Physiotherapy after a stroke in the cerebellum – Knowledge

1. What you will find in this text

This text is about practice after a stroke in the cerebellum. It describes what is examined in physiotherapy, which exercises come in which order, how much it takes, what you do at home, and how you can tell that things are moving.

It applies to both forms: to cerebellar infarction, where a vessel closes, and to cerebellar haemorrhage, where a vessel bursts. For the practice the difference does not matter. What matters is the site within the cerebellum and what is missing in everyday life.

What is not here: how a stroke in the cerebellum arises, how it is recognised, what happens in hospital during the first hours, and how a second one is prevented. That is in the sister guide Stroke in the cerebellum: infarction and haemorrhage. Anyone wanting both reads that one first and this one second.

The text addresses people affected and their relatives. It does not replace an examination. Which exercises suit you follows from what the therapist finds in you.

2. What is disturbed after a stroke in the cerebellum

This section stays short. It names only what has a consequence for the practice.

2.1 What is missing is the coordination

The cerebellum does not produce movement. It tunes it. Where the damage stays confined to the cerebellum, weakness is usually not the leading sign. Muscle strength is in principle available, and the movement becomes inaccurate all the same: it falls too far or too short, starts at the wrong moment, or is made up of parts that fit together poorly. The collective term for this is ataxia, from the Greek for "without order" [1].

Two qualifications belong with this. Anyone who has been off their feet for weeks loses strength and endurance. That is a consequence of reduced activity rather than of the cerebellum, but in everyday life it makes itself felt just as much. And where the damage reaches beyond the cerebellum, into the brainstem for instance, a genuine weakness is added. Section 13 describes what follows from that for strength training.

From this follows the first decision for treatment. Accuracy and balance stand in the foreground. Strength training is added, but it does not stand at the beginning and does not solve the core problem.

In everyday life the ataxia shows in four forms, and each has its own programme:

• Truncal ataxia. The trunk tips to one side in unsupported sitting, although arms and legs are strong. This form needs the most help during the first days.
• Ataxia of standing and walking. Wide track, irregular step length, swaying to the side, slow pace [6].
• Ataxia of arm and leg. The hand goes past the glass and corrects in small zigzag movements. The closer the target, the stronger the tremor.
• Disorder of gaze and speech. The eyes jerk, the image wobbles while walking, speech becomes slow and unevenly loud.

2.2 Why the same body side

Unlike a stroke in the cerebrum, the disorders of coordination lie on the same side as the damage. Damage in the left half of the cerebellum makes the movements of the left arm and the left leg inaccurate [1].

That has a practical use: if your discharge report says the damage is on the right and you also notice an unsteady right hand, you have a set of findings that fit together. For the practice it means that the affected side is addressed on purpose and the other one serves as a comparison.

2.3 When the brainstem is involved

The brainstem lies immediately in front of the cerebellum. If the damage reaches there, signs are added that follow other rules and that change the treatment more than anything else in this text:

• Swallowing disorder. Saliva, drinks or food can enter the airways without any coughing. Where the brainstem is involved, a swallowing disorder is more frequent and more marked than where the damage stays confined to the cerebellum. Swallowing is checked after every stroke though, brainstem involved or not – section 6 describes this. Treatment is carried out by speech and language therapy; physiotherapy contributes with upright posture, strength for coughing, and mobility of neck and chest [9].
• Pull to one side. The body is pulled towards the side of the damage. Section 8 describes what follows from that.
• Crossed sensory loss. Pain and temperature are perceived less well on one side of the face and on the other side of the body [7]. Heat applications are dosed there by the thermometer and not by what the person feels.
• Weakness. The pathways for muscle strength run through the brainstem. If they are involved, a genuine weakness is added to the inaccuracy, and the programme gains a strength component it would not otherwise have.

This combination carries the name Wallenberg syndrome. The sister guide describes it in detail under section 4.2.

3. Why cerebellar damage is practised differently

This section is the core of the text. Anyone who has understood it understands every single exercise that follows below.

3.1 The cerebellum matches movement against feedback

When you reach for a cup, your brain predicts where your hand will be in 200 milliseconds. It compares that prediction with what comes back from eyes, skin and joints. Where the two differ, a correction signal arises. The cerebellum performs this matching [2].

If the matching fails, two things happen. The ongoing movement becomes inaccurate. And adaptation to new conditions becomes slower, because it rests on the same correction signal.

3.2 Which learning is impaired and which is not

Healthy people adapt quickly to a changed environment. Anyone wearing glasses that shift the image sideways reaches correctly again after a few attempts. This form of learning is called error-based learning: the body works out how far the movement went wrong and shifts the next one by that amount. It depends on the cerebellum and is impaired after cerebellar damage [2].

Learning does not stop there. There are several routes to improving a movement, and they are affected to different degrees:

• Learning from success and failure. Here only whether a movement worked counts, not by how much it went wrong. In one study, 12 people with a cerebellar disease and 12 comparison participants of the same age practised an aimed movement in which they learned only "hit" or "missed" after each attempt. The people with cerebellar disease did learn, and they retained fully what they had learned. When the same people were shown the exact deviation instead, none of it stuck. How much they learned from success and failure varied from person to person and depended on how much their movements scattered by themselves [33].
• Learning through repetition. A movement carried out often enough becomes more reliable, even without working out the error. This route counts as largely preserved; after cerebellar damage it is less well tested than the other two.
• Deliberate strategies. Intending to start further left, or to begin more slowly, bypasses the working out of the error. This is the route the therapist's prompts aim at.

Whether more explanation also brings more learning is open. In one study, 40 people with a cerebellar disease and 40 comparison participants of the same age practised an aimed forearm movement on five days. All of them improved when they saw their movement on a screen. One group additionally received detailed verbal feedback and instruction, another targeted feedback about joint position. Neither addition brought an advantage over plain practice with sight of the movement [34].

3.3 Three consequences for the session

• Few tasks, many repetitions. Not twenty different exercises, but a few, often enough. What is meant is the task and not the identical execution: as soon as something succeeds safely, pace, direction, surface and surroundings are varied, so that what has been practised carries outside the practice as well. The recommendations for stroke rehabilitation name exactly this combination: many repetitions of everyday activities, and of ones that challenge [40].
• Feedback from outside. A mirror, a line on the floor, two strips of tape, a video on the phone, a counter, a sentence from the therapist. They supply the information that no longer arrives reliably from inside. Which form works best and how often it should come is unresolved – in the study of 40 people, more explanation was no better than less [34]. The aids are taken away when the movement succeeds without them.
• A target that can be hit or missed. Because learning from success and failure is preserved [33], exercises are given a clear mark: the cup on the spot, the foot inside the taped square, ten steps within the track. That is why so much is taped out and counted in this treatment.

