TWave Discrimination Feature

The oversensing of T-waves in addition to ventricular events may push the ventricular rate into the VT/VF zone, leading to incorrect detection of VT or VF, and delivery of inappropriate therapy. The TWave Discrimination feature is designed to withhold therapy if there is evidence that a fast ventricular rate is due to T-wave oversensing.

This feature can be found in some Medtronic ICD and CRT-D devices. Please go to or consult with your local Medtronic representative regarding device models available in your geography.

TWave Discrimination is nominally On.

To program this feature, go to Params -> Detection(V.)… -> TWave.

If the programmed VT Redetect is less than 12 and VF Redetect values are less than 12/16, TWave Discrimination may not have enough data to distinguish T-waves from R-waves.

The goal of TWave Discrimination is to reduce inappropriate therapy without compromising VT/VF detection sensitivity. It does so by differentiating R-waves from T-waves by analyzing differences in amplitude, frequency content, and pattern to distinguish R and T patterns from VT/VF. While this feature reduces the chance of inappropriate shocks due to T-wave oversensing, it does not eliminate the T-wave oversensing.

The TWave Discrimination algorithm is a fully automatic approach to identifying T-wave oversensing without compromising VT/VF sensitivity.1 TWave Discrimination is applied on initial detection and on redetection. TWave Discrimination operates on the assumption that R-waves and T-waves have different waveform characteristics. R-waves are generally higher frequency waveforms than T-waves. The algorithm utilizes a differential frequency filter that allows the device to make the size of the R-wave signal larger when compared to the size of the T-wave signal, enabling R-T pattern recognition.

For a set of 6 ventricular events, the device measures the average of the 3 largest sensed amplitudes and creates an R-wave threshold. Ventricular events below the threshold are tentatively labeled as T-waves. Identified R and T waves must then meet several rate and pattern criteria to be confirmed as a consistent T-wave oversensing pattern. If this pattern is confirmed and continues consistently, the device will withhold detection and label the episode as a “TWave” episode and label each ventricular sensed event as a “TW” on the marker channel. The rhythm will continually be monitored for any changes.

The algorithm does not eliminate T-wave oversensing. The stored episodes should be reviewed and a RV sensing test should be performed to determine the appropriate clinical steps. Only ventricular sensed events that occur above the VT or VF programmed detection zones (marked TS or FS) are analyzed.

In the example below, the patient had T-wave oversensing in the VT zone. A railroad pattern from the T-wave oversensing can be seen in the interval plot. Criteria for detection was met and therapy was withheld using the TWave Discrimination feature. The EGM shows a TW annotation on the decision channel, indicating the TWave Discrimination feature was in effect.

Although the TWave Discrimination algorithm effectively worked in avoiding inappropriate therapy, it does not eliminate the T-wave oversensing. Therefore, other programming changes and/or revising the lead could be necessary. In this example, the sensing vector was changed and RV sensitivity was reprogrammed.

Interval plot


Using the Protecta Performance Model and the RV Lead Noise Discrimination algorithm, along with ATP, TWave Discrimination, and SVT discriminators applied in the VF zone, significantly reduced shock frequency when tested on stored episodes from the SCD-HeFT Trial. This model, which combined existing and new shock reduction features, predicted 98% of patients would be free of inappropriate shocks 1 year after implant, and 92% at 5 years post-implant.1 The TWave Discrimination feature reduced inappropriate detection of VT/VF in spontaneous T-wave oversensing episodes by 97.5% while maintaining 100% sensitivity for detecting true VT/VF.2


  1. Volosin K, Exner DV, Wathen MS et al, Virtual ICD: A Model to Evaluate Shock Reduction Strategies, HRS Conference, May 2010.
  2. Cao J, Shrivastiv M, Koehler JL, Swerdlow C, Gillberg J. Heart Rhythm. May 2008; 5(5):S194.

Sources: Medtronic Protecta XT DR Clinician Manual; Medtronic Protecta XT CRT-D Clinician Manual.

Last updated: 
03 Apr 2013