Sciatica
Sciatica stretches, and how to find out which direction is yours
Most sciatica stretch lists assume everyone's back wants the same thing. Roughly three quarters of people have a direction that settles their symptoms, and it is not the same direction for everyone.
Clinically reviewed by Lee Albert, NMT · 8 August 2026
The place that hurts is not the place to work on
Sciatica is a symptom rather than a diagnosis. The sciatic nerve is being irritated somewhere along its path, and the leg is where you feel it, but the irritation is almost always higher up: where the nerve root leaves the spine, or where the nerve passes through the deep muscles of the buttock. That is the single most useful thing to know before you start stretching, because it explains why hamstring stretches so often do nothing for it, and occasionally make it louder.
It also explains the two families of movement below. One group settles the lower back, which is where the irritation usually starts. The other reaches the deep buttock, which is where the nerve passes through on its way down. Which of those helps is a question about you, not about the word sciatica.
Direction, and how to test yours
In the best-known trial on this, a directional preference was found in about three quarters of people with low back pain. Of those who had one, roughly 83 percent settled with movements that arch the back, about 7 percent with movements that bend it forward, and about 10 percent with side-bending or rotation. So the majority preference is extension, and a meaningful minority is the opposite, and nothing about the diagnosis label tells you which group you are in.
Testing it is simpler than it sounds. Pick one movement from a direction, do it gently a few times, then notice what happened afterwards rather than during. Two things are worth watching for. Symptoms drawing back up the leg toward the spine, which is a good sign and means keep going. Symptoms spreading further down the leg, which means that is not your direction today, and pushing harder into it is the specific mistake to avoid.
That last point is not a matter of comfort. In the same trial, people given the opposite direction to their preference did substantially worse: a third of them stopped within two weeks because they were not improving or were getting worse, while none of the matched group did.
Why nothing here is a loaded forward fold
Every movement below is either passive, supported, or unloaded. That is a deliberate line rather than a stylistic one. The evidence about flexion and discs is about loaded, repeated and end-range bending, not about resting a spine in a supported fold, and the difference between those two matters most in exactly this population.
So knees drawn to the chest, a strap-assisted leg raise and a supported fold over a chair are all here. A standing toe-touch and a loaded seated forward fold are not. Every movement on this page was checked against the app's own sciatica filter, which is the same filter that runs when someone selects sciatica in the assessment.
The sequence
- 1
Single knee to chest · 30 to 60 seconds each side
The gentlest place to start, and a useful first test: one leg at a time tells you whether the two sides behave differently, which they often do here.
- 2
Reclined leg raise with a strap · 30 to 60 seconds each side
The clinical library ranks this first for a nerve-sensitive low back. The floor holds your spine and the strap sets the height, so you can find the point where the leg starts to object and stop just short of it.
- 3
Figure 4 with a strap · 60 to 90 seconds each side
Reaches the deep buttock muscles the nerve passes through, which is often where the irritation actually sits rather than in the back itself.
- 4
Sphinx · 30 to 60 seconds
The extension test, and the direction the majority of people settle with. Low, supported, and easy to come out of if the leg does not like it.
- 5
Cobra, minimal lift · 5 slow lifts
A small repeated extension rather than a hold, which is how a direction is usually tested. Watch what happens to the leg afterwards, not during.
- 6
Supported forward fold over a chair · 30 to 60 seconds
The flexion option for the minority whose symptoms settle bending forward, done over a chair so the thighs carry the weight instead of the spine.
- 7
Supine twist, small range · 60 seconds each side
The rotation option, kept small. About one person in ten settles with side-bending or rotation rather than either of the other two directions.
- 8
Legs up on a chair · 5 minutes or longer
Takes the load off the lower back entirely and asks nothing of you. A reliable place to finish, and often the most comfortable position available on a bad day.
Do not do all eight. Pick the first three, then try either Sphinx or the supported fold and notice which direction your leg prefers over the next few hours. Whichever one draws symptoms back up toward the spine is the one to keep. Anything that sends them further down the leg comes out of the routine, however good it is supposed to be for sciatica.
What the research says
The reference guidance is NICE NG59, which covers low back pain and sciatica together as one population. Its recommendations are unusually open: provide advice and information tailored to the person, including encouragement to continue with normal activities, and consider a group exercise programme that is biomechanical, aerobic, mind-body or a combination, chosen around the person's needs, preferences and capabilities. It does not prescribe a direction of movement, and it does not name a best stretch. That is not an omission. It is the finding.