These three points are the reason why the exercises below look so unspectacular. A line on the floor is not a makeshift; it is the tool.

How quickly this shows varies. Clear changes come partly within the first days and weeks, further ones over months. Section 16 describes how you can tell.

3.4 What remains available for practice

The cerebellum holds reserves. If part of the tissue is preserved, neighbouring areas take over some of the tasks. For this capacity an international expert group proposed the term cerebellar reserve in 2020 [3].

The same group states that the reserve is addressed through training, and that the evidence for this comes predominantly from animal experiments and from small studies of people with progressive cerebellar diseases [3]. How large the reserve is in an individual person cannot be measured in advance. That is why one begins, and after eight to twelve weeks looks at the measurements to see what has come of it.

4. What the site means for your programme

All the building blocks of the following sections exist. The site within the cerebellum indicates which problems are likely and where a start is worthwhile. It does not decide the programme on its own. That is done by the assessment, the risk of falling, what you can tolerate, and what you are missing in everyday life.

This mapping is not derived from anatomy alone. In a study of 90 people with a circumscribed cerebellar lesion, it was tested for every image point in the cerebellum which symptoms go with damage at that place. Unsteadiness in standing and walking was tied to the midline, inaccuracy of arm and leg to the deep-lying nuclei and the adjoining cortical parts, and slurred speech to the parts beside the midline in the upper cerebellum [4].

The last row deserves an explanation. Deep within each half of the cerebellum lie the cerebellar nuclei. Everything the cerebellum passes on to the rest of the brain runs through them. In the same study of 90 people, symptoms after damage involving these nuclei persisted at every age, while damage to the cerebellar cortex was compensated better [4]. From this it can be inferred that such damage needs more time and that a walking aid comes up more often. That is an inference from the observed course and not a tested rule of treatment – and a statement about the effort, not about hopelessness.

These mappings are pointers, not a programme. Damage does not keep to boundaries, and many people have several things at once. What is actually practised follows from the examination and from what you are missing in everyday life.

5. The first assessment

5.1 What is measured, and why

Without baseline values there is little telling six months later whether anything has changed. Memory deceives in both directions: on good days the progress seems larger, on bad days it seems to have vanished. That is why measurements are taken at the beginning and at intervals afterwards.

The ataxia scale was developed and tested in 2006 on 167 people with an inherited ataxia [10]. Of its 40 points, 18 fall on gait, stance and sitting [11]. It therefore responds above all to changes in trunk and standing steadiness. Anyone mainly affected in the hand needs a measure alongside it that captures this.

Two things about these figures matter. First, they fluctuate from day to day; a single measurement says little. Second, a change in a measurement does not yet mean that everyday life has changed. That is why both are recorded: the figure, and what you can do again in everyday life.

Depending on the findings, more is added. These additions are customary in practice; a selection tested specifically for stroke in the cerebellum does not exist:

• Balance in motion. The Berg Balance Scale tests above all positions that are held. Anyone who stands safely and still wobbles while walking is captured more precisely by a test containing walking, turning, head movements and obstacles. The Mini-BESTest and the Functional Gait Assessment are in common use.
• Falls and near-falls. It is recorded how often you have fallen in recent months and how often you have only just caught yourself. Near-falls are the more sensitive figure, because there are more of them.
• Confidence in your own balance. A questionnaire on how safe you feel during particular activities. Anyone with little confidence goes out less, regardless of what the walking distance would allow.
• Dizziness. A questionnaire on how much the dizziness restricts everyday life. What you report is the actual outcome here.
• Hand and arm. Where the arm is inaccurate, a test capturing grasping and handling. The ataxia scale maps this too coarsely.

5.2 What your balance rests on

Balance arises from three sources: from what you see; from what skin, joints and muscles report; and from the balance organ in the inner ear. The cerebellum weighs these three against each other and reweights them according to the situation. In the dark the ground under your feet counts for more, on soft ground the eye does.

After a stroke in the cerebellum this weighing gets out of order. The assessment therefore includes the question of the conditions under which balance holds and under which it does not:

• standing and walking with eyes open and with eyes closed,
• firm ground against a compliant surface, a thick foam mat for instance,
• head still against head moving,
• a quiet room against busy surroundings: people, patterns, a bus pulling away in the field of view,
• daylight against dusk.

Concrete things follow from the result. Anyone who leans heavily on vision is least steady in the evening and in a crowd; that is where practice happens and where safeguarding is put in place. Anyone who gives way on compliant ground needs exercises on exactly that: carpet, grass, gravel. And anyone who starts to sway with head movements will find the matching component in section 12.

5.3 What you contribute yourself

The most useful part of the first session is not the measurements but three activities you miss most in everyday life. The programme grows out of them. "I want to walk safely again" leads to a general programme. "I want to manage the two flights of stairs to my flat without holding on" leads to one in which progress can be read off.

Bring these three activities as precisely as you can, with distance, surface and time of day. Section 19 names what else is useful.

6. The first days

An early start belongs to treatment on the stroke unit, the specialised stroke ward. On the dose during the first 24 hours there is a large trial with 2,104 people after a stroke [12].

One group was mobilised frequently and extensively within 24 hours. The other group received the usual care of the ward. After three months, 46 out of 100 people in the group with the high early dose were independent in everyday life. In the group with usual care it was 50 out of 100 [12].

From this follows today's approach: out of bed early, in short units, with breaks. The trial included all types of stroke and was not analysed separately for the cerebellum.

The Canadian rehabilitation recommendations of 2025 draw the same conclusion: no high-dose mobilisation during the first 24 hours. It begins once the person is medically stable, as a rule between 24 and 48 hours after the event [40].

During the first days five things matter:

1. sitting safely,
2. standing safely,
3. making nausea and dizziness bearable,
4. preventing pneumonia, thrombosis and pressure sores,
5. and having swallowing assessed before anything is eaten, drunk or taken as a tablet for the first time [32].

6.1 Swallowing is checked after every stroke

This point stands here and not with the brainstem, because it holds regardless of the site. The guidelines on acute treatment provide for swallowing to be checked after every stroke, before anything is eaten, drunk or taken as a tablet for the first time [32]. Until that has happened, physiotherapy too offers nothing to drink.

