Squire et al |
| 32 | baseline, 1, 3, 6 mo | Echo | 1, 3 mo | none | Webb et al | |
| 52 | baseline, 1 mo, 1 y, 4 y | CMR | 4 y | none | Orn et al | |
| 91 | baseline | Echo | 6 mo | none | Kelly et al | |
| 100 | baseline, 3, 6 mo | CMR | 6 mo | none | Weir et al | |
| Matrix metalloproteinase-3 (MMP-3) | 382 | baseline | Echo | 5 mo | positive | Kelly et al |
| 100 | baseline, 3, 6 mo | CMR | 6 mo | positive | Weir et al | |
| Matrix metalloproteinase-7 (MMP-7) | 32 | baseline, 1, 3, 6 mo | Echo | 1, 3 mo | none | Webb et al |
| Matrix metalloproteinase-8 (MMP-8) | 32 | baseline, 1, 3, 6 mo | Echo | 1, 3 mo | none | Webb et al |
| Matrix metalloproteinase-9 (MMP-9) | 32 | baseline, 1, 3, 6 mo | Echo | 1, 3 mo | positive | Webb et al |
| 91 | baseline | Echo | 6 mo | positive | Kelly et al | |
| 404 | baseline | Echo | 10 mo | positive | Kelly et al | |
| 65 | baseline, 1 y | Echo | 1 y | positive | Miyazaki et al | |
| 60 | baseline | Echo | 6 wk | none | Squire et al | |
| 52 | baseline, 1 mo, 1 y, 4 y | CMR | 4 y | none | Orn et al | |
| 100 | baseline, 3, 6 mo | CMR | 6 mo | none | Weir et al | |
| Tissue plasminogen activator antigen (t-PA) | 100 | baseline, 3, 6 mo | CMR | 6 mo | positive | Weir et al |
| GO: 0008191 metallopeptidase inhibitor activity | ||||||
| Tissue inhibitor of metalloproteinase-1 (TIMP-1) | 404 | baseline | Echo | 10 mo | positive | Kelly et al |
| 32 | baseline, 1, 3, 6 mo | Echo | 1, 3 mo | none | Webb et al | |
| 100 | baseline, 6 mo | CMR | 6 mo | none | Weir et al | |
| Tissue inhibitor of metalloproteinase-2 (TIMP-2) | 100 | baseline, 6 mo | CMR | 6 mo | positive | Weir et al |
| 32 | baseline, 1, 3, 6 mo | Echo | 1, 3 mo | none | Webb et al | |
| Tissue inhibitor of metalloproteinase-4 (TIMP-4) | 100 | baseline, 6 mo | CMR | 6 mo | positive | Weir et al |
| GO: 0007155 cell adhesion | ||||||
| Amino-terminal propeptide of type III procollagen (PIIINP) | 47 | baseline, 3 mo, 6 mo 1 y | Echo | 1 y | positive | Poulsen et al |
| 35 | baseline, 1 mo | Echo | 6 mo | positive | Radovan et al | |
| 128 | baseline, 3 mo | Echo | 3 mo | positive | Li et al | |
| Carboxy-terminal propeptide of type I procollagen (PICP) | 48 | baseline, 3 mo, 6 mo, 1 y | Echo | 1 y | positive | Poulsen et al |
| 35 | baseline, 1 mo | Echo | 6 mo | positive | Radovan et al | |
| 56 | baseline, 1, 6 mo | Echo | 6 mo | none | Cerisano et al | |
| Carboxy-terminal telopeptide of type I procollagen (ICTP) | 56 | baseline, 1, 6 mo | Echo | 6 mo | positive | Cerisano et al |
| Tenascin-C | 105 | baseline, 2 wk, 1 mo | RNA | 6 mo | positive | Sato et al |
| von Willebrand factor (vWF) | 100 | baseline, 3, 6 mo | CMR | 6 mo | positive | Weir et al |
| GO: 0005179 hormone activity | ||||||
| Aldosterone | 100 | baseline, 3, 6 mo | CMR | 6 mo | positive | Weir et al |
| Apelin | 100 | baseline, 6 mo | CMR | 6 mo | none | Weir et al |
| Atrial natriuretic peptide (ANP) | 30 | baseline, 2 wk, 1 mo | CVG | 1 mo | positive | Nagaya et al |
| 33 | baseline, 1, 3 mo | CVG | 1, 3 mo | positive | Yoshitomi et al | |
