Medtronic Rate-Adaptive AV Delay in LBBAP: Algorithm Architecture & Exercise Optimization
In any dual-chamber LBBAP system, the programmed AV delay is the primary hemodynamic tuning parameter. The Medtronic Azure XT DR MRI SureScan (model W1DR01)—and related Medtronic platforms—includes a rate-adaptive AV delay (RAAV) algorithm that automatically shortens the AV interval as heart rate increases. While the concept is physiologically sound (native PR intervals shorten with sympathetic drive), the algorithm’s linear architecture creates a systematic mismatch with the native curvilinear shortening pattern that has real consequences for athletic patients operating at high heart rates.
This guide covers the algorithm’s internal architecture, its programmable parameters, the critical interactions with other Medtronic algorithms that must be managed in LBBAP, and a four-step exercise optimization protocol.
Algorithm Architecture: Linear Interpolation
The RAAV algorithm is straightforward: it linearly interpolates the AV delay between two anchor points defined by the programmed values.
Worked Example
| Parameter | Programmed Value |
|---|---|
| Sensed AV at rest | 150 ms |
| Minimum AV (floor) | 100 ms |
| Lower rate limit | 60 bpm |
| Maximum sensor rate | 130 bpm |
| Slope | −0.71 ms / bpm |
At 60 bpm the AV delay is 150 ms. At 95 bpm it’s 125 ms. At 130 bpm it hits the floor at 100 ms. The line connecting these points is straight—constant rate of shortening per bpm gained.
The Mismatch: Linear vs. Curvilinear Shortening
The native PR interval does shorten with increasing heart rate, but the relationship is curvilinear—it shortens rapidly in the lower rate range (60→90 bpm) and then plateaus. Published data from Daubert, Wish, and other hemodynamic studies suggests the native sensed AV follows an exponential decay:
The crossover is typically around 90–100 bpm. Below that, the Medtronic linear algorithm is actually under-shortening relative to native (the AV is longer than the native PR would be). Above that, it’s over-shortening—and this is where the hemodynamic penalty begins. The mismatch grows progressively at higher rates, reaching its maximum at the maximum sensor rate where the algorithm hits its floor.
Programmable Parameters
| Parameter | Range | Increment | Nominal | LBBAP Recommendation |
|---|---|---|---|---|
| Sensed AV Delay | 10–350 ms | 10 ms | 120 ms | 120–150 ms (echo-guided) |
| Paced AV Delay | Sensed AV + offset | 10 ms | +30 ms | +30–60 ms (per atrial latency) |
| Rate Adaptive AV | ON / OFF | — | ON | ON (with adjusted slope) |
| Minimum AV Delay | 30–250 ms | 10 ms | 80 ms | 100 ms (athletic patients) |
The paced AV delay is longer than the sensed AV because atrial pacing introduces latency: the time from the pacing spike to actual atrial depolarization and mechanical contraction. If the patient has intact sinus function and is paced atrially only intermittently (e.g., rate drop response or sleep hysteresis), the sensed AV is the primary optimization target. If atrial pacing percentage is high, the PAV offset needs specific attention—too small an offset means paced beats will have functionally short AV timing even if the sensed AV is well-programmed.
Critical Algorithm Interactions for LBBAP
The Azure XT DR includes several algorithms designed to minimize unnecessary ventricular pacing. In conventional pacing, this is appropriate—less ventricular pacing preserves native conduction. In LBBAP, the opposite is true: you want 100% ventricular pacing via the LBB lead. Two algorithms must be explicitly disabled:
Search AV+ periodically extends the AV delay to 300+ ms to look for intrinsic ventricular conduction. In LBBAP, this allows the ventricle to activate via the native AV node/His pathway rather than the paced LBB route. This defeats the purpose of LBBAP, introduces intermittent beats with different activation patterns, and adds hemodynamic variability. If left ON, the patient will have periodic "fusion" or "pseudo-fusion" beats with altered QRS morphology and suboptimal hemodynamics.
MVP switches the device to AAI(R) mode to avoid ventricular pacing entirely, only reverting to DDD(R) when AV block is detected. For LBBAP, this would mean prolonged periods with no ventricular pacing—the patient receives no benefit from the LBB lead during AAI mode. MVP is designed for patients where ventricular pacing is undesirable; in LBBAP, ventricular pacing is the therapeutic intervention.
RAAV should remain ON because AV shortening with rate increase is physiologically appropriate. The key is adjusting the slope and floor to prevent excessive shortening at peak exercise rates. Use a sensed AV of 120–150 ms with a minimum of 100 ms for athletic patients, producing a gentler slope than nominal settings.
Rate response (the accelerometer-driven sensor) should be enabled if the patient has chronotropic incompetence. If sinus node function is normal, DDD without rate response is usually sufficient. For athletic patients with normal chronotropy, unnecessary rate-responsive pacing can create rate competition between the sensor and the sinus node.
Four-Step Exercise Optimization Protocol
Echo-guided iterative AV optimization at resting heart rate. Measure the velocity-time integral (VTI) across the LVOT at sensed AV delays of 80, 100, 120, 140, 160, 180, and 200 ms. The VTI peak identifies the optimal resting AV. For most LBBAP patients this falls at 120–150 ms sensed.
