Cardiac chamber sphericity on magnetic resonance imaging (MRI) in chronic obstructive pulmonary disease (COPD): SPIROMICS-HF

Presenting Author(s)

• CL

  Chia-Ying Liu, PhD

  Scientist
  Canon Medical Systems Corporation
  B, Maryland, United States

Primary Author(s)

• JL

  Joao A. C Lima, MD

  Professor
  The Johns Hopkins University
  Baltimore, Maryland, United States

Co-Author(s)

• CN

  Chikara Noda, PhD

  Research associate
  The Johns Hopkins University, United States
• 

  Bharath Ambale-Venkatesh

  Physicist
  The John Hopkins Hospital
  Baltimore, Maryland, United States
• 

  Yoko Kato, MD, PhD

  Research Associate
  The Johns Hopkins University
  Baltimore, Maryland, United States
• 

  Jing Liu, PhD

  Professor
  University of California, San Francisco
  San Francisco, California, United States
• 

  Yoo Jin Lee, MD

  Assistant Professor
  University of California, San Francisco
  San Francisco, California, United States
• 

  Prachi Agarwal, MD

  Professor
  University of Michigan
  Ann Arbor, Michigan, United States
• MP

  Martin R. Prince, MD, PhD

  Professor of Radiology
  Cornell University, New York, United States
• GB

  Graham Barr, MD, PhD

  professor
  Columbia University, New York, United States
• 

  David A. Bluemke, MD

  MD
  University of Wisconsin-Madison
  Madison, Wisconsin, United States

Background: Subpopulations and Intermediate Outcome Measures in chronic obstructive pulmonary disease, or COPD (SPIROMICS) is a multi-center observational study designed to evaluate COPD from multiple angles. We sought to investigate the relationships of the lung function and cardiovascular conditions and biomarkers including atrial fibrillation (Afib), coronary artery calcification (CAC) score, HbA1c and NT-proBNP to heart sphericity in COPD.

Methods: The SPIROMICS Heart Failure study recruited cases of COPD and controls with different severities of COPD from nine clinical centers. All participants underwent CMR on 1.5T or 3T scanners. Heart sphericity was measured in all chambers using the semi-automatic software (Olea Sphere) by manually contouring the cardiac chamber at end-diastolic phase on the 4ch CINE image. Sphericity was calculated as the ratio of the largest to the second-largest principal component axes of the area, representing how closely the shape approximates a circle. COPD was defined as a post-bronchodilator ratio of the forced expiratory volume in 1 second (FEV1) to the forced vital capacity (FVC) < 0.70. COPD severity was classified as: mild, FEV1 ≥ 80% predicted; moderate, 50–80% predicted; and severe, < 50% predicted. Multivariable logistic regression and linear regression analysis were used to examine the association of risk factors with heart sphericity (P < 0.05 for statistical significance).

Results: The sample consisted of 528 participants with CMR analysis (52% men). Pulmonary function data were available in 465 (88%) participants (54% participants with COPD, Table 1). Heart function was worse in relationship to COPD severity while sphericity was similar between groups. Left atrial (LA) maximum volume, LA global longitudinal strain, LA and right ventricular (RV) sphericity were associated with Afib (Table 2). Enlarged LA volume and severe COPD demonstrated the highest probability to have Afib in the postestimation. Aging and higher LA sphericity was also associated with Afib. For the lung function and cardiovascular risk factors in relationship to heart sphericity, only NT-proBNP, 6-minute walk test, and HbA1c showed some levels of association (Table 3). Particularly, NT-proBNP was positively associated with left ventricular (LV) and RV sphericity in both univariable and multivariable analysis. CAC score, FEV1 predicted, FVC predicted, and FEV1/FVC did not correlate with heart sphericity.

Conclusion: LA remodeling was invariably present in patients with Afib and the probability with Afib is further advanced with COPD severity and aging as indicated by higher sphericity. We further found that right atrial (RA) sphericity was independently associated with 6-minute walk test. The positive association between NT-proBNP and ventricular sphericity could enhance the assessment of ventricular remodeling in the progression of heart failure.

Funding: National Institutes of Health (NIH)/Heart, Lung Blood Institute (NHLBI) R01-HL093081, U01-HL137880

Table 1. Characteristics of study participants. Values display as mean ± SD or median (IQR).
Table 2. Association between atrial fibrillation (Afib) and heart sphericity. Values display as mean ± SD. P values were from multivariable logistic regression adjusted for age, sex, BMI, heart rate, hypertension, diabetes, and smoking pack year.
Table 3. Correlation of NT-proBNP, 6-min walk test, and HbA1c to heart sphericity. Multivariable linear regression was adjusted for age, sex, BMI, heart rate, hypertension, diabetes, and smoking pack year. Values are beta coefficients (P-value). Scatter plot shows a positive correlation between NT-proBNP and RV sphericity.