Lower back pain after Pilates: double-leg moves, rolling, swan and the reformer, and a plan back

Pilates helps back pain, and harms are rarely measuredEmerging evidence

In a Cochrane review of 10 trials (510 adults with long-lasting back pain), Pilates eased pain and disability more than minimal care. Only 2 of those trials looked for side effects: one found none and the other minor ones. Another review found only one trial that reported on safety, and in 54 women with long-lasting back or neck pain, reformer classes had no side effects linked to them.

No country collects injury statistics for Pilates, so there is no reliable injury rate, and an Australian association of Pilates instructors says the usual problems are muscle strain, flares of existing conditions and occasional falls from equipment. In an online survey of 494 Dutch adults who had a gym injury in the past year, the back was the commonest injured site among the 84 whose activities included yoga or Pilates (19 of them, 23%); that is 17% of the people surveyed, the two weren't separated, and there were no hours, so no rate. A news report describes a fall from the reformer's moving carriage and two reports of broken bones in reformer classes.

Sources 1, 2, 3, 4, 5, 6

A short rise in pain is the usual side effectStrong evidence

Across trials of exercise for long-lasting back pain (all kinds, not only Pilates), at least one side effect was reported in 33% of exercise groups and 29% of comparison groups, at about 0.14 events per person where they were measured, mostly minor ones such as muscle soreness. Across exercise for long-lasting pain, most side effects were more soreness or muscle pain that settled after a few weeks, and an umbrella review found no serious ones in back pain.

In 60 women, pain rose a little straight after Pilates exercises in those with back pain, less after the mat exercise than the equipment ones. So a sore back after a new class is common, and it usually settles.

Sources 7, 8, 9, 10

What Pilates asks of the lower backEmerging evidence

Four groups of moves, in our reading: legs held or lowered against gravity (the hundred with straight legs, the double-leg and straight-leg stretches, scissors, the teaser, the corkscrew), where the back arches if the trunk can't hold the pelvis still; rounding and rolling (the roll-up, rolling like a ball, the seal, the open-leg rocker, the roll-over family, the neck pull, the spine stretch forward and the saw); back extensions (swan, the double-leg kick, swimming, rocking and the reformer's long box); and planks on the moving carriage (the long stretch).

No lab study has measured the load on the lower back in a Pilates move; Pilates lab studies measure muscle activity only. In one, the mat versions of the teaser and long spine needed more stomach muscle work than the equipment versions (16 women), and in another, the criss-cross worked the trunk hardest, with the single-leg stretch and the dead bug similar (32 people). The load figures come from other exercises in young, healthy adults. The deep hip flexor (psoas), which attaches to the lower spine, worked at 15% to 35% of its maximum in sit-ups and under 10% in a curl-up (8 adults); linking that to back pain in Pilates leg work is our reading, as it wasn't measured in Pilates or in people with pain. Partial curl-ups gave the best challenge for the least low back load of 12 exercises, and lifting the chest against gravity loaded the spine most of four back extensions.

In pig spines bent tens of thousands of times under load, disc damage was more linked to the repeated bending than to the squeezing. That is an animal model with extreme repetition, not evidence that a class harms discs, and loading in the right dose is how backs get stronger.

Sources 11, 12, 13, 14, 15, 16, 17

Telling the causes apart, and when a scan mattersStrong evidence

The NHS says it's not always obvious what causes back pain, and it often gets better on its own. Many backs have a sore direction and an easy one: pain on arching that eases bending forward, or the reverse. Leg symptoms below the knee point to a nerve, and one-sided pain by the dimple of the pelvis may be the SI joint.

For most low back pain, the National Institute for Health and Care Excellence (NICE) says not to offer imaging routinely outside specialist settings. The exceptions are the warning signs: sudden pain with thin bones, where an X-ray looks for a fracture; a young person with lasting pain on arching; leg weakness, numbness or bladder or bowel changes; and a fall.

Sources 18, 19

How it usually goesStrong evidence

The NHS says back pain usually improves within a few weeks, though it can last longer or come back. In 33 cohorts (11,166 people), pain in recent back pain fell from 52 to 23 out of 100 within 6 weeks. It often returns: in 250 people who had recovered, 69% had another episode within 12 months, and earlier episodes are the best predictor. A flare after a class is often part of that usual course, not new damage. No one can promise a date.

Sources 18, 20, 21, 22

Keep moving, and keep doing Pilates with changesStrong evidence

The NHS says to stay active and carry on with your daily activities, not to stay in bed for long, and that activities like walking, swimming, yoga and Pilates may help ease back pain. NICE suggests a group exercise programme, which can include mind-body exercise. In the Cochrane review, Pilates reduced pain by about 14 points out of 100 against minimal care in the short term (low quality) and about 11 later on (moderate quality), but it isn't clearly better than other exercise.

A network meta-analysis of 89 studies ranked Pilates first for pain against no exercise, on low-quality evidence (the journal issued an expression of concern about the paper and later withdrew it, with no change to the paper), and an umbrella review found that 7 such analyses favoured Pilates and motor control exercise for pain (low certainty). The differences between types of exercise are small next to the difference between exercising and not.

Sources 18, 19, 1, 23, 9

Gentler classes: about as good, with fewer side effectsEmerging evidence

In 168 adults with long-lasting back pain, low-intensity Pilates helped pain and disability as much as high-intensity Pilates, and about 20 fewer people in 100 had a side effect with the gentler classes (from the trial's summary; the types of side effect weren't in it). In 296 adults, Pilates twice a week cut pain by about 2.3 points out of 10 against advice alone, against 1.2 once a week and 2.1 three times a week. So after a flare, a gentler class twice a week is a sensible choice.

