Multimodality Imaging May Better Stratify MACE Risk in Myocarditis

For patients with known or suspected myocarditis, a dual approach to imaging that uses fluorodeoxyglucose positron emission tomography/computed tomography (FDG PET/CT) may offer prognostic value over cardiac magnetic resonance (CMR) alone, new research suggests.

Individuals in the 100-patient study who had concordant CMR and FDG PET/CT findings went on to have a higher risk of MACE at 1 year compared with those who had discordant imaging results. Researchers say this divergence may present opportunities for better management.

Senior investigator Sanjay Divakaran, MD, MPH (Mass General Brigham and Harvard Medical School, Boston, MA), told TCTMD that their paper addresses a clinically relevant topic. It comes up when “we suspect a condition like cardiac sarcoidosis, but things don’t totally add up [and] there are some abnormalities that make us suspicious for inflammatory cardiomyopathy,” he explained. 

For many clinicians, it’s unclear what should happen next with these ambiguous cases, Divakaran told TCTMD. “Sometimes we empirically treat them as if they have sarcoidosis, sometimes we reimage in 6 months, and sometimes we decide to just watch.”

Lead investigator Chaitanya S. Rojulpote, MD (Mass General Brigham), reported the findings earlier this month at the American Society of Nuclear Cardiology meeting. The study, chosen as Best Clinical Abstract in this year’s Barry L. Zaret Young Investigator Awards Competition, was simultaneously published in the Journal of Nuclear Cardiology.

In suspected myocarditis, Rojulpote said, one option is to perform an endomyocardial biopsy. But “that’s an invasive procedure, and yield can be low,” he commented to TCTMD. Noninvasive imaging, here the combination of CMR and FDG PET/CT, may offer a “better yield, better bang for our buck.”

For Rojulpote, it was a surprise to see how large the increase in risk was when the imaging abnormalities matched up: the concordant group was nearly four times more likely to experience MACE at 1 year than the discordant group.

Divakaran cautioned that the multimodality strategy is not for all-comers but best suited to patients who have mildly abnormal CMR findings and might benefit from the further information offered by FDG PET/CT. The key message from their study, he said, is that “even in the absence of biopsy-proven sarcoidosis, [the presence of] matching imaging findings between these two modalities carries increased risk for these patients.

“And so,” Divakaran continued, “our takeaway is that if you have concordant or matching abnormalities on cardiac MRI and FDG PET/CT, regardless of the diagnosis, we should be thinking about those patients a little bit differently than patients who might have one abnormal and one normal study  or [for whom] the areas of the heart that seem to be abnormal are not matched.”

When Imaging Aligns

Rojulpote, Divakaran, and colleagues identified 100 patients (mean age 59 years; 74% male) at Brigham and Women’s Hospital who underwent FDG PET/CT for known or suspected inflammatory cardiomyopathy. All had a primary diagnosis of myocarditis with focal late gadolinium enhancement (LGE) suggestive of myocarditis on CMR, underwent both FDG PET/CT and CMR within a 12-month period, and had no signs of cardiac or extracardiac sarcoidosis via histology or metabolically active extracardiac sarcoidosis on FDG PET/CT.

The median time between CMR and FDG PET/CT was 31.2 days. Imaging results were considered concordant for patients with at least one matched myocardial segment with focal LGE on CMR and FDG uptake on PET/CT. They were deemed discordant for patients with no matched segments. 

Concordant findings were seen in 45 of the patients, who had a mean age of 61 years and mean LVEF of 50%. The remaining 55 patients, with a mean age of 57 years and mean LVEF of 52%, had discordant findings. Immunosuppression after imaging was significantly more common in the concordant versus discordant groups (31% vs 9.1%; P = 0.005), and patients with concordant findings were more likely to meet Japanese Circulation Society criteria for isolated cardiac sarcoidosis (JCS-ICS; 92% vs 8%; P < 0.001).

Stratified by phenotype, a total of 55 patients had low scar burden/low-intensity inflammation, 27 had high scar burden/low-intensity inflammation, 12 had high scar burden/high-intensity inflammation, and six had low scar burden/high-intensity inflammation.

The researchers defined MACE as the composite of heart failure hospitalization, sustained ventricular tachycardia/ventricular fibrillation (VT/VF), VT ablation, left ventricular assist device placement, cardiac transplant, and death. In all, 15 patients experienced MACE, including three heart failure hospitalizations, seven sustained VT/VF events, three VT ablations, and two deaths.

At 1 year, after accounting for age, sex, LVEF, and number of involved segments by LGE, patients with concordant imaging findings had a higher risk of MACE than did those with discordant findings (adjusted HR 3.61; 95% CI 1.14-11.39; P = 0.028). Moreover, patients who were JCS-ICS positive had higher risk than those who were JCS-ICS negative (adjusted HR 3.29; 95% CI 1.09-9.93; P = 0.035) and those with high scar burden/high-intensity inflammation were more vulnerable to MACE than those with low scar burden/low-intensity inflammation (adjusted HR 4.63; 95% CI 1.27-16.90; P = 0.02).

“Our study highlights the unique incremental value of using a multiparametric imaging approach over either modality alone to optimize risk stratification, refine imaging definitions, and potentially guide therapy in patients with known or suspected myocarditis, independent of the histologic confirmation that is rarely obtained in this population in routine clinical practice,” the authors conclude.

At their center, Divakaran said, their approach to managing patients with concordant findings has already changed. 

“We don’t usually think of myocarditis as having kind of arrhythmic risk, but now when we see this, we’re at least thinking about putting a loop recorder in these patients [to obtain] a little bit more durable monitoring for ventricular arrhythmia,” he said, adding that there also is more shared decision-making. It’s possible to be more open-minded and offer patients options by saying, “Look, we can observe you with a loop recorder, we can treat you empirically, or we can reimage in 6 months,” whereas in the past empiric treatment wasn’t as common.

There are few downsides to the dual imaging approach, said Divakaran, though it’s important to have skilled interpretation so that nonspecific findings aren’t inadvertently considered abnormal.

The investigators are pursuing research on novel tracers for PET/CT that would be more specific for myocardial inflammation. Another potential project, he noted, is to prospectively enroll a broader swath of patients with suspected myocarditis who can undergo multimodality imaging as part of the study design.