| 74 | 1 mo | CVG | 5–9 mo | none | Hirayama et al | |
| N-terminal pro–atrial natriuretic peptide (NT–pro-ANP) | 71 | baseline, 3 mo | Echo | 2 y | positive | Hole et al |
| Brain natriuretic peptide (BNP) | 30 | baseline, 2 wk, 1 mo | CVG | 1 mo | positive | Nagaya et al |
| 33 | baseline, 1, 3 mo | CVG | 1, 3 mo | positive | Yoshitomi et al | |
| 133 | baseline, 2 mo | Echo | 2 mo | positive | Crilley and Farrer | |
| 74 | 1 mo | CVG | 5–9 mo | positive | Hirayama et al | |
| 72 | baseline, 3 mo | Echo | 1 y | positive | Takagi et al | |
| 105 | baseline, 2 wk, 1 mo | RNA | 6 mo | positive | Sato et al | |
| 56 | baseline, 1, 6 mo | Echo | 6 mo | positive | Cerisano et al | |
| 30 | baseline, 3 mo, 6 mo, 2 y | Echo | 3 mo, 6 mo, 2 y | positive | Grybauskiene et al | |
| 34 | baseline, 1 mo | RNA | 1 mo | positive | Cerisano et al | |
| 106 | 1, 6 mo | CVG | 6 mo | positive | Hirayama et al | |
| 82 | baseline, 1, 6 mo | Echo and CMR | 6 mo | positive | Garcia-Alvarez et al | |
| 100 | baseline, 3, 6 mo | CMR | 6 mo | positive | Weir et al | |
| 246 | baseline, 1 mo, 3 mo 1 y | Echo | 1 y | positive | Fertin et al | |
| 108 | baseline | Echo | 1 y | none | Dominguez-Rodriguez et al | |
| N-terminal pro–brain natriuretic peptide (NT–pro-BNP) | 42 | baseline, 1, 3, 6 mo, 1 y | CMR | 1 y | positive | Nilsson et al |
| 106 | baseline | Echo | 3 mo | positive | Xiaozhou et al | |
| 52 | baseline, 1 mo, 1 y 4 y | CMR | 4 y | positive | Orn et al | |
| 404 | baseline | Echo | 10 mo | positive | Kelly et al | |
| 61 | baseline, 6 mo | Echo | 6 mo | positive | Giallauria et al | |
| 100 | baseline, 3, 6 mo | CMR | 6 mo | positive | Weir et al | |
| 206 | baseline | CMR | 4–6 mo | positive | Haeck et al | |
| 159 | baseline, 6 mo | Echo | 6 mo | positive | López Haldón et al | |
| 48 | baseline, 1, 3 mo | CMR | 3 mo | positive | Mather et al | |
| 60 | baseline | Echo | 6 wk | none | Squire et al | |
| 91 | baseline | Echo | 6 mo | none | Kelly et al | |
| Norepinephrine | 100 | baseline, 3, 6 mo | CMR | 6 mo | positive | Weir et al |
| Arginine vasopressin | 100 | baseline, 6 mo | CMR | 6 mo | none | Weir et al |
| C-terminal provasopressin, copeptin | 274 | baseline | Echo | 5 mo | positive | Kelly et al |
| Renin | 93 | baseline | CMR | 6 mo | positive | Weir et al |
| GO: 0008083 growth factor activity | ||||||
| Hepatocyte growth factor (HGF) | 40 | baseline, 2, 3 wk | CVG | 3 mo | positive | Soeki et al |
| 246 | Baseline, 1 mo, 3 mo 1 y | Echo | 3 mo, 1 y | positive | Lamblin et al | |
| Growth-differentiation factor 15 (GDF-15) | 97 | baseline | Echo | 1 y | positive | Dominguez-Rodriguez et al |
| Vascular endothelial growth factor (VEGF) | 40 | baseline, 2, 3 wk | CVG | 3 mo | none | Soeki et al |
| GO: 0005125 cytokine activity | ||||||
| Adiponectin | 75 | baseline | Echo | 1 y | negative | Piestrzeniewicz et al |
| sFas | 52 | baseline, 2, 3 wk | CVG | 3 mo | none | Soeki et al |
| sFas ligand | 52 | baseline, 2, 3 wk | CVG | 3 mo | positive | Soeki et al |
| 33 | baseline, 2 wk | CVG | 6 mo | none | Kondo et al | |
