Shoulder MUA: Procedure, Recovery, and Alternatives

Shoulder manipulation under anesthesia, commonly called MUA, is a procedure in which a surgeon forcefully moves a frozen shoulder through its full range of motion while the patient is sedated, breaking up the scar-like tissue that has locked the joint. It is one of the most widely used interventions for adhesive capsulitis that has not responded to months of physical therapy, stretching, and injections. Research consistently shows it produces a large, rapid improvement in shoulder movement and a meaningful drop in pain, with around 85% of patients reporting satisfaction afterward.

What Happens During the Procedure

The basic idea is straightforward. While you are under general anesthesia or a regional nerve block, the surgeon stabilizes your shoulder blade with one hand and moves your arm through specific arcs of motion with the other. The sequence typically starts with forward flexion (raising the arm overhead), then abduction (moving it out to the side), then external and internal rotation. Each movement is pushed to the point where the thickened, contracted joint capsule tears or stretches, releasing the restriction. The whole manipulation usually takes only a few minutes once you are asleep.

Interestingly, MUA does not always require a full general anesthetic. Researchers have performed it successfully under local anesthesia, using a suprascapular nerve block combined with an injection of local anesthetic and steroid directly into the joint, followed by manipulation of the shoulder.

Who Ends Up Needing It

Frozen shoulder tends to develop in stages. First comes a painful “freezing” phase where inflammation builds in the joint capsule. Then comes stiffness, where the capsule thickens and contracts, severely limiting your ability to raise or rotate your arm. Most people eventually recover on their own, but the natural course can drag on for one to three years, and a meaningful minority never regain full motion without intervention.

MUA is generally considered after conservative treatments have failed for several months. In one large study, the average duration of symptoms before patients underwent MUA was about five months.¹ Both primary frozen shoulder (where no clear cause can be identified) and secondary frozen shoulder (triggered by injury, surgery, or another condition) can be treated with MUA, though the underlying cause affects how well people respond.²

How Much Pain and Stiffness Improve

The evidence on MUA’s effectiveness is quite strong when it comes to restoring movement. A randomized controlled trial comparing MUA directly against physiotherapy for stage-two frozen shoulder found that MUA produced a significantly greater gain in forward flexion and abduction at every follow-up point measured.³ In other words, people who had their shoulder manipulated could raise and spread their arms considerably further than those who went through therapy alone, and the gap persisted over time.

A comprehensive review of MUA outcomes found that pain dropped substantially at every time horizon. On a standard 10-point pain scale, scores fell by about 3.5 points within six weeks, 4 points within six months, and roughly 5 points after a year or more.⁴ The same review reported that all included studies found a significant increase in range of motion in the short term, and the effect was retained over the long term. Satisfaction hovered around 85%.

One study of 50 patients reported even more dramatic early results. The average pain score dropped from roughly 93% on a disability scale to about 18% at three weeks post-procedure, and the average disability score went from 95% down to about 17%, with no procedure-related complications.⁵

What the Shoulder Looks Like Decades Later

One of the most reassuring pieces of evidence comes from a study that followed patients for over two decades after MUA. At seven years, forward flexion had reached about 155 degrees and abduction about 175 degrees. Over the next 16 years, range of motion declined modestly by 8 to 23 degrees, but still matched the range of motion in the opposite (unaffected) shoulder. Pain remained very low at the final follow-up, averaging just 1.5 on a 10-point scale with exertion, 0.3 at rest, and 0.8 at night.⁶ That is about as close to “back to normal” as you get with any musculoskeletal intervention, and it held up over what amounts to most of a person’s remaining working life.

What Can Go Wrong

MUA involves forcibly breaking adhesions in a joint, so there is inherent risk. The most commonly discussed serious complications include fracture of the humerus (the upper arm bone) and rotator cuff tears, though both are uncommon in clinical series. A more nuanced picture comes from MRI studies performed after manipulation. One imaging study of shoulders manipulated under nerve block found that all of them showed some form of capsular tearing, which is actually the intended effect. However, the scans also revealed that four shoulders had labrum tears and 15 had bone bruising in the humeral head, even though no actual fractures occurred.⁷

The bone bruises and labrum tears did not appear to cause clinical problems in that particular group, and patients still improved. But the findings are a reminder that MUA is not a gentle procedure. It produces real structural changes beyond the intended capsule disruption. Most clinicians view these as acceptable trade-offs in the setting of a genuinely frozen shoulder, but the findings underscore why MUA should be reserved for cases where conservative management has truly failed, not used as a shortcut.

