Determinants of exercise-induced pulmonary hypertension in rheumatic mitral stenosis: a study with exercise stress and speckle tracking echocardiography

Abstract

Abstract Backgrounds Exercise stress echocardiography is helpful in assessing hemodynamic consequence of mitral stenosis (MS) and in guiding treatment. Exercised-induced pulmonary hypertension (PH) is result of severity of MS, but myocardial function of left ventricle and clinical factors can also have effect. Purpose We aimed to evaluate the associated factors with the pulmonary artery systolic pressure (PASP) in exercise stress echocardiography through 2D, Doppler, and speckle tracking imaging in patients with rheumatic MS. Methods A total of 164 patients with rheumatic MS underwent a graded, symptom-limited, supine bicycle exercise with echocardiography. After exclusion of patients who had very severe MS (valve area <1.0 cm2), a history of surgery or recent percutaneous mitral valvotomy, combined significant aortic valve dysfunction, left ventricular (LV) ejection fraction <50%, we analyzed 113 patients (77.6% female; mean age, 56±9 years). Echocardiographic parameters at rest, each stage (25 watt increment every 3 minutes), and peak exercise were obtained. Exercised induced PH was defined as present if PASP >60 mmHg at peak exercise. LV global longitudinal strain (LV-GLS) and left atrial (LA) strain were analyzed by software. The subjects were divided into 4 groups according to mean transmitral pressure gradient (MG) (15 mmHg) and PASP (60 mmHg) at peak exercise (Group 1, MG <15 mmHg and PASP <60 mmHg, n=29; Group 2, MG <15 mmHg and PASP ≥60 mmHg, n=9; Group 3, MG ≥15 mmHg and PASP <60 mmHg, n=23; Group 4, MG ≥15 mmHg and PASP ≥6 0mmHg, n=52). Results The mean mitral valve area was 1.30±0.23 cm2. PASP increased from 30.0±8.0 mmHg at rest to 61.0±14.8 mmHg at peak exercise, along with increase MG. 61 (53.9%) subjects had PASP>60 mmHg at peak exercise. Compared to group 1, group 2 had higher incidence of diabetes mellitus (DM) and significantly elevated PASP at baseline and impaired LV-GLS. However, there was no statical difference in LA strain between the two groups. In subjects with MG above 15 mmHg (Group 3 and 4), a similar trend was observed in occurrence of exercise-induced PH. On logistic multivariate regression, exercised induced PH was independently associated with female (HR: 5.35, 95% CI: 1.51–24.95; p=0.032), DM (HR: 10.05, 95% CI, 1.35–74.45; p=0.024), MG at peak exercise (HR: 1.17, 95% CI, 1.02–1.34; p=0.002), PASP at rest (HR: 1.17, 95% CI, 1.05–1.30; p=0.002), and LV-GLS (HR: 1.45, 95% CI, 1.09–1.91; p=0.009), but not with LA strain. The predictive value for exercised induced PH was highest when adding LV-GLS to the clinical factor (age, sex, DM) and echocardiographic parameters (Figure 1). Conclusions Exercised induced PH is affected by not only hemodynamic consequence of MS, but also myocardial function of left ventricle and clinical factors. Therefore, when determining the optimal timing of intervention based on exercise-induced PH in rheumatic MS patients, LV-GLS should be evaluated comprehensively along with MS characteristics. Funding Acknowledgement Type of funding sources: Private hospital(s). Main funding source(s): Yonsei University College of Medicine