Trapezius strain is the name almost all of us give to that dull pain between neck and shoulder, where the muscle feels like a cord under the touch. It is one of the most common musculoskeletal complaints among people who work at a computer.
Yet many certainties circulate on this subject that the studies do not support: that it is caused by posture, that there are knots to release with the right pressure, that massage will fix it. This guide sets out what holds up, with the numbers and the level of certainty stated.
Please note: the exercises and advice below are general and may not suit your situation. If you have had an injury, if you have neurological symptoms, or if the pain does not behave as described here, get it checked before you start.
What really happens in the trapezius
First a note on the name, because almost everyone searches for it this way: “inflamed trapezius”. It is a borrowed word, and it describes poorly what is actually there. A trapezius strain does not show inflammation in the classic sense (redness, swelling, heat). What you find is a muscle that stays active more than it needs to, is sore to the touch and tolerates load less well. This changes the substance, not just the vocabulary: if it were inflammation, it would make sense to rely entirely on anti-inflammatories, whereas what actually shifts the outcome is movement.
The trapezius is a broad, thin muscle that covers the upper part of the back. It originates from the base of the skull and from the spinous processes of the cervical and thoracic vertebrae, and it inserts on the clavicle, the acromion and the spine of the scapula.
It is described in three portions with different jobs: the upper portion elevates the shoulder blade and takes part in neck movements, the middle portion retracts it, and the lower portion depresses and rotates it. The pain we are talking about almost always concerns the upper portion, the part you can touch between the neck and the shoulder.
The word strain suggests a muscle locked in a shortened state, as if the fibres were stuck together. It is an effective image but an imprecise one: what is actually observed is pain, tenderness on pressure and a subjective sense of tension, not an injury visible on tests.
In the literature the same picture is called trapezius myalgia or non-specific neck pain. Non-specific does not mean imaginary: it means that no single damaged structure can be identified that needs repairing. The pain is real and measurable, but a simple mechanical explanation does not exist.
This distinction changes practice. If the problem were a locked muscle, the logical solution would be to unlock it. If instead it is a sensitised tissue that is poorly trained for the load you ask of it, the logical solution is to train it. The studies, as we will see, push in the second direction.
“Swollen trapezius”: what you actually feel
People with a strain often describe the muscle as swollen, and the word deserves a clarification because it points you in the wrong direction. What you feel under your fingers is a contracted muscle, firmer and more bulky to the touch, sore at specific points, not fluid that has built up. A real swelling, with stretched or reddened skin, is something else and should be looked at by a doctor.
(The muscle is called the trapezius; “trapezoid muscle” is the same thing with the adjective, and you will sometimes see it on medical reports.)
Trigger points: what we know and what we don't
Trigger points are described as hypersensitive points within a taut band of muscle, able to reproduce the patient's pain and to refer it to other places. In the upper trapezius they are said to be very frequent, with pain referred towards the temple, eye and jaw.
It is worth knowing where these criteria come from. In 2018 an international Delphi panel of 60 experts from 12 countries established that the essential criteria are three: a taut band, a hypersensitive point and referred pain. The authors themselves open the study by admitting that there was previously no consensus and that different criteria were in use.
This is an important detail: that definition comes from agreement between experts, not from an objective test proving that the structure exists. The model remains debated in the literature, to the point that replies to the study have been published in the same journal.
The most serious point, however, concerns finding trigger points by hand. A systematic review with meta-analysis pooled six studies involving 363 patients and calculated the agreement between different examiners who palpate the same person.
The result is a Cohen's kappa of 0.452 (95% confidence interval: 0.364-0.540). The authors conclude, in their own words, that manual palpation to identify trigger points is not reliable.
Within that figure there is a useful nuance. The criteria that hold up best are the most basic ones: localised tenderness (kappa 0.676) and the patient recognising the pain (kappa 0.575). The palpable nodule, that is, precisely the element that is described most confidently, is the one on which two professionals agree least.
What to do with this in practice: if a specific point hurts and pressing it reproduces your pain, that is real information and clinically useful. The map of knots to deactivate one by one, on the other hand, is far less solid than it is sold as being.
Be wary of anyone who promises to find the knot and eliminate it in a single session. Not because they feel nothing under their fingers, but because another professional, on the same shoulder, would find others.
