Author + information
- Ify Mordi, MBChB,
- Aleksandra Radjenovic, MSc, PhD,
- Tony Stanton, MBChB, PhD,
- Roy S. Gardner, MD,
- Allan McPhaden, FRCPath,
- David Carrick, MBChB,
- Colin Berry, MD, PhD and
- Nikolaos Tzemos, MD∗ ()
- ↵∗Institute of Cardiovascular and Medical Sciences, British Heart Foundation Glasgow Cardiovascular Research Centre, University of Glasgow, Glasgow, United Kingdom G12 8TA
It has been noted recently, initially in pathological specimens, that infarcted myocardium can be infiltrated by adipose cells, known as lipomatous metaplasia (LM). It has been suggested that the presence of myocardial fat may interfere with the conduction system of the heart, perhaps providing a substrate for sudden cardiac death (1).
Recent studies have aimed to noninvasively characterize the incidence of LM using both computed tomography and cardiac magnetic resonance (CMR), although the studies have been limited by small size or retrospective nature (2,3). We aimed to prospectively evaluate the prognostic significance of LM in an unselected cohort of patients.
We prospectively evaluated 316 consecutive patients who had a remote history of type 1 myocardial infarction (MI) (>6 months) referred by their physician for CMR at our institution. All patients were evaluated using a 1.5-T scanner. Cine images were obtained for assessment of mass and function using a steady-state free precession sequence. Following this, T2-weighted imaging was conducted in 3 matched short-axis slices using a short TI inversion recovery (STIR) sequence. Then an identical sequence was performed without a third inversion pulse (T2-STIR−) to remove the inherent fat suppression. Finally, late gadolinium enhancement imaging was performed. An example is shown in Figures 1A to 1E. All patients were followed for a combined primary outcome of all-cause mortality, sustained ventricular arrhythmia, and heart failure hospitalization.
The average age of the study population was 58.6 ± 10.9 years. Most patients were male (75.3%). Median time from MI to CMR was 4.0 years. Using our diagnostic criteria, 77 patients (24.4%) had evidence of LM. LM+ patients had older infarcts (median infarct age 8.00 years; interquartile range [IQR]: 2.00 to 14.00 years vs. 3.00 years; IQR: 1.00 to 8.50; p < 0.001) that were more extensive (infarct size 31.56 ± 12.08% vs. 20.45 ± 14.91%; p < 0.001). There was no significant difference in the use of medications between the 2 groups. LM was not associated with typical clinical risk factors or prior percutaneous coronary intervention or coronary artery bypass graft.
LM+ patients had lower left ventricular ejection fraction (LVEF) (33.1 ± 13.1% vs. 42.6 ± 15.0%; p < 0.001) and higher LV end-diastolic volume (136.0 ± 50.2 ml/m2 vs. 108.1 ± 38.7 ml/m2; p < 0.001).
The primary outcome occurred in 46 patients, 23 LM+ (29.9%) and 23 LM− (9.6%). In a multivariable model containing the strongest other predictors (atrial fibrillation, LVEF, diabetes, and multivessel disease), LM remained a significant predictor of the primary outcome (hazard ratio: 2.67; 95% confidence interval: 1.44 to 4.96; p = 0.002) (Figure 1F).
We found that the presence of LM was associated with more adverse remodeling and larger infarcts. Our findings are in keeping with the original landmark study of Baroldi et al. (1).
Infarct age was the only clinical feature found to be associated with LM, suggesting that its development is progressive (1,3). Similar to previously published studies, we did not find any other clinical associations with LM (2,3). The pathogenesis of LM within myocardial infarcts is not clearly understood as of yet; however, it is known that areas of MI scar display abnormal metabolism with reduced uptake of fatty acids, leading to the replacement of collagen with adipocytes (1). It is postulated that newer medical and reperfusion therapies might reduce the development of LM. The prevalence of LM in our study was similar to that in earlier histopathologic studies despite these therapeutic advances.
The presence of LM was a stronger predictor than infarct size alone and remained significant when added to LV volumes or LVEF. A recent study in an ovine model by Pouliopoulos et al. (4) found that the presence of LM in sheep with experimentally induced myocardial infarcts was associated with more inducible ventricular tachycardia. The presence of myocardial fat has also been associated with inducible ventricular tachycardia in humans in the right ventricle. It may be that adipose tissue predisposes to re-entrant tachycardias and impairs myocardial conduction, leading to increased risk of ventricular arrhythmias and death (4).
Please note: Dr. Berry has received research funding from Siemens Healthcare. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- 2015 American College of Cardiology Foundation
- Ichikawa Y.,
- Kitagawa K.,
- Chino S.,
- et al.
- Pouliopoulos J.,
- Chik W.W.,
- Kanthan A.,
- et al.