That this holds for the cerebellum alone as well is shown by an analysis of 102 people whose infarction lay exclusively in the cerebellum. Of 100 such people, around 13 had a swallowing disorder. It was mostly mild: most often the bolus slipped back too early, while larger residues in the throat or entry into the airways were not observed in this group. One of the affected people died; in the others the swallowing improved, and one was discharged with restrictions on diet [35].

Both halves follow from this. Damage in the cerebellum alone leads less often to a severe and lasting swallowing disorder than damage elsewhere in the brain. And swallowing is checked in everyone all the same, because there is no telling in advance who belongs to the 13.

6.2 After a haemorrhage

After a haemorrhage the same principle holds with two additions. Blood pressure is followed during the first sitting up and the first standing, because it is kept low with medicines during this period. And where surgery was carried out, the start follows the clearance from neurosurgery.

During this time, increasing headache, increasing vomiting and increasing drowsiness are to be reported at once. They are the warning signs that space within the skull is running short; the sister guide describes them under section 7.

7. Trunk: sitting and standing

The build-up goes from the inside outwards and from supported to free. The trunk has to hold first, otherwise the arm has nothing to work against.

With truncal ataxia the treatment starts sitting on the edge of the bed, with support at the side. The build-up follows a fixed order:

1. Sitting with hand support, then without.
2. Sitting and reaching forward with one hand, then to the side.
3. Standing with support on both sides, then with one hand on a rail, then freely.
4. Standing with the feet closer together.
5. Standing while turning the head, moving the gaze, picking something up.

The step to the next stage follows when the previous one succeeds for about 30 seconds without a balancing step. This threshold comes from practice. There is no tested boundary for it.

Two further adjusting screws are added once the basic stages are settled: the surface, that is firm ground and later a compliant mat, and vision, that is eyes open and later briefly closed, always with safeguarding. This practises exactly what section 5.2 assessed.

Two things make the difference. First the feedback: a mirror in front, a marking on the floor between the feet, a hand of the therapist showing where the middle is. Second the number of repetitions: a stage is not practised three times in a session but thirty times.

Everything named so far practises anticipatory balance: you know what is coming and prepare for it. In everyday life the other kind often decides, namely what is left when you are pushed unexpectedly or catch your foot. Section 10 describes that part.

That this is worthwhile is well established for stroke in general. A summary of 467 studies with 25,373 people on physiotherapy after stroke names the training of sitting balance as the measure with the largest effect among all those tested [13]. For balance in the chronic stage there is a summary of 43 studies; 28 of them with 985 people could be pooled for the Berg Balance Scale. The groups with an exercise programme lay on average 2.2 points higher than the comparison groups (95% confidence interval 1.3 to 3.2) [16]. Programmes with balance, weight-shifting and gait training did best. These studies included all types of stroke.

8. When the body is pulled to one side

Some people are pulled towards one side, namely towards the side of the damage. In sitting the trunk tips that way, in standing the weight travels that way, in walking the path veers that way. The technical term is lateropulsion.

The reason lies in perception. The internal reference for "upright" is shifted. In 36 people with an infarction at the lateral edge of the brainstem, the felt vertical was tilted towards the side of the infarction throughout. In 12 of those 36, those with the strongest pull to one side, head posture and eye position were rotated in the same direction as well [8].

From this follows the most important rule of this section: telling someone to sit up straight does not help. Someone who sits crooked while being convinced of sitting straight cannot correct this from within. The vertical has to come from outside:

• a mirror in which you see your own tilt,
• a vertical edge in the room: a door frame, a window post, tape on the wall,
• a plumb line or a string with a weight,
• a hand at the shoulder showing where the middle is,
• a wall on the side towards which the pull goes.

What is practised is straightening up against the pull: first in sitting with feedback, then in standing, then without a mirror. This approach comes from practice and from the reasoning that a shifted internal vertical is replaced by an external one. A study with a comparison group testing this for the pull to one side after a stroke in the brainstem does not exist.

For relatives this section matters. "Do sit up straight" sounds like a lack of effort. What is meant is a shifted perception, and effort changes nothing about it.

9. Walking

Walking is worked on at four points: the width of the track, the speed, the posture of the trunk and the safety in turning.

These exercises are the usual ones:

• Walking along a line on the floor. The line supplies the feedback about the track from outside.
• Walking with a narrowing track. Two strips of tape on the floor set the width and are moved together over weeks.
• Walking with changes of pace, with stopping on command, with changes of direction.
• Walking while turning the head or carrying something.
• Stairs up and down, with one hand on the rail.

For gait training after a stroke there is a guideline that grew out of a review of the evidence in 2020. It recommends gait training at a high heart rate and with many steps as the most effective component for walking speed and walking distance [15]. The guideline relates to people with stroke in general and assumes that the person is safe enough to walk briskly.

The same guideline states that balance exercises in sitting and standing do not make up for gait training where walking speed and walking distance are concerned [15]. Balance is practised because it counts in its own right. Walking is what one does in order to walk better.

That places two demands side by side, alternating within the session: the precise, slow work on the track, and the brisk, sustained walking. Both belong. Anyone practising only slowly and precisely does not become faster in everyday life; anyone only walking briskly loses the track.

For people with marked ataxia, gait training begins with safeguarding: a harness system, a treadmill with handrails, or two people to guide. The safeguarding permits a pace and a number of repetitions that would be too risky without it.

10. When balance is lost unexpectedly

Up to here it has been about balance you see coming: standing up, stretching, taking a step. The body prepares for it. In technical language this is anticipatory postural control.

In everyday life the other kind often decides. Someone brushes past you on the bus, your foot catches the edge of the carpet, the dog pulls on the lead. Then fractions of a second remain in which to place a step in the right direction or to grab hold. This is reactive postural control. After a stroke in the cerebellum it is often more affected than it looks in quiet standing, and quiet standing does not make it visible either.

It is practised like this:

• Disturbances from outside. The therapist gives a short pull or push at the pelvis or the shoulder: announced at first, later without announcement; small at first, stronger later.
• Protective steps in all directions. Forwards, backwards and sideways. The sideways step across the supporting leg is the hardest one and the one most often missing in everyday life.
• Grabbing hold. Of the handrail, the door frame, the back of the chair. That too is a reaction that needs practice.
• Unexpected calls while walking. Stop, turn around, step aside, climb over something that appears only at the last moment.
• Changing ground. From a hard floor onto carpet, onto grass, onto gravel, over a threshold, down a ramp.
• Getting up from the floor. Anyone who can do this gets back up alone after a fall instead of lying there until someone comes. It is practised via four-point kneeling to a stable piece of furniture. It is not suitable for everyone, but it often comes into question earlier than is assumed.