Nerve injury is another theoretical risk, though it appears quite rare. In one randomized trial, a single patient developed transient weakness in the deltoid muscle, attributed to adhesions involving the axillary nerve, and it resolved on its own within a day.⁸

MUA Versus Arthroscopic Capsular Release

Arthroscopic capsular release (ACR) is the main surgical alternative to MUA. Instead of breaking the capsule blindly through force, the surgeon uses a tiny camera and instruments to cut through the thickened tissue under direct visualization. In theory, this is more controlled. In practice, the clinical results are remarkably similar.

A meta-analysis comparing the two approaches found that ACR produced a slightly larger improvement in forward flexion at three and six months, but by 12 months there was no significant difference. Pain scores, external rotation, and overall complication rates were statistically comparable between the two procedures.⁹ A separate systematic review with meta-analysis found that ACR offered a slightly better long-term pain score, but the difference was smaller than the minimum amount considered clinically meaningful, meaning patients likely could not feel the difference. That same review also found that the ACR group had a significantly higher rate of severe complications compared to MUA.¹⁰

So the picture that emerges is this: ACR and MUA produce very similar outcomes for most patients. ACR may have a small edge in early flexion recovery, but MUA tends to carry a lower risk of serious complications and is a simpler, cheaper procedure. Many surgeons use MUA as the first-line procedural step and reserve ACR for patients who fail MUA or who have specific structural concerns that warrant direct visualization.

MUA Versus Injections and Hydrodilatation

Not everyone with a stubborn frozen shoulder needs to be put to sleep. Steroid injections (sometimes combined with hydrodilatation, where fluid is injected under pressure to stretch the capsule) are a less invasive alternative. The evidence comparing these approaches to MUA is mixed and somewhat humbling.

A systematic review found that MUA was not superior to cortisone injections with hydrodilatation for either pain reduction or functional improvement at three months, and the two remained equivalent beyond six months as well.¹¹ A head-to-head randomized trial comparing MUA with hydrodilatation found that the hydrodilatation group had significantly better pain and functional scores over a six-month follow-up, with 94% satisfaction compared to 81% in the MUA group, though range of motion improvement was similar in both groups.¹² Another randomized trial found no statistical differences at all between MUA and intra-articular steroid injections with distension, and recommended the injection approach because it carried fewer risks.¹³

These findings do not mean MUA is unnecessary. They suggest that for many patients, especially those with moderate stiffness or predominantly pain-driven symptoms, less invasive options can achieve equivalent results. MUA’s advantage shows more clearly when the dominant problem is severe mechanical restriction, the kind of stiffness where the arm simply cannot be moved past a certain point regardless of pain control.

Diabetes, Thyroid Disease, and Other Complicating Factors

If there is one group where MUA outcomes are consistently less impressive, it is people with diabetes. A study tracking MUA outcomes across different comorbidities found that patients with diabetes had significantly lower range-of-motion values even three months after the procedure compared to controls. Recovery speed was slowest in the diabetes group, followed by patients with thyroid disorders and then those with high cholesterol. About 91% of diabetic patients experienced what the researchers called “late recovery,” meaning their improvement was delayed and drawn out.¹⁴

Recurrence is also substantially more common in diabetic patients. One study found that about 36% of patients with diabetes required a repeat MUA compared to 15% of non-diabetic controls. Among diabetic patients with primary frozen shoulder (no obvious triggering event), nearly half experienced recurrence. Insulin dependence increased the recurrence rate further.¹⁵ The likely explanation is that diabetes promotes the excessive collagen deposition and inflammation that drives frozen shoulder in the first place, so the biological tendency to re-stiffen is stronger even after the capsule has been mechanically disrupted.