| Granulocyte–macrophage colony-stimulating factor (GM-CSF) | 41 | baseline, 1 mo | Echo | 1 mo | positive | Parissis et al |
| Interleukin-6 (IL-6) | 42 | baseline, 1 wk, 2 mo | CMR | 2 mo | none | Ørn et al |
| Tumor necrosis factor-α (TNF-α) | 33 | baseline, 2 wk | CVG | 6 mo | none | Kondo et al |
| GO: 0008009 chemokine activity | ||||||
| C-C motif chemokine ligand 2 (CCL-2) or macrophage chemotactic protein-1 (MCP-1) | 100 | baseline, 3, 6 mo | CMR | 6 mo | negative | Weir et al |
| C-C motif chemokine ligand 3 (CCL-3) or macrophage inflammatory protein-1 alpha (MIP-1α) | 35 | baseline, 1 wk, 1 mo | Echo | 1 mo | positive | Parissis et al |
| 42 | baseline, 2 mo | CMR | 2 mo | none | Orn et al | |
| C-C motif chemokine ligand 4, 5, 19 (CCL-4, 5, 19) | 42 | baseline, 2 mo | CMR | 2 mo | none | Orn et al |
| C-X-C motif chemokine ligand 7 (CXCL-7) or platelet basic protein, C-X-C motif chemokine ligand 8 (CXCL-8) or interleukin-8 | 42 | baseline, 2 mo | CMR | 2 mo | none | Orn et al |
| C-X-C motif chemokine ligand 10 (CXCL-10) or interferon-inducible protein 10 | 33 | baseline, 1 mo | CVG | 1 mo | negative | Koten et al |
| 42 | baseline, 2 mo | CMR | 2 mo | none | Orn et al | |
| C-X-C motif chemokine ligand 16 (CXCL-16) | 42 | baseline, 2 mo | CMR | 2 mo | none | Orn et al |
| Soluble terminal C5b-9 complement complex (TCC) | 42 | baseline, 1 wk, 2 mo | CMR | 2 mo | none | Ørn et al |
| GO: 0006953 acute-phase response | ||||||
| C-reactive protein (CRP) | 31 | baseline | CVG | 6 mo | positive | Takahashi et al |
| 139 | baseline, 1 mo | Echo | 1 mo | positive | Uehara et al | |
| 106 | baseline | Echo | 3 mo | positive | Xiaozhou et al | |
| 42 | baseline, 1 wk, 2 mo | CMR | 2 mo | positive | Ørn et al | |
| 75 | baseline | Echo | 1 y | positive | Piestrzeniewicz et al | |
| 246 | baseline, 1 mo, 3 mo 1 y | Echo | 1 y | positive | Fertin et al | |
| 48 | baseline, 1 mo, 3 mo | CMR | 3 mo | positive | Mather et al | |
| 108 | baseline | Echo | 1 y | none | Dominguez-Rodriguez et al | |
| GO: 0006952 defense system | ||||||
| Myeloperoxidase | 160 | baseline | CVG | 6 mo | positive | Yunoki et al |
| Interleukin-1 receptor-like 1 (sST2) | 100 | baseline, 3, 6 mo | CMR | 6 mo | negative | Weir et al |
| Others | ||||||
| Glucose | 162 | baseline | Echo | 1 y | positive | Bauters et al |
| 52 | baseline | RNA | 6 mo | positive | Nicolau et al | |
| 75 | baseline | Echo | 1 y | positive | Piestrzeniewicz et al | |
| 131 | baseline | CVG | 6 mo | positive | Aoki et al | |
| 93 | baseline | CMR | 11 mo | none | Mather et al | |
| Neopterin | 108 | baseline | Echo | 1 y | positive | Dominguez-Rodriguez et al |
| Heart-type fatty acid binding protein (H-FABP) | 48 | baseline, 1, 3 mo | CMR | 3 mo | none | Mather et al |
| Low-density lipoprotein (LDL) | 109 | baseline, 17 mo | Echo | 17 mo | positive | Buono et al |
| White blood cell count | 107 | baseline | Echo | 1 y | positive | Bauters et al |
| 131 | baseline | CVG | 6 mo | positive | Aoki et al | |
| 246 | baseline | Echo | 1 y | positive | Bauters et al |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