This needs safeguarding: a harness system, a second person, mats, a rail within reach. Unannounced disturbances are not practised without safeguarding.

How much this achieves against falls is open. In one trial, 88 people whose stroke lay more than six months back received either training with unexpected balance disturbances or usual balance training over six weeks. In the year afterwards the first group fell on average 1.45 times per person, the second 1.72 times; with the numbers of this trial that difference cannot be distinguished from a chance finding. The reaction to a disturbance improved more in the group with the disturbance training, and this advantage held over twelve months. The authors call their result inconclusive in so many words [41].

For stroke in the cerebellum in particular this component is untested. It stands here because it makes the difference between "safe in the treatment room" and "safe on the pavement". On falls and their prevention there is a separate guide under Preventing falls.

11. Arm and hand

With inaccuracy of aim in the arm, the movement is taken apart and slowly built up again:

• Reaching for a fixed target, first with the elbow supported.
• Reaching without support, with visual control of the target.
• Everyday movements with a target: pouring water, doing up a button, putting a key into a lock.
• Writing and drawing, because fine tuning and feedback come together in them.

During the first weeks the arm often looks different from what one expects. In one study, 16 people with a fresh cerebellar infarction were measured while pointing at a target. In the acute stage what stood out above all was how slow the movements were: of 100 such people, around 70 had speed and acceleration below the range of the comparison group of the same age. The disorder in the sequence of the movement came second to that. Two weeks later the values had recovered markedly; between two weeks and three months less was added [36].

For practice this means: at the beginning, pace and size of movement belong in the programme too, not only accuracy. A movement that succeeds only very slowly gets in the way of everyday life just as much as an inaccurate one.

With terminal tremor, support helps many people: the elbow or the forearm rests on a surface. The swing becomes smaller because one joint less moves with it. The same is achieved by an elbow propped on the table while eating and by a glass with a lid and a straw.

The choice of exercises follows everyday life particularly closely here. A movement you need twenty times a day is the better exercise than one that occurs only in the practice.

12. Gaze and dizziness

Where the image wobbles while walking, or the dizziness increases with every turn of the head, the gaze is practised. The exercises are called gaze stabilisation. They come from the treatment of disorders of the balance organ and are also used after damage in the brain [18][19].

A usual basic exercise: you hold a letter at arm's length in front of you, fix it and turn your head slowly to the left and right. The letter stays sharp. The duration starts at about 20 seconds and is increased over weeks. Progression also comes through the speed of the head turn, through the background (first a plain wall, later a patterned shelf) and through the starting position (sitting, standing, walking).

On the effectiveness after damage in the brain there is a retrospective analysis of 48 courses of treatment. The people received on average five treatments over five months. Balance, walking and dizziness symptoms improved between start and finish. There was no comparison group without treatment, so part of the improvement would have occurred anyway. Among the subgroups, the people with cerebellar damage improved the least [18]. The detailed guideline on these exercises concerns the disorder of the balance organ itself rather than the central disorder [19]; the transfer is customary and less well grounded.

For stroke in general there is by now a summary of 15 trials with comparison groups and 769 participants. Those who received a gaze and balance programme in addition to usual rehabilitation had better balance afterwards; the certainty of this statement is rated moderate. The time for standing up, walking and turning back was on average 4.3 seconds shorter (95% confidence interval 2.0 to 6.7 seconds). The clearest results came from programmes combining gaze stabilisation with head movements, and from those running over four weeks [38]. This summary concerns stroke in general; how many of the participants had damage in the cerebellum alone cannot be read off from it.

One point holds regardless: head movements are brought back step by step, and the level follows how you react to them. Slight dizziness during the exercise is wanted; dizziness that takes you out of circulation for hours afterwards was too much. Keeping the head still permanently does not lead out of the symptoms, because the load then never rises again. How quickly it can be increased differs from person to person.

13. Strength and endurance

The ataxia concerns coordination. Strength and endurance decline all the same, because people affected move less for weeks. And both are the basis on which balance can be practised at all: anyone exhausted after ten minutes does not reach the repetition numbers that matter.

A Cochrane review summarises 75 studies with 3,017 people after a stroke. Endurance training reduced the limitation in everyday life, measured at the end of the training period. Mixed training of strength and endurance showed a smaller effect in the same direction. In addition, endurance, walking speed and balance improved. No serious adverse events were reported in any of the studies [17].

Strength training follows the same rules as elsewhere: two to three times a week, six to twelve repetitions, progression over weeks. With ataxia, guided machines or exercises in sitting are preferred, because less balance is needed for them. This allows strength to be built before balance is sufficient for it. There is a separate guide on the build-up under Strength training.

14. Aids and weights

A walking stick widens the base of support. A wheeled walker adds guidance. Both increase safety in everyday life and make longer distances possible. With them you practise more, not less – that is why an aid comes up early in this treatment.

Some people affected receive weight cuffs at the wrist or a vest with weights. The idea behind it: more mass damps the swing. The two applications are studied to different degrees and belong apart.

Weights on the arm. More speaks against them than their spread would suggest. In one study, 13 people with a cerebellar ataxia carried out aimed movements while various weights were attached to the arm without their noticing. For simple movements from a single joint, a weight could be found for almost every person with which they aimed more accurately. That very weight, the best one for them, mostly worsened the accuracy again as soon as the movement went across several joints. The authors question the benefit of arm weights as a treatment in so many words [37]. Everyday life consists of movements across several joints. We therefore do not propose weight cuffs as a routine trial.

Weights on the trunk. Here the situation is a different one; good it is not. The review of postural disorders in ataxia states that weights at the trunk showed an effect in individual small studies and that the evidence for them is thin [11]. A time-limited trial over two to four weeks is defensible, with a question settled in advance: can you do more in everyday life with them? If the answer is no, it is ended.

One observation helps with placing this. Among 12 people with a circumscribed cerebellar lesion, those with inaccurate leg placement adapted their walking to additional weights on the lower leg less well than those with good balance; in them the deep-lying cerebellar nuclei beside the midline were more often involved [5]. Anyone who becomes less steady with weights rather than steadier has a possible explanation here, and the trial is stopped.

15. How much and how often

For the amount of practice a relationship holds in stroke rehabilitation: more practice time goes along with a better result. An analysis of studies in which one group received more therapy time than the other found an advantage for the group with the longer time. The difference was small to moderate (standardised difference 0.35; 95% confidence interval 0.26 to 0.45) [14]. The summary of 467 studies points the same way: what is effective are programmes with a high number of repetitions that start from everyday activities [13].