This does not mean people with diabetes should avoid MUA. It means they should expect a potentially slower recovery, a higher chance of needing a second procedure, and possibly more aggressive post-manipulation physical therapy to maintain what was gained.

When the Stiffness Comes Back

Recurrence can happen in anyone, not just diabetic patients. In a large series of 730 patients (792 shoulders), about 18% required a repeat MUA. Encouragingly, the second manipulation worked just as well as the first, producing similar improvements in shoulder scores regardless of how well the initial procedure had gone or how long after the first MUA the stiffness returned.¹⁶ So if your shoulder locks up again after a first manipulation, a second attempt is a reasonable option.

For long-term recurrence, the numbers are reassuring. A study tracking patients more than five years after their initial MUA found that only about 1.7% reported recurrence, and fewer than 1% needed a repeat manipulation at that distance.¹⁷ Combined with the 23-year follow-up data cited earlier, this suggests that for most people, the improvement from MUA is durable.

The Cost Question

MUA is substantially cheaper than arthroscopic capsular release. A cost-effectiveness analysis from the UK FROST trial, one of the largest randomized comparisons of frozen shoulder treatments, found that ACR cost about £1,734 more than structured physiotherapy and £1,457 more than MUA. MUA itself was only £276 more expensive than structured physiotherapy. When accounting for both cost and quality-of-life outcomes, MUA had the highest probability of being cost-effective at standard willingness-to-pay thresholds, at over 86%.¹⁸

A separate cost-effectiveness analysis compared MUA, ACR, and a newer approach called transarterial embolization (where tiny particles are injected into blood vessels feeding the inflamed capsule to cut off the inflammation’s blood supply). When the primary goal was restoring range of motion rather than relieving pain alone, MUA and ACR produced more quality-adjusted benefit at lower cost than embolization.¹⁹ The practical takeaway: MUA is the most cost-efficient procedural option for the majority of patients whose primary complaint is stiffness rather than pain.

How Fear of Movement Shapes Recovery

An underappreciated factor in how well people do after any shoulder procedure is their psychological relationship with pain and movement. Kinesiophobia, the fear of movement because it might cause injury or re-injury, and pain catastrophizing, the tendency to ruminate on and magnify pain, both independently predict worse outcomes after shoulder surgery. Higher baseline scores on scales measuring these traits were associated with greater postoperative pain in a study of shoulder surgery patients.²⁰

This matters for MUA patients because the window right after manipulation is critical. The capsule has been torn open, and if you do not move aggressively through physical therapy in the days and weeks that follow, the scar tissue can re-form rapidly. A patient who is terrified of moving their arm because it hurt so badly before the procedure is at a disadvantage compared to someone who dives into rehabilitation with confidence. Some clinics now screen for these psychological factors and offer brief cognitive-behavioral coaching before the procedure, though the practice is not yet widespread.

How MUA Fits into the Treatment Ladder

Frozen shoulder treatment tends to follow a stepwise approach. The first line is physical therapy, anti-inflammatory medications, and possibly a corticosteroid injection. If stiffness and pain persist after several months, clinicians consider escalation. Where MUA sits on that ladder depends partly on what the dominant problem is. When motion restriction is severe and pain is manageable, MUA’s ability to rapidly restore range of motion in a single session is hard to beat. When pain is the leading complaint, hydrodilatation or a series of steroid injections may produce equivalent results with less procedural risk.

For patients who have failed both conservative management and MUA, arthroscopic capsular release remains the backstop. It costs more and carries a higher rate of serious complications, but it allows the surgeon to see exactly what they are cutting. In rare cases where the diagnosis is uncertain or where there may be underlying structural problems like a rotator cuff tear hiding behind the stiffness, going straight to arthroscopy can be the smarter play because it combines treatment with a diagnostic look inside the joint.

The emerging technique of transarterial embolization is still very new but offers a different biological strategy entirely, targeting the inflammatory blood supply rather than the mechanical restriction. Early data suggest it may be most useful in patients whose dominant symptom is pain rather than stiffness, essentially the opposite patient profile from the one who benefits most from MUA. Whether embolization will carve out a lasting role in the treatment ladder remains to be seen as larger trials report their results.