Our aim was to evaluate the temporal changes in left ventricular (LV) diastolic filling in relation to other LV parameters using cardiac MRI (CMR) in patients with HER2 positive breast cancer receiving trastuzumab therapy. Fourty-one women with early stage HER2+ breast cancer underwent serial CMR (baseline, 6, 12, and 18 months) after initiation of trastuzumab therapy. A single, blinded observer measured LV parameters on de-identified CMRs in random order. Linear mixed models were used to investigate temporal changes. Compared to baseline, there were significant decreases in systolic function as measured by both left ventricular ejection fraction (LVEF) (p <0.001 at 6 and 12 months) and peak ejection rate corrected for end-diastolic volume (PER/LVEDV) (p = 0.008 at 6 months, p = 0.01 at 12 months). However, these differences were no longer significant at 18 months. In contrast, significant reductions in diastolic function as measured by LV peak filling rate corrected for end-diastolic volume (PFR/LVEDV) were observed at 6 months (p = 0.012), 12 months (p = 0.031), and up to 18 months (p = 0.034). There were no significant temporal changes in the time to peak filling rate corrected for cardiac cycle (TPF/RR). The reduction in PFR/LVEDV at 18 months was no longer significant when corrected for heart rate. In conclusion, there were significant subclinical deleterious effects on both LV systolic and diastolic function among patients receiving trastuzumab. While there was recovery in LV systolic function after therapy cessation at 18 months, reduction in PFR/LVEDV appeared to persist. Thus, diastolic dysfunction may serve as a marker of trastuzumab-induced cardiotoxicity that needs to be confirmed in a larger study.