How much that is in concrete terms is stated by the recommendations for stroke rehabilitation of 2025. For people in hospital or in a rehabilitation clinic who are medically and neurologically stable, they name about three hours of targeted, task-specific therapy a day on five days of the week. What is meant is the sum of all the disciplines involved – physiotherapy, occupational therapy and speech and language therapy together – not three hours of physiotherapy. Anyone who goes home earlier and is looked after there should receive the same amount. For the time afterwards the same recommendations name 60 minutes per session per discipline on two to five days of the week, and set the duration expressly according to need, goals and progress [40].

Applied to a usual course in Switzerland this means:

• In hospital: several short units a day rather than one long one, spread across the disciplines involved.
• Outpatient afterwards: often two to three physiotherapy sessions a week. That is a realistic example and not a tested standard dose. Nothing speaks against more if you tolerate it and your goals call for it.
• On the remaining days a home programme of 10 to 20 minutes.

15.1 The home programme

The home programme decides the total amount. Two hours of physiotherapy a week is little against seven days of everyday life.

Three rules have proved themselves:

• Short and coupled. 10 to 20 minutes, attached to an existing habit: after brushing your teeth, before the news, while the kettle boils.
• Few exercises, often. Three to five exercises you know by heart are worth more than a sheet with twelve.
• With safeguarding. Balance exercises at home take place at the kitchen counter, in a door frame, or with the back of a chair within reach.

On the level of difficulty there is an observation more important for the design than the duration. In one study, 14 people with cerebellar ataxia practised an individually tailored balance programme at home for six weeks. Walking speed, stride length and the time for standing up, walking and turning back improved over the six weeks, and most improvements were still there a month later. How much the walking speed improved depended on how strongly balance was challenged in the exercises – and not on how long they practised, how old the person was, or how marked the ataxia was [25].

That study had 14 participants and no comparison group. It does not establish how effective a home programme is. What it shows is the direction: an exercise in which you never come close to wobbling is too easy. The point at which it only just works is the point at which practice happens.

15.2 After physiotherapy ends

The practice continues, in a group or alone. The reason is in the figures. In the study of people with progressive cerebellar disease who practised intensively for four weeks, the gains fell back within 24 weeks in some of the participants; in 22 of 42 people at least one measurement was still better after 24 weeks than at the start [22]. In another group the improvement held over a year in those who kept practising [21].

In practical terms: before physiotherapy ends, what continues afterwards is settled – a group, a club offering, a fixed home programme, a check-up after three months.

16. How you can tell there is progress

The first clear gains partly come within days and weeks. In the study of 16 people with a fresh cerebellar infarction the arm movement had largely recovered after two weeks [36]. In a study of 23 people with an infarction in the cerebellum alone, unsteadiness in standing had resolved completely after three months [26]. After that it continues, only more slowly: the largest steps fall into the first three months, smaller ones are added over six and twelve months [28].

Both ends of that range matter. Anyone who notices no change after two weeks has no reason to hold the practice ineffective. And anyone who experiences a large jump in the first two weeks has no reason to stop there.

Three signs say more in everyday life than the feeling on a single day:

• A stage in the build-up has become stable. What needed help four weeks ago works alone today.
• The aids fall away. The mirror is no longer needed, the tape lies narrower, the hand on the rail becomes a finger on the rail.
• One of your three everyday activities works again. That is the measure that counts.

For the measurements: compare quarter with quarter rather than week with week. And read a change in a measurement together with everyday life. An improvement of two points on a scale is an indication; whether you go shopping with it is the answer.

On recovery after rehabilitation there is an analysis of 58 people who came to a rehabilitation clinic after a cerebellar infarction or a cerebellar haemorrhage. Independence in everyday life was measured on a scale from 18 to 126 points. On admission the mean was 65.5 points, on discharge 89.8 points. Among the 45 people reached after an average of 19.5 months, the mean value was 123.5 points [27]. That analysis was retrospective and had no comparison group; part of the improvement would have occurred without rehabilitation too.

Two things slow progress down and therefore belong in the conversation. One is fatigue: of 100 people after a stroke, around 50 report lasting fatigue [30]. The other is mood: around 31 out of 100 develop depression [31]. Both change how much practice is possible, and both belong in the medical follow-up. On top of that, damage in the posterior lateral parts of the cerebellum brings changes in planning, switching and emotional regulation [29], which can make keeping to a home programme harder. Where a programme is not running, this question is worth asking before any increase in the amount of practice.

On falls and their prevention there is a separate guide under Preventing falls.

17. What the studies give – and what they do not

This section deliberately comes after the exercises and not before them. It says how certain what stands above actually is.

Little has been tested directly on stroke in the cerebellum. A German study followed 23 people with a fresh infarction confined to the cerebellum over three months. Unsteadiness in standing had resolved completely after three months; a mild gait disorder remained, above all in pace. Some of the participants received treadmill training with increasing speed over two weeks. Between this group and the comparison group no difference showed [26]. The authors conclude that two weeks are too few and that more intensive coordination training should be tested. That test is still outstanding today.

Most of the knowledge comes from progressive cerebellar diseases. Two studies are decisive for the choice of exercises:

• Four weeks of coordination training, 16 people. Balance, trunk control and aimed movements, with guided sessions and a daily home programme. The ataxia decreased, and the improvement was still there after eight weeks. People with purely cerebellar damage benefited more than people with additional damage to the feedback pathways [20].
• Four weeks of inpatient rehabilitation, 42 people. Random allocation: one group started at once, the other four weeks later. The group treated at once improved more in ataxia, walking speed and everyday independence. Truncal ataxia improved more than ataxia of arm and leg [22].

A review of 19 studies on the treatment of postural disorders in ataxia sums up: there is evidence of moderate certainty that rehabilitation improves steadiness in standing. Best studied are progressive ataxias and multiple sclerosis; two of the 19 studies concerned people after stroke [11]. A review of inherited ataxias comes to the same picture [23]. On this basis an international expert group recommends regular coordination and balance training as a fixed part of treatment [24].

A more recent summary evaluates 26 studies on exercise therapy in cerebellar ataxia; 8 of them could be pooled. On the ataxia scale the exercise groups lay on average 3.3 points lower than the comparison groups (95% confidence interval 2.8 to 3.7), on the Berg Balance Scale 2.6 points higher (1.1 to 4.2). For independence in everyday life no difference showed. The certainty of these statements is rated low to moderate [39].