Psychosomatic trapezius strain: what it means
This is the muscle on which this question comes up most often, and the short answer is that stress and tension do raise the tone of the trapezius muscles: you can see it when you hunch your shoulders towards your ears under pressure, and it happens to everyone. But “psychosomatic” does not mean imaginary: the pain is real and the muscle is genuinely contracted. We have covered this in full in psychosomatic back pain.
“Nerves pinched”: what lies behind that expression
It is an expression often heard, nervi accavallati in Italian, sometimes nervo accavallato, and it is worth saying straight away that it describes nothing anatomical: nerves do not overlap and do not tangle. Whoever uses it is almost always describing the same thing this page is about: a muscle that stays contracted, hard to the touch and sore at specific points.
The distinction matters, and it is not pedantry: if you think you have a pinched nerve, you expect someone to “put it back in place” and you look for a manoeuvre. If it is a muscle that will not let go, what changes it is movement and load over time. And a nerve that really is involved gives different signs, such as tingling or numbness running down the arm, or weakness, which you will find later in the list of signs to show a doctor.
Trapezius strain: how to sleep
There is no single position that fixes it, but there are a couple that usually make things worse at night and can be dropped.
Face down is the one that costs the most: it forces you to keep your head turned for hours. If it has always been your position, you will not change it in a single night, but it is worth trying.
On your side, with a pillow that fills the space between shoulder and ear: if the pillow is too low, the head drops downwards and the trapezius on the opposite side works all night; if it is too high, the reverse happens.
On your back with a not-too-high pillow is usually the most neutral.
The practical rule: if you wake up in a worse state than when you went to bed, the night is contributing, and the first thing to check is the height of the pillow relative to your shoulder.
The causes: what holds up and what does not
Posture matters much less than you have been told
The popular version is well known: rounded shoulders and a head pushed forward load the upper trapezius, so it hurts. It is an intuitive explanation and for that reason it circulates very widely. The data support it poorly.
The most cited systematic review with meta-analysis on the relationship between forward head posture and neck pain included 15 studies, all cross-sectional. These are snapshots taken at one moment: they can tell whether people who have pain also sit in a certain position, not whether that position caused the pain.
In the ten studies comparing people with and without pain, the mean difference was 4.84 degrees, with a confidence interval from 0.14 to 9.54. Statistically significant, but with a lower limit that is barely above zero: a thin margin for such a categorical claim.
In adolescents there was no difference at all: mean -1.05, interval from -4.23 to 2.12. And the correlations reported in adults had a negative sign, for example -0.55 with pain intensity and -0.42 with disability: a direction opposite to the one the hypothesis that posture equals pain would predict. The authors conclude that age acts as a confounding factor.
Put simply: some association exists in adults, but it is small, inconsistent, and does not demonstrate a causal link. No serious study allows anyone to tell you that you are standing badly and therefore it hurts.
This changes the tone of the whole matter. If your shoulders are a little rounded, you have not made a mistake from which your symptoms derive, and you do not need to spend the day policing your position. Blaming posture, besides being unfounded, adds tension to a picture in which tension is already part of the problem.
Exposure matters, but with modest numbers
A meta-analysis of prospective studies on workers who use a computer regularly, with at least one year of follow-up, found two significant predictors of arm, neck and shoulder disorders: using a computer or mouse for more than four hours a day, with a relative risk of 1.25 (interval 1.08-1.44), and body mass index, with a relative risk of 1.05 (1.02-1.08).
Look at the order of magnitude. More than four hours a day at the computer raises the risk by about a quarter; it does not multiply it. It is a real factor worth managing, not the single cause that explains your pain.
There are also windows in life when exposure multiplies without you having changed job. In the weeks after giving birth you spend hours with your head bent forward, often at night and without support under your arms, and the most frequent painful point is the upper part of the trapezius: what happens in that period, and what really changes, is covered in neck and shoulders stuck after childbirth.
A review of prospective studies on office workers adds a finding that is usually left out. Of 47 factors examined, strong evidence concerned only two predictors of onset: being a woman and having already had neck problems in the past. For many factors cited everywhere as causes, including low social support and high psychosocial stress, those studies showed no predictive value.
