Does Tendonitis Show Up on an MRI?

Tendonitis does show up on MRI in most cases, appearing as areas of abnormal signal intensity and often tendon thickening compared to the normally dark, uniform appearance of healthy tendons. But the relationship between what the MRI shows and what you actually feel is surprisingly messy. Abnormal signals appear in people with no pain at all, certain technical quirks can mimic disease on the scan, and for some types of tendon problems MRI is not the best imaging tool available. Understanding what MRI can and cannot tell you about your tendons is worth more than a simple yes-or-no answer.

What Healthy Tendons Look Like on MRI

To understand what tendonitis looks like on a scan, it helps to know what normal looks like first. Healthy tendons appear very dark on most standard MRI sequences because their tightly organized collagen fibers produce very little signal. The Achilles tendon, for example, shows up as a low-signal structure on standard sequences, though thin bright lines called septae that separate the individual sub-tendons are normal and expected.¹ Those normal septae are worth knowing about because radiologists sometimes mistake them for tears, a recognized interpretation pitfall.

When tendonitis or tendinopathy develops, that neat, dark appearance changes. The damaged or inflamed region lights up as a brighter area against the dark background. In the shoulder, tendonitis of the rotator cuff appears as focal or diffuse regions of increased signal, often paired with swelling of the tendon itself.² In the patellar tendon at the knee, increased signal shows up in roughly four out of five patients with clinically diagnosed tendinopathy, and the tendon is measurably thicker than in healthy controls.³ So the typical MRI signature of tendonitis is a combination of two things: brighter-than-normal signal within the tendon and tendon thickening.

How Reliably MRI Detects Tendon Problems at Different Sites

MRI’s ability to pick up tendonitis varies depending on which tendon you are looking at. The Achilles tendon is one area where MRI performs well. In a study of 118 painful Achilles tendons, abnormalities were detected in all but seven, and of 21 surgically confirmed areas of tendon degeneration, MRI identified 20.⁴ That same study found that more than one type of abnormality existed in about two-thirds of cases, including changes in the surrounding soft tissues and the nearby retrocalcaneal bursa. MRI’s ability to show everything in the neighborhood, not just the tendon itself, is one of its strengths. The retrocalcaneal bursa, a small fluid-filled sac behind the heel, tends to enlarge in people with Achilles tendon disorders compared to pain-free individuals.⁵

At the knee, MRI reliably identifies patellar tendinopathy as well. In athletes with chronic patellar tendinitis who eventually underwent surgery, every single knee showed focal thickening and abnormal signal in the upper third of the patellar tendon on MRI. When pathologists examined the tissue, those bright spots on the scan corresponded to actual structural changes in the tendon including cell overgrowth, new blood vessel formation, and microscopic tears in the collagen fibers.⁶

For lateral epicondylitis at the elbow (tennis elbow), MRI also picks up changes in the common extensor tendon. In one study, all but one patient with clinically diagnosed tennis elbow had visible tendon changes on MRI, graded from mild to severe.⁷ The severity of those MRI findings correlated with how much pain and functional limitation patients reported.⁸ That correlation matters because at some other body sites, MRI findings and symptoms do not line up so neatly.

MRI Versus Ultrasound

MRI is not the only way to image tendons, and for some tendon problems it is not the best way. Ultrasound is cheaper, faster, and in certain situations more sensitive. For patellar tendinopathy, one study found that ultrasound had an accuracy of about 83% compared to 70% for MRI, and its sensitivity was significantly higher.⁹ Ultrasound also has the advantage of being a real-time exam. The person performing it can press on the sore spot and watch the tendon move, which is something a static MRI scan cannot replicate.

That said, MRI has its own advantages. It provides a broader field of view, which means it can catch problems you were not expecting: bone marrow changes, joint fluid, ligament injuries, or other issues that might be contributing to your pain but would not show up on a focused ultrasound exam. For the rotator cuff, one comparative study found ultrasound had about 85% sensitivity for detecting tendonitis against MRI as the reference standard.¹⁰ The practical takeaway is that neither test is categorically better. Ultrasound depends heavily on the skill of the operator, while MRI depends on the scanner’s technical setup and the radiologist’s interpretation. For superficial tendons like the Achilles or patellar tendon, ultrasound often performs just as well or better. For deeper structures or when your doctor wants to rule out multiple possible diagnoses at once, MRI tends to be the preferred choice.