Six qualifications belong with this:

• A different cause. After a stroke the tissue improves by itself; with a progressive disease it deteriorates. The transfer is plausible and untested.
• Small groups. The participant numbers lie between 14 and 42 people.
• No blinding. Anyone who trains knows it. Expectation acts on the test results.
• Comparison groups are often missing. During the first weeks much improves by itself. Without a comparison group the share due to treatment cannot be determined.
• Different outcomes. What was measured was mostly scale points and walking speed. Whether what you are missing in everyday life also succeeds is a second question.
• Whole components untested. For training with unexpected balance disturbances (section 10) and for the work against the pull to one side (section 8) there is no study with a comparison group in stroke of the cerebellum. They stand in this text because they follow from the findings, and not because they have been tested.

In summary: that practice after a stroke in the cerebellum helps is well grounded and little tested for this condition. How much it helps is open. We say so here because you need it for weighing things up – and because a practice offering exercise therapy should have no interest in concealing that gap.

18. Common misunderstandings

• "I have to build strength first, then the balance comes." Where the damage lies in the cerebellum alone, weakness is usually not the leading sign [1]. Strength and endurance are trained because they decline during the weeks of reduced activity and because they carry the amount of practice. The balance does not wait for that.
• "If I try harder, the movement becomes more accurate." With ataxia, effort often increases the swing. What helps is slow execution, a clear target, support and feedback from outside.
• "The mirror is a crutch, I should practise without it." The mirror supplies the feedback that no longer arrives reliably from inside [2]. It is taken away when the movement succeeds without it, and not before.
• "Do sit up straight." With a pull to one side the felt vertical is shifted along with it [8]. The instruction goes nowhere; a visible vertical is what is needed.
• "Dizziness improves if I keep my head still." Keeping the head still permanently does not lead out of the symptoms, because the load then never rises again. Head movements are brought back step by step, at a level that follows your reaction [19][38].
• "Weights at the wrist make the movement steadier." In 13 people with cerebellar ataxia, an individually fitted weight improved simple movements from one joint and mostly worsened movements across several joints again [37]. Everyday movements belong to the second group.
• "Nothing happens in the first weeks anyway." In the study of 16 people with cerebellar infarction, a large part of the recovery of arm movement fell into the first two weeks [36]. The weeks afterwards keep bringing something, only more slowly.
• "A wheeled walker is a sign of giving up." An aid increases the distance you cover and lowers the risk of falling. With it you practise more, not less.
• "Two weeks of training achieved nothing, so training achieves nothing." In the only training study directly on cerebellar infarction the programme lasted two weeks and showed no difference [26]. The programmes that did show something lasted four weeks and continued with a daily home programme [20][22].
• "If something is left after six months, it stays for good." The largest gains fall into the first three months. Smaller ones over six and twelve months do occur [28].
• "If I stand safely in the treatment room, I am safe outside too." Quiet standing does not show what happens when you are pushed unexpectedly. That part is practised in its own right (section 10).
• "Exercises at home have to be done for a long time to work." In the study of 14 people with cerebellar ataxia the improvement depended on the level of difficulty and not on the duration [25].
• "Physiotherapy for stroke in the cerebellum is well studied." Most training studies come from people with progressive cerebellar diseases [20][22]. Little has been tested on stroke in the cerebellum itself [26].

19. What to bring to therapy

For the first session these details are useful:

• the hospital discharge report with the site and the type of the stroke,
• from it in particular which part of the cerebellum is affected and whether the brainstem was involved,
• whether a swallowing disorder was found in hospital and whether speech and language therapy was involved,
• the current list of medicines,
• three activities you miss most in everyday life,
• whether and when you have fallen, and whether you have only just caught yourself in between,
• your walking aids, if you have any,
• the shoes you wear in everyday life.

Useful questions for the therapist:

• Which site in my cerebellum is affected, and what does that mean for the choice of exercises?
• Which measurements do we record today so that we can compare in three months?
• What do I practise at home, how often and for how long?
• How do I recognise that an exercise is too easy?
• What do I do if I become dizzy while practising?
• When do we talk about a walking aid?
• When do we practise what happens with an unexpected disturbance – and when getting up from the floor?
• What continues when physiotherapy ends?

20. In summary

After a stroke that stays confined to the cerebellum, weakness is usually not the leading sign. What is missing is the coordination. Accuracy and balance therefore stand in the foreground [1].

Because the cerebellum performs the matching between movement and feedback, learning from one's own movement error is impaired [2]. Other routes remain usable: learning from success and failure [33], repetition, and the deliberate strategy. Mirror, line on the floor, video and the therapist's sentence supply the feedback that no longer arrives reliably from inside. Which form works best is open [34].

The site within the cerebellum indicates where a start is worthwhile. Midline: trunk, standing, walking. Laterally: arm, leg and speech on the same side. Below: gaze and dizziness. Where the damage reaches into the brainstem, the work on the vertical is added [4][7]. The decision follows the findings, the risk of falling, what you can tolerate and your goals. Swallowing is checked after every stroke, before anything is eaten or drunk for the first time [32][35].

The build-up goes from supported to free and is checked against measurements and against three everyday activities. Alongside anticipatory balance comes the reactive kind: protective steps, unexpected disturbances, getting up from the floor [41]. Start early, in short units [12]. The amount follows need and goals; on an outpatient basis two to three hours a week is a common example, plus a home programme of 10 to 20 minutes on the remaining days [40]. For the home programme the level of difficulty counts for more than the duration [25].

The first clear gains can come within days and weeks [36]; the largest steps fall into the first three months, smaller ones are added over the year [28]. After physiotherapy ends the practice continues, otherwise part of what was gained falls back [21][22].

And: that practice helps is well grounded. How much it helps is open for this condition [26].

The medical side – cause, assessment, acute treatment, prevention – is in the sister guide Stroke in the cerebellum: infarction and haemorrhage.

References

All Digital Object Identifiers (DOIs) were checked individually against the Crossref register in August 2026. The links in the reference list lead through the DOI service to the publishers' pages. Some of these are located outside Switzerland and the EU. When you click, your IP address is transmitted to the respective provider. On our own site this does not happen.

