Cardiac Resynchronization Therapy Programming and Troubleshooting

Adaptive CRT.

Although CRT was approved 10 years ago, nonresponder rates up to 30% remain a major issue in clinical practice. The most common reason for CRT nonresponse is suboptimal AV delays. Various clinical trials looked at improvement in echo parameters (ejection fraction [EF], LV volumes), clinical parameters (6-minute walk test, NYHA functional status, quality of life, and peak VO ₂ consumption), and heart failure outcomes (mortality, time to transplant, and heart failure hospitalizations as end points to assess CRT response). More recently, clinical composite scores (CCS) are frequently used to assess CRT response.

Several acute hemodynamic studies suggested that synchronized LV pacing resulting in QRS fusion leads to great improvements in left ventricular performance. Clinical trials suggested that LV-only pacing was not inferior to BiV pacing; some nonresponders to BiV pacing did favorably well with LV-only pacing.

An adaptive CRT algorithm was developed to automate AV and interventricular (VV) delays to promote synchronized LV pacing and prevent deleterious right ventricular pacing in appropriate patients. The algorithm is operational in the following three modes ( Fig. 39-1 ):

• Adaptive BiV and LV. This is the preferred mode in patients with intrinsic 1 : 1 conduction and a PR interval less than 200 msec. AV delays and VV delays are determined by the algorithm, and dynamic adjustments are made.

• Adaptive BiV. This is the preferred mode in patients with underlying AV block. AV and VV delays are optimized by the algorithm. Synchronized LV pacing is not applicable due to AV block.

• Nonadaptive CRT. This is the preferred mode in patients with permanent atrial fibrillation (AF). AV and VV interval optimization is not performed.

Synchronized LV pacing is operational when the patient has a regular rhythm with rates less than 100 beats per minute (bpm) with a short AV intrinsic conduction. The details of the algorithm are shown below ( Fig. 39-2 ). The AV interval is extended once every minute to look for the intrinsic AV interval ( Fig. 39-3 ). The following intervals/events are determined:

• AV Conduction Interval. The timing between the atrial electrogram and right ventricular (RV) electrogram.

• End of P Wave. The timing from the atrial electrogram to end of P wave on a far-field channel (Can-SVC coil/Can to A ring electrode).

• RV Timing with Reference to QRS Duration. The timing of the RV electrogram to the end of the QRS duration on a far-field (Can to RV coil) channel. This interval is longer in patients with right ventricular delay and shorter with normal intrinsic right ventricular conduction.

When the conditions for synchronized LV pacing are met, LV pacing is delivered to preempt native conduction by 40 msec ( Fig. 39-4 ). The algorithm converts to BiV pacing when conditions are no longer met.

The AV delay is determined after considering the end of P wave (20 msec after P wave) and to preempt the intrinsic RV conduction by 50 msec. VV interval is determined by a combination of AV interval and the RV-QRS duration interval. The algorithm considers RV sense to end of the QRS as a surrogate of QRS width. Shorter intervals mean better RV conduction and are associated with preemptive LV pacing, as shown in Figure 39-5 .

Use of this algorithm has shown that dynamic ambulatory adjustments of AV and VV delays are safe and the adaptive LV pacing is not inferior to traditional echo-optimized BiV pacing. Substudies analysis suggested that patients who had synchronized LV pacing >50% had better clinical composite scores, fewer HF hospitalizations, and lower AF prevalence and mortality when compared with patients who had synchronized LV pacing <50%.

Multipolar LV pacing leads with additional LV pacing vectors.

Attain Performa is a quadripolar LV pacing lead with steroid elution on all four electrodes. Use of narrow spacing between electrodes two and three can reduce the incidence of phrenic nerve stimulation. The lead comes with an IS4 header, which is coated with a yellow color to differentiate from a DF4 connector pin. Using Vector Express, the threshold can be determined for all the 16 possible vectors. Impedance, along with the pacing threshold, is taken into consideration for selecting a better vector ( Fig. 39-6 ).

Ventricular-sense response.

Triggered LV or BiV pacing is a programmable option. The ventricular sense response (VSR) rate can be programmed to a higher value than the maximum track rate. When dual-chamber pacing mode is enabled, VSR response initiated triggered pacing when PVC was sensed in the AV interval. Safety pacing overrides the VSR response in the safety pacing window, while VSR is active during the remainder of the AV interval. In nontracking modes, triggered ventricular pacing is available as long as it does not violate the maximum VSR rate ( Fig. 39-7 ).

Conducted AF response.

