P6469Predictive value of currently available echocardiographic work-up algorithm for previous and updated pulmonary hypertension definitions

• Published in: European heart journal Vol. 40; no. Supplement_1
• Main Authors: Tanyeri, S, Akbal, O Y, Keskin, B, Hakgor, A, Karagoz, A, Tokgoz, H C, Kulahcioglu, S, Erdogan, E, Tanboga, I H, Dogan, C, Acar, R D, Ozdemir, N, Kaymaz, C
• Format: Journal Article
• Language: English
• Published: Oxford University Press 01.10.2019
• Subjects: Pulmonary Hypertension
• ISSN: 0195-668X; 1522-9645
• DOI: 10.1093/eurheartj/ehz746.1061

Abstract Background and aim In this study we evaluated the predictive value of currently available European Society of Cardiology 2015 Pulmonary Hypertension (PH) Guidelines two-step echocardiographic (Echo) work-up algorithm for two PH definitions as invasively evaluated pulmonary arterial mean pressures (PAMP) ≥25 mmHg and >20 mmHg. Methods Study group comprised the retrospectively evaluated 1299 patients (pts) (53.1±18.8 years, female 807, 62.1%) who underwent right heart catheterisation (RHC) with different indications between 2006 and 2018. Echo data obtained from same day and pre-RHC evaluation were available in all pts. Chemla method was used for Doppler estimation of PAMP from tricuspid regurgitant jet. For two PH definitions as PAMP ≥25 mmHg (definition-A) and PAMP >20 mmHg (definition-B) predictive value of the ESC Echo algorithm was tested with two logistic regression models based on only PAMP-Echo (Model-1) and PAMP-Echo and Echo findings supportive for PH (Model-2). Results RHC revealed that criteria for PH definitions A and B were met in 891 (68.6%) and 1051 (80.9%) of overall pts. Pre-RHC Echo findings supportive for PH were noted in 529 (40.7%) of pts. Because pulmonary regurgitant velocity data were imprecise, other 6 supportive Echo findings were utilised. The % of the supportive Echo measures were as follows; 1 (11%), 2 (10%),3 (8.1%), 4 (7.2%), 5 (2.5%) and 6 (1.8%). In Model-1, for definitions A and B, odds ratio (OR) of PAMP-Echo were (1.07; 0.99–1.19, p=0.058, LRX2:139, c: 0.72) and (1.05; 0.98–1.12, p=0.102, LRX2:129, c: 0.71), respectively. In Model-2, for definition-A, OR of PAMP-Echo and number of supportive findings were (1.05; 0.98–1.12, p=0.162) and (1.58; 1.42–1.75, p<0.001) (LRX2:217, c: 0.77), respectively. Using the same model for definition-B, OR of PAMP-Echo and number of supportive findings were (1.04; 0.97–1.11, p=0.264) and (1.56; 1.40–1.73, p<0.001),(LRX2:203, c: 0.76), respectively. Receiver operating curve (ROC) analysis yielded that PAMP-Echo >39.8 mmHg had a sensitivity of 80% and a specificity of 54% for definition-A, and a sensitivity of 81% and a specificity of 52% for definition-B, respectively. Regardless of the definition tresholds for PH on RHC, number of supportive Echo findings were responsible for nearly 70% of the overall variance in the PH diagnosis. Conclusions Irrespective of the RHC definition criteria as PAMP ≥25 mmHg or >20 mmHg, number of supportive Echo findings but not PAMP-Echo only approach seems to be reliable in the diagnostic work-up for PH.