[1] Marsden J, Harris C. Cerebellar ataxia: pathophysiology and rehabilitation. Clinical Rehabilitation. 2011;25(3):195–216. https://doi.org/10.1177/0269215510382495

[2] Bastian AJ. Understanding sensorimotor adaptation and learning for rehabilitation. Current Opinion in Neurology. 2008;21(6):628–633. https://doi.org/10.1097/WCO.0b013e328315a293

[3] Mitoma H, Buffo A, Gelfo F, et al. Consensus Paper. Cerebellar Reserve: From Cerebellar Physiology to Cerebellar Disorders. The Cerebellum. 2020;19(1):131–153. https://doi.org/10.1007/s12311-019-01091-9

[4] Schoch B, Dimitrova A, Gizewski ER, Timmann D. Functional localization in the human cerebellum based on voxelwise statistical analysis: A study of 90 patients. NeuroImage. 2006;30(1):36–51. https://doi.org/10.1016/j.neuroimage.2005.09.018

[5] Ilg W, Giese MA, Gizewski ER, Schoch B, Timmann D. The influence of focal cerebellar lesions on the control and adaptation of gait. Brain. 2008;131(11):2913–2927. https://doi.org/10.1093/brain/awn246

[6] Ilg W, Timmann D. Gait ataxia—specific cerebellar influences and their rehabilitation. Movement Disorders. 2013;28(11):1566–1575. https://doi.org/10.1002/mds.25558

[7] Sacco RL, Freddo L, Bello JA, Odel JG, Onesti ST, Mohr JP. Wallenberg's Lateral Medullary Syndrome. Clinical-Magnetic Resonance Imaging Correlations. Archives of Neurology. 1993;50(6):609–614. https://doi.org/10.1001/archneur.1993.00540060049016

[8] Dieterich M, Brandt T. Wallenberg's syndrome: Lateropulsion, cyclorotation, and subjective visual vertical in thirty-six patients. Annals of Neurology. 1992;31(4):399–408. https://doi.org/10.1002/ana.410310409

[9] Martino R, Foley N, Bhogal S, Diamant N, Speechley M, Teasell R. Dysphagia After Stroke: Incidence, Diagnosis, and Pulmonary Complications. Stroke. 2005;36(12):2756–2763. https://doi.org/10.1161/01.STR.0000190056.76543.eb

[10] Schmitz-Hübsch T, du Montcel ST, Baliko L, Berciano J, Boesch S, Depondt C, et al. Scale for the assessment and rating of ataxia: development of a new clinical scale. Neurology. 2006;66(11):1717–1720. https://doi.org/10.1212/01.wnl.0000219042.60538.92

[11] Marquer A, Barbieri G, Pérennou D. The assessment and treatment of postural disorders in cerebellar ataxia: A systematic review. Annals of Physical and Rehabilitation Medicine. 2014;57(2):67–78. https://doi.org/10.1016/j.rehab.2014.01.002

[12] The AVERT Trial Collaboration group. Efficacy and safety of very early mobilisation within 24 h of stroke onset (AVERT): a randomised controlled trial. The Lancet. 2015;386(9988):46–55. https://doi.org/10.1016/S0140-6736(15)60690-0

[13] Veerbeek JM, van Wegen E, van Peppen R, van der Wees PJ, Hendriks E, Rietberg M, Kwakkel G. What Is the Evidence for Physical Therapy Poststroke? A Systematic Review and Meta-Analysis. PLoS ONE. 2014;9(2):e87987. https://doi.org/10.1371/journal.pone.0087987

[14] Lohse KR, Lang CE, Boyd LA. Is More Better? Using Metadata to Explore Dose–Response Relationships in Stroke Rehabilitation. Stroke. 2014;45(7):2053–2058. https://doi.org/10.1161/STROKEAHA.114.004695

[15] Hornby TG, Reisman DS, Ward IG, et al. Clinical Practice Guideline to Improve Locomotor Function Following Chronic Stroke, Incomplete Spinal Cord Injury, and Brain Injury. Journal of Neurologic Physical Therapy. 2020;44(1):49–100. https://doi.org/10.1097/NPT.0000000000000303

[16] van Duijnhoven HJR, Heeren A, Peters MAM, Veerbeek JM, Kwakkel G, Geurts ACH, Weerdesteyn V. Effects of Exercise Therapy on Balance Capacity in Chronic Stroke: Systematic Review and Meta-Analysis. Stroke. 2016;47(10):2603–2610. https://doi.org/10.1161/STROKEAHA.116.013839

[17] Saunders DH, Sanderson M, Hayes S, Johnson L, Kramer S, Carter DD, Jarvis H, Brazzelli M, Mead GE. Physical fitness training for stroke patients. Cochrane Database of Systematic Reviews. 2020;2020(3):CD003316. https://doi.org/10.1002/14651858.CD003316.pub7

[18] Brown KE, Whitney SL, Marchetti GF, Wrisley DM, Furman JM. Physical Therapy for Central Vestibular Dysfunction. Archives of Physical Medicine and Rehabilitation. 2006;87(1):76–81. https://doi.org/10.1016/j.apmr.2005.08.003

[19] Hall CD, Herdman SJ, Whitney SL, et al. Vestibular Rehabilitation for Peripheral Vestibular Hypofunction: An Updated Clinical Practice Guideline From the Academy of Neurologic Physical Therapy of the American Physical Therapy Association. Journal of Neurologic Physical Therapy. 2022;46(2):118–177. https://doi.org/10.1097/NPT.0000000000000382

[20] Ilg W, Synofzik M, Brötz D, Burkard S, Giese MA, Schöls L. Intensive coordinative training improves motor performance in degenerative cerebellar disease. Neurology. 2009;73(22):1823–1830. https://doi.org/10.1212/WNL.0b013e3181c33adf

[21] Ilg W, Brötz D, Burkard S, Giese MA, Schöls L, Synofzik M. Long-term effects of coordinative training in degenerative cerebellar disease. Movement Disorders. 2010;25(13):2239–2246. https://doi.org/10.1002/mds.23222

[22] Miyai I, Ito M, Hattori N, et al. Cerebellar Ataxia Rehabilitation Trial in Degenerative Cerebellar Diseases. Neurorehabilitation and Neural Repair. 2012;26(5):515–522. https://doi.org/10.1177/1545968311425918

[23] Milne SC, Corben LA, Georgiou-Karistianis N, Delatycki MB, Yiu EM. Rehabilitation for Individuals With Genetic Degenerative Ataxia: A Systematic Review. Neurorehabilitation and Neural Repair. 2017;31(7):609–622. https://doi.org/10.1177/1545968317712469

[24] Ilg W, Bastian AJ, Boesch S, et al. Consensus Paper: Management of Degenerative Cerebellar Disorders. The Cerebellum. 2014;13(2):248–268. https://doi.org/10.1007/s12311-013-0531-6

