Objective To explore the combination of ultrasound and serological indicators for predicting the cardiotoxicity of breast cancer after chemotherapy utilizing a nomogram. Methods Female breast cancer patients who underwent radical mastectomy and postoperative chemotherapy at our hospital were recruited. All patients were routinely administered with chemotherapy in two weeks after surgery. Patients were divided into toxic group and non-toxic group according to whether left ventricular ejection fraction (LVEF) decreased more than 10% occurred within one year after chemotherapy. Differences in ultrasound

serological examinations and two-dimensional speckle tracking imaging (2D-STI) during the third chemotherapy cycle were compared. Univariate and multivariate COX regression was used to find the independent influencing factors of cardiotoxicity in postoperative chemotherapy patients. Nomogram was established based on predictors to determine the probability of cardiotoxicity. Results A total of 148 patients were included in the study

and 2 patients were lost to follow-up. The incidence of cardiac toxicity was 12.3% in the remaining 146 patients. Multivariate COX regression showed that delta-high sensitive cardiac troponin (Δhs-cTnI) (P<0.05)

delta-N-terminal pro -B-type natriuretic peptide (ΔNT-proBNP) (P<0.05) and delta-Global longitudinal strain (△GLS) (P<0.05) were independent risk indicators for predicting cardiotoxicity in patients undergoing chemotherapy after breast cancer surgery. The nomogram established based on Δhs-cTnI

ΔNT-proBNP and △GLS to indicate the risk of cardiotoxicity in patients had good discrimination (C-index 0.904) and calibration (χ2= 8.170

P = 0.417

Hosmer-Lemeshow test). Conclusion The nomogram established by 2D-STI combined with Δhs-cTnI and ΔNT-proBNP are helpful for assessment of the risk of cardiotoxicity in patients undergoing chemotherapy after breast cancer surgery.