Heart inflammation scans diverge from actual injury in connective tissue disease

Patients with connective tissue disease who show signs of active myocardial inflammation on cardiovascular magnetic resonance imaging carry higher myocardial T2 values than patients with acute suspected viral myocarditis, yet they show far less structural injury on late gadolinium enhancement imaging. That discordance, reported in an open-access study published in Clinical Research in Cardiology, challenges the assumption that imaging markers of inflammatory activity translate uniformly into myocardial damage across different inflammatory heart conditions.

The analysis drew on two prospective institutional registries at a German tertiary center, enrolling consecutively imaged patients between April 2017 and September 2025. Patients with established connective tissue disease came from the MARCIE registry, which recruits people referred for CMR evaluation of suspected cardiac involvement. The comparison group came from an all-comers CMR registry, from which 137 patients fulfilled CMR-based criteria for active suspected viral myocarditis. Of 77 connective tissue disease patients screened, 44 demonstrated active myocardial inflammation under the study definition, yielding a final cohort of 181 patients.

Active myocardial inflammation was defined using a study-specific CMR definition based on the T2 component of the updated Lake Louise Criteria, requiring myocardial T2 values of at least 40 milliseconds on quantitative T2 mapping, applied identically in both groups. Crucially, tissue injury criteria such as elevated native T1, extracellular volume fraction, or late gadolinium enhancement were considered supportive but not required for the case definition, so that structural injury status was not predetermined by the inclusion rule. Connective tissue disease diagnoses, including systemic sclerosis, systemic lupus erythematosus, idiopathic inflammatory myopathies, and undifferentiated disease, were established by experienced rheumatologists according to American College of Rheumatology and European League Against Rheumatism classification criteria.

The two groups differed demographically in expected ways. Connective tissue disease patients were predominantly female, 75 percent versus 44 percent, with comparable mean ages of 56 and 52 years. Cardiac biomarkers were similar between groups, with median NT-proBNP of 347 versus 370 pg/mL and median troponin T of 20 versus 15 pg/mL, neither difference statistically significant. Systemic sclerosis was the most common underlying disease at 48 percent, followed by idiopathic inflammatory myopathies at 27 percent and lupus at 25 percent. Nearly all connective tissue disease patients, 96 percent, were on maintenance anti-inflammatory or immunosuppressive therapy at the time of imaging, predominantly low-dose corticosteroids, but no patient had received high-dose immunosuppressive therapy or steroid pulses before CMR, since treatment escalation was institutionally guided by the scan findings.

The central imaging result was a divergent pattern between inflammatory activity and structural injury. Myocardial T2 was higher in the connective tissue disease group, 42.5 versus 41.0 milliseconds, a difference that reached statistical significance and persisted after multivariable adjustment for age, sex, and symptomatic presentation, with an adjusted coefficient of 1.14 milliseconds. Yet late gadolinium enhancement, the marker of structural myocardial injury, was present in only 41 percent of connective tissue disease patients compared with 88 percent of viral myocarditis patients. In multivariable logistic regression, acute suspected viral myocarditis was independently associated with an adjusted odds ratio of 11.14 for LGE presence. Native T1 and extracellular volume fraction, by contrast, did not differ significantly between the groups, although both patient cohorts showed elevated T2 and native T1 relative to a healthy volunteer reference cohort of 64 people.

The authors applied a two-part statistical model to account for the many connective tissue disease patients without any LGE. This revealed that the group difference was driven primarily by the likelihood of having LGE at all rather than by its extent once present. Among LGE-positive patients with quantifiable scans, mean relative LGE mass was 7.96 percent in connective tissue disease versus 8.96 percent in viral myocarditis, a difference that did not reach significance after adjustment. The authors caution that this exploratory comparison rested on only 14 LGE-positive connective tissue disease patients and should be interpreted with care. A sensitivity analysis additionally adjusting for troponin T left both main associations stable, arguing that differences in acute injury severity do not explain the pattern.

Functional findings reinforced the imaging contrast. Left ventricular ejection fraction was lower in viral myocarditis, 50 percent versus 60 percent, and global longitudinal strain was reduced in the connective tissue disease group relative to healthy volunteers, indicating subtle systolic dysfunction despite preserved ejection fraction. In a subgroup restricted to patients with preserved ejection fraction of at least 50 percent, functional parameters became comparable between groups, but the tissue characterization differences persisted, with higher T2 in connective tissue disease and greater LGE prevalence and extent in viral myocarditis. Exploratory analysis of LGE patterns showed intramural enhancement predominating in connective tissue disease and subepicardial enhancement in viral myocarditis, though the small numbers again warrant caution.

The authors offer several possible explanations while emphasizing what the cross-sectional design cannot resolve. The absolute T2 difference was modest, roughly 1.5 milliseconds, comparable to the smallest detectable change between independent readings of roughly 2.1 milliseconds reported for their protocol, so the finding is meaningful at the population level rather than as an individually discriminatory threshold. Septal T2 measurement may underestimate focal lateral edema typical of viral myocarditis, and prior work suggests diffuse global edema in connective tissue disease versus regionally confined subepicardial involvement in myocarditis, a qualitative difference that could partly explain why focal concentrated inflammation carries a greater propensity for irreversible damage. Background immunosuppression in the connective tissue disease group cannot be excluded as a contributor to the lower LGE prevalence, and whether LGE reflected new or pre-existing injury could not be determined. The diagnosis of viral myocarditis rested on clinical and CMR criteria rather than routine endomyocardial biopsy, and no specific viral pathogen was confirmed.

Clinical implications follow cautiously. The authors stress that the findings should not be read as evidence that myocardial inflammation in connective tissue disease is a harmless bystander. LGE in acute myocarditis does not necessarily represent irreversible fibrosis, since studies have documented complete LGE disappearance at six-month follow-up in a substantial proportion of patients, suggesting that acute-phase enhancement may partly reflect reversible interstitial change. Exploratory one-year outcome data, covering death or heart failure hospitalization in 150 of 181 patients, showed events in 9.1 percent of connective tissue disease and 6.8 percent of viral myocarditis patients, with numerically higher rates among LGE-positive patients, but only 11 events occurred, too few for reliable modeling. The authors conclude that disease-specific contextualization of CMR inflammatory markers is warranted and that prospective longitudinal studies with serial imaging are needed to establish whether T2 elevation without LGE in connective tissue disease progresses to structural injury and how immunosuppressive therapy modifies that trajectory.

Subject of Research: Cardiovascular magnetic resonance comparison of myocardial inflammation and structural injury in connective tissue disease versus acute suspected viral myocarditis

Article Title: Myocardial inflammatory activity and structural injury are discordant in connective tissue disease compared with acute suspected viral myocarditis: a CMR study

Article References: Treiber, J. M., Wilhelmi, T., Binz, C., Backhaus, S. J., Wolter, S., Rieth, A., Rademann, M., Kriechbaum, S., Müller-Ladner, U., Sossalla, S., Klemm, P., & Rolf, A. (2026). Myocardial inflammatory activity and structural injury are discordant in connective tissue disease compared with acute suspected viral myocarditis: a CMR study. Clinical Research in Cardiology. https://doi.org/10.1007/s00392-026-03034-x

Image Credits: AI Generated

DOI: 10.1007/s00392-026-03034-x

Keywords: cardiovascular magnetic resonance, connective tissue disease, myocarditis, T2 mapping, late gadolinium enhancement, myocardial inflammation, Lake Louise Criteria, cardiac imaging, autoimmunity, structural injury, CMR, rheumatology