Worth knowing Quality score: 64/100

Long-term outcome of acromioclavicular joint injuries in an urban population

BMJ open sport & exercise medicine · 2026

KeyOSS Oxford Shoulder ScorePROM Patient-Reported Outcome MeasureSD Standard Deviation

Most AC joint injuries, even higher-grade ones, do reasonably well at 7.6 years without surgery, but the high-grade subgroup is too small to be definitive.

The paper

Prospective cohort study, n=240/287 (84% response) with AC joint injuries, Oslo urban population, followed up at a mean of 7.6 years.

What they found

Mean Oxford Shoulder Score (OSS) was 41.8/48 (SD 8.6) with no significant difference between Rockwood type I and types II, III or V (p>0.05). Younger females aged 20-30 and males aged 20-40 scored below normative OSS data (p<0.05), and radiographic degenerative changes were present in 64% (96/150) but did not correlate with PROMs. Only 6 patients (3%) had undergone surgery.

Study design

Prospectively registered cohort study, single tertiary centre, n=287 all-comers with a diagnosed AC joint injury in 2013-2014, re-contacted in 2021 for long-term follow-up (7.6 years). This sits as cohort/observational evidence, below RCT level, but strengthened by being an unselected, consecutive, population-based sample rather than a surgical case series.

Methodology

The 84% response rate and long, fixed follow-up interval are strong points for an observational cohort, and prospective registration limits recall-based selection into the study. Key weaknesses: the clinical exam, Constant-Murley score and radiographs were done in only a subset (not the full 240), so the radiograph-PROM 'no correlation' finding rests on fewer patients; Rockwood type V comprised only 4% of respondents, so comparisons involving that group are explicitly underpowered; and there is no true comparator arm (surgical vs conservative) since treatment wasn't randomised, so any apparent equivalence across grades is association, not causation, and residual confounding (why some type III/V patients got surgery and others didn't) isn't addressed.

The appraisal

The headline, that OSS doesn't differ significantly by Rockwood grade, is statistically underpowered for the type V comparison (small n) rather than strong evidence of true equivalence, so it shouldn't be read as 'grade doesn't matter.' The absolute OSS of 41.8/48 is clinically good news regardless of grade, and the finding that radiographic degeneration is common but asymptomatic echoes what's known in other joints (knee, shoulder OA), reinforcing that imaging findings shouldn't drive treatment decisions in isolation. This is a reassuring descriptive natural-history study, not a comparative-effectiveness study of surgery versus conservative care.

The gap

The type V subgroup is tiny (about 4% of 240 respondents, roughly 10 people), so the 'no difference' finding for that grade is underpowered, not proof of true equivalence. Only 6 patients (3%) had surgery overall, so differential surgical allocation is at most a minor confound here, not the main limitation; the bigger gap is that outcomes were captured at a single cross-sectional timepoint 7.6 years out, with no baseline or interim data, so it's unclear whether the good scores reflect steady resolution, an early plateau, or simply that 72% of the cohort were low-grade (I/II) injuries to begin with.

Landmark context

This builds on the Rockwood classification itself, which has long been used to guide operative versus conservative decision-making for AC injuries, and adds to the ongoing clinical debate (particularly for type III, the 'Rockwood type III conundrum') about whether higher-grade injuries actually need surgery, lending long-term, population-based support to conservative-first approaches.

What to do Monday

Reinforces that conservative management remains reasonable as the default for most AC joint injuries, including many type III cases, since long-term shoulder function was generally good and surgery rates were low; it doesn't give grounds to abandon individualised assessment, especially for younger, more symptomatic patients who scored below norms.

Read the primary source ↗

The five papers that matter — every week, free.

We read the whole sports-medicine, rehab & performance literature and appraise the papers worth your time.

Get the free weekly issue

Read in your language: English · Français · Deutsch · Español · Italiano · Nederlands · Dansk · Svenska