Use pulsed-wave Doppler at the LVOT (5-chamber view) and ensure each measurement captures 3–5 consecutive beats to account for beat-to-beat variability. The difference between optimal and suboptimal VTI is typically 10–20%.
The minimum AV delay (floor) prevents over-shortening at peak rates. A reasonable empiric starting point for athletic patients is 100 ms (vs. the nominal 80 ms). This prevents A-wave truncation at heart rates above 130 bpm while still allowing appropriate physiologic shortening.
The floor should reflect the patient’s atrial mechanical systole duration at peak exercise heart rates. In a patient with normal LA size, atrial mechanical systole takes 80–120 ms even at high rates. If the AV delay equals or is shorter than this duration, the ventricle fires before the atrium finishes contracting.
With resting AV and floor defined, the slope is determined: slope = (AVrest − AVmin) / (HRmax − HRrest). A gentler slope (0.4–0.5 ms/bpm) is safer than the steeper nominal slope. Example: AVrest 140 ms, AVmin 100 ms, HR 60→150 bpm gives a slope of 0.44 ms/bpm.
Compare this slope against the native PR shortening curve. At submaximal rates (80–110 bpm), the linear algorithm should track close to or slightly above the native curve. If the linear slope is steeper than the native curve at these intermediate rates, consider raising the floor or lowering the resting AV to reduce the slope.
The gold standard is treadmill or ergometer exercise with simultaneous echo (exercise VTI measurement at multiple heart rate stages) or at minimum arterial line / pulse oximetry waveform monitoring. Look for the onset of beat-to-beat SV alternation (pulsus alternans) or SV drop as rate increases.
If pulsus alternans appears at a specific heart rate, that rate marks where the algorithm has shortened AV past the patient’s optimal value. Widen the floor or reduce the slope accordingly and repeat. For competitive athletes, this iterative process may require 2–3 programming sessions to find the sweet spot across the full heart rate range.
Echocardiographic AV optimization at rest may not predict the optimal AV delay at exercise heart rates. The patient’s atrial mechanical systole duration, ventricular filling dynamics, and contractile reserve all change with heart rate and sympathetic tone. A resting-only optimization gives you one point on the curve; you need at least two (rest + exercise) to define the appropriate slope. For the athletic patient, the exercise data point is arguably more important than the resting one.
Putting It All Together: A Programming Checklist
| Parameter / Algorithm | Setting for LBBAP | Rationale |
|---|---|---|
| Pacing Mode | DDD or DDDR | Dual-chamber; DDDR only if chronotropic incompetence |
| Sensed AV Delay | 120–150 ms (echo-guided) | Maximize atrial kick at rest |
| Paced AV Offset | +30–60 ms | Compensate for atrial pacing latency |
| Rate Adaptive AV | ON | Physiologic AV shortening needed |
| Minimum AV (Floor) | 100 ms (athletic) / 90 ms (sedentary) | Prevent A-wave truncation at peak rates |
| Search AV+ | OFF | Prevents unwanted intrinsic conduction |
| MVP | OFF | Prevents AAI mode; LBB pacing = therapeutic |
| Upper Tracking Rate | Per patient max exercise HR + margin | Prevent 2:1 block at peak exercise |
| PVARP | Short enough to avoid limiting UTR | Ensure 1:1 tracking at peak sinus rates |
The total atrial refractory period (TARP = AV delay + PVARP) determines the maximum 1:1 tracking rate: Max tracking rate = 60,000 / TARP. If TARP is too long (e.g., AV 150 ms + PVARP 250 ms = 400 ms TARP), the maximum tracking rate is only 150 bpm. Above that, the device enters 2:1 block—suddenly dropping from 150 to 75 bpm ventricular pacing. For an athletic patient approaching 150+ bpm sinus rate, this produces abrupt hemodynamic collapse during peak exercise.
Ensure that the TARP at the minimum AV delay allows 1:1 tracking above the patient’s expected maximum sinus rate. For a competitive rower expecting sinus rates of 170 bpm, TARP must be <353 ms (60,000/170). With a minimum AV of 100 ms, PVARP must be <253 ms.
Frequently Asked Questions
The RAAV algorithm linearly interpolates the AV delay between two anchor points: the programmed AV at the rest rate and the minimum AV at the maximum sensor rate. The slope is constant in ms-per-bpm, unlike native PR shortening which follows a curvilinear pattern that shortens rapidly at lower rates and plateaus at higher rates. This mismatch means the algorithm over-shortens relative to native above approximately 95 bpm.
OFF. Search AV+ periodically extends AV delay to 300+ ms to search for intrinsic conduction, which allows ventricular activation via the native His pathway rather than the paced LBB route. This defeats the purpose of LBBAP and introduces intermittent beats with different activation patterns. MVP must also be OFF for the same reason.
A floor of 100 ms (vs. nominal 80 ms) is a reasonable starting point, producing a gentler slope that reduces A-wave truncation risk at peak rates. The optimal floor depends on the patient’s atrial mechanical systole duration at exercise heart rates, ideally assessed with exercise echocardiography. Iterative programming over 2–3 sessions may be needed for competitive athletes.