The Magic Angle Artifact and Other Technical Pitfalls

One of the most important things to know about tendon MRI is that it can show false abnormalities. The most common culprit is something called the magic angle effect. When a tendon curves at roughly 55 degrees relative to the magnetic field inside the scanner, it can appear artificially bright on certain MRI sequences, mimicking the look of tendonitis or a tear when nothing is actually wrong. This affects tendons that naturally curve, such as the peroneal tendons around the ankle or the supraspinatus at the shoulder.

Radiologists can reduce this artifact by choosing different imaging sequences. Using fat-suppressed T2-weighted images instead of standard proton-density sequences, for example, cut the magic angle effect in the peroneus brevis tendon by about 85% in one study.¹¹ But not every imaging center uses optimal sequences, and not every radiologist flags this artifact in their report. If you have been told your scan shows tendonitis in a tendon that runs at a curve, and it does not match your symptoms, the magic angle effect is worth asking about.

Abnormal Scans in People Without Pain

Here is where things get uncomfortable for anyone expecting their MRI to tell a clean story. Tendons can look abnormal on MRI in people who have zero symptoms. A study of young, asymptomatic volunteers found that every single supraspinatus and infraspinatus tendon in the shoulder showed at least mild signal changes on MRI, and about a quarter had moderate changes.¹² None of these people had shoulder pain. Similarly, high-resolution MRI of completely painless Achilles tendons revealed signal heterogeneity in many cases, with a handful showing small areas of what appeared to be early tendon degeneration.¹³

This matters for you because it means a positive MRI does not automatically explain your pain. If you get a scan that shows “tendinopathy” or “tendinosis,” that finding might be a real contributor to your symptoms, or it might be an incidental finding that has been sitting there quietly for years. Clinicians use the MRI alongside the physical exam and your history to decide whether what the scan shows is actually the problem. Treating the scan rather than the patient is a well-known trap in musculoskeletal medicine.

Distinguishing Tendonitis From a Tear

One of the more useful things MRI can do is help separate tendonitis or tendinosis from a partial or complete tendon tear, since the treatment implications differ. Quantitative MRI techniques can measure differences between these conditions. In the supraspinatus tendon, for instance, tendons with actual tears showed significantly higher T2 values on specialized mapping sequences compared to tendons with tendinosis alone, which in turn looked only mildly different from completely healthy tendons.¹⁴ On standard MRI without these advanced sequences, the distinction can be harder. Radiologists look at factors like whether the abnormal signal extends all the way through the tendon thickness, whether there is a visible gap or retraction, and whether the tendon surface is disrupted. A purely internal signal change with intact tendon surfaces generally points toward tendinopathy rather than a tear.

When MRI Falls Short on Calcific Tendonitis

Calcific tendonitis, where calcium deposits form within the tendon, is one scenario where MRI is not the best option. Calcium shows up dark on MRI, similar to tendons themselves, making it difficult to distinguish calcific deposits from other dark-signal structures or artifacts. Standard MRI sequences identified only about 59% of calcium deposits that were clearly visible on plain X-rays.¹⁵ A specialized technique called susceptibility-weighted imaging dramatically improved detection to about 98%, but it is not part of routine scanning protocols at most facilities. Plain X-rays remain the go-to for confirming calcific tendonitis because calcium is bright white and unmistakable on a radiograph.¹⁶

This is a good example of why a doctor who suspects calcific tendonitis may order an X-ray instead of, or in addition to, an MRI. No single imaging test is best for every type of tendon problem.

