Feasibility of subcostal view compared to apical four chambre view for global longitudinal strain analysis in cancer patients

Global longitudinal strain (GLS) analysis is a critical echocardiographic parameter for early detection of subclinical left ventricular (LV) dysfunction, especially in cancer patients undergoing cardiotoxic therapies. However, standard imaging windows are often limited by patient factors such as body habitus, chest wall deformities, radiation therapy or post-surgical reconstruction.

This study evaluates the feasibility and reliability of the subcostal view for Left ventricular Global Longitudinal Strain (LV GLS) assessment in cancer patients as an alternative imaging approach.

This prospective Cohort Design study evaluated the feasibility and reliability of using subcostal view for (LV GLS) measurement as an alternative to the apical four-chamber view in cancer patients undergoing chemotherapy or radiotherapy. Between March and December 2024, 216 echocardiography exams were enrolled at our Research Center. Subcostal measurements were performed manually due to software limitations, while apical four-chamber views used automated endocardial delineation. The primary outcome was the comparison of LV GLS values between subcostal and apical four-chamber views, with secondary outcomes including sensitivity, specificity, and the predictive value of subcostal-derived LV GLS for identifying abnormal LV function. The study adhered to the American Society of Echocardiography guidelines, using -20% as the normal LV GLS cutoff.

The study enrolled 108 cancer patients of total 216 Echocardiography studies with a median age of 45 years (IQR: 39–61), of whom 88.9% (n=96) were female. LV GLS measurements from the subcostal view were feasible in all patients. Agreement between average LV GLS values using subcostal view and average LV GLS values using four-chambre view was excellent, with an ICC of 0.949 (95% CI: 0.926–0.965, p<0.0001). Regarding the detection of abnormal average LV GLS (<-20), the subcostal and four-chamber views demonstrated 90.7% agreement (Cohen’s Kappa = 0.754, p<0.0001). Logistic regression showed that the subcostal average LV GLS was a strong predictor of the four-chamber average GLS, with an odds ratio of 5 (95% CI: 2.73–12.92, p<0.0001) and an AUC of 0.984 (95% CI: 0.966–1.0). The optimal subcostal average LV GLS threshold of -20.85 yielded 100% sensitivity and 92.31% specificity.

The findings of this study highlight the feasibility and reliability of the subcostal view for average LV GLS assessment in cancer patients. With excellent agreement between average subcostal and average four-chamber views and strong predictive accuracy for detecting abnormal average LV GLS values, the subcostal view emerges as a valuable alternative for echocardiographic monitoring, particularly in scenarios where apical imaging is challenging or suboptimal.