If I had to pick one piece of rehabilitation equipment that gets dramatically underused in canine physical medicine, it would not be the underwater treadmill. It would not be the balance disc or the inflatable peanut ball. It would be four to six PVC poles lying on a gymnasium floor. Cavaletti poles. They look simple to the point of being dismissible, and that appearance costs patients real recovery time.
I have incorporated cavaletti canine rehabilitation into orthopedic and neurologic protocols for years, and my clinical experience consistently shows that poles outperform treadmill-only work for a specific and important subset of deficits. In this post I want to explain the neuromechanical rationale, walk through spacing calculations I use for dogs of different stature, and describe how I build progression from the first post-operative week through return to full function.
What Cavaletti Work Actually Does to a Dog's Nervous System
To understand why cavaletti work is valuable, you have to understand what it demands from the patient at a systems level. When a dog walks over a ground pole, the visual system detects the obstacle, the motor cortex begins generating an anticipatory adjustment to limb trajectory, and the sensorimotor loop has to execute that adjustment with precise timing. This is not passive locomotion. This is active gait modification under cognitive load.
The key concept here is feedforward motor control. On a flat surface, a dog with a healed TPLO incision can compensate beautifully. The limb avoidance strategies that developed during the pain and guarding phase are deeply encoded, and a treadmill belt does nothing to disrupt those compensation patterns. The dog trots along, looks comfortable, and the rehabilitation team celebrates. But the motor program running is still the compensatory one.
A cavaletti pole placed at the correct height and spacing forces the dog to recruit the affected limb actively. The stifle must flex, the hock must flex, and the stride must reach a target length. There is no way to ghost through it. That demand is the therapeutic mechanism. The Canine Rehabilitation Institute's published course curriculum on gait analysis identifies exactly this active limb recruitment as the distinguishing feature of pole work compared to surface-only locomotion. I have seen this play out clinically more times than I can count.
Why the Treadmill Falls Short for Certain Deficits
I want to be precise here because I spend a significant portion of my clinical day operating an underwater treadmill. Hydrotherapy is indispensable for early-phase weight-bearing, cardiovascular conditioning, and pain modulation through warm water thermodynamics. I am not arguing against it. I am arguing against relying on it exclusively when the patient's primary remaining deficit is gait quality rather than pain or load tolerance.
The treadmill presents a fixed, predictable surface moving at a fixed speed. The dog's nervous system adapts to that rhythm quickly. After several sessions, the motor program running during underwater treadmill work is essentially an automated one. That is appropriate for cardiovascular conditioning and for patients who are not yet ready for cognitive challenge. It is not appropriate as the sole tool when the goal is restoring symmetrical stride length, conscious proprioception, or precise limb placement.
The specific deficits where I consistently see cavaletti outperform treadmill-only work include:
- Persistent stride length asymmetry after orthopedic repair
- Scuffing or circumduction in dogs with incomplete proprioceptive recovery
- Disuse-pattern stifle or carpal flexion insufficiency
- Trunk sway and weight-shifting avoidance in hip osteoarthritis cases
- Tardy toe clearance following spinal cord injury with partial myelopathy
For all of these, the patient needs to be asked a question their motor system has to consciously answer. Cavaletti poles ask that question every single stride.
Spacing Calculations by Dog Stature
This is where many rehabilitation practitioners default to guesswork, and guesswork produces inconsistent results. The goal of spacing is to match the natural stride length of the target gait and limb, then use height to modulate difficulty. I calculate spacing from a few reference points.
For a walk, target spacing equals approximately 25 to 30 percent of the dog's body length measured from tip of nose to base of tail. For a small dog at 45 cm body length, that puts spacing at 11 to 13 cm between pole centers. That sounds close, but remember we are targeting the walk, a gait where the stride is short and controlled. Getting that spacing wrong by even a few centimeters will cause the dog to place feet on poles rather than between them, which defeats the exercise entirely and risks a stumble.
For a trot, spacing increases to approximately 30 to 35 percent of body length. A medium dog at 65 cm body length will need poles placed at roughly 20 to 23 cm center-to-center for trotting work. A large dog at 85 cm body length moves to approximately 26 to 30 cm. These are starting points. I always watch the patient's first two passes and adjust before considering the setup finalized.
Pole height is a separate variable. I start at ground level for all acute-phase and neurologically compromised patients. Ground poles still demand stride modification without the stumbling risk of elevated poles. When I progress to height, I use quarter-inch increments on PVC risers, targeting a maximum of approximately 50 percent of the dog's carpus height for walk work and 75 percent for trot work in advanced stages. Going higher than that shifts the exercise from gait retraining into jump training, which is a different stimulus entirely.
The number of poles matters too. I use a minimum of four poles to capture at least two complete stride cycles. Six poles is my preferred working set for most patients because it gives the dog enough distance to settle into the required gait modification rather than reacting only at the first obstacle.
Indications Beyond Neurologic Cases
The dominant narrative in rehabilitation literature positions cavaletti work as a neurologic tool. Degenerative myelopathy. Intervertebral disc disease recovery. Wobbler syndrome cases. That framing is accurate but severely incomplete, and it causes orthopedic patients to miss a highly effective intervention.