Contrast Dye and Tenosynovitis

Most MRI scans for tendonitis are done without injecting any contrast agent. But when the concern is tenosynovitis, which is inflammation of the sheath surrounding the tendon rather than the tendon itself, contrast-enhanced MRI can be more revealing. In the hand and wrist, standard non-contrast MRI detected tenosynovitis with a sensitivity of only about 40% in the hand and 67% in the wrist when contrast-enhanced images were used as the benchmark.¹⁷ Injecting a gadolinium-based contrast agent makes inflamed synovial tissue light up brightly, improving the ability to grade the severity of the problem. Contrast-enhanced hand MRI has been used in large cohorts studying early inflammatory arthritis, precisely because it detects subtle tendon sheath inflammation that non-contrast scans miss.¹⁸

If your doctor suspects you have rheumatoid arthritis or another inflammatory condition affecting the tendon sheaths, they may specifically request contrast-enhanced MRI. For garden-variety overuse tendonitis of the Achilles or rotator cuff, contrast is usually not needed.

Whether Getting an MRI Actually Changes Your Treatment

Having an abnormal scan does not automatically mean you need surgery or a different treatment plan. For atraumatic shoulder pain with suspected tendinopathy, a study found that over 90% of patients who received an MRI went on to improve with conservative treatment and never needed surgery. The authors estimated that the early MRI added a substantial and unnecessary economic burden without changing the clinical course for most patients.¹⁹ In these situations, physical therapy, activity modification, and anti-inflammatory measures would have been tried first regardless of the scan result.

There are scenarios, however, where MRI findings genuinely steer management. For insertional Achilles tendinopathy, the pattern of signal changes on MRI helped predict which patients would respond to non-surgical treatment and which would not. Patients whose scans showed confluent areas of internal signal changes were unlikely to improve without surgery, and identifying them early could spare months of fruitless conservative care.²⁰ Similarly, in athletes with chronic patellar tendinitis that had not responded to conservative treatment, MRI helped localize the damaged region for targeted surgical debridement, with most achieving good or excellent outcomes afterward.²¹

The general pattern: if your tendon pain is recent, you have no major strength loss, and you have not tried physical therapy yet, an MRI is often premature. If you have failed a reasonable course of conservative treatment and your doctor is deciding between continued non-operative care and something more invasive, MRI becomes much more valuable.

Reading a Post-Surgical Tendon Scan

If you have already had tendon surgery and get a follow-up MRI, the scan will look different from a pre-surgical one in ways that can be confusing. Repaired tendons go through predictable healing phases that produce their own signal changes, swelling, and altered appearance. A radiologist who does not know the surgical technique or where you are in recovery can mistake normal post-operative healing for new pathology.²² This is one reason surgeons emphasize sharing the surgical details with the imaging center before the scan. The timeline matters too: a tendon that looks worryingly bright at six weeks post-repair may be completely on track, while the same appearance at six months might warrant concern.

Newer MRI Techniques on the Horizon

Standard MRI sequences are tuned to detect signals from tissues with longer relaxation times, like fluid and fat. Tendons, with their short relaxation times, produce very little signal and appear dark, which is why radiologists rely on indirect signs: the tendon should be uniformly dark, and anything that breaks that uniform darkness is suspicious. Newer ultrashort echo time (UTE) MRI sequences can directly capture signal from short-T2 tissues like tendons, ligaments, and bone, offering a way to assess the tendon substance itself rather than just looking for deviations from darkness.²³ Early research using UTE techniques in Achilles tendons has shown differences in quantitative measures between patients with psoriatic arthritis and healthy volunteers, suggesting these sequences can detect subtle tendon changes that conventional MRI misses.²⁴ These techniques are still largely in the research phase and not widely available in routine clinical practice, but they point toward a future where MRI can do more than just flag abnormalities and actually measure the biochemical health of the tendon.

Artificial Intelligence and Automated Tendon Analysis

One of the more practical developments in tendon imaging is the application of deep learning algorithms to MRI interpretation. A systematic review and meta-analysis of deep learning models for tendinopathy detection found pooled sensitivity of about 91% and specificity of about 95%, with an overall diagnostic area under the curve of 96%.²⁵ An automated system trained specifically on supraspinatus tendon injuries achieved performance comparable to experienced radiologists and in some comparisons outperformed them.²⁶ These tools are not replacing radiologists today, but they are increasingly being tested as second-opinion tools that flag potential tendon abnormalities the human reader might overlook, especially in busy practices where scans are read quickly. For patients, this could eventually mean more consistent and accurate reporting of tendon findings, particularly at facilities that lack subspecialty musculoskeletal radiologists.