Increase in ventricular pacing rate during an episode of atrial tachyarrhythmia, in response to the patient's intrinsic rate can promote CRT delivery. Beat-to-beat rate adjustments are made according to the algorithm. The response level and rates can be programmed based on ventricular rates during an atrial arrhythmia episode on a histogram. Figure 39-8 shows the operating algorithm of conducted AF response and its effect on ventricular pacing rate.

Atrial tracking recovery.

This algorithm is operational when the atrial rates are below the maximum tracking and mode switch rates. PVCs can lead to atrial events falling in refractory periods and loss of atrial tracing and CRT delivery. When events are sensed in the PVARP period with conduction to the ventricle, PVARP is shortened by 50 msec to promote CRT delivery ( Fig. 39-9 ). The AV interval is extended so the upper tracking rate is not violated.

Ventricular sense episodes.

Prolonged episodes of ventricular sensing with loss of CRT delivery can result in exacerbation of heart failure symptoms. The number of sensed beats to log an episode and the number of paced beats to terminate an episode are programmable parameters. Understanding these episodes can result in programming changes to promote CRT. VSR and VSP (safety pacing) are considered as sensed events even though triggered pacing is often delivered. Figure 39-10 shows a ventricular sensed episode logged by the device showing loss of CRT at rates above the upper tracking rates. Increasing upper tracking rate will effectively improve CRT delivery.

Quick opt.

The algorithm has two particular components AV delay and VV delay optimization as follows:

• AV Delays. Sensed and paced AV delays are determined after eight consecutive sensed and paced intervals. P wave duration measurement on electrograms provides an indirect measure of timing for total atrial activation. The total atrial activation in turn provides an estimate of mitral valve closure. An offset interval of 30 to 60 msec is added to the electrogram duration (30 ms if duration is >100 ms, 60 msec if duration is <100 msec) to account for electro-mechanical delay. This constitutes the optimal sensed AV delay. Optimal paced AV delay is calculated by sensed AV delay with addition of pacing related latency (often 50 msec).

• VV Delays. There are three steps involved in calculation of optimal VV delays. During sensed beats, the peak of the R wave correlates with the onset of isovolumic contraction. The delta interval is calculated by measuring the time difference between intrinsic deflections on LV and RV leads (Δ = R _LV − R _RV ) ( Fig. 39-11A ).

  

Figure 39-11B shows that the delta interval is the temporal difference of intrinsic deflections sensed on RV and LV leads. Paced interventricular delay is assessed during pacing from both chambers and assessing the delay to the alternate chamber. The difference between the two values is termed as ε. (ε = pIVCD _LR − pIVCD _RL ) ( Figs. 39-11C and D ). The optimal VV interval is calculated by averaging the sensed and paced interventricular delays as VV = 0.5 (Δ + ε). If the optimal VV delay greater than 0 msec, LV is paced first, and if the optimal VV delay less than 0 msec, RV is paced first to ensure synchronization of wavefronts in the septum.

Multipolar LV pacing leads with additional LV pacing vectors.

Quartet multipolar pacing lead offers increasing number of options for choosing a suitable LV pacing vector. All four electrodes can act as a cathode and two as an anode, providing up to 10 possible unipolar and bipolar pacing configurations. Steroid elution is available only on the lead tip electrode ( Fig. 39-12 ). VectSelect Quartet provides the interface for testing pacing thresholds and the propensity for phrenic nerve capture across all vectors. Once the appropriate vector is selected, outputs can be programmed manually or by autocapture algorithm ( Fig. 39-13 ).

RV-LV conduction time measurement.

This tool/algorithm helps in prioritizing pacing vectors to understand the site of latest activation near the four electrodes on the quartet lead. During sensed or ventricular beats, interventricular conduction delay is calculated as described in the QuickOpt algorithm ( Fig. 39-14 ). The default vector during multivector capture testing is the vector that involves the longest VV delay.

Trigger pacing.

Trigger pacing to promote CRT during sensed ventricular events can programmed up to 150 bpm. The maximum programmable triggered pacing rate should be 30 msec longer than the ventricular tachycardia (VT) detection interval.

Negative AV/PV hysteresis.

This particular feature shortens AV delay to promote forced ventricular pacing ( Fig. 39-15 ). There is a built-in lockout feature between use of the negative AV/PV hysteresis algorithm and others to minimize ventricular pacing (Ventricular Intrinsic Preference). The shortest AV delay is programmable in the algorithm (−10 msec to −120 msec, delta −10 msec). The algorithm is activated when an R wave is sensed in AV interval. Then the AV interval is shortened by programmed delta value (−10 msec) for 32 cycles. If another R wave is sensed during the AV interval, the device will shorten by another delta value for 256 cycles.