, Cihangir Kaymaz 2
, Yiğit Cengiz Erkuş 3
, Bahadir Erdem Öcal 4
, Seda Tanyeri Üzel 5
, Hacer Ceren Tokgöz 5
, Çağlar Emre Çağlıyan 6
, Berhan Keskin 7
, Çağdaş Buluş 5
, Doğan Şen 5
, İpek Akkuş 5
, Azmican Kaya 5
, Oğuz Dönmez 5
, Ömer Doğukan Ertan 5
, Muhammet Bulut 5
, Aykun Hakgör 7
, Şeyhmus Külahçıoğlu 5
, Nihal Özdemir 5
2Department of Cardiology, Faculty of Medicine, Başkent University, İstanbul, Türkiye
3Department of Pediatrics, University of Wisconsin-Madison School of Medicine and Public Health, Wisconsin, United States
4Department of Nuclear Medicine, Faculty of Medicine, İstanbul University Cerrahpaşa, İstanbul, Türkiye
5Department of Cardiology, Kartal Koşuyolu Heart Training and Research Hospital, İstanbul, Türkiye
6Department of Cardiology, Faculty of Medicine, Çukurova University, Adana, Türkiye
7Department of Cardiology, Faculty of Medicine, Medipol University, İstanbul, Türkiye
Abstract
Background: In acute pulmonary embolism (PE), abrupt right ventricular (RV) pressure overload and ventricular interdependence are primary determinants of hemodynamic deterioration and short-term mortality. Conventional computed tomographic pulmonary angiography (CTPA) markers, such as the RV/left ventricular (LV) diameter ratio, exhibit variable prognostic performance and may not fully capture early geometric remodeling. It was hypothesized that CTPA-derived apical angle metrics, reflecting biventricular geometric distortion, provide complementary prognostic information beyond established clinical scores.
Methods: In this retrospective study of 535 consecutive patients with acute PE, novel CTPA-derived apical geometry metrics were evaluated: the RV apical angle (α1), the LV apical angle (α2), and their ratio (α1/ α2). The primary outcome was in-hospital mortality; long-term all-cause mortality served as the secondary outcome.
Results: Higher α1 and α1/α₂ values were significantly associated with greater pulmonary arterial obstruction, severe RV dysfunction, and higher clinical risk status. In-hospital mortality increased significantly across α1 tertiles (P = .017). In multiple logistic regression models, after adjustment for the Pulmonary Embolism Severity Index (PESI) score and the
conventional RV/LV diameter ratio, both α1 and α1/α₂ remained independently associated with in-hospital mortality. Adding these apical metrics to the PESI-based baseline model significantly improved prognostic accuracy, demonstrating substantial incremental value for short-term risk stratification. Mortality risk increased linearly beyond thresholds of α1>32.5° and α1/α₂ > 1.17. In multiple Cox proportional hazards analysis, no imaging-derived parameter independently predicted long-term mortality.
Conclusions: Computed tomographic pulmonary angiography–derived apical angle metrics provide physiologically meaningful, continuous measures of acute RV geometric remodeling in acute PE. These metrics offer significant incremental prognostic value for in-hospital mortality beyond the PESI score and conventional CT markers.