Skip to main content
SCMR 29th Annual Scientific Sessions Main banner
Final Program


Program-At-A-Glance

Rapid Fire Session

Session: Rapid Fire Session 3: Cardiomyopathy (Monitor 3)

Validation of prognostication of estimated pulmonary capillary wedge pressure by CMR in a large cohort of patients with diverse cardiac conditions

Friday, February 6, 2026
10:22 AM - 10:29 AM  
Location: Galapagos / Aruba / Ilhabela (2nd Lower Level)

    Presenting Author(s)

  • J. Jane J. Cao, MD, MPH

    Director of Translational Research, System director of cardiac imaging
    St. Francis Hospital, The Heart Center
    Roslyn, New York, United States

    • Primary Author(s)

  • J. Jane J. Cao, MD, MPH

    Director of Translational Research, System director of cardiac imaging
    St. Francis Hospital, The Heart Center
    Roslyn, New York, United States

    • Co-Author(s)

  • KP

    Karli PIpitone, MPH

    Research Analyst
    St. Francis Hospital & Heart Center
    Huntington, New York, United States

  • JW

    Jonathan Weber, MPH

    Research Analyst
    St. Francis Hospital, The Heart Center
    Greenvale, New York, United States

  • LC

    Luis Cardoza

    Research Volunteer
    St. Francis Hospital, The Heart Center, United States

  • JO

    Jonathan Orozco-Diaz, BSc

    Research Associate
    St. Francis Hospital & Heart Center
    Roslyn, New York, United States

  • CT

    Charles Tarantino

    Clinical Research Associate
    St. Francis Hospital, The Heart Center, United States

  • JS

    Jacob Stein, BA

    Research Associate
    St. Francis Hospital & Heart Center
    Roslyn, New York, United States

  • LK

    Luke Kenny, BSc

    Research Associate
    St. Francis Hospital & Heart Center
    Roslyn, New York, United States

  • NP

    Nicolette Piscopo

    Clinical Research Volunteer
    St. Francis Hospital, The Heart Center, United States

  • AS

    Angel (AJ) Salcedo

    Clinical Research Scholar
    St. Francis Hospital, The Heart Center, United States

  • SQ

    Sophia Qiu, BSc

    Clinical Scholar
    St. Francis Hospital & Heart Center
    Roslyn, New York, United States

Background: Recent development of non-invasive estimation of pulmonary capillary wedge pressure (PCWP) by CMR has shown early promise in single centers. In this study, we aim to validate the prognostication of CMR PCWP in an independent large cohort with diverse cardiac conditions.

Methods: This study pooled consecutive subjects with diverse cardiac conditions who were prospectively recruited for a number of clinical CMR research projects. Demographic and past medical history information were collected prospectively. Cine images were used to assess cardiac chamber volumes, mass, and systolic function using commercial software. Sex-specific CMR-derived PCWP was calculated using the following equation:

CMR PCWP=5.7591+(0.07505*LAV)+(0.05289*LVM)-(1.9927*sex)[female = 0; male = 1].

CMR PCWP >15 mmHg was used as the cut-point to predict a composite outcome of hospitalized heart failure and all-cause mortality. Cox proportional hazards models were used to estimate hazards of the composite outcome, adjusted for age, gender, BSA, hypertension, diabetes, and coronary artery disease (CAD). We performed likelihood ratio testing to investigate the added value of CMR PCWP to outcome prediction from well-established CMR biomarkers known to be associated with adverse outcome.

Results: Among 1108 subjects, the average age was 57±13 years and 759 (69%) were male. Having CMR PCWP >15 mmHg was associated with higher prevalence of hypertension and heart failure (Table 1). This group also had significantly increased cardiac chamber size, left ventricular mass, decreased ejection fractions, reduced stroke volume, and increased prevalence of LGE and regional wall motion abnormalities (Table 1). The Kaplan-Meier curve demonstrated that event-free survival was reduced in subjects with CMR PCWP >15 mmHg (Figure). The adjusted hazard ratio (HR) for every 5 mmHg increase in CMR PCWP was 1.97 (95% confidence limits [CL] 1.68-2.31). Having CMR PCWP >15 mmHg was associated with a HR of 1.37 (95% CL 2.28-6.82). Likelihood ratio testing revealed that adding CMR PCWP to reduced models including risk factors and common CMR biomarkers resulted in a significantly improved model fit for all models (Table 2).

Conclusion: In this prospective cohort with diverse cardiac conditions, CMR PCWP not only independently predicted adverse clinical outcome but also improved model performance of well-established CMR based biomarkers in outcome prediction. Our findings support CMR PCWP generalizability to subjects commonly referred for CMR evaluation.

