Hypertrophic Cardiomyopathy

Refining the Lens of Cardiac Magnetic Resonance to Evaluate Late Gadolinium Enhancement

• cardiac magnetic resonance
• hypertrophic cardiomyopathy
• late gadolinium enhancement
• myocardial fibrosis

Hypertrophic cardiomyopathy (HCM) is the most common inheritable cardiomyopathy, affecting about 1 in 500 individuals. Echocardiography is most commonly used for diagnosis and the diagnostic criteria are quite straightforward: abnormally increased thickness of the left ventricular (LV) wall in the absence of other causes of hypertrophy. Indeed, there are few other myocardial diseases for which such a “simple” anatomic measurement seems sufficient for diagnosis. However, sophisticated new imaging methods have the potential to move well beyond anatomy alone for the assessment of HCM.

Cardiac magnetic resonance (CMR) can be used to assess LV wall thickening, but the cost is 3 to 4 times more than echocardiography. CMR physicians have long observed HCM patients for which echocardiography did not detect abnormal wall thickening, although the converse (echocardiography only detected cases) is much less likely. In a large series of more than 300 patients, Maron et al. (1) showed that 12% of patients had abnormal wall thickening by CMR that was underestimated or undetected by echocardiography. Do those “missed” cases by echocardiography matter in terms of patient outcome, especially in regard to the greater cost of CMR? For example, apical HCM is more likely to be missed by echocardiography than magnetic resonance imaging, but those patients have less frequent adverse events than do those with outflow tract obstruction. What additional information can CMR provide?

The substrate for arrhythmia in HCM is thought to be myocardial scar (2,3). In 2002, Choudhury et al. (4) demonstrated the use of CMR to detect enhancing areas of myocardial tissue in HCM. These late gadolinium enhancement (LGE) areas on CMR reminded the investigators of myocardial scar on postmortem specimens. Yet pathology correlation with CMR to determine the etiology of the enhancing areas is rare, requiring antemortem or pre-explant CMR evaluation. Moon et al. (5) observed a single 28-year-old male who underwent CMR and cardiac transplant 49 days later. O'Hanlon et al. (6) also described a patient with CMR prior to death and autopsy. These rare but important observations have helped to validate the relationship between LGE tissue on CMR and myocardial fibrosis in hypertrophic cardiomyopathy. At present, LGE/“fibrosis” has been reported in 60% to 70% of all HCM subjects using CMR imaging (7).

Since those early studies, LGE detected by CMR has been rapidly investigated as a potential biomarker for malignant arrhythmia in HCM. The presence of LGE is more frequent in HCM patients with tachyarrhythmia (8–10). Importantly, several studies have indicated that LGE is an independent risk factor for adverse outcome in HCM (6,11). Once again in HCM, could we have another “simple” biomarker in HCM: does simply the presence of LGE (rather than the extent or pattern) place the patient at high risk? If LGE is present, does that risk information help the clinical cardiologist manage the HCM patient? Although 60% to 70% of patients with a CMR will have LGE/fibrosis, the incidence of sudden cardiac death in HCM is only about 0.5% per year. Perhaps further detailed evaluation of LGE/fibrosis by CMR would be informative.

In this issue of the Journal, Todiere et al. (12) present important insights regarding change of LGE over time in HCM. The authors measured focal myocardial scarring in 55 HCM patients using LGE CMR. Study subjects underwent two CMR scans at an average interval of about 2 years (710 ± 410 days). The authors considered >1 g of change in LGE mass to be statistically significant. Greater LGE/fibrosis progression (expressed as grams per month) was associated with a greater amount of LGE at baseline, apical pattern of HCM, and worsening New York Heart Association (NYHA) functional class. No correlations were noted between LGE progression and LV mass index, LV volume, or ejection fraction. This study demonstrates that LGE is a late but common disease phenotype that follows the development of hypertrophy. Interestingly, Todiere et al. (12) also observed two patients with regression of LGE/fibrosis.

What do we know about LGE progression? In ischemic cardiomyopathy, infarct remodeling during the first 3 months results in slight decrease in mass of the LGE tissue. This is accompanied by tissue change of the LGE region, from myocyte necrosis to replacement fibrosis. In contrast, we know very little about change in LGE in nonischemic cardiomyopathy. The study by Todiere et al. (12) thus represents 1 of the first studies with long-term follow-up of the LGE findings. In another study, there was no significant LGE change by CMR after a mean of 1.5 years in subjects with myocarditis (13). Few other studies have addressed change in LGE over time.

The study by Todiere et al. (12) also raises important questions regarding quantitative imaging by CMR. The reproducibility of quantitative CMR for most standard measures is between 5% and 10% (14). Todiere et al. (12) defined a threshold of LGE signal, above which LGE was said to be present. Caution is warranted here, in that a threshold method is likely of value only under the exact CMR circumstances used by the authors. Common CMR variations, such as gadolinium dose, magnetic field strength, CMR scanner manufacturer, and pulse sequence parameters to name but a few, have significant effects on both the numerator (signal) and denominator (noise) of the threshold. Our experience in an HCM multicenter trial suggests that the optimal visual threshold for LGE correlates poorly with fixed, computer-defined thresholds when such thresholds are applied to multiple CMR centers and manufacturers.

The incidence and rate progression of LGE observed by Todiere et al. (12) was also almost certainly influenced by the characteristics of the study population. There was improvement in NYHA functional class status in 3 of 55 patients but worsened NYHA functional class in 13 patients following medical therapy. Little information is provided regarding concurrent medical issues as well as medications initiated between the 2 scan times. Because the study by Todiere et al. (12) was observational rather than interventional, the medications were likely heterogeneous. Angiotensin II receptor blockers and angiotensin-converting enzyme inhibitors as well as other therapies have been associated with regression of hypertrophy and myocardial fibrosis in animal models as well as humans (15–18). Thus, the relationship of medical therapy to LGE progression will need further evaluation in subsequent studies.

In conclusion, the evidence that CMR can provide clinically useful and distinct information apart from echocardiography for HCM patients continues to increase. Starting with the “simple” parameter of LV wall thickness, the presence and now the progression of LGE/fibrosis may be important parameters that can be tracked by CMR. Further efforts on CMR reproducibility need to be performed, and standardization may eventually allow CMR physicians to determine useful thresholds for LGE presence or absence. Multiple studies already point to LGE CMR as an important biomarker to indicate adverse events in HCM. Further studies are needed to determine if LGE is a modifiable marker, and if those modifications translate into better patient outcome.

• American College of Cardiology Foundation