Massive Rotator Cuff Tear With Severe Muscle Damage: Repair or Replacement?

What My Research Shows About Choosing Between Arthroscopic Repair and Reverse Shoulder Replacement

Massive rotator cuff tears are among the most complex problems I manage in shoulder surgery. When a tear is large — involving two or more tendons — and the muscles have undergone significant fatty degeneration, the surgical decision-making becomes genuinely difficult. Two fundamentally different approaches are available: try to repair the torn tendons arthroscopically, or bypass the rotator cuff entirely with a reverse total shoulder replacement.

Both operations can relieve pain and restore function. But they are not equivalent, and they are not interchangeable. Choosing between them requires weighing the patient's age, functional goals, severity of muscle damage, the duration and nature of their weakness, and whether the shoulder joint itself has developed arthritis.

A study from my practice at Oregon Shoulder Institute, published in The Orthopaedic Journal of Sports Medicine, directly compares outcomes of arthroscopic rotator cuff repair (ARCR) and reverse shoulder arthroplasty (RSA) in patients with massive tears and severe muscle degeneration — but without glenohumeral arthritis.

The Hardest Decision in Rotator Cuff Surgery

When a massive rotator cuff tear has been present long enough, the muscles attached to those tendons can begin to fill with fat — a process called fatty atrophy. This is graded on the Goutallier classification, from 0 (normal) to 4 (more fat than muscle). Grades 3 and 4 represent severe, largely irreversible muscle changes that significantly reduce the likelihood of a successful repair.

This study focused specifically on patients with Goutallier grade 3 or 4 changes. Conventional thinking has often suggested that at this level of muscle degeneration, repair is unlikely to heal and that shoulder replacement may be the better path. My research challenges that assumption, at least for appropriately selected patients without joint arthritis.

Both arthroscopic repair and reverse shoulder replacement improved pain and function significantly in patients with massive tears and severe muscle degeneration. But repair produced greater range of motion in the right patients — and even repairs that didn't fully heal still provided meaningful functional improvement.

The Study: My Patients, Compared Head to Head

This retrospective study analyzed 81 patients from my practice who underwent either arthroscopic rotator cuff repair (56 patients) or reverse shoulder arthroplasty (25 patients) for massive rotator cuff tears with Goutallier grade 3 or 4 muscle degeneration, without glenohumeral arthritis. All surgeries were performed by me, providing a consistent standard of technique across both groups.

The groups were not identical at baseline. RSA patients were older (average 71.7 years vs. 66.7 years), had more severe muscle degeneration across more tendons, higher tendon retraction, and were more likely to have pseudoparalysis. These differences reflect how I approach surgical decision-making.

All patients had a minimum two-year follow-up. ARCR patients were assessed with ultrasound at final follow-up to determine whether their rotator cuff had healed.

The Results: Both Work, But Repair Preserves More Motion

• 85.7% vs 92% — patient satisfaction, repair vs. replacement (no significant difference)
• 87.5% vs 92% — return to previous activities (no significant difference)
• 144° vs 113° — postoperative forward flexion (significant difference)
• 55% — of arthroscopic repairs healed on ultrasound at final follow-up

Pain scores, ASES shoulder function scores, and subjective shoulder value improved significantly in both groups. Where the groups diverged meaningfully was in range of motion.

Patients who underwent arthroscopic repair achieved significantly greater forward flexion — 144 degrees versus 113 degrees — and better internal rotation. RSA patients actually lost internal rotation on average (two vertebral levels), while repair patients maintained or modestly gained it.

What About the Repairs That Didn't Heal?

In this study, 45% of repairs showed incomplete healing on ultrasound. That is a meaningful retear rate — but it does not tell the full story.

When I compared patients whose repairs healed versus those whose didn't, pain scores, ASES scores, and satisfaction were statistically similar. The one area where healed repairs clearly outperformed was forward flexion: 150 degrees in healed patients versus 136 degrees in those with retears.

An unhealed repair is not a failed surgery. In this study, patients with retears still achieved significant improvements in pain and shoulder function — and were satisfied at similar rates to those whose repairs healed.

How I Decide Between Repair and Replacement

Factors that favor arthroscopic repair:

• Younger age — repair preserves future options, including conversion to replacement if needed later
• Acute or short-duration pseudoparalysis (under 6 months)
• Primary goals of pain relief — even unhealed repairs provide this reliably
• Goutallier grade 3 in one tendon with less severe changes in others
• Patient preference for avoiding joint replacement

Factors that favor reverse shoulder replacement:

• Older age with chronic pseudoparalysis lasting more than 6 months
• Goutallier grade 3 or 4 in multiple tendons with severe retraction
• Goals that include restored overhead strength and function
• Desire to avoid a second surgery if repair fails
• Overall health and activity demands that favor the predictability of replacement

Glenohumeral arthritis is a separate consideration — this study excluded those patients. When significant arthritis is present, replacement becomes the dominant consideration.

A Note on Pseudoparalysis

My own prior research has shown that arthroscopic repair reliably reverses pseudoparalysis in 90% of patients when the duration is short — around 3 months on average. However, when pseudoparalysis has been present for more than 6 months, recovery of overhead function after repair becomes less reliable. In those cases, reverse shoulder replacement is typically the more dependable choice.

In this study, 52% of RSA patients had pseudoparalysis compared to 21% in the ARCR group.

The Bottom Line

Massive rotator cuff tears with severe muscle degeneration are manageable with either approach — but they are not the same operation, and they don't produce the same outcomes. Both relieve pain reliably. Both produce meaningful functional improvement.

Where they diverge is in range of motion, which is significantly better after repair in appropriately selected patients. And while retear rates after repair are real, an unhealed repair is not a failure.

Patrick J. Denard, MD is a fellowship-trained orthopedic surgeon specializing in shoulder surgery at the Oregon Shoulder Institute in Medford, Oregon. He is the senior author of "Comparison of Clinical Outcomes Between Arthroscopic Rotator Cuff Repair and Reverse Total Shoulder Arthroplasty in Patients With Massive Rotator Cuff Tears and High-Grade Fatty Atrophy Without Glenohumeral Osteoarthritis," published in The Orthopaedic Journal of Sports Medicine (2024).

If you have questions about a massive rotator cuff tear or would like to discuss your surgical options, request an appointment at Oregon Shoulder Institute.