Sources 24, 25

Change the double-leg, rolling and extension moves for a whileEmerging evidence

No trial has tested a change to Pilates for back pain that Pilates caused. A review of 87 studies of ab exercises lists avoiding active hip flexion with fixed feet, and keeping the knees and hips bent, among its safety rules. A Pilates teaching platform says the teaser is "Not recommended for those with lower back injuries", and cues the swan to avoid lifting so far "that you feel a pinch in the lower back". Our suggestion from these: bend your knees and keep your legs higher in leg work, lower them only as far as your back stays still, use the half roll back instead of the full roll-up, and keep back extensions small.

Sources 26, 27, 11, 13

You don't need to brace hardEmerging evidence

In 152 people with long-lasting back pain, being taught to keep the stomach relaxed during Pilates improved disability slightly more than being taught to keep it contracted (1.4 points on a 0 to 24 scale), with a possible small benefit for pain. No study shows that pressing your lower back into the mat, or bracing hard, protects the spine.

Source 28

Mat or reformer: both helpEmerging evidence

In 86 adults, mat and equipment Pilates eased pain equally, with equipment slightly better for disability at 6 months; in 98 adults, both helped. Neither is clearly gentler on a sore back. Lighter springs, keeping your knees down in the long stretch and a still carriage when you get on and off are our suggestion.

Sources 29, 30, 3

Thin bones: swap deep rounding, keep doing PilatesStrong evidence

A UK consensus found that "physical activity and exercise is not associated with significant harm, including vertebral fracture", but asks for movements with sustained, repeated or end-of-range rounding to be changed, and says that "As a precaution, alternatives to exercises such as the ‘roll down ‘and ‘curl up’ in Pilates should be considered". Its review of harms found bending activities (certain yoga moves and sit-ups) were linked with more spinal fractures, but that these events were rare. The Royal Osteoporosis Society says "Generally, Pilates is very safe and won't cause a spinal fracture", and its fact sheet keeps the hundred, roll-up and leg stretches only for people with very strong stomach and back muscles from doing them regularly, swaps them for a wedge or the knee fold for most, doesn't recommend rolling like a ball, and says the roll-over "places a lot of pressure on your upper spine and neck".

The evidence behind the caution is thin. The best-known study wasn't randomised: new fractures followed in 89% of 9 women given bending exercises against 16% of 25 given back extension exercises (and 67% of 6 who did none). The case series are of yoga bending (9 people and 3 people), and we found no published spinal fracture in Pilates. Pilates trials in people with low bone density helped pain and function (70 women over a year in one) but reported no data on harms, and Pilates didn't change bone density. An international consensus favours teaching spine-sparing technique such as the hip hinge over blanket bans, with back strength for everyone.

Sources 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43

Who should check firstEmerging evidence

Australian physiotherapists who teach Pilates agreed that people with pre-eclampsia, an unstable spondylolisthesis or a fracture may not benefit (expert opinion, not trial data). With a diagnosed back condition, that guide's plan comes first.

Source 44

Your Pilates for now (written steps)

Our 3D figure shows the moves, not the springs, the straps, how low your legs go or what you tell your instructor. These steps come from back pain trials, lab studies, the UK osteoporosis consensus, the Royal Osteoporosis Society, a Pilates teaching platform and the move pages; where a step is our suggestion, it says so. None has been tested in a trial against back pain from Pilates.

• 1. A warning sign first: numbness around your genitals or bottom, bladder or bowel changes, weakness or numbness in a leg, a fall from the reformer, sudden pain with thin bones, or a teenager whose back hurts arching for more than a couple of weeks.
• 2. Tell your instructor before class what hurts and which moves you'll change (our suggestion). For example: "My lower back flared after last week's class. I'll keep my knees bent in the leg work and skip the roll-over and teaser; could you suggest options?"
• 3. Double-leg moves: knees bent or legs higher, one leg at a time, and lower them only as far as your back stays still (a review's safety rules); leave out the teaser for now, as a teaching platform advises with a lower back injury.
• 4. Rounding and rolling: the half roll back or chest lift instead of the full roll-up, and no rolling like a ball, seal, open-leg rocker, roll-over, jack knife or neck pull while it hurts (the move pages).
• 5. Extensions: smaller lifts in swan, the double-leg kick and swimming, stopping short of a pinch, and rocking left out (a teaching platform's cue and a lab study of back extensions).
• 6. Bracing: breathe and let your stomach work without gripping hard; a relaxed stomach did slightly better in a trial.
• 7. The reformer (our suggestion): lighter springs, a still carriage when you get on and off, the long stretch on your knees or left out, the long box extension kept small, and footwork with your back resting.
• 8. Choose the class: a slower, lower-intensity class for a few weeks, twice a week rather than more (two Pilates trials).
• 9. Thin bones: no rolling moves or roll-over family; the wall roll-down or knee fold instead of the roll-up, a hip hinge with a straight back instead of rounding in the spine stretch forward and the saw, and twists kept smooth and comfortable (the UK consensus and the Royal Osteoporosis Society).
• 10. Between classes, keep walking and doing your daily activities, and don't stay in bed for long, as the NHS says.

Sources 26, 27, 13, 28, 24, 25, 31, 34, 35, 18