The most solid predictor, then, is precisely the one nobody talks about: having had an episode before. Anyone who has had neck pain tends to have it again. This is not fatalism; it is a reason to work on recurrence rather than chase the single episode.
Stress is a documented predictor, not a fault
On stress the picture is more nuanced, and it has to be stated in full, because it is easy to misuse.
A systematic review of 18 longitudinal studies on psychosocial work factors found strong evidence for the incremental effect of job demands, control over one's work, social support and job strain on the development of neck and shoulder disorders. These are studies that follow people over time, so they carry more weight than a snapshot.
The authors, however, add two explicit cautions. The results were too heterogeneous to calculate pooled risk estimates, and they should be interpreted with care before claiming that psychological factors have an independent causal influence. Their practical conclusion is that prevention must address physical and psychosocial factors together.
Set this beside the review on office workers, where high psychosocial stress did not predict onset, and you get the honest reading: stress is a documented predictor of neck and shoulder pain, but it is not consistent across all studies and it is not a single cause.
From this comes a rule that I keep for myself and for my patients. Stress explains something, it never justifies the phrase “it's all in your head”, and it is not a fault to be attributed to anyone who has a difficult life.
If your work puts you under pressure and your neck hurts, you have not brought the pain on through weakness: you are living through a relationship that research has measured and that deserves to be managed, not scolded. I have written about this in stress and neck pain.
What really works: the exercises
Here the literature is more generous, and the conclusion is fairly consistent in direction even if not in certainty.
The most relevant study is a randomised trial of 42 women with a clinical diagnosis of trapezius myalgia, divided into three groups for ten weeks: specific strengthening of the neck and shoulder muscles, general fitness done on an exercise bike for the legs, and a reference group with no physical activity.
Significant changes were seen only in the specific strengthening group. Pain fell by 42-49%, electromyographic activity of the sore trapezius rose by 42-86%, and shoulder abduction strength rose by 18-53%. In the exercise bike group and the reference group, nothing changed.
The most instructive detail is precisely the exercise bike group. Moving in general, which does good for a thousand other reasons, did not reduce that pain. What reduced it was loading the affected area. It is worth keeping this in mind before falling back on the generic advice to just move more.
The limits should also be stated: 42 participants is few, they were all women, and all were in one specific work context. A clear result in a small sample points a direction; it does not close the question.
Systematic reviews confirm the direction and state candidly how little certainty there is. A meta-analysis of office workers with chronic neck pain included eight randomised trials: seven reported a significant reduction in pain intensity and five in disability after strengthening exercises.
All eight studies, however, were at high risk of bias, and the overall certainty of the evidence was judged to be low. On quality of life, only one study measured the effect, and it did not find one.
A second meta-analysis, covering 29 trials of workplace interventions, eight of them high quality, observed that neck strengthening showed the clearest drop in pain scores. The authors' conclusion speaks of low-quality evidence that neck strengthening and targeted changes to the workstation reduce neck pain in office workers.
Low certainty does not mean that it does not work. It means that the available studies are few and imperfect, that the estimated effect could change with better research, and that I am giving you the intervention with the best evidence available, not a guarantee.
How to train the trapezius properly
The difference between those who improve and those who go round in circles almost always lies in load and consistency, not in choosing the perfect exercise.
- Shoulder shrugs with dumbbells or a band: lift the shoulders towards your ears against resistance, and control the descent. These are the most direct movement for the upper trapezius.
- Rows with a band or dumbbells: pull the elbows back while keeping the shoulder blades free to move. Loads the middle trapezius and rhomboids.
- Reverse fly, that is, rear openings: semi-straight arms that open backwards, with the torso leaning forward.
- Lateral raises up to about 90 degrees: load the deltoid and the upper trapezius together.
- Scapular retraction and depression: bring the shoulder blades together and move them slightly down, holding for five seconds.
- Chin tuck against light resistance: draw the chin back while keeping the gaze horizontal. It does not correct posture; it trains the deep neck flexors.
A reasonable dose is two or three sessions a week, three sets of eight to twelve repetitions, with a load that feels challenging in the final repetitions. The point is to increase the load over time: if after a month you are using the same band you started with, you are not training, you are repeating.
Mild discomfort during the exercise that clears within a day is acceptable. Pain that increases session after session, or that wakes you at night, is not: in that case reduce the load and get it checked.