In my post-TPLO caseload, I introduce ground cavaletti poles as early as week three post-operatively when the supervising DVM has cleared the patient for controlled leash walking. The stifle flexion demand created by even a ground pole is therapeutically relevant at that stage. Quadriceps and biceps femoris recruitment increases measurably with obstacle negotiation compared to flat walking, and that early muscle activation supports the peri-implant tissue remodeling that determines long-term outcomes.
For hip osteoarthritis patients, cavaletti work achieves something I cannot replicate on a treadmill: it demands lateral weight transfer. To step over a pole, the dog shifts its center of mass over each supporting limb in sequence. That challenge directly addresses the hip abductor weakness and contralateral weight-shifting that characterize chronic hip OA. I combine this with a slight uphill incline when the patient can tolerate it, which further recruits hip extensor musculature.
Carpal hyperextension rehabilitation is another underappreciated application. Dogs recovering from carpal sprain or post-surgical carpal arthrodesis of adjacent joints need controlled flexion-extension cycling. Ground poles at tight walk spacing drive exactly that cycling in a functional, load-bearing context that passive range of motion exercises cannot replicate.
Progression Principles I Use Clinically
My progression framework moves along two axes simultaneously: mechanical demand and cognitive demand. These are related but not identical, and conflating them leads to progression that is either too conservative or that exposes patients to stumbling risk before they are ready.
Mechanical demand increases by adjusting pole height, altering the surface beneath the poles (moving from flat rubber flooring to a yoga mat to a slightly compliant foam surface), and changing the approach angle. Walking a dog through poles set at a slight curve rather than a straight line adds a rotational demand to the hip and shoulder girdles that straight-line work never achieves.
Cognitive demand increases by varying pole spacing across a single set, so the dog cannot predict the pattern. I call this my irregular spacing protocol and I use it in the late-middle phase of rehabilitation. Alternate poles are set at 10 percent shorter and longer than the baseline spacing. The dog has to actually look, process, and adjust each stride. That is the closest approximation of real-world ambulation I can create in a controlled clinical environment.
My general phase structure looks like this:
- Phase 1 (Weeks 1-3 post-operative or equivalent early phase): Ground poles at walk spacing, flat surface, straight approach, 2-3 repetitions with full rest between sets
- Phase 2 (Weeks 4-6): Ground poles at trot spacing, beginning curved approaches, mild surface variation beneath poles
- Phase 3 (Weeks 7-10): Low elevated poles at 25 percent carpus height, irregular spacing introduction, uphill approach angle
- Phase 4 (Weeks 10+, per DVM clearance): Trot-height elevated poles, combined with hydrotherapy sets, progressing toward sport-specific pattern work for canine athletes
Progression decisions always involve the supervising veterinarian. I document the dog's gait score using a standardized scale at each session so progression decisions are data-driven rather than impression-based.
Common Setup and Handler Errors That Undermine Results
The most common error I see when rehabilitation teams introduce cavaletti work is handler interference. The owner or handler walks the dog through poles with a tight lead held high, which unloads the forelimbs and alters the natural head carriage the dog needs to visualize the poles. The lead should be loose. The dog should be allowed to look down. If the dog is pulling forward, the spacing is too wide and is inviting trotting when you want walking.
The second major error is incorrect spacing for the individual dog. Teams often use breed-average spacing rather than measuring the individual patient. I have worked with giant-breed dogs who were compact for their breed and small-breed dogs with unusually long bodies. Always measure the individual. Body length from tip of sternum to ischium is my preferred reference point because it correlates more directly with stride mechanics than overall body length including head and tail.
Pole rolling is a hazard that gets underestimated. Unsecured PVC poles on a smooth floor will roll when contacted, and a dog in a post-operative recovery phase cannot safely manage a shifting obstacle. I use rubber endcaps and place poles on rubber matting whenever the patient has any limb instability. Safety infrastructure is not optional.
Integrating Cavaletti Work Into a Complete Rehabilitation Plan
Cavaletti work does not replace other modalities. It occupies a specific niche within a multimodal plan, and placing it correctly within a session matters as much as the poles themselves.
I position cavaletti exercises after therapeutic exercises designed to warm up the target musculature but before fatigue sets in. In a typical post-TPLO session, the sequence might be warm-up on the underwater treadmill, passive range of motion and soft tissue mobilization, cavaletti pole sets, then balance board or proprioceptive disc work, followed by a cool-down treadmill walk. The poles sit at the performance peak of the session where the dog is warm, mentally engaged, and not yet fatigued.
Fatigue is the enemy of this exercise. A tired dog will begin scuffing poles, which is exactly the compensatory pattern you are trying to eliminate. I cap cavaletti sets at four to six passes depending on the phase and stop immediately if I see the dog beginning to clip poles on the affected limb. Clipping means either the dog is fatigued or the spacing is wrong. Both require an immediate response.
For practitioners looking to develop their cavaletti protocols further, the Canine Rehabilitation Institute at caninerehabinstitute.com and the American College of Veterinary Sports Medicine and Rehabilitation at vsmr.org both publish continuing education resources on therapeutic exercise design that I consider foundational reading. The Veterinary Society of Sports Medicine at vsso.org also maintains position statements on exercise prescription standards that inform how I think about progression criteria.
Cavaletti poles are not glamorous. They do not have a screen or a motor or a recirculating pump. But the neuromuscular demand they place on a recovering patient is precise, measurable, and clinically powerful in ways that more technology-forward tools simply are not. They deserve a central place in any serious canine rehabilitation program.