Shoulder Replacement Risks and Complications: What Can Go Wrong and How Likely

Shoulder replacement is major joint surgery under anaesthetic, and its recognised complications are infection (roughly 1 in 100), nerve injury, dislocation or instability, loosening and wear of the components, and fracture around or near the implant, with reverse replacements carrying a higher overall complication rate than anatomic total ones. Serious complications are the exception rather than the rule, but they are real and worth understanding plainly before deciding¹.

When I signed the consent form for my reverse replacement, the surgeon read the risks out one by one, and I remember thinking the list sounded longer than the benefit. What I did not have then was any sense of how likely each item actually was, or which ones belonged to my reverse operation rather than the standard total everyone assumes. This is the account I wanted at that moment: each named complication, roughly how common it is, and what it means if it happens, checked line by line by a consultant shoulder surgeon. It sits under the main shoulder replacement guide.

How common are complications, and does the type of replacement matter?

Most people come through a shoulder replacement without a major complication, but a reverse replacement has a higher overall complication rate than an anatomic total replacement, the trade-off for working without a functioning rotator cuff. Reverse designs have expanded across cuff tear arthropathy, complex fractures and revision cases, and that broader, harder use partly explains the raised rate².

It helps to hold the risks against the result. Shoulder replacement is a high-satisfaction operation: most series report around 90% good or excellent results, and roughly 90% to 95% of patients have less pain and better function afterwards³. The list below is long because it is honest, not because most shoulders run into trouble. Which complications lean toward which operation is part of why the choice between an anatomic total and a reverse matters so much.

Infection

Infection affects roughly 1 in 100 primary shoulder replacements (about 1%, reported from 0 to 4%), and reverse replacements carry a somewhat higher rate; a deep infection reaching the implant can mean further surgery to wash out or exchange the components. A superficial wound infection may settle with antibiotics, but the joint itself is the concern, which is why any spreading redness, fever, or a weeping wound is treated as urgent¹.

The early weeks are when I watched my own wound most anxiously, because this is the complication with the clearest warning signs and the highest stakes. Shoulder infections can also be quiet, driven by low-grade organisms that take time to declare themselves. The full picture, including the slow-burning kind, is set out in shoulder replacement infection.

Nerve injury

The nerve most at risk is the axillary nerve, which runs close to the joint and can be stretched as the surgeon works around it; the usual result is temporary weakness or numbness that recovers over weeks to months, though occasionally it lasts. Nerve injury is an uncommon complication, and a full, permanent palsy is rarer still⁴.

One thing that confused me early on was telling ordinary block-related numbness apart from a nerve problem. The regional nerve block used for pain relief deliberately numbs the whole arm for the first day or so, which is expected and wears off; a genuine nerve injury is a different, longer story your team assesses at review. How the block works is set out in shoulder replacement anaesthesia.

Dislocation and instability

The artificial joint can come apart or feel unstable, and this is more common after a reverse replacement than an anatomic one. A dislocation may follow a specific movement in the early weeks or reflect the balance of the soft tissues around the implant; some are reduced (put back) without surgery, while recurrent instability sometimes needs a further operation⁴.

This is one of the clearest examples of a risk that belongs more to my reverse operation than to a standard total. The reverse design lifts the arm using the deltoid in place of a lost cuff, and that same reworked mechanics is what makes instability a little more likely, which is part of the honest bargain of a reverse shoulder replacement.

Loosening, wear and periprosthetic fracture

Over time the components can loosen or wear, a fracture can occur around or near the implant (a periprosthetic fracture), and in anatomic replacements a later rotator cuff failure or glenoid (socket) loosening can develop; each is a recognised reason for revision surgery. These are the longer-term, mechanical complications rather than the early ones¹.

A worn plastic surface or a loosening socket does not usually announce itself overnight; it shows up as returning pain or reduced function that a surgeon investigates with imaging. Because these problems accumulate with years and use, they feed directly into how durable the joint proves to be, which is the subject of how long a shoulder replacement lasts.

Acromial and scapular stress fracture

A stress fracture of the acromion or the scapular spine (bone at the top and back of the shoulder blade) is a complication specific to reverse replacements, reported in a small percentage of cases. It can cause pain and hold back the result, and it is one of the recognised downsides that long-term reverse series track alongside survivorship⁵.

This one rarely appears on the plain-English risk lists, yet it matters for anyone facing a reverse operation, because it is not a risk a standard total shares. It is another reason the reverse replacement is described honestly as carrying more to watch for, in exchange for restoring an arm that could not otherwise lift.

Stiffness, blood clots and the anaesthetic

Stiffness is a common issue that rehabilitation works against; blood clots (deep vein thrombosis or pulmonary embolism) are possible but rarer after shoulder surgery than after lower-limb operations, and the general risks of anaesthesia apply as with any major operation. These are the more general surgical risks that sit alongside the shoulder-specific ones¹.

Stiffness was the one I felt most in the first months, and it is also the one most under the patient’s influence, because the movement and strength keep improving with the rehabilitation rather than arriving on their own. That is why the physiotherapy carries so much of the eventual result, more than the operation alone.

When a complication means more surgery

Most complications are managed without a repeat operation, but a deep infection, a loosened or worn component, a periprosthetic fracture, recurrent dislocation, or a later cuff or glenoid failure can all lead to a revision; after the first decade the background risk of needing a revision is roughly 1% per year. A replacement put in at a younger age is therefore more likely to need redoing within a lifetime¹.

It is worth being clear that a revision is a bigger operation than the first, so the aim is always to avoid one, not to treat it as a routine backstop. The surgery also does not stop the underlying disease elsewhere in the body, and none of these risks can be reduced to zero. Weighing that honest list against the pain relief most people gain is the real decision from the other side of the operation.