The reason no guideline prescribes a direction is that direction is a property of a person rather than of a diagnosis. In a randomised trial of 312 people with low back pain, including people with sciatic symptoms, a directional preference was identified in 74 percent, split 83 percent extension, 7 percent flexion and 10 percent lateral. Matching the exercise to that preference produced significantly greater improvement in every outcome measured than either the opposite direction or non-directional exercise. A more recent systematic review with meta-analysis puts the prevalence of directional preference at roughly 60 percent when assessed by trained therapists, and finds matched exercise outperforms generic exercise in the short term.
On how much exercise does in general, the largest review available is the 2021 Cochrane review of exercise therapy for chronic low back pain, which pooled 249 trials and found moderate-certainty evidence that exercise beats no treatment, usual care or placebo for pain, by about 15 points on a 100-point scale, with smaller effects on function. None of this is about Vinys and none of it endorses it. Taken together it says something simple: moving helps, the direction that helps is individual, and a list of stretches that assumes one direction for everyone is guessing on your behalf.
Sources
- 1NICE NG59, low back pain and sciatica in over 16s (2016, last updated 29 July 2026)
Provide tailored self-management advice including encouragement to continue with normal activities, and consider a group exercise programme (biomechanical, aerobic, mind-body or a combination) chosen around the person's needs, preferences and capabilities.
- 2Long et al., does it matter which exercise? A randomised controlled trial of exercise for low back pain (Spine, 2004)
A directional preference was found in 74 percent of 312 people with low back pain, split 83 percent extension, 7 percent flexion and 10 percent lateral, and matching exercise to it produced significantly greater improvement in every outcome than the opposite or non-directional exercise.
- 3Hennemann et al., the McKenzie Method for chronic low back pain with directional preference, systematic review with meta-analysis (Journal of Manual and Manipulative Therapy, 2024)
Directional preference, meaning sustained symptom reduction after repeated movement in one direction, is present in roughly 60 percent of people with chronic low back pain when assessed by trained therapists.
- 4Hayden et al., exercise therapy for chronic low back pain (Cochrane Database of Systematic Reviews, 2021)
Across 249 trials, moderate-certainty evidence that exercise is more effective than no treatment, usual care or placebo for pain (about 15 points on a 100-point scale), with smaller effects on function and only small differences against other conservative treatments.
Common questions
What is the fastest way to relieve sciatica?
There is no honest fast answer, and pages that promise one are guessing. What the evidence supports is finding the direction of movement your own back settles with, and then doing gentle work in that direction rather than pushing into the one that provokes it. Legs up on a chair is usually the most comfortable position while you work that out.
Should I stretch my hamstrings for sciatica?
Cautiously, and with a strap rather than by reaching for your toes. The sciatic nerve runs through the same territory as the hamstring, so a straight-leg stretch pulls on the nerve as well as the muscle. If a hamstring stretch produces burning or electricity rather than a broad muscular pull, that is nerve tension and the answer is to lower the leg, not to push further.
Which stretches make sciatica worse?
Whichever ones send the symptoms further down your leg, and that is individual rather than universal. In general terms, loaded forward bending, deep twisting and end-range work are the most common provokers, which is why nothing on this page is loaded or end-range. The signal to watch is where the symptom travels, not how much it hurts while you are doing it.
How do I know if it is sciatica or something else?
Sciatica usually travels below the knee, follows a line rather than spreading everywhere, and is often one-sided. Buttock pain that does not travel that far may be coming from the deep hip muscles instead. Either way, the movements here are gentle enough to be a reasonable starting point, and persistent or worsening symptoms deserve an assessment rather than a longer stretching routine.
How often should I do these?
Little and often works better here than one long session, because the aim early on is to test how your back responds rather than to accumulate volume. Two or three short rounds spread through the day is a reasonable shape once you have found a direction that suits you.
Can I do these if I cannot get down to the floor?
Yes. The figure 4, the supported fold and the seated versions of the twist can all be done in a chair, and legs up on a chair only needs you to get onto a bed or a sofa. The chair yoga pages cover a full practice built that way.
Seek urgent care for numbness around the groin or back passage, any change in bladder or bowel control, or weakness in both legs, which need same-day assessment. See a doctor rather than starting a routine if the symptoms followed a significant fall, if there is weakness in the leg or foot rather than just pain, or if you have unexplained weight loss or fever alongside the pain. Stop any movement that sends symptoms further down the leg.
Also called sciatica stretches, stretches for sciatica, sciatica exercises, sciatic nerve stretches, sciatica stretches for seniors, best stretches for sciatica. Movements are drawn from the clinically authored Vinys library. This page is general movement information, not medical advice, and it is not a diagnosis. How the method works.