Table 1. Risk factor and CMR biomarker differences between strata of CMR PCWP

 

PCWP ≤15 mmHg (N=330)

PCWP >15 mmHg (N=778)

 

Variable

Mean ± SD, N (%)

Mean ± SD, N (%)

p-value*

Demographics/Past Medical History

 

 

 

Age (years)

56 ± 14

57 ± 13

0.3102

BMI (kg/m2)

26.9 ± 4.8

30.2 ± 5.5

< .0001

BSA (m2)

1.9 ± 0.2

2.1 ± 0.3

< .0001

Males

184 (56)

575 (74)

< .0001

Hypertension

123 (38)

383 (50)

0.0002

Diabetes

57 (18)

130 (17)

0.8698

Hyperlipidemia

171 (52)

395 (52)

0.8845

Smoking

  

0.2789

   Current

14 (4)

51 (7)

 

   Never

213 (65)

474 (62)

 

   Quit

101 (31)

240 (31)

 

Myocardial infarction

34 (10)

109 (14)

0.0802

Coronary artery disease

54 (17)

147 (19)

0.2968

Stents

49 (15)

123 (16)

0.7383

CABG

13 (4)

51 (7)

0.0818

Stroke/TIA

10 (3)

37 (5)

0.1806

CMR findings

 

 

 

LAV max (ml/m2)

29.7 ± 6.9

48 ± 18.7

< .0001

LAV min (ml/m2)

14.5 ± 4.5

28.1 ± 17.8

< .0001

LVEDV (ml/m2)

71.6 ± 16.4

93.4 ± 28

< .0001

LVESV (ml/m2)

33 ± 11.8

49.6 ± 28.4

< .0001

LVEF (%)

54.5 ± 8.1

49.6 ± 12.9

< .0001

LV mass (g/m2)

48.5 ± 9.9

65.4 ± 18.5

< .0001

LV stroke volume (ml/m2)

71.8 ± 16.9

90.1 ± 23.1

< .0001

RVEDV (ml/m2)

65.6 ± 14.6

77.6 ± 21.2

< .0001

RVESV (ml/m2)

29.7 ± 9.0

37.1 ± 15.4

< .0001

RVEF (%)

55.2 ± 7.1

53.1 ± 9.7

0.0004

RV stroke volume (ml/m2)

67.2 ± 17.0

83.8 ± 25.1

< .0001

Delayed enhancement

111 (34)

409 (53)

< .0001

Regional wall motion abnormalities

64 (19)

276 (36)

< .0001

CMR PCWP

13.6 ± 1.0

18.9 ± 3.7

< .0001

*P-values were calculated using Student's t-test for continuous variables and chi-square for categorical variablesTable 2. Likelihood ratio testing adding CMR PCWP group (≤15 vs. <15 mmHg)

Model* predicting composite outcome

AUC (95% LL, UL)

LR

p-value

Model 1

0.67 (0.63, 0.72)

Reference

Model 1 + CMR PCWP**

0.73 (0.68, 0.77)

32.61

< .0001

Model 1 + LVEF

0.77 (0.73, 0.81)

Reference

Model 1 + LVEF + CMR PCWP

0.79 (0.75, 0.82)

19.07

< .0001

Model 1 + RVEF

0.71 (0.67, 0.76)

Reference

Model 1 + RVEF + CMR PCWP

0.75 (0.71, 0.79)

30.53

< .0001

Model 1 + LVEDVI

0.73 (0.69, 0.77)

Reference

Model 1 + LVEDVI + CMR PCWP

0.75 (0.71, 0.79)

15.35

< .0001

Model 1 + LVESVI

0.76 (0.72, 0.80)

Reference

Model 1 + LVESVI + CMR PCWP

0.78 (0.74, 0.81)

15.57

< .0001

Model 1 + RVEDVI

0.67 (0.63, 0.72)

Reference

Model 1 + RVEDVI + CMR PCWP

0.73 (0.68, 0.77)

36.36

< .0001

Model 1 + RVESVI

0.68 (0.64, 0.73)

Reference

Model 1 + RVESVI + CMR PCWP

0.73 (0.69, 0.77)

28.86

< .0001

Model 1 + LV stroke volume

0.70 (0.66, 0.75)

Reference

Model 1 + LV stroke volume + CMR PCWP

0.76 (0.72, 0.80)

47.38

< .0001

Model 1 + RV stroke volume

0.72 (0.67, 0.76)

Reference

Model 1 + RV stroke volume + CMR PCWP

0.77 (0.73, 0.81)

48.42

< .0001

Model 1 + LGE

0.69 (0.64, 0.73)

Reference

Model 1 + LGE + CMR PCWP group

0.73 (0.69, 0.77)

30.01

< .0001

*Model adjusted for age, gender, BSA, hypertension, diabetes, and CAD

** CMR PCWP ≤15 vs. < 15 mmHgFigure. Kaplan-Meier curve of composite outcome stratified by CMR PCWP group (≤15 vs. <15 mmHg)