Be realistic about timescales. The study on trapezius myalgia measured results at ten weeks. Weeks are the right unit of measure, not sessions.
Stretching remains useful and pleasant, but it needs to be put in context: it lengthens the side of the neck and gives relief, but it is not the piece that moves the needle. If you want a short routine to pair with strengthening, you will find one in daily stretching.
Massage, heat and passive therapies: what the evidence says
People like massage, and the reason is simple: it works straight away. The serious question is how long it lasts and how much it is worth. The comparison in full, covering when one option is better than the other, is in massage or physiotherapy: which to choose.
A systematic review with network meta-analysis published in 2025 pooled 101 randomised studies involving 7,633 participants with neck pain.
Compared with no treatment, soft tissue techniques reduced pain intensity with a weighted mean difference of -28.72 points on the pain scale used by the authors, with a confidence interval from -51.65 to -4.79. Mobilisation achieved -20.23, with an interval from -39.87 to -0.06.
Before getting excited, look at three things. The confidence intervals are very wide, which means the estimates are very imprecise. 61.3% of the 101 studies were at high risk of bias, and only 20.7% at low risk. And the authors rate the certainty of the evidence as moderate to very low.
There is a fourth point, the most important. The comparison is with no treatment, not with a credible placebo. Part of that benefit is almost certainly attention, contact and expectation: things that matter, but that do not make massage a cure for the tissue.
The same analysis finds that the most effective intervention is multimodal treatment, meaning two or more techniques combined. This too points in the right direction: not the passive session on its own, but a pathway that brings several ingredients together.
Local heat deserves a brief and honest discussion. A warm pack for fifteen or twenty minutes reduces discomfort and makes it easier to move. It is an excellent way to manage the symptom, it costs little and has no significant contraindications. It is not a therapy for the cause, and it does not need to be one.
Self-massage and a tennis ball against a wall belong in the same category: temporary relief that opens a window in which you can move better. Use them as a bridge towards exercise, not as a substitute.
Signs that require a doctor
First a premise that almost nobody makes. A systematic review of 29 clinical guidelines on neck pain found that twelve of them list 114 red flags in total, and that agreement between the guidelines on which to use is practically nil, with a median Fleiss kappa of zero.
The authors conclude that these signs rest mainly on expert opinion, and that the only recommendation with a solid basis concerns the Canadian C-spine rule for fractures after trauma.
So the list that follows is a shared list of clinical caution, not a battery of proven tests. Its purpose is to lower the threshold at which you ask for an opinion, not to let you self-diagnose.
- Pain that appeared after an injury, a fall or whiplash.
- Fever, shivering or general malaise together with neck pain.
- Unexplained weight loss or a personal history of cancer.
- Loss of strength, persistent numbness, or tingling running down the arm to the hand.
- Difficulty walking, with balance or with fine hand movements, such as doing up buttons.
- Sudden and severe headache, described as the worst you have ever had.
- Visual disturbances, intense dizziness, difficulty speaking or swallowing, nausea and vomiting.
The last group deserves particular attention because it can indicate a dissection of the cervical arteries, that is, an injury to the wall of the carotid or vertebral artery. It is rare, but it is an emergency: in that case you go to the emergency department, not to the physiotherapist.
A systematic review described 334 cases of serious adverse events after physical procedures on the cervical spine. 58% were vascular in nature, mostly arterial dissections, and 75% involved manipulation. But the authors note that some cases followed much milder procedures, including gentle mobilisation, exercises, acupuncture and even massage.
The early symptoms reported are useful to know: a sudden increase in headache or neck pain, nausea, vomiting, dizziness, changes in sensation, appearing during treatment or within 48 hours, often before the true neurological signs. The practical rule is simple. If after a session on the neck one of these symptoms appears, do not wait for it to pass.
Keep this in proportion: 334 cases collected across 233 studies over decades, against millions of treatments. The risk is very low. Knowing which symptoms not to ignore is nonetheless the best way to keep it that way.
If the pain, on the other hand, comes with recurring headache and none of these signs, the picture is almost always benign, and I discuss it in detail in neck pain and headache.
How to manage the day without chasing perfect posture
Given how fragile the story of posture as a cause is, prevention needs to be rewritten accordingly.