[25] Keller JL, Bastian AJ. A Home Balance Exercise Program Improves Walking in People With Cerebellar Ataxia. Neurorehabilitation and Neural Repair. 2014;28(8):770–778. https://doi.org/10.1177/1545968314522350

[26] Bultmann U, Pierscianek D, Gizewski ER, Schoch B, Fritsche N, Timmann D, Maschke M, Frings M. Functional recovery and rehabilitation of postural impairment and gait ataxia in patients with acute cerebellar stroke. Gait & Posture. 2014;39(1):563–569. https://doi.org/10.1016/j.gaitpost.2013.09.011

[27] Kelly PJ, Stein J, Shafqat S, Eskey C, Doherty D, Chang Y, Kurina A, Furie KL. Functional Recovery After Rehabilitation for Cerebellar Stroke. Stroke. 2001;32(2):530–534. https://doi.org/10.1161/01.STR.32.2.530

[28] Stinear CM, Lang CE, Zeiler S, Byblow WD. Advances and challenges in stroke rehabilitation. The Lancet Neurology. 2020;19(4):348–360. https://doi.org/10.1016/S1474-4422(19)30415-6

[29] Schmahmann JD, Sherman JC. The cerebellar cognitive affective syndrome. Brain. 1998;121(4):561–579. https://doi.org/10.1093/brain/121.4.561

[30] Cumming TB, Packer M, Kramer SF, English C. The prevalence of fatigue after stroke: A systematic review and meta-analysis. International Journal of Stroke. 2016;11(9):968–977. https://doi.org/10.1177/1747493016669861

[31] Hackett ML, Pickles K. Part I: Frequency of Depression after Stroke: An Updated Systematic Review and Meta-Analysis of Observational Studies. International Journal of Stroke. 2014;9(8):1017–1025. https://doi.org/10.1111/ijs.12357

[32] Powers WJ, Rabinstein AA, Ackerson T, et al. Guidelines for the Early Management of Patients With Acute Ischemic Stroke: 2019 Update to the 2018 Guidelines for the Early Management of Acute Ischemic Stroke. Stroke. 2019;50(12):e344–e418. https://doi.org/10.1161/STR.0000000000000211

[33] Therrien AS, Wolpert DM, Bastian AJ. Effective reinforcement learning following cerebellar damage requires a balance between exploration and motor noise. Brain. 2016;139(1):101–114. https://doi.org/10.1093/brain/awv329

[34] Draganova R, Konietschke F, Steiner KM, Elangovan N, Gümüs M, Göricke SM, Ernst TM, Deistung A, van Eimeren T, Konczak J, Timmann D. Motor training-related brain reorganization in patients with cerebellar degeneration. Human Brain Mapping. 2022;43(5):1611–1629. https://doi.org/10.1002/hbm.25746

[35] Schaefer JH, Luft F, Seiler A, Harborth E, Kaffenberger S, Polkowski C, Foerch C, Lapa S. Prevalence, recovery and phenotype of dysphagia in patients with ischaemic cerebellar stroke. European Journal of Neurology. 2024;31:e16303. https://doi.org/10.1111/ene.16303

[36] Konczak J, Pierscianek D, Hirsiger S, Bultmann U, Schoch B, Gizewski ER, Timmann D, Maschke M, Frings M. Recovery of Upper Limb Function After Cerebellar Stroke: Lesion Symptom Mapping and Arm Kinematics. Stroke. 2010;41(10):2191–2200. https://doi.org/10.1161/STROKEAHA.110.583641

[37] Zimmet AM, Cowan NJ, Bastian AJ. Patients with Cerebellar Ataxia Do Not Benefit from Limb Weights. The Cerebellum. 2019;18(1):128–136. https://doi.org/10.1007/s12311-018-0962-1

[38] Meng L, Liang Q, Yuan J, Li S, Ge Y, Yang J, Tsang RCC, Wei Q. Vestibular rehabilitation therapy on balance and gait in patients after stroke: a systematic review and meta-analysis. BMC Medicine. 2023;21:322. https://doi.org/10.1186/s12916-023-03029-9

[39] Winser S, Chan HK, Chen WK, Hau CY, Leung SH, Leung YH, Bello UM. Effects of therapeutic exercise on disease severity, balance, and functional independence among individuals with cerebellar ataxia: A systematic review with meta-analysis. Physiotherapy Theory and Practice. 2023;39(7):1355–1375. https://doi.org/10.1080/09593985.2022.2037115

[40] Nelson MLA, Shi J, Lindsay MP, et al. Canadian Stroke Best Practice Recommendations: Rehabilitation, Recovery and Community Participation Following Stroke. Part One: Stroke Rehabilitation Planning for Optimal Care Delivery, 7th Edition Update 2025. American Journal of Physical Medicine & Rehabilitation. 2026;105(1):59–75. https://doi.org/10.1097/PHM.0000000000002802

[41] Mansfield A, Aqui A, Danells CJ, Knorr S, Centen A, DePaul VG, Schinkel-Ivy A, Brooks D, Inness EL, Mochizuki G. Does perturbation-based balance training prevent falls among individuals with chronic stroke? A randomised controlled trial. BMJ Open. 2018;8(8):e021510. https://doi.org/10.1136/bmjopen-2018-021510

Transparency

• Authorship: Roger Hilfiker
• AI support: the literature search and the text draft were produced with Claude (Anthropic). Roger Hilfiker checked all statements, figures and sources and revised the text.
• Created: 23 August 2026
• Updated: 24 August 2026. Revised after an expert review: the chapter on motor learning (sections 3.2 and 3.3), the checking of swallowing (now section 6.1, independent of brainstem involvement), the figures on the amount of therapy (section 15) and on the time course of recovery (section 16), and the account of weights (section 14). Newly added are the assessment of sensory conditions (section 5.2) and reactive balance with protective steps (section 10).
• Relation to the sister guide: this text is the physiotherapy version of Stroke in the cerebellum: infarction and haemorrhage. That one describes cause, assessment and acute treatment, this one the practice. The shared figures come from the same works; the numbers in the reference list are self-contained per article.
• Sources: the 41 works in the reference list. All DOIs were checked against the Crossref register.
• Conflict of interest: the practice offers physiotherapy for dizziness, balance disorders and neurological conditions as a service. This article recommends exercise therapy. We disclose this so that you can take it into account while reading. Section 17 names the gaps in the evidence for the same reason.
• Funding: Physiotherapie Tschopp & Hilfiker, 3902 Glis. The article was produced from the practice's own resources.
• Next review: planned for August 2028