Electrocardiogram (ECG) is a biomedical signal that provides vital information about the condition of heart. However, different artifacts such as powerline interference, electrode contact noise, motion artifacts, muscle contraction and base line drift usually distort the ECG signal leading to inaccurate diagnoses of patient ECG. The most significant interference that corrupts the ECG is powerline Interference. Hence, this paper present the design of finite impulse response (FIR) digital notch filter with optimized trapezoid window for the removal of power line noise from ECG signals. The three ECG signals records (100MLII, 124MLII and 200MLII) used for the simulation were taken from MIT-BIH Arrhythmia database. These ECG records were then corrupted with 50Hz powerline noise generated in MATLAB and filtered using the proposed filter. A comparative analysis of the FIR notch filter designed with the optimized trapezoid window with the FIR notch filters developed with the other window types shows that the proposed filter displays an improved signal to noise ratio (SNR) of 25.2070 dB, 31.9562 dB, 28.6152 dB for the three ECG signals records (100MLII, 124MLII and 200MLII) respectively.