- Change position often instead of looking for the right one. The best position is the next one: stand up, shift your weight, move your shoulders.
- Break up long sessions. Above four hours a day of computer or mouse use the measured risk rises, so breaking those hours up makes sense.
- Set up your workstation so that it serves you, not to reach an ideal alignment: monitor at eye level, forearms supported, a backrest that gives support. The practical details are in desk ergonomics.
- If your day includes breastfeeding, the same principle applies: these are long, still sessions with your arms under load, and you break them up by preparing your supports before you start, not by correcting your back halfway through a feed. The breastfeeding positions and what changes for the neck and shoulders.
- Keep up two strengthening sessions a week even when you have no pain. This is the part that acts on recurrence, which is the most solid predictor we know of.
- Look after your mental load with the tools that work for you, sleep included. Not because it is your fault, but because it is a factor over which you have some hold.
- Do not neglect a new episode that lasts more than two or three weeks without changing.
When it makes sense to be supervised
A physiotherapist is needed above all for three things, and none of the three is releasing a knot.
The first is ruling out that something else lies underneath, with an assessment that takes your history into account and not only the painful spot. The second is building a progressive loading programme tailored to you, with doses that really increase over time. The third is keeping you on track during the weeks when the results are not yet visible, which are the weeks in which most people stop.
Pain between neck and shoulder lends itself well to telerehabilitation precisely because most of the work is exercise that you do yourself. Over a video call the picture can be assessed, the technique taught, the progress checked and the load adjusted.
Start with the Ri-Hub questionnaire if the pain has been recurring for weeks or if you do not know where to start with the exercises.
Stay in touch with us
Join our dedicated Facebook group: we answer all patients, give practical advice and clear up the most common doubts.
If the discomfort is concentrated on the left and you are wondering whether you should worry, the case is covered in Pain in the left trapezius.
If you often turn to massage and the tension comes back, the reason is explained in Decontracting massage: when it is needed and when it is not.
Sources
- Interrater Agreement of Manual Palpation for Identification of Myofascial Trigger Points: A Systematic Review and Meta-Analysis - Clinical Journal of Pain, 2017
- International Consensus on Diagnostic Criteria and Clinical Considerations of Myofascial Trigger Points: A Delphi Study - Pain Medicine, 2018
- The Relationship Between Forward Head Posture and Neck Pain: a Systematic Review and Meta-Analysis - Current Reviews in Musculoskeletal Medicine, 2019
- Office workers' risk factors for the development of non-specific neck pain: a systematic review of prospective cohort studies - Occupational and Environmental Medicine, 2012
- The incremental effect of psychosocial workplace factors on the development of neck and shoulder disorders: a systematic review of longitudinal studies - International Archives of Occupational and Environmental Health, 2013
- Computer use duration and body mass index as risk factors for arm, neck, and shoulder complaints among regular computer-using workers: A systematic review and meta-analysis - Work, 2026
- Effect of physical training on function of chronically painful muscles: a randomized controlled trial - Journal of Applied Physiology, 2008
- The influence of exercise on pain, disability and quality of life in office workers with chronic neck pain: A systematic review and meta-analysis - Applied Ergonomics, 2024
- Systematic Review and Meta-Analysis Suggest Strength Training and Workplace Modifications May Reduce Neck Pain in Office Workers - Pain Practice, 2021
- Effectiveness of musculoskeletal manipulations in patients with neck pain: a systematic review and network meta-analysis - BMJ Open, 2025
- Red flags for potential serious pathologies in people with neck pain: a systematic review of clinical practice guidelines - Archives of Physiotherapy, 2024
- Serious adverse events associated with conservative physical procedures directed towards the cervical spine: A systematic review - Journal of Bodywork and Movement Therapies, 2025
Last check of the sources: August 2026.
Need a consultation?
Start with the Ri-Hub questionnaire, a few questions about your case, you sign up, and you see who will follow you: a physiotherapist registered with the Albo, whom you can write to in chat whenever you like. If you prefer to talk it through, the first consultation is free, lasts 20 minutes and commits you to nothing: they arrive having already read what you wrote, and together you decide whether and how to go on. No credit card is needed: the first consultation commits you to nothing.
